Rap1-interacting adapter molecule (RIAM) associates with the plasma membrane via a proximity detector

Abstract

Adaptive immunity depends on lymphocyte adhesion that is mediated by the integrin lymphocyte functional antigen 1 (LFA-1). The small guanosine triphosphatase Rap1 regulates LFA-1 adhesiveness through one of its effectors, Rap1-interacting adapter molecule (RIAM). We show that RIAM was recruited to the lymphocyte plasma membrane (PM) through its Ras association (RA) and pleckstrin homology (PH) domains, both of which were required for lymphocyte adhesion. The N terminus of RIAM inhibited membrane translocation. In vitro, the RA domain bound both Rap1 and H-Ras with equal but relatively low affinity, whereas in vivo only Rap1 was required for PM association. The PH domain bound phosphoinositol 4,5-bisphosphate (PI(4,5)P(2)) and was responsible for the spatial distribution of RIAM only at the PM of activated T cells. We determined the crystal structure of the RA and PH domains and found that, despite an intervening linker of 50 aa, the two domains were integrated into a single structural unit, which was critical for proper localization to the PM. Thus, the RA-PH domains of RIAM function as a proximity detector for activated Rap1 and PI(4,5)P(2).

Figure 1.

RIAM translocation to the PM in live cells. (A) GFP-tagged full-length (FL) RIAM, RIAM RA domain, RIAM RA-PH domains, or RalGDS-RBD (positive control probe for GTP-bound Rap1) were coexpressed in Jurkat T cells along with mCherry-tagged constitutively active Rap1V12 and the cells were imaged with a laser-scanning confocal microscope 24 h later. (B) Jurkat cells were homogenized by nitrogen cavitation before and after stimulation with anti-CD3 antibodies and membrane and cytosolic fractions were blotted for RIAM and RhoGDI (cytosolic control). Blots were quantified by Li-Cor and percent RIAM in the membrane fraction is indicated (n = 3, P < 0.02). (C and D) GFP-RA-PH (C) or GFP-RalGDS-RBD (D) was coexpressed with mCherry-tagged wild-type Rap1 in Jurkat T cells that were serum starved for 2 h and then imaged before and at the indicated times after the addition of cross-linking antibodies to CD3 (100 ng/ml). (E) The same experiment as in B substituting serum-starved COS-1 fibroblasts stimulated with 50 ng/ml EGF for the Jurkat T cells. Bars: (Jurkat cells) 5 µm; (COS-1 cells) 10 µm. Confocal images are representative of >80% of at least 20 cells imaged in at least three independent experiments.

Figure 2.

The N terminus of RIAM inhibits translocation to the PM. (A) The domain structure of RIAM includes a talin binding (TB) region, two coiled-coil (CC) regions, short and long polyproline (PP) regions, and the membrane targeting region consisting of RA and PH domains. The truncation mutants produced to analyze membrane association are aligned with full-length RIAM. (B) Co-expression of RIAM-GFP, tagged at the opposite end relative to GFP-RIAM, with constitutively active mCherry-Rap1V12 in Jurkat T cells. (C) Co-expression of GFP-RIAM or the indicated truncations with mCherry-Rap1V12 in Jurkat T cells. Bars: 5 µm.

Figure 3.

RIAM RA domain binding to Rap1 is required for PM recruitment. (A) Affinity purification of GFP-tagged RIAM RA-PH or RalGDS-RBD by the indicated concentrations of GST-tagged recombinant Rap1b loaded in vitro with either GDPβS or GTPγS. GFP fusion proteins were visualized by immunoblot for GFP. Immunoblots of the inputs are shown to the left. In the bottom panel, the input (lane 1) was run on a separate gel. The positions of the 55 or 43 kD markers are indicated for each gel. (B) Protein interaction assay as in A, tracking affinity purification of GFP-RIAM, GFP-RA-PH, and GFP-RA-PH K213A by 4 µM GDPβS- or GTPγS-loaded H-Ras or Rap1a. (A and B) the tiff images acquired with the Licor Odyssey were adjusted in Photoshop using the nonlinear Levels command. (C and E) Confocal localization of GFP-RA-PH with a K213A substitution predicted to inhibit RA function coexpressed with constitutively active mCherry-Rap1V12 in Jurkat T cells (C) or COS-1 cells (E). (D and F) Confocal localization of GFP-RA-PH coexpressed with wild-type mCherry-Rap1 in anti-CD3-stimulated Jurkat T cells (D) or EGF-stimulated COS-1 cells (F). (G) CFP-tagged dominant negative Hras17N, a potent and specific inhibitor of Ras signaling, was coexpressed in serum-starved COS-1 cells with GFP-RA-PH and mCherry-Rap1a and the cells were imaged before and after addition of EGF. (H) FLAG-Rap1GAP, a potent and specific inhibitor of Rap signaling, was coexpressed in serum-starved COS-1 cells with GFP-RA-PH and mCherry-Rap1 and the cells were imaged before and after addition of EGF. Bars: (Jurkat cells) 5 µm; (COS-1 cells) 10 µm.

Figure 4.

RIAM PH domain binds PI(4,5)P₂. (A) Colocalization in Jurkat T cells of mCherry-Rap1V12 and GFP-RA-PH with substitutions at the indicated basic amino acids that are predicted to be required for PIP binding. AAA indicates the triple mutation of K227A, K331A, and R333A. (B) Fluorescence polarization (FP) analysis of the binding of the indicated fluorescently labeled PIPs to recombinant RIAM RA-PH. K_d values obtained from the curve fittings are given on the right. As a result of the large protein demands of this experiment, it was performed once, although seven concentrations of RIAM RA-PH were measured independently for each PIP. (C) Effect of Wortmannin on Akt phosphorylation and GFP-RA-PH recruitment to PM in a Jurkat T cells coexpressing mCherry-Rap1V12. The tiffs shown in the immunoblot were adjusted in Photoshop with the nonlinear Levels command. The nearest molecular mass marker is shown to the left. (D) Localization of the PI(4,5)P₂ probe, PH-PLCδ-GFP, or GFP-RA-PH in COS-1 cells coexpressing Rap1V12, an FRB domain tethered to the PM, and a cytosolic PI(4,5)P₂-directed 5-phosphatase fused with FKBP12, before and after addition of rapamycin to induce heterodimerization of FKBP12 and FRB. (E) Localization of PH-PLCδ-GFP or GFP-RA-PH in COS-1 cells coexpressing mCherry-Rap1V12 and a Gq/PLCβ linked angiotensin II receptor before and after addition of angiotensin II. Bars: (Jurkat cells) 5 µm; (COS-1 cells) 10 µm.

Figure 5.

Functional RA and PH domains are required for adhesion. Jurkat T cells were electroporated with the indicated constructs and allowed to recover for 48 h. Cells were fluorescently labeled with BCECF, treated with or without 5 µg/ml anti-CD3 or 100 ng/ml PMA for 30 min at 37°C and allowed to adhere to ICAM-1–coated wells. After gentle washing of nonadherent cells, percent adhesion was determined by comparing input fluorescence to remaining fluorescence. Data shown are mean ± SEM, n = 3 (*, P < 0.02; **, P < 0.0005). The immunoblot in the lower panel shows equivalent expression of each RIAM construct. The nearest molecular mass marker is shown to the left.

Figure 6.

Crystal structure of the RA-PH domains of mouse RIAM. (A) Ribbon diagram of the crystal structure of RIAM RA-PH. The RA domain is colored yellow, the PH domain is colored cyan, and the intervening linker is colored gray. Residues in the RA domain that are predicted to interact with active GTPases are colored red, and the β1-β2 loop in the PH domain, the presumed site of PIP binding, is colored magenta. Secondary-structure elements (α helices and β strands) within the RA and PH domains are labeled, as are the N and C termini. (B) Model of the interaction between RIAM RA-PH and GTP-loaded Rap1. The model is based on the crystal structure of Grb14 RA-PH bound to H-Ras (not depicted). The same orientation and coloring as in A is shown, with Rap1 colored green and GTP colored black and shown in stick representation. The figure is rendered with PyMOL.

Figure 7.

Integrated structure of the RA-PH domains is required for PM association of RIAM. (A) Schematic representation of RIAM RA-PH domain tandem probes used in B–E. An X through a domain indicates that it has been inactivated by mutation (K213A for RA and K327A/K331A/R333A for PH). (B and C) Localization of GFP-2xRAPH expressed alone in Jurkat T cells (B) or with mCherry-Rap1V12 (C). (D) Localization of the indicated GFP-tagged tandem RA-PH probes in Jurkat T cells coexpressing mCherry-Rap1V12 at near endogenous levels or active HrasV12 expressed 10–100-fold above endogenous (grayscale images to right). (E) Chart summarizing PM recruitment. Bars: 5 µm.

Figure 8.

Model for RIAM translocation to the PM of lymphocytes. Although Rap1 constitutively associates with multiple cellular membranes by virtue of its geranylgeranyl modification, the relevant pool for LFA-1 regulation is at the PM where it is activated by one of two GEFs, C3G or CD-GEF1, that are downstream of both the TCR and chemokine receptors (A). In the resting state, RIAM is cytosolic and its N terminus inhibits the RA-RH membrane association domains. This inhibition may be alleviated by the interaction of talin with its binding region at the N terminus of RIAM (B). The exposed RA-PH domain is now free to associate with the PM where the RA-PH domains can sense the coincidence of GTP-bound Rap1 and the constitutively high concentrations of PI(4,5)P₂ (C). The domains are not drawn to scale. In particular, the C terminus that includes the proline rich (PR) region is reduced.