Cell surface protein decreases microvilli and ruffles on transformed mouse and chick cells.

Transformation of cultured fibroblasts usually results in a decrease in a high molecular weight cell surface glycoprotein (LETS protein) and often in increased numbers of surface microvilli and ruffles. We have isolated such a major cell surface glycoprotein from chick embryo fibroblasts; this protein, CSP, is decreased after transformation. Treatment of a mouse tumor cell line (SV1), L929 cells, and transformed chick fibroblasts with CSP results in a decrease in the number of microvilli and marginal ruffles, accompanied by restoration of a more normal morphology.     

Heat protectors and heat-induced preferential redistribution of 26 and 70 kDa proteins in Chinese hamster ovary cells

An overall increase of 40% in nuclear-associated protein has been shown to be one of the sequellae of exposure of eukaryotic cells to elevated temperatures. Several investigators have shown that the increased protein/DNA ratios correlated well with the degree of cytotoxicity. In previous investigations, we have shown that cycloheximide, which protects the cell from the killing effects of heat, produces a dramatic reduction of the bulk nuclear-associated proteins after heating. In this investigation, we studied a previously unobserved efflux of a 26 kDa protein after heat shock and the preferential accumulation of the 70 kDa protein. The 26 kDa protein was shown not to be a member of previously described heat shock protein families. Preferential reduction of a 26 kDa protein and accumulation of a 70 kDa protein was observed in nuclei isolated from Chinese hamster ovary cells after heating at 43 degrees C. After heat treatment, the 26 kDa protein in the nucleus was decreased to a level 0.1-0.3 times the original amount in unheated cells, and the 70 kDa protein in the nucleus increased by a factor of 1.6-1.8. The normal levels of these two proteins were restored when cells were incubated at 37 degrees C following heat shock. Cells treated with heat protectors, cycloheximide and histidinol, demonstrated approximately the same redistribution in nuclear 26 and 70 kDa proteins immediately after heating as those not exposed to these drugs. On the other hand, restoration to control levels was much faster in the protector-treated cells, suggesting that "repair" of heat-induced damage is an important factor in the cells ability to survive this insult. Return to normal protein levels did not require new protein synthesis.     

Mitochondrial protein synthesis in a mammalian cell-line with a temperature-sensitive leucyl-tRNA synthetase.

The temperature-sensitive Chinese hamster ovary cell mutant tsH1, has been shown previously to contain a temperature-sensitive leucyl-tRNA synthetase. At the non-permissive temperature of 40 degrees C cytosolic protein synthesis is rapidly inhibited. The protein synthesis which continues at 40 degrees C appears to be mitochondrial, since: (a) whole-cell protein synthesis at the permissive temperature of 34 degrees C is not inhibied by tevenel, the sulfamoyl analogue of chloramphenicol and a specific inhibitor of mitochondrial protein synthesis; however, whole-cell protein synthesis at 40 degrees C is inhibited by tevenel, (b) Protein synthesis by isolated mitochondria from tsH1 cells is not significantly inhibited at 40 degrees C. (c) At 40 degrees C [14C]leucine is incorporated predominantly into the mitochondrial fraction of tsH1 cells. (d) The incorporation of [14C]leucine at 40 degrees C into mitochondrial proteins of tsH1 cells is inh-bited by tevenel but not by cycloheximide. These results suggest that the mitochondria of tsH1 cells contain a leucyl-tRNA synthetase which is different from the cytosolic enzyme. The inhibition of cytosolic, but not of mitochondrial protein synthesis in tsH1 cells at 40 degrees C allows the selective labelling of mitochondrial translation products in the absence of inhibitors. The mitochondrial translation products labelled in tsH1 cells at 40 degrees C and at 34 degrees C in the presence of cycloheximide have been compared by sodium dodecylsulphate-polyacrylamide gel electrophoresis. Both conditions of labelling give similar profiles. The mitochondrial translation products are resolved into two components, one with an apparent molecular weight range from 40,000 to 20,000 and a second with an apparent molecular weight range from 20,000 to 10,000.     

Degradation of Connexins from the Plasma Membrane Is Regulated by Inhibitors of Protein Synthesis

Little is known about the mechanism and regulation of connexin turnover from the plasma membrane. We have used a combination of cell surface biotinylation, immunofluorescence microscopy, and scrape-load dye transfer assays to investigate the effect of the protein synthesis inhibitor cycloheximide on connexin43 and connexin32 after their transport to the plasmalemma. The results obtained demonstrate that cycloheximide inhibits the turnover of connexins from the surface of both gap junction assembly-deficient and -efficient cells. Moreover, cell surface connexin saved from destruction by cycloheximide can assemble into long-lived, functional gap junctional plaques. These findings support the concept that downregulation of connexin degradation from the plasma membrane can serve as a mechanism to enhance gap junction-mediated intercellular communication.     

Degradation of connexins from the plasma membrane is regulated by inhibitors of protein synthesis

Little is known about the mechanism and regulation of connexin turnover from the plasma membrane. We have used a combination of cell surface biotinylation, immunofluorescence microscopy, and scrape-load dye transfer assays to investigate the effect of the protein synthesis inhibitor cycloheximide on connexin43 and connexin32 after their transport to the plasmalemma. The results obtained demonstrate that cycloheximide inhibits the turnover of connexins from the surface of both gap junction assembly-deficient and -efficient cells. Moreover, cell surface connexin saved from destruction by cycloheximide can assemble into long-lived, functional gap junctional plaques. These findings support the concept that downregulation of connexin degradation from the plasma membrane can serve as a mechanism to enhance gap junction-mediated intercellular communication.     

Alkaline phosphodiesterase I of cultured mouse peritoneal macrophages. Distribution between cell surface and cytoplasm, turnover rate, and effects of colchicine

Mouse peritoneal macrophages were cultivated in vitro and analyzed for activity of alkaline phosphodiesterase I. The specific enzyme activity was 2 to 3 times higher in thioglycollate-elicited than in resident cells and raised slowly for at least 3 days of culture in both cell types. Following phagocytosis of polystyrene latex beads the enzyme activity remained unchanged. About 90% of the activity was lost by treatment of the cells with the diazonium salt of sulfanilic acid (DASA) for 30 min at 37 degrees C. At 4 degrees C the corresponding degree of inactivation was only 40 to 50%. This temperature-dependent variation in the accessibility of the enzyme to inactivation by DASA could be related to the high rate of endocytosis in macrophages. During a 30 min incubation with DASA at 37 degrees C, large amounts of surface membrane will be internalized and replaced from the interior of the cell. Surface membrane is thus transferred into intracellular membrane and vice versa. Moreover, these membranes will be exposed to DASA independent of their location at the start of incubation. At 4 degrees C no comparable membrane traffic will take place. Treatment of the cells with cycloheximide at concentrations that inhibited protein synthesis by 90% or more left the enzyme activity essentially unaltered, indicating a slow turnover rate. On the basis of these findings, it is suggested that about half of the alkaline phosphodiesterase I activity of cultured macrophages is located in the plasma membrane. The other half is believed to be present in endocytic vesicles, lysosomes, and other as yet unidentified organelles that participate in the circulation of membrane constituents between the plasma membrane and the interior of the cell.     

Correlation between redistribution of a 26 kDa protein and development of chronic thermotolerance in various mammalian cell lines

Previous studies suggested that a 26 kDa protein might play an important role in protein synthesis-independent thermotolerance development in CHO cells. To determine if this phenomenon was universal, four mammalian cell lines, viz., CHO, HA-1, murine Swiss 3T3, and human HeLa, were studied. Cells were heated at 42 degrees C, and the level of 26 kDa protein in the nucleus was measured, together with clonogenic survival and protein synthesis. The results demonstrated that 1) the 26-kDa protein was present in the four different cell lines, and 2) the level of the 26 kDa protein in their nuclei was decreased by 30-70% after heating at 42 degrees C for 1 hr. However, restoration of this protein occurred along with development of chronic thermotolerance. The protein synthesis inhibitor cycloheximide (10 micrograms/ml) neither inhibited the development of chronic thermotolerance nor affected the restoration of the 26 kDa protein in the nucleus. In fact, this drug protected cells from hyperthermic killing and heat-induced reduction of 26 kDa protein in the nucleus. Heat sensitizers, quercetin (0.1 mM), 3,3'-dipentyloxacarbocyanine iodide (DiOC5[3]: 5 micrograms/ml), and stepdown heating (45 degrees C-10 min----42 degrees C), potentiated hyperthermic killing and inhibited or delayed the restoration of the 26 kDa protein to the nucleus. These results support a correlated, perhaps causal relationship between the restoration of the 26 kDa protein and chronic thermotolerance development in four different mammalian cell lines.     

The surface content of asialoglycoprotein receptors on isolated hepatocytes is reversibly modulated by changes in temperature

We have investigated the effect of temperature on the content of surface asialoglycoprotein receptors on isolated rat hepatocytes. Receptor was determined by measuring the specific binding of 125I- or [3H] asialo-orosomucoid at 0 degrees C. As reported previously, the receptor number/cell increases 2-3-fold within 30-60 min when freshly isolated cells are warmed from 0-37 degrees C (Weigel, P. H. (1980) J. Biol. Chem. 255, 6111-6120). This increase in receptor number is not inhibited by cycloheximide and also occurs on cells which have first been treated with EDTA to expose a population of cryptic receptors on the cell surface. The rate and extent of the receptor number increase on the cell surface are proportional to the temperature above about 17 degrees C. If cells are first equilibrated at 37 degrees C and then transferred to a lower temperature, the surface receptor number decreases at a rate and to an extent dependent on the temperature. The surface receptor number can be modulated up and down by successive temperature change cycles between 25 and 37 degrees C. In this temperature range, the number of surface receptors/cell is dependent on the final temperature but independent of the pathway to that temperature and is, therefore, a function of state with respect to temperature. The results demonstrate that temperature changes reversibly modulate the number of receptors on the hepatocyte surface. We conclude that, in the absence of ligand, surface receptors can either recycle or can be reversibly internalized or sequestered to prevent access to ligand. The results may also explain why different laboratories have reported a wide range of values for the number of receptors per hepatocyte.     

Rapid turnover of endogenous endonuclease activity in thymocytes: effects of inhibitors of macromolecular synthesis

Previous work has shown that inhibitors of protein or mRNA synthesis block endonuclease activation in thymocytes undergoing programmed cell death. In the present study we used isolated nuclei to investigate the effects of cycloheximide and actinomycin D, inhibitors of protein and mRNA synthesis, respectively, on endogenous endonuclease activity in thymocytes. We observed a rapid loss of Ca2(+)-dependent endonuclease activity in nuclei isolated from thymocytes treated with these inhibitors. In contrast, pretreatment of cells with antipain and leupeptin, inhibitors of proteases, prevented the depletion of endonuclease activity in the nuclei, suggesting that proteolysis was involved. The effects of cycloheximide and actinomycin D were mimicked by incubating thymocytes with treatments known to exert their effects via activation of protein kinase C. Our results suggest that endonuclease activity in thymocyte nuclei undergoes rapid, spontaneous turnover. Agents interfering with macromolecular synthesis may therefore block DNA fragmentation in thymocytes by depleting nuclei of endogenous endonuclease activity.     

Cycloheximide increases the thermostability of proteins in Chinese hamster ovary cells

Protein denaturation resulting from temperatures between 42.0 degrees C and 50 degrees C has been observed and implicated as the lethal lesion for hyperthermic cell killing. A logical corollary is that protection against hyperthermic killing requires stabilization of cellular proteins against thermal denaturation. To test this, Chinese hamster ovary cells were treated with the heat protector cycloheximide and then subjected to differential scanning calorimetry to measure protein denaturation. Cycloheximide stabilized proteins that denatured between 42 degrees C and 52 degrees C in control cells by increasing their transition (denaturation) temperature by an average of 1.3 degrees C. In addition, cycloheximide reduced the cytotoxicity of actinomycin D and adriamycin, suggesting that protein stabilization protects cells against stresses other than hyperthermia.     

Defective plasma membrane assembly in yeast secretory mutants.

Yeast mutants that are conditionally blocked at distinctive steps in secretion and export of cell surface proteins have been used to monitor assembly of integral plasma membrane proteins. Mutants blocked in transport from the endoplasmic reticulum (sec18), from the Golgi body (sec7 and sec14), and in transport of secretory vesicles (sec1) show dramatically reduced assembly of galactose and arginine permease activities. Simultaneous induction of galactose permease and alpha-galactosidase (a secreted glycoprotein) in sec mutant cells at the nonpermissive temperature (37 degrees C) shows that both activities accumulate and can be exported coordinately when cells are returned to the permissive temperature (24 degrees C) in the presence or absence of cycloheximide. Plasma membrane fractions isolated from sec mutant cells radiolabeled at 37 degrees C have been analyzed by two-dimensional sodium dodecyl sulfate-gel electrophoresis. Although most of the major protein species seen in plasma membranes from wild-type cells are not efficiently localized in sec18 or sec7, several of these proteins appear in plasma membranes from sec1 cells. These results may be explained by contamination of plasma membrane fractions with precursor vesicles that accumulate in sec1 cells. Alternatively, some proteins may branch off during transport along the secretory pathway and be inserted into the plasma membrane by a different mechanism.     

Cell surface distribution and intracellular fate of asialoglycoproteins: a morphological and biochemical study of isolated rat hepatocytes and monolayer cultures

A combination of biochemistry and morphology was used to demonstrate that more than 95 percent of the isolated rat hepatocytes prepared by collagenase dissociation of rat livers retained the pathway for receptor-mediated endocytosis of asialoglycoproteins (ASGPs). Maximal specific binding of (125)I-asialoorosomucoid ((125)I-ASOR) to dissociated hepatocytes at 5 degrees C (at which temperature no internalization occurred) averaged 100,000-400,000 molecules per cell. Binding, uptake, and degredation of (125)I- ASOR at 37 degrees C occurred at a rate of 1 x 10(6) molecules per cell over 2 h. Light and electron microscopic autoradiography (LM- and EM-ARG) of (125)I-ASOR were used to visualize the surface binding sites at 5 degrees C and the intracellular pathway at 37 degrees C. In the EM-ARG experiments, ARG grains corresponding to (125)I-ASOR were distributed randomly over the cell surface at 5 degrees C but over time at 37 degrees C were concentrated in the lysosome region. Cytochemical detection of an ASOR-horseradish peroxidase conjugate (ASOR-HRP) at the ultrastructural level revealed that at 5 degrees C this specific ASGP tracer was concentrated in pits at the cell surface as well as diffusely distributed along the rest of the plasma membrane. Such a result indicates that redistribution of ASGP surface receptors had occurred. Because the number of surface binding sites of (125)I-ASOR varied among cell preparations, the effect of collagenase on (125)I-ASOR binding was examined. When collagenase-dissociated hepatocytes were re-exposed to collagenase at 37 degrees C, 10-50 percent of control binding was observed. However, by measuring the extent of (125)I-ASOR binding at 5 degrees C in the same cell population before and after collagenase dissociation, little reduction in the number of ASGP surface receptors was found. Therefore, the possibility that the time and temperature of the cell isolations allowed recovery of cell surface receptors following collagenase exposure was tested. Freshly isolated cells, dissociated cells that were re-exposed to collagenase, and perfused livers exposed to collagenase without a Ca(++)-free pre-perfusion, were found to bind 110-240 percent more(125)I-ASOR after 1 h at 37 degrees C that they did at 0 time. This recovery of surface ASGP binding activity occurred in the absence of significant protein synthesis (i.e., basal medium or 1 mM cycloheximide). Suspensions of isolated, unpolarized hepatocytes were placed in monolayer culture for 24 h and confluent cells were demonstrated to reestablish morphologically distinct plasma membrane regions analogous to bile canalicular, lateral, and sinusoidal surfaces in vivo. More than 95 percent of these cells maintained the capacity to bind, internalize, and degrade (125)I-ASOR at levels comparable to those of the freshly isolated population. ASOR-HRP (at 5 degrees C) was specifically bound to all plasma membrane surfaces of repolarized hepatocytes (cultured for 24 h) except those lining bile canalicular-like spaces. Thus, both isolated, unpolarized hepatocytes and cells cultured under conditions that promote morphological reestablishment of polarity maintain the pathway for receptor- mediated endocytosis of ASGPs.     

Differences in thermotolerance induced by heat or sodium arsenite: correlation between redistribution of a 26-kDa protein and development of protein synthesis-independent thermotolerance in CHO cells

In previous studies, we have demonstrated the differences in thermotolerance induced by heat and sodium arsenite (Lee et al., Radiat. Res. 121, 295-303, 1990). In this study, we investigated whether a 26-kDa protein might play an important role in evincing these differences. Chinese hamster ovary (CHO) cells treated for either 1 h with 100 microM sodium arsenite (ARS) or 10 min at 45.5 degrees C became thermotolerant to a test heat treatment at 43 degrees C administered 6 or 12 h later, respectively. After the test heating at 43 degrees C for 1.5 h, the level of 26-kDa protein in the nucleus was decreased by 92% in nonthermotolerant cells, 78% in ARS-induced thermotolerant cells, and 3% in heat-induced thermotolerant cells. Inhibiting protein synthesis with cycloheximide (CHM, 10 micrograms/ml) after ARS treatment eliminated thermotolerance to 43 degrees C and delayed restoration of the 26-kDa protein in the nucleus. In contrast, CHM neither prevented the development of thermotolerance nor inhibited the restoration of the 26-kDa protein in heat-induced thermotolerant cells. However, when cells were exposed to cold (4 degrees C), immediately after initial heating, restoration of the 26-kDa protein and development of thermotolerance did not occur. These results demonstrate a good correlation between the restoration and/or the presence of this 26-kDa protein and the development of protein synthesis-independent thermotolerance.     

The src oncogene can regulate a human glucose transporter expressed in chicken embryo fibroblasts.

When fibroblasts are transformed by the src oncogene, there is a two- to fivefold increase in glucose transport and in the level of immunoprecipitable glucose transporter protein. In chicken embryo fibroblasts (CEFs), this increase is correlated with a comparable reduction in the rate at which the glucose transporter protein is turned over. In contrast, in mammalian fibroblasts glucose transporter biosynthesis is increased by src, but there is little or no change in its turnover. To further understand the action of src on transporter turnover, we investigated whether a mammalian transporter can be stabilized by src in a chicken cell environment. The human type 1 glucose transporter protein (hGT), originally cloned from HepG2 cells, was expressed in CEFs or Rat-1 fibroblasts by using a retroviral vector. In CEFs transformed by a temperature-sensitive src mutant, tsNY68, turnover of hGT was lower at the permissive temperature (36 degrees C) than at the nonpermissive temperature (42 degrees C). When this protein was expressed in CEFs transformed by wild-type src, no difference in turnover was observed at the two temperatures. In the case of Rat-1 cells transformed by the temperature-sensitive src mutant tsLA29, turnover of hGT was the same at the permissive temperature (35 degrees C) as at the nonpermissive temperature (39.5 degrees C). These data demonstrate that a heterologous glucose transporter behaves in the same way in chicken and rat cells as the respective endogenous transporter, i.e., when src is active, the protein is stablilized against turnover in chicken cells but not in rat cells.     

Effects of cycloheximide on thermotolerance expression, heat shock protein synthesis, and heat shock protein mRNA accumulation in rat fibroblasts.

A single hyperthermic exposure can render cells transiently resistant to subsequent high temperature stresses. Treatment of rat embryonic fibroblasts with cycloheximide for 6 h after a 20-min interval at 45 degrees C inhibits protein synthesis, including heat shock protein (hsp) synthesis, and results in an accumulation of hsp 70 mRNA, but has no effect on subsequent survival responses to 45 degrees C hyperthermia. hsp 70 mRNA levels decreased within 1 h after removal of cycloheximide but then appeared to stabilize during the next 2 h (3 h after drug removal and 9 h after heat shock). hsp 70 mRNA accumulation could be further increased by a second heat shock at 45 degrees C for 20 min 6 h after the first hyperthermic exposure in cycloheximide-treated cells. Both normal protein and hsp synthesis appeared increased during the 6-h interval after hyperthermia in cultures which received two exposures to 45 degrees C for 20 min compared with those which received only one treatment. No increased hsp synthesis was observed in cultures treated with cycloheximide, even though hsp 70 mRNA levels appeared elevated. These data indicate that, although heat shock induces the accumulation of hsp 70 mRNA in both normal and thermotolerant cells, neither general protein synthesis nor hsp synthesis is required during the interval between two hyperthermic stresses for Rat-1 cells to express either thermotolerance (survival resistance) or resistance to heat shock-induced inhibition of protein synthesis.     

Purification, characterization, and rapid inactivation of thermolabile ubiquitin-activating enzyme from the mammalian cell cycle mutant ts85

Conjugation of ubiquitin to certain proteins can trigger their degradation in the in vitro reticulocyte system. In order to determine whether ubiquitin conjugation serves as an intermediate step in the turnover of cellular proteins in vivo, it is necessary to isolate proteolytic intermediates, i.e. ubiquitin-protein adducts of specific cellular proteins. While the steady-state level of conjugates of rapidly turning over proteins is relatively high, that of long-lived proteins is presumably extremely low, and therefore undetectable. Therefore, mutant cell lines with conditionally altered function(s) of the ubiquitin system can serve as powerful tools in studying the degradation of stable cellular proteins. We have characterized a temperature sensitive cell cycle arrest mutant cell (ts85) with a thermolabile ubiquitin-activating enzyme (E1; Finley, D., Ciechanover, A., and Varshavsky, A. (1984) Cell 37, 43-55). Following incubation at the restrictive temperature (39.5 degrees C), these cells fail to degrade short-lived proteins (Ciechanover, A., Finley, D., and Varshavsky, A. (1984) Cell 37, 57-66). However, involvement of the ubiquitin system in the turnover of long-lived proteins has not been addressed in these cells. A slow rate of inactivation of E1 in vivo, and significant rate of cell death following long incubation periods at the restrictive temperature, make this question difficult to address experimentally. In the present study we show that incubation of the cells for 1 h at 43 degrees C leads to rapid inactivation of ubiquitin conjugation in the intact mutant cell. Following heat treatment, the cells can be incubated at 39.5 degrees C for at least 6 h in order to study the possible involvement of the system in the turnover of long-lived cellular proteins. The viability of the cells is excellent at the end of the incubation. Following extraction, we have shown that inactivation occurs much more rapidly in the cell lysate in vitro than in the intact cell (t1/2 of 10 min compared to 4 h at 39.5 degrees C). The enzyme from both the mutant cell and the wild-type cell was purified to homogeneity. The molecular mass of the native enzyme from both cells is approximately 220 kDa with a subunit molecular mass of about 108 kDa. The structure of the enzyme is therefore very similar to that purified from rabbit reticulocytes. At the permissive temperature, the enzymes from both cells catalyze ATP-PPi and ATP-AMP exchange in similar kinetics. However, at the high temperature, the mutated enzyme is at least 7-fold less stable than the wild-type enzyme.(ABSTRACT TRUNCATED AT 400 WORDS)     

Differential transport and integration into the nuclear lamina for lamins A, B, and C

Lamins A, B, and C are the major proteins of the mammalian nuclear lamina and have been well studied in BHK-21 cells. Using in vivo labelling, cell fractionation, and immunoprecipitation, we have found that lamins have different patterns of nuclear transport and solubility. Newly synthesized lamin A is translocated to the nucleus faster than lamin C or B. It is the most tightly bound lamin and cannot be extracted from the lamina by nonionic detergent or high-salt buffers. Lamins B and C migrate more slowly to the nucleus. Partitioning between cytoskeleton and detergent-soluble fractions shows that integration of lamins B and C is not completed before a 1-h chase. For lamin C this process is dependent upon protein synthesis and can be inhibited with cycloheximide. Even though lamins A and C are almost identical, lamin C is never firmly bound to the lamina and can be partially solubilized upon high-salt treatment.     

Programmed cell death by default in embryonic cells, fibroblasts, and cancer cells.

We recently proposed that most mammalian cells constitutively express all of the proteins required to undergo programmed cell death (PCD) and undergo PCD unless continuously signaled by other cells not to. Although some cells have been shown to work this way, the vast majority of cell types remain to be tested. Here we tested purified fibroblasts isolated from developing or adult rat sciatic nerve, a mixture of cell types isolated from normal or p53-null mouse embryos, an immortalized rat fibroblast cell line, and a number of cancer cell lines. We found the following: 1) All of these cells undergo PCD when cultured at low cell density in the absence of serum and exogenous signaling molecules but can be rescued by serum or specific growth factors, suggesting that they need extracellular signals to avoid PCD. (2) The mixed cell types dissociated from normal mouse embryos can only support one another's survival in culture if they are in aggregates, suggesting that cell survival in embryos may depend on short-range signals. (3) Some cancer cells secrete factors that support their own survival. (4) The survival requirements of a human leukemia cell line change when the cells differentiate. (5) All of the cells studied can undergo PCD in the presence of cycloheximide, suggesting that they constitutively express all of the protein components required to execute the death program.