Tyrosine Kinase Inhibitor, Vatalanib, Inhibits Proliferation and Migration of Human Pterygial Fibroblasts

BASIC INVESTIGATION

Hong Kyu Kim, MD,* Ji-Young Choi, PhD,† Sang Min Park, PhD,‡ Chang Rae Rho, MD, PhD,§ Kyong Jin Cho, MD, PhD,¶ and Sangmee Ahn Jo, PhD‡k

b significantly reduced, but there was no alteration in p53 protein Purpose: Vatalanib is a small-molecule tyrosine kinase inhibitor. levels in HPFs. We investigated the effects of vatalanib on the proliferation and migration of cultured human pterygial fibroblasts (HPFs). Conclusions: These results indicate that vatalanib significantly suppressed the proliferation and migration of HPFs by decreasing Methods: Pterygium tissues were obtained after pterygium exci- vascular endothelial growth factor and transforming growth factor-b. sion surgery and subjected to primary culture. HPFs were treated Vatalanib showed less toxicity than that of MMC. Based on these with vatalanib at various concentrations. Mitomycin C (MMC) was results, vatalanib may potentially serve as a new adjuvant treatment used as a positive control. Cell proliferation and migration assays after pterygium excision surgery. were used to investigate the effects of vatalanib. Cell death was measured using flow cytometry analysis. Western blot analysis was Key Words: tyrosine kinase inhibitor, vatalanib, mitomycin C, performed to identify signaling molecules associated with the pterygium, cell growth response to vatalanib. (Cornea 2017;36:1116–1123) Results: Vatalanib inhibited both proliferation and migration of HPFs in a dose-dependent manner. Cell proliferation was signif- m terygium is one of the most common ocular surface icantly suppressed by vatalanib (10 and 100 M) and MMC diseases. It is characterized by remodeling, proliferation, (0.004% and 0.04%) treatments. Migration assays revealed signif- P fl m angiogenesis, and in ammation of penetrating triangle- icant HPF delay when treated with vatalanib (1, 10, and 100 M) shaped fibrovascular tissue from the conjunctiva into the and MMC (0.004% and 0.04%) compared with that in a negative cornea. Pterygium is pathologically benign; however, the control. Cell death analysis showed that high concentrations of m disease recurs easily and has a tendency to invade normal vatalanib (100 M) and MMC (0.004% and 0.04%) decreased cell tissue. It has also been reported to be associated with numbers. Western blot analysis of vatalanib-treated cells showed premalignant disease, possibly because of cumulative genetic vascular endothelial growth factor and transforming growth factor- damage from chronic UV exposure.1 Causes of pterygium include UV-A exposure,2 UV-B exposure, viral infection,3 and various molecular mecha- Received for publication January 15, 2017; revision received April 16, 2017; nisms. UV-A induces urokinase-type plasminogen activator accepted April 26, 2017. Published online ahead of print June 21, 2017. fl From the *Department of Ophthalmology, Institute of Vision Research, (uPA), a central mediator of extracellular matrixes, in am- 2 College of Medicine, Yonsei University, Seoul, South Korea; †Department matory processes, and angiogenesis. UV-B causes cellular of Pharmacology and Medicinal Toxicology Research Center, Hypoxia- DNA damage through 2 distinct mechanisms: direct photo- related Disease Research Center, Inha University School of Medicine, toxic effects on cellular DNA and generation of reactive ‡ Incheon, South Korea; Department of Nanobiomedical Science and BK21 oxygen species.4 A variety of oncogenic viruses have been PLUS NBM Global Research Center for Regenerative Medicine, Dankook fi University, Cheonan-si, South Korea; §Department of Ophthalmology and identi ed in pterygium tissues (eg, human papillomavirus, Visual Science, Daejeon St. Mary’s Hospital, College of Medicine, The cytomegalovirus, and herpes simplex virus), which inactivate Catholic University of Korea, Seoul, Republic of Korea; ¶Department of p53 and lead to chromosomal instability.3 Several additional Ophthalmology, College of Medicine, Dankook University, Cheonan-si, factors play a role in the pathogenesis of pterygium, South Korea; and kDepartment of Pharmacy, College of Pharmacy, – Dankook University, Cheonan-si, South Korea. including epigenetic factors, cell proliferation related fac- Supported by grants by the Korea Health Technology R&D Project, tors, inflammatory mediators, growth factors, extracellular Ministry of Health & Welfare, Republic of Korea (HI12C0003), and matrix modulators, angiogenic and lymphangiogenic factors, the National Research Foundation of Korea (NRF) grant funded by immunologic mechanisms, epithelial–mesenchymal transi- the Korean government (NRF 2014R1A2A2A005564 and NRF- tion, and alterations in cholesterol metabolism.4 2016R1C1B2016649). The authors have no conflicts of interest to disclose. Surgical excision is a standard treatment for symp- K. J. Cho and S. A. Jo contributed equally to this article and should be tomatic pterygium, although the condition frequently regarded as co-corresponding authors. (24%–89%) recurs after treatment.5,6 Adjunctive surgical Reprints: Kyong Jin Cho, MD, PhD, Department of Ophthalmology, procedures and adjuvant therapies are used to decrease the Dankook University Hospital, College of Medicine, Dankook University, #119 Dandae-ro, Dongnam-gu, Cheonan-si, Chungcheongnam-do, risk of recurrence. Conjunctival autografts and amniotic Republic of Korea, 31116 (e-mail: [email protected]). membrane grafts have been implemented as adjunctive Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved. surgical techniques. Adjunctive pharmaceutical treatments

1116 | www.corneajrnl.com Cornea Volume 36, Number 9, September 2017

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 36, Number 9, September 2017 Inhibitory Effect of Vatalanib in HPFs include mitomycin C (MMC), subconjunctival bevacizu- Korea) solution and passaged for subcultures with 1:4 split. mab,7 bevacizumab eye drops,8,9 cyclosporine,10 pirfeni- Cells at 4 to 6 passages were used for further experiments. done,11 and 5-fluorouracil.12,13 Proliferative factors related to pterygium development include basic fibroblast growth factor (bFGF), vascular Characterization of Cultured endothelial growth factor (VEGF), platelet-derived growth Pterygial Fibroblasts factor (PDGF), heparin-binding EGF, c-myc, and uPA.2,14–17 Immunofluorescence staining was performed to char- The VEGF family members (eg, VEGF and PDGF) are acterize the cultured human pterygial fibroblasts (HPFs). activated by binding to tyrosine receptors on cellular The cells were fixed with 10% neutral-buffered formalin membranes. Vatalanib is a tyrosine kinase inhibitor (TKI) (BBC Biochemical, Mount Vernon, WA) for 30 minutes at that suppresses receptor-mediated signaling. It was well room temperature (RT), then washed with washing buffer tolerated without toxicities (eg, hypertension, proteinuria, [1· PBS (Biosesang, Seongnam, Korea), 0.3% Triton X-100 nausea, diarrhea, and fatigue) in patients with advanced (Sigma-Aldrich)] 5 times (5 min per wash). The cells were colorectal cancer in a phase I study, and its biological activity then incubated with blocking solution [1· PBS, 0.3% Triton was maintained.18,19 To our knowledge, few studies have X-100, 2% bovine serum albumin (BSA) (Cellect; MP investigated the effects of TKIs on pterygial fibroblasts. Biomedicals, Santa Ana, CA), 2% goat serum (Vector The objective of this study was therefore to elucidate Laboratories, Burlingame, CA)] for 1 hour at RT. Cells the effects of vatalanib on proliferation and migration of were then incubated with antibodies against cytokeratin-13, pterygial fibroblasts and to identify the molecular mechanism vimentin, and alpha-smooth muscle actin (a-SMA) over- underlying any observed effects. night at 4°C (Table 1). After washing with PBS, cells were incubated with fluorescein isothiocyanate–conjugated secondary antibodies for 1 hour at RT. The washed cells were mountedwithVectashieldmountingmediumwithDAPI MATERIALS AND METHODS (Vector Laboratories, Burlingame, CA) and stored in a dark Study Subjects place. Prepared cells were monitored at ·200 magnification Discarded pterygium tissue was obtained from patients using a fluorescent microscope (Observer Z1, Carl Zeiss, who underwent clinically indicated pterygium excision with Oberkochen, Germany). conjunctival autograft surgery. Written informed consent was obtained; the research was performed in accordance with the Declaration of Helsinki. The study was reviewed and Cell Number Counting approved by the Institutional Review Board of Dankook HPFs were seeded in 12-well culture plates at a density University Hospital (Cheonan, South Korea) on February 23, of 15,000 cells per well. After 24 hours, cells were treated m 2015 (code: 2015-02-023). with vatalanib (0, 1, 10, or 100 M) or MMC (0.004 or 0.04%). At 0, 6, 12, 24, and 36 hours, the cells were trypsinized and inactivated with 50% fetal bovine serum. After centrifugation at 300g for 5 minutes, the cell pellet was Isolation and Primary Culture of resuspended in growth media containing 0.1% Trypan blue to Pterygial Fibroblasts detect dead cells and loaded into a hemocytometer. The cell Excised pterygium specimens were stored in 1.5-mL number was counted under microscopy. tubes containing Dulbecco Modified Eagle Medium (DMEM/ F12; Welgene, Gyeongsan, Korea); the samples were placed on ice and transferred to our laboratory within an hour. Cell Proliferation Assay Specimens were chopped using surgical blade No. 15; The cell proliferation assay was performed using the transferred to a 1.5-mL tube containing collagenase type I Cell Counting Kit-8 (Dojindo, Kumamoto, Japan). HPFs solution (250 mL, 12 mg/mL concentration; Sigma-Aldrich, were seeded in a 96-well culture plate at a density of 5000 St. Louis, MO), dispase solution (250 mL, 16 mg/mL cells per well and incubated for 24 hours in serum-free media. concentration; Gibco, Grand Island, NY), and sterile The same drug concentrations used in the cell-counting assay phosphate-buffered saline (PBS, 500 mL; Hyclone, Logan, were followed by 36 hours of incubation. The media were UT); and shaken gently at 37.0°C as described previously.20 removed, and all groups were treated with 100 mL culture After shaking, the samples were washed with serum-free media and 10-mL Cell Counting Kit-8; cells were further media [DMEM/F12, 1· penicillin streptomycin (PS; Gibco incubated for 4 hours in a dark place. Cell proliferation was Life Technologies)] and then cultured in 4-mL culture media assessed by absorbance at 450 nm using a microplate reader [DMEM/F12, 1· PS, 10% fetal bovine serum (Gibco Life (Synergy Mx; BioTek, Winooski, VT). Technologies)] in a 100p culture plate (SPL Life Sciences, Pocheon, Korea) in a CO2-regulated incubator (37°C and 5% CO2). Culture media were replaced after 4 days initially and Flow Cytometry Using Fluorescence-Activated every 2 days thereafter. When cells were 80% confluent, Cell Sorting cultured cells were detached from the plate using a 1· T/ The cytotoxicity of vatalanib on cultured HPFs was EDTA (0.25% trypsin/1 mM EDTA; Welgene, Gyeongsan, assessed using flow cytometry analyses. The cells were

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Kim et al Cornea Volume 36, Number 9, September 2017

TABLE 1. Primary and Secondary Antibodies Used in Immunohistochemistry and Western Blot Analysis Antibody Immunized Animal Dilution Company Catalog Number Immunohistochemistry Cytokeratin-13 antibody Mouse 1:100 Santa Cruz sc-101460 Vimentin antibody Mouse 1:100 Santa Cruz sc-6260 a-SMA antibody Rabbit 1:500 Abcam ab5694 488 anti-mouse IgG Goat 1:500 Life tech A11029 488 anti-rabbit IgG Goat 1:500 Vector DI-1488 Western blot analysis VEGF antibody Mouse 1:200 Santa Cruz sc-7219 TGF-b antibody Rabbit 1:100 Cell signaling 3711S TNF-a antibody Goat 1:200 Santa Cruz sc-1651 uPA antibody Mouse 1:200 Santa Cruz sc-14019 p53 antibody Rabbit 1:200 Santa Cruz sc-6243 MMP-1 antibody Rabbit 1:5000 Abcam ab38929 a-SMA antibody Rabbit 1:1000 Abcam ab5694 Vimentin antibody Mouse 1:200 Santa Cruz sc-6260 a-tubulin antibody Mouse 1:2500 Millipore 05-829 Anti-goat IgG-HRP Rabbit 1:5000 Santa Cruz sc-2768 Anti-rabbit IgG-HRP Donkey 1:5000 Santa Cruz sc-2313 Anti-mouse IgG-HRP Goat 1:5000 Santa Cruz sc-2005

HRP, horseradish peroxidase.

seeded (1 · 106 cells per well) in 100p culture plates (SPL incubated for 5 minutes, and analyzed through flow Life Sciences) and cultured in serum-free media. After 24- cytometry analyses. hour incubation, the media were removed, and the cells were incubated with either culture media with vatalanib (0, 1, 10, or 100 mM) (PTK-787; ChemieTek, Indianapolis, Cell Migration Assay IN). For the MMC-treated group, MMC (Kyowa Kirin, Migration of HPFs was assessed using the scratch- Tokyo, Japan) was diluted into sterile PBS (0.004% and wound assay previously described.11 HPFs were seeded (3 0.04%); cells were treated for 5 minutes, washed with · 105 cells per well) in a 6-well culture plate (SPL Life serum-free media for 5 times, and incubated with culture Sciences) and maintained in serum-free media (DMEM/ media. Cells were detached with trypsin/EDTA, followed F12, 1· PS). After 24 hours of incubation, a straight line by separation through centrifugation (900g for 5 min). The was made through the cells using a P200 pipette tip. The cells were washed twice with cold sterile PBS and plates were rinsed with serum-free media to remove the resuspended in 4% paraformaldehyde for 5 minutes. After suspended cells. Samples were treated as described in fixation with 4% paraformaldehyde, the cells were washed the cell-counting assay for the control, vatalanib-treated, twice with cold PBS and then resuspended in 1· binding and MMC-treated groups. Identical cell sites were buffer (100 mL). After adding 5 mL of annexin V-FITC and observed 0, 12, 24, and 36 hours after scratching using 5 mL of propidium iodide (PI), the cells were gently a light microscope (Olympus, Tokyo, Japan) incorporating vortexed and incubated for 15 minutes at RT in the dark. a camera (Canon, Tokyo, Japan) with ·40 magnification. After the addition of 1· binding buffer (400 mL) to each The color images were transformed into binary images tube, all samples were analyzed within 1 hour using the using ImageJ software (National Institutes of Health, flow cytometry BD FACSCalibur equipped with CellQuest Bethesda, MD), and migrated cells were counted using software (BD Biosciences, San Jose, CA). the analyze particle method. To further monitor the effects of vatalanib on the cell cycle, additional in vitro assays were performed. After 36 hours of treatment with vatalanib and MMC at the Western Blot Analysis concentrations previously described, cells were detached HPFs were seeded in a 10p culture plate (SPL Life and fixed with cold sterile PBS (2 mL) and cold 100% Sciences) and incubated until 80% confluence. The cells were ethyl alcohol (8 mL) for 18 hours at 4°C. Cells were then incubated in serum-free media for 24 hours to eliminate centrifuged again and the resulting supernatant was the effects of serum. Samples were treated as described above removed and treated with RNase A (40 mL, 10 mg/mL, for each group, then incubated for 36 hours. The culture Sigma-Aldrich), followed by sterile PBS (300 mL). media were removed, and the cells were washed with cold Cells were then incubated for 2 hours at 4°C. Cells were sterile PBS (4°C). The cells were then harvested and lysed on treated with PI (5 mL, 2.5 mg/mL; Sigma-Aldrich), ice using lysis buffer [RIPA buffer (Thermo Scientific,

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 36, Number 9, September 2017 Inhibitory Effect of Vatalanib in HPFs

FIGURE 1. Histochemical characterization of cultured human pterygial and conjunctival cells. Cultured pterygial cells were im- munostained with CK-13, vimentin, and a-SMA antibodies and counterstained with DAPI. Pterygial cells were strongly positive for vimentin and a-SMA but negative for CK-13. CK-13, cytokeratin-13.

Rockford, IL), 1· phosphatase inhibitor cocktail tablets manufacturer’s instructions, using a luminescent image (PhosSTOP; Roche, Mannheim, Germany), and 1· protease analyzer (Image Quant Las 4000 mini; GE Healthcare, inhibitor cocktail tablets (Complete EDTA-free; Roche)]. Milwaukee, WI). Final protein concentrations were determined using a BCA protein assay kit (Pierce; Thermo Scientific) as per the manufacturer’s specifications. Statistical Analysis Western blot analysis was performed as previously 6 – described.21 Briefly, each prepared sample was heated at Data are expressed as mean SD. The Mann Whitney test and Kruskal–Wallis test were used to 100°C for 10 minutes; a standard amount of total protein fi was added to 10% acrylamide gel and resolved by SDS- examine signi cant differences between the control and treated groups. P , 0.05 was considered statistically sig- PAGE. The separated proteins were transferred to a poly- fi vinylidene fluoride membrane (Immobilon; Millipore, ni cant. Statistical analyses were performed using Med- Billerica, MA). Nonspecific binding was blocked through Calc for Windows, version 15.6.1 (MedCalc Software, incubation with 2% blocking solution [2% BSA, 1· tris- Mariakerke, Belgium). buffered saline (TBS, SuperBlock T20; Thermo Scientific) and 0.1% Tween 20 (Sigma-Aldrich)] for 30 minutes, followed by overnight incubation with primary antibodies RESULTS (Table 1). The membrane was incubated with primary antibodies (diluted in 5% blocking solution; 5% BSA, 1· Characterization of Cultured TBS, and 0.1% Tween 20) overnight at 4°C with gentle Pterygial Fibroblasts shaking. After washing with TBS-T (1· TBS and 0.1% Immunofluorescence staining showed that the cultured Tween 20), the membrane was incubated with secondary cells were strongly positive for vimentin and a-SMA. No antibodies in 2% blocking solution for 1 hour at RT. The staining was noted for cytokeratin-13, a marker of conjunc- membranewaswashedagainwithTBS-T,anddetectionof tival epithelium (Fig. 1). The cultured cells were thereby specific proteins was performed using an ECL Western confirmed to be pterygial fibroblasts rather than conjunctival blotting kit (Advansta, Menlo Park, CA), according to the epithelial cells.

FIGURE 2. Effects of vatalanib (left) and MMC (right) on pterygial ﬁbroblast proliferation. Data repre- sent mean 6 SEM of 3 independent experiments. Mean signiﬁcant differences (P , 0.05) of different characters among the groups were determined using the Kruskal–Wallis test.

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Kim et al Cornea Volume 36, Number 9, September 2017

FIGURE 3. Apoptosis and cell cycle analysis of pterygial fibroblasts treated with vatalanib and MMC using FACS. A and B, Viable cells (left lower quadrant, annexin V and PI negative at FACS analyses) were reduced by vatalanib only at the high concentration of 100 mM. MMC treatment reduced viable cells at both concentrations tested (0.004% and 0.04%). C, No difference in cell cycle profiles was found with vatalanib treatment. *Differences were statistically significant at P , 0.05 compared with the control group.

Effects of Vatalanib on Cell Growth, Cell control cells in the G0/G1, G2/M, and S phases were 76.6%, Death, Cell Cycle, and Migration 12.2%, and 5.8%, respectively. Similar distributions were Cell numbers were counted to measure the antipro- obtained for cells treated with vatalanib for 24 hours; the liferative effects of vatalanib on HPFs. Cell numbers percentages of 1.0, 10, and 100 mM vatalanib-treated cells in decreased significantly in the vatalanib-treated groups the G0/G1 phase were 77.0%, 80.0%, and 78%; those in the (10 and 100 mM) in a dose-dependent manner. MMC G2/M phase were 13.1%, 10.5%, and 10.8%; and those in the treatment also decreased cell numbers significantly com- S phase were 5.3%, 5.7%, and 5.7%, respectively. Vatalanib pared with the control group (Fig. 2). therefore did not alter the S phase population at any To determine whether the vatalanib-induced cell concentration tested, which suggests that vatalanib has no number reduction was due to cell death, cell cycle and effect on HPF proliferation (Fig. 3C). cell death analyses were performed using fluorescence- In the migration assay, all vatalanib- and MMC-treated activated cell sorting (FACS) analysis. In the apoptosis groups showed significant decreases in migration compared analysis, viable cell numbers (left lower quadrant, annexin with the control group (Fig. 4). V and PI negative at FACS analyses) remained unchanged at the lower concentrations of vatalanib (1–10 mM) but slightly reduced at the high concentration of 100 mM(Figs. Effect of Vatalanib on Signaling Molecules 3A, B) compared with the control group. MMC treatment To understand the vatalanib-mediated reduction in reduced viable cell counts at both concentrations tested proliferation, Western blot analyses were performed using (0.004% and 0.04%) compared with the control group. antibodies against the key proteins known to be involved in The effect of vatalanib on the cell cycle was then VEGF signaling, including VEGF, transforming growth examined by FACS analysis. The percentages of untreated factor-b (TGF-b), TNF-a, uPA, p53, ERK, MMP-1, a–

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 36, Number 9, September 2017 Inhibitory Effect of Vatalanib in HPFs

FIGURE 4. Effects of vatalanib and MMC on pterygial fibroblast migration. All vatalanib- and MMC-treated groups showed a significant decrease in migration compared with the control group. *Differences were statistically significant at P , 0.05 compared with the control group.

SMA, and vimentin (Fig. 5). The vatalanib-treated groups pterygium recurrence. MMC use, however, is associated showed a significant reduction in VEGF compared with the with adverse events, including scleral ulceration and control group; the MMC-treated groups showed no difference delayed conjunctival wound healing; increased MC con- from the control group. The TGF-b protein level also centration and length of exposure increase the risk of decreased with vatalanib (10 and 100 mM) and MMC occurrence.22 It is therefore necessary to develop a new (0.004% and 0.04%) treatments. Vatalanib had no effect on adjuvant treatment that effectively decreases pterygium p53 protein levels, whereas MMC significantly increased p53. recurrence and has an acceptable safety profile. p-ERK levels increased with both vatalanib and MMC In pterygium tissues, fibroangiogenic growth factors, treatments. No statistically significant differences in TNF-a, including bFGF and VEGF, are often elevated.15,23 This uPA, MMP-1, a-SMA, or vimentin levels were noted suggests the direct or indirect involvement of various between the treated and untreated samples. growth factors in the pathogenesis of pterygium. Anti- VEGF therapy using bevacizumab has been used to suppress pterygium progression and recurrence.7–9 Side DISCUSSION effects of bevacizumab use, however, have been docu- Pterygium is a common ocular surface disorder and mented (eg, epithelial defect and epithelial healing is presumed to be associated with chronic UV exposure. delay),24 and reported clinical results provide conflicting The most common and important complication related to evidence on the efficacy of bevacizumab in suppressing pterygium is postsurgical recurrence. MMC is one of the corneal neovascularization and pterygium recurrence.25–28 most widely used adjuvant medications to decrease The lack of clear efficacy may be due to the involvement

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Kim et al Cornea Volume 36, Number 9, September 2017

FIGURE 5. Western blot analysis of pterygial cells treated with vatalanib and MMC. Pterygial cells were treated with vatalanib or MMC for 36 hours; VEGF, TGF-b, TNF-a, uPA, p53, MMP-1, a-SMA, and vimentin protein levels were measured by Western blot analysis. Quantification was performed using densitometry (ImageJ software). Results were normalized to b-actin. VEGF protein was significantly decreased in the vatalanib-treated (10 and 100 mM) group. TGF-b protein significantly reduced in the vatalanib- (1, 10, and 100 mM) and MMC-treated (0.004% and 0.04%) groups. The P53 protein level was significantly increased in the MMC-treated group. *Differences were statistically significant at P , 0.05 compared with the control group. of growth factors other than VEGF (eg, bFGF) in the They are typically activated during normal wound healing; pathogenesis of pterygium. For this reason, multiple TKIs however, dysregulation of the wound-healing response can may be good candidates for suppressing cell activation. lead to an irreversible fibrotic response.30 TGF-b is one of We selected vatalanib, a TKI, to suppress receptor- the key drivers of fibrosis. This cytokine directly induces mediated signals. Vatalanib is reported to have a high differentiation of fibroblasts into collagen-secreting my- receptor-binding affinity for all VEGF receptor and PDGF ofibroblasts.30 TGF-b staining was stronger in pterygium receptor tyrosine kinases.29 Its relatively high water than in the normal conjunctiva.23 Interestingly, vatalanib solubility makes it a plausible candidate for adjunctive decreased TGF-b expression as demonstrated by Western pterygium treatment. blot analysis. Elshal et al31 reported the effects of In this study, vatalanib elicited dramatic growth a multiple TKI, pazopanib, on liver cirrhosis. Pazopanib inhibition of HPFs in a dose-dependent manner; vatalanib decreased tissue TGF-b levels in a dose-dependent man- 100 mM almost completely inhibited cell proliferation, with ner, which is consistent with our results. As expected, inhibition effects comparable to those seen with 0.04% vatalanib treatment significantly decreased VEGF expres- MMC. Migration assays, however, showed that vatalanib sion through its TKI activity. MMC is known to cause 100 mM was more effective in inhibiting cell migration than DNA damage and to induce p53.32 MMC 0.04%. Cell death analysis suggested possible cyto- It is necessary to note the limitations of this study. toxicity of vatalanib 100 mM and MMC (0.04% and 0.004%). Safety of vatalanib on corneal epithelium and limbal stem The effects of vatalanib on the cell cycle were further cells was not tested. In vivo experiments will be necessarily investigated using FACS analysis; as illustrated in Figure 3, needed to confirm the results of these in vitro assays. the percentage of annexin V- and PI-negative cells was In conclusion, vatalanib inhibited both proliferation and significantly decreased at the high concentration (100 mM) of migration of HPFs. The suppressive effect of vatalanib was vatalanib. MMC (0.004% and 0.04%) also significantly comparable to MMC, but the cellular toxicity was decreased decreased the cell population. In summary, vatalanib and compared with MMC. Vatalanib could be developed as a new MMC demonstrated similar growth inhibitory effects, but adjuvant treatment for pterygium excision surgery. vatalanib 10 mM demonstrated a potential safety advantage compared with MMC (0.004% and 0.04%). The cytotoxicity of vatalanib and MMC on cultured REFERENCES human conjunctival fibroblasts (HCFs) was also assessed 1. Chui J, Coroneo MT, Tat LT, et al. Ophthalmic pterygium: a stem cell – (results not shown). Because of the lack of simultaneous disorder with premalignant features. Am J Pathol. 2011;178:817 827. fi 2. Chao SC, Hu DN, Yang PY, et al. Ultraviolet-A irradiation upregulated testing, the cytotoxicity ndings of vatalanib and MMC on urokinase-type plasminogen activator in pterygium fibroblasts through HCFs were not directly comparable to the results of these ERK and JNK pathways. Invest Ophthalmol Vis Sci. 2013;54:999–1007. compounds on HPFs. Overall, the results suggested that 3. Detorakis E, Drakonaki E, Spandidos D. Molecular genetic alterations and HCFs were slightly more stable when treated with vatalanib viral presence in ophthalmic pterygium. Int J Mol Med. 2000;6:35–76. 4. Cárdenas-Cantú E, Zavala J, Valenzuela J, et al. Molecular basis of and were more vulnerable to MMC compared with HPFs. pterygium development. Semin Ophthalmol. 2016;31:567–583. Myofibroblasts, which are an activated form of 5. Jaros PA, DeLuise VP. Pingueculae and pterygia. Surv Ophthalmol. fibroblasts, act as a central coordinator of tissue fibrosis. 1988;33:41–49.

Copyright Ó 2017 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited. Cornea Volume 36, Number 9, September 2017 Inhibitory Effect of Vatalanib in HPFs

6. Singh G, Wilson MR, Foster CS. Mitomycin eye drops as treatment for derived from permanent teeth by using two different methods. J Vis Exp. pterygium. Ophthalmology. 1988;95:813–821. 2012;69:e4372. 7. Mauro J, Foster CS. Pterygia: pathogenesis and the role of subconjunc- 21. Choi JY, Kim JH, Jo SA. Acetylation regulates the stability of glutamate tival bevacizumab in treatment. Semin Ophthalmol. 2009;24:130–134. carboxypeptidase II protein in human astrocytes. Biochem Biophys Res 8. Wu PC, Kuo HK, Tai MH, et al. Topical bevacizumab eyedrops for Commun. 2014;450:372–377. limbal-conjunctival neovascularization in impending recurrent ptery- 22. Kaufman SC, Jacobs DS, Lee WB, et al. Options and adjuvants in gium. Cornea. 2009;28:103–104. surgery for pterygium: a report by the American Academy of Ophthal- 9. Fallah MR, Khosravi K, Hashemian MN, et al. Efficacy of topical mology. Ophthalmology. 2013;120:201–208. bevacizumab for inhibiting growth of impending recurrent pterygium. 23. Kria L, Ohira A, Amemiya T. Immunohistochemical localization of basic Curr Eye Res. 2010;35:17–22. fibroblast growth factor, platelet derived growth factor, transforming 10. Torun-Acar B. The effect of topical 0.05% cyclosporine on recurrence growth factor-b and tumor necrosis factor-a in the pterygium. Acta following pterygium surgery. Clin Ophthalmol. 2011;5:881–885. Histochem. 1996;98:195–201. 11. Lee K, Young Lee S, Park SY, et al. Antifibrotic effect of pirfenidone on 24. Koenig Y, Bock F, Horn F, et al. Short-and long-term safety profile human pterygium fibroblasts. Curr Eye Res. 2014;39:680–685. and efficacy of topical bevacizumab (Avastin) eye drops against 12. Said D, Faraj L, Elalfy M, et al. Intra-lesional 5 fluorouracil for the corneal neovascularization. GraefesArchClinExpOphthalmol. management of recurrent pterygium. Eye. 2013;27:1123–1129. 2009;247:1375–1382. 13. Pikkel J, Porges Y, Ophir A. Halting pterygium recurrence by post- 25. Bahar I, Kaiserman I, McAllum P, et al. Subconjunctival bevacizumab operative 5-fluorouracil. Cornea. 2001;20:168–171. injection for corneal neovascularization in recurrent pterygium. Curr Eye 14. Maini R, Collison DJ, Maidment JM, et al. Pterygial derived fibroblasts Res. 2008;33:23–28. express functionally active histamine and epidermal growth factor 26. Razeghinejad MR, Hosseini H, Ahmadi F, et al. Preliminary results of receptors. Exp Eye Res. 2002;74:237–244. subconjunctival bevacizumab in primary pterygium excision. Ophthal- 15. Jin J, Guan M, Sima J, et al. Decreased pigment epithelium-derived mic Res. 2010;43:134–138. factor and increased vascular endothelial growth factor levels in pterygia. 27. Razeghinejad M-R, Banifatemi M. Subconjunctival bevacizumab for Cornea. 2003;22:473–477. primary pterygium excision; a randomized clinical trial. J Ophthalmic Vis 16. Nolan TM, Di Girolamo N, Coroneo MT, et al. Proliferative effects of Res. 2014;9:22. heparin-binding epidermal growth factor-like growth factor on pterygium 28. Nava-Castaneda A, Ulloa-Orozco I, Garnica-Hayashi L, et al. Triple epithelial cells and fibroblasts. Invest Ophthalmol Vis Sci. 2004;45:110–113. subconjunctival bevacizumab injection for early corneal recurrent 17. Cui D, Pan Z, Zhang S, et al. Downregulation of c-Myc in pterygium and pterygium: one-year follow-up. J Ocul Pharmacol Ther. 2015;31: cultured pterygial cells. Clin Exp Ophthalmol. 2011;39:784–792. 106–113. 18. Thomas AL, Morgan B, Horsfield MA, et al. Phase I study of the safety, 29. Bold G, Altmann KH, Frei J, et al. New anilinophthalazines as potent and tolerability, pharmacokinetics, and pharmacodynamics of PTK787/ZK orally well absorbed inhibitors of the VEGF receptor tyrosine kinases 222584 administered twice daily in patients with advanced cancer. J Clin useful as antagonists of tumor-driven angiogenesis. J Med Chem. 2000; Oncol. 2005;23:4162–4171. 43:2310–2323. 19. Morgan B, Thomas AL, Drevs J, et al. Dynamic contrast-enhanced 30. Wynn TA, Ramalingam TR. Mechanisms of fibrosis: therapeutic trans- magnetic resonance imaging as a biomarker for the pharmacological lation for fibrotic disease. Nat Med. 2012;18:1028–1040. response of PTK787/ZK 222584, an inhibitor of the vascular endothelial 31. Elshal M, Abu-Elsaad N, El-Karef A, et al. The multi-kinase inhibitor growth factor receptor tyrosine kinases, in patients with advanced pazopanib targets hepatic stellate cell activation and apoptosis alleviating colorectal cancer and liver metastases: results from two phase I studies. progression of liver fibrosis. Naunyn Schmiedebergs Arch Pharmacol. J Clin Oncol. 2003;21:3955–3964. 2015;388:1293–1304. 20. Karamzadeh R, Eslaminejad MB, Aflatoonian R. Isolation, characteriza- 32. Abbas T, Olivier M, Lopez J, et al. Differential activation of p53 by the tion and comparative differentiation of human dental pulp stem cells various adducts of mitomycin C. J Biol Chem. 2002;277:40513–40519.