Neuroretinal-Derived Caveolin-1 Promotes Endotoxin-Induced Inflammation in the Murine Retina

bioRxiv preprint doi: https://doi.org/10.1101/2020.01.08.899377; this version posted January 10, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC-ND 4.0 International license. Neuroretinal‐Cav1 promotes retinal inflammation

Neuroretinal-derived caveolin-1 promotes endotoxin-induced inflammation in the murine retina

Jami M. Gurley1, Grzegorz Gmyrek1, Mark E. McClellan1, Stefanie M. Hauck2, Mikhail G. Dozmorov3, Jonathan D. Wren4, Daniel J. J. Carr1,5, and Michael H. Elliott1*

1 Department of Ophthalmology/Dean McGee Eye Institute, University of Oklahoma Health Sciences Center (OUHSC)

2 Research Unit Protein Science, Helmholtz Zentrum München, German Research Center for Environmental Health (GmbH)

3 Department of Biostatistics, Virginia Commonwealth University (VCU)

4 Arthritis and Clinical Immunology Research Program, Division of Genomics and Data Sciences, Oklahoma Medical Research Foundation (OMRF)

5 Department of Microbiology and Immunology, University of Oklahoma Health Sciences Center (OUHSC)

Running title: Neuroretinal-Cav1 promotes retinal inflammation

*To whom correspondence should be addressed: Michael H. Elliott: Department of Ophthalmology/Dean McGee Eye Institute, University of Oklahoma Health Sciences Center, Oklahoma City OK 73104; [email protected] ; Tel.(405) 271-8001, ext.30024; Fax.(405) 271-8128.

Keywords: caveolin-1; caveolin; caveolae; inflammation; neural retina; retina; retinal degeneration; immune response

ABSTRACT model to deplete Cav1 specifically in the neural Chronic ocular inflammation is associated retinal (NR) compartment in order to clarify the with many retinal degenerative diseases that result role of neural retinal-specific Cav1 (NR-Cav1) in in vision loss. The immune-privileged the retinal immune response to intravitreal LPS environment and complex organization of retinal (lipopolysaccharide) challenge. Our data support tissue allow for the retina’s essential role in visual that neural retinal-derived Cav1 promotes retinal function processes, yet confound inquiries into tissue inflammation as Chx10-mediated Cav1 cell-specific inflammatory effects that lead to depletion was sufficient to suppress both retinal retinal dysfunction and degeneration. Caveolin-1 cytokine production and immune cell infiltration (Cav1) is an integral membrane protein expressed following inflammatory stimulation. Additionally, in many retinal cell populations and has been we identify Traf3 (tumor necrosis factor (TNF) implicated in retinal immune regulation. However, receptor-associated factor 3) as a highly expressed the direction (i.e., promotion or inhibition) in potential immune modulator in retinal tissue that is which Cav1 regulates inflammatory processes in upregulated with NR-Cav1 depletion. the retina (as well as in other tissues) remains Furthermore, this study highlights the importance unclear. Previously, we showed that global-Cav1 for understanding the role of Cav1 (and other depletion in the retina paradoxically resulted in proteins) in cell-specific contexts. reduced retinal inflammatory cytokine production with concurrent elevated retinal immune cell INTRODUCTION infiltration. We hypothesized that our previous Chronic ocular inflammation promotes the results could be explained by cell-specific Cav1 development and progression of retinal functions in the retina. Here, we utilized our degenerative diseases that lead to vision loss. Chx10 (visual system homeobox 2)-Cre knockout Many of these retinal degenerative diseases, such

as age-related macular degeneration, diabetic inflammatory stimulation via caveolar retinopathy, and glaucoma, are major causes of sequestration, which prevents receptor complex blindness and are strongly linked to dysregulated formation and downstream signaling (10). immune responses that damage retinal tissue and In the retina, Cav1 affects important visual decrease visual function (1). processes such as phagolysosomal digestion of The retina, which is located in the back of photoreceptor outer segments by the retinal the eye and is responsible for processing light in pigment epithelium (RPE) (16), maintenance of order to facilitate visual function, is a complex and retinal vascular blood-retinal-barrier integrity (17- delicate tissue that is highly susceptible to 19), and retinal neovascularization (20). We and inflammation-mediated damage (1,2). As it is an others have also implicated Cav1 function immune-privileged tissue, the retina can elicit an specifically in retinal inflammation and immune immune response when necessary, but can also response modulation (21-23). However, the engage immune-suppressive mechanisms in order specific role for Cav1 in retinal immunity is as to control inflammatory responses that could lead perplexing as it has proven to be in extraocular to collateral damage (3,4). Under retinal tissue immune responses, and has only begun to be degenerative disease conditions, uncontrolled investigated (21). Previously, we showed that inflammatory activation and/or dysregulation of global-Cav1 deletion resulted in suppression of ocular immune suppression are key contributors endotoxin-mediated retinal inflammatory cytokine toward the development and progression of retinal production, which suggested that Cav1 promotes disease and vision loss (1). Unfortunately, the retinal inflammatory activation (21). Surprisingly, leading treatment option for dysregulated retinal Cav1 deletion enhanced retinal immune cell inflammation is localized corticosteroid infiltration, which paradoxically implied that Cav1 administration, which increases the risk for ocular function suppresses the immune response. conditions that exacerbate vision loss (e.g., Because Cav1 is expressed throughout retinal cataracts, glaucoma). Thus, understanding how to tissue in several retinal cell populations, we control inflammatory responses in the context of reasoned that these paradoxical phenotypes could the immune-privileged retinal environment is be explained by differential cell-specific Cav1 important for developing novel therapeutic functions in the retina. interventions that will preserve retinal and visual The aim of this study was to identify the function. Cav1-expressing retinal compartment responsible Several studies have implicated Cav1 as for promotion of endotoxin-induced inflammation. an immune regulator in lung, brain, and myeloid- Here, we show that neural retinal-derived Cav1 derived immune cells (5-7). In most cell types, depletion is sufficient to suppress inflammatory Cav1 is integrated into specialized plasma processes in the retina, which highlights the membrane domains called caveolae, which are importance of the neural retinal compartment in involved in a variety of cellular processes immune activation. This study (in conjunction including clathrin-independent endocytosis, with ours and other previous work) contributes to endothelial transcytosis, lipid transport, vascular our understanding of complex cell-specific Cav1 tone and contraction, fibrosis, cell proliferation, functions. Additionally, we identify Traf3 as a monocyte/macrophage differentiation, and potential retinal immune modulator that deserves signaling pathway regulation (6,8-15). The role of future study in regards to our search for novel Cav1 and caveolae in immune regulation is therapeutic interventions aimed at suppressing the complex. Collectively, existing studies suggest retinal immune response under conditions of that Cav1-dependent inflammatory promotion or chronic ocular inflammation. inhibition can occur via various mechanisms that depend on cell-specific contexts. For example, in Results the lung, Cav1 has been shown to promote Neural retinal depletion of CAV1 protein in inflammation by inhibiting eNOS-dependent retinal tissue downregulation of proinflammatory cytokine Our previously published results presented production (6); whereas, in macrophages, Cav1 seemingly contradictory data where global-Cav1 inhibits TLR4 (Toll-like receptor 4)-mediated depletion resulted in suppression of retinal

inflammatory cytokine production with a Müller glia have a unique morphology where their concurrent elevation in retinal immune cell cellular processes span the retina from inner infiltration (21). Because we hypothesized that this limiting membrane at the edge of the RGC layer to paradoxical result was due to divergent cell- the outer limiting membrane at the edge of the specific Cav1 functions in different retinal cell outer nuclear layer (ONL); whereas, the retinal populations, we wished to separate potential cell- vasculature appears as discrete vessel cross- specific Cav1 effects using conditional knockout sections in the plexiform synaptic layers of the animals. Thus, we utilized our previously- retina (Fig. 2, arrowheads). Importantly, in all developed Chx10-Cre/Cav1flox/flox neural retinal sections analyzed, CAV1 expression in the retinal knockout animals (NR-Cav1 KO). This model vasculature was not affected by NR-Cav1 expresses Cre recombinase in neuroretinal depletion (Fig. 2, arrowheads; Fig. S1). However, progenitors resulting in Cav1 gene recombination the characteristic staining of CAV1 in the Müller in retinal neurons and Müller glia but not other glia was largely ablated with NR-Cav1 KO, retinal glia (i.e., astrocytes or microglia) (Fig. 1A) though we did observe modest mosaicism in (24). To validate specific targeting of retinal Müller glia staining (Fig. 2, arrows) as previously tissue, we first assessed CAV1 protein expression reported (34). Alternatively, RGCs of the inner (via Western blot) in various tissues collected retina (illustrated by Brn3a immunoreactivity) do from NR-Cav1 wild type (WT) and KO animals not express high levels of CAV1 (Fig. S1A). (Fig. 1B-C). Whole retinal tissue lysates from NR- Likewise, BPCs located in the inner nuclear layer Cav1 KO animals displayed a ~76% depletion of (INL) also do not express detectable levels of CAV1 protein compared to their WT counterparts. CAV1 protein (Fig. S1B). Finally, we also did not Residual CAV1 protein expression is likely detect CAV1 immunoreactivity in the ONL primarily due to the remaining CAV1 protein region, which houses retinal photoreceptors (Fig. expression in non-targeted retinal cell populations S1C) (37). Thus, our data support that CAV1 (e.g., retinal vascular cells) with some modest protein is most highly expressed in the Müller glia contribution from mosaic recombination in the and blood vasculature of the mouse retina. neural retina (Fig. 2) (34). There was no genotype- Importantly, Chx10-mediated Cre-recombination dependent difference in CAV1 protein expression of Cav1 significantly reduces Müller glia-specific in the other tissues we assessed (brain, lung, heart, CAV1 protein while retaining CAV1 protein spleen, or liver). Thus, our data show that NR- expression in retinal tissue vasculature. This specific deletion targets a significant percentage of supports that the majority of CAV1 protein total retinal Cav1 expression. depletion observed in our retinal lysates (Fig. 1) is due to Müller-derived Cav1 recombination. Thus, Neural retinal Cav1 depletion primarily targets our NR-Cav1 KO provides an ideal model to retinal Müller glia CAV1 protein assess the contribution of neural retinal Cav1 on To ensure that our model specifically the retinal immune response. targeted Cav1 in the neural retinal compartment, we assessed CAV1 protein expression in retinal Neuroretinal-Cav1 promotes retinal endotoxin- tissue sections by immunohistochemistry. In order mediated immune response to deduce retinal cell populations targeted by We have previously shown that global Chx10-mediated Cav1 depletion, we co-stained Cav1 depletion prevented endotoxin-induced retinal tissue sections with cell-specific markers as elevation of inflammatory retinal cytokines, which follows: Müller glia, glutamine synthetase (GS); suggested that Cav1 functions to promote retinal ganglion cells (RGCs), Brain-Specific proinflammatory cytokine production. Because we Homeobox/POU Domain Protein 3A (Brn3a); suspected that neural retinal-specific Cav1 bipolar cells (BPCs), Chx10; and rod expression was responsible for this effect, we used photoreceptor cells (PRCs), rhodopsin (1D4) (Fig. our NR-Cav1 KO model to assess whole retinal 2; Fig. S1). We observed the highest levels of cytokine levels following ocular immune CAV1 protein expression in Müller glia cells and challenge via intravitreal LPS injection (Fig. 3). retinal blood vessels, which agrees with previous Immune challenge in mice resulted in a robust studies (17,22,24,35,36). In retinal tissue sections, proinflammatory cytokine response in WT animals

(Fig. 3, LPS effect = *). Similar to our previous tissue. To further validate that this observation was data collected from global Cav1-KO retinas (21), due to neural retinal compartment-specific effects, whole retinal tissue from NR-Cav1 KO animals we similarly assessed immune infiltration in exhibited lower inflammatory cytokine endothelium-specific Cav1 KO animals (Tie2-Cre concentrations (pg/mL) than their WT Cav1 KO model; Endo-Cav1 KO). Endo-Cav1 KO counterparts following LPS challenge (Fig. 3: IL-6 mice did not exhibit the suppressive effect on (interleukin 6): 28.3%, 165.5 NR-KO vs 585.4 infiltration, which further supports that neural NR-WT; CXCL1/KC (C-X-C motif chemokine retinal Cav1, specifically, plays a crucial role in ligand 1): 18.1%, 792.0 NR-KO vs 4374.0 NR- retinal immune activation (Fig. 4B: total WT; CCL2/MCP-1 (monocyte chemoattractant leukocytes: 2,741 NR-KO vs 2,435 NR-WT; protein 1): 22.2% 185.9 NR-KO vs 838.4 NR-WT; PMNs: 2,164 NR-KO vs 1,799 NR-WT;INF. and IL-1β (interleukin 1 beta): 71.3%, 15.9 NR- MNCs: 163 NR-KO vs 184 NR-WT; and MΦ: 445 KO vs 22.3 NR-WT). These data suggest that NR-KO vs 441 NR-WT). neural retina-derived Cav1 function is sufficient to promote LPS-induced retinal immune activation. Immunomodulatory TRAF3 protein is highly Furthermore, our findings here support that our expressed in retinal tissue and is upregulated in previous observation of CAV1-dependent cytokine NR-Cav1 KO animals that exhibit reduced production in global Cav1 KO animals was likely immune responses due to CAV1 function in the neural retinal While this work is focused on clarifying compartment (21). the role of NR-specific Cav1 function, we Previously, in global Cav1-KO animals, ultimately hope to identify novel potential we made the paradoxical observation that retinal regulators of retinal inflammation that may be immune cell infiltration following LPS stimulation targeted in order to prevent retinal damage. Here, was elevated compared to WT mice, which our work suggests that NR-Cav1 promotes retinal supported that Cav1 functions to suppress the inflammation and that an active retinal immune response. As Cav1 plays a role in immunosuppressive mechanism is present in the several cell processes and is expressed in multiple absence of NR-Cav1 expression. Thus, we have retinal cell populations (e.g., vascular begun to investigate potential immune suppressors endothelium), we reasoned that this finding could that are upregulated in response to Cav1 depletion. be explained by differential cell-specific Cav1 CAV1 protein is primarily localized to the functions in the neural retina as opposed to Cav1 cell membrane where it participates in formation function in the endothelia and/or systemic immune and stabilization of specialized lipid raft domains, system. We, therefore, hypothesized that we which can facilitate regulatory effects on would observe reduced immune cell infiltration membrane receptor localization and activation following intraocular LPS challenge when Cav1 (38). Thus, because we suspected that Cav1 depletion is specifically targeted to the neural regulates immune pathways at the level of retina. As suspected, using flow cytometry, we membrane receptor complex formation and signal found that NR-Cav1 KO animals exhibited lower activation, we decided to narrow our search for levels of retinal immune cell infiltrate than novel immune regulators to those that interact with controls, following LPS challenge (Fig. 4A: total CAV1 protein at the cell surface. Thus, we leukocytes: 23.2%, 20,354 NR-KO vs 4,713 NR- isolated retinal membrane fractions (CAV1- WT; PMNs (polymorphonuclear leukocytes): containing membrane fractions) from whole 22.1%, 4325 NR-KO vs 19,596 NR-WT; INF. retinal tissue and performed mass spectrometry MNCs (inflammatory monocytes): 52.9% 202 NR- and quantitative proteomics analysis on KO vs 382 NR-WT; and MΦ (macrophages): membrane-associated proteins using previously 58.7%, 311 NR-KO vs 530 NR-WT). These data described methods (22). Our data show significant further suggest that neural retinal-specific Cav1 downregulation of 108 genes (including Cav1) and functions to promote the retinal immune response upregulation of 177 genes in retinal membrane to LPS, and that NR-Cav1 depletion is sufficient fractions in response to NR-Cav1 deletion. Data to prevent both proinflammatory cytokine for the top 70 differentially expressed proteins are production and immune cell infiltration into retinal represented in Fig. 5 (blue = downregulated; red =

upregulated). Retinas from our NR-Cav1 KO phosphatase non-receptor type 11), which TRAF3 animals exhibited 70% reduction in CAV1 peptide is known to inhibit. Many of these shared PPIs abundance (Fig. 5), which is consistent with the (e.g., EGF/ epidermal growth factor, 76% reduction observed by Western blotting of SRC/neuronal proto-oncogene tyrosine-protein whole retinal tissue lysates (Fig. 1). Collectively, kinase, BCAR1/breast cancer anti-estrogen our data suggest that the majority of retinal CAV1 resistance protein 1) are also associated with protein resides within the plasma membrane of anoikis (programmed cell death upon detachment neural retinal cell populations (i.e., primarily the from the extracellular matrix). Indeed, CAV1 has Müller glia; Figs. 1,2,5). After validating that NR- been reported to increase resistance to anoikis, Cav1 KO results in CAV1 protein reduction in which may be related to Cav1’s role in retinal membrane fractions, we then used the neuroprotection (45,46). But because the changed ShinyGO gene ontology enrichment analysis proteins were not significantly enriched in immune online software to identify which of our 177 functions or pathways, we cannot rule out the upregulated genes are functionally associated with possibility that the immune effects of Cav1 KO immune and stress response pathways (Table 1). may be indirect and due to alterations in the Interestingly, we identified upregulation structure of the plasma membrane. This may, in of membrane TRAF3 in NR-Cav1 retinas, which fact, help reconcile why some studies observe a we validated with retinal membrane fractionation pro- or anti-inflammatory effect, as a change in and TRAF3 Western blot analysis (Fig. 5; Fig. membrane dynamics would likely affect both. S2). Traf3 is a known inflammatory regulator in non-retinal tissues that associates with immune Discussion receptor complexes at the cell membrane in order Previous work from our lab showed that to modulate downstream signaling (39-42). Cav1 depletion in global-Cav1 KO animals Importantly, Traf3 function in the retina has not prevented inflammatory cytokine production in the yet been investigated, though existing proteomics retina, which suggested that Cav1 promotes data sets suggest that Traf3 is highly expressed in inflammation in retinal tissue (21). Our current the retina compared to other tissues (Genecards). results determine that NR- Cav1 depletion is This, together with our current data, suggests that sufficient to produce this response. Likewise, this Traf3 likely functions to regulate immune work shows that NR-Cav1 depletion also prevents responses in the retina, which we intend to immune cell infiltration. However, our previous examine in future studies. results illustrate that global depletion of Cav1 To better understand how CAV1 might actually exacerbates retinal immune cell interact with TRAF3 to modulate retinal immune infiltration. Together, this suggests that ablation of responses, we used the STRING v10 database to a non-neural retinal Cav1 pool is responsible for obtain protein-protein interaction (PPI) the enhanced presence of immune cells in the information for the most significantly changed retinal tissue of global-Cav1 KO animals. We proteins (adjusted p-value <=0.10) within our data initially suspected that the elevated immune cell set (32). Fig. 6 shows that CAV1 is known to infiltration in the global-Cav1 KO animals could share at least 19 PPIs with 7 proteins and be explained by enhanced vascular permeability illustrates how loss of Cav1 might impact the due to endothelial-Cav1 depletion. Thus, we also activity and/or upregulation of 4 proteins assessed infiltration in our endothelial (Tie2)-Cav1 (including TRAF3) (Fig. 6; Fig. S3). At least 3 of KO animals. Interestingly, there was no difference the 4 CAV1-TRAF3 shared interacting proteins in infiltration values between Endo-Cav1 KO have been associated with TNF-alpha-induced cell animals compared to WT controls. Thus, death (i.e., RIPK1/Receptor-interacting endothelial-derived Cav1 is not responsible for the serine/threonine-protein kinase 1, TRAF2/ tumor effect of elevated infiltration observed in the necrosis factor (TNF) receptor-associated factor 2, global-Cav1 KOs. We believe, rather, that immune TRADD/ tumor necrosis factor receptor type 1- cell-specific effects of Cav1 depletion in global- associated DEATH domain protein) (43,44). The Cav1 KO animals is responsible. Previous work bulk of the PPIs shared with CAV1 are among from our lab has shown that global-Cav1 KO TRAF3 and PTPN11 (tyrosine-protein animals have more CD45-positive cells (i.e.,

leukocytes) within their retinal vasculature inflammatory activation observed with NR-Cav1 compared to WT animals (21). Thus, we suspect depletion is due to altered receptor-membrane that the increased presence of circulating immune dynamics that affect immune receptor complex cells in global-Cav1 KO animals results in greater formation and propagation of downstream accessibility to the retinal vessels under basal inflammatory signaling pathways. Thus, we have conditions and primes the retina to uptake more begun to investigate novel potential inflammatory immune cells following an immune response that regulators that might associate with the cell induces vessel permeability. Alternatively, the membrane to regulate receptor activation in the enhanced infiltration phenotype observed in the absence of Cav1. Interestingly, we identified global-Cav1 KO could be explained by loss of elevated TRAF3 protein levels in retinal RPE-specific Cav1-dependent functions, which membrane fractions from our NR-Cav1 KO are also not targeted in our NR-Cav1 KO animals. animals. While Traf3 is known to play an Our data support that the majority of important immune regulatory role in other tissues, neural retinal (and total retinal) Cav1 protein the function in the retina is unknown. resides within the Müller glial cell population, Additionally, Traf3 has been shown to directly which is in agreement with ours and other interact with immune receptors, including CD40 previous work (24). In addition to their important and TLR4, which are both expressed in Müller structural and neurosupportive roles in the retina glial cells (52). Thus, we suggest that Traf3 may (i.e., metabolic homeostasis, neurotransmitter play a previously-unappreciated role in retinal recycling, and intercellular communication), inflammatory modulation. Our future studies Müller glia also play an unappreciated role in include investigation of Traf3 function in the retinal immunity (47). Müller glia have been retina, as well as further examination into the shown to express innate immune receptors and to mechanism(s) of Cav1-dependent inflammatory produce inflammatory and neuroprotective activation in Müller glia. cytokines (48,49). Intriguingly, Cav1 has been shown to be important for secretion of cytokines Experimental procedures such as IL-6 (50,51). Furthermore, previous data Animals-NR-specific Cav1 KO mice from our lab suggests a Cav1-dependent role in (Chx10-Cre; Cav1flox/flox) were generated using IL-6 family cytokine production that is upstream Cre-Lox technology (24). Animals carrying of neuroprotective signaling (via STAT3 pathway Chx10-promoter-driven Cre recombinase (stock#: activation) (24). These studies are consistent with 005105, The Jackson Laboratory, Bar Harbor, our observed suppression of LPS-induced cytokine ME) were bred with animals carrying the Cav1 production in NR-Cav1 KO mice and support a gene flanked by Lox P sites located upstream and possible role for Müller-derived inflammatory downstream of exon 2 (25). Endothelial-specific cytokine production and secretion following Cav1 knock-out mice (Tie2-Cre; Cav1flox/flox) were ocular insult. However, further studies are needed generated similarly, using endothelial cell-specific to fully understand the contribution of Müller glia- recombination (B6.Cg-Tg (Tek-cre) 1Ywa/J; derived immune responses and the role of Müller stock#: 008863, The Jackson Laboratory, Bar glia in maintenance of the retinal immune- Harbor, ME) (26). Cre-negative littermates were privileged environment. used as controls. Mice were screened for rd1 and Many studies have been published on rd8 mutations prior to this study. All procedures Cav1 since its identification in the late 1980s. were approved by the University of Oklahoma However, the precise role of Cav1 in immune Health Sciences Center (OUHSC) Institutional regulation has remained unclear. Furthermore, Animal Care and Use Committee (IACUC) and whether Cav1 generally functions to promote or comply with the Association for Research in suppress retinal immune activation is equally as Vision and Ophthalmology (ARVO) Statement for perplexing. Our study suggests that neural retinal the Use of Animals in Ophthalmic and Visual Cav1, specifically, promotes retinal inflammation. Research. Because Cav1 is known to stabilize cell surface Whole retinal tissue lysis- Total retinal receptors in specialized lipid raft membrane protein was obtained via whole retinal tissue lysis domains (38), it is conceivable that the blunted by brief sonication in 2X lysis buffer (120 mM

octyl glucoside, 2% Triton X-100, 20 mM Tris- marker, 1:100, Santa Cruz #sc-31984); sheep HCl, pH 7.4, 200 mM NaCl, 1 mM EDTA) polyclonal anti-Chx10 (for bipolar cell marker; containing a protease inhibitor cocktail (Roche). 1:100, Exalpha #X1179P); and anti- Lysates were cleared by centrifugation at 13K rhodopsin/1D4 (for photoreceptor marker; 1:100, rpm, at 4°C. Protein concentration was determined gifted from R. Molday, Univ. of British via BCA assay (ThermoSci, Cat# 23225) and Columbia). Immunoreactivity was detected with samples were diluted in 6X Laemmli buffer. Alexa Fluor-conjugated secondary antibodies Western blotting- Retinas were (1:500, Life Technologies, Grand Island, NY) and homogenized in buffer containing 100 mM sodium nuclei were stained with DAPI (4,6-diamidino-2- carbonate and 1 mM EDTA using an plastic pellet phenylindole). Images of immunostained retinal pestle and brief pulse sonication on ice. tissue sections were acquired on an Olympus Membranes were isolated by centrifugation at FV1200 laser scanning confocal system using 16,000 x g in a benchtop microcentrifuge and FluoView software (Olympus). Pseudocolors were membrane pellets were lysed in 1X lysis buffer. assigned to images as follows: Glutamine Protein content of lysates was determined by the Synthetase, Brn3a, Chx10, or 1D4 (green); Cav-1 BCA assay with bovine serum albumin as the (red); and nuclei (blue). standard. Protein separation was achieved via Endotoxin-induced uveitis model-As SDS-PAGE and proteins were transferred to previously described, local, intravitreal injection nitrocellulose for immunoblotting with the of 1 μg LPS (Salmonella typhimurium; Sigma) following primary antibody dilutions: rabbit was used to induce ocular inflammation (21,29). monoclonal anti–Caveolin-1 (1:1000; Cell Mice were anesthetized via intraperitoneal Signaling #3267), mouse monoclonal anti–Actin injection of ketamine (100 mg/kg)/xylazine (10 (1:1000; ThermoFisher #MA1-744), rabbit mg/kg). Each eye was then injected with either 1 polyclonal anti-TRAF3 (TNF Receptor-Associated μL LPS diluted in sterile PBS vehicle or PBS Factor 3; 1:500; Novus Biologicals, #NBP-88639), alone using a 10 µL glass syringe (Hamilton). and rabbit polyclonal anti-Gαt (Transducin; Cytokine/chemokine quantification Assay- 1:4000; Santa Cruz). Immunoreactivity was Retinas from euthanized animals were collected 24 detected using species-appropriate horseradish h after LPS or PBS injection. Retinal tissue peroxidase (HRP)-conjugated secondary dissection was performed using a stereo dissecting antibodies (1:5000; GE Healthcare, Cleveland, microscope (Zeiss Stemi DV4; Carl Zeiss OH, USA). Western blots were imaged via HRP Microscopy, Jena, Germany). Retinas were chemiluminescent detection (Azure Biosystems, processed and cytokines protein levels were Inc.; Dublin, CA) and densitometric analyses were measured by BioPlex suspension array (Bio-Rad performed using Image Studio Lite software (LI- Life Science, Hercules, CA, USA) as previously COR Biosciences). described (21). Immunohistochemistry & confocal Flow cytometry-Flow cytometry was microscopy-Whole eye globes from euthanized performed on retinal samples as described adult mice were fixed in Prefer fixative (Anatech, previously for neural tissues with minor Ltd., Battlefield, MI), embedded in paraffin, and modification (21,30). Twenty-four hours after LPS cut into 5-μm sections. Sections were or PBS injection, mice were perfused with PBS, deparaffinized, permeabilized with 1% Triton X- and single-cell suspensions were obtained from 100 in PBS, then incubated with blocking solution retinas by digestion of minced tissue with liberase (10% normal horse serum, 0.1% Triton X-100, in TL in RPMI 1640 at 370C for 30 minutes. Next, PBS) for one hour prior to immunohistochemistry, the cell suspensions were passed through a 40-µm which was performed as previously described nylon cell strainer (Thermo Fisher Scientific) (27,28) using the following antibodies: rabbit followed by washing with staining buffer (SB; monoclonal anti-Caveolin-1 (1:100, Cell Signaling PBS supplemented with 2% FBS, and 2 mM #3267); mouse monoclonal anti-Glutamine EDTA). Samples were blocked with anti-mouse Synthetase (for Müller glia-specific marker; CD16/32 (Fc-block, Invitrogen, San Jose, CA, 1:5000 Millipore #MAB302, clone GS-6); goat USA) for 10 minutes at 40C followed by staining polyclonal anti-Brn3a (for retinal ganglion cell with antibody cocktail containing Pacific-Blue-

conjugated anti-mouse CD45 (clone 30-F11; spectra obtained by LC-MSMS on an Biolegend, San Diego, CA, USA), PE-conjugated Ultimate3000 nano HPLC system (Dionex, anti-mouse Ly-6G (Gr-1) (clone RB6-8C5; Sunnyvale, CA) online coupled to a LTQ Biolgend), and APC-conjugated anti-mouse F4/80 OrbitrapXL mass spectrometer (Thermo Fisher (clone BM8; Biolegend). Samples were incubated Scientific) were transformed to peak lists using 20 minutes at 40C and then washed with SB Progenesis software, followed by label-free buffer. The data were acquired immediately after quantification based on peak intensities. Total list staining using a MacsQuant flow cytometer of identified peptides and proteins are available in (Miltenyi Biotec, Auburn, CA USA) and analyzed Supporting Information (Supplemental Table 1). with FlowJo software (FlowJo LLC). The events The mass spectrometry data have been deposited were gated by forward and side scatter, as well as to the ProteomeXchange Consortium via the CD45 to distinguish infiltrating immune cells PRIDE (31) partner repository with the dataset (CD45hi). identifier PXD016872 and 10.6019/PXD016872. Mass spectrometry-based proteomics- Reviewer account details: Membrane proteins were prepared from whole Username: [email protected] frozen retinal tissue as described previously (22). Password: VbXalKFX Briefly, retinal tissues were lysed in high salt Bioinformatics-Peptide abundances were buffer (2M NaCl, 10mM HEPES-NaOH, pH 7.4, log2-transformed and analyzed for differential 1mM EDTA, complete protease inhibitors). Pellets abundance using the limma R package. The row- were harvested via high-speed centrifugation and scaled abundances of the top 70 differentially were subsequently homogenized in carbonate expressed retinal proteins affected by NR-Cav1 buffer (0.1M Na2CO3, pH 11.3, 1mM EDTA, depletion were visualized as a heatmap using the complete protease inhibitors). Pellets were pheatmap R package (blue = downregulated; red = collected and re-extracted with carbonate buffer a upregulated). The resulting complete proteomics second time. Pellets were collected and re- data set was analyzed using both pathway and homogenized in urea buffer (4 M urea, 100 mM network analyses. KEGG (Kyoto Encyclopedia of NaCl, 10 mM HEPES/NaOH, PH 7.4, 1 mM Genes and Genomes) analysis and IPA (Ingenuity EDTA, complete protease inhibitors). The Pathway Analysis) were used for pathway resulting membrane fractions were collected via enrichment in order to identify known gene sets centrifugation and were used for tryptic digest as affected by NR-Cav1 depletion. We obtained the described previously (22). Label-free mass protein-protein interaction (PPI) information for spectrometry was used to measure peptide the most significantly changed proteins (adjusted abundances from membrane-enriched fractions of p-value <=0.10) from the STRING v10 database retinal samples from both NR-Cav1 WT and NR- (32). Single PPIs (nodes with only one edge) were Cav1 KO mice. For an in-depth description of the removed to reduce clutter and to focus upon LC-MSMS procedure and detailed label-free shared PPIs within this network, and visualized peptide quantification analysis parameters, please with Cytoscape 3.7 (33). refer to Hauck, et al., 2010. Briefly, peptide

Acknowledgments: Data are available via ProteomeXchange with identifier PXD016872. We thank Linda Boone in the NEI P30-supported Cellular Imaging Core for expert preparation of specimens and tissue sections for immunohistochemistry. We are also grateful to Renee Selleck from the Carr laboratory for preparation of retinal tissues for flow cytometry.

Conflict of interest: The authors declare that they have no conflicts of interest with the contents of this article. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

References 1. Whitcup, S. M., Nussenblatt, R. B., Lightman, S. L., and Hollander, D. A. (2013) Inflammation in retinal disease. Int J Inflam 2013, 724648 2. Murakami, Y., Ishikawa, K., Nakao, S., and Sonoda, K. H. (2019) Innate immune response in retinal homeostasis and inflammatory disorders. Prog Retin Eye Res, 100778 3. Streilein, J. W. (2003) Ocular immune privilege: the eye takes a dim but practical view of immunity and inflammation. J Leukoc Biol 74, 179-185 4. Zhou, R., and Caspi, R. R. (2010) Ocular immune privilege. F1000 Biol Rep 2 5. Chidlow, J. H., Jr., and Sessa, W. C. (2010) Caveolae, caveolins, and cavins: complex control of cellular signalling and inflammation. Cardiovasc Res 86, 219-225 6. Garrean, S., Gao, X. P., Brovkovych, V., Shimizu, J., Zhao, Y. Y., Vogel, S. M., and Malik, A. B. (2006) Caveolin-1 regulates NF-kappaB activation and lung inflammatory response to sepsis induced by lipopolysaccharide. J Immunol 177, 4853-4860 7. Blochet, C., Buscemi, L., Clement, T., Gehri, S., Badaut, J., and Hirt, L. (2018) Involvement of caveolin-1 in neurovascular unit remodeling after stroke: Effects on neovascularization and astrogliosis. J Cereb Blood Flow Metab, 271678X18806893 8. Drab, M., Verkade, P., Elger, M., Kasper, M., Lohn, M., Lauterbach, B., Menne, J., Lindschau, C., Mende, F., Luft, F. C., Schedl, A., Haller, H., and Kurzchalia, T. V. (2001) Loss of caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science 293, 2449-2452 9. Fu, Y., Moore, X. L., Lee, M. K., Fernandez-Rojo, M. A., Parat, M. O., Parton, R. G., Meikle, P. J., Sviridov, D., and Chin-Dusting, J. P. (2012) Caveolin-1 plays a critical role in the differentiation of monocytes into macrophages. Arterioscler Thromb Vasc Biol 32, e117-125 10. Wang, X. M., Kim, H. P., Nakahira, K., Ryter, S. W., and Choi, A. M. (2009) The heme oxygenase-1/carbon monoxide pathway suppresses TLR4 signaling by regulating the interaction of TLR4 with caveolin-1. J Immunol 182, 3809-3818 11. Ghitescu, L., Fixman, A., Simionescu, M., and Simionescu, N. (1986) Specific binding sites for albumin restricted to plasmalemmal vesicles of continuous capillary endothelium: receptor- mediated transcytosis. J Cell Biol 102, 1304-1311 12. Chaudhary, K. R., Cho, W. J., Yang, F., Samokhvalov, V., El-Sikhry, H. E., Daniel, E. E., and Seubert, J. M. (2013) Effect of ischemia reperfusion injury and epoxyeicosatrienoic acids on caveolin expression in mouse myocardium. J Cardiovasc Pharmacol 61, 258-263 13. Fernandez-Rojo, M. A., Gongora, M., Fitzsimmons, R. L., Martel, N., Martin, S. D., Nixon, S. J., Brooks, A. J., Ikonomopoulou, M. P., Martin, S., Lo, H. P., Myers, S. A., Restall, C., Ferguson, C., Pilch, P. F., McGee, S. L., Anderson, R. L., Waters, M. J., Hancock, J. F., Grimmond, S. M., Muscat, G. E., and Parton, R. G. (2013) Caveolin-1 is necessary for hepatic oxidative lipid metabolism: evidence for crosstalk between caveolin-1 and bile acid signaling. Cell Rep 4, 238- 247 14. Milici, A. J., Watrous, N. E., Stukenbrok, H., and Palade, G. E. (1987) Transcytosis of albumin in capillary endothelium. J Cell Biol 105, 2603-2612 15. Zhao, Y. Y., Zhao, Y. D., Mirza, M. K., Huang, J. H., Potula, H. H., Vogel, S. M., Brovkovych, V., Yuan, J. X., Wharton, J., and Malik, A. B. (2009) Persistent eNOS activation secondary to caveolin-1 deficiency induces pulmonary hypertension in mice and humans through PKG nitration. J Clin Invest 119, 2009-2018 16. Sethna, S., Chamakkala, T., Gu, X., Thompson, T. C., Cao, G., Elliott, M. H., and Finnemann, S. C. (2016) Regulation of Phagolysosomal Digestion by Caveolin-1 of the Retinal Pigment Epithelium Is Essential for Vision. J Biol Chem 291, 6494-6506 17. Gu, X., Fliesler, S. J., Zhao, Y. Y., Stallcup, W. B., Cohen, A. W., and Elliott, M. H. (2014) Loss of caveolin-1 causes blood-retinal barrier breakdown, venous enlargement, and mural cell alteration. Am J Pathol 184, 541-555

18. Chow, B. W., and Gu, C. (2017) Gradual Suppression of Transcytosis Governs Functional Blood- Retinal Barrier Formation. Neuron 93, 1325-1333 e1323 19. Klaassen, I., Hughes, J. M., Vogels, I. M., Schalkwijk, C. G., Van Noorden, C. J., and Schlingemann, R. O. (2009) Altered expression of genes related to blood-retina barrier disruption in streptozotocin-induced diabetes. Exp Eye Res 89, 4-15 20. Jiang, Y., Lin, X., Tang, Z., Lee, C., Tian, G., Du, Y., Yin, X., Ren, X., Huang, L., Ye, Z., Chen, W., Zhang, F., Mi, J., Gao, Z., Wang, S., Chen, Q., Xing, L., Wang, B., Cao, Y., Sessa, W. C., Ju, R., Liu, Y., and Li, X. (2017) Critical role of caveolin-1 in ocular neovascularization and multitargeted antiangiogenic effects of cavtratin via JNK. Proc Natl Acad Sci U S A 114, 10737- 10742 21. Li, X., Gu, X., Boyce, T. M., Zheng, M., Reagan, A. M., Qi, H., Mandal, N., Cohen, A. W., Callegan, M. C., Carr, D. J., and Elliott, M. H. (2014) Caveolin-1 increases proinflammatory chemoattractants and blood-retinal barrier breakdown but decreases leukocyte recruitment in inflammation. Invest Ophthalmol Vis Sci 55, 6224-6234 22. Hauck, S. M., Dietter, J., Kramer, R. L., Hofmaier, F., Zipplies, J. K., Amann, B., Feuchtinger, A., Deeg, C. A., and Ueffing, M. (2010) Deciphering membrane-associated molecular processes in target tissue of autoimmune uveitis by label-free quantitative mass spectrometry. Mol Cell Proteomics 9, 2292-2305 23. Cehofski, L. J., Kruse, A., Magnusdottir, S. O., Alsing, A. N., Nielsen, J. E., Kirkeby, S., Honore, B., and Vorum, H. (2018) Dexamethasone intravitreal implant downregulates PDGFR-alpha and upregulates caveolin-1 in experimental branch retinal vein occlusion. Exp Eye Res 171, 174-182 24. Reagan, A., Gu, X., Hauck, S. M., Ash, J. D., Cao, G., Thompson, T. C., and Elliott, M. H. (2016) Retinal Caveolin-1 Modulates Neuroprotective Signaling. Adv Exp Med Biol 854, 411-418 25. Cao, G., Yang, G., Timme, T. L., Saika, T., Truong, L. D., Satoh, T., Goltsov, A., Park, S. H., Men, T., Kusaka, N., Tian, W., Ren, C., Wang, H., Kadmon, D., Cai, W. W., Chinault, A. C., Boone, T. B., Bradley, A., and Thompson, T. C. (2003) Disruption of the caveolin-1 gene impairs renal calcium reabsorption and leads to hypercalciuria and urolithiasis. Am J Pathol 162, 1241- 1248 26. Oliveira, S. D. S., Castellon, M., Chen, J., Bonini, M. G., Gu, X., Elliott, M. H., Machado, R. F., and Minshall, R. D. (2017) Inflammation-induced caveolin-1 and BMPRII depletion promotes endothelial dysfunction and TGF-beta-driven pulmonary vascular remodeling. Am J Physiol Lung Cell Mol Physiol 312, L760-L771 27. Li, W., Liu, H., Zhou, J. S., Cao, J. F., Zhou, X. B., Choi, A. M., Chen, Z. H., and Shen, H. H. (2012) Caveolin-1 inhibits expression of antioxidant enzymes through direct interaction with nuclear erythroid 2 p45-related factor-2 (Nrf2). J Biol Chem 287, 20922-20930 28. Gu, X., Reagan, A., Yen, A., Bhatti, F., Cohen, A. W., and Elliott, M. H. (2014) Spatial and temporal localization of caveolin-1 protein in the developing retina. Adv Exp Med Biol 801, 15-21 29. Rosenbaum, J. T., Woods, A., Kezic, J., Planck, S. R., and Rosenzweig, H. L. (2011) Contrasting ocular effects of local versus systemic endotoxin. Invest Ophthalmol Vis Sci 52, 6472-6477 30. Wuest, T. R., and Carr, D. J. (2008) Dysregulation of CXCR3 signaling due to CXCL10 deficiency impairs the antiviral response to herpes simplex virus 1 infection. J Immunol 181, 7985-7993 31. Perez-Riverol, Y., Csordas, A., Bai, J., Bernal-Llinares, M., Hewapathirana, S., Kundu, D. J., Inuganti, A., Griss, J., Mayer, G., Eisenacher, M., Perez, E., Uszkoreit, J., Pfeuffer, J., Sachsenberg, T., Yilmaz, S., Tiwary, S., Cox, J., Audain, E., Walzer, M., Jarnuczak, A. F., Ternent, T., Brazma, A., and Vizcaino, J. A. (2019) The PRIDE database and related tools and resources in 2019: improving support for quantification data. Nucleic Acids Res 47, D442-D450 32. Szklarczyk, D., Franceschini, A., Wyder, S., Forslund, K., Heller, D., Huerta-Cepas, J., Simonovic, M., Roth, A., Santos, A., Tsafou, K. P., Kuhn, M., Bork, P., Jensen, L. J., and von Mering, C. (2015) STRING v10: protein-protein interaction networks, integrated over the tree of life. Nucleic Acids Res 43, D447-452

33. Shannon, P., Markiel, A., Ozier, O., Baliga, N. S., Wang, J. T., Ramage, D., Amin, N., Schwikowski, B., and Ideker, T. (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res 13, 2498-2504 34. Cepko, C. (2014) Intrinsically different retinal progenitor cells produce specific types of progeny. Nat Rev Neurosci 15, 615-627 35. Li, X., McClellan, M. E., Tanito, M., Garteiser, P., Towner, R., Bissig, D., Berkowitz, B. A., Fliesler, S. J., Woodruff, M. L., Fain, G. L., Birch, D. G., Khan, M. S., Ash, J. D., and Elliott, M. H. (2012) Loss of caveolin-1 impairs retinal function due to disturbance of subretinal microenvironment. J Biol Chem 287, 16424-16434 36. Lakk, M., Yarishkin, O., Baumann, J. M., Iuso, A., and Krizaj, D. (2017) Cholesterol regulates polymodal sensory transduction in Muller glia. Glia 65, 2038-2050 37. Elliott, M. H., Fliesler, S. J., and Ghalayini, A. J. (2003) Cholesterol-dependent association of caveolin-1 with the transducin alpha subunit in bovine photoreceptor rod outer segments: disruption by cyclodextrin and guanosine 5'-O-(3-thiotriphosphate). Biochemistry 42, 7892-7903 38. Lajoie, P., Goetz, J. G., Dennis, J. W., and Nabi, I. R. (2009) Lattices, rafts, and scaffolds: domain regulation of receptor signaling at the plasma membrane. J Cell Biol 185, 381-385 39. Bishop, G. A. (2016) TRAF3 as a powerful and multitalented regulator of lymphocyte functions. J Leukoc Biol 100, 919-926 40. Hauer, J., Puschner, S., Ramakrishnan, P., Simon, U., Bongers, M., Federle, C., and Engelmann, H. (2005) TNF receptor (TNFR)-associated factor (TRAF) 3 serves as an inhibitor of TRAF2/5- mediated activation of the noncanonical NF-kappaB pathway by TRAF-binding TNFRs. Proc Natl Acad Sci U S A 102, 2874-2879 41. He, J. Q., Saha, S. K., Kang, J. R., Zarnegar, B., and Cheng, G. (2007) Specificity of TRAF3 in its negative regulation of the noncanonical NF-kappa B pathway. J Biol Chem 282, 3688-3694 42. Hacker, H., Tseng, P. H., and Karin, M. (2011) Expanding TRAF function: TRAF3 as a tri-faced immune regulator. Nat Rev Immunol 11, 457-468 43. Dillon, C. P., and Balachandran, S. (2016) StIKKing it to a death kinase: IKKs prevent TNF- alpha-induced cell death by phosphorylating RIPK1. Cytokine 78, 47-50 44. Vy Tran, A. H., Hahm, S. H., Han, S. H., Chung, J. H., Park, G. T., and Han, Y. S. (2015) Functional interaction between hMYH and hTRADD in the TNF-alpha-mediated survival and death pathways of HeLa cells. Mutat Res 777, 11-19 45. Wang, K., Zhu, X., Mei, D., and Ding, Z. (2018) Caveolin-1 contributes to anoikis resistance in human gastric cancer SGC-7901 cells via regulating Src-dependent EGFR-ITGB1 signaling. J Biochem Mol Toxicol 32, e22202 46. Tang, W., Feng, X., Zhang, S., Ren, Z., Liu, Y., Yang, B., lv, B., Cai, Y., Xia, J., and Ge, N. (2015) Caveolin-1 Confers Resistance of Hepatoma Cells to Anoikis by Activating IGF-1 Pathway. Cell Physiol Biochem 36, 1223-1236 47. de Hoz, R., Rojas, B., Ramirez, A. I., Salazar, J. J., Gallego, B. I., Trivino, A., and Ramirez, J. M. (2016) Retinal Macroglial Responses in Health and Disease. Biomed Res Int 2016, 2954721 48. Kumar, A., and Shamsuddin, N. (2012) Retinal Muller glia initiate innate response to infectious stimuli via toll-like receptor signaling. PLoS One 7, e29830 49. Rutar, M., Natoli, R., Chia, R. X., Valter, K., and Provis, J. M. (2015) Chemokine-mediated inflammation in the degenerating retina is coordinated by Muller cells, activated microglia, and retinal pigment epithelium. J Neuroinflammation 12, 8 50. Moon, H., Lee, C. S., Inder, K. L., Sharma, S., Choi, E., Black, D. M., Le Cao, K. A., Winterford, C., Coward, J. I., Ling, M. T., Australian Prostate Cancer, B., Craik, D. J., Parton, R. G., Russell, P. J., and Hill, M. M. (2014) PTRF/cavin-1 neutralizes non-caveolar caveolin-1 microdomains in prostate cancer. Oncogene 33, 3561-3570 51. Inder, K. L., Ruelcke, J. E., Petelin, L., Moon, H., Choi, E., Rae, J., Blumenthal, A., Hutmacher, D., Saunders, N. A., Stow, J. L., Parton, R. G., and Hill, M. M. (2014) Cavin-1/PTRF alters prostate cancer cell-derived extracellular vesicle content and internalization to attenuate

extracellular vesicle-mediated osteoclastogenesis and osteoblast proliferation. J Extracell Vesicles 3 52. Wajant, H., Henkler, F., and Scheurich, P. (2001) The TNF-receptor-associated factor family: scaffold molecules for cytokine receptors, kinases and their regulators. Cell Signal 13, 389-400

FOOTNOTES This work was funded by grants from R01EY019494 to MHE; T32EY023202; P30EY021725; an unrestricted grant from Research to Prevent Blindness to the Department of Ophthalmology at the Dean McGee Eye Institute (OUHSC); the German Research Foundation (DFG: HAU 6014/5-1, priority program SPP 2127) to SMH.

Dr. Jami Gurley is the recipient of the Postdoctoral OCAST Fellowship grant HF18-008.

The abbreviations used are: Cav-1, caveolin-1; NR, neural retina; LPS, lipopolysaccharide; Traf3, TNF receptor-associated factor 3; TNF, tumor necrosis factor; TLR4, toll-like receptor 4; RPE, retinal pigment epithelium; KO, knockout; WT, wild type; GS, glutamine synthetase; RGCs, retinal ganglion cells; Brn3a, Brain-Specific Homeobox/POU Domain Protein 3A; BPCs, bipolar cells; PRCs, photoreceptors; ONL, outer nuclear layer; INL, inner nuclear layer; IL-6, interleukin 6; CXCL1, C-X-C motif chemokine ligand 1; MCP-1, monocyte chemoattractant protein 1; PMNs, polymorphonuclear leukocytes; INF. MNCs, inflammatory monocytes; MΦ, macrophages; PPI, protein-protein interaction; RIPK1, receptor- interacting serine/threonine-protein kinase 1; TRAF2, TNF receptor-associated factor 3; TRADD, tumor necrosis factor receptor type 1-associated DEATH domain protein; PTPN11, tyrosine-protein phosphatase non-receptor type 11; SRC, neuronal proto-oncogene tyrosine-protein kinase; BCAR1, breast cancer anti- estrogen resistance protein 1; DAPI, 4,6-diamidino-2-phenylindole; KEGG, Kyoto Encyclopedia of Genes and Genomes; IPA, Ingenuity Pathway Analysis.

Immune-Related Gene Ontology Enrichment Analysis for All 177 Significantly Upregulated Genes following NR-Cav1 Depletion

N High level GO category Genes 45 Response to stress DUSP3 RBM14 ANXA6 RAD21 HSPA9 ANXA5 VCP PCK2 H2AFX UBE2N ENTPD2 VAMP4 PPP3CA GSTM5 GORASP2 UCHL5 XAB2 TRAF3 OXCT1 DLG1 FEN1 CORO1B HSPD1 CAPN2 DDAH1 SFPQ NONO NEDD4 ADRA2A MACROD1 NPLOC4 SMC1A CORO2B HNRNPA1 USP14 PRKCA PRMT1 ANXA7 UCHL1 PTPN11 WDFY1 GSTM7 CAMK2G HSPA4L XRN2 17 Immune system process DUSP3 RBM14 HSPD1 VAMP4 TRAF3 HSPA9 SFPQ NONO NPLOC4 USP14 RPS17 PRMT1 PTPN11 DLG1 NEDD4 WDFY1 UBE2N 13 Immune response DUSP3 RBM14 VAMP4 TRAF3 SFPQ NONO NPLOC4 USP14 DLG1 NEDD4 WDFY1 HSPD1 UBE2N 12 Regulation of immune system process DUSP3 RBM14 HSPD1 TRAF3 HSPA9 SFPQ NONO NPLOC4 PRMT1 DLG1 WDFY1 UBE2N 9 Activation of immune response DUSP3 RBM14 SFPQ NONO NPLOC4 TRAF3 WDFY1 HSPD1 UBE2N 4 Immune effector process TRAF3 NPLOC4 DLG1 HSPD1

Table 1. Effect of neuroretinal-specific depletion on immune-related gene ontology categories for significantly upregulated genes determined by proteomics data. ShinyGO analysis was performed on all 177 genes that were identified by proteomics as being significantly upregulated following NR-Cav1 depletion. Table represents immune-associated functional category groups (“High level GO category”) defined by high-level gene ontology terms used by ShinyGO analysis software. Table also includes both the number (“N”) and gene names (“Gene”) of upregulated genes identified to contribute to the immune- associated GO categories.

Figure 1. Effect of neuroretinal-specific deletion on whole retinal tissue CAV1 protein expression (Chx10-Cre/Cav1-flox model). (A) Schematic diagram illustrating the strategy for neuroretinal-specific Cav1 deletion via Chx10-promoter driven Cre-mediated recombination of the floxed Cav1 gene. (B) Representative Western blots of whole cell lysates from various tissues showing retina-specific CAV1 protein reduction following Chx10-Cre expression in the floxed Cav1 model. Protein loads ranged from 5 µg to 35 µg, depending on tissue, in order to detect protein within the linear range of the assay. (C) Histogram of Western blot protein densitometry data quantification showing retinal tissue specific CAV1 protein reduction following NR-Cav1-depletion in Chx10-Cre/Cav1+/+ mice compared to Chx10-Cre/Cav- /- controls. β-actin was used as a loading control and for normalization. Data are mean ± SEM and were analyzed via Student’s t-test (** p<0.01; N=7). NR, neural retina/neuroretinal; CAV1, Caveolin 1; ACT = β-actin.

Figure 2. Neuroretinal-specific Cav1 depletion targets retinal Müller glial cells. Immunohistochemical staining of mouse retinal sections stained with anti-CAV1 and costained with Müller cell-specific marker glutamine synthetase (GS) shows high CAV1 protein expression in the retinal Müller glial cell population. Retinal vascular endothelial cross-sections (arrowheads) also exhibit high levels of CAV1 protein expression, but are not targeted by the NR-Cav1 KO model. Sparse mosaic Cre- mediated recombination results in occasional residual Müller glial CAV1 protein expression (arrows). WT, NR-Cav1 WT; KO, NR-Cav1 KO; GS, glutamine synthetase; Cav1, caveolin 1; RGC, retinal ganglion cell layer; INL, inner nuclear layer; ONL, outer nuclear layer.

Figure 3. Neuroretinal-specific Cav1 depletion blunts endotoxin-mediated retinal proinflammatory cytokine production. Multiplex protein suspension array analysis showing endotoxin-mediated inflammatory activation in NR-Cav1 WT and KO retinas 24h-post intravitreal LPS injection. Histograms represent measured concentrations of proinflammatory cytokines/chemokines (A) IL-6, (B) CXCL1/KC, (C) CCL2/MCP-1, and (D) IL-1β. Data are mean ± SEM and were analyzed using 2-way ANOVA, with Fisher’s LSD post-hoc. Treatment effect: *<0.05, ***p<0.001, ****p<0.0001; Genotype effect: ††p<0.01, †††p<0.001. IL-6, interleukin 6; CXCL1/KC, C-X-C motif chemokine ligand 1; CCL2/MCP- 1, monocyte chemoattractant protein 1; IL-1β, interleukin 1 beta.

Figure 4. Neuroretinal-specific (but not endothelial cell-specific) Cav1 depletion reduces immune cell recruitment to retinal tissue. Flow cytometry data representing immune cell infiltration into retinal tissue 24-hours post-intraocular LPS challenge. Animals were perfused prior to collection of whole retinal tissue from Chx10-Cav1 WT and Chx10-Cav1 KO animals. (A) Representative dot plots showing spread of all data and gating strategy represent the following cell populations: total leukocytes (CD45hi), PMNs (F4/80-/GR1+), INF. MNCs (F4/80+/Gr1+), and MΦ (F4/80+/GR1-). (B) Histograms of summarized data from control or LPS-treated Chx10-Cav1 WT and Chx10-Cav1 KO animals. Data are mean ± SEM and were analyzed using 2-way ANOVA, with Fisher’s LSD post-hoc. Treatment effect: *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001; Genotype effect: †p<0.05, †††p<0.001. PMNs, polymorphonuclear leukocytes; INF. MNCs, inflammatory monocytes; MΦ, macrophages.

Figure 5. Effect of neuroretinal-specific depletion on membrane-associated protein composition. Mass spectrometry-based proteomics heat map analysis (pheatmap R) of membrane raft fractions from whole retinal tissue and histograms representing peptide abundance quantification that shows Cav1 downregulation and TRAF3 upregulation in raft protein fractions as a result of NR-Cav1 depletion. Each replicate sample contained pooled retinas from 8 mice. Blue = downregulated; red = upregulated. WT, NR-Cav1 WT; KO, NR-Cav1 KO. Data are mean ± SEM and were analyzed via Student’s t-test (* p<0.05; *** p<0.001). WT, NR-Cav1 WT; KO, NR-Cav1 KO.

Figure 6. CAV1 protein interaction network. Shared protein-protein interactions (PPIs) between CAV1 and the most significantly changed proteins. Red=upregulated after CAV1 KO, green=downregulated. Red dashed lines indicate potential PPIs that are altered by an absence of cellular CAV1 protein.