Upregulation of P-selectin and Intercellular Adhesion Molecule-1 after Retinal Ischemia-Reperfusion Injury

Abstract

Purpose. Vascular endothelial cells and leukocytes express several inflammatory adhesion receptors, such as selectins and cell adhesion molecules, after retinal ischemia. They mediate the transmigration process of leukocytes. For further understanding of the role of leukocytes after retinal ischemia-reperfusion injury, the responses of the leukocyte-endothelial cell adhesion molecules P-selectin and intercellular adhesion molecule (ICAM)-1 during retinal ischemia and reperfusion were examined.

Methods. Male pigmented rats were subjected to retinal ischemia by a 1-hour ligation of the optic nerve followed by reperfusion. Gene and protein expression of P-selectin and ICAM-1 were studied at 0.5, 1, 6, 12, and 24 hours after the onset of reperfusion with semiquantitative polymerase chain reaction and Western blot analysis. Immunohistochemical methods were used to detect specific lesions expressing ICAM-1.

Results. Significant upregulation of P-selectin and ICAM-1 mRNA (at 6, 12, and 24 hours of reperfusion) were observed, with the expression peaks of both P-selectin and ICAM-1 mRNA occurring at 12 hours after reperfusion. P-selectin protein gradually increased and reached a maximum 6 hours after reperfusion, whereas ICAM-1 protein increased until 24 hours after reperfusion. Immunostaining with ICAM-1 antibodies was positive in endothelial cells after ischemia-reperfusion injury.

Conclusions. Retinal ischemia-reperfusion stimulates P-selectin and ICAM-1 expression. Endothelial ICAM-1 expression in retinal vessels was observed. Upregulation of P-selectin and ICAM-1, which contribute to leukocyte rolling and adhesion, were observed after retinal ischemia.

Inflammatory reactions play an important role in the pathogenesis of ischemic injury. Cytopathological features of retinal inflammation are neuronal loss, gliosis, edema, an increase in vascular permeability, inflammatory cell influx and inflammatory mediator production. The response after retinal ischemia also involves the activation and infiltration of peripheral leukocytes from the retinal circulation into the retina. ¹ Compromise of the blood-retinal barrier is a critical event leading to vasogenic edema, neutrophil-leukocyte invasion, and secondary retinal injury. Leukocyte transmigration from blood vessels requires sequential interactions of adhesion molecules between leukocytes and endothelial cells. ^{2 3} The process of leukocyte extravasation can be divided into three distinct steps: rolling, firm adhesion, and transmigration. The initial step is the transition from rapid flow of neutrophils to neutrophil rolling, which is mediated by selectins ^{4 5 6 7 8} which consist of three structurally homologous proteins expressed on leukocytes (L-selectin), platelets (P-selectin), and endothelial cells (P- and E-selectin). Neutrophil activation by chemotactic factors leads to firm adhesion that is dependent primarily on leukocyte β2-integrins and endothelial intercellular adhesion molecule (ICAM)-1. ⁹ The process of leukocyte extravasation entails sequential steps of rolling, firm adhesion, and transmigration, involving interactions of molecules in the selectin, immunoglobulin, and integrin gene families. ³ Adhesion molecules are thought to contribute to leukocyte dynamics and platelet-endothelium interactions after transient retinal ischemia, and previous studies have shown the contribution of adhesion molecules to leukocyte dynamics, leukocyte rolling, and accumulation. ^{10 11} In this article, we evaluated semiquantitatively and characterized the time course and localization of the adhesion molecules P-selectin, a member of the glycoprotein family that is thought to play an essential role in initial leukocyte-endothelium interaction, ^{5 12} and ICAM-1, a member of the immunoglobulin superfamily that plays an important role in leukocyte adhesion and transmigration, ^{7 13 14} after retinal ischemia-reperfusion in the rat. The present studies were designed to evaluate directly the expression of P-selectin and ICAM-1 in retinal inflammation associated with ischemia and reperfusion. 

Material and Methods

Animal Model

Male Long-Evans rats weighing 200 to 250 g were used (n = 48). Transient retinal ischemia was induced as previously described ¹⁵ with a slight modification. Briefly, rats were anesthetized with a mixture (1:1) of 4 mg/kg xylazine hydrochloride and 10 mg/kg ketamine hydrochloride. The pupils were dilated with 0.5% tropicamide and 2.5% phenylephrine hydrochloride. After a lateral conjunctival peritomy and disinsertion of the lateral rectus muscle, the optic nerve of the right eye was exposed by blunt dissection. A 6-0 nylon suture was passed around the optic nerve and tightened. Complete nonperfusion was confirmed through an operating microscope. After 60 minutes, retinal vessels were examined with the rat still under the operating microscope to confirm nonperfusion. Then, the ligation was removed. Reperfusion of the retinal vessels also was observed through the operating microscope. Sham-surgery rats underwent similar surgery without tightening of the suture. Procedures used in all experiments were consistent with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Reverse Transcription-Polymerase Chain Reaction

At determined postreperfusion intervals (0.5, 1, 6, 12, and 24 hours), rats were killed. The sensory retinas were dissected carefully after enucleation and frozen immediately in liquid nitrogen. Total RNA was prepared from frozen tissue samples according to the acid guanidinium thiocyanate-phenol-chloroform extraction method. Residual DNA was removed by treatment with RQ1 RNase-free DNase (Promega, Madison, WI). Five micrograms of total RNA was reverse transcribed into cDNA in a 20-μL reaction mixture containing reverse transcriptase (Superscript II; Invitrogen-Gibco, Gaithersburg, MD) and oligo(dT) primers. The cDNA was amplified using specific primers with a gene amplification system (Gene Amp PCR System 2400; Perkin-Elmer, Boston, MA). The cDNA amplification conditions were determined to be optimal: 0.5 ng cDNA, 1.5 U Taq DNA polymerase (Promega), in a total volume of 25 μL containing 12.5 picomoles of each primer. The sequences for the P-selectin and ICAM-1 were as follows: P-selectin: upstream primer, 5′-CAAGAGGAACAACCAGGACT-3′; downstream primer, 5′-AATGGCTTCACAGGTTGGCA-3′; ICAM-1: upstream primer, 5′-AGACACAAGCAAGAGAAGAA-3′; downstream primer, 5′-GAGAAGCCCAAACCCGTATG-3′. Oligonucleotide primer pairs from separate exons were prepared for P-selectin and ICAM-1. The following conditions were used: denaturing at 94°C for 30 seconds, annealing at 54°C for 30 seconds, and polymerization for 30 seconds. The reaction was performed for 30 cycles. In addition, 28S rRNA was amplified as a reference marker using the same RT-PCR technique (25 cycles: 30 seconds at 94°C, 30 seconds at 54°C, and 30 seconds at 72°C). For 28S, a pair of oligonucleotide primers, 5′-TGTTGACGCGATGTGATTTCTGC-3′(forward) and 5′-TCTACACCTCTCATGTCTCTTCA-3′ (reverse), was prepared. On the basis of genomic sequence, every primer contained no intron. We quantified PCR products during the exponential phase of amplification. As a negative control of 28S amplification, isolated total RNAs were treated in the same way, except that no reverse transcriptase was added. PCR products were run on 2% agarose gel and stained with ethidium bromide, and bands were visualized by scanning with laser densitometry (FMBIO II; Hitachi, Ltd., Tokyo, Japan). Molecular identity and homogeneity of the resultant PCR fragments were checked by DNA sequencing. 

Analysis of the DNA-stained agarose gels was evaluated by band intensity comparison of 28S expression versus each cytokine with NIH Image (available by ftp at zippy.nimh.nih.gov/ or at http://rsb.info.nih.gov/nih-image; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD). Each PCR reaction was repeated three times in all four eyes in each time group. The results are expressed as arithmetic means (±SEM). Statistical comparisons between groups were performed by unpaired t-test. The difference was considered to be significant when P < 0.05. 

Western Blot Analysis

Proteins (150 mg) extracted from the rat retinal tissue at various times after ischemia-reperfusion injury were resolved by SDS-polyacrylamide (10%) gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes (Immobilon-P; Millipore, Bedford, MA). A monoclonal antibody against mouse vimentin (Dako, Glostrup, Denmark) was used as an internal standard to check protein loading. A polyclonal antibody against mouse P-selectin (Santa Cruz Biotechnology, Santa Cruz, CA), a monoclonal antibody against mouse ICAM-1 (Seikagaku Kogyo) and vimentin antibodies were diluted in a block solution containing 20% skim milk and 0.1% Tween 20 in Tris-buffered saline and incubated with PVDF membranes overnight at 4°C. Blots were detected with an alkaline phosphatase-labeled rabbit anti-goat IgG antibody against P-selectin (Bio-Rad, Hercules, CA) and a goat anti-rabbit IgG antibody against ICAM-1 and vimentin (Roche Diagnostics, Mannheim, Germany) and visualized with a 5-bromo-4-chloro-3-indoyl phosphate (BCIP)-nitroblue tetrazolium indicator substrate. 

Immunohistochemistry

At various times after ischemia-reperfusion injury, in rats under deep anesthesia with 50 mg/kg pentobarbital sodium, the chest cavity was opened. The left heart ventricle was then accessed with a 16-gauge perfusion canula, and the right atrium was opened to achieve an outflow. The rats were first perfused with a physiological saline body weight to wash out erythrocytes. Then, they were perfused with 4% (wt/vol) paraformaldehyde (PFA) for fixation. After a 2-hour fixation in 4% PFA and phosphate-buffered saline (PBS), the sclera was dissected from eyes in PBS, and the lens and vitreous humor were removed. Retinas were flattened and stored in methanol. After a brief wash in PBS, retinal wholemounts were incubated in a blocking buffer (PBS containing 1% bovine serum albumin and 0.5% Triton-X 100) for 1 hour at room temperature. Incubation with a ICAM-1 antibody (diluted 1:1000 in blocking buffer) was performed overnight at 4°C in the case of the primary antibody and 2 hours at 4°C in the case of secondary antibody and lectin conjugate. Antibodies used were mouse monoclonal anti-rat ICAM-1 (Seikagaku Kogyo), Alexa Fluor 488 isolectin IB4 conjugate, and Alexa Fluor 594 goat anti-mouse IgG (both from Molecular Probes, Eugene, OR). 

Results

Gene Expression of P-selectin and ICAM-1 in the Retina

To characterize the regulation of P-selectin and ICAM-1 mRNA in ischemia and reperfusion retina, total RNA was isolated from retinas 0.5, 1, 6, 12, and 24 hours after ischemia-reperfusion injury, with nontreated eyes used as the control. RNA was subjected to semiquantitative RT-PCR analysis with primers specific for P-selectin, ICAM-1, and 28S (as an internal standard) genes. No detectable signals were seen at any period in any sample without reverse transcriptase (Fig. 1a) . 

For each sample, the ratio of the integrated optical density of the P-selectin and ICAM-1 bands to that of the internal standard was calculated (means ± SEM; n = 3 per each period; P < 0.05). Time-dependent induction of P-selectin and ICAM-1 mRNAs were observed. The peak mRNA expression of both P-selectin and ICAM-1 was 12 hours after reperfusion (Figs. 1b 1c) . The expression of P-selectin mRNA rapidly increased from 6 hours after reperfusion (Fig. 1b) , whereas ICAM-1 mRNA expression was gradually increased and significantly upregulated from 6 hours after reperfusion (Fig. 1c) . 

Protein Expression of P-selectin and ICAM-1 in the Retina

Figure 2 shows results of Western blot analysis of P-selectin, ICAM-1, and vimentin in the retina after ischemia-reperfusion. The molecular sizes of the bands for P-selectin, ICAM-1, and vimentin were 140, 90, and 55 kDa, respectively. Immunoreactive P-selectin and ICAM-1 proteins were faintly detected in the sham control retina, and time-dependent induction of P-selectin and ICAM-1 proteins were observed. Expression of ICAM-1 protein also gradually increased until 24 hours after reperfusion. The expression of P-selectin protein gradually intensified, reaching a maximum at 6 hours. 

Immunostaining with an ICAM-1 antibody was positive in retinal vessels 24 hours after ischemia-reperfusion injury. Intense ICAM-1 immunoreactivity was associated with the retinal vascular endothelium (Figs. 3a 3b 3c) , the marked increase in the immunoreactivity of ICAM-1 was evident in the venules and capillaries. Control retinas showed a very low level of ICAM-1 immunoreactivity (Figs. 3d 3e 3f) . 

Discussion

We collected data demonstrating the time course of P-selectin and ICAM-1 after transient retinal ischemia. Our data emphasize the upregulation of P-selectin and ICAM-1. P-selectin and ICAM-1 upregulation after ischemia-reperfusion injury has been reported in various conditions, such as cerebral, ^{16 17} cardiac, ^{18 19 20} splanchnic, ^{21 22} and renal ischemia. ^{23 24} Also, in the postischemic retina, the time course of P-selectin gene expression shown by RT-PCR has been reported. ²⁵ That study showed upregulation of P-selectin mRNA after ischemia-reperfusion injury, as did the present study. 

In the present study, P-selectin and ICAM-1 levels of both mRNA and protein increased after retinal ischemia-reperfusion, with the peak P-selectin mRNA level occurring 12 hours after injury, whereas P-selectin protein expression gradually increased and reached a maximum 6 hours after reperfusion. P-selectin is found in endothelial cells and platelets. Stimulation with factors such as histamine or thrombin immediately induces endothelial cells to release P-selectin of the storage granules (Weibel-Palade bodies), ²⁶ and activated platelets express P-selectin on their surfaces by rapid secretion from α-granules. ²⁷ The P-selectin can be regulated on two different levels: by the transport of stored P-selectin to the cell surface by regulated secretion of storage granules and by transcriptional induction. The earlier protein upregulation compared with mRNA in the present study may reflect P-selectin secretion from circulating stimulated platelets. More investigations are needed to resolve this question. 

The peak level of ICAM-1 mRNA was 12 hours after reperfusion, and the protein level increased until 24 hours after reperfusion. Clearly, P-selectin showed an earlier upregulation than did ICAM-1. Previous studies of cerebral ²⁸ and myocardial ^{29 30} ischemia also showed by immunohistochemistry that P-selectin is upregulated earlier than ICAM-1. 

Tsujikawa et al. ¹ showed that leukocyte dynamics during retinal ischemia-reperfusion injury could be evaluated quantitatively in vivo with a scanning laser ophthalmoscope (SLO). They reported that rolling leukocytes were observed in treated rats during the reperfusion period, and the number of rolling leukocytes gradually increased and peaked 12 hours after reperfusion. In addition, the number of accumulated leukocytes increased as time elapsed, peaking 24 hours after reperfusion and decreasing thereafter. They also showed that inhibition of P-selectin and ICAM-1 with a monoclonal antibody significantly attenuated retinal ischemia-reperfusion injury. A P-selectin monoclonal antibody reduced leukocyte rolling along the major retinal veins and leukocyte accumulation during the reperfusion period, whereas an ICAM-1 monoclonal antibody reduced leukocyte accumulation during the reperfusion period. ¹⁰ The combination of these results and our results regarding P-selectin upregulation indicate that P-selectin mainly acts as a mediator of leukocyte rolling. Also, the time course of ICAM-1 upregulation seemed to be consistent with the leukocyte dynamics in vivo. These results are in agreement that ICAM-1 mainly acts as a mediator of leukocyte adhesion after inflammatory reactions such as ischemia-reperfusion injury and that the leukocyte accumulation results in adhesion to the vascular wall and/or transmigration to the abluminal side of the endothelium. 

Mice deficient in ICAM-1 and P-selectin show reduced neutrophil trafficking after myocardial ischemic injury and reperfusion, but do not show a reduction in infarct size. ³¹ In retinal ischemia-reperfusion injury, blocking of adhesion molecules (P-selectin or ICAM-1) showed a neuroprotective effect that was not so dramatic in contrast to its impact on leukocyte accumulation. ¹⁰ As previously reported, ^{32 33} factors other than recruitment of leukocytes must be involved in retinal ischemia-induced injury. Further studies should be conducted to determine the whole mechanism that induces retinal damage after ischemia-reperfusion. 

To identify the localization of ICAM-1 upregulation induced by retinal ischemia-reperfusion injury, retinal wholemount immunohistochemistry studies were performed. Several studies have shown by immunohistochemical techniques the retinal expression of ICAM-1 in the diabetic eye, ³⁴ after laser photocoagulation, ³⁵ and during cerebral malaria. ³⁶ In the present study, ICAM-1 immunoreactivities were observed in the endothelium of the retinal vessels, and the marked increase of ICAM-1 immunoreactivity was evident in the venules and the vascular bed. In the diabetic retina, ³⁴ the staining was regional within the vascular bed. The difference in ICAM-1 expression may contribute to the greater leukocyte accumulation after ischemia-reperfusion injury ¹ than in the diabetic condition. ³⁷  

Miyamoto et al., ³⁷ in a rat model of early diabetic retinopathy induced by streptozotocin, has shown that diabetic retinal vascular leakage and nonperfusion are temporally and spatially associated with retinal leukocyte stasis, and they showed that the blocking of ICAM-1 with an mAb prevents diabetic retinal leukostasis and vascular leakage. McLeod et al. ³⁴ showed that ICAM-1 upregulation and subsequent leukocyte recruitment are involved in the pathogenesis of human diabetic retinopathy. Consequently, the retinal ischemia-reperfusion model that we examined reflects part of the pathogenesis in diabetic retinopathy, providing a possible biochemical basis for the mechanism of worsening retinopathy. 

In summary, the present study demonstrated concretely the upregulation of P-selectin and ICAM-1, which facilitate retinal leukocyte infiltration and damage after ischemia-reperfusion injury. 

 Submitted for publication December 24, 2002; revised May 23, 2003; accepted June 23, 2003.

 Disclosure: A. Nishiwaki, None; T. Ueda, None; S. Ugawa, None; S. Shimada, None; Y. Ogura, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

(a) Changes over time in P-selectin and ICAM-1 mRNA expression after ischemia-reperfusion injury. Expression of P-selectin (b), ICAM-1 (c), and 28S as a control was analyzed by RT-PCR. The graphic data show the density ratio by semiquantitative RT-PCR. *Statistically significant compared with the control (P < 0.05; unpaired t-test).

Figure 2.

 

Immunoblot analysis of P-selectin and ICAM-1 expression after ischemia-reperfusion injury. Vimentin was used as a loading control.

Figure 3.

 

Localization of ICAM-1 protein on a rat retina flatmounted after 24 hours of ischemia-reperfusion injury. Immunofluorescence of ICAM-1 (a, d), isolectin IB4 (b, e), and their overlay (c, f) in the same flatmounted section of retina. (a) A perivascular lesion that appeared by shape to be an endothelial cell. The localization was positive for ICAM-1 (red). (b) Isolectin IB4 (green) indicates the localization of the retinal vessels. (c) Digitally merged images of (a) and (b). (d) No positive reaction with ICAM-1 antibodies was detected in the control retina. (e) Control, isolectin IB-4. (f) Digitally merged images of (d) and (e).