Automatic segmentation of lysosomes and analysis of intracellular pH with Radachlorin photosensitizer and FLIM.

We report application of the fluorescence lifetime imaging microscopy (FLIM) for analysis of distributions of intracellular acidity using a chlorin-e6 based photosensitizer Radachlorin. An almost two-fold increase of the photosensitizer fluorescence lifetime in alkaline microenvironments as compared to acidic ones allowed for clear distinguishing between acidic and alkaline intracellular structures. Clusterization of a phasor plot calculated from fits of the FLIM raw data by two Gaussian distributions provided accurate automatic segmentation of lysosomes featuring acidic contents. The approach was validated in colocalization experiments with LysoTracker fluorescence in living cells of four established lines. The dependence of photosensitizer fluorescence lifetime on microenvironment acidity allowed for estimation of pH inside the cells, except for the nuclei, where photosensitizer does not penetrate. The developed method is promising for combined application of the photosensitizer for both photodynamic treatment and diagnostics.

Analysis of Radachlorin localization in living cells by fluorescence lifetime imaging microscopy.

Intracellular localization of photosensitizer molecules is influential on cell death pathway at photodynamic treatment and is thus an important aspect in achieving enhanced efficacy of photodynamic therapy. In this paper we performed thorough studies of the distribution of Radachlorin photosensitizer in three established cell lines: HeLa, A549, and 3T3 with fluorescence lifetime imaging microscopy through the analysis of lifetime distributions. Experiments carried out in Radachlorin solutions in phosphate buffered saline revealed the pronounced dependence of the fluorescence quantum yield and lifetime on solution pH. This finding was used for analysis of lifetime images of living cells and their phasor plot representations and allowed us to suggest that Radachlorin localized predominantly in lysosomes, known to have acidic pH values. Experiments on co-localization of Radachlorin fluorescence lifetimes and LysoTracker fluorescence intensity supported this suggestion. The results obtained show that the inhomogeneity of fluorescence quantum yield within a cell can be significant due to lower pH values in lysosomes than in other intracellular compartments. This finding suggests that the actual amount of accumulated Radachlorin can be underestimated if being evaluated solely by comparison of fluorescence intensities.

Comparative analysis of Radachlorin accumulation, localization, and photobleaching in three cell lines by means of holographic and fluorescence microscopy.

In this paper we compare the response of cells of established lines of different origin: HeLa, A549 and 3T3 to photodynamic treatment with Radachlorin photosensitizer. The analysis was performed on different aspects of the treatment procedure including photosensitizer accumulation, localization and photobleaching in cells and post-treatment dynamics of changes in cellular morphology at different treatment doses. It was shown that in the three cell lines Radachlorin accumulated in lysosomes to much greater extent than in mitochondria. The cells' response to treatment was analyzed by identification of their death pathways and evaluation of average phase shift dynamics using digital holographic microscopy. The analysis performed on the three cell lines allowed us to evaluate treatment doses specific for each pathway in each line. Among the three lines HeLa cells were found to be the most susceptible to treatment while 3T3 cells the most resistant. The comparison of these results with the data on Radachlorin accumulation, localization and photobleaching rates showed that the observed higher sensitivity of HeLa cells to photodynamic treatment correlated with higher photosensitizer uptake and more intensive photobleaching while lower sensitivity of 3T3 cells correlated with lower uptake and less intensive photobleaching.

In vitro monitoring of photoinduced necrosis in HeLa cells using digital holographic microscopy and machine learning.

Digital holographic microscopy supplemented with the developed cell segmentation and machine learning and classification algorithms is implemented for quantitative description of the dynamics of cellular necrosis induced by photodynamic treatment in vitro. It is demonstrated that the developed algorithms operating with a set of optical, morphological, and physiological parameters of cells, obtained from their phase images, can be used for automatic distinction between live and necrotic cells. The developed classifier provides high accuracy of about 95.5% and allows for calculation of survival rates in the course of cell death.

Quantitative assessment of changes in cellular morphology at photodynamic treatment in vitro by means of digital holographic microscopy.

Temporal dependence of changes in the morphological characteristics of cells of two cultured lines of cancer origin, HeLa and A549, induced by photodynamic treatment with Radachlorin photosensitizer, have been monitored using digital holographic microscopy during first two hours after short-term irradiation. The observed post-treatment early dynamics of the phase shift in the transmitted wavefront indicated several distinct scenarios of cell behavior depending upon the irradiation dose. In particular the phase shift increased at low doses, which can be associated with apoptosis, while at high doses it decreased, which can be associated with necrosis. As shown, the two cell types responded differently to similar irradiation doses. Although the sequence of death scenarios with the increase of the irradiation dose was the same, each scenario was realized at substantially different doses. These findings suggest that the average phase shift of the transmitted wavefront can be used for quantitative non-invasive cell death characterization. The conclusions made were cofirmed by commonly used test assays using confocal fluorescent microscopy.

Digital holographic microscopy in label-free analysis of cultured cells' response to photodynamic treatment.

A methodology providing noninvasive monitoring and evaluation of the effect of photodynamic treatment on live cells in vitro is presented. Variations in morphological characteristics of cells in the course and after treatment are recorded by means of digital holographic microscopy. High-precision measurements of phase shift gained by probe radiation in HeLa and human endometrial mesenchymal stem cell cultures demonstrate for the first time changes of their volume occurred in response to treatment.

EEA1 CARRYING VESICLES ARE NOT AUTOPHAGOSOMES IN SERUM-DEPRIVED HeLa CELLS.

According to current model, stimulation of EGF-receptor endocytosis results in recruitment onto early endosomes cytosolic tether protein EEA1 necessary for their further fusions. However, EEA1-positive vesicles are found in the cells not treated with growth factor, that were incubated in serum-free conditions. It is known also that prolonged serum deprivation induces autophagosomes formation, the process possibly involving endocytic compartments. To check whether EEA1-positive vesicles seen in serum-deprived HeLa cells are autophagosomes, we here evaluated colocalization of EEA1 and autophagosome marker LC3 and studied dynamics of the EEA1- and LC3-vesicles' number and size during 12­36 h cell cultivation in serum-free medium. It was found that the number of autophagosomes per cell is significantly less than the number of EEA1-vesicles. We show that serum starvation results in increase of only mean autophagosomes' size, while the number and size of EEA1-vesicles did not changed. Colocalization of EEA1 and LC3 in serum-free cells was very low during first 12­18 h of starvation and increased insignificantly only by 36 h. Biosynthetic pathway inhibition by Golgi apparatus disruption by brefeldin A, decreased the number and increased the size of EEA1-vesicles. LC3-vesicles also demonstrated an increase of mean size and growth of colocalization with EEA1. Thus, we conclude that the majority of EEA1-vesicles in serum-starved cells are not autophagosomes. More pronounced effect of brefeldin A indicates that blockade of biosynthetic pathway is more strong stress factor comparing to serum deprivation in HeLa cells. This also suggests that this pathway is involved in EEA1-vesicles biogenesis.

[INHIBITORS OF MAP-KINASE PATHWAY U0126 AND PD98059 DIFFERENTLY AFFECT ORGANIZATION OF TUBULIN CYTOSKELETON AFTER STIMULATION OF EGF RECEPTOR ENDOCYTOSIS].

To confirm the hypothesis about the involvement of EGF-stimulated MAP-kinase ERK1/2 in the regulation of microtubule (MT) system, the influence of two widely used ERK1/2 inhibitors, U0126 and PD98059, on the organization of tubulin cytoskeleton in interphase HeLa cells during EGF receptor endocytosis has been investigated. We have found that addition of U0126 or PD98059 to not-stimulated with EGF ells for 30 min has no effect on radially organized MT system. However, in the case of U0126 addition before EGF endocytosis stimulation, the number of MT per cell decreased within 15 min after such stimulation and was followed by complete MT depolymerization by 60-90 min. Stimulation of EGF endocytosis in the presence of PD98059 resulted only in insignificant depolymerization of MT and it could be detected mainly from their minus-ends. At the same time, MT regions close to plasma membrane became stabilized, which was proved by increase in tubulin acetylation level. This situation was characteristic for all period of the experiment. It has been also found that the inhibitors affect endocytosis dynamics of EGF-receptor complexes. Quantitative analysis demonstrated that the stimulation of endocytosis in the presence of U0126 generated a greater number of endosomes compared to control cells, and their number did not change significantly during the experiment. All these endosomes were localized peripherally. Effect of PD98059 resulted in the formation of lower number of endosomes that in control, but they demonstrated very slow clusterization despite the presence of some intact MT. Both inhibitors decreased EGFR colocolization with early endosomal marker EEA1, which indicated a delay in endosome fusions and maturation. The inhibitors were also shown to affect differently phospho-ERK 1 and 2 forms: U0126 completely inhibited phospho-ERK1 and 2, white, in the presence of PD98059, the two ERK forms demonstrated sharp transient activation in 15 min after stimulation, but only phospho-ERK2 could be detected after 60 min of endocytosis. In both cases, MAP-kinase activation dynamics was significantly different from the control. Our results suggest involvement of EGF-stimulated MAP-kinase pathway in cytoskeleton regulation. At the same time, they demonstrate that the two studied and widely used inhibitors are not equivalent with respect to not only the effect on MAP-kinase activity but also to such interdependent processes such as changes in cytoskeleton organization and signaling receptor' endocytosis.

Receptor-mediated endocytosis and cytoskeleton.

Receptor-mediated endocytosis is the most specific pathway for macromolecules and macromolecular complexes generally designated as ligands to enter cells. Upon binding to their transmembrane receptors, the ligands enter endocytic vesicles that fuse with each other giving rise to the so-called early endosomes. The sorting of ligand-receptor complexes internalized in these endosomes depends on their nature: metabolic receptors are recycled back to the plasma membrane, while signaling receptors and their ligands (e.g. receptor tyrosine kinases or receptors associated with tyrosine kinase) are delivered to internal vesicles of the multivesicular late endosomes and finally are degraded after interaction with lysosomes. During these processes, endosomes undergo translocation from the cell periphery to the juxtanuclear region, which is accompanied by multiple fusion, invagination, tabulation, and membrane fission events. This review considers modern concepts of the sorting mechanisms of ligand-receptor complexes, the crosstalk between endosomes, microtubules, and actin, and the role of this crosstalk in endosome maturation.

[Analysis of vesicle subpopulations carrying early endosomal autoantigen EEA1].

Confocal immunofluorescent analysis of interphase HeLa cells has demonstrated that involved in regulation of homotypic fusions early endosomal autoantigene EEA1 is associated with vesicles represented by two populations differing in apparent size, localization and the level of bound EEA1. Before analysis the cells have been preincubated in serum-deprived medium for 12 h to minimize ligand-dependent endocytosis of serum growth factors. The first subpopulation is mainly represented by large vesicles strongly decorated with EEA1. These vesicles are localized presumably in juxtanuclear region. Microtubule depolimerization experiments have shown that this localization is maintained by tubulin cytoskeleton. The second subpopulation consists of numerous small vesicles slightly stained by EEA1 antibody and localized more peripherally. Double indirect immunofluorescent ananlysis of fixed cell images has revealed that juxtanuclear vesicles enriched in EEA1 are fully colocalized with key protein of early endosomes small GTPase Rab5, whereas about 50% of slightly decorated peripheral vesicles are Rab5-negative. It is found that the number of Rab5-positive vesicles per cell is higher than that of EEA1-positive vesicles. Thus, in serum-deprivated HeLa cells with low endocytic activity two subpopulations of EEA1-vesicles are revealed: the first one carries the both EEA1 at high level and Rab5 (EEA1+++/Rab5+), and the second subpopulation oconsists of weakly decorated EEA1-vesicles, that can be both Rab5-positive and -negative (EEA1+/Rab5- and EEA1+/Rab5+). Besides, there are vesicles carrying Rab5 only (EEA1-/Rab5+). The data obtained favor different functional role of all these subpopulations, which are associated with proteins widely considered as equivalent markers of early endosomes.

[Analysis of EGF receptor endocytosis dynamics based on semiquantative processing of confocal immunofluorescent images of fixed cells].

Endocytosis of signaling receptors EGF receptor in particular, starting at the plasma membrane and finishing in perinuclear lysosomes implies endosomes' multiple interactions with homotypic endosomes and vesicles of other origin (lysosomes, trans-Golgi network), which results in endosomal size changes. In addition, the characteristics of endocytic pathway is endosomes' translocation from cell periphery to juxtanuclear region. Thus, endocytosis as a highly dynamic process develops in time and space. Obviously one of the most productive approach is light immunofluorescent microscopy that allows of estimating endocytes dynamics at the level of one or several cells. Different impacts influencing endocytic regulator components are inevitably reflected on dynamics of endosome size and (or) its translocation. This provide possibility to uncover the both crucial and secondary components of regulatory machinery. However, visual estimation of such effects is often too subjective and does not allow getting statistically reliable data. Comparison of different experiments even in the case of the same series is also impeded. In this work we use such parameters as apparent vesicle size (diameter, area or volume) and vesicle number per cell to provide quantitative estimation of fusion efficacy, as well as the coefficient reflecting vesicles' clasterization in perinuclear region as a measure of their translocation along microtubules toward nucleus (D(clust)) and present these parameters application using EGF receptor endocytosis as an example.

[Change in localization of cellular vesicular apparatus during differentiation of myoblasts into myotubules in cell culture].

The study of changes in the intracellular processes during differentiation of myoblasts into myotubules is of great importance for understanding several fundamental problems of cell biology. At first, this concerns the spatial organization of vacuolar apparatus that reflects the alterations in the properties of cell membranes, cytoskeleton elements and dynamics of vesicular transport in the course of differentiation. The distribution of acidic membrane organelles (lysosomes, late endosomes, Golgi cisternae) during the myotubule formation was revealed. It was shown that perinuclear localization of acidic organelles in myoblasts was replaced by diffuse distribution of these structures in the whole volume of myotubules. Using lipophilic fluorescent dyes, RH 414 and di-8-ANEPPS, the process of formation and dynamics of endocytic vesicles in myoblasts and myotubules was investigated. In the present work, semiconductive nanocrystals, quantum dots (QDs), conjugated with TAT-peptide, which belongs to cell-penetrating peptides, were used to characterize nonspecific endocytosis. It was shown that QDs--TAT complexes penetrate myoblasts but do not penetrate myotubules even after 24 h incubation, which might be connected with plasma membrane changes during the process of skeletal muscle differentiation.

[Acetylation of microtubules during endocytosis of epidermal growth factor receptor (c-ErbB1) in interphase HeLa cells].

Acetylation of microtubules (MT) during endocytosis of epidermal growth factor receptor, c-ErbB1, was studied by confocal immunofluorescence microscopy. It was found that stimulation of endocytosis of c-ErbB1 complexed with the epidermal growth factor (EGF), resulted in continuous raising of MT acetylation that reached its maximum at 60-90 min and then went down to the control level. Simultaneously, the receptor-containing endosomes grew in size and were redistributed into juxtranuclear region. Enlarged endosomes formed dense clusters around MTOC in 60-90 min. Another native c-ErbB1 ligand, transforming growth factor-alpha (TGF-alpha) and unlike EGF causing the receptor recycling, also initiated a wave of MT acetylation, but the effect was expressed more poorly. In this case, endosomes did not grow in size and did not form dense clusters near the MTOC. Cell treatment with deacetylase inhibitor trichostatin A (TSA) caused acetylation of the whole cellular MT population. Under these conditions, translocations of EGF-c-ErbB1-containing endosomes had the same pattern as in the cells untreated with the inhibitor, but the size of endosomes didn't increase during their redistribution into juxtranuclear region. Acetylation was especially pronounced in strongly bent MT regions positioned proximally to MTOC and forming there a dense meshwork whereas peripheral MT plus-ends were basically straight and not modified. We assume that MT acetylation is not so much crucial for preferential interaction with dynein or kinesin and, accordingly, for organization of endosome translocations in a certain direction. It is rather the result of stabilization of some MT pool which supports homotypic fusion of endosomes at early stage of their maturation.

[Untargeted quantum dots in confocal microscopy of living cells].

The problem of non-specific binding of quantum dots (QDs) with cells is very important but not fully understood taking into account the possible application of QDs in medical and fundamental studies. The interactions of untargeted CdSe/ZnS QDs with isolated frog muscle fibres, HeLa cells and J774 cells were investigated. The observations were performed on living cells using laser confocal microscopy (Leica TCS SL). The QDs covered with polyethylenglycol without any functional reactive groups with emission maximum at 565 nm were used in the study. This type of QDs is suggested to prevent an interaction of QDs with biological molecules. It has been shown that QDs do not enter HeLa cells, T-system and sarcoplasm of skeletal muscle fibres. However, during long-term incubation J774 cells can uptake QDs. The data obtained has demonstrated the diversity of interactions of untargeted QDs with different cell types and are important for understanding of the problems of non-selective uptake and cytotoxicity of QDs.

[Ubiquitination of EGF receptors with C-terminal domain deletion and point mutations during endocytosis].

Analysis of ubiquitination of EGF receptor carrying different mutations of C-terminal domain was done. The mutants differed both by the set of major autophosphorylation sites that determine the way of interaction with ubiquitin-ligase c-Cbl, and by the presence of lysine residues which can be possible acceptor sites for ubiquitin. It was found that the receptor lacking tyrosine kinase activity due to lysine for phenylalanine substitution at ATP-binding site of tyrosine kinase (TK) domain (K721) failed to be ubiquitinated as well as the receptor without all binding sites for c-Cbl (CD165), while dynamics and pattern of ubiquitination of other deletion mutants was significantly different. The mutant lacking Grb2 binding sites but able to bind c-Cbl directly (CD123) was minimally ubiquitinated and only at early stages upon EGF endocytosis stimulation. At the same time, the receptor possessing all binding sites for Cbl but lacking C-terminal domain of 63 aminoacid residues (CD63) which contains two autophosphorylation sites (Y1148 and 1173) and 4 lysines, was less ubiquitinated and had more low-ubiquitinated forms comparing to the WT one. However, these lysines are not acceptor sites for ubiquitin since the full-size receptor lacking like CD63 the same major autophosphorylation sites underwent ubiquitination similar to the deletion mutant. Thus, C-terminal region of the EGF receptor, being not a substrate for ubiquitination per se, is involved in its regulation. It was also found that ubiquitination pattern at fast endocytosis differed from those at slow one. In the first case the total level of EGFR decreased dramatically as a result of efficient lysosomal degradation. The level of receptor-associated c-Cbl was practically the same, while the total intracellular c-Cbl dropped. Treatment of cells with proteasomal inhibition MG132 blocked the loss of Cbl only partially. In the second case, total amount of both EGF receptor and c-Cbl did not notably change that suggested recycling pathway for receptors even despite them beeng ubiquitinated.

[Analysis of tyrphostin AG1478 effect on behavior of internalized EGF receptor at different stages of endocytosis].

In the present work the effect of specific inhibitor of the receptor tyrosine kinase, tyrphostin AG1478, has been analyzed for behavior of internalized EGF receptor at different stages upon stimulation of endocytosis. It was found that tyrphostin addition in 30 min after endocytosis stimulation resulted in recycling of a significant portion of 125I-EGF onto cell surface. This portion was decreasing with time. EGF-receptor complexes, being recycled under action of AG1478, however, did not dissociate possibly because of tyrphostin ability to initiate receptor oligomerization in the absence of the ligand which can possibly affect dissociation constants. It was found that only a portion of EGF receptor localized in early endosomes was able to recycle upon TK inhibition. Addition of the inhibitor in 30 and 60 min after endocytosis stimulation resulted in decrease of labeled EGF degradation. At early stages internalized EGF-receptor complexes was blocked mostly in early endosomes, while at late stages their accumulation occured in incompletely matured late endosomes. These data speak in favor of late endocytic stage existence transition through which depends on the receptor TK. Besides, tyrphostin addition in 90 min after endocytosis led not to decrease, but on the contrary, to increase in degradation. That speaks about theexistence of the mechanisms providing a time window during which receptor TK can carry out the functions which are not connected directly with endocytosis.

[Remodeling of microtubule system during EGF receptor endocytosis].

The idea of microtubules (MTs) as of passive railway tracks, along which transport vesicles travel by use of motor proteins, is widely accepted. In the present work the organization of MT system during EGF-receptor endocytosis was investigated by indirect double immunofluorescence in HeLa and A431 cell lines. Stimulation of cells with EGF resulted in formation of EGF receptor-containing peripheral vesicular endosomes. During time course of endocytosis the endosomes tended to concentrate in juxtranuclear region close to MTOC. This translocation was dependent on MTs since nocodazole treatment resulted in endosomes' scattering throughout the cytoplasm. Parallel staining of the cells with tubulin antibody has revealed significant remodeling of MTs organization during endocytosis. At early stages MTs demonstrated slight retraction at the cell periphery and the increasing intensity of tubulin fluorescence in the juxtranuclear region. Later on, long individual MTs disappeared and peripheral cytoplasm show diffuse staining in combination with a meshwork of short MT fragments. This stage correlated with EGFR localization in juxtranuclear endosomes. Disappearance of EGFR-positive staining due to its lysosomal degradation occurred in parallel to reestablishment of radial MT system. Possible functional significance of described alterations in organization of tubulin cytoskeleton is discussed.

Nocodazole inhibits macronuclear infection with Holospora obtusa in Paramecium caudatum.

Holospora obtusa is a Gram-negative bacterium inhabiting the macronucleus of the ciliate Paramecium caudatum. Experimental infection with H. obtusa was carried out under nocodazole treatment. Nocodazole has been shown to cause disassembly of the cytoplasmic microtubules radiating from the cytopharynx and postoral fibers in P. caudatum. Treatment with this drug did not prevent the ingestion of both prey bacteria and H. obtusa, but it reduced the phagosome number and affected cyclosis. In situ hybridization revealed infectious forms of this endobiont very close to the macronucleus, but never inside it. These results indicate that disassembly of microtubules does not impair transportation of the infectious forms of H. obtusa in the cytoplasm, but that it completely blocks the invasion of the nucleus by the bacteria.

[Effect of synthetic proteasomal inhibitor MG132 on dynamics of EGF-receptor complexes endocytosis in A431 cells]. / Vliianie sinteticheskogo ingibitora proteasom MG132 na dinamiku éndotsitoza EFR-retseptornykh kompleksov v kletkakh A431.

The effect of proteasomal activity suppression induced by MG132, a synthetic proteasomal inhibitor of EGF-receptor complexes endocytosis in human epidermoid carcinoma A431 cell line, was studied. Using subcellular fractionation in 17% Percoll gradient, it was demonstrated that the addition of MG132 to the cells 15 min following stimulation of EGF endocytosis resulted in a slight accumulation of 125I-EGF in early endosomes, and in much more significant accumulation of the labeled growth factor in late endosomes/lysosomes, as compared to untreated cells. The release of 125I-EGF degradation products into the incubation medium was significantly (3-12-fold) inhibited in the presence of MG132. At the same time biochemical analysis has demonstrated that the EGF receptor itself is not a direct target of proteasomes, since it is revealed as a full-length protein with native mol. mass (170 kDa) in fractions of early and late endosomes and lysosomes. Possible mechanisms of the MG132 effect on intracellular processing of EGF-receptor complexes are discussed.

[The role of phosphatidylinositol-3-kinases P85/P110 and hVPS34 in endocytosis of EGF-receptor complexes]. / Rol' fosfatidilinozitol-3-kinaz P85/P110 i hVPS34 v éndotsitoze EFR-retseptornykh kompleksov.

The previous data (Zheleznova et al., 2001) did not enable the authors to conclude which particular wortmannin sensitive PI-3-kinase--p85/p110 (I class PI-3-K) or hVPS34 (III class PI-3-K)--may be involved in the regulation of EGF-receptor endocytosis. In the present work, we have shown that upon stimulation of EGF-receptor endocytosis additional structures stained with antibody against p85 appear in A431 cells, but the p85-positive compartment never co-localized with EGF-receptor-containing compartments either in control or in wortmannin-treated cells. At the same time, wortmannin treatment prevented association of hVPS34 with endosomal membranes. We have also found that early endosomal markers--Rab5 and EEA1 (membrane association of the latter depends on Rab5 and hVPS34)--co-localized with EGF-receptor in the juxtranuclear region during late stages of endocytosis, both in control and upon wortmannin treatment. These observations favor our suggestions that the transition of EGF-receptors from early to late endosomes may occur directly in this juxtranuclear region and be tightly associated with the formation of so called multivesicular bodies (MVB), which are late endosomes per se. We suggest that wortmannin may have no effect on early EEA1-dependent stage of the receptor endocytosis but blocks a transition of EGF-receptor complexes into the late endosomes by inhibiting activity of hVPS34 and removing it from membranes. The hVPS34 product PI-3-K, according to the known data, is involved in the formation of internal vesicles of MVB. Accumulation of EGF-receptors in these vesicles is believed to be necessary for the receptor degradation.