Ellen Skarpen

Cytoplasmic retention of peroxide‐activated ERK provides survival in primary cultures of rat hepatocytes

Reactive oxygen species (ROS) are implicated in tissue damage causing primary hepatic dysfunction following ischemia/reperfusion injury and during inflammatory liver diseases. A potential role of extracellular signal‐regulated kinase (ERK) as a mediator of survival signals during oxidative stress was investigated in primary cultures of hepatocytes exposed to ROS. Hydrogen peroxide (H2O2) induced a dose‐dependent activation of ERK, which was dependent on MEK activation. The ERK activation pattern was transient compared with the ERK activation seen after stimulation with epidermal growth factor (EGF). Nuclear accumulation of ERK was found after EGF stimulation, but not after H2O2 exposure. A slow import/rapid export mechanism was excluded through the use of leptomycin B, an inhibitor of nuclear export sequence–dependent nuclear export. Reduced survival of hepatocytes during ROS exposure was observed when ERK activation was inhibited. Ribosomal S6 kinase (RSK), a cytoplasmic ERK substrate involved in cell survival, was activated and located in the nucleus of H2O2‐exposed hepatocytes. The activation was abolished when ERK was inhibited with U0126. In conclusion, our results indicate that activity of ERK in the cytoplasm is important for survival during oxidative stress in hepatocytes and that RSK is activated downstream of ERK. Supplementary material for this article can be found on the HEPATOLOGY website (http://www.interscience.wiley.com/jpages/0270‐9139/suppmat/index.html). (HEPATOLOGY 2005;42:200–207.)

Immunocytochemical Localization of Shc and Activated EGF Receptor in Early Endosomes After EGF Stimulation of HeLa Cells

After binding of epidermal growth factor (EGF), the EGF receptor (EGFR) becomes autophosphorylated via tyrosine. The ligand-activated receptor is internalized by endocytosis and subsequently degraded in the lysosomal pathway. To follow EGFR activation after EGF stimulation, we generated antisera to the EGFR phosphotyrosine sites pY992 and pY1173. The SH2 region of Shc binds to both these sites. Both antisera identified EGFR after EGF binding and did not crossreact with the unactivated receptor. The intracellular distribution of phosphorylated EGFR after ligand binding was traced by two-color immunofluorescence confocal microscopy and immunoelectron microscopy. Before EGF stimulation EGFR was primarily located along the cell surface. When internalization of activated EGFR was inhibited by incubation with EGF on ice, Y992- and Y1173-phosphorylated EGFR were located along the plasma membrane. Ten minutes after internalization at 37C, Y992- and Y1173-phosphorylated EGFR were almost exclusively located in early endosomes, as shown by co-localization with EEA1. Immunoelectron microscopy confirmed that phosphorylated EGFR was located in intracellular vesicles resembling early endosomes. After EGF stimulation, the adaptor protein Shc redistributed to EGFR-containing early endosomes. Our results indicate that EGFR activation of Shc via tyrosine-phosphorylated Y992 and Y1173 occurred in early endocytic compartments, and support a role for membrane trafficking in intracellular signaling.

Re-localization of activated EGF receptor and its signal transducers to multivesicular compartments downstream of early endosomes in response to EGF.

The rapid internalization of receptor tyrosine kinases after ligand binding has been assumed to be a negative modulation of signal transduction. However, accumulating data indicate that signal transduction from internalized cell surface receptors also occurs from endosomes. We show that a substantial fraction of tyrosine-phosphorylated epidermal growth factor receptor (EGFR) and Shc, Grb2 and Cbl after internalization relocates from early endosomes to compartments which are negative for the early endosomes, recycling vesicle markers EEA1 and transferrin in EGF-stimulated cells. These compartments contained the multivesicular body and late endosome marker CD63, and the late endosome and lysosome marker LAMP-1, and showed a multivesicular morphology. Subcellular fractionation revealed that activated EGFR, adaptor proteins and activated ERK 1 and 2 were located in EEA1-negative and LAMP-1-positive fractions. Co-immunoprecipitations showed EGFR in complex with both Shc, Grb2 and Cbl. Treatment with the weak base chloroquine or inhibitors of lysosomal enzymes after EGF stimulation induced an accumulation of tyrosine-phosphorylated EGFR and Shc in EEA1-negative and CD63-positive vesicles after a 120-min chase period. This was accompanied by a sustained activation of ERK 1 and 2. These results suggest that EGFR signaling is not spatially restricted to the plasma membrane, primary vesicles and early endosomes, but is continuing from late endocytic trafficking organelles maturing from early endosomes.

MEK1 and MEK2 regulate distinct functions by sorting ERK2 to different intracellular compartments

In this study, we provide novel insight into the mechanism of how ERK2 can be sorted to different intracellular compartments and thereby mediate different responses. MEK1‐activated ERK2 accumulated in the nucleus and induced proliferation. Conversely, MEK2‐activated ERK2 was retained in the cytoplasm and allowed survival. Localization was a determinant for ERK2 functions since MEK1 switched from providing proliferation to be a mediator of survival when ERK2 was routed to the cytoplasm by the attachment of a nuclear export site. MEK1‐mediated ERK2 nuclear translocation and proliferation were shown to depend on phosphorylation of S298 and T292 sites in the MEK1 proline‐rich domain. These sites are phosphorylated on cellular adhesion in MEK1 but not MEK2. Whereas p21‐activated kinase phosphorylates S298 and thus enhances the MEK1‐ERK2 association, ERK2 phosphorylates T292, leading to release of active ERK2 from MEK1. On the basis of these results, we propose that the requirement of adhesion for cells to proliferate in response to growth factors, in part, may be explained by the MEK1 S298/T292 control of ERK2 nuclear translocation. In addition, we suggest that ERK2 intracellular localization determines whether growth factors mediate proliferation or survival and that the sorting occurs in an adhesion‐dependent manner.—Skarpen, E., Flinder, L. I., Rosseland, C. M., Ørstavik, S., Wierød, L., Pedersen Oksvold, M., Skålhegg, B. S., Huitfeldt, H. S. MEK1 and MEK2 regulate distinct functions by sorting ERK2 to different intracellular compartments. FASEB J. 22, 466–476 (2008)

Photochemical internalization (PCI) of immunotoxins targeting CD133 is specific and highly potent at femtomolar levels in cells with cancer stem cell properties

CD133 is a putative cancer stem cell (CSC) marker for a number of different cancers and is suggested to be a therapeutic target. Since also normal stem cells express CD133 it is of paramount importance that targeting strategies provide a specific and efficient delivery of cytotoxic drugs in only CD133-positive CSCs. In this study, we have employed photochemical internalization (PCI), a minimally invasive method for light-controlled, specific delivery of membrane-impermeable macromolecules from endocytic vesicles to the cytosol, to specifically target CD133-positive cancer cells. We demonstrate that PCI increases the cytotoxic effect of an immunotoxin (IT) targeting CD133-expressing cancer cells of colon (WiDr and HCT116) and pancreas (BxPC-3) origin. The IT consisted of the mAb CD133/1 (AC133) bound to the ribosome inactivating plant toxin saporin (anti-CD133/1-sap). We show that TPCS2a-PCI of anti-CD133/1-sap is specific, and highly cytotoxic at femto-molar concentrations. Specific binding and uptake of CD133/1, was shown by fluorescence microscopy and co-localization with TPCS2a in endosomes/lysosomes was determined by confocal microscopy. CD133(high) WiDr cells, isolated by fluorescence activated cell sorting, had a 7-fold higher capacity to initiate spheroids than CD133(low) cells (P<0.001) and were resistant to photodynamic therapy (PDT). However, PDT-resistance was bypassed by the PCI strategy. Tumor initiation and aggressive growth in athymic nude mice was obtained with only 10 CD133(high) cells in contrast to CD133(low) cells where substantially higher cell numbers were needed. The excellent high efficacy and selectivity of eliminating CD133-expressing cells by PCI warrant further pre-clinical evaluations of this novel therapeutic approach.

EGF‐induced ERK‐activation downstream of FAK requires rac1‐NADPH oxidase

Reactive oxygen species (ROS) function as signaling molecules mainly by reversible oxidation of redox‐sensitive target proteins. ROS can be produced in response to integrin ligation and growth factor stimulation through Rac1 and its effector protein NADPH oxidase. One of the central roles of Rac1‐NADPH oxidase is actin cytoskeletal rearrangement, which is essential for cell spreading and migration. Another important regulator of cell spread is focal adhesion kinase (FAK), a coordinator of integrin and growth factor signaling. Here, we propose a novel role for NADPH oxidase as a modulator of the FAK autophosphorylation site. We found that Rac1‐NADPH oxidase enhanced the phosphorylation of FAK at Y397. This site regulates FAKs ability to act as a scaffold for EGF‐mediated signaling, including activation of ERK. Accordingly, we found that EGF‐induced activation of FAK at Y925, the following activation of ERK, and phosphorylation of FAK at the ERK‐regulated S910‐site depended upon NADPH oxidase. Furthermore, the inhibition of NADPH oxidase caused excessive focal adhesions, which is in accordance with ERK and FAK being modulators of focal adhesion dissociation. Our data suggest that Rac1 through NADPH oxidase is part of the signaling pathway constituted by FAK, Rac1, and ERK that regulates focal adhesion disassembly during cell spreading. J. Cell. Physiol. 226: 2267–2278, 2011.

Altered regulation of EGF receptor signaling following a partial hepatectomy.

We have studied epidermal growth factor receptor (EGFR) phosphorylation and localization in the pre‐replicative phase of liver regeneration induced by a 70% partial hepatectomy (PH), and how a PH affects EGFR activation and trafficking. When Western blotting was performed on livers after PH with antibodies raised against activated forms of EGFR autophosphorylation sites, no marked increase in EGFR tyrosine phosphorylation was observed. However, events associated with attenuation of EGFR signals were observed. Two hours after PH, we found increased EGFR ubiquitination and internalization, followed by receptor downregulation. Furthermore, EGFR phosphorylation following an injection of EGF was reduced after PH. This reduction correlated with an increased activation of PKC and a distinct augmentation in the phosphorylation of the PKC‐regulated T654‐site of EGFR. When primary cultured hepatocytes were treated with tetradecanoylphorbol acetate (TPA) to induce T654‐phosphorylation of EGFR, we found colocalization of a fraction of EGFR with EEA1, downregulation of EGF‐mediated EGFR autophosphorylation, altered ligand‐induced intracellular sorting of EGFR, and increased mitogenic signaling through the EGFR‐Ras‐Raf‐ERK pathway. Further, we found that both TPA and a PH enhanced EGF‐induced proliferation of hepatocytes. In conclusion, our results suggest that hepatocyte priming involves modulation of EGFR that enhances its ability to mediate growth factor responses without an increase in its receptor tyrosine kinase‐activity. This may be a pre‐replicative competence event that increases growth factor effects during G1 progression.

Photochemical internalization of CD133-targeting immunotoxins efficiently depletes sarcoma cells with stem-like properties and reduces tumorigenicity.

BACKGROUND The normal stem cell marker CD133 is also a putative marker of cancer stem cells (CSCs) in different types of cancers. Hence, a major challenge when targeting CD133-expressing CSCs is to prevent depletion of the normal stem cell pool. We hypothesized that the site-specific and light-controlled drug delivery method photochemical internalization (PCI) may have the potential to enhance selectivity and endosomal escape of CD133-targeting immunotoxins in stem-like sarcoma cells. METHODS We have used a sarcoma model, SW872 cells isolated from xenografts harboring CSCs within a ~2% CD133(high) subpopulation to investigate the potential of PCI of CD133-targeting toxin as a novel strategy to kill CSCs. Model immunotoxins were generated by binding the ribosome-inactivating protein toxin saporin to each of the monoclonal antibodies CD133/1 (AC133) or CD133/2 (293C), specific for individual CD133-epitopes. Cellular targeting, intracellular co-localization with the PCI photosensitizer, disulfonated meso-tetraphenylchlorin (TPCS2a), and cytotoxic efficacy of PCI of the CD133-targeting toxins were evaluated. RESULTS PCI of CD133-saporin efficiently targets CD133-expressing SW872 and HT1080 sarcoma cells and results in loss of cell viability. Following sub-toxic treatment, surviving SW872 cells, depleted of the CD133-expressing population, display reduced proliferative capacity and attenuated CSC properties, such as reduced colony-forming ability and tumorigenicity. CONCLUSION Here we present a proof-of-concept study, where PCI enables light-triggered delivery of CD133-targeting antibody-drug conjugates, resulting in decreased sarcoma tumor-initiating capacity. GENERAL SIGNIFICANCE PCI of CD133-targeting toxins may be used as a minimal invasive strategy in the treatment of sarcomas, and potentially as a therapeutic for other solid tumors expressing CD133.

Distinct functions of H-Ras and K-Ras in proliferation and survival of primary hepatocytes due to selective activation of ERK and PI3K

Ras proteins mediate signals both via extracellular signal‐regulated kinase 1 and 2 (ERK), and phosphoinositide 3‐kinase (PI3K). These signals are key events in cell protection and compensatory cell growth after exposure to cell damaging and pro‐apoptotic stimuli, thus maintaining homeostasis. By transfection techniques, we found that both H‐Ras and K‐Ras were expressed and appeared functionally active in primary hepatocytes. We compared the ability of H‐Ras and K‐Ras homologues to preferentially activate one of the two pathways, thereby differentially controlling cell survival and growth. We found that ectopic expression of dominant negative (DN) H‐RasN17, but not DN K‐RasN17, efficiently inhibited both phosphorylation and translocation of ERK to the nuclear compartment, which are prerequisites for cell cycle progression. Furthermore, ectopic expression of constitutive active (CA) H‐RasV12, but not CA K‐RasV12, potentiated EGF‐induced proliferation. We also found that expression of CA mutants of either H‐Ras or K‐Ras protected hepatocytes from transforming growth factor‐β1 (TGF‐β1)‐induced apoptosis. However, H‐Ras‐induced survival was mediated by ERK/RSK as well as by PI3K, whereas K‐Ras‐induced survival was mediated by PI3K only. In conclusion, H‐Ras and K‐Ras had differential functions in proliferation and survival of primary hepatocytes. H‐Ras was the major mediator of ERK‐induced proliferation and survival, whereas H‐Ras and K‐Ras both mediated PI3K‐induced survival. J. Cell. Physiol. 215: 818–826, 2008.

Fluorescent histochemical techniques for analysis of intracellular signaling

Intracellular signaling relies on the orchestrated cooperation of signaling proteins and modules, their intracellular localization, and membrane trafficking. Recently, a repertoire of fluorescence-based techniques, which significantly increases our potential for detailed studies of the involved mechanisms, has been introduced. Microscopic techniques with increased resolution have been combined with improved techniques for detection of signaling proteins. Transfections of fluorescently tagged proteins have allowed in vivo microscopy of their trafficking and interactions with other proteins and intracellular structures. We present an overview of general signaling principles and a description of techniques based on fluorescent microscopy suited for studies of signaling mechanisms.