Preparation and characterization of nuclear-envelope vesicles from rat liver nuclei

LUMA interacts with emerin and influences its distribution at the inner nuclear membrane

We present here a first characterization of LUMA, an unique integral inner nuclear membrane (INM) protein. LUMA is a highly conserved protein even in some bacteria and shares a PFAM domain of unknown function with orthologs from many species. Assessing LUMA topology by using protease protection of membrane-inserted LUMA and antibody epitope accessibility assays reveals that LUMA contains four transmembrane domains and a large hydrophilic domain located between membrane spans 1 and 2. The large hydrophilic domain is exposed to the perinuclear space whereas both LUMA termini reside cyto- or nucleoplasmically. Nuclear envelope targeting of LUMA mainly depends on the membrane spans. LUMA's transmembrane domains also promote homooligomerization. LUMA binds A- and B-type lamins and depends on A-type lamins for its INM localization. Furthermore, it interacts with emerin. Both downregulation of LUMA and overexpression of dominant-negative acting LUMA fragments causes redistribution of emerin. We propose that LUMA functions as a tetraspanin-like membrane organizer and has the potential to contribute to the pathomechanism of dystrophic diseases, such as Emery-Dreifuss muscular dystrophy.

A novel function for the 90 kDa heat-shock protein (Hsp90): facilitating nuclear export of 60 S ribosomal subunits

Ribosomal subunits are assembled in the nucleus, and mature 40 S and 60 S subunits are exported stoichiometrically into the cytoplasm. The nuclear export of ribosomal subunits is a unidirectional, saturable and energy-dependent process. An in vitro assay for the nuclear export of 60 S ribosomal subunits involves the use of resealed nuclear envelopes. The export of ribosomal subunits from resealed nuclear envelopes is enhanced by cytoplasmic proteins. Here we present evidence that the export-promoting activity was due to the cytoplasmic 90 kDa heat-shock protein (Hsp90). Isolated, purified Hsp90 vastly enhanced the export of 60 S ribosomal subunits from resealed nuclear envelopes, while inhibition of Hsp90 function, either with the Hsp90-binding drug geldanamycin or with anti-Hsp90 antibodies, resulted in reduced release of 60 S ribosomal subunits. To confirm these findings under in vivo conditions, corresponding experiments were performed with Xenopus oocytes using microinjection techniques; the results obtained confirmed the findings obtained with resealed nuclear envelopes. These findings suggest that Hsp90 facilitates the nuclear export of 60 S ribosomal subunits, probably by chaperoning protein interactions during the export process.

Export of ribosomal subunits from resealed rat liver nuclear envelopes

We have previously described the rat liver resealed nuclear envelope model system for the study of the selective import of nuclear proteins, and the export of poly(A)-containing mRNA [Riedel, N., Bachmann, M., Richter, H. & Fasold, H. (1987) Proc. Natl Acad. Sci. USA 83, 3540-3544]. The vesicles still respond to the importin-ATP signal for the uptake of nuclear-location-sequence (NLS)-carrying proteins. During the preparation of the vesicles and extraction of the chromatin from nuclei in cold hypotonic heparin solution, ribosomal subunits may be introduced into these envelopes, and after resealing remain stably included. Efflux from the resealed nuclear envelopes is effected by a cytoplasmatic protein fraction, and strongly enhanced in the presence of ATP. The heterogeneous nuclear RNP (hnRNP) A1, the components of importin, or GTP showed no influence on this export. The ATP-dependent efflux of mRNA is not affected by these cytoplasmic proteins in this model system.

Intranuclear distribution of 2,2',4,4',5,5'-hexachlorobiphenyl and 2,3,7,8,-tetrachlorodibenzo-p-dioxin

The in vitro intranuclear distribution of the chlorinated aromatic hydrocarbon compounds 2,2',4,4',5,5'-hexachlorobiphenyl and 2,3,7,8-tetrachlorodibenzo-p-dioxin was determined in isolated rat liver cell nuclei. Suspended nuclei were incubated with the 3H-labeled congeners, and the incubation terminated by brief UV irradiation. High-intensity UV irradiation at 254 nm changes the reversible association between macromolecules and ligands into covalent linkages and thus stabilizes the equilibrium distribution. The nuclei were then fractionated with the radioactive congeners covalently linked to the purified fractions. The intranuclear distribution of the model compounds was not uniform. The majority of either chemical was attached to the nuclear envelope and to the chromatin fraction. Much lower amounts were bound to nucleoli. The nuclear matrix was almost devoid of the chemicals. Minute amounts of either compound were detected in association with DNA, none with nuclear RNA. The substantial association of these chlorinated hydrocarbon model compounds with chromatin may bear more general biological relevance and point to detrimental effects on the genetic apparatus. The presented method yields an unequivocal profile of the genuine nuclear distribution of photoactivatable chemicals.

Nuclear-envelope nucleoside triphosphatase kinetics and mRNA transport following brain ischemia and reperfusion

STUDY HYPOTHESIS

We attempted to determine whether the reduced egress of mRNA from brain nuclei following in vivo ischemia and reperfusion is caused by direct damage to the nuclear pore-associated NTPase that impairs the system for nuclear export of polyadenylated, or poly(A)+, mRNA.

DESIGN

Prospective animal study.

INTERVENTIONS

NTPase activity and poly(A)+ mRNA transport were studied in nuclear envelope vesicles (NEVs) prepared from canine parietal cortex isolated after 20 minutes of ischemia or 20 minutes of ischemia and 2 or 6 hours of reperfusion.

RESULTS

Brain NEV NTPase Michaelis-Menten constant (Km) and maximum uptake velocity (Vmax) and the ATP-stimulated poly(A)+ mRNA egress rates were not significantly affected by ischemia and reperfusion. In vitro exposure of the NEVs to the OH. radical-generating system completely abolished NTPase activity.

CONCLUSION

We conclude that brain ischemia and reperfusion do not induce direct inhibition of nucleocytoplasmic transport of poly(A)+ mRNA. This suggests that the nuclear membrane is not exposed to significant concentrations of OH. radical during reperfusion.

Histone-poly(A) hybrid molecules as tools to block nuclear pores

Histone-poly(A) hybrid molecules were used for transport experiments with resealed nuclear envelopes and after attachment of a cleavable cross-linker (SASD) to identify nuclear proteins. In contrast to histones, the hybrid molecules cannot be accumulated in resealed nuclear envelopes, and in contrast to poly(A), the export of hybrids from preloaded nuclear envelopes is completely impaired. The experiments strongly confirm the existence of poly(A) as an export signal in mRNA which counteracts the nuclear location signals (NLS) in histones. The contradicting transport signals in the hybrid molecules impair translocation through the nuclear pore complex. The failure to accumulate hybrid molecules into resealed nuclear envelopes results from the covalent attachment of polyadenylic acid to histones in a strict 1:1 molar ratio. This was demonstrated in control transport experiments where radiolabeled histones were simply mixed with nonlabeled poly(A) or radiolabeled poly(A) mixed with nonlabeled histones. In comparison, control uptake experiments with histones covalently linked to a single UMP-mononucleotide are strongly enhanced. Such controls exclude the conceivable possibility of a simple masking of the nuclear location signal in the histones by the covalent attached poly(A) moiety. Photoreactive histone-poly(A) hybrid analogs serve to identify nuclear envelope proteins--presumably in the nuclear pore--with molecular weights of 110, 80, and 71.4 kDa.

GTP-binding proteins in bovine brain nuclear membranes

Nuclear membranes and other subcellular fractions derived from bovine brain cortex were investigated for the existence of GTP-binding proteins. By using photolytic labeling with [alpha-32P]GTP a 29 kDa GTP-binding protein was shown to be present in nuclear membranes which was not present in the plasma membranes nor in microsomal or cytosolic fractions. Two-dimensional gel electrophoresis revealed that this protein is rather acidic with a pI lower than 4.5. Members of the heterotrimeric Gi/o family are not present in the nuclear envelope: a 39 kDa protein, ADP ribosylated by pertussis toxin, was shown to originate from plasma membrane contamination.

Association of AUUUA-binding protein with A+U-rich mRNA during nucleo-cytoplasmic transport

Resealed nuclear envelope (NE) vesicles from rat liver containing entrapped exogenous RNA were used to study the effect of adenosine+uridine binding factor (AUBF), present in cytosolic cell extracts, on ATP-dependent transport of A+U-rich RNA (AU+RNA) and A+U-free RNA (AU-RNA) across the NE. This factor specifically binds to A+U-rich sequences present in the 3' untranslated regions of lymphokine and cytokine mRNAs, containing overlapping AUUUA boxes (granulocyte-macrophage colony stimulating factor, interleukin-3). Addition of AUBF to the extravesicular compartment markedly increased the efflux of the in vitro transcribed, capped and polyadenylated AU+ RNAs. Export of entrapped AU- control RNA, such as beta-globin RNA, was not affected by AUBF, in contrast to chimeric AU+ beta-globin RNA containing the A+U-rich sequence of human interferon-alpha mRNA (6 reiterated AUUUA motifs). Competition experiments revealed that AUBF enhances the affinity of poly(A)-containing AU+ RNAs to the NE poly(A)-binding component (poly(A)-recognizing mRNA carrier p106), and thereby accelerates nuclear export of these RNAs. We could demonstrate that AUBF added to the extravesicular space forms stable complexes with polyadenylated AU+ RNA with relative molecular masses of about 45,000, 62,000 and 70,000 inside the vesicles or during ATP-dependent export. In addition we determined that AUBF may affect mRNA stability by protecting A+U-rich RNA against degradation by trans-acting, nuclear matrix-associated and A+U-specific endoribonuclease V.

Evidence for a direct interaction of Rev protein with nuclear envelop mRNA-translocation system

The interaction of the Rev protein from human immunodeficiency virus type 1 (HIV-1) with the nucleocytoplasmic mRNA-transport system was investigated. In gel-shift assay, the recombinant Rev protein used in this study selectively bound to the Rev-responsive element (RRE) region of HIV-1 env-specific RNA. Nitrocellulose-filter-binding studies and Northern/Western-blotting experiments revealed an association constant of approximately 1 x 10(10) M-1. The Rev protein also strongly bound to isolated nuclear envelopes from H9 cells, containing the poly(A)-binding site (= mRNA carrier) and the nucleoside triphosphatase (= NTPase), which are thought to be involved in nuclear export of poly(A)-rich mRNA. Binding of 125I-Rev to a 110-kDa nuclear-envelope protein, the putative mRNA carrier, could be demonstrated in in vitro experiments. Both efflux of cellular poly(A)-rich RNA, such as actin RNA [but not efflux of poly(A)-free RNA] from isolated nuclei and the nuclear-envelope NTPase activity were strongly inhibited by Rev protein. On the other hand, transport of viral env RNA, containing the Rev-responsive element, was increased in the presence of Rev. Studying the release of RNA from closed nuclear-envelope vesicles containing entrapped RNA, the action of Rev was found to occur at the level of translocation of RNA through the nuclear pore. Evidence is presented that Rev down-regulates the NTPase-driven transport of mRNA lacking the RRE, most likely via binding to the mRNA carrier within the envelope. In contrast to the efflux of RRE-free RNA, ATP-dependent efflux of RRE-containing RNA from resealed nuclear-envelope vesicles was found to be increased, if the RNA was entrapped in the vesicles together with Rev protein. In addition, it was found that phosphorylated Rev, which is transported together with RRE-containing RNA out of the vesicles, becomes dephosphorylated during transport. In the vesicle experiments it is demonstrated for the first time that a protein selectively channels a specific mRNA across the nuclear-envelope pore complex.

Evidence for involvement of a nuclear envelope-associated RNA helicase activity in nucleocytoplasmic RNA transport

It seems well established that translocation of at least some mRNAs through the nuclear pore is (1) an energy-dependent process, and (2) dependent on the presence of the poly(A) segment attached to most mRNA species. We describe that RNA helicase (RNA duplex unwindase) activity is present in a nuclear envelope (NE) preparation, which also appears to be involved in nucleocytoplasmic RNA transport. This activity unwinds RNA: RNA hybrids. The helicase has a pH optimum of 7.5 and a temperature optimum of 30 degrees C. Applying the sealed NE vesicle system, it was shown that duplex RNA species are readily released from the vesicles in an unidirectional manner, in contrast to single-stranded RNA, which is much slower transported into the extravesicular space. Attachment of a poly(A) segment to the RNA duplex additionally increases the efflux rate of this RNA. Efflux of duplex RNA but not efflux of single-stranded RNA was strongly inhibited by formycin B 5'-triphosphate. Our results suggest that, besides poly(A), duplex structures, if present in a given RNA, modulate and control the export of RNA.

GTP-binding proteins in rat liver nuclear envelopes

Nuclear transport as well as reassembly of the nuclear envelope (NE) after completion of mitosis are processes that have been shown to require GTP and ATP. To study the presence and localization of GTP-binding proteins in the NE, we have combined complementary techniques of [alpha-32P]GTP binding to Western-blotted proteins and UV crosslinking of [alpha-32P]GTP with well-established procedures for NE subfractionation. GTP binding to blotted NE proteins revealed five low molecular mass GTP-binding proteins of 26, 25, 24.5, 24, and 23 kDa, and [alpha-32P]GTP photoaffinity labeling revealed major proteins with apparent molecular masses of 140, 53, 47, 33, and 31 kDa. All GTP-binding proteins appear to localize preferentially to the inner nuclear membrane, possibly to the interface between inner nuclear membrane and lamina. Despite the evolutionary conservation between the NE and the rough endoplasmic reticulum, the GTP-binding proteins identified differed between these two compartments. Most notably, the 68- and 30-kDa GTP-binding subunits of the signal recognition particle receptor, which photolabeled with [alpha-32P]GTP in the rough endoplasmic reticulum fraction, were totally excluded from the NE fraction. Conversely, a major 53-kDa photolabeled protein in the NE was absent from rough endoplasmic reticulum. Whereas Western-blotted NE proteins bound GTP specifically, all [alpha-32P]GTP photolabeled proteins could be blocked by competition with ATP, although with a competition profile that differed from that obtained with GTP. In comparative crosslinking studies with [alpha-32P]ATP, we have identified three specific ATP-binding proteins with molecular masses of 160, 78, and 74 kDa. The localization of GTP- and ATP-binding proteins within the NE appears appropriate for their involvement in nuclear transport and in the GTP-dependent fusion of nuclear membrane vesicles required for reassembly of the nucleus after mitosis.

Differential effect of insulin and epidermal growth factor on the mRNA translocation system and transport of specific poly(A+) mRNA and poly(A-) mRNA in isolated nuclei

The efficiency of efflux of rapidly labeled poly(A)-containing mRNA from isolated rat liver nuclei was found to be modulated by insulin and epidermal growth factor (EGF) in a biphasic but opposite way. At physiological concentrations (10 pM insulin and 1 pM EGF), maximal stimulation of the transport rate by insulin (to 137%) and maximal inhibition by EGF (to 69%) were obtained; at higher concentrations (greater than 100 pM and greater than 10 pM, respectively), the amount of poly(A)-containing mRNA released into the postnuclear supernatant was nearly identical with the level found in untreated nuclei (= 100%). Using mRNA entrapped into closed nuclear envelope (NE) vesicles as a model system, it was found that the modulation of nuclear efflux of mRNA by the two growth factors occurs at the level of translocation through the nuclear pore. The NE nucleoside-triphosphatase (NTPase) activity, which is thought to mediate nucleocytoplasmic transport of at least some mRNAs, responded to insulin and EGF in the same manner as the mRNA transport rate. The increase in NTPase activity caused by insulin and the decrease in NTPase activity caused by EGF were found to be due to changes of the maximal catalytic rate; the Michaelis constant of the enzyme remained almost constant. Investigating the effect of the two growth factors on transport of specific mRNAs, poly(A)-containing actin mRNA was found to display the same alteration in efflux rate as rapidly labeled, total poly(A)-containing mRNA. In contrast, efflux of histone H4 mRNA, which lacks a 3'-poly(A) sequence, decreased in response to insulin and reached minimum levels at the same concentration at which maximum levels of actin mRNA transport rate were obtained. Studying the mechanism of action of insulin and EGF on NE mRNA translocation system, insulin was found to cause an enhancement of NE-associated phosphoprotein phosphatase activity, resulting in a dephosphorylation of the NE poly(A) binding site (= mRNA carrier) and, hence, in a decrease in its affinity to poly(A) [the poly(A) binding affinity of the poly(A)-recognizing mRNA carrier within the envelope is increased after phosphorylation]. EGF, on the other hand, stimulated the protein kinase, which phosphorylates the carrier, and, hence increased the NE poly(A) binding affinity. Because the stage of phosphorylation of the mRNA carrier (which is coupled with the NTPase within the intact NE structure) is inversely correlated with the activity of the NTPase, an enhancement of poly(A)-containing mRNA transport rate by insulin and an inhibition by EGF are observed.

Interaction of a nuclear location signal with isolated nuclear envelopes and identification of signal-binding proteins by photoaffinity labeling

The nuclear envelope (NE) separates the two major compartments of eukaryotic cells, the nucleus and the cytoplasm. Recent studies suggest that the uptake of nuclear proteins into the nucleus is initiated by binding of nuclear location signals (NLSs) contained within these proteins to receptors in the NE, followed by translocation through the nuclear pore complex. To examine the binding step without interference from intranuclear events, we have used a system consisting of (i) purified rat liver NEs fixed onto glass slides and (ii) the prototype simian virus 40 large T antigen (SV40 T) NLS conjugated to nonnuclear carrier proteins, and we have visualized the receptor-ligand interaction by indirect immunofluorescence. In this system, incubation of isolated NEs with the wild-type SV40 T NLS conjugate with carrier proteins resulted in binding that was signal sequence-dependent, could be competitively blocked with excess conjugated and unconjugated wild-type peptide, did not require ATP, and was not affected by the transport-inhibiting lectin wheat germ agglutinin. In contrast, only minimal binding was observed with a mutant SV40 T NLS conjugate. These results are consistent with those obtained in other, more complex in vitro systems and suggest that binding of the SV40 T NLS is receptor-mediated. Binding is largely abolished by extraction of the NE with the nonionic detergent Triton X-100, suggesting that the receptor is soluble in detergent. We find in the Triton X-100 supernatant four major NLS-binding proteins with apparent molecular masses of 76, 67, 59, and 58 kDa by photoaffinity labeling with a highly specific crosslinker, azido-NLS. The reduced complexity of the system described here should be useful for the functional study of other potential NLSs for the identification and isolation of their binding sites and for the screening of antibodies raised against these binding sites.

Organization of the nuclear pore complex in Saccharomyces cerevisiae

Fractions enriched for nuclear pore complexes (NPCs) have been isolated from Saccharomyces cerevisiae. The sequential extraction of nuclei with detergent, nucleases, and salt reveals an organization of the yeast NPC similar to other eukaryotes. Yeast NPCs contain a 30-nm "ring" structure not previously described in other organisms. This structure appears to organize 10-nm filaments into an assembly which exhibits an eight-fold rotational symmetry. Some proteins in the NPC fraction are capable of forming intermediate-sized filaments. These studies suggest that some component of the nuclear pore complex organizes an interaction between nuclear and cytoplasmic networks of intermediate filaments.

Synthesis of polyphosphoinositides in nuclei of Friend cells. Evidence for polyphosphoinositide metabolism inside the nucleus which changes with cell differentiation

Previous work demonstrated the existence of phosphatidylinositol kinase and phosphatidylinositol phosphate kinase in rat liver nuclei, with the suggestion that these activities are in the nuclear membrane [Smith & Wells (1983) J. Biol. Chem. 258, 9368-9373]. Here we show that highly purified nuclei from Friend cells, washed free of nuclear membrane by Triton, can incorporate radiolabel from [gamma-32P]ATP into phosphatidic acid, phosphatidylinositol phosphate and phosphatidylinositol 4,5-bisphosphate. The degree of radiolabelling of phosphatidylinositol bisphosphate is highly dependent on the state of differentiation of the cells, being barely detectable in growing cells and much greater after dimethyl sulphoxide-induced differentiation; this difference is mostly due to different amounts of phosphatidylinositol phosphate in the isolated nuclei. We suggest that polyphosphoinositides are made inside the nucleus and that they have a role in chromatin function; either the phospholipids themselves play a role, or there is a possibility of intranuclear signalling by inositide-derived molecules.

Nuclear-envelope vesicles as a model system to study nucleocytoplasmic transport. Specific uptake of nuclear proteins

In the preceding paper [Riedel & Fasold (1987) Biochem. J. 241, 203-212] we have described a procedure for the preparation of nuclear-envelope vesicles (NE vesicles) from rat liver nuclei. These vesicles, which are largely free of components of the nuclear interior, were employed in an assay system in vitro to study protein translocation across the NE. We found that nuclear proteins such as histones, high-mobility-group proteins and acidic chromosomal proteins are specifically taken up and accumulated in the NE vesicles, whereas there is little or no affinity for non-nuclear proteins like immunoglobulin, myoglobin and cytochrome c. The kinetics of histone uptake into the NE vesicles are similar to those obtained for whole rat liver nuclei, and comparative studies with non-vesicular NEs prepared by deoxyribonuclease I-treatment (DNAase-NEs) indicate that the NE of the vesicles affects the uptake kinetics and increases the capacity for nuclear proteins. The uptake of histones into NE vesicles, but not the binding to DNAase-NEs, can be stimulated by GTP and GDP. Furthermore, we found that even very large molecules can be entrapped in the vesicles during their preparation. These results indicate that the NE vesicles might provide a useful system in vitro with which to investigate the structures and mechanisms involved in protein translocation across the NE.