THE Jounrn~ OF BIOLOGICAL CHEMISTRY

Val 252, No. 6, Issue of March 25, pp. 2134-2142, 1977 Prmted in U.S.A.

Isolation and Characterization of Plasma Membranes and Intact Nuclei from Lymphoid Cells*

(Received for publication, August 6, 1976)

MARTI JETT, T. M. SEED,$ AND G. A. JAMIESON

From the American National Red Cross, Blood Research Laboratory, Bethesda, Maryland 20014

A method has been developed for the rapid large scale isolation of plasma membranes and intact nuclei from RAJI lymphoid cells utilizing hypotonic lysis of cells after intra- cellular loading with glycerol followed by combined flota- tion-sedimentation within a discontinuous sucrose gradient. Nuclei may be isolated in about 1 h and plasma membranes in about 6 h from 1 to 20 g of cells.

Intact nuclei, obtained in 90 to 95% yield based on lysed cells, was isolated by differential centrifugation and con- tained 16% DNA and about 30% of total cell sialic acid.

A crude plasma membrane fraction was isolated by cen- trifugation onto a cushion of 38% sucrose (d 1.1683) and subsequently resolved into two subfractions. The less dense vesicles had an average d 1.127 and showed a ‘I-fold increase in specific activity for thymidine phosphodiesterase while the more dense (d 1.151) had a 20-fold concentration of enzyme activity. Activity of enzymes indicative of contami- nation with lysosomes, microsomes, mitochondria, and cy- toplasm was negligible in these plasma membrane fractions. The less dense vesicles had a cholesterol:phospholipid ratio of 0.97 which was higher than that of the more dense vesicles (0.69). Otherwise, the analytical values for the two types of membrane vesicles were similar as both fractions contained like percentages of protein C-30%), lipid t-30%), and carbo- hydrate (-15%) with trace amounts of RNA and DNA. Twenty-five per cent of the total cell sialic acid was in the plasma membrane fractions.

The initial event in lectin-induced transformation of lym- phocytes is the interaction of mitogen with the cell surface. In the first paper of this series (1) we have shown that one of the effects of transformation of peripheral blood lymphocytes with phytohemagglutinin (PHA) of Phase&s uulgaris is a 20- to 40-fold stimulation of incorporation of radioactive n-glucosa- mine and that three different size classes of radioactive prod- ucts could be isolated by proteolysis of a membrane fraction.

The limited amount of material available from this source precluded the further characterization of these membrane components and their possible role in initiating or modulating the response of cells to a mitogenic stimulus. However, RAJI, a Burkitt lymphoma line can also be stimulated with concana-

* This work was supported, in part, by United States Public Health Service Grant AI 09017. Publication 342 from American National Red Cross.

$ Present address, Division of Biological and Medical Research, Argonne National Laboratory, Argonne, 111. 60438.

valin A’ as can certain other cell lines with other lectins (2). We have utilized RAJI cells to show that two size classes of radioactive product were directly associated with the lymphoid cell membrane but that the third component, although tran- siently associated with the plasma membrane, was secreted from the cellular cytoplasm (3).

These studies have required the rapid and frequent isolation of plasma membranes on a large scale. While a multitude of methods for the isolation of plasma membranes has been developed in the past few years these are frequently not appli- cable to cell types other than that for which they were divel- oped or require inordinately long times to process small quan- tities of material which are insufficient for chemical character- ization.

The method reported here permits the rapid processing of from 1 to 20 g of cells. The disruption procedure is based on the glycerol-lysis method initially developed for human platelets (4), but has been extensively revised to reflect the greater permeability of lymphoid cells to glycerol. A novel method has been developed for fractionation of the subcellular organelles which involves the simultaneous flotation and sedimentation in a stepwise sucrose gradient to effect a rapid purification of plasma membrane vesicles. In addition, this method permits the isolation of intact cell nuclei in high yields from the original homogenate.

MATERIALS AND METHODS

Cell Cultures RAJI, a human Burkitt lymphoma cell line, was obtained from

American Type Culture Collection (ATCC) cell repository and had undergone 100 subcultures. This suspension cell line was described by ATCC as being virus-free as determined by electron microscopy and free of Epstein-Barr virus as determined by immunofluorescent staining. The methods for cell culture were as previously described (3). Not more than 30 subcultures were carried out; after this the lines were discontinued and new cultures established from the origi- nal stock.

The cells were harvested under aseptic conditions in the log phase of growth (1.5 x 10” cells/ml). They were gently centrifuged from the culture medium in a Sorvall RC-3 centrifuge which had been warmed to 30” at 2000 rpm (800 x g,, for 10 min, HB-4-rotor). The cells were washed twice at 37” with a volume of Earle’s balanced salt solution equal to 10% of the original culture volume.

The concentrations of all sucrose solutions were determined from their refractive indices and adjusted as required since the percentage of water in various batches of ribonuclease-free sucrose (Schwartz and Mann) was quite variable. A refractometer (Bausch and Lomb) with a 20” constant temperature bath was used. A concentrated

’ Observation to be publrshed elsewhere.

2134

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Nuclei and Plasma Membrane Isolation and Characterization 2135

sucrose solution was prepared by adding 85 g of sucrose to 100 ml of water to a give a solution with a refractive index of 1.4418 to 1.4509 corresponding to 60 to 64% sucrose solution. This solution was then adjusted to give the appropriate sucrose concentration, as calculated from refractive index values, in a final concentration of 10 rn~ Tris buffer (pH 7.4).

Lysis Method

The washed cells were resuspended in Earle’s balanced salt solu- tion equivalent to 1% of the original culture volume. A solution of glycerol (90% w/v in Earle’s balanced salt solution) was added in three equal increments 5 min apart to give a final concentration of 30% glycerol. All of the above operations were carried out at 30-37”. Five minutes after the last addition of glycerol, the cell suspension was cooled in an ice bath for 2 min and from this point on all operations were carried out at 4”. The glycerol-loaded cells were centrifuged from the suspension in 250-ml plastic bottles using a Sorvall RC-2B centrifuge with a HB-4 rotor at 2800 ‘pm (1200 x g) for 10 min and the supernatant solution was decanted and discarded.

A volume of buffer (10 rnM Tris/HCl, pH 7.4, containing 1.0 mM MgCl,, 1.0 rnM CaCl,) equal to 0.7 to 1% of the original culture volume was rapidly added to the pelleted cells with vigorous agita- tion and the mixture was allowed to stand in an ice bath for 5 min.

Fractionation Procedure

The fractionation scheme is summarized in Fig. 1.

Differential Centrifugation

In order to remove the nuclei the cell lysate was centrifuged at 700 x g,, in the Sorvall RC-2B centrifuge, HS-4 rotor at 2000 rpm for 10 min. The supernatant solution was removed by aspiration. The pelleted nuclei were resuspended in the same volume of the lysis buffer and centrifuged as before.

The supernatant solutions (“nuclei-free lysate”) combined after pelleting and washing the nuclei were centrifuged at 3300 x g,, using the Sorvall RC-ZB, HS-4 rotor for 10 min at 4200 rpm in order to remove a debris fraction comprising mitochondria. The superna- tant solution (“membrane lysate”) was carefully removed by aspira- tion prior to further purification of the membrane fraction.

Ultracentrifugation

Step 1 -The crude membrane fraction was obtained by sedimenta- tion onto a sucrose cushion: 33 ml of supernatant solution (mem- brane lysate from Part A, Fig. 1) was placed in cellulose nitrate centrifugation tube !ll/z x 3V2 inches) and 38% sucrose (5 ml) was layered underneath the lysate using a Pasteur pipette. The sample was centrifuged in the Sorvall OTD-2 ultracentrifuge using a Beck- man SW 27 swing-out rotor for l’iz h at 26,500 rpm (95,000 x g,, 1. A band, designated as crude membrane fraction, formed at the inter- face. The supernatant solution above this band was removed as completely as possible by aspiration and the crude membrane frac- tion and sucrose cushion were decanted. A residual pellet, which remained adherent to the bottom of the tube, termed “debris,” was resuspended for assay.

Step 2 -Further purification involved the simultaneous flotation and sedimentation of membrane fractions within a discontinuous sucrose gradient and their separation from contaminating subcellu- lar organelles.

The crude membrane fraction was adjusted to a refractive index of n = 1.3829 to 1.3847 (corresponding to 31 to 32% sucrose concentra- tion) by adding the calculated volume of concentrated (approxi- mately n = 1.451) sucrose solution. The following solut,ions, buffered with 10 mM TrisiHCl, pH 7.4, were pipetted into a 38.ml centrifuge tube (2.6 x 9 cm) in the following order: (a) 7 ml of 38% sucrose solution; (b) 20 ml of crude membrane suspension; (~1 15% sucrose solution was then added until the tube was balanced (about 12 ml). The sample was then centrifuged for 2 h in the Sorvall OTD-2 ultracentrifuge using a fixed angle rotor (T-8651 at 63,000 rpm (430,000 x g,,,,).

Wash

The membranes were washed, as described in the legend of Fig. 1, to remove residual contaminating material which, in the case of the floated vesicles was lysosomal particles, and with the sedimented vesicles was lysosomal and microsomal particles. This procedure has been used by other investigators to remove such contaminating structures (5).

Fractionation

Fractionation was effected by gently lowering thin (l-mm diame- ter) polyethylene tubing to the bottom of the centrifuge tube and pumping out the tube from the bottom by a Sigmamotor pump at maximum speed (100 ml/h): the fractions were monitored by use of an LKB Uvicord and recorder. Fractions (0.8 ml) were collected in 2- ml conical cups (Technicon, Tarrytown, N. Y) using an LKB Ul- trorac fraction collector. Refractive indices were used to correlate sucrose density with the 280 nm absorbance pattern. The fractions were stored at -20” until use.

Enzyme Assays

The enzyme assays were performed at 37” using the methods outlined below. In each case, the reaction was shown to be linear with time and proportional to the amount of enzyme preparation used. Calculations are based on international units of activity in which 1 unit of enzyme yields 1 pmol of product/h.

Lactic Acid Dehydrogenase (EC 1 .l .I 27) -This enzyme was used as a cytoplasmic marker (6). The activity was measured by following the decrease in 340 nm absorbance due to the conversion of NADH to NAD, a parallel reaction in the lactic acid dehydrogenase conversion of pyruvate to lactate (7).

Succinate Dehydrogenase (EC 1.3.99.1 )-This activity has been shown to be typical of mitochondria (8). Succinate dehydrogenase activity was determined using 2,6-dichloroindophenol as the electron acceptor (9). The initial change in absorbance was used to calculate units of activity.

Phosphodiesterase (Nonspecific) (EC 3.1.4.18)-This enzyme was used for screening samples. It is a plasma membrane marker (101 and was assayed by the method of Koerner and Sinsheimer (101.

Thymidine-5’.phosphodiesterase (EC 3 .I .3.35) -This enzyme was used as a marker for the plasma membrane (11). The substrate used was thymine 5’-p-nitrophenyl phosphate (Sigma Chemical Co., St. Louis, MO.) and the change of absorbance at 400 nm was observed at 30, 60, and 90 min, basically following the procedure of Erecinska et al. (11).

Glucose-6-phosphatase (EC 3 .1.3.91- Glucose-6-phosphatase was used as a microsomal marker (12) and was measured by the proce- dure of Nordlie and Arion (131 by following the release of inorganic phosphate by the method of Bartlett as modified by Ames (14).

P-N-Acetylglucosamini~se (EC 3.2.1.30)-This enzyme was used as a lysosomal marker (15) and was measured by the increase of absorbance at 420 nm at 20, 40, and 60 min using p-nitrophenyl glucosaminide (Sigma) as a substrate (16).

Esterase-This acid phosphomonoesterase (EC 3.1.3.2) was used as a marker of microsomal material (17). The method used for assay was that described by Earl and Korner (9).

Chemical Analyses

The samples were pooled and dialyzed exhaustively against dis- tilled water. Each was lyophilized, then dried in oacuo at 100” over P,O, for 18 h, cooled to 25”, and weighed.

Protein

Protein was determined for enzymatic studies by the procedure of Lowry et al. (18) using bovine serum albumin as a standard and by amino acid analysis for chemical studies, using norleucine as an internal standard.

Amino Acids

Amino acid analyses were performed after hydrolysis in constant boiling HCl at 108” under nitrogen for 24 h at a weight concentration of 1 pg of sample/ml. The sample was evaporated to dryness and dissolved in 0.2 N citrate buffer, pH 2.2. The amino acid analyses were performed using a JEOL model JLC-6AH amino acid analyzer.

Carbohydrate

Sialic acid was determined by the thiobarbituric acid assay (19) after hydrolysis in 0.1 N H,SO, at 80” for 1 h. In order to assay sialic acid in the presence of large amounts of DNA, a new method was devised using radioactive sialic acid as an internal standard and preferentially extracting the chromophore attributable to DNA.’

Total hexose was determined by the phenol/sulfuric acid method

Z M. Jett, manuscript in preparation.

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2136 Nuclei and Plasma Membrane Isolation and Characterization

of Dubois et al. (20) using equimolar amounts of galactose and mannose as a standard. Hexosamines were determined by the Elson- Morgan procedure modified by Boas (21) after hydrolysis in 4 N HCl for 2 h in sealed tubes.

Lipids

Total lipids were determined gravimetrically after extraction three times with chloroform/methanol (2:l w/v) at room temperature for 18 h and partitioning against an equal volume of distilled water by the method of Folch et al. (22) and calorimetrically by the sulfo- phosphovanillin reaction (23). Total phosphorus and organic-extract- able phosphorus were determined after perchloric acid combustion by the method of Bartlett (24). The calculations of phospholipid were based on the assumption of 25 pg of phospholipid/pg of lipid phosphorus (25). Cholesterol was determined by the Lieberman- Burchard reaction as modified by Watson (26).

fraction was obtained by centrifugation onto a 38% sucrose pad (d 1.1683). The crude membrane fraction was adjusted with concentrated sucrose to achieve a density intermediate be- tween that of the two bands (31 to 32% sucrose solution; d 1.1339 to 1.1386) and this fraction was introduced in a centri- fuge tube so as to be intermediate between sucrose solutions of 15% concentration (d 1.0610) and 38% concentration (d 1.1683). This simultaneous sedimentation-flotation step was devised to separate two populations of plasma membrane vesi- cles according to their different buoyant densities. The lighter of the two bands floated to the 32 to 15% interface while the heavier sedimented onto the cushion of 38% sucrose and lyso- somal contaminants floated to the top of the tube or were held in the 32% phase.

Nucleic Acids

DNA was determined by measuring fluorescence (Turner III fluo- rometer) with ethidium bromide (27). Incubation at 37” for 1 h in 0.3 N KOH was used to eliminate fluorescence attributable to RNA. The fluorescence after hydrolysis represented the DNA in the sample. Calf thymus DNA solutions, made fresh daily, were used for a standard. The difference before and after hydrolysis was a measure of the RNA in the sample. RNA was also determined by the Schmidt- Thannhauser method as described by Munro and Fleck (28). For this assay, type III RNA (Sigma) was used as a standard since it was equivalent to the average of solutions of AMP and UMP.

Electron Microscopy

Fig. 2 shows the profile of a typical experiment. Phosphodi- e&erase (nonspecific) activity was associated with the sedi- mented and the floated material with very little being re- tained in the middle fraction. The protein profile shows three major fractions, the sedimented and floated vesicles, and the intermediate fraction. This intermediate fraction consists of contaminating organelles, primarily lysosomal material as determined by P-N-acetylglucosaminidase activity (15) which reached a peak at an elution volume of 18 ml. The minor protein peak at 38 ml also had a large amount of P-N-acetyl- glucosaminidase activity associated with it.

Fractions were routinely fixed with 1.25% glutaraldehyde in a 4% sucrose, 0.05 M phosphate buffer (pH 7.3) for 18 to 20 h. After fixation, the samples were washed in buffer (0.05 M phosphate buffer, pH 7.3) for about 30 min and then postfixed in buffered 1% osmium tetroxide. Samples were dehydrated in ascending concentra- tions of ethanol and finally in propylene oxide and embedded in Epon 812 (29, 30). Ultrathin sections were cut and stained with uranyl acetate and lead citrate and examined in a Hitachi HU-12 transmis- sion electron microscope.

RESULTS

Lysis and Fractionation -The procedure described under “Materials and Methods,” and summarized in Fig. 1, afforded a rapid isolation of a fraction containing intact cell nuclei and the separation of two populations of plasma membrane vesi- cles. The first step, differential centrifugation, was effective in removing nuclei and cell debris. Then a crude membrane

The major contaminants left after the sedimentation-flota- tion gradient were lysosomal (P-N-acetylglucosaminidase) with both bands and microsomal (esterase) with the sedi- mented band. In order to see if the contaminating activities were intimately associated with the plasma membranes, the two fractions were each subjected, separately, to isopycnic density centrifugation on 20 to 45% sucrose gradients. Fig. 3A shows that a small amount of P-N-acetylglucosaminidase ac- tivity was coincidental with the peak of phosphodiesterase activity of the floated vesicles. Fig. 3B shows a minor peak of P-N-acetylglucosaminidase activity to coincide with the phos- phodiesterase activity of the sedimented vesicles. A character- istic pattern of the specific activities of the two enzymes for the sedimented band is shown in Fig. 3C. The specific activity of /3-N-acetylglucosaminidase associated with the plasma mem- brane fraction is slightly less than that in whole cells while

A: Differential Centrifugation

LYSATE 7ooxfio, ,Om,n r”U- PELLET .TN’JCLEI

1 , NU;zcE 3000~~ br ,Omm [DEBn’s (PELLET1

MEMBRANE LYSATE

B: Ultracentrifugation Step I

E, pr<( ‘:..\I\,..,

!

LIPID ‘HAZE’ i” ‘\x:->

;$.y 95,OOOX~ for $Om,n *.’ &Jq

’ .j;;<> MEMBRANE LYSATE

;,<;j!,$J; . .&. CRUDE MEMBRANE

38% SUCROSE DEBRIS

C: Ultracentrifugation Step 2

15% SUCROSE FLOATED BAND

CRUDE MEMBRANE FRACTlON 3oo,oOox~ for 2hr

* IN 31-3296 SUCROSE SEDlMENTED BAND

38% SUCROSE DEBRIS

FIG. 1. Fractionation procedure for the isolation of RAJI lymphoid cell nu- clei and plasma membranes. 1, all gravity forces in this scheme are re- ported as average (g). 2, “Wash” CD) represents pooled plasma membrane fractions diluted with lo-mm Tris/HCl, pH 7.4, to less than 4% sucrose (by refractive index) and then pelleted on a l-ml cushion of 38% sucrose at 36,000 X g for 1 h in the Sorvall RC-2B.

D: Wash 0

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Nuclei and Plasma Membrane Isolation and Characterization 2137

-

1.C

0.f

a E E

2

z 0.6 a

.g

5 h

0 0.4 E

0.2

1 ) . . . . .

0.20

3.15 a, a E

2

ii a m

.Tg

1.10 2

B

II.05

Elution Volume (ml )

FIG. 2. Purification of RAJI lymphoid cell plasma membrane ves- icles by use of a discontinuous sucrose gradient in which the more dense vesicles are sedimented onto a cushion of 38% sucrose and the less dense are floated to the 31 to 15% interface (see Fig. El. The tube was fractionated from the bottom and O.&ml fractions were collected which were assayed for protein, O-0, and phosphodies- terase (PDE) activity, n - - -W.

that of phosphodiesterase increased 12-fold. Since this proce- dure showed that most of the contaminants were not an inte- gral part of the plasma membranes, simply washing the plasma membrane fractions (Fig. lD). after their isolation from the sedimentation-flotation gradient was the logical pro- cedure. Indeed, 95% of the phosphodiesterase activity but only about 20% of the N-acetylglucosaminidase activity in the pooled fraction remained after washing the floated vesicles. Similarly, 92% of the phosphodiesterase activity, 12% of the /3- N-acetylglucosaminidase activity, and 20% of the e&erase activity remained after washing the sedimented vesicles. A second wash of the floated vesicles removed even more P-N- acetylglucosaminidase but resulted in additional loss of phos- phodiesterase activity.

Analytical Determinations -Enzymatic analyses were car- ried out to determine the relative purity of the washed mem- brane fractions and to see the distribution, compared to the whole cell homogenate, of a variety of marker enzymes (listed under “Materials and Methods”). Table I shows that approxi- mately 12 to 15% of all the enzymatic activities are found in the glycerol phase which is attributed to carry-over of intact cells in this fraction. Although increased centrifugation re- covers an additional 4 to 6% of the cells, their lysis properties are unsatisfactory. This may be due to the fact that the uptake of glycerol requires a fully functioning cell, therefore damaged cells are selected out by this procedure.

Thymidine-5’-phosphodiesterase, a marker for plasma membranes, was concentrated in the floated and sedimented fractions, respectively, but a high percentage of the total activity was also found in the nuclei. The floated membrane fractions showed trace contamination with p-N-acetylglucosa-

minidase, while the sedimented membrane fraction showed trace contamination with e&erase and /3-N-acetylglucosamin- idase: in neither case was there a concentration above that found in intact cells (Table II). For thymidine-5’-phosphodies- terase, the only significant increases in specific activity oc- curred in the two plasma membrane fractions. The debris fractions showed a slight enrichment of activity, probably representing some retention of plasma membranes, but they contained large amounts of contaminating organelles such as mitochondria, as indicated by succinate dehydrogenase activ- ity, and of microsomes, as indicated by glucose-6-phosphatase and esterase activities (Table II). Lactic acid dehydrogenase, esterase, and /3-N-acetylglucosaminidase were concentrated only in the cytosol fraction.

The chemical composition of certain fractions is shown in Table III. The two plasma membrane fractions contained 2.88% of the whole cell protein. The sedimented membrane fraction is the larger of the two, comprising nearly two-thirds of the total protein and of the dry weight. In both plasma membrane bands, the protein/lipid ratio is near unity while the cholesterol/phospholipid ratio approaches unity for the floated band but is 0.69 for the sedimented band. A ratio of 0.7 was found for lymphocyte plasma membranes from tonsilar tissue (31) although most reported values approach unity (5, 32). The higher cholesterol content in the floated band may account for its higher buoyant density (33). Only minute amounts of DNA and RNA, 0.06 and 0.5%, respectively, of the total cell content occur in the plasma membrane fractions. The sialic acid content of the combined plasma membrane fractions is 25% of the total cell sialic acid (Table III) and there is little difference in the overall carbohydrate content of the two bands (Table IV).

Nuclear Fraction -The recovery of intact nuclei, as deter- mined in 10 preparations by light and phase contrast micros- copy, was 75 to 80%. Since there is a loss of 15% of the cells in the glycerol phase (Table I), the yield of intact nuclei is 90 to 95% based on lysed cells.

The nuclear fraction contains 63% protein and 16% DNA (Table IV) and represents 92% of the total DNA of the intact cell when corrected for cell loss in the glycerol phase (Table III).

Electron Microscopy -Electron micrographs of various frac- tions are shown in Fig. 4. The nuclei (Fig. 4a) are somewhat oblong and are 4 to 6 pm in width and approximately 9 pm in length. For the most part, the double nuclear membrane is intact.

The “membrane lysate” (see Fig. 1A) preparation shows an abundance of plasma membrane vesicles (Fig. 4b) plus various ribosomal and lysosomal contaminants.

The sedimented plasma membrane vesicles (Fig. 4~) have an average diameter of 3 to 4 pm. Some ribosomes appear to be an integral part of the plasma membrane while a few other particles seem to be trapped inside the vesicles. Fig. 4d shows the same sample after the final washing by dilution to 4% sucrose concentration and subsequent pelleting (Fig. 1, Step D).

The floated plasma membrane vesicles (Fig. 4e) are approxi- mately 2 pm in diameter. There are a few granular particles scattered throughout the fraction.

DISCUSSION

Since the study of membrane-related phenomena requires the isolation of plasma membranes, a variety of techniques have been developed for cell disruption and subcellular frac-

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2138 Nuclei and Plasma Membrane Isolation and Characterization

PDE fx+J

0.12

PDt b-r

0.1

30 40 0 10

Elution Volume (ml )

10 20

Elution Volume (ml )

tionation. Lymphoid cells have been disrupted by most of the conventional techniques including nitrogen cavitation, Potter- Elvehjem and Dounce homogenization with and without chemical stabilization, mechanical shear devices, and chemi- cal-destabilizing agents: the merits of these various proce- dures have been reviewed (5, 32, 33).

Optimal cell disruption requires that a balance be achieved between cell breakage and preservation of intracellular organ- elles (primarily nuclei). In addition the effect on biologic func- tions such an antigenic and enzymatic activities must be considered. The method we describe results in 85% cell lysis. This degree of cell disruption has been obtained for suspension lymphoid cells using mechanical shear (32, 34) and for solid tissue disrupted by a perforated disc (35) but this later tech-

p-NG k-4

0.25

il.20

0.15

-

0.10

0.05

0

3-NG I---A

D.06

FIG. 3. Isopycnic density centrifu- gation for 18 h at 95,000 x g,, on linear 20 to 45% sucrose gradients. Phospho- diesterase activity V’DE), A-A; P-N-acetylglucosaminidase (@VG), A---A. (A) floated and (B) sedi- mented plasma membrane vesicles with enzymatic activities reported in units per 0.8 ml. (Cl sedimented plasma membrane vesicles with spe- cific activities reported in units per mg of protein.

nique resulted in only 50% lysis (5) when applied to suspen- sion cells.

In evaluating a variety of buffers for lysis of the pelleted, glycerol-loaded cells, 1.0 mM MgCl, and 1.0 mM CaCl, were used to preserve nuclear integrity. Buffers containing sodium chloride caused membrane clumping either after lysis or upon the addition of sucrose and the percentage of unlysed cells was somewhat greater by both microscopic examination and enzy- matic assay when using hypotonic sucrose as compared to 10 mM Tris/HCl buffer.

Several important points have emerged from these studies in order to obtain optimum yields. Cell damage during the washing procedure was decreased by retaining normal culture conditions including maintenance of the temperature at 30-37”

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Nuclei and Plasma Membrane Isolation and Characterization 2139

(36) and use of balanced salt solutions until the final lysis step. Upon addition of glycerol to a final concentration of lo%, the

cells immediately began to shrink, as monitored by light and phase microscopy, and their surfaces became slightly irregu- lar. By 30 s, 80% of the cells had shrunk, but by 60 s only about 20% were still shrunken and by 90 s, nearly 99% of the cells had regained their original volume and shape. The second increment of glycerol, which brought the concentration to 20%, caused similar effects but the changes in size and shape were less marked. The last addition to 30% produced only slight changes in cell size and shape. The degree of lysis, based on enzyme assays and microscopic examination, was not im- proved by further increase in glycerol concentrations up to 45%.

The outstanding feature of this method is that the nuclei are recovered intact and quantitatively. Other methods report

TABLE I

Distribution of enzymatic activities in subcell&r fractions

Per cent distribution of enzymatic activity in various fractions

Whole ____ cells Plas-

Enzyme (total activ- ityj”

‘i

“,kl!- NU- De- ,“,“,- cyto-

phas? clei bris’ branes SOl’ (COIlI-

bined)

Tymidine-5’- phosphodies- terase

P-N-Acetylglu- cosaminidase

Glucose-6-phos- phatase

Esterase Succinate dehy-

drogenase Lactate dehy-

drosenase

22 15 43 I 17 13

67 12 12 2 1 58

90 15 39 11 0.5 14 59 16 39 7 2 56

583 14 31 36 0.6 17

45 9 1 0.5 0.4 101

” Total activity is expressed as micromoles of product liberated per h per 9.5 x 10y cells.

* “Glycerol phase” is the supernatant solution resulting when the cells are centrifuged after the addition of glycerol.

c “Debris” includes the combined pellets from Steps A, B, and C in Fig. 1.

d These figures represent the sum of the activities of the two plasma membrane fractions.

’ The cytosol fraction represents the material above the crude membrane (Fig. IS), the material above the floated band and be- tween the bands (Fig. 1C) and the supernatant solutions of the wash of the membrane vesicles (Fig. ID).

nuclear breakage from 30%, using N, cavitation (31) and Pot- ter-Elvehjem (371, to 4 to lo%, using Dounce homogenization (381, N, cavitation (391, and electrical shearing devices (16, 35). A recent review compares the various methods of cell disruption and plasma membrane isolation and characteriza- tion (32).

Our preliminary data, as well as that of others (331, indi- cated that rapid separation of organelles from the cell lysate was necessary to increase recovery and decrease contamina- tion. For this reason, the use of isopycnic centrifugation was not acceptable for routine work due to the time involved and the limitations on amounts of starting material which could be used. We sought a procedure which would rapidly eliminate most contaminating structures from the plasma membrane preparation and settled on a preliminary sedimentation onto a sucrose cushion. Further purification of this crude membrane fraction was achieved by simultaneous flotation and sedimen- tation of the membrane within a carefully designed discontin- uous sucrose gradient system. First, the buoyant densities of the individual membrane bands were determined by centrifu- gation within a continuous linear sucrose gradient. This showed that the two bands had densities of 1.151 and 1.127. The crude membrane fraction was then brought to a density intermediate between that of the two bands by adding a con- centrated sucrose solution and it was introduced between su- crose solutions of higher and lower density as described under “Materials and Methods” and in Fig. 1. On centrifugation, a residual debris fraction sedimented to the bottom of the tube while one population of membrane vesicles sedimented to the interface with the 38% sucrose cushion, and the other floated to the interface with the 15% sucrose solution; some lysosomal material floated to the top of the tube.

This method provides a rapid and effective way of separat- ing and purifying two populations of membrane vesicles. It should be readily applicable to other cell types after modifica- tion of the stepwise gradients to reflect the densities deter- mined for the individual membrane fractions in pilot experi- ments.

RAJI lymphoid cells, like many tumor cells have large nuclei (average diameter 12 pm) which may also be damaged during cell disruption by mechanical shear since the diameter of the nuclei is about two-thirds that of the average cell. Attempts to monitor nuclear damage and cell disruption enzy- matically are somewhat misleading since the nuclei contain enzymatic activities usually associated with plasma mem- branes. In addition, the measure of lactic acid dehydrogenase release shows the degree of cell damage but not necessarily cell disruption.

TABLE II

Comparison of specific activities of various marker enzymes in subcellular fractions

Ratio of specific activity to that in whole cells

Whole cells (specific activity)” Glycerol” Nuclei Debris

Plasma membrane cytoso1

Floated Sed’

Thymidine-5’-phosphodiesterase 0.04 0.82 0.96 1.2 20 6.7 0.92 P-N-Acetylglucosaminidase 0.12 0.69 0.27 0.3 1.2 0.5 1.6 Glucose-6-phosphatase 0.18 0.77 0.81 1.6 0.2 0.3 0.88 Esterase 0.11 0.91 0.87 1.2 0.8 1.0 2.8 Succinate dehydrogenase 1.4 0.79 0.69 5.7 0 0.4 0.56 Lactate dehvdroaenase 0.08 0.51 0.03 0.1 0.3 0.3 3.0

a Specific activity is expressed as micromoles of product per h per mg of protein. b For description of fractions, see Table I. c Sed, sedimented.

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2140 Nuclei and Plasma Membrane Isolation and Characterization

TABLE III

Chemical composition of whole cells and subcellular fractions: total milligrams and per cent

Figures in parentheses represent the percentage of that component in the fraction compared to the whole cell homogenate. The 15% loss of cells in the glycerol phase is not taken into account.

Whole cell homog- Sedimented plasma Floated plasma Sum of plasma ellatea Nuclei membrane membrane membrane

Protein

Lipidb

Cholesterol

RNA 599

DNA 496

Neutral hexoses 245

Hexosamines 30.4

Sialic acid 7.8

Sum (mg) Total dry weight (mg) % accounted % whole cell dry weight CholesteroliohosnboliDid’

3406

656

108

5548

6613.3

84

1463 67.5 30.4 97.9 (43) (2) (=a C-3)

222 59.7 29.9 89.6

(34) (9) (5) (14) 36.5 9.2 5.9 15.1

(34) (8.5) (5.5) (14) 148 2.6 0.3 2.9

(25) (0.4) (0.05) (0.5) 372 0.2 0.1 0.3

(75) (0.04) (0.02) (0.06) 72.9 21.7 13.6 35.3

(30) (9) (6) (15) 7.3 2.9 1.9 4.8

(24) (10) (6) (16) 2.5 1.2 0.7 1.96

(31) (16) (9) (25)

2324.9 2327.49

99.9 35

165 82.8

201.15 97.47

82 85 3 1.5

0.69 0.97

247.8

298.61 83

4.5

n These figures represent the composition of 7.2 x 10”’ cells. D “Lipid” represents total neutral and phospholipids minus cholesterol. r Phospholipid was calculated assuming 25 pg of phospholipid/pg of lipid phosphorus (25).

TABLE IV

Chemical distribution in nuclei and plasma membrane fractions Plasma membrane

Sedimented vesicles Floated vesicles

Protein Lipid Cholesterol RNA DNA Neutral hexoses Hexosamines Sialic acid

Per cent accounted for 82.2 85.0 99.8

pgglmg protein

884

135.5 38.6

3.4

321

43.1 18.4

70 dry wt 33.6

29.7

4.6 1.3

0.1

10.8

1.5 0.6

pglmg protein

982 193.8

9.5 2.6

447.

61.8 23.7

% dry wt 31.2

30.7

6.1 0.3

0.1

14

1.9 0.7

Nuclei

pglmg protein

152

24.9 101.3

254.4

49.9 4.98

1.7

9% dry wt 62.9 9.6 1.6 6.3

15.9

3.1 0.3

0.1

Since the nuclear membranes contain large amounts cf typical plasma membrane enzyme activities, and these mem- branes vesiculate as do plasma membranes, they are virtually impossible to distinguish or separate from plasma mem- branes. In fact, it is commonly believed that there is continu- ity between the plasma membrane and the nuclear membrane (40). For these reasons, a method of cell disruption which leaves intact nuclei is important for any claim regarding localization of components in either nuclei or plasma mem- brane. The glycerol procedure offers an ideal system for ac- complishing this objective. Although the plasma membrane is permeable to glycerol, the nuclear membrane does not appear to be under these conditions. When the whole cells are exposed to a hypotonic buffer following glycerol-loading, the difference in osmotic pressure between the inside and outside of the plasma membrane is so great that the membrane is stripped off and forms vesicles up to about one-fourth the size of the

intact cell. The nuclei are recovered in nearly quantitative yield.

The chemical analyses previously reported for lymphoid cell plasma membranes is somewhat fragmentary and incomplete. Warley and Cook (41) using mouse leukemia cells, found 51.8% protein and 48.1% lipid, while Crumpton and Snary (5) report 42% protein, 51% lipid, and 5.5% neutral sugars and sialic acid for the BRI-8 lymphoid cell line. It may be noted that the total analyses approached 100% for the components assayed in these two reports even though certain components were not determined. The dry weight composition of the two plasma membrane fractions isolated in this study are nearly identical with regard to lipid (-30%) and protein (-32%) while the carbohydrate composition (neutral hexoses, hexosamine, and sialic acid) represents 13% of the dry weight for the sedimented and 17% for the floated vesicles. Another differ- ence between the sedimented and floated vesicles is in the

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Nuclei and Plasma Membrane Isolation and Characterization 2141

I I 5 PM 4a

1 3 P’M

I

4b

I 3 PM

t

4d

I 3 PM

I

4c

FIG. 4. a, nuclei x 10,000; b, mem- brane lysate (see Fig. 1B) X 8,000; c, sedimented vesicles x 10,900; d, washed sedimented vesicles x 18,000; e; floated vesicles x 14,500.

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

2142 Nuclei and Plasma Membrane Isolation and Characterization

RNA composition which is 9 pg/mg of protein in the case of the floated vesicles and 38 pg/mg for the sedimented vesicles (“Sum of the Plasma Membrane,” Table III, shows a total of 29 pg of RNA/mg of membrane protein). Other investigators have reported 15 to 34 pg of RNA/mg of protein (5, 31, 34, 37) for lymphoid cells.

Small quantities of DNA (0.06% of the whole cell DNA were found in the plasma membrane vesicles of RAJI cells with the DNA/protein ratio being nearly the same (-3 pg/mg of pro- tein) for both the floated and sedimented vesicles. These val- ues are slightly lower than those reported for plasma mem- branes from other lymphoid cells (6 to 10 pg of DNAimg of protein; 16, 31, 34, 37). In fact, there has always been some question regarding the presence of DNA in plasma membrane preparations, but Lerner et al. (42, 43) concluded that some DNA was truly a component of the plasma membranes of WIL 2 cultured cells.

Thymidine-5’-phophodiesterase, thought to be a marker for plasma membranes (111, is enriched in the membrane frac- tions isolated by this technique, but is not found there exclu- sively, being present also in isolated nuclei (Table I). ATPase and 5’-nucleotidase which are also thought to be plasma mem- brane markers have been reported in nuclear fractions by others (5, 41, 44, 45). In addition, 10 to 15% of these enzymes are found in soluble form (5, 45) (Table I). Although this could be due to vesicles which do not sediment, or to enzyme which is dislodged from the membrane and solubilized during cell dis- ruption, it probably represents a truly cytoplasmic form of the enzyme (5, 11). This means that greater than 50% of the “plasma membrane marker” is associated with other fractions. Therefore, recoveries of greater than 50%, based on whole cell enzymatic data, probably indicate that nuclear membranes are contaminating the plasma membrane fraction. Recovery of plasma membranes may be more meaningful if based on a nuclei-free lysate minus the cytoplasmic form of the enzyme. This requires quantitative removal of the nuclei. On such a basis, recovery of plasma membrane in this study would be approximately 70% with 22% occurring in the debris fractions and 8% remaining in the 31 to 32% sucrose phase (included in the cytosol fraction, Table I).

RAJI cell nuclei, on the other hand, contain relatively large amounts (39% of total cell activity) of the microsomal marker enzymes, glucose-6-phosphatase and esterase (Table I) which have also been reported as markers of rat liver nuclei (46) and calf liver nuclei (461, respectively. Thirty per cent of the succi- nate dehydrogenase activity (Table I) is associated with the nuclei. None of these activities are present in significant amounts in the plasma membrane fraction. Further fractiona- tion of the nuclei, with separation of the inner and outer membranes from the intranuclear material, may define the points of similarity and difference to the plasma membrane.

Acknowledgment-We are grateful to Ms. Kay Ennis for her skillful technical assistance.

REFERENCES

1. Hayden, G. A., Crowley, G. M., and Jamieson, G. A. (1970) J. Biol. Chem. 245, 5827-5832

2. Mackler, B. F. (1972) J. N&l. Cancer Inst. 49, 935 3. Jett, M., and Jamieson, G. A. (1973) Biochem. Biophys. Res.

Commun. 55, 1225-1233 4. Barber, A. J., and Jamieson, G. A. (1970) J. Biol. Chem. 245,

6357-6365 5. Crumpton, M. J., and Snary, D. (1974) in Contemporary Topics

6.

I. 8. 9.

10.

11.

12.

13.

14. 15.

16.

17.

18. 19. 20.

21. 22.

23.

24. 25.

26. 27.

28.

29. 30.

31. 32.

33.

34.

35. 36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

in Molecular Immunology (Ada, G. L., ed) Vol. 3, pp. 27-55, Plenum Press, New York

Beattie, D. S., Sloan, H. R., and Basford, R. E. (1963) J. Cell. Biol. 19, 309-316

Kornberg, A. (1955) Methods Enzymol. 1, 441-442 Weiner. N. (1970) Annu. Reu. Pharmacol. 10. 273-290 Earl, D: C. N., and Korner, A. (1965) Biochem. J. 94, 721-734 Koerner. J. F.. and Sinsheimer, R. L. (1957) J. Biol. Chem.

228, 1649-1062 Erecinska, M., Sierakowska, H., and Shugar, D. (1969) Eur. J.

Biochem. 11, 465-471 Hortnagl, H., Winkler, H., and Hortnagl, H. (1969) Eur. J.

Biochem. 10, 243-248 Nordlie, R. C., and Arion, W. J. (1966) Methods Enzymol. 9,619-

621 Ames, B. N. (1966) Methods Enzymol. 8, 115-116 Sellinger, 0. Z., Beaufay, H., Jaques, P., Doyen, A., and

deDuve, C. (1960) Biochem. J. 74, 450-456 Kornfeld, R., and Siemers, C. (1974) J. Biol. Chem. 249, 1295-

1301 Weinhold, P. A., and Rethy, V. B. (1969) Biochem. Pharmacol.

18, 677-680 Folin, 0.. and Ciocalteu. V. (1927) J. Biol. Chem. 73. 627-650 Aminoff,‘D. (1961) Biochem. J. 81, 384-392 Dubois, M., Gilles, K.. Hamilton, J., Rebers, P. A., and Smith,

F. (1956) Anal. Chem. 28, 350-356 Boas, N. F. (1953) J. Biol. Chem. 204, 553-563 Folch. J.. Lees. M.. and Sloane Stanlev. G. H. (1957) J. Bzol.

Chem. i26, 447-509 ” I

Zoellner, N., and Kirsch, K. (1962) 2. Gesamte Exp. Med. 135, 545-561

Bartlett, G. R. (1959) J. Biol. Chem. 234, 466-468 Kooaczvk. K.. Perdue. J.. and Green. D. E. (1966) Arch. Bio-

&em: Biophys. 115, 2151225 Watson, D. (1960) CZin. Chim. Acta 5, 613-615 Donkersloot, J. A., Robrish, S. A., and Krichevsky, M. I. (1972)

Appl. Microbial. 24, 179-183 Munro, H. N., and Fleck, A. (1966) Methods Biochem. Anal. 14,

158-162 Aikawa, M., and Ng, A. B. P. (1973) Cancer 31, 385-397 Seed, T. M., Aikawa, M., Sterling, C., and Rabbege, J. (1974)

Infect. Immun. 9, 750-761 Demus. H. 11973) Biochim. Bioohvs. Acta 291. 93-106 Jett, M., and Hickey-Williams: C: A. (1976) in Mammalian Cell

Membranes (G. A. Jamieson, and Robinson. D. M.. edsl Vol. 3. p. 44-87, Butterworths, London

DePierre, J. W., and Karnovsky, M. L. (1973) J. Cell. Biol. 56, 275-303

Ladoulis, C. T., Misra, D. N., Estes, L. W., and Gill, T. J., III (1974) Biochim. BioDhvs. Acta 356. 27-35

Allan, D., and Crumpton, M. J. (1970) Biochem. J. 120, 133-143 Lin. P. S., Wallach, D. F. H.. and Tsai. S. (1973) Proc. N&l.

Acad. Sci. U. S. A. 70, 2492-2496 Ferber, E., Resch, K., Wallach, D. F. H., and Imm, W. (1972)

Biochim. Biophys. Acta 266, 494-504 Marique, D., and Hildebrand, J. (1973) Cancer Res. 33, 2761-

2769 Van Blitterswijk, W. J., Emmelot, P., Hilgers, J., Kamlag, D.,

Nusse, R., and Feltkamp, D. A. (1975) Cancer Res. 35, 2743- 2751

Grundmann, E. (1966) General Cytology: An Introduction to Function and Morphology of the Cell, pp. 242, Williams and Wilkins Publishers, Baltimore, Md. - -

Warley, A., and Cook, G. M. W. (1973) Biochim. Biophys. Acta 323. 55-68

Lerner, R. A., Meinke, W., and Goldstein, D. A. (1971) Proc. N&l. Acad. Sci. U. S. A. 68, 1212-1216

Lerner, R. A. (1972) in Contemporary Topics in Immunochemis- try (Inman, F. P., ed) Vol. 1, p. 111, Plenum Press, New York

Misra, D. N., Gill, T. J., III, and Estes, L. W. (1974) Biochim. Biophys. Acta 352, 455-461

Wallach, D. F. H., and Win&r, R. J. (1974) Euoluing Strategies and Tactics in Membrane Research, Springer-Verlag, New York

Stern, H., Allfrey, V., Mirsky, A. E., and Saetren, H. (1952) J. Gen. Physiol. 35, 559-578

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from

M Jett, T M Seed and G A Jamiesonlymphoid cells.

Isolation and characterization of plasma membranes and intact nuclei from

1977, 252:2134-2142.J. Biol. Chem. 

  http://www.jbc.org/content/252/6/2134Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/252/6/2134.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 14, 2020

http://ww

w.jbc.org/

Dow

nloaded from