Background
Age-related macular degeneration (AMD) is the leading cause of irreversible blindness among those aged over 65 in industrialised nations [
1]. It is a progressive retinopathy that affects the photoreceptor population, retinal pigment epithelium (RPE), and Bruch’s membrane, and the choroid at various stages of the disease. The advanced stages of AMD are classified into two forms: exudative ‘wet’ AMD, induced by sudden choroidal neovascularisation (CNV), and geographic atrophy (GA) marked by a progressive expanding lesion of the RPE, photoreceptors, and choriocapillaris. While AMD has a complex aetiology involving multiple lifestyle and genetic risk factors, the significance of immunological processes has emerged slowly [
2‐
7]. The relatively recent confirmation in genome-wide association studies (GWAS) of the importance of the complement system and macrophages has placed inflammation among the forefront of factors influencing both the onset and progression of AMD [
8,
9].
The accumulation and activation of macrophages within the retinal tissues is a prominent feature of most retinopathies and is well documented in all forms of AMD. Microglial cells are the resident macrophages of the retina that are derived from embryonic yolk sac progenitors during development [
10,
11], while non-resident bone marrow-derived macrophages may be recruited into the retina from the vasculature in pathology [
12,
13]. In AMD lesions, macrophages form large aggregations in the photoreceptor layer, subretinal space, and RPE [
4,
6,
7,
14‐
17], which are typically free of these leukocytes in healthy individuals [
18]. Despite having some beneficial properties [
19], excessive accumulation of these cells is implicated in degeneration in animal models of AMD, retinal detachment [
20], diabetic retinopathy [
21‐
23], and glaucoma [
24,
25].
Chemokines are key regulators of macrophage activation and accumulation in AMD [
26,
27], as well as degenerations associated with retinal detachment [
20] and diabetic retinopathy [
23]. The chemokine family comprises more than 50 molecules and associated receptors that activate and recruit leukocytes to sites of damage. Transcriptome-wide analysis of AMD donor retinas reports that the leukocyte chemoattractants Ccl2, Cxcl1, Cxcl10, and Cxcl11 are up-regulated in all forms of the disease [
28]. Moreover, intraocular Ccl2 protein levels are increased in patients with CNV or GA [
16,
29,
30], and infiltrating microglia/monocytes in retinas with GA express the cognate Ccl2 receptor, Ccr2 [
16]. Chemokine signalling is implicated in exacerbating the accumulation of macrophages and photoreceptor death in a range of animal models that demonstrate some features in common with AMD, including laser-induced CNV [
31], photo-oxidative stress [
16,
32,
33], Abca4
−/−Rdh8
−/− mice [
34], and carboxyethylpyrrole (CEP)-immunised mice [
35].
While these findings underscore the importance of chemokines in shaping inflammation in the retina, therapeutic blockade of individual chemokine ligands/receptors may have limited efficacy due to chemokine receptor redundancy, and the tendency toward compensatory increases in expression of other chemokine family members [
34,
36]. Broad-spectrum chemokine inhibitors (BSCIs) are a class of chemokine modulators with the ability to inhibit signal transduction of a broad swathe of chemokine receptors [
37]. Among these, NR58-3.14.3 is a robust cyclic peptide that specifically inhibits chemokine-mediated migration of macrophages by blocking the signalling of Ccl- and Cxcl- pathways [
37]. NR58-3.14.3 has been shown to effectively inhibit macrophage accumulation and ameliorate pathology in lung ischemia [
38], obliterative bronchiolitis [
39], skin inflammation [
37], and atherosclerotic plaques [
40].
To our knowledge, there are no studies which have explored the potential of BSCIs in retinal degeneration models. Here, we investigate the effect of NR58-3.14.3 in modulation of the accumulation of macrophages in the outer retina, using a light-induced model of oxidative stress and retinal inflammation. While acute exposure to light damage does not induce some of the classical features of AMD (including drusen and CNV), it induces focal and progressive atrophy of the outer retina and models very effectively the accumulation of subretinal macrophages and up-regulation of chemokines Ccl2, Cxcl1, Cxcl10, and Cxcl11 [
41]. Furthermore, we have shown that the degenerative changes associated with light damage closely mimic naturally occurring degenerative changes in the ageing rat retina [
42].
In the current study, we show that intravitreal delivery of NR58-3.14.3 is safely tolerated in experimental animals, is long-lasting in the retinal environment, suppresses macrophage accumulation in the outer retina, and reduces retinal pathology following light damage. Additionally, we demonstrate that NR58-3.14.3 reduces the expression of pro-inflammatory factors Il6, Ccl3, and Ccl4 by macrophages, which are implicated in promoting their deleterious activity in retinal dystrophy. These data provide proof-of-principle support of the value of BSCIs in thwarting deleterious macrophage activity in retinal degeneration.
Methods
Animal handling and rearing
All experiments were in accordance with the ARVO Statement for Use of Animals in Ophthalmic and Vision Research and with approval from the ANU Animal Experimentation Ethics Committee (Ethics ID: A2014/56). Adult Sprague-Dawley (SD) rats were utilised in all experiments, which were born and reared under dim-light conditions (5 lux) and aged between 90 and 120 post-natal days at the time of use.
Preparation of NR58-3.14.3 and intravitreal injections
Animals were anaesthetised with an intraperitoneal injection of ketamine (100 mg/kg; Troy Laboratories, NSW, Australia) and Xylazil (12 mg/kg; Troy Laboratories, NSW, Australia). Intravitreal injections of consisting of 3 μL were conducted as described previously [
32]; this route was preferred over systemic administration as it minimises the potential for off-target effects which may cofound the data. The 3-uL injection volume was chosen as it offers good reproducibility, minimal disturbance in vitreal volume, and negligible cataract formation [
43]. The broad-spectrum chemokine inhibitor (BSCI) NR58-3.14.3 (Auspep, VIC, Australia) was reconstituted in endotoxin-free 0.1 M PBS to a final concentration of 66.7 μg/μl. In the treatment group, 3 μl of NR58-3.14.3 in PBS was intravitreally injected into both eyes of each animal, so that 200 μg of NR58-3.14.3 was delivered to each eye. The control group received an injection of 3 μl PBS vehicle in each eye. Following intravitreal injections, the animals were monitored closely throughout the time course of experiments. As expected, no incidences of cataract formation or other abnormalities were observed.
Light damage
Immediately after administering intravitreal injections, animals were exposed to 1000 lux light for a period of 24 h, as described previously [
44]. During light damage, animals were placed into transparent Perspex open-top cages under a light source (COLDF2, 2 × 36 W, IHF, Thorn Lighting, Australia) with access to food and water ad libitum. After light damage, animals were either euthanized for tissue collection (0 day time point) or were placed into recovery under dim-light conditions (5 lux) for 7 days.
Retinal localisation of injected NR58-3.14.3 via fluorescein
The fluorescein-labelled NR58-3.14.3 was prepared using an NHS-Fluorescein Antibody Labelling Kit (Pierce, Thermo Fisher Scientific, MA, USA), applied according to the manufacturer’s protocol. In brief, the NR58-3.14.3 was reconstituted in endotoxin-free PBS to a concentration of 1 mg/ml. The NR58-3.14.3 was then incubated with the NHS-Fluorescein for 1 h at room temperature, before being purified through a column to remove excess dye from the final product. The fluorescein-NR58-3.14.3 was stored at 4 °C. Using the intravitreal procedure described earlier, each animal was injected with 3 μl fluorescein-NR58-3.14.3 in one eye only; the contralateral eye was used as either a fluorescein-only-injected control, or a non-injected control. Following the injection, animals were placed under dim-light conditions to recover. Both eyes of each animal were collected for cryosectioning at several times after the injection (20 min, 3 h, 24 h, and 7 days). The levels of fluorescein-NR58-3.14.3 in injected retinas were measured on cryosections. The fluorescence was visualised with a laser-scanning A1+ confocal microscope (Nikon, Tokyo, Japan), and images were acquired at consistent regions using the NIS-Elements AR software (Nikon, Tokyo, Japan). From these images, fluorescence intensity was quantified with ImageJ software (NIH, MD, USA), using integrated density for a fixed area.
Tissue collection and processing of whole retinas
Animals were euthanized using an overdose of barbiturate, which was administered via intraperitoneal injection (Valabarb; Virbac, NSW, Australia). The left eye from each animal was marked for orientation then enucleated for cryosectioning, while the retina from the right eye was excised through a corneal incision for RNA extraction. Both cryosectioning and RNA extraction of the excised retinas were performed according to our previous methodology [
45].
Fluorescence-activated cell sorting (FACS) of retinal macrophages
Rats in each treatment group were euthanized immediately after light damage (0 day) using barbiturate overdose as previously described. Retinas from both eyes were promptly removed through a corneal incision. Retinas from each animal were pooled and immediately placed in chilled Hank’s balanced salt solution (HBSS) and then subjected to light mechanical separation using a razor blade. Samples were transferred into a 0.2 % papain digestion cocktail as described in our previous protocol [
41], with minor modifications. Samples were incubated at 37 °C for 7 min then 8 °C for 30 min. After neutralisation and resuspension, the resulting homogenate was incubated in staining buffer containing a CD11b antibody conjugated to Alexa 647 (Biolegend, San Diego, CA, USA) for 30 min at 4 °C, and then washed in HBSS and resuspended in staining buffer. The resultant CD11b-stained samples were run through a fluorescence-activated cell sorter (FACS) (BD FACSAria II; BD Biosciences, Franklin Lakes, NJ, USA). The isolated CD11b+ macrophages were collected in staining buffer and kept chilled on ice until RNA extraction could be commenced. RNA extraction was conducted via a protocol we have previously established [
41]. Isolated total RNA was analysed for quantity and purity with a ND-1000 spectrophotometer (Nanodrop Technologies, Wilmington, DE, USA).
TUNEL assay and quantification of photoreceptor survival
To quantify levels of photoreceptor cell death in the retina, retinal cryosections were stained for apoptosis using a terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) kit (Roche Diagnostics, NSW, Australia) as previously described [
46,
47]. TUNEL+ cells in the ONL were counted throughout the full length of each retinal section along the para-sagittal plane (supero-inferior) and also included the optic nerve for consistency. These counts were subsequently averaged for each experimental group. To quantify photoreceptor survival, sections were stained with bisbenzimide (Sigma-Aldrich Co., MO, USA) and the number of rows of photoreceptor cell bodies in the ONL were counted. Five counts were made for each retinal section and then were averaged for each experimental group.
Immunohistochemistry
Cryosections were used for immunohistochemical analysis, using antibodies against IBA1 (1:500, Wako, Osaka, Japan) and GFAP (1:500, Dako, Agilent Technologies, CA, USA) as described in our previous study, with minor modifications [
41,
44]. Fluorescence in retinal sections was captured with a laser-scanning A1
+ confocal microscope (Nikon, Tokyo, Japan). Images panels were analysed and assembled using ImageJ (NIH, MD, USA) and Photoshop CS6 software (Adobe Systems, CA, USA).
Quantification of infiltrating macrophages
Macrophage counts were performed on sections immunolabelled with the IBA1 marker, a label for retinal microglia and macrophages. Counts of IBA1+ cells were carried out along the full length of retinal sections cut in the para-sagittal plane (supero-inferior) within the vertical meridian, and each section also included the optic nerve to ensure consistency. Counts were made of all IBA1+ cells throughout the retina, which were split into two groups: the inner retina (ILM-OPL) and the outer retina (ONL-RPE) where activated macrophages are known to aggregate during retinal degeneration [
19,
20,
48].
Polymerase chain reaction
cDNA was prepared from 1 μg of RNA of each sample using a Tetro cDNA Synthesis Kit (Bioline Reagents, London, UK). The quantitative real-time polymerase chain reactions (qPCRs) were conducted using Taqman hydrolysis probes (Table
1) and Taqman Gene Expression Master Mix (Applied Biosystems, Life Technologies, CA, USA). These were applied following the manufacturer’s instructions and run on a QuantStudio Flex 12K instrument (Applied Biosystems), at the Biomolecular Resource Facility, JCSMR, ANU. Data analysis was performed using Expression Suite v1.0.3 software (Life Technologies, CA, USA); the fold change was determined for each gene and sample using the comparative cycle threshold (
C
t) method (ΔΔ
C
t). All data was normalised to the GAPDH reference gene, which does not change in expression over the course of retinal light damage [
49,
50]. ACTB was utilised as an additional reference gene in qPCRs involving the CD11b+ isolates.
Table 1
TaqMan hydrolysis probes used for qPCR
Ccl2 | Chemokine (C–C motif) ligand 2 | Rn01456716_g1 | 24770 |
Ccl3 | Chemokine (C–C motif) ligand 3 | Rn00564660_m1 | 25542 |
Ccl4 | Chemokine (C–C motif) ligand 4 | Rn00587826_m1 | 116637 |
Ccl7 | Chemokine (C–C motif) ligand 7 | Rn01467286_m1 | 287561 |
IL-6 | Interleukin 6 | Rn01410330_m1 | 24498 |
IL-1B | Interleukin 1β | Rn00580432_m1 | 24494 |
TNF | Tumour necrosis factor α | Rn00562055_m1 | 24835 |
IL-12B | Interleukin 12β | Rn00575112_m1 | 64546 |
GAPDH | Glyceraldehyde-3-phosphate dehydrogenase | Rn99999916_s1 | 24383 |
ACTB | Actin, beta | Rn00667869_m1 | 81822 |
Standard PCR was performed on RNA samples purified from FACS-isolated CD11b+ macrophages, using primers specific to chemokine receptor genes (Ccr1, Ccr2, Ccr5, Cxcr3) which are detailed in our previous investigation [
41]. First-strand cDNA synthesis was performed from 50 ng of RNA using the Tetro cDNA Synthesis Kit (Bioline), and standard PCR was then performed using MyTaq DNA polymerase (Bioline); negative controls had cDNA omitted from the reaction cocktail. The presence of PCR product and specificity of the reaction were assessed by gel electrophoresis.
In situ hybridisation
Ccl3 was cloned from a 540 bp PCR product using cDNA prepared from rat retinas (as described in the qPCR section), and which was synthesised into a digoxigenin (DIG)-labelled riboprobe according to our previous publication [
41]. The in situ hybridisation was performed using our established protocol [
51]; the Ccl3 riboprobe was hybridised on retinal cryosections overnight at 55 °C and then washed in decreasing concentrations of saline sodium citrate (pH 7.4) at 60 °C. The bound probe was visualised with either NBT/BCIP or HNNP/Fast-Red (Roche).
Statistical analysis
All graphing and statistical analysis was performed using Prism 6 (GraphPad Software, CA, USA). A two-way ANOVA with Tukey’s multiple comparison post-test or an unpaired Student t test was utilised to determine the statistical outcome; a P value of <0.05 was considered statistically significant.
Discussion
The findings of the current study demonstrate the efficiency of the broad-spectrum chemokine inhibitor NR58-3.14.3 in modulating the pathogenic activity of macrophages/microglia and ameliorating photoreceptor death, in a model of photo-oxidative damage and outer retinal inflammation. First, our data demonstrate that the intravitreal delivery of NR58-3.14.3 is tolerated within the retinal environment, and its persistence in the retina provides for a generous window of activity. Second, we show that injection of NR58-3.14.3 suppresses the chronic accumulation of macrophages/microglia within the photoreceptor layer and subretinal space following damage. Third, NR58-3.14.3-treated macrophages demonstrated reduced pathogenic activation, indicated by decreased expression of Ccl3, Ccl4, and Il6. Finally, we show that photoreceptor degeneration is reduced across the time course of damage in retinas injected with NR58-3.14.3, which is accompanied by an increase in photoreceptor survival at the site of focal and progressive atrophy.
NR58-3.14.3 is thought to exert its inhibitory effect on Ccl- and Cxcl- families though disruption of intracellular signal transduction, which effectively renders leukocytes blind to directional cues provided by Ccl- and Cxcl- ligands [
60,
61]. This broad suppression is desirable, as the blockade of singular chemokine ligands, such as Ccl2, can induce compensatory increases in the expression other chemokines which exacerbate retinal inflammation and degeneration [
34,
36]. Up-regulation of Ccl2, Cxcl1, Cxcl10, and Cxcl11 are present in all forms of AMD [
28], and the Ccl2/Ccr2 axis is implicated in the pathogenic accumulation of macrophages to the outer retina in models that feature aspects of either CNV or geographic atrophy, such as light damage [
16,
31‐
35]. Increases in ocular Ccl- and Cxcl- chemokines are also found in individuals affected by retinal detachment [
62,
63], uveitis [
64], glaucoma [
65], diabetic retinopathy, and retinitis pigmentosa [
66], while blockade of Ccl2 signalling has proven beneficial in thwarting macrophage aggregation and cell death in experimental retinal detachment, diabetic retinopathy [
23], and retinitis pigmentosa [
67]. While the light damage paradigm does not model features of CNV or degenerations such as retinal detachment, our findings allude to the potential of BSCIs in targeting a broad swathe of chemokine activity across a spectrum of retinal degenerations.
To our knowledge, this is the first study to examine the potential of BSCIs as a potential therapeutic agent in retinal disease. The findings from our study are consistent with previous investigations using NR58-3.14.3 in inflammatory models ranging from atherosclerotic plaques to obliterative bronchiolitis, wherein an inhibition of macrophage accumulation and amelioration of tissue damage were observed [
61,
68,
69]. Our own data show that intravitreal NR58-3.14.3 reduces the accumulation of macrophages within the outer retina, including ONL and subretinal space. The proportion of these subretinal macrophages that are either resident microglia or recruited bone-marrow-derived monocytes is unclear, although the effect of intravitreal NR58-3.14.3 on these subpopulations is beyond the scope of this study. Nevertheless, we speculate that intravitreal NR58-3.14.3 could impact the infiltration of both subpopulations to the subretinal space, despite the local nature of its administration. This is suggested by the localisation fluorescein-NR58-3.14.3 within the retinal vasculature, where blood borne monocytes are known to recruit into the retina following light-induced injury [
36].
Our data indicate that NR58-3.14.3 modulates the transcriptional phenotype of macrophages following light damage. The expression of Ccl3, Ccl4, and Il6 were all down-regulated in macrophages exposed to NR58-3.14.3 and are consistent with the down-regulation of Il6 observed in experimental obliterative bronchiolitis following administration of NR58-3.14.3 [
39]. Ccl3, Ccl4, and Il6 are markers of M1 polarisation, which is typically associated with deleterious pro-inflammatory responses [
70]. Although macrophage polarisation is considered a simplification of extremes which form part of a phenotypic spectrum, the suppression of these genes within the macrophage population in NR58-3.14.3-injected retinas do point toward a lower inflammatory state among these cells as a result. How NR58-3.14.3 elicits this transcriptional change is unclear, though we speculate it may occur via its inhibition of macrophage chemotaxis to the outer retina, wherein emergent photoreceptor death may stimulate up-regulation of cytokines such as Ccl3, Ccl4, and Il6. This is supported by our in situ hybridisation data, wherein Ccl4 mRNA is expressed only in macrophages that have migrated to the outer retina.
Several lines of evidence suggest that Ccl3, Ccl4, and Il6 are important mediators of pathogenic activity by macrophages in retinal degeneration, which may account for the neuroprotective action of NR58-3.14.3. Increased levels of Il6 are associated with the incidence and progression of AMD, and the cytokine is considered a potential therapeutic target [
71‐
73]. A pathogenic role of Il6 is more recently highlighted by Levy and colleagues, who show that accumulation of the AMD biomarker apolipoprotein E (APOE) in subretinal macrophages up-regulates their expression of Il6, which in turn promotes their increased survival and induction of retinal pathology in light-damaged Cx3cr1−/− mice [
59]. Additionally, a deleterious role of Il6 has been characterised in experimental ocular toxoplasmosis via antibody neutralisation, which considerably improved retinal morphology compared to controls [
74].
Ccl3 is implicated in animal models of AMD/Stargardt disease (Abca4−/−Rdh8−/−), retinitis pigmentosa (Mertk−/−) [
34], and oxygen-induced retinopathy [
75]. Ccl3 was expressed by macrophages/microglia as shown by Kohno and colleagues—consistent with our current investigation—and its ablation reduced the extent of photoreceptor degeneration in both aged Abca4−/−Rdh8−/− mice and young Mertk−/− mice [
34]. In experimental oxygen-induced retinopathy, mice were treated with the combination therapy of Ccl3- and Ccl2-neutralising antibodies, rather than a singular therapy. Regardless, the neovascular pathology was reduced by 30 % in these animals [
75]. The precise role of Ccl4 in retinal degeneration has yet to be examined, although protein levels are increased in the aqueous humour of AMD patients [
76], and its expression is up-regulated concomitantly with emergence of retinal degeneration in experimental light damage [
41], diabetic retinopathy, and Abca4−/−Rdh8−/− and Mertk−/− mice [
34]. Moreover, ablation of Ccl3 induced compensatory increases in Ccl4 expression in light-damaged Abca4−/−Rdh8−/− mice, which was postulated to contribute to the increased pathology observed in this paradigm (rather than the age paradigm where there was protection).
We speculate that NR58-3.14.3 may also confer retinal protection by inhibiting other macrophage processes, such as aberrant phagocytosis. This phenomenon has been recently demonstrated in the rd10 model of retinitis pigmentosa, in which infiltrating microglial cells were found to exacerbate cell death by phagocytosing non-apoptotic photoreceptors [
77]. In suppressing the infiltration of macrophages within the outer retina, NR58-3.14.3 may also reduce the erratic phagocytosis of photoreceptors by activated macrophages. This may be a contributing factor to the increases in ONL thickness in NR58-3.14.3-injected retinas, although additional investigations are required to firmly establish this in the future.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
NF conducted the experiments and analysis and wrote the paper; RN designed and conducted the experiments; KV designed the experiments and assisted with the corrections; JP designed the experiments and assisted with the corrections. MR designed and conducted the experiments, conducted the analysis, and wrote the paper. All contributing authors have read and approved the final version of the manuscript.