Alterations in the corneal nerve and stem/progenitor cells in diabetes: preventive effects of insulin-like growth factor-1 treatment

Abstract

This study aimed to investigate whether corneal nerve and corneal stem/progenitor cells are altered in insulin-like growth factor-I (IGF-I-) treated individuals with diabetes. A group consisting of db/db mice with type 2 diabetes mellitus (DM) and a wild-type group were assessed by neural and corneal stem/progenitor cell markers immunostaining and real-time PCR. Moreover, the expression of corneal nerve and stem/progenitor cell markers was examined in IGF-1-treated diabetic mice. Compared with a normal cornea, swelling and stratification of the corneal epithelium were noted in db/db mice. Beta-III tubulin immunostaining revealed that the corneal subbasal plexuses in diabetic mice were thinner with fewer branches. mRNA expression levels of Hes1, Keratin15, and p75 (corneal stem/progenitor cell markers) and the intensity and number of positive cells of Hes1 and Keratin19 immunostaining diminished in the diabetic corneas. Compared with the subbasal nerve density in the normal group, a decrease in the diabetic group was observed, whereas the corneal subbasal nerve density increased in IGF-1-treated diabetic group. The decreased expression of Hes1 and Keratin19 was prevented in IGF-1-treated diabetic group. Our data suggest that corneal nerve and stem/progenitor cells are altered in type 2 DM, and IGF-I treatment is capable of protecting against corneal damage in diabetes.

Figure 1

Ocular properties in ob/ob mice. Representative histological photomicrographs of hematoxylin and eosin-stained globe sections that were obtained from normal and 12-week-old diabetic mice. Swelling and stratification of the diabetic corneal epithelium are noted in comparison with the normal cornea. Magnification: 200x. Representative biomicroscopy of a normal cornea ((a) and (c)) and of a diabetic mouse cornea after 12 weeks ((b) and (d)). Representative histological photomicrographs of corneal nerves that were immunostained with an anti-β-III-tubulin fluorescein isothiocyanate-conjugated antibody showing the epithelial nerve (left column), subbasal nerve plexus (middle column), and stromal nerves (right column) in the normal eye ((e), (f), and (g)) and in a 12-week-old diabetic mouse eye ((h), (i), and (j)). Note the development of marked nerve degeneration after 12 weeks when the nerves appear faint and interrupted. Magnification: 200x.

Figure 2

Diminished mRNA expression levels of stem cell markers in diabetic cornea. The transcript levels of p75, hairy enhancer of split 1 (Hes1), Abcg2, and Keratin15 in the cornea are shown. The mRNA expression levels of p75, Hes1, and Keratin15 significantly decreased in diabetic corneas. A reduction of Abcg2 was also observed, although this was not statistically significant. 18S ribosomal RNA expression showed no significant difference between diabetic corneas and normal corneas. Statistical significance for mRNA levels is calculated on fold-change values over those of diabetic corneas. The dotted line represents the mRNA expression levels in normal corneas (normal expression of mRNA: 1.0). Statistical evaluation of data was performed with the unpaired t-test. The bars represent the mean ± standard error of the mean (SEM) (*P < 0.05).

Figure 3

Decreased expression levels of corneal progenitor/stem cell markers in diabetes. Representative immunostained micrographs of the cross-sections of the normal ((a) and (c)) and diabetic ((b) and (d)) corneas showing the expression of stem cell markers. Expression levels of Hes1 (b) and Keratin19 (d) decreased in diabetic eyes after 12 weeks compared with those in the normal eyes ((a) and (c)).

Figure 4

Effect of diabetes and treatment with recombinant human insulin-like growth factor-I (rhIGF-1) on corneal innervation and corneal progenitor/stem cells. Representative histological photomicrographs of corneal nerves that were immunostained with β-III-tubulin showing epithelial nerve (a), the subbasal nerve plexus (b), and stromal nerve trunks. (c) The rhIGF-I administration significantly accelerated recovery of the corneal subbasal nerve and epithelial branches. (d) Histogram of the quantification of the corneal subbasal nerve density. The subbasal nerve plexuses of the diabetic eyes completely diminished compared with that in the normal eyes, whereas those of rhIGF-1-treated diabetic corneas recovered. Statistical comparison of data between groups was performed using the nonparametric Mann-Whitney U test. The bars represent the mean ± SEM (*P < 0.00001, **P < 0.01). Representative immunostained micrographs of cross-sections of diabetic and rhIGF-1-treated corneas showing the expression of progenitor/stem cell markers. The decrease in Hes1 (e) and Keratin19 (f) staining was prevented in rhIGF-1-treated diabetic mice compared with diabetic mice (Figures 3(b) and 3(d)).