Investigative Ophthalmology & Visual Science Cover Image for Volume 56, Issue 3
March 2015
Volume 56, Issue 3
Free
Cornea  |   March 2015
Poly(ADP-Ribose) Polymerase Inhibition Improves Corneal Epithelial Innervation and Wound Healing in Diabetic Rats
Author Affiliations & Notes
  • Yong-Soo Byun
    Department of Ophthalmology and Visual Science, Seoul St. Mary's Hospital, Catholic University of Korea, College of Medicine, Seoul, Korea
    Catholic Institute for Visual Science, Catholic University of Korea, College of Medicine, Seoul, Korea
  • Borami Kang
    Division of Endocrinology and Metabolism, Department of Internal Medicine, Seoul St. Mary's Hospital, Catholic University of Korea, College of Medicine, Seoul, Korea
  • Young-Sik Yoo
    Catholic Institute for Visual Science, Catholic University of Korea, College of Medicine, Seoul, Korea
  • Choun-Ki Joo
    Department of Ophthalmology and Visual Science, Seoul St. Mary's Hospital, Catholic University of Korea, College of Medicine, Seoul, Korea
    Catholic Institute for Visual Science, Catholic University of Korea, College of Medicine, Seoul, Korea
  • Correspondence: Choun-Ki Joo, 222 Banpo-daero, Seocho-gu, 137-701 Seoul, Korea; [email protected]
Investigative Ophthalmology & Visual Science March 2015, Vol.56, 1948-1955. doi:https://doi.org/10.1167/iovs.14-16259
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Yong-Soo Byun, Borami Kang, Young-Sik Yoo, Choun-Ki Joo; Poly(ADP-Ribose) Polymerase Inhibition Improves Corneal Epithelial Innervation and Wound Healing in Diabetic Rats. Invest. Ophthalmol. Vis. Sci. 2015;56(3):1948-1955. https://doi.org/10.1167/iovs.14-16259.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: We evaluated the effect of poly(ADP-ribose) polymerase (PARP) inhibition by using 1,5-isoquinolinediol (ISO) on corneal epithelial innervation in diabetic rats.

Methods.: ISO (3 mg/kg, intraperitoneal) or vehicle was administered to rats with diabetes induced by streptozotocin for 4 weeks. Epithelial innervation, epithelial wound healing, and corneal sensation were evaluated in diabetic rats (DM rats), diabetic rats treated with ISO (DM-ISO rats), and nondiabetic (non-DM) rats. The density of epithelial innervation was calculated separately as nerve terminals and sub-basal nerve plexus by analyzing the images of whole-mount corneas. Healed areas of epithelial defect were measured at 0, 18, and 36 hours after creating a 4-mm wound on the cornea. Corneal sensitivity test was conducted using a Cochet-Bonnet handheld esthesiometer. Additionally, PARP1 and poly(ADP-ribosyl)ated polymers (pADPr) as its products, were identified in trigeminal ganglions (TGs) by Western blot analysis and immunofluorescence staining.

Results.: In DM rats, the density of nerve terminals (5.57% ± 0.94%) and sub-basal nerve plexus (22.08 ± 1.78 mm/mm2) was significantly reduced in comparison with that in DM-ISO rats (8.64% ± 1.42%, 30.82 ± 2.01 mm/mm2, respectively) and non-DM rats (9.02 ± 1.14%, 34.77 ± 4.45 mm/mm2, respectively). The percentages of healed area of the epithelial defects at 18 and 36 hours were significantly smaller in DM rats (23.8 ± 5.2%, 53.2 ± 4.6%, respectively) than in DM-ISO rats (43.2 ± 1.4%, 75.8 ± 2.2%, respectively) and non-DM rats (48.1 ± 8.6%, 86.1 ± 3.3%, respectively). Corneal sensitivity decreased in DM rats (51.1 ± 0.3 mm) but not in DM-ISO rats (57.8 ± 0.2 mm). There were no differences between parameters in DM-ISO rats and those in non-DM rats.

Conclusions.: Diabetic corneas showed loss of epithelial innervation, resulting in delayed epithelial healing and decreased corneal sensitivity. Inhibition of poly(ADP-ribose) polymerase (PARP) with 1,5-isoquinolinediol alleviated these diabetes-induced alterations in the corneal epithelium in the diabetic rats.

Corneal innervation plays a critical role in protecting corneas from mechanical, chemical, and thermal stimuli, as well as in producing trophic factors that are necessary for the maintenance of a healthy ocular surface.1,2 Decreased corneal nerve function is closely related to reduced lacrimal gland secretion and epitheliopathy, or poor epithelial healing. 
Diabetic neuropathy is a common complication of diabetes.3,4 Population-based studies have reported rates of diabetic neuropathy ranging from 8% to 54% for type 1 diabetes and from 13% to 46% for type 2 diabetes.5 An early and common finding associated with diabetic neuropathy is peripheral neuropathy, which is characterized by dysfunction and loss of Aδ- and C-type sensory small nerve fibers.6,7 The cornea is the most innervated part of the human body and receives Aδ- and C-type fibers derived from the ophthalmic division of the trigeminal nerve. For this reason, the cornea is highly affected in diabetic patients.8,9 Decreased corneal innervation and sensitivity in diabetic patients lead to impairment of epithelial wound healing, predisposing patients to sight-threatening complications such as stromal opacification, surface irregularity, and microbial infection.1014 Decreased corneal sensation with a loss of innervation is one of the main causes of corneal complications in diabetic patients. 
With an increase in interest and understanding of diabetic neuropathy, poly(ADP-ribose) polymerase (PARP) activation is emerging as a fundamental mechanism in the pathogenesis of diabetic complications, including diabetic neuropathy.1517 Several studies have shown that PARP plays a key role in the pathogenesis of diabetic neuropathy and that PARP inhibitors or gene deficiency reversed functional and metabolic abnormalities associated with diabetic neuropathy.1822 
To date, no study has examined the effect of PARP inhibition on corneal alterations induced by diabetes. Therefore, in the present study, we determined whether the PARP inhibitor 1,5-isoquinolinediol (ISO) protects against or reverses streptozotocin (STZ)-induced loss of epithelial innervation and promotes epithelial wound healing in rats with STZ-induced diabetes. Epithelial innervation, epithelial wound healing, and corneal sensation were evaluated in STZ-treated rats. Additionally, PARP1 and its products, poly(ADP-ribosyl)ated polymers (pADPr), were identified in trigeminal ganglions (TGs) by Western blot analysis and immunofluorescence staining. 
Methods
STZ-Induced Diabetes in Rats and PARP Inhibition With ISO Treatment
Eight-week-old male Sprague-Dawley (SD) rats weighing approximately 300 g (Oriental Bio, Inc., Seongnam, Korea) were used in this study. A single dose of STZ (55 mg/kg in 0.01 M citrate buffer, pH 4.5; Sigma-Aldrich, Munich, Germany) was administered intraperitoneally to induce diabetes.23 Four weeks after the administration of STZ, blood glucose levels were measured from the tail vein, using a commercial glucometer and disposable strips (Accu-Chek; Roche Diagnostics, Mannheim, Germany), and rats with glucose levels of 250 mg/dL (13.8 mM) or more were considered to be diabetic. In the nondiabetic, healthy control rats, the same volume of vehicle (0.01 M citrate buffer) was injected intraperitoneally. Treatment with the PARP inhibitor ISO (3 mg/kg, every other day, intraperitoneally) or vehicle continued from 8 to 12 weeks after STZ injection, for a total of 4 weeks.22 In this study, the experimental groups consisted of diabetic (DM) rats, diabetic rats treated with ISO (DM-ISO), and nondiabetic (non-DM) rats, which were age and sex matched. All experiments were performed in accordance with the Research in Vision and Ophthalmology statement for the Use of Animals in Ophthalmic and Vision Research. The Institutional Animal Care and Use Committee at the College of Medicine, Catholic University of Korea, approved the animal protocol. 
Abundance and Immunohistochemistry of PARP and Poly(ADP-Ribosyl)ated Polymers in TGs
After rats were euthanized, TGs were quickly dissected, immersed immediately in liquid nitrogen, and stored at −80°C to determine the abundance of PARP and its products, poly(ADP-ribosyl)ated polymers (pADPr), by Western blot analysis. TGs were homogenized in radioimmunoprecipitation assay buffer (20 mM Tris-HCl, pH 7.5, 0.1% [w/v] sodium lauryl sulfate, 0.5% [w/v] sodium deoxycholate, 135 mM NaCl, 1% [v/v] Triton X-100, 10% [v/v] glycerol, and 2 mM EDTA) supplemented with a protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific, Rockford, IL, USA). Tissue homogenates were centrifuged at 14,000g for 15 minutes, and the supernatant was collected. The total protein concentration was determined by bicinchoninic acid assay protein assay. Lysates were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was placed in blocking buffer (5% nonfat milk in 1× Tris-buffered saline with 0.1% Tween-20 [TBST]) for 1 hour at room temperature (RT) with shaking and then incubated with primary antibodies overnight at 4°C with shaking. Following incubation, the membrane was washed 3 times with TBST and incubated with horseradish-peroxidase-conjugated secondary antibodies in TBST for 1 hour at RT with shaking. After three washes with TBST and soaking with the chemiluminescent reagent, protein bands were detected using a ChemiDoc MP imaging system (Bio-Rad Laboratories, Hercules, CA, USA). ImageJ software (version 1.47; US National Institutes of Health, Bethesda, MD, USA) was used to calculate the signal intensity of the bands. The values were normalized relative to β-actin. The primary antibodies used were rabbit anti-PARP1 (Santa Cruz Biotechnology, Dallas, TX, USA), mouse anti-pADPr (Santa Cruz Biotechnology), and rabbit anti-β-actin (Cell Signaling Technology, Danvers, MA, USA). 
Some of the collected TGs were used for immunohistochemistry. After overnight fixation in 4% paraformaldehyde (PFA) at 4°C, the tissues were soaked in 30% sucrose for 24 hours and then frozen in optimal cutting temperature compound (Tissue-Tek, Sakura Fine Technical Co., Ltd., Tokyo, Japan) at −80°C. Cryostat sections (10 μm) were mounted on silane-coated slides (Matsunami, Osaka, Japan), which were permeabilized in 0.1 M phosphate-buffered saline with 0.1% Triton-X (PBST) and blocked in 0.1% PBST containing 1% bovine serum albumin and 5% donkey serum for 1 hour at RT. The slides were incubated with anti-PARP and anti-pADPr primary antibodies in 0.1% PBST overnight at 4°C and incubated with secondary antibodies in 0.1% PBST for 2 hours at RT. After being washed three times, slides were mounted with a 4′,6-diamidino-2-phenylindole (DAPI)-supplemented mounting medium and examined using fluorescence microscopy (Axio Observer; Carl Zeiss Meditec, Hamburg, Germany). 
Whole-Mount Cornea Staining and Quantitative Analysis of Epithelial Innervation
To evaluate the effects of diabetes and PARP inhibitor on epithelial innervation, whole-mount corneas were prepared according to a modified version of the method described by Dvorscak and Marfurt.24 After rats were euthanized, the eyeballs were extracted and immersed in 4% PFA on ice for 1 hour. The cornea, with 1 mm of the surrounding scleral rim, was dissected, and the iris and endothelium were removed. Corneas were fixed in 4% PFA for 1 hour at RT. After being washed in 0.1 M PBS, the corneas were incubated in 0.1% EDTA (Sigma-Aldrich) and 0.02% hyaluronidase (type IV-S; Sigma-Aldrich) at 37°C overnight. The corneas were rinsed 3 times for 15 minutes each and permeabilized in 0.3% PBST with 2% bovine serum albumin for 2 hours at RT, after which they were incubated with an NL637-conjugated anti β-III tubulin antibody (1:10 dilution; R&D Systems, Minneapolis, MN, USA), a pan-neuronal marker, in 0.1% PBST for 48 hours at 4°C. After incubation, the corneas were washed extensively for 1 hour in PBS. Whole corneas were cut into quadrants, flattened, mounted on glass slides with DAPI-supplemented mounting media (Vectashield, Vector Laboratories, Burlingame, CA, USA), and examined with a confocal laser microscope (model LSM700; Carl Zeiss Meditec). 
Using a modified version of a method from previous studies,25,26 we examined the density of epithelial innervation by using images taken from whole-mount corneas (n = 6). The densities of nerve terminals and sub-basal nerve plexuses were determined separately. The densities of nerve terminals were represented as the area fraction (percentage) occupied by epithelial nerve terminals, whereas the density of sub-basal nerve plexuses was expressed as the total length per unit area (mm/mm2) in the observed fields (640 × 640 μm; 2560 × 2560 pixels). Three representative areas 1 mm from the limbal margin were observed in each cornea, using a confocal laser microscope (LSM700; Carl Zeiss Meditec). Z-stack images with 2-μm intervals across the whole epithelium were collected and processed into all-in-focus images corresponding to nerve terminals (Fig. 1A) and sub-basal nerve plexuses (Fig. 1D), using the maximal intensity projection tool of the Zeiss software (ZEN, black 2010; Carl Zeiss Meditec). Following conversion to 8-bit gray scale images (Fig. 1B, 1E), the area occupied by the nerve terminals was measured using ImageJ software (Fig. 1C) and the fraction (percentage) of the total observed field (2560 × 2560 pixels) was calculated. The length of each fiber consisting of the sub-basal nerve plexus was measured by tracing it using NeuronJ software (version 1.42, a plug-in for ImageJ), and the total length of the sub-basal nerve plexuses in the observed field was presented (Fig. 1F). 
Figure 1
 
Quantitative analysis of epithelial innervation. Epithelial nerve terminals (A–C) and sub-basal nerve plexuses (D, F) were analyzed separately. All-in-focus images of nerve terminals (A) and sub-basal nerve plexuses (D) were acquired from corresponding Z-stacks with a maximal intensity projection method. After conversion to 8-bit gray scale images (B, E), the area occupied by nerve terminals was measured with the analyzing particles tool in ImageJ software (C). Sub-basal nerve plexuses in the observed area were traced semiautomatically (purple lines), and the total length of the traced nerves was calculated using NeuronJ, a plug-in for ImageJ software (F). Each area (160 × 160 μm) shown above accounts for 1/16 of the observed area (640 × 640 μm), which was analyzed in the study.
Figure 1
 
Quantitative analysis of epithelial innervation. Epithelial nerve terminals (A–C) and sub-basal nerve plexuses (D, F) were analyzed separately. All-in-focus images of nerve terminals (A) and sub-basal nerve plexuses (D) were acquired from corresponding Z-stacks with a maximal intensity projection method. After conversion to 8-bit gray scale images (B, E), the area occupied by nerve terminals was measured with the analyzing particles tool in ImageJ software (C). Sub-basal nerve plexuses in the observed area were traced semiautomatically (purple lines), and the total length of the traced nerves was calculated using NeuronJ, a plug-in for ImageJ software (F). Each area (160 × 160 μm) shown above accounts for 1/16 of the observed area (640 × 640 μm), which was analyzed in the study.
Epithelial Wound Healing Study
To evaluate the effects of diabetes and the PARP inhibitor on epithelial wound healing, unilateral, 4-mm-diameter, round epithelial wounds were made in the eyes of another experimental group (n = 6) after ISO or vehicle treatment for 4 weeks. A commercial excimer laser device (VISX S4, Santa Clara, CA, USA) was used to make an epithelial wound of an exact size and to uniformly remove the subepithelial nerves beneath the epithelial wound. According to modified protocols,27 rats were anesthetized with 50 mg/kg tiletamine plus zolazepam (Zoletil; Virbac, Carros, France) and 15 mg/kg xylazine hydrochloride (Rompun; Bayer, Leverkeusen, Germany) by intraperitoneal injection. Lidocaine ophthalmic solution drops (Alcon, Fort Worth, TX, USA) were administered to the eyes prior to the procedure. After a stainless washer with a 4.0-mm inner diameter was placed on the rat cornea, the corneal surface was ablated to a depth of 70 μm within a 4-mm zone, using the excimer laser (Fig. 2). Prophylactic antibiotic eye drops (Cravit, levofloxacin 0.5%; Santen, Osaka, Japan) were applied to the eye to prevent infection. Images of the epithelial defect were taken at 0, 18, and 36 hours following fluorescein dye instillation. The area of the epithelial defect was measured using ImageJ software, and the healed area (percentage) of the epithelial defects at each time point was calculated using the following equation: healed area (%) = [(original defect area − current defect area)/original defect area]. 
Figure 2
 
Epithelial wound creation by excimer laser. (A) A stainless steel washer of 4.0-mm inner diameter was laid on the rat cornea. (B) The epithelium was ablated to a depth of 40 μm in epithelial ablation mode, followed by a 30-μm-deep stromal ablation in phototherapeutic keratectomy mode, for a total depth of 70 μm (pulse rate, 10 Hz; laser energy, 160 mJ/cm2). All procedures were conducted under sterile conditions.
Figure 2
 
Epithelial wound creation by excimer laser. (A) A stainless steel washer of 4.0-mm inner diameter was laid on the rat cornea. (B) The epithelium was ablated to a depth of 40 μm in epithelial ablation mode, followed by a 30-μm-deep stromal ablation in phototherapeutic keratectomy mode, for a total depth of 70 μm (pulse rate, 10 Hz; laser energy, 160 mJ/cm2). All procedures were conducted under sterile conditions.
Corneal Sensitivity Test
Corneal sensation was measured in DM and DM-ISO rats after ISO (or vehicle) treatment for 4 weeks, as well as in non-DM rats, using a handheld esthesiometer (Cochet-Bonnet, Luneau Ophtalmologie, Chartres, France). According to previous protocol,28 rats were sedated by anesthetic inhalation (isoflurane; ChoongWae Pharmaceutics, Seoul, Korea) and restrained just before measurement. The nylon filament of the device, with a full extension of 60 mm, was allowed to contact the corneal center and retracted 5 mm until a positive blink was observed. The filament length (millimeters) that elicited a positive blinking response was recorded, and each test was repeated three times (n = 6). 
Data Analysis
Data are means ± SEM. Comparisons among DM, DM-ISO, and non-DM rats were performed using one-way ANOVA with Bonferroni post hoc test (SPSS version 19 software; IBM Corp., Armonk, NY, USA). Differences of P < 0.05 were considered statistically significant. 
Results
Body Weight and Blood Glucose Level
Body weights and blood glucose levels were monitored in all groups over a 12-week period (n = 6 per group). The body weights of DM rats (511.5 ± 30.2 g) and DM-ISO rats (499.8 ± 46.4.5 g) 12 weeks after STZ administration were significantly lower than that of non-DM rats (585.2 ± 29.5 g, P < 0.05) (Fig. 3A). DM rats and DM-ISO rats had weight gain restricted by diabetes 12 weeks after STZ injection, in contrast to that of non-DM rats, which showed gradual weight gain. 
Figure 3
 
Changes in body weight and blood glucose level. (A) Body weight of the non-DM rats gradually increased, whereas weight gain in the DM and DM-ISO rats was significantly restricted 12 weeks after STZ injection. (B) Blood glucose levels in the DM and DM-ISO rats were significantly higher than that in non-DM rats, and the DM and DM-ISO groups maintained blood glucose levels greater than 250 mg/dL (13.8 mM) through the experimental period. There were no differences between body weight or blood glucose levels in DM and those in DM-ISO rats. *P < 0.05, compared to non-DM rats; n = 6 in all groups.
Figure 3
 
Changes in body weight and blood glucose level. (A) Body weight of the non-DM rats gradually increased, whereas weight gain in the DM and DM-ISO rats was significantly restricted 12 weeks after STZ injection. (B) Blood glucose levels in the DM and DM-ISO rats were significantly higher than that in non-DM rats, and the DM and DM-ISO groups maintained blood glucose levels greater than 250 mg/dL (13.8 mM) through the experimental period. There were no differences between body weight or blood glucose levels in DM and those in DM-ISO rats. *P < 0.05, compared to non-DM rats; n = 6 in all groups.
Blood glucose concentrations were higher than 250 mg/dL (13.8 mM) in the DM (552 ± 47.2, 480 ± 50.5, and 564 ± 98.1 mg/dL at 4, 8, and 12 weeks, respectively) and DM-ISO rats (449 ± 52.4, 524 ± 56.7, and 462 ± 51.2 mg/dL at 4, 8, and 12 weeks, respectively) from 4 weeks after STZ injection (Fig. 3B). There were no differences between blood glucose levels in DM rats and those in DM-ISO rats, suggesting that the PARP inhibitor ISO did not affect blood glucose in the experimental diabetic rats, similar to the results for body weight. 
Effect of Diabetes and ISO Treatment on PARP Activity in TGs
Western blot analysis was used to determine the abundance of PARP1 and pADPr, an indicator of the activity of PARP, in TGs (Fig. 4A). Compared to that in non-DM rats, there were no changes in the abundance of PARP1 in the DM rats (142.7 ± 20.7%, P = 0.208) and the DM-ISO rats (114.7 ± 17.3%, P = 0.754). The abundance of pADPr was increased significantly in DM rats (257.4 ± 5.71%, P = 0.004) in comparison with that of non-DM rats, but not in comparison with that of DM-ISO rats (107.2 ± 6.9%, P = 0.631). This result suggests that diabetes and the PARP inhibitor ISO do not affect the expression of PARP1, whereas PARP activity, as measured by pADPr abundance, is increased by diabetes and inhibited by ISO. 
Figure 4
 
Quantification of PARP activity in TGs. (A) Representative bands of PARP1 and pADPr in TGs. Densitometric analysis of the bands was conducted using ImageJ software, and protein abundance in the DM and DM-ISO rats was represented as relative values (percentage) in comparison with that of non-DM rats. *P < 0.05, n = 3. (B) Representative immunofluorescence staining of TGs with PARP1 (FITC [green]) and pADPr (rhodamine [red]). PARP1 was immunolocalized dominantly in the nuclei of TG neuronal cells in all groups. Most TG neuronal cells (white arrow head) in DM rats showed strong pADPr immunoreactivity, whereas TG neuronal cells, which have no pADPr immunofluorescence or weak pADPr immunofluorescence (white arrows), were observed predominantly in TGs from DM-ISO and non-DM rats. Nuclei were counterstained with DAPI (blue). Scale bars: 50 μm.
Figure 4
 
Quantification of PARP activity in TGs. (A) Representative bands of PARP1 and pADPr in TGs. Densitometric analysis of the bands was conducted using ImageJ software, and protein abundance in the DM and DM-ISO rats was represented as relative values (percentage) in comparison with that of non-DM rats. *P < 0.05, n = 3. (B) Representative immunofluorescence staining of TGs with PARP1 (FITC [green]) and pADPr (rhodamine [red]). PARP1 was immunolocalized dominantly in the nuclei of TG neuronal cells in all groups. Most TG neuronal cells (white arrow head) in DM rats showed strong pADPr immunoreactivity, whereas TG neuronal cells, which have no pADPr immunofluorescence or weak pADPr immunofluorescence (white arrows), were observed predominantly in TGs from DM-ISO and non-DM rats. Nuclei were counterstained with DAPI (blue). Scale bars: 50 μm.
Immunohistochemistry experiments with TGs demonstrated PARP1 staining in the nuclei of all types of cells, and especially in the nuclei of TG neuronal cells, of which the afferent arms distribute in the corneas. The immunofluorescence of pADPr in the cytoplasm of TG neuronal cells in DM rats was much stronger than that in DM-ISO rats and non-DM rats, both of which showed only minor staining for pADPr (Fig. 4B). This immunostaining pattern was consistent with the Western blot results. 
Effect of Diabetes and ISO Treatment on Epithelial Innervation
The densities of nerve terminals and sub-basal nerve plexuses of the DM, DM-ISO, and non-DM rats were compared. In the DM rats, nerve terminals were distributed sparsely, and a loss of fibers consisting of sub-basal nerve plexuses was seen, whereas the nerve terminals and sub-basal nerve plexuses of the DM-ISO rats were as dense as those of the non-DM rats (Fig. 5A). We represent the density of nerve terminals as the fraction (percentage) of the area occupied by nerve terminals and the density of sub-basal nerve plexuses as the total length per unit area (mm/mm2) (Fig. 5B). DM rats showed significantly decreased density of nerve terminals (5.57 ± 0.94%) in comparison with that in non-DM (9.02 ± 1.14%, P = 0.001) or DM-ISO (8.64 ± 1.42%, P = 0.001, respectively) rats. There were no differences in densities of nerve terminals between the DM-ISO and non-DM rats (P = 0.34). The densities of the sub-basal nerve plexuses of the DM rats (22.08 ± 1.78 mm/mm2) were significantly lower than that of the non-DM (34.77 ± 4.45 mm/mm2, P = 0.002) or DM-ISO rats (30.82 ± 2.01 mm/mm2, P = 0.02). 
Figure 5
 
The effect of diabetes and ISO treatment on epithelial innervation. (A) Representative images of nerve terminals and sub-basal nerve plexuses in DM, DM-ISO, and non-DM rats. Nerve terminals ascend tortuously from the sub-basal nerve plexuses and terminate near the superficial layer of the epithelium. Nerve fibers consisting of sub-basal plexuses run parallel to each other, horizontal to the basement membrane of the basal cells of the epithelium. (B) Using image analysis, we quantified nerve density as the area fraction (percentage) occupied by nerve terminals and the total length per unit area (mm/mm2) of the sub-basal nerve plexus in the observed area (640 × 640 μm). n = 6 in all groups. *P < 0.05.
Figure 5
 
The effect of diabetes and ISO treatment on epithelial innervation. (A) Representative images of nerve terminals and sub-basal nerve plexuses in DM, DM-ISO, and non-DM rats. Nerve terminals ascend tortuously from the sub-basal nerve plexuses and terminate near the superficial layer of the epithelium. Nerve fibers consisting of sub-basal plexuses run parallel to each other, horizontal to the basement membrane of the basal cells of the epithelium. (B) Using image analysis, we quantified nerve density as the area fraction (percentage) occupied by nerve terminals and the total length per unit area (mm/mm2) of the sub-basal nerve plexus in the observed area (640 × 640 μm). n = 6 in all groups. *P < 0.05.
Effect of Diabetes and ISO Treatment on Epithelial Wound Healing
Images of epithelial defects stained with fluorescent dye were obtained at 0, 18, and 36 hours after wound creation (Fig. 6A). There were no complications related to the wound creation procedure. Following image analysis, the healed areas (percentages) of the epithelial defects of the DM, DM-ISO, and non-DM rats were compared at each time point (Fig. 6B). At 18 hours, the healed area of the DM rats (23.8 ± 5.2%) was significantly lower than that of the DM-ISO rats (43.2 ± 1.4%, P = 0.03) and non-DM rats (48.1 ± 8.6%, P = 0.007). At 36 hours, the healed area of the DM rats (53.2 ± 4.6%) was also significantly lower than that of the DM-ISO rats (75.8 ± 2.2%, P = 0.02) and non-DM rats (86.1 ± 3.3%, P = 0.004). DM-ISO rats showed no differences in healed areas at each time point in comparison with those in non-DM rats. 
Figure 6
 
Comparison of epithelial wound healing in DM, DM-ISO, and non-DM rats. (A) Defect areas are shown stained by fluorescence dye at 0, 18, and 36 hours. (B) DM rats showed a significantly lower percentage of healed areas at 18 and 36 hours than DM-ISO and non-DM rats. n = 6 in all groups; *P < 0.05.
Figure 6
 
Comparison of epithelial wound healing in DM, DM-ISO, and non-DM rats. (A) Defect areas are shown stained by fluorescence dye at 0, 18, and 36 hours. (B) DM rats showed a significantly lower percentage of healed areas at 18 and 36 hours than DM-ISO and non-DM rats. n = 6 in all groups; *P < 0.05.
Effect of Diabetes and ISO Treatment on Corneal Sensitivity
Corneal sensitivities were measured after ISO or vehicle treatment for 4 weeks, and the corneal sensitivities of the DM, DM-ISO, and non-DM rats were compared (Fig. 7). The corneal sensitivity of DM rats was significantly decreased (51.1 ± 0.3 mm) in comparison with that of the non-DM rats (59.3 ± 0.1 mm, *P < 0.001), and the corneal sensitivity of the DM-ISO rats was significantly increased (57.8 ± 0.2 mm, *P < 0.001) in comparison with that of the DM rats. There was no difference in corneal sensitivity between the DM-ISO and non-DM rats (P = 0.134). 
Figure 7
 
Comparison of corneal sensitivity in DM, DM-ISO, and non-DM rats. Values are presented as the longest length (millimeters) of the filament leading to a positive blinking response. The test was performed on both eyes three times (n = 6 in all groups). *P < 0.001.
Figure 7
 
Comparison of corneal sensitivity in DM, DM-ISO, and non-DM rats. Values are presented as the longest length (millimeters) of the filament leading to a positive blinking response. The test was performed on both eyes three times (n = 6 in all groups). *P < 0.001.
Discussion
Diabetic neuropathy manifests with a wide range of sensory, motor, and autonomic deficits. Peripheral and thin fibers, including Aδ- or unmyelinated C-fibers undergo the earliest damage during diabetes. Corneal innervation has become a widely used clinical marker for early detection of diabetic neuropathy, as well as for clinical assessment of diabetes.2937 Many researchers have investigated corneal innervation in diabetic animal models as a means to understand the pathogenesis of diabetic neuropathy. 
Our data revealed that the densities of epithelial nerve terminals and sub-basal nerve plexuses were significantly reduced in DM rats. We analyzed epithelial innervation by separately measuring the densities of nerve terminals and sub-basal nerve plexuses. Measuring the length (millimeter) by tracing each fiber is suitable for evaluating the sub-basal nerve plexuses, which run parallel to each other on a horizontal plane at the bottom of the epithelium. However, this method cannot be used to evaluate nerve terminals, which pass through the epithelium perpendicular to the horizontal plane of the epithelium in a tortuous pattern. Accordingly, the density of nerve terminals was measured as an area fraction (%) occupied by each terminal relative to the observed field, using an analyzing particle tool in ImageJ software.24,25 As a result, the density of epithelial innervation was reduced approximately 60% (61.8% for nerve terminals, 63.5% for sub-basal nerve plexus) in comparison with non-DM rats. These reductions in epithelial innervation were accompanied by delayed epithelial healing and decreased corneal sensitivity. These diabetes-induced alterations are consistent with the results reported by previous studies, although there are minor differences in outcomes due to differences in measurement and interpretation that are inherent in the use of animal models.26,35,3840 
The major finding of our study was that PARP inhibition prevented the loss of epithelial innervation and promoted epithelial wound healing, which was impaired in diabetic rats. PARP activation has been established to play a critical role in the pathogenesis of diabetic neuropathy. PARP, a nuclear enzyme involved in DNA repair and programmed cell death, is activated by oxidative-nitrosative injury and DNA single-strand breakage under hyperglycemic conditions. Hyperactivated PARP leads to NAD+ depletion and energy failure,17 changes in transcriptional regulation and gene expression,41 poly(ADP-ribosyl)ation, and inhibition of glyceraldehyde 3-phosphate dehydrogenase, resulting in activation of several signaling pathways related to diabetes complications.42 Furthermore, PARP inhibition via administration of PARP inhibitors or gene deficiency reversed diabetic neuropathy-induced alterations in animal studies.1922,4345 Diabetic or galactose-fed mice showed slower motor or sensory nerve conduction, enhanced poly(ADP-ribosyl)ation in nerve fibers, and nerve energy failure, whereas PARP-deficient mice were not affected.17 PARP inhibitors reversed the functional and metabolic abnormalities associated with diabetic neuropathy, such as loss of intraepidermal nerves, abnormal sensory response, delayed nerve conduction, and decreased nerve blood flow, NAD+-to-NADH ratio, phosphocreatine concentrations, and phosphocreatine-to-creatine ratio.1822 In the present study, the effect of PARP inhibition on diabetic neuropathy was identified in the cornea, especially in the epithelium, where the diabetic neuropathy manifested earliest. We demonstrated that ISO treatment for 4 weeks prevented the loss of epithelial innervation and normalized epithelial healing and corneal sensitivity in DM rats. The morphological findings were supported by the Western blot results and immunofluorescence staining of TGs, which showed that PARP activity was increased by diabetes and inhibited by ISO. 
There are two possible interpretations for the relationship of PARP inhibition to epithelial wound healing. First, the PARP inhibitor may protect against the loss of epithelial innervation due to diabetes, which sequentially affects epithelial wound healing. Corneal sensation and trophic factors released by nerves are critical for wound healing and homeostasis.46,47 Second, the PARP inhibitor may also affect the structural alterations or cellular mechanisms related to epithelial wound healing. As our data indicate so far, ISO treatment seems not to reverse or protect the structural alterations of epithelial cells in diabetic corneas (see Supplementary Fig. S1 for ultrastructure of corneal epithelial basement membrane), unlike its effectiveness on corneal innervation. PARP inhibitor, meanwhile, is more likely to play a role in the cellular response related to epithelial wound healing. There is some evidence that PARP is involved in EGFR signaling pathway-related epithelial healing,48,49 which is impaired in diabetes.50 We will soon report whether PARP activation is involved in the cellular mechanisms related to epithelial wound healing in diabetic corneas, and the effectiveness of PARP inhibitor, if so. 
In the present study, ISO was administered intraperitoneally at a dose of 3 mg/kg every other day for 4 weeks, according to a modified protocol.22 We did not observe any side effects related to ISO, which was administered intraperitoneally for 4 months. This result was consistent with those of previous studies. Administration of a PARP inhibitor for 9 months in STZ-induced diabetic rats did not produce any side effects,49 and PARP gene-deficient mice did not develop phenotypic changes.45 However, PARP is an important enzyme for DNA repair, and the long-term consequences of complete PARP inhibition under pathological conditions associated with oxidative stress, including diabetes, have not been determined, even in animal studies. To advance clinical research, comprehensive understanding of the consequences of long-term inhibition of PARP is necessary. In addition, an alternative strategy for the administration of PARP inhibitors, such as topical instillation or local injection, may be required for administration to the eye. 
In summary, our study showed that loss of epithelial innervation, impairment of epithelial healing, and decreased corneal sensation were prevented by treatment with a PARP inhibitor in DM rats. Further studies are needed to establish the molecular mechanism through which PARP inhibition produces its preventive effects, as well as to examine the possibility of topical use of PARP inhibitors. 
Acknowledgments
The authors thank Jun-Sub Choi and Jee-Won Mok for help with experimental advice at CIVS, Young-Shin Joo for help with animal care at IACUC, and Sang-Kyu Lee for help with setting the confocal laser microscope at CRCID. 
The study was partially supported by Grant HI14C3417 from the Korean Health Technology R and D Project, Ministry for Health and Welfare, Republic of Korea. The English edit was performed by professional editors at Editage, a division of Cactus Communications. 
Disclosure: Y.-S. Byun, None; B. Kang, None; Y.-S. Yoo, None; C.-K. Joo, None 
References
He J, Bazan HE. Mapping the nerve architecture of diabetic human corneas. Ophthalmology. 2012; 119: 956–964.
Muller LJ, Marfurt CF, Kruse F, Tervo TM. Corneal nerves: structure, contents and function. Exp Eye Res. 2003; 76: 521–542.
Won JC, Kim SS, Ko KS, Cha BY. Current status of diabetic peripheral neuropathy in Korea: report of a hospital-based study of type 2 diabetic patients in Korea by the diabetic neuropathy study group of the Korean diabetes association. Diabetes Metab J. 2014; 38: 25–31.
Tesfaye S, Selvarajah D. The Eurodiab study: what has this taught us about diabetic peripheral neuropathy?? Curr Diab Rep. 2009; 9: 432–434.
Ziegler D, Zentai CP, Perz S, et al. Prediction of mortality using measures of cardiac autonomic dysfunction in the diabetic and nondiabetic population: the MONICA/KORA Augsburg Cohort Study. Diabetes Care. 2008; 31: 556–561.
Tavee J, Zhou L. Small fiber neuropathy: a burning problem. Cleve Clin J Med. 2009; 76: 297–305.
Efron N. The Glenn A. Fry award lecture 2010: ophthalmic markers of diabetic neuropathy. Optom Vis Sci. 2011; 88: 661–683.
Lutty GA. Effects of diabetes on the eye. Invest Ophthalmol Vis Sci. 2013; 54: 81–87.
Schultz RO, Van Horn DL, Peters MA, Klewin KM, Schutten WH. Diabetic keratopathy. Trans Am Ophthalmol Soc. 1981; 79: 180–199.
Goldich Y, Barkana Y, Gerber Y, et al. Effect of diabetes mellitus on biomechanical parameters of the cornea. J Cataract Refract Surg. 2009; 35: 715–719.
Friend J, Ishii Y, Thoft RA. Corneal epithelial changes in diabetic rats. Ophthalmol Res. 1982; 14: 269–278.
Bikbova G, Oshitari T, Tawada A, Yamamoto S. Corneal changes in diabetes mellitus. Curr Diabetes Rev. 2012; 8: 294–302.
Cousen P, Cackett P, Bennett H, Swa K, Dhillon B. Tear production and corneal sensitivity in diabetes. J Diabetes Complications. 2007; 21: 371–373.
Dogru M, Katakami C, Inoue M. Tear function and ocular surface changes in noninsulin-dependent diabetes mellitus. Ophthalmology. 2001; 108: 586–592.
Obrosova IG, Drel VR, Pacher P, et al. Oxidative-nitrosative stress and poly(ADP-ribose) polymerase (PARP) activation in experimental diabetic neuropathy: the relation is revisited. Diabetes. 2005; 54: 3435–3441.
Obrosova IG, Julius UA. Role for poly(ADP-ribose) polymerase activation in diabetic nephropathy, neuropathy and retinopathy. Curr Vasc Pharmacol. 2005; 3: 267–283.
Obrosova IG, Li F, Abatan OI, et al. Role of poly(ADP-ribose) polymerase activation in diabetic neuropathy. Diabetes. 2004; 53: 711–720.
Drel VR, Lupachyk S, Shevalye H, et al. New therapeutic and biomarker discovery for peripheral diabetic neuropathy: PARP inhibitor, nitrotyrosine, and tumor necrosis factor-(alpha). Endocrinology. 2010; 151: 2547–2555.
Sharma SS, Kumar A, Kaundal RK. Protective effects of 4-amino1,8-napthalimide, a poly(ADP-ribose) polymerase inhibitor in experimental diabetic neuropathy. Live Sci. 2008; 82: 570–576.
Li F, Drel VR, Szabo C, Stevens MJ, Obrosova IG. Low-dose poly(ADP-ribose) polymerase inhibitor-containing combination therapies reverse early peripheral diabetic neuropathy. Diabetes. 2005; 54: 1514–1522.
Li F, Szabo C, Pacher P, et al. Evaluation of orally active poly(ADP-ribose) polymerase inhibitor in streptozotocin-diabetic rat model of early peripheral neuropathy. Diabetologia. 2004; 47: 710–717.
Ilnytska O, Lyzogubov VV, Stevens MJ, et al. Poly(ADP-ribose) polymerase inhibition alleviates experimental diabetic sensory neuropathy. Diabetes. 2006; 55: 1686–1694.
Mordes JP, Rossini AA. Animal models of diabetes. Am J Med. 1981; 70: 353–360.
Dvorscak L, Marfurt CF. Age-related changes in rat corneal epithelial nerve density. Invest Ophthalmol Vis Sci. 2008; 49: 910–916.
Zhivov A, Winter K, Hovakimyan M, et al. Imaging and quantification of subbasal nerve plexus in healthy volunteers and diabetic patients with or without retinopathy. PLoS One. 2013; 8: e52157.
Wang F, Gao N, Yin J, Yu FS. Reduced innervation and delayed re-innervation after epithelial wounding in type 2 diabetic Goto-Kakizaki rats. Am J Pathol. 2012; 181: 2058–2066.
Varela JC, Goldstein MH, Baker HV, Schultz GS. Microarray analysis of gene expression patterns during healing of rat corneas after excimer laser photorefractive keratectomy. Invest Ophthalmol Vis Sci. 2002; 43: 1772–1782.
Kim J, Kim NS, Lee KC, Lee HB, Kim MS, Kim HS. Effect of topical anesthesia on evaluation of corneal sensitivity and intraocular pressure in rats and dogs. Vet Ophthalmol. 2013; 16: 43–46.
Tavakoli M, Petropoulos IN, Malik RA. Assessing corneal nerve structure and function in diabetic neuropathy. Clin Exp Optom. 2012; 95: 338–347.
Shtein RM, Callaghan BC. Corneal confocal microscopy as a measure of diabetic neuropathy. Diabetes. 2013; 62: 25–26.
Tavakoli M, Kallinikos P, Iqbal A, et al. Corneal confocal microscopy detects improvement in corneal nerve morphology with an improvement in risk factors for diabetic neuropathy. Diabetes Med. 2011; 28: 1261–1267.
Tavakoli M, Quattrini C, Abbott C, et al. Corneal confocal microscopy: a novel noninvasive test to diagnose and stratify the severity of human diabetic neuropathy. Diabetes Care. 2010; 33: 1792–1797.
Pritchard N, Edwards K, Shahidi AM, et al. Corneal markers of diabetic neuropathy. Ocul Surf. 2011; 9: 17–28.
Kallinikos P, Berhanu M, O'Donnell C, Boulton AJ, Efron N, Malik RA. Corneal nerve tortuosity in diabetic patients with neuropathy. Invest Ophthalmol Vis Sci. 2004; 45: 418–422.
Davidson EP, Coppey LJ, Kardon RH, Yorek MA. Differences and similarities in development of corneal nerve damage and peripheral neuropathy and in diet-induced obesity and type 2 diabetic rats. Invest Ophthalmol Vis Sci. 2014; 55: 1222–1230.
Hossain P, Sachdev A, Malik RA. Early detection of diabetic peripheral neuropathy with corneal confocal microscopy. Lancet. 2005; 366: 1340–1343.
Edwards K, Pritchard N, Vagenas D, Russell A, Malik RA, Efron N. Standardizing corneal nerve fibre length for nerve tortuosity increases its association with measures of diabetic neuropathy. Diabetes Med. 2014; 31: 1205–1209.
Yin J, Huang J, Chen C, Gao N, Wang F, Yu FS. Corneal complications in streptozocin-induced type I diabetic rats. Invest Ophthalmol Vis Sci. 2011; 52: 6589–6596.
Leppin K, Behrendt AK, Reichard M, et al. Diabetes mellitus leads to accumulation of dendritic cells and nerve fiber damage of the subbasal nerve plexus in the cornea. Invest Ophthalmol Vis Sci. 2014; 55: 3603–3615.
Davidson EP, Coppey LJ, Holmes A, Yorek MA. Changes in corneal innervation and sensitivity and acetylcholine-mediated vascular relaxation of the posterior ciliary artery in a type 2 diabetic rat. Invest Ophthalmol Vis Sci. 2012; 53: 1182–1187.
Ha HC, Hester LD, Snyder SH. Poly(ADP-ribose) polymerase-1 dependence of stress-induced transcription factors and associated gene expression in glia. Proc Natl Acad Sci U S A. 2002; 99: 3270–3275.
Du X, Matsumura T, Edelstein D, et al. Inhibition of GAPDH activity by poly(ADP-ribose) polymerase activates three major pathways of hyperglycemic damage in endothelial cells. J Clin Invest. 2003; 112: 1049–1057.
Negi G, Kumar A, Sharma SS. Concurrent targeting of nitrosative stress-PARP pathway corrects functional, behavioral and biochemical deficits in experimental diabetic neuropathy. Biochem Biophys Res Commun. 2010; 391: 102–106.
Jangra A, Datusalia AK, Sharma SS. Reversal of neurobehavioral and neurochemical alterations in STZ-induced diabetic rats by FeTMPyP, a peroxynitrite decomposition catalyst and 1,5-Isoquinolinediol a poly(ADP-ribose) polymerase inhibitor. Neurol Res. 2014; 36: 619–626.
Obrosova IG, Xu W, Lyzogubov VV, et al. PARP inhibition or gene deficiency counteracts intraepidermal nerve fiber loss and neuropathic pain in advanced diabetic neuropathy. Free Radic Biol Med. 2008; 44: 972–981.
Sigelman S, Friedenwald JS. Mitotic and wound-healing activities of the corneal epithelium; effect of sensory denervation. AMA Arch Ophthalmol. 1954; 52: 46–57.
Mittag TW, Mindel JS, Green JP. Trophic functions of the neuron. V. Familial dysautonomia. Choline acetyltransferase in familial dysautonomia. Ann NY Acad Sci. 1974; 228: 301–306.
Hassa PO, Hottiger MO. A role of poly (ADP-ribose) polymerase in NF-kappaB transcriptional activation. Biol Chem. 1999; 380: 953–959.
Zheng L, Szabo C, Kern TS. Poly(ADP-ribose) polymerase is involved in the development of diabetic retinopathy via regulation of nuclear factor-kappaB. Diabetes. 2004; 53: 2960–2967.
Xu K, Yu FS. Impaired epithelial wound healing and EGFR signaling pathways in the corneas of diabetic rats. Invest Ophthalmol Vis Sci. 2011; 52: 3301–3308.
Figure 1
 
Quantitative analysis of epithelial innervation. Epithelial nerve terminals (A–C) and sub-basal nerve plexuses (D, F) were analyzed separately. All-in-focus images of nerve terminals (A) and sub-basal nerve plexuses (D) were acquired from corresponding Z-stacks with a maximal intensity projection method. After conversion to 8-bit gray scale images (B, E), the area occupied by nerve terminals was measured with the analyzing particles tool in ImageJ software (C). Sub-basal nerve plexuses in the observed area were traced semiautomatically (purple lines), and the total length of the traced nerves was calculated using NeuronJ, a plug-in for ImageJ software (F). Each area (160 × 160 μm) shown above accounts for 1/16 of the observed area (640 × 640 μm), which was analyzed in the study.
Figure 1
 
Quantitative analysis of epithelial innervation. Epithelial nerve terminals (A–C) and sub-basal nerve plexuses (D, F) were analyzed separately. All-in-focus images of nerve terminals (A) and sub-basal nerve plexuses (D) were acquired from corresponding Z-stacks with a maximal intensity projection method. After conversion to 8-bit gray scale images (B, E), the area occupied by nerve terminals was measured with the analyzing particles tool in ImageJ software (C). Sub-basal nerve plexuses in the observed area were traced semiautomatically (purple lines), and the total length of the traced nerves was calculated using NeuronJ, a plug-in for ImageJ software (F). Each area (160 × 160 μm) shown above accounts for 1/16 of the observed area (640 × 640 μm), which was analyzed in the study.
Figure 2
 
Epithelial wound creation by excimer laser. (A) A stainless steel washer of 4.0-mm inner diameter was laid on the rat cornea. (B) The epithelium was ablated to a depth of 40 μm in epithelial ablation mode, followed by a 30-μm-deep stromal ablation in phototherapeutic keratectomy mode, for a total depth of 70 μm (pulse rate, 10 Hz; laser energy, 160 mJ/cm2). All procedures were conducted under sterile conditions.
Figure 2
 
Epithelial wound creation by excimer laser. (A) A stainless steel washer of 4.0-mm inner diameter was laid on the rat cornea. (B) The epithelium was ablated to a depth of 40 μm in epithelial ablation mode, followed by a 30-μm-deep stromal ablation in phototherapeutic keratectomy mode, for a total depth of 70 μm (pulse rate, 10 Hz; laser energy, 160 mJ/cm2). All procedures were conducted under sterile conditions.
Figure 3
 
Changes in body weight and blood glucose level. (A) Body weight of the non-DM rats gradually increased, whereas weight gain in the DM and DM-ISO rats was significantly restricted 12 weeks after STZ injection. (B) Blood glucose levels in the DM and DM-ISO rats were significantly higher than that in non-DM rats, and the DM and DM-ISO groups maintained blood glucose levels greater than 250 mg/dL (13.8 mM) through the experimental period. There were no differences between body weight or blood glucose levels in DM and those in DM-ISO rats. *P < 0.05, compared to non-DM rats; n = 6 in all groups.
Figure 3
 
Changes in body weight and blood glucose level. (A) Body weight of the non-DM rats gradually increased, whereas weight gain in the DM and DM-ISO rats was significantly restricted 12 weeks after STZ injection. (B) Blood glucose levels in the DM and DM-ISO rats were significantly higher than that in non-DM rats, and the DM and DM-ISO groups maintained blood glucose levels greater than 250 mg/dL (13.8 mM) through the experimental period. There were no differences between body weight or blood glucose levels in DM and those in DM-ISO rats. *P < 0.05, compared to non-DM rats; n = 6 in all groups.
Figure 4
 
Quantification of PARP activity in TGs. (A) Representative bands of PARP1 and pADPr in TGs. Densitometric analysis of the bands was conducted using ImageJ software, and protein abundance in the DM and DM-ISO rats was represented as relative values (percentage) in comparison with that of non-DM rats. *P < 0.05, n = 3. (B) Representative immunofluorescence staining of TGs with PARP1 (FITC [green]) and pADPr (rhodamine [red]). PARP1 was immunolocalized dominantly in the nuclei of TG neuronal cells in all groups. Most TG neuronal cells (white arrow head) in DM rats showed strong pADPr immunoreactivity, whereas TG neuronal cells, which have no pADPr immunofluorescence or weak pADPr immunofluorescence (white arrows), were observed predominantly in TGs from DM-ISO and non-DM rats. Nuclei were counterstained with DAPI (blue). Scale bars: 50 μm.
Figure 4
 
Quantification of PARP activity in TGs. (A) Representative bands of PARP1 and pADPr in TGs. Densitometric analysis of the bands was conducted using ImageJ software, and protein abundance in the DM and DM-ISO rats was represented as relative values (percentage) in comparison with that of non-DM rats. *P < 0.05, n = 3. (B) Representative immunofluorescence staining of TGs with PARP1 (FITC [green]) and pADPr (rhodamine [red]). PARP1 was immunolocalized dominantly in the nuclei of TG neuronal cells in all groups. Most TG neuronal cells (white arrow head) in DM rats showed strong pADPr immunoreactivity, whereas TG neuronal cells, which have no pADPr immunofluorescence or weak pADPr immunofluorescence (white arrows), were observed predominantly in TGs from DM-ISO and non-DM rats. Nuclei were counterstained with DAPI (blue). Scale bars: 50 μm.
Figure 5
 
The effect of diabetes and ISO treatment on epithelial innervation. (A) Representative images of nerve terminals and sub-basal nerve plexuses in DM, DM-ISO, and non-DM rats. Nerve terminals ascend tortuously from the sub-basal nerve plexuses and terminate near the superficial layer of the epithelium. Nerve fibers consisting of sub-basal plexuses run parallel to each other, horizontal to the basement membrane of the basal cells of the epithelium. (B) Using image analysis, we quantified nerve density as the area fraction (percentage) occupied by nerve terminals and the total length per unit area (mm/mm2) of the sub-basal nerve plexus in the observed area (640 × 640 μm). n = 6 in all groups. *P < 0.05.
Figure 5
 
The effect of diabetes and ISO treatment on epithelial innervation. (A) Representative images of nerve terminals and sub-basal nerve plexuses in DM, DM-ISO, and non-DM rats. Nerve terminals ascend tortuously from the sub-basal nerve plexuses and terminate near the superficial layer of the epithelium. Nerve fibers consisting of sub-basal plexuses run parallel to each other, horizontal to the basement membrane of the basal cells of the epithelium. (B) Using image analysis, we quantified nerve density as the area fraction (percentage) occupied by nerve terminals and the total length per unit area (mm/mm2) of the sub-basal nerve plexus in the observed area (640 × 640 μm). n = 6 in all groups. *P < 0.05.
Figure 6
 
Comparison of epithelial wound healing in DM, DM-ISO, and non-DM rats. (A) Defect areas are shown stained by fluorescence dye at 0, 18, and 36 hours. (B) DM rats showed a significantly lower percentage of healed areas at 18 and 36 hours than DM-ISO and non-DM rats. n = 6 in all groups; *P < 0.05.
Figure 6
 
Comparison of epithelial wound healing in DM, DM-ISO, and non-DM rats. (A) Defect areas are shown stained by fluorescence dye at 0, 18, and 36 hours. (B) DM rats showed a significantly lower percentage of healed areas at 18 and 36 hours than DM-ISO and non-DM rats. n = 6 in all groups; *P < 0.05.
Figure 7
 
Comparison of corneal sensitivity in DM, DM-ISO, and non-DM rats. Values are presented as the longest length (millimeters) of the filament leading to a positive blinking response. The test was performed on both eyes three times (n = 6 in all groups). *P < 0.001.
Figure 7
 
Comparison of corneal sensitivity in DM, DM-ISO, and non-DM rats. Values are presented as the longest length (millimeters) of the filament leading to a positive blinking response. The test was performed on both eyes three times (n = 6 in all groups). *P < 0.001.
Supplementary PDF S1
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×