One of the best-characterized examples of specialized signaling domains where the dynamic assembly of both small- and large-scale protein associations have a crucial role, is the contact region formed between a T cell and a target or an antigen-presenting cell (APC), termed the immunological synapse (IS) [63,64]. Formation of the IS is initiated by recognition of the peptide antigen presented in complex with MHC glycoproteins on the APC or target cell by the T cell receptor complex (TCR) on the appropriate T cell [65,66]. Recognition of the antigen causes the redistribution of numerous molecules in both cells and leads to the formation of well-defined junctional structures at the interface of the two cells. Participation of the TCR along with adhesion and co-stimulatory molecules, co-receptors and associated cytosolic signaling elements on T cells as well as recruitment of MHC-peptide complexes and adhesion molecules on APCs can be observed in all ISs [67-69]. Supramolecular organization of these molecules is generated by their clustering/segregation at both the micrometer and submicrometer (nanometer) scale. ISs mature through discrete stages characterized by high-order temporal and spatial cooperation of multiple elements (membrane proteins, signaling molecules) required for appropriate function [66-68].
Involvement of lipid rafts in T cell signaling in general, and in the IS formation in particular, has been extensively studied and firmly established [19,29,70,71]. It is hypothesized that upon synapse formation small, individual rafts are assembled into raft macrodomains in a process regulated by the actin cytoskeleton . Assembly of raft macrodomains occurs in three stages. Engagement of TCR initiates signals for actin polymerization followed by actin- and myosin-dependent migration of rafts to the site of cell signaling. Raft microdomains coalesce and, as a consequence, signaling proteins residing in discrete domains are now proximal within the same microenvironment, which enhances their interactions and amplifies the initial signals resulting in further raft assembly and signal amplification. The cascade continues until arrested by stop signals, which might include protein tyrosine phosphatases .
Despite the common features, ISs display diversity both in function and architecture, depending on the nature of interacting cells and as a consequence, that of the interacting molecular partners [67-69,72]. In the prototypic "bull-eye" structure, a central supramolecular activation cluster (cSMAC) containing TCR and MHC-peptide is surrounded by a pSMAC (peripheral SMAC) of adhesion molecules [63,64,73,74]. The "secretory synapse" formed between cytotoxic T cells (CTLs) and their targets has a double cSMAC containing the secretory apparatus juxtaposed with the TCR cluster . The formation of several small, distinct TCR clusters instead of a large, single one (multifocal synapses) or nonclustered, diffuse distribution of TCR (nonsegregated IS) in the contact zone was also detected in many cases [69,72]. Dynamic contacts allowing migration (amoeboid movement) of T cells across the surface of the counterpart cells are presumably non-segregated but at the same time mobile structures (migratory synapses) .
Fig. 7.1 Measurement of MHC I homoasso-ciation on OCM-3 uveal melanoma cells. The solid line denotes the distribution of FRET efficiency values between MHC I molecules measured by flow cytometry on a cell-by-cell basis. MHC I was targeted by Cy3- and Cy5-conjugated L368 mAbs specific for p2-micro-globulin (p2m, light chain of MHC I). The
Fig. 7.1 Measurement of MHC I homoasso-ciation on OCM-3 uveal melanoma cells. The solid line denotes the distribution of FRET efficiency values between MHC I molecules measured by flow cytometry on a cell-by-cell basis. MHC I was targeted by Cy3- and Cy5-conjugated L368 mAbs specific for p2-micro-globulin (p2m, light chain of MHC I). The positive FRET value (peaking at ~10%) indicates homoassociation of MHC I. As a positive control, the intramolecular FRET efficiency between the heavy chain of MHC I and p2m targeted by Cy3-W6/32 and Cy5-L368 mAbs, respectively, was also determined (dotted line).
The IS may fulfill numerous functions with varying importance for a particular cell-cell interaction [67,68]. For example, it may enhance and/or prolong signaling, integrate different signaling pathways, direct granule release and cytokine secretion, terminate signaling processes, and balance enhancing and terminating signals.
Whereas antigen-induced redistribution of the relevant molecules on T cells has been extensively characterized (for reviews, see [63,67-69,72]), much less is known about the behavior of MHC and adhesion molecules or the mechanisms controlling their accumulation on target cells or APCs. Therefore, in the following section focus is centered on data regarding the distribution of MHC as well as intracellular adhesion molecules (ICAM-1) on APCs and target cells.
By using FRET and lateral diffusion (single particle tracking) experiments, clustering (self-association) of MHC I glycoproteins was observed at the surface of various human cell types [77-80] (see Fig. 7.1). A recent detergent-solubility analysis of their self-association properties also confirmed that homotypic association is an inherent property of MHC I (and MHC II) molecules , in accordance with earlier observations of their spontaneous clustering after reconstitution into liposome model systems . Electron and scanning force microscopic experiments also disclosed the nonrandom (clusterized) organization of MHC I molecules at a higher hierarchical level: immunogold-labeled MHC I molecules were observed to form domains of several hundred nanometers diameter . The degree of MHC I oligomerization showed good correlation with the expression of free MHC I heavy chains [lacking b2-microglobulin (b2m); "FHC"]: both of these were significant on activated or transformed/tumor cells . The two forms of heavy chains participated in common small- and large-scale clusters at the surface of human B cells, as revealed by FRET and SNOM experiments . Culturing cells with b2m resulted in a decreased homotypic association of intact MHC I heterodimers and their reduced co-clustering with free heavy chains . According to these data, FHCs likely have an important contribution to MHC I clustering: their involvement seems to stabilize MHC I clusters and vice versa, their functionally active conformation, which is still capable of rebinding b2m, may also be stabilized by participation in these clusters [83,84]. Otherwise, FHCs would undergo irreversible de-naturation and become functionally inactive and/or would be either internalized or released in a soluble form from the cell surface [85,86].
Clustered cell-surface distribution and anomalous diffusion of MHC II glyco-proteins as well as their heteroassociation with MHC I were also reported in numerous cell types, including APCs [77,87-89]. Atomic force and electron microscopic data showed that MHC II molecules form homoclusters not only on the nanometer scale attainable by FRET, but also at a higher hierarchical level, in the micrometer distance range . Electron microscopy revealed that a fraction of MHC II molecules was heteroclustered with MHC I at the same hierarchical level . Molecular associations were detected between the ICAM-1 adhesion molecules and MHC glycoproteins at the surface of human T and B lymphoma cell lines [77,90]. In addition, a high degree of ICAM-1 self-association was found on HUT102 B2 human T lymphoma cells . The above-mentioned association motifs of ICAM-1 and MHC glycoproteins were also observed on uveal melanoma and colon carcinoma cells. Interferon (IFN)-g changed the expression levels of MHC and ICAM-1 molecules as well as inducing the re-arrangement of their spatial distribution/association patterns on these cells [78,91].
Since clustering of MHC and ICAM-1 molecules could be observed in the IS [64,74], it is reasonable to assume that in vivo formation of the aforesaid association patterns of MHC I, MHC II and ICAM-1 proteins may promote IS formation, and their high local concentration can significantly increase the avidity of APC-T cell interaction [63,92]. Indeed, diminishing MHC I oligomerization on target cells by b2m treatment considerably reduced the efficiency of activation and effector function of allospecific cytotoxic T lymphocytes . This hypothesis is also supported by studies with soluble MHC:peptide multimers (dimers, trimers or tetra-mers) showing that aggregation of MHC molecules may significantly increase the efficiency of activation/immune response of T cells [93-95].
Although to a different extent, constitutive or inducible association of MHC glycoproteins and ICAM-1 molecules with lipid rafts could be observed in various cell types [39,78,90,96-101]. Disruption of lipid raft integrity with filipin or methyl-b-cyclodextrin caused dispersion of large-scale MHC clusters [39,97] and translocation of MHC to the soluble membrane fractions [90,96]. Dissociation of small-scale MHC II clusters was also detected upon cholesterol depletion . These data
Fig. 7.2 Triple co-localization of lipid rafts, ICAM-1 and MHC I on OCM-3 human uveal melanoma cells, as detected by CLSM. Lipid rafts were labeled by Alexa Fluor 488-conju-gated cholera toxin B subunit (panel A). ICAM-1 (panel B) and MHC I (panel C) molecules were targeted by Alexa 546-MEM111 and Cy5-W6/32 mAbs, respectively. Membrane areas where two or three membrane species co-localize are indicated with mixed colors in the overlay image (panel D). The pairwise cross-correlation coefficients between the fluorescence distributions indicate a high level of co-localization between the observed markers: Cab~0.7, Cac~0.7, Cbc = 0.65. The applied colors are pseudocolors. (Scale bar= 2 mm.)
indicate that lipid raft association is one of the underlying mechanisms responsible for MHC clustering.
Co-localization of MHC I and ICAM-1 could be observed within lipid rafts of colon carcinoma and uveal melanoma cells [78,91] (Fig. 7.2). On human B lympho-blastoid cells, physical association of MHC I and ICAM-1 could be detected both in detergent-insoluble and in detergent-soluble membranes . The disruption of raft integrity resulted in a significant loss of MHC I and ICAM-1 from the raft fraction, but their association was still detectable, implying that this interaction does not critically depend on the structure of rafts.
The accommodation of MHC and ICAM-1 in lipid rafts has important functional consequences in the process of antigen presentation. Lipid raft-assisted compartmentation of MHC II was shown to enhance the efficiency of antigen presentation to CD4+ T cells [96,97,102]. This effect was more prominent at low antigen doses, suggesting that rafting MHC domains are critical for T cell activation by rare MHC II-peptide complexes and less important when the antigen density is high [96,97,102,103]. It was also demonstrated that APC lipid rafts, raft-associated relevant MHC II-peptide complexes and even immunologically irrelevant MHC II molecules accumulate at the IS . Upon maturation of the IS, relevant MHC II-peptide complexes were sorted to the central region of the interface, while irrelevant MHC II molecules were excluded from this site . Similar to T cells, remodeling of the APC surface after the initial TCR signal seems to be also cytoskeleton-dependent .
Co-clustering of ICAM-1 with MHC I in lipid rafts of B lymphoblastoid cells was also shown to facilitate efficient presentation of viral peptides to CTLs . Raft accommodation of MHC I-ICAM-1 assemblies enabled specific recruitment of Src kinases harbored in lipid rafts to these complexes. Since activity of Src kinases along with preserved integrity of rafts was critical for the CTL response, it can be suggested that engagement of raft-included MHC I and ICAM-1 initiates intracellular signaling, leading to the concomitant migration of rafts and MHC I-ICAM-1 assemblies to the area of the initial target cell-CTL contact . This mechanism could provide the linkage between antigen recognition and early immunological synapse formation [90,92].
Was this article helpful?
Complete Guide to Preventing Skin Cancer. We all know enough to fear the name, just as we do the words tumor and malignant. But apart from that, most of us know very little at all about cancer, especially skin cancer in itself. If I were to ask you to tell me about skin cancer right now, what would you say? Apart from the fact that its a cancer on the skin, that is.