Phosphorylation of the growth arrest-specific protein Gas2 is coupled to actin rearrangements during Go-->G1 transition in NIH 3T3 cells.

Growth arrest-specific (Gas2) protein has been shown to be a component of the microfilament system, that is highly expressed in growth arrested mouse and human fibroblasts and is hyperphosphorylated upon serum stimulation of quiescent cells. (Brancolini, C., S. Bottega, and C. Schneider. 1992. J. Cell Biol. 117:1251-1261). In this study we demonstrate that the kinetics of Gas2 phosphorylation, during Go-->G1 transition, as induced by addition of 20% FCS to serum starved NIH 3T3 cells, is temporally coupled to the reorganization of actin cytoskeleton. To better dissect the relationship between Gas2 phosphorylation and the modification of the microfilament architecture we used specific stimuli for both membrane ruffling (PDGF and PMA) and stress fiber formation (L-alpha-lysophosphatidic acid LPA) (Ridley, A. J., and A. Hall. 1992. Cell. 70:389-399). All of them, similarly to 20% FCS, are able to downregulate Gas2 biosynthesis. PDGF and PMA induce Gas2 hyperphosphorylation that is temporally coupled with the appearance of membrane ruffling where Gas2 localizes. On the other hand LPA, a specific stimulus for stress fiber formation, fails to induce a detectable Gas2 hyperphosphorylation. Thus, Gas2 hyperphosphorylation is specifically correlated with the formation of membrane ruffling possibly implying a role of Gas2 in this process.

Gas2, a growth arrest-specific protein, is a component of the microfilament network system.

In this report we analyze the protein product of a growth arrest-specific gene, gas2, by means of an affinity-purified antibody raised against the protein produced in bacteria. The regulation of Gas2 biosynthesis reflects the pattern of mRNA expression (Schneider, C., R. King, and L. Philipson. 1988. Cell. 54:787-793): its relative level is tightly associated with growth arrest. Gas2 seems to be regulated also at the posttranslational level via a phosphorylation mechanism. Gas2 is well conserved during the evolution with the same apparent molecular mass (36 kD) between mouse and human. We also demonstrate that Gas2 is a component of the microfilament system. It colocalizes with actin fiber, at the cell border and also along the stress fiber, in growth-arrested NIH 3T3 cells. The pattern of distribution, detected in arrested cells, can also be observed in growing cells when they are microinjected with the purified GST-Gas2 protein. In none of the analyzed oncogene-transformed NIH 3T3 cell lines was Gas2 expression induced under serum starvation.

Dismantling cell-cell contacts during apoptosis is coupled to a caspase-dependent proteolytic cleavage of beta-catenin.

Cell death by apoptosis is a tightly regulated process that requires coordinated modification in cellular architecture. The caspase protease family has been shown to play a key role in apoptosis. Here we report that specific and ordered changes in the actin cytoskeleton take place during apoptosis. In this context, we have dissected one of the first hallmarks in cell death, represented by the severing of contacts among neighboring cells. More specifically, we provide demonstration for the mechanism that could contribute to the disassembly of cytoskeletal organization at cell-cell adhesion. In fact, beta-catenin, a known regulator of cell-cell adhesion, is proteolytically processed in different cell types after induction of apoptosis. Caspase-3 (cpp32/apopain/yama) cleaves in vitro translated beta-catenin into a form which is similar in size to that observed in cells undergoing apoptosis. beta-Catenin cleavage, during apoptosis in vivo and after caspase-3 treatment in vitro, removes the amino- and carboxy-terminal regions of the protein. The resulting beta-catenin product is unable to bind alpha-catenin that is responsible for actin filament binding and organization. This evidence indicates that connection with actin filaments organized at cell-cell contacts could be dismantled during apoptosis. Our observations suggest that caspases orchestrate the specific and sequential changes in the actin cytoskeleton occurring during cell death via cleavage of different regulators of the microfilament system.

Twist is substrate for caspase cleavage and proteasome-mediated degradation.

Twist is a member of the basic helix-loop-helix family of transcription factors. An aberrant Twist expression has been found in diverse types of cancer, including sarcomas, carcinomas and lymphomas, supporting a role for Twist in tumor progression. Twist is known to be essential for mesodermal development. However, since a prolonged Twist expression results in a block of muscle, cartilage and bone differentiation, Twist has to be excluded from somites during late embryogenesis for terminal differentiation to occur. This implies that Twist expression must be target of a tight control. Here we provide evidence that Twist undergoes post-transcriptional regulation. Twist is substrate for cleavage by caspases during apoptosis and its cleavage results in ubiquitin-mediated proteasome degradation. Our findings suggest that Twist post-transcriptional regulation may play an important role in tissue determination and raise the possibility that alterations in the protein turnover may account for Twist overexpression observed in tumors.

The protein encoded by a growth arrest-specific gene (gas6) is a new member of the vitamin K-dependent proteins related to protein S, a negative coregulator in the blood coagulation cascade.

A set of growth arrest-specific genes (gas) whose expression is negatively regulated after serum induction has previously been described (C. Schneider, R. M. King, and L. Philipson, Cell 54:787-793, 1988). The detailed analysis of one of them, gas6, is reported here, gas6 mRNA (2.6 kb) is abundantly expressed in serum-starved (48 h in 0.5% fetal calf serum) NIH 3T3 cells but decreases dramatically after fetal calf serum or basic fibroblast growth factor stimulation. The human homolog of gas6 was also cloned and sequenced, revealing a high degree of homology and a similar pattern of expression in IMR90 human fibroblasts. Computer analysis of the protein encoded by murine and human gas6 cDNAs showed significant homology (43 and 44% amino acid identity, respectively) to human protein S, a negative coregulator in the blood coagulation pathway. By using an anti-human Gas6 monospecific affinity-purified antibody, we show that the biosynthetic level of human Gas6 fully reflects mRNA expression in IMR90 human fibroblasts. This finding thus defines a new member of vitamin K-dependent proteins that is expressed in many human and mouse tissues and may be involved in the regulation of a protease cascade relevant in growth regulation.

Exposure at the cell surface is required for gas3/PMP22 To regulate both cell death and cell spreading: implication for the Charcot-Marie-Tooth type 1A and Dejerine-Sottas diseases.

Gas3/PMP22 is a tetraspan membrane protein highly expressed in myelinating Schwann cells. Point mutations in the gas3/PMP22 gene account for the dominant inherited peripheral neuropathies Charcot-Marie-Tooth type 1A disease (CMT1A) and Dejerine-Sottas syndrome (DSS). Gas3/PMP22 can regulate apoptosis and cell spreading in cultured cells. Gas3/PMP22 point mutations, which are responsible for these diseases, are defective in this respect. In this report, we demonstrate that Gas3/PMP22-WT is exposed at the cell surface, while its point-mutated derivatives are intracellularly retained, colocalizing mainly with the endoplasmic reticulum (ER). The putative retrieval motif present in the carboxyl terminus of Gas3/PMP22 is not sufficient for the intracellular sequestration of its point-mutated forms. On the contrary, the introduction of a retrieval signal at the carboxyl terminus of Gas3/PMP22-WT leads to its intracellular accumulation, which is accompanied by a failure to trigger cell death as well as by changes in cell spreading. In addition, by substituting the Asn at position 41 required for N-glycosylation, we provide evidence that N-glycosylation is required for the full effect on cell spreading, but it is not necessary for triggering cell death. In conclusion, we suggest that the DSS and the CMT1A neuropathies derived from point mutations of Gas3/PMP22 might arise, at the molecular level, from a reduced exposure of Gas3/PMP22 at the cell surface, which is required to exert its biological functions.

Rho-dependent regulation of cell spreading by the tetraspan membrane protein Gas3/PMP22.

Gas3/PMP22 plays a crucial role in regulating myelin formation and maintenance, and different genetic alterations in gas3/PMP22 are responsible for a set of human peripheral neuropathies. We have previously demonstrated that Gas3/PMP22 could regulate susceptibility to apoptosis in NIH3T3 cells but not in REF 52 cells. In this report we demonstrate that when the apoptotic response triggered by gas3/PMP22 was counteracted by Bcl-2 coexpression, morphological changes were observed. Time-lapse analysis confirmed that Gas3/PMP22 can modulate cell spreading, and this effect was strengthened after inhibition of phosphoinositide 3-kinase. Using the active form of the small GTPase RhoA, we have been able to dissect the different Gas3/PMP22 biological activities. RhoA counteracted the Gas3/PMP22-dependent morphological response but was unable to neutralize the apoptotic response. Treatment of NIH3T3 cells with cytotoxic necrotizing factor 1, which activates endogenous Rho, also counteracted Gas3/PMP22-mediated cell shape and spreading changes. Treatment of REF 52 cells, which are unresponsive to Gas3/PMP22 overexpression, with the C3 exoenzyme, inhibiting Rho activity, renders REF 52 cells responsive to Gas3/PMP22 overexpression for cell shape and spreading changes. Finally, assembly of stress fibers and focal adhesions complexes, in response to lysophosphatidic acid-induced endogenous Rho activation, was impaired in Gas3/PMP22-overexpressing cells. We hypothesize that cell shape and spreading regulated by Gas3/PMP22 through the Rho GTPase might have an important role during Schwann cells differentiation and myelinization.

Caspase activation and apoptosis in response to proteasome inhibitors.

Several studies have indicated that proteasome inhibitors (PIs) are promising anticancer agents. We have discovered that PIs have the unique ability to activate effector caspases through a mitochondrial Bcl-2 inhibitable but caspase-9 independent pathway. Stabilization of released Smac induced by blockade of the proteasome could explain the apoptosome-independent cell death induced by PIs. In fact, Smac/DIABLO critically supports this PIs-dependent caspase activation. By using a new assay, we confirm that at a single cell level both Smac and PIs can activate caspases in the absence of the apoptosome. Moreover, we have observed two PIs-induced kinetics of caspase activation, with caspase-9 being still required for the rapid caspase activation in response to mitochondrial depolarization, but dispensable for the slow DEVDase activation. In summary, our data indicate that PIs can activate downstream caspases at least in part through Smac/DIABLO stabilization.

Sensitivity of human multiple myelomas and myeloid leukemias to the proteasome inhibitor I.

The proteasome inhibitor PSI is potently cytotoxic in vitro against human chronic myeloid leukemia (CML) and acute myeloid leukemias (AML). Here, we have tested proteasome inhibitor I (PSI) in a panel of 11 human multiple myeloma (MM) cell lines and found that it has antiproliferative activity, with an IC50 between 4.5 and 557 nM at 48 h. PSI potentiated the toxicity of a number of chemotherapeutic agents in myeloid leukemia but not in MM cell lines, while in combination with therapeutic proteasome inhibitor PS-341 (Bortezomib) it had a synergistic effect. PSI suppressed the growth of AML cell lines more effectively than PS-341. CFU-GM colony assays revealed that CD34+ bone marrow progenitors from CML and AML patients were more sensitive to PSI than those from normal subjects (IC50: 5, 15 and 50 nM for AML, CML and normal, respectively). Moreover, the growth of normal primitive progenitors (LTC-IC) was unaffected by 15 nM PSI (P=0.576). PSI-induced cell death required RNA transcription and protein synthesis, but not DNA replication, was accompanied by the upregulation of Bcl-2 and modest reduction of Bax and Bcl-XL proteins, and involved the activation of caspases 2, 3, 7 and 8. These findings lend additional support to preclinical investigations with PSI.

Dissection of c-myc domains involved in S phase induction of NIH3T3 fibroblasts.

The product of the c-myc proto-oncogene is an important regulator of cell proliferation and apoptosis in murine fibroblasts. Addition of the tumor promoter, phorbol myristate acetate (PMA), prevents apoptotic cell death induced by low serum concentrations in NIH3T3 cells that constitutively express and are transformed by v-myc. The protective effect of PMA allowed us to analyse the ability of normal c-Myc and Myc deletion mutants to induce serum starved, untransformed NIH3T3 cells to enter S phase. By microinjecting these quiescent cells with wild type and mutant human c-myc plasmids, we showed that full length c-myc is able to induce S phase entry in presence of PMA, but that c-Myc mutants that delete amino acids delta 7/91, delta 41/53, delta 56/103, delta 106/143, delta 265/317 and delta 414/433 are totally inactive. c-Myc did not shorten the period before entry into S phase, since Myc overexpressing cells entered S phase with the same kinetics as control cells when both were stimulated with 20% fetal calf serum (FCS). However, c-Myc overexpression did increase the percentage of cells entering S phase when these cells were stimulated with 2% fetal calf serum. Interestingly, this ability to enhance stimulation by a suboptimal concentration of FCS was retained to a significant degree by Myc mutants that delete amino acids delta 41/53, delta 56/103 or delta 265/317. Finally, Myc mutants that delete delta 106/143 or delta 414/433 exerted a dominant negative effect on S phase entry both in quiescent cells stimulated with 2% FCS and in unsynchronized, cycling cells.

Susceptibility to p53 dependent apoptosis correlates with increased levels of Gas2 and Gas3 proteins.

p53 dependent apoptosis is a critical regulator of tumorigenesis. In this paper we demonstrate that BALB/c cells transformed with a LT mutant perturbing pRb but not p53 functions (LT-2809) show unrestrained cell division under low serum condition which is actively counterbalanced by apoptosis. BALB/c cells transformed with a LT mutant perturbing p53 but not pRb functions (LT-K1), show similar unrestrained cell division but no evident signs of apoptosis when grown in low serum. Such apoptotic response of LT-2809 cells is characterised by increased expression of Gas2 which becomes proteolytically processed. Similarly Gas3 expression is markedly increased in LT-2809 cells with respect to LT-K1. Since both Gas2 and Gas3 have been previously associated with the apoptotic response at growth arrest, our observations suggest that they could also contribute to the regulation of cellular susceptibility to p53 dependent apoptosis.

Proteolytic processing of the adherens junctions components beta-catenin and gamma-catenin/plakoglobin during apoptosis.

Apoptotic cells undergo specific morphological changes that include loss of cell-cell interactions. Cellular adhesiveness is dependent on members of the cadherin family of adhesion receptors and on the cytoplasmic adaptor proteins alpha-catenin, beta-catenin and gamma-catenin/plakoglobin. The caspase family of cystein proteases play a key role during the execution phase of the apoptotic program. These proteolytic enzymes, once activated, cleave cellular proteins which are important for the maintenance of cell integrity. Here we report that gamma-catenin is cleaved at different sites during apoptosis in various cell lines. The major apoptotic product of gamma-catenin still retains the ability to bind alpha-catenin but loses the carboxy-terminal region. We also show that gamma-catenin is cleaved by caspase-3 in vitro although with lower affinity when compared to PARP or beta-catenin. These findings indicate that multiple proteolytic events regulate the dismantling of the cell-cell junctional complexes during apoptosis.

Tissue transglutaminase is a caspase substrate during apoptosis. Cleavage causes loss of transamidating function and is a biochemical marker of caspase 3 activation.

Tissue transglutaminase (tTG) is a Ca2+-dependent cross-linking enzyme that participates in the apoptotic machinery by irreversibly assembling a protein scaffold that prevents the leakage of intracellular components. In the present study a single-chain antibody fragment (scFv) detecting tTG is described. We demonstrate that TG/F8 scFv, selected from a phase display library of human V-gene segments by binding to guinea-pig liver tTG, can react with human tTG both in Western blot and in immunohistochemistry. The specific detection of tTG by TG/F8 in human thymocytes is verified by mass spectrometric analysis of the purified protein. Furthermore, we demonstrate that in lymphoid cells tTG is cleaved by caspase 3 during the late phase of apoptotic death, concomitant to DNA fragmentation, and that such cleavage causes loss of cross-linking function. We propose tTG cleavage as a valuable biochemical marker of caspase 3 activation during the late execution phase of apoptosis.

Identification and characterization of a new member of the gas3/PMP22 gene family in C. elegans.

The Gas3/PMP22 protein family is characterized by tetraspan transmembrane proteins. The gas3/PMP22 gene is highly expressed in Schwann cells of the peripheral nervous system, and different alterations of this gene are associated with hereditary demyelinating neuropathies, such as the Charcot-Marie-Tooth type 1A, the Dejerine-Sottas syndrome and the Hereditary Liability to Pressure Palsies (HNPP).Here, we report on the identification of at least one member of the Gas3/PMP22 family in the nematode C. elegans (C01C10.1b). C01C10.1b shares 36% of identical amino acids with the human Gas3/PMP22 and is characterized by four hydrophobic putative transmembrane domains. It lacks the typical N-linked glycosylation consensus in the first extracellular loop. C01C10.1b is transcribed as an operon downstream to the gene C01C10.1a, which encodes for a putative tetraspan protein with less conserved homology with the Gas3/PMP22 family. Interestingly, C01C10.1a contains three N-glycosylation sites at the C-terminus. Both genes are expressed in different nematode developmental stages and in the adults. The characterization of one member of the gas3/PMP22 family in C. elegans gives the opportunity to use this model organism to investigate the role of gas3/PMP22 in the regulation of cell proliferation and differentiation and its relation to the hereditary neurodegenerative diseases in humans.

USP18 is a key regulator of the interferon-driven gene network modulating pancreatic beta cell inflammation and apoptosis.

Type 1 diabetes (T1D) is an autoimmune disease targeting pancreatic beta cells. Genome-wide association studies and gene expression analysis identified interferon (IFN)-driven gene networks as crucial pathways in the pathogenesis of T1D. IFNs are linked to the response to viral infections and might contribute to the initiation of the autoimmune process in T1D. We presently analyzed the role of ubiquitin-specific peptidase 18 (USP18), an interferon-stimulated gene 15-specific protease, on IFN-induced pancreatic beta cell inflammation and apoptosis. Our findings indicate that USP18 inhibition induces inflammation by increasing the STAT signaling and exacerbates IFN-induced beta cell apoptosis by the mitochondrial pathway of cell death. USP18 regulates activation of three BH3-only proteins, namely, DP5, Bim and PUMA in pancreatic beta cells, suggesting a direct link between regulators of the type I IFN signaling pathway and members of the BCL-2 family. USP18 depletion increases the expression of the T1D candidate gene MDA5, leading to an upregulation of double-stranded RNA-induced chemokine production. These data suggest a cross talk between the type I IFN signaling pathway and a candidate gene for T1D to increase pro-inflammatory responses in beta cells. The present study shows that USP18 is a key regulator of IFN signaling in beta cells and underlines the importance of this pathway in beta cell inflammation and death.

Change in the expression of a nuclear matrix-associated protein is correlated with cellular transformation.

We have identified a monoclonal antibody that recognizes a nuclear matrix-associated protein in NIH 3T3 cells. The immunofluorescence pattern consists predominantly of bright nuclear granule clusters distributed throughout the nucleoplasm, with the exclusion of nucleoli. It recognizes a protein of 190 kDa that is down-regulated to various degrees in a panel of single-oncogene-transformed NIH 3T3 cells. Its localization is similar, but not identical, to the spliceosomal speckles. p190 shows a coordinate expression during the growth cycle of nontransformed NIH 3T3 cells; it is synthesized at the highest level under growth arrest conditions. It is expressed in adult mouse brain and is also present in human IMR-90 fibroblasts.

Microfilament reorganization during apoptosis: the role of Gas2, a possible substrate for ICE-like proteases.

Gas2, a component of the microfilament system, belongs to the class of gas genes whose expression is induced at growth arrest. After serum or growth factor addition to quiescent NIH 3T3 cells, Gas2 is hyperphosphorylated and relocalized at the membrane ruffles. By overexpressing gas2wt and a series of deletion mutants of the C-terminal region, we have analysed its role in the organization of the actin cytoskeleton in different cell lines. Overexpression of Gas2 deleted at its C-terminal region (delta 276-314 and delta 236-314), but not its wild-type form, induces dramatic changes in the actin cytoskeleton and cell morphology. These effects are not due to interference of the deleted forms with the endogenous Gas2wt function but could be ascribed to a gain of function. We demonstrate that during apoptosis the C-terminal domain of Gas2 is removed by proteolytic cleavage, resulting in a protein that is similar in size to the described delta 276-314. Moreover, by using in vitro mutagenesis, we also demonstrate that the proteolytic processing of Gas2 during apoptosis is dependent on an aspartic acid residue at position 279. The evidence accumulated here could thus represent a first example of a mechanism linking apoptosis with the co-ordinated microfilament-dependent cell shape changes, as possibly mediated by an interleukin-1 beta-converting enzyme (ICE)-like dependent proteolytic cleavage of the Gas2 protein.

Apoptotic phenotype induced by overexpression of wild-type gas3/PMP22: its relation to the demyelinating peripheral neuropathy CMT1A.

Although the Gas3/PMP22 protein is expressed at highest levels in differentiated Schwann cells, its presence, albeit at lower levels, in non-neuronal tissues and in NIH-3T3 growth-arrested fibroblasts argues for a more general function of this protein that is uncoupled to myelin structure. We show that gas3/PMP22 overexpression in NIH-3T3 growing cells leads to an apoptotic-like phenotype, which is suppressed by antioxidants and characterized by typical membrane blebbing, rounding up, and chromatin condensation, but with no evidence of DNA fragmentation. REF-52 fibroblasts seem to be completely refractive to gas3/PMP22 overexpression. Recently, several point mutations of the human gas3/PMP22 gene have been associated with Charcot-Marie-Tooth type 1A (CMT1A), a common hereditary demyelinating neuropathy. When gas3/PMP22 point mutations (L16P, S79C, T118M, and G150D) are similarly overexpressed in NIH-3T3 cells, the induced apoptotic-like phenotype as compared to the wild-type is significantly reduced. Both of the dominant mutations (L16P, S79C) for CMT1A behave as dominant negatives with respect to the wild type, whereas T118M, the only recessive mutant described, behaves as recessive under the same coexpression experiments. These data suggest a role for altered Schwann cell apoptosis in the pathogenesis of CMT1A.

Expression of growth arrest-specific (Gas) genes in murine keratinocytes: Gas2 is specifically regulated.

In order to elucidate a possible role of growth arrest-specific (gas) genes in the regulation of tissue proliferation, we analyzed their expression in keratinocytes isolated from murine back skin. On the mRNA level gas1, gas5, and gas6 were found to be significantly expressed whereas there was a relatively low expression of gas2, gas3, and gas4. Using keratinocytes fractionated according to their density resulted in subpopulations of cells: differentiating suprabasal cells in fractions I and II; proliferative basal cells in fractions IIIa, III and IV. We found gas2 protein to be expressed more strongly in the proliferative cells than in the differentiating cells. Stimulation of hyperproliferation by 12-O-tetradecanoylphorbol-13-acetate (TPA) resulted in a transient increase of gas2 protein content concomitantly with the time of maximal cell renewal. In this respect the murine keratinocyte cell line MSCP5 resembled freshly isolated keratinocytes. There was a higher expression of gas2 protein during exponential growth than during growth arrest, induced either by serum starvation or by TGFbeta treatment. Since, in contrast to the results reported for 3T3 cells, growth arrest within these cells was not accompanied by an elevation of gas2 protein, we suggest a cell-specific regulation of its expression.

Caspase-2-induced apoptosis is dependent on caspase-9, but its processing during UV- or tumor necrosis factor-dependent cell death requires caspase-3.

Mammalian caspases are a family of cysteine proteases that plays a critical role in apoptosis. We have analyzed caspase-2 processing in human cell lines containing defined mutations in caspase-3 and caspase-9. Here we demonstrate that caspase-2 processing, during cell death induced by UV irradiation, depends both on caspase-9 and caspase-3 activity, while, during TNF-alpha-dependent apoptosis, capase-2 processing is independent of caspase-9 but still requires caspase-3. In vitro procaspase-2 is the preferred caspase cleaved by caspase-3, while caspase-7 cleaves procaspase-2 with reduced efficiency. We have also demonstrated that caspase-2-mediated apoptosis requires caspase-9 and that cells co-expressing caspase-2 and a dominant negative form of caspase-9 are impaired in activating a normal apoptotic response and release cytochrome c into the cytoplasm. Our findings suggest a role played by caspase-2 as a regulator of the mitochondrial integrity and open questions on the mechanisms responsible for its activation during cell death.