Abstract
Purpose :
Several diagnoses are associated with lacrimal gland dysfunction and dry eye disease, but the recognized immunopathologies do not account for the known spectrum of histopathological presentations. As the existence of unrecognized disease phenotypes may hinder design and testing of therapeutics, we devised new protocols for inducing lacrimal gland disease models in rabbits and a new approach to assessing disease phenotype diversity.
Methods :
A gland lysate post-nuclear membrane fraction was used to immunize rabbits. Microparticles (exosomes, MP) secreted by primary acinar cells were isolated from supernatant media and loaded into bone marrow monocyte-derived dendritic cells (mDC). Peripheral blood lymphocytes (PBL) were activated in ex vivo mixed cell reactions with LPS-stimulated, MP-loaded mDC. Activated PBL were used to transfer disease to autologous lacrimal glands (primary adoptively transferred disease, 1°ATD). Glands were profiled for T cell- and bone marrow-derived cell numbers, immune response-related gene transcript abundances, and atrophy. Principal Component Analysis (PCA) was used to form hypotheses for immune cell- and epithelial cell functional clusters determining normal- and disease phenotypes.
Results :
Adoptive transfer to naïve animals, systemic immunization, and adoptive transfer to systemically immunized animals led to a variety of disease phenotypes. The most severe developed in 1°ATD glands of animals that had been systemically immunized. PCA suggested that the major source of variation among these was the balance between two competing functional clusters, one comprising cells that, collectively, expressed CD4, perforin, and IL-17A, the other comprising cells that, collectively, expressed CD8, TNF-α, IL-1, IL-18, and MMP-9. Perforin- and IL-17A-expressing functional clusters were strongly associated with increasing atrophy; development of such clusters was related to robustness of cellular autoimmune responses, rather than humoral autoimmune responses, elicited by systemic immunization.
Conclusions :
The newly-devised protocols are suitable for large-scale studies. Each induces a diversity of disease phenotypes, which may be described as combinatorial products of distinct immune cell functional clusters. Characterization of predominant pathological functional clusters may reveal new therapeutic targets. Prospectively stratifying study participants according to disease phenotype may improve clinical trial outcomes.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.