Isolation and Characterization of Neurospora Plasma Membranes*

SUMMARY The isolation and characterization of plasma membranes from a cell wall-less mutant of Neurospora CTUSSU are described. The plasma membranes are stabilized against fragmentation and vesiculation by treatment of intact cells with concanavalin A just prior to lysis. After lysis, the concanavalin A-stabilized plasma membrane ghosts are isolated by low speed centrifugation techniques and the purified ghosts subsequently converted to vesicles by removal of the bulk of the concanavalin A. The yield of ghosts is about 50% whereas the yield of vesicles is about 20%. The isolated plasma membrane vesicles have a characteristically high sterol to phospholipid ratio, Mg2+-dependent ATPase activity and (Na+ + I(+)-stimulated Mg2+ATPase activity. Only traces of succinate dehydrogenase and 5’-nucleotidase are present in the plasma membrane preparations. reports this laboratory and others

From the Department of Biochemistry, University of Colorado School of Medicine, Denver, Colorado 80220 SUMMARY The isolation and characterization of plasma membranes from a cell wall-less mutant of Neurospora CTUSSU are described.
The plasma membranes are stabilized against fragmentation and vesiculation by treatment of intact cells with concanavalin A just prior to lysis. After lysis, the concanavalin A-stabilized plasma membrane ghosts are isolated by low speed centrifugation techniques and the purified ghosts subsequently converted to vesicles by removal of the bulk of the concanavalin A. The yield of ghosts is about 50% whereas the yield of vesicles is about 20%.
The isolated plasma membrane vesicles have a characteristically high sterol to phospholipid ratio, Mg2+-dependent ATPase activity and (Na+ + I(+)-stimulated Mg2+ATPase activity. Only traces of succinate dehydrogenase and 5'-nucleotidase are present in the plasma membrane preparations.
Earlier reports from this laboratory and others have described two glucose transport systems in wild t.ype cells of Neurospora crassa (l-10).
One, a facilitated diffusion system which is present in cells grown in a medium containing high levels of glucose and another, an active transport system which is derepressed in cells grown in a medium containing little or no glucose. With these two systems as models for eukaryote membrane transport, the primary goal in this laboratory is an understanding of the molecular events underlying the two fundamental aspects of transport, translocation and energy-coupling. Remarkable progress toward this goal has been realized by bacterial transport workers utilizing isolated bacterial plasma membrane vesicles (11).
That progress made obvious the need for a similar preparation of Neurospora plasma membrane vesicles. The isolation of plasma membranes from free living cells other than erythrocytes, isolated fat cells, and bacteria, in high yield and purity, is a task replete with difficulties.
The specific problems which arose in our early attempts to isolate N. crassa plasma membranes were the unavailability of a suitable plasma membrane marker, the presence of a rigid cell wall which made gentle lysis impossible, the tendency for the Neurospora plasma membranes to fragment and vesiculate immediately upon lysis trap-* This investigation was supported by grants from the National Institutes of Health (AM14479, GM19971) and the National Science Foundation (GB38801).
ping other cellular constituents, and the variable densities of the plasma membrane particles w-hich lead to extensive smearing in standard isopycnic centrifugation procedures.
Most of these difficulties have been experienced by others w-ith a variety of cell types (12)(13)(14).
This communication describes procedures which eliminate all of these problems in the isolation of Neurospora plasma membranes.
The chemical and enzymatic properties of the plasma membranes obtained are also presented. Although the methods described here were designed specifically for the isolation of Neurospora plasma membranes, the principles of the isolation procedure should be applicable to a variety of eukaryotic cells. are harvested by centrifugation at 700 X g for 10 min, resuspended in 80 ml of ice-cold Buffer A, divided into four 20-ml aliquots, and washed five times with 80 ml of ice-cold Buffer A (20 ml each) by alternate centrifugation and resuspension (50.ml glass tubes in a swinging bucket clinical centrifuge at 140 X g for 6 min).
Prior to the final centrifugation, 5 ml of the 80-ml cell suspension are withdrawn, pelleted by centrifugation, and resusnended in 5 ml of ice-cold 0.01 M Tris-HCl. oH 7.5. This is the "ceils" preparation referred to at various points in the text. If the washed cells are to be surface-labeled with diazotized [35S]sulfanilic acid (experiments described in Fig. 1 and Table I only) they are resuspended in a total of 20 ml of diazotized sulfanilic acid in Buffer A prepared as described above and incubated at 25" for 5 min.
After surface-labeling, the cells are then washed five more times in 80 ml of Buffer A as described above. Plasma membranes derived from surface-labeled cells behave the same as plasma membranes derived from nonsurface-labeled cells in all steps of the isolation procedure.
The washed cells are then resuspended in a total of 20 ml of Buffer A (25"), mixed with 20 ml of 0.5 mg per ml of concanavalin A in Buffer A, and incubated with occasional gentle agitation for 10 min at 25". The concanavalin A agglutinates the cells during this period. The concanavalin A-agglutinated cells are then chilled, centrifuged at 140 X g for 1 min, resuspended gently in 40 ml of ice-cold Buffer A, and centrifuged at 140 X g for 6 min. The resulting cell pellet is then resuspended in 50 ml of ice-cold 0.01 M Tris-HCl, pH 7.5, containing 5 mM MgS04 and 50 mg of DNase, and homogenized in a glass-Teflon tissue homogenizer (50 passes over a IO-min period; clearance approximately 0.008 inch). This is the lysate fraction referred to at various points in the text. Twelve-milliliter portions of the lysate are layered over 35 ml of ice-cold Buffer B (0.1 M Tris-HCl, pH 7.5, containing 0.5 M mannitol) and the resulting two-phase systems are centrifuged at 140 X g for 30 min in a swinging bucket clinical centrifuge at 4". The supernatant fluids containing most of the cell contents are removed by aspiration and the plasma membrane pellets are resuspended in a total of 20 ml of ice-cold 0.01 M Tris-HCl, pH 7.5, and again homogenized in a glass-Teflon tissue homogenizer (20 passes, 4").
Ten-milliliter aliquots of the resulting suspension are layered over 35 ml of Buffer B and the resulting two-phase systems are centrifuged at 250 X g for 30 min in a swinging bucket clinical centrifuge (4').
The supernatant fluids are removed by aspiration and the pellets containing the plasma membrane ghosts are resuspended in a small volume of 0.01 M Tris-HCl, pH 7.5. This is the plasma membrane ghost fraction referred to at various points in the text. The plasma membrane ghost fraction is contaminated with significant amounts of a nonmembrane carbohydrat,e material and small amounts of succinate dehydrogenase, 5'.nucleotidase, and RNA, but the high yield, near-purity, and ease of preparation make it a suitable starting material for a variety of applications. membrane vesicles, which remain on top of the Buffer C, are removed with a Pasteur pipette, diluted to 50 ml with ice-cold 0.01 M Tris-HCl, pH 7.5, and centrifuged at 12,000 X g for 30 min (4").
The supernatant fluid is decanted and the pellet conta.ining the plasma membrane vesicles is resuspended in a small volume of 0.01 M Tris-HCl, pH 7.5. This is the plasma membrane vesicle preparation referred to at various points in the text.

Chemical Analyses
Protein and carbohydrate determinations were carried out on "cells," and plasma membrane vesicle preparations which had been dialyzed against 1 liter of 0.01 M Tris-HCl, pH 7.5, overnight (4").
Lipid and nucleic acid determinations were carried out on undialyzed material. Protein-Since the plasma membrane vesicles contain significant amounts of concanavalin A, direct determination of plasma membrane protein was not possible. For this reason, the protein content of the plasma membrane vesicles was determined from the tritium content.
The amount of protein in the "cells" fraction was determined by the method of Lowry et al. (18) and, in addition, the tritium content of the cells was determined by liquid scintillation counting (in PG).2 From the ratio of counts per min per mg of protein in the "cells" fraction, the milligrams of protein in the plasma membrane vesicle preparation were calculated.
This method relies upon the assumption that the total cellular proteins and the plasma membrane proteins are equally labeled, and, in an overnight culture, this is a reasonable assumption.
Lipid-The total lipids present in the "cells" and plasma membrane vesicle preparations were extracted by the method of Bligh and Dyer (19)

Electron Microscopy
Small aliquots of intact cells were withdrawn immediately before and after the concanavalin A treatment and pelleted by centrifugation at 250 X g for 5 min (4').
The cell pellets were then fixed in 3% glutaraldehyde in Buffer A for 2 hours at 4", and postfixed in 1% 0~04 in three-quarters strength Buffer A for 1 hour at 4". The nellets were then washed once with cold Buffer A, dehydrated in a graded acetone series, and embedded in Spurr's resin (28). Thin sections, cut on a Porter-Blum MT-2B ultramicrotome fitted with a diamond knife, were picked up on copper grids, double stained in uranyl acetate and lead citrate, and photographed with a Philips EM300 electron microscope. Plasma membrane ghosts and vesicles prepared as described above were pelleted by centrifugation at 12,000 X g for 30 min (4'), treated with Karnovsky's formaldehyde-glutaraldehyde fixative (29) for 2 hours at 4", rinsed overnight in 0.2 M sodium cacodylate buffer, pH 7.2, postfixed for 1 hour in 1% 0~04 in 0.2 M sodium cacodylate buffer, pH 7.2, and subsequently treated as described above for the intact cells.

Identification
of sl Plasma Membranes-The experiment presented in Fig. 1 shows the behavior of surface-labeled plasma membranes during the first step in the isolation procedure. A lysate of %-surface-labeled, tritiated cells was obtained essentially as described under "Experimental Procedure" (one-tenth scale), and 3.5 ml were layered over a two-phase system containing 13 ml of Buffer B over 1 ml of 80% (w/v) sucrose in an l&ml cellulose nitrate centrifuge tube. The 80% sucrose cushion was included to stop the plasma membrane fraction from pelleting, thus facilitating the collection of all the material in the tube. The resulting discontinuous gradient was centrifuged at 140 x g for 30 min in a swinging bucket clinical centrifuge (4"), fractionated from the top in an ISCO density gradient fractionator, and the fractions obtained were dialyzed and counted in PG.
It can be seen that the majority of the Wsurface label sediments to the bottom of the Buffer B whereas most of the intracellular constituents (3H) remain at the top.
A second centrifugation of the sedimented material through Buffer B results in a similar distribution of surface label and further purification. Yields-Unlike the marker enzyme approach for plasma membrane identification, the surface-labeling approach permits estimation of the yield of plasma membranes obtained in a given procedure. Table I  See text for details.
were obtained from %S-surface-labeled, tritiated cells as described under "Experimental Procedure," dialyzed, and counted for 35S and 3H content.
The amounts of the total 3% surface label recovered in the ghost and vesicle preparations indicate a yield greater than 50% for t.he plasma membrane ghosts and a yield greater than 20% for the plasma membrane vesicles. Electron Microscopy-Presented in Fig. 2 are electron photomicrographs of samples taken at various points in the plasma membrane isolation procedure. Fig. 2A represents the intact cell surface just prior to the concanavalin A treatment, and Fig.  2B represents the cell surface just after the concanavalin A step. The effect of the accumulation of concanavalin A on the cell surface can be clearly seen. The size of the particles suggests that they are aggregates rather than individual concanavalin A molecules. Fig. 2, C and D, are low and high magnification pictures of the plasma membrane ghost preparation. The presence of concanavalin A particles (on one side only) is quite evident and plasma membranes at this stage exist primarily as nonvesiculated sheets. Fig. 2, E and F, are low and high magnification pictures of the plasma membrane vesicle preparation.
Most of the concanavalin A aggregates are removed and the plasma membranes exist predominantly as small vesicles. Chemical Composition-The chemical composition of the "cells" and plasma membrane vesicle preparations is presented in Table  II.
Since the plasma membrane ghost preparation is contaminated with significant amounts of a nonmembrane carbohydrate material, and small amounts of RNA, succinate dehydrogenase, and 5'-nucleotidase, the chemical analysis of the ghost preparation is not presented.
The plasma membrane vesicles contain large amounts of sterol and the molar ratio of sterol to phospholipid is around 1.3. A high sterol to phospholipid ratio is characteristic of all eukaryote plasma membranes studied so far (12,30). From the yield data in Table I it can be calculated that the plasma membranes of intact cells contain approximately 60% of all the cellular sterol, which is clearly not t,he case for any other chemical constituent of the plasma membranes measured. This observation is suggestive of a sine qua non relationship between eukaryote plasma membranes and sterols. The carbohydrate content of the plasma membrane vesicle preparation represents about 147, of the mass of the plasma membrane vesicles. Only very small amounts of RNA and DNA are present in the plasma membrane vesicle preparation. Enzyme Activities- Table  III summarizes the enzyme activities thus far detected in the plasma membrane ghost and plasma membrane vesicle preparations. The enzyme activities present in the "cells" preparation are included for comparison. The plasma membrane ghosts and vesicles have a Mg"+ATPase ac-  There is little contamination of the ghost and vesicle preparations by nuclei as indicated by the extremely low DNA content of these preparations.
Contamination by mitochondria is minimal as indicated by the low succinate dehydrogenase activities in the preparations. Significant contamination by endoplasmic reticulum is unlikely due to the extremely low g forces employed in the isolation procedure and the low RNA content of the preparations. The electron photomicrographs shown in Fig. 2 support the above conclusions.
In addition, another criterion of purity arises from the yield experiment described in Table I. Since the amount of Wlabeled protein and % surface label are lost at a constant ratio in the conversion of ghosts to vesicles it can be concluded that there is little Wlabeled protein in the ghost preparation which is not plasma membrane in origin.

DISCUSSION
As pointed out in the Introduction, the isolation of plasma membranes from most eukaryotic cells in high yield and purity is a difficult task.
The major problems which impeded our efforts to isolate Neurospora plasma membranes were the unavailability of a suitable plasma membrane marker, the presence of a rigid cell wall \\hich makes gentle lysis impossible, the tendency for the Neurospora plasma membranes to fragment and vesiculate immediately upon lysis trapping other cellular constituents, and the variable densities of the plasma membrane particles which lead to extensive smearing in standard isopycnic centrifugation procedures.
The unavailability of a suitable plasma membrane marker was the first problem to be surmounted.
In a previously uncharacterized free living cell, t,he only predictable difference between the plasma membrane and other cellular membranes is that the plasma membrane is exposed to the esternal environment. Thus, surface labeling by membrane impermeable reagents (see Ref. 12 for a review) seems to be the most dependable method for identifying plasma membranes in any isolation procedure involving free living cells.
In addition to providing a method for confidently identifying plasma membranes, the surface-labeling approach allows for an accurate estimation of the yield of plasma membranes in a given procedure. This is not possible with the marker enzyme approach. The surface-labeling approach to plasma membrane isolation has been applied recently in the isolation of a yeast plasma membrane fraction (31). The rigid cell wall problem was obviated simply by switching from the wild type strain to a cell wall-less mutant of Neurospora.
The extreme problems which arise when plasma membranes fragment and vesiculate immediately upon lysis can be solved only by somehow stabilizing the plasma membranes prior to lysis so that they do not spontaneously fragment and vesiculate and may be isolated as sheets or "ghosts" which represent the intact plasma membranes freed of intracellular contents. Treatment with Zn2+ or fluorescein mercuric acetate have been reported as methods for plasma membrane stabilization (32) but those methods were unsatisfactory for Neurospora plasma membrane stabilization.
Treatment of the intact sl cells with concanavalin A just prior to lysis leads to a remarkable stabilization of the plasma membranes and the resulting stabilized plasma membrane ghosts are visible entities in the light microscope and can be isolated in a nearly pure state by simple low speed centrifugation procedures.
During chemical analyses of the plasma membrane