CA1256795A - Anti-idiotype antibodies induced by synthetic polypeptides - Google Patents

Abstract

ABSTRACT OF THE DISCLOSURE
Chemically synthesized polypeptides having amino acid residue sequences that substantially correspond to the amino acid residue sequences of particular variable or hypervariable regions of immunoglobulins, when administered alone or as polymers or as conjugates bound to carriers, induce the production of anti-idiotype antibodies of predetermined specificities.

Description

' ' ' ' ~,~j~9 ANTI-IDIOTYPE ANTIBODIES
INDUCED BY SYNTHETIC POLYP~PTIDES
Description Technical Fi~ld The present invention relates to chemically synthe~ized polypeptides having amino acid residue sequences that substantially orrespond to thP amino acid residue sequenc~s of particular variable or hypervariable regions of imnnunoglobulins, and which, when administered alone or as polymers or as conjugates bound to carriers, induce the production of anti-idiotype antibodies of predetermined specificities.

An antibody is an immunoglobulin molecule that has a specific amino acid .equence and thus binds only with the antigen that induced it~
~ynthe~is (it~ immun4gen) or with a closely related antiyen ~or immunoyen)~ Immunoglobulin molecules include two kinds o polypeptide chains. Each molecule consist~ o larger identical polypeptide chalns referred to as heavy chains lH chains) and wo identical smaller ones referred to as light chains IL
chainsj. These polypeptide chains are h~ld together by disulfide bonds and by noncovalent bonds, which are primarily hydrophobic~ ThP heavy and light polypeptide chains are synthesized ln Vlvo on separate ribosomes, assembled in the cell, and are secreted as an intact immunoglobulin molecule.
The understanding.of the s~ructure and function of immunoglobulins ha~ been facilitated by ~tudies of fragment~ produced by enzymatic cleavage of the antibody molecule. For example, treatment of an antibody molecule with the enzyme papain produces 3S two antigen-binding fragments (designa~ed "Fab") and ~ ".

~ ~ ~r~r a complement-binding fragment ~designated "Fc") 7 which contains no antigen-binding capability but determines important biological ~haracteristics of the intact immunoglobulin molecule. Treatment of an antibody molecule with the enzyme pepsin on ~he other hand, produces a single antigen~binding fragment [designated "F(ab')2"] and a somewhat smaller complement-binding fragment (also designated "Fc").
The constant region of H chains allows the differentiation of immunoglobulins into classes and subclasses and confers certain biological properties such as the ability to activate complement, to cross the placenta, and to bind tv polymorphonuclear leukocytes or macrophagesO
In particular, five immunoglobulin classes tIYG, I~A, IgM, IgD, and IgE) are r~cognized on the basis o~ structural dif~erences of their heavy chains including the am.~no acid sequence and length of the polypeptide chain. The antigenic determinants on the heavy chains also permit the identification and quantitation of the immunoglobulin classes by immunochemical assay techniques.
The amino-terminal half of the light chains and the amino-terminal quarter of the heavy chains of an immunoglobulin molecule vary in their amino acid sequence and are termed the variable regions (V regions) of the po~ypeptide chains. Portions of the V region of one heavy and of one light polypeptide chain constitute the site for antigen binding. A considerable vari~tion of the amino acid sequence of the variable region of an immunoglobulin molecule can exist which produces the many difff?rent antibody specificities. A region of extreme variability in the primary se~uence within a variable 67~3~

region is called a hypervariable region. Capra e~
al., Proc. Natl~ Acad. Sci. USA, 71, 845 (1974).
The specificity of the molecular binding site of an antibody is termed its idiotype. The term idiotype denotes the unique variable (V) region sequences produced by each t~pe of antibody-iorming cell. An antibody having a binding site sp~cificity for the binding site of another antibody is termed an anti-idiotype antibody~
The same amino acid sequence variation that produces the antigen binding specificity of an immunoglobulin also determines which idiotypic determinants are present. Thus, particular idiotypes are almost invari~bly associat0d with immunoglobulins of a par~icular specificity~ As such, idiotypes can serve as antigenic markers for immunoglobulins with a particular specificity and, by virtue o~ their surace imrnunoglobulin, ~ lymphocytes o~ the s~me ~pecificlty~
The term~ "cross-reactive" and "cro~s-reActivity" refer to the ability of an antibody to bind antigens other than its idio-specific antigen. Cross-reactive anti-idiotype antibodies can be divided into two major groups. One group comprises those anti-idiotype antibodies that recognize idiotypic antigenic determinants that are associated with specific amino acid sequences in the he~vy and light chain variable regions.
Anti-idiotype antibodies of this group often reflect the action of inherited immunoglobulin structural genes, Conse~uently, these antibodies do not cross-react in subjects that are not genetically similar.
The second group includes anti-idiotype antibodies that are cross-reactive to the internal i~$~

image of the antigen. This type of anti-idiotypic antibody is elicited by immuni~ation with an intact immunoglobulin and usually recognizes idiotypic antigenic determinants as a result of a particular quaternary interaction of the light and heavy chains. The antigenic site recognized by this group of anti idiotype antibodies, however, is not associated with a particular ligh~ or heavy chain amino acid sequence~ Because the antibody binding site bears the internal image of the antigen; i.e., mimics the size, shape, charge and/or van der Waals attraction the antigen, this group of anti-idiotype antibody binds to many different antibodies of the same specificity. The idiotypes recognized by such - 15 antibodies can be produced by individuals with different genetic backgrounds and are controlled by genes that bear no ~peclal rela~ionship.
~ nti-idiotype immunotherapy can be very u~eful in the treatment of autoimmune disease, by neutralizing pathological auto-antibodie~. Moreover, anti-idiotye immunotherapy can be used in the treatment of certain B cell malignancies.
Current therapeutic techniques in this area include chemotherapy, active immunizatio~ with non-specific stimulants of the immune system and pASSive immunization with antibodies directed against specific markers including idiotypes. The major disadvantage of the first two techniques is the lack of specificity which, in the case of chemotherapy, usually results in general tissue toxicity~
On the other hand, anti-idiotypic therapy can be highly specific~ But such therapy suffers from the disadvantages associated with passive administration since the anti-idiotype antibodies must be produced in a non-human species. Therefore, ~25~$

a signiicant possibility exists that an individual so treated will develop an immune response against the passively admi~istered antibodies, which response can negate any potential therapeutic effect. This i5 par~icularly true because the antibsdies must be administered many times to produce the desired result.
Moreover, all anti-idiotype antibodies have previously been generated by immunizing the host with the target immunoglobulin. The resulting polyclonal antisera must then be extensively purified to produce antibodies having the desired anti-idiotype specificity. The selected, purified, "monoclonal"
antibodies must then be carefully tes~ed to determine their specificities.
The structural coLrelakes of idiotypes have been ~ought in ~everal well~defined antibody systems~ Se~ Kunkel et al., Science, 140, l218 (l~63); Capra et aL.I P c. _ _ Acad Sci. USA, 71, 4032 (l974); Weigert et al., ~ , 139, 137 (l974) Klapper e~ al., Ann. Imrnunol. (Inst.
Pasteur), l27C, 261 ~1976~; Schilling et al., Nature, 2839 35 (1980); Capra et alu-, Immunol. Today/ 3, 332 (l982); and Capra et al., ~ , 4, 177 (1983). These studies suggest that a hypervariable region (also referred to as a csmplementarity-determining region or a CDR~ of an immunog].obulin is the structural correlate of an idiotypic determinant.
In particular, in the murine anti-dextran system, on~ private (or individual~ idiotype and one pu~1ic (or cross reactive) idiotype were assigned t9 the third and second hypervariable regions, respectively, of the heavy chain. Schilling et al.9 ~ 0 However, in most systems, it has proven extremely difficult to associate a particular $

idiotypic determinant with a specific amino acid sequence, Capra et al~, Immunol. Today, 4, su~a.
Rather, anti-idiotypic antibodies elicited by immunization with an intact immunoglobulin usually recognize determinants dependent upon a particular quaternary interaction or "internal image" of both of the light and heavy chains Capra et al., Id.
Lerner e~ al. have been successful in obtaining protection o~ animals by the use of vaccines against pathogens that utilize synthetic polypeptides having amino acid residue sequences of short to moderate length as immunogens. See Sutcliffe et al., Science, 219, 495 (1983). Such synthetic polypeptides induce antibodies specific for predefined determinants of intact proteins~
As described hereinl synthetic polypeptide technology can avoid the previously described dif~iculties associated with conventional an~i-idiotype therapy. ~ccordin~ to the present invelltion, described in detail herelna~ter, polypeptide6 having relatively short amino acid residue xequences that substantially correspond to the portion of the immunoglobulin primary sequence that forms the idiotype can be synthesized, coupled to an appropriate carrier and inoculated into animal hosts including humans as immunogens to raise antibodies. 'The resulting antisera recognize the immunoglobulin(to a portion of which the polypeptide corresponds in amino acid residue sequence)and are idiotype specific~ Such antisera produced by synthetic polypeptides are thus of predetermin~d specificity and the necessity for extensive purification and specificity testing is eliminated.
In addition, such synthetic polypeptides alone~ as conjugates or as polymers can be ~S67~

administered to individuals to raise antibodies that immunoreact with the particular idiotypes of that individual. Autologous anti-idiotype antibodies are well documented and are widely believed to be very important in immunoregulationD One advantage in the use of synthetic polypeptide-rontaining antigens (immunogens) is that antibodies reactive with otherwise non-immunogenic determinants can be elicitedO Therefore, appropriate synthetic polypeptides can induce anti-idiotype antibodies in an individual that are directed against a particular idiotype of that individual whereas this could not be achieved by immunizing with the intact immunoglobulin.
Thus, an individual can be actively immunized against a p~thological idiotype and the number of th~rapeutic interventions required can be sub~tantial`Ly reduced compared to conventiona~
lmmunization with An intact immunoglobulin. Also, the possibility of an immune response against the anti-idiotype antibodies can be reduced substantially.
Polypeptides can also be synthesized to mimic an an~igen under a~tack by pathological auto-antibodies. These polypeptides can block or inhibit the interaction between the antigen and the undesired auto-antibodies, thereby ~i~nificantly impeding the disease process.
It is believed that certain idiotypes occur very freguently in particular syndromes. Synthetic polypeptides, corresponding to such idiotypes may be used to elicit antibodies of predetermined specificity for such syndromes, and may then be applied in the diagnosis o that syndrome.

BrieF 5ummar~ of the Invention The present invention contemplates the use of synthetic polypeptides to mimic idiotypic antigenic determinants on immunoglobulin molecules and to elicit the production of antibodies (anti-idiotype antibodies) of predetermined specificity which are reactive with those idiotypes.
Such antibodies can be useful in the treat~ent o F
autoimmune disease and certain diseases of B-lymphocytes. Such antibodies can also be used in the diagnosis of disease where a particular idiotype occurs.
A synthetic polypeptide in accordance with this invention has an amino acid residue sequence of contiguous amino acids that substantially corresponds immunologically to the ~mino acid sequence o~ an idiotypic antigenic determinant; i~e., a varl~ble or hypervariable region, o~ a~ immunoglobulin. The polyp~ptide contains from about 6 to about ~0 amirlo acid residues and preferably from about 8 to about 20 amino acid residues. The polypeptide, when administered alone or as polymer or as a conjugate bound to a carrier such as keyhole limpet hemocyanin (KL~) or the like and in~roduced in an effective amount in a physiologically acc~ptable vehicle into a host animal, induces the production of anti-idiotypic antibodies in the host.
Throughout the application, the terms "peptide" and "polypeptidel' are used interchangeably. The te~m "synthetic polypeptide"
means a chemically derived chain of amino acid residues that is free o~ naturally occurring proteins and fragments thereof. Such synthetic polypeptides can elicit production of anti-idiotype antibodies in a ho~t.

:~.Z5i6~

The phrase 'limmunologically corresponds substantially" in its various grammatical forms is used herein and in the claims in relation to polypeptide sequencesto mean that the pol~peptide sequence described induces production of antibodies that bind to the polypeptide as well as to the idiotypic antigenic determinant.
The term "substantially correspon~s" in its various grammatical forms is used herein and in the claims in relation to polypeptide sequences to mean the Polypeptide sequence described plus or minus up to three amino acid residues at either or both of the amino- and carboxy-termini and containing only conserva~ive substitutions in particular amino acid residues along the polypeptide sequence.
The term "conservative substitution" as used above is meant to denote that one amino acid re~idue has heen replaced by another, biologically slmilar residue~ Examples o~ conservative substitutions include the substitution of one hydrophobic residue such as Ile, Val, Leu or Met for another, or the substitution of one polar residue for another such as between Arg and Lys, between Glu and Asp or between Gln and Asn, and the like~
In some instances, the replacement of an ionic residue by an oppositely charged ionic residue such as Asp by Lys has been termed conservative in the art in that those ionic groups are thought to merely provide solubility assistance. In general, however, since the replacements discussed herein are on relatively short synthetic polypeptide antigens, as comp~red to a whole protein, replacement of an ionic residue by another ionic residue of opposite charge is considered herein to be "radical replacement", as are replacements between nonionic ~2S~ $

and ionic residues, and bulky residues such as Phe, Tyr or Trp and less bulky residues such as Gly, Ile and Val~
The terms "nonionic" and "ionic" resid~es are used herein in their usual sense to mean those amino acid residues that normally either bear no charge or normally bear a charge, respectively, at physiological pH values. Exemplary nonionic residues include Thr and Gln, while exemplary ionic residues include Arg and Asp.
According to the method of the present invention, a suitabl~ host is treated with an ~ffective amount of a synthetic polypeptide in a physiologically acceptable vehicle, the polypeptide havi.ng an amino acid residue sequence that immunologically corresponds substantially to an amino acid sequence of a variable or hypervariabl~ region (an ~diotypic antigenic determinant) of an immunoylvbulln. ~nti-ldiotype antibodie~ are thus produced which carl bind to the variable or hypervariable region, respectively, of the immunoglobulin.
In this manner, khe resulting l'anti-antibody" or "anti-idiotype antibody" has a predetermined specificity and has substantially the same configuration as an antigen that binds to the variable or hypervariable region of the immunoglobulinO ~ntibodies of-this type provide an improved means for defining the structure of idiotypes, as well as providing means for diagnostics and therapy.
The method of the present invention produces antibodies ~ainst eynthetic polypeptides that mimic an idiotypic antigenic determinant of a naturally occuring protein and result in a l~rge fraction of j7~

the elicîted se~a being reactive against the natural and the denatured protein.
It remains unclear whether in a given anti-polypeptide serum the same antibody molecules are responsible for the interaction with both natiYe and denatured proteins or whether different antibody molecules in the antiserum react preferentially with one or the other of the two protein states. Whether same or different antibodies react with the denatured and native proteins may depend on the characteristics of the particular epitope or idiotope being considered.
Another aspect of this invention is that the method of producing antibodies against synthetic polypeptides can be used ~o raise antibodies against idiotypic antigenic determinants that are not naturally ~munogenic in the host. That is, certain por~ions of a macromolecule have the ability to be bound by an antlhody (i.e., are antiyenic) but clo not elicit the production of antibodies ti.e.~ are not immunogenic). Certain idiotopes are an example of such determinants which are antigenic but not immunogenic. Thus, the polypeptides of the present invention may be used to terminate tolerance and thereby target the immune response to restricted regions of self-proteins.
Anti-idiotype antibodies produced according to this invention have several distinct advantages over anti-idiotype antibodies produced by conventional immunization with an intact immunoglobulin.
Conventional anti-idiotype sera distinguish - the idiotype on the ~asis of the quarternary structure of the immunoglobulin. $hat is, the anti-idiotype antibody recognizes a three dimensional protein structure created by the folding of the primary sequence and the juxtalocation of non-contiguous regions of the primary sequence.
Anti idiotype antibodies produced according to this inventior. can distinguish idiotopes on the basis of continuous, sequence defined determinants.
The recognition site does not require the juxtalocation on non-contiguous regions of the prlmary sequence. This results in an ability to generate anti-idiotype antibodies with a high degree of specificity to a predetermined region of the primary sequence. This was not possible wi~h conventional methodology.
Another advancement of this invention over conventional methodology is that anti-idiotype antibodies can be produced against a particular idiotype without the need ~or purification.
Conventional techniques or raising anti-idiotype ~ntlhodie~ involves immuniæing a ho6t with the appropriate immunoglobulin or a fragment thereof.
This re~ults in a polyclonal response against the various idiotypic antigenic determinants on the immunoglobulin. The sera must then be passed over an absorp~ion column to separate and to isolate the particular anti-idiotope of interest, and to produce serum with specificity for an idiotype.
In contrast~ the present invention requires no absorption purification. By the method of this invention, highly ~pecific anti-idiotype antibodies can be produced ayainst a predetermined and predefined idiotype.
Still further, synthetic polyp~ptide technology provides new ana:.yl:ical tools which may play a paramoun~ role in answering questions about the structural correlates of idiotype.

s ~nti-polypeptide antisera directed against certain idiotypic determinants located in the antigen binding regions of antibodies may be a way to rel~te protein structure to antigen binding. For example, one could induce a set of antibodies to different regions in the vicinity of the binding site and determine which perturb antigen-antibody union.
The word "antigen" has been used historically to mean the entity that is bound by an antibody as well as to mean the entity that induces the production of the antibody. More current usage limits the meaning of antigen to that entity bound by an antibody, while the word "immunogen" is used for the entity that induces antibody production. In some instances, the antigen and immunogen are the sa~e entity as where a synthetic polypeptide is utillæed ~o induce produc~ion of antibodies ~hat bind to the polypeptlde. E~owever, the same polypeptide can also be utiLized to induce antibodie~ that also bind to a whole protein such as immunoglobulin, in which case the polypeptide is both immunogen and antigen, while the immunoglobulin is an antigen. Where an entity discussed herein is both immunogenic and antigenic, it will generally be termed an antigen.
Brief Description of the Drawin~
In the drawings, which constitute a portion of this disclvsure-Figure 1 illustrates the amino acid sequPnceof myeloma protein M104 from positions 90 to 110.
The D-segment amino acids located at positions 100 and 101 are indicated in bold type. The line for the s~uence of myeloma protein J558 indicates a similarity with the sequence of M104 exi~ept, as shown, for the D~segment. Polypeptides 3MM and 3JN
have shorter amino acid se~uences than the ~256~

corresponding sequences of polypQptides hV3M and hV3J, respectively, but include additional amino acids shown in italics to space the se~uence ~rom the carrier protein and to increase solubility.
Figure 2 shows nitrocellulose (Wes~ern) blots of ~odium dodecyl sulfate (SDS~-polyacrylamide ge~s including anti-polypeptide sera. Lanes 1-12 illustrate the results obtained with the individu`al antisera generated as described with reference to 10 Figure 1. Lane 13 ~hows the result with pooled, normal rabbit serum. Lane 14, on the other hand~
shows the total protein in each le~s. The upper lanes contain MlD4 myeloma protein and the lower lanes contain J558 my~loma protein. Electrophoresis was performed ~rom top to bottom with 11 micrograms of protein being loaded per lane. Sera l~B and 13 were diluted 1 in 100 in a 1~ by weight solutinn of powdered milk in borate buffered saline (BBS)~see Jvhnson et al., "Improved Technique Utiliz.ing NonEat Dry M:ilk for Analys.is of Protein~ and Nucleic Acids Transferred to Nitrocellulose", Gene Analysis ~echn~qLues (1983)~ before use while sera 9-12 were diluted 1 in 10 in the same medium. The letters M
and J indicate which protein sequence corresponds to the immunizing polypeptide (see Table 1 and Figure 1).

~:56~

Figure 3 shows Western blots of non-dissociating gels probed with anti polypeptide sera. The details are as described for Figure 2 except that 13 micrograms M104 protein and 14 micrograms J558 protein were loaded per lane. The bands of J558 protein at the top of the lower set of gel lanes indicate material which did not enter the resolving gel. This may comprise IgA (immunoglobulin A) dimers and perhaps higher oligomers~
Figure 4 demonstrat~s the inhibition o~ the polypeptide-induced anti-idiotype antibody binding to the polypeptide (A) and ~o the intact IgM-RF (Sie;
B). Antibody binding was inhibited by: polypeptide ~567~

(~ ), IgM-RF (Sie;~ ) and control polypeptide (~), at the indicated concentrations.
Figure 5 comprises a Western blot of 5 ~onoclonal human rheumatoid factors and pooled human IgG, developed with separate polypep~ide-induced anti-idiotype antisera from two i~nune rabbits (panels a and b). ~ach antibody identifies primarily a band of about 70,000 daltons which corresponds to the heavy chain of IgM-RF (Sie). Control studies with a polyvalent anti-heavy chain antibody indicate that the minor bands of lower molecular weight represented minor proteolytic degradation products of the heavy chain. The markers used are: -fluoresceinated-bovine serum albumin (F-BSA; 68k), fluoresceinated~gamma H chain (F-HC; 53k), fluore seinated L chain (F-LC; 2~k) and ~ ch~in (2$k).
Figure 6 identi~ies the amino acid resldue sequence3 (corresponding to PSI,2) o~ certain reported rheumatoid factor~. The regions and residue numbers have been assigned by Kabat et al., "Sequence of Proteins of Immunological Interest", U.5. Department of Health and Human Services (1983). The public or cross-reactive idiotypes have been determined by Kunkel et al., J. Exp. MPd~, 137, 331 (1~73). The amino acid rsidue sequences for IgM-RF ~Sie1 and Ig~RF (Wol) have been reported by Andrew et al., Proc. Natl. Acad. Sci. USA~ 78, 3799 (1981), whereas -the amino acid residue sequences for IgM-RF ~Pon) and IgM-RF (Lay) have been reported by Klapper et al., Ann. Immunol. (Inst. Pasteur), 127C, 261 (1976).
Figure 7 is a Western blot analysis of the antibody activity of the anti-P5L2 antiserum. About 2U micrograms of each indicated sample were loaded on each gel. After electrophoresis on sodium dcdecyl sulfate polyacrylamide gel and electrophretical ~Z567~$

transfer to nitrocellulose p~per, the samples were reacted respectively with anti-IgM (A, ~), anti-PSH3 (B) and anti-PSL2 (C, E) antisera. After subsequent development with the 125I-protein A (Staphylococcus aureus), the papers were finally exposed to film overnight, except that (D~ and (E) were exposed for three days.
~ igure 8 illustrate that PS~2 inhibits the binding of PSL2-induced antibodies to the IyM-RF
(Sie) (panel A); and to the isolated light chains of RF-Glo (o; panel B). In addition, inhibition by the control PSH3 (~ ) is shown in panel B.
Figure 9 illustrates inhibition of the IgM-RF Sie binding to the bound polypeptide-induced anti idiotype antibody. The PS~2 and the control P5H3 were added at the indicated concentrations to well~ precoated with ~finity-purified anti-PS~2 ant~bodle~. ~fte~ incubation ~or one hour at room temperature (23C), AP-IgM-RE' Sle (10 micrograms per milliliter) wa8 ~dded to each well and the plate was incubated for another 1.5 hours at room temperature.
Thereafter, the plate was washed, and the absorbance at 405 nanometers was measured one hour after the addition of substrate to the wells.
Deta led ~escr~ption I. Introduction Idiotypic determinants are generally believed to be involved in immunoregulation as described by Jerne, Ann. Immonol. (Inst Pasteur), 1255l 373 (1974) and Binion et al., J Exp. Med., -156, 860 (1982). Control of the system also appears to involve idiotype specific T-cells as describ~d by Milburn et al., ~. Exp. Med., 155, 852 ~1982).
Evidence shows that the predominant expression of certain idiotypes may be the result of a regulatory process rather than a restricted immunological repertoire as described by 5~senza et al., Immunological Rev., _ , 3 (1977).
Since idiotype and anti-idiotype antibodies are involved in immune regulation, it is possible to manipulate the immune response by inducing autologous anti-idiotype antibodies as described in Cosenza et al., supra. This manipul~tion with anti-idiotype antibodies is believed to have considerable medical significance in certain B-cell malignancies and autoimmune diseases. In those autoimmune diseases where the injurious antibody is of restricted origin, it may be possible to use synthetic immunogens to modulate or even eliminate the B cell clones producing the antibody.
Such autologous anti-idiotype antibodies to manipulate the immune re~pon~e can be produced in an anirnal according to the method o~ the present invention. I~ a ~peci~ic clonal type i8 sought to be regulated, then an anti-idiotype antibodies against a private iditope should be raised. But, a private idiotope is an idiotope found on only one or a few clones of an antibody of a given specificity. If regulation of all antibodies of a given speci~icity is sought, then anti-idiotype antibodies against a public idiotope should be raised.
Anti-idiotype antibodies have a second mode of function to combat autoimmune diseases and transplant or graft regulations. For example, an anti-idiotype antibody may block or obstruct the binding site so as to preclude union between the injurious antibody and its antigen.
An "antigenic determinant" is a portion of the structural configuration of macromolecule which has the capability to be bound by an antibody.

~25~i7~3~

Further explanation of an "antigenic determinant" is best accomplished by way of example. A simple protein is comprised of a linear chain of amino acid residues. This chain folds into a three dimensional structure. Certain portions of the chain are internal and other portions are external. In addition, amino acid residues that are distant apart in the primary sequence are brought into close proximity by the folding. A protein structural arrangement (configuration) thereby results. Certain portions of this configuration are such that they have the ability to be bound by an antibody of appropriate specifity; that is, they are antigenic determinants. These portions of the configuration can be thought of as having the right "shape", proper neighboring molecular environment, to bind an antibody. The antigenic determinantg o particular intere~t in this invent~on are located in th~
idiotypic reglon oE an immunoglobulin and are

2~ there~ore termed idiotypic antigenic determinants~
The preRence of an antigenic determinant on a molecule is not limited to simple proteins. The ability is general to most natural macromolecules which include, for example, glycoproteins~ dextrans, multipolypeptide chain proteins and the like.
The word "immunogenic" is used to describe the ability of an antigenic determinant to elicit an immune response; i.e~, initiate production of antibodies against the determinant. Not all 30 antigenic determinants are immunogenic, as no~ed generally befvre. This is so because of several reasons. First, the animal may have tolerance to the antigenic determinant. Second, the molecular environment around the determinant may not be right 3S for eliciting antibody production~ Third, there may ~25~i79~i be certain factors in the sera that inhibit antibody production, for example, shedded "tumor specific transplantation antigens".
The antigen binding site of an immunoglobulin is, in many respects, no different from any other macrsmolecule. Thus, the present invention illustrates that with respect to antigenicity and immunogenicity, portions of the structural configuration of the binding site and of he surrounding region can be antigenic and immunogenic. Such antigenic determiants found in the region of the binding site are called idiotopes.
Often, a binding site region will have several idiotopes, As explalned earlier, "idiotype" is the term used to de~cribe the set o~ idiotopes expressed in a binding site region. Idiotopes have been shown to be ~ound in the pocket o the binding slte and on the ~urrounding ~ur~ace~ This finding has been confirmed by x-ray di~raction techniyue~ conducted by Gi Int. Review of Biochem, 23, 71 (lg7g).
Studies of idiotopes have revealed that idiotopes lend themselves to be classified into two natural categories. In 1968, Williams showed that antibodies with similar binding specifities pos~essed idiotopes unique to themselves as well as idiotopes which they shared with the other antibodies of the same specifity. A study by Schilling et alO, Nature, 283, 35 (1980) involved 10 hybridoma clones and three myeloma proteins against alpha-1,3-dextran. This study revealed that more than one half the anti~
dextran antibodies ~hared an idiotope. The s~udy also revealed idiotypes unique to one or a few clones. Idiotype~ thus lend themselves to classi-fication into two catagories based on this phenomenon.

;67~i ~21 The first category of idiotope is th~"cross-reactive" or "public~ idiotope. Thecross-reactive idio~ope is an idiotope shared by several an~ibodies of the same specifity. Antibodies to a given antigen generated in different strains of the same species or 4ven different species have been shown to possess the same idiotope. The second category of idiotope is the "private" or "individual"
idiotope. This idiotope is found on only one of a few clones of the antibodies to a given antigen.
Changes in the amino acid composition of an idiotope do not necessarily change the binding site specificity. Rajewsky et al., Ann. Rev~ Immunol., 1, 569 (1983). A single amino acid change in the D gene segment led to Loss of one idiotype, modification of a second and loss of ~ix others in the parent molecule. However, the ~n~ibody still re~ained its binding ~ite ~peci~itl. Thi~ flndlng even applied if the idiotop~ was locat~d in or near binding site pocket On a f inal note, an animal has over one million antibodies of different blnding site specifities. Associated with each binding site specifity is a set of idiotopes. Thus in an animal there is an enormous number of idiotopes.
The present inv ntion employs the technology of raising antibodies against naturally occurring aminal proteins using short chemically syntheszed polypeptides. The basic scheme of this technology is as follows.
To begin, the primary se~uence of a protein or a portion thereof is determined. ThiS is accomplished by one of several ways. First, the sequence information may have already been determined 35 and is available from the literature. Second, the ~2~;~i7~

protein itself can be isolated and directly sequenced, using methods well known in the art.
Third, the gene which codes for the protein can be identified using genetic techniques well known in the art.
The identified gene can then be cloned and isolated using recombinant DNA ~echnology. The primary sequence of the protein can be determined from the cloned gene in either of two ways. First, the gene itself can be directly sequenced and this information translated into primary sequence, all using techniques and information well known in the art. Second, the protein coded from the gene can be sequenced .
The next step after determining the primary sequence is to analyze the sequence for regions included in antlgenic determinants. This i5 accompli~hed using the genqral knowledge of biochemi~try and immunoloyy and ~pecific information on the behavior of the particular protein~ Next, short polypeptide portions of the regions associated with an idiotypic antigenic determinant are chemically synthesized using methods such as those described in Merrified et al., ~
Society, 86, 2149 [1963~. The chemically synthesized polypeptide is typically bound to a carrier as a conjugate and the conjugate i5 injected in an effective amount as a vaccine into a host animal.
~ntisera are raised against the conjuga~e in which the specific antigenic determinant is the synthetic polypeptide The present invention provides a method for producing an:isera specific for a defined idiotype of an antibody using synthetic polypeptides~ That is, through us~ of the present invention, synthetic ~:Z 567~i polypeptide technology can be employed ~o produce anti-idiotype antibodies. The antibodies formed have at least the following characteristics.
First,the antibodies have the capability of distinguishing immunizing polypeptides and determinants on natural proteins (idivtypes) whose corresponding sequences differ by only two amino acid residues. Second, antibodies can be raised to a predetermined specific idiotype. This has not previously been possible.
Furthermore, antisera can be raised against idiotypes that are not naturally immunogenic in the animal. That is, the idiotype has the capability of being bound by an antibody (i.e. is antigenic) but does no~ initiate antibody production ayainst itself (i.e. is not immunogenic). Fourth, the antisera have the capability of reacting with both the natural and denatured protein. Fi~th, a anti-idiotype sera c~n be generated wlthout need Eor absorptlon.
~ne o~ th~ use3 and benefits o~
anti-idiotype antibodies is a possible means of medicinal lmmunoregulation. Immunoregulation using anti-idiotype antibodies and the reasoning behind selecting individual idiotypes will be discussed in greater detail below.
It will be understood that while there are many procedural steps utilizing many materials in the manufacture of ths vaccines and anti-idiotype antibody preparations of this inventionl as discussed in detail hereinafter, the invention is not limited to the utilization of any particular steps or reagents or conditions, but rather the invention is conceptually as ~tated above and as defined with particularity in the claims appended hereto.

~2~

-~4-II. _eptide Synthes s and Selection A. Syntheses Polypeptides were synthesized by solid phase methods as described in Merrifield et al., J. Amer.
Chem. Soc., 85, 2149 (1963) and Houghten et al., Int.
_ J. Peptide Protein Res., 16, 311 tl930)o The ; relatively short polypeptides used herein substantially correspond to the hypervariable regions of selected immunoglobulin molecules. [Also see, Marglin et al., Ann. Rev. Biochem~, 39, 841 (1970).]
The polypeptides were coupled to the protein carrier KhH through a cysteine residue which was added to the carboxy-terminus of the polypeptide.
The step of linking the polypeptide to the protein carrier was performed by addition of the cysteine sulfur atom to the double bond of the reaction product between the carrier and N~rnal¢imidobenxoyl~N~hydroxy succinimide ester (MBS), ~ollowlng the procedure de~cribed by I.iu et al., B ~ , 18, 690-697 (1979).
B. ~YE~E___e5 Related to M104 and J558 Proteins A myeloma protein is simply an antibody produced by a B cell which has undergone a transformation to a cancer-causing cell. It is essentially the same as an antibody produced by a normal healthy B cell. M104 and J558 are myeloma proteins produced by the MOPC104E and .~558 murine myelomas. Like any antibody, they have a binding site specificity. These antibodie~ bind the alpha-1,3-glycoside linkage of Bl355 dextran. Leon et al., _ ochemi~try, ~, 1023-1030 (1982) and Londblad et al., Inwun--he Le~L~ 9- 535,544 (1972)-Both the heavy and th~! ligh'c chains of M104 and J58 myeloma proteins h~ve been sequenced. ~he light chains of both myeloma proteins are of the ~679S

lambda allotype and are identical in sequence.
Appella, Proc. Natl Acad. Sci. USA, 68, 540-544 (1971) and Weigert et al., Nature, 228, 1045-1047 (1970). The heavy chain variable regions of both myeloma proteins have identical sequences throughout the variable domain except for the D-segment in the middle of the third hypervariable region. It is to be noted that the M104 myeloma is a mu isotype and the J558 is an alpha isotype. Kehry et al. r Proc.
atl. Acad. Sci. USA, 76, 2932-2936 (1979~ and_ Schilling et al., Nature, 283, 35-40 (1980).
The D segment sequences have been shown to determine an "individual idiotype" expressed by the M104 and J558 myeloma proteins. Clevinger et al., ICN-UCLA Svm~ Molec. Cell~ Biol., 30, 159-168 -(1981). An additional idiotype, determined by amino acids at positions 54 and 55 in the heavy chain, is expressed by both M104 and J558 myelomas and also by most of the clonotypes elicited by Dextran B1355 in the Balb/c mouse. Clevinger et al. (1981~ supra, Clevinger et al., J~ Exp. Med., 151, 1059-1070 ~1~80), and Blomberg et al., Science, 177, 179-189 (1972).
The light chains of the M104 and J558 are identical in sequence. The heavy chain variable regions are identical except for the D segment. Thus it appears that the individual idiotype found in the third hypervariable region is determined by the heavy chainO Notwithstanding, this does not imply that the light chain does not play a role in the expression of the idiotype. The light chain may be required to stabilize the conformation of the determinant.
Carson et alO, Proc. Natl. Acad Sci. U'~A, 70, 235-239 (1973)o i79S

Several synthetic polypeptides were used to induce the production of antibodies specific to a particular idiotype~ In particular, an idiotype comprising the D segment of the third hypervariable region of M104 and J558 myeloma proteins was investigated. The D segment is comprised of two amino acid residues located at positions 100 and 101 from the amino-terminus. The D segment is bordered on the amino-terminal side by the V segment and on the carboxy-terminal side by the J segment (see Figure 1).
Two sets of two synthetic polypeptides were prepared in this study. The first set comprised two polypeptides corresponding in sequence to the first seven adjacent amino acid residues in the ~ segment and 3 adjacent amino acids in the V segment of M104 and J558. These synthetic polypeptides were designed to be long enough to ensure eliciting an anti-polypeptide immune response. The synthetic polypeptides were designated hV3M and hV3J and corresponded, respectively, to the above-described portions of the M104 and J558 myeloma proteins.
The hV3M and hV3J polypeptides represent a substantial portion of the sequence encoded by the Jl gene segment which is common to both polypeptides.
Thus, there was the possibility that much of the anti-polypeptide response would be directed against - this common sequence. This could perhaps result in antibody activity specific for ~he different D-segments being obscure~
The second set of synthetic polypeptides employed in-this study included about a minimum cognate sequence necessary to form an antigenic determinant as described by ~erner et al., Na ure, 299~ 592 (1982)o These synthetic polypeptides were ~67~5i designated 3MN and 3JN corresponding, respectively, to the M104 and J558 myeloma proteins, and are shown in Figure 1, herein.
In order to prevent steric interference with the immune presentation of the second set of polypeptide antigens with the carrier, three proline residues were added as a spacer between the cognate sequence and the RLH carrier proteinO Proline, glycine and other groups have been used successfully as spacers as described in Stevens et al., Am. J.
Reprod. Immunol, 1, 307 (1981). Glutamic acid residues were added to the sequence to increase solubility.
C. Polypeptides Related to IgM-RF
For the most part, antibody response is directed against invading organisms and altered self cells or self antigens. For example, rheumatoid factor (RF) is an IgM or IgG antibody with specificity for the Fc fragment of IgG. Rheumatoid factor usually results from an abnormal condition and causes synovial inflamation and vascolitis. IgM-RF
is abundant in the sera of most individuals with rheumatoid arthritis~ IgM-RFs in rheumatoid sera are polyclonal and react with a number of different antigenic determinants in the Fc region of IgG.
IgM-RFs from unrelated individuals show cross-reactive idiotypes.
A polypeptide (designated PSH3) was synthe-sized that corresponded to the third hypervariable region (amino acid residues 99-111) of the heavy chain of the monoclonal IgM-RF (Sie) antibody according to the sequence reported by ~ndrews et al.
Proc. Natl. Acad. Sci. US~, 73, 3799 (1981). The polypeptide had the sequence from left to right and Bo25679~i in the direction from amino-terminus to carboxy-terminus of: GluTrpLysGlyGlnValAsnValAsnProPheAspTyr.
A tripeptide GlyGlyCys was added to the carboxy-terminal (C terminal) end of the above polypeptide sequence to facilitate coupling to a protein carrier as described by Green et al., Cell, 28, 477 (1982). The resulting polypeptide had the following sequence from let to right and in the direction from amino-terminus to carboxy-terminus:
GluTrpLysGlyGlnValAsnValAsnProPheAspTyrGlyGlyCys.
~ polypeptide (designated PSL2) was synthe-sized that corresponded to the second hypervariable region (amino acid residues 49-61) of the L chain of the monoclonal IgM-RF (Sie), according to the sequence reported by Andrews et al., supra. PSL2 has the amino acid residue sequence from left to right and in the direction from amino-terminus to carboxy-terminus of: TyrGlyAlaSerArgAlaThrGlyIleProAspArg. ~n additional Cys was added to the C-terminal end of the seguence to facilitate coupling to the protein carrier as referenced above.
III. Immunization Procedures The vaccines used herein contain the tated amount of polyp~ptide alone, as a polymer of individual polypeptides linked together through reaction with glutaraldehyde or linked to a carrier.
Polymeric polypeptides can also be prepared by addition of cystei~e residues at both polypeptide termini followed by oxidation as with atmospheric oxygen at moderate pH values such as between about p~
7 and pH 10. The stated amounts of polypeptides refer to the weight of polypeptide without the weight of a carrier, when a carrier was used.
The vaccines also contained a physiologically acceptable vehicle such as water or ~25~

saline, and further typically included an adjuvant.
Complete Freund's adjuvant (CFA) and incomplete Freund's adjuvant ~IFA) are materials well known in the art and are available commercially from several sources.
Vaccine stock solutions were prepared with IFA or CFA as follows. An amount of the synthetic polypeptide, polymeric polypeptide or conjugate suficient to provide the desired amount of polypeptide per inoculation was dissolved in borate-buffered saline (BBS). Equal volumes of CFA
or IFA were then mixed with the polypeptide solution to provide a vaccine containing po~ypeptide, water and adjuvant in which the water-to-oil ratio was 1:1. The mixture was thereafter homogenized to provide the vaccine stock solution.
Vaccine stock solutions can also be prepared with keyhole limpet hemocyanin (KLH), KLH in IFA
(incomplete Freund's adjuvant), alum, K~H-alum absorbed, KLH alum absorbed-pertussis, edestin, thyroglobulin, tetanus toxoid and tetanus toxoid in IFA.
Upon injection or ~ther introduction of the antigen or vaccine into the host, the host's system responds by producing large amounts of antibody to the antigen. Since the specific idiotypic antigenic determinant of the manufactured antigen, i.e., the antigen formed from the synthetic polypeptide and the carrier, is the same as or is an immunological surrogate for the determinant ~f the natural antig~n of interest, the host becomes immune to the natural anti~en. In the case where the invention is used as a vaccine, this is the desired result.
The effective amount of polypeptide per inoculation tas discussed in the accompanying tables) ~L~Z567~

depends, inter alia, on the animals inoculated, body weight of such animals and the chosen inoculation regimen Vaccines are typically prepared from the dried solid polypPptide by suspending the polypeptide in water, saline or adjuvant, or by binding the polypeptide to a carrier and suspending the carrier-bound polypeptide (conjugate~ in a similar physiologically tolerable vehicle such as an adjuvant. An effective amount of polypeptide present in a vaccine may be from about 20 micrograms to about 500 milligrams per inoculation, exclusive of any carrier used.
It is frequently convenient to add one or more additional amino acids to the amino- or carboxy-termini of the synthetic polypeptide to assist in binding the synthetic polypeptide to a carrier to form a conjugate. As discussed before, cysteine residues added at the carboxy-terminus of the synthetic polypeptide have been found to be particularly useful for forming conjugates via disulfide bonds and Michael addition reaction products, but other methods well known in the art for - preparing con~ugates may be used. Exemplary binding procedures include the use of dialdehydes such as glutaraldehyde and the like, or the use of carbodiimide technology as in the use of a water-soluble carbodiimide, e.g.
l-ethyl-3-(3-dimethylaminopropyl) carbodiimide.
Useful carriers are well known in the art and are generally proteins themselves. Exemplary of such carriers are keyhole limpet hemocyanin (KLH), edestin, thyroglobulin, albumins such as bovine serum albumin or human serum albumin ~BSA or HSA, respectively), red blood cells such as sheep erythrocytes ~SRBC), tetanus toxoid, as well as ~679~

polyamino acids such as poly(D-lysine:D-glutamic acid), and the like.
As is also well known in the art, it is often beneficial to bind the synthetic polyp~ptide to its carrier by means of an intermediate, linking group. As noted above, glutaraldehyde is one such linking group, while when cysteine is utilized, the intermediate linking group is preferably a m-maleimidobenzoyl N-hydroxysuccinimide ester (MBS).
MBS is typically first added to the carrier by an ester-amide interchange reaction. Thereafter, the above Michael reaction may be followed, or the addition may be followed by addition of a blocked mercapto group such as thiolacetic acid (CH3COSH) across the maleimido-double bond. Cleavage of the blocking group follows, and then the disulfide bond is formed between the deblocked linking group mercaptan and the mercaptan of the added cysteine residue of the synthetic polypeptide.
The choice of carrier is more dependent upon the ultimate intended use of the antigen than upon the determinant of the antigen, and is based upon criteria not particularly involved in the present invention. For example, if a vaccine is to be used in animals, a carrier which does not generate an untoward reaction in the particular animal will be selected. If a vaccine is to be used in man, then the overriding mat~ers relate to lack of immunochemical or other side reaction of the carrier and/or the resulting antigen, safety and efficacy--the same considerations which apply to any vaccine intended for human use.
It is very often desirable to determine if a particular antigen is present as an aid, for example, in the diagnosis vf a particular diseaseO Because ~æs6~$

the synthetic antigen is mono-specific to the single specific antigenic determinant of interest, antibodies to the antigen are also mono-specific to the antigen of interest. Perfect mono-specificity may not always be accomplished but cross-referencing to other antigenic portions of the antigen is avoided because only one immune response is possible by the antibody.
Antibodies are harvested and prepared in any conventional procedure for use in diagnostic tests.
It is common, for examplel to label the antibody for identification and quantitative determination.
Radiolabelling by the method of Greenwood, for example, lSee Hunter~ Br Med. Bull. 30rl8 (1974)]
and fluorescent dye labelling are commonly used in immunoassays.
Vaccines are prepared by suspending the carrier-bound (conjugate) polypeptide in a physiologically tolerable vehicle. Other acceptable suspension media are well known in the art and include water, saline, and the like. The suspended carrier-coupled synthetic polypeptide will be referred to herein as a vaccine.
IV. Results -A. Immunization With M104 and J550 Related Pol e tides ~P P
The vaccines were injected into host animals in an effective amount to elicit an immune response.
Three immunization procedures were employed to generate anti-polypeptide sera against the polypeptides as described in Table 1. Two rabbits were taken through the immunization process for each synthetic polypeptide. Antisera against ~he hV3M and hV3J polypeptide were raised in New Zealand white rabbits and antisera against the 3MN and 3JN
polypeptides were raised in red New Zealand Rabbits.

~2567~5 Table 1 _ ~ - ~ ~ 3 ~abbit iPolypeptidel First ~nte~val ¦Second Interval Serum Number ~Iniection! ~njection : 1 and 2 , h~3M I CFA IFA
l 3 and 4 , hV3J I sc + fp 14 sc ~ f 7 . , __ . ~ P
5 and 6 hV3M CFA 72 IFA 42 7 and 8 hV3J sc sc . --1 9 and 10 j 3MN CFA I IFA
11 and 12 I 3~N sc + lfp 62 Isc ~ r~ _ ____ ¦ Rabbit ~ Polypeptid~ Third ~ Fourth ~erum Numberl Injection njection _ _ . l.and 2~same as Alum Alum

3 and 4ab~e) ip 28 ip _ 5 and 6(same as IFA __ __ 7 and 8above) sc _ _ _ _ 9 and 10 (same as IFA IFA
11 and 12 above) Rc ~ lfP 38 sc ~ rfF .
,.

_. _ __ 1. The various immunization protocols employed to generate the anti-polypeptide sera using synthetic polypeptides related to myeloma ~xoteins M104 and J558 are shown. Rabbits 1-8 were ~emale New Zealand White and rabbits 9-12 were female New Zealand Red. Each animal received 200 micrograms peptide coupled to KLH at each injection. ~abbits 1-8 were bled 7 days after the last injection and rabbits were bled 9 day~ aftex the last injection.
2. The adjuvant used and the injection sites are shown for each injection. The following abbreviations are used:
CFA - complete Freund's adjuvant; IFA ~ incomplete Freund's adjuvant; ~c - 4 8 multiple Gubcutaneous injection~ in the back; fp - injection into both rear footpads; lfp or rfp -injection into one rear footpad, left or right, respec~ively;
and ip - intraperitoneal injection.
3. Intervals between successive injection~ are ~hown in days.

~Z$~79$

The immunization procedures involved either three or four innoculations of vaccine. In any inoculation, the animal received two hundred (200) micrograms of synthetic polypPptide coupled to keyhole limpet hemocyanin (KLH) and the conjugate was either emulsified in Freund's adjuvant or alum. The vaccines were administered in either multiple subcutaneous injections in the back and rear footpads vr intraperitoneally. Subsequent inoculations followed between 7 and 62 days after the prior inoculation. The rabbits immunized with HU3J and hV3M polypeptides were bled 7 days after the last inoculation and the rabbits immunized with 3MN and 3JN polypeptide were bled 9 days after the last inoculation.
1. Reactivity with denatured M104 and J558 proteins The antisera were first evaluated for their ability to descriminate between denatured Ml04 and J558 myeloma proteins. The determination was accomplished by first running Ml04 and J558 myeloma proteins on vertical slab denaturing sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PA OE) using a five to ten percent polyacrylamide gradient as described in Laemnli, Nature, 227, 680 (1970).
The proteins were then electrophoretically transferred to nitrocellulose (Schleider and Shuell Inc., of Keene, N~ as described in Towbin et al~, Proc. Natl. Acad. Sci~ USA, 76, 4350 (1979). Next, the blots were stained with amino black and were cut into strips corresponding to the gel lanes.
Non-specific binding was blocked wi~h a reagent rec~ntly de~ised for this purpose as described in Johnson et al. t SU~o Finally, the strips were exposed to anti polypeptide sera for ~wo hours at ~Z~67g~

room temperature and positive reactions were then visualized with I125 - Protein A (Sta~ylococcus aureus) and autoradiography as described in Johnson et al., supra.
The results of this determination are illustrated in Figure 2~ Two rows of a plurality of strips are shownO The top row of strips illustrates binding of the various antisera to the denatured M104 protein. The bottom row illustrates binding o~ the various antisera to the J588 protein.
Fourteen columns of strips are illustrated.
Each column is labeled with a number at the top.
Columns l through 12 correspond to antisera raised in the twelve immunized rabbits, respectively (Table 1). Column thirteen in a blank and column fourteen is a standard showing the migration positions of the myeloma protein.
Below the column number is the letter M or the letter J. The letter M indicates that the rabbit was immunized with a polypeptide containing the D
segment of a M104 myeloma protein. The J indicates that the rabbit was immunized with a polypeptide containing the D segment of a J558 myeloma protein.
The antisera raised in eight of the twelve rabbits (Numbers 1, 3/ 5, 6, 7, 8, 10, and ll) descriminated between the M104~and J558 proteins and displayed a high level of specificity for the protein corresponding to the immunizing polypeptide.
Furthermore most, if not all, of the reactivity was confined to the heavy chain bands. This is consistent with the predetermined specificity of the anti-polypeptide sera. The faint additional bands which are apparen~ in lanes l and ~ are presumably due to either breakdown products or differences in glycosylation or both.

~;~56'7~i Antisera raised against the hV3M and hY3J
polypeptides in RAbbits 1, 2, 3, and 7 displayed cross reactivity between the two myeloma proteins.
This cross reactivity is presumed to be due to extensive stretch of Jl gene segment sequence shared by both proteins.
Attention is also called to the observation that the antisera raised in rabbits 9 and ll did not react with the denatured M104 myeloma proteins. The next section discusses a determination of the reactivity of antisera to the native myeloma proteins. There it will be shown that anti-sera raised in rab~its 9 and 11 react with the native M104. These two observations coupled together tend to indicate that different antibody molecules in the antiserum react preferentially with or the other of the two protein states. This finding will be more thoroughly discussed in the next section.
2. Reactivity of the Anti-sera With Native M104 and J558 Proteins .
The antisera were evaluated for ~heir ability to descriminate between native M104 and J558 myeloma proteins. This determination was accomplished by first running M104 and J558 myeloma proteins on vertical slab ~5 gel electrophoresis. The same apparatus was used as in the determination with denaturing gels. The following gel systems were employed:
1. M104 protein was run on a 7 ~ercent agarose gel in a buffer gradient of 10 millimolar to 100 millimolar tris~borate pH 90 0 f or 20 5 hours at 3 milli-amps.
2. The J558 protein was run i~ a 5 to 12 percent polyacryLamide gradient in 70 millimolar tris-borate pH 9.0 for 2 hours at 10 milli-amps.

~2S6'7~5 ~37-The proteins were transferred o nitrocellulose (Schleider and Shuell), for the non-dissociating gels by way of a Bio-Rad (Richmond, CA) Trans-Blot cell utilizing seventy millimolar Tris-borate pH 9.0 as the transfer buffer. Next, the blots were stained with amido black, and were then cut into strips corresponding to gel lanes.
Non-specific binding was blocked with a reagent described in Johnson et al., supra. Finally, the strips were exposed to anti-polypeptide sera for two hours at room temperature and positive reactions were then visualized with I125 - Protein A and autoradiography as described in Johnson et al., supra.
The results of this determination are illustrated in Figure 3. Two rows of a plurality of strips are shown. The top row of strips illustrates binding of the various antisera to the native M104 protein. The bottom row illustrates binding of the various anti-sera to the native J558 protein. The column number designations and immunizing polypeptide designations tM or J) are the same as that described in Figure 2.
The bands in Figure 3 are morP diffuse than those in the denaturing gel system (Figure 2). This is because of an absence of stacking forces in the non-denaturing gels of Figure 3~ The denaturing gels showed that ~he protein preparations used contained little, if any, extraneous protein material.
Therefore, despite the diffuse appearance of the bands, the bands in Figure 3 comprise just M104 or J558 protein.
The antisera raised in six of the twelve rabbits (numbers 1, 5, 61 8, 9, and 10~ ~escriminated between the M104 and J558 native protein and displayed specifity for the protein corresponding to ~s~9~

-3~-the immunizing polypeptide. It is to be recalled that the antisera raised in rabbits 1, 3, 5, 6, 7, 8, 10, 11 demonstrated the same abilities with respect to the denatured protein. Thus, at least half of the animals produced antibodies capable of descriminating the individual idiotypes of the M104 and J558 proteins, irrespective of whether the proteins were native or denatured.
These results represent the first production of antibodies specific for defined idiotype elicited by synthetic immunogens.
The antiserum raised in rabbit 9, and to a lesser extent the antiserum raised in rabbit 11, reacted with the native myeloma protein. However, the antiserum from these rabbits did not react with the denatured proteins (Figure 2). It remains unclear whether, in the given anti-polypeptide serum, the same antibody molecules are responsible for the interaction with both native and denatured proteins or whether different molecules in the antiserum react preferentially with one or the other or the two protein states. The precise situation may well depend on the particular epitopes being considered.
The above data tends to indicate that different antibody molecules may be involved.
Antisera raised against the hV3~ and hV3J
polypeptides in rabbits 1, 2, 3, 6, and 7 displayed cross reactivity between the two myeloma proteins.
This cross reactivity is presumed to be due to the extensive stretch of J gene segment sequence shared by both proteins.
B. Immuni~tion with I~M-RF_Related Poly~tides 1. Immunizations Synthetic polypeptides PSL2 and PSH3 were conjugated in separate reactions via their terminal ~Z567~5 cysteines to keyhole limpet hemocyanin (KLH) with m~maleimidobenzoyl N-hydroxysuccinamide ester, as described previously. Green et al., Cell, 28, 477 l982) and Liu et al., Biochemistry, 18, 690 (1979).
Each of two rabbits was injected subcutaneously with 2.5 mg (milligrams) of the conjugates emulsified in complete Freund's adjuvant. The injection was repeated two months later. Three weeks after the second immunization, the rabbits were boosted again with 2.5 mg glutaraldehyde cross-linked polypeptide in incomplete Freund's adjuvant. The latter reagent was prepared by the addition of glutaraldehyde (final concentration 0.25 % v/v) to a 5 mg/ml solution of polypeptide in isotonic phosphate buffered saline, followed by l-hour incubation at room temperature, the rabbits were bled, and the sera were stored at 20~C until analyzed.
2. Purification of Proteins Plasma or purified proteins from patients with monclonal IgM cryoglobulins were purified by repeated precipitation at 4C, followed by chromatography on Sephadex G-200 or Ultrogel AcA 22 in 0.2 molar sodium acetate at pH 3.5. IgM and IgG
peaks were identified by immunodiffusion, and then the appropriate fractions were pooled and stored at a temperature of -20C. Human IgG was prepared from Cohn fraction II (Sigma, St. Louis, MO) by DEAE
cellulose chroma~ography in 0.01 molar sodium phosphate at pH 8Ø
The heavy and light chains of the IgM-RF
protein Sie were separated on a Sephadex G-100 column with 1 molar acetic acid, after complete reduction and alkylation. Bridges et ala, Blochemistr~, 10, 2525 (1971). The polypeptides were stored frozen at 1 mg/ml~ Using a radioimmunoassay specific for IgM

~ ~R~D~ ~Rk 7~S

heavy chains or kappa chains, it was estimated that the heavy chains contained less than 5% light chains while the light chains contained less th~t 2~ heavy chains.
3. Enzyme Linked Immunosorbent Assay (ELISA) The synthetic polypeptide ~100 micrograms/ml), various purified monclonal IgM-RF ~10 micrograms/ml), and isolated heavy and light chains from the IgM-RF proteins (10 micrograms/ml) were dissolved in borate buffered saline (BBS) including 0.1 molar borate and 0.2 molar NaCl at pH 8.2. The mixture was then added to wells of polyvinyl chloride microtiter plates (Costar ~3590) at a concentration of about 100 microliters per well. After overnight incubation at 4C., the plates were washed twice with BBS containing 0.5% Tween-~0 (Sigma #P-1379;
BBS/Tween-20) and were quenched with BBS containing 1% bovine serum albumin (BSA) for one hour at room temperature. Then, 100 microliters of sera diluted with BBS containing 0.5~ BSA were distributed to wells in duplicate. The plates were incubated for 3 hours at room temperature. Subsequently, each well was washed 3 times with BBS/Tween-20. Then, 100 microliter aliquots of a 1:800 dilution of alkaline-phosphatase labeled goat anti-rabbit IgG
(Kirkegaard and Perry, Gaithersburg, MD), that had been previously adsorbed with human IgG-Sepharose 4Bt were dispensed to the wells. ~fter another one hour incubation at room tempera~ure, the plates were washed 5 times with BBS/Tween-20. Thereafter, 100 microliters of ~-nitrophenyl phosphate (1 mg/ml) in 0.05 molar sodium carbonate pH 9 8 was added to the wells, and the absorption at 405 nanometers was measured in a Titertek Multiscan meter after on~ hour at room temperature, and 16 hours at 4~C.

R t'r D ~

~2S~5

4. Inhibition Assay The inhibition of the anti-polypeptide antibody binding to plates coated with IgM RF (Sie) or the polypeptide was assessed by the previously described ELISA methods, but with the following modifications. The antiserum from rabbit number 2 diluted 1:1,000 in BBS/0.5% BSA, was first mixed with an equal volume of inhibitors (e.g., Sie, immunizing polypeptide, or control polypeptide) at the concentration specified in Figure 1, and then was distributed to wells in duplicate 100 microliter aliquots. A polypeptide corresponding to the third hypervariable region of the heavy chain of the monoclonal human IgM-RF (Wol) having the amino acid residue sequence from left to right and in the direction from amino-terminus to carboxy-terminus of: GluTyrGlyPheAspThrSerAspTyrTyrTyrTyrTyrGlyGlyCys was used as a control. Andrew~ et al., supra.
SO Adsorption and Elution of the IgM-RF (Sie)-binding activ tx The globulin fraction of anti-polypeptide antisera was precipitated twice with 40~ ammonlum sulfate, and then was digested with 3~ (w/w) pepsin for 16 hours at 37C and pH 4.1. After neutraliza-tion, the~digest was recirculated over a protein-~Sepharose 4B column (Pharmacia Fine Chemicals, Piscattaway, NJ) to remove undige~ted IgG. Subse-quently, the F(ab')2 fragments were recirculated over a polypeptide-coupled Sepharose 4B affinity column (ca. 6.6 mg/ml gel x 5 ml), that had been prepared with cyanogen bromide activated Sepharose 4B
(Sigma, St. Louis, MO)o After removal of non-bound material with BBS the F(ab')2 antipolypep~ide antibody was eluted with 0.1 molar glycine ~Cl, pH
3.0, and was then dialyzed against BBS.
~ D~ ~ ~ R~

~256~9~

6. Protein Blotting The reactivity of th~ anti-polypeptide antibody with immunoglobulin light and heavy chain polypeptides was tested by the Western blot method (Towbin et al., Proc. Natl. Acad. Sci. USA, 76, 4350 (1979) as modified by Billings, et aL., J. Immunol., 128, 1176 (1982). Briefly, about 20 micrograms of individual monoclonal IgM-RF proteins [as discussed in Carson et al., Mol. Immunol., 20, 1081 (1983)] or pooled human IgG in 25 microliters of sample buffer supplemented with 0.01% 2-mercaptoethanol, was loaded onto each slot of a 10% polycacrylamide slab gel, containing 0.1~ sodium dodecyl sulfate [Laemelli, Nature, 227, 680 (1970).~. After electrophoresis for 3 hours at 30 milliamperes, the proteins in the gel were transferred electrophoretically to nitrocellulose paper. Protein binding sites on the paper were quenched with PBS containing both BSA ~5%) and ovalbumin (5~) for one hour at room temperature.
Thereafter, the paper was overlaid with the anti-polypeptide antiserum (1:100 dilution in PBS
containing 2~ of both BSA nd ovalbumin) for one hour. After washing, the paper was developed with the I-labelled protein A (1 mCi/mg, 2 x 105 cpm/ml) for another hour. After extensive washing, the paper was dried and finally exposed to XAR-5 film (Eastman Kodak Co., Rochester, NY) overnight at -70C.
7. Induction of the Anti-pe~tlde An After receiving two subcutaneous injections of polypeptide-KLH conjugates in CFA, and one injection of glutaraldehyde cross-linked polypeptide in incomplete Freund's adjuvant, the rabbits were bled and the sera were analy~ed for anti-polypeptide activity by the ELISA method. As shown in Table 2, sera from both immunized rabbits contained ~:25~3i7~

anti-polypeptide antibody detectable at dilutions as high as 1:100,000. Control ~era from normal rabbits did not bind significantly to the polypeptide coated plates.

INDUCTIO~ OF ANTI-IDIOTYPE ANTIBODIES BY A
POLYPEPTIDE SUBSTANTIALLY CORRESPONDING TO THE
SECOND HYPERVARIABLE ~EGION OF MONOCLONAL
10 HUMAN IgM RHEUMATOID FACTOR (Sie) Absorption at 405 nanometers (x103) Serum Dilution 10 3 10 4 10 5 Control 15 Normal Rabbit25 0 0 4 Immune Rabbit 11023 497 95 5 Immune Rabbit 2937 530 85 0 The activities of polypeptide-induced anti-idiotype antibodies (as described above) of two immune rabbit sera were assayed as described herein by a solid phase ELISA on a polypeptide-coated polyvinyl chloride microtiter plate. Duplicate microtiter wells were coated with the peptide (100 micrograms per milliliter of BBS~ and various dilutions of the rabbit antisera were ad~ed. Control wells contained buffer only~ After incubation or three hours followed by washing, the amount of bound antibody was determined with alkaline phosphatase conjugated anti-rabbit IgG. The absorption at ~05 nanometers of the enzyme substrate was measured after one hour.

~25~7~5 8. Reactivity of Anti-polypeptide Antibody with the Intact antibody Molecule I~M-RF (Sie) The anti polypeptide antisera were assayed for direct binding to plates coated with intact IgM-RF (Sie)~ Table 3 ~A and B~ shows that both anti-polypeptide antisera, but not control sera, reacted with the intact antibody molecule. Even at l:lOO,Q00 dilution, the anti-polypeptide sera bound significantly to the intact IgM protein.

2~

- ~L;25~7~S

-~5-TAB~E 3 REACTIVITY OF THE POLYPEPTIDE-INDUCED
ANTI-IDIOTYPE ANTIBODY WITH T~E INTACT
IgM-RF ~Sie) ANTIBODY MOLECULE*

Absorbance at 405 nanometers (x103) lAfter 1 hour Incubation at Room Temperature t23C)]

10 Serum Dilution 10 3 10 3_ 10 5_ Control ~ormal Rabbit 35 13 3 14 Immune Rabbit 1 292 106 47 46 Immune Rabbit 2 307 164 75 0 Absorbance at 405 nanometers tX103 [After Overnight Incubation at 4C]

20 Serum Dilution__ 10 3 10 410_5 Control Normal Rabbit 96 19 3 14 Immune Rabbit 1 1041 376.78 14 Immune Rabbit 1 824 38075 9 ~5 * The solid phase ELIS~ was performed as in Table 2, except that the wells were coated with 10 micrograms intact IgM-RF per milliliter BBS. The absorbance readings at 405 nanometers were taken after one hour incubation at room temperature (23 C), and again after overnight incubati~n at 4~C.

3~

~2S6~

Several IgM-RF paraproteins, including IgM-RF (Sie), have been shown to interact with rabbit IgG. Kunkel et al.~ J. Exp. Med., 137, 331 (1973).
Hence, it was necessary to prove that the interaction between the anti-polypeptide antibody and the IgM-RF
(Sie) was due to the specific binding activity of the anti-polypeptide antibody, and not to a non-specific interaction of the IgM-RF with rabbit IgG in the antiseraO Table 4 (A and B) ~hows that the anti-polypeptide antibody bound significantly to isolated heavy chains prepared from IgM-RF (Sie), which lacked detectable ability to bind rabbit IgG.

~5 ~56~795 REACTIVITY OF THE POLYPEPTIDE-INDUCED
ANTI-IDIOTYPE ANTIBODY WITH THE
ISOLATED HEAVY CHAIN OF IgM-RF (Sie)*

Absorbance at 405 nanometers (x103) [After 1 hour Incubation at Room Temperature (23C)]

Serum Dilution10 3 10 =10Control Normal Rabbit 16 9 3 10 Immune Rabbit 1437 105 10 11 Immune Rabbit 2324 92 6 0 Absorbance at 405 nanometers (x103) ~After 1 hour Incubation at Room ~ Temperature (4C)]

Serum Dilution 10 10 _ 10 Control_ Normal Rabbit40 13 4 11 Immune Rabbit 1 1443 404 67 12 Immune Rabbit 1 1084 319 50 5 _ * In this determination`, the wells were coated with the isolated heavy chain of the IgM-RF (Sie) protein ~10 micrograms protein per milliliter BBS). Otherwise, the conditions were the same as those described for the trial shown in Table 2.

Morever, the F(ab')2 fragments of the anti-polypeptide antibo,~, but not those of normal rabbit IgG, bound to the intact IgM-RF (Sie~ protein.
(See Table 5).

~2~i;6~95 ADSORPTION AND ELUTION OF THE
IgM-RF (Sie) BINDING ACTIVITY FROM A
POLYPEPTIDE-COUPLED AFFINITY COLUMN

Antigens Coated: BSA Polypeptide IgM-RF (Sie) Samples: Absorbance at 405 nanometers (x103) Original 17 1900 587 Effluent 33 374 47 Eluate 28 1900 459 Normal Rabbit IgG 59 136 89 Crude F(ab')2 fragments (200 milligrams) of the polypeptide anti-idiotype antibody were added to a 3 milliliter peptide~coupled column (5 mg/ml gel).
After incubation for 15 minutes at room temperaturee (23C), the effluent was collected. After washing, the bound material was eluted with 0.1 molar glycine-HCl (pH3)y and was neutralized. All samples were assayed at a concentration of 25 microgram per milliliter to microtiter wells coated wi~h BSA, the polypeptide, or intact IgM-RF ~Sie) in the standard ELISA procedure as ~5 described above.

9. The Anti-Peptide Antibody Recognizes A
P4ptide-Determ ned Epitope_on IgM-RF (SieL
To prove that the antibody bound to a specific polypeptid~-determined epitope on the intact IgM-RF (Sie) molecule, two types of studies were performed. First, as shown in Table 5~ most of the IgM-RF (Sie) binding activity W2.5 adsorbed by, and eluted from, a polypeptide-coupled immunoadsorbent column. Second, the antibody binding activity to ~2567gs - IgM-RF (5ie) coated plates was inhibited completely by the free polypeptide in solution (Figure 4).
Under the same conditions, a control polypeptide corresponding to the third hypPrvariable region of the heavy chain o the monoclonal IgM-RF (Wol), did not have significant inhibitory activity, even at a l,000-fold higher concentration.
10. The Anti-polypeptide Antibody Recognizes a Private Idiotope on IgM-RF tsie)_ _ The observation that the anti-polypeptide antibody bound efficiently to isolated IgM-RF (Sie) heavy chains enabled development of a sensitive protein blotting method for the detection of the epitope-bearing IyM-RF that avoided non-specific interactions between human IgM-RF and rabbit IgG. A panel of IgM-RF
- paraproteinsr as well as pooled human IgG (Cohn Fraction II), were fractioned by SDS polyacrylamide gel electrophoresis under reducing conditions, and then were transferred onto nitrocellulose paper. After incubation with the anti-polypeptide antibody and final development with 125I-label~d protein-A, autoradiographs were prepared.
Figure 5 shows that the anti-polypeptide antibody reacted with the heavy chain of IgM-RF
paraproteins, two of which (Glo and Teh) react with a monoclonal antibody against a cross-reactive (public) idiotype on the Sie molecule (Carson et al., ~ol.
Immunol~, su~ra).
Notably, the anti-polypeptide antibody did not react detectably with the heavy chains of pooled human IgG.
These results demonstrate that the anti-polypeptide antibody recognizes a private idiotype on the heavy chain of IgM-~F (~ie).

~2S~795i 11. Polypeptide-Induced ~ntibodics Reco~nize a Public Idiotype ~f ~um~n RF~

After receiving two subcutaneous injections of PSL2-KLH conjugates and one injection of glutaraldehyde cross-linked polypeptide, the sera were obtained one week after the last injection and were analyzed for anti-polypeptide activity by the herein described ELISA techinique.
As shown in Table 6(A), both immune sera contained high titers of anki-PSL2 antibodies. Coated sera pooled from normal rabbits did not bind significantly to the polypeptide-coated plates. In addition, the immune sera did not react with polypeptide PSH3 which was used as a control.
Thereafter, the anti-polypeptide antisera were assayed for reactivity with IgM-~F Sie. Table 6(B) shows that both antisera, but not the control sexa, reacted speciically with the intact antibody molecules.
However, the reactivities of both antisera with IgM-RF
Sie were relatively weak. This suggests that either only a small fraction of PSL2 induced antibodies reacted with the intact i~munoglobulin, or that the interactions between the PS~2-induced antibodies and the Igm-RF Sie were of very low affinity.

.

~L~256795 ~ABLE 6 INDUCTIO~ OF U~ ~F ~NTI~OD~S ~IT~ POLYP~P~ID~
Detecti~n o~ Pslypeptl~e-Inaucea An~i-Idiotype Ant$~ca~s.

~bsorbance at 405_n~nometer~ 5x10 ) ~olypeptide 5~SL2~ 8erum Dllut~on 10-310-4 10 5 Control ~or~l Xab~i~23 6 3 5 I~mune Rnbbit 1 1036 302 53 15 Immune ~ bbit 2 94B 370 54 ~) Detectlon of ~-Sndured ~nti-~diotype An~i~odie~.2'3 b~or~nce ~t 4~5 nanometer3 ~103) RF txgM-RF I~y) ~erum Dilutlon 10-310 10 5 CDntrol Normal ~bbit35 13 3 4 Immu~e Rab~it 1 136 2B 5 12 I~mune R~bbit 2 92 21 13 14 1. The polyp~pt$de-lnduced an~i-ia~ Dtyp~
antibody ~c~ivitl@s ~ore ~s~sa by ~ sollaoph~e ~LI8A on ~ polyp~ptide-coat0d ~cro~lger plate. Vas~ou~ ~ilutions o~ the ~ra ~n~ cQntrol (bu~er only~ ~ere aadea to w~
in duplicate ~d ibcubnt~ ~or 3 hour6 ~t roD~ te~per~tur~
(23C.). Sub~quently, ~he bOuna ant~boai~ r~
~uan~ t~d ~h the en~y~-con~uga~@a ~ntl-r~b~lt I~G An~
the sub~trat@. ~he ~b~orptions ~t A~05 a ter l~hour ~t ro~m temperature ~ers det~rsi~
2. The sp~c~fici~ o bo~h sas~y~ were sh~w~
~y that lmmune ~erum ~1~ not reac~ ~lth the oontrol P~3 ~A~ ~d the ~ontrol ~gM RF ~ay ~B) re8p~1v~1y (~Sa no~
~own ) .
3. ~ame ~ o e 1 ~b~ve, ~Xoept th~t the well3 w~se c~te~ with ~he IgM~F 8ie.

~L25~i79~i Since PSL2 includes ~n amino ~cid residue ~equance that ~orresponds ~o a hypervariable reglon ~h~red by L chains of two human monoclonal IgM-RF~
(Andrews et ~1., upra), the reac~ivi~y ~f the ~SL2-induced antibody ~as determined against a panel of human monoclonal RFs. Table 7 (line~ 2 ~nd 3) shows that PSL2-~nduced antibodie~ react with Ig-~F
Glo, and to a le~er degree with IgM-RF P~m, but not I~M RF Lay. Again, the binding was extremely weak.

~2S~795 --53--.
TA~3 7 RE~CTIVI~IES GF ~SL2-INDUCED A~TI~OD~S ~ITH
C~RT~IN ~UH~N ~O~CLONAL I~-R~

Absorb~nce ~t 405 n~nometerc (x103) eoat~ng RF
_ ~ieGlo Po~ LAY
&l~pl#~:
Antl-IgM 568735 859 3~5 I~mune ~erum3 In~tial bleeding 38 109 23 0 Subsequent ~l~eaing 116 ~D5 23 ~D

1. ~a~e B6 in ~able 6~), except: 1) the listea RF~ were coated ~t 2 ~lc~ogram8 RF per ~ill~llter BB~. I~mune ~era were a~n~yed ~t the 1:1000 ~ilution, ~nd ~he absorbance at 40~ nano~eter~ w~ ~easured a~t~r overnight lncubation ln ~ cold r~om ~t 6 ~-7 3) Rnti-~gM ~rom the r~bbit immunlzo~ w~th IgM-~F 8ie, ~nd the ~b~srb~nce at ~05 n~no~eter~ wa~ mea~urea ~ter l-hous incu~ation at room temperst~re.
2. ~ntt~IgM was u~e~ ~o ~easure ~he rel~tlve qu~n~ltle~ o~ ~ch Ig~-RF.
3. ~he rabbi~ S~l) was init~lly ~le~ one ~sek ~fter the ~irst ln~ct~on o~ gl~taraldehyae cro~-llnkea PSL2. ~ub~equen~ly; the r~bblt ~a~ furthor boo~d ~ith two ~ore ~n~e~tion~ ~one of cross-linkea P~L2 ana ~hen o~e of PSL2-XL~ con~ugato~) to lncrea~e ~he ~nti~R~ 8i~ t~t~ry ~nfl blea one month a~er ~h~ la~t in~tlon.
4 ~he nu~bes~ ~pre~ea ~ro A~05 ~ ~he i~m~ne 3eru~ ~ter ~btractlng ~4~ of ~he c~ntrvl norm~l rabbit ~eru~.

5~ ~D~t aete~M~ nea.

~æ5~i79s Xt b~d been ~hown th~t several human IgM-~F
par~protein reacted with rabbit IgG Igunkel et al., 1 , 129 (1974)1. ~ence, it was necessary to prove that ~he ~n~eractions between PSL2-~nduced antibody ~nd those reactive IgM-RFs were due to the specific binding activity of the PSL2-induced ~ntibodies, ~nd not to the non-~pecific interactions of the human IgM-RFs with rabbit IgG in ~he antisera. ThuS, the reaGtivity of the PSL2-~ndu~ed antibody with the i~olated chains of these IgM-R~s was determined by the We t~rn blot method (Towbin et al., upra), As shown in Figure 75C), the PSL2-indu~ed ~ntibodies react~d egually wPll with the light (L) chains o$ IgM-~F Sie, Gl~ ~nd Teh. Due to the extremely small quantity of IgM-RF
Teh that we had, snly 1/5 eguivalent we~gh~ o I9M-RF
Teh was used.
Thu~, an~ther Western blot was performed ~o ~onfirm the ~ntibody reac~ivity with ~he RF-~eh ~Figures 7(D) ~nd 7~E)3. One ~t of ~amples w~
react~d wi~h anti-IgM ~nti~erum to ~how the rel~t~ve ~uantities o~ difere~t RFS IFi9Ure 7 ~D~
~utoradiogr~phs were developed for ~ 3-~ay periodJ

: L2567~:;

These results demonstrated that PSL2-induced antibodies reacted with RF-Teh. In addition, the antibody reacted very weakly with the L chains of IgM-RF Pom and pooled IgG. However, most importantly, it reacted neither with the L chain of the IgM-RF Lay, nor- with the heavy (H) chains of all IgM-RFs. In contrast, PSH3-induced antibodies reacted only with the H chain of the IgM-RF Sie.
These results collectively demonstrated that PSL2-induced antibodies recognized a public idiotype on the L chains of some human IgM-~Fs.
12. The Structural Correlate of the RF Public Idiotype Defined b the PSL2-Induced Antibodie~s To prove that PSL2-induced antibodies bound to a specific PSL2-determined idiotype on the reactive IgM-RFs, inhibitions of the antibody bindings to both intact RF tsie), and the isolated L
chain of RF-Glo by the PSL2 peptide were determined.
Figure 8(A) shows that the antibody binding to RF-Sie was completely inhibited by the free PSL2 peptide in solution, while the binding was not affected at all by the control PSH3 at the same concentrations (da~a not shown). Moreover, the control PSH3 at 5000 ng~ml completely inhibited the bindings of PSH3-induced antibodies to RF-Sie. Thus, these data indicated that the PSL2-induced antibodies recognized a specific PSL2-determined idiotype on the RF-Sie.
Subse~uently, isolated L chains were prepared from the Ig~-~F Glo. As expected, PSL2-induced antibodies reacted with the L chains of RF-Glo lFigure 8(B)~. Moreover, the binding was completely inhibited by the free PSL2 in solution, but not PSH3 lFigure 8(B)l. This suggests that the L
chains of RF-Sie and RF-Glo share an homologous ~2~6~795 second complementarity-determining region (CDR-2~, and that the L chain-C~R-2 is the structural correlate of the RF public idiotype defined by the PSL2-induced antibodyO
13. The Anti-~F Activity Resides ln the Anti-peptide Antibod~
To further characterize PSL2-induced antibodies, the IgG fraction was prepared from the antiserum containing high titer of anti-RF activity (e.g., the bleeding from immune rabbit #l one month after one more injection of glutaraldehyde cross-linked polypeptides and another injection of polypeptide-protein conjugates)O Then, the IgG
preparation was adsorbed with BSA to remove non-specific binding activities~ and the specific antibodies were purified by a PSL2-coupled column.
Table 8 shows that anti-RF activities were adsorbed by, and eluted from the PSL2-coupled column, indicating that anti-RF antibodies of the immune sera were induced by peptide directly, but not indirectly through some pathways of the immune network.

~2~i79~;;

ADSORPTION AND ELUTION OF THE ANTI-RF ACTIVITY

Absorbance at 405 nanometers ~x103) Antigens Coated Bovine Serum Albumin Polypeptide IgM-RF Sie (BSA) (PSL2) Samples2 Original 24 1319 303 BSA-adsorbed 0 1214 250 BSA, peptide adsorbed 0 38 25 Eluate 12 1859 730 Eluate (1 microgram/ 0 1616 375 milliliter) 1. An IgG fraction (100 milligrams) of the antiserum was first adsorbed with a 3 ml BSA-coupled column (5 mg/ml gel), and then loaded onto a 3 ml PSL~-coupled column (1.6 mg/ml gel). After incubation for 15 minutes at room temperature (23C.), the effluent was collected. Subsequently, the column was washed extensively, and the bound material was eluted with 0.1 M glycine-HCA at pH 2.5.
2~ Unless indicated otherwise, all samples at a concentration of 10 micrograms per milliliter were assayed as described with reference to Table 6~B).
Furthermore, Table 9 shows that the affinity-purified anti-PSL2 antibody reacts strongly with Sie and Glo but reacts weakly with Pom.

~:2S67~5 ANTIBODIES WITH OE RTAIN HUMAN MONOCLONAL IgM-RFsl Absorbance at 40S nanometers (x103) Coatin~ RF _ _ Sie Glo Pom Anti-PSL2 Antibody1136 1900 256 Antibody-depleted Ig57 393 56 Differential binding 1179 1507 200 1. The anti-PSL2 antibody (the eluate in Table 8) and the antibody-depleted Ig (the peptide-adsorbed fraction in Table 8) were assayed for their reactivities with various IgM-RFs, coated onto microtiter plates at 2 micrograms per the absorbance at 405 nanometers was deter.mined.
2, The absorbance at 405 nanometers was determined after overnight incubation at room temperature (23C.).
3. Differential binding equal to (A~05 f antibody - A405 of antibody-depleted immunoglobulin).
14. The Anti-peptide Antibody Reacts with Intact Antibody Molecules in their Native For~s During this study, it was suspected that the denatured IgM-RFs coated onto wells were responsible for interactions with anti peptide antibodies. To rule out this possibility~ it was shown that IgM-RF
5ie in liquid l~hase was still recvgnized by anti-peptide antibodies. Table 10 shows that enzyme-conjugated IgM-RF Sie (AP-IgM-RF Sie~ bound to ~2~;67~5 wells precoated with anti-PSL2 antibodies, but not the rabbit IgG depleted of anti-PSL2 antibodi~s.

NATIVE IgM-RF SIE ~IN THE LIQUID PHASE) REACTS WITH POLYPEPTIDE-INDUCED
ANTI-IDIOTYPE ANTIBODIES COATED ONTO WELLS

Absorbance at 405 nanometers (x103) .
_ Proteins Coatedl Anti-peptide Antibody-antibody depleted Ig ~P-IgM Sie:2 ~Micrograms/Milliliter) 2~5 603 22 _ _ ~__ 1. Polypeptide-induced anti-idiotype antibodies ~the eluate of Table 8) and antibody-depleted IgG ~the peptide-adsorbed IgG of Table 8) at 3 micrograms/milliliter were used to coat separate wells~
2. IgM-RF Sie was labeled with alkaline phosphatase by glutaraldehyde.
3. AP IgM-RF Sie of specified concentrations were distributed to wells in duplicate and the plate was incubated for 3 hours at room temperature (23C). After washing, the substrate was added to wellst and the absorbance at 40~ nanometers was measured after 1 hour at room temperature (23C).

Furthermore, as shown in FIGURE 9, th~ binding of AP-IgM-RF Sie ~o the bound anti~PSL2 antibodies was 35 specifically inhibited by ~he PSL2, but not the SEi7~5 control PSH3. This demonstrates that the binding of the intact IgM-RF Sie to anti-PSL2 antibodies is due to its PSL2-determined epitope, and is not due to its RF activity. Taken together, these data demonstrate that the anti-PSL2 antibody did react specifically with the PSL2 determined idiotype on an intact IgM-RF
Sie in its native form.
V. Diagnostics The above polypeptides can also be used as a portion of a diagnostic composition for detecting the presence of antigenic proteins and antibodies.
A diagnostic reagent system embodying this invention is useful for the determination of the presence of an increased amount of protein compared to the amount of that protein normally present. This system comprises in separate containers ~a) a first reagent and (b) a second, reagent, both in biologi-cally active form, along with an indicating group.
The first reagent contains one of the before-described synthetic polypeptides, a combination of such polypeptides, or conjuyates prepared therefrom. The second reagent includes polyamides containing idiotypic regions of antibodies raised to the synthetic polypeptides or their conjugates. The idiotype-containing polyamides of the second reagent may be substantially intact antibodies or Fab or F(ab')2 antibody fractions whose preparations were described before. An indicating group may also be provided to the syst~m and can be initially bound to or free from either of the two reagents.
Admixtur2 of predetermined amounts of the first and second reagents in the l?resence of a predetermined amount of a body component ~o be assayed results in an immunoreaction. The degree or ~25~5 amount of the immunoreaction so produced is different from a known immunoreaction amount when an increased amount of protein is present in the body component.
The amount of immunoreaction i~ typically diminished when an increased amount of the protein is present in the body componentr compared to the usual amount present.
Another diagnostic system comprises the before-discussed anti-idiotype antibodies in biochemically active form and an indicating means.
In this system, the anti-idlotype antibodies or their Fab or F(ab')~ fractions react with an antigen in a sample to be assayed to form an immuno-reactant whose presence is signalled by the indicating means.
The above system may also include second antibodies raised to antibodies of the same class and from the same species as the anti-idiotype antibodies as part of the indicating means. For example, where the anti-idiotype antibodies are raised in rabbits, ~0 commercially available ~oat anti-rabbit antibodies may be used. Such second antibodies conveniently include a label such as a linked enzyme like horseradish peroxidase or a radioactive element like 125I as the signal indicator.
Exemplary diagnostic reagent systems include enzyme-linked immunosorbent assays (E~ISA) wherein the indicator group is an enzyme such as horseradish peroxidase which is bound to the above antibody or another antibody, and radioimmunoassays in which the indicating group is a radioactive element such as 1~5I present in either the synthetic polypeptide or the antibody raised thereto.
VI. Discussion The antigen binding site of an antibody is formed from the three dimensional folding of the ~56~:~

variable regions of the heavy and light chains.
Padlan et al., Nature, New Biol, 245, 165 (1973).
Diversity in antibody specificity is derived from variation in the primary (linear) amino acid sequences of the variable regions. The primary amino acid sequence dictates the three dimensional folding pattern. Thus, different "shape" binding sites with different active residues in the binding site are generated from different primary amino acid sequences. There are three regions of extreme variability in the primary sequence within the variable region. These sites are termed hypervariable regions. Capra et al., Proc. Nat.
Acad. Sci. U.S A., 71, 845 ~1974).
As described herein, the anti-idiotype antibodies of the present invention were raised against polypeptides that correspond to sequences in the heavy chain o~ a particular immunoglobulin. The antibodies raised could recognize and distinguish the corresponding individual idiotype, for example, in the third hypervariable region of that immunoglobulin. The data further strengthen the generally accepted conclusion that the idiotope is defined principally by the heavy chain.
Notwithstanding, it is not possible to exclude the light chains as playing a role in the expression of individual idiotopes. The individual idiotope of the J558, for examplet is not expressed by separated J558 heavy and light chains nor by J558 heavy chains reconstituted with an in~ppropriate light chain as described in Carson et al., Proc. NatlO Acad. Sci.
USA, 70, 235 (1973). Thus, in the native immunoglobulin or Fab or F(ab')2 fragments, a light-heavy chain interaction may be required to stabilize the heavy chain variable region in the ~256795 conformatiOn necessary for expression of the particular idiotope~ -The present invention can be applied tomanipulating the immune response to cure or combat autoimmune diseases. In a like manner, this invention can be used to reduce transplant rejections. Such application of the invention is further explained below.
Invasion by an antigenic substance generally results in a polyclonal response to the anti~en as described by Hansburg et al., J. Immunol., 194, 1406 (1977); Briles et al., J. Ex~. Med., 152, 151 ~1980) and Ceney et al., J. Immunol., 128, 1885 (1982).
A polyclonal response is likely to occur for two reasons. First, an antibody or the receptor on a B-cell has a precise specificity. However, antibodies and B-cell receptors have the capability of reacting to antigenic determinants which do not exactly match the binding site "shape" but are related in "shape". This is known as cross-reactivity~ Thus, a particular antigen is likely to fall within the range of cross-reactivity of several B-cell clones. Consequently, several antibodies with different variable regions are produced against an antigenic determinant. Second, an antigen is likely to p~ssess several antigenic determinants. Each of these antigenic determinants is likely to elicit its own antibody response, thereby producing antibodies that contain different variable regions.
In summary, invasion by an antigenic foreign substance results in a polyclonal antibody responseO
These antibodies do not all share identical variable regions. This in turn results in a variety of ~2567~;

idiotypes being expressed by the antibodies directed ; against an antigen.
It was desirable to raise anti-idiotype antibodies to the individual idiotype found in the third hypervariable region of J558 and M104 myeloma proteins, as previously stated. Two different sets of synthetic polypeptides were investigated. The first set included the first seven adjacent amino acid residues in the J segment and 3 adjacent amino acids in the V segment. This synthetic polypeptide was designed to be long enough to ensure the eliciting of an anti-polypeptide immune response.
The synthetic polypeptides were designated hV3M and hVBJ and correspond respectively to the M104 and J558 myeloma proteins.
The amino acid residue sequences selected or hV3M and hV3J polypeptide result in a substantial portion of the sequence encoded by the Jl gene segment being common to both polypeptides. Thus, there was the possibility that much of the anti-polypeptide response would be directed against this common sequence. This could possibly result in an obscuring of the antibody activity that is specific for the different D-segments.
The second set of synthetic polypeptides used in that study includes the minimum cognate sequence necessary to form an antigenic determinant as described by I,erner et al., Nature; 299~ 59~
. (1982)o These synthetic polypeptides were designated 3MN and 3JN and correspond respectively to the M104 and J558 myeloma proteins.
Antisera were raised against the various synthetic polypeptides. The ability of these antisera to distinguish between the two myeloma proteins was determined. Eight of the twelve ~;67~i antisera were able to distinguish between the two myeloma proteins in a denatured form (Figure 2). In addition, these antibodies displayed a high level specificity for the natural protein that corresponds to the immunizing polypeptide~ Six of the twelve antisera could distinguish between the natural myeloma proteins (i.eO nct denatured~ and exhibited a high level of specificity for the protein corresponding to the immunizin~ agent.
In summary, at least half of the immunized animals produced antibodies capable of descriminating the individual idiotypes of the M104 and J558 proteins, irrespective of whether the proteins were native or denatured. Thus, these results show that anti~idiotype antibodies specific for a predetermined particular idiotypic antigenic determinant can be raised using synthetic polypeptides. That is, the - antibodies distinguished between two proteins in-a specific manner corresponding to the immunizing polypeptide. Furthermore, these anti-idiotype antibodies have the capability of descriminating between idiotypes which differ by as few as two amino acid residues.
Three of the antisera (numbers 5~ 8 and 10) appear to have absolute specificity within experimental limits for their target idiotype. The other antisera which distinguished the two myeloma proteins displayed a high level specificity for their target idiotype It may be possible to increase the frequency of antisera with absolute specificty by utilizing shorter cognate sequences and spacers; as was done with immunizing polypPptides 3MN and 3JN.
Not enough animals have yet been immuni~ed to make the appropriate comparison ~256~95 It should also be noted that antibodies raised against the shorter cognate seqùence polypeptides, 3MN and 3JN, were generally of lower titre than the antibodies raised against synthetic polypeptides with a longer cognate series, hV3M and hV3J (data not shown). The binding site affinity of the various antisera for their corresponding immunizing polypeptide was determined using enzyme linked immunosorbant assay as described in Green et al., Cell, 28, 477 (1982)o All the polypeptides induced antibodies which bound the immunizing polypeptide. However, the antibodies elicited by the 3MN and 3JN were generally of lower titre. It is not clear whether this is due to the use of a shorter conjugate.
As also described herein, a synthetic polypeptide, corresponding to the third hypervariable region of the heavy chain of the human monoclonal IgM-RF Sie, has also been used to induce the production of an anti-hypervariable region antibody.
The anti-idiotype antibody binds to the intact immunoglobulin M molecule (Sie) and to its isolated chains, but does not bind to other IgM paraproteins or to pooled human IgG. Moreover, the binding of the antibody to the intact IgM was inhibited specifically by the free polypeptide.
These results also demonstrate that a specific anti-idiotype antibody of predefined specificity; i.e., that bind to a particular idiotypic antigenic determinant, can be induced by a synthetic polypeptide, and that such an anti~idiotype antibody recognizes an idiotypic antigenic determinant formed by the known hypervariable region. In one embodiment of the present invention, that predefined specificity does not extend beyond ~25~ii795 recognition of a private idiotype of a single antibody molecule.
In a preferred embodiment, however, the use of synthetic polypeptides to induce the production of anti-idiotype antibodies has been extended to generate an anti-idiotype antibody of a public or cross-reactive idiotype. Among human monoclonal IgM
rheumatoid factors tIgM-RF), two major cross-reactive idiotypes (e.g., Wa and Po) have been describedO
[Kunkel et al., J. Exp. Med., 137, 331 (1973)]. The Wa group includes 60% of monoclonal IgM-RFs, and the expression of the Wa cross-reactive idiotype seems to depend on the L chains of reactive IgM-RF.
lKunkel et al., J. Exp. Med., 139, 128 (1974) and Andrens et al., supra.].
A murine monoclonal antibody (designated mab 17-109) has been prepared that reacts with two cross-reactive (~) idiotypic IgM-RFs (e.g., Sie and Glo, but not with two Wa-cross-reactive (-) idiotypic (e.g., Lay and Pom). [Carson et al., Mol. Immunol., 20, 1983)~. In addition, mab 17-109 reacted with the light chains, but not with the heavy chains, of Wa-cross-reactive (~) idiotypic IgM-RFs. A
comparison of the reported amino acid sequences of the L chains of Wa~cross-reactive (~) idiotypic IgM-RFs (See Figure 6) indicates that these L chains have the same amino acid sequence in the second hypervariable region. Thus, a synthetic polypeptide corresponding to that region was prepared and was used to induce the production of an anti-idiotype antibody of a cross-reactive hypervariable region.

~567~3S

~ he foregoing is intended as 111ustratiYe of the present invention but is not limiting. Numerous variations and modifications.can be effected withou~
departing frcm the ~pirit and 5cope of the novel ~oncepts of ~he invention. It is to be understood that no limitation with respect to the 5pecific ~ntibodies, compositions and uses described herein i8 intended or should be inferredO

Claims (14)

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

1. A synthetic polypeptide containing 6-40 amino acid residues and having an amino acid residue sequence that substan-tially corresponds to an amino acid residue sequence of an idiotypic antigenic determinant of an immunoglobulin and free of naturally occurring protein or a fragment thereof, said synthetic polypeptide having the capacity alone, as a polymer or as a conjugate of said polypeptide bound to a carrier, when injected into a host in an effective amount and in a physiolog-ically tolerable vehicle of inducing the production of anti-bodies to said antigenic determinant of said immunoglobulin.

2. The synthetic polypeptide according to claim 1 wherein said synthetic polypeptide contains from about 8 to about 20 amino acid residues.

3. The synthetic polypeptide according to claim 1 including the sequence of amino acid residues, taken from left to right and in the direction from amino-terminus to carboxy-terminus, represented by a formula selected from the group consisting of:
CysAlaArgAspTyr(Arg)Asp(Tyr)TrpTyr PheAspValTrpGly;
GluGluCysProProProAlaArgAsp Tyr(Arg)Asp(Tyr)TrpTyrPhe;
GluTrpLysGlyGlnValAsnValAsn ProPheAspTyr;
GluTrpLysGlyGlnValAsnValAsn ProPheAspTyrGlyGlyCys; and TyrGlyAlaSerSerArgAlaThrGlyIleProAspArgCys wherein each amino acid residue in paretheses is an alternative to the immediately preceding amino acid residue.

4. The synthetic polypeptide according to claim 1 where-in said immunoglobulin is a human rheumatoid factor of class IgM.

5. A vaccine against infection by a pathogen having an immunoglobulin with a particular idiotypic antigenic determinant comprising an effective amount of a snythetic polypeptide contain-ing 6-40 amino acid residues and having an amino acid residue sequence that immunologically corresponds substantially to an amino acid sequence of an antigenic determinant of an immunoglob-ulin and free of naturally occurring protein or a fragment thereof, and a physiologically acceptable diluent, said vaccine, when introduced into a host, being capable of inducing the production of antibodies in the host that immunoreact with said antigenic determinant and protect the host from infection.

6. The vaccine according to claim 5 wherein said synth-etic polypeptide is bound to a carrier selected from the group consisting of keyhole limpet hemocyanin, keyhole limpet hemocy-anin in incomplete Freund's adjuvant, alum, keyhole limpet hemocyanin-alum adsorbed, keyhole limpet hemocyanin-alum absorbed-pertussis, edestin, thyroglobulin, tetanus toxoid and tetanus toxoid in incomplete Freund's adjuvant.

7. An antibody raised in an animal host to a synthetic polypeptide containing 5-40 amino acid residues and having an amino acid sequence that immunologically corresponds substant-ially to an amino acid sequence of an idiotypic antigenic deter-minant of an immunoglobulin and free of naturally occurring protein or a fragment thereof, said antibodies having the capac-ity to immunoreact with said antigenic determinant and protect the host from said immunoglobulin.

8. A diagnostic kit for assaying for the presence of an antigenic determinant of an immunoglobulin comprising in biochemically active form the antibody of claim 7, said antibody immunoreacting with an admixed sample to be assayed to form an immunoreactant whose presence is signalled by an indicating means.

9. The diagnostic kit according to claim 8 further including an indicating means comprising enzyme-linked second antibody, said second antibody being raised to an antibody of the same class and from the same species as the first named anti-body, and signalling said immunoreaction by binding to said first named antibody present in said immunoreactant, said signal being indicated by the reaction of said linked enzyme with an added substrate.

10. The diagnostic kit according to claim 8 wherein said indicating means comprises a radioactive element bonded to said antibody and said immunoreaction causes precipitation of said immunoreactant containing said radioactive element.

11. A diagnostic kit for assaying for the presence of an idiotypic antigenic determinant of an immunoglobulin in a body component comprising in separate containers (a) a first reagent that contains in biologically active form a synthetic polypeptide containing a sequence of about six to about forty amino acid residues that immunologic-ally corresponds substantially to an amino acid residue sequence of said antigenic determinant and free of naturally occurring protein or a fragment thereof said polypeptide, when linked to a carrier as a conjugate and introduced in an effective amount as a vaccine into a host animal, being capable of inducing production of antibodies in the host that immunoreact with said antigenic determinant and protect the host from said immuno-globulin; and (b) a second reagent that contains in biochemically active form an anti-idiotype antibody that immunoreacts with said synthetic polypeptide; along with a means for indicating the presence of an immunoreaction between said first and second reagents;
said first and second reagents, when admixed in predetermined amounts in the presence of a predetermined amount of body component to be assayed, providing an amount of immuno-reaction signalled by said indicating means, the amount of said immunoreaction being different from a known immunoreaction amount when an immunoglobulin including said antigenic deter-minant is not present in said body component.

12. A method of forming an anti-idiotype antibody comprising the steps of:
(a) administering to a host a synthetic polypeptide containing 6-40 amino acid residues and having an amino acid residue sequence that substantially corresponds to an amino acid residue sequence of an idiotypic antigenic determinant and free of naturally occurring protein or a fragment thereof in an amount sufficient to induce the production of an antibody to an antigenic determinant of an immunoglobulin, said synthetic polypeptide having an amino acid residue sequence that substan-tially corresponds to the amino acid residue sequence of an idiotypic antigenic determinant of said immunoglobulin; and (b) recovering the anti-idiotype antibody so produced.

13. The method according to claim 12 wherein said idio-typic antigenic determinant is a variable or hypervariable region of said immunoglobulin.

- 72a -

14. A method for assaying for the presence of an idio-typic antigenic determinant of an immunoglobulin in a sample comprising:
(a) providing a synthetic polypeptide of claim 1;
(b) admixing a predetermined amount of said poly-peptide with a predetermined amount of sample to be assayed for the presence of the idiotypic antigenic determinant to which the polypeptide binds;
(c) maintaining that admixture for a period of time sufficient for said polypeptide to bind to idiotypic anti-genic determinant present in the admixed sample; and (d) determining the amount of binding between said synthetic polypeptide and said idiotypic antigenic determinant of said immunoglobulin.