Isolation and Characterization of Nuclear Envelopes from the Yeast

We have developed a large scale enrichment procedure to prepare yeast nuclear envelopes (NEs). These NEs can be stripped of peripheral proteins to produce a heparin-extracted NE (H-NE) fraction highly enriched in integral membrane proteins. Extraction of H-NEs with detergents revealed previously uncharacterized ring structures associated with the NE that apparently stabilize the grommets of the nuclear pore complexes (NPCs). The high yields obtained throughout the fractionation procedure allowed balance-sheet tabulation of the subcellular distribution of various NE and non-NE proteins. Thus we found that 20% of endoplasmic reticulum (ER) marker proteins are localized at the NE. Using a novel monospecific mAb made against proteins in the H-NE fraction and found to be directed against the pore membrane protein POM152, we showed that while the majority of POM152 is localized in the NE at the NPC, a proportion of this protein is also present in the ER. This ER pool of POM152 is likely to be involved in the duplication of nuclear pores and NPCs during S-phase. Both the NEs and H-NEs were found to be competent for the in vitro posttranslational translocation of preproc~-factor. They may also be suitable to investigate other ERand NE-associated functions in cell-free systems. T HE nuclear envelope (NE) 1 defines the boundary of the nucleus in eukaryotic cells and is composed of two distinct membranes enclosing a lumenal (perinuclear) space. Facing the nucleoplasm is the inner nuclear membrane, which in higher eukaryotes is often lined by a filamentous network called the nuclear lamina. Towards the cytoplasm is the outer nuclear membrane which is continuous with the ER membranes and is thought to perform rough ER functions. The inner and the outer nuclear membranes join to form specialized circular apertures containing the nuclear pore complexes (NPCs), which regulate the exchange of material between the nucleus and cytoplasm. It is widely assumed that the NE plays a role in the control of nuclear architecture both during interphase and at mitosis. More specifically, it has been proposed that the NE could contribute in defining the spatial distribution of specific segments of the genome such as the telomeres inside the nucleus, thereby facilitating the regulation of DNA transcription and replication. Some enzymatic activities may also be restricted to the NE; for example it is likely that proteins involved in the nucleus-speAddress correspondence to Dr. G. Blobel, Laboratory of Cell Biology, Howard Hughes Medical Institute, The Rockefeller University, New York, NY 10021. Ph.: (212) 327-8096; Fax: (212) 327-7880. 1. Abbreviations used in this paper. CM, crude microsome; gpaF, glycosylated pro-a-factor; H-NE, heparin-extracted NE; MT, microtubule; NE, nuclear envelope; NPC, nuclear pore complex; ppaF, prepro-a-factor; PVP, polyvinylpyrrolidone; SPB, spindle pole body. cific phosphoinositide metabolism are localized to the inner nuclear membrane. Despite considerable progress in the past few years, the molecular details of many important functions of the NE still remain poorly defined (for reviews see Hurt et al., 1992; Gilson et al., 1993; Rout and Wente, 1994; Moore and Blobel, 1994; Kilmartin, 1994). In the yeast Saccharomyces, morphometric studies indicate that the NE represents roughly 30% of the functional rough ER (Preuss et al., 1991). Thus, the isolation of NE from yeast would be useful for the study of both NE and rough ER functions. Furthermore, yeast present numerous advantages over higher eukaryotes as a system to study these functions. They have neither the complications of developmental regulation of nuclear processes, nor of nuclear disassembly, having a closed mitosis; in addition, the genetics and molecular biology of yeast are better understood than in any other eukaryote, and a large program is under way to complete the entire yeast genome sequence by the end of the decade (Maddox, 1992). Unfortunately the cell biological and biochemical characterization of cellular membranes and compartments in budding yeast remains incomplete, and would benefit from the development of rigorous cellular fractionation techniques comparable to the ones available for higher eukaryotes. We describe here a procedure for the preparation of a highly enriched NE fraction from the yeast Saccharomyces. NEs were prepared from yeast nuclei on a large scale and in high yield. To understand the relationship between peripheral and integral membrane components that define © The Rockefeller University Press, 0021-9525/95/10/19/13 $2.00 The Journal of Cell Biology, Volume 131, Number 1, October 1995 19-31 19 on A ril 2, 2006 w w w .jc.org D ow nladed fom the various functions of the NE, we prepared a highly enriched nuclear membrane fraction by stripping the NE fraction with heparin. The fractionation pattern of representative markers throughout the procedure was used to tabulate the distribution of various cellular organelles and functions within the cell. Heparin-extracted NEs (H-NEs) were used to raise a panel of mAbs, one of which is described in this paper. Both the NE and H-NE fractions were shown to be functional in an ER protein translocation assay and thus retain one of numerous potentially testable functions that are associated with the NE in vivo. Detergent extraction of H-NEs showed that ring structures associated with the NE are likely to be responsible for anchoring the NPCs to their grommets and stabilizing the pore membrane domain. Materials and Methods Yeast Subcellular Fractionation: Preparation of Enriched Nuclei and Highly Enriched NPCs The yeast strain Saccharomyces uvarum (NCYC 74, ATCC 9080; American Type Culture Collection, Rockville, MD), considered a strain of Saccharomyces cerevisiae (Mortimer and Johnson, 1986), was used throughout the procedure. Enriched nuclei were prepared as previously described (Rout and Kilmartin, 1990, 1994). Briefly, 70-90 g of mid-log phase cells were obtained from a 36 liter yeast culture. Cells were harvested and converted to spheroplasts in 1.1 M sorbitol (Rout and Kilmartin, 1994). Spheroplasts were harvested by centrifugation and then lysed in 300 ml of polyvinylpyrrolidone (PVP) solution (8% PVP, 20 mM K-phosphate, pH 6.5, 0.75 mM MgCl2). The cell lysate (fraction 1) was separated by centrifugation (15 min at 10,000 g) into crude cytosol (fraction 2) and a crude nuclei pellet (fraction 3). The nuclei were resuspended in 144 ml of 1.7 M sucrose in PVP solution and this suspension was divided into 12 equal aliquots. Each aliquot was overlayered over a three step sucrose/PVP gradient (8 ml each of 2.01 M sucrose, 2.10 M sucrose, and 2.30 M sucrose in PVP solution) in a SW28 tube (Beckman Instruments, Palo Alto, CA). The gradients were centrifuged in a rotor (Beckman SW28 ) at 28,000 rpm (103,000 g) for 4 h at 4°C. After centrifugation fractions were collected from the top. The load zone, including a thick layer at the top of the tube (fraction 4), and a thick yellowish band at the load/2.01 M interface (fraction 5), both contained intact mitochondria, vesicles and microsomes (as judged by EM). Very little material was present in the third gradient fraction at the 2.01/2.10 M interface (fraction 6). A dense white band at the 2.10 M/2.30 M interface contained the bulk of the nuclei (fraction 7). The bottom of the gradient (fraction 8) included a pellet composed mainly of cells remnants. Highly enriched NPCs were prepared from nuclei (fraction 7) exactly as described in Rout and Blobel (1993). Yeast Nuclear Envelope Preparation NEs were prepared from the enriched nuclear fraction (fraction 7). The OD at 260 nm of the nuclear fraction was measured after 1 in 100 dilution in 1.0% SDS; approximately 1,000-2,000 OD260nms were obtained from a 36 liter preparation. The nuclear fraction was adjusted to a refractive index of 1.4340 with PVP solution and centrifuged at 145,000 g for 1 h at 4°C. The supernatant was carefully but thoroughly removed by aspiration, and the tubes placed on ice. Typically, 20 ml of freshly prepared, ice cold bt-DMSO (bt buffer [10 mM bisTris-Cl, pH 6.5; 0.1 mM MgC12] containing 20% [vol/vol] DMSO) in the presence of 20 Ixg/ml DNase I-EP (Sigma Chemical Co., St. Louis, MO) and 1% (vol/vol) solution P (18 mg/ml PMSF, 0.3 mg/ml pepstatin A in absolute ethanol), were added to 2,000 OD260 nm of nuclei. This was followed immediately by vigorous vortexing at 4°C until the pellet was completely resuspended. The suspension was then incubated at room temperature (~25°C) for 5-10 min. After incubation, the lysed nuclei were placed back on ice and 60 ml of 2.1 M sucrose, 20% Nycodenz (Accudenz; Accurate Chemical and Scientific, Westbury, NY) in bt buffer in the presence of 0.1% (vol/vol) solution P, were added, and thoroughly mixed. The suspension was divided into six tubes (Beckman SW28) and overlayered with 12 ml of 2.0 M sucrose (R.I. = 1.4295) and 12 ml of 1.5 M sucrose (R.I. = 1.4055) in bt buffer containing 0.1% (vol/vol) solution P. The tubes were centrifuged in a rotor (Beckman SW28) at 28,000 rpm (103,000 g) for 24 h at 4°C. The tubes were unloaded from the top using a hand-held pipette. A faint white band at the top of the tube was completely removed (~6.0 ml collected per tube; fraction 9). The NEs were found at the 1.5 M/2.0 M interface, appearing as a broad, white, slightly flocculent band (,'~12.0 ml collected per tube; fraction 10). Next was a dense, sharp yellowish/white band containing a few NEs, chromatin, and cell remnants (~12.0 ml collected per tube; fraction 11). The final ~7.0 ml collected (fraction 12), including a dense brownish/white pellet, contained soluble and particulate matter mainly derived from chromatin. Extraction of Yeast Nuclear Envelopes For heparin extraction, 0.6 ml (~0.4 mg of protein) of the yeast NE fraction were mixed with 2.4 ml of a solution containing 10 mg/ml heparin (Sigma Chemical Co.), 0.1 mM DT]? and 0.5% (vol/vol) solution P in bt buffer. After 1 h on ice, 50 iLg/ml RNase A was added and the incubation was continued for 15 min at 10°C. The sample was over-layered onto two 1-ml layers of 1.0 and 2.0 M sucrose in bt buffer containing 0.1% (vol/vol) solution P, and centrifuged in a rotor (Beckman SW55) at 45,000 rpm (~192,000 g) for 30 min at 4°C. The tube was unloaded from the top using a hand-held pipette. The first fraction (~2 ml; fraction 13) contained the bulk of the solubilized proteins. The next fraction (~1.8 ml; fraction 14) contained some of the soluble proteins together with a few of the NE membranes. The bulk of H-NE membranes was recovered at the 1.0 M/2.0 M sucrose interface and appeared as a tight white band (~0.4 ml; fraction 15). The last fraction (~0.8 ml; fraction 16) sometimes contained small amounts of H-NEs. The yeast NE fraction was extracted with sodium carbonate using a previously described method (Wozniak et al., 1994). Posttranslational Translocation Assay The procedures for the preparation of yeast nuclei and NEs described above were modified to maintain the ER-translocation activity throughout the fractionation procedure. Firstly, yeast spheroplasts were allowed to recover in YPD medium (1% yeast extract, 2% peptone, 2% dextrose) containing 1.0 M sorbitol for 30 min at room temperature before lysis. Secondly, all the solutions starting from the lysis buffer and including all the gradient solutions were supplemented with 2 mM DTT. Finally, the MgC12 concentration in the nuclear lysis buffer and in the solutions used for the NE flotation gradient (fractions 9-12) was raised from 0.1 to 0.5 mM. The degree of enrichment of "active" NEs was shown to be similar to that obtained with the original method (data not shown). Just before the in vitro protein translocation reaction the "active" NE fraction was concentrated 20-fold by pelleting at 70,000 g for 30 min, and gently resuspending in solution A (20 mM Hepes-KOH, pH 7.4, 100 mM KOAc, 2 mM Mg[OAc]2, 2 mM DTT) containing 0.25 M sucrose. The heparin extraction of the "active" NE fraction was carried out as described above except that 2 mM DTT was added to all solutions and gradients, the RNase A digestion step was omitted and the H-NEs were pelleted through a 1.0 M sucrose cushion (1.0 M sucrose, 2 mM DTI', 0.5 mM MgCl:, 0.5% [vol/vol] solution P in bt buffer) instead of being recovered over 2.0 M sucrose. Heparin traces were removed by resuspending the "active" H-NEs pellet obtained from 2.4 ml of NEs, in 2.4 ml of 0.5 M KCI, 2 mM DTT, 0.25 M sucrose, 0.5% (vol/vol) solution P and incubating the suspension for 1 h on ice. The membranes were recovered by centrifugation through a 0.5 ml, 0.6 M sucrose cushion (0.6 M sucrose; 10 mM bisTris-C1, pH 6.5; 0.5 mM MgC12; 0.1% [vol/vol] solution P), at 39,000 rpm (~100,000 g) in a TLS-55 Beckman rotor for 1 h at 4°C. The supernatant from the 0.5 M KCI wash was shown not to contain significant amounts of extracted proteins (data not shown). Finally, the sample was resuspended in a volume of 0.25 M sucrose in solution A equal to roughly 2.5% of the initial NEs volume. Yeast crude microsomes (CMs), used as a positive control for the ER translocation reaction, and yeast crude cytosol were prepared as described (Waters et al., 1986; Waters and Blobel, 1986). [35S]Methionine labeled prepro-a-factor (ppaF) was synthesized using a wheat germ in vitro translation kit (Promega Biotec, Madison, WI) following the specifications of the manufacturer. Immediately before use, the translation mixture containing ppuF was diluted with 3 vol of 8 M urea and incubated for 10 min at 20°C. The translocation reaction and the protease protection assays were performed as described (Waters and Blobel, 1986; Chirico et al., 1988). Typically, the translocation mix (total volume 150 pA) consisted of the following: 43.6 }xl of "master mix" (14.4 mM Hepes-KOH, pH 7.4; 276 The Journal of Cell Biology, Volume 131, 1995 20 on A ril 2, 2006 w w w .jc.org D ow nladed fom mM KOAc; 1.0 mM Mg[OAc]2; 1.0 mM DTT; 1.7 mM ATP; 86 mM creatine phosphate; 0.7 mg/ml creatine kinase; 0.07 mM GDP-mannose; 0.07 mM UDP-glucose; 0.07 mM UDP-N-acetylglucosamine; 1.4% glycerol), 90 ~1 of yeast crude cytosol in solution A containing 1.0 mM Mg-ATt', and 14 ILl of CM, NE, or H-NE membranes in solution A containing 0.25 M sucrose. This mixture was pre-incubated at 20°C for 5 rain and the import reaction was then started by the addition of 2.4 ILl of urea-denatured translation product. At the end of the reaction the sample was divided in three equal aliquots. 30 ixl of water were added to the first aliquot. 10 ill of 8 mM CaCI 2, 10 Ixl of water and l0 Ixl of 800 ixg/ml trypsin were added to the second aliquot. The third aliquot was treated as the second except that 10 ixl of water were substituted with 10 ixl of 8% (vol/vol) Triton X-100. All aliquots were incubated on ice for 30 min and the reactions were stopped by the addition of 10 ~1 of 50 mM PMSF. After an additional 10 min on ice, the samples were TCA precipitated and analyzed by SDSPAGE and fluorography.