Introduction

Human endogeneous uveitis is an intraocular inflammatory disease causing severe visual loss, and even blindness. There are various animal models of uveitis including EIU and EAU. EIU is an animal model for acute ocular inflammation induced by injection of the LPS, a component of gram-negative bacterial cell wall [36]. EIU is characterized by protein leakage in the anterior chamber and by infiltration of macrophages and neutrophils into the eye, with a peak at 24 h after LPS injection [9, 28]. Acute inflammation develops mainly in the anterior chamber (iridocyclitis) and the inflammatory cells infiltrate also in the vitreous as well as the retina [47]. It has been reported that inflammatory cytokines play an essential role in development of the EIU. Elevated expression of cytokines and chemokines such as TNF-α and MCP-1 has been observed concomitant with the peak of EIU [13, 34]. Local angiotensin II (AII) expression has also been found to be elevated during retinal inflammation in the EIU model [26].

EAU is another animal model for relatively chronic endogeneous uveitis such as Behçet disease, birdshot retinochoroidopathy, sympathetic ophthalmia and Vogt–Koyanagi–Harada disease [8]. EAU represents a Th1 cell-mediated and organ-specific autoimmune disease that is induced by immunization with retinal proteins, such as soluble antigen (SA) or hIRBP-derived peptide, or by the adoptive transfer of specific T cells for these antigens into naïve syngeneic recipients [7, 18, 30].

RAS is well known to play an important role in regulating blood pressure and body fluid regulation. It has been considered that the function of RAS is to control vascular tonus and maintain the fluid homeostasis by regulating water and electrolyte absorption and/or excretion in the kidney and gastrointestinal tract [27]. Recent studies, however, have demonstrated that AII is also involved in broad biological actions such as apoptosis, remodeling, and inflammation of vascular wall [37, 51, 52]. In addition, peripheral AII may be involved in an attack of fever and the peripheral interleukin (IL)-1β production following systemic injection of LPS in rats [29, 42]. When angiotensin-converting enzyme (ACE) inhibitor was administered to an experimental myocarditis model to block RAS, an anti-inflammatory effect was observed [17]. Similarly, RAS inhibitor showed a significant attenuation of hepatic IL-1β production in rats treated with LPS [29]. We previously reported that captopril, an ACE inhibitor, suppressed the inflammation in a rat EIU model [21].

AT1 receptor antagonists have been developed as drugs that selectively and markedly blocked the RAS at the receptor level, differently from ACE inhibitors [46]. Indeed, it has been reported that the AT1 receptor antagonist effectively attenuates various inflammatory processes [11, 16, 33, 35]. One of the AT1 receptor blockers, telmisartan, showed a neuroprotective effect via modulating AT1 receptor and AT2 receptor signaling in retinal inflammation in a mouse EIU model [26]. It was also reported in a retinal inflammation model that telmisartan reduced the number of infiltrating cells in aqueous humor [32]. Another AT1 receptor antagonist, losartan, is now widely used to treat hypertension. Notably, it was reported that losartan acted not only as an AT1 receptor antagonist but as an anti-inflammatory agent via pathways perhaps other than modulation of AT1 and AT2 receptor signaling [24].

Recent studies have collectively demonstrated the protective effect of RAS antagonism against immune-mediated inflammatory diseases such as myocarditis, chronic allograft rejection, antiglomerular basement membrane nephritis, colitis, and arthritis [1, 2, 6, 14, 15, 20, 22, 35, 38, 43]. However, no report has clearly demonstrated beneficial contribution of the RAS antagonism in ocular inflammatory conditions. The purpose of the present study was to investigate the anti-inflammatory effect of losartan on an animal model of acute uveitis (EIU). We analyzed the effects of losartan on cellular infiltration, extravasation of protein, and levels of TNF-α and MCP-1 in the aqueous humor of EIU rats. To further clarify the mechanism of the anti-inflammatory effect, we also examined the expression of NF-κB in the EIU lesion. Finally, to examine and compare the effect of losartan on a relatively chronic inflammatory model (EAU) with that on EIU, we analyzed the clinical score and generation of antigen-specific T cell proliferative response in the EAU mice, either in the presence or absence of losartan.

Materials and methods

Experimental animals and reagents

Seven-week-old male Lewis rats (220–250 g) were obtained from Clear Japan (Tokyo, Japan). Six- to 8-week-old C57BL/6 female mice were obtained from Japan SLC (Hamamatsu, Japan). All rats and mice were bred and maintained in a specific pathogen-free condition. All experimental animals were treated in accordance with the ARVO Statement for Use of Animals in Ophthalmic and Vision Research. All experiments were conducted with the approval and supervision of Hokkaido University Animal Care and Use Committee.

Losartan was provided by Merck and Co Inc (Rahway, NJ, USA). LPS from Salmonella typhimurium was purchased from Sigma-Aldrich (St. Louis, MO, USA). hIRBP peptide sequence 1–20 (GPTHLFQPSLVLDMAKVLLD) was synthesized by Sigma-Genosys (Ishikari, Japan). Purified PTX was purchased from Sigma-Aldrich (St. Louis). CFA and Mycobacterium tuberculosis strain H37Ra were purchased from Difco (Detroit, MI, USA).

Induction of EIU and collection of the aqueous humor

EIU was induced by subcutaneous injection of 200 μg of LPS from Salmonella typhimurium in 0.1 ml of phosphate buffered saline (pH 7.4, PBS) [21]. At the same time, these rats were injected intravenously with 1 mg/kg or 10 mg/kg of losartan (losartan potassium) diluted in 0.1 ml of PBS. Control EIU rats or naïve rats (negative control) were intravenously administered PBS alone (no losartan). Twenty-four hours after LPS injection, the rats were sacrificed and the aqueous humor (15–20 μl/rat) was collected as described below.

Histopathologic evaluation

Rats were euthanized 24 h after LPS injection. The eyes were enucleated immediately, stored in 10 % buffered formalin for 24–48 h, and embedded in paraffin. Sagittal sections (5 μm) were cut near the optic nerve head and stained with hematoxylin and eosin (H&E). Anterior chamber, iris-ciliary body (ICB), vitreous and retina were observed under light microscopy.

Number of infiltrating cells and protein concentration in aqueous humor

At 24 h after LPS injection, the rats were euthanized, and the aqueous humor was collected immediately from both eyes by an anterior chamber puncture (15–20 μl/rat), using a 30-gauge needle under a surgical microscope. The aqueous humor was then accurately diluted 10-fold with PBS. For cell counting, the aqueous humor sample was suspended in an equal amount of Türk stain solution, and the cell number was counted with a hemocytometer under a light microscope. The number of cells per field (an equivalent of 0.1 μl) was manually counted by two independent researchers.

Measurement of cytokines

The total protein concentration in the aqueous humor samples was measured with a bicinchoninic acid protein assay kit (Pierce, Rockford, IL, USA). The aqueous humor samples were stored on ice until using. The levels of TNF-α and MCP-1 concentrations in aqueous humor were measured with enzyme-linked immunosorbent assay (ELISA) kit (R&D Systems, Minneapolis, MN, USA) according to the manufacturer’s instruction.

Immunohistochemical studies for NF-κB

At 3 h after LPS injection, rats were anesthetized and the eyes were fixed by an intracardiac perfusion of 4 % paraformaldehyde in 0.1 M PBS. The eyes were enucleated and immersed in the same fixative for 12 h. After dehydration and paraffin embedment of the eyes, 5-μm sagittal sections near the optic nerve head were obtained. The sections were rinsed in PBS twice and incubated with normal goat serum before staining with anti-p65 antibody (Santa Cruz Biotechnology, Santa Cruz, CA). Binding of the primary antibody was visualized using Cy-3 conjugated goat anti-rabbit IgG (Jackson Immuno-Research Laboratories, West Grove, PA, USA). Nuclei were then stained with PBS containing YO-PRO-1 (Molecular Probes, Eugene, OR, USA) for 5 min. The sections were examined by laser scanning confocal microscopy (MRC-1024: Bio-Rad, Richmond, CA, USA; and LSM 510: Carl Zeiss, Oberkochen, Germany). Two areas of ICB were randomly photographed and the number of activated NF-κB positive cells was counted. The results were averaged for each sample in each group. This analysis was performed in the six eyes from three rats in each group.

Induction and assessment of EAU

Mice were immunized subcutaneously in the back with hIRBP peptide as described elsewhere [23]. IRBP 1–20 (200 μg) in 0.2 ml was emulsified in CFA (1:1 v/v) that had been supplemented with M.tuberculosis to 5 mg/ml. These mice were given 0.1 μg of PTX in 100 μl of PBS intraperitoneally as an additional adjuvant. Losartan was dissolved in PBS solution for in vivo injections. Mice were administered orally with losartan (20 mg/kg) or vehicle alone for 14 days. Two groups of mice were given oral administration of losartan or PBS on days 0–14, and other two groups of mice received the oral administration on days 14–28 after immunization. A disposable feeding needle was used for the oral administration.

Fundus examination was performed every 3 or 4 days from day 7 post-immunization by two ophthalmologists [23], and the severity of EAU in each eye was evaluated according to EAU clinical score (grade 0–5) as described previously [44].

Antigen specific T cell proliferative assay

Mice were immunized with 100 μg of hIRBP1-20 that was emulsified in CFA (1:1 v/v). Immunized mice orally received either losartan (20 mg/kg) or vehicle alone every day. Ten days after immunization, cells were collected from draining lymph nodes, suspended at 5 × 105 /well and cultured in vitro in the presence of graded concentrations of hIRBP peptide for 48 h [23]. Then, the cells were pulsed-labeled with 3H-thymidine and cultured for additional 16 h. Quantitation of these responses was accomplished by measuring incorporation of 3H-thymidine.

Statistical analysis

Data in EIU are presented as means±SD. Data were analyzed by analysis of variance (ANOVA), and statistical significances were analyzed using a Student’s t-test with Bonferroni’s collection. EAU scores were compared by Mann-Whitney U-test. A p-value was considered statistically significant when lower than 0.05.

Results

Histopathological findings in eyes of EIU rats

The eyes of control group rats that had been administered PBS alone showed no inflammation (Fig. 1a,e). Representative histological changes in eyes of LPS-treated EIU group but untreated with losartan were shown as a positive control (Fig. 1b,f). A large number of inflammatory cells were found in the anterior and posterior segment at 24 h after LPS administration. In contrast, only a few inflammatory cells were infiltrating in the eyes of rats treated with losartan (1 or 10 mg/kg) (Fig. 1c,g or d,h). These findings demonstrated that inflammation of eyes was ameliorated in the losartan-treated rats compared with the losartan-untreated rats.

Fig. 1
figure 1

Histologic changes of ICB, vitreous and retina 24 h after LPS injection. Eyes were enucleated 24 h after LPS injection, and fixed. Then they were sectioned, and stained with H & E. Photographs on the left side show ICB legion and those on the right side show vitreous and retina in rat. Rats of control group (a, e) were not injected with LPS. No inflammation is observed. Severe inflammatory cell infiltration is observed in EIU rats (b, f). In the group of EIU rats treated with losartan (1 mg/kg ; c, g , 10 mg/kg ; d, h), reductions of cell infiltration are observed compared to EIU rats. Inflammatory cells around the ICB and in the anterior chamber and vitreous cavity are present (arrows in b, c, f, g and h). Original magnification, ×200

Full size image

Effect of losartan on cellular infiltration and protein concentration in the aqueous humor in eyes of EIU rats

In the EIU group, the mean number of inflammatory cells in aqueous humor 24 h after LPS administration was 75.3 ± 45.6 × 105 cells/ml. Aqueous cell numbers of rats injected with 1 mg/kg or 10 mg/kg losartan were 27.9 ± 8.1 or 41.3 ± 30.9 × 105 cells/ml respectively. Thus, treatment of EIU rats with losartan significantly reduced the number of inflammatory cells in the anterior segment (p < 0.01, Fig. 2a).

Fig. 2
figure 2

Effect of losartan administration on cellular infiltration and protein concentration in the aqueous humor. Rats were injected with LPS and the aqueous humor was collected 24 h later. a Mean cell number in the aqueous humor. b Mean protein concentration. All data show the mean±SD from 8–13 rats of each group. ** Significantly different from the LPS group (p < 0.01)

Full size image

Next, we examined aqueous protein levels. The protein concentrations in aqueous humor collected from rats treated with 0, 1, or 10 mg/kg of losartan were 72.3 ± 3.4, 53.0 ± 5.5, or 42.9 ± 6.0 mg/ml respectively. The aqueous protein levels were significantly lower in either dose of losartan-treated rats than those in losartan-non-treated rats (p < 0.01). Further, the decrease in the protein level was losartan dose-dependent (Fig. 2b). In naïve rats administered PBS alone, the protein concentration in the aqueous humor was negligible (1.3 ± 0.13 mg/ml).

Effect of losartan administration on aqueous TNF-α and MCP-1 levels in EIU rats

Then, TNF-α level in aqueous humor of EIU rats was measured. No detectable TNF-α and MCP-1 was observed in naïve rats (data not shown). The concentration of TNF-α collected from LPS-treated EIU rats reached 2.91 ± 0.36 ng/ml at 24 h after LPS administration, while EIU rats injected with low-dose (1 mg/kg) or high-dose (10 mg/kg) losartan showed 2.50 ± 0.26 or 1.91 ± 0.78 ng/ml TNF-α respectively (Fig. 3a). Thus, aqueous TNF-α levels were reduced by losartan treatment in a dose-dependent manner, and the reduction by losartan (10 mg/kg) was regarded as significant compared to TNF-α levels in control EIU rats (p < 0.01).

Fig. 3
figure 3

Effect of losartan administration on TNF-α and MCP-1 concentration in the aqueous humor. Rats were injected with LPS and the aqueous humor was collected 24 h after LPS injection. a TNF-α concentration. b MCP-1 concentration. Data are the mean±SD of results from 8–13 rats. ** Significantly different from the LPS group (p < 0.01)

Full size image

Next, we quantified aqueous levels of a chemokine, MCP-1. Aqueous humor of losartan-untreated EIU rats contained 1164.3 ± 276.5 pg/ml of MCP-1 (Fig. 3b). MCP-1 levels in the aqueous humor collected from rats injected with 1 or 10 mg/kg of losartan were 847.6 ± 265.9 or 595.6 ± 256.7 pg/ml respectively. Thus, aqueous MCP-1 concentrations were also significantly decreased (p < 0.01) in rats treated with losartan in a dose-dependent fashion compared to those in control EIU rats.

Immunohistochemistry of NF-κB p65 in the ICB after LPS injection in EIU

NF-κB activation is involved in various inflammatory responses. We then analyzed immunohistochemically activated NF-κB expression in the lesion. No activated NF-κB-positive nuclei were found in the ICB before EIU induction (Fig. 4a[A]). Three hours after LPS injection, a considerable expression of activated NF-κB p65 was observed in the ICB of EIU rats (Fig. 4a[B]). By contrast, only a few NF-κB p65 positive nuclei were detected in losartan (10 mg/kg)-injected EIU rats (Fig. 4a[C]). The mean proportion of activated NF-κB-positive cells in EIU rats was 27.6 ± 14.1%, whereas those in rats treated with 1 or 10 mg/kg of losartan were 12.5 ± 2.7% or 10.6 ± 4.5%, respectively (Fig. 4b). The proportion of activated NF-κB positive cells was significantly lower in EIU rats treated with either dose of losartan (1 or 10 mg/kg) than that in untreated EIU rats (p < 0.01).

Fig. 4
figure 4

Effect of losartan administration on NF-κB p65 activation in the ICB 3 h after LPS injection. a Immunohistochemistry of NF-κB p65 (red) in the ICB of the rats 3 h after LPS injection. Dual-immunofluorescence labeling shows the colocalization of p65 (yellow) in nuclei (green). [A], control group; rats were not injected with LPS. [B], LPS group; rats were injected with LPS and 0.1 ml PBS. [C], losartan-treated group; rats were injected with LPS and 10 mg/kg of losartan diluted in 0.1 ml PBS. Magnification: ×400. Arrows: activated NF-κB positive cells. b Quantitative analysis of NF-κB-positive cells in the ICB. Control rats, LPS-injected EIU rats, and EIU rats treated with 1 mg/kg or 10 mg/kg losartan are shown. Data are the mean±SD (n = 6). ** Significantly different from LPS group (p < 0.01)

Full size image

Effect of losartan on EAU mice

To determine the effect of losartan on Th1-induced inflammation, EAU, mice were immunized with hIRBP peptide emulsified in CFA as described in “Materials & methods”. Two groups of experimental mice were prepared. In the first group (“early”), for 14 days after immunization with hIRBP B6 mice were orally administered with losartan (20 mg/kg) or PBS alone, every day (Fig. 5a). Another group of mice was treated with losartan or PBS alone from day 14 to 28 after immunization (“late”) (Fig. 5b). The funduscopic examination was performed from day 7 after immunization every 3–4 days.

Fig. 5
figure 5

Clinical score of EAU in mice treated with losartan. EAU was induced in B6 mice by injection of hIRBP peptide in CFA. Mice were treated with losartan (○) or PBS alone (●). Daily administration of losartan or PBS was performed for 2 weeks after immunization, days 0–14 (“early” group, a) or days 14–28 (“late” group, b). Results are presented as the mean score for all eyes of each group of mice (8 mice per group) ±SEM. Closed triangles (▲) indicate daily administration of losartan or PBS. * Significance was determined using Mann-Whitney U-test (p < 0.05)

Full size image

As shown in Fig. 5a,b, similar levels of EAU were observed in both groups of losartan-treated and vehicle-treated mice (n = 8 in each experimental group of “early” and “late” groups). In the “early” group, however, mean EAU severity of the losartan-treated mice was significantly milder (0.25 ± 0.11, mean±SE) than that in the controls (0.75 ± 0.11, p < 0.05) on day 10 after immunization (Fig. 5a). However, significant differences in the severity were no longer detected between losartan- and vehicle-treated mice on day 14 or later stages. In mice of the “late” group, no difference in EAU severity was seen between losartan- and vehicle-treated mice for the entire period of the experiment (Fig. 5b). These findings suggest that losartan is effective only in an early phase.

Next, we examined whether losartan showed a suppressive effect on the antigen-specific proliferative response of hIRBP-specific T cells. Th1 cells play an essential role in development of EAU. Draining lymph nodes were collected from hIRBP plus CFA-immunized mice treated with either losartan or PBS alone (control) for 10 days after immunization. Lymph node cells were cultured in vitro in the presence of graded concentrations of hIRBP peptide for 48 h. Comparable levels of antigen-specific proliferation were seen in a dose-dependent manner in both losartan or PBS-treated groups (Fig. 6). No difference was observed in the T cell proliferation between these two groups of EAU mice.

Fig. 6
figure 6

Antigen-specific T cell proliferative response. 3H-thymidine incorporations of lymphocytes collected from losartan-treated (○) or PBS alone (●) mice were demonstrated. Cells were obtained from draining lymph nodes of B6 mice 10 days after immunization. The results are presented as the mean±SEM (n = 3)

Full size image

Discussion

In the present study, we examined the therapeutic effect of a blocker of AT1 receptor, losartan, on LPS-induced acute ocular inflammation (EIU). We found that a single administration of losartan significantly ameliorated EIU in rats. Aqueous cell infiltration and levels of protein, inflammatory cytokines and chemokines in the aqueous humor were significantly decreased in EIU rats treated with losartan as compared to those in non-treated EIU rats. Activation of NF-κB in the legion was also significantly suppressed by the treatment with losartan. These results suggest that the AT1 receptor antagonist, losartan, exerts an anti-inflammatory influence in ocular tissue via directly inhibiting activation of NF-κB pathway, which leads to subsequent reduction of inflammatory cytokine production.

NF-κB is a key transcription factor that regulates various inflammatory processes [4]. The activated form of NF-κB is a heterodimer, usually consisting of two proteins, p65 (relA) and p50 subunit. In unstimulated cells, NF-κB is found in cytoplasm and bound to IκBα and IκBβ. When cells are stimulated, NF-κB is released from IκB by specific kinases phosphorilate IκB, and moves into the nucleus, where it binds to specific sequences in promoter regions of the target genes [3]. With various stimuli including LPS, proinflammatory cytokines and reactive oxygen species activate NF-κB [4]. It has been reported that AII also rapidly activates NF-κB [25] through AT1 and AT2 receptors [12, 31].

The present study showed that LPS-induced NF-κB p65 nuclear translocation in the ICB was significantly suppressed by an administration of losartan. The activation of NF-κB leads to an increase in the expression of numbers of genes, including TNF-α and IL-1β, that mediate inflammation and immune responses. These cytokines sometimes activate and, on another occasion, are activated by NF-κB [5]. It has been reported that losartan inhibited the LPS-induced production of IL-1β [42]. In the present study, we demonstrated that increased levels of aqueous TNF-α in EIU rats was significantly down-regulated by injection of a high dose of losartan. Thus, it seems that losartan inhibits not only IL-1β production but also the positive cycle of NF-κB and TNF-α, which results in the anti-inflammatory effect in EIU rats.

It has been reported that various chemokines are involved in ocular inflammation. In the present study, we focused on MCP-1 expression as a representative inflammatory chemokine. MCP-1 is an important mediator of monocyte infiltration [19] and shown to be over-expressed in human eyes during acute anterior uveitis [49] and during EIU in rats [13]. NF-κB up-regulates transcription of MCP-1 gene [48, 50]. Indeed, we found augmented production of MCP-1 in the aqueous humor of EIU rats. The augmented MCP-1 level was significantly decreased when treated with losartan in a dose-dependent manner. We consider that the suppression of MCP-1 production by an administration of losartan leads to reduced monocyte recruitment in the inflamed ocular tissue.

It has recently been reported that the AT1 receptor antagonist shows an anti-inflammatory effect by inhibiting retinal ICAM-1 upregulation, leukocyte adhesion and infiltration [26, 32]. It seems that the AT1 receptor antagonist has several pathways to suppress ocular inflammation. Our present results demonstrated a new pathway by which an AT1 blocker directly inhibited NF-κB activation and ameliorated acute ocular inflammation in the EIU system.

On the other hand, losartan administration suppressed a clinical score of EAU only at an early phase. As mentioned previously, losartan has an anti-inflammatory effect by inhibiting the expression of adhesion molecules. Thus, it seems possible that this very limited suppression of EAU is attributable to the inhibitory effect of losartan on infiltration of the inflammatory cells into the eyes that express reduced adhesion molecules. However, we found that oral administration of losartan showed no suppressive effect on generation of hIRBP-specific T cell in the lymph nodes. These results suggest that losartan did not affect activation of hIRBP-specific T cells by the antigen presenting cells in lymph nodes. There are various differences between EIU and EAU models examined in the present study. EIU was induced rapidly by LPS in rats, whereas EAU was induced by immunization of mice with hIRBP, an analogue protein antigen for self murine IRBP, and it took a relatively long period to induce inflammatory manifestation in the latter model. Furthermore, a single intravenous administration was given to EIU rats, whereas multiple oral administrations were performed in EAU mice. We considered that the oral administration of losartan did not affect NF-κB activation of antigen presenting cells in the regional lymph nodes, and stimulation of hIRBP-specific Th1 cells normally occurred in the EAU mice. These points, especially the status of NF-κB activation of antigen-presenting cells in the regional lymph nodes in the EAU model in the presence of losartan should be pursued in further studies.

We have reported that a high-potency NF-κB inhibitor, pyrrolidine dithiocarbamate (PDTC), ameliorated EAU when intraperitonealy administered for a long period [23]. It is possible that PDTC is more potent inhibitor of NF-κB than losartan. Indeed, PDTC inhibited the translocation of NF-κB into the nucleus of EAU retina [23]. However, PDTC has never been used yet in clinical practice, since there are concerns about its safety. Meanwhile, losartan is used widely in safety, and a clinical dose of losartan showed the anti-inflammatory effect on EIU in the present study. Thus, it should be examined whether the losartan (clinical dose) really suppresses the NF-κB activation in the antigen-presenting cells in the regional lymph nodes and retina in the EAU model as well, when administered intraperitoneally.

Another point between EIU and EAU may be species. We induced EIU in rats and EAU in mice in this study. Human, rat, and mouse AT1 receptor isoforms are pharmacologically indistingishable [10]. Rat and mouse ANGII receptors have coevolved, and cloned mammalian ANGII receptors share their AT1 subtypes, which can recognize losartan with high affinity [40, 41]. It means that species may not be really acknowledged as a problem in this study.

There is growing evidence suggesting that among various AT1 receptor antagonists only losartan has AT1 receptor-independent actions primarily related to anti-inflammatory and antiaggregatory mechanisms. These properties are not shared by other angiotensin receptor blockers, such as candesartan, valsartan, and ACE inhibitors [39]. Losartan is a prodrug that is converted to active form in vivo. Thus, losartan undergoes the firstpass metabolism in the liver and is converted into EXP3174, whose affinity to the AT1 receptor is ten times higher than that of losartan. Another metabolite of losartan, EXP3179, seems to possess anti-inflammatory and anti-aggregatory properties, though it does not interfere with ligand binding to the AT1 receptor [24]. Since EXP3179 is structurally similar to indomethacin, a conventional nonsteroidal anti-inflammatory drug (NSAID), it seems possible that these metabolites function like as NSAID [45] in suppression of EIU in the present study. Though the exact mechanism of anti-inflammatory effect of losartan still remains unclear, the anti-inflammatory and anti-aggregatory activities of losartan metabolite may be responsible for the suppression of EIU.

In the present study, rats were injected with a usual dose (1 mg/kg) and a 10-times dose (10 mg/kg) of losartan. Though aqueous cytokines and chemokines were suppressed in a dose-dependent manner, the numbers of aqueous humor cells were not significantly different between 1 mg/kg and 10 mg/kg. Since rats have a very small volume of aqueous humor (5 μl/eye), some errors may be observed in case of counting the aqueous cells. Another possible cause may be that a common dose of losartan is good enough to decrease the cell infiltration in the aqueous chamber. In any case, losartan administration ameliorated EIU and suppressed the inflammatory parameters in aqueous humor definitely. When viewed from the point of view of cytokines and chemokines, a higher dose (10 mg/kg) of losartan might be better to suppress the acute ocular inflammation than the common dose of the drug. Since the treatment with losartan induced no marked changes in rat behavior or food and water consumption in the present study, a higher dose of losartan administration may be one option of future clinical application.

In conclusion, we demonstrated that RAS was involved in acute ocular inflammation, and inhibition of AT1 receptor ameliorated EIU in rats. Further, we showed that losartan exerted a suppressive effect on acute inflammation by inhibiting the NF-κB pathway. Although the AT1 receptor blocker did not show an impressive regulatory effect on the antigen-specific T cell proliferative response in EAU model mice, modification of the method of losartan administration, including the dose, route and timing to the mice, should be pursued to eventually determine whether the AT1 receptor blocker is effective in relatively chronic ocular inflammation like EAU. At the present time, we consider that RAS/ AT1 receptor antagonism is one of the prophylactic strategies for ocular acute inflammatory disorders.