Molecular Vision 2019; 25:266-282
Received 03 January 2019 | Accepted 25 April 2019 | Published 27 April 2019
1Department of Ophthalmology, Oregon Health & Sciences University, Portland, OR; 2Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL
Correspondence to: Mary K. Wirtz, Department of Ophthalmology, Oregon Health & Sciences University, Portland, OR 97239 Phone: (503) 494 4698; email: email@example.com
Purpose: Inflammatory responses may be involved in the glaucomatous process. Our previous studies mapped a T104M mutation in interleukin-20 receptor beta (IL-20RB) in a family with primary open angle glaucoma (POAG). IL-20RB can heterodimerize with IL-20RA to propagate signals from IL-20 family cytokines, IL-19, IL-20, and IL-24 (the type I receptor complex), or it can heterodimerize with IL-22RA to propagate signals from IL-20 and IL-24 (type II receptor complex). In this study, we investigated IL-20 heterodimeric receptor complexes in the trabecular meshwork (TM) compared to dermal fibroblast cell cultures, and examined the phosphorylation of signal transducer and activator of transcription (STAT)-1, -3, and -5 following exposure to IL-20 family cytokines. Additionally, we determined the effects of IL-20 family cytokines on outflow rates in anterior segment perfusion culture, an in vitro model of intraocular pressure (IOP) regulation.
Methods: Primary human TM (HTM) cells were grown from dissected TM tissue, and IL-20 receptor expression was investigated with PCR. A Duolink assay was performed to investigate in situ IL-20 receptor protein interactions in HTM or dermal fibroblasts, and Imaris software was used to quantitate the association of the heterodimeric complexes. Phosphorylation of STAT-1, -3, and -5 were evaluated in HTM or dermal fibroblasts using Western immunoblotting after exposure to IL-10, IL-19, IL-20, IL-22, or IL-24. Anterior segment perfusion culture was performed in human cadaver and porcine eyes treated with IL-20, IL-19, or IL-24.
Results: All of the IL-20 receptors, IL-20RA, IL-20RB, and IL-22RA1 were expressed in HTM cells. Two isoforms of IL-20RA were expressed: The V1 variant, which is the longest, is the predominant isoform, while the V3 isoform, which lacks exon 3, was also expressed. The Duolink assay demonstrated that the type I (IL-20RA–IL-20RB) and type II (IL-22RA1–IL-20RB) receptors were expressed in HTM cells and dermal fibroblasts. However, in the HTM cells, the type I receptor was present at significantly higher levels, while the type II receptor was preferentially used in the dermal fibroblasts. The HTM cells and the dermal fibroblasts predominantly phosphorylate the Ser727 site in STAT-3. The dermal fibroblasts had higher induction of phosphorylated STAT-1 compared to the HTM cells, while neither cell type had phosphorylated STAT-5 in the cell lysates. The outflow rates in the human anterior segment cultures were increased 2.3-fold by IL-20. However, IL-19 and IL-24 showed differential responses. For IL-19 and IL-24, 50% of the eyes responded with a 1.7- or 1.5-fold increase, respectively, while the other half did not respond. Similarly, perfused porcine anterior segments showed “responders” and “non-responders”: IL-20 responders (2.3-fold increase in outflow, n=12) and non-responders (n=11); IL-19 responders (2.1-fold increase, n=7) and non-responders (n=5); and IL-24 responders (1.8-fold increase, n=12) and non-responders (n=5).
Conclusions: Type I and type II IL-20 receptor complexes are expressed in human TM cells with predominant expression of the type I receptor (IL-20RA and IL-20RB), which propagates signals from all three IL-20 family cytokines. However, there was a variable response in the outflow rates following perfusion of cytokines in two different species. This may explain why some people are more susceptible to developing elevated IOP in response to inflammation.
Glaucoma is a group of diseases resulting from damage to the optic nerve that leads to a distinctive pattern of vision loss. Nearly 112 million individuals will develop glaucoma by 2040, making glaucoma the leading cause of global irreversible blindness . Primary open angle glaucoma (POAG) is the most common form of glaucoma in Western and African nations . Because the symptoms are subtle in the early stages of the disease, many individuals remain undiagnosed until severe visual field loss has occurred. Vision loss from glaucoma has a significant impact on health-related quality of life, and the overall burden increases as glaucomatous damage and vision loss progress . The exact etiology of POAG remains unknown, although it is clear that it is a heterogeneous group of disorders and is frequently associated with elevated intraocular pressure (IOP) [3-5]. Although lowering IOP often reduces the rate of vision loss, many patients continue to go blind despite apparently “successful” pressure control [4-6]. Elevated IOP is caused by dysfunction of the aqueous humor drainage pathway in the trabecular meshwork (TM) tissue in the anterior of the eye , which eventually damages optic nerve axons, resulting in retinal ganglion cell death and visual impairment [8-10].
There are potentially many disease mechanisms behind glaucoma. Recent studies suggested that inflammatory responses may be involved in the glaucomatous process, as shown by human studies and in rodent models [11-15]. Drugs such as glucocorticoids are often used to treat inflammatory and autoimmune disease, but in certain individuals, these drugs can have serious side effects, such as the development of ocular hypertension and glaucoma [16-18]. Becker and Mills found that normal patients fell into two distinct subgroups: those exhibiting a moderate increase in IOP and those exhibiting no increase in IOP . Follow-up studies refined the characterization of response to glucocorticoids in normal patients, recognizing three distinct subgroups: those exhibiting a high response (a change >15 mmHg), those exhibiting a moderate response (a change of 6–15 mmHg), and those exhibiting a low response (a change <6 mmHg) [20,21]. Subsequently, individuals with a predisposition to develop elevated IOP when treated with glucocorticoids systemically or locally in the eye were referred to as “steroid responders,” a term coined by Armaly and Becker .
Other studies have demonstrated a genetic component of glaucoma [22-26]. In 1997, our group mapped the GLC1C (Gene ID 50604; OMIM 601682) locus in a large family with POAG to a region on chromosome 3 . Subsequently, a T104M mutation in the interleukin-20 receptor beta (IL-20RB; Gene ID 50604; OMIM 605621) gene was identified in 11 members of this family who had POAG . All 11 individuals have high IOPs ranging from 22 to 46 mmHg . The T104M mutation identified in the family with POAG is located in the IL-20/IL-20RB binding interface . Therefore, it is likely that patients with glaucoma with this mutation cannot respond appropriately to cytokine signaling, leading to an inadequate homeostatic response to elevated IOP. IL-20 family cytokines (IL-19, IL-20, and IL-24) and receptors (IL-20RA, IL-20RB, IL-22RA1) are members of the larger IL-10 family, which were grouped together based on conserved receptor sequences . IL-19 signals exclusively through the type I receptor, a heterodimer consisting of IL-20RA and IL-20RB (Figure 1) . IL-20 and IL-24 can signal through either the type I receptor or the type II receptor, which is composed of IL-22RA1 and IL-20RB [27,28]. IL-22 and IL-26 do not signal through IL-20RB, but instead use heterodimeric complexes consisting of IL-22RA1/IL-10RB, and IL-20RA/IL-10RB, respectively .
The IL-20 family has been proposed to have proinflammatory, angiogenic, and chemoattractive effects on chronic inflammatory diseases, including psoriasis, atherosclerosis, and rheumatoid arthritis [29,30]. This family of cytokines enhances tissue remodeling and wound-healing activities, maintaining tissue integrity and restoring homeostasis of the epithelial layers during infection and inflammatory responses . Interestingly, IL-24 was first reported as an antitumor molecule, and was originally named melanoma differentiation-associated 7 (mda-7) . The IL-20 family is known to be expressed in myeloid, epithelial, and T cells, as well as innate lymphoid cells. However, little is known about IL-20 family signaling in the TM .
Our previous genetic study reported a T104M mutation in the extracellular domain of the IL-20RB in a family with glaucoma, but it remains unclear which IL-20 receptors are expressed in normal TM cells, and whether the IL-20 receptor complexes are functional. The purpose of this study was to investigate expression of IL- 20 heterodimeric receptor complexes in TM cell cultures, and to measure the effects of IL-20 family cytokines on outflow rates in anterior segment perfusion culture, an in vitro model of IOP regulation.
Human TM (HTM) cells were grown from TM tissue from cadaver eyes (Lions VisionGift, Portland, OR) or from corneal rims discarded following corneal transplant surgery (Casey Eye Institute, Portland, OR) . A summary of the donors for the cell strains used in this manuscript is included in Appendix 1. HTM cells were grown in a 1:1 mixture of high and low glucose Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS; Fisher Scientific, Waltham, MA) and 1% penicillin-streptomycin-Fungizone (Gibco, Waltham, MA). HTM cell cultures were split 1:3, and used up to passage 6. All TM cell cultures were examined morphologically and tested for their ability to induce myocilin protein after exposure to dexamethasone for 2 weeks, the current standard for characterizing primary TM cells . Adult human dermal fibroblasts were purchased from three commercial sources: PromoCell (age: 60, sex: male, cell source: cheek skin; Heidelberg, Germany), MilliporeSigma (age: 38, sex: female; cell source: abdominal skin; St. Louis, MO), and ATCC (age: 50, sex: female, cell source: abdominal skin; Manassas, VA). All fibroblasts were maintained in DMEM containing 10% FBS and 1% penicillin-streptomycin-Fungizone.
HTM cells (n=5) or human dermal fibroblasts were grown until confluent. RNA was isolated in TRIzol (ThermoFisher, Grand Island, NY) and purified with the Direct-zol RNA miniprep kit (Zymo Research, Irvine, CA). Total RNA concentration and purity were quantitated with a NanoDrop 2000 (Wilmington, DE). Superscript III reverse transcriptase (ThermoFisher) was used to generate cDNA using 300–600 ng RNA as a template. PCR products were amplified (94 ºC for 15 min, then 32 cycles of 94 ºC for 30 s, 60 ºC for 30 s and 72 ºC for 45 s, followed by a final extension of 72 ºC for 10 min) on a thermal cycler using the primers listed in Appendix 2. To investigate the alternative splicing of IL-20RA, primers (602S and 603AS) were designed to flank exon 3, and reverse transcriptase (RT)–PCR was performed. Products were separated on an agarose gel and purified, and DNA sequencing was performed to confirm the identity of the products.
To investigate IL-20 receptor protein interactions in cells, an in situ Duolink proximity ligation assay (PLA; Duolink Red starter kit for mouse and goat; MilliporeSigma) was performed. This assay uses primary antibodies to each receptor subunit, which were raised in different host species, and detected with secondary antibodies conjugated to PLA probes. When the two primary antibodies are in close proximity (<40 nm), the DNA strands on the probes are joined by ligation into a circular DNA molecule, which is then amplified with “rolling-circle amplification” (RCA) primed by one of the probes. The RCA product was detected with hybridization of fluorescently labeled oligonucleotides. HTM or human dermal fibroblasts were grown to approximately 70% confluence and then fixed in 4% paraformaldehyde. The assay was then performed following the manufacturer’s instructions: Briefly, after blocking at 37 °C for 1 h, the cells were incubated with primary antibody overnight at 4 °C. Antibody combinations were IL-20RB (goat polyclonal; R&D Systems, Minneapolis, MN) and IL-20RA (Mouse monoclonal; R&D Systems), or IL-20RB and IL-22RA1 (mouse monoclonal; R&D Systems). Negative controls were performed by substituting the primary antibody with PBS (1X; 155 mM NaCl, 3 mM Na2HPO4, 1 mM KH2PO4, pH 7.4; Gibco, Waltham, MA) After washing, the primary antibodies were incubated with the PLA probes, which were subsequently ligated with incubation with ligase. Signals were amplified by the addition of polymerase for 1 h 40 min at 37 °C. After final washes, coverslips were mounted in mounting medium containing 4′,6-diamidino-2-phenylindole (DAPI). Confocal z-stack images were acquired with confocal microscopy (Olympus Fluoview, Center Valley, PA), and the number of punctate dots and nuclei per field were counted using the “spots” module of Imaris software (Bitplane, Concord, MA). Results are reported as the number of spots/nuclei in a field, and seven to 14 fields were counted for each receptor type in each cell type, and then averaged. Data were analyzed with one-way ANOVA to determine significance with Bonferroni correction. A p value of less than 0.025 was considered statistically significant.
HTM cells or dermal fibroblasts were grown to confluence in DMEM + 10% FCS, washed with PBS, and placed in serum-free media overnight. The next day, the serum-free media were replaced, and the following recombinant human cytokines were added for 15 min: IL-10, IL-19, IL-20, IL-22, or IL-24 (all 100 ng/ml; R&D Systems). An untreated control was also included. Cell lysates were harvested in RIPA buffer, and the proteins were separated on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) gels and transferred to nitrocellulose membranes. The membranes were then probed with the following signal transducer and activator of transcription (STAT) antibodies (all antibodies from Cell Signaling Technology Inc., Danvers, MA, unless stated otherwise): phospho-tyrosine701 STAT-1 (mouse monoclonal; ThermoFisher, 33–3400), total STAT-3 (mouse monoclonal; 9139), phospho-serine727 STAT-3 (rabbit polyclonal; 9134), phospho-serine705 STAT-3 (rabbit polyclonal; 9131), total STAT-5 (rabbit polyclonal; 25,656), and phospho-TyrosineY694 (mouse monoclonal; 9356). After washing and incubation with infrared (IR)-dye-conjugated secondary antibodies, the membranes were imaged on the Li-Cor Odyssey (Lincoln, NE) imaging system. Densitometry was used to measure band intensity using FIJI software (provided in the public domain). Experiments were performed at least three times using at least three biologic replicates for each cell type. Data from each experimental replicate (n=3–6) were normalized to the total STAT-3 loading control, made a percentage of the untreated control. Then, the data were averaged for each treatment, and a standard error of the mean calculated. Statistical significance was set at a p value of less than 0.05.
Human cadaver eyes were acquired from the Lions VisionGift eye bank. Information on donor eyes can be found in Appendix 1. The eye globes were bisected, and the lens, ciliary body, and iris were removed from the anterior segment. After dissection, the human anterior segments were maintained in serum-free stationary organ cultures to allow recovery of the cells post-mortem. Porcine eyes were acquired from a local abattoir (Carlton Farms, Carlton, OR), and dissected within 4 h of death. For both species, anterior segment containing the cornea, trabecular meshwork, and sclera was clamped into a perfusion chamber and perfused with serum-free DMEM at a constant pressure of 8 mmHg . After 24 h, any anterior segment that was not stabilized at a flow rate of 1–8 µl/min was discarded . Flow rates were measured for 4 h before 100 ng/ml of human recombinant IL-20, IL-19 or IL-24 (R&D Systems, Minneapolis, MN) were added in 100 µl of PBS to the intake port. Perfusion was resumed for a further 70–73 h, and the flow rates were measured. For each eye, the flow rates after treatment were normalized to the average flow rate before the cytokines were added. Individual eye data were combined and averaged, and a standard error of the mean calculated. Statistical significance was determined using a one-way ANOVA, where a p value of less than 0.05 was considered statistically significant.
Amino acid sequences of human and porcine IL-20 receptors were obtained from the NCBI. The sequences were aligned using Clustal Omega . Residues involved in IL-20–IL-20RA, IL-20–IL-20RB, IL-20RA–IL-20RB, and IL-22–IL-22RA1 interactions were identified using the crystal structures of the IL-20–IL-20RA–IL-20RB complex (pdbid 4doh) and the IL-22–IL-22RA1 complex (pdbid 3dgc). Contact residues that bury surface area in the interfaces were identified using the PISA server  and validated with computer graphics using the program pymol.
In this study, we investigated IL-20 receptor family expression in human TM cells and dermal fibroblasts. With RT–PCR, all of the IL-20 receptors, IL-20RA, IL-20RB, and IL-22RA1 were expressed in human TM cells, as well as the related receptor, IL-10RB (Figure 2A). Four alternatively spliced isoforms of IL-20RA are reported in the NCBI database (V1: NM_014432.2), V2: NM_001278722.1, V3: NM_001278723.1, and V4: NM_001278724.1), which vary in the untranslated region (UTR) and extracellular binding region of the receptor. Therefore, we investigated which splice forms are expressed by HTM cells and dermal fibroblasts. In both cell types, the longest V1 variant was the predominant isoform expressed, but the V3 isoform, which lacks exon 3 encoding an extracellular domain, was also expressed (Figure 2B). Using appropriate primer sets, we did not detect isoforms V2 or V4 in either the HTM cells or the dermal fibroblasts (data not shown).
Next, we investigated which receptor complex, that is, the type I receptor (IL-20RA–IL-20RB heterodimer) or the type II receptor (IL-22RA1–IL-20RB heterodimer), was used in dermal fibroblasts and HTM cells. We used the in situ Duolink proximity ligation assay, which investigates protein interactions in fixed cells . In this assay, IL-20 receptor heterodimers were visualized as punctate dots with confocal microscopy (Figure 3A–D). The type I and type II receptors were identified in both cell types. However, after quantification of the number of dots, there was a statistically significant difference between the type I and type II receptors in the fibroblasts compared to the HTM cells. In the dermal fibroblasts, there were statistically significantly more dots detected with the IL-20RB and IL-22RA1 antibodies, whereas in the HTM cells, there were statistically significantly more dots detected with the IL-20RB and IL-20RA antibody pairings (Figure 3E). A negative control where primary antibodies were replaced by PBS is shown (Appendix 3). This suggests that HTM primarily signal via type I receptors, while dermal fibroblasts prefer type II receptors.
Once IL-20 family cytokines bind to the heterodimeric IL-20 receptor complexes, STATs become activated (Figure 1). The potential for STAT activation is based on the finding that IL-22 induces STAT-1, -3, and -5 through the IL-22RA1/IL-10RB complex in hepatoma cells . However, it is unknown which STATs are used in TM cells. Therefore, we investigated STAT phosphorylation in the HTM cells compared to the dermal fibroblasts (Figure 4). STAT-1 is phosphorylated at Tyr701 . In the HTM cells, there were lower levels of phosphorylated STAT-1 in cell lysates compared to the levels in the dermal fibroblasts. Although all cytokines, and IL-22 in particular, induced Tyr701 STAT-1 phosphorylation in the dermal fibroblasts, there was no such induction by IL-22, or any of the cytokines, in the HTM cells. The major two phosphorylation sites of STAT-3 are Ser727 and Tyr705 . The present results showed that the HTM cells and dermal fibroblasts predominantly phosphorylate STAT-3 on the Ser727 site. The phospho-Ser727 antibody detected two bands in the serum-free controls, but there was a marked shift to the upper band after 15 min. In the HTM cells, the upper band was increased compared to that in the serum-free controls, which suggests activation via this phosphorylation site. Densitometry was used to quantitate Ser727 phosphorylation of STAT-3 (Figure 4C). In the HTM cells (n=6), all the cytokines tested induced phosphorylation of Ser727. However, in the dermal fibroblasts (n=3), IL-20, IL-22, and IL-24 induced phosphorylation, whereas IL-10 and IL-19 did not. The STAT-3 Tyr705 phosphorylation site was not used, apart from when the dermal fibroblasts were stimulated with IL-22. STAT-5 is phosphorylated at Tyr694 . Much lower amounts of STAT-5 were detected in lysates from both cell types compared to STAT-3, and none of the cytokines stimulated phosphorylation of Tyr694 in either cell type. Thus, the IL-20 family cytokines appear to predominantly signal via phosphorylation of STAT-3 at the Ser727 site in HTM cells.
Anterior segment perfusion culture is a standard ex vivo technique for measuring the effects of agents on the outflow rates through the TM [34,35,43-45]. The T104M mutation in IL-20RB found in patients with POAG suggests that it may disrupt signaling from IL-20, IL-19, or IL-24, but not from IL-10, IL-22, or IL-26. Therefore, we tested the impact of IL-20, IL-19, and IL-24 signaling on the outflow rates using human (Figure 5) and porcine anterior segments (Figure 6). For human eyes, data from three eyes were discarded as the initial flow rates were below 1 µl/min. Application of IL-20 to human anterior segments (n=7) caused an average flow rate increase of 2.3-fold, in all eyes tested, over the 70 h experiment, compared to the vehicle control (Figure 5A). However, application of IL-19 or IL-24 had variable effects. Fifty percent of eyes tested showed increased flow rates (responders) while an equal number showed no effect (non-responders). Non-responders were defined as having a flow rate that did not go above 1.2 µl/min, while responders had flow rates >1.23 μl/min. For eyes that responded, the fold increase was lower than in the IL-20 treatment: IL-19 induced a 1.7-fold increase (Figure 5B), while IL-24 caused a 1.5-fold increase over 70 h (Figure 5C). Raw flow rate data for human eyes are shown in Table 1.
To further investigate the effects of these cytokines on outflow, we used porcine eyes, because they are more readily available. In perfused porcine anterior segments (Figure 6), there were responders and non-responders with all three cytokine treatments. For IL-20, 12 anterior segments increased the outflow to approximately 2.1-fold, whereas 11 anterior segments did not respond (Figure 6A). One outlier was excluded from the data analysis. For IL-19, seven eyes responded with a 2.1-fold increase in outflow, while five eyes had no response (Figure 6B). For IL-24, 12 eyes responded with a 1.8-fold increase in outflow, while five eyes did not respond (Figure 6C). The raw flow rate data for porcine eyes are shown in Table 2, Table 3, and Table 4. We also analyzed the final flow rate data at the 73 h time point (Figure 7). When we grouped the data together as “responders” versus “non-responders,” a significant difference was observed for each cytokine (Figure 7).
For the porcine eyes, we hypothesized that the human recombinant cytokine may not bind porcine receptors. Therefore, we analyzed the porcine IL-20 receptor protein sequences, and found that all of the residues that make direct contact with the cytokines were conserved (Figure 8) . Thus, human recombinant cytokine should bind effectively to porcine receptors.
In this study, we investigated the expression and function of IL-20 receptor complexes in regard to aqueous outflow regulation in the trabecular meshwork, which is dysfunctional in glaucoma . We initially became interested in the IL-20 family cytokines when we discovered a T104M mutation in IL-20RB in a large family with POAG . This mutation is predicted to impact cytokine binding to the IL-20RB chain . The IL-20 cytokine family is unusual, because of the promiscuity of the type I and type II receptors . However, all three ligands (IL-20, IL-19, and IL-24) can signal through the type I receptor (IL-20RA–IL-20RB), while only IL-20 and IL-24 signal through the type II receptor (IL-22RA1–IL-20RB). This raises the question of how specificity is engendered by type I and type II receptors. In this study, we showed that IL-20RA, IL-20RB, IL-22-RA1 and IL-10RB are expressed in TM cells, as well as in skin fibroblasts. Additionally, we showed that the IL-20RA isoforms V1 and V3, which lacks exon 3, are expressed by TM cells and fibroblasts. Exon 3 contains the critical YG (Tyr-76 and Gly-77) interaction motif, as well as Glu105, Tyr-109, and Arg-128, which are involved in the binding of IL-20 cytokines by IL-20RA . Domain 1 is completely absent in the IL-20RA V3 protein. Thus, the V3 isoform may act as a competitive inhibitor reducing the amount of cytokine available to the receptor .
Although it appears that HTM cells and dermal fibroblasts express all the IL-20 receptor subunits, each cell type may preferentially use different receptor complexes. The Duolink assay, which measures the colocalization of proteins, showed increased numbers of type-I complexes in the HTM cells. Conversely, in fibroblasts, type II receptor complexes predominated. Thus, the present results suggest that HTM cells can respond to all three cytokines, whereas dermal fibroblasts may preferentially respond to IL-20 and IL-24. This suggestion is supported by the western immunoblot data, where all cytokines induced phosphorylation of Ser727 of STAT-3 in the HTM cells, but IL-19 did not affect Ser727 phosphorylation in the dermal fibroblasts. Although this is the simplest interpretation, it remains possible that a lower number of spots in the Duolink assay may be due to intracellular dissociation of the heterodimeric complex following endocytosis. Future studies using overexpression of GFP-tagged IL-20RB will investigate the potential of IL-20R endocytosis. In addition, although cultured TM cells seem to utilize the Ser727 phosphorylation site of STAT-3, the significance of STAT phosphorylation in outflow experiments has not yet been tested. Nonetheless, these results may be important for other inflammatory diseases that involve IL-20 signaling, such as psoriasis, inflammatory bowel disease, liver inflammation, and rheumatoid arthritis [28,30]. Psoriasis is a chronic inflammatory disease of the skin with sharply demarcated, erythematous, and slightly elevated skin plaques covered by silvery white scales . High levels of IL-19, IL-20, and the IL-20 type I receptor (IL-20RA–IL-20RB) are found in psoriatic skin [47-49]. The epidermis of IL-20 and IL-24 transgenic mice displays severe skin abnormalities resembling human psoriatic skin [48,50,51]. In contrast, IL-20RB knockout mice are born healthy with no apparent skin abnormalities, and are protected from stimuli that induce psoriasis [52,53]. The development of psoriasis in the skin, which can be triggered by overexpression of IL-20 cytokines, can be considered an uncontrolled wound-healing response, similar to acute reepithelialization . The IL-20 subfamily participates in amplifying inflammatory responses, particularly during autoimmune and chronic inflammation, and alternatively, in anti-inflammatory responses, such as tissue protection and regeneration [28,30].
One of the most direct in vitro methods for studying the aqueous outflow pathway is to use perfusion-cultured anterior segments . Therefore, we studied the effects of IL-19, IL-20, and IL-24 on outflow in human and porcine eye models. The results of the experiments revealed some anterior segments were cytokine “responders,” and a nearly equal number were “non-responders.” This was not due to cytokine batch differences, because porcine perfusion experiments run on the same day showed that some eyes responded while others did not (Table 2, Table 3, and Table 4). We also hypothesized that human recombinant cytokines might not be recognized by porcine receptors. However, the comparison of the human and porcine IL-20 receptor amino acid sequences showed all residues that made direct contact with the cytokines were conserved. Thus, we conclude that porcine IL-20 receptors have the capacity to bind and respond to human IL-20 cytokines. Analyzing the human data shows that in certain individuals, both eyes of a pair did not respond demonstrating congruence in response within an individual (Table 1). All eyes were screened routinely with hematoxylin and eosin staining of sections, and all TM tissue contained cells (not shown). However, it remains possible that there were lower cell numbers in the non-responding eyes, thus lowering the potential to bind and propagate cytokine signals. However, one of the non-responders (2016–1705) was 42 years old at the time of death, and was unlikely to have reduced cell loss, which is more often seen in elderly TM tissue . Thus, the data suggest that certain individuals are more responsive to IL-20 family cytokine signals than others.
This study is not the first to report differences in response to treatments. Interestingly, “responders” and “non-responders” have also been described for glucocorticoids, such as dexamethasone. Glucocorticoids can induce ocular hypertension in approximately 40% of the general population, and at a much higher rate in patients with glaucoma [18,57,58]. Glucocorticoids induce their effects via the glucocorticoid receptor (GR), which undergoes alternative splicing to produce GRα and GRβ . The GRβ isoform is a dominant negative inhibitor of GRα function, and their relative ratio regulates glucocorticoid responsiveness in TM cells and in mice [18,60].
In conclusion, the present results demonstrated that ocular TM cells express all IL-20 receptors, and predominantly signal through the type I receptor, which can propagate signals from IL-20, IL-19, and IL-24 cytokines. Finally, the outflow results suggest that responses to IL-20 cytokines are variable. As a result, some individuals are unable to alter outflow rates when exposed to IL-20 cytokines. Thus, these results may explain why some individuals are more susceptible to developing elevated IOP in response to inflammation.
Appendix 1. Summary of the donors for the cell strains.
Appendix 2. PCR primers used to determine alternative splicing of IL-20RA.
Appendix 3. Negative control for Figure 3 with replacement of primary antibodies by PBS
The authors would like to thank Lions VisionGift, Portland, OR for their help in the procurement of human eye cadaver tissue. This work was supported by NIH/NEI grants R21EY027079 (MKW), R01EY11650 (MKW) and P30EY010572 (Casey Eye Institute Core facility grant) and an unrestricted grant to the Casey Eye Institute from Research to Prevent Blindness, New York, NY.