|Home | About | Journals | Submit | Contact Us | Français|
TLRs are a family of pattern recognition receptors that recognize conserved molecular structures/products from a wide variety of microbes. Following recognition of ligands, TLRs recruit signaling adapters to initiate a pro-inflammatory signaling cascade culminating in the activation of several transcription factor families. Additionally, TLR signals lead to activation of PI3K, affecting many aspects of the cellular response, including cell survival, proliferation and regulation of the pro-inflammatory response. The recent discovery of BCAP as a TLR signaling adaptor, crucial for linking TLRs to PI3K activation, allows new questions of the importance of PI3K activation downstream of TLRs. Here, we summarize the current understanding of signaling pathways activated by TLRs and provide our perspective on TLR mediated activation of PI3K and its impact on regulating cellular processes.
Toll-like receptors (TLRs) are a major family of pattern recognition receptors that play a critical role in innate host defense as well as in initiation of adaptive immune responses. Since the discovery of the first human TLR,11 there have been significant advances in the field of TLR biology, both in identification of the receptors, their diverse ligands and the signaling pathways that lead to responses necessary for protection of the host. There are currently 10 described TLRs in humans and 12 in mice, ligands for most of which have been identified. This family of TLRs can be broadly divided into two classes, depending upon the cellular location where they engage their respective ligands. TLR1, TLR2, TLR4, TLR5 and TLR6 engage their respective ligands from the cell surface, while TLR3, TLR7, TLR8 and TLR9 bind their cognate ligands in endosomes. Initiation of TLR signaling occurs when the Leu-rich repeat ectodomains of the TLR engage and complex with their respective ligand. A subset of the human TLRs have crystallographically elucidated complexes of their Leu-rich repeat (LRR) ectodomains and their respective ligands, showing the remarkable plasticity of the LRR recognition platform for unrelated pathogen molecules.12,13 For a more detailed review of TLRs and their ligands as well as the importance of their cellular signaling location, please refer to other reviews.14-16
Ligand binding by TLRs initiates recruitment of TLR signaling adaptors through homotypic interaction of the cytosolic Toll-Interleukin-1 receptor (TIR) homology domain found at the C terminus of all TLRs as well as the C terminus of the cytosolic adaptors. Until recently, there were only five known TLR signaling adaptors; MyD88, TRIF, TRAM, TIRAP and SARM. All TLRs, except TLR3, utilize MyD88 for signal transduction upon ligand binding, whereas TLR3 signals through the adaptor TRIF. TLR4 uniquely utilizes both a MyD88-dependent and MyD88-independent pathway, whereby TRIF is the critical signaling adaptor for the MyD88-independent node. TIRAP and TRAM serve as shuttling adaptors for TLR2 and TLR4, with TIRAP mediating recruitment of MyD88 and TRAM mediating recruitment of TRIF. SARM is the fifth member of the TLR signaling adaptor family. SARM was initially implicated in suppressing TRIF-dependent signaling through TLR3 and TLR4 in human cells.17,18 Studies using SARM-deficient mice demonstrated SARM is mainly expressed in the brain, and SARM-deficient macrophages were functionally normal when stimulated with TLR ligands.17,19 Therefore, it is proposed that it could play different roles between the two species.17 Indeed, SARM-deficient mice were found to be highly susceptible to infection with West Nile virus and displayed enhanced viral replication in the brain, decreased activation of microglia and, ultimately, increased mortality.20 Recently, we discovered BCAP (B-cell adaptor for PI3K, also known as Pik3ap1) as a sixth TLR signaling adaptor.21 The role of BCAP as a TLR signaling adaptor as well as the differential usage of signaling adaptors by TLRs, is discussed in more detail below.
Upon binding to their respective ligands, TLRs initiate a signaling cascade resulting in activation of the responding cell. The outcome of TLR activation depends on both the TLR that is activated as well as the type of responding cell. For instance, in macrophages and neutrophils, TLR activation enhances phagocytosis and increases the oxidative burst that facilitates rapid uptake and killing of microbes. In addition, resident macrophages responding to TLR ligands secrete chemokines that recruit additional neutrophils and monocytes to the site of infection. Activation of TLRs in dendritic cells (DCs) leads to their migration to the draining lymph nodes, allowing priming of pathogen-specific T cells. Additionally, TLR stimulation of DCs causes their maturation, including increased expression of co-stimulatory molecules important for activation of naïve T cells. Further, TLR ligands induce macrophages and DCs to secrete pro-inflammatory cytokines important for activation and differentiation of naïve T cells (an event critical for development of adaptive immunity), and regulating the host acute phase response, thereby further mediating the immediate clearance of pathogens, as well as contributing to the repair response by providing critical tissue remodeling and regeneration factors.22 B cells also respond to TLR ligands, and this response includes increasing expression of co-stimulatory molecules, inducing proliferation, class switch recombination machinery and differentiating into antibody-producing cells.23,24 Thus, the impact of TLRs induce multiple cellular outcomes, which vary depending upon the cell type responding to the stimuli.
Although TLR activation is important for host defense, an exaggerated innate immune response with very high circulating levels of pro-inflammatory cytokines can lead to septic shock and death of the host. In addition, chronic activation of TLRs can also lead to development of autoimmune diseases, such as systemic lupus erythematosus and inflammatory bowel disease, in genetically pre-disposed individuals. Consequently, it is important for responding cells to regulate the TLR signaling pathway to contain inflammation. Several negative regulators of TLR signaling, including IRAK3 (also known as IRAKM), SIGIRR and A20, to name just a few, have been previously identified and described.25-28 In addition, many studies implicate phosphoinositide 3-kinases (PI3K) and its substrate AKT in regulation of TLR signaling.5-7,29,30 The PI3K pathway integrally regulates important cellular processes, including cell survival and proliferation. Although we have a thorough understanding of the signaling pathways and transcription factors involved in induction of a pro-inflammatory response, the molecular players and events involved in TLR mediated PI3K are, by comparison, poorly understood. Our perspective on recent advances pertaining to TLR signaling leading to PI3K activation and its effects on the cell and the host response is discussed in further detail below.
TLR signaling leads to multiple outcomes (cell differentiation, induction of inflammatory/regulatory genes, cell-surface expression of co-stimulatory molecules, cellular proliferation, antibody class-switching production, etc.) dependent upon the cell type responding to the stimuli. These outcomes are a direct result of activation of several transcription factors including nuclear factor kappa light-chain enhancer of activated B cells (NFκB), activator protein-1 (AP-1) and interferon regulatory factors (IRFs). All TLRs activate NFκB and AP-1, while activation of IRF3 and IRF7 is regulated by differential usage of the signaling adapters as well as the cellular compartment where the signaling originates. The role of these different transcription factors in driving the pro-inflammatory response is discussed below.
Activation of NFκB is a major outcome of TLR signaling and contributes to most of the pro-inflammatory response induced by TLR ligands. Activation of NFκB is a multi-step process beginning with recruitment of the proximal TLR adaptors MyD88 and/or TRIF to the TLR membrane complex. Based on the initial choice of adaptor proteins, the TLR signaling network is broadly classified as the MyD88-dependent pathway or the TRIF-dependent pathway, though both lead to activation of NFκB.14 The MyD88-dependent pathway is engaged by all known TLRs, with the exception of TLR3, while the TRIF-dependent pathway is used by both TLR3 and TLR4. Uniquely, TLR4 activates both the MyD88- and TRIF-dependent signaling pathways.14 The MyD88-dependent pathway is important for early activation of NFκB, and the TRIF-dependent pathway is involved in late phase activation of NFκB in response to TLR4 stimulation.31 Activation of both of these pathways is important for induction of pro-inflammatory cytokine secretion downstream of TLR4.14
Following recruitment of MyD88 to TLRs through their TIR domains, MyD88 via its death domain engages a complex of IRAK4/IRAK1/IRAK2 through interaction with death domains found in these adaptors. A structural understanding has emerged about the assembly of this critical complex, termed the “Myddosome.”32 The IRAK-containing complex then engages and activates TRAF6, an E3 ubiquitin ligase. TRAF6, along with the E2 ubiquitin ligases UBC13 (also known as UBE2N) and UEV1A (also known as UBE2V1), catalyzes the formation of Lys 63-linked polyubiquitin chains upon itself as well as free ubiquitin chains.33 Synthesis of Lys 63-linked polyubiquitin chains by TRAF6 mediates recruitment of TAB2 and TAB3, which regulate the kinase TAK1, thus permitting TAK1 activity.33 TAK1 phosphorylates IKKβ, which is part of a complex with IKKα and NEMO. The IKK complex then phosphorylates IκBα (which enforces sequestration of NFκB inside the cytosol). Phosphorylated IκBα is targeted for ubiquitin-mediated proteasome degradation, freeing NFκB to translocate to the nucleus and act upon target genes.34
The TRIF-dependent pathway also leads to NFκB activation. This pathway includes a complex of TRADD and RIP-1, which mediates cleavage of the zymogens capsase-8 and caspase-10 into their active forms, which, in turn, activate NFκB.14,35-38 TRIF also leads to NFκB activation through interaction with TRAF6. Activation of both of these pathways is important for induction of pro-inflammatory cytokine secretion downstream of TLR4.14 NFκB activation driven by the TLR3 mediated TRIF-dependent pathway is not very robust at driving inflammatory cytokine secretion and instead leads to activation of IRF3 and a subsequent interferon response, which is discussed in greater detail below.
Signaling through TLRs also leads to activation of the AP-1 family of transcription factors, which are activated by mitogen-activated protein (MAP) kinases. MAP kinases are composed of a sequential three-component system involving activation of a MAP 3-kinase, which phosphorylates a MAP 2-kinase, which, in turn, phosphorylates a MAP kinase.39-42 The mammalian MAP kinase family consists of four groups: extracellular signal regulated protein kinases (ERK1/2), c-JUN NH2 terminal kinases (JNK1/2/3), p38 kinases (p38α/β/γ/δ) and ERK5.39,41 Both the MyD88-dependent and -independent pathways activate MAP kinases and AP-1, and the two pathways converge upon TRAF6. TAK1, a MAP 3-kinase, is activated by TRAF6 and transduces signals leading to the activation of JNK and p38.42-45 Activation of ERK1/2 proceeds through the activity of TPL2 (also known as MAP3K8 or COT), which is regulated through its interaction with NFκB p105.39,42 Upon TLR activation, TPL2 is released from p105 and activates MEK1 and MEK2, which then activate ERK1/2. However, ERK1/2 activation can proceed independent of TPL2 in some cell types, suggesting the existence of an alternate pathway.42,46 Activation of MAP kinases then leads to activation of the AP-1 family of transcription factors, which further feeds into regulation of the pro-inflammatory response, including inducing inflammatory cytokines and regulating susceptibility to endotoxin shock.39-43
Another consequence of TLR activation is the production of type I interferons, which are critical for induction of anti-viral immunity. The induction of type I interferons depends upon activation of the IRF family of transcription factors.47 This family includes nine members in both mice and humans, the most important of which are IRF3, IRF5 and IRF7.47 Upon activation, IRFs form homo- or hetero-dimers and translocate to the nucleus for binding to target DNA sequences. Activation of IRF transcription factors through TLR stimulation is accomplished through both the MyD88-dependent and TRIF-dependent signaling pathways upon ligand binding.48 The MyD88-dependent pathway leads to IRF7 activation upon stimulation of TLR7, TLR8 or TLR9.49 Ligand binding leads to an interaction of IRF7 with MyD88 through its death domain.50 This pathway further depends upon IRAK1 and IKKα, as cells lacking expression of these proteins are defective for IRF7 activation and subsequent IFN production.50-52 Likewise, stimulation of TLR7, TLR8 or TLR9 leads to phosphorylation of IRF5; however, the biochemical mechanism mediating IRF5 activation is incompletely understood.
Stimulation of TLR3 or TLR4, through the TRIF-dependent pathway, induces phosphorylation of IRF3 leading to IFNβ induction. TRIF-mediated activation of IRF3 depends upon upstream activation of TRAF3. Upon TLR4 stimulation, TRAF3 is modified by Lys 63-linked ubiquitination, which is required for subsequent activation of TANK-binding kinase 1 (TBK1) and IKKε.53,54 Together, TBK1 and IKKε then mediate phosphorylation of IRF3.55,56 Consequently, cells deficient in TRAF3, TBK1 or IKKε are severely compromised in their ability to activate IRF3 and induce IFNβ upon TLR3 stimulation.53,55-58
Phosphoinositide 3-kinases (PI3K) are a family of lipid kinases consisting of three classes (class I, class II and class III) based upon substrate specificity as well as several structural characteristics.3,4,9,10,59 Class I PI3K are further sub-divided into class IA and class IB depending upon which regulatory subunit they engage.59 The class IA PI3K, which plays a major role in cells of the immune system, consists of p110α, p110β and p110δ, and are all regulated by either p85α or p85β.8 The class IB PI3K, p110γ, instead binds the regulatory subunits p101 or p87. A detailed description of PI3K classes and their roles in regulating cellular responses can be found elsewhere.3,4,9,10
One mechanism mediating class I PI3K activation occurs through recruitment of the PI3K complex through the SH2 domain of the regulatory p85 subunit to phosphorylated tyrosine residues.3 Engagement of p85 to phosphorylated tyrosines releases the inhibitory pressure p85 places upon the p100 PI3K subunit. Further, recruitment of the PI3K complex to phosphorylated tyrosines brings PI3K to the lipid membrane and in close proximity with its lipid substrate. Upon activation, PI3K phosphorylates phosphatidylinositol-4,5-bisphosphate [PtdIns(4,5)P2] to yield phosphatidylinositol-4,5-trisphosphate [PtdIns(3,4,5)P3].3,4,9,10 Increased concentrations of PtdIns(3,4,5)P3 at the membrane leads to recruitment of PDPK1 (3-phosphoinositide dependent protein kinase-1) and AKT (also known as PKB). PDPK1 phosphorylates the serine/threonine kinsase AKT, which consists of three isoforms derived from three separate genes.3,4,9,10 The kinase AKT then acts upon its extensive complement of targets.
The precise role the PI3K pathway plays in regulation of the cell or host response to pathogenic insult is not completely clear. Studies using pharmacological inhibitors of PI3K activation have yielded controversial results. Some studies suggested PI3K contributes to NFκB activation and, thus, the inflammatory response, while other studies advocated a role for PI3K in inhibiting the inflammatory response.5-7 These discrepancies may be incompletely explained by usage of different pharmacological inhibitors as well as non-standard dosages.
Conversely, genetic evidence suggests the PI3K pathway acts as a critical negative regulator of the pro-inflammatory response. Dendritic cells from p85α-deficient mice produced more Il12 upon TLR stimulation, correlating with greater p38 activity in vitro, leading to increased resistance to Leishmania major infection with an enhanced Th1 effector phenotype in vivo.30 Intriguingly, thioglycollate macrophages have increased ERK1/2 and JNK activity, but not p38 activity, and increased message levels of Tnf and Il6, suggesting a cell type-specific role for p85 in regulating p38 MAP kinase activation.60
Pdpk1, which functions through PI3K to coordinate activation of multiple downstream effectors, including a critical role in Akt phosphorylation, suppressed production of Il6 and Tnf from macrophages upon stimulation through TLR2 or TLR4.61 Pdpk1-deficient macrophages had enhanced levels of phosphorylated IκBα, but in contrast to p85α-deficient DCs, had normal levels of MAP kinase activity.61 As a result, mice with myeloid-specific deletion of Pdpk1 in vivo contain higher levels of serum Tnf upon systemic administration of LPS. Consequently, Pdpk1-deficient mice exhibited increased end organ pathology and succumbed more rapidly to systemic LPS challenge.61
Furthermore, macrophages from Akt1-deficient mice had increased inflammatory responses to TLR4 stimulation, including enhanced production of Tnf, Il6 and Ccl3 (also known as MIP1α).29 Mechanistically, the authors found that signals provided by Akt1 were critical for induction of several miRNA species, including miRNA-155. Importantly, miRNA-155 was found to directly target the 3′ UTR of an important negative regulator of PI3K, SHIP1 (Src homology-2-domain containing inositol 5-phosphatase 1), blocking its expression.29,62 Importantly, detection of phosphorylated Akt in macrophages stimulated through TLR4 is largely abrogated in Akt1-deficient cells but not Akt2-deficient cells, suggesting that Akt1 is the most critical isoform utilized by the TLR-PI3K signaling axis.63
Phosphatase and tensin homolog (PTEN), a 3-phosphatase counterbalancing PI3K by catalyzing conversion of PtdIns(3,4,5)P3 back into PtdIns(4,5)P2, plays a positive role in inducing inflammation in TLR-stimulated cells. PTEN-deficient macrophages secreted decreased quantities of Tnf and Il6 upon stimulation with TLR ligands.60,64,65 Consequently, in a pneumococcal pneumonia model, mice deficient for PTEN in the myeloid compartment have less Tnf, Il6 and Cxcl1 (also known as KC), but more Il10 in their bronchiolar lavage fluid.66 This phenotype corresponded with increased phagocytosis and elimination of intracellular bacteria by alveolar macrophages infected in vitro with Streptococcus pneumonia.66 In contrast, Kuroda, et al. found a requisite role for PTEN in elimination of Leishmania parasites in vivo in a manner suggested to be dependent upon Tnf-mediated induction of nitric oxide.65 This discrepancy details the instructive role of inflammatory cytokines in an autocrine/paracrine fashion for eliciting the effector capacity of local immune cells.
Another important regulatory step imposed upon the class I PI3K pathway is through lipid hydrolysis of PtdIns(3,4,5)P3 by the phosphatase SHIP1, which acts as a 5′ phosphatase, mediating PtdIns(3,4)P2 conversion from PtdIns(3,4,5)P3. The role SHIP1 plays in TLR-stimulated cells is currently unclear, as conflicting results exist in the literature. In some reports, macrophages from SHIP1-deficient mice secreted less Tnf and Il6 and had a corresponding increase in PI3K activity through Akt phosphorylation.67,68 Paradoxically, in a series of reports from another group, SHIP1 blocked Tnf and Il6 secretion upon TLR3 or TLR4 stimulation, but induced Il12 and expression of surface co-stimulatory molecules upon TLR stimulation, suggesting a more complex role for SHIP1 in downstream TLR signaling.69-72 Consequently, SHIP1-deficient DCs were functionally incompetent for priming Th1 effector T cells.69,70 These conflicting reports warrant further and more detailed investigation into the molecular mechanism of SHIP1-mediated control of the TLR pathway. Indeed, the ability of SHIP1 to suppress inflammation was found to be independent of its phosphatase activity.73 Further work remains to fully understand the role SHIP1 plays in TLR signaling in a manner dependent and independent of its role in suppressing PI3K activity.
The mechanism for TLR-mediated activation of the PI3K pathway has been explored and a direct linkage of TLRs themselves, through phosphorylated tyrosine residues, is suggested to mediate activation of PI3K. Phosphorylation of tyrosine residues at the C terminus of TLR2 was required for recruitment of PI3K-p85 subunit and subsequent activation by RAC1.74 Similarly, TLR3 and TLR8 had phosphorylated tyrosine residues upon ligand engagement, which led to the recruitment of PI3K-p85 subunit.75,76 However, these studies were performed in the context of overexpression and the detailed biochemistry as well as genetic confirmatory studies are yet to be performed. In addition to recruitment to TLRs themselves, roles for adaptor-mediated recruitment of PI3K to the TLR signaling complex have also been described. MyD88 itself contains a YXXM motif within its TIR domain and was found to associate with p85. However, truncation and mutagenesis studies demonstrated that neither the YXXM motif of MyD88, nor its TIR domain, was required for permissive association with p85, suggesting the existence of other mechanisms promoting complex assembly.77 The adaptor TIRAP was also found to mediate Akt phosphorylation post-stimulation with the TLR2 ligand MALP2.78 Receptor-interacting protein 1 (RIP1)-deficient splenocytes were impaired in their ability to phosphorylate Akt upon LPS or CpG stimulation compared with control cells.79 MyD88-deficient macrophages were also found to be defective for Akt phosphorylation through several TLR ligands, as reported by us and others.21,63,77 Similarly, TRIF signals were required by macrophages for Akt phosphorylation upon LPS stimulation and likely serves as the upstream adaptor allowing RIP-mediated PI3K activation.63 Together, these data suggest that TLR adaptors may serve to recruit a previously unknown downstream target mediating PI3K activation.
We explored the possibility that other TIR-domain containing adapters could also be servicing TLR/IL-1R family members. Since the discovery of SARM as the fifth TIR adaptor in 2003, the current family of human TIR adapters,17,80,81 comprised by MyD88, TRIF, TRAM, TIRAP and SARM, has remained rather stable. The TIR domain superfamily is represented in every branch of the tree of life and has remarkably ancient origins in bacteria.82,83 A primitive TIR module structures from bacteria84 shows that the TIR domain fold is quite well conserved in spite of great sequence divergence; furthermore, TIR domain virulence factors from pathogenic bacteria are capable of binding host (MyD88 and TRIF) TIR domains, suggesting that their TIR interaction function remains intact.85,86 In addition, more thorough sequence searches have also revealed the presence of distant TIR-like domains in the intracellular segments of IL17 receptors (called STIR or SEFIR domains87,88) that interact with an IL17R-dedicated adaptor, ACT1, that also possesses a STIR/SEFIR domain.89
The plasticity of the TIR domain fold to sequence variations (either in the mammalian, plant or bacterial guises) suggested that there could be novel TIR domains cryptically embedded in other human proteins outside of the five known adapters for TLR and IL1R pathways that were not readily identifiable from routine sequence searches. To this end, we employed more sensitive methods utilized in structure prediction, fold recognition and modeling to canvass the human proteome for unrecognized TIR adaptors, and discovered two potential TIR candidates in the form of two paralogous chains, human BCAP (B-cell adaptor for PI3K, also known as PIK3AP1) and BANK1 (B-cell scaffold protein with ankyrin repeats 1) (Fig. 1), that are orthologs of the Drosophila DOF protein (also known as stumps).21 These novel TIR domains were located at the N terminus of BCAP/BANK1, followed closely by a transcription factor-Ig (TIG) domain that is common to many DNA-binding proteins including NFκB, p53 and STATs (Fig. 1).90 There is a short, conserved helical module after the TIG, and two additional helical domains in the C terminal half of BCAP/BANK that have a repeat-like nature (Fig. 1).21 Protein engineering of BCAP produced truncated versions that consisted only of the fused TIR/TIG/helical domains, and this engineered protein retained binding function for MyD88 and TIRAP adapters.21
Previous work implicated BCAP in positively regulating B cells signaling through their receptor and CD19 via activation of PI3K. The ability of BCAP to mediate PI3K activation through the B-cell receptor or CD19 depended upon phosphorylation of critical tyrosine residues in YXXM motifs within BCAP.91-93 In addition to B cells, BCAP is expressed by macrophages and NK cells, and acts as a negative regulator of inflammation through TLRs and NK cell receptors respectively.21,94-96 Indeed, BCAP is critical in mediating PI3K/Akt activation upon stimulation of cells with TLR ligands (Fig. 2).21,94,97 BCAP mechanistically connects TLRs to PI3K through a cryptic TIR domain at the N terminus of BCAP, which is utilized for recruitment to the TLR signaling apparatus (Fig. 1).21 Consequently, macrophages deficient for BCAP produced increased quantities of pro-inflammatory cytokines and BCAP-deficient mice were more susceptible to LPS toxicity assays as well as a model of inflammatory colitis.21,94,95 PI3K/Akt activation in macrophages also depended upon the TLR adaptors MyD88 and TIRAP, thus, it is possible that BCAP acts downstream of these two adaptors and is recruited to the TLR signaling adaptors and not the TLR itself, though this remains to be tested. In support of that idea, BCAP interacted with both MyD88 and TIRAP dependent upon the TIR domain of BCAP.21 Alternatively, the more classical TLR adaptors may mediate recruitment of an unknown intermediate factor required for permissive tyrosine phosphorylation of BCAP and, thus, mediating its downstream capacity for PI3K/Akt activation. Previous work showed that tyrosine phosphorylation of BCAP is critical for recruitment of PI3K as mutation of the tyrosine residues to phenylalanine abrogated the ability of PI3K to be activated upon stimulation of B cells through their receptor.91-93 Similarly, BCAP undergoes phosphorylation upon TLR stimulation, which mediated recruitment of PI3K to BCAP in a kinetic fashion upon TLR stimulation (our unpublished observations). Critically, as is the case for BCAP mediated PI3K activation through BCR stimulation, tyrosine phosphorylation of BCAP is required for optimal restriction of inflammatory cytokine production, suggesting the requirement of tyrosine phosphorylation of BCAP for TLR mediated PI3K activation.94 At present, the tyrosine kinase mediating BCAP phosphorylation in TLR-stimulated cells remains unknown. Notably, the SRC family kinase Lyn is implicated in phosphorylation of BCAP in B cells stimulated through their receptor or CD19 crosslinking.92 Further, Lyn also is implicated in mediating the linkage of TLRs to PI3K activation, with a consequential increase in Il6 and Tnf production by cells genetically deficient for Lyn.68 Thus, it is plausible to speculate that Lyn serves as the tyrosine kinase imparting permissive phosphorylation upon BCAP and subsequent induction of TLR mediated PI3K activation. This hypothesis is further supported by the overlapping phenotype of BCAP-deficient and Lyn-deficient macrophages stimulated with TLR agonists.21,68,94 However, this hypothesis remains to be experimentally tested.
Although PI3K is required to negatively regulate pro-inflammatory cytokine production by TLR stimulated myeloid cells, it is critical for survival and proliferation of B cells. The outcome of PI3K activation downstream of immune receptors depends on the cell type being activated. PI3K activation clearly dampens the inflammatory response in monocytes, macrophages and dendritic cells, but enhances the immune response in mast cells and potentially other granulocytes. Furthermore, PI3K activation is integral for promoting immune responses via its ability to promote survival and proliferation of B and T lymphocytes. In agreement with this, our unpublished observations demonstrate that BCAP-deficient B cells have a drastically reduced ability to proliferate in response to TLR ligands.
All the conflicting and controversial data point to an idea that perhaps PI3K is not a negative regulator of TLR signaling, nor a positive regulator of TLR signaling, but instead plays a critical role in orchestration and fine tuning of an ongoing immune response with pleiotropic roles as varied as the immune response itself. Indeed, the idea of PI3K as a negative or positive regulator is vastly over-simplified in the light of the varied roles it plays. Instead of settling the controversy, it is our opinion here that the most benefit to the field would come instead from increasing the understanding of the complicated role PI3K plays in a biological response through TLRs.
Recent studies have provided key insights into TLR mediated activation of PI3K. Although BCAP can now be considered as a critical TIR domain containing signaling adaptor serving as a molecular link for activation of PI3K upon TLR stimulation, many cellular and molecular outcomes of TLR mediated PI3K activation remain to be explored. Although BCAP acts as a negative regulator of TLR signaling due to its ability to activate PI3K, it may influence other aspects of a cell’s response to inflammatory stimuli. Selective genetic disruption of TLR mediated activation of PI3K, through BCAP deficiency in specific cell types via Cre-Lox technology, could give more insight into the particular role of BCAP and PI3K in regulating inflammation and other cellular processes. Furthermore, additional work detailing the individual PI3K molecules affected by BCAP deficiency could provide evidence into how these various PI3K components shape the TLR response.
We would like to thank members of the Pasare lab for their critical review of the manuscript and insightful discussions.
Previously published online: www.landesbioscience.com/journals/cc/article/21572