Combination of Peroxisome Proliferator-Activated

International Journal of Molecular Sciences

Article Combination of Peroxisome Proliferator-Activated Receptor (PPAR) Alpha and Gamma Agonists Prevents Corneal Inﬂammation and Neovascularization in a Rat Alkali Burn Model

Yuji Nakano 1,2, Takeshi Arima 1,2 , Yutaro Tobita 1,2, Masaaki Uchiyama 1,2, Akira Shimizu 2 and Hiroshi Takahashi 1,*

1 Department of Ophthalmology, Nippon Medical School, Tokyo 113-8602, Japan; [email protected] (Y.N.); [email protected] (T.A.); [email protected] (Y.T.); [email protected] (M.U.) 2 Department of Analytic Human Pathology, Nippon Medical School, Tokyo 113-8602, Japan; [email protected] * Correspondence: [email protected]; Tel.: +81-3-3822-2131

Received: 12 June 2020; Accepted: 15 July 2020; Published: 19 July 2020

Abstract: Peroxisome proliferator-activated receptor alpha (PPARα) and gamma (PPARγ) agonists have anti-inflammatory and anti-neovascularization effects, but few reports have tested the combination of PPARα and PPARγ agonists. In this study, we investigated the therapeutic effects of ophthalmic solutions of agonists of PPARα, PPARγ, and the combination in a rat corneal alkali burn model. After alkali injury, an ophthalmic solution of 0.05% fenofibrate (PPARα group), 0.1% pioglitazone (PPARγ group), 0.05% fenofibrate + 0.1% pioglitazone (PPARα+γ group), or vehicle (vehicle group) was topically instilled onto the rat’s cornea twice a day. After instillation, upregulation was seen of PPAR mRNA corresponding to each agonist group. Administration of agonists for PPARα, PPARγ, and PPARα+γ suppressed inflammatory cells, neovascularization, and fibrotic changes. In addition, the PPARγ agonist upregulated M2 macrophages, which contributed to wound healing, whereas the PPARα agonist suppressed immature blood vessels in the early phase. Administration of PPARα+γ agonists showed therapeutic effects in corneal wound healing, combining the characteristics of both PPARα and PPARγ agonists. The results indicate that the combination of PPARα and γ agonists may be a new therapeutic strategy.

Keywords: PPARα; PPARγ; combination of PPARα and PPARγ; cornea; alkali burn injury; neovascularization; inﬂammation; ﬁbrotic changes

1. Introduction Transparency of the cornea is an essential factor for normal vision. An alkali burn can cause severe damage to the corneal surface, resulting in loss of its transparency [1,2]. Thus, suppression of scar formation and neovascularization during corneal wound healing is important for a good visual outcome. Peroxisome proliferator-activated receptors (PPARs) are a group of nuclear receptors that belong to the steroid hormone receptor superfamily and act as transcription factors to control activation of genes [3,4]. PPARs have three isoforms—PPAR alpha (α), PPAR beta/delta (β/δ), and PPAR gamma (γ), and they are important factors in adipocyte differentiation and lipid metabolism [5]. Recent studies have shown their roles in lipid metabolism and also in the control of inflammation, angiogenesis, and fibrosis [6,7]. Furthermore, we previously showed anti-inflammatory and anti-neovascularization effects by PPARα and PPARγ agonists in a rat corneal alkali burn model [8,9].

Int. J. Mol. Sci. 2020, 21, 5093; doi:10.3390/ijms21145093 www.mdpi.com/journal/ijms Int. J. Mol. Sci. 2020, 21, 5093 2 of 16

Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 2 of 16 Several studies have shown that the combination therapy of PPARα and PPARγ agonists is more Several studies have shown that the combination therapy of PPARα and PPARγ agonists is more effective than the use of either alone [10,11]. For example, addition of a PPARγ agonist to a PPARα effective than the use of either alone [10,11]. For example, addition of a PPARγ agonist to a PPARα agonistagonist further further reduces reduces the the expression expression ofof genesgenes involvedinvolved in insulin resistance, resistance, fatty fatty acid acid synthesis, synthesis, andand fibrosis fibrosis than than when when either either agonist agonist is is appliedapplied alonealone [[11].11]. However, However, as as fa farr as as we we know, know, little little has has beenbeen reported reported on on the the e ffeffectsects of of combined combined PPARPPARαα andand PPARγ agonists in in the the field field of of ophthalmology. ophthalmology. InIn the the present present study, study, we we compared compared the the anti-inflammatory, anti-inflammatory, anti-neovascularization,anti-neovascularization, and and anti-fibrosis anti-fibrosis effectseffects of of fenofibrate, fenofibrate, a a selective selective agonist agonist ofof PPARPPARαα,, pioglitazone,pioglitazone, a a selective selective agonist agonist of of PPAR PPARγγ, and, and theirtheir combination combination in in a rata rat corneal corneal alkali alkali burn burn model. model.

2. Results2. Results

2.1.2.1. Corneal Corneal Wound Wound Healing Healing after after Alkali Alkali Burn Burn EffEffectsects of of each each PPAR PPAR agonist agonist ophthalmic ophthalmic solutionsolution werewere examinedexamined by macroscopic macroscopic observation observation (Figure(Figure1) and1) and histological histological analysis analysis using using hematoxylin hematoxylin and and eosin eosin (HE)(HE) stainingstaining (Figure(Figure2 2).). InIn thethe macroscopicmacroscopic photographs, photographs, blood blood vessels vessels in in the the irisiris disappeared,disappeared, and the corneal corneal epithelial epithelial defect defect was was seenseen as as a circlea circle line line immediately immediately after after the the injuryinjury (Figure(Figure1 1a).a). OnOn day 7, the cornea lost lost transparency, transparency, andand bleeding bleeding in in thethe anterioranterior chamber chamber occurred occurred in inthe the vehicle vehicle group group (Figure (Figure 1b). 1Onb). the On other the hand, other hand,the corneal the corneal opacity opacity and neovascularization and neovascularization were less were obvious less obvious in all PPAR in all instillation PPAR instillation groups (Figure groups (Figure1c–e).1 c–e).Furthermore, Furthermore, corneal cornealopacity opacitywas minimal was minimalin the cornea in the in the cornea PPAR inα the+γ group PPAR (Figureα+γ group 1e). (FigureMicroscopically,1e). Microscopically, various infiltrating various infiltratinginflammatory inflammatory cells, including cells, macrophages including macrophages and neutrophils, and neutrophils,were seen werein the seen peripheral in the peripheral cornea at 6 cornea h after at the 6 h alkali after theburn alkali in all burn groups in all (Figure groups 2b,e,h,k). (Figure2 Byb,e,h,k). day By7, day the 7, infiltration the infiltration of inflammatory of inflammatory cells had cells decreas had decreaseded in the inperipheral the peripheral cornea cornea (Figure (Figure 2c,f,i,l),2c,f,i,l), and andthese these cells cells had had migrated migrated to to the the center center of of the the cornea cornea (Figure (Figure 2a,d,g,j).2a,d,g,j). In In addition, addition, neovascularization neovascularization waswas observed observed in in the the corneal corneal limbus limbus (Figure (Figure2 c,f,i,l).2c,f,i,l). AllAll PPARPPAR groupsgroups showedshowed a lesser degree degree of of infiltratinginfiltrating inflammatory inflammatory cells cells and and neovascularization neovascularization comparedcompared toto thethe vehicle group group on on day day 7. 7.

Figure 1. Anterior segment photographs of the cornea in each group after alkali burn. In the cornea immediatelyFigure 1. Anterior after the segment burn, blood photographs vessels of thethe iriscornea disappeared, in each group and af ater circular alkali epithelialburn. In the defect cornea line wasimmediately observed ( aafter). Anterior the burn, chamber blood vessels bleeding of the occurred iris disappeared, in the vehicle and a group circular (b ).epithelial The central defect opacity line was observed (a). Anterior chamber bleeding occurred in the vehicle group (b). The central opacity and neovascularization of the cornea were less severe in all peroxisome proliferator-activated receptor and neovascularization of the cornea were less severe in all peroxisome proliferator-activated receptor (PPAR) groups than in the vehicle group on day 7 (c–e). In the PPARα+γ group in particular, the (PPAR) groups than in the vehicle group on day 7 (c–e). In the PPARα+γ group in particular, the corneal opacity was quite slight (e). corneal opacity was quite slight (e).

Int. J. Mol. Sci. 2020, 21, 5093 3 of 16

Int.Int. J.J. Mol.Mol. Sci.Sci. 20202020,, 21,, xx FORFOR PEER REVIEW 3 3 of of 16 16

FigureFigure 2. 2.Histological Histological analysis analysis followingfollowing hematoxylin and and eosin eosin (HE) (HE) staining. staining. The The central central andand Figure 2. Histological analysis following hematoxylin and eosin (HE) staining. The central and peripheralperipheral cornea cornea of of the the vehicle vehicle group group ((aa––cc),), PPARPPARα groupgroup (d (d––f),f), PPAR PPARγ γgroupgroup (g– (gi),– iand), and PPAR PPARα+γα +γ peripheral cornea of the vehicle group (a–c), PPARα group (d–f), PPARγ group (g–i), and PPARα+γ groupgroup (j– l()j– atl) 6at h6 andh and day day 7. 7. Bar, Bar, 50 50µ μm.m. VariousVarious inflammatoryinflammatory cells cells in infiltratedfiltrated from from the the peripheral peripheral group (j–l) at 6 h and day 7. Bar, 50 μm. Various inflammatory cells infiltrated from the peripheral corneacornea at 6at h 6 andh and migrated migrated to to the the central central corneacornea by day day 7. 7. Furthermore, Furthermore, lumina luminall structures structures that that appear appear cornea at 6 h and migrated to the central cornea by day 7. Furthermore, luminal structures that appear to be blood vessels were seen in the peripheral cornea on day 7. In all PPAR instillation groups, low to beto be blood blood vessels vessels were were seen seen in in the the peripheral peripheral corneacornea on on day day 7. 7. In In all all PPAR PPAR instillation instillation groups, groups, low low numbers of infiltrating cells and luminal structures, and corneal stromal edema were seen. numbersnumbers of of infiltrating infiltrating cells cells and and luminal luminal structures, structures, and corneal stromal stromal edema edema were were seen. seen. 2.2.2.2. PPAR PPARα andα and PPAR PPARγ mRNAγ mRNA Expression Expression in in the the CorneaCornea 2.2. PPARα and PPARγ mRNA Expression in the Cornea Real-time reverse transcription polymerase chain reaction (RT-PCR) was performed to Real-timeReal-time reverse reverse transcription transcription polymerase polymerase chain chain reaction reaction (RT-PCR) (RT-PCR) was performed was performed to investigate to investigate mRNA expression levels of PPARα and PPARγ in the burned cornea at 6 h after mRNAinvestigate expression mRNA levels expression of PPAR levelsα and of PPAR PPARγ αin and the burnedPPARγ corneain the atburned 6 h after cornea instillation at 6 h after of each PPARinstillation (Figure3 of). PPAReach PPARα and (Figure PPARα +3).γ treatmentPPARα and increased PPARα+ PPARγ treatmentα mRNA increased expression PPAR comparedα mRNA with instillationexpression ofcompared each PPAR with (Figure the vehicle 3). PPAR and αPPAR andγ PPAR treatmentα+γ treatment(Figure 3a). increased PPARγ PPARand PPARα mRNAα+γ the vehicle and PPARγ treatment (Figure3a). PPAR γ and PPARα+γ treatment also increased PPARγ expressiontreatment also compared increased with PPAR theγ vehiclemRNA andexpression PPAR γin treatment comparison (Figure to the 3a).vehicle PPAR groupγ and (Figure PPAR 3b).α+ γ mRNAtreatment expression also increased in comparison PPARγ tomRNA the vehicle expression group in comparison (Figure3b). to the vehicle group (Figure 3b).

Figure 3. Expression of PPARα and PPARγ mRNA. At 6 h, real-time RT-PCR showed upregulation Figure 3. Expression of PPARα and PPARγ mRNA. At 6 h, real-time RT-PCR showed upregulation of Figureof mRNA 3. Expression expression of levels PPAR ofα andPPAR PPARα inγ themRNA. alkali-burned At 6 h, real-time cornea RT-PCRafter PPAR showedα and upregulation PPARα+γ mRNA expression levels of PPARα in the alkali-burned cornea after PPARα and PPARα+γ treatment oftreatment mRNA comparedexpression with levels the vehicleof PPAR andα thein thePPAR alkali-burnedγ group (a). Incornea addition, after mRNA PPAR expressionα and PPAR levelsα+γ γ γ comparedtreatmentof PPAR withγ comparedwere the increased vehicle with andinthe the vehicle the PPAR PPAR andγ and thegroup PPAR PPAR (αa+γ).γ group Ingroups addition, (a in). In comparison addition, mRNA expression mRNAto the vehicle expression levels group of levels PPAR(b). wereofData PPAR increased areγ presented were in increased the as PPAR the inmeanγ andthe PPAR± PPARstandardγα and+γ error PPARgroups (nα =+ inγ8 samples/group).groups comparison in comparison to the**p < vehicle 0.01, to the *p vehicle group< 0.05. (groupb). Data (b). are presentedData are as presented the mean as thestandard mean ± errorstandard (n = error8 samples (n = 8 /samples/group).group). **p < 0.01, **p < *p 0.01,< 0.05. *p < 0.05. ±

Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 4 of 16

2.3. Anti-Inflammatory Effects of the PPAR Agonist Ophthalmic Solution To investigate infiltrating neutrophils, naphthol AS-D chloroacetate esterase (EST) staining was performed. EST-positive neutrophils were noted in the corneal limbus at 6 h and increased by day 1 after the alkali burn (Figure 4). On day 1, the number of neutrophils peaked in the injured cornea in all groups. The ophthalmic solution of PPARγ and PPARα+γ agonists decreased the degree of infiltrating neutrophils compared to vehicle treatment at 6 h. Instillation of all PPAR agonists reduced the number of neutrophils in the cornea on day 1 (Figure 4e). Real-time RT–PCR analysis was performed at 6 h after alkali injury to assess the expression of genes associated with proinflammatory cytokines, including interleukin-1β (IL-1β), IL-6, nuclear factor-kappa B (NF-κB), nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha (IκB-α), and tumor necrosis factor-α (TNF-α). All PPAR treatments equally suppressed expression of IL-1β and IL-6 mRNA (Figure 5a,b). Instillation of PPARγ and PPARα+γ agonists also reduced the mRNA expression of TNF-α and NF-κB in comparison with vehicle instillation (Figure 5c,d). On the other hand, only PPARα treatment increased the mRNA expression of IκB-α compared with vehicle treatment (Figure 5e). Therefore, we performed immunohistochemical analysis to investigate expression of NF-κB and IκB-α in corneal epithelium cells at 6 h and four days after the injury (Figure 6). NF-κB and IκB-α staining was observed at a similar location in the corneal epithelium. At 6 h after the injury, NF-κB expression was high and localized in the nuclear area (Figure 6a), and IκB-α was slightly expressed in the cytoplasm in the vehicle group (Figure 6b). PPARα instillation significantly enhanced IκB-α expression (Figure 6d). In the PPARγ and PPARα+γ treatment groups, less NF-κB expression was present than in the vehicle group and the PPARα group (Figure 6e,g). At four days after the injury, epithelial proliferation was observed in the vehicle group, and NF-κB and IκB-α expression was localized in nearly identical cells (Figure 6i,j). After PPARα treatment, NF-κB was mainly expressed Int. J. Mol. Sci. 2020, 21, 5093 4 of 16 in the cytoplasm, and IκB-α expression was localized in the nuclear area (Figure 6k,l). PPARγ instillation reduced NF-κB expression, and little IκB expression was seen in comparison with 2.3.instillation Anti-Inflammatory of vehicle E fforects PPAR of theα PPAR (Figure Agonist 6m,n). Ophthalmic PPARα+ Solutionγ treatment strongly decreased NF-κB expression compared to the other three groups (Figure 6o,p). ToWe investigate also investigated infiltrating macrophage neutrophils, infiltration naphthol using AS-D immunohistochemical chloroacetate esterase analysis (EST) staining with CD68 was performed.antibody (ED-1) EST-positive and CD163 neutrophils antibody were (ED-2) noted (Figur in thee 7). corneal In the limbusvehicle atgroup, 6 h and the increased number of by ED-1- day 1 afterpositive the alkali macrophages burn (Figure increased4). On in day the 1, peripheral the number cornea of neutrophils at 6 h after peaked the injury, in the and injured peaked cornea on day in all 1, groups.similar Theto EST-positive ophthalmic solutionneutrophils. of PPAR PPARγ andγ and PPAR PPARα+αγ+γagonists treatment decreased significantly the degree suppressed of infiltrating ED-1- neutrophilspositive macrophages compared to on vehicle day 1 treatmentcompared atwith 6 h. vehicle Instillation treatment of all (Figure PPAR agonists 7c). On reducedthe other the hand, number ED- of2-positive neutrophils cells in increased the cornea in on the day PPAR 1 (Figureγ and 4PPARe). α+γ groups on day 4 (Figure 7f).

Figure 4. Esterase (EST) staining was performed to evaluate inﬁltrating cells in the rat cornea after

alkali injury. Immunohistochemical images of the corneal limbus at 6 h are shown (a–d). Bar, 50 µm. The black arrow indicates EST-positive cells (a). PPARγ and PPARα+γ treatment reduced the number of neutrophils in the peripheral cornea at 6 h. Bar chart of the number of EST-positive cells shows a statistically significant difference between vehicle and PPAR treatments on day 1 (e). At 6 h after alkali burn, instillation of PPARγ and PPARα+γ agonists significantly suppressed neutrophils compared with vehicle and PPARα instillation (e). Data are presented as the mean standard error (n = 8 ± samples/group). **p < 0.01, *p < 0.05.

Real-time RT–PCR analysis was performed at 6 h after alkali injury to assess the expression of genes associated with proinﬂammatory cytokines, including interleukin-1β (IL-1β), IL-6, nuclear factor-kappa B (NF-κB), nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha (IκB-α), and tumor necrosis factor-α (TNF-α). All PPAR treatments equally suppressed expression of IL-1β and IL-6 mRNA (Figure5a,b). Instillation of PPAR γ and PPARα+γ agonists also reduced the mRNA expression of TNF-α and NF-κB in comparison with vehicle instillation (Figure5c,d). On the other hand, only PPARα treatment increased the mRNA expression of IκB-α compared with vehicle treatment (Figure5e). Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 5 of 16

Figure 4. Esterase (EST) staining was performed to evaluate infiltrating cells in the rat cornea after alkali injury. Immunohistochemical images of the corneal limbus at 6 h are shown (a–d). Bar, 50 μm. The black arrow indicates EST-positive cells (a). PPARγ and PPARα+γ treatment reduced the number of neutrophils in the peripheral cornea at 6 h. Bar chart of the number of EST-positive cells shows a statistically significant difference between vehicle and PPAR treatments on day 1 (e). At 6 h after alkali burn, instillation of PPARγ and PPARα+γ agonists significantly suppressed neutrophils compared with vehicle and PPARα instillation (e). Data are presented as the mean ± standard error (n = 8 Int. J. Mol. Sci. 2020, 21, 5093 5 of 16 samples/group). **p < 0.01, *p < 0.05.

FigureFigure 5. Expression 5. Expression of ofmRNA mRNA for for proinflammatory proinﬂammatory cytokines cytokines in thein corneathe cornea at 6 h at after 6 h alkali after burn. alkali We burn. We measuredmeasured mRNA mRNA expression expression levels levels of IL-1 ofβ (IL-1a), IL-6β (a (b),), IL-6 TNF- (αb),(c ),TNF- NF-καB( (dc), andNF- IκBB- α(d(),e). and Treatment IκB-α (e). Treatmentwith all with PPAR all agonists PPAR equallyagonists suppressed equally suppressed expression of expression IL-1β and IL-6 of mRNAIL-1β and (a and IL-6b). mRNA Instillation (a and of b). γ α γ α κ InstillationPPAR ofand PPAR PPARγ +andagonists PPARα also+γ reducedagonists the also mRNA reduced expression the mRNA of TNF- expressionand NF- Bof in TNF- comparisonα and NF- with vehicle instillation (c), (d). On the other hand, only PPARα treatment increased the mRNA κB in comparison with vehicle instillation (c), (d). On the other hand, only PPARα treatment increased expression of IκB-α compared with vehicle treatment (e). Data are presented as the mean standard the mRNA expression of IκB-α compared with vehicle treatment (e). Data are presented± as the mean error (n = 8 samples/group). **p < 0.01, *p < 0.05. ± standard error (n = 8 samples/group). **p < 0.01, *p < 0.05. Therefore, we performed immunohistochemical analysis to investigate expression of NF-κB and IκB-α in corneal epithelium cells at 6 h and four days after the injury (Figure6). NF- κB and IκB-α staining was observed at a similar location in the corneal epithelium. At 6 h after the injury, NF-κB expression was high and localized in the nuclear area (Figure6a), and I κB-α was slightly expressed in the cytoplasm in the vehicle group (Figure6b). PPAR α instillation signiﬁcantly enhanced IκB-α expression (Figure6d). In the PPAR γ and PPARα+γ treatment groups, less NF-κB expression was present than in the vehicle group and the PPARα group (Figure6e,g). At four days after the injury, epithelial proliferation was observed in the vehicle group, and NF-κB and IκB-α expression was localized in nearly identical cells (Figure6i,j). After PPAR α treatment, NF-κB was mainly expressed in the cytoplasm, and IκB-α expression was localized in the nuclear area (Figure6k,l). PPAR γ instillation reduced NF-κB expression, and little IκB expression was seen in comparison with instillation of vehicle or PPARα (Figure6m,n). PPAR α+γ treatment strongly decreased NF-κB expression compared to the other three groups (Figure6o,p).

Int. J. Mol. Sci. 2020, 21, 5093 6 of 16 Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 6 of 16

FigureFigure 6. 6. ImmunostainingImmunostaining for NF- for NF-κB andκB I andκΒ-α I κinB- epitheliumα in epithelium cells in cellseach group in each at group6 h and at four 6 h days and four afterdays the after injury. the injury.Representative Representative sections sections at 6 h (a at–h 6) hand (a –fourh) and days four (i–p days) are (shown.i–p) are At shown. 6 h after At the 6 h after injury,the injury, NF-κ NF-B wasκB wasstrongly strongly expressed expressed in the in cell the nuclei cell nuclei in the in vehicle the vehicle group group (a), whereas (a), whereas it was it was expressedexpressed at at low low levels levels in in the the nuclei nuclei in inthe the PPAR PPARα groupα group (c). (Icκ).B- Iακ B-wasα was expressed expressed in the in cytoplasm the cytoplasm afterafter PPAR PPARαα instillationinstillation (d ().d ).In Inthe the PPAR PPARγ andγ and PPAR PPARα+γα +groups,γ groups, less lessNF-κ NF-B expressionκB expression was seen was seen comparedcompared to to the the vehicle vehicle group group and and the the PPAR PPARα groupα group (e),( e(g),). ( Theg). The PPAR PPARα+γα group+γ group exhibited exhibited a lesser alesser degreedegree of of NF- NF-κκB-positiveB-positive inflammatory inﬂammatory cell cell infiltration inﬁltration compared compared to the to other the other three three groups groups on day on 4 day 4 afterafter the the injury injury (i (–ip–p). ).Bar, Bar, 50 50 μmµm (a– (ap–).p ).

We also investigated macrophage inﬁltration using immunohistochemical analysis with CD68 antibody (ED-1) and CD163 antibody (ED-2) (Figure7). In the vehicle group, the number of ED-1-positive macrophages increased in the peripheral cornea at 6 h after the injury, and peaked on day 1, similar to EST-positive neutrophils. PPARγ and PPARα+γ treatment signiﬁcantly suppressed ED-1-positive macrophages on day 1 compared with vehicle treatment (Figure7c). On the other hand, ED-2-positive cells increased in the PPARγ and PPARα+γ groups on day 4 (Figure7f).

2.4. Suppression of Neovascularization by Instillation of PPAR Agonists Next, immunohistochemical analysis of nestin and aminopeptidase P (JG12) was conducted to investigate neovascularization. Nestin-positive newly formed endothelial cells were observed in the peripheral cornea on day 4 and day 7 after the alkali burn (Figure8a–c). Although each PPAR agonist ophthalmic solution reduced nestin-positive endothelial cells on day 7, the PPARα agonist suppressed neovascularization from the early phase on day 4 (Figure8c). JG-12-positive capillary lumens were observed on day 4, and these were increased by day 7 after the corneal alkali injury in the vehicle group. The diﬀerent PPAR treatments equally suppressed the increase in capillary lumens by day 7 (Figure8f).

Int. J. Mol. Sci. 2020, 21, 5093 7 of 16 Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 7 of 16

FigureFigure 7. Infiltration7. Infiltration of macrophages of macrophages in the cornea in the after corn alkaliea injury. after Representativealkali injury. photomicrographs Representative ofphotomicrographs infiltrating macrophages of infiltrating in the macrophages corneal limbus in the after corneal instillation limbus ofafter the instillation vehicle (a of) and the vehicle PPARα (+aγ) (band) on PPAR day 1α+ showedγ (b) on prominentday 1 showed infiltration prominent of infiltration ED-1-positive of ED-1-positive macrophages macrophages on day 1 in on the day vehicle 1 in group,the vehicle but significantly group, but significantly suppressed suppressed infiltration ininfiltration the PPAR inα +theγ groupPPARα (+cγ). group Black ( arrowc). Black indicates arrow anindicates ED-1-positive an ED-1-positive cell (a). PPAR cell (γa).and PPAR PPARγ andα+ γPPARtreatmentα+γ treatment significantly significantly suppressed suppressed ED-1-positive ED-1- macrophages on day 1 compared with vehicle treatment (c). Data are presented as the mean standard positive macrophages on day 1 compared with vehicle treatment (c). Data are presented as± the mean error± standard (n = 8 samples error (n/group). = 8 samples/group). *p < 0.05. Although *p < 0.05. few Although ED-2-positive few ED-2-p M2 macrophagesositive M2 macrophages were present inwere the peripheralpresent in cornea the peripheral on day 4 aftercornea alkali on burnday in4 theafter vehicle alkali groupburn (ind ),the PPAR vehicleα+γ treatmentgroup (d), upregulated PPARα+γ M2treatment macrophages upregulated (e). Red M2 arrowsmacrophages indicate (e). ED-2-positive Red arrows indicate cells ( dED-2-positive), (e). Bar chart cells of (d), the (e). number Bar chart of ED-2-positiveof the number cells of ED-2-positive shows a statistically cells shows significant a statistically difference significant between difference the vehicle between and PPAR theγ vehiclegroups, similarand PPAR to theγ groups, PPARα similar+γ instillation to the PPAR groupα+γ (f instillation). Bar, 50 µ groupm (a), ( fb).), Bar, (d), 50 (e ).μm Data (a), are(b), presented(d), (e). Data as are the meanpresentedstandard as the errormean ( n± standard= 8 samples error/group). (n = 8 samples/group). **p < 0.01, *p < 0.05. **p < 0.01, *p < 0.05. ±

2.4. Real-timeSuppression RT–PCR of Neovascularization analysis was by performed Instillation inof PPAR the cornea Agonists at 6 h after the alkali burn to assess the expression of genes associated with neovascularization, including vascular endothelial growth Next, immunohistochemical analysis of nestin and aminopeptidase P (JG12) was conducted to factor-A (VEGF-A), angiopoietin-1 (Ang-1), and Ang-2 (Figure9a–c). A significant di fference was investigate neovascularization. Nestin-positive newly formed endothelial cells were observed in the present in the mRNA expression levels of VEGF-A between each PPAR treatment group and the peripheral cornea on day 4 and day 7 after the alkali burn (Figure 8a–c). Although each PPAR agonist vehicle group (Figure9a). Furthermore, instillation of all PPAR agonists significantly downregulated ophthalmic solution reduced nestin-positive endothelial cells on day 7, the PPARα agonist mRNA expression of Ang-2 at 6 h (Figure9c). No significant di fferences were found in Ang-1 mRNA suppressed neovascularization from the early phase on day 4 (Figure 8c). JG-12-positive capillary expression among all groups (Figure9b). The expression level of Ang-2 mRNA was comparatively lumens were observed on day 4, and these were increased by day 7 after the corneal alkali injury in higher than that of Ang-1. the vehicle group. The different PPAR treatments equally suppressed the increase in capillary lumens by day 7 (Figure 8f). Real-time RT–PCR analysis was performed in the cornea at 6 h after the alkali burn to assess the expression of genes associated with neovascularization, including vascular endothelial growth factor-A (VEGF-A), angiopoietin-1 (Ang-1), and Ang-2 (Figure 9a–c). A significant difference was present in the mRNA expression levels of VEGF-A between each PPAR treatment group and the vehicle group (Figure 9a). Furthermore, instillation of all PPAR agonists significantly downregulated mRNA expression of Ang-2 at 6 h (Figure 9c). No significant differences were found in Ang-1 mRNA expression among all groups (Figure 9b). The expression level of Ang-2 mRNA was comparatively higher than that of Ang-1.

Int. J. Mol. Sci. 2020, 21, 5093 8 of 16 Int.Int. J.J. Mol.Mol. Sci.Sci. 20202020,, 2121,, xx FORFOR PEERPEER REVIEWREVIEW 8 8 of of 16 16

FigureFigureFigure 8. 8.8.Immunohistochemical ImmunohistochemicalImmunohistochemical analysisananalysisalysis of of nestin nestin (( aa––ccc))) andand and JG12JG12 JG12 ((d (dd––f–f))f )inin in thethe the vehiclevehicle vehicle ((aa), (),a ),((dd ())d andand) and PPARPPARPPARα+αγα++(γγb (),(bb (),),e ()(ee groups.)) groups.groups. Bar, Bar,Bar, 50 5050µ μμm.m.m. Nestin-positive Nestin-positiveNestin-positive endothelialendothelial cellscells cells werewere were observedobserved observed duringduring during thethe the earlyearly early phasephasephase in in thein thethe corneal cornealcorneal limbus limbuslimbus (a ((),aa),), ( b ((b).b).). The TheThe black blackblack arrowarrowarrow indicatesindicates nestin-positivenestin-positive endothelialendothelial endothelial cellscells cells ofof ofaa a newnewnew blood bloodblood vessel vesselvessel (a ().(aa).). Bar BarBar charts chartscharts of ofof the thethe number numbernumber ofofof nestin-positivenestin-posnestin-positiveitive cellscells cells indicateindicate indicate thatthat that allall all PPARPPAR PPAR agonistagonist agonist ophthalmicophthalmicophthalmic solutions solutionssolutions reduced reducedreduced nestin-positive nestin-positivenestin-positive endothelial endotheendotheliallial cells cells inin thethe the laterlater later phasephase phase (day(day (day 7).7). 7). InInIn thethe the earlyearly early phasephasephase (day (day(day 4), newly4),4), newlynewly formed formedformed blood bloodblood vessels vesselsvessels were werewere signiﬁcantly significantlysignificantly lower lowerlower in the inin PPAR thethe PPARPPARα andαα PPAR andand PPARαPPAR+γ groupsαα++γγ thangroupsgroups in the thanthan vehicle inin the groupthe vehiclevehicle (c). Data groupgroup are ((cc presented).). DataData areare as presentedpresented the mean asas standard thethe meanmean error ±± standardstandard (n = 8 samples errorerror ((n/ngroup). == 88 ± **psamples/group).samples/group).< 0.01, *p < 0.05. ****pp < JG-12-positive< 0.01,0.01, **pp << 0.05.0.05. capillaryJG-12-positiveJG-12-positive lumens capillarycapillary were lumenslumens observed werewere on observedobserved day 7 (d onon), ( daydaye). 7 Red7 ((dd),), arrow ((ee).). indicatesRedRed arrowarrow a JG12-positive indicatesindicates aa JG12-positiveJG12-positive cell (d). The cellcell number ((dd).). TheThe of numbernumber capillaries ofof capillariescapillaries increased increasedincreased by day byby 7 dayday after 77 afterafter corneal cornealcorneal alkali injuryalkalialkali in injuryinjury the vehicle inin thethe group. vehiclevehicle Each group.group. PPAR EachEach treatment PPARPPAR trtreatmenteatment signiﬁcantly significantlysignificantly suppressed suppressedsuppressed the increase thethe increaseincrease in capillary inin lumenscapillarycapillary by day lumenslumens 7 compared byby dayday 7 to7 comparedcompared the vehicle toto group thethe vehiclevehicle (f). Data groupgroup are ( presented(ff).). DataData areare as presented thepresented mean asasstandard thethe meanmean error ±± ± (n =standardstandard8 samples errorerror/group). ((nn == 88 samples/group).samples/group). **p < 0.01, *p < 0.05. ****pp << 0.01,0.01, **pp << 0.05.0.05.

FigureFigureFigure 9. 9.9.Expression ExpressionExpression of ofof mRNA mRNAmRNA for forfor molecules moleculesmolecules involved involvedinvolved inin neovascularization.neovascularization.neovascularization. Real-timeReal-time Real-time RT–PCRRT–PCR RT–PCR analysisanalysisanalysis was waswas performed performedperformed to toto investigate investiginvestigateate the thethe level levellevel ofofof VEGF-AVEGF-A mRNA mRNA (( (aaa),),), Ang-1Ang-1 Ang-1 mRNAmRNA mRNA ((bb (),b), ),andand and Ang-2Ang-2 Ang-2 mRNAmRNAmRNA (c ) ((c inc)) inin the thethe cornea corneacornea at atat 6 6 h6 hh after afterafter alkali alkalialkali burn. burn.burn. InstillationInstInstillationillation ofof allall all PPARPPAR PPAR agonistsagonists agonists suppressedsuppressed suppressed mRNAmRNA mRNA expressionexpressionexpression of VEGF-Aofof VEGF-AVEGF-A (a ). ((aa All).). AllAll PPAR PPARPPAR treatments treatmtreatmentsents significantly significantlysignificantly reduced reducedreduced mRNA mRNAmRNA expression expressionexpression of ofof Ang-2 Ang-2Ang-2 at atat 6 h (c).66 In hh contrast,((cc).). InIn contrast,contrast, no significant nono significantsignificant differences differencesdifferences were seen werewere in seenseen the degreeinin thethe degreedegree of Ang-1 ofof mRNAAng-1Ang-1 mRNAmRNA expression expressionexpression between eachbetweenbetween group eacheach (b). Thegroupgroup mRNA ((bb).). TheThe expression mRNAmRNA expressionexpression level of Ang-2 levellevel of mRNAof Ang-2Ang-2 was mRNAmRNA relatively waswas relativelyrelatively higher than higherhigher that thanthan of Ang-1thatthat mRNA.ofof Ang-1Ang-1 Data mRNA.mRNA. are presented DataData areare as presentedpresented the mean asas thethestandard meanmean ±± error standardstandard (n = error8error samples ((nn == 88/group). samples/group).samples/group). **p < 0.01, ****pp * p << < 0.01,0.01,0.05. **pp << 0.05.0.05. ±

Int. J. Mol. Sci. 2020, 21, 5093 9 of 16

2.5. Anti-Fibrotic Effects of PPAR Agonist Instillation Int. J. Mol. Sci. 2020, 21, x FOR PEER REVIEW 9 of 16 Immunohistochemical analysis of collagen type III was performed to evaluate fibrotic changes at seven days2.5. after Anti-Fibrotic the alkali Effects injury of PPAR (Figure Agonist 10 Instillationa–f). On day 7 after injury, we noted the accumulation of collagen typeImmunohistochemical III in the corneal stroma analysis (Figure of collagen 10a–d). type III All was PPAR performed instillation to evaluate groups fibrotic exhibited changes a lower volume ofat collagen seven days type after IIIthe (Figure alkali injury 10b–d) (Figure and 10a–f). a decrease On day 7 inafter the injury, area we of noted collagen the accumulation type III expression of collagen type III in the corneal stroma (Figure 10a–d). All PPAR instillation groups exhibited a comparedlower to the volume vehicle of collagen (Figure type 10 e).III (Figure Interestingly, 10b–d) and percentages a decrease in of the collagen area of collagen type III type in theIII corneal regions wereexpression significantly compared lower to the vehicle in the (Figure PPAR 10e).α+γ Interegroupstingly, compared percentages to eitherof collagen PPAR type alone III in the group. We also investigatedcorneal regions expression were significantly of mRNA forlower transforming in the PPARα growth+γ group factor- comparedβ (TGF- to eitherβ), whichPPAR alone is involved in the fibroticgroup. reaction. We also Instillation investigated of expression PPARγ ofand mRNA PPAR forα transforming+γ agonists growth reduced factor- theβ (TGF- mRNAβ), which expression of is involved in the fibrotic reaction. Instillation of PPARγ and PPARα+γ agonists reduced the mRNA β TGF- 1 inexpression comparison of TGF- withβ1 in vehicle comparison instillation with vehicle (Figure instillation 10f). (Figure 10g).

Figure 10.FigureImmunohistochemical 10. Immunohistochemical analysis analysis of of collagen collagen type type IIIIII to toevaluate evaluate fibrotic ﬁbrotic changes changes at seven at seven days afterdays the alkaliafter the injury alkali injury (a–f). (a Collagen–f). Collagen type type III III waswas observed observed in the in theinjured injured area du arearing duringthe healing the healing process in the vehicle group (a). Instillation of each PPAR reduced collagen type III expression (b–d). Bar chart of percentages of collagen type III expression area/cornea area showed that the PPARα+γ group had a signiﬁcantly lower percentage of collagen type III compared to other groups (e). Real-time RT-PCR showed marked suppression of TGF-β mRNA expression in the burned cornea after PPARγ or PPARα+γ agonist treatment compared with the vehicle group (f). Bar, 100 µm. Data are presented as the mean standard error (n = 8 samples/group). **p < 0.01, *p < 0.05. ± Int. J. Mol. Sci. 2020, 21, 5093 10 of 16

3. Discussion Corneal alkaline injury is one of the most severe ophthalmic emergencies and requires prompt treatment because it usually results in a serious outcome accompanied with corneal neovascularization and opacity [2,12]. Topical therapy with PPAR agonists is expected to be a new treatment strategy for alkali-burned corneas [13]. Our present study demonstrated the following: (1) PPARα and PPARγ agonist treatment upregulated the mRNA expression of PPARα and PPARγ, respectively. In addition, PPARα+γ agonist instillation increased the mRNA expression of both PPARα and PPARγ. (2) Administration of all topical PPAR agonists suppressed infiltrating inflammatory cells in the early phase after the alkali burn and mRNA expression of proinflammatory cytokines. On the other hand, PPARγ and PPARα+γ significantly increased the number of M2 macrophages on day 4. (3) Instillation of all PPAR agonists prevented neovascularization by reducing VEGF-A and Ang-2 expression. The strong inflammation induced by alkali injury to the mouse cornea leads to corneal epithelial defects and upregulation of various cytokines such as IL-1β and IL-6, which play important roles in wound healing after an alkali burn [14]. In the present study, the PPARα agonist and PPARγ agonist significantly suppressed these cytokines, suggesting that both agonists induced significant anti-inflammatory effects in the alkali-burned cornea. In addition, the PPARγ agonist significantly reduced mRNA expression of TNF-α and NF-κB, whereas the PPARα agonist increased mRNA expression of IκB-α. TNF-α is involved in the inflammatory pathway, and thus decreased TNF-α mRNA expression by PPARγ treatment seems effective in corneal wound healing after alkali injury. Furthermore, NF-κB plays a major role in regulation of the expression of a number of genes involved in cell growth, inflammation, and apoptosis [15,16]. Both the PPARα agonist and PPARγ agonist induce IκB[17–20], which is the inhibitory protein of NF-κB. IκB inhibits the effect of NF-κB in the cytoplasm by blocking NF-κB binding to DNA [16]. In this study, only PPARα increased IκB-α expression at 6 h after the alkali injury, suggesting downregulation of NF-κB. Immunohistochemical analysis of NF-κB and IκB-α showed that both NF-κB and IκB-α were highly expressed in the cytoplasm after PPARα instillation in the early phase (Figure6). On the other hand, immunostaining revealed that PPAR γ instillation strongly reduced NF-κB expression, but rarely activated IκB-α expression. These results suggested that PPARγ has an anti-inflammatory effect by reducing NF-κB expression without upregulation of IκB. Thus, PPARα agonists and PPARγ agonists may have a different mechanism of anti-inflammatory effects. Macrophages are classified as either classically activated (M1) or alternatively activated (M2) cells and can change their subtype [21]. M1 macrophages are tissue injury-type macrophages involved in the development of inflammation [22]. Conversely, M2 macrophages play immunoregulatory and immunosuppressive roles. M2 macrophages reduce inflammation by producing anti-inflammatory factors, and they affect tissue remodeling and repair [23]. Activation of PPARγ polarizes monocytes to become M2 macrophages in the circulating blood [24]. Our previous study also revealed that a PPARγ agonist was one factor underlying the M1/M2 balance [8]. Our current study showed an increase in M2 macrophages after PPARγ treatment on day 4, suggesting that administration of the PPARγ agonist played another critical role in tissue remodeling and wound healing. Neovascularization is mediated by molecular factors. VEGF-A is an important molecule for angiogenesis [25]. Ang-1 and Ang-2 are the ligands of the Tie2 receptor. Ang-1 is the major ligand for the Tie2 receptor and may induce maturation and stabilization of developing blood vessels [26]. On the other hand, Ang-2 has opposite effects of Ang-1 and may cause destabilization of developing blood vessels [27]. Although blood vessel formation is regulated by VEGF along with Ang-1 and Ang-2, neither Ang-1 nor Ang-2 alone can promote angiogenesis [28]. The combination of Ang-1 and VEGF increases macroscopically evident perfusion of corneal neovascular structures. In contrast, the combination of Ang-2 and VEGF promotes neovascularization [29]. PPARα and PPARγ agonists prevent neovascularization in the eye by suppressing inflammatory factors and neovascularization factors [9,30,31]. In our present study, both the PPARα and PPARγ agonists reduced expression of VEGF and Ang-2 mRNA at 6 h after the alkali burn. The number of immature lumens on day 4 and Int. J. Mol. Sci. 2020, 21, 5093 11 of 16 mature blood vessels on day 7 after instillation of PPARα and PPARγ agonists was lower than after vehicle instillation. These results indicated that neovascularization was prevented by suppressing the levels of VEGF and Ang-2 mRNA expression at 6 h after the alkali injury. Regarding the PPARα agonist, immature juvenile blood vessels were suppressed from the early phase. The normal cornea mainly consists of collagen type I, the expression of which is tightly regulated, that maintains transparency [32]. Collagen type III plays an important role in corneal wound healing, and unregulated accumulation of collagens can result in an opaque cornea [33]. The present study showed deposition of collagen type III in the cornea at seven days after the alkali injury and that inflammation suppressed by PPARα or PPARγ instillation resulted in less scar formation. TGF-β is also an important profibrotic element, especially in fibrosis [34,35]. Rosglitazone, which is a PPARγ agonist, suppresses TGF-β1 expression and prevents conjunctival fibrosis after glaucoma filtration surgery [36]. In our current study, PPARγ treatment suppressed fibrotic changes in the corneal stroma, which may be induced by anti-TGF-β effects of PPARγ. A few studies have reported the efficiency of combined PPARα and PPARγ agonist therapy for diabetes and dyslipidemia [10,11]. The combination of half doses of fenofibrate and pioglitazone improves nonalcoholic steatohepatitis-related disturbances, similar to a full dose of fenofibrate [11]. In another report, pre-treatment with fenofibrate, pioglitazone, and their combination in rats with cerebral ischemia improves inflammatory and apoptotic markers, such as TNF-α [37]. However, little has been reported on the effects of the combination of PPARα and PPARγ agonist therapy in the field of ophthalmology. In this study, subtle differences were observed in the anti-inflammatory and anti-angiogenic effects of each PPAR agonist. Fenofibrate, a PPARα agonist, suppressed angiogenesis from the early phase, whereas pioglitazone, a PPARγ agonist, enhanced the anti-inflammatory effect by upregulating M2 macrophages. Furthermore, the combination of fenofibrate and pioglitazone significantly prevented corneal accumulation of collagen type III compared to either alone. NF-κB expression in the corneal epithelium was strongly reduced by the combination of PPARα and PPARγ at four days after the injury. A different mechanism involved in NF-κB and IκB-α expression after fenofibrate and pioglitazone treatment may contribute to the significant downregulation of NF-κB, thereby suppressing scar formation. Therefore, instillation of a drug combination may be more effective than a single PPAR agonist for corneal alkali burn. Recently, a few reports have been published on a PPARα/γ dual agonist [38,39]. Kaur et al. reported that trans-acrylic acid derivatives have binding affinity for PPARα and PPARγ and significantly reduce the plasma glucose level and total cholesterol level in diabetic rats compared to pioglitazone [39]. Further studies are needed to investigate the therapeutic effect of a dual agonist following corneal alkali burn.

4. Materials and Methods

4.1. Animals and Ethics Statement Eight-week-old male Wistar rats (n = 8 per group/per endpoint; Sankyo Laboratory Service, Tokyo, Japan) were used for all experiments in the present study. All animal experiments were performed in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vison Research. The Experimental Animal Ethics Review Committee of Nippon Medical School, Tokyo approved the current research procedures (approval number: 2019-15, 1 April 2019).

4.2. Alkali Burn Model and TREATMent with PPAR Agonist Ophthalmic Solution The corneal alkali burn was made by placing a circular piece of ﬁlter paper, 3.2 mm in diameter, which was soaked in 1 M NaOH, on the central cornea for 1 min under general isoﬂurane anesthesia. After alkali exposure, the cornea was rinsed immediately with 40 mL physiological saline. The procedure was performed in the right eye of each rat. Int. J. Mol. Sci. 2020, 21, 5093 12 of 16

Immediately after the above treatment, vehicle solution (vehicle/control group), 0.05% fenofibrate solution (PPARα group), 0.1% pioglitazone solution (PPARγ group), or 0.05% fenofibrate + 0.1% pioglitazone solution (PPARα+γ group) was applied to the ocular surface of each rat. Topical administration was continued in each group twice a day until the endpoint. Macroscopic photographs of each group were obtained at each time point. Each eye drop solution was prepared as follows. Vehicle solution was made using 0.1 mL polyoxyethylene sorbitan monooleate (Wako Pure Chemical Industries, Osaka, Japan) and 100 mL NaCl-based phosphate-buffered saline (0.01 M; pH 7.4), which was prepared with disodium hydrogen phosphate dodecahydrate (232 g), sodium dihydrogen phosphate dihydrate (23.7 g), and distilled water (4000 mL). The 0.05% fenofibrate solution was made by adding 10 mg fenofibrate (Wako Pure Chemical Industries) to 20 mL of vehicle solution. The 0.1% pioglitazone solution was made by adding 20 mg pioglitazone (Wako Pure Chemical Industries) to 20 mL of vehicle solution. The combination of 0.05% fenofibrate and 0.1% pioglitazone was made by adding 10 mg fenofibrate and 20 mg pioglitazone to 20 mL of vehicle solution. All solutions were kept at 4 ◦C. Rats were euthanized at each time point by exsanguination under general isoflurane anesthesia. The eyeballs were enucleated for histological and immunohistochemical analysis and real-time RT-PCR. The corneal tissues for RT-PCR were immediately placed in RNAlater solution (Life Technologies, Carlsbad, CA, USA).

4.3. Histological and Immunohistochemical Analysis At 6 h and on days 1, 4, and 7 after the alkali burn, the eyes were enucleated for histological and immunohistochemical analysis (n = 8 per time point), fixed in 10% buffered formalin, and embedded in paraffin for light microscopic analysis. Deparaffinized tissues were stained with HE for histopathological examination. EST staining was performed to detect infiltrating neutrophils [40]. Primary antibodies used for immunohistochemical analysis included: polyclonal rabbit anti-NF-κB/P65 (Santa Cruz Biotechnology, Dallas, TX, USA), monoclonal mouse anti-IκB-α (Santa Cruz Biotechnology), monoclonal mouse anti-rat ED-1 (BMA, Nagoya, Japan; to detect infiltrating macrophages), monoclonal mouse anti-rat ED-2 antibody (BMA; to detect M2 macrophages), monoclonal mouse anti-nestin (Nestin; Merck Millipore, Darmstadt, Germany; to detect endothelial cells of immature blood vessels [41]), monoclonal mouse anti-JG12 (Thermo Fisher Scientific, Tokyo, Japan), which is specifically expressed by endothelial cells of blood vessels [42], and polyclonal goat anti-type III collagen (Southern Biotech, Birmingham, AL, USA). For all immunohistochemical analyses, paraffin-embedded tissues that were fixed in 10% buffered formalin were used. The tissues were stained using a standard avidin-biotin-peroxidase complex technique. The numbers of immunohistochemically positive cells were counted by one researcher, and the average of five fields in the cornea at 400 magnification was calculated. Cornea areas in three × fields of the center cornea at a magnification of 200 were measured using ImageJ software. × 4.4. Real-Time RT-PCR To examine the mRNA expression levels of PPARα, PPARγ, IL-1β, IL-6, NF-κB, IκB, TNF-α, TGF-β1, VEGF-A, Ang-1, and Ang-2, we used real-time RT–PCR (n = 8 per time point). Total RNA was extracted from corneas using ISOGEN II (Nippon Gene, Tokyo, Japan) in accordance with the manufacturer’s protocol. To measure the RNA concentration and ensure purity (A260/A280), a NanoDrop ND-1000 V3.2.1 Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) was used. cDNA libraries were synthesized from total RNA using the High Capacity cDNA Reverse Transcription kit (Thermo Fisher Scientific) in accordance with the manufacturer’s protocol. Target genes were amplified (2 min at 50 ◦C, 10 min at 95 ◦C, and 45 cycles of denaturation at 95 ◦C for TM 15 s and annealing at 60 ◦C for 60 s) using the QuantStudio 3 Real-Time PCR System (Thermo Fisher Scientific), THUNDERBIRD SYBR qPCR Mix (TOYOBO, Osaka, Japan), and specific primers. Normalized values for mRNA expression were calculated as the relative quantity of the housekeeping gene, β-actin. The primers used in this experiment are described below (Table1). Int. J. Mol. Sci. 2020, 21, 5093 13 of 16

Table 1. Primer pairs used in this study.

Gene Forward Primer Sequence (50-30) Reverse Primer Sequence (50-30) β-actin GCAGGAGTACGATGAGTCCG ACGCAGCTCAGTAACAGTCC PPARα TCGTGGAGTCCTGGAACTGA GAGTTACGCCCAAATGCACC PPARγ GCGAGGGCGATCTTGACA ATGCGGATGGCCACCTCTTT IL-1β TACCTATGTCTTGCCCGTGGAG ATCATCCCACGAGTCACAGAGG IL-6 GTCAACTCCATCTGCCCTTCAG GGCAGTGGCTGTCAACAACAT NF-κB TGGACGATCTGTTTCCCCTC TCGCACTTGTAACGGAAACG IκB-α TGACCATGGAAGTGATTGGTCAG GATCACAGCCAAGTGGAGTGGA TNF-α AAATGGGCTCCCTCTCATCAGTTC TCTGCTTGGTGGTTTGCTACGAC TGF-β1 TGGCCAGATCCTGTCCAAAC GTTGTACAAAGCGAGCACCG VEGF-A GCAGCGACAAGGCAGACTAT GCAACCTCTCCAAACCGTTG Ang-1 CACCGTGAGGATGGAAGCCTA TTCCCAAGCCAATATTCACCAGA Ang-2 CTTCAGGTGCTGGTGTCCA GTCACAGTAGGCCTTGACCTC

4.5. Statistical Analyses All results are expressed as the mean standard error. Diﬀerent groups were compared using ± one-way analysis of variance followed by the Tukey–Kramer post-analysis test. A p-value less than 0.05 was considered statistically signiﬁcant. All analyses were calculated using GraphPad Prism software (Version 8.4.2, GraphPad Software, San Diego, CA, USA).

5. Conclusions The combination ophthalmic solution of fenofibrate and pioglitazone suppressed inflammation, neovascularization, and fibrotic changes by a dual therapeutic effect of the PPARα and PPARγ agonists in the alkali-burned cornea, suggesting a new candidate therapeutic strategy for corneal disease.

Author Contributions: Conceptualization, Y.N.; methodology, Y.N.; formal analysis, Y.N.; investigation, Y.N. and Y.T.; data curation, Y.N.; writing—original draft preparation, Y.N.; writing—review and editing, T.A., M.U., A.S., and H.T.; visualization, Y.N.; supervision, T.A., M.U., A.S., and H.T.; project administration, Y.N., Y.T., and T.A.; funding acquisition, A.S. and T.A. All authors have read and agreed to the published version of the manuscript. Funding: This work was supported by a Grant-in-Aid for Scientific Research (KAKENHI, No. 19K18859) from the Japan Society for the Promotion of Science. Acknowledgments: We thank N. Kuwahara and K. Wakamatsu of the Graduate School of Medicine, Nippon Medical School, for their special support with this project. Conflicts of Interest: The authors declare no conflict of interest.

Abbreviations

PPAR Peroxisome proliferator-activated receptor HE Hematoxylin and eosin RT-PCR Reverse transcription polymerase chain reaction EST Esterase IL Interleukin NF-κB Nuclear factor-kappa B IκB-α nuclear factor of kappa light polypeptide gene enhancer in B-cells inhibitor alpha TNF-α Tumor necrosis factor-α TGF-β Transforming growth factor-β VEGF Vascular endothelial growth factor Ang Angiopoietin Int. J. Mol. Sci. 2020, 21, 5093 14 of 16

References

1. Wagoner, M.D. Chemical injuries of the eye: Current concepts in pathophysiology and therapy. Surv. Ophthalmol. 1997, 41, 275–313. [CrossRef] 2. Sharma, N.; Kaur, M.; Agarwal, T.; Sangwan, V.S.; Vajpayee, R.B. Treatment of acute ocular chemical burns. Surv. Ophthalmol. 2018, 63, 214–235. [CrossRef][PubMed] 3. Abdelrahman, M.; Sivarajah, A.; Thiemermann, C. Beneficial effects of PPAR-gamma ligands in ischemia-reperfusion injury, inflammation and shock. Cardiovasc. Res. 2005, 65, 772–781. [CrossRef] [PubMed] 4. Evans, R.M. The steroid and thyroid hormone receptor superfamily. Science 1988, 240, 889–895. [CrossRef] [PubMed] 5. Lee, C.-H.; Olson, P.; Evans, R.M. Minireview: Lipid metabolism, metabolic diseases, and peroxisome proliferator-activated receptors. Endocrinology 2003, 144, 2201–2207. [CrossRef][PubMed] 6. Barbier, O.; Torra, I.P.; Duguay, Y.; Blanquart, C.; Fruchart, J.-C.; Glineur, C.; Staels, B. Pleiotropic actions of peroxisome proliferator-activated receptors in lipid metabolism and atherosclerosis. Arterioscler. Thromb. Vasc. Biol. 2002, 22, 717–726. [CrossRef] 7. Zhang, S.; Gu, H.; Hu, N. Role of Peroxisome Proliferator-Activated Receptor γ in Ocular Diseases. J. Ophthalmol. 2015, 2015, 275435. [CrossRef] 8. Uchiyama, M.; Shimizu, A.; Masuda, Y.; Nagasaka, S.; Fukuda, Y.; Takahashi, H. An ophthalmic solution of a peroxisome proliferator-activated receptor gamma agonist prevents corneal inflammation in a rat alkali burn model. Mol. Vis. 2013, 19, 2135–2150. 9. Arima, T.; Uchiyama, M.; Nakano, Y.; Nagasaka, S.; Kang, D.; Shimizu, A.; Takahashi, H. Peroxisome proliferator-activated receptor alpha agonist suppresses neovascularization by reducing both vascular endothelial growth factor and angiopoietin-2 in corneal alkali burn. Sci. Rep. 2017, 7, 1–11. [CrossRef] 10. Tanabe, J.; Tamasawa, N.; Yamashita, M.; Matsuki, K.; Murakami, H.; Matsui, J.; Sugimoto, K.; Yasujima, M.; Suda, T. Effects of combined PPARgamma and PPARalpha agonist therapy on reverse cholesterol transport in the Zucker diabetic fatty rat. Diabetes Obes. Metab. 2008, 10, 772–779. [CrossRef] 11. Abd El-Haleim, E.A.; Bahgat, A.K.; Saleh, S. Effects of combined PPAR-γ and PPAR-α agonist therapy on fructose induced NASH in rats: Modulation of gene expression. Eur. J. Pharmacol. 2016, 773, 59–70. [CrossRef][PubMed] 12. Pargament, J.M.; Armenia, J.; Nerad, J.A. Physical and chemical injuries to eyes and eyelids. Clin. Dermatol. 2015, 33, 234–237. [CrossRef] 13. Zhou, H.; Zhang, W.; Bi, M.; Wu, J. The molecular mechanisms of action of PPAR-γ agonists in the treatment of corneal alkali burns (Review). Int. J. Mol. Med. 2016, 38, 1003–1011. [CrossRef][PubMed] 14. Sotozono, C.; He, J.; Matsumoto, Y.; Kita, M.; Imanishi, J.; Kinoshita, S. Cytokine expression in the alkali-burned cornea. Curr. Eye Res. 1997, 16, 670–676. [CrossRef][PubMed] 15. Baeuerle, P.A.; Baltimore, D. NF-kappa B: Ten years after. Cell 1996, 87, 13–20. [CrossRef] 16. Verma, I.M.; Stevenson, J.K.; Schwarz, E.M.; Van Antwerp, D.; Miyamoto, S. Rel/NF-kappa B/I kappa B family: Intimate tales of association and dissociation. Genes Dev. 1995, 9, 2723–2735. [CrossRef] 17. Consoli, A.; Devangelio, E. Thiazolidinediones and inflammation. Lupus 2005, 14, 794–797. [CrossRef] 18. Scirpo, R.; Fiorotto, R.; Villani, A.; Amenduni, M.; Spirli, C.; Strazzabosco, M. Stimulation of nuclear receptor peroxisome proliferator-activated receptor-γ limits NF-κB-dependent inflammation in mouse cystic fibrosis biliary epithelium. Hepatology 2015, 62, 1551–1562. [CrossRef] 19. Ajmone-Cat, M.A.; Bernardo, A.; Greco, A.; Minghetti, L. Non-Steroidal Anti-Inflammatory Drugs and Brain Inflammation: Effects on Microglial Functions. Pharmaceuticals 2010, 3, 1949–1965. [CrossRef] 20. Delerive, P.; Gervois, P.; Fruchart, J.-C.; Staels, B. Induction of IkappaBalpha expression as a mechanism contributing to the anti-inflammatory activities of peroxisome proliferator-activated receptor-alpha activators. J. Biol. Chem. 2000, 275, 36703–36707. [CrossRef] 21. Ricardo, S.D.; Van Goor, H.; Eddy, A.A. Macrophage diversity in renal injury and repair. J. Clin. Investig. 2008, 118, 3522–3530. [CrossRef][PubMed] Int. J. Mol. Sci. 2020, 21, 5093 15 of 16

22. Mosser, D.M. The many faces of macrophage activation. J. Leukoc. Biol. 2003, 73, 209–212. [CrossRef] [PubMed] 23. Bouhlel, M.A.; Derudas, B.; Rigamonti, E.; Dièvart, R.; Brozek, J.; Haulon, S.; Zawadzki, C.; Jude, B.; Torpier, G.; Marx, N.; et al. PPARgamma activation primes human monocytes into alternative M2 macrophages with anti-inflammatory properties. Cell Metab. 2007, 6, 137–143. [CrossRef][PubMed] 24. Charo, I.F. Macrophage polarization and insulin resistance: PPARgamma in control. Cell Metab. 2007, 6, 96–98. [CrossRef][PubMed] 25. Amparo, F.; Sadrai, Z.; Jin, Y.; Alfonso-Bartolozzi, B.; Wang, H.; Shikari, H.; Ciolino, J.B.; Chodosh, J.; Jurkunas, U.; Schaumberg, D.A.; et al. Safety and efficacy of the multitargeted receptor kinase inhibitor pazopanib in the treatment of corneal neovascularization. Investig. Ophthalmol. Vis. Sci. 2013, 54, 537–544. [CrossRef] 26. Suri, C.; Jones, P.F.; Patan, S.; Bartunkova, S.; Maisonpierre, P.C.; Davis, S.; Sato, T.N.; Yancopoulos, G.D. Requisite role of angiopoietin-1, a ligand for the TIE2 receptor, during embryonic angiogenesis. Cell 1996, 87, 1171–1180. [CrossRef] 27. Maisonpierre, P.C.; Suri, C.; Jones, P.F.; Bartunkova, S.; Wiegand, S.J.; Radziejewski, C.; Compton, D.; McClain, J.; Aldrich, T.H.; Papadopoulos, N.; et al. Angiopoietin-2, a natural antagonist for Tie2 that disrupts In Vivo angiogenesis. Science 1997, 277, 55–60. [CrossRef] 28. Daly, C.; Wong, V.; Burova, E.; Wei, Y.; Zabski, S.; Griffiths, J.; Lai, K.-M.; Lin, H.C.; Ioffe, E.; Yancopoulos, G.D.; et al. Angiopoietin-1 modulates endothelial cell function and gene expression via the transcription factor FKHR (FOXO1). Genes Dev. 2004, 18, 1060–1071. [CrossRef] 29. Asahara, T.; Chen, D.; Takahashi, T.; Fujikawa, K.; Kearney, M.; Magner, M.; Yancopoulos, G.D.; Isner, J.M. Tie2 receptor ligands, angiopoietin-1 and angiopoietin-2, modulate VEGF-induced postnatal neovascularization. Circ. Res. 1998, 83, 233–240. [CrossRef] 30. Qiu, F.; Matlock, G.; Chen, Q.; Zhou, K.; Du, Y.; Wang, X.; Ma, J.-X. Therapeutic Effects of PPARα Agonist on Ocular Neovascularization in Models Recapitulating Neovascular Age-Related Macular Degeneration. Investig. Ophthalmol. Vis. Sci. 2017, 58, 5065–5075. [CrossRef] 31. Xin, X.; Yang, S.; Kowalski, J.; Gerritsen, M.E. Peroxisome proliferator-activated receptor gamma ligands are potent inhibitors of angiogenesis In Vitro and In Vivo. J. Biol. Chem. 1999, 274, 9116–9121. [CrossRef] [PubMed] 32. Chen, S.; Mienaltowski, M.J.; Birk, D.E. Regulation of corneal stroma extracellular matrix assembly. Exp. Eye Res. 2015, 133, 69–80. [CrossRef][PubMed] 33. Meek, K.M.; Knupp, C. Corneal structure and transparency. Prog. Retin. Eye Res. 2015, 49, 1–16. [CrossRef] [PubMed] 34. Shah, M.; Foreman, D.M.; Ferguson, M.W. Neutralisation of TGF-beta 1 and TGF-beta 2 or exogenous addition of TGF-beta 3 to cutaneous rat wounds reduces scarring. J. Cell. Sci. 1995, 108, 985–1002. [PubMed] 35. Ask, K.; Bonniaud, P.; Maass, K.; Eickelberg, O.; Margetts, P.J.; Warburton, D.; Groffen, J.; Gauldie, J.; Kolb, M. Progressive pulmonary fibrosis is mediated by TGF-beta isoform 1 but not TGF-beta3. Int. J. Biochem. Cell Biol. 2008, 40, 484–495. [CrossRef][PubMed] 36. Zhang, F.; Liu, K.; Cao, M.; Qu, J.; Zhou, D.; Pan, Z.; Duan, X.; Zhou, Y. Rosiglitazone Treatment Prevents Postoperative Fibrosis in a Rabbit Model of Glaucoma Filtration Surgery. Investig. Ophthalmol. Vis. Sci. 2019, 60, 2743–2752. [CrossRef] 37. Shehata, A.H.F.; Ahmed, A.-S.F.; Abdelrehim, A.B.; Heeba, G.H. The impact of single and combined PPAR-α and PPAR-γ activation on the neurological outcomes following cerebral ischemia reperfusion. Life Sci. 2020, 252, 117679. [CrossRef] 38. Balakumar, P.; Mahadevan, N.; Sambathkumar, R. A Contemporary Overview of PPARα/γ Dual Agonists for the Management of Diabetic Dyslipidemia. CMP 2019, 12, 195–201. [CrossRef] 39. Kaur, P.; Bhat, Z.R.; Bhat, S.; Kumar, R.; Kumar, R.; Tikoo, K.; Gupta, J.; Khurana, N.; Kaur, J.; Khatik, G.L. Synthesis and evaluation of new 1,2,4-oxadiazole based trans-acrylic acid derivatives as potential PPAR-alpha/gamma dual agonist. Bioorganic Chem. 2020, 100, 103867. [CrossRef] 40. Masuda, Y.; Shimizu, A.; Mori, T.; Ishiwata, T.; Kitamura, H.; Ohashi, R.; Ishizaki, M.; Asano, G.; Sugisaki, Y.; Yamanaka, N. Vascular endothelial growth factor enhances glomerular capillary repair and accelerates resolution of experimentally induced glomerulonephritis. AJPA 2001, 159, 599–608. [CrossRef] Int. J. Mol. Sci. 2020, 21, 5093 16 of 16

41. Mokrý, J.; Nemecek, S. Immunohistochemical detection of intermediate filament nestin. Acta Med. (Hradec Kral.) 1998, 41, 73–80. 42. Matsui, K.; Nagy-Bojarsky, K.; Laakkonen, P.; Krieger, S.; Mechtler, K.; Uchida, S.; Geleff, S.; Kang, D.-H.; Johnson, R.J.; Kerjaschki, D.; et al. Lymphatic microvessels in the rat remnant kidney model of renal fibrosis: Aminopeptidase p and podoplanin are discriminatory markers for endothelial cells of blood and lymphatic vessels. J. Am. Soc. Nephrol. 2003, 14, 1981–1989. [CrossRef][PubMed]

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).