Cytokine and Complement Response in the Glaucomatous βB1-CTGF Mouse Model

Glaucoma is a complex neurodegenerative disease leading to a loss of retinal ganglion cells (RGCs) and optic nerve axons. An activation of the complement system seems to contribute to cell loss in this disease. Hence, we investigated a possible initiation of the complement system and the cytokine response in the βB1-CTGF glaucoma model. In these mice, intraocular pressure is elevated, which is the main glaucoma risk factor in patients, and RGC loss occurs at 15 weeks of age. Therefore, quantitative real-time PCR and immunohistological experiments were performed in 5-, 10-, and 15-week-old βB1-CTGF animals and their corresponding wildtypes (WT) to analyze the expression of several complement system factors. We could show that mRNA levels of the terminal complement pathway components C3 and C5 (Hc) were upregulated at 10 weeks. In accordance, more C3+ and membrane attack complex+ cells were observed in transgenic retinae. Further, the C5a receptor anaphylatoxin receptor (C5ar) and the complement component C5a receptor 1 (C5ar1; CD88) mRNA levels were upregulated in 10- and 15-week-old βB1-CTGF mice. Interestingly, all three activation routes of the complement system were elevated in βB1-CTGF mice at some age. Especially C1q, as a marker of the classical pathway, was significantly increased at all investigated ages. Furthermore, mRNA expression levels of interferon-γ (Infg) were upregulated at 5 weeks, while Cxcl1 and Cxcl2 mRNA levels were upregulated at 10 and 15 weeks. The mRNA levels of the chemokines Cxcl10 were increased at all ages in βB1-CTGF mice. These results lead to the assumption that in these transgenic mice, a complement activation mainly through the classical pathway as well as a cytokine response plays a major role in cell death.


INTRODUCTION
The complement system forms an integral part of the early immune response. Three complement cascades, the classical, the lectin, and the alternative, can activate the terminal pathway including the membrane attack complex (MAC) (Walport, 2001;Bayly-Jones et al., 2017). It could be demonstrated that in nucleated target cells, membrane disruption leads to cell death by apoptosis or by lysis if enough MAC is present (Koski et al., 1983;Kim et al., 1989;Nauta et al., 2002).
In the eye, the complement system is usually activated at low levels and provides immune tolerance (Sohn et al., 2000;Niederkorn, 2007). Recent data indicates that all retinal cells are able to express components of the complement system (Pauly et al., 2019). Albeit it resembles a double-edged sword since its activation can either benefit or harm the host and is therefore probably involved in several retinal diseases. In the last years, a contribution of the complement in glaucoma, a leading cause of blindness worldwide, was described. An elevated intraocular pressure (IOP) is the main risk factor for this disease and hallmarks are a loss of retinal ganglion cells (RGCs) as well as a degeneration of optic nerve axons (Casson et al., 2012). In donor eyes of glaucoma patients, an upregulation of proteins associated with the complement system, for example C1s, C3, C9, and mannose-binding serin proteases (MASPs) 1 and 2, were detected (Tezel et al., 2010;Doudevski et al., 2014). Furthermore, C1q, a component linked to the classical pathway, was found in human glaucoma retinae (Stasi et al., 2006). In addition, proteome analyses identified lectin pathway proteins (Tezel et al., 2010). Assembling of MAC can result in an activation of pro-inflammatory cytokines, such as interleukin (IL)-1, tumor necrosis factor α (TNFα), CCL2, interferon-γ (INF-γ), or IL-8 (CXCL8) (Schonermark et al., 1991;Liu et al., 2011;Xie et al., 2019). In aqueous humor samples of glaucoma patients, certain cytokines, like IL-1α, IL-8, CCL2, and INFγ, were upregulated (Chua et al., 2012;Chono et al., 2018). Furthermore, a recent study suggests that elevated levels of IL-8 can be considered a risk factor used for detecting and managing glaucoma (Chono et al., 2018).
In the DBA2/J model as well as in a surgical induced ocular hypertension (OHT) model, an activated complement system could be revealed (Kuehn et al., 2006;Stasi et al., 2006). Similar results were shown in an IOP-independent glaucoma model (Reinehr et al., , 2018b. However, the contribution of the complement system in the transgenic βB1-CTGF mouse model has not been investigated yet. Previous studies showed that in these transgenic mice, IOP elevation is accompanied by a degeneration of optic nerve axons and apoptotic RGC loss making it a reliable model for primary open-angle glaucoma (POAG) (Junglas et al., 2012;Reinehr et al., 2019b;Weiss et al., 2021). Particularly, already at the ages of 5 and 10 weeks, more cleaved caspase 3 + RGCs and TUNEL + apoptotic cells were detected in the transgenic βB1-CTGF mice. This was accompanied by a loss of retinal neurofilament H in 10-week-old mice (Weiss et al., 2021).
Hence, we now aimed to determine, which mechanisms lead to the apoptosis and subsequent cell loss in this model. Here, we investigated a possible contribution of the complement system. In addition, a potential corresponding cytokine response was investigated in βB1-CTGF mice at different ages. We revealed, for the first time, an activation of the complement system, mainly via the classical pathway, as well as a response of proinflammatory cytokines. This was noted even before a notably RGC loss occurred in the βB1-CTGF model as shown previously (Reinehr et al., 2019b).

Animals
All procedures concerning animals adhered to the ARVO statement for the use of animals in ophthalmic and vision research. All experiments involving animals were approved by the animal care committee of North Rhine-Westphalia, Germany. Mice were kept under environmentally controlled conditions with free access to chow and water.
The used βB1-CTGF and wildtype (WT) animals in this study had a CD1 background (Reinehr et al., 2019b;Weiss et al., 2021). All animals were bred in-house at the animal facility at the Ruhr-University Bochum. WT CD1 mice for breeding were obtained from Charles River (Sulzfeld, Germany). βB1-CTGF mice for breeding were kindly provided by Prof. Fuchshofer (University Regensburg, Germany). Then, all animals for this study were bred and housed at the animal facility at the Ruhr-University Bochum (Bochum, Germany). Potential βB1-CTGF mice were screened by isolating genomic DNA from tail biopsies and testing for transgenic sequenced by PCR, using the following primer sequences: 5 -GGAAGTGCCAGCTCATCAGT-3 and 5 -GTGCGGGACAGAAACCTG-3 . 5-, 10-, and 15-week-old female and male mice were included in the current study.

Quantitative Real-Time PCR
Both retinae of an animal at each age (5, 10, and 15 weeks; n = 5-7 animals/group) were pooled for RNA preparation and cDNA synthesis as previously described (Reinehr et al., 2019b). Total RNA was extracted with the Gene Elute Mammalian Total RNA Miniprep Kit, including a digestion with RNAase free DNase according to the manufacturer's instructions (Sigma-Aldrich, St. Louis, MO, United States). The quantity and quality of the RNA were determined utilizing a NanoDrop ONE (Thermo Fisher Scientific, Schwerte, Germany). Total RNA of 1 µg was applied for reverse transcription using a cDNA synthesis kit (Thermo Fisher Scientific). The designed oligonucleotides for RT-qPCR are shown in Table 1. Actb and Cyclophilin (Ppid) served as reference genes. The RT-qPCR was performed using DyNAmo Flash SYBR Green (Thermo Fisher Scientific) on the PikoReal RT-qPCR Cycler (Thermo Fisher Scientific) (Palmhof et al., 2018;Wilmes et al., 2018). No template controls, where PCR grade water (Roche Diagnostics, Basel, Switzerland) was applied instead of cDNA, were used as negative controls (Adams, 2020). For quantification, values were transferred to REST© software (Qiagen, Hilden, Germany) for further analysis.

Tissue Preparation for Immunohistology
At 5, 10, and 15 weeks, eyes (n = 7-9/group) were enucleated and fixed in 4% paraformaldehyde for 1 h. Thereafter, the eyes underwent a 30% sucrose treatment overnight and got embedded in a Neg-50 compound (Tissue-Tek; Thermo Fisher Scientific). 10 µm thick cross-sections were cut with a cryostat (Thermo Fisher Scientific) for further staining (Casola et al., 2016). The listed oligonucleotide pairs were used in quantitative real-time PCR experiments, while Actb and Cyclophilin (Ppid) served as housekeeping genes. The predicted amplicon sizes are given. F, forward; R, reverse; acc. no., accession number; bp, base pair.

Immunohistology
In order to identify different complement markers, specific immunofluorescence antibodies were applied (n = 7-9 eyes/group; 6 sections/animal, Table 2; Reinehr et al., 2016). Briefly, retina cross-sections were blocked with a solution containing 10-20% donkey, 2-3% BSA and/or goat serum and 0.1% or 0.3% Triton-X in PBS. Sections were incubated with primary antibodies at room temperature overnight. Incubation using corresponding secondary antibodies was performed for 1 h on the next day. Nuclear staining with 4 ,6 diamidino-2-phenylindole (DAPI, Serva Electrophoresis, Heidelberg, Germany) was included to facilitate the orientation on the slides. Negative controls were performed for each stain by using secondary antibodies only.

Histological Examination
The single plane photographs were taken using a fluorescence microscope using 400x magnification (Axio Imager M1 or M2, Zeiss, Oberkochen, Germany). Two photos of the peripheral and two of the central part of each retina cross-section were captured. The images were transferred to Corel Paint Shop Pro (V13, Corel Corporation, Ottawa, Canada) and equal excerpts were cut out. Afterward, C3 + , MAC + , Factor B + , and C1q + cells were counted in the ganglion cell layer (GCL) using ImageJ software (NIH, Bethesda, MD, United States) (Reinehr et al., 2019b). MASP2 + staining area in the nerve fiber layer, GCL, and inner plexiform layer, was evaluated using ImageJ software. Briefly, images were transformed into grayscale. To minimize interference with background labeling, a defined rolling ball radius of 150 pixels was subtracted. The percentage (%) of the labeled area was then measured between defined thresholds, which were obtained when the grayscale and the original picture corresponded the most (lower threshold: 13.76; upper threshold: 264.8) (Casola et al., 2016;Reinehr et al., 2018a).

Statistics
Regarding RT-qPCR, the relative expression values are presented as median ± quartile + minimum/maximum and were assessed via Pair Wise Fixed Reallocation Randomization Test using REST© software (Qiagen) (Pfaffl, 2001;Reinehr et al., 2019b,c). Immunhistological data are presented as mean ± SEM. The βB1-CTGF animals were compared to the WT group via twotailed Student's t-test using Statistica Software (Version 13, Dell, Tulsa, OK, United States). Control values were set to 100%. P-values below 0.05 were considered statistically significant, with * p < 0.05, * * p < 0.01, and * * * p < 0.001.

Upregulation of C3 mRNA Levels
C3 mRNA levels were evaluated via RT-qPCR at all points in time. In 5-week-old βB1-CTGF mice, no alterations in C3 mRNA expression levels were noted (1.03-fold expression; p = 0.880). At 10 weeks, a significant upregulation of C3 mRNA could be revealed in transgenic mice (1.55-fold expression; p = 0.048). At 15 weeks, the mRNA expression of C3 was normalized again and remained unchanged (1.55-fold expression; p = 0.123; Figure 1A).
Further, to gain an insight which cells might produce C3 in the retina, co-stainings with antibodies against anti-NeuN (neurons), anti-GFAP (astrocytes), and anti-Iba1 (macrophages/microglia) were performed in 10-week-old WT and βB1-CTGF mice. A co-localization of C3 + cells was mainly found with astrocytes and macrophages/microglia (Supplementary Figures 1A,C,E).

Enhancement of Terminal Complex Components
The terminal complement pathway consists of several proteins, which in the end assemble MAC. Here, the mRNA expression levels of C5 (Hc), as part of MAC, were analyzed through RT-qPCR. With increasing age, the Hc mRNA expression levels in transgenic mice continuously enhanced. At 5 weeks, no changes could be noted in Hc mRNA expression levels in transgenic mice (0.81-fold expression; p = 0.205). In 10-week-old βB1-CTGF mice, a trend toward an upregulation was noted (2.26-fold expression; p = 0.059). A significant upregulation of Hc mRNA expression levels was observed in 15-week-old transgenic animals compared to WT (2.32-fold expression; p = 0.025; Figure 2A).

Activation of Alternative Pathway
The activators of the alternative complement pathway, Cfb and Cfd as well as the regulator Cfh were examined via RT-qPCR analysis. At 5 and 10 weeks, no alterations were noted in Cfb mRNA expression levels in βB1-CTGF retinae compared to WT (5 weeks: 1.11-fold expression; p = 0.542; 10 weeks: 1.27-fold Frontiers in Cellular Neuroscience | www.frontiersin.org FIGURE 1 | Increased C3 expression in transgenic retinae. (A) C3 mRNA levels were increased in βB1-CTGF retinae at 10 (p = 0.048), but not at 5 or 15 weeks. (B) Retinal cross-sections were stained with an antibody against C3 (red), while cell nuclei were labeled with DAPI (blue) at all ages. (C) The number of C3 + cells in the ganglion cell layer was not altered at 5 and 15 weeks. At 10 weeks, significantly more C3 + cells could be observed in retinae of βB1-CTGF mice compared to WT (p = 0.009). Values for RT-qPCR are median ± quartile ± minimum/maximum (n = 5-7/group). The dotted line in panel (A) represents the relative expression of the WT group. Values for immunohistology are mean ± SEM (n = 7-9/group). GCL, ganglion cell layer; IPL, inner plexiform layer. Scale bar: 20 µm. *p < 0.05, **p < 0.01. expression; p = 0.173). In 15-weeks old transgenic mice, Cfb mRNA expression levels were significantly upregulated (7.01-fold expression; p = 0.013; Figure 3A).

Mild Activation of Lectin Pathway
Masp2 mRNA expression levels were analyzed through RT-qPCR. At 5 weeks, the Masp2 mRNA expression was not altered in βB1-CTGF mice compared to WT (1.35-fold expression; p = 0.297). In 10-week-old transgenic animals, a significant upregulation of Masp2 mRNA was noted (1.79-fold expression; p = 0.047). At 15 weeks, the mRNA expression of Masp2 went back to WT level (0.72-fold expression; p = 0.077; Figure 5A).
The mRNA expression levels of Cxcl2 were not altered in 5-week-old transgenic mice (0.98-fold expression; p = 0.940).
Retinal cross-sections were labeled with an antibody against IP-10 (CXCL-10) and RBPMS. Representative pictures show that CXCL-10 + cells were predominantly localized in the GCL and inner nuclear layer of βB1-CTGF mice (Figure 6E).

DISCUSSION
Glaucoma is a complex neurodegenerative disease, which can lead to blindness when untreated. It often remains asymptomatic for a long time, since progressive visual field loss is peripheral The number of Factor B + cells in the ganglion cell layer was significantly higher in βB1-CTGF animals compared to WT (p = 0.046). No changes were seen at 10 and 15 weeks. Values for RT-qPCR are median ± quartile ± minimum/maximum (n = 5-7/group). The dotted line in panels (A-C) represents the relative expression of the WT group. Values for immunohistology are mean ± SEM (n = 7-9/group). GCL, ganglion cell layer; IPL, inner plexiform layer. Scale bar: 20 µm. *p < 0.05. and typically asymmetric, which allows for compensation from the overlapping visual field of the other eye (Gupta and Chen, 2016). Unfortunately, the only currently modifiable factor is the IOP. Therefore, there is an urgent need for new and additional therapeutic approaches. Hence, other mechanisms besides the main risk factor IOP elevation need to be investigated.
For instance, previous studies concluded that an activation of the complement system seems to contribute to cell death in glaucoma. In the study presented here, we aimed to determine complement activation as well as cytokine expression in different ages of the transgenic βB1-CTGF mouse model. This model enables to study underlying pathomechanisms occurring in high-pressure glaucoma. In contrast to other OHT models, no surgical interventions are needed. While the IOP was not altered in 5-and 10-week-old βB1-CTGF mice, elevated IOP was accompanied by an apoptotic loss of RCGs in 15-week-old animals (Reinehr et al., 2019b). Furthermore, more apoptotic RGCs and TUNEL + apoptotic cells were already noted in 5-and 10-week-old transgenic mice (Weiss et al., 2021). The relation of apoptotic cells and complement activation should be investigated more precisely in future studies.
In the aqueous humor of patients suffering from POAG, the complement factor C3 was elevated (Liu et al., 2020). Recently, Hubens et al. (2021) demonstrated an upregulation of the C3a/C3 ratio in aqueous humor from progressive, but not from stable POAG patients. In our study, we could demonstrate that the terminal complement pathway components C3 and MAC were increased in 10-week-old βB1-CTGF mice before an RGC death was noted. Especially, the formation of MAC could lead to the apoptosis of cells, which was shown, for example, in rat mesangial cells (Nauta et al., 2002). Regarding the initiation of the complement system, similar results were obtained in an IOP-independent experimental autoimmune glaucoma (EAG) model. Here, immunization with ocular antigens led to an activation of the complement system before RGC and optic nerve degeneration (Reinehr et al., , 2018b. Also, in DBA/2J.Wld 5 mice, which are protected from axon dysfunction, early C3 upregulation could be observed (Harder et al., 2017). Further, the C5a anaphylatoxin receptor (C5ar) as well as the complement component C5a receptor 1 (C5ar1; CD88) were found upregulated transgenic mice at 10 and 15 weeks of age. C5a itself is a powerful mediator of inflammation and is important in the recruitment and chemotaxis of glia to the site of damage, through its anaphylatoxin G-protein coupled receptor (Yao et al., 1990;Miller and Stella, 2009;Woodruff et al., 2010). In accordance with studies in other glaucoma models, our results reveal that complement activation is an early event in glaucoma pathology.
To elaborate, which cells might be the source of C3 and MAC in our model, we performed co-stainings with markers for neuronal cells, astrocytes, and macrophages/microglia in 10week-old WT and βB1-CTGF mice. These stainings show that, most likely, astrocytes and/or macrophages/microglia are the main source of local complement in our model. After lightinduced damage and in the aging retina, microglia were identified to express the complement component C3 (Rutar et al., 2011(Rutar et al., , 2014. In an Alzheimer's disease mouse model, the authors showed that under physiological conditions, C3 is expressed primarily by astrocytes (Lian et al., 2016). Another study revealed a direct involvement of Mueller cells in the transcript expression of the retinal complement components, and it is likely that in the retina neurons and glia cells orchestrate complementmediated maturation (Tenner et al., 2018;Pauly et al., 2019). Nonetheless, the origin and sources of the complement proteins in the βB1-CTGF model should be explored in more detail in further experiments.
As mentioned, the complement system can be initialized through three different pathways. In the βB1-CTGF mouse model, we observed an activation of all three pathways throughout the study. Especially C1q, as a member of the classical route, was upregulated at all investigated ages, namely in 5-, 10-, and 15-week-old βB1-CTGF mice. Activation of the classical pathway was also previously described in other glaucoma models. For example, Stasi et al. detected that a C1q upregulation in the retina of DBA/2J mice preceded RGC death (Stasi et al., 2006). After injection of microbeads in mice, which leads to OHT, elevated levels of C1q mRNA were detected (Krishnan et al., 2019). Also, in rats with OHT, elevated C1q levels were noted (Kuehn et al., 2006). Additionally, more C1q was found in laser-induced glaucomatous monkey eyes (Stasi et al., 2006). Furthermore, Howell et al. (2011) showed that mice with a C1q mutation were protected from glaucomatous damage. But not only in animal models, also in human glaucoma donor eyes, C1q depositions were observed (Stasi et al., 2006). Hence, the classical pathway plays an important role in complement system activation during glaucomatous damage.
Interestingly, we also observed an upregulation of the lectin pathway component MASP2 via RT-qPCR and immunohistology. Although the lectin pathway recognizes carbohydrate structures mostly on bacteria, viruses, fungi, and parasites, it is also able to bind to similar structures on (C) The MASP2 + area was not altered in transgenic mice compared to WT at 5 and 10 weeks. At 15 weeks, a significant increase of MASP2 + area was revealed in βB1-CTGF animals (p = 0.04). Values for RT-qPCR are median ± quartile ± minimum/maximum (n = 5-7/group). The dotted line in panel (A) represents the relative expression of the WT group. Values for immunohistology are mean ± SEM (n = 7-9/group). GCL, ganglion cell layer; IPL, inner plexiform layer. Scale bar: 20 µm. *p < 0.05. apoptotic or necrotic cells (Neth et al., 2000;Turner, 2003;Stuart et al., 2005). This was previously also noted in the EAG model, where IOP is not increased. Here, an activation of the lectin pathway was observed (Reinehr et al., , 2018b. In addition, increased levels of several complement components, including MASP2 were found in glaucoma patients (Tezel et al., 2010). Furthermore, in sera of POAG patients, higher levels of the mannose binding lectin 2 were detected (Dursun et al., 2012).
The alternative pathway of the complement system is spontaneously activated and often associated with processes occurring in age-related macular degeneration (Cipriani et al., 2020). Regarding glaucoma, little is known about the contribution of this pathway. In our βB1-CTGF mice, more Factor B + cells were counted in the GCL at 5 weeks, while enhanced mRNA levels of Cfb and Cfd were observed at 15 weeks. Recently, Fernandez-Vega Cueto et al. (2020) investigated several immune components in sera of DBA2/J mice and their ability as biomarker. They noted that a five protein-panel, amongst others this included complement factor H, as another component of the alternative pathway, predicted the transition to glaucoma in about 78% of these FIGURE 6 | Upregulated cytokine expression in βB1-CTGF retinae. (A) RT-qPCR analyses showed an upregulation of Infg mRNA levels in βB1-CTGF retinae at 5 (p = 0.009), but not at 10 and 15 weeks of age. (B) The mRNA expression levels of Cxcl1 were not altered in 5-week-old transgenic mice. At 10 and 15 weeks, significantly higher mRNA levels of Cxcl1 were noted in βB1-CTGF mice (10 weeks: p = 0.01; 15 weeks: p = 0.04). (C) RT-qPCR analyses revealed no changes in the mRNA expression levels of Cxcl2 in 5-week-old transgenic animals. A significant upregulation of Cxcl2 mRNA levels was noted in βB1-CTGF mice at 10 (p = 0.03) and 15 weeks (p = 0.04). (D) Cxcl10 mRNA levels were significantly upregulated in βB1-CTGF retinae at 5 (p = 0.03), 10 (p = 0.02), and 15 weeks (p = 0.002). (E) Representative pictures of retinal cross-sections labeled with an antibody against IP-10 (CXCL-10; red) and RBPMS (RGCs; green), while DAPI counterstained cell nuclei (blue). IP-10 + cells were predominantly observed in the ganglion cell layer and inner nuclear layer of βB1-CTGF mice. Values are median ± quartile ± minimum/maximum (n = 5-6/group). The dotted line in panels (A-D) represents the relative expression of the WT group. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer. Scale bar: 20 µm. *p < 0.05, **p < 0.01. animals (Fernandez-Vega Cueto et al., 2020). Also, in the IOPindependent EAG model an early upregulation of Cfb mRNA levels was observed (Reinehr et al., 2018b).
It is notable to mention the discrepancy of some results regarding immunohistological and RT-qPCR findings. For example, the results of the alternative and lectin complement pathway seem to differ between the two methods. Immunohistological stainings, particularly in the retina, have the advantage to show the distribution of different cell types within the retinal layers. Further, with this method we were able to count complement positive cells only in the GCL since this is the most affected one during glaucoma. In contrary, for RT-qPCR analyses, RNA and cDNA syntheses were performed using the whole retina. This could explain the discrepancy in the results. In addition, the inconsistency could be explained with different posttranscriptional and translational regulations. Nonetheless, using both methods enabled us to determine the complement response in βB1-CTGF mice quite well.
Overall, these results reveal that the complement system is active in early stages in βB1-CTGF mice and especially the classical pathway plays a role in its initiation. Therefore, therapeutic approaches should be focused on components like C5 to inhibit the terminal pathway, as shown in previous studies (Howell et al., 2013;Reinehr et al., 2019a;Gassel et al., 2021).
Complement activation not only harms cells directly, but its activation is also engaged in the regulation of an inflammatory reaction (Markiewski and Lambris, 2007). C5a, for example, can act directly on neutrophils and monocytes to increase their ability to adhere to vessel walls, their migration toward sites of antigen deposition as well as to ingest particles (Murphy et al., 2008). In osteoblasts, C5a induced migration and expression of IL-8 (Ignatius et al., 2011). Furthermore, C5aR1, in interaction with toll-like-receptor 2, resulted in an upregulation of CXCL-10 (Modinger et al., 2018). We could determine that the mRNA levels of the examined C5a receptors, C5ar and C5ar1, were significantly upregulated in 10-and 15-week-old transgenic animals. An upregulation of C5ar was also noted in mouse retinae 2 days after light damage and a C5aR knockout was able to diminish microglia cells (Song et al., 2017). Additionally, Zhang et al. showed that C3aR/C5aR double knockout mice developed less severe uveitis, which supports the role of these receptors in retinal inflammation (Zhang et al., 2016). Further, we found upregulated Cxlc10 mRNA levels in βB1-CTGF retinae at 5, 10, and 15 weeks of age, while an elevation of Cxcl1 and Cxcl2 (IL-8) mRNA levels was noted at 10 and 15 weeks. At 5 weeks, an upregulation of Infg was also demonstrated in transgenic mice. iPS-retinal pigment epithelium (RPE) cells can express complement factors, such as C3, C5, and MAC, especially after INF-γ exposure (Sugita et al., 2018). RPE cells with activated T-cells express INF-γ associated chemokines, such as CXCL-10 (Sugita et al., 2006). Under the influence of IFN-γ, CXCL-10 is secreted by several cell types including endothelial cells, fibroblasts, keratinocytes, thyrocytes, or preadipocytes (Antonelli et al., 2014). It seems likely that the upregulation of these cytokines correlates with the activation of the complement system and contribute to cell death in transgenic βB1-CTGF mice.

CONCLUSION
Cell death in βB1-CTGF mice seems to be associated with an activation of the complement system, especially through the classical pathway and corresponding cytokine response (Figure 7). These results further support the hypothesis that an altered immune response plays a crucial role in glaucomatous neurodegeneration.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by the Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen. and analyzed the data. RF and HD revised the manuscript. SJ conceived the study and revised the manuscript. All authors read and approved the final manuscript.

FUNDING
This study was in part supported by the Deutsche Forschungsgemeinschaft (Germany, RE-4543/1-1), FoRUM (Ruhr-University Bochum, Germany), and the Ernst and Berta Grimmke foundation (Germany). We acknowledge the support by the Open Access Publication Funds of the Ruhr-University Bochum.