T Helper 1 Cellular Immunity Toward Recoverin Is Enhanced in Patients With Active Autoimmune Retinopathy

Abstract

Autoimmune retinopathy (AIR) causes rapidly progressive vision loss that is treatable but often is confused with other forms of retinal degeneration including retinitis pigmentosa (RP). Measurement of anti-retinal antibodies (ARA) by Western blot is a commonly used laboratory assay that supports the diagnosis yet does not reflect current disease activity. To search for better diagnostic indicators, this study was designed to compare immune biomarkers and responses toward the retinal protein, recoverin, between newly diagnosed AIR patients, slow progressing RP patients and healthy controls. All individuals had measurable anti-recoverin IgG and IgM antibodies by ELISA regardless of disease status or Western blot results. Many AIR patients had elevated anti-recoverin IgG1 levels and a strong cellular response toward recoverin dominated by IFNγ. RP patients and controls responded to recoverin with a lower IFNγ response that was balanced by IL-10 production. Both AIR and RP patients displayed lower levels of total peripheral blood mononuclear cells that were due to reductions of CD4⁺ T_H cells. A comparison of messenger RNA (mRNA) for immune-related genes in whole blood of AIR patients versus RP patients or controls indicated lower expression of ATG5 and PTPN22 and higher expression of several genes involved in T_H cell signaling/transcription and adhesion. These data indicate that an immune response toward recoverin is normal in humans, but that in AIR patients the balance shifts dramatically toward higher IFNγ production and cellular activation.

Keywords: Nanostring technologies; autoimmunity; cancer immunotherapy; cell mediated immunity; interferon-gamma; interleukin 10; recoverin; retinal diseases.

Figure 1

Anti-retinal antibodies are more common in serum of AIR patients than in non-autoimmune RP patients or controls. Serum was collected from newly diagnosed and untreated AIR (A#) patients, RP (R#) patients, or healthy control (C#) individuals. Human retinas were isolated from cadaver eyes provided by the Michigan Eye Bank/Eversight. Retinal proteins were prepared by homogenization and 30 μg of protein was separated on a 10% NuPAGE gel by using standard electrophoresis methods. Proteins were transferred to a nitrocellulose membrane that was used to capture anti-retinal antibodies from a 1:100 dilution of patient serum. Binding was detected using biotinylated anti-human IgG antibody, horseradish peroxidase-conjugated streptavidin, and SuperSignal West Pico Chemiluminescent Substrate. Exposure to auto-radiography film was for 10 minutes, and SeeBlue Plus 2 molecular weight standard was run on each gel to align protein sizes. Representative blots are shown for RP patients and controls, but all individuals were tested.

Figure 2

Anti-recoverin antibodies are frequently present in blood of patients and controls. Plasma was separated from heparinized blood of untreated AIR patients (n = 15), RP patients (n = 15), and healthy controls (n = 14). Half of the wells of a flat-bottomed 96 well EIA plate were coated overnight with 2.5 μg/mL recombinant human recoverin (Recov) in carbonate buffer, washed, and blocked with PBS 2% BSA for 2 h. Control wells were similarly treated without addition of recoverin (NoAg). A 1:40 dilution of plasma was added in triplicate to Recov and NoAg wells for each sample for 2 h. Plates were then washed and binding of patient antibodies was detected in replicate wells using biotin-conjugated anti-human IgM-, anti-human IgG-, or anti-human IgG1-specific antibodies, followed by addition of streptavidin-HRP and TMB substrate. Each dot represents the mean 450 nm absorbance for one individual run in triplicate. Black lines and hash marks show the median and 95% confidence interval for each group and the exact P-values as determined by Mann-Whitney tests are shown for the indicated pairwise comparisons.

Figure 3

Comparison of PBMC cytokine production in response to recoverin. Isolated PBMC from AIR patients (n = 15), RP patients (n = 15) and healthy controls (n = 14) were cultured for six days in the presence or absence of 2.5 μg/mL of recombinant human recoverin. Culture supernatants were collected and the levels of released IFNγ and IL-10 were determined by commercial sandwich ELISA. Each dot represents the mean cytokine level from triplicate wells for each individual. The group median and 95% confidence intervals are shown with horizontal lines and hash marks, respectively. Exact P-values determined by non-parametric Mann-Whitney tests are expressed below the title of each graph. The ratio of IFNγ to IL-10 production for each individual is shown on the lower graph.

Figure 4

Comparison of major lymphocyte subsets in blood of AIR or RP patients and controls. Isolated PBMC from 15 AIR patients, 15 RP patients and 14 healthy controls were counted and stained with fluorochrome-conjugated antibodies specific for human CD3, CD4, CD8, CD56, and CD19 (A–C). (A) Peripheral blood mononuclear cells were counted using trypan blue and then stained for flow cytometry. Lymphocytes were gated as small, non-granular cells using forward scatter and side scatter. Percentages of CD56⁺CD3^neg natural killer cells, CD19⁺CD3^neg B cells, CD8⁺CD3⁺ cytotoxic T cells, and CD4⁺CD3⁺ T helper cells within the lymphocyte gate were determined for each individual. Horizontal lines mark the median and 95% confidence intervals for each group. Exact P-values from Mann-Whitney tests are shown below the title of each graph. (B) The percentages of lymphocytes, monocytes and granulocytes were determined using forward scatter and side scatter and multiplied by total PBMC counts to determine the number of each cell type per microliter of blood. Stacking bars represent the mean for each patient group. (C) Bar graph of average numbers of lymphocytes in each subset which were determined by multiplying the percentages and total lymphocyte counts for each sample. (D) A separate staining panel consisting of antibodies against CD3, CD4, CD8, CD45RA, CD45RO, and CCR7 was used to find percentages of naïve, central memory and effector memory T helper cell subsets (see Materials and Methods for criteria) which were multiplied by total numbers of CD4⁺CD3⁺ T_H cells in each sample.

Figure 5

Differences in immune mediator gene expression between AIR patients and control groups. Total RNA was isolated from 2.5 mL of whole blood collected in PAXgene tubes from 14 individuals in each group. Quality of RNA was confirmed in house prior to sending the sample for analysis of mRNA copy numbers using the NanoString Human Immunology Panel. Statistical analysis was performed on the top 445 expressed genes of the 579 immune-associated genes on the panel, of which 53 were very differentially expressed between AIR patients and healthy controls (adjusted P-value <0.025). The graphs show normalized and batch-corrected mRNA copy numbers for twelve of the most different genes with each dot representing a different patient. (A) Expression of the transcription factors: nuclear factor of activated T cells 1, signal transducer and activator of transcription 5A and 5B, and interferon response factor 5. (B) Expression of regulators of cell activation: protein tyrosine phosphatase non-receptor type 22, autophagy related 5, caspase 3, and retinoic acid receptor responder 3. (C) Expression of the two subunits (CD11a and CD18) of the cell adhesion molecule: lymphocyte function-associated antigen 1 (LFA-1); expression of CXCR4, the receptor for the chemokine stromal cell-derived factor-1 (SDF-1, CXCL12); and expression of the chemokine interleukin-8 (CXCL8) that is an attractant for neutrophils to sites of infection or inflammation.