Antigenic Modification of Polypeptides

It is well known that antibodies are generated in humans and in other animals in response to the presence of foreign antigens. It is also known to confer immunity on an animal by administering an antibody formed elsewhere. For instance, patents to Michaelson (U.S. Pat. No. 3,553,317), Friedheim (U.S. Pat. No. 2,388,260), Reusser (U.S. Pat. No. 3,317,400) and Peterson (U.S. Pat. No. 3,376,198) relate to production of antibodies, which when injected into an animal of a different species or into a human being cause passive immunization. In patents to Fell (U.S. Pat. Nos. 2,301,532 and 2,372,066), the patentee refers to active immunization using modified histamine in such animals as horses, cows, etc. In a paper by R. G. Edwards in the British Medical Journal, Vol. 26, pages 72 to 78, published in 1970 on "Immunology of Conception and Pregnancy", he surveys the literature regarding the possibilities of utilizing immunological methods to influence or control fertility, surveying first production of antibodies against testes or spermatozoa. Much of the literature surveyed is directed to the production of foreign antibodies which are injected into the subject (passive immunization).

Hormone antibodies have been studied for a long time and the effect of specific antisera have been recorded for many years. It is known that administration of certain antibodies during pregnancy can suppress implantation or cause fetal resorption. Several different approaches have been tried ranging from the induction of near permanent infertility in the case of agglutination of spermatozoa in the male to the disturbance of a single pregnancy by passive immunization with antibodies.

There are serious limitations to the use of passive immunization procedures for human therapy. Since the antibodies are practically produced only in non-human animals, the repeated injection of animal proteins into humans is known to produce serious reaction in many individuals.

British patent Specification No. 1,058,828 discloses that small molecules, referred to as "serological determinant peptides", can be coupled to large protein molecules, such as cattle albumin and the resultant conjugate then may be injected into animals for antibody production. The document lists proteins from which the seriologically determinant peptides may be isolated prior to being used in the process taught, the collection including viruses and bacteria whose surface component has the characteristics of a protein, toxins and hormones having protein structure and enzymes. No specific hormone is named in the document and no utility of anti-hormone immunization is described. The patent specification references a publication entitled: "The Specificity of Serological Reactions", Dover Publications, Inc., New York, 1962, Chapter V, "Artificial Conjugated Antigens" by K. Landsteiner. This publication outlines various chemical methods and applies them passively to bind various toxic substances in the blood such as arsenic. Thyroxine data provided in the publication suggests that such methods may be applied to hormones without indicating the therapeutic application, the publication teaching that specific antibodies may be formed to the small molecules and these antibodies are capable of neutralizing the biological action of a large protein from which the small peptide was a part.

Recently it has been discovered that doses of certain steroids consisting of synthetic non-protein hormones ("The Pill") when administered at stated intervals usually confer protection against pregnancy for a short time (possibly a month). This medication has sometimes been found to create undesirable side effects in creating undesirable metabolic changes and sometimes changes in the blood clotting mechanisms. Moreover, the effect of each dose is of such short duration that often it is of limited application, particularly in remote areas to persons not readily instructed on proper and continuing use.

There is need therefore of an effective safe method of creating a temporary but relatively long-time immunity against pregnancy which does not have serious side effects. There is also a need for an effective safe method of terminating a pregnancy soon after conception which does not have serious harmful side effects. Such need may be met by the neutralization of a reproductive protein which is necessary for the normal events of conception and/or gestation.

There is also a need for a means for control of various disease states or maladies caused or influenced by unusual excesses of certain polypeptides such as gastrin, angiotension II, or somatomedian. It is believed that this invention meets this need safely and effectively.

This invention is concerned (1) with the production of antigens for the purpose of active immunization, (2) with the antigens so produced, and (3) with the use of said antigens. More particularly, the invention relates to antigens consisting of natural protein reproductive hormones, non-hormonal proteins, specific fragments of such hormones and proteins and synthetically derived portions of said hormones and proteins, all modified as will be indicated more fully hereinafter. For the sake of simplicity, hereinafter in this specification and in the claims, these antigens are collectively referred to as modified polypeptides.

The invention is directed in one aspect to the use of modified polypeptides in actively immunizing an animal, particularly mammals, against the biological action of endogenous unmodified non-hormonal natural protein and/or hormone. The state of immunity arises because of the creation of antibodies which act against both the antigenic modified polypeptide and its endogenous counterpart which is neutralized (rendered biologically ineffectual) as a result of the existence of said antibodies. The immunity may take place because of the inability of the antibody to distinghish between the modified polypeptide and the naturally existing protein, but it is uncertain that this is in fact the situation. In effect, the invention provides, in one aspect, for the isoimmunization of a primate animal.

A more specific aspect of this invention relates to the modification of protein reproductive hormones by adding certain numbers of foreign moieties to each hormone molecule, or hormonal fragment. The modification must be sufficient to cause the body to create antibodies to the modified hormones which will neutralize or inhibit the biological action of the natural hormones produced by the body. Thus, the modified hormones become antigenic and cause the production of antibodies which disrupt the natural processes of conception and/or gestation. The term "protein reproductive hormones" includes those hormones essential to the normal events of the reproductive process.

According to a further aspect of this invention, a disease state which can be treated by application of the technique of the instant invention is the digestive disorder known to those skilled in the medical field as the Zollinger-Ellison Syndrome. This syndrome or disease state is generally described as a condition in which a hypersecretion of the polypeptide gastrin, which is produced in the pancreas, brings about a state of hyperacidity in the stomach which results in a chronic digestive disorder. Heretofore, the only effective treatment for this disease state was the surgical removal of a part or total removal of the subject's stomach. Although survival of such patients is usually not threatened, the medical state and life style of such individuals is severely affected by such treatment.

Treatment of such subjects with hapten coupled (produced according to the general method described herein) or otherwise chemically modified gastrin can be used to enhance the production of antibodies against the hypersecretion of gastrin and thereby alleviate or reduce the symptons of this disease without surgical intervention. Sufficient reduction by immunological means of this substance in the system of the body would be sufficient to avoid the complicated and serious consequences of the surgical treatment currently in use. In practice, an effective amount of modified gastrin is simply injected into the patient as required to accomplish the control of the flow or presence of gastrin.

Another serious medical problem which is treatable by the application of the technique of the instant invention is that of hypertension. In general terms, the state of hypertension is the abnormal level or fluctuation of one's blood pressure. The blood pressure in an individual is controlled by many physiological processes in the body. However, one major substance affecting the regulation of such pressure is the hormonal polypeptide known as angiotension II. In certain states of high blood pressure (hypertension) it is difficult to medically control the secretion into, and therefore the level of angiotension II in, the circulatory system. By appropriate modification of this hormone and subsequent immunization with this altered modified proteinaceous hormone, it is possible to reduce the secretion of angiotension II in patients with chronically elevated hormone levels. The predictable and controlled reduction of this substance is beneficial to certain patients with chronic problems of hypertension. Modified angiotension II can be produced by the general protein modification technique described herein. The resultant modified angiotension II is simply injected into the patient in an amount sufficient to induce antibody response sufficient to control or regulate unmodified angiotension II to the desired degree.

A further embodiment of the present invention is the treatment of diabetes and associated micro and macro vascular diseases. Currently, the treatment of diabetes is limited to dietary and/or drug treatment to regulate blood glucose levels. Recent scientific data support the concept that growth hormone and somatomedian (both polypeptides) are intimately involved in the disease syndrome. These substances can be modified by the technique described herein and used in an effective amount to control the progress of this disease. In practice, modified growth hormone or modified somatomedian is injected into the body to develop antibodies for control of the normally secreted hormones.

Another health problem that can be treated by the use of the concepts of this invention is that of certain endocrine or hormone dependent breast tumors or cancers. Certain of these cancers have been shown to be dependent upon the abundant secretion of the hormone prolactin for their continued survival. The inhibition of the secretion of prolactin has been shown to diminish the growth rate and the actual survival of certain of these tumors. The immunization of such subjects with the hapten coupled or otherwise altered prolactin produced as described herein, would result in the systematic reduction of the level of this hormone circulating in the system and consequently, may result in the regression or remission of tumor growth. The consequence of this treatment would be far more favorable in terms of effective treatment of this disease since surgical removal of the breasts is a principal method of treatment currently available. It should be understood that this treatment should be effective for only those tumors that are dependent upon the secretion of prolactin for survival.

Investigators also have determined, for example, that certain polypeptide entities are supportive factors to and secretions of neoplastic diseases in both man and other animals. These entities have biochemically, biologically and immunologically close resemblances to hormones, particularly to Chorionic Gonadotropin (CG), as well as to Luteinizing Hormone (LH). By applying the isoimmunization techniques of the invention, the function of such polypeptides or endogenous counterparts can be neutralized to carry out regulation of the malignancy. For example, tumors in both male and female primates may be treated by isoimmunization procedures developing antibodies to Chorionic Gonadotropin or Luteinizing Hormone or the noted entity analogous thereto. Further, neoplasms in primate females may be regulated by isoimmunization procedures developing antibodies to endogenous Follicle Stimulating Hormone (FSH). This hormone, when associated with a tumor state, tends to aggravate the tumorous condition.

Example XXXIV below shows that a beta-HCG/tetanus toxoid modified polypeptide of the invention confers upon rats substantially complete protection against an injection of tumor cells of the virulent rat mammary adenocarcinoma R 3230 AC, which is associated with CG-like material. The aforementioned modified polypeptide of the invention, when given prior to injection of a dose of tumor cells which causes 100% mortality in unprotected rats, reduces the mortality to zero.

The immunochemical control asserted, as noted, neutralizes the naturally occurring hormone or the above-described entity biologically analogous thereto. As a consequence, the hormone or entity will not be available as would normally be the case, for example, the stimulation of some action of a target tissue. Conversely, the neutralization of the biological activity of the hormone or analogous entity may serve to take away an inhibitory action which it otherwise might assert.

There are certain other disease states that may be treatable by the use of altered or modified hormonal or non-hormonal proteins as antigens. The disease states and the associated substances that may be used as modified antigens for immunological treatment of these diseases will be listed as follows:

(1) modified parathyroid hormone for the treatment of kidney stones,

(2) modified insulin and/or glucagon for the treatment of hyperinsulinoma,

(3) modified thyroid stimulating hormone (TSH) for the treatment of hyperthyroidism, and

(4) modified secretion for the treatment of irritable bowel syndrome.

Another group of polypeptides which can be altered by the procedures described herein and used in the field of human fertility control are specific non-hormonal protein antigens isolated from placental tissue. There is direct evidence that inhibition of substances that are specific to the placental tissue and do not have similar antigenic properties with other antigens from organs in other parts of the body, can result in the disruption of pregnancies by passive immunization. Such specific placental substances when modified to form modified polypeptides by the procedures described herein can be injected into the body of an animal of the same species as an effective fertility control means with the mechanism being active immunization similar to that described for the antigenic modification of hormones. The particular advantages of these substances is that placental antigens are foreign to the non-pregnant female human subject and therefore are unlikely to cause any cross-reaction or disruption of normal body function in the non-pregnant female.

While the invention is useful for the human species it will be appreciated that it is also useful in connection with other animals. Similarly, while the reference herein with respect to fertility control is primarily directed to females, such described techniques may be applicable to males, i.e. FSH, its beta subunit and fragments thereof. Such immunization represents an effective fertility control procedure, providing no physiological consequences are encountered which may be found to react adversely to the performance of other body constituents.

Whether the concerned hormone, non-hormonal protein or specific fragment thereof which is modified is naturally occurring or is a synthetic product is clearly immaterial. A synthetic protein molecule will perform the same function as the naturally occurring one, inasmuch as the body will react in an equivalent antigenic manner.

It has accordingly been discovered by virtue of this invention that it is possible to interfere with or treat various disease states or medical problems which are caused or influenced by certain polypeptides by active immunization of a male or female animal by the production and use of antigens formed by administration of modified polypeptides. The modification of the polypeptides forms antigens which are then administered into an animal in which immunization is to be developed. Said modification is accomplished by attaching to a polypeptide one or more foreign reactive (modifying) groups and/or by attaching two or more polypeptides to a foreign reactive group (i.e., a carrier) or both of the above, so that the body of the animal, recognizing the modified polypeptide as a foreign object, produces antibodies which neutralize not only the modified protein but also the natural protein which is responsible for the disease or medical problem being regulated. In order to produce an effective quantum of antibodies to the antigen or targeted functional polypeptide, it may be advantageous to administer the modified polypeptide together with an immunological adjuvant. The term "adjuvant" is commonly referred to by those engaged in the field at hand as being a substance which will elevate the total immune response of an animal or person to any immunization thereof, i.e. the adjuvant is a nonspecific immuno-stimulator.

FIG. 1 is a chart describing the results of mating four baboons three times following the administration thereto of a fertility controlling antigen according to the invention;

FIG. 2 shows two plots illustrating the antifertility antibody levels maintained within two baboons following the administration of antigens thereto formulated in accordance with the invention;

FIG. 3 shows three dose response lines illustrating the specificity of antibody response to a CG antigen formulated in accordance with the invention;

FIG. 4 shows the levels of antibody to .beta.-HCG produced in rabbits by various modified peptides in Example XXXII below;

FIG. 5 shows the levels of antibody to HCG produced in rabbits by various tetanus-toxoid-coupled modified peptides in Example XXXII below;

FIG. 6 shows the levels of antibody to HCG produced in rabbits by various tetanus-toxoid-coupled modified peptides having differing peptide:carrier ratios in Example XXXII below;

FIG. 7 shows the levels of antibody to HCG produced in rabbits by vaccines containing a tetanus-toxoid-coupled modified peptide and various adjuvants in Example XXXIII below; and

FIG. 8 shows the levels of antibodies to HCG and peptide produced in mice of various species by a vaccine comprising a tatanus-toxoid-coupled modified peptide coupled to various adjuvants in Example XXXIII below.

In an effort to better define the modified polypeptides with which this invention is concerned, it is first considered appropriate to set out more precisely than hereinabove, examples of the natural hormones and natural non-hormonal proteins modified according to this invention. They include Follicle Stimulating Hormone (FSH), Luteinizing Hormone (LH), Chorionic Gonadotropin (CG), e.g. Human Chorionic Gonadotropin (HCG), Placental Lactogen, e.g. Human Placental Lactogen (HPL), Prolactin, e.g. Human Prolactin (all of which are proteinaceous reproductive hormones) gastrin, angiotension II, growth hormone, somatomedian, parathyroid hormone, insulin, glucagon, thyroid stimulating hormone (TSH), secretin, and other polypeptides which could adversely affect body function.

The hormone, Chorionic Gonadotropin (CG) has been the subject of extensive investigation, it being demonstrated in 1927 that the blood and urine of pregnant women contained a gonad-stimulating substance which, when injected into laboratory animals, produced marked gonadal growth. Later, investigators demonstrated with certainty that the Placental Chorionic villi, as opposed to the pituitary, were the source of this hormone. Thus, the name Chorionic Gonadotropin or, in the case of humans, Human Chorionic Gonadotropin (HCG) was given to this hormone of pregnancy. During the more recent past, a broadened variety of studies have been conducted to describe levels of HCG in normal and abnormal physiological states, indicating its role in maintaining pregnancy. The studies have shown the hormone's ability to induce ovulation and to stimulate corpus luteum function, and evidence has been evoked for showing its ability to suppress lymphocyte action. The immunological properties of the HCG molecule also have been studied widely. Cross-reaction of antibodies to HCG with human pituitary Luteinizing Hormone (LH), and vice-versa, has been extensively documented, see for example:

Paul, W. E. & Ross, F. T., Ummunologic Cross Reaction Between HCG and Human Pituitary Gonadotropin. Endocrinology, 75, 352-358 (1964)

Flux, D. X. & Li C. H. Immunological Cross Reaction Among Gonadotropins. Acta Endocrinologic, 48, 61-72 (1965)

Bagshawe, K. D.; Orr, A. H. & Godden J. Cross-Reaction in Radio-Immunoassay between HCG and Plasma from Various Species. Journal of Endocrinology, 42, 513-518 (1968)

Franchimont, P. Study on the Cross-Reaction between HCG and Pituitary LH. European Journal of Clinical Investigation, 1, 65-68 (1970)

Dorner, M.; Brossmer, R.; Hilgenfeldt, U. and Trude, E. Immunological reactions of Antibodies to HCG with HCG and its chemical derivatives.

In Structure-Activity Relationships of Proteins and Polypeptide Hormones (ed. M. Margoulies & F. C. Greenwood), pp. 539, 541 Amsterdam: Excerpta Medica Foundation (1972)

Further, these cross-reactions have been used to perform immunoassays for both CG and LH hormones. See:

Midgley, A. R. Jr. Radioimmunoassay: a method for HCG and LH. Endocrinology, 79, 10-16 (1966)

Crosignani, P. G., Polvani, F. & Saracci R. Characteristics of a radioimmunoassay for HCE-LH. In Protein and Polypeptide Hormones (ed. M. Margoulies) pp. 409, 411 Amsterdam: Excerpta Medica Foundation (1969)

Isojima, S; Nake, O.; Kojama, K. & Adachi, H. Rapid radioimmunoassay of human L. H. using polymerized antihuman HCG as immunoadsorbent. Journal of Clinical Endocrinology and Metabolism, 31, 693-699 (1970)

In addition to providing for the modification of the entire hormone or selected polypeptide, the invention further provides for the utilization of modified subunits, for example the beta subunit of Chorionic Gonadotropin. Of particular interest, such subunits may be fragmented into smaller components herein termed "fragments". The latter can be produced synthetically to exhibit an amino acid sequence sufficiently in analogous correspondence to a predetermined portion of the parent subunit. Such fragments generally are conjugated with a larger molecule or component foreign to the body, which may be termed a "carrier", in order to effectively evoke or raise a sufficient quantum of antibodies. The use of the fragments, as thus conjugated, advantageously provides a high degree of specificity of antigenic reaction to the targeted hormone or its biochemical equivalent, i.e. the antibodies will not react with other body constituents. Of particular interest, the above-discussed cross reaction of HCG and LH can be avoided by utilization of fragments of the respective hormone due to the desirable specificity of response thereto. Thus, when interested in obtaining an immunological reaction against the hormone, HCG, the undesirable immune reaction to the naturally occuring body constituent, LH, may be eliminated. Synthetic equivalents of the fragments offer enhanced practicality both from the standpoint of production costs and necessary maintenance of purity.

As indicated in the above discussion, when considered in isolation with respect to conception and pregnancy, CG only is present in female primates when they are in a post-conception state. However, as discussed above and later herein, an entity at least analogous thereto (having similar immunological properties to HCG) is seen to be present in conjunction with malignancies.

Subunits and fragments of the proteinaceous reproductive hormones include the beta subunit of natural Follicle Stimulating Hormone, the beta subunit of natural Human Chorionic Gonadotropin, fragments including, inter alia, a 20-30 or 30-39 amino acid peptide consisting of the C-terminal residues of natural Human Chorionic Gonadotropin beta subunit, as well as specific unique fragments of natural Human Prolactin and natural Human Placental Lactogen, which may bear little resemblance to analogous portions of other protein hormones. Further with respect to the type of novel chemical entities with which this invention is concerned, one may note for instance the chemical configuration of the beta subunit of HCG. That structure is as follows: ##STR1##

For specifity of antibody action it is necessary that distinctive peptides be isolated or prepared that contain molecular structures completely or substantially completely different from the other hormones. The beta-subunit of HCG possesses a specific chain or chains of amino acid moieties which differ either completely or essentially from the polypeptide chain of Human Luteinizing Hormone. These chains or fragments, when conjugated with a carrier, represent an additional aspect of this invention. Accordingly, the polypeptide Structures (II) and (III) [C-terminal portion of structure I]. ##STR2## whether obtained by purely synthetic methods or by enzymatic degradation from the natural or parent polypeptide, [Carlson et al., J. Biological Chemistry, 284 (19), 6810, (1973)] when modified according to this invention, similarly provide materials with antigenic properties sufficient to provide the desired immunological response. It will be understood, for example, that addition of a polytyrosine chain or a protein macromolecule (carrier) may assist in rendering Structure (II) antigenic so that the resulting administration of modified Structure (II) will provide the desired immunological action against natural HCG.

The beta subunit set forth at structure (I) is seen to represent a chemical sequence of 145 amino acid components. This structure has a high degree of structural homology with the corresponding subunit of Luteinizing Hormone (LH) to the extent of the initial 110 amino acid components. As indicated above, it may be found desirable, therefore, to evoke a high specificity to the Chorionic Gonadotropin hormone or an analogous entity through the use of fragments analogous to the C-terminal, 111-145 amino acid sequence of the subunit. Structure (II) above may be observed to represent just that sequence. Structure (III) is slightly shorter, representing the 116-145 amino acid positions within the subunit sequence.

Further polypeptide chains useful in promoting antibody buildup against natural HCG include the following structures labeled Structures (IV) through (XIV). When modified according to this disclosure, such as by coupling to Ficoll 70 a synthetic copolymer of sucrose and epichlorohydrin having an average molecular weight of 70,000.+-.10,000, good solubility in water, Stokes radius about 5.1, stable in alkaline and neutral media, available from Pharmacia Fine Chemicals, Pharmacia Laboratories, Inc., 800 Centennial Ave., Piscataway, N.J. 08854) or other modifier-carriers such as protein macromolecules described herein, these polypeptides provide immunogenic activity with which this invention is concerned. All of these polypeptides are considered fragments of HCG by virtue of their substantial resemblance to the chemical configuration of the natural hormone and the immunological response provided by them when modified as indicated herein. ##STR3## Structure (IV) will be recognized as incorporating a Cys component at the amino or N terminal which is associated with a Proline spacer sequence. These spacers serve to position the sequence which follows physically distant from the carrier-modifier. The latter sequence may be observed to represent the 138th to 145th amino acid component sequence of the subunit Structure (I). Structure (V) on the other hand, represents an initial sequence corresponding with the 111th to 118th components of the subunit Structure (I) followed by a sequence of six Proline spacer components and a carboxyl terminal, present as Cysteine. The rationale in providing such a structure is to eliminate the provision of sites which may remain antigenically neutral in performance. Structures (IV) and (V) represent relatively shorter amino acid sequences to the extent that each serves to develop one determinant site. Consequently, as alluded to in more detail hereinafter, they are utilized in conjunction with a mixed immunization technique wherein a necessary two distinct determinants are provided by the simultaneous administration of two such fragments, each conjugated to a corresponding, separate carrier macromolecule. Structure (VI) represents the 115th through 145 th component sequence of structure (I). Structure (VII) represents a portion of Structure (I), however, essentially, a sequence of the 111th to 130th components thereof is formed.

Structure (VIII) incorporates two sequences, one which may be recognized in Structure (V) and the other in Structure (IV). These two sequences are separated by two spacer sequences of Proline components and one is joined with an intermediately disposed Cysteine component which serves a conjugation function as described later herein. With the arrangement, two distinct determinant sites are developed in physically spaced relationship to avoid the development of an unwanted artificial determinant possibly otherwise evolved in the vicinity of their mutual coupling. Structure (VIIIa) represents Structure (VIII) with additional Pro spacer residues to provide a widened spacing of determinant sites.

Structure (IX) mimics sequences from Structure (I) with the addition of a Proline Spacer Sequence, A Cysteine Component at the C-terminal, and an Aba substituted for Cys at the 110 position. The Aba designation is intended herein to mean alpha-aminobutyric acid of Cysteine. Structure (X) will be recognized as a combination of Structure (II) with a six residue Proline spacer sequence and a Cysteine component at the C-terminal. Similarly, Structure (XI) combines Structure (II) with a Cysteine component at the C-terminal without a Proline spacer sequence. ##STR4##

Structure (XII) will be recognized as having the sequence of Structure (II) with the addition of Thr-Cys components at its N terminal. Structure (XIII) is similar to Structure (IX) but does not contain the spacer conjugate. Structure (XIV) will be recognized as being similar to Structure (II) with the addition of spacer components at the N terminal and a Cys component for conjugation purposes.

Particularly where the larger whole hormone or subunit type molecular structures are concerned, the number of foreign reactive groups which are to be attrached to the polypeptide and the number of polypeptides to be attached to a foreign reaction group depends on the specific problem being treated. Basically, what is required is that the concerned polypeptide be modified to a degree sufficient to cause it to be antigenic when injected in the body of the host. If too little modification is effected, the body may not recognize the modified polypeptide as a foreign body and would not create antibodies against it. If the number of foreign molecules added to the polypeptide is too great, the body will creat antibodies against the intruder antigen, but the antibodies will be specific to the injected antigen and will not neutralize the action of the concerned natural endogenous hormone or non-hormonal protein, i.e. they will be specific to the modifier.

In general, again considering the larger molecule subunit or whole hormone, it has been found that about 1-40 modifying groups per molecule of polypeptide will be useful in modifying the polypeptide adequately so as to obtain the desired immunological effect of this invention. As will be appreciated by one skilled in the art, this ratio of modifying groups per polypeptide will vary depending upon whether an entire hormone is utilized for modification or whether for instance a relatively small synthetic fragment of said hormone is to be modified. Generally for the larger molecules, it is preferred that 2-40 modifying groups per molecule of polypeptide be used according to this invention. In the instance where the polypeptide is the beta-subunit of HCG it is particularly preferred that about 5-30 and more preferably 10-26 modifying groups per molecule of polypeptide be used. The importance consideration with respect to each modified polypeptide is that the degree of modification be adequate to induce generation by the body of antibodies adequate to neutralize some of the natural hormone or non-hormonal protein against which neutralization is desired, and this will vary with each polypeptide involved, and the degree of correction or change desired for the body function involved.

Modification of the polypeptide is accomplished by attaching various kinds of modifying groups to proteinaceous hormone, non-hormonal proteins, subunits or specific fragments thereof according to methods known in the art.

As is apparent, structures (II)-(XIV) are relatively smaller fragments, usually produced synthetically. To render them capable of eliciting antibody production, it becomes necessary to conjugate them with larger carrier-modifier molecules. Generally about 5-30 peptide fragments will be coupled with one carrier molecule. The body will, in effect, recognize these foreign carriers as well as the sequences represented by the fragments and form antibodies both to the carrier and to the sequences of the coupled fragments. Note that the carrier-modifiers are foreign to the body and thus antibodies to them will not be harmful to any normal body constituents. In the latter regard, it may be found preferable to utilize a carrier which, through the development of antibodies specific to it, will be found beneficial to the recipient.

As indicated earlier herein, it also is preferred that the modification constitute two or more immunological determinants represented on the native hormone as polypeptide structures to which it is desired to evoke an antibody response. The effect is one of heterogeneity of antibody development. Thus, several fragment structures have been described above having at least two distinct amino acid sequences represented in the HCG beta subunit. These sequences may be so spaced as to derive the determinants in mutual isolation, while the spaced sequence fragment is conjugated with a larger, macromolecular carrier. Alternately, the noted mixed immunization arrangement may be utilized wherein a first fragment developing one determinant is conjugated with a first carrier molecule and is administered in combination with a second, distinct fragment which is conjugated with a second carrier molecule, the latter of which may be the same as or different in structure from the first carrier. Thus, each macromolecular carrier must be conjugated with hormone fragments such that each fragment represents two or more immunological determinants. These two necessary determinants can be evolved by mixing, for example, separate conjugate structures, for example Structures (IV) and (V) each of which, through forming antibodies separately to the distinct determinants, will provide a population of antibodies reacting with two separate determinants on the natural endogenous hormone.

Inasmuch as the noted fragments are relatively small as compared, for instance, to a whole hormone or subunit thereof, a criterion of size is imposed upon the selection of a carrier. The carrier size must be adequate for the body immune system to recognize its foreign nature and raise antibodies to it. Additionally, carrier selection preferably is predicated upon the noted antibody heterogeneity requirement, i.e. the carrier must itself evoke a heterogeneous immune response in addition to the fragments. For example, improved response may be recognized where the carrier is varied in structure, e.g. incorporating branching chains to enhance the recognition of both the carrier and the attached polypeptide as being of a foreign nature.

As one example of whole hormone modification, modified diazo groups derived from sulfanilic acid may be attached to the subject polypeptides, see the Cinander et al. and Phillips et al. references cited subsequently for instruction on how this "attachment" is accomplished, and to the extent necessary for an understanding of this invention, such is incorporated herein by reference.

Additional modifying groups for modifying whole hormones or their subunits are those groups obtained by reaction of the polypeptides with dinitrophenol, trinitrophenol, and S-acetomercaptosuccinic anhydride, while, suited for utilization as a carrier-modifier in conjunction with fragments, are polytyrosine in either straight or branched chains, polyalanines in straight or branched chains, biodegradable polydextran, e.g. polymerized sugars such as sucrose copolymerized with epichlorohydrin, e.g. Ficoll 70 and Ficoll 400 (a synthetic copolymer of sucrose and epichlorohydrin having an average molecular weight of 400,000.+-.100,000 intrinsic viscosity of 0.17 dl/g. specific rotation [.alpha.].sub.D.sup.20 of +56.5.degree. available from Pharmacia Fine Chemicals, Pharmacia Laboratories, Inc. 800 Centennial Ave., Piscataway, N.J. 08854) or a polyglucose such as Dextran T 70 (a glucan containing alpha-1, 6-gluscosidic bonds and having an average molecular weight of approximately 70,000, synthesized microbiologically by the action of Leuconostoc mesenteroides strain NRRL B-512 on sucose), serum proteins such as homologous serum albumin, hemocyanin from Keyhole limpet, a marine gastropod mollusk, viruses such as influenza virus (type A, B, or C) or poliomyelitis virus, live or killed, Types 1, 2 and 3 of tetanus toxoid, diphtheria toxoid, cholera toxoid or somewhat less preferably, natural proteins such as thyroglobulin, and the like. Generally, synthetic modifiers are preferred over the natural modifiers. However, carrier-modifiers found particularly suitable for conjugation with the above-discussed fragment structures are Flagellin, tetanus toxoid and an influenza subunit, for example, the preparation of which is described by Bachmeyer, Schmidt and Liehi, "Preparation and Properties of a Novel Influenza Subunit Vaccine", Post-Graduate Medical Journal (June, 1976) 52:360-367. This influenza subunit was developed as a vaccine which incorporates essentially only the two viral proteins, Haemagglutinin and Neuraminidase. Containing substantially only these two essential immunogens, the subunit represents a preparation which does not contain other protein and lipid antigens which may be found to cause undesired side reactions. A secondary benefit may be realized through the utilization for example, of the influenza subunit, poliomyelitis virus, tetanus toxoid, diphtheria toxoid, cholera toxoid or the like as a modifier-carrier, inasmuch as beneficial antibodies will be raised to that modifier-carrier as well as the hormonal fragment conjugated thereto.

Flagellin is a protein described as forming the wall of the main spiral filament of the flagellum. Bacterial flagella, in turn, have been known as the active organelles of cell locomotion, individual flagella (flagellum) occurring in suspension as individual spirals which, upon drying, collapse into filaments which describe a sine wave with a wave length of 2-3 microns and an amplitude of 0.25-0.60 microns. Generally, the flagellum consists of three morphologically distinct parts: a basal structure that is closely associated with the cytoplasmic membrane and cell wall, a hook and the noted main spiral filament.

Purified flagellum is readily obtained by solubilization of flagellar filaments below a pH value of about four, and subsequent removal of the insoluble material by centrifugation or filtration. As a group of related proteins, flagellins from different bacterial species have been predicted to have similar amino-acid compositions. However, the amino acid composition of each flagellin species in unique. Essentially all flagellins are described as containing no or only a few residues of cysteine, tryptophan, tyrosine, proline and histidine. Thus, when conjugated with fragments in accordance with the invention, a thiolactonization procedure or the like is carried out as described later herein.

The molecular weights of various flagellin have been calculated, in all cases the values thereof of the monomeric subunits falling in the range of 30,000 to 50,000. From an immunological standpoint, a flagellin molecule is highly immunogenic. For further and more detailed discourse describing bacterial flagella and flagellin, reference is made to "Advances in Microbial Physiology" 6, 219 (1971), "Bacterial Flagella" by R. W. Smith and Henry Coffler, which publication is incorporated herein by reference.

Tetanus toxoids have been the subject of study and production for many years. The toxoid generally is evolved from a formalinization of tetanus toxin, the latter being a protein synthesized by Clostridium tetani. Immunization currently is carried out utilizing soluble and absorbed tetanus toxoids and suggestions have been made concerning the utilization of fluid tetanus toxoid in complex with antitoxin. Publications describing the toxin and toxoid are numerous, reference being made to the following:

1. Immunochemistry of Tetanus Toxin, Bizzini, et al., Journal of Biochemistry, 39, 171-181 (1973).

2. Early and Enhanced Antitoxin Responses Elecited with Complexes of Tetanus Toxoid and Specific Mouse and Human Antibodies, Stoner et al., Journal of Infectious Diseases, 131, (3), 230-238 (1975).

3. Differences in Primary and Secondary Immunizability of Inbred Mice Strains, Ipsen, Journal of Immunology, 83, 448-457 (1959).

4. Antigenic Thresholds of Antitoxin Responses Elecited in Irradiated Mice with Complexes of Tetanus Toxin and Specific Antibody, Hess et al., Radiation Research, 25, 655-667 (1965).

5. Early and Enhanced Germinal Center Formation and Antibody Responses in Mice After Primary Stimulation with Antigen-isologous Antibody Complexes as Compared with Antigen Alone, Laissue et al., Journal of Immunology, 107, 822-825 (1971).

6. Distinctive Medullary and Germinal Center Proliferative Patterns and Mouse Lymph Nodes after Regional Primary and Secondary Stimulation, with Tetanus Toxoid, Buerki et al., Journal of Immunology, 112, (6), 1961-1970 (1974).

Modification by removal of moieties is also contemplated by this invention. Thus, for example, where certain of the natural proteins have carbohydrate moieties, these carbohydrate moieties may be removed according to methods known in the art by, for instance, N-acetyl neuriminidase or N-acetyl glucosidase, materials useful for removal of specific carbohydrate moieties.

These various means for modification are, as indicated above, known to persons skilled in the art. Certain of these means may be found in the following list of literature references, whereas various others of them may be found elsewhere in the literature by art-skilled persons:

1. Klotz et al., Arch. of Biochem. and Biophys., 96, 605-612, (1966).

2. Khorana, Chem. Rev. S3 145 (1953).

3. Sela et al., Biochem. J., 85, 223 (1962).

4. Eisen et al., J. Am. Chem. Soc. 75, 4583 (1953).

5. Centeno et al., Fed. Proc. (ABSTR) 25, 729 (1966).

6. Sokolowsky et al., J. Am. Chem. Soc. 86, 1212 (1964).

7. Tabachnick et al., J. Biol. Chem. 234 1726, (1959).

8. Crampton et al., Proc. Soc. Exper. Biol. & Med. 80, 448 (1952).

9. Goodfriend et al., Science 144, 1344 (1964).

10. Sela et al., J. Am. Chem. Soc., 78, 746 (1955).

11. Cinader et al., Brit. J. Exp. Pathol. 36, 515 (1955).

12. Phillips et al, J. of Biol. Chem. 240 (2), 699-704 (1965).

13. Bahl, J. of Biol. Chem., 244, 575 (1969).

Methods for preparing the modified polypeptides of this invention also include the following.

In one preferred modification approach, the polypeptide fraction, for example, Structure (XII), is activated first following which it is conjugated with a carrier, for example the influenza subunit described above, tetanus toxoid or Flagellin. An activating reagent may be utilized which exhibits differing functionality at its ends and by choice of reaction conditions, these end components can be made to react selectively. For example, the following activators A and B, having a maleiimido group and a substituted acid group, may be provided: ##STR5## "maleiimido" group where X is a non-reacting group made up of a substituted, or unsubstituted phenyl or C.sub.1 -C.sub.10 alkylene moiety, or a combination thereof. In this regard, the moiety substituted on the phenyl should be non-interfering as is the remainder of the "X" grouping.

X may, inter alia, be selected from the following: ##STR6##

The maleiimido grouping of the above reagents will react with sulfhydryl (SH) groups in the polypeptide fragments under conditions whereby the opposite end (active ester end) of the reagent does not react with the amino groups present in the fragment sequences. Thus, for example, polypeptide fragments such as Structure (XII), containing a Cys amino acid and hence, an SH group react as follows: ##STR7## Following the above, upon adjusting the pH to a slightly alkaline condition, e.g. 8, and adding the carrier protein accomplishes the following conjugation: ##STR8##

Preferably, a carrier protein such as the above-noted Flagellin which does not contain SH groups, but does contain NH.sub.2 groups, may first be treated with activator A or B at pH 7 or lower at the active ester end, giving: ##STR9## Following the above, the activated carrier is reacted with a polypeptide fragment containing a SH group to derive a product similar to that discussed immediately above.

Should the polypeptide fragment not contain an SH group, e.g. Structures (II), (III), (VI) and (VII), such structures can be modified first to introduce such a grouping by standard methods such as "thiolactonization", following which they are conjugated utilizing the above-discussed selective bi-functional reagents. For a more detailed description of these reagents, reference is made to the following publications:

O. Keller and J. Rudinger, Helv. Chim. Acta 58, 531-541 (1975).

W. Trommer, H. Kolkenbrock and G. Pfleiderer, Hoppe-Seyler's Z. Physiol. Chem., 356, 1455-1458 (1975).

Further description of preferred embodiments of the above-described utilization of bi-functional reagents is provided hereinbelow at Examples XXVII and XXVIII.

It is well known to those skilled in the art that, in many natural proteins containing cysteine residues, these residues are not present in their thiol form containing a free-SH group; instead pairs of cysteine residues are linked by means of disulfide bridges to form cystine. Such disulfide bridges are very important in determining the conformation of the protein. In most cases, the disulfide bridges present in the natural form of the protein are easily reduced to pairs of --SH groups by means of mild reducing agents under conditions which leave the remaining parts of the protein molecule unchanged. Accordingly, when it is desired to produce free-SH groups in proteins in order to carry out the coupling reactions discussed above, one convenient way of providing such free-SH groups may be to cleave disulfide bridges naturally present in the protein or other polypeptide which is desired to couple. In particular, the twelve cysteine residues present in the beta subunit of HCG (Structure (I) above) are, in the natural form of the subunit, coupled together to form six disulfide bridges, so that the natural form of the protein has no free-SH groups. To produce free-SH groups for coupling reactions, any number of these bridges, from 1 to 6, may be broken using known techniques, as set out for example in:

Bahl et al., Biochem. Biophys. Res. Comm. 70, 525-532 (1976).

This particular article describes cleaving 3-5 of the six disulfide bridges in the beta subunit of HCG, but the same techniques may be used to break all six bridges if this is so desired.

The generation of free-SH by reduction of disulfide bridges in naturally occuring forms of proteins may also effect the cross-reactivity of the antibodies produced when a modified polypeptide produced from the protein is injected into an animal. Frequently, an antibody recognizes its corresponding antigen not only by the amino acid sequence in the antigen by also by the conformation (shape) of the antigen. Accordingly, an antibody which binds very strongly to a protein or other polypeptide in its natural conformation may bind much less strongly, if at all, to the same protein or polypeptide whose conformation has been drastically altered by breaking disulfide bridges therein.

Accordingly, the breaking of disulfide bridges in proteins or other polypeptides may provide a basis for reducing the cross-reactivity between antibodies to antigens having the same amino acid sequence along parts of the molecule. For example, it has been pointed out above that cross-reaction is frequently encountered between antibodies to beta-HCG and HLH because the first 110 residues in the beta-HCG sequence are the same as the corresponding part of the HLH molecule, and in the natural forms of both molecules, the conformations are also presumably very similar. It has been suggested above that one means of producing in an animal antibodies to beta-HCG which do not substantially cross-react with HLH is to supply to the animal an antigen of the invention derived from a polypeptide which contains all or part of the residues 111-145 of beta-HCG but which lacks all or substantially all of the residues 1-110 of beta-HCG. In effect, this approach avoids antibody cross-reaction with HLH by chemically removing from the modified polypeptide of the invention the sequence of residues which is common to beta-HCG and HLH. As an alternative approach, by cleaving the appropriate number of disulfide bridges in the natural form of beta-HCG, it may be possible to so alter the conformation of residues 1-110 thereof that the antibodies formed when a modified polypeptide of the invention based upon this altered-conformation beta-HCG is administered to an animal will no longer cross-react significantly with HLH. In other words, instead of chemically severing the common sequence of residues from beta-HCG in order to prevent cross-reaction, it may be possible to leave this common sequence of residues in the beta-HCG but to so alter the conformation of this common sequence that to an antibody the altered-conformation common sequence does not "look" like the natural form of the common sequence, so that an antibody which recognizes the altered-conformation common sequence will not recognize the natural-conformation common sequence in HLH. Moreover, once the natural conformation of the sequence of residues 1-110 has been destroyed by breaking the disulfide bridges, this common sequence will probably assume the helical conformation common in polypeptides lacking disulfide bridges, so that this part of the beta-HCG will not be strongly immunogenic and most of the antibodies formed by a antigen of the invention based upon the altered-conformation beta-HCG will be antibodies to the sequence 111-145 which is not common with HLH.

As an alternative approach to the utilization of the maleiimido group reagents discussed above, an alkylation step may be used to cause conjugation. Conditions can be chosen such that in the presence of amino groups, essentially only SH groups will be alkylated. With this approach, a generalization of the reactions carried out may be expressed as follows: ##STR10##

With this approach, the larger molecule carrier, e.g. Flagellin, tetanus toxoid or the influenza subunit described herein is first modified by reaction of a fraction of its amino groups with an active ester of chloro, dichloro, bromo or iodo acetic acid, such as: ##STR11## and this modified carrier is then reacted with the sulfhydryl group in a polypeptide fraction, or a polypeptide fraction which has been modified to contain the SH group (e.g. thiolactonization) if it does not already have such a group. Such modification is described in Example XXV below. The present approach produces a thio ether linkage by alkylation of a free thiol (sulfhydryl group).

With the instant procedure, the roles of the fragment and carrier may be reversed, the fragment being modified to contain the halomethyl alkylating group which would then react with sulfhydryl groups in the carrier, or a carrier suitably modified to exhibit a sulfhydryl group. More description of this selective alkylation of sulfhydryl groups is provided in conjunction with Example XXX below.

It may be seen from an observation of the formulae of Structures (IV), (V), (IX), (X), (XI), (XII), (XIII), and (XIV) that a Cys amino acid, which in a reduced state provides an SH reactive group, is located at either the C terminal or N terminal of the peptide structure. This location permits the peptide to be chemically linked to carrier molecules at either terminus. And some Structures (XIV), (X), (IX), (X), (IV) have a six-Proline spacer chain (Pro).sub.6 between the Cys residue and the remainder of the peptide sequence. This latter arrangement provides a chemical spacer between the coupled carrier and the sequences representing a fragment of the natural hormone. A six-Proline spacer can be added as a side chain spacer, for example at position 122 (Lys) in Structure (II), by initially adding an SH group (thiolactonization) to the free or unblocked epsilon amino group on this (Lys) residue, as set out in Example XXIX below. Then, utilizing the activator A, B above in which the component "X" is a chain of six Proline amino acids, conjugation can be carried out. In the latter case, a spacer is provided between the carrier and peptide linked at an intermediate site, for example at position 122 in Structure (II). In the former case, only the space represented by conjugating reagent links the carrier and peptide.

Modifying groups, such as hemocyanin from Keyhole limpet, containing free amino groups, are prepared in buffer solution such as phosphate buffer, in sodium chloride solution at a pH of 6-8. To this solution, tolylene diisocyanate (T.D.I.C.) reagent diluted from about 1-10 to about 1-40 times with dioxane is added to the modifying group. The general procedure was disclosed by Singer and Schick, J. Biophysical and Biochem. Cytology 9, 519 (1961). The amount of T.D.I.C. added may range from 0.075 to 1,000 molar equivalents of the modifier used. The reaction may be carried out at about -5.degree. to about +10.degree. C., preferably 0.degree. to 4.degree. C., for about 1/2 to 2 hours. Any excess T.D.I.C. may be removed by centrifugation. The precipitate may be washed with the above-mentioned phosphate buffer and the supernatants combined.

This activated modifying group solution may then be combined with the hormonal or non-hormonal polypeptide to be conjugated. Polypeptide is dissolved in the same phosphate buffer (5-30 mg/ml) and the volume of modifier and polypeptide combined according to the molar ratio of the two desired in the conjugate. Combined solutions are reacted at 30.degree.-50.degree. C., preferably 35.degree.-40.degree. C., for 3-6 hours.

Separation of modified polypeptide and free unconjugated polypeptide may be accomplished by conventional techniques, such as gel filtration.

Picogram amounts of I.sup.125 labeled polypeptide may be added as a tracer to the reaction mixture at the time of conjugation, and a quantity of polypeptide conjugated to modifying groups (molar ratio) may be determined by the amount of radioactivity recovered.

Included in the methods for modifying the hormones, non-hormonal proteins and their fragments (unmodified polypeptides) are conjugation by use of water-soluble carbodiimide. The amino groups of the unmodified polypeptide are first preferably protected by acetylation. This (acetylated) unmodified polypeptide is then conjugated to modifier, such as natural protein modifier, e.g. hemocyanin from Keyhole limpet, homologous serum albumin, and the like, or Dextrans, Ficolls, or polytyrosine, preferably in the presence of guanidine such as guanidine HCl, using 10-ethyl-3-(3-dimethylamino propyl) carbodiimide as activating agent. This method is generally disclosed by Hoare and Koshland, Jr., J. of Biological Chemistry 242, 2447 (1967). In the instant where Ficoll 70 is used, it is preferred that it be first treated with ethylenediamine so as to render the final coupling more efficient. This treatment with ethylenediamine may be performed in solvent such as saline and dioxane at about room temperature and a pH of about 9-12, preferably 10-11 for about 1/4 to about 2 hours. The conjugation itself between the unmodified polypeptide and the modifier may be performed in solvent such as glycine methyl ester while maintaining the pH at about 4-5, preferably about 4.5-4.8. The temperature of reaction is conveniently about room temperature and the reaction may be allowed to proceed for about 2-8 hours, preferably 5 hours. The resulting modified polypeptide with which this invention is concerned may be purified by conventional techniques, such as column chromatography.

The immunogenic substances for this invention may also be provided by polymerization of unmodified polypeptide using bi-functional imidoester. The imidoester, such as dimethyl adipimidate, dimethyl suberimidate and diethyl malonimidate, may be used to form the polymer in a manner similar to the generally described methods of Hartman and Wold, Biochem. 6, 2439 (1967). The polymerization may take place conveniently at room temperature in aqueous solvent at a pH of about 9-12, preferably about 10-11, over a period of 1/4-2 hours.

Said immunogenic substances may also be prepared by dimerization through a disulfide bond formed by oxidation of the thio group on a Cys-residue using iodosobenzoic acid and methods corresponding to known methods, such as room temperature reaction for about 10-40 minutes.

The modified polypeptides of the invention may also be formed by forming linear polymers of the polypeptides. In a preferred method of forming such linear polymers, the polymer chain is begun with a first peptide lacking a C-terminal cysteine and having an amino group only at its N-terminal (peptides containing amino groups other than at the N-terminal may of course be used if all non-terminal amino groups are blocked with any conventional blocking agent). This starting peptide is reacted with 6-maleimido caproic acyl N-hydroxy succinimide ester (MCS) at a pH of 6.6, whereby the ester portion of the reagent reacts with the N-terminal amino group of the peptide. After removal of excess MCS, the thus-activated peptide is reacted under the same conditions with an equivalent amount of a second peptide having a C-terminal cysteine in a reduced stated (i.e. having a free SH group). This reaction will cause the coupling of the N-terminal of the activated peptide to the C-terminal of the second peptide. The resultant dimer is purified by gel filtration and then again reacted with MCS and a second equivalent of the second peptide, thereby producing a trimer. This polymerization procedure can be repeated until the desired chain length has been achieved (polymers containing at least 12 peptides have been prepared in this manner). It is critical that high-resolution chromatography be employed in the purification of the polymer at each step of the polymerization in order in ensure that all the final polymers have the same chain length and are not missing one or more of the units.

In this method of forming linear polymers, the first and second peptides may be identical in chemical configuration except that in the first peptide the C-terminal cysteine has a blocked thiol group. Where the modified polypeptide is to be used to form antibodies to HCG, preferred fragments for this purpose are (111-145)-Cys and (105-145)-Cys, where the figures refer to the bases in the beta subunit of HCG (Structure I above). It will be appreciated that, when these fragments are to be used in forming the linear polymers, the lysine residue at position 122 must have its amino group blocked and, in the case of the (105-145)-Cys fragment, the non-terminal cysteine at position 110 must also have its thiol group blocked, preferably with an acetamidomethyl group.

The linear polymers preferably contain from about 4 to about 14 fragments.

Modified polypeptides may also be prepared using glutaric dialdehyde as conjugating agent. According to a theory proposed by Richards and Knowles [J. Mol. Biol. 37, 231 (1968)], commercial glutaric dialdehyde contains virtually no free glutaric dialdehyde, but rather consists of a very complex mixture of polymers rich in .alpha.,.beta.-unsaturated aldehydes. Upon reaction with natural protein modifiers such as homologous serum albumins, these polymers form a stable bond through the free amino group, leaving aldehyde groups free. This intermediate product then reacts with unmodified polypeptide in the presence of alkali metal borohydride, such as sodium borohydride. This intermediate is formed at pH 7-10, preferably 8-9, at about room temperature. The modified polypeptide is also conveniently obtained at about room temperature after about 1/4-2 hours' reaction time. The resulting product is recovered in pure form by conventional techniques, such as gel filtration, dialysis and lyophilization.

Polymerized sugar modifiers such as Ficoll 70 or Dextran T 70 may also be prepared for conjugation by treatment with a cyanuric halide such as cyanuric chloride to form a dihalotriazinyl adduct. The process may be performed in solvent such as dimethylformamide at about 0.degree.-20.degree. C., preferably 10.degree.-15.degree. C., for about 1/2-4 hours. The resulting intermediate product may then be dialyzed until essentially halogen ion free, and lyophilized and treated with unmodified polypeptide at pH 8-11, preferably about 9-10, for about 1/2-12 hours at about 15.degree.-35.degree. C., conveniently at room temperature. The resulting modified polypeptide may be recovered as indicated above.

Said polymerized sugar modifiers may also be treated with alkali metal periodate, such as sodium periodate, at a pH of 3-6 at about 30.degree.-60.degree. C. for about 1/2-4 hours, and the resulting intermediate conjugated with unmodified polypeptide at a pH of about 7-11, preferably about 8-10, for about 1/4 to about 2 hours at a temperature of about 15.degree.-80.degree. C., preferably 20.degree.-60.degree. C. The resulting immunogenic substance according to this invention may be separated as indicated previously.

The modifying groups may vary in chemistry and number for any given polypeptide structure. However, they will attach to only certain amino acid moieties. In particular, when modifying with diazo groups they will chemically bond to only the histidine, arginine, tyrosine and lysine moieties or sites. Other modifying groups will bond to peptide molecules at different sites and in different numbers. Consequently, depending upon the size and chemical make-up of a particular modified polypeptide desired, one skilled in the art will readily be able to calculate the maximum possible number of modifying groups associable with a polypeptide. It is also recognized that several modifying groups may attach themselves to each other which in turn attach to a single amino acid moiety, but as used herein, reference to a number of modifying groups means the number of reaction sites to which a modifier has been attached.

As indicated above, a theory leading to this invention was that the chemical modification of an essential reproductive hormone would alter it such that it would exhibit antigenic properties so that when injected into an animal (including humans) it would cause the formation of antibodies which in turn would not only react to the injected modified hormone but also to the natural unmodified endogenous hormone as well. With this theory in mind, reproductive hormones of various species were modified and tested in baboons. The results illustrated that modified hormones of unrelated species do not produce the desired results, whereas modified hormones of the same or closely related species do produce the desired results. It will accordingly be clear that the polypeptide to be modified should be so related to the endogenous hormone or non-hormonal protein as to be either from the same animal species or be the immunological equivalent thereof as modified.

Additional experiments were conducted to test the validity of this concept in humans, i.e. modified human reproductive hormones injected into humans. Collectively, the results prove the conclusion drawn from the experiments with the baboons, namely, isoantigenic immunization using modified human reproductive hormones does produce contraception or interruption of gestation. Detailed examples which follow illustrate this result.

It is known that fragments of endogenous hormones exhibit essentially no antigenic properties. However, should a large enough fragment of an endogenous hormone be slightly modified as indicated above, then antibodies will be formed which will react in the same way as if the modification is on a whole hormone, provided the large fragment is sufficiently distinctive in chemical and physical make-up as to be recognized as a specific part of the whole.

Whether the hormone or specific fragment thereof is naturally occurring or is a synthetic product is clearly immaterial. A synthetic hormone molecule will perform the same function as the naturally occurring one, being equivalent for the purpose of this invention. In this connection, it will be noted that natural substances with which this invention is concerned possess carbohydrate moieties attached at certain sites thereof whereas the contemplate corresponding synthetic polypeptides do not. Nevertheless, for the purpose of the instant specification and claims, the synthetic and natural polypeptides are treated as equivalents and both are intended to be embraced by this invention. Reference in the above regard is made to Table No. 3 herein as read in conjunction with Example XXIX.

Thus, where the word "hormone" or "hormone molecule" is used herein, the work "synthetic" may be added before "hormone" without changing the meaning of the discussion. Similarly, the word, "fragment" may be inserted after "hormone" or "molecule" without changing the meaning, whether or not "synthetic" has been inserted before "hormone".

Throughout the above specification, the term "modified" has been utilized in referring to the chemical reaction by which the foreign molecules become chemically attached to specific sites on the usually much larger polypeptide molecule. Although specific mechanisms by which this is accomplished are described herein in detail, other appropriate mechanisms may be used if desired. It is clear that the modifier, i.e., the substance which modifies the concerned protein, can be a physically larger molecule or fragment thereof than the molecule or fragment which it modifies. As noted above, such large molecules are deemed herein to be "carriers". Clearly, physical size of the fragment is not always critical; the criterion for effectiveness being that the body reaction generate antibodies in sufficient quanta and specific to the targeted hormone or endogenous substance.

The modified polypeptides of this invention may be administered parenterally to the animals to be protected, preferably with a pharmaceutically acceptable injectable vehicle. They may be administered in conventional vehicles with other standard adjuvants, as may be desirable, in the form of injectable solutions or suspensions. As indicated earlier, the adjuvant serves as a substance which will elevate total immune response in the course of the immunization procedure. Lipasomes have been suggested as suitable adjuvants. The insoluble salts of aluminum, that is, aluminum phosphate or aluminum hydroxide, have been utilized as adjuvants in routine clinical applications in man. Bacterial endotoxins or endotoxoids have been used as adjuvants as well as polynucleotides and polyelectrolytes and water soluble adjuvants such as muramyl dipeptides. The adjuvants developed by Freund have long been known by investigators; however, the use thereof is limited to non-human experimental procedures by virtue of a variety of side effects evoked. The usual modes of administration of the entire vaccine are intramuscular and sub-cutaneous.

The amount of modified polypeptide to be administered will vary depending upon various factors, including the condition being treated and its severity. However, in general, unit doses of 0.1-50 mg in large mammals administered one to five times at intervals of one to five weeks provide satisfactory results. Primary immunization may also be followed by "booster" immunization at one to twelve month intervals.

Adult female baboons were studied for at least one menstrual cycle for patterns of urinary estrogens, plasma, progestin, and in some cases urinary LH. Only those animals displaying normal patterns of these hormones were immunized. The criteria for normality and the procedures for housing animals are well known and will not be described.

Human Luteinizing Hormone (HLH)--partially purified preparation from human pituitaries with a biological potency of 2.5 units per mg. (NIH-LH-SI).

Human Follicle Stimulating Hormone (HFSH)--a partially purified preparation from human pituitaries with a biological potency of 86 units per mg. (NIH-FSH-SI).

Human Chorionic Gonadotropin (HCG)--a highly purified preparation from human pregnancy urine with biological potency of 13,200 IU/mg. (2nd IRP-HCG).

Monkey Luteinizing Hormone (MLH)--a crude preparation from rhesus monkey pituitaries with a biological potency of 0.75 units per mg. (NIH-LH-SI).

Ovine Luteinizing Hormone (OLH) (NIH-LH-S5).

Baboon Luteinizing Hormone (BLH)--partially purified baboon pituitary preparation with a biological potency of 1.1 units per mg. (NIH-LH-Sl).

All prepatations, excepting the OLH, were prepared in the inventor's laboratory. LH and HCG biological activity was determined by the ovarian ascorbic acid depletion test and the FSH preparation assayed by the ovarian augmentation assay.

Hormones were altered as antigens by coupling with a hapten in varying ratios of hapten to hormone as described by Cinander et al., supra. For convenience, the Cinader process is discussed herein although Phillips, supra, may provide a more stable bond under certain circumstances. In this procedure, the protein hormone serves as a carrier and the hapten is coupled to it by diazo bonds. Although a variety of hapten groups were coupled to different hormones, the same basic procedures was used for any combination. Fifteen to thirty-five haptenic groups per hormone molecule were found most useful for preparing immunizing antigens. The basic reaction consisted of diazotizing the hapten (sulfanilic acid) by adding it to a solution of 0.11 N.HCl and then slowly adding this solution dropwise to a 1 percent solution of NaNO.sub.2 with constant stirring at 4.degree. C. Diazotization was considered complete when free HNO.sub.2 was detected in the reaction mixture. Although the above reaction was accomplished at 4.degree. C., optimum temperatures for the reaction normally are about 0.degree.-6.degree. C., although 4.degree. C. is preferred.

The hapten-protein coupling was performed by dissolving the protein hormone in an alkaline buffer, pH 8.0. The diazotized hapten was added slowly to the hormone solution with continuous stirring at 4.degree. C. The pH of the reaction was constantly monitored and kept near 8.0. After all the hapten was added, the pH was finally adjusted to 8.0, stirred for 1-2 hours and allowed to stand at 4.degree. overnight. The mixture was thoroughly dialyzed for 6-8 days against distilled water to remove unreacted hapten.

Although the number of diazo groups per hormone molecule could be regulated by the number of moles of hapten and hormone reacted, a parallel control experiment with S.sup.35 labelled sulfanilic acid to evaluate the precise composition of the hapten-protein samples was performed with each diazotization. The same hormone preparation to be used for immunization was used in the control experiment. After the reaction was completed, an aliquot was taken from the reaction mixture and the remainder thoroughly dialyzed. Equal volumes of the dialyzed and undialyzed solutions were counted by liquid scintillation. By comparing the counts of the dialyzed and undialyzed samples, the moles of hapten coupled to each mole of hormone was calculated since the unreacted hapten was removed by dialysis. For this calculation, a molecular weight of 30,000 was assumed for all gonadotropin preparations.

Following dialysis, hapten-hormones were lyophilized and stored at 4.degree. C. Diazo-HCG (35 groups/molecule) and HLH (26 groups/molecule) were bioassayed by the ovarian ascorbin acid depletion method and found to retain 62 and 85 percent respectively of the activity of the unaltered hormones from which they were derived. None of the other hormones were assayed for biological activity.

Female baboons received their initial immunization on days 3-5 of the menstrual cycle and the second and third injections one week apart. The fourth injection was given 2-3 weeks after the third. A few animals received a fifth injection at 70-80 days after the first injections. All antigens were administered subcutaneously in a suspension of mannide manoleate or peanut oil. Doses of antigens for each injection varied between 3 and 5 mg. Injection sites were inspected daily for 5 days after each immunization for local reactions.

Daily 24-hour urine specimens and frequent serum samples were collected during at least one menstrual cycle prior to immunizations and following immunizations until the effects of treatment were assessed. Urinary LH, urinary estrogens and plasma progestins were measured. Antibodies were detected in post-immunization serum samples by reacting 0.2 ml. of a 1:1000 dilution of serum in phosphate-buffered saline (pH 7.4) 0.5 percent normal baboon serum with 250 pg of 1.sup.131 labelled hormone. Sera were reacted with both the unaltered immunizing hormone and unaltered baboon LH for antibody detection. A purified baboon LH preparation (1.9.times.NIH-LH-Sl) was used as a tracer antigen. Antigen-antibody complexes were precipitated with ovine anti-baboon gamma globulin after a 24-hr. incubation at 4.degree. C. Antibody levels were expressed as pg of labelled hormone bound. Significant antibody levels were considered to be those that would bind 5.0 pg or more of the 1.sup.131 labelled antigen.

Antisera were fractionated by gel filtration on Sephadex G-200 according to the procedure of Fahey and Terry (at p. 36, Experimental Immunology, F. A. Davis Co., Philadelphia, PA., 1967, incorporated by reference to the extent necessary to understand the invention) to determine the proportion of IgM and IgG antibodies in the baboon sera. Since the IgG fraction in this procedure contained a portion IgA and IgD antibodies, only IgM and total titers were determined. The IgM fraction from the column was reacted with 1.sup.131 hormones and the binding capacity determined. The volumes of the fractionated sera were adjusted so that antibody levels would be comparable to those of whole serum.

No significant reactions were observed at the site of injection following any immunization. On 4 occasions, a slight induration (2-3 cm in diameter) was seen when mannide manoleate was used as a vehicle but the redness and swelling disappeared within 4-5 days. Antibodies were detected against the immunizing antigen within 3-5 weeks in all animals. The extent, duration and cross reactivity of these antibodies is recorded. Generally speaking, higher levels were observed to heterologous gonadotropin immunization than to homologous ones.

The cross-reactivity of induced antibodies with baboon LH was studied on each animal. Cross-reactivity of antisera at peak levels was recorded. Although relatively high antibody activity against human LH and HCG were seen, relatively little reaction with baboon LH occurred. An intermediate cross-reaction was noted with anti-ovine LH and a high degree of cross-reactivity was seen with anti-monkey LH. Diazo-human FSH was weakly antigenic in the baboon. The duration of antibody production was generally longer with the human and sheep gonadotropin immunization than with those of monkey or baboon origin.

Peak antibody levels usually occurred at the time when the antibodies had shifted to principally the IgG type. Early antibodies had a larger proportion of IgM type and were generally more cross-reactive with baboon LH. The change in the proportion of the total antibody population that was IgM was recorded from the time antibodies were first detected. Significant cross-reactivity to baboon LH was observed in anti-human gonadotropins when IgM was abundant but dropped sharply as the antisera shifted to nearly all IgG. This drop in cross-reactivity did not occur with monkey and baboon immunizations. Again, the ovine LH immunizations produced an intermediate change in reactivity with the shift from IgM to IgG.

The effects of immunization upon the event of the menstrual cycle were determined by observing changes in sex skin turgescence and levels of pituitary and/or ovarian hormones. Based on these parameters, the delay or retardation of ovulation from the expected time, as judged by the control cycle, was calculated. One animal immunized with HCG had no interruption in ovulation and another immunized with HFSH was delayed for only one cycle. Two animals injected with HLH and two injected with HCG had ovulation delays equivalent to two menstrual cycles. A third animal immunized with HLH was delayed a calculated 86 days. Ovine LH immunizations produced an 88 day delay in ovulation.

Immunizations with diazo-monkey or baboon LH resulted in longer disruption of the menstrual cycle. Calculated delays in ovulation for the two animals receiving monkey LH was 146 and 122 days whereas the animals receiving altered baboon LH were retarded from ovulation 224 and 210 days.

Effects on specific hormone patterns following immunization with HLH in one animal were recorded. The interval between menses was considered to represent a "cycle". Urinary estrogens and plasma progestin patterns indicated that no ovulation occurred during the cycle of immunization which was 85 days in duration. Urinary estrogens were elevated during treatment but did not reflect a typical pattern. Plasma progestins were not elevated until about day 19 of the first post-treatment cycle. Patterns of both estrogens and progestins were within normal limits during the second post-treatment cycle. Antibody levels were elevated from about day 35 of the treatment cycle until 289 days from the first detection of antibodies. An LH assay was not available when this animal was studied and no data on plasma or urinary levels of this hormone were obtained.

Hormonal patterns following an immunization with diazo-baboon LH were recorded. In this animal, antibody levels were lower and persisted, in general, for a shorter period than did immunizations with human gonadotropins. During the treatment cycle, levels of urinary estrogens and plasma progestins followed a normal pattern but were quantitatively lower than normal. Urinary LH patterns fluctuated markedly due to the injections of diazo-LH during this period. No conclusive evidence of ovulation was obtained for the treatment cycle. The first post-treatment cycle lasted 246 days. During this cycle urinary LH and estrogens were elevated on days 35-41 but there was no subsequent elevation in plasma progestins that would indicate ovulation had occurred. Following day 42 of this cycle, there was no significant elevation in any of the three hormone levels until day 231 when significant elevations of urinary estrogens and LH occurred. These rises were followed 3 days later by an elevation in plasma progestins indicating the presence of a functioning corpus luteum. A second post-treatment menstrual cycle was of normal duration and the endocrine patterns were normal.

Antibodies so unaltered baboon LH attained maximum levels by about day 70 of the post-treatment cycle and remained relatively constant until day 190 when a steady decline was observed. By day 215 of this cycle, antibody levels were barely detectable. Approximately 16 days after this time, a peak of LH commensurate with a normal midcycle elevation was observed. From this point the animals appeared to have the normal function of the pituitary-ovarian axis. Hormonal patterns in animals with other heterologous gonadotropin immunizations were similar to animal receiving HLH and other animals receiving monkey or baboon LH were similar in response to animal receiving baboon LH.

These results in baboons indicated that the modification of a reproductive hormone, by the procedures outlined, did render it antigenic and the antibodies thus formed did neutralize natural endogenous hormones if the natural hormone was obtained from the species receiving the immunizations with modified hormone.

HCG is a hormone naturally present only in pregnant women with the exception that an entity at least analogous thereto has been found to be present in humans in conjection with neoplasms. HCG is also commercially available. Human LH is immunologically and biologically identical to HCG even though there are chemical differences. Since they are biologically identical and HCG is readily available from commercial sources it was presumed that the effectiveness of this immunological procedure could be evaluated by injecting modified HCG into nonpregnant women and monitoring the blood levels of LH. Antibodies formed will neutralize both the LH and the modified HCG. Reference in the above regard is made to the publications identified earlier herein.

Women have a pattern of LH levels; the level is substantially constant until the middle period between menstrual cycles, immediately prior to ovulation; at that point the LH level rises greatly and helps induce the ovulation. Monitoring the LH level and the antibody level will show that the procedure used did or did not cause the production of antibodies capable of neutralizing the endogenous reproductive hormone, namely LH.

A woman aged 27 years were selected for study. Hormone was obtained, purified and modified. The modified human hormone (HCG) was injected into the subject. It is well known that antibodies to HCG react identically to LH as well as HCG. The effect of the immunization was evaluated, principally by monitoring blood levels of LH. Finally the results were evaluated.

Clinical grade HCG derived from pregnancy urine was obtained from the Vitamerican Corp., Little Falls, New Jersey. This material has an immunological potency of 2600 IU/mg. Contaminants were detected in this preparation. Purification consisted of chromatography and elution. Fractions were dialyzed and lyophylized. The most potent fraction contained approximately 7600 IU/mg.; however, it was heterogeneous on polyacrylamide gel electrophoresis.

The fraction was further purified by gel filtration. The elution profile revealed two major protein peaks. The most potent HCG was found in the first peak and had an immunological potency of 13,670 IU per mg. This fraction was subjected to polyacrylamide gel electrophoresis. Further purification by gel filtration showed no evidence of heterogeneity of the HCG at this stage. Consequently, materials for study were processed according to the above procedure.

The contamination of this purified HCG was tested with I.sup.131 used for identification and a sample was reacted with antisera against several proteins offering potential contamination. Those proteins were follicle stimulating hormone, human growth hormone, whole human serum, human albumin, transferin, alpha one globulin, alpha two globulin and orosomucoid. No detectable binding of the purified HCG was observed with any antisera at a dilution of 1:50 of each. These negative results, calculated against potential binding of the respective proteins, indicated that contamination if any was less than 0.005 percent.

Hormone was altered by coupling with a hapten (sulfanilazo). This method couples the hapten molecules to the protein via the amino group of the aliphatic or aromatic portion of the hapten. The number of hapten molecules coupled to each HCG molecule (Ha-HCG) can be regulated and for this study, forty haptenic groups per HCG molecule were used for preparing the immunizing antigen.

Following the hapten-coupling process, the Ha-HCG was sterilized and tested.

The subject was multiparous and had terminated her reproductive capabilities by prior elective bilaterial salpingectomy. She was in good health and had regular cyclic menstruation. She underwent complete history, physical examination and laboratory evaluation including blood count, urinalysis, latex fixation and Papanicolau smear. She had no history of allergy.

To demonstrate normal functioning of the pituitary-ovarian axis prior to immunization, blood samples were obtained every other day from the first day of menses for 10 days, then daily for 10 days and finally, every other day until the next menses. Serum determinations of FSH, LH, estrone, estradiol and progesterone were performed. These studies indicated an ovulatory pattern.

Ten mg. of the Ha-HCG antigen were dissolved in 1.0 ml. of saline and emulsified with an equal volume of oil. Prior to injection, scratch tests to antigen and vehicle were performed. Immunizations were begun in the luteal phase of the treatment cycle to prevent superovulation from the administered HCG. Four injections at two week intervals were given to the subject. The first two of these were administered in oil subcutaneously (1.0 ml in each upper arm); the final two injections were given in saline only via the intradermal route. Following each injection, blood pressure readings were taken and the subject observed for allergic reactions.

Blood samples were collected at weekly intervals beginning two weeks after the initial injection to test for the presence of humoral and cellular antibodies. Following completion of the immunization schedule, blood samples were collected in the same manner as in the control cycle to assess effects of immunization on hormonal patterns of the menstrual cycle. Since antibodies to HCG react identically to LH as with HCG, LH was monitored as an index of effectiveness of the procedure. A third cycle was similarly studied six months after initial immunization. Upon completion of the study, physical and pelvic examinations and laboratory evaluations were repeated.

Serum samples from the control and post-treatment cycles were assayed for FSH, LH, estrone, estradiol and progesterone.

The subject was tested for delayed hypertensivity before immunization and at two week intervals until the injection schedule was completed by an in vitro lymphocyte transformation test.

Temporal relationships of serum pituitary and gonadal hormones in the control cycles of the subject were recorded. Antibody titers to HCG were detected in the subject after two injections. Menses occurred at regular intervals during the immunizations.

Following the initial injection in mannide manoleate, some itching and swelling at the injection site occurred. Subsequent intradermal injections in saline produced no reactions and it was concluded that the local reactions were induced by the mannide manoleate. Lymphocyte transformation tests on plasma samples were negative.

In the post-treatment cycle, baseline follicular and luteal phase LH levels were not noticeably changed in the subject. Very small midcycle elevations in LH levels were observed as compared to the normal large increases. FSH patterns in the post-treatment cycle were normal. This indicated that the antibodies were neutralizing the action of endogenous LH.

The subject showed an ovulatory progesterone pattern but attained relatively high antibody titers to LH and HCG after only two injections of Ha-HCG.

The subject was studied during another cycle approximately six months from the first immunization. Significant antibody titers were found. LH patterns indicated a small midcycle elevation. FSH patterns were essentially normal. Thus, the specificity of anti-HCG antibodies to LH was shown but not to FSH.

Another woman aged 29 years was selected for further study. Hormone was obtained, purified, and modified as in Example II. This modified hormone was injected into this subject in the same way as in Example II. The subject was monitored and tested as in Example II.

The results were similar to the results found in Example II except that (1) the levels of estrone and estradiol were substantially normal, (2) the subject acquired significant antibody titers late in the post-immunization cycle, and (3) in the cycle studies after six months this subject showed no significant midcycle elevation in LH patterns.

Another woman aged 29 years were selected for further study. Hormone was obtained and purified and modified as in Example II. This modified hormone was injected into this subject in the same way as in Example II. The subject was monitored and tested as in Example II.

The results were similar to the results found in Example II except that (1) baseline follicular and luteal phase LH levels were noticeably depressed in the post-treatment cycle, (2) no midcycle elevations were observed in LH, (3) estrone levels were elevated during the follicular phase of the post-immunization cycle, and (4) during the six-months' study there was no significant midcycle elevation in LH patterns.

Another woman aged 35 years was selected for further study. Hormone was obtained, purified, and modified as in Example II. This modified hormone was injected into this subject in the same way as in Example II. The subject was monitored and tested as in Example II.

The results were similar to the results found in Example II except that (1) baseline follicular and luteal phase LH levels were noticeably depressed in the post-treatment cycle, (2) a very small midcycle elevation of LH was observed, (3) levels of FSH patterns in the post-treatment cycle were depressed, and (4) levels of both estrone and estradiol were reduced during the follicular phase of the post-immunization.

Another woman aged 28 years was selected for further study. Hormone was obtained, purified, and modified as in Exaple II. This modified hormone was injected into this subject in the same way as in Example II. The subject was monitored and tested as in Example II.

The results were similar to results found in Example II except that (1) baseline follicular and luteal phase LH levels were depressed in the post-treatment cycle, (2) no peaks were observed in midcycle levels of LH, (3) estrone levels appeared elevated in the follicular phase of the post immunization cycle, and (4) LH patterns indicated no significant midcycle elevation in the six-month post-immunization cycle.

Another woman aged 28 was selected for further study. Hormone was obtained, purified, and modified as in Example II. This modified hormone was injected into this subject in the same way as in Example II. The subject was monitored and tested as in Example II.

The results were similar to results found in Example II except that (1) antibody titers to HCG were not detected until after three injections, (2) baseline follicular and luteal phase LH levels were depressed in the post-treatment cycle, (3) no peaks nor midcycle elevation in the LH were observed, (4) estrone levels were elevated during the follicular phase, and (5) no significant antibody titers were found in the six month cycle.

All the above examples show the practicality of injecting modified hormones for the purpose of neutralizing an endogenous reproductive hormone and thereby offering a procedure for the prevention of conception or the disruption of gestation.

Data obtained in earlier experiments and discussed in Examples I-VII showed that a modified natural reproductive hormone, when injected into an animal of species from which it was derived, would produce antibodies that would neutralize the action of the unmodified endogenous natural hormone in the body of the animal. Hormones used in Examples I-VII were FSH, LH and HCG. New experiments were performed, based on this knowledge, to identify another reproductive hormone (placental lactogen) that could be used in a similar fashion.

A purified preparation of Placental Lactogen was prepared from placentae of baboons since it was intended to use modified placental lactogen to immunize baboons. Placentae were extracted and purified on column chromatograph according to previously published procedures. The purity was tested by polyacrylamide gel, electrophoresis and by radioimmunoassay. The material obtained showed a high degree of purity on electrophoresis, and radioimmunoassay showed no contamination with other placental hormones.

The baboon placental lactogen (BPL) was altered by coupling with the diazonium salt of sulfanilic acid as outlined for other hormones in Example I. The number of diazo molecules per BPL molecule in this instance was 15. Immunization procedures were also similar to those described in Example I for other hormones.

Within 4-6 weeks after the first injection of diazo-BPL, antibody levels to nautral unmodified BPL in vitro were detected in 6 femal baboons. Levels rose to a plateau within 8-10 weeks and remained there for several months. Hormonal measurements indicated that there were no effects on the normal events of the menstrual cycle due to the immunizations. Since BPL is normally secreted only in pregnancy, this was not a surprising observation.

All six females were mated with a male of proven fertility three times (once each in three different cycles during the fertile period). Pregnancy diagnosis by hormonal measurement was performed after each mating. From the 18 matings, there were 13 conceptions as judged by pregnancy tests. The animals that were pregnant had menstrual bleeding 7-12 days later than was expected for their normal menstrual cycles. Subsequent hormonal measurements confirmed that these 13 pregnancies were terminated by abortions approximately one week after the time of expected menses.

These findings suggest that the antibodies formed in the animal's body after immunization had no effect on the nonpregnant menstrual cycle but when pregnancy was established, they neutralized the baboon placental lactogen in the baboon placenta and the result was abortion very early after conception.

When in Examples I-VIII above Structures (I), (II), and (III) are modified by use of a diazosulfanilic acid, dinitrophenol, or S-aceto mercaptosuccinic anhydride or structures (II) and (III) are modified by addition of polytyrosine or polyalanine, according to known methods, the results obtained should be similar to those in said Examples.

Similarly, when FSH, somatomedian, growth hormone or angiotension II are modified by use of diazosulfanilic acid or trinitrophenol, the results obtainable upon administration of the purified modified polypeptide into a male or female human or animal would indicate the stimulation of antibodies which neutralize all or some of the modified polypeptide as well as corresponding endogenous polypeptide.

The subjects used in the studies reported in the example are female baboons. All baboons were adults of reproductive age. A description of subjects and the conditions of experimentation have been described in Example I. The animals have been studied using highly purified beta submits of HCG using a preparation with a biological activity of less than 1.0 IU/mg. Animals were immunized with 14-26 moles/mole of polypeptide of diazosulfanilic acid coupled subunits in mannide manoleate.

Antibody levels were assessed by determining the binding of serum dilutions with I.sup.125 labelled antigens. Cross-reactivity of antisera was measured by direct binding of labelled antigens and by displacement radioimmunoassays. Antifertility effects in actively immunized animals were tested by mating females with males of proven fertility. Effects in pregnant baboons passively immunized with either sheep or baboon anti-P-HCG were determined by monitoring serum levels of gonadotropins and sex steroid hormones before and after immunizations.

Eight female baboons were immunized with the modified beta subunit of HCG. Significant antibody levels were attained in all animals.

Baboon immunizations with the modified beta subunit of HCG resulted in high antibody levels reacting to HCG, human LH and baboon CG but not to baboon LH. All animals remained ovulatory; however, no pregnancies resulted from numerous matings with males of proven fertility. Passive immunization of non-immunized pregnant baboons with sheep anti-P-HCG serum produced abortions within 36-44 hours.

Hemocyanin from Keyhole limpet (KLH) solution (7 mg/ml) in 0.05 M sodium phosphate buffer in a 0.2 M NaCL, pH 7.5, is prepared. Insoluble particles are removed by centrifugation. To one ml of this solution, tolylene diisocyanate (T.D.I.C.) reagent is added (20 .mu.l) diluted to 1/30 with dioxane, the amount being essentially the equivalent of the moles of lysyl residues in the KLH molecules. After 40 minutes at 0.degree. C., the T.D.I.C. activated KLH solution is combined with 0.5 mg of synthetic .beta.-HCG peptide having the following structure: ##STR12## which is first dissolved in 25 .mu.l of 0.05 M sodium phosphate buffer in a 0.2 M NaCl, pH 7.5. The mixture is incubated at 37.degree. C. for four hours. The resulting product is purified by gel filtration.

One g. of Ficoll 70 is dissolved in 1 ml each of normal saline and 2 M ethylene diamine (adjusted to pH 10 with hydrochloric acid) solution. The solution is kept at room temperature in a water bath and stirred with magnetic stirrer. Cyanogen bromide, 4 g, dissolved in 8 ml of dioxane, is added to the Ficoll 70 solution. The acidity of the mixture is maintained at pH 10-10.5 for 8 minutes by adding drops of 2 N sodium hydroxide solution. An additional 2 ml of 2 M ethylene diamine, pH 10, solution is added, and stirring at room temperature is continued for 30 more minutes. The product is purified by passing it through a Bio-Gel p-60 column.

Two mg of the compound of Structure (II) containing a picogram amount of I.sup.125 labeled adduct and KLH (1.6 mg) is dissolved in 1 ml. of 1.0 M glycine methyl ester in 5 M guanidine hydrochloride. Ethyl dimethylamino propylcarbodiimide (E.D.C.) 19.1 mg is added to this solution. The acidity is adjusted to and maintained at pH 4.75 with 1 N HCl at room temperature for 5 hours. The KLH-peptide conjugate is purified by passing it through a Bio-Gel p-60 2.2.times.28 cm column equilibrated with 0.2 M NaCl.

Solid bifunctional imidoester dihydrochloride (3 mole) is added in 2 mg portions at 5-minute intervals to a constantly stirred solution of 1 mole of polypeptide of Structure (II) (1-20 mg/ml) in 0.1 M sodium phosphate, pH 10.5, at room temperature. Sodium hydroxide 0.1 N is added to maintain the acidity at pH 10.5. One hour after the addition of the dimidoester has been completed, a polymerized product according to this invention is obtained.

To a 20 mg/ml solution of homologous serum albumin in 0.1 M borate buffer, pH 8.5, 1000% mole excess of 25% aqueous solution of glutaric dialdehyde is added at room temperature. The excess dialdehyde is removed by gel filtration in water using Bio-Gel p-2. The material collected at the void volume is lyophilized, and the dried product is redissolved in 0.1 M borate buffer, pH 8.5 (20 mg/ml), mixed with the required amount of polypeptide of the following Structure: ##STR13## (20 mg/ml) in the same buffer at room temperature. Twenty minutes later, sodium borohydride in 250 percent molar excess of polypeptide XVI is added. The reaction is terminated after one hour. The conjugated product is purified by gel filtration on Bio-Gel p-60 column, dialyzed free of salt and lyophilized.

Ficoll 701 g, NaHCO.sub.3 500 mg, cyanuric chloride 3 g, H.sub.2 O 20 ml, and dimethylformamide 80 ml., are stirred at a temperature below 16.degree. C. for 2 hours. The product is dialyzed against distilled water until Cl-free, then lyophilized. A polypeptide of Structure (XV) (2 mg) containing a minute quantity of I.sup.125 labeled analogue is incubated with 1 mg of this product in 0.25 ml of 0.2 M. sodium borate buffer, pH 9.5, for one hour at 20.degree. C., and the product is recovered from a Bio-Gel p-60 2.2.times.28 cm column.

When the above procedure is carried out and Dextran T 70 is used in place of Ficoll 70, the corresponding modified polypeptide, useful according to this disclosure, is obtained.

Ficoll 70 1 g, NaIO.sub.4 1.2 g, and KCl 0.42 g are dissolved in 1.5 ml of 1 M sodium acetate buffer, pH 4.5, and incubated at 37.degree. C. for 1 hour.

Two mg (=588 .mu.moles) of polypeptide of Structure (XV) above mixed with a minute quantity of I.sup.125 -labeled analogue is incubated with 2 mg of the product obtained above in 0.3 ml of 0.2 M borate buffer, pH 9.5 at 55.degree. C. for 1 hour. The reaction mixture is then chilled in an ice water bath and NaBH.sub.4 1 mg is then added into this solution. The reduction reaction is terminated by passing the product through the Bio-Gel p-60 2.2.times.28 cm column equilibrated and eluted with 0.2 M NaCl.

Numerous rabbits are immunized with a variety of synthetic peptides conjugated to different modifying groups. Following two or three immunizations at 3-5 week intervals, sera from animals are assessed by determining their ability to bind in vitro to radiolabeled HCG. The specificity of this binding is studied by reacting the same sera against similarly labeled other protein hormones, particularly pituitary LH. Sera are further assessed by determining their ability to inhibit the biological action of exogenously administered HCG in bioassay animals. Thus, the increase in uterine weight of the immature female rat in response to a prescribed dose of HCG is noted. The dose of HCG is administered subcutaneously in saline in five injections over a three day period and the animal is sacrificed for removal of the uterus on the fourth day. The weight of the uterus increases in dose response fashion to the hormone injections. When assessing the effects of antisera in this response, varying quantities of test serum are administered intraperitoneally separately from the subcutaneous injection of hormone during the assay. This procedure permits the antiserum to be absorbed rapidly into the rat's bloodstream and will permit interaction of it with hormone when the latter likewise enters this fluid. If the antiserum is capable of reacting with the hormone in a manner preventing stimulation of the uterus, the antiserum is considered to be effective for biological inhibition of hormone action.

The frequency of animals showing a positive response to immunological binding and neutralization of biological activity is presented in

Iodosobenzoic acid dissolved in a slight excess of 1 N potassium hydroxide in 10% molar excess is added to the peptide of Structure (II) in phosphate buffer with normal saline at pH of 7.0. After thirty minutes at room temperature, the product polypeptide dimer is purified by gel filtration.

To an ice water bath cooled and vigorously stirred 0.23 ml. of bovine gamma globulin (10 mg/ml) in 0.05 M phosphate buffer with normal saline (PBS) pH 7.5, 50 .mu.l of 1/10 T.D.I.C. in dioxane is added. After 40 minutes, the excess T.D.I.C. is removed by centrifugation (0.degree. C., 10 minutes, 10,000 g) and the precipitate is washed twice with 0.1 ml. of PBS. The combined supernatants are added to 7.7 mg. of the peptide of Structure (II) dissolved in 0.8 ml. of PBS, pH 7.5. The mixture is stirred at room temperature for 10 minutes, then incubated at 37.degree. C. for 4 hours. The conjugate product is purified by dialysis.

BSA (10 mg/ml) in PBS solution (0.25 ml.) is treated with 50 .mu.l of 1/10 T.D.I.C. dioxane solution and conjugated to 7.5 mg. of synthetic .beta.-HCG peptide of Structure (III) in 0.8 ml. of PBS (pH 7.5) as in Example XIX to obtain the product.

To an ice water bath cooled and vigorously stirred 0.6 ml. of .beta.-HCG peptide of Structure (III) (10 mg/ml) in phosphate buffered saline, pH 7.5, is added 30 .mu.l of 1/10 T.D.I.C. dioxane solution. After 40 minutes, the excess T.D.I.C. is removed by centrifugation (10,000 g, 0.degree. C., 10 minutes) and the precipitate is washed twice with 0.1 ml. PBS. The combined supernatants are added to 3 mg. of poly (D, L-Lys-Als) dissolved in 0.3 ml. of PBS. The mixture is incubated at 37.degree. C. for 4 hours. The product is then dialyzed and lyophilized.

The results set out in Table I provide further evidence of the broad applicability of this invention as indicated previously in this specification.

Using standard methods of testing in rabbits, both immunological binding response and neutralization of biological activity were established for the modified polypeptides indicated with the result as set out in Table I.

Antigen was prepared by reacting a Diisocyanate (T.D.I.C.--see above) coupling reagent with carrier (tetanus toxoid), extracting excess reagent and incubating activated carrier with peptide Structure (II). Baboons were immunized with the antigen and results of mating 4 animals three times are shown in FIG. 1. The figure shows that from 12 exposures (matings) one pregnancy resulted even though relatively low levels of immunity from the antigen were achieved. Non-immunized baboons of the same colony had a fertility rate of approximately 85%.

Referring to FIG. 2, baboons were immunized initially with a beta subunit of HCG modified by diazotization in a manner similar to that described in conjunction with Example II. Following this initial administration, the baboons were injected 21 and 42 days later with Structure (II) above having been modified by the same diazotization process. FIG. 2 shows plots representing the levels of antibodies generated in consequence of these administrations. Such quantities of antibodies are expressed as micrograms of isotopically-labeled HCG that will bind each milliliter of serum from the baboons at specified days after the initial injection. The levels shown were maintained for a period of over one year.

Referring to Table 2, the results of breeding the two baboons represented in FIG. 2 is revealed in tabular form. The table presents the results of mating these animals ten times over a period of approximately one year. These data suggest that the animals ovulated in every cycle; however, no pregnancy was observed, as indicated by the animal having a menstrual period at or before the expected time therefor. While the results tabulated demonstrate the efficacy of the entire procedure, it was observed for the particular structure utilized in the primary immunization, i.e. Structure (I), antibody cross reactivity with LH was observed. Such cross reactivity may be avoided by the utilization of the fragment conjugation procedures set forth in detail hereinabove.

The specificity of antibody response to a CG fragment-macromolecular carrier is represented by the instant experiment. A 35 amino acid sequence [Sructure (II), herein "synthetic peptide"] of the HCG beta subunit was conjugated with bovine gamma-globulin and administered to a baboon. Varying doses of each of these three hormones were tested for their ability to compete with I.sup.125 -labeled synthetic peptide [Structure (II)] bound to the anti-serum. The results are set forth in FIG. 3. Note from the figure that Human LH was ineffective for displacement of tracer antigen at doses up to 2.5 IU (international units). Since HCG displaced antigen at a dose of 20 mIU, the cross-reactivity with HLH in this assay system was less than 0.8%. Baboon CG also displaced I.sup.125 -labeled antigen in this assay and, based on biological potency of the two hormones, was about 20% as effective as HCG.

The following experiments were carried out to determine whether the carbohydrate chains contained in the C-terminal 37 residues of .beta.-HCG influence the immunogenicity of that peptide.

A peptide representing amino acid residues 109-145 of .beta.-HCG was isolated from a chymotryptic digest of reduced and carboxymethylated .beta.-HCG by procedures reported by Keutmann, H. T.; Williams, R. M., J. Biol. Chem. 252, 5393-5397 (1977). This peptide is identified in Table 3 as P-1. The purity of the peptide was confirmed by amino acid and terminal end group analyses. A portion of the isolated peptide was treated with anhydrous hydrofluoric acid (HF) to remove carbohydrate moieties and repurified by column chromatography according to methods described by Sakakibara S. et al., Bull. Chem. Soc. Japan, 40, 2164-2167 (1967). This portion of the isolated peptide is identified in Table 3 as P-2. Complete removal of the sugar chains was confirmed by carbohydrate analysis; See Nelson Norton, J. Biol. Chem. 153, 375-380 (1944). A third peptide with the amino acid sequence 109-145 of .beta.-HCG was prepared synthetically using the solid state synthesis procedure of Tregear, G. W. et al., Biochem. 16, 2817 (1977). This third peptide is identified in Table 3 as P-3. Highly purified HCG was used in all immunological experiments where reference was made to intact HCG.

Conjugates of the three peptides were prepared to keyhole-limpet hemocyanin (KLH) using tolylene diisocyanate. A peptide-carrier ratio of 4-6 peptides per 100,000 daltons of carrier was obtained for different conjugates prepared according to amino acid analyses. Rabbits were immunized with conjugates by three multiple site intramuscular injections of 1.0 mg. of conjugate in 0.5 ml. of saline emulsified with an equal volume of Freund's complete adjuvant. Injections were given at 3 week intervals and weekly blood samples were collected from 3-20 weeks of immunization.

Antisera to all conjugates were monitored for antibody levels by reacting dilutions of sera with I.sup.125 labeled HCG (chloramine T method) at 4.degree. C. for 5 days and precipitating immune complexes with sheep anti-rabbit gamma globulin serum. Antibody levels were determined by assessing dilution curves in which a linear correlation between dilution and binding of labelled antigen at equilibrium occurred. At least 3 points in each curve were used in calculating levels. These levels were expressed as .mu.g. HCG bound per ml. of undiluted serum calculated by multiplying mass of labelled antigen bound by serum dilution.

A radioimmunoassay system employing I.sup.125 HCG and antisera raised to peptide conjugates was used to determine the relative ability of HCG and peptides to compete with labeled HCG. Peak antibody levels from each rabbit were evaluated in these studies. Antigens and antisera contained in phosphate-buffered saline (pH 7.4) BSA (1%) were added to test tubes and incubated at 4.degree. for 5 days. Separation of free and bound tracer HCG was accomplished by the addition of sheep anti-rabbit gamma globulin serum and further incubated for 48 hours followed by centrifugation. Assessment of parallelism of dose response curves was accomplished using methods described by Rodbard, D. in: Odell, W. D. and Daughaday, W. H., eds., "Competitive Protein Binding Assays", J. B. Lippincott, Phila. Pa. (1971). The ability of unlabeled HCG and peptides to compete with I.sup.125 HCG for antibody binding sites was expressed as moles of unlabeled antigen, per mole of unlabeled HCG, required to reduce the binding of labeled HCG by 50%. For this purpose molecular weights for HCG, P-1, P-2, and P-3 of 38,000, 7,000, 3,990, and 3,990 respectively were used. The molecular weight of the P-1 peptide was an estimate since the contribution of the 4 carbohydrate chains to its size was not determined. Four radioimmunoassays were performed with each of the 11 antisera studied and the results presented as the mean of the four values.

Parallel dose response curves of HCG and peptides were observed in all radioimmunoassays. In the assay system employed, 200-400 moles of unlabeled HCG was required per mole of labeled HCG at 50% binding of the latter to antisera. There was no detectable difference among antisera to the 3 peptide conjugates in the ability of intact HCG to compete with labeled hormone for antibody binding sites.

Data obtained from comparing the ability of HCG and peptides to compete with I.sup.125 HCG for binding to anti-peptide sera revealed some qualitative differences in the antisera (Table 3). Much larger quantities of P-2 peptide and P-3 peptide were required to reduce I.sup.125 HCG binding than was required by P-1 peptide when sera against the P-1 peptide was tested. While similar quantities of P-2 and P-3 peptides were required to inhibit one mole of labeled HCG binding, these were 2-10 times the amounts required by the P-1 peptide.

Differences in the quantities of peptides required to compete with an equivalent mass of labeled HCG were less using antisera raised to carbohydrate-free natural peptide (P-2). More P-1 peptide was needed for an equal reduction in binding than the other 2 peptides. No significant difference could be detected in the quantities of P-2 or P-3 peptides required among the 3 antisera tested.

Approximately 1.5-2.0 times as much P-1 peptide was required to compete equally with I.sup.125 HCG for antibodies raised to the P-3 peptide but P-2 peptide reacted nearly as well as did the synthetic peptide.

Despite low levels of antibodies obtained in this study, the carbohydrate-containing peptide was not more immunogenic than those without this moiety when conjugates to both were prepared in the same manner.

From these studies, it can be concluded that although antibodies to carbohydrate free peptides are qualitatively different than those to the natural peptide, antisera generated to the synthetic peptide reacted with HCG as well as anti-sera to natural peptides and equivalent to natural and synthetic peptides elicited similar anti-HCG levels in rabbits.

In this Example, a polypeptide fragment structure having an --SH group is activated utilizing the following reagent: ##STR14## A solution of the reagent (1.2 eq. per --SH group in the polypeptide) in a suitable water miscible organic solvent, such as dioxane, is added to a solution of the polypeptide fragment structure, e.g. Structure (XII) (which has had its amino groups blocked) in aqueous buffer at pH 6.5. After 2 hours, the solvent is removed at a temperature of less than 30.degree. C. under vacuum, and to the residue are added water and ethyl ether (1:1). The aqueous layer is separated and its pH adjusted to approximately 8.5 by the addition of sodium hydroxide solution and this alkaline mixture is added rapidly to an aqueous solution of the carrier, e.g. the above described influenza subunit, maintained at pH 8.5 by a suitable buffer. After a further 4 hours, the conjugate is isolated by gel filtration.

With the following reagent: ##STR15## a solution or suspension of a carrier containing no sulfhydryl groups such as Flagellin in a suitable aqueous buffer at a pH 6.5 is treated with the required (1.2 eq/--NH.sub.2 desired to be reacted) amount of a solution of the reagent in dimethylformamide. After 1 hour, the modified carrier is isolated by column chromatography and added to buffer at pH 6-7. This is then treated with a solution of the selected fragment (containing sulfhydryl groups) in the same buffer and the reaction is allowed to proceed for 12 hours before the conjugate is isolated by column chromatography.

Modification of non-sulfhydryl containing peptide fragments [e.g. Structure (II)] or a carrier such as Flagellin to a sulfhydryl containing one via "thiolactonization" is carried out as follows:

The peptide is dissolved in a 1 M aqueous solution of imidazole containing 0.5% of ethylenediamine tetraacetic acid at a pH of 9.3 under an atmosphere of nitrogen, and a 100 fold excess of N-acetylhomocysteine thiolactone is added in three portions at eight hour intervals. After a total of 30 hours, the pH is adjusted to 3-4 with acetic acid and the modified peptide is isolated by gel chromatography and elution with 0.5 M acetic acid.

The carrier protein is reacted with the N-hydroxy-succinimide ester of a halo-(either chloro, bromo or iodo) acetic acid in the general procedure described in the first part of Example XXVIII thus yielding a modified carrier containing the required number of halomethyl alkylating groups as desired.

To a solution of the sulfhydryl containing peptide [e.g. Structure (XII)] in a phosphate buffer at pH 6.5-7.0 under nitrogen at room temperature is added an aqueous solution or suspension of the modified carrier prepared above. The mixture is stirred for 12 hours. It is then washed with ethyl acetate and the conjugate contained in the aqueous phase is purified by dialysis, gel chromatography and lyophilization.

Should neither the carrier nor polypeptide fragment contain a sulfhydryl group, one may be introduced into either of them by the standard procedures such as "thiolactonization" described above under Example XXIX.

This example illustrates the use of a modified polypeptide of the invention in repressing fertility in baboons.

A polypeptide of Structure (XII) above, identical to the residues 109-145 of .beta.-HCG was prepared by total synthesis using the solid-phase synthesis method described in Tregear et al, Biochem. 16, 2817 (1977). The purity of the peptide was assessed using thin-layer chromatography, high-voltage electrophoresis and amino acid analysis. The peptide was conjugated to the amino groups of tetanus toxoid via the cysteine residue at position 110 by the method described in Lee et al., Mol. Immunol. 17, 749 (1980). The resultant conjugated polypeptide contained 22 peptides per 100,000 daltons of the toxoid.

Male and female baboons (obtained from Primate Imports, Inc., Port Washington, N.Y.) were housed individually in metabolic cages measuring 89.times.73.times.114 cm (L.times.W.times.H) for for females and 168.times.94.times.165 cm for males, each cage being equipped with a "squeeze-bar" mechanism for restraining the baboon. Each male baboon was housed in a separate room in which were also housed six to ten females. The room temperature was maintained at 21.degree. C. with artificial light for 12 hours daily and the baboons were fed on Purina Monkey Meal (product of Ralston Purina, St. Louis, Mo.) mixed with corned beef, corn syrup and a vitamin mineral supplement; fresh fruit was given daily as a conditioning aid and water was provided ad libitum. Daily observations were made of each female baboon to establish the pattern of sex skin turgescence/deturgescence and menstrual bleeding.

In order to ensure that the results of the test described below were statistically valid, the number of baboons required in each of the control and immunized groups was determined using arcsine transformatin of the projected fertility rates and the resultant values applied to probability tables, as described in Sokal et al., Biometry, W. H. Freeman and Co., San Francisco (1969), page 609. For an alpha level of p=0.05 and a 90% confidence of detecting a significant reduction in fertility rate when the control rate is at least 70% and the immunized is not greater than 10%, a group size of 15 animals per group was determined to be required, and thus this was the group size used in the experiment.

Accordingly, 30 female baboons whose cycle length varied by no more than three days each side of its mean and who had exhibited progesterone levels of at least 3.0 ng/ml. for each of their last three menstrual cycles (thus indicating ovulation) were randomly assigned to each of two groups of 15 animals. Six male baboons, who had each proved their fertility by siring several offspring were selected for use in the experiments.

The control group of 15 baboons were immunized with pure tetanus toxoid while the other group received the aforementioned modified polypeptide conjugated with this toxoid. The antigens were dissolved in physiological saline, mixed with an equal volume of Complete Freund's Adjuvant (supplied by Difco Laboratories, Detroit, Mich.) and emulsified just prior to each immunization. The modified polypeptide/toxoid conjugate was dissolved at a concentration of 4.0 mg/ml. and a dose of 2.0 mg given to each baboon in the second group, whereas the pure toxoid was dissolved in a concentration of 2.0 mg/ml. and a dose of 1.0 mg. given to each baboon in the control group; since the conjugate comprises approximately 50% of the toxoid by weight, the dose of toxoid administered to each animal was substantially the same. Each dose of the pure toxoid or conjugate in the adjuvant, approximately 1.0 ml. in volume, was injected intramuscularly into four separate sites into the animal, two in each thigh. The first (primary) immunization was given during the first five days of the menstrual cycle, with subsequent immunizations given at 28 day intervals thereafter or until a pregnancy was confirmed. Females that did not become pregnant received five or six immunizations during the course of the study depending upon the length of their individual menstrual cycles.

Blood samples were collected from the female baboons without anesthesia via the cubital vein, five to six ml. of blood being drawn at weekly intervals beginning at 21 days after the primary immunization, and also immediately before and after mating for antibody determinations. Blood samples for progesterone determinations were drawn five and seven days after sex skin deturgescence and, in cycles in which mating occurs, samples for pregnancy testing were drawn daily commencing 12 days after deturgescence and continuing until pregnancy was confirmed or menstruation began. After the blood samples were withdrawn, the serum was removed and samples not immediately tested were stored at -20.degree. C.

The serum samples thus obtained were tested for the presence of antibodies to .sup.125 I-labelled HCG, Structure (XII), baboon chorionic gonadotropin (bCG) and tetanus toxoid by the methods described in Powell et al., J. Reprod. Immunol. 2, 1 (1980). As described in this Powell et al paper, Structure (XII) can be labelled with .sup.125 I only after introduction of a tyrosine residue into the peptide. The HCG preparation, which was highly purified, had a biological potency of 10,800 IU/mg., while the bCG preparation, which was only partially purified, had a biological potency of 850 IU/mg. Concentrations of labelled antigen capable of saturation of antibody combining sites at equilibrium were reacted with dilutions of serum for five days at 4.degree. C., followed by separation of the antigen-antibody complex from the unbound labeled antigen using the double-antibody method. With the exception of tetanus toxoid, the concentrations of labeled antigens were adjusted so that equimolar quantities were reacted with the serum. The molecular weight of bCG, which has not yet been established, was assumed to be the same as that of HCG, namely 38,000. The labeled antigen binding for HCG, bCG and Structure (XII) was expressed as moles/liter (M/L).times.10.sup.-10, whereas for tetanus toxoid, due to its molecular heterogeneity, binding was expressed as micrograms/ml.

All 30 female baboons were mated during the course of their third menstrual cycle following the primary immunization and for the next two consecutive cycles if they did not become pregnant. Based upon their previously established menstrual histories, each female baboon was transferred to a male's cage three days prior to expected ovulation. Cohabitation was continued until the day of sex skin deturgescence, 2-3 days after ovulation and the female was then transferred back to her individual cage. If the first mating did not make the female pregnant, subsequent matings were conducted with a different male baboon.

The serum levels of steroid hormones and bCG were determined by the radioimmunoassay methods described in Powell et al, Clin. Chem. 19, 210 (1973) and Hodgen et al., J. Clin. Endo. Metabol. 39, 457 (1974). The assay for bCG was conducted with an antiserum to Ovine-LH.beta. supplied by Dr. Gary Hodgen, Bethesda, Md.; the supplier has previously shown, by means of unpublished data, that the binding of .sup.125 -I-HCG to this antiserum can be displaced sensitively with bCG, but not with baboon LH. The sensitivity of this assay was 5 mIU of HCG/ml. Since antibodies produced in the female baboons immunized with the Structure (XXII) conjugate were capable of binding the labeled HCG used in the HCG assay, this assay system could not be used to determine pregnancy in the baboons immunized with the conjugate. Accordingly, pregnancy testing in the conjugate-immunized baboons was performed by measuring estradiol and progesterone levels only. In the tetanus toxoid-immunized baboons, however, pregnancy was tested using the CG assay as well as the steroid assays.

Data obtained on antibody levels, cycle levels and progesterone concentrations were evaluated by various methods for randomized design experiments, as set out in Ostle, B., Correlation Methods, in Statistics in Research, Ames, I. A., Iowa State College Press (1954), page 174, while assessment of the mating data was accomplished by the chi-squared procedure set forth in Mantel, Cancer Chemotherapy Reports 50, 163 (1966); this procedure compares the mating data in its entirety and not only in terms of individual matings.

The tetanus toxoid antibody level in both the baboons immunized with the pure tetanus toxoid and those immunized with the conjugate are shown in Table 4. High antibody levels against tetanus toxoid were produced in both groups of baboons 60 days after primary immunization, with peak levels being reached in 90-120 days. The diferences between the tetanus toxoid antibody levels in the baboons immunized with the pure tetanus toxoid and with the conjugate were not significant at the p=0.05 confidence level. Thus, it will be seen that, in addition to the anti-fertility effects observed below, the instant conjugate conferred a significant degree of protection against tetanus. Accordingly, by careful choice of the hapten to which the polypeptide is conjugated in the instant modified polypeptide, the invention provides a method of protecting against a disease linked with the presence of the hapten as well as against pregnancy.

The mean antibody levels produced to HCG, bCG and Structure (XII) in the group of baboons immunized with the conjugate are shown in Table 5. Antibody levels to Structure (XII) reached a maximum during the luteal phase of the second menstrual cycle, approximately 60 days after the primary immunization, as did antibodies to HCG. While the mean antibody level to HCG and Structure (XII) were maintained by repeated booster immunizations, the responses of individual baboons varied considerably. There was a very close correlation between antibody levels to HCG and Structure (XII), r=0.97. The mean levels of antibodies which were reacted with HCG were only 71% of those reacting with Structure (XII). However, because of variation in levels between the individual animals, this difference in levels is not significant at the p=0.05 level. Although the mean levels of antibody reacting with bCG were only 4.5% of those reacting with Structure (XII) and 6.3% of those reacting with HCG, these bCG antibody levels reached maximum levels by the first mating cycle and remained close to that level during the next two cycles. There is a significant positive correlation (r=0.78) between the bCG antibody level and the Structure (XII) antibody level during these three cycles.

Moreover, the antibody levels for HCG and bCG shown in Table 6 reveal significant correlation between these two antibody levels during the three mating cycles. The correlation coefficients (r) are 0.55, 0.89, and 0.85 for the first, second, and third mating cycles respectively, the latter two correlation coefficients being significant at the 1% level.

Table 7 below compares the cycle lengths and progesterone levels in the luteal phase of the menstrual cycles before and after the immunizations of the two groups of baboons. The pre-immunization portion of this table shows that the two randomly assigned groups of baboons showed no significant differences at the p=0.05 level in the cycle length or progesterone levels for the three cycles immediately proceeding immunization. Even though both groups of baboons were immunized using complete Freund's, no change significant at the p=0.05 level in the cycle lengths or progesterone levels was apparent when the three pre-immunization cycles were compared with the post-immunization cycles.

Table 8 shows the highly significant difference in the fertility rate between the two groups of baboons. On the first mating, 10 out of 15 of the tetanus toxoid-immunized baboons became pregnant, 4 of the 5 remaining baboons became pregnant after the second mating and the single remaining baboon became pregnant after the third mating. Thus, of 21 matings, 15 resulted in pregnancy, giving a fertility rate of 71.4%.

Of the 15 baboons immunized with the conjugate, one became pregnant after the first mating, one of the remaining 14 became pregnant after the second mating and 2 of the remaining 13 baboons became pregnant after the first mating. Thus, 42 matings resulted in four pregnancies for a fertility rate of 9.5%.

Chi-squared analysis of this data shows that this difference in fertility rate is highly significant (p<0.0005) even after adjustment for the small sample size.

The antibody levels to tetanus toxoid from sera obtained during the three mating cycles were assessed for correlation to the outcome of mating for all animals. No correlation significant at the p=0.05 level was found for either group of baboons. Similarly, although the post-mating antibody levels for menstrual cycles 3, 4, and 5 were quite variable, as shown by the rather large 95% confidence interval, no correlation significant at the p=0.05 level was found between the antibody levels to Structure (XII), HCG or bCG and the fertility of the conjugate-immunized group. However, there was a significant difference (p<0.025) between the mean bCG antibody level in the four pregnant conjugate immunized baboons (1.4 M/L.times.10.sup.-10) and the mean (9.8) and the 95% confidence interval (5.5.-14.1) levels of the same antibodies for all matings, thus suggesting that these four baboons became pregnant because their bCG antibody levels were insufficiently raised.

I have shown in the following published papers:

Excerpta Medica International Congress Series No. 402, pp. 379 (1976); and

Physiological Effects of Immunity Against Reproductive Hormones, R. G. Edwards and M. H. Johnson (ed.), Cambridge University Press, pp. 249 (1975);

that immunizations with synthetic peptides containing a C-terminal portion of .beta.-HCG result in the production of antibodies capable of binding a neutralizing the biological activity of intact HCG and that these antibodies exhibit a low degree of reactivity with baboon CG. This is in accord with the results of this example, in which the conjugate based on Structure (XII), the 109-145 sequence of .beta.-HCG, produced high levels of antibody to HCG but relatively low levels of antibody to bCG. Nevertheless, despite the low levels of bCG antibody, the conjugate was highly effective in preventing pregnancy. The pregnancy-preventing action of the conjugate demonstrated in this example provides the statistically valid proof of the feasibility of this approach to fertility regulation in humans. It may reasonably be anticipated that the anti-fertility effects which the conjugate would produce in humans would be considerably greater than that in baboons, given the much higher level of antibodies to HCG produced in the baboons, as compared to the levels of antibody to bCG.

The exact mechanism of action of the conjugate is not known, although presumably antibody neutralization of CG occurs in the peripheral blood soon after implantation and disrupts trophic hormone support to the corpus luteum of pregnancy and causes early abortion. However, since the duration of the menstrual cycle is not significantly effected, it appears that pregnancy must be disrupted almost immediately after implantation.

The following peptides, each having a sequence derived at least in part from that of .beta.-HCG were prepared by the same method as in Example XXXI (the numbers given refer to the sequence of bases in the full .beta.-subunit of HCG, Structure (I) above:

a. 138-145, hereinafter referred to as Structure (XVI)

b. 126-145, hereinafter referred to Structure (XVII)

c. 115-145, Structure (VI) above

d. 111-145, Structure (II) above

e. 109-145, Structure (XII) above

f. 106-145, hereinafter Structure (XVII)

g. 105-145, hereinafter Structure (XIX)

h. Cys-(Pro).sub.6 -(111-145), Structure (XIV) above;

i. (111-145)-(Pro).sub.6 -Cys, Structure (X) above; and

j. Cys-(105-145), with the SH group of the Cys residue at position 110 blocked with an acetamidomethyl (ACM) group, hereinafter referred to Structure (XX).

The peptides of Structures (XVIII), (XIX) and (XX) contained a protected sulfhydryl group at the 110 cysteine position. The purity of all the peptides was demonstrated using high-voltage electrophoresis, thin-layer chromatography and amino acid analysis.

The following carriers were used in the experiments: tetanus toxoid, polymerized flagellin, poly-DL-alanine/lysine (polyalanine), (poly DAL) poly(tyrosine, glutamic acid)/poly(alanine, lysine) (TGAL) and polymerized sucrose (Ficoll), described above. The carriers used were diphtheria toxoid obtained from Connaught Laboratories, Swiftwater, Pa.) bovine gamma globulin (obtained from Swartz-Mann Laboratories, Orangeburg, N.Y.), Cornebacterium parvum (obtained from Burroughs-Wellcome, London, England), uncapsulated meningococcal protein and pneumococcus polysaccharide.

A thiol group on each peptide was coupled to an amino group on the carrier by the same method as in Example XXXI. The site at which the peptide was coupled to the carrier depended upon the point at which a thiol group existed or could be created on the peptide. Structure (XII) contains a cysteine at position 110 and accordingly, this peptide was conjugated by means of the thiol group produced at this position after cleavage of the synthetically produced disulfide dimer of the peptide. Structures (XVI), (XVII), (XVIII) and (XIX) had a thiol group introduced at the amino terminal group and were coupled to the carrier via this introduced thiol group. Structures (II) and (VI) had thiol groups at the terminal amino group and also at the amino group at the lysine residue at position 122; under the conditions used, approximately equal numbers of peptides were attached to the carrier at each of the two attachment sites. Structures (X) and (XIV) were coupled to carriers using the thiol group on the terminal cysteine residue.

After the coupling of the peptide to the carrier, the resulting conjugates were all purified by gel filtration or ultrafiltration and the ratio of the peptide to carrier was determined.

The animals used in these experiments were genetically heterogenous rabbits of the New Zealand White variety, weighing 2-4 kg. and inbred female mice of the C3H/He strain, 8-10 weeks old or retired breeders, weighing 20-30 gm. each and obtained from Jackson Laboratories, Bar Harbor, Ma. For use in the rabbits, the conjugates were dissolved in saline and emulsified with an equal volume of Complete Freunds' adjuvant, exactly as in Example XXXI. However, for immunization into mice, an adjuvant was prepared by mixing 1.5 parts by volume of Arlacel A (obtained from Hilltop Research, Miamiville, Ohio) with 8.5 parts by volume of Klearol (Bate Chemical, Don Mills, Ontario, Canada) and autoclaving for 20 minutes at 15 lb. pressure. Thereafter, heat-killed desiccated acetone-washed BCG bacteria (obtained from Connaught Laboratories, Willowdale, Ontario, Canada) were added at the rate of 5 mg. per 10 ml. of adjuvant. Apart from the change of adjuvent, the solution used to immunize the mice was prepared in the same way as that used to immunize the rabbits.

The mice were immunized three times, the primary immunization being given at day 0 with booster immunizations given at 21 and 38 days; the dose of conjugate injected on each occasion into mice was usually 100 .mu.g, though multiples of this dose were given where stated. Blood samples were collected from the mice at weekly invervals starting at day 14.

The rabbits were immunized three times at 21 day intervals (the primary immunization at day 0 and booster immunizations at 21 and 42 days), 1 mg. of the conjugate being given at each immunization. Blood samples were collected from the rabbits weekly beginning at day 21. The sera of the blood samples drawn from both the mice and the rabbits were separated from the cells and stored at -20.degree. C. prior to analysis.

The levels of antibody to both the peptide used and HCG were measured by isotopically labeling the antigen with .sup.125 I and reacting it with various dilutions of the antisera. In most cases, 250 pg. of labeled antigen was incubated with 200 .mu.l of diluted serum for 120 hours at 4.degree. C. Antigen binding was determined at three or more serum dilutions using a double antibody technique, the results being expressed as nanograms (ng) of antigen bound per milliliter of undiluted serum. The minimum sensitivity was 1 ng/ml. In some experiments, the level of antigen binding was tested using two concentrations of antigen with two dilutions of serum and expressing the results as M/L.times.10.sup.-10 by the same methods as in Example XXXI.

The antigen binding levels found in the various groups of rabbits and mice were compared using the two-tailed Mann-Whitney U-test described by S. Siegel, "The Case of Two Independent Variables", in Nonparametric Statistics, McGraw-Hill Book Company, New York (1956), p. 116. Results where p<0.05 were considered significant. In mice, in a few cases where the level of antigen binding was too low to be detected by the methods used, an arbitrary value of 0.1 ng/ml was assigned for statistical comparisons of values that were undetectable.

Mice were injected by the procedures set out above with doses of 100, 40, 8 and 1.6 .mu.g. of conjugates of Structure (II) and (XII) with tetanus toxoid. Table 9 below shows the levels of antibodies to both the peptide used and HCG at 21 and 35 days after the primary immunization. Analysis of these results shows that the differences between peptide binding in the sera between the mice injected with the two conjugates were significant at 21 days but not at 35 days. An increase in antibody levels to peptides was observed at 35 days with increasing doses of antigen injected, but again there was no significant difference between the antibody levels of the mice injected with the two conjugates at the same dosage levels.

To determine the effect of chain-length of the peptide on antibody response, groups of four rabbits were injected with conjugates of tetanus toxoid with the peptides of Structures (XVI), (XVII), (VI), (II), (XII), (XVII), and (XIX). Sera from animals immunized with these conjugates were reacted separately with equimolar quantities of labeled peptide and labeled HCG, and the proportion of antibodies reactive to peptides which were also reactive to HCG determined. The results are shown in FIG. 4.

Contrary to what might be expected, the maximum reactivity to HCG is not a simple function of the chain-length of the peptide. Maximum reactivity to peptide and HCG was obtained by rabbits receiving the conjugate of the peptide of Structure (XII), representing residues 109-145 of .beta.-HCG. The peptide of Structure (II), representing residues 111-145 of .beta.-HCG, produced antibodies nearly as reactive as the peptide of Structure (XII), but the antibody levels produced by the longer peptides of Structures (XVII) and (XIX), representing respectively residues 106-145 and 105-145 of .beta.-HCG, were lower than those produced by the peptides of Structures (XII) and (II). Not surprisingly, the shorter peptides also resulted in a lower proportion of antibody reacting to HCG.

A further series of tests were effected to determine the effect of the hexaproline cystine spacer sequences in the peptides of Structures (XIV) and (X) on antibody production. Conjugates of tetanus toxoid were prepared coupled to the peptides of Structures (II) (the 111-145 sequence without any spacer), (XIV) (the 111-145 sequence with a N-terminal spacer) and (X) (the 111-145 sequence with a C-terminal spacer), all conjugates containing a peptide:carrier ratio of approximately 20-22 peptides/10.sup.5 daltons of toxoid.

Table 10 below shows the antibody levels to HCG and peptides obtained in mice. Twenty-one days after primary immunization, the HCG antibodies were significantly higher in the mice immunized with the conjugates containing either of the spacer peptides than with the non-spacer peptide, Structure (II). Thirty-five days after primary immunization, the HCG antibody levels in mice immunized with the conjugate containing the C-terminal spacer peptide of Structure (X) were significantly greater than those of mice immunized with the non-spacer peptide, but the difference in antibody levels between the latter and the mice receiving the conjugate of the N-terminal spacer peptide of Structure (XIV) was not significant.

However, very different results were obtained when the serum antibodies were evaluated in terms of their ability to bind peptides. More of the labelled peptide was bound by the antibodies produced by the conjugate of the non-spacer peptide of Structure (II) 35 days after the primary immunization than by the antibodies produced by mice immunized with the conjugate of the N-terminal spacer peptide of Structure (XIV). There was no significant difference between the peptide binding abilities of the sera from mice immunized with the conjugates of the N-terminal and C-terminal peptides of Structures (XIV) and (X) respectively at this time.

The tests in rabbits were carried out with the same three conjugates as in mice, and also with the conjugate of tetanus toxoid and the peptide of Structure (XII) (the 109-145 sequence of .beta.-HCG) used in the preceding series of tests. The antibody levels 10-13 weeks after the primary immunization are shown in FIG. 5. These results show that the mean antibody levels in rabbits immunized with the conjugate non-spacer peptide of Structure (II) were lower than those of rabbits immunized with the spacer peptides of Structures (XIV) and (X), but the rabbits injected with the conjugate of the peptide of Structure (XII) had mean antibody levels to HCG comparable to the rabbits immunized with the conjugates of the spacer peptides of Structures (XIV) and (X).

A further series of tests was carried out to determine the effects of different carriers on antibody production. The peptide of Structure (XII), representing the 109-145 sequence of .beta.-HCG, was coupled to various carriers in a ratio of 15-28 peptides per 10.sup.5 daltons of carrier and mice and rabbits were immunized with these conjugates. The antibody levels to the peptide and to HCG were tested in the sera 21 and 35 days after the primary immunization. The results obtained in mice are shown in Table 11 below. Although the large standard deviations makes the detection of significant differences difficult, the results do show that the tetanus toxoid conjugate elicited antibody levels to both the peptides and HCG which were significantly higher than those produced by the conjugates of all the other carriers. Mean antibody levels in the groups injected with the peptide conjugated with flagellin and bovine gamma globulin were higher than those in which the peptide was linked to synthetic sugar (Ficoll) or polypeptide carriers. Table 12 shows the U-test analysis of the date presented in Table 11.

The results obtained in rabbits are shown in Table 13 below. The highest antibody levels were obtained in the rabbits immunized with conjugates of bovine gamma globulin, tetanus toxoid and diphtheria toxoid, there being no significant difference between the peak antigen titers of these three carriers. Significantly lower antibody levels were found in rabbits immunized with bacterial carriers, while synthetic polypeptide and sugar carriers produced antibody levels which were significantly lower than those of bacterial carriers.

A final series of experiments were performed to determine the effect of the peptide:carrier ratio on antibody production. Conjugates of tetanus toxoid and the peptide of Structure (XII) with peptide:carrier ratios of from 5-33 peptides per 10.sup.5 daltons of carrier were prepared and groups of four rabbits were immunized with these conjugates. The mean antibody levels to HCG produced 42, 63, and 84 days after primary immunization are shown in FIG. 6. Statistical analysis of the data in FIG. 6 shows no significant difference with peptide:carrier ratio 42 days after primary immunization, but 63 days after primary immunization the antibody responses to conjugates containing 23 or more peptides per 10.sup.5 daltons of toxoid are significantly greater than those of conjugates with lower peptide:carrier ratios. A similar comparison 84 days after primary immunization shows that the antibody levels produced by conjugates containing 16 or more peptides per 10.sup.5 daltons of carrier are significantly greater than those of conjugates with a lower peptide:carrier ratio. Accordingly, it is believed that it is advantageous to use a conjugate containing between 20 and 30 peptides per 10.sup.5 daltons of carrier.

When mice were immunized with the same conjugates, the responses were more variable and no linear dose-antibody response was observable, but the highest antibody levels were obtained in mice receiving conjugates containing 28-33 peptides per 10.sup.5 daltons of carrier.

The above results show that antibodies formed to peptides with 30 or more amino acids bind HCG better than those to peptides with fewer residues. However, since the 40 and 41 residue peptides (Structures (XVIII) and (XIX) above) were not as reactive to HCG as the 35 or 37 residue peptides of Structures (II) and (XII) above, it appears that no immunological determinant of HCG is present in the 105-109 region of the beta subunit thereof. Based upon the foregoing results, the preferred peptides for use in forming conjugates to produce antibodies to HCG are the peptides of Structures (XII) (the 109-145 sequence without spacers), (XIV) and (X) (the 111-145 sequence with N-terminus and C-terminus spacers respectively). The addition of the seven-residue spacer sequence to either the N-terminus or the C-terminus of the 111-145 peptide of Structure (II) produced higher antibody levels than the same peptide without spacer. It appears likely that a similar advantage can be produced by attaching similar spacer sequences to the 109-145 peptide of Structure (XII) since this peptide without spacers elicited responses similar to the 111-145 peptide with spacer (cf. results for Structures (XIV) and (XII)). On the other hand, it appears disadvantageous to attach peptides to the carrier at both position 122 and the N-terminus (see results from peptides of Structures (II) and (VI) above) since peptides attached to the carrier at both positions did not elicit levels equivalent to those attached at either terminus alone. Probably coupling of the peptide at both its midpoint and its N-terminus affects its conformation and creates an immunological determinant dissimilar to that found on intact HCG.

Moreover the results presented above strongly suggest that the best carriers for use in humans or other primates are tetanus toxoid and diphtheria toxoid. While the antibody levels in rabbits for bovine gamma globulin, tetanus toxoid and diphtheria toxoid are not significantly different, the antibody levels produces in mice with conjugates of the bovine gamma globulin are not as high as those produced by conjugates of the two toxoids. Immunization of humans or other primates with tetanus and diphtheria toxoids is acceptable and even advantageous (since a single vaccination can then provide protection against tetanus or diphtheria as well as an isoimmunogenic action), whereas injections of non-primate gamma globulins may not prove safe. Conjugates of either tetanus toxoid or diphtheria toxoid with a peptide:carrier ration of 20-30 peptides per 10.sup.5 daltons of carrier evoked large titers of antibody reactive to HCG and would therefore appear to be suitable for an anti-HCG vaccine.

This example illustrates the variations in antibody levels produced by changes in the adjuvant and vehicle used in conjunction with a modified polypeptide of the invention.

Based upon the results in Example XXXII above, the conjugates of tetanus toxoid with the peptides of Structures (XIV) and (XII) were selected as most efficacious in generating antibodies to HCG and were thus used in these experiments to select the optimum adjuvant and vehicle. The tetanus toxoid/peptide conjugates were prepared in exactly the same manner as in Example XXXII and purified by gel filtration and lyophilization. The conjugate of the peptide of Structure (XII) contained 21-25 peptides per 10.sup.5 daltons of carrier, while the conjugate of the peptide of Structure (XIV) contained 20-27 peptides per 10.sup.5 daltons of carrier. A further conjugate was prepared by conjugating the same tetanus toxoid to both the peptide of Structure (XIV) and the synthetic muramyl dipeptide CGP 11637 (manufactured by Ciba-Geigy Limited Basle, Switzerland--Formula (a) below). Using carbodiimide, the carboxyl group of the dipeptide was coupled to the amino groups of the tetanus toxoid, whereafter the peptide of Structure (XIV) was coupled to the tetanus toxoid via the remaining amino groups using the same procedure as in Example XXXII. The resultant conjugate contained five muramyl dipeptides and 31 peptides of Structure (XIV) per 10.sup.5 daltons of carrier respectively.

A total of eight different adjuvants were tested. The first five of these adjuvants were synthetic muramyl dipeptide hydrophilic analogues obtained from Ciba-Geigy Ltd., these five dipeptides being:

(a) CGP 11637, of the formula NAc-nor Mur-L. Ala-D. iso Gln;

(b) CGP 14767, of formula NAc-nor-Mur-L-Abu-D.isoGln;

(c) CGP 18177 of formula NAc-Mur(6-0-stearoyl)-L.Ala-D.isoGln;

(d) CGP 18741 of formula NAc-nor-Mur(6-0-stearoyl)-L.Ala-D.isoGln;

(e) CGP 19835 of formula NAc-Mur-L.Ala-D.isoGln-L.Ala-Cephalin.

The sixth adjuvant was another muramyl dipeptide obtained from Syntex Corporation, Palo Alto, Calif. being:

(f) DT-1, of formula NGlycol-Mur-L..alpha.-Abu-D.isoGln.

The last two adjuvants were lipophilic adjuvants manufactured by Ciba-Geigy Ltd., as follows:

(g) CGP 16940, of formula N-Palmitoyl-S-[2(R,S)-3-dipalmitoyloxy-propyl]-L-Cys-L-Ser-L-Ser-L-Asn-L-A la-L-Glu; and

(h) CGP 12908, a highly purified lipoprotein from the cell membranes of E. Coli B.

The vehicles used in these experiments were:

(a) an aqueous solution of 0.01 M sodium phosphate and 0.14 M sodium chloride, of pH 7.0, hereinafter designated BPS;

(b) Incomplete Freunds' adjuvant comprising 1.5 parts by volume of Arlacel A (mannide monooleate) and 8.5 parts by volume of Klearol, both reagents being obtained from the same sources as in Example XXXII, the adjuvant being referred to hereinafter as IFA;

(c) Squalene-Arlacel A, comprising four parts by volume Squalene (obtained from Sigma Chemicals, St. Louis, Mo. and one part by volume Arlacel A;

(d) Squalane-Arlacel A, comprising four parts by volume Squalane (obtained from Eastman Dodak, Rochester, N.Y.) and one part by volume Arlacel A;

(e) Peanut oil adjuvant, comprising 10 parts by volume peanut oil (obtained from Merck, Munich, West Germany) and one part by volume egg lechithin;

(f) Liposomes adjuvant, comprising 12 parts by weight egg lecithin and 1.6 parts by weight cholesterol; and

(g) Alum adjuvant, comprising 10 percent by weight potassium alum precipitated with 1 N sodium hydroxide, as described in M. W. Chase and C. A. Williams (eds.), Methods in Immunology and Immunochemistry, Vol. I, Preparation of Antigens and Antibodies, Academic Press, New York (1967), pp. 201-202.

The experimental animals used in these studies were the four inbred strains of mice C3H/He, C57BL/6, DBA/1 and SJL, obtained from the same source as in Example XXXII. The mice were retired breeders of more than 32 weeks of age and weighed 25-30 grams. Also used were genetically heterogenous New Zealand White rabbits weighing 2-4 kg. obtained from the same source. The mice were immunized subcutaneously with a primary immunization at day 0 and boosters at 21 and 38 days and, in some experiments, at day 55. Unles otherwise stated, each injection comprised 200 .mu.g. of conjugate and 100 .mu.g. of adjuvant. Blood samples were collected from the mice 28, 35, 45, 52, 62, and 69 days after the primary immunization.

The rabbits were immunized intramuscularly three times at 21 day intervals, each injection comprising 500 .mu.g. of conjugate and 500 .mu.g. of adjuvant unless otherwise stated. Blood samples from the rabbits were collected weekly on the 21st day after the first immunization. In the case of the blood samples from both the mice and the rabbits, the serum was separated from the remaining components of the blood and stored at -20.degree. C. prior to analysis. Complete Freund's adjuvant purchased or prepared as in Example XXXII above was used as a reference adjuvant. The levels of antibody in the blood sera reacting to the peptide in the conjugate or HCG were measured using the same double-antibody technique as in Example XXXII and the test results were evaluated using the same Mann-Whitney U-test as in that Example. In experiments in which antigen binding levels were pooled within each experimental group, Chi-squared analysis was used to determine significant differences.

In a first series of tests, various adjuvants were evaluated using IFA as the vehicle. The Structure (XII)/tetanus toxoid conjugate was combined separately with the adjuvants CGP11637, CGP 14767, CGP 12908 and CGP 16940. The resultant conjugate/adjuvant mixtures were incoporated into IFA emulsions and a parallel series of emulsions were prepared using the conjugate (without any adjuvant) in Complete Freunds' Adjuvant (CFA). Each separate emulsions was administered to four groups each comprising five mice from one of the four inbred strains. Antibody levels to Structure (XII) and HCG were determined 52 days after the primary immunization and the results are shown in Table 14 below.

The data in Table 14 reveal significant differences among the four strains of mice immunized with CFA. The differences between the levels of antibody to Structure (XII) between the DBA/1 and C3H/He groups on the one hand and the C57BL/6 group on the other are significant at the p=0.05 level. Moreover, the differences between the HCG antibody levels of the SJL group on the one hand and the DBA/1 and the C57BL/6 group on the other are also significant at the p=0.05 level.

Table 15 below presents statistical comparisons of antibody levels produced in mice receiving each of the four adjuvants in comparison with the antibody levels levels in mice receiving the CFA.

Of the four adjuvants tested, only the lipophilic adjuvant CGP 12908 induced responses in all four strains of mice that were not significantly different from those of the respective CFA groups, as regards antibody levels to either HCG or Structure (XII). The other lipophilic adjuvant CGP 16940, produced significantly lower levels of antibody to Structure (XII) in the mouse strains DBA/1 and C3H/He, and significantly lower levels of antibody to HCG in all four mouse strains. The hydrophilic adjuvant CGP 14767 produced Structure (XII) antibody levels significantly lower than those produced by CFA in only one mouse group, the strain DBA/1, while the Structure (XII) antibody levels produced by the other hydrophilic adjuvant CGP 11637 were significantly lower in the three mouse groups DBA/1, C3H/He and SJL than in the corresponding CFA groups. The HCG antibody levels produced by both the hydrophilic adjuvants CCG 11637 and 14767 were not significantly different from those of the corresponding CFA groups, only the SJL strain producing significantly different results in both cases.

Since the responses of the mice to immunization with the Structure (XII)/tetanus toxoid conjugate were more dependent upon the adjuvant used than the strain of mouse injected, genetic differences among the various strains were difficult to assess. However, the C57BL/6 strain mice did not show any significant differences in Structure (XII) antibody levels with any of the four adjuvants tested, as compared with the CFA immunized mice. On the other hand, mice of the DBA/1 strain did show significantly lower Structure (XII) antibody levels with three of the four adjuvants tested, as compared with the CFA immunized mice. Accordingly, in some of the later experiments only these two strains of mice were used for assessing genetic differences.

A parallel series of tests using the same four adjuvants, as well as the hydrophilic adjuvant DT-1 in conjunction with Structure (XII)/tetanus toxoid conjugate and IFA, produced somewhat different results, as shown in FIG. 7, which shows the mean HCG antibody levels averaged over the groups of four rabbits used, from 3-13 weeks after primary immunization. As with the mouse tests, a group of four rabbits was immunized with the conjugate incorporated into CFA for comparison. FIG. 7 shows that the HCG antibody levels in all groups were substantially constant from the 9-13 weeks after primary immunization, and therefore statistical evaluation of the antibody levels was conducted after pooling data within each group during this five week period.

As compared with the CFA immunized rabbits, only the rabbits receiving the adjuvant CGP 16940 produced significantly lower HCG antibody levels (p<0.05). The rabbits receiving adjuvants CGP 11637 and DT-1 produced HCG antibody levels significantly greater than the rabbits receiving adjuvants CGP 16940 and 12908 (p<0.05), but not significantly higher than the rabbits receiving adjuvant CGP 14767 or the rabbits receiving CFA.

Further tests were conducted in mice and rabbits to determine the effects of the various vehicles on antibody levels. In the mice tests, groups of five mice from each of the strains C57BL/6 and DBA/1 were immunized with the conjugate of tetanus toxoid and Structure (XIV) (the 111-145 sequence of .beta.-HCG with the N-terminal spacer sequence) in conjunction with one of the vehicles PBS, IFA, liposomes and Squalene/Arlacel. No adjuvants were used in immunizing the mice receiving the vehicles, but a group of mice from each strain were immunized with the conjugate incorporated into CFA for comparison purposes.

The mean Structure (XIV) and HCG antibody levels for each group of mice 35 and 52 days after the primary immunization are shown in Table 16, along with the standard deviations. Comparisons of the data in Table 16 as between the CFA Control group with the groups receiving the various other vehicles by the aforementioned U-test are shown in Table 17 below. Table 17 shows that the Structure (XIV) antibody levels in the DBA/1 mice were significantly higher than those in the C57BL/6 mice at either 35 or 52 days after the primary immunization. In terms of HCG antibody levels the DBA/1 mice were significantly lower than those of the C57BL/6 mice 35 days after immunization, but not significantly lower 52 days after primary immunization.

The U-test factors in Table 17 show that, in the C57BL/6 mice groups 35 days after primary immunization, only the group receiving the IFA did not have significantly lower levels of antibodies to both Structure (XIV) and HCG than those of the mice receiving CFA. However, the results 52 days after primary immunization are strikingly different: no significant differences between Structure (XIV) or HCG antibody levels existed at that time between the mice receiving CFA and those receiving BPS, IFA or liposomes vehicles. Indeed, the Squalene/Arlacel vehicle immunized mice had levels of antibodies to both Structures (XIV) and HCG which were significantly higher than those of any other vehicle including CFA. The responses of the DBA/1 mice were not similar to those of the C57BL/6 mice. Thirty-five days after primary immunization, the four groups of DBA/1 mice immunized with the vehicles under test had Structure (XIV) antibody levels significantly greater than those of the mice receiving CFA, but only the group of mice receiving liposomes vehicle had HCG antibody levels significantly lower than the CFA group. In all the groups of DBA/1 mice 52 days after primary immunization, the antibody levels to both Structure (XIV) and HCG were not significantly different from the CFA group.

A similar series of tests were run in rabbits using the four vehicles Squalane/Arlacel A, Squalene/Arlacel B, peanut oil and alum. These tests were conducted using the same (Structure XIV)/tetanus toxoid conjugate, but the adjuvant DT-1 was used. Again, a control group of rabbits were immunized with the conjugate emulsified in CFA without a vehicle. Antibody levels to HCG were measured throughout the 13 week immunization period and the data obtained from each group of rabbits at the week of peak antibody level were pooled with the values obtained from the same group one week earlier and one week later. The results are shown in Table 18 below. The HCG antibody levels in rabbits receiving Squalene/Arlacel were significantly higher than those of all other groups, including the rabbits receiving CFA. No significant differences in HCG antibody levels existed between the rabbits receiving Squalane/Arlacel and those receiving CFA, while the rabbits receiving peanut oil or alum vehicles had significantly lower HCG antibody levels than those receiving CFA.

The efficacy of the Squalene/Arlacel A vehicle in increasing the antibody levels in the injected animals is surprising, especially since Squalane/Arlacel has not previously been used as a vehicle in a vaccine, although Squalene is used in topical preparations such as ointments and cosmetics. Although the Squalane/Arlacel A vehicle was not as effective as the Squalene/Arlacel vehicle, it was as efficacious as CFA. Both the Squalane/Arlacel and the Squalane/Arlacel vehicles should be clinically acceptable for use in human beings, since they appear to produce little or no irritation at the sight of injection, whereas CFA is known not to be clinically acceptable for use in human beings since it tends to produce intense irritation, absesses, etc. at the point of injection.

To evaluate simultaneously combinations of various adjuvants and various vehicles, and thus to detect any second order effects due to the interactions of particular adjuvants with particular vehicles, groups of five or six mice from each of the strains C57BL/6 and DBA/1 were immunized with preparations comprising the same Structure (XIV)/tetanus toxoid conjugate used in the preceding tests in combination with one of the synthetic adjuvants CGP 18177, 18741 or 19834 and one of the three vehicles Squalene/Arlacel, liposomes and peanut oil. As in the preceding test described above with reference to Tables 16 and 17, HCG antibody levels were determined in sera collected 35 and 52 days after primary immunization, and the results are shown in Table 19 below. The data in Table 19 reveal no significant differences between the levels of antibodies in the groups of mice injected with the same adjuvant in different vehicles, nor between the groups receiving the same vehicle and different adjuvants, at either 35 or 52 days after primary immunization (i.e. p>0.05 in all cases). Furthermore, no differences significant at the p=0.05 level were detected between the levels between corresponding groups of mice of difference strains.

Table 20 below shows data similar to those in Table 19 but relating to Structure (XIV) antibody levels instead of HCG antibody levels. Analysis of the date in Table 20 shows that 35 days after primary immunization in the C57BL/6, mice, mice receiving the adjuvant CGP 18177 produced significantly higher levels of antibodies than the group receiving adjuvant CGP 19835 (p<0.05). However, the similar differences at 52 days after primary immunization were not significant at the p=0.05 level. Comparison of the three adjuvants of the DBA/1 mice 35 days after primary immunization shows no significant differences between the adjuvants in Squalene/Arlacel or peanut oil vehicles but the Structure (XIV) antibody levels produced by CGP 18177 in liposomes vehicle were significantly higher than those produced by CGP 19835 in the same vehicle (p<0.05). The results 52 days after primary immunization show the same pattern of significant differences as the results 35 days after primary immunization.

While mean HCG antibody levels were generally higher in mice immunized with Squalene/Arlacel vehicle and (Structure XIV) antibody levels were consistently higher using this vehicle, regardless of adjuvant or mouse strain, these differences were not significant in view of the variability in the responses of the mice. Although no significant differences were found between the two strains of mice as regards to HCG antibody levels, the DBA/1 mice produced higher mean Structure (XIV) antibody levels in all vehicle and adjuvant groups than the C57BL/6 mice although the differences were not statistically different.

A similar series of tests were then carried out in rabbits using the same Structure (XIV)/tetanus toxoid conjugate, the adjuvants CGP 18177, 18741, and 19835 tested in mice, plus the additional adjuvants CGP 11637 and DT/1, using groups of four rabbits and Squalene/Arlacel as the vehicle. Again, a control group of rabbits were immunized with the conjugate incorporated into CFA without a vehicle. The mean HCG antibody levels in sera collected from the various groups of rabbits weekly from 3-12 weeks after primary immunization are shown in Table 21 below, together with the corresponding standard deviations. The data in Table 21 show that the highest antibody levels were achieved in rabbits receiving the adjuvants CGP 11637 and DT/1 10 weeks after primary immunization. Comparison of the antibody levels produced by these adjuvants with the control group receiving CFA showed that the increases in antibody levels with both adjuvants were significant at the p=0.05 level. Adjuvant CGP 18177 also produced higher antibody levels than CFA, the difference being significant at the p=0.05 level, and the differences between the antibody levels produced by the three adjuvants CGP 11637, DT/1 and CGP 18177 are not significant at the p=0.05 level. The adjuvants CGP 19835 and 18741 produced antibody levels which were lower than the CFA rabbits, the difference being significant at the p=0.05 level.

A final series of tests were run in mice to determine the effect of coupling one of the adjuvants to the peptide/tetanus toxoid conjugate instead of merely administering the peptide/tetanus toxoid conjugate mixed with the adjuvant. Groups of five mice from each of the strains C57 BL/6 and DBA/1 were immunized with the Structure (XIV)/tetanus toxoid/CGP 11637 conjugate described above, separate groups of mice being immunized with the conjugate in association with each of the vehicles PBS, IFA, liposomes and Squalene/Arlacel. As before, a Control group from each strain was immunized with the conjugate incorporated into CFA for comparison purposes.

FIG. 8 shows the mean HCG and Structure (XIV) antibody levels in the mice as a function of time after primary immunization (the coding for the various vehicles in FIG. 8 is the same as that in FIG. 7). FIG. 8a shows the HCG antibody levels of the C57BL/6 mice. The elevated antibody levels produced by CFA and IFA 35 days after primary immunization are not significant at the p=0.05 level and were not sustained after the booster immunizations at 33 and 55 days. The mice receiving the conjugate in Squalene/Arlacel vehicles did not respond significantly until the third bleeding, 45 days after primary immunization, but there HCG antibody levels increased progressively thereafter and were higher by an amount significant at the p=0.05 level as compared with those of the other mouse groups. The Structure (XV) antibody levels of the same mice shown in FIG. 8b parallel those of the HCG antibody levels and statistical analysis shows that the same differences are significant.

The HCG antibody levels of the DBA/1 mice shown in FIG. 8c are considerably lower than those of the corresponding C57BL/6 mice. Fifty-two days after primary immunization, the HCG antibody levels in the DBA/1 mice receiving the Squalene/Arlacel vehicle were not significantly different from those of the CFA Control group (p>0.05), but were significantly higher than the levels achieved in the mice receiving the other vehicles (p<0.0). Sixty-nine days after primary immunization, there were no differences significant at the p=0.05 level between any of the groups of DBA/1 mice. However, as shown in FIG. 8d, unlike the results obtained with the C57Bl/6 mice, in the DBA/1 mice, the antibody levels to Structure (XIV) do not follow the antibody levels to HCG. The Structure (XIV) antibody levels in the DBA/1 mice receiving CFA or IFA were elevated 28 days after the primary immunization and thereafter declined until day 52. Later, the CFA immunized mice levels rose in response to the booster injection administered at day 55, while the antibody levels in the IFA mice continued to fall. No differences significant at the p=0.05 level were found between the antibody levels in the mice receiving CFA IFA or Squalene/Arlacel 52 days after the primary immunization, but at this time the differences between the mice receiving CFA and those receiving PBS and liposomes vehicles were significant at the p=0.05 level. After 69 days from primary immunization, while the difference between the CFA groups and the remaining vehicles were significant at the p=0.05 level, the differences between the other groups were not significant at this level.

The experimental results described above clearly demonstrate the efficacy of some of the synthetic adjuvants in increasing antibody production when administered in conjunction with the modified polypeptide of the invention. Some of the adjuvants, in particular CGP 11637 and 18177 and DT/1, when administered in certain vehicles, especially IFA and Squalene/Arlacel produced antibodies exceeding those produced by CFA. In general, the hydrophilic muramyl peptide adjuvants were superior to the lipophilic adjuvants in enhancing the antibody production caused by the hydrophilic Structure (XIV)/tetanus toxoid conjugate. The experimental results also show that the delivery system used to administer the antigen and adjuvant is of critical importance in producing an enhanced response to the conjugate. For example, the antibody responses of the C57BL/6 mice which received the conjugate in Squalene/Arlacel without adjuvant were greater than those of the mice receiving the conjugate in CFA. Adding an adjuvant to this vehicle produced only a slight increase in the antibody levels. These observations are confirmed by those in rabbits receiving the conjugate and the synthetic adjuvant in different vehicles, and different adjuvants in the same vehicle. Squalane/Arlacel and peanut oil emulsions were efficacious, though less effective than Squalene/Arlacel.

To sum up, based upon all the foregoing data it appears that the optimum vehicle was Squalene/Arlacel, which was superior to the others in almost every formulation and that the best adjuvants are CGB11637 and DT/1, which were more efficacious in various vehicles than other adjuvants tested simultaneously. Squalane/Arlacel was also a highly acceptable vehicle although somewhat less efficacious than Squalene/Arlacel.

This example illustrates a use of a modified polypeptide of the invention in repressing a carcinoma which produces a chorionic gonadotropin-like material. This material by John A. Kelen, Arnost Kolin and Hernan F. Acevedo is in press and will shortly appear in "Cancer".

The rat mammary adenocarcinoma R 3230 AC has been shown, using immunocytochemical, radioimmunoassay (RIA) and bioassay methods in tissue sections and cell cultures to product a chorionic gonadotropin (CG)-like material; however, this CG-like material cannot be detected in the sera of the animals bearing the carcinoma. The detection of the CG-like material is described in:

(A) Malkin et al., Immunohistochemical Detection of Ectopic Hormones in Experimental Rat Tumors, in F. G. Lehmann (ed.), Carcino-embryonic Proteins, Vol. 2, Elsevier-North Holland Biomedical Press, New York, Amsterdam (1979), pp. 751-758.

This adenocarcinoma can be propagated in cell culture and its cells retain their morphology and malignant characteristics. Intravenous injection of these cells into isologous Fischer 344 rats gives rise to numerous foci of this neoplasm in the lungs within 8-10 days, the animals then becoming very sick and then dying within 12-15 days.

The R 3230 adenocarcinoma (obtained from Dr. A. Bogden, Mason Research Institute, Worcester, Mass.) was cultured from explants of subcutaneous tumors grown for 20-21 days in Fischer 344 female rats. The cultures were maintained in RPMI 1640 medium supplemented with 10% fetal cough serum. Upon reaching confluency, the cells were dispersed by means of trypsin-EDTA (1:250) and passaged five times. During the early stationary growth phase, the cultured cells were dispersed for in vivo administration. The cell density was determined using a hemocytometer and viability was tested by the Trypan blue exclusion method. One million cells were then injected into the tail vein of each of the experimental animals, which were 100 female Fischer 344 rats weighing 130-150 g. The animals were divided into two matched control groups of 15 animals each and a test groups of 70 animals. They were individually caged, fed Purina (Registered Trade Mark) chow and tap water ad libitum. All animals except six from the test group were sacrificed at intervals by cervical dislocation for necropsy studies. The lungs, livers, spleen and kidneys were examined and processed for paraffin sections and H&E staining in order to detect neoplastic foci, the diameter of the foci being measured with an eye piece micrometer. All animals were also bled before sacrifice and the serum tested for the presence of antibodies to CG.

The test group of 70 rats received, prior to the injection of carcinoma cells, a .beta.-HCG/tetanus toxoid modified polypeptide of the invention. This .beta.-HCG conjugate, and its method of preparation, is described in U.S. Pat. No. 4,161,519 to Talwar and in the following papers written by the same author and others:

(b) Proc. Natl. Acad. Sci. U.S.A. 73, 218-222 (1976);

(c) Contraception, 18, 19-21 (1978);

(d) Fertil. Steril. 34, 328-335 (1980).

The conjugate was administered in saline once per week for a period of three weeks, approximately 1.3 .mu.g. being given at each injection and the last injection being given two weeks before administration of the carcinoma cells. In order to study the effect of the conjugate over an extended period a number of animals selected at random were sacrificed for autopsy, as follows (the day when the carcinoma cells were injected being taken as day 0):

One of the two control groups was administered highly purified tetanus toxoid (obtained from Connaught Laboratories, Toronto, Ontario, Canada) at a rate of 0.7 .mu.g. per injection, following the same injection schedule as the test group followed with the conjugate. Thus, this control group received a dose of tetanus toxoid substantially equal to that received in the conjugate by the test animals. The second control group received no treatment other than the tumor suspensions. All the animals from the test group were sacrificed for autopsy on day 10.

Testing for the presence of HCG antibodies was performed using the HCG-.beta.-RIA kit, Quantitative Method II (manufactured by Serono Laboratories, Braintree, Massachusetts. To the cold standard HCG, HCG free serum and .sup.125 I-HCG, an aliquot of the rat serum was added. After overnight incubation, the original protocol of the kit was followed in the subsequent procedure. If all the labeled HCG was preciptated (thus indicating a high antibody titer) the test was repeated with appropriately diluted rat serum.

All the animals in the two control groups, sacrificed 10 days after injection of the tumor cells, showed multiple lung foci of neoplastic cells in accordance with the normal progress of this carcinoma. None of the 30 animals in the control groups showed neoplasms in any other organs. The qualitative histological appearance of the pulmonary tumor deposits was practically identical in every animal. The neoplasm is a moderately differentiated adenocarcinoma rarely producing glandular lumina. The mitotic rates are high, but tumor necrosis and inflammatory reactions are absent. The tumor nodules range in size from a few cells to more than one millimeter in diameter and are uniformly distributed throughout the long parenchyma. Using a 2.5.times.objective, from 5-10 tumor nodules were present in every low power camera field in the controlled animals. In contrast, the test animals that had received the .beta.-HCG/tetanus toxoid conjugate rarely displayed more than 1 to 4 metastatic nodules not exceeding 0.3 mm. in diameter in the entire lung sections 8-10 days after administration of the tumor cells, so that only a few camera fields could be found containing a single metastatic focus. Most lung parenchyma were completely free of neoplasms. The histological appearance of the neoplasms in the test animals was the same as that in the controls.

In all the rats receiving the conjugate, a significant antibody titer was found, antibody levels capable of precipitating 200 mIU of HCG standard per milliliter of serum being consistently determined. This titer persisted for up to 120 days after receipt of the tumor cells and continued to protect at that time against formation of metastatic lung foci after a new intravenous seeding with the tumor cells. In contrast, neither of the control groups showed any measurable HCG antibody levels, regardless of whether the rats had or hot not received the tetanus toxoid.

Thus, the protective effects of immunization with the conjugate prior to injection with the tumor cells were demonstrated by the test animals being alive 20 days after injection of the tumor cells (at which time, 100% mortality would be expected in unprotected animals), by the absence of lung pathology in the animals sacrificed more than 20 days after injection of the tumor cells, by the long term survival (more than six months) and lack of deterioration in the six test animals that were not sacrificed, and by the fertility of these six surviving test animals, several of which produced normal litters after termination of the 120 day observation period. This long term survival of all the six animals and absence of any deterioration therein is especially surprising in view of the virulence of the R 3230 A carcinoma chosen for study, since previous work has indicated that no rats receiving the intravenous injection of tumor cells received by these test animals can be expected to survive for more than about 20 days.

Since there are no detectable levels of serum HCG associated with this carcinoma in rats, the absence of pathological changes in the organs investigated and the maintenance of the reproductive functions in the surviving test rats suggest that the antibodies act at the level of the cell membrane, probably in a cytotoxic manner, although the above experimental results are not sufficient to prove this hypothesis.

This example illustrates the use of a modified polypeptide of the invention coupled to diphtheria toxoid in repressing fertility in baboons.

A modified polypeptide of the invention based upon Structure (XII) above was prepared in exactly the same manner in Example XXXI except that the peptide was conjugated to diphtheria toxoid instead of tetanus toxoid. Also, the resultant conjugate, again containing about 22 peptides per 100,000 daltons of the toxoid, was dissolved in a solution of the muramyl dipeptide CGP 11637 used in Example XXXIII, instead of the Complete Freunds' adjuvent used in Example XXXI. The resultant conjugate/adjuvant mixture was emulsified with an equal volume of the 4:1 v/v Squalene/Arlacel A vehicle used in Example XXXIII. Again, the control animals received diphtheria toxoid in an amount equal to that received as part of the conjugate by the test animals. All details regarding animal feeding, group size, animal housing, mode of administration of vaccine and blood testing were exactly as in Example XXXI.

The length of the menstrual cycles and the luteal phase progesterone levels for both the test and control groups were measured for three menstrual cycles before immunization and five menstrual cycles after immunization. The results are shown in Table 22 below (which is directly comparable with Table 7 of Example XXXI). The data in Table 22 shows that, as in the previous experiment, no significant differences existed between the lengths of menstrual cycle and luteal phase progesterone levels of the two groups of baboons either before or after immunization.

Table 23 below shows the antibody levels to Structure (XII), baboon chorionic gonadotropin and human chorionic gonadotropin produced in the test animals injected with the conjugate during the five menstrual cycles following immunization. As in Example XXXI, the antibody levels were determined from serum drawn from the early luteal phase of each menstrual cycle. Thus, the data in Table 23 may be compared directly with those in Table 5 of Example XXXI. Comparing Tables 5 and 23, it will be seen that the diphtheria-toxoid-containing conjugate produced slightly lower levels of antibodies to baboon chorionic gonadatropin than did the tetanus-toxoid-containing conjugate (though this difference does not appear to be significant), but rather higher levels of antibodies to human chorionic gonadatropin (and the differences between the two groups for menstrual cycles Nos. 4 and 5 would appear to be significant).

As in Example XXXI, each of the baboons was mated during the course of the third menstrual cycle following immunization and also mated during the two subsequent menstrual cycles unless, of course, a previous mating produced a pregnancy. The results are shown in Table 24, which is directly comparable with Table 8 of Example XXXI. Of the control animals immunized only with diphtheria toxoid, 11 out of 15 became pregnant on the first mating, three of the remaining four became pregnant on the second mating and the remaining baboon was still not pregnant after the third mating. Thus, a total of 20 matings produced 14 pregnancies and a fertility rate of 70%. On the other hand, in the test animals immunized with the conjugate, none of the 15 baboons became pregnant during the first mating, only one of the remaining baboons became pregnant at the second mating, and at the third mating, another one of the 14 remaining baboons became pregnant. Thus, a total of 44 matings produced only two pregnancies for a fertility rate of 4.6 %. Although statistical comparisons are difficult because of the very small numbers of pregnancies involved in the conjugate-immunized baboons, it thus appears that the vaccine used in this Example containing Structure (XII)/diphtheria toxoid conjugate is more effective in preventing conception in baboons than the vaccine used in Example XXXI.

It will be appreciated that numerous changes and modifications can be made in the embodiments of the invention described above without departing from the scope of the invention. Accordingly, the foregoing description is to be construed in an illustrative and not in a limitative sense, the scope of the invention being defined solely by the appended claims.