GB2242903A - Protein purification by multiphase chromatography - Google Patents

Abstract

A method for the fractionation of proteins in effective amounts on one column in one operation comprises (i) reversibly modifying species of proteins, Pi, in solution through the epsilon-amino group of a lysyl residue of each protein with a modifying compound D which reacts selectively and reversibly with said amino group; (ii) subjecting the modified proteins, PiD to column chromatographic separation to form a new stationary phase, M.D.Pi; (iii) dissociating the individual proteins and removing them to the mobile phase, according to the principle of chemical equilibria in multicomponent systems; and (iv) readsorbing the individual proteins on the original matrix, from which they are removed to the mobile phase according to the partition coefficient between the Matrix.Solute.Solvent and eluted. The compound D may be phenosulphonephthalein, luciferin, a polyphenolic aldehyde, a folate or folate analogue or heparin.

Description

PROTEIN PURIFICATION BY MULTIPHASE CHROMATOGRAPHY The invention: Multiphase chromatography is a novel method for fractionation of proteins in effective amounts on one column in one operation. In this method firstly: i species of proteins, P. in solution are reversibly modified through the àmino group of a lysinyl residue of the macromolecule with a compound of distinct chemical properties; subsequently the modified proteins, PiD, are subjected to column chromatographic separation on commercially available adsorbent material.PiD desposits on the top of the adsorbent matrix (2 to 5% of the adsorbent material) forming a new stationary phase, M.D.P; from this phase the individual proteins are dissociated and removed to the mobile phase, according to the principle of chemical equilibria in multicomponent systems; they are readsorbed on the original matrix from where they are removed to the mobile phase according to the partition coefficient between the Matrix.Solute.Solvent, and eluted in the front.

By the above method all the proteins, initially subjected to the column in the form of PiD, are recovered, and fractionated in the eluates Due to the intrinsic properties of the system, 20 to 50-fold more protein can be applied on the same size of column and the same type of adsorbent material compared with the conventional chromatographic method. The fractionation of the individual protein takes place on the principle of the equilibrium constant of the individual proteins in the PiD complex on a new, self-created stationary phase with i number of plates. This is in addition to the fractionation of the individual protein in the conventional adsorbent material.

The method of Multiphase Chromatography permits: a) purification of potent, biologically active proteins, namely Lymphotoxines, Human Necrosis Factor(s), Cytotoxic Killer Factor(s), Macrophage Activating Factor(s) secreted by the cells (established cell lines) to the culture media, where they are present among a large number of other proteins in large volumes; b) separation, purification and quantitative determination of the amounts of isoenzymes and variant forms of enzymes in blood samples of patients with various infectious diseases and malignancies. Both the purification and quantitative measurements of these protein-enzymes have to be made on blood samples taken from individual patients, consequently the amounts of these samples are restricted and irreplaceable.

INTRODUCTION This invention relates to Multiphase Chromatography for the separation and purification of proteins and protein factors and to the production of biologically important proteins in pure form.

(a) Portent; biologically active proteins present in minute amounts in physiological fluids are of great interest for basic biological knowledge and therapeutic use. In recent years, a number of laboratories have been involved in the study of human lymphotoxins, human necrosis factors, human macrophage activating factors, but progress has been limited, primarily because of the difficulties of their production in pure form in sufficient amounts for biochemical and physical characterizations.

The main difficulties in their production are: That they are found in small quantities amongst a large number of other proteins in large volumes; there has hitherto been only very little information on their biochemical properties.

A problem which has been encountered, therefore, is to separate a protein Factor(s), F, present in a minute amount amongst a large amount of other proteins e.g.

in physiological fluids, cell culture media.

(b) Isoenzymes and variant forms of the enzymeprotein appearing in the circulating blood and in the cells of the circulating blood of patients with various types of infectious diseases or malignancies. The amounts of these biological materials are limited.

The Methodology of the invention: Multiphase Chromatography involves: 1. Reversible modification of i protein species, denoted Pi, where i stands for the n th number of the individual proteins composing the physiological fluid and of soluble protein fractions of cell extracts through the amino group of the lysinyl residue of the macromolecule, with a compound, denoted D, having the following structures: (Scheme 1)

Scheme 1 Below pH 5.0 the unionized form, I exists, which form cannot react with the i-amino group of lysinyl residue of the macromolecule at an appreciable rate. In the pH range 5.0 to 6.2 the monoionized forms II and III are structures in resonance. The reactions of P. are with the structures

In the pH range 7.4 to 8.2 the di-ionized forms IV and the resonance structures V and VI can exist.

2. In the pH range 5.0 to 6.2, at 0.05 At buffer concentration the i individual modified, proteins PiD, are in reversible equilibrium with the reactants. The equilibrium constant, K, of each individual protein is different and dependent on pH and ionic strength at constant temperature. In cases where stoichiometric amounts (1:1, ratios of the reactants) were used, the reaction between P. and D can be described by the set of equations given in Eq.l.

i = n th protein, n values are 1, 2, 3....

stands for pH or ionic strenght.

3. Consequently to 2. the bonds between P..... PiD are susceptible to dissociation.

4. By raising the pH of the solution containing PiD to pH 8.0 (retaining the buffer concentration 0.05 M) D in P.D is transformed to the structures illustrated in Eq. 2.

Eq .2 5. The solution described in 4, is transferred to a column containing conventional adsorbent material: DEAE - cellulose, Sephadex(agarose); Sepharose, crosslinked dextran; Glass beads coated with amino ethanol.

6. Adsorption of P.D (IV)

The adsorptin of PiD to the stationary phase containing the conventional matrix, M, pre-equilibrated with buffer pH 8.0, 0.05 M, creates a new stationary phase with i plates, Phase II in the upper part of the column (2 to 5% of the dry weight of the adsorbent material).

7 PiD is attached to the adsorbent through D in PiD, according to Eq. 3.

M + DPi(IV,V,VI structures)

M.D.Pi Eq. 3 8. The relationship between the adsorption contant, mKiv of of DPi to the matrix, and the dissociation constant of PiD, KY can be described by Eq. 4.

mKiu Ki Eq. 4 9. The attachment between the Matrix and PiD complex described in Eqs. 3 and 4 provides means of removing the individual proteins from Phase II before D is removed from that phase.

10. The removal of the individual proteins to the mobile phase takes place after the dissociation of the individual proteins from the complex M.D P 11. The dissociation of the individual proteins from Phase II takes place according to K1.

12. The Kf are dependent on pH and ionic strength.

Therefore, the dissociation of the individual proteins from Phase II and their removal from this phase to the mobile phase are taking place dependently on the composition of the solvent, buffer7 passing through the column.

13. Consequently, the number, n, of proteins removed at a certain pH gradient and ionic strength, is dependent on the equilibrium constant, K? of the individual proteins in M.D.P. - complex and not on their relative concentrations in the physiological fluid.

14. Property number 13 of Multiphase Chromatography is very important since the Factor(s) is(are) present in small concentrations relative to the other proteins in the above fluids.

15. The proteins are removed from Phase II to the mobile phase adsorbed on the original matrix in Phase I. The number, n, of the protein species to the mobile phase at a certain pH and ionic strength can be relatively small and, consequently., interference is avoided on the adsorbent.

16. Proteins, n number of individual proteins7 adsorbed on the matrix in Phase I are removed to the mobile phase according to the partition coefficient of the Matrix.protein complex and the mobile phase.

17. All the proteins are eluted in full, amounting to a 100% recovery from the column in close relationship to their native form present in the biological fluid.

The statements put forward in paragraphs 1 17 are supported by experimental findings and they are in accordance with physical and chemical principles.

Description of Multiphase Chromatography in Detail Multiphase chromatography, MC, is a preparative chromatographic method (a) for processing large quantities of physiological fluids, cell culture media with effective recovery of the purified protein(s), the Factor(s) of interest; (b) to isolate protein(s), appearing in the circulating blood and in the cells of the circulating blood of an individual patient with various infectious diseases and with malignancies. The amounts of the blood available of the individual patients are restricted.

The method invented: Multiphase Chromatographic System, MCS, consists of two stages, namely, Stage A: Reversible modification of i species of protein, Pi, of physiological fluids (type a, type =) in solution; Stage B: Fractionation of PiD, the modified proteins by column chromatography.

Stage A. Modification of i protein species, Pi of physiological fluids.

Introduction of a compound, D, (five different compounds were used) to i species of proteins, present in physiological fluids, D reacts selectively and reversibly in the pH range 5.0 to 6.0 with the -amino group of lysinyl residue of the polypeptide chain.

A-l Protein(s) of physiological fluid investigated: A-l-l Lymphotoxins, Factor I and Factor II produced by a human lymphoblastoid cell line Karpas 160 (1977) Br. J. Cancer, 35, 152-160. Neumann, H. and Karpas, A. (1981). Biochem, J.

A-1-2 Human Macrophage Activating Factor, V G 37, Wallenbery Uppsala, Kenneth Wilson 1971: cell line of a patient with histocytic leukaemia.

A-1-3 Isoenzymes variant forms a leukocyte alkaline phosphatase in transformed lymphoblast of a patient with acute lymphoblastic leukaemia and in myeloblast of patients with acute and chronic myelogenous leukaemia.

A-1-4 The name of modifying compound is as listed below: Da: phenosulphonephthalein, sulphenthal Db: Luciferin Dc: Gossypol is a polyphenolic aromatic aldehyde: 1,1',6,6',7,7'-hexahydroxyl 3,3'-dimethyl- 5,5'-diisopropyl-2,2'-bis-naphthyl-8,8'- dialdehyde Dd: Folate and Folate analogues De: Heparin is a sulphated glycosaminoglycans polysaccharine.

A-2 The reacting group of the proteins: The g -amino group of lysinyl residue of i protein species of physiological fluids was chosen to be modified by D. Even though the reactions take place with the same amino acid residue due to the versatile nature, structure, micro-environ ment and their pK value, the reactions of this group with the same modifying compound, D, should yield equilibrium constant, K. accordingly.

A-3 The modifying compounds and the chemical and physical properties of the modifying compounds: The modifying compounds, D, selected have: two dissociation constants, at least 1.2 pH units part in the pH range 5.0 to 8.5; 2. D is a molecule in which the parent structure I is in reversible equilibrium with two sets of substructures, in hierarchical order, depending on pH and ionic strength according to the Scheme given below:

Scheme A-3 3. The structures I, II, III, IV, V, VI respond delicately to changes both of pH and of ionic strength.

A-4 The reactions between P. and D A-4-l The g -amino group of the lysinyl residue with the unique and different pKa value characteristic of the individual proteins reacted with the structures

A-4-2 The reactions between the # amino group of the lysinyl residue of i species of proteins in 0.08 M buffer solution in the pH range 5.2 to 6.2 at 20"C and the modifying compound D can be described in the Scheme and set of equations given below, provided that the reactants are in stoichiometric amounts: A-4-2 Scheme

D can be any of the modifying compounds listed in A-1-4; 9 stands for the pH and ionic strength in which the reactions were carried out.

A-5 Reactions take place in solution, and are monophase reactions which can be analysed and quantitatively monitored by measuring spectral changes, pH titra tions, from which measurements the stoichiometry of the reaction and the equilibrium contents, K1v can be established. The compounds used are well characterised both chemically and physically and preferably the interactions of proteins are also known. Ail the reactions are reversible so that all the interacting compounds may be recovered in their "parent" form. It is also of the utmost importance that all the reactions be hierarchical.

A-6 Determination and Calculation of Kvi Computer calculations of chemical equilibria and stability constants using various numerical and analytical methods and programs are available.

References for the calculations of the equilibrium constants Kv1 , Kv2...... kvi; P. Gans, Coord. Chem.

Rev. 19, 99 (1976) F. Gaiper, ibid, 27 195 (1979); Moter Eartis, R. J. Martell, A. E. Can.

J. Chem. 60, 2403 (1982).

In these methods the pH values (or ionic strength) at constant temperature, are the measured dependent variables that reflect the unique set of equilibrium constants, while the interacting i species of proteins, Pi, and the modifying compound and their concentrations are the input parameters.

The knowledge of the equilibrium constants has great importance in separations of the individual proteins, Factors, by using the Multiphase Chroma tography.

A-7 The properties of P D A-7-l The reactions in solution of i species of proteins, Pi, of one physiological fluid with one of the modifying compounds, Da were studied in detail in 0.08 M buffer at pH 8.0 at 20"C.

Spectral Properties of P D A-7-1 The reactions of P. with Da resulted in the development of distinctive sharp features in the ultraviolet and visible light spectrum 250 nm to 600 nm. Protein concentrations (Lowry Method) and spectral analysis of several different prepara tions of PiDa (same physiological fluid and the same modifying compound, but different protein concentrations) were made (see Fig. 1). The adsorbancy at 280 nm vs 559nm gave a mean value of 2.9, and were well correlated with the protein content of each preparation.

A-7-2 The stoichiometry of the reactions between P.

and Da was established from the isosbestic point, which is the point of intersection of the adsorp tion curve of PiDa and the adsorption curve of Das one of the reactants, in excess.

A-7-3 Fractionation of the modified proteins, PiDa, according to their molecular weight and buoyant density S20w.

Sucrose - gradient - ultracentrifugation of PiDa was performed as described in the Experimental Section. The modified proteins separated according to their molecular weight to 10 fractions.

In each fraction the molecular weight the sedimentation velocity constant, S20 W were estimated; the adsorption spectra, protein content were measured.

The following statements, based on the results of the experiments can be made: 1. Da , as a free compound, Mr = 500, was not present in the fractions in the original sample and did not form in sucrose-gradient-ultracentrifugation.

2. Free proteins were also not present or did not form in the conditions of which the sucrose-gradientultracentrifugations were carried out.

3. P.D fractionated to 10 subclasses with distinct ia Mr values ranging from 10,000 to 125,000 or more, as established both from known markers in sucrose-gradientultracentrifugation and from SDS polyacrylamide gel electrophoresis.

4. The findings presented in 1, 2, 3 show that PiDa are stable complexes under the experimental conditions (pH, ionic strength and temperature).

5. The absorption spectra, 250 nm to 600 nm, of the 10 fractions were measured and given in Fig. 2a and b. The curves of the various fractions show a similar pattern, maximum absorptions in the ultraviolet range at 280 nm and in the visible range at 559 nm.

The absorbancy at 280 nm of the l0fractions found to be linearly related to the protein content (measured by the method of Lowry) and of the adsorption of Da at 559 nm.

6. From the above results presented in 5 it could be concluded that one functional group - the -amino of the lysinyl residue of the i species of the proteins is modified by D involving one lysinyl residue in each a protein.

A-7-4 Evidence that the amino group of the lysinyl residue of the polypeptide chain of i species of proteins, P. and only one of each protein reacted with Da i species of the modified proteins, PiD were fractionated to subclasses according to their buoyant density (S20 ; Mr) by sucrose gradient ultracentrifugation (see: Methods).

In the subclasses so separated the protein content, and the adsorptions in the wavelength range of 250 nm to 600 nm were measured and the Mr values were estimated by polyacrylamide gel electrophoresis.

The absorbancy, OD, at 280 nm and 559 nm were linearly proportional to the protein content of each fraction, whereas the ratio of the absorp tion at the two wavelengths was found to be a function of the average Mr of the proteins in each subclass. These results could be inter preted that one functional group, and only one of the individual proteins has reacted with Da A-7-5 No special shift of the absorption curves of subclasses was observed. This finding strongly suggests that only one type of functional group - probably the f-amino group of the lysinyl residue of the individual proteins is reacting with Da Stage B Protein purification on a column.

B-l The pH of the solutions containing PiD was raised from 6.2 to 8.0. At the latter pH the following transition of PiD takes place.

PiD \ PiD(IV2V2VI) 1 1 PiD(IVVVI) B-2 Conventional column adsorbent materials, Matrix 1 ion exchange materials: Sephadex Matrix 2; Sepharose Matrix 3; Glass Beads coated with amino-ethanol Matrix 4; were equilibrated with a buffer pH 7.8, 0.05 M for 48 hours at 4"C.

B-3 The modified protein solution (pH-7.8) P.D8 was applied to the columns.

B-4 P.D deposited on the upper art of the column (occupying 2 to 5% of the dry weight of the adsorbent materials), and doing so a new solid phase with i number of plates M.D.Pi was formed, according to set of equations; Eq.2 e Matrix + PiD (1V,V,V1)

Matrix.D.P.

where K are the binding constants between the Matrix, M, and i individual proteins at varying pH (ionic strength) 77, at a given temperature.

B-5 The binding of the modified proteins to the Matrix occurred through the modifying compound, D in P. D.

B-6 D in PiD has a greater affinity to the Matrix compared to the dissociation constant Kvi of the individual V proteins in PiD in the experimental condition: MKv2 Kiv...

B-7 Consequently to B-6, i species of proteins in M.DlPi will be dissociated from Phase II before D dissociates from the Matrix.

B-8 Proteins are removed to the mobile phase after dissociating from the Matrix.DlPi complex.

They are dissociated according to K. depending on the pH (or ionnic strength) of the solvent passing through the column. Therefore the number of proteins, Pnt removed at a certain buffer concentration from Phase II is dependent not on relative concentrations of the individual proteins, but on Ki9. The pH gradient of the mobile phase, the buffer solution, can be made up according to the dissociation constant of the individual protein(s), the Factor(s).

B-9 The proteins, Pnt removed from Phase II are readsorbed on the original Matrix. The number of the individual protein, n, removed at a certain pH (pH changed from 7.8 0.05 M to 1.0 M at a given temperature), can be relatively small, n i. The reduction of the number of proteins readsorbed (this is like initially only a few individual proteins being applied) has great importance in chromatographic separations (ideal Langmuir adsorption avoiding interference of the proteins).

B-1O The proteins form Phase I. Pn are removed to the mobile phase in accordance with the partition coefficient of Phase I. Pn and the mobile phase. Better separations of the individual proteins were achieved when Phase I, Sephadex or Sepharose, was used. In that case the elution from Phase I is independent from ionic strength.

BF l Measurements in the eluate tubes.

In the eluate tubes the concentrations of PiDl Pi D, pH and ionic strength were measured and the absorption spectrum was recorded in the wavelength range 260 nm to 650 nm. See Fig.4.

In general; in the cases of the four absorbent materials M1,M2,M3,M4, with the modified i species of proteins of physiological fluid investigated the following results were obtained: at pH 8.0, ionic strength 0.05 M AE/HC1 at 40C, independently from the volume of the buffer PiD(IV,V,VI structures) passed through the column, no PiD, P. and D could be detected. Additionally, the band initially deposited on the top of the column did not widen and did not diffuse (visible observation in case of Da). Fram these results it was concluded that a new stationary phase, Phase II was created on the top of the column on the four adsorbent materials.In experiments where the unmodified proteins in solution were subjected to chromato graphic separations using Sephadex or Sepharose, all the proteins were found to be eluted in the same volume passing through the column. On ion exchange adsorbent materials, 15 to 20% were found in the eluates using buffer solution 0.05M pH 7.8.

B-ll-2 By changing the pH and ionic strength of the elution buffer using non-linear gradient elutions with the compositions of the buffer listed hereunder: a. pH 8.0 to 7.8 ionic strength 0.05M to 0.10M, at 40C b. pH 7.8 to 7.6 ionic strength 0.10M to 0.25M, at 40C c. pH 7.6 to--7.4 ionic strength 0.25M to 0.50M, at 40C d. pH 7.9 -------- ionic strength 0.50M to 1.00M, at 40C the following observations were made:PiD never appeared; all the proteins, Pi, eluted in accordance to Kç (the dissociation partition coefficient K of constants of the individual proteins in PiD) and the original adsorbant Matrix (M1,M2,M3,M4). They were fractionated to various numbers of peaks, according to the nature of the Matrix, with 100% recovery of the proteins subjected to chromatographic separations; D was eluted in full at pH 7.95, 1.00M AE/HCl buffer concentration. Furthermore, D was eluted in a pattern specific to the nature of the Matrix in which the chromatographic separation was performed. All these results strongly support the claim made and described in Part B Sections 1-10.

B-12 In the eluates the biological activity of the Factor(s) was (were) assayed. The fractions, eluate tubes, where the Factor(s) appeared were further analysed for protein content, molecular size, pI, binding to lectins of different kinds.

The above characteristics are important for the production of the Factor(s) sought for in its (their) pure form.

C. Advantages of MCS 0-1 Twenty to fifty fold amounts of protein in the form of PiD(IV,V,VI) can be applied in one injection on the same adsorbent Matrix (M1 ,M2,M3,M4), in the same size of column compared to the unmodified i species of protein, Pi.

0-1-1 Consequently to C-1, the effective recovery of the Factor(s) is considerably greater: fewer column manipulations, measurements, biological assays are needed to be performed.

C-1-2 In more concentrated fractions the biological assays can be performed with higher accuracy; C-2 All the proteins are dissociated from Phase II and removed to Phase I, to the original Matrix according to Ki and not according to the concentration ratio of the individuals proteins present in the original fluid.

C-2-1 Accordingly, the number, n, of the individual proteins removed from Phase II can be chosen, and controlled by the pH and ionic strength of the mobile phase. Therefore n can be small compared to i (nci).

C-2-2 Due to C-2-1, the interference of the proteins on the column can be avoided (up to 5 to 10 solutes), consequently nearly ideal isotherms (Langmuir) were observed. The removal of the individual proteins from Phase II takes place according to Ki, the number of proteins, n, reaching the stationary phase ! M1, where the actual separations are taking place, greatly reduced.

C-3 All the proteins are recovered from MCS. The proteins from the Matrix D.Pi are dissociated before D. The dissociation constants of P. in the above complex, K. 1 1 M where M are K. were K. are the association constants between the 1 1 Matrix and D in P.D.

C-4 The proteins are recovered with their parent structures and the biological activity present in the original fluid before separation. This may be due to the D protecting the proteins from other interactions, digestions etc.

C-S Multiphase Chromatography, MC, is constituted of two reactions described in Part A and Part B. Since the reactions in Part A are taking place in monophase and the reactions in Part B are taking place in heterogeneous phases, the two types of reactions can be well separated from each other.

C-6 Choice of adsorbent materials: As much as conventional chromatographic substances are available and can be adapted to the properties of the Factor(s) of interest.

C-7 It is a general method. Any protein mixture can be fractionated on a MCS.

C-8 All the reactions on the column take place in hierarchical order.

C-9 The quantitative relationship between isoenzymes and variant forms, can be established since all the proteins initially transferred to the columns are recovered in the eluate tubes.

E. Experimental Part Comparative experiments on four different adsorbent materials.

E-l Comparative experiments on the efficiency of the fractionation of i species of proteins of physiological fluids on four different adsorbant materials, M1,M2,M3, M4.

E-2 Same amounts of PiD (0.8 g)solution were applied on four column packed with M1, M2, M3, M4. The columns were preequilibrated with 0.05M AE/HCl buffer pH 7.8 at 40C.

E-2-l PiDa solution deposited on the top of the Matrix (Ml, M2, M3, M4) and formed a narrow band (2 to 5% of the dry weight of the adsorbant).

E-2-2 Buffer pH 7.8, 0.05M at 40Cas allowedtoflow through the columns for 48 hours. In the eluate tubes no PiD no P. and no D could be detected.

1 1 a E-3-l Elutions of the columns with each buffer with the indicated compositions were performed for 48 hours, 3ml per hour flow rate. pH and ionic strength in the elution buffer used had the following compositions: a. pH 7.8 - 7.6 0.05 to 0.10M AE/HC1 b. pH 7.6 - 7.5 0.10 to 0.25M AE/HCl c. pH 7.5 - 7.40 0.25 to 0.50M AE/HC1 d. pH 7.95 - > 0.50 to 1.00M AE/HCl E-3-2 The eluate tubes.r-were analysed for the following parameters: pH, ionic strength, protein concentration Lowry method, absorbancy in the wave-length range 250nm to 600 nm and bioloaical activity. Molecular weight and pl were estimated by polyacrylamide gel electrophoresis only in tubes where the biological activity was present.

E-3-3 The modified proteins, PiD were not eluted from any of the columns, M1, M2, M3, M4. See figures.

E-3-4 All the proteins, P. originally subjected for column separations eluted from the column before Da, i.e. they have to be dissociated from the new solid phase M.Da.Pi before their removal to the mobile phase, the buffer passing through.

E-3-5 n individual proteins were eluted at a given gradient of the buffer from the four columns, with pattern "fitted" to the nature of the original Matrix 1 M2, M3, M4). See figures.

E-4 Da eluted from the column only at pH 7.95, 1.OM AE/HCl buffer at 4"C.

E-5 The analysis of the tubes for the parameters indicated in Section E-3-2 illustrated in figures E-3-5:1 elutions patterns of the proteins vs eluate volume on DEAE - cellulose; Fig.9 E-3-5-:2 elution pattern from Sephadex G-25, Figs.5-8 E-3-5-:3 elution pattern from Sepharose 4-B.

E-6 Biological activity vs elution volume. See figures.

E-7 With this particular Factor the separation and extent of purification were found to be optimal on Sepharose 4-B.

Legend for Table 1: Equal amounts of PiDa (1 gr. - specific activity 20 units) were transferred to two identical DEAE-cellulose and Sepharose 4B columns. In the eluate tubes biological activity and protein content were measured and specific activity calculated in units (see methods).

lA*, 1B* columns eluted under similar conditions with measurements in combined and separate tubes.

Similarly 2A** and 2B** Sepharose 4B columns eluted under similar conditions but with and without combination of eluate tubes.

Table 1 Purification of FA by Multiple Chromatograph

Absorbant material Recovered of column Sp A Total A Unit Unit 1A* DEAE-cellulose 4,000 50,000 column size 2x40 cm 11 ml combined eluates 1B* DEAE-cellulose No. tube 26 lml 2,800 column size 2x40 cm 27 lml 12,000 28 lml 16,800 29 lml 3,500 30 lml 3,200 31 lml 2,800 32 lml 1,200 33 lml 600 34 lml 400 35 lml 400 36 lml 200 total 11 ml 43,900 2A** Sepharose - 4B I 6,500 32,500 column size 1.5x100 cm II 1,500 7,500 15 ml combined eluates III 640 3,200 43,200 2B** Sepharose - 4B No. tube 51 lml 200 column size 1.5x100 cm 52 lml 800 53 lml 3,000 54 lml 20,000 55 lml 6,000 56 lml 2,000 57 lml 500 58 lml 400 59 lml 600 60 lml 200 61 lml 400 62 lml 180 34t280 Materials and Methods Chemicals. p-Nitrophenyl phosphate (p-NPP) and 5,5-dithiobis(2-nitrobenzoic acid) were purchased from Sigma Chemical Co. and Aldrich Chemical, Co., respectively.

DEAE-cellulose (DE-23) was purchased from Whatman Co.; Sephadex and Sepharoe, from Pharmacia Fine Chemicals; Tris, from Sigma Co. Aminoethanol-O-phosphate was purchased from British Drug Houses; O-phosphoryl-L-serin (Lot 60622), from Calbiochem. Stone, Kent, UK; Sephadex G-25, G-lOO and G-200, Sepharose 4B, DEAE-Sepharose and CM-Sepharose were from Pharmacia Fine Chemicals (Uppsala, Sweden); concanavalin A-Sepharose was from Miles-Yeda, Rehovot, Israel; acrylamide (gelelectrophoresis grade) ethanolamine, (NH4)2SO4, EDTA (disodium salt), ammonium persulphate,acrylamide (specially purified for electrophoresis), NN'-methylenebisacrylamide and 2-aminoethanol were from BDH Chemicals (Poole, Dorset, U.K.); dithiothreitol (Clelands reagent), guanidinium chloride (crystalline), urea and Tris were from Bethesda Research Laboratories, Bethesda, MD, U.S.A.: NNN'N'tetramethylethylenediamine was from Koch-Light Laboratories (Colnbrook, Bucks.,U.K.); Trizma base was from Sigma Chemical Co. (Poole, Dorset, U.K.); [14C]methylated protein mixture, [125I]iodide and [3H]aminoethanol were from The Radiochemical Centre (Amersham, Bucks. U.K.), and Aquacide III was from Calbiochem (San Diego, CA, U.S.A.).

Other materials used were of analytical grade purchased elsewhere.

Aminoethanol was distilled freshly before preparing the buffers.

DEAE-cellulose chromatography DEAE-cellulose was treated as described by Peterson & Chiazze (1962) and adjusted to pH 7.2 with dilute HCl.

The absorbent was then equilibrated with 5 mM-aminoethanol/ HCl buffer, pH 8.0. The sample, dialysed against the starting elution buffer, was applied to the column (3cm x 80cm . Elution with a stepwi,] -onntratiOn gradient was then performed with equal volumes of starting buffer and the next chosen concentration (50-l00ml of each concentration; the concentration steps were 5-50, 50-100, 100-180, 180250, 250-500 and 500-800mM). Recycling was performed in a small column of DEAE-tellulose, usually 10% of the initial column size. [A linear concentration gradient of buffer was used to elute the purified protein in a conveniently small volume from the second column].

Preparations of aminoethanol-glass affinity-chromatography materials [3H]Aminoethanol was coupled to activated glass beads to monitor the amounts of covalently bound [3H]aminoethanol.

Activated glass beads were prepared by the procedure of Weetall (l'969).

Methods Used, Measurements Molecular-weight determinations (a) Sedimentation analysis. Boundary-sedimentation, including sedimentation-equilibrium, studies were performed in a Beckman model E ultra-centrifuge equipped with u.v.-absorption optics, photoelectrical scanning system and schlieren optical system as described by Chervenka (1969) and Yphantis (1960).

Protein solutions for ultracentrifugal analysis were dialysed at 40C againstO.lM sodium phosphate buffer, pH7.2, or against the same buffer containg 4M guanidinium chloride for 24h. All samples were sedimented at 56 OOOrev./min. far 18h. Equilibrium sedimentation was performed at 18 OOOrev./min or, at concentrations below 0.15mg of protein/ml, at 24 000 rev./min for leh. All runs were performed at 200C.

(b) Sephadex G-200 and Sepharose-4B gel filtration.

14 Two sets of marker proteins were used: (1) [ C] methylated protein mixture; (2) calibration mixture composed of Blue Dextran (Mr2000000), fructose bisphosphate aldolase (Mr, 125 000), alkaline phosphataes (Escherichia coli) (M 80 000), bovine serum albumin (Mr 67 000), egg albumin (Mr 46000),pepsinogen (Mr40800), pepsin (Mr35000), soya-bean trypsin inhibitor (Mr21 500) and cytochrome c(Mrll 500). The eluates were monitored at 650nm for Blue Dextran, at 435nm for cytochrome c and at 278nm for the other proteins.

Enzymic activities of fructose bisphosphate aldolase, alkaline phosphatase and activated pepsinogen were also measured.

Isoelectric focusing Isoelectric focusing of the purified Factor (s) was performed in an LKB 8101 isoelectric-focusing column (28 ml) with 2% Ampholines of pH3.5-lO.0 at 800 V for 16h. Fractions of volume l.Oml were collected.

Protein content and pH were measured in each fraction.

Polyacrylamide-gel electrophoresis Electrophoresis in polyacrylamide gels in the presence of sodium dodecyl sulphate and absence of reducing agent was performed in column ad in slab gels (7.5%, pH 9.0) as described by Laemmli (1970). The gels were stained with Coomassie Blue R Protein determination Protein content ,was measured routinely at 278 nm, the absorption coefficients A1cm being taken to be 10 t 0.5. Calibration values were determined for the most purified material by the method of Lowry et al. (1951), with bovine serum albumin as standard.

Concentrations and separation of the proteins from biological fluids.

I-l Potent, biologically active protein(s), Factor (s) which are present among large number of other materials in large volume are separated and concentrated by various percentage of (NH4)2S04, 40 to 60% loss of the proteins and biological activities due to this procedure (including dialysis and concentrations) is reported, with physiological fluids namely Lymphotoxins, Necrosis Factors, Macrophage Activating Factor (MAF).

I-2 The inventor concentrated on the initial volume of the cell culture medium of Karpas 160 line, and the cell culture medium of MAF.

1-3 A fifty-fold concentration of the above media was achieved with the recovery of 80% of both biological activities and protein con tent by mo} .-yi-z the -amino group of the lysinyl residue of the proteins with multiple ionizable derivatives of chromogenic aromatic compound.

1-4 The above reaction induced changes in; a. polarizability of the proteins; b. protonation; I- 5 The changes in the above physical properties of the proteins were exploited for phase separation in a thermal field.

Ref: Kinetic and Thermodynamic Studies of Electrolytes in Solutions. B.G. Cox An. Rev. Vol: 81, 1984 p.43 120 Physical Chemistry Section.

A-l Proteins of physiological fluids investigated in details.

A-l-l Lymphotoxin Lymphotoxin is a lymphokine which specifically inhibits tumor cell growth in vivo and in vitro and is usually less cytostatic against nontumoregenic cells from the same species. Lymphotoxin has been shown to inhibit the transformation of cells (induced by carcinogens or ultraviolet radiation) and displays cytostatic and cytocidal activities against several transformed cell lines from various sources.

The sources of lymphotoxin reported are: human tonsils, human peripheral blood lymphocytes and lymphocyte cell lines.

Most of the biological studies have been carried out with relatively crude lymphotoxin preparations until recently.

A-l Two lymphotoxins, Factor I and Factor II, produced by a human lymphoblastoid cell line (Karpas (1977) Br. J.

Cancer, 35 152-160). The above Factors were purified and characterized by Neumann & Karpas (1981) Biochem. J. 194, 847 0 856.

A-A-2 Detailed description of the modifying compound used.

Four compounds were experimented with. They are denoted Da,Db,DC,Dd, A-2-1 Da: phenolsulphonephthalein, sulphenthal; C19H1505S; molar mass 354.4 structural formula:

The structural changes taking place with a change of pH causing colour changes are as illustrated below.

II yellow. III alkaline form red A-2 In Da, the splitting off of the first proton changes the symmetrical structure into an asymmetrical one, while splitting off of the second proton again forms a symmetrical structure.

The gradual dissociation is mainly caused by the charge, which remains on the molecule after the dissociation of the first protein and which hinders the splitting off of the second protein (one-sided quinonoid ring system). The bis phenolate ion formed by further dissociation, corresponds to two alternative quinonoid ring systems, which is in accordance with the dark red colour. The various structures of Da, depending on proton concentrations, accompanied by matching absorbancies, make this compound suitable for the present invention.

The properties of Da are described in Indicators, edited by Edmund Bishop, Pergamon Press 1972 international series of monographs in analytical chemistry, general editors R. Belcher and R. Frieser. Da as a crystalline powder dissolves in alcohol or dilute alkali hydroxide solution. The pH transition interval of Da lies between pH 5.0 to 6.4 (yellow) and pH 7.4 to 8.2 (red). The pKi values in water depend on the ionic strength; and temperature.

Ionic strength 0 0.05 0.10 0.50 pKi (20) 8.0-8.2 7.84 7.81 7.6 (KCL) REF: Kolthoff, I.M., Guss, L.S: J.Am.Chem.Soc. (1938, 20,2516 Guss, L.S., Kolthoff, I.M.: J.Am.Chem.Soc., (1940) 62, 249.

D Luciferin structural formula:

Db-Luciferin D(LH2). Luciferin reacts with proteins in the presence of Mg + RCHO. The modifications of proteins by this compound were first investigated and D(LH2) was found to give fluorescent (spectra) products, PiDa, which are easy and sensitive to monitor in the procedures described in Stage A and Stage B.

Dc. Gossypol. Gossypol is a polyphenolic aromatic aldehyde: 1;l,6,6',7,7' hexahydroxyl 3,3'-dimethyl-5,5' diisopropyl-2,2'-bis-naphthyl-8,8'-dialdehyde. The molecule has two PKa and several resonance structure. It is therefore capable of reacting with proteins and adsorbant materials by ionic and other type of reactions.

Gossypol reacts with -NH2 of the lysinyl residue of proteins selectively, in the ratio 1:2, and differently from other aldehydes including benzaldehyde.

Tanksley, T.D., Neumann, H., Lymann, C.M., Pace, N.C.

and Prescot: J. Biol.Chem., (1970) 245, 6456-6461; Wong, Q.C., Nakagawa, Y. and Perlmann, G.E. J.Biol.Chem., 1972) 247, 1625-2631.

Dd Folate and folate analogues Folate, (+) - L - tetrahydrofolate, (+)-L-N5 10 Methylene tetrahydrofolate (commercially available) and 5 10 (+)-L-N5,N Methylene tetrahydrofolate (which was synthesised) were experimented with. Some of these compounds were coupled for affinity. Affinity chromatographic separation of F1 previously was performed on Sepharosealbumin-folate and on Sepharose-albumin-folate affinity columns. This was done to isolate folate binding proteins from chronic myelogeneous leukemic cells (CML) and attempts were made to purify Factor a by this affinity column material. Better results were obtained when the cell culture medium containing F a were modified and processed by the M.C. method.

o Heparin is a sulphated glycosaminoglycans polysaccharide.

heparin is a macromolecule random polymer of repeating a unit variously sulphated. The basically simple carbohydrate backbone consists of hexuronic (D-glucuronic or L-iduronic) acid and D-glucosamine units, joined in alternating sequence by 1,4-glycosidic linkages.

The variable location of the sulphate substituents on these three units will give rise to at least ten different monosaccharide building blocks. The two outstanding structural features of the heparin molecule are its high negative charge density, mainly due to the high sulphate content, and its heterogeneous composition.

Structural unit of Heparin.

Heparin interacts with other macromolecules, ranging from cooperative, electrostatic, and the fine structure of the carbohydrate backbone.

All the compounds used are well characterized, both chemically and physically; some of the compounds suggested to use are compounds with which interactions of proteins are also known (dissociation -association reaction and dependency).

B : Solid phase adsorbent material used are: o Adsorbant (1) 1. DEAE-cellulose (Whatman) prepared for chromatographic separation according to Peterson and Chiaze (ref).

Adsorbant (2) and (3) (2)(3)Sephadex G-25, Sephadex G-100; Sepharose-4B.

Adsorbant (4) 4. Glass beads coated by amino ethanol: This type of adsorbant material has been prepared and used in the purification procedure reported by Neumann & BR< Karpas (1981) for the killer Factor, lymphotoxin (A).

The description of a preparation of affinity chromatogram using glass beads" instead of cellulose Step 1. 2.5g. of glass beads (120-200 mesh size 2046 mean diameter) were treated with 70% nitric acid in a boiling water bath for 1 hour.

Step 2. The -beads were rinsed with H20 until all the acid was removed and the pH of the water was neutral.

Step 3. The pH was adjusted to pH 4.0, and the beads kept in 10% 3-aminopropyltrietoxy silane for 3 hours at 800C, the unabsorbed (or unreacted) 3-aminopropyltrietoxy silane was separated (by draining) and the beads were held at 1100C overnight.

Step 4. The dry beads were washed with acetone and then with H20- Step 5. Excess of succinic anhydride was added and held at pH 6.0 using a solid NaOH trap for one hour.

Step 6. The beads were then washed with H20 and acetone.

Step 7. E g. of the beads was placed in a column and 20ml of pH 4.5 0.1M l-ethyl-3(3-dimethyl-aminopropyl) carbodiimide was recirculated for one hour.

Step 8. The beads were then reacted with the required substance for affinity chromatography (for APase and N APase purification with a:aminoethanol,b a :aminoethanol,bcysteamine-S- phosphate and cphenylalanine) at 40C and held in the presence of the compound for 2 hours. Radioactively.

labelled compounds were used to determine the quantity of aminoethanol etc. bound to the beads.

Step 9. The beads were drained from the excess of the "reactant" and washed on a funnel connected to vacuum and dried.

The reactions involved in the preparation of the glass beads-aminoethanol chromatogram:

A 4-B sepharose L-phenylalanine affinity chromat gam was also prepared and canpared to that in which activated glass beads were used instead of 4-B sepharose. The separation of APase on glass beads - L-phenylalanine column was much better inasmuch as the peaks of the compounds were eluted in a smaller volume providing the same concentration both in respect of activity and protein content. Consequently with glass beads affinity chromatography a better separation and purification of the substance of interest would be achieved.

To the Method: Fractionation of P D by sucrose-gradientultracentrifugation.

PiDa, the modified proteins, were fractionated according to their molecular weight, by sucrose-gradient-ultracentrifugation.

1. Gradients of 6-22% w/w sucrose in 0.08M AE/HC1 pH 8.0 buffer were made (6.0ml). On the top of each gradient tube 0.1 mg protein in 15P1 solution of PiDa was layered.

2. The tubes were centrifuged at lOO,OOOg for 36 hours at 150C.

3. Ten fractions, 10 visible bands, were observed in the tubes. The fractions were separated, using Gradient-Mixer, and fraction collector, to tubes (rate of the removal of the layers: 15 1/5mien).

4 The amount of the fractionated PiDa, 15 pl is not enough for the analysis of the required parameters.

Therefore, PiD fractions obtained from the same sucrose-gradient ultracentrifugation in 4 tubes were separated and collected in the same way as described in E-II-3.

5. The spectra The absorbancy was recorded at the wavelength 250nm - 600nm.

6. The protein content of each tube was measured by the method of Lowly.

7. Polyacrylamide gel electrophoresis of the aliquots of the 10 fractions was carried out in the presence of SDS usina Marker proteins with known molecular weight.

Protein Purification by Multiphase Chromatography Brief description of the method: Multiphase Chromatography is a preparative chromatographic method for fractionation of NC protein components of physiological solution in effective amounts on one column in one operation. This is done on the principle of chemical equilibria between the reactants and the products where the reactants are the protein components, PNC, and the modifying compound, L, and the product PNCL.

On this method the protein components PNC of the physiological solution are reversibly modified through the -amino group of lysinyl residue of the macromolecule in stoichiometric amount with a compound, L, of distinct chemical properties; subsequently, PNCL are subjected to column chromatographic separation on commercially available adsorbent material;PNCL deposit on the top of the adsorbent matrix forming a new stationary phase, M2, with NC distinct plates; in M2 the ligand, L, in the complex PNCL is treated as immobilized lisand on the adsorbent surface, whereas PNC as the solute molecules adsorbed on M2; from this phase the individual proteins are dissociated according to their dissociation constant KD in the complex PNCL; and thereby removed to the mobile Phase; the proteins are readsorbed on the original matrix M1; and removed from the mobile phase according to the partition coefficient between M1.Solute.Solvent; and eluted in the front.

By the above method most of the proteins initially subjected to column separation are recovered with their little change in their native structure.

Multiphase Chromatography is a versatile and a specific adsorbtion method for the purification of biomolecules of physiological solutions on an analytical or a preparative scale.

The interactions of PNCL with the commercially available adsorbent matrix were studied by frontal analysis at various pH and ionic strength using the same initial concentration of PNCL. The chromatographic data so obtained served as a basis for evaluating some relevant affinity chromatography parameters by adapting previously reported equations described for multi-component system in solution (see equations (1-8) and for immobilized metal ion affinity chromatogry (IMAC), see equations [9-11]).

Frontal analysis of the proteins eluted from MC base columns also offers ideal conditions for studying adsorption equilibria. In adition, it is a reliable method for the determination of binding constant of selective proteins with biospecifical ligand, using equation (4)[10-12).

The method of Multiphase Chomatography permits to purify: (a) lymphotoxins, Human Necrosis Factor(s), Cytotoxic Killer Factors, Macrophage Activating Factors secreted by cells (or established cell lines) to the culture medium, where they are present in minute amounts among a great number of other proteins in large volume; (b) to separate, purify and quantitatively determine the amount of isoenzymes, or variant forms in blood sample of patients with minimum amounts of blood (10-20ml).

I have recently purified by the Multiphase Chromatography method antithrombine from human plasma using stoichiometric amount of heparin as a modifying agent. Heparin, a polyanionic compound interacts due to its charge with proteins and biospecifically with selective number of proteins ( ). Using the above method I have achieved recently a thousand-fóld purification of antithrombine from human plasma.

Other proteins can be similarly purified by the Multiphase Chromatography method adopting the appropriate chromatographic parameters.

The method of Multiphase Chromatography uses the following approach: Part A. Reversible modification of the protein components of physiological solution.

In physiological solution NC protein components PNC=T, are modified in stoichiometric amounts with a compound of known structure, ligand = L. The various possible equilibria of these reations is represented in Scheme 1: Scheme 1.

For monophasic solution of NC components, NS species, the mass-balance equations are described in (1-8]. The mass-balance quantities, PNC and L, are treated as known parameters, while the pH and ionic strength are the measured dependent variables, reflecting the unique set of equilibrium constants that most accurately characterize the system.

The complexes between the proteins and the ligand at varying pH and ionic strength can be treated as coming from the components: protein = M, ligand = L proton = H, hydroxyl ion = OH, whose concentrations are constrained by the mass-balance equations [1-3] and by the equilibrium constant [4]. Different symbols are employed for each interacting component (1,2):

-he explicit equations [1-33 have been replaced below by a perfectly general mass-balance equation''[7] which states that the analytical concentration of a given component, see equation [5], is always balanced by its distribution in all the equilibrating species, defined in equation [6). The sum totals of each component are taken over all the species present, while the product of each species is taken over the range of all components.

[5] T component = #&gamma; [species] [6 ] [species] = ss # [component]&gamma; For a monophasic solution of NC components and NS species the generalized mass-balance equation of each component of concentration Tcj is:

K=l and t nj is the stoichiometric coefficient of component n in species i, ES] is the concentration of species Si, and > gi is the overall formation constant for species i from its componet C.

In solution the stoichiometry of the reactants and the equilibrium constants of the reactions can be determined from the data obtained from: poteptiometric titrations; sucrose-gradientultracentrifugation; spectrophotometric titrations; radioactive labelling of the reactants; or indirectly from frontal analyses of the critical protein(s) obtained from a certain physiological solution after chromatographic separation(s).

Part B Purification of proteins in solution by chromatography is based on the adsorption of the solute and development (3,4).

B.1. For the treatment and development of chromatographic separation, the following conditions should be fulfilled: (a) instantaneous equilibrium is reached between the solution and the adsorbent material; (b) the equlibrium between the two is always maintained; (c) the band containing the solute molecules, which is already formed, develops by passage through the column of a volume V of a solvent is not necessarily the same solvent used for the forming band; (d) the volume of interstices between the particles of the adsorbent per unit length of the column is negligible.

The adsorption isotherm of the i-th solute in a solution containing n solutes is a function of concentration of all the solutes. The isotherm equation for the i-th solute is [9) qi/m = fi (cl , c2, ... , cn), where g is the number of m moles adsorbed on m grams in equilibrium with a solution whose concentratioW is c moles per liter.

B. 2. Adsorption of the solution of the modified proteins PNCL, on commercially available matrix.

The modified protein in solution, PNCL, applied to column separation is adsorbed on the matrix M1, according to the isotherm equation [9). The observations and results described in paragraph C permit to state that on the top of the adsorbent column (5-108 of length of the original column) a new stationary phase is formed. In this new stationary phase, denoted M2, the ligand, L, in the complex PNCL, is treated as immobilized lisand on the adsorbent surfaces, whereas the protein components, iNCf r as the solute molecules adsorbed on the new stationary phase, M2. The interaction of PNC with the ligand, L, on the new stationary phase, M2, was studied by frontal analysis.The chromatographic data so obtained served as a basis for evaluating relevant chromatographic parameters of MC by adapting the massbalance and equilibrium equations [1-4] and equations for interactions of [10] multicomponent protein solution with immobilized ligand on adsorbent surface

where A is the n-th component in PNCL modified with L in solution; L=Bis the concentration of the immobilized ligand on the adsorbent matrix, M1 Kd is the dissociation constant of A in the complex PNCL from the new stationary phase ; Bo is the amount of ligand bound to the adsorbent matrix in M2 ;B is the total amount of the new stationary phase ;V is the variable elution volume of A; and Vo is the elution volume of A, without modification in solution, eluted under the same chromatographic parameters Kd = KNC is given in equation 4.

Alternatively, Kax can be calculated from experimental data according to equations [11] and [12]. 11 [A] 1 (V - %) (V0 7rn)WXThKax + (\i% - '/rn)FXJ (V- - K0 ) [12] kax = where A is the critical protein; X is the concentration of the immobilized ligand, L, on the adsorbtion site; V is the variable elution volume of A ; V0 is the elution of A, where its interaction with X is abolished. In MC method the most valid determination of Vo is the chromatography of the unmodified protein mixture of the physiological solution using the same chromatographic parameters; Vm is the elution volume of a neutral and completely excluded large molecule; Vo - Vm is the volume of the new stationary phase, M2 available to the solute; Kax is the association constant of A with the immobilized ligand X.The volume of the adsorbent column available for the proteins removed from M2 is Y - (Vo - Vm), where Y is the original volume of the adsorbent column.

According to equation 110 supported by experimental findings, in conditions where the relationship K > > Kd is maintained, the proteins are dissociated from M2 before the ligans, L=B, and thereby they are removed to the mobile phase. The removal of the individual proteins from M2 are constrained to the law of mass action, equations tl- 3), and equations of equilibria t4] and 110].

Dissociation of the proteins from phase M2 is achieved by changing the pH, ionic strength of the solvent front or by raising the temperature.

B. 3. The protein components, Ps, removed from M2 to the solvent front, readsorb on M1 according to the isotherm equation t9] M1 consists of approximately 90% of the original volume of the adsorbent volume. Since the number of protein components removed from M2 can be coDtrolled, by choosing appropriately the pH, ionic strength and gradient volume, the interference of the solute protein molecule can be avoided. This phase can be treated as a new application of reduced number of solutes on M1. This is an additional fractionation of the proteins on commercially available adsorbent (SEC, IES, HIC).

Claims (2)

1. A method of protein purification by multiphase chromatography wherein: (i) species of protein, Pi, in solution are reversibly modified through the -amino group of a lysinyl residue of the macromolecule with a modifying compound D which reacts selectively and reversibly with the -amino group of the lysinyl residue of the polypeptide chain; (ii) subsequently, the modified proteins, PiD, are subjected to column chromatographic separation on commercially available adsorbent material, PiD depositing on the top of the adsorbent matrix (2 to 5% of the adsorbent material) to form a new stationary phase, M.D.P.i; (iii) from this phase, the individual proteins are dissociated and removed to the mobile phase, according to the principle of chemical equilibria in multicomponent systems; and (iv) the individual proteins are readsorbed on the original matrix from which they are removed to the mobile phase according to the partition coefficient between the Matrix.Solute.Solvent, and eluted in the front.

2. A method according to Claim 1, wherein the said modifying compound is selected from: Da: phenosulphonephthalein, sulphenthal Db: Luciferin Dc:Gossypol, polyphenolic aromatic aldehyde: 1,1',6,6',7,7'-hexahydroxyl 3,3' dimethyl-5, 5 '-diisopropyl-2, 2 '-bis- naphthyl-8, 8' - dialdehyde Dd: Folate and Folate analogues De: Heparin, a sulphated glycosaminoglycans polysaccharine.