Purpose: The purpose of this study was to investigate the role of substance P (SP) in patients affected with corneal neovascularization (CNV) and in three different Tac1-knockout (KO) murine models of CNV. Methods: SP levels in tears were measured with a multiplex bead assay. The extent of human CNV was quantified as number of affected corneal quadrants. Murine CNV was induced in both strains by means of total disepithelization, alkali burn, and intrastromal sutures. After death, CNV (blood and lymphatic) and leukocyte infiltration were quantified by CD31, LYVE1, and CD45 immunofluorescence, respectively. Trigeminal ganglions were collected for quantitatitive PCR IL1β quantification. Hematoxylin-eosin corneal cross sections and whole-mounted β-3-tubulin nerve staining were used to compare anatomy and nerve density of wild-type (WT) versus Tac1-KO normal mice. Results: SP tear levels correlate positively with CNV extension in patients (r = 0.49, P = 0.03). After disepithelization, Tac1-KO corneas showed reduced blood and lymphatic vascularization (-34% and -51% respectively) compared with the WT counterpart. CD45+ leukocytes infiltrating the cornea were reduced in Tac1-KO mice as opposed to WT in the disepithelization (P = 0.0001), alkali burn (P = 0.0258), and suture (P = 0.0149) models. Tac1-KO mice showed reduced IL1β expression in the trigeminal ganglion. Normal WT and Tac1-KO corneas did not show significant differences in transparency, thickness, and nerve density. Conclusions: Our results suggest (1) the involvement of SP in human CNV; (2) the key role of SP in promoting inflammatory CNV in three different mouse models; and (3) that absence of SP is not associated with obvious ocular surface pathology in a KO model.
Purpose: The purpose of this study was to investigate the role of substance P (SP) in patients affected with corneal neovascularization (CNV) and in three different Tac1-knockout (KO) murine models of CNV. Methods: SP levels in tears were measured with a multiplex bead assay. The extent of human CNV was quantified as number of affected corneal quadrants. Murine CNV was induced in both strains by means of total disepithelization, alkali burn, and intrastromal sutures. After death, CNV (blood and lymphatic) and leukocyte infiltration were quantified by CD31, LYVE1, and CD45 immunofluorescence, respectively. Trigeminal ganglions were collected for quantitatitive PCR IL1β quantification. Hematoxylin-eosincorneal cross sections and whole-mounted β-3-tubulin nerve staining were used to compare anatomy and nerve density of wild-type (WT) versus Tac1-KO normal mice. Results: SP tear levels correlate positively with CNV extension in patients (r = 0.49, P = 0.03). After disepithelization, Tac1-KO corneas showed reduced blood and lymphatic vascularization (-34% and -51% respectively) compared with the WT counterpart. CD45+ leukocytes infiltrating the cornea were reduced in Tac1-KO mice as opposed to WT in the disepithelization (P = 0.0001), alkali burn (P = 0.0258), and suture (P = 0.0149) models. Tac1-KO mice showed reduced IL1β expression in the trigeminal ganglion. Normal WT and Tac1-KO corneas did not show significant differences in transparency, thickness, and nerve density. Conclusions: Our results suggest (1) the involvement of SP in human CNV; (2) the key role of SP in promoting inflammatory CNV in three different mouse models; and (3) that absence of SP is not associated with obvious ocular surface pathology in a KO model.