SEARCH

SEARCH BY CITATION

Keywords:

  • ischemia;
  • lipopolysacharide;
  • microglia;
  • neuroprotection;
  • toll-like receptor 4

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Reprogramming of toll-like receptor 4 (TLR4) by brief ischemia or lipopolysacharide (LPS) contributes to superintending tolerance against destructive ischemia in brain. However, beneficial roles of TLR4 signaling in ischemic retina are not well known. This study demonstrated that preconditioning with LPS 48 h prior to the retinal ischemia prevents the cellular damage in morphology with hematoxylin and eosin (H&E) staining and functions of retina with electroretinogram (ERG), while post-ischemia treatment deteriorated it. The preventive effects of LPS preconditioning showed the cell type-specificity of retinal cells. There was complete rescue of ganglion cells, partial rescue of bipolar and photoreceptor cells or no rescue of amacrine cells, respectively. LPS treatment caused the proliferation and migration of retinal microglia and its preconditioning prevented the ischemia-induced microglial activation. Preventive actions from cell damages following LPS preconditioning prior to retinal ischemia were abolished in TLR4 knock-out mice, and by pre-treatments with anti-TLR4 antibody or minocycline, a microglia inhibitor, which themselves had no effects on the retinal ischemia-induced damages or microglia activation. Thus, this study revealed that TLR4 mediates the LPS preconditioning-induced preventive effects through microglial activation in the retinal ischemia model.

Abbreviations used
Chx10

ceh-10 homeodomain-containing homolog

ERG

electroretinogram

GCL

the ganglion cell layer

GFAP

glial fibrilary acidic protein

H&E

hematoxylin and eosin

i.vt.

intravitreously

Iba-1

ionized calcium binding adaptor molecule 1

INL

inner nuclear layer

IPL

inner plexiform layer

LPS

lipopolysacharide

NeuN

neuronal nuclei

ONL

outer nuclear layer

OPL

outer plexiform layer

pp38 MAPK

phospho-p38 mitogen-activated protein kinase

TLR4

toll-like receptor 4

WT

wild-type

Ischemia in the central nervous system including retina is one of the most well-known pathophysiological condition, which leads to extensive neuronal damages and functional disorders by triggering diverse types of self-reinforcing destructive mechanisms, such as necrosis and apoptosis (Lipton 1999; Bernstein et al. 2003; Danton and Dietrich 2003; Osborne et al. 2004; Ueda and Fujita 2004; Arumugam et al. 2006; Kaur et al. 2008; Lakhan et al. 2009; Neroev et al. 2010; Iadecola and Anrather 2011). These neuronal damages in retina are caused by ischemia-induced activation of detrimental cascades including up-regulation and secretion of injury-related cytokines derived from retinal glial cells (Neufeld et al. 2002; Osborne et al. 2004; Langmann 2007; Kaur et al. 2008; Neroev et al. 2010; Cervia and Casini 2012). Among these deleterious pathways, toll-like receptor 4 (TLR4) functions as a key component to mediate ischemic damages in the brain and retina through generation of several cytotoxic mediators, and mice lacking functional TLR4 shows partial neuroprotection against brain ischemia (Cao et al. 2007; Caso et al. 2007; Hua et al. 2007; Lehnardt et al. 2008; Marsh et al. 2009b; Dvoriantchikova et al. 2010; Hyakkoku et al. 2010; Ko et al. 2011; Wang et al. 2011). Similar cytotoxic effects are caused by the treatment with lipopolysaccharide (LPS), an endotoxin and known specific ligand for TLR4 (Guha and Mackman 2001; Sheng et al. 2003; Kawai and Akira 2007; Liang et al. 2007; Qin et al. 2007; Litvak et al. 2009; Jeong et al. 2010; Pang et al. 2012). However, it is also reported that the LPS preconditioning leads to robust neuroprotection against lethal cerebral ischemia through TLR4 signaling (Tasaki et al. 1997; Bastide et al. 2003; Rosenzweig et al. 2004; Marsh et al. 2009a, b; Vartanian and Stenzel-Poore 2010; Stevens et al. 2011; Vartanian et al. 2011). Although such contradictory cytotoxic and preventive roles of TLR4-mediated actions against ischemia may be in part attributed to the involvements of secondary produced glial cytokines (Obrenovitch 2008; Shpargel et al. 2008; Pradillo et al. 2009; Vartanian and Stenzel-Poore 2010; Wang et al. 2011; Chen et al. 2012), detailed and sequential machineries through glial cells remain elusive. One of difficulties in the study using cerebral ischemia model and systemic LPS treatment is ascribed to multiple parameters, including actions on peripheral immune cells, vascular endothelial cells, and brain glial cells. Even if using the intracerebroventricular administration of LPS, sequential analyses through glial cells seem to be difficult because of its regionally uneven brain distribution.

To answer this problem, we decided to use the retinal ischemia model, since it is very convenient, in terms of the simplicity of histological and functional studies (Fujita et al. 2009; Perlman 2009; Ueda et al. 2010, 2012; Halder et al. 2013). In addition, retinal cell layers are organized in proper orders, along with availability of biomarkers for specific cells (Sherry et al. 2006; Rhee et al. 2007; Buckingham et al. 2008; Masland 2012; Halder et al. 2013). This cell layer function is also evaluated by use of electroretinogram (Fujita et al. 2009; Perlman 2009; Halder et al. 2013). Most importantly, in the experiment using intravitreous injection, we can discuss the effects on all retinal cell layers as well as glial cells in closed and very small volume of space. In this study, we attempted to characterize the LPS-induced protection of retinal cells and its glial involvements in the retinal ischemia model.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals

Male C57/BLJ mice weighing 20–30 g were purchased from Tagawa Experimental Animals (Nagasaki, Japan). Toll-like receptor 4 knock-out (TLR4−/−) mice were from Laboratory of Host Defense, World Premier International Immunology Frontier Research Center, Osaka University, Japan (Hoshino et al. 1999). Mice were kept in a room maintained at constant temperature (21 ± 2°C) and relative humidity (55 ± 5%) with an automatic 12 h light/dark cycle with free access to standard laboratory diet and tap water. Animal care and all experimental procedures were formally approved by Nagasaki University Animal Care and Use Committee (Animal Experiments Approval Number: 1104190914).

LPS administration

LPS (Escherichia coli serotype 0111:B4, catalog # L2630, Sigma-Aldrich, St Louis, MO, USA) was dissolved in autoclaved and K+-free phosphate buffered saline (PBS). Following the protocol of injection in the eye as described previously (Halder et al. 2013), LPS was intravitreously injected (1 μg/μL, i.vt.) in the eye 48 h, 24 h, and 30 min before retinal ischemia. Mice were treated (i.vt.) with LPS at doses of 0.01, 0.1, 1, and 5 μg/μL 48 h before retinal ischemia (preconditioning). LPS was injected (1 μg/μL; i.vt.) in the eye of non-ischemic mice and killed at 3 h, 6 h, 12 h, day 1, day 2, and day 3 after LPS treatment. In addition, LPS was administered (1 μg/μL; i.vt.) in the eye at 3 h after retinal ischemia (post-conditioning). Respective vehicles were prepared by treatment (i.vt.) with equal volume of PBS.

TLR4 antibody and minocycline treatments

Monoclonal antibodies (clones: Sa15-21 and MTS510) against TLR4 were developed by Dr. Kensuke Miyake, Institute of Medical Science, University of Tokyo, Japan (Akashi-Takamura et al. 2006). Both MTS510 and Sa15-21 antibodies were diluted in K+-free PBS and individually injected (0.5 μg/μL, i.vt.) in the eye 4 h before LPS treatment. For combined antibodies treatment, 0.5 μg MTS510 was mixed with 0.5 μg Sa15-21 and injected (1 μg/μL; i.vt.) 4 h before LPS and PBS administration. In addition, combined antibodies was administered (1 μg/μL; i.vt.) in the eye 30 min before retinal ischemia.

Minocycline (Catalog # M9511-1G; Sigma-Aldrich) was dissolved in sterile saline, fresh solution being made daily and injected intraperitoneally (90 mg/kg, i.p.) in mouse 24 h and 0 h before LPS treatment in the ischemic mice. Minocycline was injected (90 mg/kg, i.p.) twice (at 0 h as well as at 24 h after the first delivery) in non-ischemic mice. Following similar ways, respective vehicles were prepared by treatment with equal volume of PBS.

Retinal ischemia

Retinal ischemia was performed following the method as described previously (Halder et al. 2013). Briefly, mice were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and pupils were fully dilated with 1% atropine sulfate drops (Nitten, Nagoya, Japan). The anterior chamber of the eye was cannulated with a 33-gauge needle attached to an infusion container of sterile intraocular irrigating solution (BSS PLUS dilution buffer; Alcon, Fort Worth, TX, USA). Retinal ischemia was induced by elevating the intraocular pressure (IOP) to generate a hydrostatic pressure of 130 mm Hg for 45 min by lifting the container. Following 45 min after retinal ischemic stress, the needle was withdrawn and 0.3% ofloxacin (Santen Pharmaceutical Co. Ltd., Osaka, Japan) was applied topically into the eye to avoid infection.

Electroretinogram (ERG)

ERG study was performed following the protocol as previously described (Fujita et al. 2009; Halder et al. 2013). Briefly, mice were dark-adapted for 3–4 h, then anesthetized with pentobarbital sodium (50 mg/kg, i.p.) and pupils were dilated with 1% atropine. A contact electrode (KE–S; Kyoto contact lenses, Kyoto, Japan) was placed topically on the corneal apex and reference electrode was placed near the ipsilateral eye. The ground was a subdermal platinum needle electrode near the abdominal area. ERGs were produced by 20 J flash intensities. The flash stimulus source (SLS-3100; Nihon Kohden, Tokyo, Japan) illuminated the eye by diffuse reflection off the interior surface of the ganzfeld. Maximum flash luminance was measured with detector (MEB–9104; Nihon Kohden). After the intensity series, an incandescent background light sufficient to desensitize the rod system was turned on, and ERGs produced by the standard stimulus were recorded every 2 min for 20 min. The background was then turned off, and ERGs were produced by the standard stimulus every 2 min for the first 30 min of dark adaptation. The a- and b-wave amplitudes were measured online (Neuropack m, QP-903B; Nihon Kohden). ERG was performed at day 4 and 7 after retinal ischemia.

Retinal tissue preparation

For retinal tissue preparation, mice were deeply anesthetized with sodium pentobarbital (50 mg/kg, i.p.). Eye was quickly isolated, washed with saline and 4% paraformaldehyde (PFA). Eye was then nicked through pupil, post-fixed in 4% PFA for 24 h and finally transferred to 25% sucrose solution (in 0.1 M K+-free PBS) overnight at 4°C for cryoprotection. Eye was frozen in cryoembedding compound and retinal sections were cut at 10 μm thickness.

Morphological assessment of retinal damages

For hematoxylin and eosin (H&E) staining, frozen retinal sections were washed with 0.1 M K+-free PBS, immerged in Mayer's hematoxylin solution (WAKO, Osaka, Japan) for 5 min at 25°C and then washed with tap water for 20 min. Following brief treatment with 95% ethanol, sections were immerged in eosin-alcohol solution (WAKO) for 4 min at 25°C. Sections were dehydrated through a series of ethanol solutions, xylene, and over-slipped with Permount (Fisher Scientific, Waltham, MA, USA). Sections were then analyzed using a BZ-8000 microscope with BZ Image Measurement Software (KEYENCE, Osaka, Japan).

Immunohistochemical analysis

To perform fluorescence immunohistochemistry, retinal sections were washed with 0.1 M K+-free PBS and incubated with 50% methanol followed by 100% methanol for 10 min. Following treatment with blocking buffers [1–3% bovine serum albumin (BSA) as well as 10% goat serum with 0.1% Triton X-100 in phosphate buffered saline (PBST)], retinal sections were incubated overnight at 4°C with following primary antibodies: anti-NeuN (1 : 100, mouse monoclonal IgG1, clone A60; Chemicon, Temecula, CA, USA), anti-syntaxin-1 (1 : 500, mouse monoclonal; Sigma-Aldrich), anti-glial fibrilary acidic protein (GFAP) (1 : 500, rabbit polyclonal; Glostrup, Dako, Denmark), anti-Iba-1 (1 : 1000, rabbit polyclonal; WAKO), anti-Chx10 (1 : 300, sheep polyclonal; Exalpha Biologicals, Shirley, MA, USA), anti-phosho-p38 MAPK (1 : 1000, mouse monoclonal; Cell Signalling, Danvers, MA, USA), and anti-TLR4 (1 : 100, clone: MTS510; developed by Dr. Kensuke Miyake, Institute of Medical Science, University of Tokyo). Sections were then incubated with Alexa Fluor 488-conjugated anti-mouse IgG, Alexa Fluor 594-conjugated anti-rabbit IgG and Alexa Fluor 488-conjugated anti-sheep IgG secondary antibodies (1 : 300; Molecular Probes, Eugene, OR, USA). The nuclei were visualized with Hoechst 33342 (1 : 10 000; Molecular Probes). Samples were then washed thoroughly with PBS and cover-slipped with Perma Fluor (Thermo Shandon, Pittsburgh, PA, USA). Images were collected using a BZ-8000 microscope with BZ Image Measurement Software (KEYENCE).

Cell counting

Measurements of Iba-1-, NeuN-, Chx10-, and Syntaxin-1-positive cells in the retina were performed using the BZ Image Measurement software (KEYENCE). Cell counts were carried out in bright field images following the protocol as reported previously (Halder et al. 2012). Briefly, the number of cells in each image was counted (in between 1 mm distance from the optic nerve of retina) in random square fields (approximately 200 μm × 200 μm) in the control and ischemic retina of different treatment groups. However, we carried out the counts using H&E staining and specific cell markers with clearly visible nuclei by Hoechst staining to determine the Iba-1-, NeuN-, Chx10-, Syntaxin-1-, and photoreceptor-positive cells in each image. The quantification was expressed as average percentage of the number of cells in the one to three retinal sections per mouse to obtain a mean value for each animal. Experiments were carried out using three mice, unless otherwise stated for each group.

Quantitative real-time polymerase chain reaction

Total RNA was extracted from retina using Total RNA Extraction Miniprep System (VIOGENE, New Taipei, Taiwan), and 500 ng of RNA was used for cDNA synthesis. Quantitative real-time polymerase chain reaction (RT-PCR) was performed with qPCR MasterMix Plus for SYBR Green I (Eurogentec, Seraing, Belgium) using the Light Cycler ® 480 II (Roche, Penzberg, Upper Bavaria, Germany). Primer sequences were as follows: IL1RN (interleukin 1 receptor antagoninst), 5′-AAGCCTTCAGAATCTGGGA-3′ (forward) and 5′-GTCCTTGTAAGTACCCAGC-3′ (reverse); IFIT1 (interferon-induced protein with tetratricopeptide repeats 1), 5′-CTGAGATGTCACTTCACATGGAA-3′ (forward) and 5′-GTGCATCCCCAATGGGTTCT-3′ (reverse); SOCS1 (suppressor of cytokine signaling 1), 5′-CTGCGGCTTCTATTGGGGAC-3′ (forward) and 5′-AAAAGGCAGTCGAAGGTCTCG-3′ (reverse); SOCS2, 5′-TGCGGATTGAGTACCAAGATGG-3′ (forward) and 5′-CTGTCCGTTTATCCTTGCACA-3′ (reverse); SOCS3, 5′-TGCGCCTCAAGACCTTCAG-3′ (forward) and 5′-GCTCCAGTAGAATCCGCTCTC-3′ (reverse); TNFα (tumor necrosis factor α), 5′-CCCTCACACTCAGATCATCTTCT-3′ (forward) and 5′-GCTACGACGTGGGCTACAG-3′ (reverse); IL-1β (interleukin-1β), 5′-TTCAGGCAGGCAGTATCACTC-3′ (forward) and 5′-GAAGGTCCACGGGAAAGACAC-3′ (reverse); IL-6, 5′-CATAGCTACCTGGAGTACATGA-3′ (forward) and 5′-CATTCATATTGTCAGTTCTTCG-3′ (reverse); cyclooxygenase-2 (COX-2), 5′-TTCAACACACTCTATCACTGGC-3′ (forward) and 5′-AGAAGCGTTTGCGGTACTCAT-3′ (reverse); MCP-1 [monocyte chemoattractant protein-1, Ccl2: chemokine (C-C motif) ligand 2]. 5′-TTAAAAACCTGGATCGGAACCAA-3′ (forward) and 5′-GCATTAGCTTCAGATTTACGGGT-3′ (reverse), and GAPDH, 5′-TATGACTCCACTCACGGCAAAT-3′ (forward) and 5′-GGGTCTCGCTCCTGGAAGAT-3′ (reverse). The cycling conditions for all primers were 5 min at 95°C, followed by 45 cycles of 15 s at 95°C and 60 s at 60°C. In all cases, the validity of amplification was confirmed by the presence of a single peak in the melting temperature analysis and linear amplification against the PCR cycles.

Statistical analysis

All results are shown as means ± standard error of the mean (SEM). Two independent groups were compared using the Student's t-test. Multiple groups were compared using Dunnett's multiple comparison tests after a one-factor analysis of variance (anova) or a repeated measure anova. < 0.01 and 0.05 were considered significant.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Prevention of retinal ischemic damages by LPS preconditioning

To examine the time-dependent preventive action of LPS preconditioning in ischemic retina, LPS was intravitreously injected (1 μg/μL, i.vt.) in the eye 48 h, 24 h, and 30 min before retinal ischemia. The H&E staining data showed that the cellular loss and decrease in retinal thickness are significantly blocked at day 7 after the ischemic stress in mice treated with LPS 48 h and 24 h before ischemia, but the maximum prevention was observed by LPS treatment 48 h before ischemia, whereas LPS injection 30 min before ischemic stress induced no preventive action (Fig. 1a and b). The ERG analysis revealed the maximum increase in amplitudes of a- and b-waves at day 7 after retinal ischemia in mice treated by LPS 48 h before the ischemic stress (Fig. 1c and d). These a- and b-wave amplitudes were also significantly rescued in ischemic mice treated with LPS 24 h, but not 30 min before retinal ischemia (Fig. 1c and d)

image

Figure 1. Time-dependent beneficial effects of lipopolysacharide (LPS) preconditioning on retinal ischemia-induced damages. Following LPS injection intravitreously (1 μg/μL, i.vt.) and phosphate buffered saline (PBS) treatment (vehicle, veh) 48 h, 24 h, and 30 min before retinal ischemia, representative histological data using hematoxylin and eosin (H&E) staining of retinal sections (a), measurement of retinal thickness extended from ganglion cell layer (GCL) to outer nuclear layer (ONL) (b), a-wave (c), and b-wave (d) amplitudes of electroretinogram (ERG) analysis are performed at day 7 after the ischemic stress. Data are mean ± SEM. (*< 0.05, vs. Control, #< 0.05, vs. Veh) from experiments using three mice for each group.

Download figure to PowerPoint

Next, we determined the LPS dose-dependent prevention of ischemic retina by LPS treatment (i.vt.) with doses of 0.01, 0.1, 1, and 5 μg/μL 48 h before retinal ischemia. Following retinal ischemia, the findings using H&E staining revealed that the number of cells in different retinal layers and total retinal thickness extended from ganglion cell layer (GCL) to outer nuclear layer (ONL) are significantly decreased in the PBS (vehicle)-pre-treated mice at day 7 after ischemic stress, compared to control (Fig. 2a and d). In contrast, 0.1 μg, but not 0.01 μg LPS preconditioning significantly blocked this cellular loss and decrease in retinal thickness at day 7, followed by maximum prevention by 1 μg LPS, and then declined by 5 μg LPS (Fig. 2a and d). The ERG analysis revealed that a- and b-wave amplitudes are significantly decreased at day 7 after retinal ischemia in vehicle-pre-treated mice, whereas 1 μg LPS pre-treatment maximally prevented the retinal ischemic damages at day 7, though the significant protection was observed in ischemic mice pre-treated with 0.1 and 5 μg LPS, but not with 0.01 μg (Fig. 2b and c). In addition, 1 μg LPS completely prevented the cells in the GCL (Fig. 2e), whereas dose-dependent prevention of inner plexiform layer (IPL) and inner nuclear layer (INL) in ischemic retina was observed between 0.01 and 1 μg LPS, but there was decline with 5 μg (Fig. 2f and g, respectively). However, the maximum prevention of outer plexiform layer (OPL) and ONL in ischemic retina was incomplete, but no decline was observed when pre-treated with 0.1, 1, and 5 μg LPS (Fig. 2h and i, respectively).

image

Figure 2. Lipopolysacharide (LPS) dose-dependently prevents the retinal ischemia-induced damages. LPS is injected (i.vt.) with doses of 0.01, 0.1, 1, and 5 μg/μL in the eye 48 h before retinal ischemia. (a–d) Following LPS and phosphate buffered saline (PBS) (veh) pre-treatments, representative data using H&E staining of retinal sections (a), measurement of retinal thickness from ganglion cell layer (GCL) to outer nuclear layer (ONL) (b), a-wave (c), and b-wave (d) amplitudes of electroretinogram (ERG) study are performed at day 7 after the ischemic stress. (e–i) LPS dose-dependent prevention of different layers of ischemic retina. (e) Complete prevention of cells in the GCL is observed at day 7 after retinal ischemia in mice pre-treated with 1 μg LPS. (f, g) Dose-dependency in inner plexiform layer (IPL) and inner nuclear layer (INL) of ischemic retina is observed at day 7 in between 0.01 and 1 μg LPS, but there is a decline in thickness by 5 μg LPS. (h, i) Maximal prevention of outer plexiform layer (OPL) and ONL is incomplete, but no decline is observed at day 7 after retinal ischemia. Data are mean ± SEM. (*< 0.05, vs. Control, #< 0.05, vs. Veh) from experiments using three mice for each group.

Download figure to PowerPoint

LPS-induced retinal cell type-specific survival against ischemic damages

To determine LPS preconditioning-induced cell type-specific survival in ischemic retina, LPS was injected (1 μg/μL, i.vt.) in the eye 48 h before retinal ischemia. The fluorescence immunostaining data showed that NeuN-positive neurons in the GCL (Buckingham et al. 2008) are significantly decreased at day 7 after retinal ischemia, whereas LPS pre-treatment completely recovered these NeuN-positive ganglion cells at day 7 after ischemic stress (Fig. 3a and g). In addition, the partial but significant blockade of the loss of Chx10-positive bipolar cells (Rhee et al. 2007) in the INL and photo-receptor cells in the ONL was observed at day 7 in LPS-preconditioned ischemic mice, compared to respective vehicle-treated mice (Fig. 3b, d, respectively, and g). However, LPS preconditioning caused no prevention of ischemia-induced decrease in syntaxin-1-positive amacrine cells (Sherry et al. 2006), of which the cell bodies and processes are located in the INL and IPL, respectively (Fig. 3c and g).

image

Figure 3. Retinal cell type-specific prevention by lipopolysacharide (LPS) preconditioning. LPS is injected (1 μg/μL, i.vt.) in the eye 48 h before retinal ischemia and fluorescence immunohistochemical analysis of retinal sections is performed at day 7 after the ischemic stress. (a–d) The immunostaining of ganglion cells (a) in the ganglion cell layer (GCL) (NeuN, green), bipolar cells (b) in the inner nuclear layer (INL) (Chx10, green), amacrine cells (c) of which the cell body and process are located in the INL and inner plexiform layer (IPL), respectively, (syntaxin–1, green), photoreceptor cells (d) in the outer nuclear layer (ONL) (Hoechst: blue), microglia (e: Iba–1, red), and astrocytes (f: GFAP, red) in the retinal sections are carried out at day 7 after retinal ischemia. The higher magnification views of lower panels in (a–f) indicate the expression of retinal cell types noted by dotted rectangles (respective upper panels). (g) Quantitative analysis of different cell types in control, phosphate buffered saline (PBS)-, and LPS-pre-treated ischemic retina at day 7. Data represent the means ± SEM. (*< 0.05, vs. Control, #< 0.05, vs. Veh, ##p < 0.01, vs. Veh) from experiments using three mice for each group.

Download figure to PowerPoint

The immunohistochemical analysis against Iba-1, a microglial marker, showed that Iba-1-positive microglia is normally located in the GCL and IPL of control retina, whereas ischemic stress-induced activation and migration of Iba-1-positive microglia was observed throughout the different retinal cell layers at day 7 (Fig. 3e). This microglial activation was significantly diminished at day 7 after ischemic stress in LPS-preconditioned mice. On the other hand, GFAP-positive astrocytes were normally expressed in the GCL of control retina, followed by retinal ischemia-induced activation of these astrocytes from GCL to ONL at day 7 after the ischemic stress (Fig. 3f). There was no significant change in ischemia-induced astrocyte activation between vehicle and LPS pre-treatment at day 7 after ischemia, indicates that LPS preconditioning has no significant effect on the retinal astrocytes upon ischemia (Fig. 3f).

Involvement of TLR4 in LPS-induced prevention of retinal ischemic damages

According to the fact, LPS is one of the specific ligand for TLR4, we examined whether TLR4 is implicated in LPS preconditioning-induced retinal protection against ischemic damages using TLR4 knock-out (TLR4−/−) mice. Following LPS injection (1 μg/μL, i.vt.) in TLR4 knock-out (TLR4−/−) mice 48 h before retinal ischemia, the H&E staining also showed the significant decrease in cells number in the different layers and the total retinal thickness extending from GCL to ONL at day 7 after retinal ischemia in TLR4−/− mice, whereas the total retinal thickness was slightly, but not significantly increased in TLR4−/− ischemic retina than the WT ischemic retina at day 7 (Fig. 4a and b). The ERG analysis data revealed that a- and b-wave amplitudes are markedly decreased in TLR4−/− mice at day 7 after the ischemic stress, showing the slight, but insignificant increase in these amplitudes at the same time point in TLR4−/− ischemic mice, compared to amplitudes in WT ischemic mice (Fig. 4c and d). From detailed morphological analysis using H&E staining, we found that the thickness of IPL and INL is markedly protected in TLR4−/− mice at day 7 after ischemia, though ischemia-induced loss of cells in the GCL as well as the thickness of OPL and ONL was not significantly rescued at day 7 in TLR4−/− mice (Fig. 4e–i). Interestingly, both the H&E staining and ERG analysis data showed that the effect of LPS preconditioning in terms of retinal protection is abolished in TLR4−/− mice at day 7 after ischemia (Fig. 4a–i).

image

Figure 4. Blockade of lipopolysacharide (LPS) preconditioning-induced retinal protection in toll-like receptor 4 (TLR4)−/− mice. (a–d) LPS preconditioning-induced retinal protection is abolished in TLR4−/− mice. Following LPS treatment (1 μg/μL, i.vt.) 48 h before retinal ischemia in wild-type (WT) and toll-like receptor4 knock-out (TLR4−/−) mice, the H&E staining of retinal sections (a), measurement of total retinal thickness extending from ganglion cell layer (GCL) to outer nuclear layer (ONL) (b), a-wave (c), and b-wave (d) amplitudes of electroretinogram (ERG) analysis are performed at day 7 after retinal ischemia. (e–i) TLR4−/− mice show significant retinal protection against ischemic damages. Following day 7 after retinal ischemia, measurement of cells in the GCL (e), thickness of inner plexiform layer (IPL) (f), inner nuclear layer (INL) (g), OPL (h), and ONL (i) in the retina of WT and TLR4−/− mice. The thickness of IPL and INL is significantly protected in TLR4−/− mice upon retinal ischemia Data are mean ± SEM. (*< 0.05, vs. each Control, #< 0.05, vs. each Veh, < 0.05, vs. each WT Veh and LPS) from experiments using three mice for each group.

Download figure to PowerPoint

Previous studies reported that two monoclonal antibodies Sa15-21 and MTS510 against mouse TLR4/myeloid differentiation protein 2 (MD-2) are reacted with TLR4 through distinct epitopes, because these two antibodies do not cross-block with each other (Akashi et al. 2003; Akashi-Takamura et al. 2006). Sa15-21 reacts with the mouse N-terminal leucine-rich repeat, and MTS510 reacts with rest of the extracellular domain of mouse TLR4. MTS510 binds to TLR4/MD-2 in the absence of LPS and Sa15-21 reacts with TLR4/MD-2 after stimulation with LPS, because the MTS510 epitope is disappeared by LPS stimulation without any effect on the epitope for Sa15-21. It is also noted that these antibodies against TLR4/MD-2 protects mice from LPS-induced acute lethal hepatitis (Akashi-Takamura et al. 2006). On basis of the previous study, monoclonal antibodies (MTS510 and Sa15-21) against functional TLR4 were injected (i.vt.) in the eye 4 h before LPS treatment (1 μg/μL, i.vt.) to confirm the LPS-caused retinal protection against ischemic damages. Following retinal ischemia 48 h after LPS injection, the H&E staining data showed that LPS-induced prevention of the loss of retinal cells in the different layers and the decrease in retinal thickness is partially blocked at day 7 after ischemic stress by treatment with MTS510 antibody (0.5 μg/μL), but not with 0.5 μg of Sa15-21 (Fig. 5a and b). Treatment with 1 μg combined antibodies (0.5 μg of each antibody, i.vt.) completely blocked the LPS-induced recovery of this functional damages at day 7 after ischemic stress (Fig. 5a and b), being consistent with the previous study that TLR4 is blocked by MTS510 antibody, whereas Sa15-21 antibody itself exerts additive effects on the TLR4 (Akashi et al. 2003; Akashi-Takamura et al. 2006). In addition, anti-TLR4 antibody treatment also showed the similar blocking effect on LPS action in the ERG analysis (Fig. 5c and d). These results indicate the LPS preconditioning-induced retinal protection is mediated by TLR4-signaling. However, no protective effect against ischemic damages was observed at day 7 when treated with combined anti-TLR4 antibody at the same time point in PBS-preconditioned ischemic mice (Fig. 5a–d).

image

Figure 5. Toll-like receptor 4 (TLR4) antibody blocks lipopolysacharide (LPS) preconditioning-induced prevention of retinal ischemic damages. Mice are treated (i.vt.) with TLR4 antibodies Sa15-21 (Sa) and MTS510 (MTS) with doses of 0.5 μg/μL of each antibody as well as 1 μg/μL of combined antibodies 4 h before 1 μg LPS injection. (a–d) Following retinal ischemia 48 h after LPS treatment, H&E staining of retinal sections (a), measurement of total retinal thickness extending from ganglion cell layer (GCL) to outer nuclear layer (ONL) (b), a-wave (c), and b-wave (d) amplitudes of electroretinogram (ERG) analysis are performed at day 7 after retinal ischemia. Data are mean ± SEM. (*< 0.05, vs. Veh, #< 0.05, vs. LPS) from experiments using three mice for each group.

Download figure to PowerPoint

Altered gene expression of TLR4 signaling outcomes by LPS preconditioning

To observe the LPS-induced expression of neuroprotective/anti-inflammatory genes in retina, LPS was injected (1 μg/μL, i.vt.) in the eye of non-ischemic mice. Following real-time polymerase chain reaction (RT-PCR) 24 h after LPS treatment, the significant increase in mRNA levels of anti-inflammatory (neuroprotective) genes IL1RN and IFIT1 was observed in retina at this time point after LPS administration, compared to vehicle-treated mice (Table 1). We also found that mRNA levels of SOCS 1 and SOCS 3, but not SOCS 2, suppressor genes of the myeloid differentiation factor (MyD)88 – nuclear factor κB (NF-κB) pathway of TLR4 signaling, are significantly increased at 24 h after LPS treatment (Table 1). Interestingly, treatment with combined anti-TLR4 antibodies (1 μg/μL, i.vt.) 4 h before LPS administration, the LPS-induced up-regulation of the mRNA levels of IL1RN and IFIT1 as well as SOCS 1 and SOCS 3 was markedly decreased at 24 h after LPS treatment (Table 1).

Table 1. LPS preconditioning-induced up-regulation of neuroprotective genes in retina
Name of genesLPS mRNA change (ratio of veh)LPS + anti-TLR4 mRNA change (ratio of veh)
  1. LPS (1 μg/μL) and PBS (vehicle) are administered (i.vt.) in the non-ischemic eye and retina is collected at 24 h after LPS treatment. Following extraction of total mRNA from retina, quantitative real-time polymerase chain reaction (RT-PCR) is carried out to measure mRNA levels of interleukin-1 receptor antagonist gene (IL1RN), interferon-induced protein with tetratricopeptide repeats 1 (IFIT1) as well as suppressor of cytokine signaling (SOCS): SOCS 1, SOCS 2, and SOCS 3. The quantification was indicated as average ratio of the vehicle (veh) in the absence of ischemia. Data represent as the means ± SEM. (*< 0.05, vs. Veh, **< 0.05, vs. LPS) from experiments using four mice for each group.

IL1RN3149.42 ± 1862.47*1.58 ± 0.67**
IFIT1126.17 ± 32.31*3.16 ± 0.82**
SOCS 17.52 ± 1.47*0.65 ± 0.13**
SOCS 21.14 ± 0.201.19 ± 0.29
SOCS 327.10 ± 9.92*0.52 ± 0.27**

Next, to analyze the pattern of LPS preconditioning-induced expression of inflammatory/injury genes upon ischemia, LPS (1 μg/μL) and PBS were administered (1 μg/μL, i.vt.) in the eye 48 h before retinal ischemia. The RT-PCR data showed that mRNA levels of injury genes TNFα, IL-1β, IL-6, COX-2, and MCP-1 are markedly increased in retina at 24 h after the ischemic stress in vehicle-treated mice, whereas LPS preconditioning caused significant decrease in mRNA levels of these injury genes 24 h after ischemia (Table 2).

Table 2. LPS preconditioning blocks the ischemia-induced up-regulation of injury genes
Name of genesVeh + ischemia mRNA change (ratio of control)LPS + ischemia mRNA change (ratio of control)
  1. LPS (1 μg/μL) and PBS (vehicle) are administered (i.vt.) in the eye 48 h before retinal ischemia, and retinal tissue is collected at 24 h after the ischemic stress. Following extraction of total mRNA from retina, quantitative real-time polymerase chain reaction (RT-PCR) is performed to measure mRNA levels of tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), interleukin-6 (IL-6), cyclooxygenase 2 (COX2), and monocyte chemoattractant protein-1 (MCP-1). The quantification was indicated as average ratio of the control (in the absence of ischemia). Data represent as the means ± SEM. (*< 0.05, vs. Control: In the absence of ischemia, **< 0.05, vs. Veh + ischemia) from experiments using three mice for each group.

TNFα32.04 ± 19.99 × 105*4.55 ± 1.09**
IL1β245.00 ± 0.00*2.00 ± 0.78**
IL630.60 ± 3.87 × 105*244.03 ± 27.32**
COX266.43 ± 25.19 × 103*19.07 ± 2.55**
MCP-113.04 ± 7.35 × 103*0.54 ± 0.11**

Expression of TLR4 in retinal microglia

LPS preconditioning significantly blocked the ischemia-induced activation of microglia in retina (Fig. 3e), therefore we analyzed expression pattern of TLR4 in microglia. To observe LPS- and ischemia-induced expression levels of TLR4 in retinal microglia, double fluorescence immunohistochemical analysis of control, LPS-treated (1 μg/μL), and ischemic retinal sections were carried out using antibodies against Iba-1 and TLR4. The findings revealed that TLR4 is expressed in Iba-1-positive microglia in the GCL and IPL of control retina, whereas the number of TLR4-positive microglia was significantly increased throughout the different retinal cell layers at day 2 after treatment with 1 μg of LPS (Fig. 6a and b). On the other hand, the number and level of TLR4 was increased gradually in activated Iba-1-positive microglia in different cell layers in retina through day 1 (data are not shown), day 2, and day 7 after the ischemic stress (Fig. 6a and b). However, the expression of TLR4 was also observed in Iba-1-negative retinal cells in different layers at day 2 and day 7 after ischemia (Fig. 6a). It is interesting to note that microglia in LPS preconditioning showed the rod shape, a resting state; on the other hand, the microglia in the ischemic retina was ameboid in shape, a cytotoxic-activated state.

image

Figure 6. Lipopolysacharide (LPS) and ischemia increase the number of toll-like receptor 4 (TLR4)-expressing microglia. (a) The double fluorescence immunohistochemical analysis of retinal sections is performed using antibodies against ionized calcium-binding adaptor molecule 1 (Iba-1), a microglial marker, and TLR4. Immunostaining data show the expression of TLR4 in Iba-1-positive microglia in the ganglion cell layer (GCL) and inner plexiform layer (IPL) of control retina (Iba-1, red; TLR4, green). The number of TLR4-positive microglia is increased at day 2 after LPS treatment (1 μg/μL, i.vt.) in non-ischemic retina. Following retinal ischemia, TLR4-positive microglia is increased at day 2 and 7 after the ischemic stress. The higher magnification views of lower panels in (a) indicate the expression of TLR4 in Iba-1-positive microglia noted by dotted rectangles (respective upper panels). (b) Quantitative analysis of TLR4-positive microglia in control, LPS-treated, and ischemic retina. Data represent the means ± SEM. (*< 0.05, vs. Control) from experiments using three mice for each group.

Download figure to PowerPoint

Microglia is implicated in LPS-TLR4 signaling

Several reports suggested the common and significant up-regulation of p38 mitogen-activated protein kinase (p38 MAPK) and its phosphorylation (pp38 MAPK) in activated microglia in the central nervous system including retina and brain following injury/ischemia and disease conditions (Koistinaho and Koistinaho 2002; Jin et al. 2003; Ibrahim et al. 2011), though ischemia-induced expression of pp38 MAPK in other retinal cell types has also been reported (Roth et al. 2003). In addition, pp38 MAPK is associated with NF-κB-mediated expression of IL-1β, IL-6, TNFα, COX-2, and inducible nitric oxide synthase/iNOS (Koistinaho and Koistinaho 2002). On the basis of previous studies, we used pp38 MAPK as a marker for activated microglia in retina. For the examination of time-dependent microglial activation by LPS in vivo, mice were treated with LPS (1 μg/μL, i.vt.) and double immunostaining of Iba-1 and pp38 MAPK was carried out at 3 h, 6 h, 12 h, day 1, day 2, and day 3 after LPS injection. The findings revealed the expression of Iba-1-positive microglia in the GCL and IPL of control retina (Fig. 7a and c). Following LPS treatment, Iba-1-positive microglia was gradually activated mildly, proliferated, and migrated throughout different retinal cell layers from 3 h to day 1 and persisted on day 2, then declined at day 3 (Fig. 7a and c). We observed that pp38 MAPK is expressed in some Iba-1-positive microglia from 6 h to 12 h after LPS injection, followed by decrease in pp38 MAPK signals in microglia through day 2–3 (Fig. 7a). The LPS-caused proliferation and migration of Iba-1-positive microglia were completely blocked at day 2 when treated with combined anti-TLR4 antibodies (1 μg/μL, i.vt.) 4 h before LPS injection (Fig. 7b and d). In addition, these LPS effects on retinal microglia were also completely abolished in TLR4−/− mice (Fig. 7b and d). These results indicate that LPS-induced microglial priming is mediated through TLR4.

image

Figure 7. Involvement of toll-like receptor 4 (TLR4) in lipopolysacharide (LPS) preconditioning- and ischemia-induced microglial activation. (a–d) LPS causes mild microglial activation, proliferation and migration in retina via TLR4 signaling. (a) LPS is injected (1 μg/μL, i.vt.) in non-ischemic mice and subsequent immunohistochemical analysis of retinal sections using antibodies against Iba-1 and phospho-p38 mitogen-activated protein kinase (pp38 MAPK), a marker for activated microglia, is carried out at 0 h (control), 3 h, 6 h, 12 h, day 1, day 2, and day 3 (Iba-1, red; pp38 MAPK, green). (b) Following combined anti-TLR4 antibodies treatment (Sa15-21 and MTS510; 1 μg/μL, i.vt.) 4 h before LPS injection (1 μg/μL, i.vt.) as well as minocycline (mino) treatment intraperitoneally (90 mg/kg, i.p.) 24 and 0 h before LPS injection, immunostaining (Iba-1 & pp38 MAPK) of retinal sections is performed at day 2 after LPS treatment. The LPS-induced mild activation and migration of retinal microglia is completely abolished by anti-TLR4 antibody and minocycline pre-treatments or in TLR4−/− mice. (c) Quantitative analysis of Iba-1-positive microglia in control, LPS-treated non-ischemic retina. Data are mean ± SEM. (*< 0.05, vs. Control, **< 0.01, vs. Control) from experiments using three mice for each group. (d) Quantitative analysis of Iba-1-positive microglia in control, LPS-treated, anti-TLR4 antibody- and LPS-treated, minocycline- and LPS-treated WT non-ischemic retina. Data are mean ± SEM. (*< 0.05, vs. Control, #< 0.05, vs. LPS) from experiments using three mice for each group. In the experiments using TLR4−/− mice, data were shown using five pieces of section from two mice for control and LPS-treated group. (e) TLR4 is implicated in retinal ischemia-induced microglia activation. Combined anti-TLR4 antibody is injected (1 μg/μL, i.vt.) 4 h before LPS treatment (1 μg/μL, i.vt.), and minocycline is injected (90 mg/kg, i.p.) 24 and 0 h before LPS treatment. Vehicles are treated with phosphate buffered saline (PBS) in similar manners. Following retinal ischemia 48 h after LPS treatment, immunostaining of retinal sections (Iba-1 & pp38 MAPK) is performed at day 2 after the ischemic stress. Experiments are carried out using three mice for each group.

Download figure to PowerPoint

In animal research on neurodegenerative diseases, minocycline, a semisynthetic second-generation tetracycline, was mostly used by intraperitoneal injection and its single dosage per day ranged from 5 mg/kg to 90 mg/kg (Du et al. 2001; Blum et al. 2004). The effective dose of minocycline used in the present retinal study (90 mg/kg, i.p.) was fairly consistent with previous reports showing the effect of this agents in models of retinal disorders through penetration into the central nervous system via the blood–brain barrier or blood–retinal barrier (Shimazawa et al. 2005; Xiao et al. 2012). Therefore, this dosage of minocycline (90 mg/kg, i.p) was acceptable for retinal research in mice. In addition, we decided to perform minocycline administration through the conventional i.p. route to avoid retinal tissue damages by repeated i.vt. injections. Minocycline has emerged as a potent inhibitor of microglial activation and has no direct action on astrocytes or neurons, and its anti-inflammatory property is completely separate from its antimicrobial action (Tikka et al. 2001). Minocycline induces protective action against several types of neurological and retinal diseases as well as ischemic damages by the inhibition of microglial activation and proliferation (Raghavendra et al. 2003; Krady et al. 2005; Zeng et al. 2005; Bye et al. 2007). Since minocycline inhibits the LPS-induced activation of microglia in vitro and in vivo (Fan et al. 2005; Wang et al. 2005; Yang et al. 2007), minocycline was injected (90 mg/kg, i.p.) 24 h and 0 h before LPS treatment to examine whether LPS-induced activation of retinal microglia is blocked by minocycline. We found that LPS-induced proliferation and migration of retinal microglia are completely inhibited at day 2 after LPS injection in mice pre-treated with minocycline (Fig. 7b and d).

Following retinal ischemia, the immunostaining data showed that the activation (pp38 MAPK-positive), proliferation, and migration of Iba-1-positive microglia are significantly and maximally increased throughout the retinal cell layers at day 2 after the ischemic stress (Fig. 7e) and persisted on day 7 (data are not shown). In contrast, LPS treatment 48 h before retinal ischemia significantly blocked the ischemia-induced cytotoxic activation of retinal microglia at day 2 (Fig. 7e), being consistent with the previous finding that cerebral ischemia-induced microglia activation in the infarct regions of brain is markedly decreased by LPS pre-treatment (Rosenzweig et al. 2004). On the other hand, treatment with combined anti-TLR4 antibodies (1 μg/μL, i.vt.) 4 h before LPS injection or minocycline (90 mg/kg, i.p.) 24 and 0 h before LPS injection significantly rescued the LPS-induced blockade of Iba-1-positive microglial activation at day 2 after ischemic stress (Fig. 7e). However, anti-TLR4 antibodies or minocycline treatments at respective time points before PBS (vehicle) injection (48 h before retinal ischemia) failed to block the ischemia-induced microglial activation in retina at day 2 after the ischemic stress (Fig. 7e).

Minocycline blocks LPS-induced prevention of ischemic retina

To determine whether minocycline inhibits the LPS preconditioning-induced prevention of retinal ischemic damages, mice were treated with minocycline (90 mg/kg, i.p.) 24 and 0 h before LPS (1 μg/μL, i.vt.) and PBS (vehicle) administration. Following retinal ischemia 48 h after LPS and PBS treatment, the findings using H&E staining revealed that LPS pre-treatment-induced recovery of the loss of retinal cells in the different layers and the decrease in retinal thickness is completely blocked at day 7 after ischemic stress in minocycline-pre-treated mice (Fig. 8a and b). The ERG analysis showed the complete inhibition of LPS-caused rescue of a- and b-wave amplitudes against retinal ischemic damages at day 7 by minocycline pre-treatment (Fig. 8c and d). Similarly, the experiments using H&E staining and ERG analysis revealed that the ischemia-induced functional damages in retina are not recovered at day 7 by minocycline injection at same time points in PBS-preconditioned ischemic mice (Fig. 8a–d). However, no effect of minocycline treatment in non-ischemic mice was observed at day 7 after the last treatment (Fig. 8a–d).

image

Figure 8. Minocycline abolishes lipopolysacharide (LPS) preconditioning-induced prevention of ischemic damages. (a–d) Minocycline (mino) is injected (90 mg/kg, i.p.) 24 h and 0 h before LPS treatment (1 μg/μL, i.vt.), and. vehicle is prepared by treatment with minocycline at same time points in phosphate buffered saline (PBS)-treated mice in a similar manner. Minocycline is also injected twice (0 h and 24 h) in non-ischemic mice. Following retinal ischemia 48 h after LPS injection, H&E staining of retinal sections (a), measurement of retinal thickness (b), a-wave (c), and b-wave (d) amplitudes of electroretinogram (ERG) analysis are performed at day 7 after the ischemic stress. The H&E staining and ERG analysis reveal no effect of minocycline at day 7 after last treatment in non-ischemic mice. Data are mean ± SEM. (*< 0.05, vs. Veh, #< 0.05, vs.) from experiments using three mice for each group.

Download figure to PowerPoint

On the other hand, the immunohistochemical analysis showed that LPS pre-treatment-induced complete recovery of NeuN-positive neurons in the GCL as well as partial rescue of Chx10-positive bipolar cells in the INL and photoreceptor cells in the ONL is significantly blocked by minocycline injection 24 and 0 h before LPS administration, without any effect on the syntaxin-1-positive amacrine cells, compared to respective controls and vehicles (Table 3).

Table 3. Minocycline blocks LPS preconditioning-induced cell type-specific prevention in ischemic retina
 Ganglion cells (%)Bipolar cells (%)Amacrine cells (%)Photoreceptor cells (%)
  1. Minocycline (Mino) is injected intraperitoneally (90 mg/kg, i.p.) 24 h and 0 h before LPS treatment (1 μg/μL, i.vt.). Following retinal ischemia 48 h after LPS injection, the inhibition of LPS-induced cell type-specific prevention by minocycline was observed at day 7 after the ischemic stress. The quantification was expressed as average percentage of the number of cells (% of control). Data represent as the means ± SEM. (*< 0.05, vs. Control, **< 0.05, vs. Veh) from experiments using three mice for each group.

PBS (veh) + Ischemia27 ± 329 ± 433 ± 335 ± 4
LPS + Ischemia92 ± 5*53 ± 5*29 ± 267 ± 4*
Mino + LPS + Ischemia30 ± 2**32 ± 3**31 ± 438 ± 2**

LPS post-treatment deteriorates ischemia-induced retinal damages

To see whether LPS post-treatment protects the ischemic damages in retina, mice were treated with LPS (1 μg/μL, i.vt.) at 3 h after retinal ischemia. The H&E staining data showed that the cell number in different retinal layers and total retinal thickness are maximally decreased at day 7 after the ischemic stress in LPS post-treated mice, as like as vehicle-treated ischemic mice (Fig. 9a and b). The ERG analysis revealed the maximum decrease in a- and b-wave amplitudes at day 7 after retinal ischemia in LPS post-treated mice (Fig. 9c and d). To confirm whether LPS post-treatment deteriorates the retinal ischemia-induced damages through TLR4 signaling, combined anti-TLR4 antibodies (1 μg/μL) was injected (i.vt.) 30 min before retinal ischemia, followed by LPS injection 3 h after the ischemic stress. Detailed morphological analysis using H&E staining revealed that the thickness of IPL and INL is significantly rescued at day 7 after ischemia by treatment with anti-TLR4 antibody in mice post-treated with both LPS and PBS (vehicle), though ischemia-induced decrease in cell number in the GCL as well as the thickness of OPL (data are not shown) and ONL was not markedly protected at day 7 after ischemia by this antibody treatment in LPS and PBS post-treated mice (Fig. 9a, b and e–h). However, our ERG analysis data showed no significant recovery of ischemia-induced decline in a- and b-wave amplitudes at day 7 when injected with this anti-TLR4 antibody at the same time point in LPS or PBS (vehicle) post-treated ischemic mice (Fig. 9c and d).

image

Figure 9. Lack of preventive effect by post-ischemia treatment of lipopolysacharide (LPS). (a–p) LPS post-treatment deteriorates the ischemic damages through toll-like receptor 4 (TLR4) signaling. Combined anti-TLR4 antibody is injected (1 μg/μL, i.vt.) 30 min before retinal ischemia, followed by LPS injection (1 μg/μL, i.vt.) 3 h after retinal ischemia. (a–h) Following anti-TLR4 antibody pre-treatment and LPS post-treatment, H&E staining of retinal sections (a), measurement of total retinal thickness from ganglion cell layer (GCL) to outer nuclear layer (ONL) (b), a-wave (c), and b-wave (d) amplitudes of ERG study, cell number in the GCL (e), thickness of inner plexiform layer (IPL) (f), INL (g), and ONL (h) in the retina is performed at day 7 after the ischemic stress. Data are mean ± SEM. (*< 0.05, vs. Control, #< 0.05, vs. LPS, < 0.05, vs. Veh) from experiments using three mice for each group. (i–p) Early deterioration of ischemic damages by LPS post-treatment via TLR4 signaling. Following anti-TLR4 antibody pre-treatment and LPS post-treatment, H&E staining of retinal sections (i), measurement of total retinal thickness (j), a-wave and b-wave amplitudes of ERG study (k, l, respectively), cell number in the GCL (m), thickness of IPL (n), inner nuclear layer (INL) (o), and ONL (p) in the retina is performed at day 4 after the ischemic stress. Data are mean ± SEM. (*< 0.05, vs. Control, **< 0.05, vs. Veh, #< 0.05, vs. LPS, < 0.05, vs. Veh) from experiments using three mice for each group.

Download figure to PowerPoint

Since LPS post-treatment showed the maximum functional damages at day 7 after ischemia, we decided to perform experiments at day 4 after retinal ischemia using LPS post-treated mice. The findings using H&E staining revealed that the retinal cell number and thickness is partially decreased at day 4 after ischemic stress in vehicle-treated mice, whereas LPS post-treatment caused the maximum decline in retinal cells and thickness at day 4 (Fig. 9i, j, and m–p). The ERG analysis showed the maximum decrease in a- and b-wave amplitudes at day 4 after retinal ischemia in LPS post-treated mice, compared to partial decline in a- and b-wave amplitudes in vehicle-treated ischemic mice (Fig. 9k and l). Interestingly, detailed morphological study and ERG analysis showed the significant rescue of LPS post-treatment-induced cellular loss in GCL in the retina, decrease in thickness of IPL and INL as well as the decline in a- and b-wave amplitudes at day 4 by treatment with anti-TLR4 antibody 30 min before retinal ischemia (Fig. 9i–p). In addition, anti-TLR4 antibody pre-treatment at the same time point markedly ameliorates the ischemic retina at day 4 in vehicle post-treated mice (Fig. 9i–p).

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Retinal ischemia is associated with multiple sight-threatening disorders and blindness including macular degeneration, retinopathy and glaucoma, in which the event of functional damage and neuronal loss are mediated by a large array of injury-related signaling, such as excitatory neurotransmitter release, Ca2+ overload, formation of reactive oxygen species and free radicals, and secretion of potentially toxic mediators from glial cells, depending on the strength and duration of the ischemic insult (Osborne et al. 2004; Kaur et al. 2008; Hernandez et al. 2009; Neroev et al. 2010; Cervia and Casini 2012). Innate immunity system through TLR4 also contributes to this retinal ischemic damages (Dvoriantchikova et al. 2010), its beneficial roles have been reported in case with cerebral ischemia model, in which preconditioning with lower doses of LPS prevents the ischemia-induced brain damages through TLR4 signaling (Marsh et al. 2009a, b; Vartanian et al. 2011). In this study, we confirmed that the preconditioning of LPS significantly prevented the cell damages in all retinal layers after the ischemia. The best prevention in the evaluation with H&E histology and ERG was obtained when the preconditioning was given 48 h before the ischemic stress. The local intravitreous administration of LPS at as low as 0.1 and 1 μg significantly rescued the cellular loss and decrease in total thickness of ischemic retina at day 7, while there was a slow decline with 5 μg. As the LPS-induced preventive effect on GCL was maximal at 1 μg LPS, the decline with 5 μg seems to be attributed to its toxic effect, since it is reported that the activation of TLR4 by higher dose of LPS induces the robust production of inflammatory molecules through different cascades including myeloid differentiation factor (MyD)88-nuclear factor κB (NF-κB) pathway and MyD88-deficient cells failed to produce inflammatory cytokines in response to LPS (Palsson-McDermott and O'Neill 2004; Broad et al. 2007; Kawai and Akira 2007; Piao et al. 2009; Maitra et al. 2012). Furthermore, MyD88-dependent pathway is involved in an early response to LPS, followed by late activation of TIR-domain-containing adapter-inducing interferon-β (TRIF)-dependent cascade of TLR4 (Palsson-McDermott and O'Neill 2004). In contrast, lower dose of LPS is mainly associated with the activation of MyD88-independent interferon regulatory factor 3 (IRF3)-TRIF pathway of TLR4 signaling, leading to induction of anti-inflammatory molecules and type I interferons (IFNs), such as IFN-β and concomitant suppression of NF-κB-mediated signaling, though several reports suggested the initial induction of low-grade pro-inflammatory cytokines through MyD88-dependent pathway and other unknown mechanisms upon this LPS preconditioning (Marsh et al. 2009b; Vartanian and Stenzel-Poore 2010; Stevens et al. 2011; Vartanian et al. 2011). This suppression of MyD88-NF-κB pathway is caused by several negative regulators, nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha (IκBα), including interleukin-1 receptor-associated kinase (IRAK)-M and tripartite motif family (TRIM)-30, which are produced by low-dose LPS-induced TLR4 signaling (Vartanian and Stenzel-Poore 2010; Stevens et al. 2011; Wang et al. 2011). Previous reports suggested the up-regulation downstream anti-inflammatory genes of IRF3-TRIF pathway and down-regulation of downstream inflammatory genes IL-6, IL-1β, COX-2, and TNFα of MyD88-NF-κB pathway in LPS-preconditioned cerebral ischemic mice (Stevens et al. 2011; Vartanian et al. 2011; Wang et al. 2011). We found that the downstream neuroprotective genes of IRF3-TRIF pathway IL1RN and IFIT1 as well as the suppressor genes of MyD88-NF-κB pathway SOCS 1 and SOCS 3 are increased by LPS treatment, and this up-regulation was blocked by pre-treatment with anti-TLR4 antibody. On the other hand, the ischemia-induced up-regulation of downstream injury genes TNFα, IL-1β, IL-6, COX-2, and MCP-1 of MyD88-NF-κB pathway was markedly inhibited by LPS preconditioning, being consistent with the previous study against brain ischemia (Vartanian et al. 2011). Thus, low-dose LPS preconditioning causes the reprogramming of transcriptional response to TLR4 by enhancing the TRIF signaling and suppressing the NF-κB-induced production of cytotoxic molecules. However, detailed studies of anti-TLR4 antibody treatment or in TLR4−/− mice of LPS preconditioning-induced alternation of protective and injury genes upon ischemia, and exhaustive gene expression profile by use of microarray technique should be the next subjects.

Photosignal processing in the retina is mediated by sequential activation of three cell layers, ONL including rod and corn cells, INL including bipolar, amacrine, and horizontal cells, and GCL including ganglion cells, and the damage of any cell layers causes crucial problems in the signal processing. Previous investigators described that ischemia-induced cytotoxic signaling causes the damages of all these retinal cells, while ganglion cells are more vulnerable to ischemic stress than others (Osborne et al. 2004; Uckermann et al. 2005; Langmann 2007; Jehle et al. 2008; Kaur et al. 2008; Neroev et al. 2010; Cervia and Casini 2012). In this study, ganglion cells were seriously damaged by the ischemia, while this damage was perfectly prevented by LPS preconditioning, being in a good contrast with the case of amacrine cells, whose damages were not at all prevented by LPS. As reported in the case with cerebral ischemia (Marsh et al. 2009a; Stevens et al. 2011; Vartanian et al. 2011; Wang et al. 2011), LPS preconditioning effects in the retinal model were abolished in TLR4−/− mice or by the intravitreous pre-treatment with anti-TLR4 antibody. It should be noted that the thickness of IPL and INL was partially, but significantly prevented when injected this antibody alone 30 min before ischemia, whereas the basal retinal ischemia-induced damages was not affected when treated with anti-TLR4 antibody alone 2 days before ischemia in vehicle-preconditioned WT ischemic mice. This difference may be explained by the possibility that anti-TLR4 antibody is degraded in vivo during the period of two days. Furthermore, TLR4−/− mice also showed the significant retinal protection underlying the recovery of IPL and INL against retinal ischemic damages, being consistent with cerebral ischemia, in which partial prevention is observed in TLR4−/− mice (Cao et al. 2007; Hyakkoku et al. 2010). In addition, mice that have been preconditioned showed a marked reduction in cytotoxic mediators including TNFα, iNOS, and COX-2 in the brain of TLR4−/− mice (Pradillo et al. 2009). Thus, no alteration of retinal ischemia-induced damages by long-term pre-treatment of anti-TLR4 antibody alone or partial protection by treatment of anti-TLR4 antibody immediately before ischemia and in TLR4−/− mice may be attributed to higher contribution of TLR4-independent toxicity.

Several reports suggested the expression of TLR4 on the cell surface in murine microglia in the brain, while astrocytes express it intracellularly (Bsibsi et al. 2002; Olson and Miller 2004; Jack et al. 2005; Marsh et al. 2009b). In this study, the double fluorescence immunohistochemical data showed that TLR4 is expressed in retinal microglia in the GCL and IPL, whereas 1 μg LPS treatment caused a migration of TLR4-positive microglia throughout the different retinal cell layers and increased its number at day 2 after injection. It should be noted that there was no change in morphology of LPS-treated microglia, while a marked hypertrophy in nucleus and processes of ischemia-treated microglia. This difference may suggest that retinal ischemia causes TLR4-independent toxicity to a high degree as well as a slight degree of TLR4-dependent toxicity, which was evidenced by the experiment using anti-TLR4 antibody.

On the other hand, pre-treatment with minocycline, a microglial inhibitor caused the blockade of LPS-induced mild activation and migration of retinal microglia via TLR4. We found that minocycline pre-treatment causes the complete blockade of LPS preconditioning-induced prevention of retinal ischemia-induced functional damages and survival of ganglion, bipolar, and photoreceptor cells. However, minocycline alone failed to protect the ischemic damages in retina, when it was treated 48 h before ischemia. This may be explained by the decrease in efficacy during long-term incubation, since the average half-life of minocycline is reported as 15 h (Xiao et al. 2012). Thus, it is suggested that LPS preconditioning-induced protection against ischemia is mediated by microglia. Microglia is highly activated by its robust change in morphology upon ischemia in brain and in retina, and subsequently causes tissue damages by releasing a multitude of noxious mediators, but minocycline reverses these effects (Mabuchi et al. 2000; Tikka et al. 2001; Lai and Todd 2006; Langmann 2007; Plane et al. 2010). In contrast, microglia have neuroprotective actions by producing anti-inflammatory factors including neurotrophins (Streit 2002; Denes et al. 2007; Lalancette-Hebert et al. 2007; Shpargel et al. 2008; Lambertsen et al. 2009; Ransohoff and Perry 2009; Chen et al. 2012). Retinal microglia is normally localized in the GCL and IPL in retina and is superficially exposed by intravitreal LPS administration. It is noted that LPS preconditioning did not affect the astrocyte activation with or without ischemia. Thus, the mild activation of microglia following LPS preconditioning may be responsible for retinal protection against ischemia.

In contrast, we found that post-treatment with 1 μg LPS, as applied in preconditioning experiments deteriorates the ischemia-induced cellular and functional damages in retina at day 4, followed by maximum damages at day 7 after the ischemic stress. LPS post-treatment-induced retinal damage was significantly reversed by treatment with anti-TLR4 antibody, suggesting that LPS post-ischemia treatment induces cytotoxic effects through TLR4 signaling. This LPS action is not attributed to the TLR4 reprogramming, because brief ischemia- or LPS preconditioning-induced TLR4 reprogramming provides neuroprotection against lethal ischemia (Marsh et al. 2009a; Pradillo et al. 2009; Vartanian et al. 2011). Taken together, this study indicates the LPS preconditioning caused functional and cell type-specific protection of retina against ischemic damages, whereas its post-ischemia treatment failed to protect the ischemic damages. This LPS preconditioning-induced retinal protection is mediated by TLR4 signaling in microglia.

In conclusion, LPS treatment 48 h before retinal ischemia prevented the functional damages of ischemic retina, whereas its post-ischemia treatment failed to protect the ischemic damages. LPS preconditioning completely rescued the ischemia-induced ganglion cell loss, along with partial recovery of bipolar and photoreceptor cell damages. This LPS-induced retinal protection against ischemia was mediated through microglial TLR4 signaling. Therefore, detailed mechanisms underlying LPS preconditioning caused protection could give us novel information for the discovery of new drugs to prevent the ischemic disorders in the central nervous system.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Parts of this study were supported by Grants-in-Aid for Scientific Research (to H.U., B: 13470490 and B: 15390028) on Priority Areas-Research on Pathomechanisms of Brain Disorders (to H.U., 17025031) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT). We have no conflict interest to report.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Akashi S., Saitoh S., Wakabayashi Y. et al. (2003) Lipopolysaccharide interaction with cell surface Toll-like receptor 4-MD-2: higher affinity than that with MD-2 or CD14. J. Exp. Med. 198, 10351042.
  • Akashi-Takamura S., Furuta T., Takahashi K., Tanimura N., Kusumoto Y., Kobayashi T., Saitoh S., Adachi Y., Doi T. and Miyake K. (2006) Agonistic antibody to TLR4/MD-2 protects mice from acute lethal hepatitis induced by TNF-alpha. J. Immunol. 176, 42444251.
  • Arumugam T. V., Chan S. L., Jo D. G. et al. (2006) Gamma secretase-mediated Notch signaling worsens brain damage and functional outcome in ischemic stroke. Nat. Med. 12, 621623.
  • Bastide M., Gele P., Petrault O., Pu Q., Caliz A., Robin E., Deplanque D., Duriez P. and Bordet R. (2003) Delayed cerebrovascular protective effect of lipopolysaccharide in parallel to brain ischemic tolerance. J. Cereb. Blood Flow Metab. 2, 399405.
  • Bernstein S. L., Guo Y., Kelman S. E., Flower R. W. and Johnson M. A. (2003) Functional and cellular responses in a novel rodent model of anterior ischemic optic neuropathy. Invest. Ophthalmol. Vis. Sci. 44, 41534162.
  • Blum D., Chtarto A., Tenenbaum L., Brotchi J. and Levivier M. (2004) Clinical potential of minocycline for neurodegenerative disorders. Neurobiol. Dis. 17, 359366.
  • Broad A., Kirby J. A. and Jones D. E. (2007) Toll-like receptor interactions: tolerance of MyD88-dependent cytokines but enhancement of MyD88-independent interferon-beta production. Immunology 120, 103111.
  • Bsibsi M., Ravid R., Gveric D. and van Noort J. M. (2002) Broad expression of Toll-like receptors in the human central nervous system. J. Neuropathol. Exp. Neurol. 61, 10131021.
  • Buckingham B. P., Inman D. M., Lambert W., Oglesby E., Calkins D. J., Steele M. R., Vetter M. L., Marsh-Armstrong N. and Horner P. J. (2008) Progressive ganglion cell degeneration precedes neuronal loss in a mouse model of glaucoma. J. Neurosci. 28, 27352744.
  • Bye N., Habgood M. D., Callaway J. K., Malakooti N., Potter A., Kossmann T. and Morganti-Kossmann M. C. (2007) Transient neuroprotection by minocycline following traumatic brain injury is associated with attenuated microglial activation but no changes in cell apoptosis or neutrophil infiltration. Exp. Neurol. 204, 220233.
  • Cao C. X., Yang Q. W., Lv F. L., Cui J., Fu H. B. and Wang J. Z. (2007) Reduced cerebral ischemia-reperfusion injury in Toll-like receptor 4 deficient mice. Biochem. Biophys. Res. Commun. 353, 509514.
  • Caso J. R., Pradillo J. M., Hurtado O., Lorenzo P., Moro M. A. and Lizasoain I. (2007) Toll-like receptor 4 is involved in brain damage and inflammation after experimental stroke. Circulation 115, 15991608.
  • Cervia D. and Casini G. (2012) Recent advances in cellular and molecular aspects of mammalian retinal ischemia. World J. Pharmacol. 1, 3043.
  • Chen Z., Jalabi W., Shpargel K. B., Farabaugh K. T., Dutta R., Yin X., Kidd G. J., Bergmann C. C., Stohlman S. A. and Trapp B. D. (2012) Lipopolysaccharide-induced microglial activation and neuroprotection against experimental brain injury is independent of hematogenous TLR4. J. Neurosci. 32, 1170611715.
  • Danton G. H. and Dietrich W. D. (2003) Inflammatory mechanisms after ischemia and stroke. J. Neuropathol. Exp. Neurol. 62, 127136.
  • Denes A., Vidyasagar R., Feng J., Narvainen J., McColl B. W., Kauppinen R. A. and Allan S. M. (2007) Proliferating resident microglia after focal cerebral ischaemia in mice. J. Cereb. Blood Flow Metab. 27, 19411953.
  • Du Y., Ma Z., Lin S. et al. (2001) Minocycline prevents nigrostriatal dopaminergic neurodegeneration in the MPTP model of Parkinson's disease. Proc. Natl Acad. Sci. USA 98, 1466914674.
  • Dvoriantchikova G., Barakat D. J., Hernandez E., Shestopalov V. I. and Ivanov D. (2010) Toll-like receptor 4 contributes to retinal ischemia/reperfusion Injury. Mol. Vis. 16, 19071912.
  • Fan L. W., Pang Y., Lin S., Rhodes P. G. and Cai Z. (2005) Minocycline attenuates lipopolysaccharide-induced white matter injury in the neonatal rat brain. Neuroscience 133, 159168.
  • Fujita R., Ueda M., Fujiwara K. and Ueda H. (2009) Prothymosin-alpha plays a defensive role in retinal ischemia through necrosis and apoptosis inhibition. Cell Death Differ. 16, 349358.
  • Guha M. and Mackman N. (2001) LPS induction of gene expression in human monocytes. Cell. Signal. 13, 8594.
  • Halder S. K., Matsunaga H. and Ueda H. (2012) Neuron-specific non-classical release of prothymosin alpha: a novel neuroprotective damage-associated molecular patterns. J. Neurochem. 123, 262275.
  • Halder S. K., Matsunaga H., Yamaguchi H. and Ueda H. (2013) Novel neuroprotective action of prothymosin alpha-derived peptide against retinal and brain ischemic damages. J. Neurochem. doi:10.1111/jnc.12132 (in press).
  • Hernandez M., Rodriguez F. D., Sharma S. C. and Vecino E. (2009) Immunohistochemical changes in rat retinas at various time periods of elevated intraocular pressure. Mol. Vis. 15, 26962709.
  • Hoshino K., Takeuchi O., Kawai T., Sanjo H., Ogawa T., Takeda Y., Takeda K. and Akira S. (1999) Cutting edge: Toll-like receptor 4 (TLR4)-deficient mice are hyporesponsive to lipopolysaccharide: evidence for TLR4 as the Lps gene product. J. Immunol. 162, 37493752.
  • Hua F., Ma J., Ha T. et al. (2007) Activation of Toll-like receptor 4 signaling contributes to hippocampal neuronal death following global cerebral ischemia/reperfusion. J. Neuroimmunol. 190, 101111.
  • Hyakkoku K, Hamanaka J., Tsuruma K., Shimazawa M., Tanaka H., Uematsu S., Akira S., Inagaki N., Nagai H. and Hara H. (2010) Toll-like receptor 4 (TLR4), but not TLR3 or TLR9, knock-out mice have neuroprotective effects against focal cerebral ischemia. Neuroscience 171, 258267.
  • Iadecola C. and Anrather J. (2011) The immunology of stroke: from mechanisms to translation. Nat. Med. 17, 796808.
  • Ibrahim A. S., El-Remessy A. B., Matragoon S., Zhang W., Patel Y., Khan S., Al-Gayyar M. M., El-Shishtawy M. M. and Liou G. I. (2011) Retinal microglial activation and inflammation induced by amadori-glycated albumin in a rat model of diabetes. Diabetes 60, 11221133.
  • Jack C. S., Arbour N., Manusow J., Montgrain V., Blain M., McCrea E., Shapiro A. and Antel J. P. (2005) TLR signaling tailors innate immune responses in human microglia and astrocytes. J. Immunol. 175, 43204330.
  • Jehle T., Wingert K., Dimitriu C., Meschede W., Lasseck J., Bach M. and Lagrèze W. A. (2008) Quantification of ischemic damage in the rat retina: a comparative study using evoked potentials, electroretinography, and histology. Invest. Ophthalmol. Vis. Sci. 49, 10561064.
  • Jeong H. K., Jou I. and Joe E. H. (2010) Systemic LPS administration induces brain inflammation but not dopaminergic neuronal death in the substantia nigra. Exp. Mol. Med. 42, 823832.
  • Jin S. X., Zhuang Z. Y., Woolf C. J. and Ji R. R. (2003) p38 mitogen-activated protein kinase is activated after a spinal nerve ligation in spinal cord microglia and dorsal root ganglion neurons and contributes to the generation of neuropathic pain. J. Neurosci. 23, 40174022.
  • Kaur C., Foulds W. S. and Ling E. A. (2008) Hypoxia-ischemia and retinal ganglion cell damage. Clin. Ophthalmol. 2, 879889.
  • Kawai T. and Akira S. (2007) TLR signaling. Semin. Immunol. 19, 2432.
  • Ko M. K., Saraswathy S., Parikh J. G. and Rao N. A. (2011) The role of TLR4 activation in photoreceptor mitochondrial oxidative stress. Invest. Ophthalmol. Vis. Sci. 52, 58245835.
  • Koistinaho M. and Koistinaho J. (2002) Role of p38 and p44/42 mitogen-activated protein kinases in microglia. Glia 40, 175183.
  • Krady J. K., Basu A., Allen C. M., Xu Y., LaNoue K. F., Gardner T. W. and Levison S. W. (2005) Minocycline reduces proinflammatory cytokine expression, microglial activation, and caspase-3 activation in a rodent model of diabetic retinopathy. Diabetes 54, 15591565.
  • Lai A. Y. and Todd K. G. (2006) Microglia in cerebral ischemia: molecular actions and interactions. Can. J. Physiol. Pharmacol. 84, 4959.
  • Lakhan S. E., Kirchgessner A. and Hofer M. (2009) Inflammatory mechanisms in ischemic stroke: therapeutic approaches. J. Transl. Med. 7, 97.
  • Lalancette-Hebert M., Gowing G., Simard A., Weng Y. C. and Kriz J. (2007) Selective ablation of proliferating microglial cells exacerbates ischemic injury in the brain. J. Neurosci. 27, 25962605.
  • Lambertsen K. L., Clausen B. H., Babcock A. A. et al. (2009) Microglia protect neurons against ischemia by synthesis of tumor necrosis factor. J. Neurosci. 29, 13191330.
  • Langmann T. (2007) Microglia activation in retinal degeneration. J. Leukoc. Biol. 81, 13451351.
  • Lehnardt S., Schott E., Trimbuch T., Laubisch D., Krueger C., Wulczyn G., Nitsch R. and Weber J. R. (2008) A vicious cycle involving release of heat shock protein 60 from injured cells and activation of toll-like receptor 4 mediates neurodegeneration in the CNS. J. Neurosci. 28, 23202331.
  • Liang H., Brignole-Baudouin F., Labbé A., Pauly A., Warnet J. M. and Baudouin C. (2007) LPS-stimulated inflammation and apoptosis in corneal injury models. Mol. Vis. 13, 11691180.
  • Lipton P. (1999) Ischemic cell death in brain neurons. Physiol. Rev. 79, 14311568.
  • Litvak V., Ramsey S. A., Rust A. G., Zak D. E., Kennedy K. A., Lampano A. E., Nykter M., Shmulevich I. and Aderem A. (2009) Function of C/EBPdelta in a regulatory circuit that discriminates between transient and persistent TLR4-induced signals. Nat. Immunol. 10, 437443.
  • Mabuchi T., Kitagawa K., Ohtsuki T., Kuwabara K., Yagita Y., Yanagihara T., Hori M. and Matsumoto M. (2000) Contribution of microglia/macrophages to expansion of infarction and response of oligodendrocytes after focal cerebral ischemia in rats. Stroke 31, 17351743.
  • Maitra U., Deng H., Glaros T., Baker B., Capelluto D. G., Li Z. and Li L. (2012) Molecular mechanisms responsible for the selective and low-grade induction of proinflammatory mediators in murine macrophages by lipopolysaccharide. J. Immunol. 189, 10141023.
  • Marsh B., Stevens S. L., Packard A., Gopalan B., Hunter B., Leung P. Y., Harrington C. A. and Stenzel-Poore M. P. (2009a) Systemic lipopolysaccharide protects the brain from ischemic injury by reprogramming the response of the brain to stroke: a critical role for IRF3. J. Neurosci. 29, 98399849.
  • Marsh B. J., Williams-Karnesky R. L. and Stenzel-Poore M. P. (2009b) Toll-like receptor signaling in endogenous neuroprotection and stroke. Neuroscience 158, 10071020.
  • Masland R. H. (2012) The neuronal organization of the retina. Neuron 76, 266280.
  • Neroev V. V., Zueva M. V. and Kalamkarov G. R. (2010) Molecular mechanisms of retinal ischemia. Vestn. oftalmol. 126, 5964.
  • Neufeld A. H., Si Kawai, Das S., Vora S., Gachie E., Connor J. R. and Manning P. T. (2002) Loss of retinal ganglion cells following retinal ischemia: the role of inducible nitric oxide synthase. Exp. Eye Res. 75, 521528.
  • Obrenovitch T. P. (2008) Molecular physiology of preconditioning-induced brain tolerance to ischemia. Physiol. Rev. 88, 211247.
  • Olson J. K. and Miller S. D. (2004) Microglia initiate central nervous system innate and adaptive immune responses through multiple TLRs. J. Immunol. 173, 39163924.
  • Osborne N. N., Casson R. J., Wood J. P., Chidlow G., Graham M. and Melena J. (2004) Retinal ischemia: mechanisms of damage and potential therapeutic strategies. Prog. Retin. Eye Res. 23, 91147.
  • Palsson-McDermott E. M. and O'Neill L. A. (2004) Signal transduction by the lipopolysaccharide receptor, Toll-like receptor-4. Immunology 113, 153162.
  • Pang Y., Fan L. W., Zheng B., Campbell L. R., Cai Z. and Rhodes P. G. (2012) Dexamethasone and betamethasone protect against lipopolysaccharide-induced brain damage in neonatal rats. Pediatr. Res. 71, 552558.
  • Perlman I. (2009) Testing retinal toxicity of drugs in animal models using electrophysiological and morphological techniques. Doc. Ophthalmol. 118, 328.
  • Piao W., Song C., Chen H., Diaz M. A., Wahl L. M., Fitzgerald K. A., Li L. and Medvedev A. E. (2009) Endotoxin tolerance dysregulates MyD88- and Toll/IL-1R domain-containing adapter inducing IFN-beta-dependent pathways and increases expression of negative regulators of TLR signaling. J. Leukoc. Biol. 86, 863875.
  • Plane J. M., Shen Y., Pleasure D. E. and Deng W. (2010) Prospects for minocycline neuroprotection. Arch. Neurol. 67, 14421448.
  • Pradillo J. M., Fernandez-Lopez D., Garcia-Yebenes I., Sobrado M., Hurtado O., Moro M. A. and Lizasoain I. (2009) Toll-like receptor 4 is involved in neuroprotection afforded by ischemic preconditioning. J. Neurochem. 109, 287294.
  • Qin L., Wu X., Block M. L., Liu Y., Breese G. R., Hong J. S., Knapp D. J. and Crews F. T. (2007) Systemic LPS causes chronic neuroinflammation and progressive neurodegeneration. Glia 55, 453462.
  • Raghavendra V., Tanga F. and DeLeo J. A. (2003) Inhibition of microglial activation attenuates the development but not existing hypersensitivity in a rat model of neuropathy. J. Pharmacol. Exp. Ther. 306, 624630.
  • Ransohoff R. M. and Perry V. H. (2009) Microglial physiology: unique stimuli, specialized responses. Annu. Rev. Immunol. 27, 119145.
  • Rhee K. D., Ruiz A., Duncan J. L., Hauswirth W. W., Lavail M. M., Bok D. and Yang X. J. (2007) Molecular and cellular alterations induced by sustained expression of ciliary neurotrophic factor in a mouse model of retinitis pigmentosa. Invest. Ophthalmol. Vis. Sci. 48, 13891400.
  • Rosenzweig H. L., Lessov N. S., Henshall D. C., Minami M., Simon R. P. and Stenzel-Poore M. P. (2004) Endotoxin preconditioning prevents the cellular inflammatory response during ischemic neuroprotection in mice. Stroke 35, 25762581.
  • Roth S., Shaikh A. R., Hennelly M. M., Li Q., Bindokas V. and Graham C. E. (2003) Mitogen-activated protein kinases and retinal ischemia. Invest. Ophthalmol. Vis. Sci. 44, 53835395.
  • Sheng J. G., Bora S. H., Xu G., Borchelt D. R., Price D. L. and Koliatsos V. E. (2003) Lipopolysaccharide-induced-neuroinflammation increases intracellular accumulation of amyloid precursor protein and amyloid beta peptide in APPswe transgenic mice. Neurobiol. Dis. 14, 133145.
  • Sherry D. M., Mitchell R., Standifer K. M. and du Plessis B. (2006) Distribution of plasma membrane-associated syntaxins 1 through 4 indicates distinct trafficking functions in the synaptic layers of the mouse retina. BMC Neurosci. 7, 54.
  • Shimazawa M., Yamashima T., Agarwal N. and Hara H. (2005) Neuroprotective effects of minocycline against in vitro and in vivo retinal ganglion cell damage. Brain Res. 1053, 185194.
  • Shpargel K. B., Jalabi W., Jin Y., Dadabayev A., Penn M. S. and Trapp B. D. (2008) Preconditioning paradigms and pathways in the brain. Cleve. Clin. J. Med. 75, 7782.
  • Stevens S. L., Leung P. Y., Vartanian K. B., Gopalan B., Yang T., Simon R. P. and Stenzel-Poore M. P. (2011) Multiple preconditioning paradigms converge on interferon regulatory factor-dependent signaling to promote tolerance to ischemic brain injury. J. Neurosci. 31, 84568463.
  • Streit W. J. (2002) Microglia as neuroprotective, immunocompetent cells of the CNS. Glia 40, 133139.
  • Tasaki K., Ruetzler C. A., Ohtsuki T., Martin D., Nawashiro H. and Hallenbeck J. M. (1997) Lipopolysaccharide pre-treatment induces resistance against subsequent focal cerebral ischemic damage in spontaneously hypertensive rats. Brain Res. 748, 267270.
  • Tikka T., Fiebich B. L., Goldsteins G., Keinanen R. and Koistinaho J. (2001) Minocycline, a tetracycline derivative, is neuroprotective against excitotoxicity by inhibiting activation and proliferation of microglia. J. Neurosci. 21, 25802588.
  • Uckermann O., Uhlmann S., Pannicke T. et al. (2005) Ischemia-reperfusion causes exudative detachment of the rabbit retina. Invest. Ophthalmol. Vis. Sci. 46, 25922600.
  • Ueda H. and Fujita R. (2004) Cell death mode switch from necrosis to apoptosis in brain. Biol. Pharm. Bull. 27, 950955.
  • Ueda H., Matsunaga H., Uchida H. and Ueda M. (2010) Prothymosin alpha as robustness molecule against ischemic stress to brain and retina. Ann. N. Y. Acad. Sci. 1194, 2026.
  • Ueda H., Matsunaga H. and Halder S. K. (2012) Prothymosin α–a novel endogenous neuroprotective polypeptide against ischemic damages, in Neuropeptides in Neuroprotection and Neuroregeneration (Nyberg F. J., ed.), pp. 128143. CRC Press Taylor & Francis Group, Abingdon.
  • Vartanian K. B. and Stenzel-Poore M. (2010) Toll-like receptor tolerance as a mechanism for neuroprotection. Transl. Stroke Res. 1, 252260.
  • Vartanian K. B., Stevens S. L., Marsh B. J., Williams-Karnesky R., Lessov N. S. and Stenzel-Poore M. P. (2011) LPS preconditioning redirects TLR signaling following stroke: TRIF-IRF3 plays a seminal role in mediating tolerance to ischemic injury. J. Neuroinflammation 8, 140.
  • Wang A. L., Yu A. C., Lau L. T., Lee C., le Wu M., Zhu X. and Tso M. O. (2005) Minocycline inhibits LPS-induced retinal microglia activation. Neurochem. Int. 47, 152158.
  • Wang Y. C., Lin S. and Yang Q. W. (2011) Toll-like receptors in cerebral ischemic inflammatory injury. J. Neuroinflammation 8, 134.
  • Xiao O., Xie Z. L., Lin B. W., Yin X. F., Pi R. B. and Zhou S. Y. (2012) Minocycline inhibits alkali burn-induced corneal neovascularization in mice. PLoS ONE 7, e41858.
  • Yang L., Zhu X. and Tso M. (2007) Minocycline and sulforaphane inhibited lipopolysaccharide-mediated retinal microglial activation. Mol. Vis. 13, 10831093.
  • Zeng H. Y., Zhu X. A., Zhang C., Yang L. P., Wu L. M. and Tso M. O. (2005) Identification of sequential events and factors associated with microglial activation, migration, and cytotoxicity in retinal degeneration in rd mice. Invest. Ophthalmol. Vis. Sci. 46, 29922999.