Skip to main content
NCBI home page
The FASEB Journal logoLink to The FASEB Journal
. 2011 Apr;25(4):1284–1294. doi: 10.1096/fj.10-160051

Noninvasive molecular imaging reveals role of PAF in leukocyte-endothelial interaction in LPS-induced ocular vascular injury

Rebecca C Garland *,†,‡,§,1, Dawei Sun *,†,‡,§,‖,1, Souska Zandi *,†,‡,§, Fang Xie *,†,‡,§, Sepideh Faez ‡,§, Faryan Tayyari ‡,§, Sonja A F Frimmel ‡,§, Alexander Schering *,†,‡,§, Shintaro Nakao ‡,§, Ali Hafezi-Moghadam *,†,‡,§,2
PMCID: PMC3058708  PMID: 21257713

Abstract

Uveitis is a systemic immune disease and a common cause of blindness. The eye is an ideal organ for light-based imaging of molecular events underlying vascular and immune diseases. The phospholipid platelet-activating factor (PAF) is an important mediator of inflammation, the action of which in endothelial and immune cells in vivo is not well understood. The purpose of this study was to investigate the role of PAF in endothelial injury in uveitis. Here, we use our recently introduced in vivo molecular imaging approach in combination with the PAF inhibitors WEB 2086 (WEB) and ginkgolide B (GB). The differential inhibitory effects of WEB and GB in reducing LPS-induced endothelial injury in the choroid indicate an important role for PAF-like lipids, which might not require the PAF receptor for their signaling. P-selectin glycoprotein ligand-1-mediated rolling of mouse leukocytes on immobilized P-selectin in our autoperfused microflow chamber assay revealed a significant reduction in rolling velocity on the cells' contact with PAF. Rolling cells that came in contact with PAF rapidly assumed morphological signs of cell activation, indicating that activation during rolling does not require integrins. Our results show a key role for PAF in mediating endothelial and leukocyte activation in acute ocular inflammation. Our in vivo molecular imaging provides a detailed view of cellular and molecular events in the complex physiological setting.—Garland, R. C., Sun, D., Zandi, S., Xie, F., Faez, S., Tayyari, F., Frimmel, S. A. F., Schering, A., Nakao, S., Hafezi-Moghadam, A. Noninvasive molecular imaging reveals role of PAF in leukocyte-endothelial interaction in LPS-induced ocular vascular injury.

Keywords: platelet-activating factor, leukocyte recruitment, endotoxin-induced uveitis, P-selectin glycoprotein ligand-1, ginkgolide B, WEB 2086

Inflammation, the underlying condition in a host of diseases, is characterized, in its early stages, by the recruitment of leukocytes to the activated endothelium. On injury, endothelial cells express adhesion molecules, such as P-selectin and intracellular adhesion molecule-1 (ICAM-1), that facilitate the multistep recruitment cascade (1–3). The binding of endothelial P-selectin to its ligand, P-selectin glycoprotein ligand-1 (PSGL-1), on the leukocyte surface initiates the leukocyte adhesion cascade through tethering and rolling of leukocytes (4–6). Firm leukocyte adhesion depends on and is followed by these first events.

To detect the endothelial molecules that facilitate the recruitment cascade in living animals, we (7–9) recently introduced noninvasive molecular imaging in the eye. In this technique, fluorescent microspheres (MSs), of subcellular dimensions, are conjugated with ligands or antibodies to 1 or more endothelial surface molecules of interest (8, 10). After systemic injection, the interactions of these MSs with the endothelium of the retinal and choroidal vessels of live animals are studied by scanning laser ophthalmoscopy (SLO) under normal or inflammatory conditions.

Ocular inflammation or uveitis, often a manifestation of systemic disorders, is a major cause of vision loss (11). An established model of uveitis is the endotoxin-induced uveitis (EIU) in rats in which, hours after a footpad injection of lipo-polysaccharide (LPS), a considerable infiltration of polymorphonuclear granulocytes and monocytes occurs in the retina, cornea, and other ocular tissues (12, 13). Because of the cornea's transparency, the eye is uniquely suited for in vivo investigation of the contribution of specific mediators of the inflammatory immune response in the choroidal or barrier-privileged retinal vasculature.

Platelet-activating factor (PAF) is a signaling phospholipid, synthesized in endothelial and other cells in response to inflammatory stimuli. After synthesis, it resides on the endothelial surface, where it activates adhering polymorphonuclear neutrophils by binding to the G-protein-linked receptor on the polymorphonuclear neutrophil membrane (14). PAF also plays an important part in the rapid translocation of P-selectin from cytoplasmic Weibel-Palade bodies to the endothelial surface (15, 16). However, the mechanistic details of PAF's immune-regulatory actions in vivo remain to be elucidated. In an autoperfusion chamber system, where endogenous mouse leukocytes interact with immobilized adhesion molecules, we examined the direct effect of PAF on leukocyte rolling and adhesion (13, 17).

A competitive inhibitor of PAF, WEB 2086 (WEB; IC50: 23±10.4 nM; refs. 18–20), is anti-inflammatory when administered systemically and does not bind to tissues of PAFR−/− mice (21). Ginkgolide B (GB; IC50: 2.2±0.8 μM), a PAF inhibitor isolated from the extract of ginkgo tree leaves (22), reduces LPS-induced leukocyte accumulation in mesenteric vessels of rats (23). Although acting like a competitive inhibitor of PAF, based on biological assays cited above, GB has also more recently been shown to accelerate the activity of PAF-acetyl hydrolase (PAF-AH), the endogenous inactivator of PAF (24). Plasma PAF-AH is a lipoprotein that circulates mainly in HDL and LDL particles and is responsible for the short half-life of PAF in the blood (25, 26). PAF activities are mimicked by physiological analogues, “PAF-like lipids,” derived from the oxidation of certain phospholipids, which also bind to the PAF receptor (14, 27) but with a lower affinity than that of PAF (28, 29). LPS administration to rats is accompanied by oxidative stress, which results in increased formation of PAF-like lipids in these animals and, at the same time, increases PAF-AH expression in the liver, primarily in macrophages. This increase is reduced by half in the presence of PAF inhibitors (30). PAF-like lipids are associated with lipoprotein (31) that circulates in the bloodstream and, unlike most PAF, are not bound to the endothelium and therefore are not localized to specific inflammatory loci (32).

Using recombinant (r)PSGL-1-conjugated MSs as imaging agents in EIU animals, we investigated the contribution of PAF-induced P-selectin translocation to areas of ocular vascular injury. To evaluate the effect PAF in retinal and choroidal vessels, we intravitreally injected PAF and compared leukocyte and MS adhesion in the retina and choroid with that in the contra lateral control eye. By measuring the relative effects of 2 differentially acting PAF inhibitors on leukocyte and MS accumulation in the retinal and choroidal vessels of EIU rats, we addressed the possibility that PAF or PAF-like lipids, through a PAFR-independent pathway, produce additional vascular injury in LPS-treated animals.

MATERIALS AND METHODS

EIU

All experiments were performed in accordance with the Association for Research in Vision and Ophthalmology statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the Animal Care and Use Committee of the Massachusetts Eye and Ear Infirmary. Male Lewis rats (8 to 10 wk old) were obtained from Charles River (Wilmington, MA, USA). Uveitis was induced in rats by injecting 100 μg of LPS (Salmonella typhimurium; Sigma Chemical, St. Louis, MO, USA) diluted in 0.1 ml sterile saline into 1 hind footpad of each animal (12). Control animals received a footpad injection of vehicle (saline). Animals were maintained in an air-conditioned room with a 12-h light-dark cycle and were given free access to water and food until used for the experiments.

The inhibitors ginkgolide B (BN 52021) and WEB 2086 (Biomol, Plymouth Meeting, PA, USA) were administered intraperitoneally, 2.5 mg/animal.

Preparation of the molecular imaging probes

Carboxylated, fluorescent MSs (2 μm; Polysciences, Warrington, PA, USA) were covalently conjugated to protein G (Sigma) using a carbodiimide-coupling kit (Polysciences; ref. 10). rPSGL-Ig (Y's Therapeutics, San Bruno, CA, USA) was incubated with the MSs at 0.4 mg/ml overnight at room temperature. MSs were washed in PBS with 1% BSA before use in vivo. We injected 6 × 108 fluorescent MSs intravenously into each rat, 30 min before perfusion.

Evaluation of MS adhesion in vivo

To evaluate MS adhesion in the retinal and choriocapillaris vessels in normal and EIU animals, a scanning laser ophthalmoscope (HRA2; Heidelberg Engineering, Dossenheim, Germany), coupled with a computer-assisted image analysis system, was used to make continuous high-resolution fundus images. An argon blue laser was used as the excitation light, with a regular emission filter for fluorescein angiography, as the excitation (441 nm) and emission (488 nm) maxima of the MSs are comparable with those of sodium fluorescein. SLO images were obtained at a 30° angle at 15 frames/s and digitally recorded for further analysis. MSs were injected 24 h after LPS treatment, and images were recorded 30 min after MS injection.

Rats were anesthetized with xylazine hydrochloride (10 mg/kg) and ketamine hydrochloride (50 mg/kg), and their pupils were dilated with 0.5% tropicamide and 2.5% phenylephrine hydrochloride. A contact lens was used to retain corneal clarity throughout the experiment. Animals were placed on a platform, allowing flexible positioning of the animals in relation to the SLO. With a 30.5-gauge needle, MSs (6×108/ml in saline) were injected slowly (within 60 s) into the tail vein.

At 30 min after initial injection of the conjugated MS, the number of free-flowing MS in the retinal vessels and the choriocapillaris of normal and EIU rats was substantially diminished, presumably due to the interaction of the MS with the endothelium of the vessels throughout the body. This allowed us to conveniently identify and quantify the number of accumulated MSs in the retinal vessels and choriocapillaris as distinct stationary fluorescent marks with high contrast against the nonfluorescent background. ImageJ 1.41 software (U.S. National Institutes of Health, Bethesda, MD, USA) was used for image analysis. For automated quantification of the number of bound MSs in the choriocapillaris microvasculature, the confocal images were merged, and subsequently the area of the bright spots was measured in ImageJ 1.41.

Quantification of firm leukocyte and MS adhesion

At 24 h after LPS injection, animals were perfused with 50 ml of PBS to remove intravascular content, including nonadherent leukocytes and MSs. Under deep anesthesia [0.1 ml/kg of xylazine (100 mg/ml)/ketamine (20 mg/ml), 1:1], perfusion was performed with 30 ml rhodamine-conjugated concanavalin A (ConA; 40 μg/ml in PBS, pH 7.4, 5 mg/kg BW) to label adherent leukocytes and vascular endothelial cells, followed by flushing with 10 ml of PBS to remove residual unbound material. The eyes were enucleated, and retinas were carefully removed and flat mounted in a mounting medium (Vector Laboratories, Burlingame, CA, USA). Each retina was imaged with an epi-fluorescence microscope (DM RXA; Leica, Solms, Germany), and the number of adherent leukocytes in veins and arteries, as well as the total number of MSs in each retina, was counted.

Autoperfused microflow chamber assay: evaluation of leukocyte adhesion

Leukocyte rolling and adhesion to immobilized P-selectin was analyzed using our autoperfused microflow chamber assay (17). Briefly, translucent microslides (inner diameter 0.4×0.04 mm; VitroCom, Mountain Lakes, NJ) were coated with recombinant murine P-selectin (5 μg/ml; R&D, Minneapolis, MN, USA) at 4°C overnight. P-selectin-coated microslides were connected to biocompatible polyester tubing (PE10; Becton Dickinson, Franklin Lakes, NJ, USA) at both ends. The tubing and microslide were incubated with 1% BSA (Sigma-Aldrich) for 1 h to block nonspecific leukocyte interactions with the inner surfaces. Subsequently, the tubing ends were microsurgically connected to the right carotid artery and the left jugular vein of an anesthetized mouse, as described previously (17).

In this model, blood flows from the carotid artery into the biocompatible inlet tubing, then through the microslides, proceeds through outlet tubing, and reenters the animal's body via the jugular vein. To regulate the blood velocity, the diameter of the inlet tubing is altered by adjustable screw valves. To continuously measure the blood pressure, microtransducers (Harvard Apparatus, Holliston, MA, USA) are attached to the chamber by 3-way connectors embedded before the microslide. The analog output of the pressure control unit was digitalized by an A/D converter (ML785 PowerLab/8SP; ADInstruments; Colorado Springs, CO, USA) connected to a Macintosh computer (Apple, Inc., Cupertino, CA, USA) running CHART software (ADInstruments). The pressure values were used to calculate the shear stress (τ) using our previously described custom software, Rectflow (17).

Under an upright, fixed-stage intravital microscope (Leica), leukocyte rolling on immobilized P-selectin was videotaped using a color charge-coupled device camera (Dage, Stamford, CT, USA) for subsequent analysis. Leukocyte rolling velocity was defined as displacement of individual cells interacting with the chamber surfaces over time (Δd/Δt). The displacement of leukocytes was determined using Exbem 3.0 software (Pixlock, Muenster, Germany).

Statistical analysis

Student's t test was used for comparison of groups. All values were expressed as means ± se. Values of P < 0.05 were considered statistically significant.

RESULTS

PAF-induced leukocyte adhesion in retinal and choroidal vessels

To study the direct effect of PAF on leukocyte-endothelial interaction in the fundus vasculature, PAF (5 μg) was injected into the vitreous, and leukocyte adhesion was quantified in retinal and choroidal vessels of the injected eyes and in the contralateral control eyes (Fig. 1A). Animals were perfused with rhodamine-conjugated ConA to stain the vasculature and firmly adhering leukocytes, and subsequently retinal and choroidal flat mounts were prepared for quantification of firmly adhering leukocytes. PAF significantly increased leukocyte adhesion to retinal arteries (Fig. 1B) and veins (Fig. 1C), compared with the constitutive levels in control eyes. However, no increase was seen in the choroid, possibly because of exclusion by the outer blood/retinal barrier (Fig. 1D).

Figure 1.

Figure 1.

Open in a new tab

PAF-induced leukocyte adhesion in retinal and choroidal vessels. PAF (5 μg) was injected into the vitreous of Lewis rats; 2 h later, animals were perfused, and the vasculature was ConA stained. Subsequently, retinal and choroidal flat mounts were prepared, and number of firmly adhering leukocytes was quantified by counting. A) Representative micrographs of firmly adhering leukocytes (arrows) in retinal vessels. Higher magnification micrographs (bottom panels) show sections of the retinal veins. B) Numbers of leukocytes in retinal arteries. C) Numbers of leukocytes in retinal veins. D) Numbers of firmly adhering leukocytes in choroidal vessels. Bars are means ± se.

PAF inhibition substantially reduces retinal vascular inflammation

To investigate the role of PAF in uveitis, we quantified firm leukocyte adhesion in retinal vessels of normal and EIU animals with or without inhibitor treatment. To inhibit PAF, EIU rats were intraperitoneally injected with 2 differently acting inhibitors, WEB (2.5 mg), GB (2.5 mg), or both, 30 min before LPS injection. The amounts of inhibitors were those that gave maximum inhibition of leukostasis. After 24 h, flat mounts of retinas from ConA-injected rats were prepared, and adherent leukocytes in the retinal arteries and veins were counted. Compared with normal animals, uveitic rats showed a substantial increase in leukocyte accumulation both in retinal arteries and veins. Pretreatment with WEB or GB significantly reduced the number of firmly adhering leukocytes in retinal arteries and veins of EIU animals (Fig. 2A). The absolute numbers of adherent leukocytes were substantially lower in flat mounts of retinal arteries (Fig. 2B) than in veins (Fig. 2C). In the choroid, the pattern of leukocyte inhibition by the inhibitors was similar to that in the retinal veins (Fig. 2D).

Figure 2.

Figure 2.

Open in a new tab

Effect of PAF inhibition on leukocyte accumulation in retinal and choroidal vessels of EIU rats. To inhibit PAF, animals were intraperitoneally treated with WEB (2.5 mg), GB (2.5 mg), or both; 30 min minutes later, EIU was induced by toepad injection of LPS (100 μg/kg), and 24 h later, animals were perfused with rhodamine-conjugated ConA to remove the intravascular content and stain vessels and firmly adhering leukocytes. A) Representative micrographs of adherent leukocytes (arrows) in retinas from control and PAF-inhibitor-treated EIU rats. Higher magnification micrographs (bottom panels) show sections of the retinal veins. B) Quantification of firmly adhering leukocytes in retinal arteries. C) Firmly adhering leukocytes in retinal veins. D) Firmly adhering leukocytes in choroidal vessels.

Molecular imaging reveals PAF-induced retinal and choroidal vascular injury in vivo

To investigate the role of PAF in ocular vascular injury, we used our molecular imaging approach to quantify the amount of endothelial P-selectin expression in vivo. MSs, conjugated with the P-selectin ligand PSGL-1, were injected into rats 24 h after intravitreal injection of PAF (5 μg) or vehicle. With the use of SLO to image the fundus (Fig. 3A), accumulation of the MSs was evident as bright spots in the retinal or choroidal vessels, depending on the depth of focus (Fig. 3B). Intravitreal PAF caused a substantial increase in leukocyte adherence in retinal vessels, indicating ocular damage (Fig. 3C). Compared with normal controls, PAF had no effect on MS accumulation in choroidal vessels (Fig. 3D). As in the case of leukocytes, PAF may not have crossed the outer blood/retinal barrier to increase P-selectin expression in choroidal vessels. Ex vivo quantification of MSs in retina (Fig. 3E) and choroids (Fig. 3F) in normal and PAF-treated eyes confirmed the in vivo MS binding results.

Figure 3.

Figure 3.

Open in a new tab

Molecular imaging with rPSGL-1-conjugated fluorescent MSs reveals role of PAF in endothelial injury in vivo. Endothelial injury in the fundus vasculature in vivo was quantified using our adhesion-molecule-conjugated imaging agents in normal animals with and without intravitreal PAF injection. At 24 h after PAF-injection, rPSGL-1-conjugated MSs were systemically injected; 30 min later, the fundus was imaged using SLO (30° angle at 15 frames/s). Number of MSs in this region was quantified. A) Schematic of our in vivo molecular imaging approach. B) Representative SLO still images showing adhering MSs in retinas of live animals. Bright spots indicate adhering MSs that resisted the blood flow. C) In vivo quantification of the number of accumulated MSs in retinal vessels, indicating the level of endothelial injury. D) In vivo quantification of MS accumulation in the choroidal vessels of control and PAF-injected animals. E) Ex vivo quantification of the number of MSs in flat mounts of retinal vessels. F) Ex vivo quantification of the number of MSs in flat mounts of choroidal vessels.

Effect of PAF inhibition in retinal and choroidal vascular injury in EIU

To evaluate the contribution of endogenous PAF to leukostasis in EIU animals, we intraperitoneally injected PAF inhibitors 30 min before LPS administration. Using in vivo imaging, we quantified adherent MSs in vessels of normal and inhibitor-treated animals (Fig. 4A).

Figure 4.

Figure 4.

Open in a new tab

Effect of PAF inhibition on retinal and choroidal P-selectin up-regulation. To quantify the level of P-selectin expression in retinal and choroidal endothelium of normal and EIU rats, with and without PAF inhibition, we injected rPSGL-1-conjugated MSs, 24 h after LPS injection, and performed in vivo molecular imaging of the fundus vessels or ex vivo epifluorescence microscopy of retinal and choroidal flat mounts. A) In vivo binding analysis in choroidal vessels was performed 30 min after MS injection. B) Quantification of MS binding to choroidal microvessels in normal and EIU rats with or without inhibitor pretreatment. C) Representative retinal flat mounts from normal and EIU animals with or without inhibitor treatments. Bright green spots, adhering MSs; round red cells in vessels, firmly adhering leukocytes. D) Quantification of the number of accumulated rPSGL-1-conjugated MSs in the retinal vessels of normal and EIU animals with or without PAF inhibitor treatment. E) Representative choroidal flat mounts from normal and EIU animals with or without inhibitor treatments. F) Quantification of the number of accumulated rPSGL-1-conjugated MSs in the choroidal vessels of normal and EIU animals with or without PAF inhibitor treatment. Bars are means ± se.

In choroids, measured in vivo, WEB treatment of EIU animals only slightly reduced accumulation of rPSGL-1-conjugated MSs, while GB reduced it almost to control levels (Fig. 4B).

Ex vivo retinal flat mounts showed a significant increase in rPSGL-1-conjugated MS binding in EIU animals (Fig. 4C). More MSs adhered in retinal veins than in arteries, matching the differences in leukocyte adhesion between the 2 types of vessels. Pretreatment with WEB, GB, or both inhibitors reduced MS accumulation to control levels (Fig. 4D).

In choroidal flat mounts, MS binding was significantly higher in the EIU animals (Fig. 4E). In line with the in vivo results, WEB treatment did not reduce accumulation of the rPSGL-1-conjugated MSs in the choroids of EIU animals, while there was a significant reduction with GB (Fig. 4F). These results suggest that PAFR-independent entities, perhaps PAF-like lipids, played a significant role in choroidal accumulation of MSs in these animals.

In vivo detection of firmly adhering leukocytes by our molecular imaging agents

In addition to endothelial P-selectin, rPSGL-1-conjugated MSs detect adhering leukocytes (8), since PSGL-1 is also a ligand for leukocyte L-selectin (Fig. 5A). Therefore, we quantified the number of leukocyte-bound MSs expressed as ratios of total adhering MSs, MSLeu/MStotal, in retinal vessels of EIU animals with the various treatments. The data show that much of the increase in MS binding in EIU animals is due to MS adherence to the increased number of bound leukocytes. PAF inhibition significantly reduced MSLeu/MStotal in EIU animals (Fig. 5B).

Figure 5.

Figure 5.

Open in a new tab

Ratios of number of MSs binding to leukocytes to that of total MSs counted. To distinguish the binding of rPSGL-1-conjugated MSs to activated endothelial cells, from those bound to adherent leukocytes, we calculated the ratio between them under various conditions. A) Confocal micrograph showing adhesion of an MS to a firmly adhering leukocyte. B) Ratios of MSleu to MStotal showing the effect of EIU and PAF inhibitors on MS binding to leukocytes.

Role of PAF in constitutive endothelial function in retinal and choroidal vessels

To explore the role of PAF in the normal retina, we first treated rats with the PAF inhibitors and quantified the amount of leukocyte recruitment and endothelial injury in their fundus vessels. In retinal arteries, inhibitors caused a slight increase in the few adhering leukocytes, while the difference was only significant in GB-treated animals (Fig. 6A). In the retinal veins, both inhibitors slightly increased the control value; however, the difference was not statistically significant (Fig. 6B). The binding of rPSGL-1-conjugated MSs to retinal and choroidal vessels did not differ significantly between the vehicle and the PAF-inhibitor-treated animals (Fig. 6C, D). This suggests that PAF inhibition in normal eyes may not affect P-selectin expression. However, in retinal arteries GB might up-regulate other adhesion molecules that increase leukocyte adhesion. Even though the result was statistically significant, such small differences might not have physiological significance.

Figure 6.

Figure 6.

Open in a new tab

Effect of PAF inhibitors on leukocyte and MS binding to retinal and choroidal vasculatures of normal rats. To investigate the constitutive role of PAF in endothelial function, the level of endothelial activation and leukocyte adhesion was quantified in animals 24 h after treatments. Adherent leukocytes were counted in retinal and choroidal flat mounts 24 h after intraperitoneal injection of PAF inhibitors to normal rats. A) Number of firmly adhering leukocytes in retinal arteries after GB intraperitoneal injection. B) Number of firmly adhering leukocytes in retinal veins after GB intraperitoneal injection. C) Number of bound rPSGL-1-conjugated MSs in retinal vessels after injection of WEB or GB. D) Number of bound MSs in the choroid of control animals and in those pretreated with WEB or GB.

Reduced PSGL-1-mediated leukocyte rolling velocity with PAF activation

PAF causes rapid PSGL-1 shedding from isolated leukocytes (33). Rapid proteolytic shedding of adhesion molecules sensitively regulates leukocyte rolling velocity (34) and adhesion (10). We hypothesized that reduced PSGL-1 expression on peripheral blood leukocytes might counter the effect of endothelial P-selectin expression. To study the isolated effect of PAF activation on PSGL-1-mediated rolling, we used our in vivo autoperfusion system (17) (Fig. 7A). Murine P-selectin was immobilized on the chamber surfaces. The inlet of the chamber was attached via biocompatible tubing to the carotid artery of a live mouse, and the outlet was connected to the jugular vein. Constitutive rolling of PBLs on the chamber surfaces was visualized using intravital microscopy (Fig. 7B). Thereafter, PAF (0.6 μg in 100 μl saline) was injected through the sideport in the inlet tubing leading to the chamber. Based on the prior literature (33), a higher rolling velocity would have been expected. However, PAF significantly reduced the rolling velocity of PBLs through the chamber (Fig. 7C). Furthermore, 22% of the leukocytes assumed a change in shape, characteristic of activation that was not observed in normal rolling cells (n=4; P=0.03; Fig. 7D). These findings suggest that, analogous to the LPS-induced up-regulation of PSGL-1 function, (35), PAF might modify PSGL-1 and therewith decrease leukocyte rolling velocity. In vivo this would further enhance recruitment.

Figure 7.

Figure 7.

Open in a new tab

Direct effect of PAF on PSGL-1-mediated rolling in vivo. Velocity of the rolling leukocytes was analyzed in vivo as they passed through the transparent P-selectin-coated, chamber. Blood is pumped by the mouse heart through a biocompatible tube from the carotid artery to the chamber and then to the jugular vein on the opposite side. Ports in the tube provide a site for blood pressure monitoring and control as well as one for injecting experimental agents. A) Schematic of the experimental design illustrating the study of the immediate effect of PAF on rolling leukocytes, originally introduced by Hafezi-Moghadam et al. (17). B) Successive in vivo images of a rolling leukocyte on immobilized P-selectin before (top panel) and one after PAF infusion (0.6 μg in 100 μl saline; bottom panel). C) Cumulative histogram of the velocity of rolling leukocytes on immobilized P-selectin without, and 10 min after, PAF infusion. D) Representative micrographs of activated leukocytes (arrows) bound to P-selectin 10 min after PBS control or PAF infusion.

DISCUSSION

PAF, synthesized in response to specific signaling, is a potent instigator of the inflammatory response. Here, we used EIU, a well-characterized model of acute ocular inflammation (12), in combination with our recently introduced molecular imaging approach (7–9) and inhibitors to understand the contribution of PAF to vascular injury. In vivo and ex vivo imaging of the rat retina show that PAF and, probably, PAF-like lipids significantly contribute to leukocyte recruitment in the retina 24 h after EIU induction. Timing appears critical, as PAF injection, immediately after LPS challenge, mitigates, rather than enhances, the inflammatory response in mice (36). However, PAF inhibitors, under the conditions of this study, strongly antagonize LPS-induced leukocyte accumulation in retinal and in choroidal vessels.

The differences between the effectiveness of GB and WEB treatment in reducing leukocyte accumulation may be due to differences in their modes of PAF inhibition. GB also increases the activity of certain PAF-AHs, which inactivate PAF-like lipids as well as PAF. Unlike PAF-like lipids, PAF does not generally enter the circulation but remains on activated endothelium (14). WEB binds exclusively to PAFR, yet our data indicate that it is not as potent an inhibitor of leukostasis, or of P-selectin translocation, as is GB. This suggests the existence of a PAFR-independent pathway that is unaffected by WEB, but one that is inhibited by GB. However, the specificity of PAF-AH is not well established. For instance, PAF-AH inactivates lipoteichoic acid, an inflammatory factor in the endotoxemia induced by gram-positive bacteria (37).

Using our novel molecular imaging approach, we show the role of PAF in adhesion of PSGL-1-cojugated MSs to retinal and choroidal vessels. While intravitreal administration of PAF causes a profound increase in MS binding to the retinal vasculature, it does not affect binding in the choroid. The possible explanation that PAF does not cross the outer blood-retinal barrier, which separates the retinal from choroidal vessels, is supported by the observation that choroidal leukocyte adhesion is also not affected by PAF injection. Our in vivo imaging further reveals that rPSGL-1 MSs bind to the activated endothelium, as well as to adherent leukocytes, presumably through leukocyte L-selectin. By plotting the ratio of leukocyte-bound MSs to total MSs, with and without WEB and GB pretreatment, we gained insight into the differential effect of PAF on these parameters in EIU animals. LPS alone doubled the total count of MSs in retinal vessels and brought the relative numbers of leukocyte-bound to total-bound MSs to over 70%. This shows that PAF-induced MS adhesion is due more to increased leukocyte adherence than to direct adhesion to the endothelium.

A previous report by Davenpeck et al. (33), showing that activation of isolated leukocytes by exogenous PAF causes PSGL-1 shedding, would have predicted an increase in PSGL-1-mediated leukocyte rolling velocity by PAF. Contrary to that expectation, we found that direct interaction of PAF with rolling leukocytes under our, more physiological flow conditions, significantly reduces their rolling velocity on immobilized P-selectin in vivo. This suggests that PAF might also cause other changes in the PSGL-1 molecule that would intensify its binding to other ligands that could reduce the velocity of the rolling cells. Since PSGL-1 is post-translationally modified, these changes could be in the realm of rapid enzymatic deglycosylation. As Davenpeck et al. (33) have suggested, in vivo, PAF-induced PSGL-1 shedding might be a mechanism to modulate the transition to integrin-mediated firm adhesion. In our experiments, there appears to be a competing balance between the PSGL-1 changes that reduce the rolling velocity, such as the proposed enzymatic modifications, and the rapid shedding that would increase the rolling velocity. Our experiments reveal that the result of these 2 opposed effects is, under the physiological flow conditions of the autoperfused microflow chamber, a slowing down of the rolling cells. The lower rolling velocity on P-selectin is in line with our findings on PAF-induced leukocyte accumulation in vivo. Furthermore, in our autoperfusion flow experiments, circulating leukocytes not only slow down in the presence of exogenous PAF but also firmly adhere to immobilized P-selectin and undergo morphological changes characteristic of activation. This indicates that integrin interaction with endothelial ICAM-1 may not be required for the firm adhesion step in the recruitment cascade, in contrast to the accepted paradigm. A better understanding of these results will require future investigations.

In summary, the data on the effect of PAF inhibitors in EIU animals, as well as the direct effect of PAF on retinal and choroidal leukostasis, indicate a prominent role for PAF and PAF-like lipids in acute ocular inflammation in vivo. A novel finding is that circulating leukocytes, in the presence of exogenous PAF, firmly bind to P-selectin-coated surfaces and become activated without integrin participation. We infer that PAF-induced retinal and choroidal inflammation, in EIU rats, is due to both P-selectin translocation and leukocyte activation. However, as in most complex in vivo scenarios, contribution of other inflammatory molecules cannot be excluded. PAF inhibition might become a new strategy in the treatment of ocular inflammatory diseases.

Acknowledgments

Research-grade YSPSL (recombinant soluble P-selectin glycoprotein ligand IgG fusion protein rPSGL-Ig) was a kind gift of Y's Therapeutics (San Bruno, CA, USA).

This work was supported by U.S. National Institutes of Health grant AI-050775, the American Health Assistance Foundation, and the Malaysian Palm Oil Board.

REFERENCES

  • 1. Suzuma K., Mandai M., Kogishi J., Tojo S. J., Honda Y., Yoshimura N. (1997) Role of P-selectin in endotoxin-induced uveitis. Invest. Ophthalmol. Vis. Sci. 38, 1610–1618 [PubMed] [Google Scholar]
  • 2. Miyamoto K., Ogura Y., Hamada M., Nishiwaki H., Hiroshiba N., Honda Y. (1996) In vivo quantification of leukocyte behavior in the retina during endotoxin-induced uveitis. Invest. Ophthalmol. Vis. Sci. 37, 2708–2715 [PubMed] [Google Scholar]
  • 3. Devine L., Lightman S. L., Greenwood J. (1996) Role of LFA-1, ICAM-1, VLA-4 and VCAM-1 in lymphocyte migration across retinal pigment epithelial monolayers in vitro. Immunology 88, 456–462 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Norman K. E., Moore K. L., McEver R. P., Ley K. (1995) Leukocyte rolling in vivo is mediated by P-selectin glycoprotein ligand-1. Blood 86, 4417–4421 [PubMed] [Google Scholar]
  • 5. Moore K. L., Patel K. D., Bruehl R. E., Li F., Johnson D. A., Lichenstein H. S., Cummings R. D., Bainton D. F., McEver R. P. (1995) P-selectin glycoprotein ligand-1 mediates rolling of human neutrophils on P-selectin. J. Cell Biol. 128, 661–671 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Geng J. G., Bevilacqua M. P., Moore K. L., McIntyre T. M., Prescott S. M., Kim J. M., Bliss G. A., Zimmerman G. A., McEver R. P. (1990) Rapid neutrophil adhesion to activated endothelium mediated by GMP-140. Nature 343, 757–760 [DOI] [PubMed] [Google Scholar]
  • 7. Miyahara S., Almulki L., Noda K., Nakazawa T., Hisatomi T., Nakao S., Thomas K. L., Schering A., Zandi S., Frimmel S., Tayyari F., Garland R. C., Miller J. W., Gragoudas E. S., Masli S., Hafezi-Moghadam A. (2008) In vivo imaging of endothelial injury in choriocapillaris during endotoxin-induced uveitis. FASEB J. 22, 1973–1980 [DOI] [PubMed] [Google Scholar]
  • 8. Sun D., Nakao S., Xie F., Zandi S., Schering A., Hafezi-Moghadam A. (2010) Superior sensitivity of novel molecular imaging probe: simultaneously targeting two types of endothelial injury markers. FASEB J. 24, 1532–1540 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Xie F., Sun D., Schering A., Nakao S., Zandi S., Liu P., Hafezi-Moghadam A. (2010) Novel molecular imaging approach for subclinical detection of iritis and evaluation of therapeutic success. Am. J. Pathol. 177, 39–48 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Hafezi-Moghadam A., Thomas K., Prorock A., Huo Y., Ley K. (2001) L-selectin shedding regulates leukocyte recruitment. J. Exp. Med. 193, 863–872 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Suttorp-Schulten M. S., Rothova A. (1996) The possible impact of uveitis in blindness: a literature survey. Br. J. Ophthalmol. 80, 844–848 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Rosenbaum J. T., McDevitt H. O., Guss R. B., Egbert P. R. (1980) Endotoxin-induced uveitis in rats as a model for human disease. Nature 286, 611–613 [DOI] [PubMed] [Google Scholar]
  • 13. Hafezi-Moghadam A., Noda K., Almulki L., Iliaki E. F., Poulaki V., Thomas K. L., Nakazawa T., Hisatomi T., Miller J. W., Gragoudas E. S. (2007) VLA-4 blockade suppresses endotoxin-induced uveitis: in vivo evidence for functional integrin up-regulation. FASEB J. 21, 464–474 [DOI] [PubMed] [Google Scholar]
  • 14. Prescott S. M., Zimmerman G. A., Stafforini D. M., McIntyre T. M. (2000) Platelet-activating factor and related lipid mediators. Annu. Rev. Biochem. 69, 419–445 [DOI] [PubMed] [Google Scholar]
  • 15. Rollin S., Lemieux C., Maliba R., Favier J., Villeneuve L. R., Allen B. G., Soker S., Bazan N. G., Merhi Y., Sirois M. G. (2004) VEGF-mediated endothelial P-selectin translocation: role of VEGF receptors and endogenous PAF synthesis. Blood 103, 3789–3797 [DOI] [PubMed] [Google Scholar]
  • 16. Maliba R, B. A., Neagoe P. E., Villeneuve L. R., Sirois M. G. (2008) Angiopoietin-mediated endothelial P-selectin translocation: cell signaling mechanisms. J. Leukoc. Biol. 83, 352–360 [DOI] [PubMed] [Google Scholar]
  • 17. Hafezi-Moghadam A., Thomas K. L., Cornelssen C. (2004) A novel mouse-driven ex vivo flow chamber for the study of leukocyte and platelet function. Am. J. Physiol. Cell. Physiol. 286, C876–892 [DOI] [PubMed] [Google Scholar]
  • 18. O'Donnell S. R., Barnett C. J. (1988) pA2 values for antagonists of platelet activating factor on aggregation of rabbit platelets. Br. J. Pharmacol. 94, 437–442 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Stewart A. G., Dusting G. J. (1988) Characterization of receptors for platelet-activating factor on platelets, polymorphonuclear leukocytes and macrophages. Br. J. Pharmacol. 94, 1225–1233 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Yue T. L., Rabinovici R., Farhat M., Feuerstein G. (1991) Inhibitory effect of new PAF antagonists on PAF-induced rabbit platelet aggregation in vitro and ex vivo. J. Lipid. Mediat. 3, 13–26 [PubMed] [Google Scholar]
  • 21. Ishii S., Kuwaki T., Nagase T., Maki K., Tashiro F., Sunaga S., Cao W. H., Kume K., Fukuchi Y., Ikuta K., Miyazaki J., Kumada M., Shimizu T. (1998) Impaired anaphylactic responses with intact sensitivity to endotoxin in mice lacking a platelet-activating factor receptor. J. Exp. Med. 187, 1779–1788 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Nunez D., Chignard M., Korth R., Le Couedic J. P., Norel X., Spinnewyn B., Braquet P., Benveniste J. (1986) Specific inhibition of PAF-acether-induced platelet activation by BN 52021 and comparison with the PAF-acether inhibitors kadsurenone and CV 3988. Eur. J. Pharmacol. 123, 197– 205 [DOI] [PubMed] [Google Scholar]
  • 23. Schmidt H., Ebeling D., Bauer H., Bohrer H., Gebhard M. M., Martin E. (1996) Influence of the platelet-activating factor receptor antagonist BN52021 on endotoxin-induced leukocyte adherence in rat mesenteric venules. J. Surg. Res. 60, 29–35 [DOI] [PubMed] [Google Scholar]
  • 24. Bonin F., Ryan S. D., Migahed L., Mo F., Lallier J., Franks D. J., Arai H., Bennett S. A. (2004) Anti-apoptotic actions of the platelet-activating factor acetylhydrolase I alpha2 catalytic subunit. J. Biol. Chem. 279, 52425–52436 [DOI] [PubMed] [Google Scholar]
  • 25. Gardner A. A., Reichert E. C., Topham M. K., Stafforini D. M. (2008) Identification of a domain that mediates association of platelet-activating factor acetylhydrolase with high density lipoprotein. J. Biol. Chem. 283, 17099–17106 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Proudfoot J. M., Barden A. E., Loke W. M., Croft K. D., Puddey I. B., Mori T. A. (2009) HDL is the major lipoprotein carrier of plasma F2-isoprostanes. J. Lipid Res. 50, 716–722 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Stremler K. E., Stafforini D. M., Prescott S. M., McIntyre T. M. (1991) Human plasma platelet-activating factor acetylhydrolase. Oxidatively fragmented phospholipids as substrates. J. Biol. Chem. 266, 11095–11103 [PubMed] [Google Scholar]
  • 28. McIntyre T. M., Zimmerman G. A., Prescott S. M. (1999) Biologically active oxidized phospholipids. J. Biol. Chem. 274, 25189–25192 [DOI] [PubMed] [Google Scholar]
  • 29. Smiley P. L., Stremler K. E., Prescott S. M., Zimmerman G. A., McIntyre T. M. (1991) Oxidatively fragmented phosphatidylcholines activate human neutrophils through the receptor for platelet-activating factor. J. Biol. Chem. 266, 11104–11110 [PubMed] [Google Scholar]
  • 30. Howard K. M., Olson M. S. (2000) The expression and localization of plasma platelet-activating factor acetylhydrolase in endotoxemic rats. J. Biol. Chem. 275, 19891–19896 [DOI] [PubMed] [Google Scholar]
  • 31. Tellis C. C., Tselepis A. D. (2009) The role of lipoprotein-associated phospholipase A2 in atherosclerosis may depend on its lipoprotein carrier in plasma. Biochim. Biophys. Acta 1791, 327–338 [DOI] [PubMed] [Google Scholar]
  • 32. Gelb M. H., Min J. H., Jain M. K. (2000) Do membrane-bound enzymes access their substrates from the membrane or aqueous phase: interfacial versus non-interfacial enzymes. Biochim. Biophys. Acta 1488, 20–27 [DOI] [PubMed] [Google Scholar]
  • 33. Davenpeck K. L., Brummet M. E., Hudson S. A., Mayer R. J., Bochner B. S. (2000) Activation of human leukocytes reduces surface P-selectin glycoprotein ligand-1 (PSGL-1, CD162) and adhesion to P-selectin in vitro. J. Immunol. 165, 2764–2772 [DOI] [PubMed] [Google Scholar]
  • 34. Hafezi-Moghadam A., Ley K. (1999) Relevance of L-selectin shedding for leukocyte rolling in vivo. J. Exp. Med. 189, 939–948 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Almulki L., Noda K., Amini R., Schering A., Garland R. C., Nakao S., Nakazawa T., Hisatomi T., Thomas K. L., Masli S., Hafezi-Moghadam A. (2009) Surprising up-regulation of P-selectin glycoprotein ligand-1 (PSGL-1) in endotoxin-induced uveitis. FASEB J. 23, 929–939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Jeong Y. I., Jung I. D., Lee C. M., Chang J. H., Chun S. H., Noh K. T., Jeong S. K., Shin Y. K., Lee W. S., Kang M. S., Lee S. Y., Lee J. D., Park Y. M. (2009) The novel role of platelet-activating factor in protecting mice against lipopolysaccharide-induced endotoxic shock. PLoS One 4, e6503. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Seo H. S., Kim J. H., Nahm M. H. (2006) Platelet-activating factor-acetylhydrolase can monodeacylate and inactivate lipoteichoic acid. Clin. Vaccine Immunol. 13, 452–458 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from The FASEB Journal are provided here courtesy of The Federation of American Societies for Experimental Biology

RESOURCES