Expression of HLA-B27 causes loss of migratory dendritic cells in a rat model of spondylarthritis

Abstract

Objective: In rats transgenic for human HLA-B27 and β(2) -microglobulin (B27-transgenic rats), colitis and peripheral inflammation develop spontaneously. Therefore, B27-transgenic rats provide a model of spondylarthritis. Because inflammation in these rats requires CD4+ T lymphocytes and involves intestinal pathology, we hypothesized that dendritic cells (DCs) that migrate from the intestine and control CD4+ T cell differentiation would be aberrant in B27-transgenic rats.

Methods: Migrating intestinal lymph DCs were collected via thoracic duct cannulation from B27-transgenic and control (HLA-B7-transgenic or nontransgenic) rats. The phenotypes of these DCs and of mesenteric lymph node DCs were assessed by flow cytometry. The ability of DCs to differentiate from bone marrow precursors in vitro was also assessed.

Results: Lymph DCs showed increased activation and, strikingly, lacked the specific DC population that is important for maintaining tolerance to self-antigens. This population of DCs was also depleted from the mesenteric lymph nodes of B27-transgenic rats. Furthermore, in vitro culture of DCs from bone marrow precursors revealed a defect in the ability of B27-transgenic rats to produce DCs of the migratory phenotype, although the DCs that were generated induced enhanced interleukin-17 (IL-17) production from naive CD4+ T cells.

Conclusion: We describe 2 different mechanisms by which HLA-B27 may contribute to inflammatory disease: increased apoptotic death of B27-transgenic DCs that normally function to maintain immunologic tolerance and enhanced IL-17 production from CD4+ T cells stimulated by the surviving B27-transgenic DCs.

Figure 1

Systemic disease in HLA–B27–transgenic rats on a PVG background. B27-transgenic rats were backcrossed to a PVG background for >10 generations. Representative images of the hind feet, head, and back of wild-type rats (A, C, and E) and 4-week-old B27-transgenic rats (B, D, and E) are shown. The B27-transgenic rats clearly displayed nail dystrophy, hair loss on the head and back, and reddening of the feet.

Figure 2

Presence of inflammation in the intestines of HLA–B27–transgenic (B27-TG) rats on a PVG background. Intestinal tissue specimens from nontransgenic and B27-transgenic rats were obtained under terminal anesthesia and fixed in formalin. Fixed specimens were paraffin-embedded, sectioned, and stained with hematoxylin and eosin. B27-transgenic rats had shortened, inflamed villi in duodenal, jejunal, and ileal sections. Inflammatory infiltrates were observed in the lamina propria throughout the small intestine of B27-transgenic rats (arrowheads).

Figure 3

Characteristics of dendritic cells (DCs) migrating from the intestine of HLA–B27–transgenic (B27-TG) rats. Thoracic duct lymph was collected from nontransgenic and B27-transgenic rats and analyzed by flow cytometry. A, DCs were identified by their expression of CD103 and class II major histocompatibility complex (MHCII). Lymph was collected for 24 hours, either immediately after cannulation (0–24 hours) or for the subsequent 24 hours (24–48 hours). Both the percentage and the absolute number of DCs from each rat were calculated. Values are the percent total cells represented by DCs. B, Data for multiple nontransgenic and B27-transgenic rats at both time points were compiled and analyzed. Bars show the mean. C, Representative histograms with gating of DCs stained for CD80, CD86, or CD25 from nontransgenic rats (shaded), B27-transgenic rats (thick line), or isotype controls (thin line) are shown. D, CD80 and CD25 data for multiple nontransgenic and B27-transgenic rats at 0–24 hours and 24–48 hours were pooled and compared. Bars show the mean ± SEM. * = P < 0.05 by Student’s t-test. E, Allogeneic mixed lymphocyte reactions using flow-sorted lymph DCs from B27-transgenic rats and nontransgenic littermates were performed. Proliferation was measured by incorporation of ³H-thymidine after 5 days. Values are the mean ± SEM of triplicate wells for DCs from nontransgenic and B27-transgenic rats; the results are representative of 5 experiments.

Figure 4

Absence of a population of migratory DCs in B27-transgenic rats. Thoracic duct lymph was collected from nontransgenic, B7-transgenic, and B27-transgenic rats and analyzed by flow cytometry. A, Class II MHC–positive CD103+ cells were analyzed for their expression of CD172a and CD11b. Values in the dot plots are percentages. The experiment was repeated 3 times, and both the percentage and the absolute numbers of DCs in each subset were calculated. Bars show the mean ± SEM. B, Mesenteric lymph nodes were collected from nontransgenic, B7-transgenic, and B27-transgenic rats, and the proportion of CD172a− cells within the class II MHC–positive CD103+ lineage–negative (T cell receptor α/β−, CD45RA−, Igκ− CD45RC−) cells were analyzed by flow cytometry. Values in the dot plots are percentages. C and D, Flow cytometric analysis was performed to determine the number (C) and percent (D) of CD172a− cells. A representative histogram shows the expression of CD172a in CD103+ cells from nontransgenic rats (shaded), B7-transgenic rats (thin line), and B27-transgenic rats (thick line) (C). The percentages of CD172a− cells, as gated in C, from 3 individual experiments were compared (D). Data in A were analyzed by one-way analysis of variance. Data in D were analyzed by Student’s t-test. * = P < 0.05; ** = P < 0.01; *** = P < 0.001. See Figure 3 for other definitions.

Figure 5

Effects of HLA–B27 expression on cultured bone marrow–derived DCs. Bone marrow cells from nontransgenic and B27-transgenic rats were cultured for 7 days and analyzed by flow cytometry. A, Flt-3 ligand-supplemented cultures of DCs from B27-transgenic rats contained a higher proportion of small granular cells, as shown by the forward scatter (FSC) and side scatter (SSC) characteristics of the cells. B and C, The proportion of live (annexin V–negative/DAPI-negative) cells in cultures of B27-transgenic rat bone marrow–derived DCs was significantly lower than that in cultures of DCs derived from both nontransgenic and B7-transgenic rats. D and E, Treatment with Q-VD-OPh (Q-VD) restored the proportion of live cells in the B27-transgenic rat bone marrow–derived DC cultures (D), but the effect of HLA–B27 on bone marrow–derived DCs was not observed when bone marrow–derived DCs were treated with with granulocyte-macrophage–colony-stimulating factor (GM-CSF) (E). Values in or adjacent to the quadrants in the dot plots represent the percentage of cells therein. Bars in C show the mean ± SEM of 3 independent experiments. * = P < 0.05; *** = P < 0.001, by two-way analysis of variance. See Figure 3 for other definitions.

Figure 6

Functions of bone marrow–derived DCs in B27-transgenic rats. Bone marrow cells from nontransgenic and B27-transgenic rats were cultured with Flt-3 ligand (Flt-3L) for 7 days and analyzed by flow cytometry. A, Cultures of bone marrow–derived DCs from B27-transgenic rats contained a smaller proportion of live (annexin V–negative/DAPI-negative) CD103+ class II MHC–positive cells. B, The addition of lipopolysaccharide (LPS) to the cultures for the final 12 hours before analysis resulted in increased expression of class II MHC. Values in the dot plots represent the percentage of CD103+ cells. C, Flt-3L–supplemented bone marrow–derived DCs from B27-transgenic rats were significantly less efficient at stimulating proliferation from naive CD4+ T cells compared with nontransgenic bone marrow–derived DCs. **** = P < 0.0001 versus nontransgenic, by two-way analysis of variance (ANOVA). The results are representative of 4 experiments. D and E, The concentration of cytokines in coculture supernatants was measured. Nontransgenic and B27-transgenic DCs induced similar concentrations of interleukin-2 (IL-2), IL-6, IL-10, and tumor necrosis factor α (TNFα), while nontransgenic DCs induced higher concentrations of interferon-γ (IFNγ), and B27-transgenic DCs induced higher concentrations of IL-17. Values in C–E are the mean ± SEM of 3 experiments. * = P < 0.05 by ANOVA. See Figure 3 for other definitions.