A novel germline hyperactivating JAK2 mutation L604F.

Somatic JAK2 mutations are the main molecular cause of the vast majority of polycythemia vera (PV) cases. According to a recent structural model, the prevalent acquired V617F mutation improves the stability of the JAK2 dimer, thereby enhancing the constitutive JAK2 kinase activity. Germline JAK2 mutations usually do not largely alter JAK2 signaling, although they may modulate the impact of V617F. We found an unusual germline JAK2 mutation L604F in homozygous form in a young PV patient, along with a low allele burden JAK2 V617F mutation, and in her apparently healthy sister. Their father with a PV-like disease had L604F in a heterozygous state, without V617F. The functional consequences of JAK2 L604Fmutation were compared with those induced by V617F in two different in vitro model systems: (i) HEK293T cells were transfected with plasmids for exogenous JAK2-GFP expression, and (ii) endogenous JAK2 modifications were introduced into HeLa cells using CRISPR/Cas9. Both mutations significantly increased JAK2 constitutive activity in transfected HEK293T cells. In the second model, JAK2 modification resulted in reduced total JAK2 protein levels. An important difference was also detected: as described previously, the effect of V617F on JAK2 kinase activity was abrogated in the absence of the aromatic residue F595. In contrast, JAK2 hyperactivation by L604F was only partially inhibited by the F595 change to alanine. We propose that the L604F mutation increases the probability of spontaneous JAK2 dimer formation, which is physiologically mediated by F595. In addition, L604F may contribute to dimer stabilization similarly to V617F.

Expression of circular RNAs in myelodysplastic neoplasms and their association with mutations in the splicing factor gene SF3B1.

Mutations in the splicing factor 3b subunit 1 (SF3B1) gene are frequent in myelodysplastic neoplasms (MDS). Because the splicing process is involved in the production of circular RNAs (circRNAs), we investigated the impact of SF3B1 mutations on circRNA processing. Using RNA sequencing, we measured circRNA expression in CD34+ bone marrow MDS cells. We defined circRNAs deregulated in a heterogeneous group of MDS patients and described increased circRNA formation in higher-risk MDS. We showed that the presence of SF3B1 mutations did not affect the global production of circRNAs; however, deregulation of specific circRNAs was observed. Particularly, we demonstrated that strong upregulation of circRNAs processed from the zinc finger E-box binding homeobox 1 (ZEB1) transcription factor; this upregulation was exclusive to SF3B1-mutated patients and was not observed in those with mutations in other splicing factors or other recurrently mutated genes, or with other clinical variables. Furthermore, we focused on the most upregulated ZEB1-circRNA, hsa_circ_0000228, and, by its knockdown, we demonstrated that its expression is related to mitochondrial activity. Using microRNA analyses, we proposed miR-1248 as a direct target of hsa_circ_0000228. To conclude, we demonstrated that mutated SF3B1 leads to deregulation of ZEB1-circRNAs, potentially contributing to the defects in mitochondrial metabolism observed in SF3B1-mutated MDS.

Nucleolar phosphoprotein modifications as a marker of apoptosis induced by RITA treatment.

Reactivating p53 and Inducing Tumor Apoptosis (RITA) has been reported to increase the p53 activity and to trigger p53-dependent apoptosis in cancer cells with wild-type p53. Tumor suppressor p53 interacts with nucleolar phosphoproteins nucleophosmin (NPM) and nucleolin (NCL), which have crucial role in many cellular processes. Specific NPM mutations associated with acute myeloid leukemia (AML) cause aberrant localization of NPM and p53 in the cytoplasm with possible impact on the p53 function. We tested an effect of RITA on primary cells, and we found significant RITA-induced changes in NPM and NCL phosphorylation associated with apoptosis in cells of AML patients, but not that of healthy donors. Subsequent screening of several AML cell lines revealed heterogeneous response to RITA, and confirmed an association of the specific phosphorylation with apoptosis. While decreased NCL phosphorylation at Threonines T76 and T84 could be attributed to RITA-induced cell cycle arrest, enhanced NPM phosphorylation at Threonine T199 was not accompanied by the cell cycle changes and it correlated with sensitivity to RITA. Simultaneously, inverse changes occurred at Serine S4 of the NPM. These new findings of RITA mechanism of action could establish the NPM pT199/pS4 ratio as a marker for suitability of RITA treatment of AML cells.

Two-photon lifetime-based photoconversion of EGFP for 3D-photostimulation in FLIM.

Enhanced green fluorescence protein (EGFP) is a fluorescent tag commonly used in cellular and biomedical applications. Surprisingly, some interesting photochemical properties of EGFP have remained unexplored. Here we report on two-photon-induced photoconversion of EGFP, which can be permanently converted by intense IR irradiation to a form with a short fluorescence lifetime and spectrally conserved emission. Photoconverted EGFP thus can be distinguished from the unconverted tag by the time-resolved detection. Nonlinear dependence of the two-photon photoconversion efficiency on the light intensity allows for an accurate 3D localization of the photoconverted volume within cellular structures, which is especially useful for kinetic FLIM applications. For illustration, we used the two photon photoconversion of EGFP for measurements of redistribution kinetics of nucleophosmin and histone H2B in nuclei of live cells. Measurements revealed high mobility of fluorescently tagged histone H2B in the nucleoplasm and their redistribution between spatially separated nucleoli.

Cytoplasmic localization of Mdm2 in cells expressing mutated NPM is mediated by p53.

Specific C-terminal nucleophosmin (NPM) mutations are related to the acute myeloid leukaemia and cause mistargeting of mutated NPM (NPMmut) to the cytoplasm. Consequently, multiple NPM-interacting partners, e.g., the tumour suppressor p53, become also mislocalized. We found that ubiquitin ligase Mdm2 mislocalizes to the cytoplasm in the presence of NPMmut as well. Since p53 interacts with Mdm2, we searched for the NPMmut-p53-Mdm2 complex and interactions of its constituents in live cells and cell lysates using fluorescently tagged proteins, fluorescence lifetime imaging and immunoprecipitation. We proved existence of the ternary complex, which likely adopts a chain-like configuration. Interaction between Mdm2 and NPMmut was not detected, even under conditions of upregulated Mdm2 and p53 induced by Actinomycin D. We assume that p53 serves in the complex as a bridging link between Mdm2 and NPMmut. This conclusion was supported by disruption of the Mdm2-p53 interaction by Nutlin-3A, which resulted in relocalization of Mdm2 to the nucleus, while both NPMmut and p53 remained in the cytoplasm. Importantly, silencing of p53 also prevented mislocalization of Mdm2 in the presence of NPMmut.

AML-Related NPM Mutations Drive p53 Delocalization into the Cytoplasm with Possible Impact on p53-Dependent Stress Response.

Nucleophosmin (NPM) interaction with tumor suppressor p53 is a part of a complex interaction network and considerably affects cellular stress response. The impact of NPM1 mutations on its interaction with p53 has not been investigated yet, although consequences of NPMmut-induced p53 export to the cytoplasm are important for understanding the oncogenic potential of these mutations. We investigated p53-NPM interaction in live HEK-293T cells by FLIM-FRET and in cell lysates by immunoprecipitation. eGFP lifetime-photoconversion was used to follow redistribution dynamics of NPMmut and p53 in Selinexor-treated cells. We confirmed the p53-NPMwt interaction in intact cells and newly documented that this interaction is not compromised by the NPM mutation causing displacement of p53 to the cytoplasm. Moreover, the interaction was not abolished for non-oligomerizing NPM variants with truncated oligomerization domain, suggesting that oligomerization is not essential for interaction of NPM forms with p53. Inhibition of the nuclear exporter XPO1 by Selinexor caused expected nuclear relocalization of both NPMmut and p53. However, significantly different return rates of these proteins indicate nontrivial mechanism of p53 and NPMmut cellular trafficking. The altered p53 regulation in cells expressing NPMmut offers improved understanding to help investigational strategies targeting these mutations.

Dominant monoallelic variant in the PAK2 gene causes Knobloch syndrome type 2.

Knobloch syndrome is an autosomal recessive phenotype mainly characterized by retinal detachment and encephalocele caused by biallelic pathogenic variants in the COL18A1 gene. However, there are patients clinically diagnosed as Knobloch syndrome with unknown molecular etiology not linked to COL18A1. We studied an historical pedigree (published in 1998) designated as KNO2 (Knobloch type 2 syndrome with intellectual disability, autistic behavior, retinal degeneration, encephalocele). Whole exome sequencing of the two affected siblings and the normal parents resulted in the identification of a PAK2 non-synonymous substitution p.(Glu435Lys) as a causative variant. The variant was monoallelic and apparently de novo in both siblings indicating a likely germ-line mosaicism in one of the parents; the mosaicism, however, could not be observed after deep sequencing of blood parental DNA. PAK2 encodes a member of a small group of serine/threonine kinases; these P21-activating kinases (PAKs) are essential in signal transduction and cellular regulation (cytoskeletal dynamics, cell motility, death and survival signaling and cell cycle progression). Structural analysis of the PAK2 p.(Glu435Lys) variant that is located in the kinase domain of the protein predicts a possible compromise in the kinase activity. Functional analysis of the p.(Glu435Lys) PAK2 variant in transfected HEK293T cells results in a partial loss of the kinase activity. PAK2 has been previously suggested as an autism-related gene. Our results show that PAK2-induced phenotypic spectrum is broad and not fully understood. We conclude that the KNO2 syndrome in the studied family is dominant and caused by a deleterious variant in the PAK2 gene.

NSC348884 cytotoxicity is not mediated by inhibition of nucleophosmin oligomerization.

Nucleophosmin (NPM) mutations causing its export from the nucleoli to the cytoplasm are frequent in acute myeloid leukemia (AML). Due to heterooligomerization of wild type NPM with the AML-related mutant, the wild-type becomes misplaced from the nucleoli and its functions are significantly altered. Dissociation of NPM heterooligomers may thus restore the proper localization and function of wild-type NPM. NSC348884 is supposed to act as a potent inhibitor of NPM oligomerization. The effect of NSC348884 on the NPM oligomerization was thoroughly examined by fluorescence lifetime imaging with utilization of FRET and by a set of immunoprecipitation and electrophoretic methods. Leukemia-derived cell lines and primary AML cells as well as cells transfected with fluorescently labeled NPM forms were investigated. Our results clearly demonstrate that NSC348884 does not inhibit formation of NPM oligomers neither in vivo nor in vitro. Instead, we document that NSC348884 cytotoxicity is rather associated with modified cell adhesion signaling. The cytotoxic mechanism of NSC348884 has therefore to be reconsidered.

Group I p21-activated kinases in leukemia cell adhesion to fibronectin.

P21-activated kinases (PAK) regulate processes associated with cytoskeleton dynamics. PAK expression in leukemia cells was measured on protein and mRNA levels. In functional assays, we analyzed the effect of PAK inhibitors IPA-3 and FRAX597 on cell adhesivity and viability. PAK2 was dominant in cell lines, whereas primary cells also expressed comparable amount of PAK1 transcription isoforms: PAK1-full and PAK1Δ15. PAK1Δ15 and PAK2 levels correlated with surface density of integrins ß1 and αVß3. PAK1-full, but not PAK2, was present in membrane protrusions. IPA-3, which prevents PAK activation, induced cell contraction in semi-adherent HEL cells only. FRAX597, which inhibits PAK kinase activity, increased cell-surface contact area in all leukemia cells. Both inhibitors reduced the stability of cell attachment and induced cell death.

PAK1, PAK1Δ15, and PAK2: similarities, differences and mutual interactions.

P21-activated kinases (PAK) are key effectors of the small GTPases Rac1 and Cdc42, as well as of Src family kinases. In particular, PAK1 has several well-documented roles, both kinase-dependent and kinase-independent, in cancer-related processes, such as cell proliferation, adhesion, and migration. However, PAK1 properties and functions have not been attributed to individual PAK1 isoforms: besides the full-length kinase (PAK1-full), a splicing variant lacking the exon 15 (PAK1Δ15) is annotated in protein databases. In addition, it is not clear if PAK1 and PAK2 are functionally overlapping. Using fluorescently tagged forms of human PAK1-full, PAK1Δ15, and PAK2, we analyzed their intracellular localization and mutual interactions. Effects of PAK inhibition (IPA-3, FRAX597) or depletion (siRNA) on cell-surface adhesion were monitored by real-time microimpedance measurement. Both PAK1Δ15 and PAK2, but not PAK1-full, were enriched in focal adhesions, indicating that the C-terminus might be important for PAK intracellular localization. Using coimmunoprecipitation, we documented direct interactions among the studied PAK group I members: PAK1 and PAK2 form homodimers, but all possible heterocomplexes were also detected. Interaction of PAK1Δ15 or PAK2 with PAK1-full was associated with extensive PAK1Δ15/PAK2 cleavage. The impedance measurements indicate, that PAK2 depletion slows down cell attachment to a surface, and that PAK1-full is involved in cell spreading. Altogether, our data suggest a complex interplay among different PAK group I members, which have non-redundant functions.

Lifetime-based photoconversion of EGFP as a tool for FLIM.

BACKGROUND: EGFP is a fluorescent tag extensively used in biological and biomedical research. Over the years many researches have gathered collections of cell lines bearing specific EGFP-tagged proteins. Despite its popularity some photochemical properties of EGFP remain undocumented and unused. We report on so far unexplored lifetime photoconversion of EGFP usable in FLIM. METHODS: Fluorescence lifetime imaging and spectral FLIM has been used for characterization of the EGFP photoconversion and protein tracking. RESULT: Our data suggest that EGFP can be permanently photoconverted to a short-fluorescence-lifetime form (PC-EGFP) by intense blue irradiation. PC-EGFP cannot be reverted back by 405 nm light and exhibits the same spectral emission properties with blue-shifted absorption compared to the unconverted EGFP. Fluorescence of PC-EGFP is pH-independent and the photoconversion efficiency decreases with the solvent viscosity. Utilization of the EGFP photoconversion was demonstrated by tracking of a nucleophosmin mutant in live HEK-293 T cells during its cytoplasm-nuclear relocalization induced by Leptomycin B. CONCLUSIONS: Besides potential FLIM artifacts caused by an unintended EGFP photoconversion, the controlled photoconversion turns EGFP to an excellent tool for kinetic FLIM applications. Since the photoconversion occurs in the lifetime domain, PC-EGFP can be easily distinguished from the unconverted tag by time-resolved detection while all other spectral channels stay free for multicolor labeling. GENERAL SIGNIFICANCE: The reported lifetime photoconversion lines up EGFP with other photoconvertible fluorescent proteins with special advantage for fluorescence lifetime imaging where lifetime-photoconvertible labels are scarce.

AML-associated mutation of nucleophosmin compromises its interaction with nucleolin.

C-terminal mutations of the nucleolar protein nucleophosmin (NPM) are the most frequent genetic aberration detected in acute myeloid leukemia (AML) with normal karyotype. The mutations cause aberrant cytoplasmic localization of NPM and lead to loss of functions associated with NPM nucleolar localization, e.g. in ribosome biogenesis or DNA-damage repair. NPM has many interaction partners and some of them were proved to interact also with the mutated form (NPMmut) and due to this interaction thereby to be withdrawn from their site of action. We analyzed the impact of the mutation on NPM interaction with nucleolin (NCL) which is also prevalently localized into the nucleolus and cooperates with wild-type NPM (NPMwt) in many cellular processes. We revealed that the NCL-NPM complex formation is completely abolished by the mutation and that the presence/absence of the interaction is not affected by drugs causing genotoxic stress or differentiation. Deregulation resulting from changes of NCL/NPMwt ratio may contribute to leukemogenesis.

Monitoring of nucleophosmin oligomerization in live cells.

Oligomerization plays a crucial role in the function of nucleophosmin (NPM), an abundant nucleolar phosphoprotein. Two dual-color methods based on modern fluorescence confocal microscopy are applied for tracking NPM aggregates in live cells: cross-correlation Number and Brightness analysis (ccN&B) combined with pulsed interleaved excitation (PIE) and fluorescence-lifetime imaging microscopy (FLIM) utilizing resonance energy transfer (FRET). HEK-293T cells were transfected with mixture of plasmids designed for tagging with fluorescent proteins so that the cells express mixed population of NPM labeled either with eGFP or mRFP1. We observe joint oligomers formed from the fluorescently labeled NPM. Having validated the in vivo methods, we study an effect of substitutions in cysteine 21 (Cys21) of the NPM N-terminus on the oligomerization to demonstrate applicability of the methods. Inhibitory effect of mutations of the Cys21 to nonpolar Ala or to aromatic Phe on the oligomerization was reported in literature using in vitro semi-native electrophoresis. However, we do not detect any break-up of the joint NPM oligomers due to the Cys21 mutations in live cells. In vivo microscopy observations are supported by an in vitro method, the GFP-Trap immunoprecipitation assay. Our results therefore show importance of utilizing several methods for detection of biologically relevant protein aggregates. In vivo monitoring of the NPM oligomerization, a potential cancer therapy target, by the presented methods offers a new way to monitor effects of drugs that are tested as NPM oligomerization inhibitors directly in live cells.

Adhesion structures in leukemia cells and their regulation by Src family kinases.

Interaction of leukemia blasts with the bone marrow extracellular matrix often results in protection of leukemia cells from chemotherapy and in persistence of the residual disease which is on the basis of subsequent relapses. The adhesion signaling pathways have been extensively studied in adherent cells as well as in mature haematopoietic cells, but the adhesion structures and signaling in haematopoietic stem and progenitor cells, either normal or malignant, are much less explored. We analyzed the interaction of leukemia cells with fibronectin (FN) using interference reflection microscopy, immunofluorescence, measurement of adherent cell fraction, real-time microimpedance measurement and live cell imaging. We found that leukemia cells form very dynamic adhesion structures similar to early stages of focal adhesions. In contrast to adherent cells, where Src family kinases (SFK) belong to important regulators of focal adhesion dynamics, we observed only minor effects of SFK inhibitor dasatinib on leukemia cell binding to FN. The relatively weak involvement of SFK in adhesion structure regulation might be associated with the lack of cytoskeletal mechanical tension in leukemia cells. On the other hand, active Lyn kinase was found to specifically localize to leukemia cell adhesion structures and a less firm cell attachment to FN was often associated with higher Lyn activity (this unexpectedly occurred also after cell treatment with the inhibitor SKI-1). Lyn thus may be important for signaling from integrin-associated complexes to other processes in leukemia cells.

Localization of AML-related nucleophosmin mutant depends on its subtype and is highly affected by its interaction with wild-type NPM.

Mutations of the gene for nucleophosmin (NPM1) are the most frequent genetic aberration in patients with acute myeloid leukemia (AML). The mechanism of leukemic transformation in this leukemia subtype is not fully understood, but aberrant cytoplasmic localization of mutated NPM (NPMmut) is widely considered as an important factor for leukemia manifestation. We analyzed the subcellular localization of three types of NPM with a C-terminal mutation (A, B and E). Genes for the individual NPM forms were fused with a gene for one of fluorescent protein variants in plasmids, which were transfected into three cell lines with different endogenous NPM expression. Subcellular localization of the fluorescent protein-labeled NPM was further correlated with the relative expression of all NPM forms. We confirmed a high cytoplasmic expression of NPMmutA and NPMmutB whereas a substantial fraction of NPMmutE was found to be localized in nucleoli. Moreover, we revealed that the localization of fluorescently labeled NPM is affected by the interaction between various forms of the protein.

Low-Dose Actinomycin-D Induces Redistribution of Wild-Type and Mutated Nucleophosmin Followed by Cell Death in Leukemic Cells.

Specific mutations involving C-terminal part of the nucleolar protein nucleophosmin (NPM) are associated with better outcome of acute myeloid leukemia (AML) therapy, possibly due to aberrant cytoplasmic NPM localization facilitating induction of anti-NPM immune response. Actinomycin D (actD) is known to induce nucleolar stress leading to redistribution of many nucleolar proteins, including NPM. We analyzed the distribution of both wild-type and mutated NPM (NPMmut) in human cell lines, before and after low-dose actD treatment, in living cells expressing exogenous fluorescently labeled proteins as well as using immunofluorescence staining of endogenous proteins in fixed cells. The wild-type NPM form is prevalently nucleolar in intact cells and relocalizes mainly to the nucleoplasm following actD addition. The mutated NPM form is found both in the nucleoli and in the cytoplasm of untreated cells. ActD treatment leads to a marked increase in NPMmut amount in the nucleoplasm while a mild decrease is observed in the cytoplasm. Cell death was induced by low-dose actD in all the studied leukemic cell lines with different p53 and NPM status. In cells expressing the tumor suppresor p53 (CML-T1, OCI-AML3), cell cycle arrest in G1/G0 phase was followed by p53-dependent apoptosis while in p53-null HL60 cells, transient G2/M-phase arrest was followed by cell necrosis. We conclude that although actD does not increase NPM concentration in the cytoplasm, it could improve the effect of standard chemotherapy in leukemias through more general mechanisms.

Decitabine and SAHA-induced apoptosis is accompanied by survivin downregulation and potentiated by ATRA in p53-deficient cells.

While p53-dependent apoptosis is triggered by combination of methyltransferase inhibitor decitabine (DAC) and histone deacetylase inhibitor suberoylanilide hydroxamic acid (SAHA) in leukemic cell line CML-T1, reactive oxygen species (ROS) generation as well as survivin and Bcl-2 deregulation participated in DAC + SAHA-induced apoptosis in p53-deficient HL-60 cell line. Moreover, decrease of survivin expression level is accompanied by its delocalization from centromere-related position in mitotic cells suggesting that both antiapoptotic and cell cycle regulation roles of survivin are affected by DAC + SAHA action. Addition of subtoxic concentration of all-trans-retinoic acid (ATRA) increases the efficiency of DAC + SAHA combination on viability, apoptosis induction, and ROS generation in HL-60 cells but has no effect in CML-T1 cell line. Peripheral blood lymphocytes from healthy donors showed no damage induced by DAC + SAHA + ATRA combination. Therefore, combination of ATRA with DAC and SAHA represents promising tool for therapy of leukemic disease with nonfunctional p53 signalization.

Group I PAK inhibitor IPA-3 induces cell death and affects cell adhesivity to fibronectin in human hematopoietic cells.

P21-activated kinases (PAKs) are involved in the regulation of multiple processes including cell proliferation, adhesion and migration. However, the current knowledge about their function is mainly based on results obtained in adherent cell types. We investigated the effect of group I PAK inhibition using the compound IPA-3 in a variety of human leukemic cell lines (JURL-MK1, MOLM-7, K562, CML-T1, HL-60, Karpas-299, Jurkat, HEL) as well as in primary blood cells. IPA-3 induced cell death with EC50 ranging from 5 to more than 20 µM. Similar range was found for IPA-3-mediated dephosphorylation of a known PAK downstream effector, cofilin. The cell death was associated with caspase-3 activation, PARP cleavage and apoptotic DNA fragmentation. In parallel, 20 µM IPA-3 treatment induced rapid and marked decrease of the cell adhesivity to fibronectin. Per contra, partial reduction of PAK activity using lower dose IPA-3 or siRNA resulted in a slight increase in the cell adhesivity. The changes in the cell adhesivity were also studied using real-time microimpedance measurement and by interference reflection microscopy. Significant differences in the intracellular IPA-3 level among various cell lines were observed indicating that an active mechanism is involved in IPA-3 transport.

Combined treatment with low concentrations of decitabine and SAHA causes cell death in leukemic cell lines but not in normal peripheral blood lymphocytes.

Epigenetic therapy reverting aberrant acetylation or methylation offers the possibility to target preferentially tumor cells and to preserve normal cells. Combination epigenetic therapy may further improve the effect of individual drugs. We investigated combined action of demethylating agent decitabine and histone deacetylase inhibitor SAHA (Vorinostat) on different leukemic cell lines in comparison with peripheral blood lymphocytes. Large decrease of viability, as well as huge p21WAF1 induction, reactive oxygen species formation, and apoptotic features due to combined decitabine and SAHA action were detected in leukemic cell lines irrespective of their p53 status, while essentially no effect was observed in response to the combined drug action in normal peripheral blood lymphocytes of healthy donors. p53-dependent apoptotic pathway was demonstrated to participate in the wtp53 CML-T1 leukemic cell line response, while significant influence of reactive oxygen species on viability decrease has been detected in p53-null HL-60 cell line.

Live fluorescence and transmission-through-dye microscopic study of actinomycin D-induced apoptosis and apoptotic volume decrease.

The effect of actinomycin D on HeLa cells was studied by live fluorescence and transmission-through-dye microscopy-a recently developed technique that permits volume measurements in live cells. In particular, it is well suited for the observation and quantification of the apoptotic volume decrease (AVD), which is widely viewed as an essential feature of apoptosis. The main results from our study are as follows. (1) Apoptosis caused in HeLa cells by actinomycin D proceeds in two morphologically distinct stages: the early stage is characterized by extensive blebbing, and the late stage by a more compact shape. The loss of mitochondrial membrane potential occurs at about the same time as blebbing, and chromatin condensation follows 30-90 min later. Caspase-3 and 7 become activated during the late stage. (2) Because blebbing occurs before activation of caspase-3, it has to be initiated by a different mechanism. Although blebbing is one of the earliest observable changes, it can be selectively inhibited without affecting other apoptotic reactions. (3) The majority of cells experience a temporary volume increase after the appearance of blebs. Eventually, AVD takes over and the cells shrink by approximately 40 % of their initial volume; the volume loss becomes noticeable at the end of the blebbing phase and continues through the late stage. Sometimes, at the end of long incubations, shrinkage gives way to swelling, possibly indicating secondary necrosis. (4) Both early and late apoptosis are accompanied by intracellular accumulation of Na(+), while low-sodium medium prevents apoptosis. Except for a partial protective effect of quinine, all of the tested blockers of Na(+), K(+) and Cl(-) channels failed to prevent apoptosis or AVD.