Meeting ReportCell Biology

Receptors, Signaling Networks, and Disease

See allHide authors and affiliations

Science Signaling  22 Feb 2011:
Vol. 4, Issue 161, pp. mr3
DOI: 10.1126/scisignal.2001687
A report on the INSINET symposium held in Madrid, Spain, 23 to 24 September 2010.

Abstract

Over the past years, a holistic approach has been applied to the study of the field of receptor signaling, permitting the analysis of how the interaction between receptors and their cellular environment determines receptor function and the study of the role of these receptors, under both normal and pathophysiological conditions, in whole organisms. This has been facilitated by the development of high-resolution microscopy techniques, which allow single-molecule or spatiotemporal resolution, or both, of signaling processes at the cellular and organismal levels. Concurrently, the role of these signaling pathways can be tested in increasingly sophisticated murine disease models. Finally, computational approaches aid in predicting and understanding receptor behavior. The program of the Madrid meeting reflected this integrated approach, highlighting signaling by and dynamics and regulation of immune cell receptors, the T cell receptor and B cell receptor, and signaling by and regulation of G protein–coupled receptors.

Nanoclusters, Conformational Spread, and Signaling in Immune Cells

The T cell receptor (TCR) is a multiprotein complex composed of a clonotypic αβ dimer recognizing peptide-bound major histocompatibility complex (MHCp) molecules and the CD3γ, δ, ε, and ζ subunits that are responsible for the activation of the downstream signaling pathways. It is now well established that upon recognition of activating MHCp complexes on antigen-presenting cells (APCs), multiple TCRs coalesce with key downstream signaling molecules, such as the adaptors LAT and Slp76, into actively signaling microclusters and finally gather into the immunological synapse (IS) (1) (Fig. 1). However, part of the TCRs on resting T cells (in absence of activating MHCp) is present as clusters of multiple TCRs that are preferentially activated upon stimulation with small amounts of antigen (2, 3). These clusters have been named TCR protein islands, or TCR nanoclusters, to distinguish them from the MHCp-induced and larger TCR microclusters. The speakers in the first session discussed whether and how these preexisting clusters may influence TCR activation properties.

Fig. 1

Nanoclusters and microclusters converge at the immune synapse. (A) TCR organization in resting and activated T cells. (I) Monovalent TCRs and TCR nanoclusters may be found at the surface of resting T cells. The TCR nanoclusters could be equivalent to or form membrane microdomains or “protein islands.” These preexisting islands would allow rapid signaling when a specific antigen is found. (II) Microclusters rapidly form upon specific TCR stimulation, for example, by interaction with an activating surface such as a lipid planar bilayer or an APC. (III) During the process of maturation of the immune synapse, the different microclusters move centripetally from the periphery (pSMAC) to form a central cluster, where the microclusters accumulate (cSMAC). (B) Molecules involved in the formation of the IS. PLCγ1, phospholipase C γ1.

CREDIT: Y. HAMMOND/SCIENCE SIGNALING

Dennis Bray (Cambridge University) introduced the concept of conformational spread, the mechanism through which allosteric proteins in a large complex can undergo conformational changes. This process is likely to occur both in intracellular protein “machines” and in proteins clustered in cellular membranes. He explained a thermodynamic model of bacterial flagellar motor function and showed that the inclusion of a free energy term due to cooperativity between proteins in the rotary motor can explain its switchlike response (4). Experimental work confirms that the switching of the flagellar motor closely follows the pattern predicted by conformational spread (5). Moreover, a similar mechanism is likely to occur in clusters of membrane receptors in bacteria and other cells, providing a coordinated response upon encounter with graded amounts of their ligands.

Balbino Alarcón (Centro de Biología Molecular Severo Ochoa, Madrid) described a mutant form of the CD3ε subunit that impairs the conformational change in CD3ε induced by MHCp-TCR interaction (6). The mutant exerts a dominant-negative effect on T cell activation, which may reflect the disruption of cooperative signaling between adjacent TCRs. In mice homozygous for this mutation, T cell differentiation is blocked at an early stage in which the β chain of the TCR associates with an invariant pre-Tα chain and forms together with the CD3 subunits the pre-TCR (7). Pre-TCR function is critical for the lineage decision between αβ and γδ T cells, rescues the differentiating T cells from cell death, and permits progression to the next developmental stage where the mature αβ chain will be expressed. The defect in T cell differentiation of mice with the mutant CD3ε supports the hypothesis that there is a critical role for cooperative signaling, a conformational change in CD3ε, or both in pre-TCR function. Interestingly, activation of the pre-TCR is most likely ligand independent and may signal by clustering (8).

The ability of the TCR to discriminate among ligands of different affinity plays a key role in the outcome of TCR engagement. However, different techniques for measuring the kinetics of the TCR-MHCp interaction have produced divergent results, as highlighted by Mark Davis (Stanford University), who elaborated on published work in which the binding kinetics between fluorescently tagged TCRs from intact T cells and MHCp ligands in a lipid bilayer was resolved through single-molecule fluorescence resonance energy transfer (FRET) measurements (9). These more “cellular” measurements contrasted remarkably with previous measurements made for soluble TCRs with the surface plasmon resonance (SPR) technique. There was a 5- to 10-fold reduction in the half-life of binding compared with that observed by SPR, which was due to a counteractive action of the actin cytoskeleton. Such counteractive action of actin was greater in the case of activation with weaker ligands, suggesting that this mechanism helps T cells to increase discriminatory power in addition to the pure affinity for different MHCp ligands. In addition, the on-rate of the TCR-MHCp interaction was increased by almost 100-fold compared with that observed by SPR of soluble proteins. Davis speculated that this reflected cooperative binding of TCRs with their ligands, which could be facilitated by preexisting TCR protein islands (2, 10).

This concept of preexisting TCR nanoclusters was supported by the quantitative model that calculates the distribution and number of TCRs bound by multimeric MHCp complexes described by Thomas Höfer (German Cancer Research Center, Heidelberg). The requirements for TCR nanocluster formation was the topic of the work presented by Wolfgang Schamel (MPI for Immunobiology and University of Freiburg) who showed, using purified monomeric TCRs and liposomes, that sphingomyelin and cholesterol within the lipid environment induce TCR nanoclusters (2). The interactions with cholesterol appear to be direct, because the TCR complexes could be pulled down using immobilized cholesterol as a bait.

The quantitative model described by Höfer assumes that T cell stimulation only occurs upon binding of dimeric or higher-order ligands and that TCR nanoclusters exist before T cell stimulation. The model fits well with experimental binding data and predicts that binding of multimeric MHCp ligands to TCR nanoclusters permits T cells to discriminate between ligands of different affinity, irrespective of ligand concentration (that is, even high concentrations of a low-affinity ligand cannot activate T cells). This inability of abundant low-affinity ligands to activate the TCR may be important to ensure self-tolerance, given that antigenic MHCp complexes at the surface of the APC are vastly outnumbered by self-MHCp complexes. However, the model of TCR protein islands could allow serial engagements and the recognition of rare antigens by the TCR.

In the context of self-tolerance acquisition, negative selection in the thymus, which eliminates potentially autoimmune T cells and is guided by the affinity of the TCR for the self-MHCp ligands, was highlighted by Ed Palmer (University of Basel). Palmer argued that the duration of  TCR-MHCp interaction allows developing T cells to discriminate between high- and low-affinity ligands (11). Mechanistically, the time of interaction determines to what extent the Src family kinase Lck is recruited to the TCR by either the CD8 or CD4 co-receptor for class I or class II MHC molecules, respectively. Lck governs the extent of phosphorylation of the CD3 immunoreceptor tyrosine-based activating motifs and recruitment of downstream signaling proteins. At least for CD8, the recruitment of Lck depends on an association of CD8 with elements of the TCRα chain.

Andrés Alcover (Institut Pasteur, Paris) also focused on molecular trafficking of proteins associated with T cell signaling, particularly the trafficking patterns of TCRζ, the kinase Lck, and the adaptor linker for activation of T cells (LAT) to the IS. Taking advantage of the observation that the HIV protein Nef can sequester Lck in an endosomal recycling compartment (12), Alcover’s lab observed that Lck controls recruitment of TCRζ and LAT to the IS upon T cell stimulation, and his work suggested that a hierarchical recruitment appears important for controlling signal amplification in the IS. Additionally, outside-in signaling through the TCR may be regulated from the intracellular side by differential delivery of new molecules through polarized sorting. Michael Dustin (New York University) also discussed the role of endocytosis at the IS as a mechanism to modulate the output of the TCR. His lab has proposed a model in which signaling is initiated and sustained by microclusters organized around a central supramolecular cluster, the cSMAC, where signaling is terminated. Short interfering RNA–mediated knockdown of the ubiquitin-binding protein Tsg101, which forms part of the ESCRT-1 (endosomal sorting complexes required for transport 1) machinery, eliminated TCR delivery to the cSMAC and resulted in increased total tyrosine phosphorylation (13). Surprisingly, the cSMAC was composed of extracellular microvesicles apparently released into the synaptic space. MHCp ligands with weaker stimulating activity trigger less down-regulation of the TCR (14) than ligands with greater stimulating activity. Dustin reported that these weaker ligands do not recruit Tsg101 and do not induce formation of the cSMAC, although they formed microclusters that depended on the availability of a ligand for CD28. This lack of cSMAC formation in response to weaker MHCp ligands, which likely allows persistent TCR signaling, may explain how weaker MHCp ligands are able to activate T cells. A role for engagement of CD2, which is a surface receptor for CD58 that co-stimulates the T cell in presence of pMHC, in retarding microcluster transport to the cSMAC was also demonstrated. These studies raised several questions about the relationship between microclusters and cSMAC microvesicles and rules governing Tsg101 recruitment.

Following up with the concept of the IS as a site of membrane dynamics, Francisco Sánchez-Madrid (Hospital La Princesa, Madrid) described the IS as a focal point for both endocytosis and exocytosis. He showed that small vesicles can be delivered by the T cell to the APC in an antigen-specific manner, suggesting that T cells can modulate the APC functionality. These vesicles may be related to those released from the cSMAC described by Dustin. The extensive membrane and cytoskeletal dynamics occurring at the IS require energy, and Sánchez-Madrid reported that T cell mitochondria polarize to the IS upon specific stimulation of the TCR, once the microtubule-organizing center (MTOC) is reorientated to the IS. Dynamin-related protein 1, a mitochondrial profission factor, is responsible for mitochondrial movement. Sánchez-Madrid showed that TCR clustering at the cSMAC depends on mitochondria positioning and activity at the pSMAC, mainly through the regulation of the contractile activity of the actomyosin ring beneath the pSMAC (Fig. 1). Therefore, these organelles fuel the IS architecture and cSMAC formation, thus favoring TCR endocytosis (15).

Cytoskeletal dynamics of immune cells was another highlighted topic, with Andrey Shaw (Washington University, St. Louis) describing signals that control recruitment of the MTOC to the IS of T cells and Facundo Batista (Cancer Research UK, UK) describing how the actin cytoskeleton regulated activation of the B cell receptor (BCR). Shaw showed that Ksr1 (kinase suppressor of Ras) recruits phosphorylated extracellular signal–regulated kinase (ERK) to the IS. Both the scaffold function of Ksr1 and ERK activation are important to allow Ksr1 ERK phosphorylation of stathmin and its recruitment to the IS. Stathmin-deficient T cells show less efficient recruitment of the MTOC to the IS upon TCR stimulation and less cytotoxic capacity relative to wild-type T cells, indicating a clear functional role for the microtubule cytoskeleton in T cell signaling.

Like T cells, B cells have a multisubunit receptor, the BCR, responsible for B cell activation by the binding of a specific antigen (ligand), which is endocytosed with the receptor and processed into peptides. Batista used dual view total internal reflection fluorescence microscopy to show that the actin cytoskeleton restricts the diffusion rate of the BCR (16). Disruption of this actin network is sufficient to initiate signaling, likely mediated through the BCR, in the absence of specific antigen. This suggests that the actin cytoskeleton restrains BCR signaling and serves as a break on the low-level antigen-independent signaling important for cell survival (17). Mechanistically, actin might prevent BCRs from interacting with activating proteins necessary for signaling. In line with this type of signal-limiting mechanism, a similar phenomenon has been observed in T cells in which previously segregated nanoclusters containing different signaling proteins coalesce upon T cell activation (3). Alternatively, it could be that actin stabilizes and prevents the release of BCR from inhibitory proteins or domains.

GPCR Signaling Networks

G protein–coupled receptor (GPCR) signaling affects most aspects of physiology, and the components of these networks represent clinically relevant drug targets. Not only are the GPCR downstream effector and signaling cascades diverse and complex, the mechanisms regulating GPCR activity and signaling specificity also contribute to the complexity of GPCR systems (Fig. 2). New insights into the functions of arrestins (GPCR scaffolding proteins) and the GPCR kinases (GRKs), both of which have roles in signaling specificity and receptor desensitization, were described. GPCR signaling is associated with many pathological conditions. Highlighted in this meeting were mechanisms of virally encoded GPCR signaling and their implications for virus-induced cancer, GPCR signaling in inflammation mediated by eicosanoids, and GPCR signaling and GPCR regulators in cardiovascular and metabolic diseases.

Fig. 2

Signaling by GPCRs and the GPCR interactome. New techniques and model systems have uncovered previously unknown roles for the proteins implicated in GPCR signaling networks, especially GRKs and arrestins. The GPCR signalosome and GPCR-mediated signal transduction exhibits complex spatiotemporal dynamics, including receptor endocytosis, receptor conformational changes, and receptor oligomerization. Some of the main ideas presented at the meeting are indicated.

CREDIT: Y. HAMMOND/SCIENCE SIGNALING

The increasing complexity of the cellular roles and expanding targets of the GPCR regulators GRKs and arrestins (scaffolding proteins) (18) were addressed by Jeff Benovic (Thomas Jefferson University, USA), Federico Mayor Jr. (Centro de Biología Molecular Severo Ochoa, Madrid), and Walter J. Koch (Jefferson Medical College, USA). Benovic described his group’s work on the functions of arrestins, showing that the abundance of arrestin can influence centrosomal function (19) and that a correlation exists between the abundance of specific isoforms of arrestins and survival of breast cancer patients. He also discussed his lab’s studies with the nematode Caenorhabditis elegans indicating that arrestins play a role in regulating longevity.

Mayor and Koch focused on the GRKs. Mayor showed data identifying GRKs as regulators of receptors that do not belong to the GPCR family, such signaling mediated by insulin and platelet-derived growth factor (PDGF), which activate receptor tyrosine kinases, and signaling mediated by transforming growth factor–β (TGF-β), which activates a receptor serine-threonine kinase. GRK2, acting as a negative regulator of insulin signaling, is involved in the control of insulin resistance and obesity in animal models (20), whereas reduced GRK2 abundance promotes a decrease in PDGF signals and an imbalance in TGF-β signaling in endothelial cells leading to impaired angiogenesis in vivo. Koch emphasized the roles of GRKs in cardiovascular disease. GRK2 is a key modulator of both chronic and acute ventricular dysfunction (21), and Koch introduced a previously unknown role for this protein in myocyte death from ischemia-reperfusion injury by inhibiting Akt-dependent phosphorylation of endothelial nitric oxide synthase, which reduced its activation. He also discussed GRK5-dependent regulation of heart hyper‐trophy (22), which involves phosphorylation of histone deacetylase and activation of transcription of the gene encoding the transcription factor myocyte enhancer factor 2, and showed data proving a role for endogenous GRK5 in mediating maladaptive heart hypertrophy.

Pathogen-encoded GPCRs can be used to exploit the host’s signaling systems to promote infection and, in some cases, to cause secondary pathologies, such as virus-induced cancer. Martine Smit (VU University Amsterdam, The Netherlands) talked about the GPCR US28, which is encoded by the human cytomegalovirus and binds chemokines such as CCL2, CCL5, and CX3CL1. When overexpressed in NIH-3T3 cells, US28 promotes angiogenesis and tumor formation in a Cox-2–dependent manner (23). US28-dependent proliferation and transformation in 3T3 cells occurs through interleukin-6 (IL-6) secretion and signaling by the transcriptional regulator STAT3 (24), and increased phosphorylated STAT3 correlates with poor prognosis in human glioblastoma patients, putting forward another role for viral GPCR overexpression in tumor progression in addition to its previously reported actions mediated, for example, by Cox-2.

Not only are the details of downstream signaling and regulatory events becoming better characterized for GPCRs, but also the application of microscopic techniques, as described by Martin Lohse (University of Würzburg, Germany), has provided insights into the conformations adopted by the receptors, receptor oligomerization, and spatial confinement of receptors and how each of these processes influence cellular responses. Lohse showed how FRET experiments have uncovered the kinetics of the different steps of GPCR signal activation and deactivation and how inserting fluorophores at different sites within the receptors has revealed that they can adopt different conformations with varying signaling consequences or that specific ligands induce specific conformations to produce ligand-specific effects (25). Fluorescence recovery after photobleaching can be used to study dimerization and oligomerization of GPCRs, as well as the stability of receptor oligomers. These techniques also serve to examine how the spatial confinement of receptors contributes to the signal transduction profile (26).

Eicosanoids, such as prostaglandins, thromboxanes, leukotrienes, and lipoxins, are lipid mediators that signal through GPCRs and regulate inflammatory responses and reproductive functions. Prostaglandins and thromboxanes are collectively identified as prostanoids. Several eicosanoids are potent proinflammatory mediators implicated in inflammatory diseases such as rheumatoid arthritis, inflammatory bowel disease, and lupus. However, some of them, such as resolvins and protectins, display potent anti-inflammatory actions. Cells of the immune system, such as leukocytes and T cells, are influenced by eicosanoid signaling. Antonio Recchiuti (Brigham and Women’s Hospital, USA) focused on the role of lipoxins, resolvins, and protectins in the resolution of inflammation (27) and described more specifically the details of resolvin D1 (RvD1) signaling pathway through the activation of ALX and GPR32 receptors on human phagocytes (28). This revealed that RvD1 enhances the phagocytic and clearance functions of human macrophages specifically through these receptors and, thus, agonists of ALX and GPR32 may prove useful in resolving acute inflammation. Shuh Narumiya (Kyoto University, Japan) presented data suggesting that prostanoids collaborate with cytokines in the regulation of T cell differentiation (29), describing evidence that differentiation of TH1 cells [a type of T helper cell that produces interferon γ, IL-2, and tumor necrosis factor (TNF) and evokes cell-mediated immunity and phagocyte-dependent inflammation] and expansion of TH17 (a subset of IL-17–producing T helper cells that are involved in the recruitment of neutrophils at the early stages of infection) are mediated by prostaglandin E2 (PGE2). PGE2 signals mainly through EP2 and EP4 receptors in T cells. These actions are exerted through different signaling modules after the activation of EP2 and EP4. The TH1-differentiating action of PGE2 is mediated by the activation of phosphoinositide 3-kinase, whereas the TH17-expanding action is mediated by the synthesis of cyclic AMP (cAMP).

Nonsteroidal anti-inflammatory drugs inhibit Cox-2, an enzyme responsible for the synthesis of prostanoids. Garret FitzGerald (University of Pennsylvania, USA) explained that the adverse cardiovascular effects of nonsteroidal anti-inflammatory drugs are mainly due to the suppression of Cox-2–dependent cardioprotective prostaglandins (30, 31). As an example, he described that inhibiting prostaglandin D2 (PGD2) signaling by deletion of its receptor DP1 in mice resulted in detrimental cardiovascular phenotypes. Manuel Fresno (Centro de Biología Molecular Severo Ochoa, Madrid) described the coordinated regulation of membrane-associated PGE synthase-1 (mPGES-1) and Cox-2 in macrophages (32) and how Cox-2 deficiency inhibits macrophage activation patterns and chemotactic properties. Cox-2 may also have a role in atherosclerosis. Although this is a subject of controversy, the work of Fresno’s laboratory suggests a protective role for macrophage Cox-2 in atherosclerosis.

Signaling Implicated in Metabolic Diseases

Insulin resistance is a common consequence of obesity and contributes to metabolic syndrome. Mouse models of obesity have been useful for understanding the molecular mechanisms by which insulin resistance occurs and in understanding the role of inflammation in metabolic syndromes. Pablo Garcia-Roves (Karolinska Institutet, Sweden) focused on the role of the gain-of-function R225Q mutation in AMP-activated protein kinase γ3 subunit (AMPKγ3). The overexpression of the R225Q-AMPKγ3 protein in transgenic mice increases mitochondrial biogenesis in glycolytic skeletal muscle (33) and confers protection against insulin resistance, induced by a high-fat diet or by aging, in skeletal muscle. Interestingly, similar phenotypes have been observed in human families displaying the homologous mutation (34).

Antonio Vidal-Puig (Cambridge University, UK) reviewed the possible mechanistic links between obesity and insulin resistance, emphasizing that lipid accumulation in nonadipocyte cells, once the adipose tissue expansion limit has been exceeded, leads to lipotoxicity (35, 36). Lipotoxicity may be a trigger for macrophage polarization toward an M1 (proinflammatory) phenotype (opposite to the M2 anti-inflammatory phenotype). He also described the role of infiltrating macrophages as contributors to inflammation in obesity. Several factors derived not only from adipocytes but also from infiltrated macrophages seem to contribute to the pathogenesis of insulin resistance.

New Approaches

Advances in drug delivery and therapies with dendrimers, which are globular nanostructures engineered to carry molecules, as well as advances in molecular modeling with molecular dynamics and application of intravital microscopy, were the topics of talks focused on “new technologies.”

Jean-Pierre Majoral (CNRS, Toulouse, France) described the synthesis of dendrimers. Their unique architectural design—which includes a high degree of branching, multivalency, and globular architecture—and their well-defined molecular weight clearly distinguish these structures as potential carriers in medical applications, such as drug delivery, gene transfection, and tumor therapy. The highly branched structures resulting from the chemical addition of groups produces homostructural layers between the focal points (branching points). The number of focal points when going from the core toward the dendrimer surface is the generation number of the dendrimer. Majoral discussed an example of phosphonate-capped dendrimers in promoting the ex vivo proliferation of natural killer cells by the specific inhibition of CD4+ T cells through a putative dendrimer receptor and the potential use for immunotherapy of cancer (37).

Marek Maly (J. E. Purkinje University, Czech Republic) explained a computational model that analyzes the possible interactions between dendrimers and oligonucleotides or proteins at the atomic level. The kinetics of the interactions between these molecules were studied using molecular dynamics in conditions mimicking physiological conditions (temperature, pressure, pH, and Na+ and Cl concentrations), allowing the determination of binding energy and structural properties of these interactions (38, 39). Current computational advances permit calculation of complex formation between two molecules, but calculations at atomic resolution of larger-scale assemblies are hampered by technological limitations. This limitation can be solved by grouping individual atoms into virtual particles and performing the calculations using these particles. Maly presented several applications of this model to optimize interactions between bioactive compounds and delivery vehicles, such as dendrimers used in therapies for Alzheimer’s disease or HIV infection.

The application of dendrimers as versatile nonviral vectors for gene therapy against HIV infection and inflammation was the topic of Ma Angeles Muñoz-Fernández (Hospital Gregorio Marañón, Madrid), who highlighted the use of a carbosilane dendrimer named 2G-NN16 to inhibit HIV replication in human peripheral blood mononuclear cells, macrophages, and dendritic cells or human astrocytes in culture (40, 41). Gene therapy with dendrimers is a potential mechanism to prevent HIV infection.

Intravital multiphoton microscopy may also reveal new treatment strategies for human disease. Peter Friedl (Nijmegen University, The Netherlands) presented the basic mechanisms of deep collective cancer cell invasion in mouse models monitored by intravital multiphoton microscopy (42) and the role of integrins in mediating invasion. Radioresistance, which is resistance to radiation therapy, of the invasion niche was also shown to be mediated by integrins. These findings implicate differential integrin signals in microenvironmental control of cancer resistance that might be amenable for combination therapy.

This meeting brought together a diverse group of scientists who gathered to discuss the molecular mechanisms of activation of two types of membrane receptors: antigen recognition receptors and GPCRs. Their specific spatial organization within the cellular membrane and the signal transduction pathways triggered by their activation were reviewed, with a special focus on their implications in diverse pathophysiological processes, mainly cardiovascular, inflammatory, and metabolic diseases. The meeting fulfilled its primary goal of bringing together researchers, working in distinct fields and using different scientific approaches, who could potentially establish a network of collaborations and mutually help each other in the search for new tools in biomedicine. It is evident that the new mathematical models and imaging technologies make possible single-molecule and single-cell analysis in physiological contexts.

References and Notes

  1. Funding: This meeting was financially supported by a project from the Community of Madrid.
View Abstract

Stay Connected to Science Signaling

Navigate This Article