Ubiquitin-dependent control of class II MHC localization is dispensable for antigen presentation and antibody production

Abstract

Controlled localization of class II MHC molecules is essential for proper class II MHC-restricted antigen presentation and the subsequent initiation of an adaptive immune response. Ubiquitination of class II MHC molecules on cytosolic lysine (K225) of the β-chain has been shown to affect localization of the complex. We generated mice in which the endogenous β-chain locus is replaced with a GFP tagged mutant version that lacks the cytosolic lysine residue (I-A-β-K225R-EGFP). These mice have elevated levels of class II MHC as compared to I-A-β-EGFP mice, and immature bone marrow-derived dendritic cells show redistribution of class II MHC to the cell surface. Nonetheless, in these same cells efficiency of antigen presentation is unaffected in I-A-β-K225R-EGFP mice, as assayed for presentation of ovalbumin to appropriately specific T cells. The I-A-β-K225R-EGFP animals have normal CD4 T cell populations and are capable of generating antigen-specific antibody in response to model antigens and viral infection. We therefore conclude that in our experimental system modulation of trafficking by ubiquitination of residue K225 of the β-chain is not essential for the function of class II MHC products in antigen presentation or antibody production.

Figure 1. Generation of I-A-β-K225R-EGFP mice.

A, Schematic of the approach used to target the I-A^b-β locus. The exon structure of the I-A^b-β locus is shown with the locations of the signal peptide (SP), extracellular domain (Extracellular), transmembrane domain (TM), and cytoplasmic tail (Cytoplasmic) indicated. The targeting construct consisted of a long arm of homology (LAH) consisting of exons 2–4, the K225R mutation introduced into exon 4, exons 5 and 6 fused to each other and to EGFP, a floxed self-excisable neomycin resistance cassette and a short arm of homology (SAH) consisting of the 3′UTR. Not drawn to scale. B, Splenocytes from I-A-β-EGFP or I-A-β-K225R-EGFP mice were stained with antibodies to detect B cell (B220), dendritic cell (CD11c) and T cell (CD3) populations and were analyzed by flow cytometry. Class II MHC, as detected from EGFP fluorescence, was present in B cells and dendritic cells, and class II MHC levels were increased in cells isolated from I-A-β-K225R-EGFP mice. C, Total splenocytes were isolated from heterozygous I-A-β-EGFP/HA-Ub and I-A-β-K225R-EGFP/HA-Ub mice and subjected to immunoprecipitation with an antibody against the MHC II α chain and subsequently to immunoprecipitation with an antibody against GFP. Total lysates were subjected to SDS-PAGE and immunoblot analysis with anti-GFP and anti-HA. All experiments were performed at least twice; representative experiments are shown.

Figure 2. Total and cell surface levels of class II MHC are increased in I-A-β-K225R-EGFP mice.

A, Splenocytes from B6 (+), I-A-β-EGFP (WT) and I-A-β-K225R-EGFP (KR) mice were analyzed by flow cytometry for class II levels. Total levels of class II MHC were determined from GFP fluorescence and the levels of cell surface class II MHC was assessed by staining with antibodies against I-A-β. Total and cell surface levels of class II MHC are increased in both B cells and dendritic cells of I-A-β-K225R-EGFP mice relative to I-A-β-EGFP and B6. This analysis was performed 3 times with a total of 8 mice of each genotype; representative data are shown. B, Splenocytes were lysed in NP40 and analyzed by SDS-PAGE. Samples were either boiled, or incubated at room temperature (non-boiled) to distinguish between the class II MHC complex and free I-Aα and I-Aβ chains. GFP-tagged I-Aβ was detected with antibodies against GFP, while the untagged I-Aβ was detected with the JV2 antibody. Levels of the mutant I-Aβ increased in I-A-β-K225R-EGFP cells while wild type I-A-β levels in the same cells remained constant. C, Cells were analyzed as in B, and immunoblotted with antibodies directed against I-Aα. Levels of I-Aα increase in I-A-β-K225R-EGFP cells showing that both components of the class II MHC molecule are affected. Immunoblotting analysis was performed at least twice; representative data are shown.

Figure 3. Synthesis and maturation of class II MHC is normal in I-A-β-K225R-EGFP mice.

A, Splenocytes were pulsed with [³⁵S]-cystine/methionine for 30 minutes and chased for the indicated times. Immunoprecipitations were performed with the I-Aα-specific antibody JV1, precipitates were eluted at room temperature (left panel) or boiled (right panel). B, Experiments were performed as in (A), immunoprecipitates were boiled to elute samples. C and D, Quantification of the β-EGFP and Ii-p31 data in (B). Closed circles: I-A-β-EGFP, open circles: I-A-β-K225R-EGFP. These experiments were performed three separate times; representative data is shown.

Figure 4. I-A-β-K225R-EGFP is redistributed to the cell surface.

BMDCs from I-A-β-EGFP were transduced with I-A-β-K225R-Cherry and the localization of both molecules was visualized by confocal microscopy (top panels). This experiment was performed on four separate occasions. 39 cells expressing both proteins were visualized and 85% of these showed increased surface localization of I-A-β-K225R-Cherry compared to I-A-β-GFP. Representative pictures are shown. The bottom panels show the converse experiment in which I-A-β-K225R-EGFP BMDCs were transduced with I-A-β-Cherry. This experiment was performed twice and the results were similar to those in the converse experiment. Representative pictures are shown.

Figure 5. Immune cell populations are not affected in I-A-β-K225R-EGFP mice.

Splenocytes from B6 (+/+), I-A-β-EGFP (WT/WT) and I-A-β-K225R-EGFP (KR/KR) mice were isolated, stained with the indicated antibodies and analyzed by flow cytometry. Shown are data from a representative experiment. This analysis was performed 3 separate times on a total of 8 mice of each genotype.

Figure 6. Antigen presentation efficiency is normal in I-A-β-K225R-EGFP BMDCs.

A, BMDCs were incubated with the indicated concentrations of ovalbumin (Ova) for one hour and subsequently fixed. T cells from OTII mice were incubated with the fixed BMDCs overnight and upregulation of CD69 on T cells was assessed by flow cytometry. B, Assays were performed as in A. BMDCs were incubated with 2 µg/ml Ova for the indicated times prior to fixation. These experiments were performed twice with a total of 4 mice of each genotype; representative data are shown.

Figure 7. Basal and antigen specific antibody production are not altered in I-A-β-K225R-EGFP mice.

A, Serum from peripheral blood was analyzed for immunoglobulin titers for each of the indicated isotypes by ELISA. B and C, Mice were immunized with 10 µg hen egg lysozyme (HEL) and 10 µg Ova on day 1 and day 21. Serum from peripheral blood was obtained on day 1 prior to immunization and from the same mice on day 28. Ova (B) and HEL (C) specific IgGs were detected by ELISA. D, Mice were infected with the influenza virus on day 1. Serum from peripheral blood was obtained on day 0 and on day 14. Flu specific IgGs were detected by ELISA. All experiments were performed two separate times with a total of 8 mice of each genotype; representative data are shown.