Monitoring of chromatin organization in live cells by FRIC. Effects of the inner nuclear membrane protein Samp1.

In most cells, transcriptionally inactive heterochromatin is preferentially localized in the nuclear periphery and transcriptionally active euchromatin is localized in the nuclear interior. Different cell types display characteristic chromatin distribution patterns, which change dramatically during cell differentiation, proliferation, senescence and different pathological conditions. Chromatin organization has been extensively studied on a cell population level, but there is a need to understand dynamic reorganization of chromatin at the single cell level, especially in live cells. We have developed a novel image analysis tool that we term Fluorescence Ratiometric Imaging of Chromatin (FRIC) to quantitatively monitor dynamic spatiotemporal distribution of euchromatin and total chromatin in live cells. A vector (pTandemH) assures stoichiometrically constant expression of the histone variants Histone 3.3 and Histone 2B, fused to EGFP and mCherry, respectively. Quantitative ratiometric (H3.3/H2B) imaging displayed a concentrated distribution of heterochromatin in the periphery of U2OS cell nuclei. As proof of concept, peripheral heterochromatin responded to experimental manipulation of histone acetylation. We also found that peripheral heterochromatin depended on the levels of the inner nuclear membrane protein Samp1, suggesting an important role in promoting peripheral heterochromatin. Taken together, FRIC is a powerful and robust new tool to study dynamic chromatin redistribution in live cells.

Mitotic spindle assembly and γ-tubulin localisation depend on the integral nuclear membrane protein Samp1.

We have investigated a possible role for the inner nuclear membrane protein Samp1 (also known as TMEM201) in the mitotic machinery. Live-cell imaging showed that Samp1a-YFP (Samp1a is the short isoform of Samp1) distributed as filamentous structures in the mitotic spindle, partially colocalising with ß-tubulin. Samp1 depletion resulted in an increased frequency of cells with signs of chromosomal mis-segregation and prolonged metaphase, indicating problems with spindle assembly and/or chromosomal alignment. Consistent with this, mitotic spindles in Samp1-depleted cells contained significantly lower levels of ß-tubulin and γ-tubulin, phenotypes that were rescued by overexpression of Samp1a-YFP. We found that Samp1 can bind directly to γ-tubulin and that Samp1 co-precipitated with γ-tubulin and the HAUS6 subunit of the Augmin complex in live cells. The levels of HAUS6, in the mitotic spindle also decreased after Samp1 depletion. We show that Samp1 is involved in the recruitment of HAUS6 and γ-tubulin to the mitotic spindle. Samp1 is the first inner nuclear membrane protein shown to have a function in mitotic spindle assembly.

Publisher Correction: Spindle associated membrane protein 1 (Samp1) is required for the differentiation of muscle cells.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

RanGTPase regulates the interaction between the inner nuclear membrane proteins, Samp1 and Emerin.

Samp1, spindle associated membrane protein 1, is a type II integral membrane protein localized in the inner nuclear membrane. Recent studies have shown that the inner nuclear membrane protein, Emerin and the small monomeric GTPase, Ran are direct binding partners of Samp1. Here we addressed the question whether Ran could regulate the interaction between Samp1 and Emerin in the inner nuclear membrane. To investigate the interaction between Samp1 and Emerin in live cells, we performed FRAP experiments in cells overexpressing YFP-Emerin. We compared the mobility of YFP-Emerin in Samp1 knock out cells and cells overexpressing Samp1. The results showed that the mobility of YFP-Emerin was higher in Samp1 knock out cells and lower in cells overexpressing Samp1, suggesting that Samp1 significantly attenuates the mobility of Emerin in the nuclear envelope. FRAP experiments using tsBN2 cells showed that the mobility of Emerin depends on RanGTP. Consistently, in vitro binding experiments showed that the affinity between Samp1 and Emerin is decreased in the presence of Ran, suggesting that Ran attenuates the interaction between Samp1 and Emerin. This is the first demonstration that Ran can regulate the interaction between two proteins in the nuclear envelope.

Spindle associated membrane protein 1 (Samp1) is required for the differentiation of muscle cells.

Muscles are developed and regenerated in a differentiation process called myogenesis, which involves components of the nuclear envelope. We have investigated Samp1 (Spindle Associated Membrane Protein 1), a transmembrane nuclear envelope protein, which interacts with emerin and lamin A, both of which are linked to Emery-Dreifuss muscular dystrophy (EDMD). We found that the levels of Samp1 increased seven-fold during differentiation of mouse C2C12 muscle progenitor cells. To test if Samp1 could have a role in myogenesis we developed stable C2C12 knockdown cell lines expressing short hairpin RNA targeting Samp1 expression. The Samp1 depleted C2C12 cells displayed normal mobility and normal distribution of emerin and lamin A. However, Samp1 depletion increased ERK signaling and completely blocked differentiation of C2C12 cells, which failed to express myogenic marker proteins and failed to form myotubes. The block in myogenesis in Samp1 depleted cells was completely rescued by ectopic expression of RNAi resistant human Samp1, showing that Samp1 is required for muscle differentiation.

Molecular basis for the dual subcellular distribution of microsomal glutathione transferase 1.

Microsomal glutathione transferase 1 (MGST1) is a membrane bound enzyme involved in the detoxification of reactive electrophiles and protection of membranes from oxidative stress. The enzyme displays an unusual and broad subcellular distribution with especially high levels in the endoplasmic reticulum (ER) and outer mitochondrial membrane (OMM). Here we examined the molecular basis for this dual distribution. We hypothesized that the amphipathic properties of the first transmembrane segment (TMS), that contains a positively charged lysine (K25), is a central feature guiding dual targeting. The lysine-25 was substituted to alanine by site directed mutagenesis. We also increased the amphipathic character of the helix by inserting an additional lysine either one turn above or below K25. Expressing these constructs in simian COS cells, and analyzing subcellular distribution by immunocytochemistry, we observed an increased ER targeting of K25A-MGST1. In contrast I22K-MGST1 and F28K-MGST1 displayed pronounced mitochondrial targeting. By using in vitro transcription-translation we examined whether insertion of WT-MGST1 into ER is co- or post-translational and provide evidence for the former. In the same experimental set-up, mitochondrial insertion was shown to depend on the positive charge. Together these results show that removing the positive charge of lysine-25 promotes ER incorporation, but counteracts mitochondrial insertion. In contrast, introducing an extra lysine in the first TMS of MGST1 had opposite effects. The amphipathic character of the first TMS thus constitutes a molecular determinant for the dual targeting of MGST1. Broad subcellular distribution is consistent with a physiological role in protection from reactive intermediates and oxidative stress.

Samp1, a RanGTP binding transmembrane protein in the inner nuclear membrane.

Samp1 is a transmembrane protein of the inner nuclear membrane (INM), which interacts with the nuclear lamina and the Linker of Nucleoskeleton and Cytoskeleton (LINC) complex in interphase and during mitosis, it localizes to the mitotic spindle. Samp1 was recently found to coprecipitate a protein complex containing Ran, a GTPase with fundamental regulatory functions both in interphase and in mitosis. To investigate the interaction between Samp1 and Ran in further detail, we have designed and expressed recombinant fusion proteins of the Chaetomium thermophilum homolog of Samp1 (Ct.Samp1) and human Ran. Pulldown experiments show that Samp1 binds directly to Ran and that Samp1 binds better to RanGTP compared to RanGDP. Samp1 also preferred RanGTP over RanGDP in living tsBN2 cells. We also show that the Ran binding domain is located between amino acids 75-135 in the nucleoplasmically exposed N-terminal tail of Samp1. This domain is unique for Samp1, without homology in any other proteins in fungi or metazoa. Samp1 is the first known transmembrane protein that binds to Ran and could provide a unique local binding site for RanGTP in the INM. Samp1 overexpression resulted in increased Ran concentrations in the nuclear periphery supporting this idea.

Anchoring of FRET Sensors-A Requirement for Spatiotemporal Resolution.

FRET biosensors have become a routine tool for investigating mechanisms and components of cell signaling. Strategies for improving them for particular applications are continuously sought. One important aspect to consider when designing FRET probes is the dynamic distribution and propagation of signals within living cells. We have addressed this issue by directly comparing an anchored (taFS) to a non-anchored (naFS) cleavable FRET sensor. We chose a microtubule-associated protein tau as an anchor, as microtubules are abundant throughout the cytosol of cells. We show that tau-anchored FRET sensors are concentrated at the cytoskeleton and enriched in the neurite-like processes of cells, providing high intensity of the total signal. In addition, anchoring limits the diffusion of the sensor, enabling spatiotemporally resolved monitoring of subcellular variations in enzyme activity. Thus, anchoring is an important aspect to consider when designing FRET sensors for deeper understanding of cell signaling.

MCLIP Detection of Novel Protein-Protein Interactions at the Nuclear Envelope.

The organization and function of the nuclear envelope (NE) involves hundreds of nuclear membrane proteins and myriad protein-protein interactions, most of which are still uncharacterized. Many NE proteins interact stably or dynamically with the nuclear lamina or chromosomes. This can make them difficult to extract under nondenaturing conditions, and greatly limits our ability to explore and identify functional protein interactions at the NE. This knowledge is needed to understand nuclear envelope structure and the mechanisms of human laminopathy diseases. This chapter provides detailed protocols for MCLIP (membrane cross-linking immunoprecipitation) identification of novel protein-protein interactions in mammalian cells.

A convergent uptake route for peptide- and polymer-based nucleotide delivery systems.

Cell-penetrating peptides (CPPs) have been used as vehicles to deliver various cargos into cells and are promising as tools to deliver therapeutic biomolecules such as oligonucleotides both in vitro and in vivo. CPPs are positively charged and it is believed that CPPs deliver their cargo in a receptor-independent manner by interacting with the negatively charged plasma membrane and thereby inducing endocytosis. In this study we examine the mechanism of uptake of several different, well known, CPPs that form complexes with oligonucleotides. We show that these CPP:oligonucleotide complexes are negatively charged in transfection-media and their uptake is mediated by class A scavenger receptors (SCARA). These receptors are known to promiscuously bind to, and mediate uptake of poly-anionic macromolecules. Uptake of CPP:oligonucleotide complexes was abolished using pharmacological SCARA inhibitors as well as siRNA-mediated knockdown of SCARA. Additionally, uptake of CPP:oligonucleotide was significantly increased by transiently overexpressing SCARA. Furthermore, SCARA inhibitors also blocked internalization of cationic polymer:oligonucleotide complexes. Our results demonstrate that the previous held belief that CPPs act receptor independently does not hold true for CPP:oligonucleotide complexes, as scavenger receptor class A (SCARA) mediates the uptake of all the examined CPP:oligonucleotide complexes in this study.

MCLIP, an effective method to detect interactions of transmembrane proteins of the nuclear envelope in live cells.

Investigating interactions of proteins in the nuclear envelope (NE) using co-immunoprecipitation (Co-IP) has previously been difficult or even impossible due to their inherent resistance to extraction. We have developed a novel method, MCLIP (Membrane protein Cross-Link ImmunoPrecipitation), which takes advantage of a cell permeable crosslinker to enable effective detection and analysis of specific interactions of NE proteins in live cells using Western blot. Using MCLIP we show that, in U2OS cells, the integral inner nuclear membrane protein Samp1 interacts with Lamin B1, the LINC (Linker of nucleoskeleton and cytoskeleton) complex protein, Sun1 and the soluble small GTPase Ran. The results show that the previously detected in vitro interaction between Samp1 and Emerin also takes place in live cells. In vitro pull down experiments show, that the nucleoplasmic domains of Samp1 and Emerin can bind directly to each other. We also, show that MCLIP is suitable to coprecipitate protein interactions in different stages of the cell cycle.

Microtubule-associated nuclear envelope proteins in interphase and mitosis.

The LINC (linker of nucleoskeleton and cytoskeleton) complex forms a transcisternal bridge across the NE (nuclear envelope) that connects the cytoskeleton with the nuclear interior. This enables some proteins of the NE to communicate with the centrosome and the microtubule cytoskeleton. The position of the centrosome relative to the NE is of vital importance for many cell functions, such as cell migration and division, and centrosomal dislocation is a frequent phenotype in laminopathic disorders. Also in mitosis, a small group of transmembrane NE proteins associate with microtubules when they concentrate in a specific membrane domain associated with the mitotic spindle. The present review discusses structural and functional aspects of microtubule association with NE proteins and how this association may be maintained over the cell cycle.

Samp1 is functionally associated with the LINC complex and A-type lamina networks.

The transmembrane inner nuclear membrane (INM) protein Samp1 is required for anchoring centrosomes near the nuclei. Using high-resolution fluorescence microscopy we show that Samp1 is distributed in a distinct and characteristic pattern in the nuclear envelope (NE), where it partially colocalizes with the LINC complex protein Sun1. By studying the localization of Samp1 deletion mutants and fusion proteins, we conclude that the cysteine-rich N-terminal half of Samp1 is nucleoplasmically exposed and is responsible for targeting to the INM. It contains four conserved CxxC motifs with the potential to form zinc fingers. The distribution of cysteine-to-alanine substitution mutants, designed to prevent zinc finger formation, showed that NE localization of Samp1 depends on intact CxxC motifs. Overexpression of Samp1 zinc finger mutants produced an abnormal dominant phenotype characterized by disrupted organization of a selective subset NE proteins, including emerin, Sun1, endogenous Samp1 and, in some cases, lamin A/C, but not lamin B, Sun2 or nucleoporins. Silencing of Samp1 expression showed that emerin depends on Samp1 for its correct localization in the NE. Our results demonstrate that Samp1 is functionally associated with the LINC complex protein Sun1 and proteins of the A-type lamina network.

Anchored FRET sensors detect local caspase activation prior to neuronal degeneration.

BACKGROUND: Recent studies indicate local caspase activation in dendrites or axons during development and in neurodegenerative disorders such as Alzheimer's disease (AD). Emerging evidences point to soluble oligomeric amyloid-ß peptide as a causative agent in AD. RESULTS: Here we describe the design of fluorescence resonance energy transfer (FRET)-based caspase sensors, fused to the microtubule associated protein tau. Specific caspase sensors preferentially cleaved by caspase-3, -6 or -9 were expressed in differentiated human neuroblastoma SH-SY5Y cells. The anchoring of the sensors resulted in high FRET signals both in extended neurites and soma and made analysis of spatiotemporal signal propagation possible. Caspase activation was detected as loss of FRET after exposure to different stimuli. Interestingly, after staurosporine treatment caspase-6 activation was significantly delayed in neurites compared to cell bodies. In addition, we show that exposure to oligomer-enriched amyloid-ß peptide resulted in loss of FRET in cells expressing sensors for caspase-3 and -6, but not -9, in both soma and neurites before neurite degeneration was observed. CONCLUSIONS: Taken together, the results show that by using anchored FRET sensors it is possible to detect stimuli-dependent differential activation of caspases and to distinguish local from global caspase activation in live neuronal cells. Furthermore, in these cells oligomer-enriched amyloid-ß peptide induces a global, rather than local activation of caspase-3 and -6, which subsequently leads to neuronal cell death.