WO1991004326A1

WO1991004326A1 - Free electrophoresis method for isolating ribosomes and natural membrane vesicles

Info

Publication number: WO1991004326A1
Application number: PCT/US1990/005330
Authority: WO
Inventors: Robert Clyde Lawson, Jr.; Paul Wilson Todd
Original assignee: Cell Technology Inc
Current assignee: Cell Technology Inc
Priority date: 1989-09-19
Filing date: 1990-09-19
Publication date: 1991-04-04
Anticipated expiration: 1992-03-19
Also published as: AU6410090A

Abstract

The invention discloses a method for isolating ribosomes and natural membrane vesicles from a cell lysate comprising subjecting the cell lysate to free electrophoresis. The method is especially useful for separating the ribosomes and natural membrane vesicles needed to make pharmaceutical products such as biologic response modifiers.

Description

FREE ELECTROPHORESIS METHOD FOR ISOLATING RIBOSOMES AND NATURAL MEMBRANE VESICLES

Field of the Invention

This invention relates generally to methods for isolating subcellular particles useful for making pharmaceutical products. More specifically, the invention relates to a method for isolating ribosomes and natural membrane vesicles needed to make pharmaceutical products such as biologic response modifiers.

Background of the Invention Biologic response modifiers (BRMs) are agents that modify the relationship between a disease and a host by modifying the host's biological response to the disease with resultant therapeutic effects. The BRMs can be divided into two categories: (1) biologic or chemical agents that can stimulate or otherwise alter one or more of the host's resistance mechanisms and (2) purified cellular products that demonstrate direct effects on a particular disease. The present invention relates to a method that can be used to isolate ribosomes and natural membrane vesicles needed to make certain of the biologic response modifiers.

A BRM may be used alone, or in combination with other agents, to enhance resistance to or recovery from invasion by pathogens, to modify or induce tolerance to grafts of foreign tissue, to enhance tumor rejection or stabilization and to inhibit tumor recurrences following other forms of therapy, to restore normal helper suppressor mechanisms, or otherwise promote a normal immune response. USSN 057,344 and which is assigned to the assignee of the present invention, discloses a novel biologic response modifier that is useful for human or animal treatment. (USSN 057,344 is equivalent to PCT/US87/01397 which has been published as International Publication No. O87/07503.) The novel BRM comprises natural membrane vesicles and ribosomes in a suspending buffer. The membrane vesicles are comprised of cellular membrane material endogenous to a selected microorganism which does not evoke a significant immune-deviating response in patients and is substantially non-pathogenic in humans. The ribosomes are also endogenous to the selected microorganism, which in addition to the attributes listed above is a microorganism in which membrane vesicles can be formed from cell membrane material. The USSN 057,344 and PCT/US87/01397 biologic response modifier is substantially free of endotoxin, intact cells, cell walls, and cell membrane fragments, and is readily endocytosed by the patient's monocyte- macrophage cell line.

USSN 057,344 and PCT/US87/01397 further discloses that the ribosomes and the membrane vesicles are preferably produced when bacterial cells are lysed mechanically, for example in a suitable cell disrupter or French pressure cell, and that the natural membrane vesicles and ribosomes are preferably separated from smaller fractions (such as DNA fragments, RNA fragments, protein fragments, cell wall fragments, membrane fragments, etc. ) by passage of the lysate (which preferably has been subjected to centrifugation to "clear" it of larger subcellular particles) through an appropriate linear or discontinuous density gradient material, or a suitable chromatographic column. These separation methods have been traditionally used with reliability and success in bench-scale biochemical purifications, but their intrinsic limits in capacity and their labor-intensive operation reduce their cost- effectiveness in the industrial production of the BRMs. It has surprisingly been discovered that the substantially pure ribosomes and natural membrane vesicles, such as those needed to make biologic response modifiers like those disclosed in USSN 057,344 and PCT/US87/01397, can be produced by subjecting the cell lysate to separation by free electrophoresis. Furthermore it has been discovered that the free electrophoresis method of the present invention does not suffer the drawbacks of the density gradient centrifugation method disclosed in USSN 057,344 and PCT/US87/01397. For example, unlike the density gradient centrifugation method, the new free electrophoresis method can use relatively simple equipment, or it can use sophisticated equipment that can automated and operated in continuous mode. The possibility of continuous operation strengthens the utility of free electrophoresis as an industrial process. Free flow electrophoresis, which, like density gradient electrophoresis is a subset of free electrophoresis, is an example of a continuous electrophoresis process. The amount of product dilu¬ tion, and hence the need for reconcentration, could be less than that encountered in chromatography. A major advantage of the invention is its ability to completely separate ribosomes from vesicles allowing their quantification and consistent recombination to form the final useful BRM product.

Drawings The disclosure of the present invention includes a written description and six drawings, of which:

FIGURE 1 is a drawing that illustrates density gradient electrophoresis apparatus that can be used in the method of the present invention. FIGURE 2 is a drawing that illustrates distribution of components at the start of an electro¬ phoresis procedure using the apparatus shown in Figure 1. FIGURE 3 is a graph that illustrates the ability of free electrophoresis to separate components of BRMs from starting material. The starting material was a BRM ready for clinical testing.

FIGURE 4 is a graph that illustrates the char- acterization of BRMs by density gradient electrophoresis. The starting material was cleared bacterial lysate.

FIGURE 5 (A and B) is a pair of graphs showing that bacterial ribosomes and membrane vesicles, when subjected to electrophoresis independently, have different electrophoretic mobilities and are separable from one another. This FIGURE (parts A and B) shows that the peaks in other electrophoresis experiments are correctly identified. FIGURE 6 is a graph that illustrates the electrophoresis of bacterial ribosomes and vesicles together with human erythrocytes, whose electrophoretic mobility is known.

Definitions

In the present specification and claims, reference will be made to phrases and terms of art which are expressly defined for use herein as follows:

As used herein, "non-toxic" means within a level of toxicity which is tolerable by the mammalian host receiving biologic response modifier therapy. As used herein, "non-immunogenic" means evoking a sufficiently low immunogenic response, or no response at all, such that undesired immune deviating, chronic inflammatory and hypersensitivity responses are not elicited, significantly, in the mammalian host. As used herein, "mean diameter" means the mean diameter of MSD Particle Size Distribution Analysis as measured on a BI-90 (Brookhaven Instrument Corp.) particle sizer. This measurement involves an intensity weighting of the size averaging process and is explained more fully in the Operator's Manual for the instrument, Chapter 6, which is incorporated herein by reference.

As used herein, "substantially non-pathogenic in humans" means not or rarely associated with disease in humans of normal health. Since most microorganisms are capable of causing opportunistic infections under the right circumstances, such as in persons whose immune systems have been compromised, this definition excludes only those organisms which typically cause non-opportunistic infections.

As used herein, "substantially free of endotoxin, intact cells, cell walls, and cell membrane fragments" means a low enough level of biologic activity of such fractions to maintain a non-toxic characteristic as defined herein.

As used herein, "natural membrane vesicles" mean membrane vesicles prepared from membranes which are derived from living or dead natural cells.

As used herein, ImuVert* is the registered trade mark of Cell Technology, Inc. , Boulder, CO, USA. ImuVert^® indicates the source of the biologic response modifier (comprised of natural membrane vesicles and ribosomes in a suspending buffer) disclosed in USSN 057,344 and PCT/US87/01397 and produced by Cell Technology, Inc.

As used herein, "free electrophoresis" means any method whereby solutes or particles are separated on the basis of charge in free solution, in the absence of a support matrix such as a gel, hydrocolloid beads, paper or fiber matrix similar to paper. "Free electrophoresis" includes, but is not limited to, density gradient electrophoresis, reorienting density gradient electrophoresis, free flow (or "continuous flow") electrophoresis, density gradient isoelectric focusing, free flow isoelectric focusing, recycling isoelectric focusing, and recycling electrophoresis.

As used herein, in free electrophoresis, "separands" are solutes or particles separated on the basis of charge in free solution, in the absence of a support matrix. As used herein, in the formula "cm²v^"1s^"1", cm means centimeters, v means volts and s means seconds.

As used herein, temperatures are in degrees Centigrade unless specified otherwise.

Summary of the Invention

In its broadest aspects the present invention is a method for separating ribosomes and natural membrane vesicles from other subcellular materials in a cellular lysate that has been subjected to mechanical forces sufficient to produce a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers. The method is comprised of subjecting the cellular lysate to free electrophoresis under conditions, and for an amount of time, sufficient to accomplish separation of the ribosomes and natural membrane vesicles from the other subcellular materials in the cellular lysate. In its more narrow aspects the present invention is a method for separating ribosomes and natural membrane vesicles from other cellular components in a bacterial cell lysate that has been subjected to mechanical forces sufficient to produce a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers. This aspect of the invention is especially useful in processes for producing biologic response modifiers (BRMs) comprised of ribosomes and natural membrane vesicles in a suspending buffer. When compared with prior methods for isolating ribosomes and natural membrane vesicles from other subcellular particles in a cellular lysate, the method of the invention has the advantage of potential for scaling to industrial BRM production level, automation, and continuous (as opposed to batch) operation. Once the ribosomes and natural membrane vesicles are isolated according to the method of the invention they can be mixed to produce useful pharmaceutical products such as biologic response modifiers.

According to the invention, known procedures are used to prepare a cell lysate from cells of a suitable source, eg. , cells from a bacterial species such as Serratia marcescens . The cell lysate is subjected to conditions that result in the production of a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers. Preferably the lysate is further treated to "clear" the lysate of large subcellular contaminants. The lysate is then subjected to the method of the present invention, namely use of free electrophoresis to separate the natural membrane vesicles and ribosomes from additional subcellular contaminants. In the production of BRMs, known procedures are used to mix the separated ribosomes and natural membrane vesicles in an appropriate combination to make a biologic response modifier that has the desired particle size and activity. Detailed Description of the Invention

Starting Material The starting material for use in the method of the present invention is a cell lysate or suspension of particulate material created when suitable cells are broken apart under conditions that produce natural membrane vesicles and ribosomes. This starting material is preferably "cleared", for example, by subjecting it to an initial centrifugation step, to separate out large cellular contaminants, and then subjected to the free electrophoresis procedures of the present invention for further purification of particles, which are useful for producing pharmaceutical products.

Production of the Cell Lysate Using Known Procedures Bacterial cells can be used to generate the cell lysate used as the starting material in the method of the present invention. A preferred isolation procedure for preparing lysates from bacterial cells is disclosed in USSN 057,344 and PCT/US87/01397. The procedure involves cultivating a seed lot of suitable bacterial cells in a suitable culture medium at a suitable temperature (about 37° C.) to a log phase culture; rapidly chilling the log phase culture to 0-4° C. ; harvesting the bacterial cells; washing and suspending the harvested cells by centrifugation at 0° C. (for example at about 8000 rpm in a Beckman JA10 rotor) to a prescribed density (of around 16% volume of cells per total volume) in a suitable buffering system which maintains an environment suitable for the formation and stability of the membrane vesicles and for the stability of the ribosomes; breaking open (lysing) the cells in a suitable cell disrupter or French pressure cell (at pressure between 10,000-20,000 psi) to produce membrane vesicles with a diameter in excess of about 110 n or 0.11 microns (preferably the disruption of the cells occurs in the presence of a suitable detergent for facilitating endotoxin dissociation) ; and then clearing the bacterial cell lysates of cellular debris, including intact cells, cell wall fragments, and large ribosomal aggregates and polysomes, preferably by centrifugation at 30,000 x g in a Beckman #30 rotor with ω²t = 3.55 x

10⁹ (approx. 21 min) .

In carrying out the above-described isolation procedure, rapid chilling of the log phase culture to 0-

4° C. may be carried out by any suitable means, such as, for example, by employing a dry ice-alcohol mixture, an acetone-ice mixture, an alcohol-ice mixture, or special devices such as cooling coils. All subsequent steps are preferably carried out at 0-4° C. Cell harvesting may be carried out by centrifugation, or cell harvesters/concentrators may be used. Initial clearing of cellular debris may be carried out by centrifugation.

The supernatant is collected by aspiration of the fluid from above the pellet while being careful not to disturb the pellet. The supernatant can then be filtered with a filter of about 0.45 micron pore size.

Biologic response modifier is made by measuring the concentration of ribosomes and vesicles and then combining them in the correct proportion to provide the appropriate size and activity.

Suitable Buffer A suitable buffering system for isolation of the membrane vesicle and ribosomal fractions according to the method of the invention is comprised of 10 mM MgS0₄, 10 mM NH₄C1, 4 mM Tris base, pH 7.2 - 7.6, 50 mM glycine, and 0-0.8M sucrose. For final suspension the above buffer or Tris or phosphate buffered isotonic saline of the same pH may be used. The magnesium sulphate may be replaced by any other suitable source of magnesium ions, such as magnesium acetate. The components of the buffering system and their specific concentrations may be varied as long as the integrity of the membrane vesicle and ribosomal fraction isolated via the described procedure is maintained. Tris-Hcl may be replaced by Trizma 7.1, Trizma 7.2 or any other suitable Tris buffering agent adjusted to pH 7.0 to 7.6. Any buffering system/agent that does not alter the integrity of the membrane vesicles or ribosomes and which is acceptably tolerated by tissues and the intact organism at the concentration employed may be utilized. The actual pH of the buffering system must be compatible with the maintenance of the vesicles and ribosomes, and of the tissues into which the material is injected.

The USSN 057.344 and PCT/US87/01397

Biologic Response Modifier The biologic response modifier disclosed in PCT/US87/01397 contains two distinct particle classes, namely natural membrane vesicles and ribosomes. The ribosomes may exist as monomers, dimers, trimers, or as larger polymers, but the mean diameter of the ribosome population is less than the mean diameter of the membrane vesicle population. The relative amounts of the two populations appear to affect the efficacy of the product as determined by standard NK cell assays conducted in vitro . The relative populations also affect the measured mean diameter of the total population of particles. It is believed that a mean diameter in excess of 170 n is necessary to achieve the desired efficacy. Below this level, efficacy of product as observed in standard NK cell assays appears to drop significantly. (The size of the natural membrane vesicles can be measured by at least two methods: (1) direct measurement of circularized cross-section seen in electron micrographs; and (2) mathematically, using measurement of particle diffusion coefficients obtained from light scatter analysis using a BI90 (Brookhaven Instrument Corp.) particle sizer.) The relative populations of vesicles and ribosomes also affect the ratio of the absorbance values at 260 nm/280 nm. The ribosomes by themselves have a ratio of around 2.0. The vesicles have a ratio of < 1.4 and ImuVert^® (which is comprised of a combination of ribosomes and natural membrane vesicles) has a ratio of around 1.8. It has also been observed that the size of the vesicles in the vesicle population have an effect on the efficacy of the biologic response modifier. Thus, in a preferred form, substantially all of the vesicles exceed 110 nm in diameter and the mean diameter of the vesicle population is at least 180 nm and is preferably about

210 nm. Preparations with diameters below these stated figures have been observed to be of less than desirable efficacy in in vitro NK cell assays. It is also clear that preparations containing membrane vesicles alone and preparations containing ribosomes alone fall below the desired levels of efficacy, suggesting possible synergism in presence of the two populations.

The ribosomes and natural membrane vesicles used to produce the USSN 057,344 and PCT/US87/01397 BRMs are preferably derived from (i.e., are endogenous to) the gram negative bacterium, Serratia marcescens . Serratia marcescens is a well-known organism and many strains are available from a number of sources. Sixty strains are available from the American Type Culture Collection, Rockville, Maryland 20852. (For example, Serratia marcescens , ATCC No. 60.) This organism provides a particularly suitable source for manufacturing BRMs exhibiting a high level of immunomodulating/immuno-therapeutic activity as compared to other bacterial sources, and is substantially free of toxicity. The desired bacterial membrane vesicles and ribosomes may be conveniently and economically isolated from a suitable source of stationary phase or log phase Serratia marcescens bacterial cells by means of the methods disclosed in USSN 057,344 and PCT/US87/01397, especially when used in conjunction with the improvements taught by the present invention. Reagents are employed which supply the necessary conditions for the maintenance of the integrity and conformation of the specific fractions isolated. Any reagents which might by themselves be toxic (unacceptably tolerated) or in any way influence or otherwise alter an immune response are avoided.

Separation of Ribosomes and Natural Membrane Vesicles

Using Free Electrophoresis

Summary It has surprisingly been discovered that free electrophoresis such as density gradient electrophoresis (DGE) can be used to separate natural membrane vesicles and ribosomes from a cell lysate. The DGE is preferably performed in a vertical cylindrical column to which a vertical electric field is applied through a density gradient in aqueous electrolyte solution. The sample (eg. , the cellular lysate, which preferably has been "cleared", or a control) is placed at the bottom of the density gradient, and separands migrate upward upon application of the electric field (5-20 hr at 2-15 A) . The technique is of sufficiently high resolution to separate natural membrane vesicles and ribosomes from the unwanted components of the cleared lysate. Procedures for preparing samples and operating the necessary apparatus are given below. Density Gradient Electrophoresis Apparatus Figure 1 is an assembly drawing (almost to scale) that illustrates a modular glass-and-teflon density gradient electrophoresis column that can be used in the method of the present invention. Its components consist of a water-jacketed vertical cylinder, two electrode side-arm vessels, a lower inlet cone and a capillary outlet at the top. All glass pieces are held together by teflon blocks with captured O-rings. In the column shown in Figure 1, the polyacrylamide gel bridges support the mass of the fluid column and provide conductive contact to the electrodes. Any glass modules (column, top, bottom, side-arm) with the correct diameter may be inserted into the teflon block to modify the geometry of the apparatus. Details of the stopcock system, gradient maker and electrode assemblies are given below.

Solutions Used in Density Gradient Electrophoresis Figure 2 is a schematic drawing of a density gradient electrophoresis apparatus indicating where solutions are placed. The composition of each of the solutions preferred for separation of bacterial subcellar particles are given in Table 1.

TABLE 1. Compositions of solutions used in density gradient electrophoresis of particulate biological response modifier, corresponding to solutions designated in Figure 2. Amounts given are in moles/liter.

10

15

The density gradient solution contains 9% (w/v) sucrose at the top and 25% (w/v) at the

bottom; the ionic strength of top solution is 0.052 g-ions/liter.

Procedures for Preparing Samples Samples (eg. , cleared cell lysate or ImuVert^® control) are preferably in a buffered solution, to which is added solid sucrose to a final concentration of 40% (w/w) . This solution has a density intermediate between that of the floor and that of the bottom solution.

Procedures for Operating Density Gradient Electrophoresis Apparatus

Glassware and teflon blocks are cleaned by normal laboratory procedures, including thorough rinsing with distilled water. The O-rings in the teflon blocks are lubricated with glycerol, and glass side-arms of the electrode vessels are fitted into the O-rings. A teflon rod is inserted into the teflon blocks in the position normally occupied by the column. A solution of freshly-mixed 15% polyacrylamide, 0.4% N, N'-methylbisacrylamide, 10% ammonium persulfate, 0.1% N, N, N', N'-tetramethyl- ethylenediamide is poured into the sidearms and allowed to polymerize. The teflon rods are removed and the assembly of the apparatus is completed. The entire system is filled with distilled water and checked for leaks. Ions are removed from the gel plugs by overnight electrophoresis with at least one change of distilled water. In other variations of this invention semipermeable membranes or ion-exchange membranes are used in place of polyacrylamide gel plugs.

Electrode vessels are filled, first with saturated NaCl, about 175 ml, then with top solution. Cooling water flows through the jacket at 4° C. Top solutions are inserted into the bottom of the column via the three-way stopcock (Figure 1) . Top solution and bottom solution are added to the respective compartments of a gradient maker, and 35-60 ml (depending on column length) of linear gradient is formed immediately below the "ceiling" formed by the original top solution. Without adding bubbles of air, sample is inserted via the three-way stopcock and a disposable syringe. This is followed immediately by enough floor solution to raise the top solution to the outlet capillary and the density gradient to the jacketed portion of the column, then all electrode connections are made.

The power supply is switched on at a constant current of about 4 - 30 mA, and voltage is monitored as a function of time. The position of the sample band (if visible) is also monitored as a function of time. After a suitable time (in these solutions about 5 - 25 hr) the current is switched off. Pumping solution is admitted slowly at the bottom inlet, and fractions are collected out the top of the column.

The fraction collecting stream may, if desired, pass through an optical monitor on its way to the automated fraction collector, which typically collects 0.5 - 3.0 ml fractions. These are subsequently examined by spectrophotometry and activity test.

Tests By measuring the optical densities (OD) of fractions at 260 nm and 280 nm the composition of the particles in the fractions is inferred. Fractions are selected on the basis of the ratio of OD at 260 nm to that at 280 nm and of mean particle size (ribosomes have a ratio of 2.0, and vesicles have a ratio of <1.4) . The appropriate fractions are recombined to give a ratio of 1.8 and concentrated if necessary to form the specific BRM. Concentration of the BRM may be by filtration, freeze-drying or centrifugation. To illustrate how this can be done, the centrifugation technique is described.

The fractions containing the BRM are combined in a Beckman #50.2 rotor tube and centrifuged at 106,000 x g for 2 hr at 0° C. with ω²t = 7.28 x 10¹⁰. Pellets are resuspended in suitable buffer to the appropriate concentration.

The fractions of interest have been identified by their electrophoretic mobility, gel electrophoresis and by spectrophotometry. The approximate electrophoretic mobility (EPM) is measured relative to the EPM of fixed human erythrocytes, which are electrophoresed along with cleared lysate or control (ImuVert^®) . The data are as follows:

TABLE 2

Polyacrylamide gel electrophoresis of individual fractions showed the proteins and their molecular weights associated with those fractions. These data are used to tell which fractions contain ribosomal proteins, vesicle proteins, and unrelated proteins. This technique is useful for verifying what is in each fraction and to assess the purity of each fraction. Sizes of individual fractions are determined by using the BI photon correlation spectroscopy system, BI-90 particle sizer.

Using ImuVert^® as an example, ribosomal RNA quantitation is based on nucleic acid content which is determined by the following formulas:

E260 = 0.0373 - 0.0079 (A260/A280)

Micrograms Nucleic Acid per ml = A260/E260

A 0.05 ml sample of the resuspended product concentrate is diluted in the appropriate resuspending buffer so that the A260 is between 0.4 and 0.5 for standardization purposes. The absorbances at 280 nm, 260 nm, 225 nm and 215 nm are then determined using the suspending buffer as the blank. Once the nucleic acid content has been determined the product is diluted to a final concentration of 1.0 mg nucleic acid per 0.5 ml buffer. The protein content of the vesicles can be estimated by using the following formula:

Micrograms protein/ML = 144 (A₂₁₅ - A₂₂₅)

Spectrophotometry Examination of Fractions A cleared cellular lysate (from Serratia marcescens bacteria) and ImuVert^® (both of which were prepared according to the procedures disclosed in USSN 057,344 and PCT/US87/01397) were subjected to free electrophoresis (density gradient) according to the method of the present invention in order to isolate the natural membrane vesicle fractions and the residual ribosome fraction. The resulting ribosomal and natural membrane vesicle fractions were examined by spectrophotometry. The results of the examinations are shown graphically in Figures 3 (ImuVert^® control) and 3 (cellular lysate) .

As indicated earlier, Figure 3 is a graph that illustrates the characterization of BRMs by density gradient electrophoresis. The starting material was a BRM ready for clinical testing. Each plotted point represents the amount of ultraviolet light having wavelengths of 260 nm (+) or 280 nm (squares) that is absorbed by material in the aqueous solution of each collected fraction (number shown on the abscissa) after separation by electrophoresis. Chemical analyses have indicated that, in the case presented in Figure 3, fractions 0 through 28 contain very little dissolved material, fractions 30 through 38 contain ribosomes used in constituting BRMs, and fractions 39 through 53 contain vesicles used in constituting BRMs. The graph itself also illustrates the electrophoretic separ¬ ation of the ribosomes and membrane vesicles which are the major component of ImuVert^®. The ratio of optical density at 260 nm to that at 280 nm is typically 2.0 for ribosomes and for vesicles < 1.4. This indicates that fractions 30- 36 contained ribosomes and fractions 40-53 contained vesicles. Thus ribosomes and vesicles can be separated from one another by free electrophoresis. Figure 4 is a graph that illustrates the ability of free electrophoresis to separate components of BRMs from starting material. The starting material was cleared bacterial lysate. Each plotted point represents the amount of ultraviolet light having wavelength of 260 nm (+) or 280 nm (squares) that is absorbed by material dissolved in aqueous solution in each collected fraction (number shown on the abscissa) after separation by electrophoresis. Chemical analyses have indicated that, in the case presented in Figure 4, fractions 0 through 20 contain dissolved proteins and nucleic acids not of use in the BRMs, fractions 20 through 28 contain ribosomes used in constituting BRMs, and fractions 30 through 44 contain vesicles used in constituting BRMs. Figure 4 thus demonstrates that the separation of the major components of cleared lysate, which consists of vesicles, ribosomes, dissolved protein, nucleotides and nucleic acids.

As the graphed data illustrates material with the high ratio of OD at 260 to OD at 280 nm (namely 2.0) appears in the tallest peaks; this consists of ribosomes.

The peaks corresponding to larger fraction numbers (25 - 43 in Figure 4 and 39 - 53 in Figure 3) have a lower ratio of OD at 260 to OD at 280 nm (namely <1.4) and contain natural membrane vesicles. In preparing BRMs, the natural membrane vesicle peaks and the ribosome peak are preferably combined to a concentration of about 2 mg/ml nucleic acid and a mean diameter of around 160 nm. Filtering is done with a 0.22 μm filter to sterilize the final product.

Activity Tests The ability of fractions to stimulate IL-2 production by human peripheral blood mononuclear cells was assayed. The test that was run to indicate activity of the fractions was a Genzyme IL-2 ELISA kit. In the case of both chromatographically prepared and electrophoretically prepared ribosomes and vesicles, both are required for full BRM activity. Data are summarized in Table 3. TABLE 3

Electrophoresis of Pure Ribosomes and Pure Vesicles FIGURE 5 (A and B) is a pair of graphs that proves that bacterial ribosomes and membrane vesicles, when subjected to electrophoresis independently, have different electrophoretic mobilities and are separable from one another. This FIGURE also proves that the peaks in all other electrophoresis experiments are correctly identified.

Absolute Electrophoretic Mobilities of Ribosomes and Vesicles

FIGURE 6 is a graph that illustrates the electrophoresis of bacterial ribosomes and vesicles together with human erythrocytes, whose electrophoretic mobility is known. From independent laboratory data, determined by microscopic electrophoresis, at the ionic strength of the density gradient, namely 0.052 equivalents/Liter, human erythrocytes have electrophoretic mobility of -1.32 ± 0.14 x 10^" cm²v^"1s^"1 (wherein cm = centimeters, v = volts and s = seconds), and they migrated 21 fractions in the experiment of FIGURE 6. Ribosomes migrated 28 fractions, and the vesicles migrated 15 fractions, so their electrophoretic mobilities are, 1.76 and 0.94 x 10^*4 cmVs^"1, respectively. This large difference in electrophoretic mobilities, which is not expected on the basis of any published work, implies that satisfactory separation of these components is possible by any method that uses electrophoresis.

Other Electrophoresis Methods Given the data of the previous section, the quality of separation by continuous flow electrophoresis (CFE) and reorienting gradient electrophoresis (RGE) can be calculated. The distance migrated in CFE is given by x = pEτ , in which μ is the electrophoretic mobility, E is electric field strength and T is residence time, the time required for a particle to complete its traverse of the separation chamber, from entrance to exit. The latter two variables are under the control of the operator. In a flow chamber 10 cm wide, vesicles would migrate 4.7 cm if ribosomes migrated 8.8 cm, so that the two separands would be separated by 4.1 cm and possibly 40 fractions in some separators (having 10 outlets per cm) . In RGE, with typically 2 cm between electrodes, vesicles would migrate 0.94 cm if ribosomes migrated 1.76 cm; this 8 mm separation, which requires less than 20 min, results in a 2 - 3 cm separation upon reorientation.

Isoelectric Focusing Separation experiments using isoelectric focusing demonstrated that the components of ImuVert^® precipitate on any pH below 6.0, and the isoelectric pH's of the particulates (ribosomes and vesicles) are considerably below pH 6.0. References The following references are expressly incorporated by reference into the present specification.

1. PCT International Application PCT/US87/01397; Applicant: Cell Technology, Inc. , Boulder, CO, USA; Inventor: Richard E. Urban; Title: Biologic Response Modifier; published 17 December 1987 as International Publication No. WO87/07503; claims priority from USSN 057,344.

2. R.C. Boltz, P. Todd, M.J. Streibel and M.K. Louie. "Preparative Electrophoresis of Living Mammalian Cells in a Stationary Ficoll Gradient". Preparative Biochem. 3.(4), 383-401 (1973).

3. R.C. Boltz, Jr., P. Todd, R.A. Gaines, R.P. Milito, J.J. Docherty, C.J. Thompson, M.F.D. Notter, L.S. Richardson, and R. Mortel. "Cell electrophoresis research directed toward clinical cytodiagnosis". J. Histochem. Cytochm. 24., 16- 23 (1976) .

4. R.C. Boltz and P. Todd. "Density gradient electrophoresis of cells in a vertical column".

In Electrokinetic Separation Methods (eds. P.G.

Righetti, C. J. van Oss, and J. Vanderhoff)

Elsevier/North-Holland Biomedical Press,

Amsterdam, 1978, pp. 229-250. Summary Various modifications of the invention in addition to those shown and described herein will become apparent to those skilled in the art from the foregoing description and accompanying drawings. In particular, any form of free electrophoresis can be used in the preparation of particles from living material for use in BRMs. In addition, this method is useful for the isolation of ribosomes from mixtures from cell lysates. Such modifications are intended to fall within the scope of the appended claims.

Claims

WHAT IS CLAIMED IS:

1. A method for isolating ribosomes and natural membrane vesicles from other subcellular materials in a ribosome containing lysate that has been subjected to mechanical forces sufficient to produce a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers, said method comprising subjecting said lysate to free electrophoresis under conditions and for an amount of time sufficient to separate said ribosomes and said natural membrane vesicles from said other subcellular materials in said lysate.

2. A method according to Claim 1 wherein said lysate is subjected to free electrophoresis selected from the group consisting of density gradient electrophoresis, reorienting density gradient electrophoresis, free flow or continuous flow electrophoresis, density gradient isoelectric focusing, free flow isoelectric focusing, recycling isoelectric focusing, and recycling electrophoresis.

3. A method according to Claim 1 wherein the ribosomes and the natural membrane vesicles that are separated as the result of the lysate being subjected to free electrophoresis are combined in a suitable buffer to make a biologic response modifier that has the desired particle size and activity.

4. A method according to Claim 1 wherein said lysate that has been subjected to mechanical forces sufficient to produce said particle population is cleared of larger subcellular particles before being subjected to free electrophoresis.

5. A method according to Claim 4 wherein said clearing of said lysate is accomplished by centrifugation.

6. A method according to Claim 1 wherein said lysate is a lysate produced as the result of lysing viruses or cells selected from the group consisting of prokaryotic cells and eukaryotic cells.

7. A method according to Claim 6 wherein said lysate is a lysate produced as the result of lysing pro¬ karyotic cells.

8. A method according to Claim 7 wherein said prokaryotic cells are bacterial cells.

9. A method according to Claim 8 wherein said bacterial cells are gram negative bacterial cells.

10. A method according to Claim 9 wherein said gram negative bacterial cells are Serratia marcescens cells.

11. A method according to Claim 7 wherein said eukaryotic cells are selected from the group consisting of yeast cells, vertebrate cells and invertebrate cells.

12. A method for isolating ribosomes and natural membrane vesicles from other subcellular materials in a ribosome containing lysate that has been subjected to mechanical forces sufficient to produce a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers, said method comprising subjecting said lysate to density gradient electrophoresis under conditions and for an amount of time sufficient to separate said ribosomes and said natural membrane vesicles from said other subcellular materials in said lysate.

13. A method according to Claim 12 wherein said density gradient electrophoresis is performed in a vertical column to which a vertical electric field is applied through a density gradient in aqueous electrolyte solution.

14. A method according to Claim 13 wherein said lysate is placed at the bottom of the density gradient and the separands migrate upward upon application of the electric field.

15. In a method for producing a biologic response modifier, wherein said method comprises cultivating bacterial cells of the strain Serratia marcescens, harvesting the cultivated cells, dissociating endotoxin with an appropriate detergent, subjecting the cellular concentrate to a treatment sufficient to produce ribosomes and to produce a particle population which includes natural membrane vesicles and at least some ribosomes in the form of monomers, dimers and trimers, separating said ribosomes and membrane vesicles from the remaining cellular material in the cellular lysate, and combining said ribosomes and said natural membrane vesicles in a suitable buffer to make a biologic response modifier that has the desired particle size and activity, the improvement which comprises subjecting said lysate to free electrophoresis under conditions and for an amount of time sufficient to separate said ribosomes and said natural membrane vesicles from said other subcellular materials in said lysate.

16. A method according to Claim 15 wherein said lysate is subjected to free electrophoresis selected from the group consisting of density gradient electrophoresis, reorienting density gradient electrophoresis, free flow or continuous flow electrophoresis, density gradient isoelectric focusing, free flow isoelectric focusing, recycling isoelectric focusing, and recycling electrophoresis.