Roger Mosselmans

Comparison of methods based on annexin‐V binding, DNA content or TUNEL for evaluating cell death in HL‐60 and adherent MCF‐7 cells

Abstract. HL‐60 and MCF‐7 cells were treated with 0.15 μM camptothecin (CPT) or with the solvent dimethylsulfoxide (DMSO) for the controls, for 2, 3 and 4 h or for 24, 48 and 72 h, respectively. The apoptotic index (AI) was then evaluated in parallel by the following flow cytometric methods: (1) double staining of unfixed cells with fluoresceinated annexin V and propidium iodide (PI), this after detachment by trypsinization in the case of MCF‐7 cultures; (2) prefixation in 70% ethanol, extraction of degraded, low molecular weight DNA with 0.2 M phosphatecitrate buffer and analysis of the DNA content stained with PI; (3) TUNEL, i.e. labelling of DNA strand breaks with biotin‐dUTP, followed by standing with streptavidin‐fluorescein and counterstaining with PI. In HL‐60 cells, the three methods gave similar results for the AI (3‐4% in the controls and at 2 h of CPT treatment, and 35‐43% at 3 and 4 h after CPT). This indicates that CPT‐induced membrane alteration and DNA fragmentation occurred concomitantly in those cells. For MCF‐7 cells, CPT‐induced apoptosis developed more slowly, the AI, whether based on annexin V or on DNA content, remained unchanged at 24 h, then was increasing to 8% at 48 h and to 25% at 72 h of treatment. In these cells, the TUNEL index did not increase prior to 72 h and the increase was minor (up to 9% vs. 2‐3% in the controls) at 72 h of the treatment. This indicates that in MCF‐7 cells DNA strand breaks cannot be effectively labelled, which may be due to inaccessibility of 3′‐OH ends in the breaks to exogenous terminal deoxynucleotidyl transferase. The mechanism of endonucleolytic DNA fragmentation thus may be different, depending on the cell type.

Antisense inhibition of the 27 kDa heat shock protein production affects growth rate and cytoskeletal organization in MCF-7 cells.

MCF‐7 cells were co‐transfected with the human HSP27 antisense cDNA and the neomycin resistance gene, included in the constitutive expression vector pSVL, and the phenotypical changes associated with decreased expression of the HSP27 protein were analysed. Three out of 10 neomycin‐resistant clones obtained proliferated normally and showed a normal HSP27 content (Western blot). The seven other clones (designated as αHSP27 clones) were characterized by a dramatic growth inhibition associated with alterations in cellular morphology. Cells became progressively hypertrophied, exhibited lamellar protrusions and tended to lose contact with each other. They also acquired characteristics of secretory cells, namely the presence of numerous refractile granules and secretory canaliculi. Among the αHSP27 clones, two were immunocytochemically analysed for HSP27 content. Both clones were immunonegative for HSP27, contrary to parental cells and neo transfectants. Actin immunostaining in one of these HSP27 negative clones revealed that microfilament organization changed from diffuse to punctate distribution. Our data support the current concept of a role for HSP27 in cell growth and differentiation and further suggests that this might occur through a control on actin polymerization‐depolymerization.

Actin depolymerization and polymerization are required during apoptosis in endothelial cells

In order to understand the role of actin microfilaments in the apoptotic process, we followed their evolution during tumor necrosis factor‐α (TNF)‐induced apoptosis in bovine aortic endothelial (BAE) cells. Using Western blotting analysis and immunofluorescence microscopy, we observed that the actin microfilaments network was disrupted in apoptotic cells. Depolymerization of F‐actin was concomitant with internucleosomal DNA degradation and with the morphological changes associated with apoptotic cell death. However, using the actin microfilament disrupting agent, cytochalasin, we present evidence that the formation of blebs leading to apoptotic cell fragmentation requires neopolymerization of actin. Indeed, in the presence of cyochalasin, induction of apoptosis (internucleosomal DNA degradation) in BAE cells by TNF and cycloheximide was not associated with these classical morphological markers of apoptosis. Moreover, when added to BAE cells showing incipient apoptotic fragmentation, cytochalasin E reversed this process. We also observed an accumulation of actin at the basis of the apoptotic bodies in formation in these cells. Together, these results suggest that the actin network of flattened cells is disrupted concomitantly to the morphological modifications associated to the apoptotic cell death, and that the cytochalasin‐sensitive reorganisation of actin is required to the formation of apoptotic blebs. J. Cell. Physiol. 184:239–245, 2000.

Anti-sense inhibition of small-heat-shock-protein (HSP27) expression in MCF-7 mammary-carcinoma cells induces their spontaneous acquisition of a secretory phenotype.

This work was aimed at testing the hypothesis (hitherto supported only by indirect evidence) that, besides contributing to resistance to stress, the small heat‐shock‐protein HSP27 might be involved in the control of growth and differentiation in mammary‐tumour cells, where it is known to be oestrogen‐regulated. Therefore, MCF‐7 cells were transfected with a modulatable human hsp27 anti‐sense cDNA. Clones of transfectants (designated αhsp27) were selected which, upon expression of the anti‐sense, exhibited a decline in HSP27 accumulation, associated with a decrease in resistance to heat shock and in proliferation rate, the degree of the latter reflecting their respective reduction in HSP27 content. The effects of anti‐sense inhibition of HSP27 production were similar to those exerted on parental cells by phorbol myristate (TPA). Both resulted in growth inhibition, accumulation of lipid droplets in the cytoplasm, formation of secretory microvesicles with internal microvilli and increased release of several proteins, including the isoforms of a 52‐kDa protein, which we identified as the oestrogen‐regulated protein cathepsin D, all this without noticeable change in actin organization. These data constitute the first direct support for the hypothesis that, at least in some cell types, HSP27 might play a modulatory role in cell differentiation and (perhaps by this) in proliferation. While allowing dissociation of this role from the known action of HSP27 on actin polymerization, they suggest similar modulation of the function of some protein(s) implicated in the acquisition of the secretory phenotype by MCF‐7 cells, with HSP27 also exerting an inhibitory action that can be alleviated either by its phosphorylation (as occurs with TPA) or by inhibition of its production. Int. J. Cancer 82:574–582, 1999.

Expression of HSP27 results in increased sensitivity to tumor necrosis factor, etoposide, and H2O2 in an oxidative stress‐resistant cell line

The role of HSP27 in cell growth and resistance to stress was investigated using murine fibrosarcoma L929 cells (normally devoid of constitutively expressed small HSPs) and human osteoblast‐like SaOS‐2 cells stably transfected with a human hsp27 expression vector. Our data showed that our L929 cells were more resistant to oxidative stress than generally observed for this line. Production of HSP27 in these cells led to a marked decrease in growth rate associated with a series of phenotypical changes, including cell spreading, cellular and nuclear hypertrophy, development of an irregular outline, and a tremendous accumulation of actin stress fibers. By contrast, none of these changes was observable in SaOS‐2/hsp27 transfectants overexpressing the protein product. Together, these observations are consistent with a cause‐to‐effect cascade relationship between increased (or induced) HSP27 expression, changes in cytoskeletal organization, and decreased growth. On the other hand, whereas the transfection of the hsp27 gene increased the cell resistance to heat in both cell lines, only in SaOS‐2 cells was this associated with protection to the cytotoxic action of tumor necrosis factor‐alpha (TNF‐α) and etoposide. Unexpectedly, L929/hsp27 transfectants exhibited an increased sensitivity to both agents and also to H2O2. These data thus imply that different mechanisms are involved in the cell resistance to heat shock and to the cytotoxic action of TNF‐α, etoposide, and H2O2. They also plead against the simple view that overexpression of a phosphorylatable HSP27 would necessarily be beneficial in terms of increased cell resistance to any type of stress. Our data further indicate that the role of HSP27 in cellular resistance to stress and in cell proliferation involves different targets and that the ultimate result of its interference with these processes depends on the intracellular context in which the protein is expressed. J Cell Physiol 177:606–617, 1998.

CHANGES IN THE PHOSPHORYLATION STATUS OF THE 27 KDA HEAT SHOCK PROTEIN (HSP27) ASSOCIATED WITH THE MODULATION OF GROWTH AND/OR DIFFERENTIATION IN MCF- 7 CELLS

We have used human mammary cells of the MCF‐7 strain, which constitutively express high levels of the small heat shock protein HSP27 and we have compared the changes in the phosphorylation status of this protein together with changes in cell growth and/or morphology induced by the action of one of the following agents: (1) TPA (12‐O‐tetradecanoylphorbol‐13‐acetate), known as a differentiation inducer in MCF‐7 cells; (2) OH‐TAM (hydroxytamoxifen), which exerts a cytostatic and cytotoxic action; or (3) TNFα (tumour necrosis factor), which induces apoptotic cell death in this cell line. Our data show that TPA and TNF stimulate an immediate and massive phosphorylation of HSP27, whereas OH‐TAM affect the phosphorylation status of the protein only after a 3 day delay. In the case of TPA, high levels of HSP27 phosphorylation were maintained for at least 4 days, along with growth inhibition and acquisition by the cells of a secretory phenotype. TPA and OH‐TAM exerted similar immediated effects on cell growth, despite the different time course of their action on HSP27 phosphorylation. This excludes the possibility that the latter is a necessary consequence of, or an absolute requisite to, growth inhibition. With OH‐TAM and TNF the increase in HSP27 phosphorylation was concomitant with the appearance of apoptosis, not observed with TPA. This indicates that increased phosphorylation of HSP27 is not specifically associated with the triggering or the execution of apoptosis in these cells. Altogether, our data support the concept that phosphorylated HSP27 is involved (and might then be rate limiting in some instances) in the execution of vital cell programmes (including resistance to stress, proliferation and differentiation), as well as in that of cell death. This is consistent with its role in actin polymerization and its position downstream of the p38/RK‐type MAPkinase, itself a point of convergence for diverse signal transduction pathways.

Retinal pigment epithelial cells phagocytosis of T lymphocytes: possible implication in the immune privilege of the eye.

Aim: To investigate the capability of retinal pigment epithelium (RPE) cells to phagocytose T lymphocytes and to further analyse the immunobiological consequences of this phagocytosis. Methods: Human RPE cells pretreated or not by cytochalasin, a phagocytosis inhibitor, were co-cultured with T lymphocytes for different time points. Phagocytosis was investigated by optic microscopy, electron microscopy, and flow cytometry. T cell proliferation was measured by 3H thymidine incorporation. RPE interleukin 1β mRNA expression was quantified by real time PCR. Results: RPE cells phagocytose apoptotic and non-apoptotic T lymphocytes, in a time dependent manner. This is an active process mediated through actin polymerisation, blocked by cytochalasin E treatment. Inhibition of RPE cell phagocytosis capabilities within RPE-T cell co-cultures led to an increase of lectin induced T cell proliferation and an upregulation of interleukin 1β mRNA expression in RPE cells. Conclusions: It is postulated that T lymphocyte phagocytosis by RPE cells might, by decreasing the total number of T lymphocytes, removing apoptotic lymphocytes, and downregulating the expression of IL-1β, participate in vivo in the induction and maintenance of the immune privilege of the eye, preventing the development of intraocular inflammation.

Iron induces Bcl-2 expression in human dermal microvascular endothelial cells

Iron is suspected to be involved in the induction and/or progression of various human tumors. The present study was designed to investigate the effects of iron on endothelial cells, keeping in mind that the homeostasis of microvessels plays a critical role in neo-angiogenesis. Applying a model of human dermal microvascular endothelial cell terminal differentiation and death induced by serum deprivation, we found that iron salts (iron chloride and ferric nitrilotriacetate) provided a survival advantage to endothelial cells. Using immunohistochemistry and Western Blot analysis, we found that the extended cellular life span induced by iron was paralleled by an increase of Bcl-2 protein expression. Taken together, these observations suggest that iron may give a survival advantage to endothelial cells and represent a novel mechanism through which iron may contribute to tumorigenesis.

Effect of the serine protease inhibitor N-tosyl-L-phenylalanine-chloromethyl ketone (TPCK) on MCF-7 mammary tumour cells growth and differentiation

Previous studies from this and other laboratories indicated that the oestrogen‐regulated heat shock protein HSP27 is involved in the control of MCF‐7 cells growth and differentiation, as it also appears to be in other cell types, including osteoblasts and HL‐60 cells. In the latter instance, induction of differentiation is associated with the downregulation of myeloblastin, a serine protease now identified as proteinase 3 (hence its designation as PR3/Mbn), mirrored by an increase in the cellular content of the small heat shock protein HSP27, a substrate to this enzyme. Besides, antisense inhibition of PR3/Mbn production sufficed for inducing HL‐60 cells monocytic differentiation. This prompted us to examine the hypothesis that a post‐translational control on HSP27 levels (and by this on differentiation) by a serine protease might also be operating in human mammary tumour cells. As part of our attempt to evaluate this hypothesis, the present work consisted of testing the effects of a treatment of MCF‐7 cells with the serine protease inhibitor N‐tosyl‐L‐phenylalanine‐chloromethyl ketone (TPCK). Our data show that this resulted in a four‐fold increase in HSP27 content, associated with a 2.5‐fold decrease in growth rate, the formation of cytoplasmic vesicles and increased secretion of 52kDa peptides, identified by Western immunoblot as the isoforms of the oestrogen‐regulated protein, cathepsin D. TPCK only affected growth in MDAMB‐231 cells (in which HSP27 levels are very low and remained below MCF‐7 cells basal levels after treatment) and failed to affect L929 cells, in which the hsp27 gene is silent. This provides circumstantial support for the assumption that effects of TPCK on the MCF‐7 cells phenotype are linked to the associated increase in HSP27 content. Our recent demonstration that MCF‐7 cells do in fact express PR3/Mbn fits with our concept and opens the way to test it directly, using antisense strategy.

Morphology of isolated and cultured hepatocytes from normal liver and in hepatocarcinogenesis

In a wide variety of experimental hepatocarcinogenic models, the administration of chemical carcinogens induces the same sequence of morphological and functional alterations preceding tumour development (see review in reference 1).