Each week we showcase a hot new cell signaling article from a Nature Publishing Group journal. Free full text access to the paper will be maintained for three months, after which the research highlight will be accessible via the Updates page.
Sepsis-induced coagulation triggers PAR1–S1P3 signaling in dendritic cells, resulting in systemic inflammation.
Excessive activation of the innate immune system in response to sepsis produces systemic coagulation. The coagulation cascade generates thrombin, which stimulates the protease-activated receptor 1 (PAR1). While PAR1 is known to promote inflammation, the biological mechanism linking PAR1 activation to inflammation has not yet been described. In Nature, Niessen et al. report that sphingosine 1-phosphate (S1P) receptor 3 (S1P3) is a component of PAR1-mediated pro-inflammatory signaling, and that loss of PAR1 or S1P3 on dendritic cells (DCs) blocks the toxic inflammatory response to bacterial sepsis.
Mice treated with lipopolysaccharide (LPS) developed lethal sepsis characterized by increased thrombin-antithrombin (TAT) and pro-inflammatory cytokines including interleukin-1β (IL-1β). Genetic loss or pharmacologic inhibition of PAR1 resulted in an initial normal increase followed by a precipitous decrease in cytokines and coagulation activation markers. PAR1-/- mice were better able to survive LPS treatment than their wild-type counterparts. S1P signaling promotes inflammation, and sphingosine kinase-null (SphK1-/-) or S1P3-/- mice displayed the same attenuated cytokine levels seen in PAR1-knockout mice. Treatment with S1P3 agonist restored the amplified pro-inflammatory response in PAR1-/- mice, suggesting that S1P3 is downstream of PAR1 in a common pro-inflammatory pathway.
Pro-inflammatory signaling during sepsis was thought to originate from vascular cells. However, transfer of wild-type bone marrow into PAR1-/- mice reestablished the inflammatory response. DCs are derived from bone marrow and produce pro-inflammatory and pro-coagulant factors. Adoptive transfer of wild-type DCs into SphK1-/-, S1P3-/- or PAR1-/- mice reinstated the release of pro-inflammatory cytokines. Furthermore, transfer of wild-type, but not PAR1-/- DCs into S1P3-/- mice impaired survival, suggesting that the dendritic cell PAR1–S1P3 signaling axis is necessary for the late-stage systemic inflammatory response.
How does loss of PAR1 and S1P3 block systemic inflammation? S1P signaling was required for systemic IL-1β release from DCs, and IL-1β levels were elevated in the lungs of wild-type mice. In S1P3-/- and PAR1-/- mice, DCs were contained in lymph nodes, resulting in localized IL-1β release. Therefore, loss of PAR1 attenuates both S1P3 signaling and widespread IL-1β release, and prevents the systemic inflammatory response that leads to lung and organ damage. Furthermore, fibrin, the end product of coagulation, was detected in lymphatic ducts in wild-type mice, but was restricted to the lymph node parenchyma in PAR1-/- mice, suggesting that loss of PAR1 also inhibits systemic coagulation. Indeed, adoptive transfer of wild-type DCs was able to restore peripheral coagulation activation markers in PAR1- or S1P3-deficient mice following LPS challenge.
Together, these data indicate that PAR1–S1P3 signaling in DCs mediates the long-range dissemination of inflammation and coagulation in severe sepsis. Development of therapies that specifically block PAR1 or S1P3 signaling may be effective in treating the systemic inflammation, coagulation and multi-organ failure that can result from sepsis.
Emily J. Chenette Signaling Gateway
Original Reference:
Niessen, F. et al. Dendritic cell PAR1–S1P3 signalling couples coagulation and inflammation Nature452, 654-658 (2008) Full text | PDF | Subscribe to Nature
The small GTPases Rac1 and Rac2 regulate mDia2 to promote asymmetric cell division and enucleation in developing erythrocytes.
A critical step in the formation of mature mammalian erythrocytes from progenitor cells is an asymmetric cell division in which the nucleus is compacted, extruded and enveloped by the plasma membrane. The enucleated daughter cell, known as a reticulocyte, eventually matures into an erythrocyte. While other facets of erythropoeisis have been extensively characterized, the signaling pathways that regulate enucleation remain poorly understood. In Nature Cell Biology, Lodish and colleagues now show that the Rac1 and Rac2 GTPases signal through mDia2 to promote enucleation in mouse erythrocytes.
Erythropoeisis can be observed in vitro by harvesting erythroid progenitor cells from E13.5 embryonic mouse livers and culturing them with erythropoietin and fibronectin. Previous studies have indicated that actin filaments form a cortical actin ring (CAR) between the condensed nucleus and the incipient reticulocyte. The CAR then contracts and pinches the nucleus off to form a reticulocyte. Rho family GTPases are well-known regulators of the actin cytoskeleton and govern CAR formation during cytokinesis. To evaluate a role for Rho proteins in enucleation, erythroid progenitor cells were infected with dominant-negative Rac, Rho and Cdc42. Dominant-negative RhoA and Cdc42 had very little effect on enucleation, whereas pharmacologic inhibition of Rac, or expression of dominant-negative Rac1 or Rac2, blocked both formation of the CAR and enucleation. Rac-deficient cells displayed no other defects in cell proliferation or differentiation.
The RhoA effector mDia nucleates the CAR during cytokinesis. Interestingly, mDia2 and Rac GTPases were found to localize to the CAR during enucleation. Short hairpin RNA-mediated downregulation of mDia2 blocked enucleation and produced a phenotype similar to Rac-deficient cells. Furthermore, Rac1 and Rac2 interacted with mDia2 in a GTP-dependent manner in vitro, suggesting that mDia2 is involved in Rac-mediated enucleation. Indeed, expression of constitutively active mDia2 rescued the enucleation defect observed upon expression of dominant-negative Rac1 or Rac2. It remains unclear if mDia2 is a direct Rac effector, as a robust direct interaction could not be verified in vivo; nonetheless, Rac–mDia2 signaling is required for formation of the CAR and enucleation.
These data are consistent with the idea that small GTPases are required for the establishment of polarity and asymmetric cell division in vertebrates, and provide the first insights into how enucleation of erythroid progenitor cells is regulated at the molecular level. Future studies will be important to understand the signals that regulate Rac and promote enucleation in immature mammalian erythroblasts.
Emily J. Chenette Signaling Gateway
Original Reference:
Ji, P., Jayapal, S. R. & Lodish, H. F. Enucleation of cultured mouse fetal erythroblasts requires Rac GTPases and mDia2 Nature Cell Biology10, 314-321 (2008). Full text | PDF | Subscribe to Nature Cell Biology
Pyruvate kinase: Energizing tumor cell growth
A tumor-specific isoform of pyruvate kinase is regulated by binding to phosphotyrosine-containing peptides, which balances energy production and macromolecule anabolism to promote tumorigenesis.
Normal cells and cancer cells differ in their preferred glucose metabolism pathway; normal cells preferentially use oxidative phosphorylation, whereas cancer cells prefer glycolysis. This pro-oncogenic metabolism switch, known as the Warburg effect, has been linked to an increase in glycolytic pathway gene expression, including pyruvate kinase. Paradoxically, tyrosine kinase signaling — commonly upregulated in tumors — inhibits pyruvate kinase, although whether inhibition is via direct phosphorylation remains controversial. Two studies by Lewis Cantley and colleagues in Nature resolve this controversy by showing that the M2 splice variant (PKM2) is important for aerobic glycolysis in tumors, but is negatively regulated by binding to phosphotyrosine-containing peptides. This novel mode of binding-mediated regulation supports cancer cell proliferation by promoting macromolecule biosynthesis.
Cantley and colleagues analyzed the expression of pyruvate kinase isoforms and found that PKM2 predominated in tumors. PKM2 expression provided a growth advantage to tumors by facilitating the switch to glycolysis. In a separate study, phosphotyrosine-containing peptides robustly bound to PKM2, which displaced its allosteric activator fructose-1,6-bisphosphate (FBP). This suggests that the inhibitory effect of tyrosine kinase signaling is achieved via displacement of FBP rather than direct phosphorylation. Overexpression of constitutively active Src kinase, stimulation of insulin-like growth factor signaling or pervanadate treatment — which stabilizes phosphoproteins — also led to a 30% reduction of PKM2 activity and decreased the binding of PKM2 to FBP in lung cancer cells in vivo. Consistent with these observations, a phosphotyrosine-containing peptide caused a 20–30% reduction in PKM2 enzymatic activity in vivo. However, a phosphopeptide binding-deficient mutant of PKM2 (PKM2K433E) retained enzymatic activity following pervanadate treatment. Although the phosphotyrosine-containing proteins that bind PKM2 in vivo were not identified, the authors did show that phosphotyrosine peptides from enolase and lactate dehydrogenase were able to inhibit PKM2 in vitro.
Phosphopeptide-mediated regulation is important for cell proliferation, as re-expression of PKM2K433E in PKM2-deficient lung cancer cells did not overcome the proliferation defect that resulted from PKM2 depletion. It is possible that PKM2 inhibition provides this proliferative advantage by shunting metabolites to an anabolic pathway for macromolecule synthesis. Indeed, pervanadate treatment in PKM2-expressing cells facilitated efficient incorporation of glucose metabolites into lipids. These cells also exhibited a 36% decrease in oxygen consumption upon pervanadate treatment, which would permit operation of this pathway in the low-oxygen environments found in solid tumors.
These data show that PKM2 is important for aerobic glycolysis in tumors and is regulated by phosphotyrosine peptide binding, which enables cancer cells to strike a balance between energy production and macromolecule anabolism. It will be interesting to determine whether such regulation occurs as the result of a dedicated tyrosine–kinase pathway.
Emily J. Chenette Signaling Gateway
Original References:
Christofk, H. R. et al. The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth Nature452, 230-233 (2008). Full text | PDF | Subscribe to Nature
Christofk, H. R., Vander Heiden, M. G., Wu, N., Asara, J. M. & Cantley, L. C. Pyruvate kinase M2 is a phosphotyrosine-binding protein Nature452, 325-330 (2008). Full text | PDF | Subscribe to Nature
The platelet protein Kindlin-3 activates integrins by binding to their cytoplasmic β tails upon blood vessel injury, thus allowing efficient platelet aggregation and preventing blood loss.
Platelets aggregate at injury sites to seal damaged blood vessels and stop bleeding. Platelet integrins must be activated to ensure firm adhesion and aggregation, switching from a low- to a high-affinity state upon contact with a wounded vessel. In Nature Medicine, Reinhard Fässler and colleagues report that the Kindlin-3 protein activates platelet integrin in vivo by binding to its cytoplasmic β tail.
Kindlin-3 (Kind3) is exclusively expressed in hematopoietic cells and localizes to integrin adhesion sites. Reinhard Fässler and colleagues observed that mice lacking Kindlin-3 (Kind3-/-) died within one week of birth and showed severe bleeding due to impaired platelet function. The authors also found that loss of Kindlin-3 offered protection from vessel occlusion after mesenteric arteriole injury, which normally leads to thrombosis.
Platelet receptors interact with collagen and von Willebrand factor (vWF) at sites of blood vessel injury. Together with locally released thrombin and soluble platelet agonists, the platelet receptors trigger integrin αIIbβ3 activation and platelet aggregation. The authors showed that Kind3-/- platelets failed to aggregate upon exposure to platelet agonists, but could still change from a discoid to a spherical shape, suggesting a specific defect in αIIbβ3-dependent aggregation rather than a general impairment of signaling. Antibody staining showed that integrin αIIbβ3 was no longer activated, which was confirmed by the inability of stimulated Kind3-/- platelets to bind fibrinogen.
Integrin αIIbβ3 normally interacts with collagen-bound vWF, whereas integrin α2β1 binds directly to collagen fibers. Kind3-/- platelets were unable to stably adhere to collagen fibers, which, together with a loss of staining for activated β1 integrin-specific antibodies, suggested that neither αIIbβ3 nor α2β1 integrins were activated.
The cytoskeletal protein Talin interacts with the tail of integrin β through its phosphotyrosine-binding (PTB) domain — a final step in integrin αIIbβ3 activation that is common irrespective of the upstream signaling pathway. Similar to Talin, the presence of a region of Kindlin-3 containing a PTB domain is sufficient for integrin binding. However, unlike Talin, Kindlin-3 directly bound β1 and β3 tails at their distal rather than their proximal membrane residues.
Together, the findings of Reinhard Fässler and colleagues highlight an essential role for Kindlin-3 in platelet integrin activation during hemostasis and thrombosis, which changes the current view of Talin being sufficient for integrin activation. As Kindlin-3 can activate both αIIbβ3 and α2β1 integrins, it may be a general regulator of integrin function similar to Talin; however, the relative roles of Kindlin-3 and Talin, and the hierarchy of their binding to integrins remain to be determined.
Kim Baumann Cell Migration Gateway
Original Reference:
Moser, M., Nieswandt, B., Ussar, S., Pozgajova, M. & Fässler, R. Kindlin-3 is essential for integrin activation and platelet aggregation Nature Medicine14, 325-330 (2008). Full text | PDF | Subscribe to Nature Medicine
