CA1205399A - Antibodies against interferon proteins and method of producing same - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE

Two novel processes are provided for producing antibodies.
One process comprises the steps of first immunizing an immunizable animal against immunological determinants of human Le form interferon protein or proteins, and then obtaining antiserum from that animal.
The other process comprises the steps of first culturing a hybridoma cell clone producing antibodies directed against immunological determinants of human Le form interform protein or proteins, and then recovering the antibodies from the culturing medium. In each process, the human Le form interferon protein or proteins must be those which have been specifically described in the specification. These antibodies are strictly specific to the active species.

Description

~2(:!~i3~9 The present invention relates to processes for the preparation of antibodies, and to the antibodies so formed~
The application is a division of application Serial No. 350,293 filed April 21, 1980.
That parent application was directed to human interferon, to a purified form of human interferon, and to purification and preparation procedures relative thereto. As used therein, the term "protein" included "glyco-protein".
As taught in the above-defined parent application attempts had been made to purify human interferon. The objectives of such purification attempts included a complete characterization of the interferon species for standardization purposes~ To date, none of the attempts to purify human Le form interferon have been completely successful.
The invention in the above-defined parent application was based on the discovery of purification methods which permitted the preparation, for the first-time, of all the components of human Le form interferon protein or proteins substantially free of inactive and otherwise undesirable impurities. `
B. Berman, C.A. Ogburn, K. Berg, K. Paucker, and J. Vilcek, Proc. Nat.
Acad. Sci. USA, 72, 2185 - 2187 (1975).
According to broad aspects of the invention provided by the above-identified parent application, pure human leukocyte interferon proteins were prepared from crude human leukocyte interferon through a number of special purification steps, and the pure human leukocyte inter-feron was characterized by stained protein bands in SDS PAGE (sodium dodecylsulfate polyacrylamide gradient electrophoresis)~
Some of the products and procedures involved in the preparation s3~g and characterization of the pure interferon proteins were novel per se and constituted aspects of the invention provided by the above-identified parent application of generic applicability within interferon technology and, in a broader sense, in protein purification technology. The pure human interferon proteins, and especially, pure human Le form interferon proteins, now made available and characterized according to some aspects the invention provided by the above identified parent application con-stituted the key to further new developments which are still additionalt~
the invention provided by the above identified parent application.

The complete purification of interferon proteins made it possible, for the first time, to produce anti-interferon which was strictly specific to the active species simply by immunizing animals with the pure interferon preparation or one or more of its components, Such strictly monospecific anti-interferon was extremely useful for antibody affinity chromatography for purification of crude or partially purified interferon to obtain, in a simple and economic way, large amounts of pure interferon or highly purified interferon for clinical purposes, for standardization, for chemical studies, for sequence studies, and as immunogen for repeated pre~
paration of monospecific anti-interferon. It is within the scope of other aspects of the invention provided by the above identified parent applica-tion not only to purify human leukocyte interferon by means of the mono-specific antibody raised against the pure human leukocyte interferon, but also to purify other interferon types which cross-react immunologically with the monospecific anti-interferon, e.g. "Namalva'l interferon (human lymphoblastoid interferons; the Le form interferon constitutes ~5~ of the biological activity of human lymphoblastoid or Namalva interferon, vide E.A. Havell, Y,~. Yip, and J. Vilcek, I'Characterization of human lympho-S39~3 blastoid (Namal~a) interferon", J. Gen. Virol., 38, 51 - 59, (1977)), and interferon containing the Le form obtained by cultivation of a microorgan-ism carrying DNA coding for the production of interferon proteins (or proteins having the significant biological interferon activity determin-ants).
The monospecific anti interferon of aspects of the invention provided by the above identiEied parent application was also useful for establishing, in a manner known per se, a genetic engineering system for the production of interferon protein: In accordance with known methods within genetic engineering, the first stage is the isolation of messenger RNA from interferon producing cells in which the interferon synthesis had been triggered by means of an interferon inducer and had reached a degree of completion of the synthesis of interferon proteins at which the immunological determinants (or parts thereo'f) of the interferon had been expressed, while at the same time, the interferon was still attached to the ribosomes and tha messenger RNA. A high clone producing Namalva cell suspension grown in the usual way or buffy coats (or lumphocytes isolated by Ficoll techni~ue) was preferred'as the interferon~producing cells. The messenger RNA-was isolated from such cells by lysing the cells in a manner known _r se a~d passing the lysate through an antibody affinity column where the antibody bound covalently was the monospecific anti-interferon. The antibody column selectively retained not only the interferon, but also the attached messenger RNA. By known methods, e.g.
salt elution, the messenger RNA was isolated from the eluate from the column and was also, by known methods, treated with reverse transcriptase to obtain the corresponding DNA. Alternatively, immunoprecipitation methods (known per se), possibly combined with double immunoprecipitation ~Z~S399 techniques, could be used. In accordance with known methods with genetic engineering techniques, such DNA coding for interferon or important parts thereof was incorporated in a suitable cloning vector, preferably a mini-plasmid, and transformed into a microorganism, the culturing of which produced interferon and/or interferon derivatives released in the culturing medium, from which the interferon was obtained. The purification of such interferon obtained by cultivation of the microorganism could suitably be performed in the same manner as described above by passing the crude preparation through an antibody affinity column made by means of mono-specific anti-interferon. Radio-labelled monospecific anti-interferon could be a valuable tool in the assessment of which clones of the micro-organism had received the DNA and were capable of producing interferon or parts o~ derivatives thereof.
Expressed with reference to specific activity, the invention provided by the above identified parent application, related to human interferon or species thereof having a specific activity of 2 x 10 ~ 2 x 10 IFU per mg protein. However, since the methodology concerning the protein determination varie~ considerably, the actual figure of the specific activitywas of less importance compared to the clear de nstration, by SDS PA~E, of the individual species.
By one broad aspect of the present divisional invention, however, a process is provided for producing antibodies, which comprises one of the following: (I) (a) immunizing an immunizable animal against immunological determinants of human Le form interferon protein or proteins, and (b) ob-taining antiserum from the animal; or (II) (a) culturing a hybridoma cell clone producing antibodies directed against immunological determinants of human De form interferon protein or proteins, and (b) recovering the . .

53~99 antibodies frvm the culturing medium. In each of these processes, the hurnan Le form interferon protein or proteins must be selected from the following:
(a) human Le form interferon protein or protein which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9 x 106 IFU, show two major sharp stained protein bands having antiviral interferon a.ctivity at 18,400 and 20,100 Daltons, respectively, and a mi.nor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, toge-ther with small peaks of antiviral interferon activity at 19,500, 21, 130 and 23,440 Daltons (the Dalton molecular weights being subject to an experirnental accuracy of +200 Daltons), the sodium dodecylsulfate polyacrylamide gel electro-phorese acrylamide gradient showing essentially no other stained protein regions; (b) human Le form interferon pro-tein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8 x 1~ IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively (the Dalton molecular weights being subject to an experimen-tal accuracy of +200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, the sodium dodecyls~lfate polyacrylami.de gel electrophorese acrylamide gradient showing essentially no other stained protein regions;
(c) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interEeron load of 0.9 x 106 IFU, show two major sharp stained protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a rninor stained protein band havi.ng an~iviral interferon activity between 20,300 and 20,400 Daltvns, together ~ $

with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (the Dalton molecular weights being subject to an experi-mental accuracy of +200 Daltons), the sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other protein regions, but having a specific activity of at least 10 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; (d) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9 x 10 IFU, show two major stain protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (the Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), the sodium dodecylsulfate polyacrylamide gel electro-phorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2 x 10 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamlde gel electrophorese staining; (e) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese stain-ing at a total interferon load of 3.8 x 10 IFU, show six stained protein bands having antiviral interEeron activity, viz. strong bands at 18,410 Daltons and 20,1~0 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Daltons, resL~ectively (the Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), the peaks of antiviral ` - 6 -~,~2~5~399 interferon activity coinciding exactly with the stained protein bands, the sodium dodeylsulfate polyacrylamide gel electrophorese acrylamide gradient shGw essentially no other stained protein regions but having a specific activity of at least 10 IFU per mg protein as assessed by cornparative sodium dodecylsulfate polyacrylamide gel elec-trophorese staining; and (f) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8 x 106 IE'U, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltsons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130~Daltons, and 23,440 Daltsons, respectively (the Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, the sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2 x 1~ IFU per mg protein, as assessed by ccmparative sodium dodecylsulfate polyacrylamide gel electrophorese staining.
It is preferred, by one variation thereof, that the lluman Le form interferon pro-tein or proteins ccmprise the 18,400 +200 Daltons human Le form of interferon protein component of the protein or proteins, the 20,100 +200 Daltons human Le form of interferon protein component of the protein or proteins, or a combination of the 18,400 +200 Daltons human Le form of interferon protein component of the protein or proteins and the 20,100 +200 Daltons human Le form interferon protein component of the protein or,proteins.

:`~ ~ - 7 -S3~9 The process as described above may alsc include the additional step of immobilizing the antibodies in a matrix, e.g. by covalently binding the antibodies to the matrix or by using as the matrix, a cross-linked agarose.
The process as described may also include the further step of substantially freeing the antibodies fr~ proteolytic enzyme activity, e.g. by treatment with enzyme inhibitors or with enzyme destructors, or by treatment with a matrix-immobilized enzyme inhibitor or with a matrix-imrr,obilized enzyme destructor.
The process as described above may also include the step of passing the matrix-immobilized through at least one of~ (i) a column of matrix-imrnobilized poly-L-lysin; (ii) a coiumn of matrix-immobilized soyabean trypsin inhibitor; and (iii) a column of matrix-immobilized kallikrein inactivator.
In a specific variant of this invention, the immunizable animal is a pig, and the antibodies are pig IgG immunoglobulins.
By another aspect of this invention, antibodies are provided herein which are raised against, or directed substantially only against, immunological determinants of human Le form interferon protein or proteins, in which such human Le form interferon protein or proteins are selected from one of the following: (a) Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese ~ .O~;i3~g staining at a total interferon load of 0.9 x 10 IFU, show two major sharp stained protei.n bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (the Dalton molecular weights being subject to an experi-mental accuracy of +200 Daltons), the sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions; (b) human Le form interferon protein or proteins which under sodium dodecylsulfate polyacrylamide gel electrophorese stain-ing at a total interferon load of 3.8 x 10 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively (the Dalton molecular weights being subject to an experimental accyracy of +200 Daltons, the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, the sodium dodecylsulfate polyacrylamide gel electrophorese acryla-mide gradient showing essentially no other stained protein regions; (c) human Le form interferon protein or proteins which, under sodium dodecyl-sulfate polyacrylamide gel electrophorese staining at a total interferon laod of 0.9 x 10 IFU, show two major sharp stained protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (the Dalton rnolecular weights being subject to an experi.mental accuracy of +200 Daltons), tlle sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions , ......

~IL~S3~3~

but having a specific activity of at least 10 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electro-phorese staining; (d) human Le form interferon proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9 x 10 IFU, show two major sharp stained protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (the Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), the sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2 x 10 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; (e) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8 x 10 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons 20 and 20,180 Daltons, respectively, a medium-strong band at 20~420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Dal-tons, respectively (the Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), the peaks of antiviral interferon activity coinciding with the stained protein bands, the sodium dodecyl-sulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity ,' ~? - 10 ~2~5399 of at least 10 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; and (f) human Le form interferon proteins which, under sodium dodecylsulfate poly-acrylamide gel electrophorese staining at a total interferon load of 3.8 x 10 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 1~,410 Daltons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Daltons, respectively (the Dalton mulecular weights being subject to an experimental accuracy of +200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, the sodium dodecylsulfate poly-acrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2 x 10 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamide gel electr,ophorese staining.
In one preferred variant thereof, the human Le form interferon protein or proteins may comprise the 18,400 +200 Daltons hurnan Le form of interferon protein or proteins component of the protein; the 20,000 +200 Daltsons human Le forrn of interferon protein component of the protein or proteins; or a combination of the 18,400 ~200 Daltsons human Le form of interferon protein component of the protein or proteins and the 20,100 +200 Daltons human Le form interferon protein component of the protein or proteins. In another preferred variant thereof, the human Le form protein or proteins comprises human Le form interferon protein or proteins may be obtained from the respective band or bands cut from sodium dodecylsulfate or polyacrylamide gel electrophorese, especially wherè the bands have been cut from the sodium dodecylsulfate , ~ . - 1 1 -~LZ~539~

polyacrylamide gel electrophorese gel after staining or the gel and a short wash in distilled water.
The antibodies or fragrnents or derivatives thereof retaining the essential antiinterferon determinants may be immobilized on a matrix, e.g. by being covalently bound to the matrix, especially where the matrix is cross-linked agarose, e.g. is selected from the group consisting of CNBr-activated cross-linked agarose, CH-activated cross-linked agarose, and epoxy-activated cross-linked agarose.
The antibodies may be substantially free from proteolytic enzymatic activity, especially where such freeing is by treatrt~nt with enzyme inhibitors or with enzyme destructors or by treatmen-t with a matrix-immobilized enzyme inhibitor or with a matrix-immobilized enzyme destructor By another aspect of this invention covalently bound matrix-immobilized antibodies are provided herein of the type described above from human Le form interferon protein or proteins, such antibodies having further been passed through a columrl of at least one (i) matrix-immobiliæed poly-L-lysin; (ii) matrix-immobilized soyabean trypsin inhibitor; and (iii) matrix-immobilized kallikrein inactivator.
The proteins as defined herein may be additionally characterized in that they exhibit anticellular activity and potentiate the Natural Killer cell system; or they may be additionally characterized in that ~ZlD5399 they will neutralize antibodies raised against a protein as described in the various aspects and variant of the invention as described above.
They also may be described according to other aspects of this invention as human leukocyte interferon protein(s) or may be described as lympho-blastoid (Namalva) interferon protein(s).
It is important to note that the individual components in the above-mentioned bands of the SDS PAGE gel show biological interEeron activity, ability to neutralize anti-human leukocyte interferon, and anticellular activity, etc. The invention provided by the above identified parent application also provided each of the individual com-ponents represented by each of the above-mentioned individual SDS PAGE
bands, as well as to any protein having the significant biological inter-feron activity determinant(s) possessed by the individual components, and to any protein having the significant immunological determinant(s) possessed by the individual components.
With respect to origin, the human interferon proteins may be derived from human leukocyte interferon prepared using human lymphoblas-toid (Namalva) cells, or from proteins prepared by cultivation of a micro-organism containing DNA coding for the interferon or an important part thereof, e.g., as described above, but also human Le form interferons of other origin, but conforming with the above characteristics.
It is well known that human Le form interferon shows a number of important therapeutic aspects in man, including antiviral and antitumor activity, and the provision of the pure human Le form interferon makes it possible to exploit these useful properties further. One additional aspect of the invention provided by the above identified parent appli-cation provided a formulation comprising the pure human Le form interferon 0, .

Q~i399 protein or proteins adapted for administration to human beings or animals for prophylactic, therapeutic, or immunization effect. Such a formula-tion may, e.g., be adapted for parenteral, intranasal, or topical adminis-tration.
A most useful formulation of the pure interferon proteins of an aspect of the invention provided by the above identified parent appli-cation was an aqueous solution. Pure interferon proteins in aqueous solution should be stabilized, and the choice of stabilizer depended upon the use of the solution. When the solution was to be used for adminis-tration to human beings, e.g., parenteral administration, the stabilizer should be a physiologically acceptable stabilizer. A suitable stabilizer was a protein or combination of proteins which was non-toxic and non-immunoyenic in human beings, e.g. human serum proteins and fractions thereof, and human albumin. A typical preferred stabilizer was 1% human albumin. ~he normal concentration of pure interferon proteins in com-positions for parenteral administration to human beings were in the range corresponding to 1 - 20 million IFU per ml. and a normal daily dose was 3 to 10 million, e.g. 5 to 10, million IFU totally, preferably administered once or twice a day by intramuscular injection. When preparing solutions of pure interferon for administration to human beings, normal pharmaceu-tical precautions which were customarily taken in connection with the preparation of parenteral compositions, were observed, e.g., precautions to ensure sterility and freedom from pyrogenicity.
~len the stabilized formulation was an aqueous solution of pure human Le form interferon protein(s) to be used for immunization of animals for the preparation of monospecific anti-interferon, stabilization with , 53~5~

SDS (sodium dodecylsulfate) to Eorm an SDS complex of the human Le form interferon protein(s) was a preferred stabilization in view of the above-mentioned fact that SDS increases the antigenicity and/or stability of interferon. As explained in greater detail below, the pure interferon-SDS combination or complex could be formed simply by adding SDS to the a~ueous pure interferon proteins, preferably in a concentration of 0.1%
by weight, calculated on the solution, at pH 7.2. The SDS complex of the human Le form interferon protein or proteins constitutes, in itself, a valuable composition because of the stability thereof, and a most interesting form of such complex, w~ll suited for storage and transport (suitably at low temperature, e.g., at a temperature of, at the most 4~C
or preferably -20~C), is when isolated in solid form e.g. as described below. The use of other stabilizers of the detergent type for this pur-pose is within the scope of the teachings of the invention provided by the above identified parent application. A further preferred form of the pure human Le form interferon proteins is a form in which they are bound to CIB~CRON Blue F3GA (Trade Mark) or another ligand capable of binding the interferon proteins according to the mechanism exhibited by CIBACRON
Blue F3GA, e.g. as will be explained in greater detail below.
The pH of the pure interferon protein solution for immunization of animals to prepare the monospecific anti-interferon was preferably 7.2, and a suitable buffer was PBS (phosphate buffered saline).
The stabilized pure interferon protein preparation for immuni-æation of animals may additionally comprise an adjuvant, e.g., Freund's adjuvant.
It was also within the scope of the invention provided by the above identified parent application to increase and/or stabilize the ` ~,05399 antigenicity of the pure Le form interferon proteins or each member thereof by coupling to an immunogenic carrier (so as to present the pure inter-feron protein or proteins as a sort of "hapten") in accordance with well-known principles. Examples of immunogenic carriers included PPD (Purified Protein Derivative) and BCG (Bacille Calmette Guerin).
For immunization purposes, mouse, rabbit, goat and sheep are pre-ferred animals, but it is also within the scope of other aspects of the invention provided by the above identified parent application to use other animals, and as described below, pig IgG immunoglobulins show distinct advantages for certain purposes.
In principle, the immunization of animals against the pure inter-feron according to an aspect of the present invention is performed in accordance with known methods for preparation of anti-interferon, e.g., as described, for example in Acta Path. Microbiol. Scand. Section B, 83, 443-460 (1975), but the fact that the interferon proteins of aspects of the invention provided by the above identified parent application are pure gives rise to minor variations with respect to the concentration of the immunogen and the immunization time and intervals. Examples of immuni-zation schedules appear from the "Experimental Section".
The bleeding of the animal and the isolation of antiserum in order to carry out the process of aspects of the present invention are per-formed in accordance with well-known methods.
The antibodies of other aspects of the present invention prepared as described above, apart from the trivial fact that they show a natural background characteristic of the animal immunized, are substantially specific to the interferon proteins characterized by the above-mentioned SDS PAGE bands. An extremely small amount of impurities not seen as .

539~

stained bands in SDS PAGE together with the interferon protein cannot be ruled out. Such proteins which may represent small amounts, corres-ponding to 1 - 5% of the total protein content in the pure interferon protein preparation, might trigger antibodies against the corresponding impurities. One way of checking such a possibility is to construct an anti-interferon column of the relevant antiserum, obtained by immunizing a rabbit with the pure interferon (that is, interferon of the above-described eharacterization whieh in SDS PAGE gives the visible interferon protein bands at a load of 1 - 4 x 10 IFU in total). The column is construeted without any absorption at all. Crude human leukoeyte inter-feron is loaded to the column, and a normal antibody affinity chroma-tography is performed, vide below.
The eluate is analyzed in an SDS PAGE (vide below), and only the interferon bands should then he seen possibly together with 1 - 4 other proteins (impurities). This (the three proteins) was in fact seen on rabbit anti-serum with a titer of 500,000 IFU-NU/ml in a 2 ml eolumn, loading 2 - 3 x 10 IFU of erude human leukocyte interferons.
The "foreign" proteins might also appear by simple spontaneous eross reaction whieh by ehance takes plaee.
The above-mentioned method of checkiny whether particular anti-bodies are monospeeifie is believed to be novel per se. This method for cheeking whether a particular antibody preparation (e.g., an antiserum), is monospecifie to its partieular antigen, comprises constructing an anti-body affinity ehromatography eolumn by means of the antibody preparation to be ehecked, loading a solution eontaining the antigen plus impurities to the eolumn, and analyzing the eluate from the column to ascertain the presence of any protein different from the antigen~ Preferably, the latter .

~LZ(~3~3~

analysis is performed by SDS PAGE gradient in the same manner as discussed in connection with the present use of the method in determin-ing the monospecificity of the anti-interferon, and the occurance of bands corresponding to a maximum of four impuri-ty proteins in the eluate will generally be considered a satisfactory indication of monospecificity for most practical uses of the antibody preparation.
As the stained interfexon protein bands in SDS PAGE have preserved their antigenicity completely or to a considerable extent, it is also possible to use the stained interferon proteins directly cut out from SDS

PAGE as antigen preparations for immunizing immunizable animals, e.g.
rabbits. When the stained band cut out from SDS PAGE is used for the immunization (after preparation describea below), a possible cross-over reaction (or contamination from extremely small amounts of impurities) as discussed above is less likely (compared to the total eluate repre-senting 5 interferon species). Thus, antibodies versus the individual species of interferon (primarily the two major species at 18,400 and 20,100 Daltons) with optimum specificity may be produced according to the following protocol:
l. 4 - 5 x 10 IFU human leukocyte interferon (as CIF) is puri~
fied completely (by means of the "tandem" affinity chromatography described below) and is subjected to SDS PAGE.

2. The gel is only stained for 10 - 15 minutes at room tempera-ture and is partially destained for 10 minutes followed by a wash in distilled water three times, done in l - 2 minutes with 0C
distilled water. The exact location of the protein bands is noted (for example by means of a POLAROID [trade mark] photo), and the -two major interferon protein species are specifically ` 1~0~399 removed by cutting out with a sharp knife. Each slice is minced by means of a TEFLON [trade mark of polytetrafluoroethy]ene]
rod in 1 ml 0.01% SDS (in PBS, pH 7.2) and is thereafter injected subcutaneous]y into a rabbit. By following this procedure every second week, low titered antibodies against the human leukocyte interferon proteins are developed in 2 - 4 months. As soon as a low titer against interferon is detected, Freund's adjuvant is added to the irnmunogenic mixture every fourth time (every 4th to 6th week) depending on the development of the titers. This prodedure is continued for 3 - 12 months and anti-interferon against the interferon species is developed tlO,OOO - 1,000,000 IFU/ml).
Thus, the term "monospecific anti-interferon" is used both in relation to anti-interferon produced by means of the pure interferon proteins as described above without the step of cutting out from SDS PAGE, and in relation to the antibodies raised against the stained interferon band or bands cut out from the SDS PAGE.
A further process for producing monospecific antibodies against interferon proteins according to another aspect of this invention is the so-called hybridoma technique. The hybridoma technique is a well-known method for preparing antibody-producing lymphocytes/myeloma hybrids (com-pare, for example, "Current topics in Microbiology and Immunology, Vol. 81, Lymphocyte Hybridomas, Eds. F. Melchors, M. Potter, and N.L. Warner, Springer Verlag, 1978). However, it was not previously known or obvious that it would be possible to obtain an anti-interferon-producing hybridoma cell clone. In the hybridoma technique, using, for example, mouse as the animal immunized, mice are immunized with human Le form interferon and lzl~3~

spleen cells from the immunized mice are fused with myeloma cells, whereafter the fused hybridoma are cloned, antibody-producing clones are selected and cultured, and antibodies are obtained from the culturing medium.
Antibodies prepared by hybridoma technique in a mouse system are strictly monospecific and are therefore especially advantageous in radioimmunoassays or other similar tests.
The antibodies so produced constitute, as described above, a further aspect of this invention. Variants of such antibodies are those antibodies (or fragments or derivatives thereof retaining the estial anti-interferon determinants) immobilized on a matrix.
In the hybridoma technique described above, one particular way of obtaining the antibody is to culture the selected clone in vivo in the animal species from which the spleen cells were derived, and harvesting antibody from the ascites fluid of the animal, and such ernbodiment is within the scope of yet another aspect of the present invention.
The selection of positive hybridoma clones may be performed by the usual interferon neutralization test. ~owever as the usual interferon neutralization test, as a prerequisite requires that the antigenic deter-minant of the interferon be located very close to the center(s) of the biological activity (within a distance of 1 IgG molecule length), it is likely that antigenic determinants located further a~ay from the center (more than 1 IgG molecular length) of the biological activity/activities will not be detected by this test, and it is, hence, likely that "positive"
hydridoma clones (producing antibodies against antigenic determinants the interferon protein which are located at a distance from biological center which is greater than the length of 1 IgG molecule) will escape detection :. -~IZ~)S3~3~

in the test. Therefore, a more advant~geous technique for testing Eorpositive hybridoma clones is to use radio-labelled pure human Le interferon proteins of aspects of the invention provided by the above identified parent application in a radioimmunoassay. The radio-labelled pure human Le interferon proteins can be made by radio-labelling human Le interferon, e.g., a gel filter made by the gel filtration technique described below, by means of a standard radio-labelling technique, e.g.
using lactoperiodase and iodine 135, and then purifying the interferon protein in the manner described herein, subjecting the purified interferon proteins to SDS PAGE and eluting the radio-labelled pure interferon proteins from the SDS PAGE gel. Another method for selecting the positive hydridoma clones in a manner which will also detect such clones that are not detected in the usual interferon neutralization test comprises sub-jecting an amount, e.g., 500 ~1, of the supernatant from each clone culti-vation to immobilization on a matrix, e.g., immobilization in CNBr-activated SEP~IAROSE (trade mark) according to the method described in the section "materials and Methods". The human Le form interferon, e.g. crude human leukocyte interferon, is then applied to the resulting treated matrix, e.g., by mixing the resulting matrix gel suspension corresponding to each clone with the interferon and allowing the mixture to stand for a period, e.g., 1 hour at 37C. The unbound interferon is effectively separated from the matrix material, e.g., by centrifugation and washing with PBS, and thereafter each matrix gel portion is subjected to elution to release any bound interferon, e.g. by mixing with elution buffer (ph 2.4) and centrifugating, selecting. The clones corresponding to the matrix gel portions from which the eluting buffer portions are then selected. In particular, the last eluting buffer portions, contain ....~

31LZ(:~S3~5~

interferon, as the yielding of interferon in the elution is an indi-cation of a positive clone. The two above-mentioned advantageous methods for detecting positive hybridoma clones may be applied not only to anti-interferon-producing hybridoma clones, but with evident modi-fications, also to the detection of positive hybridoma clones producing antibody directed against other proteins.
Interestingly, it has been found that antibodies of aspects of this invention raised against one of the purified interferon proteins of aspects of the invention provided by the above identified parent appli-cation are capable of neutralizing the other purified proteins of other aspects of the invention provided by the above identified parent applica-tion. Thus~, as will become apparent, the monospecific antibodies of aspects of the invention or raised against a combination of purified in-terferon proteins of still other aspects of the invention, are equally effective for purification of human Le form interferon-containing solutions.
In accordance with well-known principles, the monospecific anti-interferon can be used for determination of the corresponding interferon or interferon component in biological fluids, e.g. by radioimmunoassay or related techniques. However, as alluded to above, an interesting and important utility of the monospecific antibodies is for antibody affinity chromatography purification of interferon-containing solutions.
For this purpose, the antibodies are immobilized on a matrix in a manner known per se, suitably covalently bound to a suitable antibody affinity chromatography matrix, e.g. a cross-linked agarose, for example, SEPHAR-OSE 4B ttrade mark of Pharmacia). The antibody affinity chromatography purification of interferon-containing solutions may be performed according : .

~Z~S3~

to any of the well-known methods, either batchwise or, preferably, using the matrix-immobilized antibody arranged in a column.
The affinity chromatography may comprise a combination of ligand affinity chromatography and subsequent antibody affinity chroma-tography may be CIBACRON F3GA or may be another ligand capable of binding the human Le form interferon proteins according to the mechanism exerted by CIBACRON F3CA Alternatively, the matrix-immobilized ligand may be selected from the group consisting of Blue DEXTRAN 2000, Blue SEPHAROSE
Cl-6B, and Blue DEXTRAN coupled to SEPHAROSE 4B. Also, the ligand may be immobilized on dextran of molecular weight of 2 million Daltons, coupled to cross-linked agarose, or the matrix-immobilized ligand material may be Blue DEXTRAN coupled covalently to SEPHAROSE 4B.
The solution applied may be an aqueous solution containing Le form interferon in a form having a specific activity of at least 50,000 -100,000 IFU per mg protein, such solution being buffered to a pH of 6.5 -8 and having an ionic strength of a 10 - 100 mM, in particular 20 mM, phosphate buffer, pH 7.2, solution, optionally together with water misci-ble organic solvent, e.g. ethanol in amounts of 5 - 80~.
The elution may be performed with an aqueous solution buffered to a pH of 6.5 - 8 and having an ionic strength corresponding to 0.5 - 0.7, in particular 0.5 - 0.65, molar NaCl solution. The elutant preferably has an ionic strength corresponding to 0.6 molar NaCl and is buffered to pH
7.2. The pH in the solution applied and in the eluate is preferably 7.2.
The interferon applied to the ligand affinity chromatography matrix may have a specific activity of 50,000 - 1,100,000, for example, 100,000 - 1,000,000, e.g. 200,000 - 1,000,000, in particular 500,000 -~ ,.

539~

1,000,000, IFU per mg protein. The human Le foLm interferon of the solution applied to the antibody affinity matrix is selected from the group consisting of hu~an leukocyte interferons, human lymphblastoid inter-ferons, and proteins as described above, including such proteins when produced by cultivation of a microorganism carrying DNA coding for the production of such proteins. The interferon-containing solution may desirably comprise the interferon-containing protein fraction obtained by a protein precipitation treatment of a crude, unconcentrated interferon preparation. Pre~erably, the protein precipitation treatment includes the addition of KSCN and the adjustment of the solution pH to ~.5.
The solution applied to the antibody affinity matrix may alternatively be an interferon solution which essentially only contains proteins in the 10,000 - 20,000 Daltons range. It is preferred that the solution be 10,000 - 20,000 Daltons eluate from a gel filtration performed with a buffer solution which contains 25~ by volume of a glycol, e.g.
ethylene glycol and has an ionic strength corresponding to 1 M NaCl, pH
7.2. Still more preferably, the eluate from the ligand affinity chromatography is loaded directly on the anbibody affinity matrix.
The preparation of antibody affinity columns using the mono-specific anti-interferon, and the operation of such columns are performed in a manner known per _. The interferon-containing solution applied on such columns may be a crude, unconcentrated interferon preparation, or it may be a concentrated or partially purified interferon preparation.
The interferon preparation applied on tlle column may be any interferon preparation containing human Le form interferon, that is, human leukocyte interferons, human lymphoblastoid interferons (Namalva interferons), or interferon (or important parts thereof) produced by cultivation of a ~;Z(J~3~3~

microorganism containing DNA coding for interferon, e.g. as described above. The use of antibodies against partially purified human leukocyte interferon in antibody affinity chromatography for purifying Namalva interferon and leukocyte interferon has already been described (vide, e.g. Scand. J. Immunol., 8, 429 - 436 (1978)). However, the important improvement is that monospecific anti-interferon will retain substantially only human Le form interferon protein, the remaining proteins of the pre-paration passing through the column. Very small amounts of impurities due to spontaneous cross-reactivity cannot be ruled out, not even when the antibodies used are antibodies produced by hybridoma technique which must, apart from this, be expected to "produce" (react with) only pure interferon proteins.
At suitable dimensions ofsuch antibody columns (which can be designed in accordance with well-known principles for antibody affinity chromatography columns), the columns may be used for large scale indus-trial purification of interferon from a crude interferon preparation to result in pure (or highly purified) interferon proteins in the column elate. The pure (or highly purified) interferon proteins prepared in this way are stabilized with suitable stabilizers according to the intended use thereof, e.g. as described above.
As the interferon of the interferon preparations applied on the monospecific anti-interferon columns is usually present in very low con-centrations, on a weight basis, and as great amounts as possible of the valuable interferon are to be isolated, it is of importance to minimize any deterioration of the interferon proteins which might be caused due to the presence of proteolytic activity in any biological substance with which the interferon is contacted. Consequently, one aspect of the present , ::

:lZ~S3~

invention comprises removing any proteolytic activi-ty from any biological material with which the interferon to be purified is in contact.
One important utility of this aspect is the removal of pro-teolytic activity from the anti-interferon antibodies (immoblobulins).
According to this aspect of the invention, this removal is suitably performed by treating the antibodies, prior to their binding to the matrix, with matrix-immobilized enzyme inhibitor or enzyme destructor which is not harmful to immunogolbulins (or the important fragments thereof).

A method is also provided for purifying human interferon. The method includes applying an aqueous solution containing human Le form interEeron protein in a form having a specific activity of at least 50,000 - 100,000 IFU per mg of protein, the solution being buffered to a pH of 6.5 - 8 and having an ionic strength substantially not exceeding the ionic strength of a 10 - 100 mM, in particular 20 mM phosphate buffer, pH 7.2 solution, optionally together with a water miscible organic sol-vent, e.g., ethanol in amounts of 5 - 80%, on a matrix-immobilized ligand capable of binding the interferon according to the mechanism exerted by CIBACRON Blue F3GA, and thereafter eluting interferon thus bound.

The human Le form interferon produced thereby has a specific activity of at least 30 x 10 IFU per mg protein, the protein determina-tion being based on the Lowry procedure using pure human albumin serum as standard. Such human Le form interferon may have a specific activity in the range from 30 x 10 to 10 IFU per mg protein, or preferably such human Le form interferon may have a specific activity of 30 x 106 _ 70 x 10 IFU per mg protein.
The antibodies may be passed through a column of matrix-)53g9 immobilized poly-L-lysin and/or matrix-immobilized soyabcan trypsin inhibitor, and/or matrix-immobilized kallikrein inactivator. An example of a suitable treatment of the antibodies, is passage through a column of poly-L-lysin covalently bound to cross-linked agarose, e.g., SEPHAROSE
4B, (trade mark) followed by passages through a column of soyabean trypsin inhibitor covalently bound to the same matrix. It has been found that this removal of proteolytic activity increases the recovery of interferon activity in antibody affinity chromatography purification of interferon-containing solutions.
The monospecific anti-interferon, when covalently bound to a matrix e.g., cross-linked agarose, is preferably bound to such an extent that the total amount of antibody covalently bound to the matrix corresponds to at the most 85% of the immunoglobulins used at the covalent binding stage, e.g., as described by the present inventor in Scand. J.
Immunolog., 6, 77 - 86 (1977). This results in the highest recovery of interferon from the column.
When the eluate from the monospecific anti-interferon affinity chromatography column is to be used for administration in human beings, it is important that it does not contain any component which might be immunogenic in man. One risk which might be associated with antibody affinity chromatography is that immunoglobulins or immunoglobulin fragments liberate from the column and become eluted together with the desired protein or proteins.
According to another aspect of the invention, such immunoblobu-lins or fragments thereof which are immunogenic in man are removed by passage of the eluate through an antibody affinity column in which the antibodies are directed against the anti-interferon immunoglobulins and lZ(~S399 are of a kind which is non-immunogenic on parenteral administration to human beings. (Prior to the passage of the eluate through the column, it should be adjusted to a neutral pH, e.g. by dialysis against PBS, pH
7.2).
lmmunoglobulins which are non-immunogenic on parenteral adminis-tration to human beings are primate immunoglobulins, but the access to primate immunoglobulins directed against the immunoglobulins of the animal used for the preparation of the monospecific anti-interferon may be limited or completely precluded, for legal or ethical reasons. Therefore, it is important to note that pig IgG immunoglobulins have been found to be non-immunogenic in man, e.g., as is described in U.S. Patent No. 4,132,769.
Antibodies produced in a human hybridoma system, when available, would constitute another alternative.
The removal of any anti-interferon immunoglobulin or immuno-globulin fraction from the eluate of an anti-interferon affinity chroma-tography may preferably be performed by passing the eluate (after adjust-ment of pH to neutrality) through a column of matrix-immobilized pig IgG
directed against the anti-interferon immunoglobulins.
Pig IgG immunoglobulins directed against the anti-interferon immunoglobulins may be prepared in a manner known per se by immunizing a pig with immunoglobulins from the anti-interferon immunoglobulin-pro-ducing animal species, and isolating the IgG immunoglobulin fraction from the antiserum harvested from the pig, in accordance with the methods dis-closed in the above-mentioned U.S. Patent No. 4,132,769.
In a more generalized manner, a method is provided herein for removing proteins from which are immunogenic in man from a protein solution which is to be administered to human beings, comprising subject-~'Z~Q53'.39 ing the protein solution to an antibody affinity chromatography treat-ment where the antibodies are immunoglobulins directed against the immunogenic proteins, the immunoglobulins being of a kind which is non-immunogenic on parenteral administration to human beings. As will be evident from the above explanation, the non-immunogenic immunoglobulins preferred are primate immunoglobulins or pig IgG immunoglobulins.
A method is also provided herein for producing interferon pro-teins or proteins having the significant biological interferon-activity determinants thereof, comprising cultivating a microorganism carrying DNA
coding for the production of Le form interferon proteins as described above and recovering the proteins from the culture medium. The DNA
coding for the production of interferon proteins may be present on a plasmid.
Preferably, the DNA coding for the production of the interferon proteins was prepared by treatment, with reverse transcriptase, of messen-ger RNA isolated from interferon-producing cells in a manner known per se and comprising the stage of antibody affinity chromatography and/or immunoprecipitation on a lysate of the cells, the antibody being used in the antibody affinity chromatography and/or immunoprecipitation being an antibody as described above. It is preferred that the interferon-producing clones of the microorganism be selected by means of radio-labelled monospecific antibodies described above.
A method is also provided herein for removing proteins which are immunogenic in man from a protein solution which is to be administered to human beings, comprising subjecting the protein solution to an anti-body affinity chromatography using antibodies against the immunogenic proteins, the antibodies used in the antibody affinity chromatography .

lZ~0531~9 being of a kind which are non-immunogenic on parenteral administration to human beings.
The antibodies used in the antibody affinity chromatography may be primate immunoglobulins or pig IgG immunoglobulins. The protein solu-tion tion subjected to the antibody affinity chromatography may be eluate from an antibody affinity chromatography using animal immunoglobulins which are immunogenic in man, e.g. sheep or rabbit immunoglobulins.
According to the teachings of the present disclosure the matrix-immobilized human interferon may be immobilized on a matrix selected from the group consisting of CNsr-activated cross-linked agarose, CH-activated cross-linked agarose, and epoxy-activated cross-linked agarose.
It is preferred that the immunoglobulins first be passed through a column of crude concentrated human leukocyte interferon or depleted crude interferon solutions where the interferon has been removed in a com-paratively specific manner, e.g. by antibody affinity chromatography, immobilized on an agarose gel selected from the group consisting of CNBr-activated agarose and epoxy-activated agarose and thereafter be passed through a column of crude concentrated human leukocyte interferon immobilized on CH-activated agarose. The anti-interferon may be passed several times through each of the types of column. For example, the immunoglobulins are passed 4 - 6 times through the column selected from the group consisting of CNBr-activated agarose and epoxy-activated agarose and 1 - 3 times through the column of crude concentrated human leukocyte interferon immobilized on CH-activated agarose.
It is preEerred -that the immunoglobulins, prior to passing through any of the columns of matrix-immobilized crude concentrated human leukocyte interferon, be absorbed by passing several times through a

3''~

column of matrix-immobilized human serum.
By another aspect of this invention, this method includes the further step of removing any proteolytic activity from the immunoglobulins by passing the immunoglobulins several times through a column of matrix-immobilized en~yme destructor.
The purification stages performed according to the above described aspects of the invention provided by the above-identified parent application to prepare the pure human leukocyte interferon pro-teins (human Le form interferon proteins) from crude human leukocyte in-terferon comprise concentration by precipitation of proteins with KSCN, gel filtration, ligand affinity chromatography, and antibody affinity chromatography. Although such stages are known per se in the interferon art, the particular combination thereof and the particular conditions applied in certain of the operations constitute novel features, some of which are in themselves aspects of the invention provided by the above identified parent application. The particular way in which the stages are performed, and the particular combination of operations have resulted in optimal purification and concentration of the interferon, with minimum loss of interferon proteins during the sequence.
The KSCN precipitation is preferably performed by lowering the pH of the crude interferon containing a KSCN concentration of 0.5 M
to pH 4.5 instead of the conventional lowering to pH 3.5. This results in a considerably lower amount of protein in the precipitate, thus facilitat-ing the later purification steps.
The gel filtration is performed with a buffer solution containinq 25~ by volume of ethylene glycol and being 1 molar with respect to NaCl, incl. PBS (pH 7.2). This results in a much better resolution than when - 31 ~

.... .

9.zos399 using PBS or low pH (2.4) alone, or when using urea, PBS at pH 7 2.
The eluate fractions containing essentially only proteins in the 10,000 -20,000 Daltons range are collected.
The ligand affinity chromatography is performed in a novel and extremely advantageous way and constitutes one important aspect of the present invention:
The ligand affinity chromatography is performed under specified conditions on an interferon having a specific activity of at least 50,000-100,000 IFU per mg. protein, using immobilized cIsAcRoN Blue F3GA as the ligand. The use of CIBACRON Blue F3GA as the ligand for affinity chroma-tography of interferon was known in the art, but according to aspects of this invention, it has been found that the selectivity of this ligand increases drastically when particular combinations of conditions are used:
the interferon applied should have a much higher specific activity i.e. a specific activity of at least 50,000 - 100,000 IFU per mg protein, than in the conventional uses of this ligand type twhere crude human leukocyte interferon of a specific activity of 3 - 5 x 10 IFU per mg of protein is applied), and the solution in which the interferon is applied on the column should be in the pH range of 6.5 - ~3 and should have an ionic strength which does not essentially exceed the ionic strength of a 10 -100, in particular 20, mM phosphate matrix preferably is a cross linked agarose, e.g. SEPHAROSE 4B.
By another aspect of this invention matrix-immobilized antibodies are provided which are substantially free from proteolytic enzymatic activity. Such antibodies preferably are those which have been sub-stantially freed from any proteolytic enzymatic activity by treatment with enzyme inhibitors or enzyme destructors.

~Z~53~

Another aspect of this invention provides such antibodies which, prior to their covalent binding to the matrix, have been paSsed through a column of matrix-immobilized poly-L-lysin and/or matrix-immobilized soyabean trypsin inhibitor, and/or matrix-immobilized kallikrein inac-tivator.
According to an aspect of this invention, it is preferred to use, as the immobilized CIBACRON F3GA composition, Blue DEXTRAN 2000 coupled to SEPHAROSE 4B (by means of CNBr-activated SEPHAROSE 4B).
The elution of the interferon from this type of immobilized ligand has been found, according to another aspect of the invention, to be extremely selective when using 0.6 M NaCl buffered to pH 7.2, and pH
7.2 is also the preferred pH of the interferon-containing solution applied.
The interferon which is purified by the affinity chromatography according to this aspect of the invention provided by the above identified parent applieation is typically an interferon containing human Le form interferon proteins, e.g. human interferons (apart from human fibroblast interferons), that is, e.g., human leukoeyte interferons, human lympho-blastoid interferons and human Lè form interferon proteins or important parts thereof when produced by cultivation of a microorganism clone containing DNA coding for the production of such interferon protein.
(The fact that human lymphoblastoid interferon buffer, pH 7.2. When such a relatively high specific activity of the interferon is applied, the speeifieity of the ligand changes, and a higher degree of seleetive bind-ing of the interferon proteins to the ligand o~curs. cIsAcRoN F3GA is believed to interact with interferon prot~ns in a way which indicates the existenee of a "dinucleotide fold: and in this interaction, it seems idO S 39~

to have the same binding site as polyribonucleotides. It is believed that the particular advantageous properties shown by CIBACRON F3GA under particular critical conditions as discussed above will also be exhibited by the other members of the class to which this ligand pertains. The presen-t aspect of the invention provided by the above identified parent application, therefore, is constituted by a method of purifying human interferon, comprising applying an aqueous solution contanining human Le form interferon protein in a form having a specific activity of at least 50,000 - 100,000 IFU per mg of protein, the solution being buffered to a pH of 6.5 - 8 and having an ionic strength substantially not exceeding the ionic strength of a 10 - 100, in particular 20, mM phosphate buffer, pH
7.2 solution, optionally together with a water miscible organic solvent, e.g. ethanol in amounts of 5 - 80%, on a matrix-immobilized ligand capable of binding the interferon according to the mechanism exerted by CIBACRON
F3GA, and thereafter eluting the interferon thus bound.
Examples of materials which are matrix-immobilized CIBACRON
F3GA are "Blue DEXTRAN 2000" (matrix: dextran having a molecular weight of 2 million), and Blue SEPHAROSE CL-6B. Further details concerning these and other materials and their use in the conventional interferon puri-fication appear from Bollin et al., Preparative Biochemistry, 8(4), 259 -27~ (1978).
By another aspect of this invention, antibodies are provided which are covalently bound to the matrix. The (Namalva) contains a minor proportion of interferon of fibroblast character (F form - corresponding to 15% of the biological activity) does not detract from the fact that human lymphoblastoid interferon is, with respect to its major interferon activity, a human Le form interferon in that it contains human Le form l~d(~ S 399 interferon proteins (corresponding to 85% of the biological activity) having determinants identical with determinants of human leukocyte inter-feron proteins, such as has been shown according to the present inven-tion.) It is preferred that the specific activity of the interferon preparation applied on the affinity column be 100,000 - 1,000,000 for example 200,000 - 1,000,000, e.g. 500,000, e.g. 500,000 - 1,000,000 IFU
per mg protein.
The eluate from the affinity chromatography column operated in accordance with this aspect of the invention provided by the above identified parent application may also be a product adapted for thera-peutic use. It will often have a specific activity of at least 30 x 10 IFU per mg protein, based on the Lowry procedure using pure human albumin serum as standard, e.g. 30 x 10 - 10 , e.g., 30 x 10 - 70 x 10 IFU per mg protein. For administration to human beings, this preparation is sub~
ject to normal pharmaceutical precautions, e.g. precautions to ensure sterility and freedom from pyrogenicity. The dosagc of the preparation will correspond to the dosage stated above for the'pure interferon, on a total activity basis.
As explained in the "Experimental section", the eluate from the affinity chromatograph column was, in the original experiments leading to the pure interferon, subjected to final purification by passage through an absorbed antibody affinity column in which the antibodies are immuno-globulins raised against partially purified human leukocyte interferon and then subjected to removal of antibodies against contaminating pro-teins by several passages through columns of matrix-immobilized crude human leukocyte interferon. As appears from the more detailed explanation ~2053~

be~ow, the covalent binding of crude interferon to a matrix (e.g., SEPHAROSE 4B) destroys the immunological determinants of the interferon itself, (~ 98%), but apparently not the determinants of the major part of the impurities, and this means that when immunoblobulins raised against partially purified leukocyte interferon are passed (normally several timesj through the column, the anti-impurities thereof will be retained on the column, while the anti-interferon will pass the column. Such absorbed anti-interferon (absorbed several times) was used in the antibody afflnity chromatography stage following the affinity chromatography.
As appears from the "Experimental section", a preferred way of operating the affinity columns, that is, t.ne Blue.DEXTRAN SEPHAROSE
column and the antibody affinity column, is to connect the two columns so that the eluate from the Blue DEXTR~N column at the same time loads the antibody affinity column. This prevents any loss which might otherwise occur if the eluate fractions from the Blue DEXTRAN column were handled separately.
In the final concentration of the human Le interferon proteins, a unique method of concentrating proteins by precipitation with SDS was used. This method constitues a further aspect of the invention providea by the above identified parent application and comprised precipitating SDS or a salt thereof from a solution of the protein with contains SDS, preferably in a concentration of O.l - 4 per cent by weight, in particular 0.1 per cent by weigh-t to obtain a precipitate comprising a complex or complexes of SDS or a salt thereof with the protein, separating the precipitate from the solution, preferably by centrifugation at O - 4C, and redissolving the precipitate in a smaller liguid volume. The ~;~OS3~

precipitation of the SDS may suitably be obtained by either (a) lowering the temperature to 0C for 15 minutes or (b) adding a salt, e.g., a K
salt, which forms a precipitate with SDS or with SDS-protein complexes.
This method is a valuable method for concentrating aqueous solutions of pure or purified interferons, and, as indicated above, has been found to ben an excellent way of concentrating human Le form interferon proteins.
The total purification sequence performed in accordance with the invention provided by the above identified parent application was found to be extremely activity-preserving: From a starting amount of proteins of 7 x 10 gamma, the pure interferon isolated was less than or equal to 1 gamma, (as determined by comparison of protein bands on SDS
PAGE). Yet, the overall decrease in total interferon activity from the starting batch of crude interferon to the pure interferon was only from

4 x 10 IFU to 1.85 x 10 IFU (i.c., 50'). This emphasizes the unique character of the purification sequence and the above-mentioned critical stages thereof.
As indicated above, it has been found that the covalent binding of crude interferon to a cross-linked agarose (SEPHAROSE? matrix can be performed under conditions which will substantially destroy the immuno-logical determinants of the interferon itself. Accordingly, non-anti-interferon immunoglobulins are absorbed from antiserum raised against a partially purified human interferon, by subjecting the antiserum to an absorption treatment by means of a matrix-immobilized crude concentrated human non-fibroblast interferon which has been bound covalently to the matrix in such a way that the major part of the interferon activity thereof has been destroyed. The extent to which the interferon activity has been destroyed by the binding to the matrix can be assessed using the normal interferon assay methods described in the section "materials , . , 3g~

and Methods". It is preferred that at least 90% of the interferon activity be destroyed, preferably at least 97 or 98~ thereof, in order to avoid excessive removal of anti-interferon activity from the immunoglobulins.
It has been found that when crude concentrated human leukocyte interferon is bound to cross-linked agarose ("SEPHAROSE") (which may be CNBr-acti-vated, epoxy-activated, or CH-activated3), the interferon activity there-of is destroyed to a large extent if the binding is performed in such a manner that at least 50% of the proteins present are bound on the matrix.
Apparently, the major part of the antigenic determinants as far as the impurities are concerned survive the treatment.
The most suitable way of performing the absorption is to build up a column of the matrix-immobilized crude concentrated human interferon and pass the anti-interferon through the column, It is often preferable to pass the anti-interferon through a column of matrix-immobilized human serum prior to the passage through the column of matrix-immobilized crude concentrated human interferon, because human serum will in itself bind a considerable part of the antibodies present in the antiserum raised against partially purified human interferon. However, it has been found that a human serum alone is not capable of yielding such a high degree of puri-fication as is possible with matrix-immobilized crude concentrated human interferon, and this is believed to be due to the fact that some of the impurities in the partially purified human interferon against which the antiserum has been raised are not normal cellular proteins, or are normal cellular proteins present in "abnormally" high concentrations in the induced cells. Therefore, according to the principles of this aspect of the invention provided by the above identified parent application, it was attractive to absorb antibodies to impurities from an anti-interferon .

~,'2QS39~

serum by means of a crude interferon preparation having a relatively low concentration of interferon and a high concentration of impurities. This unique concept, combined with the finding that it was possible to destroy the interferon activity and, concomitantly, the interferon determinants, of the crude concentrated human interferon immobilized on a matrix, e.g.
cross-linked agarose, forms the basis of a most useful technique accord-ing to which the absorbed an~iserum, which contains a much lower concen-tration of non-anti-interferon immunoglobulins, is used for antibody affinity chromatography to purify human interferons, including human leukocyte interferon and human Namalva interferon, to a high degree of pur _ le and economic way.

\\\

.

~2~5~99 In the experiments so far conducted, it appears that CH-activated cross-linked agarose (SEPH~ROSE) is the matrix on which crude concentrated human leukocyte interferon is capable of yielding the highest degree of purification of an antiserum raised against partially purified human leukocyte interferon, but that the interferon when bound to this matrix results in a somewhat higher reduction of the antiinterferon activity than when the crude concentrated human leukocyte inter-feron is bound to CNBr-activated SEPHAROSE or epoxy-activated SEPHAROSE. Therefore, a preferred way of performing the ab-sorption is to pass the antiserum (~hich has preferably already been absorbed in a column with matrix-immobilized human serum) first through a column of epoxy-activated SEPHAROSE or CNBr-activated SEPHAROSE, and thereafter through a column of crude eoncentrated human leukocyte interferon bound to CH-activated SEPHAROSE. The antiserum is passed several times through the columns until the desired purification thereof has been obtained e.g. as can be assessed by eluting the column subse~uent to passing the antiserum and determining the amount of proteins eluted. A typical proeedure is to pass the antiserum 4 - 6 times through a column of human serum proteinjfollowed by 4 -times through a column of human crude concentrated leukocyte interferon covalently bound to epoxy-activated SEPHAROSE, thereafter 1 - 3 tirnes through a colurnn of the human crude con-centrated leukocyte interferon covalently bound to epoxy-activ-ated S~PHAROSE and thereafter 1 - ^ times through a column of the human crude concentrated leu}:oc~te interferon bound to CH-aetivated SEPHAP~OSE.

~. .

12,053~9 During these absorption procedures, the anti-interferon is ob-tained in the wash, and the eluate is discarded. However, the amount of protein in the eluate is indicative of the degree of purification obtained.
In the final stage of purification, the degree of protein elutable from the column after the last passage of the antiserum should preferably be at the most 30 ,ug when the column is of a size of the order of 10 - 50 ml.
The degree to which the anti-interferon activity is recovered in the wash upon passage through the column is another indication to what extent the crude concentrated human leukocyte interferon has successfully been covalently bound to the matrix in such a way that the immunological interferon determinants thereof have been destroyed. The anti-interferon should preferably pass through the column with retainment of at least 90%, preferably at least 95% of its anti-interferon activity.
In analogy with what has been stated above, the immunoglobulins absorbed in accordance with this aspect of the invention provided by the above identified parent application are preferably freed of any proteoly-tic activity prior to being bound to the matrix. This may be obtained in the manner described above, preferably by passing the immunoglobulins through a column of matrix-immobilized enzyme inhibitor or enzyme destru-tor, preferably several times, e.g. 3 times.
A further aspect of the invention provided by the above identi-fied parent application is provided by a method of purifying human inter-feron by subjecting a human interferon-containing solution to antibody affinity chromatography using as antibody anti-interferon immunoglobulins from which non-anti-interferon immunoglobulins have been absorbed by the method herein described, and eluting the antibody-bound interferon. In this aspect, the anti-interferon is immobilized~n a suitable matrix in ~lZ1~5399 accordance with the general principles described further above. As appears from the experimental section, any human Le form interferon-containing solution can be purified in this manner, including human leuko-cyte interferon, and the Le form of human Namalva interferon. The result-ing purified interferon solutions may have a very high degree of purity and are then suitably stabilized in the same way as described above for the pure or purified human leukocyte interferons prepared in accordance with other aspects of the invention provided by the above identified par-ent application and may be used for the same purposes as the other highly purified interferon preparations mentioned further above.
The particular experimental conditions used for the first preparation and characterization of the pure human leukocyte interferon proteins appear from the below sections "Materials and Methods" and "Experimental Section".
In several repeated experiments, it has been established that under the SDS PAGE and staining conditions described in the section "materials and Methods", at a total interferon load of 0.9 x 10 IFU, pure human leukocyte interferon shows essentially only two sharp stained protein bands at 18,400 +200 and 20,100 +200 Daltons, respectively, and a minor stained protein band between 20,300 +200 and 20,400 +200 Daltons.
As determined by the protein determination described below, the pure human leukocyte interferon has a specific activity of 10 IFU per mg or protein; the specific activity found may vary to some extent depending upon the protein determination method employed, and the specific activity on a protein weight basis is judged to be 2 x 10 - 2 x 10 IFU per mg or protein. The fact that the pure interferon shows two major distinct bands is in accordance with prior art findings using crude or partially .....

~LZOS3~3~

purified interferon preparations which indicated that human leukocyte interferon comprises at least two major species. At a higher total interferon load, e.g., of 3.8 x 10 IFU, the above-mentioned SDS PAGE
system has been found to be capable of showing a more differentiated protein pattern comprising six interferon protein bands, i.e. the two strongly stained bands at 18,400 +200 Daltons and 20,180 +200 Daltons, respectively, a medium-stained band at 20,420 +200 Daltons (corresponding to the above-mentioned minor stained band) and just visible protein bands at 19,500 +200 Daltons, 21,130 +200 Daltons, and 23,440 +200 Dal-tons, respectively. Each of the individual components in the above-mentioned bands of the SDS PAGE acrylamide gradient gel has been found to show biological interferon activities: antiviral activity, ability to neutralize only anti-human leukocyte interferon (but not anti-human fibro-blast interferon), and anticellular activity, plus a variety of so-called non-viral activities, as exemplified by potentiation of Natural Killer cells, potentiation of MLC-CML, increase of ~A antigens, etr.
Before discussing the materials and methods, a brief reference to the accompanying drawings will be given in which:
Fig. 1 is a drawing of a stained SDS PAGE gradient gel slab;
Fig. 2 is a graph of log Mw as ordinate and gel length in CM
as abscissa of another SDS slab;
Fig. 3 is a graph of log Mw as ordinate and gel length in CM
of yet another SDS slab;
Fig. 4 is an elution pattern of a Blue DEXTRAN-SEPHAROSE 4B
column loaded with partially purified human leucocyte interferon;
Fig. 4a is a drawing of another stained SDS PAGE gradient gel slab;

......

lZC~53'~

Fig. 5 is a graph of a gel filtration curve of HuLe CIF with interferon units per ml as ordinate and fraction number as abscissa;
Fig. 6 is a graph of a gel filtration curve of NaClF with interferon units per ml as ordinate and fraction number as abscissa;
Fig. 7 is a graph of a Blue DEXTRAN chromatography of Namalva interferon with fraction number as abscissa;
Fig. 8 is a graph of the eluate from Blue DEXTRAN chromatography of Namalva interferon loaded to the absorbed antibody column with fraction number as abscissa;
Fig. 9 is a drawing of the stained slab from the SDS PAGE of the pure Namalva interferon proteins tA) and of the eluate from the Blue DEXTRAN column (B); and Fig. 10 shows the antiviral profile as assessed on an SDS PAGE
for the potentiation of the Natural Killer cell system (NK system).
Now referring briefly to some of these drawings, at an inter-feron load of 0.9 x 10 IFU, the pure human leukocyte interferon proteins appear as three individual protein bands in the SDS PAGE acrylamide gradient gel, together with five-six biological peaks. Whether five or six biological bands are found depends on the exact places at which the gel slice is cut. Fig. 1 shows a stained SDS PAGE gradient gel slab prepared at this load as described in the section "Materials and Methods" below. Each of the protein bands has been shown to possess dis-tinct interferon activity. Fig. 2 is a drawing of an SDS slab from another experiment at the same interferorl load, Fig. 2 also showing the interferon activity profile associated with the bands, determined as explained under "Materials and Methods" below. Five biological inter-feron peaks are seen, together with three distinct stained proteins.

12VS39~

From Fig. 2 it can been seen, unambiguously, that the protein bands coincide strictly with the peaks of interferon activity. This proves that the proteins are interferon proteins. It is important to note that the interferon activity profile will, of course, depend upon the exact position of the individual slicings of the gel. In Fig. 2, the individual interferon activity of the minor band at 20,410 +200 Daltons is not so evident, but in other experiments at the same inter-feron load, it was shown that the minor band itself possesses interferon activity, and in experiments with a high load, the minor band was found to be a distinct interferon sub-species. The amount of interferon activity from SDS PAGE found in the corresponding interferon protein slices corresponds unearly with the amount of protein as assessed from the intensity of the staining of the protein bands. Thus, the unambigu-ous existence of the two major interferon proteins and the minor band has been demonstrated in the experiments illustrated in Figs. 1 and 2.
In experiments where the interferon load in the system was higher, the above-mentioned more detailed band pattern was demonstrated, e.g. as illustrated in Fig. 3 (six interferon proteins together with six biolo-gical peaks determined after staining and destaining). As it is known that interferon treated with SDS will retain its immunological deter-minants and even expresses ( or preserves) its antigenicity in a more distinct way compared to non-SDS-treated interferon (as shown by immunizations of mice wi-th human leukocyte interferon preparations of Paucker et al. (Dalton,B.F., Ogburn, C.A., Paucker, K., Production of antibodies to human interferons in mice, Infect. Immun. 19(2), 570 - 574 (1978), pp 4; 25 - 30), preparative SDS PAGE makes it possible not only to obtain each of the components in isolated form, but also to perform 0539~

immunization with the isolated components, e.g. as illustrated in greater detail below.
Fig. 4 shows the elution pattern of a Blue DEXTRAN-SEP~ROSE 4B
column loaded with partially purified human leukocyte interferon, l ml, specific activity 500,000 IFU per mg protein, subsequent to throughout dialysis versus 20 mM phosphate buffer (PB), pH 7.4. The size of the fractions was 5 ml, and the flow rate was 35 - 40 ml/h. The column was washed with 20 mM PB for 2 hours before it was eluted stepwise with 0.2, 0.4, 0.6, 0.8, and 1.0 M NaCl in PB 7.4, respectively. The total eluate (I + II + III) contained 754,000 IFU (in 30 ml), the originally applied amount being determined to 750,000 IFU. Hence, the recovery was 100%.
The specific activity of the eluate was 2.1 x 10 IFU per mg or protein.
The purification factor was 42. When checking the eluates in an SDS
PAGE, most of the eluted proteins (~ 98%) appeared above 50,000 Daltons (impurities) as seen in Fig. 4a which shows an SDS PAGE of the input, wash, and eluate of Fig. 4. Although, as will appear from the above, 0.6 M NaCl buffered to pH 7.2. is a most preferred eluant for the affinity column, it will also be noted that a broader concentration range is quite selective, and the invention in a further aspect comprises the elution with aqueous NaCl solution of a concentration of 0.5 - 0.7, in particular 0.5 - 0.65 molar and buffered to a pH of 6.5 - 8, or other aqueous solution buffered to a pH of 6.5 - 8 and having an ionic strength corresponding to such NaCl.solution. The use of other eluants is also within the scope of yet further aspects of the invention provided by the above identified parent application. Examples include salts and/or ethylene glycol in stepwise and/or gradient-wise increasing ccncentration up to 50%, aminoacids, artificial aminoacids, ampholines, and proteins ., .

- 1;205399 and protein mixtures. As mentioned above, the interferon solution may be applied together with a water miscible organic solvent, e.g. alcohol, in particular ethanol.
MATERIALS AND METHODS
Interferon assays were performed according to the well-known standard method (Berg K., Sequential Antibody Affinity Chromatography of Human Leukocyte Interferon, Scand. J. Immunol. 6, 77 - 86 (1977)) using VERO cells (monkey kidney cells) and Vesicular Stomatitis Virus (VSV) as a challenge virus. All interferon ~I.Z053~

units (IFU) are e:~pressed in international reference units (69/19 B
units) (69/19 B reference was obtained from MRC, Mill Hill, U.K.) lnterferon. Crude human leukocyte interferon t~as produced accor-ding to the method as described by Cantell (Cantell, K., Hirvo-nen, S., Mogensen,' K. E. and Pyhala, L., Human leukocyte inter-feron: production, purification, stability and animal experiments.
In: The Production and use of Interferon for the Treatment and Prevention of Human Virus Infections pp. 35 - 38, Waymouth, C.
(ed . ) . Proceedings of a Tissue Culture Association Workshop held at Lake Placid, 1973 (In Vitro Monograph, volume 3), Tissue Cul-ture Association, Rockville, Md.) using Sendai virus as interferon inducer. Partially purified interferon (PIF) with a specific activity of 5 x 105 IFU/mg protein was obtained from crude concentrated '1. human ]eukocyte interferon (CIF) by ethanolic precipitation as de-scribed by Cantell, K., Hirvonen, S., Mogensen, ~. E. and Pyhala, L., loc. cit.

Crude Namalva interferon was produced substantially as described by Strander et al ., Production of human lymphoblastoid interferon , J. Clin. Microbiol. 1, 116 - 124 (1975), using Sendai virus as inter-feron inducer.

Interferon neutralization for determining anti-interferon was per-formed in a micl~o-assay system in the follo~ing manner: 20,000 VERO
cells per well ~vere seeded in 100 l~l medium and kept at 5% C2 in - a humidified cabinet. On day 2 the medium ~Yas removed from the cells, and each ~!ell received 100 ~l of a dllution (in medium) of Lhe anLiserum, containing an interferon concentration of Ç - 8 IFU/ml (the serum and interferon had been preincubated at 37C for 1 h).
On day 3 the medium vas removed, and all the ~ells received 100 7S~7 (diluted to 10 3'5 in medium). On day 4 ihe CPE (cyto-pathogenic effect) -as determined, and 50% destruction ~as taken as the end point for the determination of the anti-interferon titer.
The titel s are e~;pressed as in~erferon neutl-alization units (IFU-NU) per ml.

lZ~S3~3~

Non-monospecific anti interferon against PIF ~vas produced, according to Mogensen, K. E., Pyhala, Liisa and Cantell, K., Acta path. micro-biol. scand. Sect. B, 83, 443 - 450, (1975), part1~r in sheep, partly in rabbits. The titer of the sheep anti-interferon was 100 - 250,000 IFU-NU/ml. For the preparation of the rabbit anti-interferon, a rabbit ~vas injected ~veekl~,r, s.c. with PIF (2 x 105 IFU) for more than t~vo years. The titer of the rabbit anti-interferon was 15,000 -30,000 XFU-NU/ml. All immunoglobulins were isolated by 50% ammonium sulphate precipitation, follo~ ed b~,r a dialysis versus phosphate buffer saline (PBS), pH 7.2.

Chemicals. CNBr was- from Fluka (stored at -20C). Sodium dode-c~rlsulphate (SDS), special~y pure for electrophorese, was purch-ased through British Drug House (BDH). So~rabean Trypsin Inhi-bitor (STI) and L-Lysine were obt~ined from Sigma. SEPP~ROSE 4B, CNBr-~ctivated SEPHAROSE 4B, CH-activated SEPHAROSE 4B, and Epoxy-activated SEPHAROSE 6B all designating trade marks, were all purchased from Pharmacia (De~,lmark).
Bindingr Procedures. The covalent binding of the immunoglobulins to SEPHAROSE 4B was done as previously described b~r I~. Berg in Scand. J. Immunolog., 6, ~7 - 8G, (1977). Only 80 - 85% of the immunoglobulins were deliberately bound.

P tein determinations ~ere made by a modification of the Lo~vry procedure (Berg K., Sequential Antibod~r Affinit~r Chromatograph~,r of Human Leukocyte lnterferon , Scand . J . Immunol ., 6 , 77 - 86 (1977)) ~vhich permitted detection of 1 - 2 llg/ml as the lo~vest level of pl oteins detectable (using an LKB Calculation Absorptioner Ultralab System). Crystalline bovine serum albumin ~vas used as a standard protein. To determine the protein concentration of the purified interferon (l - 5 ,ug in total) the fol~olving procedul-e ~vas adopted: SDS ~vas added to a final concentration of 0.1%. The l~ro-philized protein sample ~-~as ful-thel e~;amined on an SDS-polyacl yl-amide gel electrophorese (SDS PAGE, see later), subsequent to a dialysis versus distillcd ~vater. The intensity of the stained protein bands ~vas compared ~vith kno~-n standards in different amounts (see later, undel SDS PAGE), and the total amounts of proteins - a,g _ ~Z(~S399 were es~imated The deviations ~vere 5 - 10%, vith the ~o~vest detecl-able level of proteins being 0.l ,ug (in total). The results from this method will serve as a rough estimate, rather than as an actual measurement .

Affinit~ chromatographies ~vere performed at 4C. The gel suspen-sions ~vere degassed before packed into the columns. Packing was performed by washing l~-ith 3 - 5 bed volumes of loadmg buffer, using a peristaltic pump. Samples (100 ~1) for interferon titrations ~ere taken from either pools or individual fractions and titrated on the same day or frozen in plastic tubes (-20C) and titrated later.
The dilutions were made in medium (incl. 10% calf serum).

Antibod~7 affinit~ chromatography ~vas essentially done as described b~ Berg (Sequential Antibody Affinit~ Chromatography of Human Leukocyte Interferon, Scand. J. Immunol., 6, 77 - 86 (lS77)). As loadin~ buffer ~vas used 0.1 M NaOA/0.3 M NaCl at pH 7.2 (flow rate 40 ml/h). Step-vise elution was performed ~ith 0.1 M HOAc/0.3 M NaCl including a minute amount of citric acid (enough to keep the pH firmly at 2 . 4) . When not operated, the co]umn was stored at 4C in PBS I M NaCl inc]uding Penicillin, streptom~Tcin, Gentam~Tcin and Chloramphenicol (1% of each). Before using the column for puri-fication purposes, it ~vas first ~vashed ~vith 100 ml of loading buffer follo~ved b~T 10 ml of eluting buffer and finally equilibrated ~vith 20 - 30 ml of loading buffer. This waslling-cycle ~vas necessar~T to a~Toid "spontaneous" proteins, especiall~ Yhen ~vorking ~vith inter-feron of specific activities above 10 IFU/mg proteins. The plastic tubes used for collecting the interferon eluate were pre-wetted ~vith 100 ~1 of 1% SDS.

SDS PAGE. The purified, concentrated interferon preparations ~vere anal~zed for pol~peptides components on SDS PAÇE slab gels usin~
20 cm long separating gels 0.75 mm thick (known by the trade mark of B10 RAD
modcl 221: Dual ~ertical slab gel electrophoresis cell) and 7 - 10 cm long stacking gel. Exponential gradient gels of 9 - 22% polyacrylamide were prepared by mixing 11 ml 22~ acrylamide solution with abont 32 ml 9% solution in a simple ice-cooled gradient-device, as described in Knight E. Interfero:
Puriication and initial characterization from _ 50 -~.205399 human diploid cells. T'roc. natn. Acad. Sci. USA 73, 520 - 523 (1976).
The discontinuous buffer s~stem~ as described by Laenomli (Laenomli, U. K., Cleavage of Structural Proteins During Assembly of the Head of Bacteriophage T4, Nature 227, 680 - 685 (1970)) ~vas used. The gel ~vas pre-cooled for 2 h (10C) before starting the actual electro-phorese ~-~hich was performed overnight (10C) at constant effect (LI~B power supply), starting out ~vith 10 mA (and 20 V).
Samples to be analyzed were dissolved (or diluted) in 0.1 M Tris, HCl (pH 6.8) 2.5% SDS and 5% glucose including a tracking dye (sample buffer).The gel ~vas stained in Comassie Blue (1.25 mg/ml in 50% methanol, 40% H2O and 10% acetic acid), without prior fix-lû ation, for 15 minutes at room temperature.under constant rocking, and destained in 7æ acetic acid (5% methanol). The gels were dried on paper of a good quality [for example, that known by the trade mark l~1HATMAN CHROMATOGRAP~IC p~per (17 1~) ] under heat and vacuum using a gel dryer (BIO RAD, gel slab dryer, model 224). Solutions of five different mo-lecular markers, from 0.1 llg to 10 ,ug of each marker per 20 ~
Lactalbumin (14,400 Daltons); Soyabean Trypsin Inhibitor (20,100 Daltons); Carbonic Anh~drase (30,000 Daltons); Ovalbumin (43,000 Daltons); Bovine Serum Albumin (67,000 Daltons); Phosphorylase (9~,000 Daltons) (obtained as an elec-rophoresis calibration kit (Pharmacia, Denmark)) - were subjected to SDS PAC~E and stained.
It should be noted that molecular ~veights assessed in this manner 2~ are subject to e:Yperi~nental accurac~ of +200 Daltons. The stained protein bands ~ere compared ~vith the corresponding bands obtained Irom a parallel SDS PAGE of a purified interferon prepa-ration and the total concentration of interferon proteins ~vas esti-mated. For obtainiulg a biological profile from an SDS PAGE, the pal L of the gel intended for interferon determination, ~vas cut from the retl~ainder gel and kept at 4C (in a humidified bo~) Oll a g]ass plate. The main part of the gel ~as stained for 15 minutes;
after destaining for additionall~ 3 - 5 minutes, wcalc bands ~vere clearl~ seen on a blue baclcground, whereb~ the precise location of the protein bands corresponding to 14,000 and 30,000 Daltons could be established. The unstained part of the gel -as cut, so it onl~ contained proteins bet~-;een 14,000 and 30,000 Da]tons and ~as furiher subdivided in 1 mm pieces b~ shal p knives . Thc interfel on from these slices ~as eluted ~ith 0.5 ml 0.1 1~1 SDS subsequent to _ 51 ~2a!S399 a complete mincing by means of a TEFLO~ rod. After 5 h at room tcmperature (rocking) the interferon activity of the supernatant Yas determined. The individual fractions vere frozen at -20~C
~rithout any additives.

E~PERIMENTAL SECTION.

Preparation of Pure Human Leukocyte and Lymphoblastoid Interferon Proteins.

Concentration of crude human leukoc~te interferon. To 3 liters of crude human leucocyt interferon was added KSCN up to a concen-tration of 0.5 M at pH 7.2. The pH was lowered by addition of lN
HC1 to 4 . 5 (magnetic stirring) whereby a protein precipitate con-tainmg the interferon (and part of the impurities) ~Yas obtained.
The precipitate was dissolved in 150 ml of PBS (phosphate buffered saline, pH 7.2) including 1 M NaCl and 25% by volume of ethylene glycol and dialyzed thoroughly versus 3 times 2 liters of the same buffer at 4C. The specific activity of the crude concentrated human leukocyte interferon (HuLeCIF) was 5 - 10 x 103 IFU/mg protein. The recovery was 93%.

Concentration of crude Namalva interferon. To 1 liter of crude Namalva interferon, ~ith a titer of 8000 IFlJ/ml, ~as added KSCN up to a concentration of 0.5 M at pH 7.2. The pH ~vas lo~-ered by addition of lN HCI to 4 . 5 (magnetic stirring) ~hereby a protein precipitate containing the interferon (and part of the impurities) ~vas obtained. The precipitate was dissolved in 50 ml of PBS, pH
7.2, including l M NaCl and 25% b3~ volume of e~hylene gl~rcol and dial~zed thoroughly versus 3 times 2 liters of the same buffer at 4C. The specific acLivity of the crude concentrated Namalva inter-feron (NaCIF) ~vas 10 - 12 ~; 103 IFU/mg protein. The recovery ~vas 90% .

Ge] fi]Lration . A l00 cm long column (2. ~ cm in diameter, Pharmacia K 2.6/100) was ~acked with an ahsorbell~ known by the trade mark X ULTROGEI. ~cA 5/4 (LKB Denmark) in PB~ containing 1 M ~aCl and 25~ by volume of etllylene glycol at _ 52 ~

S ~

4C (p]I 7 . 2) . After ~tTashing the column with 3 bed volumes of burfer, the column was stabilized. 10 - 15 ml of HuLeCIF (pre-pared as described above in 25% by vo]ume of ethylene glycol, 1 M
NaCI in PBS, pH 7 . 4) ~vere loaded to the co]umn, and the column ltTas "eluted" ~.~ith the loading buffer, the fractions being assayed for interferon activity. The interferon-containing fractions were pooled, and 95% of the original interferon activity was re-covered. The specific activity of the gél filtered human leukocyte interferon-containing eluates ~as close to 1,000,000 IFU/mg protein, corresponding to a purification fac~or of 200. As determined b~
means of molecular markers, the molecular weight of the interferon corresponds to a range of 10, 000 - 20, 000 Daltons . Titrations of individual fraction revealed only one broad peak, ~-rith a maximum at 18, 000 Da3tons .

In the same manner as described above, 10 ml of NaCIF (prepared as described above in 25% by volume of ethylene glycol, lM NaCl in PBS, pH 7.4) were loaded to the column, and the "elution" was performed in the same manner as described above. The recovery 2n was 90%. The specific activity of the gelfiltered Namalva inter-feron-containing eluate was close to 1,000,000 IFU/mg protein, cor-responding to a purification factor of 100. As determined by means of molecu]ar markers, the molucular \tTeight of the interferon corre-sponds to a range of 10,000 - 20,000 Daltons. Titrations of the in-dividual fractions revealed a broad peak, with a ma~imum at 18,000 Dalton s .

The gel filtration curves for the above-described gel filtration of HuLeCIF and NaClF are sho~Tn in Figs. 5 and 6, respectively, and "HULEIF`I' indicates the-~human leu]cocyte interferon, ~Thereas "NALY~F"
indicates the Namalva (]~mp]~oblastoid) interferon. IL is clear]y seen that the interferon activity is effective]y separated from the major part of the proteins.

Blue De.~tran chromato~raph~. The gel-fi]tered human leukocyte interferon so]ution, obtained as described above, ~-as e~;haustively dial~ ed against 200 volumes of 20 ml\l PB, pH 7 . 2 at 4~C . l`he dia-]~sis ~as performed t~tice, the tota] dialysis time ~eing 24 `~:

~2QS399 hou; s The dial~T~zed solution (25 ml, containin,, 1 8 x 106 IFU) was loaded on a column of Blue DEXl'RAN-SEPHARO~ 4B The diameter of the column ~vas 1 cm, and the length of the column was 10 cm The column was pre-washed with 200 - 300 ml of 20 mM PB (phosphate buffer) at pH 7 4 The dialyzed interferon preparation was loaded to the equilibrated column, and the column was washed with 75 ml of PB 4500 IFU was found in the ~vash I'he column ~as eluted with 0 6 M NaCl, 20 mM PB, pH 7 2 ~vhereby more than 95% of the interferon activity was recovered in 6 ml of eluate, as determined by interferon titration In e~actly the same manner, the above-mentioned gel-filtered Namalva interferon solution vas exhaustively dial~zed and thereafter subjected to Blue DExTRAN chromatography The input in the Blue J)E~TE~A~
chromatography vas l,600,000 IFU The ~ash consisted of 70,000 IFU in 50 ml The eluate was obtained by means of 0 6 M NaCl in PB (pH 7 2) The Blue DE2~'1'RAN chromatography of Namalva inter-feron is i~lustrated in Fig 7 The fibroblast part of the Namalva interferon was not eluted from the column under the above con-ditions, but is expected to be eluted using, e g, 25% ethylene glycol in 1 M NaCl, pH 7 2 The above-mentioned Blue DEXTRAN column was a ~olumn of Blue ~XTRAN ( CIBACRON Blue F3GA ~mobilized on ~X~AN 2000 (molecular weight 2 millions)) coup]ed to cyanogen bromide-activated agarose (SEPHAROSE 4B) Thus, the more complete designation of the column - is Blue DEXTRAN-SEPHAROSE 4B This type of column is described b~ Bollin et al , loc cit After elution, the co]umn was cleaned b~T
elution with 25 - 30 ml 25% eth~,T]ene gl~col, 1 5 M NaCl in 20 ml~l PB Tlle column ~vas stored in this buffer at 4C wllen not in use As melltioned above, the loading conditions could also involve the use of h~Tdl^ophobic reagents, e . g. alcohols in various amounts (O - 50%).

The 0 6 I\l NaCl eluates from the Blue Dl X~RAN chromatogrraph~ sho~
a specific activit~T~ of 70 ~; lO~ IFU/mg of protein, both for the human leul;oc~te in~erferon and for the N Imalva interrel-oll Thus, these are canclidates for parenteral administraLion in human bein~s for _ 54 --53~

therapeutic purposes and, in this regard, are much more pure pre-parations lhan the commonly used PIF preparations. For this use, the eluates are stabilized with ph~siologicall~,~ acceptable stabilizers e.g., as described further above, for example 1% of human albumin.

For further purification and for preparation of pure interferon, the eluates from the E~lue DEXTRAN column are directl~7 transferred to an antibody affinit~r chromatography column. In the rnost advan-tageous embodiment, the antibody affinity chromatography column is combined with the Blue DEXTRAN column iIl a "tandem s~rstem" as described below:
.
Tandem Affinit~ Chromatograph~. Instead of eluting the Blue ~XTRAN
column as described above, the ~3lue DEXTRAN column is combined with the equilibrated antibod~ column prior to the elution, b~ con-necting the outlet of the Blue DEXTRAN column with the inle,t of th~.
antibod~r column. In this manner, the eluate from the Blue. DEXTR~
column is immediatel~r "caught" b~r the antibody column. This com-bination makes use of the fact that the elution conditions (0.6 M
NaCl, 20 ml~ PB, pH 7.2) can be used as loading conditions of the antibody column. After the elution/loading using 20 ml of the eluate/"loading buffer" (this "loading buffer", of course. at the same time contains the interferon eluted from the Blue DEXT~AN
column), the t~vo columns are disconnected, and the antibody column is washed further before eluted as described above. The human leukocyte interferon eluate from the antibod~r column con-- tains pure interferon proteins sho~ing a specific activit.~r of more than 109 I~U/mg of protein (as assessed by the determination method discussed above). I;or stabilization of the pure interferon proteins, the tubes in ~ hich the eluate from the antibod~ co]umn is collected (fraction size 2 ml) have been pre-~vetted with 100 ~l Or ~% SDS each. After pooling of the inLerferon-containing eluate, addi~ional SDS is added up to a total concentration of 0.1% b~ eight.

The pooled interferon-containing eluates stabilized -ith 0.1% SDS
are transferred to a 20 ml stainless steel tube pre-cooled to 0C in an ice bath. After 15 minu~es, a precipitate is formed. The preci-pitate is iso~ated b~ centrifugation a~ 20~000 rpm at 4C for 20 :~205399 minutes. The supernatant is discarded (no interferon activity), and the precipitate is redissolved in 4 ml of 8 M urea and transferred to a concentra-tion cell known by the trade mark of MILLIPORE, size 8 ml, filter 10,000 molecular weight cut, and concentrated to 100JU1 at room temperature. There-after, additional 4 ml M urea (p.a.) was added to the concentrate, and the solution was concentrated to 100 pl at room temperature. 1 - 3 ml of distilled water was added, and the solution was concentrated again to a volume of 20~ul and mixed with 20 ~uL SDS sample electrophoresis buffer. 20 ~1 of the resulting solution was used for characterization as described in the section "SDS PAGE"
below.
The above-mentioned antibody affinity chromatography column had been prepared in accordance with "Binding Procedures" using non~monospecific anti-PIF which had been absorbed as follows: a total amount of 10 IFU-NU of anti-interferon immunoglobulins (corresponding to 4 ml sheep anti-interferon serum) was absorbed three times on a 150 ml column of human serum bound to SEPHAROSE
4B followed by 4 absorptions on a ClF-epoxy SEPHAROSE column and 2 absorptions on a CIF CH-activated SEPHAROSE 4B as described in the below section "Absorp-tion of Anti-interferon" and in Scand. J. Immunol. 8, 429 - 436 (1978). Finally, the immunoglobulins had been absorbed on a poly-L-lysine-SEPHAROSE column (once) and on a Soyabean Trypsin Inhibitor-SEPHAROSE column (twice).
The eluate from the Blue DEXTRAN chromatography of Namalva interferon was divided in two portions. One portion was used for SDS PAGE electrophoresis as described below. 250,000 IFU were loaded to the absorbed antibody column as shown in Fig. 8. No interferon was found in the wash. The interferon was eluted as usual by lowering pH to 2.4,and 235,000 IFU (collected in the presence of 0.1% SDS) were recovered. This eluate was concentrated as described above and further examined in SDS PAGE.

S3t39 SDS PAGE The SDS PAGE electrophoresis was performed as described under "MATERIALS AND METHODS" above. The stained slab of the electrophoresis of the pure human leukocyte interferon proteins is shown in Fig. 1. Fig. 2 shows, schematically, the - 56 a -~L,ZG!5399 stained s]ab from another experiment, together with the corrc--sponding interferon activity eluted from an unstained parallel gel strip. The striking reproducibility between the t~o experiments appears from the two Figures, the difference bet~veen 20,100 and 20,180 being within the e~;peri~nental accuracy. As mentioned pre-viously, the biological peaks coincide exactl~7 with the proteins.

From Fig. 1, it appears that the inter.feron preparation is com-p]etely pure by SDS PAGE. There is no other protein band whatsoever visible.

Fig. 9 sho~vs the stained slab from the SDS PAGE (load 0 . 9x106 I~U), of the pure Namalva interferon proteins (A), and of the eluate from the Blue De~itran column (B ) . By comparison with Fig .
1, it ~Yill be noted that the bands of the pure Namalva interferon are essentially identical with the bands of pure human leukocyte interferon applied in the same amount.

2G Establishment of H~,rbridoma Cells with Activity Directed Against Interferon .

3 female Balb/c mice, age t~o months, ~vere immunized with human leukocyte interferol~ in the follo~ing w ay:

The first injection (40,000 I~ vas perforn ed subcutaneousl~r in - the back of each mouse. The immunization was continued ever~T
week tvith subcutaneous injection of 70,000 IFU. The ]ast injection ~vas given intravenousl~ the 9th ~veek (mouse 1) and 10th week (mouses 2 and 3), resp-ectivel~.

The development of anti-intel feron ~vas determined on serum samples from the mice, using the interferon neutrallzaLion test. As a laborator~
check of tile intel-fel on neutl-alization test s~stem, an internal anti-interferon IgG preparation (raised b~ in jecting rabbils ~- ith part;.~ purified human leul;oc~le interferon preparations) ~vas, as usuall~T, included. I'he scrum samp]es from the mice sho~ed no anLi-intel feron activitt- the firs~ si~; ~veeks. Thel eafter, distinct ~h;r anti-interfel-on activit~ ~vas found:
_ 57 --' lZQ539g Tab]e ]. IFU-NU per ml 7th ~eek 3th ~-eel; 9th ~eek 10th ~veek mouse 1 160 160 120 mouse 2 200 1280 2500 1200-2500 mouse 3 80 40 40 5-10 The mice were killed by breaking their necks two to four days after the last injection, and their spleens ~vere removed under sterile conditions. After homogenization of each spleen in PBS, the homogenized cell suspension ~t as transferred to centrifuge tubes and centrifugated for 5 minutes at 170 g at 4~C. The cells ~rere resuspended in PBS~ and after a second centrifugation,,they were resuspended in serum-free DMEM ( 0 5 ml per spleen). The total amount of cells -as 108 (mouse 1), 0.8 x 10~ (mouse 2), and O . 8 ~; 103 (mouse 3) The viabilit~ t~ as around 85 - 90%.

B~ treatment with polyeth~rlene gl~col in the manner described below, the spleen cell suspension from each mouse ~as fused with 107 ~G3Ag8 (HPRT-) myeloma cells in the following manner: 108 mouse spleen cells and 10 8-azaguanin-resistent myeloma cells (~;63Ag8; NSI/lAg 4-1; SP 2/0-A~ 14) were mi~ed in a 50 ml conical plastic centrifuge tube (known by the trade mark FALCON 2070).
The tube was filled up with serum-free D~IEN and centrifuged for 10 minutes at 170 - 200 g and 4C. The supernatant was carefully removed, and at 37~C, a total of 0.7 ml of 50% polyethylene glycol solution having a temperature of 37C ~- as added drop~Yise over a period of 1 minute -tith gentle r otation After incubation for 90 seconds at 37~C, 15 ml of ~-~arm serun~-free Di~Ei~ ere added ver,y slot~rl~ (in the course of 1 ^ 2 minutes). Thereafter, the mi.~;ture was centri-fugated for lO minutes at 200 ~ and the cell pellet -as resuspen-ded in 50 m] comr~ete D~EN-FCS for seeding in t~ays, known by the trade mark COSTAR

~ , .

~2~539g From each of the fusions, ~8 cultures, each of 1 ml, were seeded in C0STAR trays (2 trays, each with 2a~ holes per spleen = 43 cultures per mouse). After 10 - 15 days, gro~Yth was noted in 21 cultures (mouse 1), 0 cultures (mouse 2) and (after further seeding out) 150 cultures (mouse 3).

The cells ~vere transferred to 5 ml cultures in 25 ml NUNC bottles which, lil~e the C0sTAR trays, contained a "feeder layer" of macro-phages. On shift of medium, the supernatants were obtained, and from these dense cultuI~es, cells were frozen in liquid nitrogen.

The supernatants of the individual cultures from mouse 1 were subjected to detection of positive clones using the interferon neutralization test. In this manner, one positive clone was found, although with a ver~r 10W titer ( 2 - 3 IFU-NU per ml).

Production of ~nti-Interferon b~r Means of Pure Interferon Proteins ~Pure by SDS PAGE).
The eluate from the above-described tandem affinit~r chromatograph~r, as characterized by SDS PAGE, was used for immunization of rabbits as follows:

" l, 000, 000 IFU units were concentrated to l ml and dial~rzed against PBS at 10C overnight. Two rabbits were injected - subcutaneousl~r ~vith each l,000,000 IFU prepared in this manner.
The injection was repeated each second week. The development of antibodies appears from Table II:

_ 59 --'X

lZa~53~9 TABLE II:

NEUTRALIZATION UNITS (IFU-NU) Freun d's adjuvant Rabbit Ixx~ 1 3 5 7 9 11 13 15 xxx) 0 0 2 100 2000 ND~) 20,000 20,000 _ _ . . _ . _ . .

Freund's adjuvant Rabbit I xx) 17 19 21 23 25 27 ~x~) 20,000 ~00,000 600,000 500,000 600,000600,000 Freund's adjuvant ~7 .
Rabbit II xx) 1 3 5 7 9 11 13 15 xxx) 0 0 0 20 500 NDX)20,000 10,000 Freund's adjuvant . . _ Rabbit II xx) 17 ~ - l9 21 23 25 27 x~x) 8000 ]00,000150,000200,000 lS0,000 1~5,000 x) not dctermined x~ veek x~:;) antiinterfel on titers (IFU-NU/ml) _ 60 -`` 3LZ,OS3':39 Production of Anti-lnterferon b~ eans of Pure Stained Interferon . .
Proteins Cut Out From an SDS PAGE

- The immunization - as performed according to the protocol e~;plained hereinbefore, using the minced interfer~n-containing (and partial~r washed and destained) gel suspension directly as the immunogenic preparation One rabbit (III) was immunized ~ith the 18~400 ~200 Daltons species, and another rabbit (IV) was immunized with the 20,100 ~200 Daltons species (died after week 15) Good results were obtained, vide Table III

TABLE III

NEUTRALIZATION UNITS

Freun dl s adjuvant I

Rabbit III week 1 3 5 7 9 11~ 15 (18,400 Dal-tons species) IFU-NU/ml 0 0 0 1-2 2 2 200 Rabbit IV week 1 3 5 7 9 11 15 (20, 100 Dal-- tons species) IFIl-NU/ml O O O O 1 2 200 Antigenicity of the 18,400 Da]tons Species ~Tersus the 20,100 Daltons Species and Vice Versa In order to show that the antigenic determinants of the above-mentioned t~o species are identical, the follo~-~ing cross-neutralization e~periments w ere performed ~Z(~53~9 Interferon protein was eluted from the 18,400 +200 Daltons species SDS PAGE band and the 20,100 +200 Daltons species SDS PAGE
band in the manner described above, and solutions containing 5 ~
lO IFU of the t-Yo species were prepared $olutions of anti-inter-feron from the two species, prepared in rabbits as described above, were diluted to contain 20 IFU-NU in total/ml Aliquots (1 ml) of pure interferon species containing lO IFU of the 18,400 Daltons interferon species and 10 IFU of the 20,100 Daltons interferon species, respectively, were mixed with 1 ml solution of the 18,400 Daltons species anti-interferon and l ml solution of the 20,100 Daltons species anti-interferon, respectively, in all possible permutations, that is, the anti-interferon of each species was mi~;ed ~vith the interferon of both species separately After 1 hour at 37C, any remaining interferon activity was determined b~r performing the usual interferon titration (vide "Materials and Methods" above) No interferon activity was found Thus, when mixing the 18,400 +200 Daltons species and the 20,100 ~200 Daltons species, respectively, with each of the anti-18,400 *200 Daltons species and the anti-20,100 +200 Daltons species, separately, and vice versa, no interferon was detected, in other words, a complete neutralization had occurred Therefore, it can be concluded that the t~vo interferon species e~:hibit the same antigenic determinants This implies that the anti-18,400 +200 Daltons species ~ill be useable as a monospecific antibody for purification of both ]nter-feron species, and the same applies for the anti-20,100 +200 Daltons species, and for a mi~;ture of the two species Further e~periments - performed in the same manner sho~ved that each of the si~ biological peal~s was completely neutralized b~ each of the antisera raised against the t~o major species , It is highly likel~ that the two major species isolated from the Namalva SDS PAGE ~vill give the same result, in other ~vords, that the~ also Cl oss-react and show identity to HuLeIf in terms of anti-genicity (~uLeIF 18,400 +200 being identical to Namalva 1~,400 ~200, both ~vith respect to antigenicit~ and molecular ~veight, and HuLeIF 20,100 +200 being similarly identical ~vith Namalva 20,100 +200) .i !..

~1 /Z(~ 5 3 ~g ical ~ffects of the rure Interferon Proteins.

Antiviral activity.

The antiviral activity of each of the six stained protein bands shown in Fig. 3 was determined. The gel was loaded in two slots, both of ~vhich were stained. The stained bands in one of the slots is shown at A in Fig. 3. The other slot slot was then briefly destained (in 50% methanol, 45% H2O2, 5% acetic acid), the exact location of the interferon proteins in the wet gel was recorded, and the gel was rinsed in water and was thereafter sliced as sho~vn at B in Fig. 3. The number of gel slices is indicated at C
in Fig. 3. In this manner, each interferon protein band was exactly cut out of the gel, without being mixed ~vith the adjacent one. Each slice was eluted in the same manner as desc`ribed in the section "Materials and Methods", and the biological profile shown in Fig. 3 was constructed using the usual interferon titration described in "Material and Methods". The neutralizing activity of each of the six species cut out and eluted from the SDS PAGE was checked against anti-leukocyte interferon, and it was found that all of the species were completely neutralized by the same anti-serum. The recovery of ;nterferon in Fig. 3 ~vas rather low (20%) compared ~vith normal "SDS PAGE elution" without pre-staining (e~icept for the 18,400 +200 Daltons species), which indicates that the biological activity of most of the interferon species was select-ively destro~ed compared ~vith the antigenicity. In the neutrali-- zation tests against anti-leukocyte interreron, the interferon proteins l'eluted from Fig. 3" ~vere able to neutralize the anti-leukocyte interferon 3 - 5 times more effectively than nalive (crude) human leukocyte interferon, calculated on interferon activity basis, ~vhich indicates a selective destruction of determinants responsible for the biological activity.

Non-antiviral effects.
-The non-antiviral effects of the pure human leukocyte interferon species were checked in 3 systems:

539~

1) Anti-cellu]ar activity.

The anticellular activity of the pure interferon proteins was in-vesLigated by incubating Daudi cells with 1 1000 dilutions (in medium) of pure interferon proteins obtained from the eluted SDS PAGE fractions as sho~vn in Fig. 2, by ascertaining the relative depression of Tritium ]abelled Th~midine (I. Heron and I~. Berg, The actions of interferon are potentiated at elevated temperature, Nature, 27~, 508 - 510 (1978)) compared to controls ~ithout interferon (Fig. 2, upper part, where "% G-I" designates % growth inhibition). As can be clearly seen, `the "anticellular curve" follows the antiviral curve very strictly.
This proves that all the five species of pllre native human leukocyte interferon contain both the antiviral activity and the anticellular activity. The peak size of the different "anticellular peaks" does not vary linearly Ivith the correspondin~ size of the "interferon peaks", which probably reflects the sensiti~ity of the Daudi cell system (J. Hilfenhaus, H. Damm, H.E. Karges and I~.E. Manthey, Gro~th inhibition of human lymphoblastoid Daudi cells in vitro by interferon preparations, Arch. ~7irol. 51, 87 - 97 (1976~. The small interferon peak at 19,500 Daltons gave no rise to a corresponding peak in the anticellular curve. At a 10-fold lo~ver di~ution (1:100), ho~vever, a small but dist.inct peak of anticellular activity was also observed (not shown).

2) The e~pression of major histocompatibility antigen (MHC) on lym-phocytes and monocytes.
.
The selective increase in ~2-associated l\qHC (major histocompatibility antigen) e~;pressioll was observed using partially purified human Ieul;ocyte interreron, e.g- described by I. Heron, M. Hokland & li. Berg (1978), "Enhanced e~pression of ~2 microglobulin and l-ILA on human lympl1oid cells by interferon", Proc. Natl. Acad.
Sci. 75: 6215 - G222 (referred to belo~v as PNAS 75). Each of the t~vo major human leulcocyte interferon species (18,~00 and 20,100 Da]tons, vide Fig. 1), ~vas assayed in doses OI' 100 IF11 per ml culture medium. The abo~e-mentioned effect ~vas found using these pure molecular species, ~vhereas e]uates from gel slices outside the regions ~vhel e antiviral activity ~-as recol-ded had no effect. It has, 3L;Z(~53'3~

thus, been proved that the effect of selective enhancement of MHC antigen expression on lymphoid cells is an inherent feature of the interferon molecules.
3) The potentiation of the Natural Killer cell system (NK system).
Fig. 10 shows the antiviral profile (as assessed on an SDS PAGE
in the same manner as described in connection with Fig. 2). Each of the species from the gel was assessed for NK enhancing activity, using the method described in PNAS 75. Fractions that have antiviral activity as shown in the lower curve gave increased NK, such as illustrated in the upper curve, whereas "base line" fractions did not. One arrow indicates only saline added as a negative control, and two arrows indicate partially purified human leukocyte interferon (PIF) used as a positive control.
100 IFU antiviral units of each interferon preparation was added per ml.
ABSORPTION OF ANTI-INTERFEP~ON.
In accordance with one aspect of the invention provided by the above identified parent application, the absorption of anti-interferon by means of matrix-immobilized crude human non-fibroblast interferon con-stitutes a method of broad applicability for establishing highly specific anti-interferon which, when immobilized on a matrix, is an extremely valuable tool for purifying interferon to a high degree. In the experi-ments reported below, the features of this aspect of the invention provided by the above identified parent application are illustrated in greater detail:
A column of CIF-epoxy-SEPHAROSE was constructed using 20 ml of CIF with a titre of 300,000 IFU/ml according to the procedure described by the manufacturer (Pharmacia), using 0.1 M NaHCO3/0.3 M NaCl as a coupling buffer. 55% of the proteins were coupled. 0.5 ml of anti-J~z~s3g~

interferon (150,000 lnterferon-neutralizing units (IFu-Nu/ml) was applied to the column, and the qualitative results were similar to those obtained with the CIF-CH-activated SEPHAROSE column below: almost all the anti-interferon appeared in the wash; 5000 IFU-NU were found in the first eluate (pH 2.4).
A column of crude concentrated human leukocyte interferon (CIF) on activated Ch-SEPHAROSE 4B was constructed using 5 ml CIF in the manner described by the manufacturer (Pharmacia), using 0.1 M NaHCO3/0.1 M NaCl as coupling buffer. 53% of the proteins were coupled. The column was loaded with 1 ml anti-interferon with a titre of 150,000 IFU-NU per ml, using as loading buffer 0.1 M NaOAc/0.3 M NaCl, pH 7Ø The major part of the anti-interferon activity passed through the column (more than 95%).
The column was eluted with 0.1 M ~IOAc/0.3 M NaCl, pH 2.4. 5 mg of immuno-globulins were removed in this eluate together with 10,000 IFU-NU. There-after, it was tried to elute with 4 M urea in PBS including 0.1% ethanola-mine (pH 6.0), and this eluted more proteins together with 3000 - 4000 IFU-NU, but the capacity of the column was partially destroyed by urea, and elution with urea was therefore abandoned, and a new column was made.
The CIF used in the construction of the two above-mentioned columns was prepared by concentrating crude interferon made as described under MATERIALS AND METHODS" above 10 - 20 times by means of 0.5 M KSCN
at pH 3.5 followed by a dialysis vs. coupling buffer.
Instead of CIF it would be within the scope of other aspects of the invention provided by the above identified parent application to use "washes" from "interferon antibody affinity chromatographies" in particu-lar interferon antibody affinity chromatographies performed using the monospecific antibodies of aspects of the invention provided by the above :lZ053~5~

identified parent application. Also, all the impurities could be pooled together from several interferon antibody affinity chromatographies, and these impurities could be used instead of or combined with CIF.
Absorption of Anti-interferon. 2 ml of anti-interferon, containing 450,000 IFU-NU/ml and 100 mg immunoglobulins, were first loaded on the CIF-epoxy-SEPHAROSE column whereby 30~ of the immunoglobulins were removed together with 25,000 IFU-NU of the anti-interferon activity ti.e., 2.7% of the input). The wash was concentrated, dialysed vs. loading buffer (same as described above) and reapplied to the equilibrated column; this time only 3.5 mg of immunoglobulins were removed, while the same amount of anti-interferon was retained. The antibodies were passed through the column eight times.
40% of the immunoglobulins had been re ved at this stage together with 15% of the anti-interferon activity (which could be re-covered in the eluates). This absorbed anti-interferon preparation was then applied to the CIF-CH-activated SEPHAROSE column, and 1600,ug were removed together with 35,000 IFU-NU. The procedure was repeated three times. The last wash had a total content of 320,000 IFU-NU and 22 mg proteins~ This purified anti-interferon was absorbed on a poly-L-lysine and soya bean trypsin Inhibitor column as described above and was then coupled covalently to SEPHAROSE 4B in accordance with "Binding Procedures", binding 85% of the immunoglobulins. In addition, two other sheep-anti-interferon batches (titres 120,000 IFU-NU/ml and 100,000 IFU-NU/ml) were also absorbed with similar results. These sera were absorbed only eight times in total, and above 35% of the immunoglobulins were removed together Wit'l 10 - 15% of the anti-interferon activity. Rabbit anti interferon . ~ .

~S3~

serum was also absorbed. Since the titres thereof were rather low (10,000 and 25,000 IFU-NU/ml), the batches were first concentrated 10-fold and 5-fold, respectively. After five absorptions, 45% of the immunoglobulins were removed together with 25% of the anti-interferon activity. A11 anti-interferon preparations were made as described in the above section "Non-monospecific anti interferon".
Purification of Various Interferons by Antibody Affinity Chromatography Using Absorbed Anti-interferon.
A "contorl" column (not according to aspects of the invention provided by the above identified parent application was constructed with unabsorbed anti-interferon and loaded with 2.5 ml of CIF containing in total 1.2 x 10 IFU. The wash contained 5.5 x 10 IFU/20 ml, and the eluate contained 3 x 10 IFU/10 ml. There was an obvious discrepancy between the protein peak and the interferon peak in the eluate. A great amount of impurity was eluted under very mild elution conditions (pH 4.5).
The quantitative data of the experiment appear from the below Table IV.
The specific activity of the protein in the eluate was only 0.9 x 10 IFU/
mg protein.
In accordance with aspects of the invention provided by the above identified parent application, the SEPHAROSE ~B column coupled with the purified anti-interferon as described in the above section llAbsorption of Anti-interferon" was loaded with 5 ml CIF (crude concentrated human leukocyte interferon). The input contained 3 x 106 IFU in total, the wash contained 3 x 10 IFU/20 r~, and the eluate contained 4 x 10 IFU/6 ml. Like in the control experiment described above, the fraction size was 2 ml, and the loading and elution buffers were the same as mentioned above, i.e., loading buffer 0.1 M NaOAc/0.3 M NaCl, pH 7.0, elution l~S3~

buffer 0.1 M HOAc/0.3 M NaCl, pH 2,4. The specific activity in the eluted material was found to be 43 x 10 IFU/mg.
Crude Namalva interferon was also purified up to the same level as normal crude leukocyte interferon, using the same column (Table IV, Experiments 5 and 6).
Human fibroblast interferon was purified (Table IV, Experiment 7) in one step up to 2 x 10 IFU/mg protein (peak fraction), with a recovery of 90~. On a subsequent SDS-PAGE, it was revealed that the preparation contained 4 bands with a faint band barely visible at the interferon region.

Partially purified human interferon (PIF) with an initial specific activity close to 1 x 10 IFU/mg protein was loaded to the column. The results are shown in Table V: Only 3000 IFU were found in the wash, which corresponds to less than 0.25%; the recovery obtained was 80% based on the eluate pool. The highest specific activity was 10 IFU/
mg protein (fraction 25, Table V).

_ 69 -~53~

. ~ U) Q, O ~ ~ O O 0~ ~ O o O (J_ , ,~ O
v c~-- ~ m ~D

E~ ~ ~ . v ~ C
>~ O .!~ O ~ ~ ~ - ~ 1-- t ) U~ . ~ C U
. . U ~ ~_ , " ,, , . ,, .~ ~ c . ~ ~ . ' o~ . ~ ~ W G) Q) ~ . ~D ~~) ~ h . O 4 0 - X o ~ . O O IJ
~ . ~ ~V ~ ~ o~ ~ o ~, . o ~ ~ ~ . . m m z H ~:

~ ~ . . ,, ~ ~S ~ .
H '. 0 t ~ J~ H ~ -1 ~ N O r-l . . ~ '' .
.~) ooooooo' ., ~ V --. X X X X X X X . .

~ ~ . t)J~_ , ~

~,, ~ C a . . ' x x x . . .
" U~ O~ ~ ~ 0 0 C~

~O E ~1 O U~ o. ~1 E ~ ~
W . . . C h . ~ ~ C
.,~ . _ . . ,~ a) o Q) O
V _l O ~ ~ rd h C~ V
. - ~: L O X ~ r~ 1 ~ ~ ~ O ra ;1 O U) J O ~) CVJ ~ Q ~ ll) O S
11 )~ ~ U~ C ~ O C V
Cl ' ~ ' ( I U U --I ~( ~ ~ ~ O
~ aJ ~ a) QJ 0 a ~ S c ~. ~ S ~ ~, .~v C v r~ O O
F CJ C C Q)~ C

O 0 1~ W ~ O C ~ --Q. o X ~ ,,,, ~,. L,-, ~ r` ~- ~ O ~

~ --7 0 ~Q1~3g9 ., Table V. J

Purification of par.ially purified human inter-feron (PIF~ . . . ................. .. ...... ...

IFU Vol.... Total Spec. act. . p~
. (~.1) proteins (IFU/mg protein) p (/ug) .. ... -:.--- --- .

Input 1.5 x 10 0.9 1300 ~ 1.2 x 10 . 7 ~7as~ 300020 N.D.: N.D. . 7 l~o. 23 8 ~: 105 2 . 10 2.5 x 107, 4.7 No. 24 8 x 10 2 16 5 x 10 3.3 ~'o. 25 1.2 x 105. 2 6 . 1.3 x 108 2.7 ~o. 26 .1.2 x 106. 2 . <2 >6 x 107 2.5 Eluate pool 1.9 x 10 10 35 3.4 x 10 N.D.
Eluate pooll 10 32 5.9 x 107 N.D.

E::periment No. 313 (PIF = partially purified human leucocyte interferon). - . . . ..
t Based on individual frac.ion : l7~D. = not detected. . ~
Ir'~ = inter.eron units in international re~erence .
units 69/19B. ...
.
-. .

Claims (22)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:

1. A process for producing antibodies, which comprises one of the following:
(I) (a) an immunizing an immunizable animal against immunological determinants of human Le form interferon protein or proteins and (b) obtaining antiserum from the animal;
or (II) (a) culturing a hybridoma cell clone producing antibodies directed against immunological determinants of human Le form interferon protein or proteins, and (b) recovering said antibodies from the culturing medium;
wherein said human Le form interferon protein or proteins are selected from:
(a) human Le form interferon protein or protein which, under sodium dodecyl-sulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9 x 106 IFU, show two major sharp strained protein bands having anti-viral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral inter-feron activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions; (b) Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 2.8 x 106 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons, the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions;
(c) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total inter-feron load of 0.9x106 IFU, show two major sharp stained protein bands hav-ing antiviral interferon activity at 18,400 and 20,100 Daltons, respec-tively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of +200 Daltons), said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions, but having a specific activity of at least 109 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electro-phorese staining; (d) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9x106 IFU, show two major stain protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2x106 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; (e) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8x106 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Dal-tons, respectively (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of at least 109 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; and (f) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8x106 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Daltons, respectively (said ?200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, said sodium dodecylsulfate poly-acrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2x109 IFU
per mg protein, as assessed by comparative sodium dodecylsulfate poly-acrylamide gel electrophorese staining.

2. A process as claimed in claim 1 wherein said human Le form inter-feron protein or proteins comprises the 18,400 +200 Daltons human Le form of interferon protein component of said protein or proteins, the 20,100 +200 Daltons human Le form of interferon protein compound of said protein or proteins, or a combination of the 18,400 +200 Daltons human Le form of inter-feron protein component of said protein or proteins and the 20,100 ?200 Daltons human Le form interferon protein component of said protein or pro-teins.

3. The process of claim 1 including the additional step of immobi-lizing said antibodies in a matrix.

4. The process of claim 3 including the step of covalently binding said antibodies to said matrix.

5. The process of claim 4 wherein said matrix is a cross-linked agarose.

6. The process of claim 3 including the further step of substantially freeing said antibodies from proteolytic enzyme activity.

7. The process of claim 6 wherein said further step is carried out by treatment with enzyme inhibitors or with enzyme destructors.

8. The process of claim 6 wherein said further step is carried out by treatment with a matrix-immobilized enzyme inhibitor or with a matrix-immobilized enzyme destructor.

9. The process of claim 3 including the step of passing said matrix-immobilized enzymes through at least one of:
(i) a column of matrix-immobilized poly-L-lysin;
(ii) a column of matrix-immobilized soyabean trypsin inhibitor;
and (iii) a column of matrix-immobilized kallikrein inactivator.

10. The process of claim 1 (I) wherein said immunizable animal is a pig, and wherein said antibodies are pig IgG immunoglobulins.

11. Antibodies raised against, or directed substantially only against, immunological determinants of human Le form interferon protein or proteins said human Le form interferon protein or proteins being selected from: (a) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total inter-feron load of 0.9 x 106 IFU, show two major sharp stained protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respectively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), said sodium dodecylsulfate polyacrylamide gel electrophorese acryl-amide gradient showing essentially no other stained protein regions; (b) human Le form interferon protein or proteins which under sodium dodecyl-sulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8 x 106IFU, show six stained protein bands having antiviral inter-feron activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons, the peaks of antiviral interferon activity coin-ciding exactly with the stained protein bands, said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions;
(c) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total inter-feron load of 0.9x106 IFU, show two major sharp stained protein bands hav-ing antiviral interferon activity at 18,400 and 20,100 Daltons, respec-tively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), said sodium dodecylsulfate polyacrylamide gel electro-phorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of at least 109 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; (d) human Le form interferon proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 0.9x106 IFU, show two major sharp stained protein bands having antiviral interferon activity at 18,400 and 20,100 Daltons, respec-tively, and a minor stained protein band having antiviral interferon activity between 20,300 and 20,400 Daltons, together with small peaks of antiviral interferon activity at 19,500, 21,130 and 23,440 Daltons (said Dalton molecular weights being subject to an experimental accuracy of ?200 Daltons), said sodium dodecylsulfate polyacrylamide gel electro-phorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2x109 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamide gel electro-phorese staining; (e) human Le form interferon protein or proteins which, under sodium dodecylsulfate polyacrylamide gel electrophorese staining at a total interferon load of 3.8x106 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Daltons, respectively (said Dalton molecular weights being subject to an experimen-tal accuracy of ?200 Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, said sodium dodecyl-sulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of at least 109 IFU per mg protein as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining; and (f) human Le form interferon proteins which, under sodium dodecylsulfate polyacryla-mide gel electrophorese staining at a total interferon load of 3.8x106 IFU, show six stained protein bands having antiviral interferon activity, viz. strong bands at 18,410 Daltons and 20,180 Daltons, respectively, a medium-strong band at 20,420 Daltons and just visible bands at 19,500 Daltons, 21,130 Daltons, and 23,440 Daltons, respectively (said ?Daltons), the peaks of antiviral interferon activity coinciding exactly with the stained protein bands, said sodium dodecylsulfate polyacrylamide gel electrophorese acrylamide gradient showing essentially no other stained protein regions but having a specific activity of 2x109 IFU per mg protein, as assessed by comparative sodium dodecylsulfate polyacrylamide gel electrophorese staining.

12. Antibodies as claimed in claim 11 wherein said human Le form interferon protein or proteins comprises the 18,400 ?200 Daltons human Le form of interferon protein or proteins component of said protein; the 20,100 ?200 Daltons human Le form of interferon protein component of said protein or proteins; or a combination of the 18,400 ?200 Daltons human Le form of interferon protein component of said protein or proteins and the 20,100 ?200 Daltons human Le form interferon protein component of said protein or proteins.

13. Antibodies as claimed in claim 11 wherein said human Le form interferon protein or proteins comprises human Le form interferon protein or proteins obtained from the respective band or bands cut from sodium dodecylsulfate polyacrylamide gel electrophorese.

14. Antibodies as claimed in claim 13 wherein said bands were cut from said sodium dodecylsulfate polyacrylamide gel electrophorese gel after staining of said gel and a short wash in distilled water.

15. Antibodies as claimed in claim 11 (or fragments or derivatives thereof retaining the essential antiinterferon determinants) immobilized on a matrix.

16. Matrix-immobilized antibodies as claimed in claim 15 covalently bound to the matrix.

17. Matrix-immobilized antibodies as claimed in claim 15 in which the matrix is a cross-linked agarose.

18. Matrix-immobilized antibodies as claimed in claim 17 in which said matrix is selected from the group consisting of CNBr-activated cross-linked agarose, CH-activated cross-linked agarose, and epoxy-activated cross-linked agarose.

19. Matrix-immobilized antibodies as claimed in claim 15 which are substantially free from proteolytic enzymatic activity.

20. Matrix-immobilized antibodies as claimed in claim 19 which have been substantially freed from any proteolytic enzymatic activity by treat-ment with enzyme inhibitors or with enzyme destructors.

21. Matrix-immobilized antibodies as claimed in claim 19 which have been substantially freed from any proteolytic enzymatic activity by treat-ment with a matrix-immobilized anzyme inhibitor or with a matrix-immobilized enzyme destructor.

22. Covalently bound, matrix-immobilized antibodies of claim 11 said antibodies further having been passed through a column of at least one of (i) matrix-immobilized poly-L-lysin; (ii) matrix-immobilized soyabean trypsin inhibitor; and (iii) matrix-immobilized kallikrein inactivator.