The use of the vaccinia virus complement control protein (VCP) in the rat retina

The complement system is highly implicated in both the prevalence and progression of Age-Related Macular Degeneration (AMD). Complement system inhibitors therefore have potential therapeutic value in managing excessive activation of the complement pathways in retinal degenerations. The vaccinia virus complement control protein (VCP) has been shown to be effective as a complement inhibitor in neuroinflammatory models including traumatic brain injury and spinal cord injury. We aimed to investigate the potential of VCP as a therapeutic molecule for retinal degenerations. In this study, we investigated the effect, localisation and delivery of VCP to the rodent retina. Complement inhibition activity of VCP was tested using a hemolytic assay. Photoreceptor cell death, inflammation and retinal stress were assayed to determine if any retinal toxicity was induced by an intravitreal injection of VCP. The effect of VCP was investigated in a model of photo-oxidative retinal degeneration. Localisation of VCP after injection was determined using a fluorescein-tagged form of VCP, as well as immunohistochemistry. Finally, a copolymer resin (Elvax) was trialled for the slow-release delivery of VCP to the retina. We found that a dose equivalent to 20μg VCP when intravitreally injected into the rat eye did not cause any photoreceptor cell death or immune cell recruitment, but led to an increase in GFAP. In photo-oxidative damaged retinas, there were no differences in photoreceptor loss, retinal stress (Gfap) and inflammation (Ccl2 and C3) between VCP and saline-injected groups; however, Jun expression was reduced in VCP-treated retinas. After VCP was injected into the eye, it was taken up in all layers of the retina but was cleared within 1–3 hours of delivery. This study indicates that a method to sustain the delivery of VCP to the retina is necessary to further investigate the effect of VCP as a complement inhibitor for retinal degenerations.


Introduction
Dysregulation of the complement system, a key component of the innate immune response, is highly linked to the prevalence and progression of neurodegenerative conditions, including Age-Related Macular Degeneration (AMD) (reviewed in [1][2][3]). AMD is the leading cause of blindness worldwide [4], and is a disease that primarily affects the photoreceptors and retinal a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 University Animal Experimentation Ethics Committee (Ethics ID: A2012/07). Adult Sprague-Dawley (SD) and Wistar albino rats from the Australian Phenomics Facility aged between 120-150 post-natal days and held under dim light conditions (5 lux) were utilised throughout the study. To induce retinal degeneration, animals were exposed to 1000 lux light for 24 hours to induce photo-oxidative damage as described previously [48]. Animals were euthanized for tissue collection at 0 days (immediately after light exposure) or at 7 days post-exposure (returned to 5 lux dim conditions).

Hemolytic assay for complement activity
The ability of VCP to inhibit complement was tested using a hemolytic assay, according to a previously described protocol [49]. In brief, sheep red blood cells (RBCs; Applied Biological Products Management, SA, Australia) were sensitised with a hemolysin antibody (Sigma-Aldrich, MO, USA). Recombinant VCP (GeneBalance Inc., MA, USA) was reconstituted using 0.9% NaCl sterile saline (Pfizer, NY, USA) to varying concentrations (0-0.6μg/μl). The antibody-sensitised red blood cells were then incubated with rabbit serum (Applied Biological Products Management) and VCP for 1 hour at 37˚C. After centrifugation of the cell suspension, the supernatant was transferred to a 96-well plate in duplicate. The absorbance of hemoglobin at 540nm was measured using a spectrophotometer, and the % hemolysis of the red blood cells for each concentration of VCP was determined. Two controls were used for comparison of hemolytic activity; a 100% lysis control (RBCs with serum, no VCP) and a 0% spontaneous lysis control (RBCs only, no serum).

Intravitreal injections and tissue collection
Using a dose of 20μg per eye, the VCP protein was reconstituted in 0.9% NaCl (saline) to a concentration of 4μg/μl. Intravitreal injections were carried out using a protocol described previously [44]. Both eyes of each animal were intravitreally injected with either 5μl VCP in saline, or 5μl saline only. Animals were left for 7 days in dim conditions, or were exposed to photooxidative damage for 24 hours following the injections and recovery from anaesthesia. After photo-oxidative damage, animals were immediately euthanized with an intraperitoneal overdose of barbiturate (Valabarb, Virbac, NSW, Australia). One eye of each animal was enucleated, fixed in 4% paraformaldehyde, and stored in 15% sucrose overnight prior to embedding in Tissue-Tek OCT Compound (Sakura Finetek, CA, USA). The retina was excised from the other eye and stored in chilled RNAlater for RNA extraction. Eyes were then cryosectioned at 16μm. Non-injected eyes/retinas from dim-reared control animals were collected for comparison.

RNA extraction, cDNA synthesis and qPCR
RNA extraction was carried out using a combination of TRIzol (Thermo Fisher Scientific) and an RNAqueous Micro Total RNA Isolation Kit (Thermo Fisher Scientific) as previously described [17]. cDNA was prepared using a Tetro cDNA Synthesis Kit (Bioline, London, UK) according to the manufacturer's instructions. Quantitative real-time polymerase chain reaction (qPCR) was carried out using Taqman hydrolysis assays (Table 1) and Taqman Gene Expression Master Mix (Thermo Fisher Scientific). Reactions were carried out on a QuantStudio Flex 12K instrument (Thermo Fisher Scientific) at the Biomolecular Resource Facility (JCSMR, ANU). The expression change of all target genes was compared to the Gapdh reference gene using the comparative C t method (ΔΔC t ).

TUNEL and immunohistochemistry
Retinal sections were assayed in duplicate per animal for photoreceptor cell death using a terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) kit (Roche, Basel, Switzerland) according to previously described protocols [50,51]. TUNEL+ photoreceptors in the outer nuclear layer (ONL) were counted, or ONL thickness measured at the lesion site (2mm superior to the optic nerve) using bisbenzimide (1:10000; Sigma-Aldrich) labelling and an LSM 5 PASCAL confocal microscope (Zeiss, West Germany). Relative ONL thickness was calculated as a ratio of the ONL thickness to the distance between the inner and outer limiting membranes. To detect microglia and macrophages in the retina, immunohistochemistry using an α-IBA1 primary antibody (1:500, #019-19741, Wako, Osaka, Japan) was performed on retinal sections according to previously described protocols [52,53]. IBA1+ cells in the outer retina (ONL-RPE) were counted for each retinal section. To detect retinal stress, immunohistochemistry using an α-GFAP primary antibody (1:500, #Z033429-2, Agilent Technologies, CA, USA) was performed on retinal sections. Fluorescence intensity for GFAP labelling was measured using ImageJ (NIH, MD, USA).

Detection of VCP in the retina using fluorescein
The fluorescein-labelled VCP was prepared using an NHS-Fluorescein Antibody Labelling Kit (Pierce, Thermo Fisher Scientific, MA, USA). In brief, the VCP protein was reconstituted in 0.9% NaCl (saline) to a concentration of 2μg/μl. The VCP was then incubated with the NHS-Fluorescein for 1 hour, and was stored at 4˚C until required for the injections. Animals were injected with 5μl fluorescein-VCP into one eye, whilst the fellow eye served as a noninjected control. Eyes were then collected at several time points (20mins, 1hr, 3hrs, 6hrs, 14hrs and 24hrs) after the injection, and were sectioned. Fluorescein-VCP was visualised in retinal sections under the confocal microscope.

Detection of VCP in the retina using immunohistochemistry
To verify the localisation of VCP in the retina in addition to fluorescein labelling, an α-VCP (anti-SPICE hybridoma KL5.1, 1.64mg/ml) mouse monoclonal antibody [54] was utilised to detect VCP in retinal sections. Specificity of the α-VCP antibody was verified using dot blot analysis. In summary, the VCP protein was transferred onto a 0.22μm PVDF membrane. A mixture of isolated proteins from the retina was also blotted onto the membrane as a control. The membrane was then blocked in 3% bovine serum albumin (BSA) for 3 hours before being incubated with the α-VCP antibody (1:100) overnight at 4˚C. The next day, the membrane was incubated with a HRP-conjugated secondary antibody (BioRad, CA, USA) for 2 hours, and a peroxidase colour substrate (Vector VIP; Vector Labs, CA, USA) was added to visualise the reaction. Dots on the membrane were imaged using the confocal microscope. Immunohistochemistry was performed on retinal sections using the α-VCP antibody (1:100) according to previously described protocols [52,53].

Delivery of VCP using Elvax
To test the slow-release delivery of VCP using an Elvax (ethylene vinyl acetate) copolymer resin implant in the eye, a VCP-Elvax drug-resin complex was prepared according to previously described protocols [55]. In brief, 40μl of fluorescein-VCP (2μg/μl) was incorporated into the Elvax resin to form a drug-resin complex. After freeze-drying the resin to form a block, a fluorescein-VCP-Elvax implant was prepared from the block, and was implanted into one eye of each animal, following previously published procedures [55]. The contralateral eye was intravitreally injected with 5μl fluorescein-VCP (2μg/μl) as described previously. Eyes were collected at several time points (20mins, 6hrs, 24hrs and 7 days) after the implantation or injection and were sectioned. Fluorescein-VCP was visualised in retinal sections under the confocal microscope.

Imaging and statistical analysis
For each retinal section, images of fluorescence in the superior retina (2mm from the optic nerve) were acquired using an LSM 5 PASCAL confocal microscope (Zeiss, West Germany) or an A1 + laser-scanning confocal microscope (Nikon, Tokyo, Japan). Images were analysed and processed using ImageJ and Photoshop CS6 software (Adobe Systems, CA, USA). All graphing and statistical analysis was performed using Prism 6 (GraphPad Software, CA, USA), using unpaired Student's t-tests, or 1-way analysis of variance (ANOVA) with Tukey's multiple comparisons post-test to determine statistical significance ( Ã P<0.05).

Effective dose of VCP required for complement inhibition
To establish an effective dose of VCP for complement inhibition, complement activity of rabbit serum was tested in a hemolytic assay, where varying concentrations of VCP (0-0.6μg/μl) was added. The % lysis of RBCs was determined for each concentration of VCP in the cell suspension, which was 100% when only serum (no VCP) was added, indicating maximal complement activity of the serum (Fig 1). The spontaneous lysis control (no serum) had 0% lysis. Addition of 0.1μg/μl VCP caused a reduction of hemolysis by approximately 10%, whereas addition of 0.2μg/μl VCP decreased hemolysis by approximately 50%, demonstrating increasing complement inhibition relative to additional VCP. By addition of 0.4μg/μl VCP, there was a near-complete inhibition of complement serum activity. This indicates that a concentration of 0.2-0.4μg/μl VCP was effective at inhibiting complement. This dose is equivalent to 20μg in the rat eye, in which the total vitreal volume is 50-55μl.

Assessment of safety of VCP in the retina
To explore any toxicity that may arise from delivering 20μg VCP in 5μl saline to the rat eye, retinal histology was assessed using TUNEL as a measure of photoreceptor cell death, bisbenzimide to visualise the retinal layers, IBA1 immunohistochemistry to quantify the recruitment of retinal microglia/macrophages to the outer retina (ONL-RPE), and GFAP immunohistochemistry to determine general retinal stress. At 7 days after an intravitreal injection of 20μg VCP in 5μl saline to the rat eye, there was no significant difference in the number of TUNEL+ photoreceptors compared to non-injected controls, which were only 0-1 cells per retinal section (P>0.05, Fig 2A, 2C and 2D). There was no visible difference in photoreceptor layer organisation, or overall retinal histology between VCP-injected and non-injected groups (Fig 2C  and 2D). No IBA1+ cells were detected in the outer retina of both VCP-injected and noninjected groups at 7 days, indicating little to no retinal inflammation following the intravitreal injection (Fig 2E and 2F). However, at 7 days after VCP was injected, there was a significant increase in GFAP staining in the VCP-injected retinas compared to non-injected controls, indicating that mild stress was induced in the retina following the VCP injection (P<0.05, Fig  2B, 2G and 2H).

Effect of VCP on the photo-oxidative damaged retina
We assessed the effect of delivering VCP to the retina prior to photo-oxidative damage. Animals were intravitreally injected with 20μg VCP or saline immediately prior to exposure to photo-oxidative damage for 24 hours, and eyes were collected both immediately and at 7 days after light exposure. Following retinal damage, there was a trend towards a reduction in photoreceptor cell death in the VCP-injected retinas compared to the saline-injected controls; however, this change was not significant (P>0.05 Fig 3A). Outer nuclear layer (ONL) thickness was measured at the lesion site (2mm superior to the optic nerve in retinal cryosections) at 7 days after photo-oxidative damage to investigate any cumulative photoreceptor cell loss that may have occurred. There was no significant difference in photoreceptor layer thickness between the two groups (P>0.05, Fig 3B). Changes in whole retinal gene expression was determined using qPCR. There was a significant reduction in Jun, a marker of apoptosis, in VCPinjected retinas compared to saline controls (P<0.05, Fig 3C). There was no significant change in the retinal stress marker Gfap, or inflammatory markers Ccl2 and C3 in the retina between VCP and saline groups (P>0.05, Fig 3D and 3F).

Localisation of VCP in the retina
We investigated the retinal localisation of VCP following an intravitreal injection, where 10μg of fluorescein-VCP was injected into each eye (concentration at 2mg/ml optimised for the fluorescein labelling kit). At 20 minutes after the injection, fluorescein-VCP was detected in all layers of the inner and outer retina (Fig 4B and 4C). This was observed in nasal, central and temporal parasagittal sections, and the fluorescein-VCP was distributed evenly from the central retina to the peripheral retina. However, from 1-3 hours after injection onwards (Fig 4B), fluorescein-VCP was cleared from the retina, and only background levels of fluorescence were detected up until 24 hours (Fig 4F). An α-VCP antibody, confirmed to have immunoreactivity for VCP (Fig 4A), was used to verify VCP localisation in the retina, which co-localised at all time points with the fluorescein-VCP (Fig 4B, 4D, 4E, 4G and 4H). This data indicates that VCP is cleared from the retina within 1-3 hours after intravitreal delivery.

Slow-release delivery of VCP using Elvax
To provide an extended drug delivery of VCP to the retina, a slow-release delivery vehicle (Elvax) was used, where the drug-resin complex is inserted into the eye [55]. Fluorescein-VCP was delivered through this complex, and fluorescence intensity was measured across a time course. While we observed fluorescein-VCP release from the complex, we were unable to find levels of VCP in retinal cryosections above background (Fig 5, dashed black line). At 20 minutes, fluorescein-VCP intravitreally injected into the eye was present in all layers of the retina as observed previously in Fig 4. The contralateral eye, which contained the VCP-Elvax complex had only background levels of fluorescence at 20 minutes, which was significantly less than the injected eye (P<0.05, Fig 5). Eye injected with fluorescein-VCP and implanted with VCP-Elvax had only background levels of fluorescence in the retina at 6 hours, 24 hours and 7 days after the drug was delivered, indicating that minimal VCP was detected in the retina at each time point following Elvax drug delivery.  The use of the vaccinia virus complement control protein (VCP) in the rat retina The use of the vaccinia virus complement control protein (VCP) in the rat retina

Discussion
This study aimed to investigate the use of the complement inhibitor VCP in the eye for the treatment of retinal degenerations, where excessive complement activation is a key feature of disease progression. We demonstrated that a concentration of 0.2-0.4μg/μl VCP is effective at inhibiting complement, which is equivalent to a dose of 20μg VCP in the rat eye. This dose did not induce any cell death or inflammation when delivered intravitreally to the eye, however led to an increase in GFAP levels, indicating mild retinal stress. In photo-oxidative damaged retinas, there was no significant change in photoreceptor loss, and expression of retinal stress (Gfap) and inflammation (Ccl2 and C3) between VCP and saline-injected groups. However, Jun expression was significantly reduced in VCP-injected retinas compared to saline controls. We found that after delivering VCP to the eye, it was cleared from all layers of the retina within 1-3 hours. The findings indicate that if VCP is to be pursued as a candidate for the treatment of retinal degenerations, a slow-release delivery vehicle or method to extend the life of VCP in the eye is necessary. VCP is a novel C3 inhibitor known to carry out complement inhibition by acting on the C3 convertases of all pathways, preventing C3 from being hydrolysed [40]. In the current study, we have shown via a hemolytic assay that complement serum activity is inhibited following incubation with 0.2-0.4μg/μl VCP. The equivalent dosage in the rat eye is approximately 20μg VCP, which would vary between rats, mice and humans due to differences in vitreal volume. In the rat retina, this dose was found to be non-toxic when delivered intravitreally, causing negligible photoreceptor cell death. In previous studies using various neuroinflammatory disease models, a similar dose of VCP has been administered. In a model of traumatic brain injury, 22.5μg VCP was injected into the brain which improved spatial memory compared to saline-treated controls [42]. Another study found that two bilateral intraparenchymal The use of the vaccinia virus complement control protein (VCP) in the rat retina injections of 24μg VCP was able to inhibit macrophage accumulation and improve function in a model of spinal cord injury [43]. It is indicated that this dose of VCP may be effective against neuroinflammation caused by retinal degenerations.
The key role of the complement system in the progression of retinal degenerations including AMD is well-established [3]. In the rodent photo-oxidative damage model that recapitulates several features of the atrophic form of AMD [48,56], excessive complement activation is known to be linked to a progressive loss of photoreceptors within a focal retinal lesion [46]. This complement activation is associated with activated microglia and macrophages, which are recruited to the lesion site and are known to produce complement components that may lead to further photoreceptor cell death [17,46]. We wanted to investigate whether the effect of VCP could be tested using the photo-oxidative damage model. There were no significant differences in photoreceptor loss and the expression of Gfap, Ccl2 and C3 between VCP and saline-injected groups.
In contrast, Jun gene expression was significantly decreased in the VCP-treated retinas compared to saline controls. Jun encodes the transcription factor AP-1 involved in apoptosisrelated mechanisms. Together, with the trend towards a reduction in photoreceptor cell death immediately after photo-oxidative damage in the VCP-treated retinas, it is indicated that there could be a decrease in photoreceptor death, possibly through complement inhibition. However, at 7 days after photo-oxidative damage, VCP had no effect on photoreceptor loss. In our previous studies, we have shown that the delivery of C3 siRNA to the retina can ameliorate complement-mediated photoreceptor loss over time [46], demonstrating that C3 inhibition as a therapeutic strategy may be beneficial. Although VCP has the ability to control complement, our present data suggest that its half-life in the retina is very short, and consequently does not lead to the expected outcomes in vivo.
Delivery of fluorescein-tagged VCP to the eye indicated that VCP was taken up by all layers of the retina, but cleared from the retina within 1-3 hours. The mechanism for the rapid clearance of VCP from the retina is not known. It may be cleared through the vitreous via the anterior chamber, or alternatively across the blood-retinal barrier in the posterior chamber. Compounds with a larger molecular weight and increased water-solubility tend to have a longer half-life in the vitreous [57]. Although VCP is water-soluble, its smaller molecular weight might contribute to its fast clearance from the eye. When delivered intraperitoneally at a concentration of 20mg/kg [41], VCP activity in serum lasts 3-4 days. Further, it is thought that heparin-binding sites of VCP may allow for sequestering of the compound in specific tissues, thereby extending its half-life [40,58,59]. The possibility for heparin-binding as a mechanism for VCP's rapid clearance may warrant further investigation.
We also investigated the use of a VCP-Elvax implant, to provide extended release of fluorescein-labelled VCP into the retina. This technique has been used previously to effectively slowrelease compounds into the eye, with a functional effect on the retina [55]. While we observed the release of fluorescein-VCP from the implant into the eye in tissue sections, we did not detect an increase in fluorescein-VCP in the retinal layers. It is likely that small amounts of VCP were being released from the complex into the eye, but were cleared from the retina resulting in minimal detection of the compound.
Other particle carriers including biodegradable microspheres, nanoparticles and liposomes have proven efficacious at delivering a sustained release of drugs to the eye [60][61][62], but may not extend the availability of VCP in the retina itself beyond 1-3 hours. As our photo-oxidative retinal degeneration model requires 24 hours of light exposure, which induces photoreceptor cell death, inflammation and complement over the ensuing 7 days [46,48], the half-life of VCP in the retina needs to be extended to appropriately assess its effect against retinal damage. Further investigation into prolonging the availability of VCP in the retina itself is necessary before VCP can be considered further than a theoretical candidate for use in models of retinal degeneration.