Testing the biocompatibility of a glutathione-containing intra-ocular irrigation solution by using an isolated perfused bovine retina organ culture model - an alternative to animal testing

The effects of a glutathione-containing intra-ocular irrigation solution, BSS Plus?, on retinal function and on the survival of ganglion cells in whole-mount retinal explants were studied. Evidence is provided that the perfused ex vivo bovine retina can serve as an alternative to in vivo animal testing. Isolated bovine retinas were prepared and perfused with an oxygen-saturated standard irrigation solution, and an electroretinogram was recorded to assess retinal function. After stable b-waves were detected, the isolated retinas were perfused with BSS Plus for 45 minutes. To investigate the effects of BSS Plus on photoreceptor function, 1mM aspartate was added to the irrigation solution in order to obtain a-waves, and the ERG trace was monitored for 75 minutes. For histological analysis, isolated whole retinal mounts were stored for 24 hours at 4°C, in the dark. The percentages of cell death in the retinal ganglion cell layer and in the outer and inner nuclear layers were estimated by using an ethidium homodimer-1 stain and the TUNEL assay. General swelling of the retina was examined with high-resolution optical coherence tomography. During perfusion with BSS Plus, no significant changes in a-wave and b-wave amplitudes were recorded. Retinas stored for 24 hours in BSS Plus showed a statistically significant smaller percentage (52.6%, standard deviation [SD] = 16.1%) of cell death in the retinal ganglion cell layer compared to the control group (69.6%, SD = 3.9, p = 0.0031). BSS Plus did not seem to affect short-term retinal function, and had a beneficial effect on the survival of retinal ganglion cells. This method for analysing the isolated perfused retina represents a valuable alternative for testing substances for their retinal biocompatibility and toxicity.     

Extracellular K+ activity changes related to electroretinogram components. II. Rabbit (E-type) retinas

Electroretinogram (ERG) and extracellular potassium activity (K+o) measurements were carried out in isolated superfused rabbit eyecup preparations under control conditions and during the application of pharmacological agents that selectively modify the light-responsive retinal network. Light-evoked K+o changes in the rabbit (E-type) retina resemble those previously described in amphibian (I-type) retinas. Different components of the light-evoked K+o changes can be distinguished on the bases of retinal depth, V vs. log I properties, and their responses to pharmacological agents. We find two separable sources of light-evoked increases in extracellular K+: a proximal source and a distal source. The properties of the distal light-evoked K+o increase are consistent with the hypothesis that it initiates a K+-mediated current through Müller cells that is detected as the primary voltage of the electroretinographic b-wave. These experiments also support previous studies indicating that both the corneal-positive component of c-wave and the corneal-negative slow PIII potential result from K+-mediated influences on, respectively, the retinal pigment epithelium and Müller cells.     

Gas chromatographic–mass spectrometric identification and quantitation of benzyl alcohol in serum after derivatization with perfluorooctanoyl chloride: a new derivative

Benzyl alcohol is commonly used as an antibacterial agent in a variety of pharmaceutical formulations. Several fatalities in neonates have been linked to benzyl alcohol poisoning. Most methods for measuring benzyl alcohol concentrations in serum utilize direct extraction followed by high-performance liquid chromatography. We describe here a novel derivatization of benzyl alcohol using perfluorooctanoyl chloride after extraction from human serum for analysis by gas chromatography–mass spectrometry (GC–MS). The derivative was eluted at a significantly higher temperature respective to underivatized molecule and the method was free from interferences from more volatile components in serum and hemolyzed specimens. Another advantage of this derivatization technique is the conversion of low-molecular-mass benzyl alcohol (M_r 108) to a high-molecular-mass derivative (M_r 504). The positive identification of benzyl alcohol can be achieved by observing a distinct molecular ion at m/z 504 as well as the base peak at m/z 91. Quantitation of benzyl alcohol in human serum can easily be achieved by using 3,4-dimethylphenol as an internal standard. The within run and between run precisions (using serum standard of benzyl alcohol: 25 mg/l) were 2.7% (mean=24.1, S.D.=0.66 mg/l, n=8) and 4.2% (mean=24.3, S.D.=1.03 mg/l, n=8), respectively. The assay was linear for the serum benzyl alcohol concentrations of 2 mg/l to 200 mg/l and the detection limit was 0.1 mg/l. We observed no carry-over (memory effect) problem in our assay as when 2 μl ethyl acetate was injected into the GC–MS system after analyzing serum specimens containing 200 mg/l of benzyl alcohol, we observed no peak for either benzyl alcohol or the internal standard in the total ion chromatogram.     

Simultaneous ex vivo Functional Testing of Two Retinas by in vivo Electroretinogram System

An In vivo electroretinogram (ERG) signal is composed of several overlapping components originating from different retinal cell types, as well as noise from extra-retinal sources. Ex vivo ERG provides an efficient method to dissect the function of retinal cells directly from an intact isolated retina of animals or donor eyes. In addition, ex vivo ERG can be used to test the efficacy and safety of potential therapeutic agents on retina tissue from animals or humans. We show here how commercially available in vivo ERG systems can be used to conduct ex vivo ERG recordings from isolated mouse retinas. We combine the light stimulation, electronic and heating units of a standard in vivo system with custom-designed specimen holder, gravity-controlled perfusion system and electromagnetic noise shielding to record low-noise ex vivo ERG signals simultaneously from two retinas with the acquisition software included in commercial in vivo systems. Further, we demonstrate how to use this method in combination with pharmacological treatments that remove specific ERG components in order to dissect the function of certain retinal cell types.     

Dissociation of field potential from neuronal activity in the isolated retina: Failure of the b-wave with normal ganglion cell response

The b-wave of the isolated rabbit retina was compared with the ganglion cell response to light before and after modification of the retina's incubating medium. Marked diminution of the b-wave, with no reduction in ganglion cell response, was observed under three experimental conditions: (1) following a short period of anoxia; (2) following a short period in 0.2 mM Ca⁺⁺; (3) in a small percentage of preparations, simply as a result of prolonged incubation in control medium. In contrast, a short period in 50 mM K⁺ led to a parallel fall and parallel recovery of both responses. It is apparent that under selected conditions the field potentials which constitute the b-wave are poorly correlated with the retina's neural activity.     

Four spectral classes of cone in the retinas of birds

Summary The spectral sensitivity of 15 species of birds has been measured by recording transretinal voltages from opened eyecups. With suitable combinations of colored adapting lights, we find that a variety of passerines have four peaks of photopic sensitivity, with maxima at 370, 450, 480, and 570 nm. Additional sensitivity maxima at 510 nm are found in some species. The spectral sensitivity functions are not altered by bathing the retinas in 50 mM sodium aspartate, suggesting that they reflect the properties of cones and do not result from inhibitory interactions between retinal interneurons.Comparison of the results with a general mathematical model that describes spectral sensitivity functions recorded extracellularly from populations of receptors in different states of adaptation (Goldsmith 1986) shows that the retinal spectral sensitivity functions are consistent with the presence of (at least) four types of cone, but indicate as well that many of the cones that are maximally sensitive in the blue and violet likely contain oil droplets that attenuate the deep violet and near uv.     

Effects of Cardiac Glycosides on Electrical Activity in the Isolated Retina of the Frog

Ouabain added to physiological salt solutions bathing the isolated frog retina irreversibly abolishes the electrical response to light (the electroretinogram or ERG). The time course of abolition depends on the concentration of ouabain in the medium and the surface of the retina to which it is applied. When the glycoside is placed on the receptor surface, in 7 min the ERG is completely eliminated by 10^-4 M ouabain and more than 90% inhibited by 3 x 10^-5 M ouabain. The effect is slower at lower concentrations and when the solution is applied to the vitreous surface of the retina. The evidence suggests that abolition of the ERG by ouabain is due principally to inhibition of the active transport of sodium: (a) Structurally modified glycosides which are considerably less potent inhibitors of alkali cation-activated ATPase activity in preparations of frog retinal outer segments are also poorer inhibitors of electrical activity in isolated retinas. (b) Replacing much of the sodium in the medium bathing the retina by choline, Tris, or sucrose significantly protects the retina from ouabain. It is suggested that in a standard sodium environment essentially constant activity of the sodium pump is required to prevent rapid and irreversible change. The cellular sites most critically dependent on the sodium pump have not been identified.     

P2X<Subscript>7</Subscript>R modulation of visually evoked synaptic responses in the retina

P2X₇Rs are distributed throughout all layers of the retina, and thus, their localisation on various cell types puts into question their specific site(s) of action. Using a dark-adapted, ex vivo mouse retinal whole mount preparation, the present study aimed to characterise the effect of P2X₇R activation on light-evoked, excitatory RGC ON-field excitatory post-synaptic potentials (fEPSPs) and on outer retinal electroretinogram (ERG) responses under comparable conditions. The pharmacologically isolated NMDA receptor-mediated RGC ON-fEPSP was reduced in the presence of BzATP, an effect which was significantly attenuated by A438079 and other selective P2X₇R antagonists A804598 or AF27139. In physiological Krebs medium, BzATP induced a significant potentiation of the ERG a-wave, with a concomitant reduction in the b-wave and the power of the oscillatory potentials. Conversely, in the pharmacologically modified Mg²⁺-free perfusate, BzATP reduced both the a-wave and b-wave. The effects of BzATP on the ERG components were suppressed by A438079. A role for P2X₇R function in visual processing in both the inner and outer retina under physiological conditions remains controversial. The ON-fEPSP was significantly reduced in the presence of A804598 but not by A438079 or AF27139. Furthermore, A438079 did not have any effect on the ERG components in physiological Krebs but potentiated and reduced the a-wave and b-wave, respectively, when applied to the pharmacologically modified medium. Therefore, activation of P2X₇Rs affects the function in the retinal ON pathway. The presence of a high concentration of extracellular ATP would most likely contribute to the modulation of visual transmission in the retina in the pathophysiological microenvironment.     

Extracellular K+ activity changes related to electroretinogram components. I. Amphibian (I-type) retinas

Electroretinographic (ERG) and extracellular potassium activity measurements were carried out in superfused eyecup preparations of several amphibians. Light-evoked changes in extracellular K+ activity were characterized on the bases of depth profile analysis and latency measurements and through the application of pharmacological agents that have selective actions on the retinal network. Three different extracellular potassium modulations evoked at light onset were identified and characterized according to their phenomenological and pharmacological properties. These modulations include two separable sources of light-evoked increases in extracellular K+: (a) a proximal source that is largely post-bipolar in origin, and (b) a distal source that is primarily or exclusively of depolarizing bipolar cell origin. The pharmacological properties of the distal extracellular potassium increase closely parallel those of the b-wave. A distal light-evoked decrease in extracellular potassium appears to be associated with the slow PIII potential, based on a combination of simultaneous intracellular Müller cell recordings and extracellular ERG and potassium activity measurements before and during pharmacological isolation of the photoreceptor responses. The extracellular potassium activity increases are discussed with respect to the Müller cell theory of b-wave generation.     

Light-evoked increases in extracellular K+ in the plexiform layers of amphibian retinas

Recordings of light-evoked changes in extracellular K+ concentration (delta[K+]o) were obtained in the retinas of frog and mudpuppy. In eyecup preparations, various recording approaches were used and provided evidence for a K increase near the outer plexiform layer (distal K increase). This distal K increase could be pharmacologically dissociated from the well-known, large K increase in the proximal retina by the application of ethanol and gamma-aminobutyric acid. The distal K increase also often showed surround antagonism. A retinal slice preparation was used to permit electrode placement into the desired retinal layers under direct visual control and without the risk of electrode damage to adjacent layers. In the slice, a distinct distal K increase was found in the outer plexiform layer, in addition to the prominent K increase in the inner plexiform layer. Compared with eyecups, only weak K increases were found in the nuclear layers of the slice. This suggests that the K responses observed in the nuclear layers of eyecups may be generated by K+ diffusing along the electrode track from the plexiform layers. In the context of current models of ERG b-wave generation, the magnitude of the recorded distal K increase, compared with the proximal K increase, seems too small to give rise to the b-wave. However, the distal K increase may be differentially depressed by electrode dead space. It is also possible that if certain aspects of the models of b-wave generation were modified, then the observed distal K increase could give rise to the b-wave.     

Quantum Relations of the Rat Electroretinogram

The rat retina is uniform and contains almost exclusively rods. Therefore the rat eye, when uniformly illuminated, produces a gross electroretinogram (ERG) which is simply related to the activity of the individual retinal sources of the ERG. Characteristics of ERG's are shown on an intensity scale of the average number of quanta absorbed per rod per stimulus flash obtained by direct accurate measurement of all quantities involved. An independent check on the accuracy of these measurements is applied to pigment-bleaching data reported by Dowling (1963). When ERG characteristics are placed on this scale it is found that: (a) The b-wave can usually be observed when fewer than one out of two hundred rods absorbs a quantum, the threshold being determined by the noise of the preparation. (b) Near threshold the b-wave amplitude is proportional to intensity. (c) The a-wave appears when there are more than two to four absorptions per rod per flash. (d) The b-wave latency decreases with intensity, and the amplitude becomes proportional to the logarithm of intensity when fewer than one out of ten rods absorbs a quantum. This implies that the b-wave sources must combine excitation from more than one rod (probably more than seven). Therefore the b-wave cannot arise from independent rods or rod-bipolar synapses, but probably reflects activity of entire inner nuclear layer cells.     

RETINAL CIRCADIAN RHYTHMS IN HUMANS*

Circadian rhythms in the retina may reflect intrinsic rhythms in the eye. Previous reports on circadian variability in electrophysiological human retinal measures have been scanty, and the results have been somewhat inconsistent. We studied the circadian variation of the electrooculography (EOG), electroretinography (ERG), and visual threshold (VTH) in subjects undergoing a 36h testing period. We used an ultrashort sleep-wake cycle to balance effects of sleep and light-dark across circadian cycles. Twelve healthy volunteers (10 males, 2 females; mean age 26.3 years, standard deviation [SD] 8.0 years, range 19–40 years) participated in the study. The retinal functions and oral temperature were measured every 90 min. The EOG was measured in the light, whereas the ERG and the VTH were measured in the dark. Sleep was inferred from activity detected by an Actillume monitor. The EOG peak-to-peak responses followed a circadian rhythm, with the peak occurring late in the morning (acrophase 12:22). The ERG b-wave implicit time peaked in the early morning (acrophase 06:46). No statistically significant circadian rhythms could be demonstrated in the ERG a-wave implicit time or peak-to-peak amplitude. The VTH rhythm peaked in the early morning (acrophases 07:59 for blue and 07:32 for red stimuli). All retinal rhythms showed less-consistent acrophases than the temperature and sleep rhythms. This study demonstrated several different circadian rhythms in retinal electrophysiological and psychophysical measures of healthy subjects. As the retinal rhythms had much poorer signal-to-noise ratios than the temperature rhythm, these measures cannot be recommended as circadian markers. (Chronobiology International, 18(6), 957–971, 2001)     

Sulfisoxazole, an endothelin receptor antagonist, protects retinal neurones from insults of ischemia/reperfusion or lipopolysaccharide

Endothelins exert pathological effects in the eye and much interest centres on their role in causing retinal neuronal death in ischemic diseases like glaucoma. In the present study the influence of the non-selective endothelin antagonist, sulfisoxazole on raised intraocular pressure-induced ischemia to the rat retina was investigated. Moreover, in vitro studies on primary rat retinal cultures were undertaken to see whether sulfisoxazole is able to blunt the toxic effect of lipopolysaccharide (LPS) to retinal neurones.

In order to determine whether sulfisoxazole provides protection to the retina the a- and b-wave amplitudes of the electroretinogram (ERG), the localisation of retinal choline acetyltransferase (ChAT), nitric oxide synthase (nNOS) and Thy-1 and the retinal mRNA levels of Thy-1 and FGF-2 were deduced in retinas subjected to ischemia in the absence or presence of sulfisoxazole. The results showed that the ischemia-induced changes to the a- and b-wave amplitudes of the ERG and changes associated with the localisation of ChAT, nNOS and Thy-1 to be significantly blunted by sulfisoxazole. However, while the ischemia-induced changes to Thy-1 and FGF-2 mRNAs were reduced by sulfisoxazole, the reduction was non-significant.

The in vitro studies provided support for the protective effect of sulfisoxazole. Here, it was clearly shown that sulfisoxazole attenuated the elevation of nitric oxide (deduced by measuring nitrite) and the reduction in numbers of GABA-containing neurones caused by LPS.

The present study provides evidence for the first time that endothelin antagonist can protect the retina from ischemic-like insults as occurs in glaucoma.     

Arginase 2 deletion reduces neuro-glial injury and improves retinal function in a model of retinopathy of prematurity

Background

Retinopathy of prematurity (ROP) is a major cause of vision impairment in low birth weight infants. While previous work has focused on defining the mechanisms of vascular injury leading to retinal neovascularization, recent studies show that neurons are also affected. This study was undertaken to determine the role of the mitochondrial arginine/ornithine regulating enzyme arginase 2 (A2) in retinal neuro-glial cell injury in the mouse model of ROP.

Methods and Findings

Studies were performed using wild type (WT) and A2 knockout (A2−/−) mice exposed to Oxygen Induced Retinopathy (OIR). Neuronal injury and apoptosis were assessed using immunohistochemistry, TUNEL (terminal deoxynucleotidyl transferase dUTP nick end) labeling and Western blotting. Electroretinography (ERG) was used to assess retinal function. Neuro-glial injury in WT ROP mice was evident by TUNEL labeling, retinal thinning, decreases in number of rod bipolar cells and glial cell activation as compared with room air controls. Significant reduction in numbers of TUNEL positive cells, inhibition of retinal thinning, preservation of the rod bipolar cells and prevention of glial activation were observed in the A2−/− retinas. Retinal function was markedly impaired in the WT OIR mice as shown by decreases in amplitude of the b-wave of the ERG. This defect was significantly reduced in A2−/− mice. Levels of the pro-apoptotic proteins p53, cleaved caspase 9, cytochrome C and the mitochondrial protein Bim were markedly increased in WT OIR retinas compared to controls, whereas the pro-survival mitrochondrial protein BCL-xl was reduced. These alterations were largely blocked in the A2−/− OIR retina.

Conclusions

Our data implicate A2 in neurodegeneration during ROP. Deletion of A2 significantly improves neuronal survival and function, possibly through the regulation of mitochondrial membrane permeability mediated apoptosis during retinal ischemia. These molecular events are associated with decreased activation of glial cells, suggesting a rescue effect on macroglia as well.     

The effect of linoleic acid and benzyl alcohol on the activity of glycosyltransferases of rat liver golgi membranes and some soluble glycosyltransferases

The effects of the membrane perturbing reagents linoleic acid and benzyl alcohol on the activities of four rat liver Golgi membrane enzymes, N-acetylglucosaminyl-, N-acetylgalactosaminyl-, galactosyl-, and sialytransferases and several soluble glycosyltransferases, bovine milk galactosyl- and N-acetylglucosaminyltransferases and porcine submaxillary N-acetylgalactosaminyltransferases have been studied. In rat liver Golgi membranes, linoleic acid inhibited the activities of N-acetylgalactosaminyl- and galactosyltransferases by 50% or greater, sialyltransferase by 10–15%, and N-acetylglucosaminyltransferase not at all. The isolated bovine milk N-acetylglucosaminyltransferase and porcine submaxillary N-acetylgalactosylaminyltranferase were not inhibited but bovine milk galactosyltransferase was inhibited by 95% or greater. The inhibition by linoleic acid on Golgi membrane galactosyltransferase appears to be a direct effect of the reagent on the enzyme. Incorporation of bovine milk galactosyltransferase into liposomes formed from saturated phospholipids, DMPC, DPPC, and DSPC (dimyristoyl-, dipalmitoyl-, and distearoylphosphatidylcholine) prevented inhibition of the enzyme activity suggesting that the lipid formed a barrier which did not allow linoleic acid access to the enzyme. The water soluble benzyl alcohol was more effective in inhibiting enzymes of the isolated rat liver Golgi complex. All four glycosyltransferases were inhibited, the N-acetylglucosaminyl- and N-acetylgalactosaminyltransferases by more than 95%. A higher concentration of benzyl alcohol was necessary to inhibit the galactosyltransferases than was required for the other Golgi enzymes. Benzyl alcohol also inhibited the isolated bovine milk N-acetylglucosaminyl- and galactosyltransferases 90% to 95%, respectively, but did not affect the isolated porcine submaxillary gland N-acetylgalactosaminyltransferase. Benzyl alcohol did not inhibit the milk galactosyltransferase incorporated into DMPC or DPPC liposomes but showed a complex effect on the activity of the enzyme incorporated into DSPC vesicles, a stimulation of activity at low concentrations followed by an inhibition. A lipid environment consisting of saturated lipids appears to present a barrier to inhibiting substances such as linoleic acid and benzyl alcohol, or lipid may stabilize the active conformation of the enzyme. The different effects of these reagents on four transferases of the Golgi complex suggest that the lipid environment around these enzymes may be different for each transferase.     

Electrical activity of the retina in different axial lengths of the eye in humans

In subjects with normal vision, we studied the relationship between the axial length of the eye and the amplitude-temporal characteristics of different types of ganzfield ERG. The amplitude of the ERG had the highest variability in the middle of the range of the axial eye lengths. The variability decreased at both ends of the range. With increasing axial length, the amplitude of the b-wave of all types of ERG decreased, and the latency decreased concurrently. The b-wave reflects the activity of the retinal ON-neurones depolarizing in response to the light stimulus. We assume that the decrease in the amplitude of b-wave may be related to the decreased number of photoreceptors and of neurons in the following retinal levels and/or increased inhibition in proximal retina, as well as an increase in relative activity of retinal OFF-neurones hyperpolarizing in response to the light stimulus.     

Nitrate Reduction in Different Grass Species

Optimal extraction conditions, assay conditions, and levels of nit rate reductase activity (NRA) were determined for eight forage grass species adaptable to growing conditions in western North Dakota. Optimal pH for extraction of the enzyme nitrate reductase (NADH: nitrate oxidoreductase) for these species ranged from 7.0 to 9.5, whereas assay pH was 7.6 in all eight species. Substrate concentrations ranged from 1.0 to 10.0 mM for maximum NRA with higher concentrations (100 mM) significantly inhibiting NRA. The enzyme was NADH₂ (0.1 to 0.2 mM for maximum activity) specific. Enhancement of maximum activity with the addition of cysteine during extraction was species dependent; six species required high cysteine concentrations between 5 mM and 10 mM and one species required only a 2.5 mM concentration. The degree of sulfhydryl protection offered by cysteine also varied. Comparisons were made between in vivo and in vitro assay methods. Ratios of in vitro to in vivo NRA ranged from 2.2. to 10.8. Use of bovine serum albumin as a protein stabilizer during extraction increased the measurable NRA in some species. Applications of nitrate reductase assay techniques to field work will be discussed.     

Valproate Increases Glutaminase and Decreases Glutamine Synthetase Activities in Primary Cultures of Rat Brain Astrocytes

Abstract: It has been proposed that hyperammonemia may be associated with valproate therapy. As astrocytes are the primary site of ammonia detoxification in brain, the effects of valproate on glutamate and glutamine metabolism in astrocytes were studied. It is well established that, because of compartmentation of glutamine synthetase, astrocytes are the site of synthesis of glutamine from glutamate and ammonia. The reverse reaction is catalyzed by the ubiquitous enzyme glutaminase, which is present in both neurons and astrocytes. In astrocytes exposed to 1.2 mM valproate, glutaminase activity increased 80% by day 2 and remained elevated at day 4; glutamine synthetase activity was decreased 30%. Direct addition of valproate to assay tubes with enzyme extracts from untreated astrocytes had significant effects only at concentrations of 10 and 20 mM, When astrocytes were exposed for 4 days to 0.3, 0.6, or 1.2 mM valproate and subsequently incubated with l -[U-¹⁴C]glutamate, label incorporation into [¹⁴C]glutamine was decreased by 11, 25, and 48%, respectively, and is consistent with a reduction in glutamine synthetase activity. Label incorporation from l -[U-¹⁴C]glutamate into [¹⁴C]aspartate also decreased with increasing concentrations of valproate. Following a 4-day exposure to 0.6 mM valproate, the glutamine levels increased 40% and the glutamate levels 100%. These effects were not directly proportional to valproate concentration, because exposure to 1.2 mM valproate resulted in a 15% decrease in glutamine levels and a 25% increase in glutamate levels compared with control cultures. Intracellular aspartate was inversely proportional to all concentrations of extracellular valproate, decreasing 60% with exposure to 1.2 mM valproate. These results indicate that valproate increases glutaminase activity, decreases glutamine synthetase activity, and alters Krebs-cycle activity in astrocytes, suggesting a possible mechanism for hyperammonemia in brain during valproate therapy.     

Phosphomannopentaose sulfate (PI-88) inhibits retinal leukostasis in diabetic rat

Retinal leukostasis, mediated by intracellular adhesion molecule-1 (ICAM-1) and vascular endothelial growth factor (VEGF), has been implicated in the pathogenesis of early diabetic retinopathy. Phosphomannopentaose sulfate (PI-88) is a highly sulfonated oligosaccharide which inhibits heparanase activity and competes with heparan sulfate binding to growth factors. In this study, we evaluated whether PI-88 could inhibit retinal leukostasis in strepotzotocin(STZ)-induced diabetic rat and elucidated the possible mechanisms. Diabetes was induced in Sprague-Dawley rats by intraperitoneal injection (i.p.) of STZ. Three months after induction, diabetic rats were administered PI-88 (25 mg/kg body weight) or vehicle solution daily via i.p. for 14 consecutive days. Leukostasis was analyzed on retinal flatmounts by concanavalin A and CD45 immunofluorescence staining. Retinal function was analyzed by electroretinography (ERG). ICAM-1 and VEGF levels in retinas were studied by Western blot and enzyme-linked immunosorbent assay (ELISA) respectively. The systemic administration of PI-88, but not vehicle, significantly decreased the number of adherent leukocytes in retinas by 52.24% (P < 0.001) and led to significant preservation (about 50%, P < 0.001) of scotopic ERG a- and b-wave amplitudes in treated diabetic rats as compared to those of diabetic control rats. These changes were associated with downregulation of ICAM-1 (45%, P < 0.001) and VEGF (26.83 ± 2.01 versus 40.8 ± 3.24 pg/mg, P < 0.01) in retinas of PI-88 treated diabetic rats as compared to those of diabetic control rats. PI-88 significantly inhibited retinal leukostasis and reversed retinal dysfunction by a mechanism that may include decreased ICAM-1 and VEGF expression in diabetic rats. Our data suggests that PI-88 is a promising agent for the treatment of diabetic retinopathy.