Cis and trans regulatory mechanisms control AP2-mediated B cell receptor endocytosis via select tyrosine-based motifs

Abstract

Following antigen recognition, B cell receptor (BCR)-mediated endocytosis is the first step of antigen processing and presentation to CD4+ T cells, a crucial component of the initiation and control of the humoral immune response. Despite this, the molecular mechanism of BCR internalization is poorly understood. Recently, studies of activated B cell-like diffuse large B cell lymphoma (ABC DLBCL) have shown that mutations within the BCR subunit CD79b leads to increased BCR surface expression, suggesting that CD79b may control BCR internalization. Adaptor protein 2 (AP2) is the major mediator of receptor endocytosis via clathrin-coated pits. The BCR contains five putative AP2-binding YxxØ motifs, including four that are present within two immunoreceptor tyrosine-based activation motifs (ITAMs). Using a combination of in vitro and in situ approaches, we establish that the sole mediator of AP2-dependent BCR internalization is the membrane proximal ITAM YxxØ motif in CD79b, which is a major target of mutation in ABC DLBCL. In addition, we establish that BCR internalization can be regulated at a minimum of two different levels: regulation of YxxØ AP2 binding in cis by downstream ITAM-embedded DCSM and QTAT regulatory elements and regulation in trans by the partner cytoplasmic domain of the CD79 heterodimer. Beyond establishing the basic rules governing BCR internalization, these results illustrate an underappreciated role for ITAM residues in controlling clathrin-dependent endocytosis and highlight the complex mechanisms that control the activity of AP2 binding motifs in this receptor system.

Figure 1. CD79b Y195 Controls BCR Internalization.

Panel A, Amino acid sequences of the cytoplasmic domains of CD79a and CD79b. YxxØ putative AP2 binding motifs underlined. Panel B, Confocal laser scanning microscopic analysis of the endocytosis of indicated MHC class II-CD79 chimeric proteins. Plasma membrane is yellow (anti-MHC class II-Alexa 488+ post-endocytic labeling of external MHC class II-CD79 with an Alexa 594-labeled antibody). Internalized MHC class II-CD79 is green (anti-MHC class II-Alexa 488 only). Panel C, Quantification of MHC class II-CD79 endocytosis. 100+ cells from across 3 independent experiments were scored for internalization. Reported is the percent of cells showing internalized MHC class II-CD79 for each construct. Statistical comparisons were made between the construct with both CD79 cytoplasmic domains and cells expressing other constructs.

Figure 2. Ultrastructural Colocalization of Ligand-BCR Complexes and AP2 in Clathrin-Coated Pits.

A20µWT B cells or primary murine splenocytes were pulsed with anti-IgM-btn followed by anti-biotin-15 nm gold (arrow heads), incubated 2 minutes at 37° and then plasma membrane rips were prepared as previously reported . The exposed cytoplasmic face of the plasma membrane was stained with anti-AP2 and Protein A-5 nm gold (arrows). The percent of BCR-containing CCP that also stained for AP2 is indicated below each image. Inset: AP2 and BCR co-localization within electron dense membrane regions lacking discernable CCP architecture. Shown are representative images from 1 of 3 experiments, with 1,000+ BCR-bound gold particles or 100+ CCP photographed cumulatively.

Figure 3. AP2µ Binds to Isolated CD79a but not CD79b.

Panel A, Amino acid sequences of CD79a and CD79b cytoplasmic domains. YxxØ putative AP2 binding motifs underlined. Panel B, AP2µ expressed as a Gal4 activation domain fusion protein was assayed for specific interaction with the cytoplasmic domain of either CD79a or CD79b fused to the Gal4 DNA binding domain. Growth on histidine deficient (His-) plates indicates an AP2–CD79 interaction. The cytoplasmic domain of TGN38 contains a known AP2 binding YxxØ motif and served as a positive control, while the cytoplasmic domain of OCA2 contains a dileucine motif (which does not bind AP2µ) and served as a negative control. Data are representative of 2 experiments. Panel C, Diagram of the GST-CD79 cytoplasmic domain–AP2µ direct binding assay. Panel D, The cytoplasmic domains of CD79a and CD79b were expressed as GST fusion proteins in BL21 E. coli cells. GST-fusion proteins were captured from cell lysates on glutathione beads and the resulting matrix was tested for binding to in vitro translated, biotin-labeled AP2µ. The AP2 binding motif from TGN38, (SDYQRL)₃, and a non-AP2-binding derivation containing a tyrosine to glycine substitution, (SDGQRL)₃, fused to GST served as positive and negative controls, respectively. Binding is expressed as a percentage of (SDYQRL)₃–AP2 interactions. Data is the mean of 3 independent experiments ± S.E.M. Statistical comparisons were measured between SDYQRL and other samples.

Figure 4. Receptor Endocytosis of Isolated CD79a but not CD79b.

Endocytosis of the indicated MHC class II-CD79 fusion proteins was analyzed (Panel A) and quantitated (Panel B) as in Figure 1. Statistical comparisons were made between the reporter proteins with both CD79 cytoplasmic domains and other reporter proteins.

Figure 5. The Membrane-Proximal YxxØ Motif of Isolated CD79a Mediates AP2 Binding and Endocytosis.

Panel A, The binding of AP2µ-btn to bead-captured GST bearing full length CD79a with the indicated mutations was determined as in Figure 3. Panels B and C, Endocytosis of the indicated MHC class II-CD79 fusion proteins was analyzed and quantitated as in Figure 1. Statistical comparisons were made between the reporter protein expressing the wild type CD79a cytoplasmic domains and other reporter proteins.

Figure 6. AP2µ Directly Binds the YxxØ Motif Centered on the Membrane-Proximal Tyrosine of Isolated CD79a.

Panel A, The binding of AP2µ-btn to bead-captured GST bearing 21 amino acid peptides centered on the five CD79 YxxØ motifs was determined as in Figure 3. Binding is expressed as a percentage of CD79a–AP2 interactions and represents the mean of 3 independent experiments ± SEM. Statistical comparisons were measured between CD79a and other samples. Panel B, For both CD79a and CD79b, a positive or negative regulatory motif lies within +/−10 amino acids of the tyrosine residue of the membrane-proximal YxxØ AP2 binding motif. In this example, the motif is arbitrarily depicted as being downstream of the YxxØ motif.

Figure 7. Hierarchical Binding of AP2µ to BCR-derived Minimal YxxØ Motifs.

Panel A, Amino acid sequences of CD79a and CD79b cytoplasmic domains. YxxØ putative AP2 binding motifs underlined. Panel B, Diagram of the synthetic peptide–AP2µ binding assay. Panel C, 18 amino acid long peptides of the form [xxYxxØ]₃ and corresponding to each of the five putative AP2µ binding motifs of CD79 were synthesized and used at a final concentration of 250 µM to block the binding of AP2µ-btn to beads coated with GST-TGN38 (GST–[SDYQRL]₃). Peptides were also tested across a decreasing range of concentration (Figure S2). Data is presented as the level of BCR-derived inhibitor peptide binding to AP2µ-btn (percent inhibition of binding to bead-associated target) and is the mean of 3 independent experiments ± S.E.M. and data were normalized to the background binding of AP2 to the non-AP2 binding target GST–[SDGQRL]₃.

Figure 8. C-terminal Deletions Localize the Motif that Drives AP2 Binding to the Membrane Proximal YxxØ Motif of CD79a.

Binding of AP2µ-btn to bead-captured GST bearing the indicated C-terminal deletions of CD79a was determined as in Figure 3. Data is the mean of 3 independent experiments ± S.E.M. and data were normalized to the binding of AP2 to full length CD79a.

Figure 9. AP2-mediated CD79-driven Endocytosis is Regulated in Cis via ITAM-embedded DCSM and QTAT motifs.

Panel A, Binding of AP2-btn to bead-captured GST bearing the indicated CD79a or CD79b cytoplasmic domains with the indicated mutations was determined as in Figure 3. Binding is expressed as a percentage of wild type CD79a binding and represents the mean of 3 independent experiments ± SEM. Statistical comparisons were measured between the non-mutated CD79a or CD79b_DCSM cytoplasmic domains and other samples. Panels B and C, Endocytosis of the indicated MHC class II-CD79 fusion proteins was analyzed and quantitated as in Figure 1. Statistical comparisons were made between the reporter protein expressing the wild type CD79a or CD79b cytoplasmic domains and other reporter proteins. Panel D, Diagrammatic representation of cis and trans regulation of BCR AP2 binding and endocytosis. The open arrows indicate cis regulation by the DCSM and QTAT regulatory motifs. The closed arrows represent trans regulation of the endocytic activity of each cytoplasmic domain by the presence of the partner cytoplasmic domain.