ALBERT

Description: Key Points Different isoforms of PIP5KIγ fulfill unique functions in platelets.

Description: Platelets release different types of secretory granules into their local environment, and this allows them to contribute to a variety of physiologic processes. These platelet granules include alpha granules, dense granules, and lysosomes, which all derive from the endosomal-lysosomal system. However, the mechanism of the biogenesis of each type of granule is not completely understood. Phosphatidylinositol-3,5-bisphosphate [PtdIns(3,5)P2] is a membrane phosphoinositide that is essential for the regulation of membrane homeostasis, as well as for vesicle trafficking and cargo transport along the endolysosomal system in mammals. PtdIns(3,5)P2 is synthesized on endosomes by the lipid kinase PIKfyve. Given the role of PIKfyve-mediated PtdIns(3,5)P2 production in the endolysosomal pathway in other mammalian cells, we hypothesized that PtdIns(3,5)P2 was an essential regulator for the biogenesis of granules within platelets. To analyze the contribution of PtdIns(3,5)P2 to platelet granule biogenesis, we generated mice lacking PIKfyve kinase activity specifically in their platelets and in their megakaryocytes (PIKfyveflox/flox Pf4-Cre). We found that when compared with the control mice, PIKfyveflox/flox Pf4-Cre mice contained higher levels of the lysosomal enzyme, β-hexosaminidase, within their platelets and within their plasma. The PIKfyve-ablated platelets also released excessive β-hexosaminidase ex vivo upon stimulation with ADP, collagen or thrombin. However, the percentage of the total cellular β-hexosaminidase that was released from the PIKfyve-ablated platelets was comparable with that of control platelets. This suggests that the increased release of β-hexosaminidase from the PIKfyve-ablated platelets is due to the excessive storage of this enzyme within these platelets and not because of increased efficiency of lysosome secretion. In addition, we observed that PIKfyve-ablated platelets expressed increased amounts of Lysosomal Associated Membrane Protein 1 (LAMP-1), a marker of late endosomes and lysosomes. However, PIKfyve-ablated platelets expressed normal amounts of Early Endosome Antigen 1 (EEA-1), a marker of early endosomes. These results suggest that PIKfyve is critical for a component of platelet lysosome biology that occurs after the maturation of early endosomes. Together, these data demonstrate that PIKfyve is essential for the homeostasis of the endolysosomal system in platelets. Notably, the generation and secretion of the alpha granule components were intact in the PIKFyve-ablated platelets. This was shown by the normal expression of von Willebrand factor, platelet basic protein, and platelet factor 4. Likewise, secretion of ATP stored in the dense granules was similar between the PIKfyveflox/flox Pf4-Cre mice and their control littermates. Together, these results suggest that PIKfyve plays an essential regulatory role along the endolysosomal pathway in platelets, and the loss of PIKfyve in platelets can lead to an abnormality of lysosomal storage. Unexpectedly, we also found that the loss of PIKfyve exclusively within platelets triggers an inappropriate inflammatory response. This is shown by the massive tissue infiltration of aberrant vacuolated macrophages. In turn, this leads to multiple organ defects that impair development, body mass, and survival in mice. Moreover, mice lacking PIKfyve within their platelets developed accelerated arterial thrombosis in vivo, despite having normal platelet aggregation ex vivo. It is also remarkable that mice lacking PIKfyve in their platelets attenuated their organ defects when the secretion of their platelet lysosomes were inhibited in vivo. Collectively, our study demonstrates that PIKfyve is an essential regulator of platelet lysosome biogenesis. This study also highlights the previously unrecognized and important contributions of platelet lysosomal storage to inflammation, arterial thrombosis, and macrophage biology. Disclosures: No relevant conflicts of interest to declare.

Description: Abstract 697 Phosphatidylinositol kinases are crucial in the generation of diverse membrane phosphoinositides that regulate the intracellular signaling for essential cellular processes including membrane trafficking. PIKFyve is a kinase that generates phosphatidylinostol 3,5 bisphosphate (PtdIns(3,5)P2) from PtdIns(3)P. Yeast mutants lacking the PIKFyve homologue were unable to synthesize PtdIns(3,5)P2 and exhibited enlarged vacuoles, defective membrane recycling and abnormal protein sorting to multivesicular bodies. In human platelets, PtdIns(3,5)P2 synthesis increased following thrombin stimulation, and our data indicated that murine platelets express PIKFyve mRNA. We hypothesized that PIKFyve and its phospholipid product, PtdIns(3,5)P2 would be critical in the membrane trafficking during the biogenesis of platelet granules. To test this hypothesis, we engineered mice to lack PIKFyve specifically within their platelets and megakaryocytes. The resulting PIKFyvefl/fl mice contained a conditional targeting mutation within two exons encoding for a critical component of its kinase domain. The conditionally targeted mice were crossed with transgenic mice expressing Cre recombinase under the control of platelet factor 4 (PF4) promoter. As expected, RT-PCR analysis confirmed a complete loss of normal PIKFyve mRNA in platelets derived from PIKFyvefl/fl PF4 Cre+ mice. Surprisingly, by 3 weeks of age, PIKFyvefl/fl PF4 Cre+ mice started to display a robust pleomorphic phenotype characterized by dorsal hair loss and progressive weight gain due to generalized body swelling. By 5–6 months of age, the body weight of the PIKFyvefl/fl PF4 Cre+ mice was about 30% higher than that of the control littermate mice (PIKFyvefl/fl PF4 Cre- or PIKFyvefl/+ PF4 Cre+) and appeared acutely ill with decreased ambulation and feeding. A DEXA scan revealed that PIKFyvefl/fl PF4 Cre+ mice had approximately 50% lower body fat content compared to their control littermates. Necropsy of PIKFyvefl/fl PF4 Cre+ mice at about 6 months of age demonstrated that their weight gain was due to massive infiltration of largely vacuolated F4/80 expressing macrophages in multiple organs including skin, muscle, lung, heart, liver, pancreas, small and large intestines, kidney, genital organs, thymus and bone marrow. The vacuoles within these macrophages occupied the entire cytoplasm and were reminiscent of PIKFyve null yeast mutant cells. Unexpectedly, PCR analysis of genomic DNA from tissues infiltrated by these vacuolated macrophages revealed that PIKFyve deletion had occurred within these cells. This indicated that the PF4 Cre was expressed in these macrophages and led to targeting of PIKFyve and ultimately to the development of vacuolated macrophages that infiltrated multiple organs. Despite the pleomorphic phenotype, the platelet counts of the PIKFyvefl/fl PF4 Cre+ mice were normal. Furthermore, platelets lacking PIKFyve normally exposed P-selectin from alpha granules and released ATP from dense granules in response to agonist stimulation. Nevertheless, the PIKFyvefl/fl PF4 Cre+ platelets exhibited a three-fold increase of basal b-hexosaminidase, a lysosomal hydrolase. This data suggests a role for PIKFyve in the biogenesis of platelet lysosomes. Furthermore, in vivo platelet thrombus formation analysis using the ferric chloride induced carotid artery injury model revealed that PIKFyvefl/fl PF4 Cre+ mice made an occluding thrombus significantly faster than the control littermate mice demonstrating that they are prothrombotic. Histological analysis of the carotid artery specimen at the site of thrombus demonstrated that the intravascular side did not contain vacuolated macrophages. This suggests that the enhanced thrombosis formation was due to the increased lysosomal content within the platelets, and not due to infiltration by abnormal macrophages. We conclude that PF4 promoter driven selective deletion of PIKFyve kinase is associated with defective biogenesis of platelet lysosomes and a prothrombotic effect in vivo. Furthermore, PIKFyvefl/fl PF4 Cre+ mice display a pleomorphic phenotype associated with vacuolated macrophage accumulation in multiple organs and suggest that PF4 promoter driven Cre expression is not restricted to megakaryocytes and platelets. This study also highlights a potential problem in previous and current work using the PF4 Cre transgenic model. Disclosures: No relevant conflicts of interest to declare.

Description: GATA-1 controls the development of erythroid cells and megakaryocytes through its ability to activate and repress gene transcription. GATA-1 binds many nuclear proteins, but only a few of these associations have been examined in vivo. One important example is FOG-1, a critical cofactor that contributes to both gene activation and repression by GATA-1. Loss of FOG-1 generally phenocopies GATA-1 deficiency, impairing both erythroid and megakaryocytic differentiation. We reported previously that FOG-1 directly binds the NuRD protein complex, which contains histone deacetylase and chromatin remodeling activities. This provides one mechanism for GATA-1/FOG-1-mediated gene repression. Accordingly, ChIP profiling of the NuRD proteins MTA-2, RbAp46 and Mi-2β revealed the presence of these molecules at the Kit and Gata2 genes both of which are directly repressed by GATA-1 in a FOG-1-dependent manner. NuRD proteins were spread broadly across the Kit and Gata2 genes but were further enriched at sites occupied by GATA-1 and FOG-1 in vivo. Unexpectedly, we also observed NuRD components at GATA-1-activated genes including β-globin and Ahsp. Moreover, the ability of FOG-1 to augment GATA-1-induced transcription in transient transfection assays required NuRD binding. Hence, NuRD may be bi-functional, contributing to either gene activation or repression, depending on the transcriptional and cellular context. To study the role of the FOG-1/NuRD interaction in vivo we generated mice bearing missense mutations in the Fog-1(Zfpm1) gene that disrupt NuRD binding in the FOG-1 protein. Homozygous mutant mice are born at reduced Mendelian ratios. Surviving animals display ineffective erythropoiesis marked by splenomegaly and impaired erythroid maturation. In addition, homozygous mutant animals display macrothrombocytopenia with impaired platelet function. Thus, recruitment of NuRD by GATA-1 and FOG-1 is essential for both erythropoiesis and megakaryocytopoiesis. Ongoing studies include further phenotypic analysis of the mutant mice, including comparative gene expression analysis in stage-matched wild-type and mutant erythroid cells to identify critical NuRD-dependent GATA target genes, and to resolve whether NuRD is essential for both activation and repression by GATA-1 and FOG-1 in vivo. An important open question under investigation is how recruitment of the NURD complex can lead to suppression of some genes and the enhanced expression of others. The FOG-1/NuRD mutant mice provide useful tools to dissect transcription pathways initiated by GATA-1. Moreover, given the role of GATA-1 mutations in congenital anemias and megakaryoblastic leukemias, enzymatic components of the NuRD complex may provide novel targets for pharmacologic manipulation to treat these disorders.

Description: Chemokines acting through G-protein coupled receptors play an essential role in both the immune and inflammatory responses. Phosphatidylinositol 3-kinase (PI3K) and phospholipase C (PLC) are two distinct signaling molecules that have been proposed as potential candidates in the regulation of this process. Studies with knockout mice have demonstrated a critical role for PI3Kγ, but not PLCβ, in Gαi-coupled receptor-mediated neutrophil chemotaxis. We compared the chemotactic response of peripheral T-cells derived from wild type mice with mice containing loss-of-function mutations of either PI3Kγ, or both of the two predominant lymphocyte PLCβ isoforms (PLCβ2 and PLCβ3). In contrast to neutrophils, loss of PI3Kγ did not significantly impair T-cell migration in vitro, although PI3K pharmacologic inhibitor experiments suggest that another isoform of this enzyme might contribute to T-cell migration. However, loss of PLCβ2β3 decreased chemokine-stimulated T-cell migration in vitro. Chelation of intracellular calcium by BAPTA-AM and Quin-2 AM decreased the chemotactic response of wild type lymphocytes, but pharmacologic inhibition of PKC isoforms by GF109203x did not impair T-cell migration. This suggests that the T-cell migration defect seen in the PLCβ2β3-null T-cells may be due to an impaired ability to increase the cytoplasmic calcium concentration, while there appears to be little requirement for PKC activity. Indeed, SDF-1α-induced calcium efflux was not detected in the PLCβ2β3-null lymphocytes. Compared to fluorescently labeled wild type T-cells, labeled PLCβ2β3 knockout T-cells migrated less efficiently into secondary lymphoid organs of recipient mice. This demonstrates that PLCβ is also required for migration in vivo. PLCβ2β3-null mice develop spontaneous skin ulcers starting around 3 months of age. Histological examination of the lesions revealed a dense inflammatory infiltrate composed of neutrophils, macrophages, and plasma cells, consistent with acute and chronic inflammation. Remarkably, lymphocytes, typical of chronic inflammation, were rare to absent by histology and by paraffin immunohistochemistry for CD3, also consistent with an in vivo migratory defect of T-cells. These results show that phospholipid second messengers generated by PLCβ and isoforms of PI3K, other than PI3Kγ, play a critical role in lymphocyte chemotaxis. Collectively, our data demonstrate that although PLCβ-mediated signaling plays no role in neutrophil chemotaxis, it makes a substantial contribution to this process within T-lymphocytes.

Description: Acute lymphoblastic leukemia (ALL) in adult patients is often resistant to current therapy, making the development of novel therapeutic agents paramount. We investigated whether mTOR inhibitors (MTIs), a class of signal transduction inhibitors, would be effective in primary human ALL. Lymphoblasts from adult patients with precursor B ALL were cultured on bone marrow stroma and were treated with CCI-779, a second generation MTI. Treated cells showed a dramatic decrease in cell proliferation and an increase in apoptotic cells, compared to untreated cells. We also assessed the effect of CCI-779 in a NOD/SCID xenograft model. We treated a total of 68 mice generated from the same patient samples with CCI-779 after establishment of disease. Animals treated with CCI-779 showed a decrease in peripheral-blood blasts and in splenomegaly. In dramatic contrast, untreated animals continued to show expansion of human ALL. We performed immunoblots to validate the inhibition of the mTOR signaling intermediate phospho-S6 in human ALL, finding down-regulation of this target in xenografted human ALL exposed to CCI-779. We conclude that MTIs can inhibit the growth of adult human ALL and deserve close examination as therapeutic agents against a disease that is often not curable with current therapy.

Description: ALPS is a recently described disorder of disrupted lymphocyte homeostasis. Patients with ALPS have mutations in the Fas apoptotic pathway, leading to abnormal lymphocyte survival. Clinical manifestations in patients with ALPS vary but typically include autoimmune cytopenias, organomegaly, lymphadenopathy, and increased risk of development of lymphoma and certain leukemias. A similar constellation of findings can be seen in Evans Syndrome (ES), a hematologic disorder defined by autoimmune destruction of at least two hematologic cell types. We hypothesized that a subset of patients diagnosed with ES may have ALPS. We have designed a study to screen children with the diagnosis of ES syndrome for ALPS with a three tiered approach. First, a retrospective chart review was performed on all patients treated at CHOP in the past five years for ES who are still followed by the Division of Hematology. Next, patients with a clinical diagnosis of ES without another underlying diagnosis were screened for ALPS by flow cytometric analysis for double negative T cells (DNTs) (CD3+, CD4−, CD8−, TCRαβ+). DNTs are elevated in patients with ALPS and may provide the best current screening assay for this disease. DNTs analysis is easy to perform at low cost. Finally, patients were tested for ALPS by the gold standard test, defective in-vitro Fas mediated apoptosis. This test is labor intensive and not clinically available. 21 patients met entry criteria for the study, and, of these, 19 consented to enrollment. 16/19 had immune mediated destruction of all three hematologic cell lines. 2/19 had only had destruction of two cell types and 1 patient had immune thrombocytopenia with a positive DAT but no hemolysis or neutropenia. 14/19 had chronic disease with frequent hematologic exacerbations. The other 5 had occasional flares with illness. 10/19 had significant lymphadenopathy and 11/19 had organomegaly at some point in their history. 14/19 had either lymphadenopathy or organomegaly. 17/19 patients were screened for DNTs and 10 had elevated DNTs. Thus far, 7 patients have been tested for defective Fas mediated apoptosis. Every patient was tested in tandem with a normal control who demonstrated normal Fas mediated apoptosis. Five of the patients, all of whom had elevated DNTs, had defective Fas mediated apoptosis, confirming the diagnosis of ALPS. The other two patients did not have defective Fas mediated apoptosis: one did not have elevated DNTs and the other had a borderline result (2.7%; upper limit normal 2.6%). One of the patients with defective Fas mediated apoptosis and elevated DNTs had no history of lymphadenopathy or organomegaly. The lack of clinically identifiable lymphoproliferation in a patient with defective apoptosis is an unexpected finding, in apparent contradiction of the accepted NIH definition of ALPS which lists lymphoproliferation as a mandatory diagnostic criterion. This patient is three years old and may develop clinically identifiable lymphoproliferation with age. In summary, in the group of 17 ES patients screened to date, 59% had elevated DNTs suggestive of ALPS, with functional confirmation in 5/5 tested. Our data suggest that DNTs may be a sensitive first line-screening test and may serve as a marker that identifies patients who require definitive testing. Our preliminary findings suggest a high prevalence of ALPS among ES patients, a novel finding with important implications.