The Physiological and Pathophysiological Role of Nerve Growth Factor at the Ocular Surface

Purpose: Nerve growth factor (NGF) has been considered as a pleiotrophic factor that is expressed in the nervous, immune and structural tissues of the body, and affects nerve growth, regulates immune responses and influences the survival and functions of structural cells. However, the role of NGF at the ocular surface is less well understood compared with it effects in most other systems of the body. Therefore the purposes of the present study were: (1) To investigate the concentration of NGF in normal human tear film and determine if it is correlated with TGF-β1 in dry eye associated with established contact lens wear. (2) To investigate if exogenous NGF alters the expression of cytokines and growth factors by corneal and conjunctival epithelial cells, and affects cell viability in a dry eye culture model. (3) To determine the effect of NGF on migration and differentiation of murine monocytic lineage cells in vivo and in vitro. Methods: (1) The concentration of NGF in the tear film of normal subjects and established contact lens wearers was measured by enzyme-linked immunosorbent assay (ELISA). The correlation of NGF levels to ocular surface dry eye index (OSDI), Schirmer test, corneal sensitivity and corneal innervation was determined. Tear film TGF-β1 was also measured by ELISA to test if its levels also correlated with NGF. (2) SV40 transformed human corneal epithelial cells (SV40-HCEC), primary cultured HCEC (P-HCEC) or a conjunctival cell line (IOBA-NHC) were treated for 24 or 48 hrs with serum-free growth-media of varying osmolality alone or also containing NGF (10 ng/ml

• 2 - or 100 ng/ml). Culture media was collected and RNA was extracted and real time reverse transcription polymerase chain reaction (RT-PCR) was performed to detect mRNA expression of IL-13, IL-16, IL-1α, MCP-1, TLR4, 9, HLA-DR, TGF-β1 and 2. ELISA was performed to quantitate TGF-β1 and 2 levels in culture media and cell lysate. TrkA expression was assessed with real-time RT-PCR and flow cytometry. Cell viability was assessed by trypan blue exclusion test and MTT assay. (3) Murine monocytic lineage cells were isolated from peripheral blood, spleen and peritoneal cavity by a density gradient and negative selection with immunomagnetic beads against lymphocyte markers (B220/CD4/CD8). Purity was assessed by flow cytometry. Monocyte migration after 3hrs of treatment using various concentrations of NGF (1~200ng/ml) was tested with a blind-well migration chamber. The expression of surface markers (CD11c, CD11b) was evaluated by flow cytometry. For in vivo studies, intrastromal injection of NGF (1μg/ml, 2μl) was performed and then expression of CD11c, CD115 and CD11b were assessed by immunohistochemistry. The percentage of CD11c, Ly6C, CD115 positive cells in the epithelium and in the stroma was determined with flow cytometry 4 hours after NGF injection. The concentration of MCP-1 in the cornea was measured by ELISA. Results: (1) There was a statistically significant decrease of corneal sensitivity in contact lens wearers compared with normal controls. The nerve density in the subepithelial plexus of contact lens wearers with dry eye was 538.8 ± 39.3 μm/image (3.959 ± 0.28 pm/ μm2), and 537.1 ± 30.9 μm/image (3.947 ± 0.27 pm/ μm2) in those without dry eye. Both of these values were significantly (p = 0.032) lower than in the normal controls (4.412 ± 0.21 pm/ μm2). The concentration of tear NGF was increased in
• 3 - contact lens wearers with dry eye, and was significantly greater (p = 0.048) compared with contact lens wearers without dry eye. TGF-β1 increased one fold in contact lens associated dry eye, and correlated with NGF levels. (2) IL-1α, IL-13, IL-16, MCP-1, and TLR4 mRNA expression was up-regulated by hyperosmolar stress and this increase was abrogated by NGF. TLR9 expression was not significantly changed by hyperosmolar stress or NGF treatment. HLA-DR decreased with hyperosmolar stress but did not change significantly with NGF. The expression of TGF-β mRNA and protein in all cell types was up-regulated by hyperosmolar stress 1.5 to 9 fold, and this was further enhanced in the presence of NGF. TrkA mRNA and protein were also up-regulated by NGF. Hyperosmolar stress induced approximately 35% conjunctival epithelial cell death but this was reduced to only 18% when NGF was present. (3) The purity of isolated monocytes was more than 90% as tested by flow cytometry. Blind-well chamber assays showed that NGF induced migration of monocytes in a concentration dependent fashion, with the highest concentration (200 ng/ml) being as effective as MCP-1 (5ng/ml). In vitro, NGF also increased expression of CD11c, CD11b, and MHCII on blood monocytes, suggesting that they differentiated into dendritic like cells. In vivo experiments showed that NGF induced migration of mouse monocytic lineage cells into the corneal stroma and this may be explained by increased MCP-1 secretion by stromal keratocytes. Conclusion: (1) Tear film NGF levels were elevated in contact lens related dry eye, likely in response to anti-inflammatory factors such as TGF-β1. (2) The results for these studies imply that NGF may modulate ocular surface inflammation in dry eye by increasing TGF-β expression and by reducing epithelial cell death. Further these data
• 4 - support a potential role for NGF in treating dry eye inflammation. (3) The specific studies showed that NGF induced migration of monocytes and promoted differentiation into a more mature antigen presenting cell phenotype. This suggests that NGF may facilitate corneal immunity through its effect on monocytes and monocyte-derived cells.