Citation
The identification of anti-idiotypic antibody during an immune response in dogs

Material Information

Title:
The identification of anti-idiotypic antibody during an immune response in dogs
Alternate title:
Anti-idiotypic antibody during an immune response in dogs
Creator:
Schultz, Kevin T., 1951-
Publication Date:
Language:
English
Physical Description:
vii, 150 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Anti idiotypic antibodies ( jstor )
Antibodies ( jstor )
Antigens ( jstor )
Canines ( jstor )
Dogs ( jstor )
Haptens ( jstor )
Immune response ( jstor )
Immunization ( jstor )
Monoclonal antibodies ( jstor )
Standard deviation ( jstor )
Dissertations, Academic -- Immunology and Medical Microbiology -- UF ( mesh )
Histocompatibility Antigens Class II ( mesh )
Hypersensitivity, Immediate ( mesh )
Immunoglobulin E ( mesh )
Immunoglobulin Idiotypes ( mesh )
Immunology and Medical Microbiology thesis Ph.D ( mesh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida.
Bibliography:
Includes bibliographical references (leaves 140-148).
Additional Physical Form:
Also available online.
General Note:
Photocopy of typescript.
General Note:
Vita.
Statement of Responsibility:
by Kevin T. Schultz.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
0029406388 ( ALEPH )
10187483 ( OCLC )

Downloads

This item has the following downloads:


Full Text












THE IDENTIFICATION OF ANTI-IDIOrYPIC ANTIBODY
DURING AN IMMUNE RESPONSE IN DOGS









BY





KEVIN T. SCHULTZ











A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY


UNIVERSITY OF FLORIDA 1983



























This dissertation is dedicated to -my wife Nancy. Without her love and support (and typing), it would never have been.












ACKNOWLEDGEMEWTfS



I wish to express my deep appreciation and thanks to some very important people in my life. First of all, to my

parents, my wife and my family for their constant love and support.

Very special thanks are extended to Dr. Richard

Halliwell for his help, guidance and friendship throughout my graduate experience.

To Drs. Gail Kunkle, Robert Mason and Bill Holloman, my deep appreciation for their unwavering friendship (and occasional needed prods) along my journey.

I also wish to thank Drs. K.I. Berns, M.D.P. Boyle, R.B. Crandall, A.P. Gee, G.E. Gifford, M.J.P. Lawman, and P.A. Small, Jr., for their suggestions, help and

encouragement.

Last (but mt least), I would also like to thank my fellow graduate students and all my friends for making this experience a nemrable one.








iii












TABLE OF CONTENTS

page

ACKNOWLEDGEMENTS ............................ iii
ABSTRACT .................................... vi

CHAPTER ONE INTRODUCTION .................. 1


CHAPTER TW0 THE INDUCTION AND KINETICS OF AN
ANI-DNP IGE RESPONSE
Introduction .......................... 17
Materials and Methods................. 18
Results ............................... 31
Discussion ............................ 55
SUrrary ..................................57
Conclusions ........................... 57

CHAPTER THREE ATTEMPTS TO REGULATE AN ANTIBODY
RESPONSE WITH AUIOLOGOUS ANTIBODY
Introduction .......................... 59
Materials and Methods ................. 60
Results ............................... 62
Discussion ............................ 74
Sunmmry and Conclusions ............... 76

CHAPTER FOUR THE IDENTIFICATION OF ANTIIDIOIYPIC ANTIBODY
Introduction .......................... 77
Materials and Methods .....................78
Results ............................... 82
Discussion ............................ 94
Summary .............................. 99
Conclusions ........................... 99

CHAPTER FIVE DETECTION OF ANTI-IDIOTYPIC
ANTIBODY USING AUITOLOGOUS ANTI-DNP F(AB)' FRAGMENTS AS THE IDIOTYPIC
ANTIGEN
Introduction .......................... 99
Material and Methods .................. 100
ReSults ............................... 103
Discussion ............................. 130
Summary and Conclusions .............. 135


iv










CHAPIER SIX CDNCLUSIONS ....................... 137

REFERENCES ..................................... 140

BIOGRAPHICAL SKETCH ............................ 149













































V











Abstract of Dissertation Presented to the Graduate
Council of the University of Florida
In Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy


THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
DURING AN IMmUNE RESP ONSE IN THE DOG


By

Kevin T. Schultz

August, 1983

Chairman: Richard E. Halliwell
Major Departmnent: Immnunology and Medical Microbiology

This investigation commensed with the development of an animal model to study the synthesis of IgE antibody.

Repeated himmnunizations with a haptenated parasite extract (dinitrophenol coupled ascaris) in young dogs resulted in

the production of anti-DNP antibody of the IgE, IgG and IgM class.

Although attempts to regulate this anti-hapten antibody response by administration of autologous anti-DNP antibody were unsuccessful, such therapy did result in the production of anti-idiotypic antibodies. These anti-idiotypic antibodies were demonstrable using mouse hybridoma-derived anti-DNP antibodies.



vi











This antibody was shown to be anti-idiotypic rather than an internal image of antigen because it bound to only two of four monoclonal anti-DNP antibodies and failed to inhibit the id anti-id interaction with hapten. Antiidiotypic antibodies were detected during the immunization schedule in three of five dogs using autologous anti-DNP F(ab)'2 fragments as the source of idiotypes.

The anti-idiotypic antibodies identified using the

mouse monoclonal antibody were the result of the immunization procedure and did not appear to be physiologically relevant to regulation of the immune response. On the other hand, the anti-idiotypic antibodies identified with the autologous source of idiotypes appear to be produced during the DNP-ASC immune response and were detected before autologous antibody immunization. The antigens that induced the anti-idiotypic response appeared to be, in this case, the idiotypes on the anti-DNP antibody that were produced from the DNP-ASC immunization.









vii











GiAPTER ONE
INMI'ODUCTION




Allergic diseases of the immediate type are very important pathologic disorders in both man and dogs. Clinical signs are initiated by an interaction of antigen and IgE antibody with resultant me~diator release fron mast cells and basophils. In man and dogs, allergic reactions cause considerable morbidity and can be fatal (1,2). Anaphylaxis fron a bee sting is a classic example for both species.

Atopy in man is an inherited disease which is

associated with an antigen specific IgE response against environmental allergens. This disease is expressed clinically as asthma, hay fever, atopic dermatitis or any combination of these three (3). 'The dog is an excellent experimental animal model to study IgE me~diated hypersensitivity for atopic diseases of man because of the similarity of the allergic reaction in both species (1,2,4). Canine IgE shares many physicochemical properties with human IgE (5,6,7). Doigs like mn, develop spontaneous disease associated with increased synthesis of IgE antibody (3,4) and in dogs the disease is also familial (4).









2



There are a number of unique features of IgE anti body synthesis. Firstly, IgE circulates in very small amounts as compared to other antibody classes. In iran, the serum level of this antibody is about 1/10,000 the level of serum IgG

(3), and in dogs serum IgE is about 1/100 the level of serum IgG (7). Serum IgE levels of internally parasitized people and dogs are elevated as compared to non-parasitized individuals (7,8,9). 'The higher IgE level in dogs is felt to be the result of a greater parasite burden in this species (7).

Secondly, IgE is produced predominantly locally by

lymph nodes in the respiratory and gastrointestinal tracts as well as in regional lymph nodes (10). These observations have led to the suggestion that this immunoglobulin is important in host defense of mucosal surfaces and particularly against parasites. Furthermo~re, IgE has been shown to participate in parasite killing through antibody dependent cell me~diated cytotoxicity (11). Thirdly, the antigens that stimulate IgE antibody are usually very complex and heterogenous substances such as allergens or parasites and their extracts. When an animal is exposed to these antigens, the antibody response usually includes high titer IgE antibody whereas bacteria and viruses usually do not induce IgE antibody in spite of being very immunogenic (12).









3

The induction of IgE antibody experimentally requires special conditions. For example, high doses of antigen and strong adjuvants such as complete Freund's adjuvant are

unfavorable to the development of an IgE response whereas low doses of antigen in an adjuvant such as aluminum hydroxide tend to favor IgE production (13). Furthernrre, if haptens are coupled to parasite extracts, high titer anti-hapten IgE antibody responses will frequently develop. However, if the same hapten is coupled to a T-independent antigen or a different T-dependent carrier there is usually no IgE response (13,14). This suggests that IgE production is dependent on both the carrier and T-cells. The reason why parasites and their extracts are efficient inducers of IgE antibody is not fully understood. It is known that this enhancing effect is modulated through factors produced by T cells. Ishizaka's group (15) have shown that T cells derived from (N. brasiliensis) parasitized rats produce an IgE-potentiating factor which selectively potentiates a non-specific IgE antibody response. This factor has affinity for IgE, binds to IgE-bearing B cells through surface IgE and enhances th e differentiation of these cells into IgE forming cells. A factor with similar properties has been produced from T cells obtained fran patients with hyper-IgE syndrome, suggesting that the regulatory factors









4

and pathways for enhancing IgE antibody production for parasites and IgE in general might be similar (16).

A number of approaches have been used in an attempt to control allergic disease. Avoidance of the antigen is one approach, but this is rarely possible. Drugs that inhibit mediator release or control the effects of that release are also employed, but they have side effects and often require continual therapy. However, the ideal approach would be to regulate the production of the unwanted IgE antibody.

The mechanisms used to regulate IgE antibody responses involve either the inactivation of B cell precursors or the manipulation of T cell populations. To this end, primary and ongoing antibody responses, including IgE antibody, have been suppressed in mice by antigen coupled to non-inmunogenic carriers such as d-glutamine-d-lysine (dGL) or polyvinyl alcohol (17,18). Both of these carriers inactivate hapten-specific B cells and can induce hapten-specific suppressor T cells. Moreover, when dGL, is coupled to proteins rather than haptens, the resultant suppression in mice is isotype specific (i.e. suppresses IgE alone) (19).

Unfortunately, there are no published results of the use of this compound in dogs or man.

Hyposensitization has also been used in an effort to control allergies in both man and dogs (20-22). The mechanism by which it works is not clear. It is known that









5
IgG antibody can have a role in regulating allergic symptoms

(22). An IgG response can be induced by administration of

allergen either by the normal route of exposure or by a route other than for normal exposure (i.e. subcutaneous versus inhalation). This IgG antibody presumably completes the allergen-IgE antibody interactions (21). However, this therapy is not without side effects (20). Furthermore, only about 65 per cent of patients treated with hyposensitization have clinical improvement (20).

An alternate type of hyposensitization involves

modifying the allergen, usually by mild denaturation. Studies in mice with urea denatured ragweed showed that such treatment reduced allergenicity while maintaining immunogenicity of the allergen. If large doses of urea denatured ragweed were given to mice previously sensitized to unmodified ragweed, such therapy resulted in antigenspecific T suppressor cell induction without the development of anaphylaxis (12). These cells suppressed the antiragweed IgE response. A controlled study is underway to determine if this form of immunotherapy is any more effective in controlling allergic symptoms than conventional hyposensiti zation.

Another approach is to regulate the response with

products of the immune system. Smith (23), in 1909, was the first person to recognize that antibody could suppress the









6

develoment of an immune response. In these experiments he showed that certain mixtures of diphtheria toxin and antitoxin could be very immunogenic in guinea pigs, but if there was a large excess of antitoxin, the immunized guinea pig wnuld fail to -mount an immune response against the toxin. Numerous studies in the 1950's and 1960's verified this observation and also demonstrated that the isotype, amount, affinity and time of administration were important variables in determining the degree of suppression that passive antibody had on the immune response (reviewed in 24). For example, IgG antibody given after antigenic exposure was more effective in inducing antibody suppression than IgM antibody. Further, the suppressed state was longer lived using IgG than IgM antibody. An interesting report by Chan and Sinclair (25) stated that the administration of antiSRBC antibody given to mice after antigenic challenge led to a suppression of this response and this tolerant state could be transferred from one mouse to another with T-cells from

the toleri zed mouse. They suggested that the regulatory action of antibody operated through some sort of "induced pathway or secondary immune response" (25 p. 977).

In the early 1970's it was likewise shown that IgE antibody could be regulated by passively administered antibody (26-28). Rabbits were immunized to produce high titer IgE antibody and were given passive antibody 24 hours









7
after antigenic challenge. A complete inhibition of the passive cutaneous anaphylaxis titer and a marked decrease in the hemagglutination titer of these rabbits resulted as compared to controls (26). It as shown by Tada and Okumura

(27) that, in the rat, the administration of anti-DNP ascaris antibody resulted in marked suppression of a preexisting IgE antibody response and this suppression was maintained for an extended period of time. This was in contrast to studies in the muse in which administration of anti-ovalbumin IgG had little effect on the preexisting anti-ovalbumin IgE response (28). These differences were explained as species variation. Alternatively, they ay be due to the difference in the antigenic system employed.

One explanation for the mechanism of regulation by

passive antibody is that the administration of this antibody acted as an antigen and stimulated an anti-antibody response. Lahss et al. (29) were the first to show that some anti-antibodies would bind to structures on antibody close to or within the antigen combining site. These determinants have been named idiotypes (id) and the immune response directed to them is termed an anti-idiotypic (anti-id) response. In 1974, Jerne (30) proposed his network hypothesis of antibody regulation. The basic premise of this theory is that the immune system is regulated by a network of interactions between id and anti-id. A number of








8
assumptions are crucial premises to this theory. Firstly, most idiotypes exist at a level too low to induce tolerance. Thus, antigenic stimulation and expansion of these id will stimulate the production of a reciprocal set of anti-id. The id is then regulated directly by the anti-id, indirectly by the anti-id on T-cellsor or by anti-id acting on T-cells. As the concentration of anti-id reaches some critical threshold, a second anti-id response develops which is specific for the id of the anti-id. This anti-id would, therefore, be an anti-(anti-id) and would then stimulate a fourth response and so on, thereby resulting in an interrelated network of regulation between antibody molecules. Jerne also stated that id determinants can be present not only on antibody molecules of one specificity, but may be present on unrelated antibody molecules. Thus, antibody against antigen x might share some ids with antibody against antigen y. Lastly, although anti-id usually suppresses the corresponding id, it can be stimulating for the id as well. The anti-id would be expected to have a three dimensional structure similar or identical to the specific antigenic determinant. This type of anti-id is termed an internal image of antigen.

The characteristics of idiotypes of antibody molecules have been described (31-35). In many instances, idiotypes are located in or very near to the antigen binding site.








9

This has been demonstrated by hapten inhibition studies. Brient and Nisonoff (31) induced anti-p-azobenzoate antibodies in rabbits. These antibodies were purified and

injected into allotypically matched rabbits and the resultant antiserum bound to determinants present on some rabbit anti-p-azobenzoate antibodies. They then studied the effects that adding increasing concentration of hapten would have on the reaction between radiolabelled anti-azobenzoate antibodies and the anti-idiotypic antiserum. They found that the binding affinities of the benzoate derivatives correlated closely with their ability to inhibit the antibody/anti-id interaction. In many other studies (32-34), anti-id was induced in animals immunized with an anti-hapten antibody. This anti-id was purified froman the sera by initial adsorption to an affinity column having the

iTmunizing antibody bound to it and was then eluted with the appropriate hapten. This purification process then would select for anti-idiotypic antibodies which were directed to those idiotypic determinants very close to or within the antigen binding site and it would be expected that hapten could inhibit the id/anti-id interaction.

On the other hand, it is not always possible for hapten to inhibit id/anti-id. For example, Sher and Cohn (35) showed that there was variation in the ability of hapten to

inhibit id/anti-id interaction. Hapten was not able to









10
inhibit the interaction by 100 percent, maximum inhibition ws only 68 percent (35). The most extrene example in which hapten cannot inhibit id/anti-id interactions are in those studies in which cross reactive ids are present on antibody molecules of widely different specificity. For example, Eichmann et al. (36) showed that one half of the A5A id producing clones in A/J mice immunized with a streptococcal carbohydrate lacked the ability to bind this antigen. Obviously then, antigen would not be expected to inhibit this id-anti-id interaction. In other studies, Bona et al.

(37) showed that not all the id positive antibody following immunization with inulin could be removed with an inulin iTnmunoabsorbent. In these experiments, the anti-inulin antibody produced following antigenic stimulation bears a predominant id. However, some immunoglobulin following antigenic stimulation had this id but lacked specificity for inulin. These experiments therefore suggest that some mechanism exists naturally in which id positive clones of immunoglobulin producing cells are expanded following antigen stimulation but that not all the id positive ijnmunoglobulin is specific for the immunizing antigen. These experiments clearly show that although id/anti-id can usually be hapten inhibited, this property is not a requirement for an antibody to be anti-idiotypic.










Identical ids have been found irrespective of the

isotype of the antibody. The mechanism by which IgE and IgG antibody can have identical idiotypes relates to the gene rearrangement that occurs during differential expression of heavy chain genes (38). As a single clone of cells goes

through isotypic shift, a single variable region of the genes which includes the idiotype, will become linked to various heavy chain gene fragments. A single cell will differentiate into plasma cells which express different heavy chain genes but the same variable gene sequence (39). Therefore, it is possible for ids to be shared between antibodies of the same binding ability irrespective of the isotype. This implies that regulation of IgG antibody by anti-id networks may also result in IgE antibody regulation.

Idiotypic determinants are usually defined serologically. There are a number of different ways to produce anti-id (reviewed in 40-42). Anti-id can be produced across the species barrier, within the same species, within the same strain, or more importantly, even within the sair individual that produced the id. Anti-id have been used to determine if the id of the antibody molecule may have a function other thian to bind antigen. This has been done by examining what functional significance the presence of anti-id had on the corresponding id in vivo.









12
There are numerous reports that have shown that the

passive administration of anti-id or the active induction of anti-id results in the suppression of the corresponding id (reviewed in 40-46). This modulation acts directly on B-cells or indirectly through T-cells. For example, in a B-cell tumor model, Balb/c mice immunized with MOPC 315 myeloma protein produced antibody with specificity for the id of MOPC 315. Subsequently these mice were injected with a PMDCP 315 bearing plasmacytoma and the tumor growth was inhibited. It has also been shown that the immunization of N4DPC 315 protein also induces idiotype specific T-suppressor cells that inhibit the MOPC 315 tumors secretion in vivo

(47). Cosenza and Kohler (48) demonstrated that anti-id can act as an anti-antigen receptor antibody and specificially inhibit the induction of a primary immune response. In other studies by this same group, anti-id, which was specific for anti-phosyphorylcholine (PC) antibody, significantly inhibited anti-PC plaque forming cells to a degree similar to the inhibition seen with antigen (49).

These studies show that experimentally, the administration of anti-id or immunization with id to induce anti-id can result in id suppression. However, if anti-id regulates id during a normal immune response, auto-anti-id

should be part of the response.









13

A number of studies have shown the presence of autoanti-id during a normal immune response to an antigen (50-57). Bankert and Pressman (50) showed that an antibody with auto-anti-id activity could be detected in rabbits during primary and secondary immune response to both sheep red blood cells and to the hapten, 3-iodo-4-hydroxy-5nitrophenyl-acetic acid. Kelsoe and Cerny (51) have demonstrated a reciprocal expansion of antigen activated idiotype bearing clones of lymphocytes followed by expansion of

clones which bear anti-id receptors in Balb/c mice immunized to Streptococcus pneumonia. They hypothesized that the out of phase expansion of the reciprocal cell sets was the result of interactions of id and anti-id. The production of auto-anti-id in man has been demonstrated to occur during the immune response against tetanus toxoid. The presence of this anti-id was associated with the loss of some of the anti-tetanus toxoid idiotypes (52). Naturally occurring anti-id has also been demonstrated in myasthenia gravis patients using, as the idiotype probe, a mouse monoclonal antibody. Those patients with the highest titer of antireceptor antibody had the lowest level of anti-id, while in patients with the lowest titer of anti-receptor antibody

(id), the highest titer of anti-id was detected (53). Comparable findings have been reported in patients with anti-DNA antibody and reciprocal anti-id in systemic lupus









14

erythematosus (54) and in some IgA-deficient people in terms of anti-casein antibody and its reciprocal anti-id (55).

In these later experiments the anti-id was detected using homologous antibody as the id probe.

These experiments suggest that because anti-id is

present during a normal immune response and regulates the expression of ids, anti-id may be an important part of the regulation of the immune response.

In reference to IgE, Geczy and his associates (58) have shown that in guinea pigs, the administration of syngeneically derived antibody led to a marked suppression in the IgE level as measured by passive cutaneous anaphylaxis. This treatment also resulted in the production of anti-id and if this anti-id was given to a guinea pig followed by antigen stimulation, there was a marked suppression in the subsequent response. This group has shown that in the mouse, the preexisting anti-hapten IgE and IgG antibody response could be suppressed with either anti-hapten or anti-carrier anti-idiotypic antibody (59-61).

These experiments and others like them show that

id/anti-id interaction results usually in suppression of the

immune response. However, this is not always the case. For example, Eichmann and Rajewsky (62) showed that the injection of guinea pig IgG1 anti-id would enhance the expression of id designated A5A when stimulated with








15

Streptococcus whereas if the anti-id was an IgG2, the expression of A5A id was suppressed. Recently, Forni et al.

(63) showed that the injection of anti-SRBC IgM into normal mice induced plaque forming cells of the same specificity as the injected antibody. Further analysis established that the mechanism of this enhanced responsiveness %es based on id/anti-id interactions (63,64). The authors state that "these results support network concepts. Thus if an antigen specific response can be induced solely by using components of the immune system itself, it follows that, in its basic economy, this system is autonomous and does not depend on the introduction of antigen to adjust to new dynamic states" (63 p. 1127). In this case anti-id most probably acted as an internal image of antigen. There have been other examples that demonstrated the mimicry of antigen by antibody. For example, Sege and Peterson (65) showed that anti-id prepared against antibody to insulin could mimic the action of insulin in cells. Schreiber et al. (66) showed that anti-id against rabbit antibodies to alprenolol would conpete with alprenolol for the binding site on turkey red blood cells. This anti-id could also stimulate adenylate cyclase activity in the cells.

This discussion raises the possibility that the administration of autologous antibody might regulate antigen specific IgE response in the dog through id/anti-id









16

networks. Therefore the objectives of the work presented here were

1) To develop a consistent IgE antibody response in the dog and to study the kinetics of this response.

2) To examine the effects that autologous antibody administration had on an ongoing IgE response.

3) To determine if an anti-id response occured at any point during the experiment and if so, to examine the relationship between ids and anti-ids.











CHAPTER TWO
THE INDUCTION AND KINETICS
OF AN ANTI-DNP IGE RESPONSE


Introduction



The value of the dog as an experimental model to

study atopy has been described. However, the expense and difficulty of obtaining atopic dogs necessitated the development of a system in which antigen-specific IgE could be consistently induced. The use of a hapten-coupled carrier as an antigen was felt to be more convenient than a more complex, heterogenous substance such as an allergen to study the synthesis and regulation of IgE antibody. Furthermore, Halliwell (7) and Schwartzman et al. (67) have shown that two dogs imnunized with dinitrophenol coupled to ascaris antigen and administered in aluminum hydroxide as the adjuvant, developed anti-DNP IgE antibody. However, it is not known a) if all dogs so inmunized produce IgE antibody, b) how long the detectable IgE response remains, and c) what the immune response in terms of other isotypes might be. The purpose of the following experiments, then, was to induce a consistent anti-hapten IgE antibody



17







18
response and to examine the kinetics of the IgE, IgG and IgM anti-hapten antibody response.


Materials and Methods


Protein Concentration Determination

The concentration of immunoglobulin was determined

from known mlar extinction coefficients and by its ability to absorb light at 280 nm. Alternatively, the protein concentration was determined at 595 nm using Bradford's reagent (68) and interpolated from a standard curve derived

from the absorption values of a series of dilutions of a similar freeze dried purified protein of known concentration. The nmasurements with both techniques gave concordant results.



Antigens

Azobenzenarsonate coupled to keyhole limpet hemocyanin (ABA-KLH) was a gift from Dr. Mark Greene, Harvard University. Ascaris antigen was prepared from adult Toxocara canis by the method of Strejan and Campbell

(69) and modified as follows: Fifty adult T. canis

were obtained from the gastrointestinal tract of euthanized dogs. The worms were washed with phosphate buffered saline (PBS), pH 7.2, containing 0.02 percent sodium azide, ground with a mortar and pestle and incubated for 48 hours at 40 C. Large particulate matter was removed by centrifugation




19


at 1000 x g for ten minutes in an IEC centra-7R centrifuge (International Equipment Co.) The supernatant was then centrifuged at 49000 x g for one hour in an L8-70 ultracentrifuge (Beckman Instrument Co., Norcross, Ga.) to remove fine particles and was then chromatographed through a Sephadex G-100 column (Pharmacia Fine Chemicals, Piscataway, N.J.). The first peak was pooled, concentrated by negative pressure dialysis, dialyzed against PBS, pH

7.2, passed through a filter having 0.2 micron pores filter (Acrodisc, Gelman Co., Ann Arbor, Mi.) and used as the

ascaris antigen (ASC). Human serum albumin (HSA) fraction V was obtained from Sigma Chemical Co. (St. Louis, Mo.). Bovine gamma globulin (BGG) was prepared from serum of an adult cow by precipitation with 40 percent saturated ammonium sulfate. The precipitate was dialyzed against 0.035 M phosphate buffer, pH 8.0 and was then chromatographed through a diethylaminoethyl cellulose (DEAE) ion exchange column (DEA, DE52, Whatman Chemicals, Kent, England) equilibrated with this same buffer. The effluent protein was concentrated by negative pressure dialysis and dialyzed against PBS, pH 7.2.



Dinitrophenylation of Proteins

Dinitrophenylation of protein was performed by mixing

equal weights of protein, potassium carbonate (Fisher Scientific Co., St. Louis, Mo.) and






2D


2,4-dinitrobenzenesulphonic acid (DNP) (Eastman Kodak Co., Rochester, N.Y.) were mixed in distilled %nter (70). This %as then incubated while gently stirring for 18 hours at

roan temperature. The solution was chramatographed through a Sephadex G-25 column (Pharmacia Fine Chemicals, Piscataway, N.J.) to separate bound from free DNP. The dinitrophenylated protein was concentrated by negative pressure dialysis and extensively dialyzed against PBS, pH 7.2. The extent of substitution was estimated by measuring light adsorption at 360 nm and assuming a molar extinction coefficient of 1.75 x 104 for the dinitrophenyl group. The average epitope density expressed as molecules of DNP per molecule carrier was DNP27-HSA, DNP14.8-BGG. Since ASC extract was a complex mixture of proteins, the extent of substitution was expressed as moles DNP/mg ASC and was

6.32 x 10-5 DNP/ASC. A single batch of each of these antigens was prepared and used throughout the experiment. These antigens, when not in use, were stored at -700C. The degree of substitution did not change due to storage.



Aluminum hydroxide Precipitation of Protein

Aluminum hydroxide precipitation of protein was performed by mixing one part of a 5 percent sterile solution of aluminum potassium sulfate (AK (S04)2), Mallincrodt, Paris, Kentucky) with five parts of 1 mg/ml









solution of protein (70). The pH was then adjusted with 0.1 N NaOH to pH 6.3 to ensure adequate precipitate.



Affinity Chromatography

Sepharose 4B beads (Pharmacia Fine Chemicals) were

activated using cyanogen bromide (CnBr) by adding 1.5 grams CnBr in 20 ml distilled water to 10 ml of washed Sepharose 4B beads and adjusted to pH 11 with I N NaOH. This mixture was maintained on ice at pH 11 for 6 minutes after which the beads were washed with 100 volumes of iced cold water. Ninety milligrams of protein in 6 ml PBS, pH 7.2 were added and incubated for 12 hours at 40C. Alternatively preactivated Sepharose 4B beads were obtained (Pharmacia Fine Chemicals) and protein was bound to these beads as described by the manufacturer. To remove unbound protein in both cases, the beads were washed with five alternate cycles of 0.1 M Tris buffer, pH 8.3 containing 0.5 M NaCl followed by 0.1 M glycine HC, pH 2.8. Any renaining sites

were blocked by incubating the beads in 0.1 M Tris buffer, pH 8.3 for four hours at room temperature. The column was then flushed with normal canine serum and washed as described above.







22



Pepsin Digestion and Purification of F(ab)'2 Antibody
Fragments.

The usual procedure for F(ab)' digestion of iumuno1 2
globulin as to digest the antibody with 6 percent pepsin (w/v) in 0.2 M acetate buffer, pH 4.5 for 18 hours at 37 OC. However, this process resulted in some loss of antigen binding of the F(ab)'2 presumably fran the prolonged incubation time at pH 4.5. Where maintenance of this activity was critical, protein as digested with 20 percent pepsin w/v in 0.2 M acetate buffer pH 4.5 for five hours at 370C. The digested protein as separated from Fc pieces and intact antibody by passage through a cyanogen bromide-activated heavy chain specific immunoabsorbent column followed by passage through a Staphylococcus protein A affinity column (Pharmacia Fine Chemicals). The effluent as concentrated by negative pressure dialysis and dialyzed against PBS, pH 7.2.



Antisera

a) Preparation and purification of anti-IgG. Normal canine serum (NCS) was precipitated with a 40 percent saturated solution of ammonium sulfate. The precipitate was dialyzed against 0.035 M phosphate buffer, pH 8.0 and applied to a DEAE ion exchange column equilibrated with this same buffer. The effluent protein was concentrated by







23

negative pressure dialysis. One milligram of this material was emulsified in complete Freund's adjuvant (CFA) and administered intramuscularly to rabbits at two week intervals four times. Fifty milliliters of blood were obtained from the rabbit by ear vein venapuncture every two weeks starting after the second immunization. All serum which gave visible precipitation reactions by agar-gel diffusion against canine IgG was pooled. This antisera

was passed through a cyanogen bromide-activated sepharose 4B F(ab)'2 affinity column, to remove light chain activity, followed by adsorption to and elution with alycine HC (Osl M), pH 2.8 from a canine IgG bound affinity column. This anti-IgG detected three subclasses of canine IgG (IgGl, IgG2ab, IgG2c) but no other

protein as measured in an hmunoelectrophoresis (70) of NCS (figure 1). To determine if this antiserum detected IgE, the antiserum was radiolabelled and used in a radioimmunoassay. The serum sample tested contained both antiDNP IgG and anti-DNP IgE. Therefore, an aliquot of this serum was heat inactivated and the level of anti-DNP IgG was compared in this aliquot to a second aliquot of this serum that was not heat inactivated. Additionally, anticanine IgE was added to an aliquot of this sample to determine if this unlabelled anti-IgE might compete with the anti-IgG for Fc binding sites. Heating serum for four hours at 56 0C destroys the heavy chain antigenic









determinants of canine IgE (7). The level of anti-DNP antibody increased both when the serum was inactivated and when non-labelled anti-IgE antiserum was added to the sample. This indicates that this anti-IgG antiserum has minimal, if any, anti-IgE activity.



b) Preparation and purification of anti-IgE. A 40 percent saturated armnium sulfate precipitate of serum obtained, from a dog that was heavily parasitized and presumed to have high levels of IgE, as dialyzed against

0.035 M phosphate buffer pH 8.0 and applied to a DEAE cellulose column equilibrated with this same buffer. The effluent protein was concentrated by negative pressure dialysis and applied to a set of three in series Sephacryl S-200 columns (Pharmacia Fine Chemicals, Piscataway, N.J.).

The first one-third of the second protein peak, which was the IgE-rich fraction as determined by agar-gel immunoprecipitation, was collected, concentrated by pressure dialysis and reapplied to these columns. The resulting IgE-rich fraction was collected and used to immunize rabbits as described previously. The rabbits were bled as described above. Serum that produced visible precipitation lines against the immunizing antigen in an agar gel immunodiffusion were pooled. The resulting antiserum detected both IgE and IgG by immunoelectrophoresis. It was rendered specific for the former protein by passage through an















Figure 1.

The specificity of anti-canine IgG as assayed in an immunoelectorphoresis against normal canine serum.





Figure 2.

The specificity of anti-canine IgE as assayed in an immunoelectrophoresis against normal canine serum
(bottom well) and this same serum after heat inactivation (top well).





Figure 3.

The specificity of anti-canine IgM as assayed in an
Lmmunoelectrophoresis against normal canine serum. The anti-canine IgM in the bottom through is before adsorption with the supernatant of a 50 percent saturated ammonium sulfate precipitate of normal canine serum. The top trough has the anti-canine IgM antiserum after this treatment.






26







27
affinity column made with the heat inactivated irnrmunogen which remved all antibody except anti-IgE antibody. Purified antibody was then prepared by adsorption to and elution from an IgE-rich affinity column. This purified antiserum detected a single heat-labile protein by immunoelectrophoresis (figure 2), produced reverse cutaneous anaphylaxis in dogs at a high dilution of serum (10-6) and was unable to detect canine anti-DNP IgG in a RIA indicating that it had no specificity for this antibody.



c) Preparation and purification of anti-IgM. Canine IgM myelama serum, which contained approximately 58 mg/ml IgM was chromatographed on Sephacryl S-200 and the void volume was collected to obtain IgM. Two milligrams of this material was emulsified in CFA and injected intramuscularly at four sites into sheep. This was repeated at two week intervals five times. Five hundred milliliters of blood were collected by jugular vein venapuncture every two weeks. Sera that produced precipitation lines against the immunizing antigen, in an agar-gel diffusion against the immunizing antigen, were pooled. Light chain activity was removed from the antiserum by passage through a canine IgG affinity column. Antibody was then purified by adsorption to and elution from an IgM affinity column. The eluted proteins produced two bands on irmunoelectrophoresis of

NCS, one of which was IgM and the other an unknown protein.






28

This second activity was removed by adsorption with the supernatant of a 50 percent saturated ammonium sulfate precipitation of NCS (figure 3). This antiserum was assayed for anti-DNP IgE and IgG activity by RIA. Serum that was used contained both of these antibody isotypes. No antibody was detected indicating the antiserum did not have activity for IgE or IgG.



Isotope Labelling of Protein

Twu methods were used to label proteins with radioactive iodine. In the first method, between 1 and 2 mg of protein in 0.1 ml PBS pH 7.0 without azide and .5 mCi 1251 (Arrersham, Chicago, If.) was incubated on ice with 15 pl (0.1 M) and 30 V1 chloramine T (10 i M) for 15 minutes. After this incubation, 25 pi sodium metabisulphate (10 m 4) and 50 p1 KI(100 rM) were added to stop the reaction. To separate bound and free iodine, the material was chromatographed through a G-25 Sephadex column. The first peak containing radiolabel was pooled, concentrated and dialysed against PBS, pH 7.2. Alternatively, one IodobeadR (Pierce Chemical Co., Rockford, Ii.) was added to 100 pg of protein in PBS, pH 7.0 and 0.5 mCi 1251. After a fifteen minute incubation, the bound and unbound 1251 was separated as described earlier. The

specific activity of the radiolabelled antibody was usually about 300 pCi/mg protein (range 212-496).










Radioimmunoassay for the Detection of DNP Specific
Antibody

Microtiter wells (Imulon Remov-a-well StripsR,

Dynateck, Richmond, Va.) were coated with 50 pl of 20 pg/ml dinitrophenylated bovine gamma globulin (DNP-BGG) in Tris buffer, pH 8 (0.1 M Tris, 0.15 M NaCI). After incubating for 12 hours at 40C the wells were then washed three times with this buffer. Any remaining sites were blocked with 2.0 percent HSA in PBS containing 0.5 percent Tween 20 for three hours at room temperature. Phosphate buffered saline, pH 7.2 containing 0.5 percent Tween 20 and

2.0 percent HSA is referred to as RAST+ and this same buffer without HSA is called RAST-. Serum samples were diluted in PBS, pH 7.2 added to the appropriate wells, incubated for three hours at 4 0C followed by five washes with RAST-. Approximately 50,000 counts per minute (cpm) of radiolabelled antiserum in RAST- was added to the well, incubated for three hours at 4 0C and washed three times with RAST-. The radioactivity associated with each well was determined in a Searle-Packard gamma counter (Chicago, Ii.). Each sample was assayed in triplicate and each sample was counted for one minute. The maximum number of cpm bound was about 20% of the amount added. The background activity was determined by including in each assay the following controls: 1) A set of triplicate wells in which BGG rather than DNP-BGG was used as the antigen, 2) A







30

triplicate set of wells in which PBS rather than serum was added. The mean cpm from these controls were subtracted from the cpm of the test sample. Although the values varied from experiment to experiment, the maximum cpm of these controls were consistently lower than the lowest values obtained for test samples.

A standard serum sample %es included with each assay as an internal reference. An arbitrary antibody

concentration was determined by assigning a value of 64 units to the undiluted standard IgG and IgE sample and 32 units to the undilute IgM standard. By interpolating from the linear portion of the standard curve, the relative units of antibody for test samples were calculated.



Animals and Immunization Schedule

Outbred pregnant female dogs were obtained from the

Division of Animal Resources, University of Florida. Serum froi these dogs was screened by RIA to ensure that they did not have anti-DNP antibody at the time of whelping. The puppies of these bitches were used as experimental animals. Serum samples were obtained on the day of birth and weekly therafter. Each puppy received 100 pg aluminum hydroxide precipitated dinitrophenol coupled ascaris antigen by the intraperitoneal route on the day of birth and at two week intervals on three further occasions. Each







31
dog received a distemper-hepatitis modified live virus vaccination at week four and eight.




Results

Standard Curve

The relative antigen-specific antibody concentration was determined by interpolation from the linear portion of the standard curve included with each assay. This serum sample contained high levels of the isotype under investigation. An example of a standard curve for anti-DNP IgE, IgG and IgM is given in figures 4, 5 and 6.



Antibody Response

Twenty-eight dogs immunized with 100 pg DNP-ASC in

aluminum hydroxide developed an IgE, IgG and IgM serum antibody response. The mean relative antibody concentration for the three isotypes is depicted in figure 7, 8 and 9. The IgM response usually was highest in samples taken seven days after the first injection of antigen. However, as seen in table 1, eight of the dogs (5,7,8,12,14,16,23,25)

had anti-DNP IgM concentrations that were greatest in samples obtained at two weeks and three dogs (9,17,22) after three weeks. Four weeks after the first antigenic challenge, five dogs had no detectable IgM antibody and



























>

rO
(D



41
-4 LL-4
0

0
-4
4-)
-r-4
>


U) rc


rs


4J

.H
N4





33








0 c
0

0 co

0 f4T
0
0 C\j C)
0 0 L.QL

LO


0 0 0 0
ro K) N N
001 x wdo

































ro
(1) >1









L(w
4
(1) c 4J



rs


4.J




rs Q) It :5 41 V) 4-)



44



0
-4

r--l





35





0 0
It

0 (Dc
00


ot0 00

0 0 L CL




00 w0 Cd
000! x wdo























Figure 6.

Dilutions of the standard anti-DNP IgM serum sample assayed by RIA. The bars represent the standard deviation of the man.




37



22 20 18 16
14
0 '12
- IO
8

6
4 2

5 10 20 40 80 160320
Reciprocal of Dilution






38

after six weeks the levels of 1g-M antibody fell to background despite maintenance of the immunizing protocol.

The anti-DNP IgE and IgG antibody responses followed similar kinetics to each other. There was an initial lag of two weeks before antibody of these classes was detected (figures 8 and 9). At the time of the second immunization (two weeks after the primary immunization), there was a sharp rise in the antibody levels which continued for one additional week. Thereafter, the antibody concentration was maintained at that level or started to gradually decline. As was the case in the IgM antibody response, sane dogs deviated from the general trend. Two dogs had detectable IgE antibody levels one week after primary immunization (Table 2) whereas three dogs failed to develop a detectable response until after the third wAeek and, in the case of one dog, IgE antibody was not detected until the fifth week from primary immunization. The IgE antibody response persisted through the seven week course of the experiment in all dogs. Anti-DL1P IgG antibody was detected in nine dogs one week after primary immunization (table 3) and by the fourth week, all dogs had an IgG antibody response. Detectable IgG persisted throughout the immunization schedule but there was a gradual decline in IgG antibody levels towards the end of the immunizing schedule (See figure 9, table 3).




























>4








to
ro
U)
ONO
'"i ca 0
r_ 4-J -A

tZ -4 0 0 rO 4 4-) .-1 4-4 4-) cz
W --q 4
>1 a) .0 4-j 4 r-4 A-- -4 4J 4-)

44 C
41
WO
7 4 4
IW 14 tn N 4J 44
x rO
0 w ro
4J 4J

4
4j U) r--l ro -,-q


0) 0) LO
> r8
41
u 0 0 V
d) a -C: -4 -,4 M
-W 4J 4-) H
4 aj 4J 4J Z
CD r)

2:





A. b





o
c I
ob 66000

mo
4 Ki -' -14 -H -H -H +1
0 -K)Mnoo Lo C\j Q
0- C\j m q- Lo W N



00


(D (n





C\j


0
C\j 0 co o It C\j

uoilw1unuoo Apoq!IuVGAIIIDI@8




























r@
4
0


-4
= C/)
4j 0 w
4 r--l 44


ro >
cc 4 Q)
Is Cl) 1: 4-) 0
.'q Q)
4-)
0 0 4
-4 4-4 W Ul rO 4-1 41
u c
4-) r -4 u
a a) 0 41
co w u m w
2 4
rq >1 41

z
-4 14 r@ -,, a M
Z4 tD).N 4-) 8 I



E- rO
4-1 4j CO



i> r8
4-J Z 4 fO
41
LW
0
4 4J 4
o
CD (13 41

4J
4 ---q
4.4 4-)






4 2



ow-% M
(D CD C\j (\j N?
CD d 06 0i
C 0,
00 000 K?(P cQ

1: 0) 0
C\j W5 ff) W5 rr) Lo
-H +1 44 44 4q -H
c 0 (D 0)
C\j (c) Lb (.0 ;4- (.0 2 0 C\j K) It LO (0 N.



co


(D cn






C\j



C\j 0 00 0 C\j -0

si! un ApOq!I'UV @AIIDI@8-.

































W r4 Z
4J Q
4 U4 In

T
U)
co
C.4
r8
C 4-) 0 tr
-4 (1) C 4j r- -4 m rl
0 0 4 a5
-,-I w 41 u
41 a 4-J
m >1 0 c c
4
-tj
4J
Q) u t 0
U 4 0)
4J -i

u
T ..., c m
rM4 CF) 4-)
H C 8
04 >1
z m





18
Z 4 ft
M 41 -4
U -,4 LI-4
c 0
m c ra
0
--4 4
41 a) ri
41

4J 4.4












Relative Antibody Units

m oo FD \)










CD
CD
;$Z
(n
0)



00
-lj (3) (11 JS w N) -0 CD
w W w -lj
N)d) r\) 0 0 Ff- FF FF F FF i+ FF 0) 7-1 (3) N (3) Cn
(j) Go N (51 o *y)

lz a iG 0-% 0-1% e-% to 0) -.4 0 0 0 N 0
0) OD (:Y) 0 --J 01 CD
0 0









45
Table 1
The Relative Anti-DNP IgM Concentration in 28 Dogs

Animal
Number 1 2 3 4
Weeks
0 0 0 0 0
1 16.6+.87 4.4+.39 6.5+.47 4.5+.06
2 10.6+1.23 3.2+.23 3.4+.24 4.4+.27
3 .3+0 1.4+.56 0 .6+.14
4 .2+.01 .2+0 0 .1-.03
5 0 0 0 0
6 0 0 0 0
7 0 0 0 0
Animal
Number 5 6 7 8
Weeks
0 0 0 0 0
1 2.9+.25 6.4+.82 6.8+1.01 3.1+.28
2 5.8+.90 5.6+.39 7.8+.32 5.2+.17
3 0 5.1+.07 3.4+.31 0
4 0 3.9+.24 0 0
5 0 0 0 0
6 0 0 0 0
7 0 0 0 0
Animal
Number 9 10 11 12
Weeks
0 0 0 0 0
1 0 10.9+1.03 7.0+.02 5.0+.25
2 2.0+.14 3.0+.41 2.8+.46 5.1+.37
3 3.37+.29 .8+.01 1.8+.19 2.07+.21
4 0 0 1.0+.07 1.8+.27
5 0 0 0 0
6 0 0 0 0
7 0 0 0 0
Animal
Number 13 14 15 16
Weeks
0 0 0 0 0
1 5.5+.16 1.8+.18 10.1+.79 1.3+.12
2 5.0+.23 5.2+.33 6.1+.43 2.6+.37
3 4.8+.07 2.3+.16 1.8+109 2.5+.14
4 .9+.13 .5+.22 1.6+.17 1.2+.20
5 0 0 0 0
6 0 0 0 0
7 0 0 0 0










46
Table 1 Continued

Animal
Number 17. 18 19 20
Weeks
0 0 0 0 0
1 1.1+.03 14.8+2.01 3.9+.39 2.7+.22
2 4.4+.17 5.6+.81 2.1+.09 2.1i+.40
3 6.0+.31 3.1+.23 .8+.02 0
4 1.7+.15 2.1+.10 0 0
5 .8+.0 0 0 0
6 0 0 0 0
7 0 0 0 0
Animal
Number 21 22 23 24
Weeks
0 0 0 0 0
1 8.7+.83 .7+.15 8.1+.44 5.1+.46
2 7.9+.09 4.5+.18 12.6+.99 .6+.06
3 3.2+.48 6.2+.46 2.7+.23 0
4 l.6+.06 2.i+.05 0 0
5 .1+0 0 0 0
6 0 0 0 0
7 0 0 0 0
Animal
Number 25 26 27 28
WeeKs
0 0 0 0 0
1 4.0+.09 13.8+.1.35 5.2+.29 10.6+.87
2 6.2+.21 9.1+1.37 .7+0 x
3 2.5+.30 3.4+.40 1.0+.16 4.2+.36
4 0 2.1+.11 0 .8 -.023
5 0 0 0 0
6 0 0 0 0
7 0 0 0 0

a) Each dog was immunized with DNP-ASC in adjuvant at weeks 0,2,4 and 6.
b) The units were calculated from a relative antibody
concentration scale derived from the titration of a serum sample
containing anti-DNP IgM. A value of zero indicates no detectable anti-DNP IgG. The data was the mean antibody concentration of a sample run in triplicate + the standard deviation from the mean. This was calculated by adding and subtracting the standard deviation to the mean cpm and calculating the relative antibody
concentration for these numbers. These numbers were then subtracted from the mean concentration.











47
Table 2
The Relative Anti-DNP IgE Concentration In 28 Dogs
Following lImunization with DNP-ASC a)

Animal
Mnumezr 1 2 3 4
Weeks0 0 b) 0 0 0
1 0 0 0 0
2 1.2+.31 10.8+1.90 3.2-i,.36 3.2+.21
3 1.5+.27 l.3+.27 10.0+.76 4.8+.07
4 1.7+.16 9.2+.36 8.5+.44 4.6+.29
5 1.3+.49 11.6+.24 10.1+.83 5.1+.36
6 1.0+.22 5.6+.70 3.3+.12 3.3+.16
7 3.0+.96 7.7+1.01 7.0+.41 4.9+.37
Animal
Number 5 6 7 8
Weeks
0 0 0 0 0
1 .6+.03 0 0 0
2 3.6+.46 1.9+.17 3.2+.10 .6+.07
3 11.9+.99 13.4+1.35 11.8+.77 5.0+1.00
4 7.8+1.04 8.C+.61 6.7+.83 3.9+.43
5 9.0+.63 12.17+.62 11.8+2.45 15.2+.18
6 11.0+.71 16.1+.90 5.5+.69 7.6+.45
7 14.1+1.42 12.8+.88 2.8+.25 6.6+.51
Animal
Number 9 10 11 12
Weeks
0 0 0 0 0
1 1.7+.39 0 0 0
2 5.6+.26 0 2.1+.24 7.5+.88
3 18.2+2.73 3.1+.69 8.3+.74 13.2+1.86
4 8.4+.97 5.5+.43 7.8+.79 9.1+1.02
5 10.6+.64 5.2+.47 3.9+.81 26.8+1.30
6 7.1+.31 3.d+.30 2.4+.65 7.0+1.51
7 8.7 +.51 12.6+.76 4.8+2.38 15.4+.36
AnimalNumber 13 14 15 16
Weeks
0 0 0 0 0
1 0 0 0 0
2 0 1.8+I.36 1.0+.11 0
3 0 9.5+.37 3.4+.48 .7+.12
4 0 8.4+.67 7.5+1.06 3.9T.61
5 .3+.l1 12.0+.87 5.1+.56 3.9+.14
6 .8+.23 8.5+.21 .6+.01 3.7+.37
7 1.6+.52 19.3+1.98 5.2T.50 4.5+.66










48
Table 2 Continued

Animal
Number 17 18 19 20
Weeks
0 0 0 0 0
1 0 0 0 0
2 .2+.07 10.1+.21 0 3.3+.09
3 11.0+.96 9.3+.49 1.4+.08 4.2+.15
4 7.7+.14 8.9+.47 .3+.04 4.8+.33
5 9.8+.34 9.4+.67 1.0+.20 5.2+.40
6 2.2 +.26 6.9+.52 1.5+.08 5.8;.26
7 6.7.31 15.3+.89 2.3+.21 7.1+.43
Animal
Number 21 22 23 24
Weeks
0 0 0 0 0
1 0 0 0 0
2 3.0+.10 1.5+.17 2.1+.13 .1+.04
3 4.5+.11 2.6+.12 1.7+.09 1.0+.06
4 4.0+.27 3.0+.31 l.8+.14 1.0+.17
5 5.6+.36 2.6+.21 1.0+.13 1.6+.31
6 2.3+.12 2.8+.32 0.8+.07 0.9+.07
7 4.8+.22 2.6+.60 0.9+.21 1.3+.29
Animal Number 25 26 27 28
Weeks
0 0 0 0 0
1 0 0 0 0
2 1.0+.29 1.7+.37 .6+.05 x
3 2.7+.31 2.1.05 3.2+T.22 1.1+.06
4 2.0+.26 3.0+.42 2.0+.23 7.3+.26
5 I.8+.19 1.3+.12 1.0+.15 8.8+.51
6 .7+.09 4.3+.27 1.5+.13 3.3+.16
7 1.2+.10 1.7+.28 0.87+.06 8.4+.42

a) Each dog was immunized with DNP-ASC in adjuvant at weeks 0,2,4 and 6. b) The units were calculated from relative antibody concentration scale derived from the titration of a serum sample containing anti-DNP IgE. A value of zero indicates no detectable anti-DNP IgE. The data was the mean antibody concentration of a sample run in triplicate + the standard deviation from the mean. This was calculated by adding and subtracting the standard deviation to the mean and calculating the relative antibody concentration for this number. The relative
concentration for this number was subtracted from the mean concentration.







49

As would be expected for outbred animals, there As considerable variation in the Timune response between dogs. If however, the IgG antibody concentration of dogs within single litters are examined, a more homo~geneous trend is observed (table 4). There vas, however, considerable animal-to-animal variation within a litter in the level of antigen specific IgE and 1gM (tables 5 and 6). If these two litters a-re compared statistically, at each time point, using a student T test, there is a significant difference

in the mean antibody level between the two groups in the IgG antibody after the first week (P is less than 0.001 in all instances).

When all the animals are considered, it appears that some are generally high responders to the antigenic stimulation whereas the response in other dogs is low. The high response or low response is seen for both IgG and IgE antibodiy classes in a single animal. For example, dogs 2 and 6 have a strong IgE and IgG response whereas dogs 15 and 24 have very weak responses. Although this trend predominates, this association of high responses or low responses is not always consistent, and a regression analysis comparing the level of anti-DNP IgG to the anti-DNP IgE failed to show a statistically significant correlation (p greater than 0.05).










50


Table 3
The Relative anti-DNP IgG Concentration in 28 Dogs
Following Immiunization with DNP-ASC a)

Animal
Number 1 2 3 4
Weeks
1 0 b) 0 0 0
2 0 0 .8+.02 0
3 4.1+.21 9.1+1.4 2.4+.21 3.7+.11
4 3.7T.20 15.5+.20 8.0+.10 7.4+.66
5 6.6;.64 16.4+1.26 4.8+.26 7.4+.14
6 8.il+.33 26.0+1.33 8.6+.23 4.8+.26
7 7.9+ 1.05 23.3+.14 6.7T.16 6.l+.88
Animal
Number 5 6 7 8
weeks
1 0 0 0 0
2 0.3+.02 0 0 0.1+0
3 3.f+. 12 9.1+1.21 12.2+.77 25.4+ 1.31
4 8.6+.17 15.3+.86 ll.i+.39 27.0+1.92
5 9.1+.34 17.0+1.41 9.2+.64 30.0+3.21
6 7.4+.23 25.6+.73 9.3+1.07 20.1+.49
7 8.J+.61 24.1+1.72 5.9+;.26 8.7+1.26
Animal
Nkmiber 9 10 .11 12
Weeks
1 0 0 0 0
2 0 0 0 0
3 7.3+.69 4.3+.05 8.3+.80 6.3+.29
4 15.4+1.71 4.3+.12 12.4+.64 22.5+1.10
5 18.5+.86 5.6+.09 7.9+.32 20.8+.93
6 16.0+.46 7.l+.51 7.5+.93 17.8+1.79
7 8.3+.62 6.87+.19 5.5+.68 10.4T.23
Animal
Number 13 14 15 16
Weeks
1 0 0 0 0
2 0 0 0 .3+.02
3 0 5.3+.31 3.8+.43 6.0+.ll
4 10.1+.61 16.7+1.1 8.8+.16 4.47+.70
5 9.3+.75 14.'3+.42 9.9+.63 11.6+1.60
6 7.8+.26 14.7+.32 8.9+.91 11.1+1.13
7 5.0+.16 l0.4+.6 9.7+.46 17.2+.32










51
Table 3 Continued

Animal
Number 17 18 19 20
Weeks
1 0 0 0 0
2 2.6+.13 .9+.04 .3+0 0
3 1.7+.14 1.7+.13 4.8+. 17 11.7+.l1
4 4.8+.36 22.7+.81 16.9+.65 12.8+.96
5 6.2+.58 24.9+1.65 21.3+.17 8.9+.04
6 6.9+.31 21.8 +-.77 20.8+1.16 7.6.47
7 5.3+.40 28.2+1.24 21.9+.84 9.8+.36
Animal
Number 21 22 23 24
Weeks
1 0 0 0 0
2 0 7.1+.33 12.9+.92 3.6+.24
3 5.6+.46 7.1+.33 12.9+.92 3.6+.24
4 5.4+.27 10.7+.75 18.0+.75 8.8+.43
5 8.5+.12 13.8+.99 13.9+.51 9.7+.81
6 10.6+.93 19.1+1.53 13.5+.87 9.0+.29
7 7.6+.46 18.2+.75 12.7+.60 6.6+.64
Animal
Number 25 26 27 28
Weeks
1 0 0 0 0
2 0 0 .8+.07 .4+0
3 10.9*.47 3.3+.33 9.6+.48 x
4 10.0+1.79 16.3+.84 10.3+.68 12.9+.85
5 8.2+.97 15.4+.93 12.1+.06 22.8 +1.68
6 7.8+.11 14.7+.43 18.7+1.78 26.9+1.41
7 5.0+.36 10.5+.70 19.6+.96 28.7+1.92

a) Each dog received immunization with DNP-ASC in adjuvant at weeks 0,2,4 and 6.
b) The units were calculated from a relative antibody
concentration scale derived from the titration of a serum sample containing anti-DNP IgG. A value of zero indicates no detectable anti-DNP IgG. The data was the mean antibody concentration of a sample run in triplicate + the standard deviation from the mean. This was calculated by adding and subtracting the standard deviation to the mean cpm and calculating the relative antibody concentration for these numbers. This number was then subtracted from the mean concentration.











52

Table 4
The Relative Anti-DNP IgG Concentration In Two Litters of Dogs
Following Imunization with DNP-ASC a)


Litter 1
3 4 17 20 24 25 M +S.D b)
Weeks
0 0 0 0 0 0 0 0
1 .8 0 2.6 0 0 0 .6 + 1
2 2.4 3.7 1.7 11.7 3.6 10.9 5.7 + 4.3
3 8.0 7.4 4.8 12.8 8.8 10.0 8.6 + 2.7
4 4.8 7.4 6.2 8.9 9.7 8.2 7.5 + 1.8
5 8.6 4.8 6.9 7.6 9.0 7.8 7.5 ; 1.5
6 6.7 6.1 5.3 9.8 6.6 5.0 6.6 + 1.7
7 5.4 6.0 3.1 11.9 7.2 6.4 6.7 + 2.9


Litter 2
2 6 14 19 23 28 M + S.D.
Weeks
0 0 0 0 0 0 0 0
1 0 0 0 .3 0 .1 0
2 9.1 9.1 5.3 4.8 12.9 x 8.2 + 3.3
3 15.5 15.3 16.7 16.9 18.0 12.9 15.5+ 1.8
4 16.4 17.0 14.3 21.3 13.9 22.8 17.6+ 3.7
5 26.0 25.6 14.7 20.8 13.5 26.9 21.3+ 5.9
6 23.3 24.1 10.4 21.9 12.7 28.7 20.2+ 7.1
7 26.8 25.8 18.9 24.8 23.0 24.7 24.0+ 2.8


a) The relative antibody concentration ues determined by extrapolation of a standard serum sample. A value of zero indicates no detectable antibody activity. Each dog received DNP/ASC in adjuvant at weeks 0,2,4 and 6. b) Mean + standard deviation











53


Table 5
The Relative Anti-DNP IgE Concentration In 2 Litters of Dogs
Following flrnunization with DNP-ASC a)

Litter 1
3 4 17 20 24 25 M+ S.D. b)
Weeks
0 0 0 0 0 0 0 0
1 0 0 0 0 0 0 0
2 3.7 3.2 .2 3.3 .1 1.0 1.9 + 1.7
3 10.0 4.8 11.0 4.2 1.0 2.7 5.6 + 4.0
4 8.5 4.6 7.7 4.8 1.0 2.0 4.8 + 3.0
5 10.1 5.1 9.8 5.2 1.6 1.8 5.6 + 3.7
6 3.3 3.3 2.2 5.8 .9 .7 2.7 + 1.9
7 7.0 4.9 6.7 7.1 1.3 1.2 4.7 + 2.8


Litter 2
2 6 14 19 23 28 M +S.D.
Weeks
0 0 0 0 0 0 0 0
1 0 0 0 0 0 0 0
2 10.8 1.9 1.8 0 2.1 x 3.3 + 4.3
3 11.3 13.4 9.5 1.4 1.7 1.1 6.5 + 5.6
4 9.2 8.0 8.4 .3 1.8 7.3 5.8 + 3.8
5 11.6 12.7 -12.0 1.0 1.0 8.8 7.9 + 5.6
6 5.6 16.1 8.5 1.5 .8 3.3 6.0 + 5.7
7 7.7 17.8 19.3 2.3 .9 8.4 9.4 + 7.7

a) The relative antibody concentration was determined by extrapolation of a standard serum sample. A value of zero indicates no detectable antibody activity. Each dog received DNP-ASC in adjuvant at wek,- 0,2,4 and 6. b) Mean + Standard Deviation











54


Table 6
The Relative Anti-DNP 1gM Concentration in 2 Litters of Dogs
Following Immniization with DNP-ASC a)

Litter 1
3 4 17 20 24 25 M +S.D. b)
Weeks
0 0 0 0 0 0 0 0
1 6.5 4.5 1.1 2.7 5.1 4.0 4.0 + 1.9
2 3.4 4.4 4.4 2.1 .6 6.2 3.5 + 2.0
3 0 .6 6.0 0 0 2.5 1.5 + 2.4
4 0 .1 1.7 0 0 0 .3+.
5 0 0 .8 0 0 0 .1T.
6 0 0 0 0 0 0 0
7 0 0 0 0 0 0 0

Litter 2
2 6 14 19 23 28 M +S.D.
Weeks
0 0 0 0 0 0 0 0
1 4.4 6.4 1.8 3.9 8.1 10.6 5.9 + 3.2
2 3.2 5.6 5.2 2.1 12.6 x 5.7 + 4.1
3 1.4 5.1 2.3 .8 2.7 4.2 2.8 + 1.6
4 .2 3.9 .5 0 0 .8 .9 + 1.5
5 0 0 0 0 0 .1 .01 +.04
6 0 0 0 0 0 0 0
7 0 0 0 0 0 0 0

a) The relative antibody concentration wAas determined by
extrapolation of a standard serum sample. A value of zero indicates no detectable antibody activity. Each dog received immunization with DNP-ASC in adjuvant at weeks 0,2,4 and 6. b) Mean + Standard Deviation







55



Discussion


The purpose of the experiments in this chapter was to induce an anti-DNP antibody response which included IgE and to examine the kinetics of this response. As described in the results section of this chapter, the anti-DNP IgE, IgG and IgM antibody response followed expected kinetics (71-73). The 1gM response ues present before IgE or IgG antibody was detected and disappeared after the sixth week in spite of continued antigenic challenge. The IgE and IgG production had a two week lag period in general, but once they developed, they were maintained throughout the immnunizat ion course.

If inbred laboratory animals such as mice are immunized with an antigen, a homogeneous response ordinarily results (74). However, in species that are genetically heterogeneous, such as nan and dogs, the immune response to the antigen would be expected to be more highly variable (74). In these outbred dogs there was -marked variation in the kinetics and magnitude of the antibody responses. The marked variations seen in these dogs are most probably the result of the genetic differences between them. In this context, it Twas noteworthy that the IgG response was more homogeneous within the same litter than between litters.







56

The genetic makeup of the animal also influences the class of antibody produced following antigen challange. In certain inbred animal strains, it is very difficult to mount an IgE antibody response without some type of rranioulative process to eliminate T-suppressor cells (75,76). Furthermore, if a comparison is -made between allergic and non-allergic people, a marked difference in antigenspecific IgE responsiveness to certain antigens is seen. Those individuals with allergic tendencies will have an enhanced IgE response to allergens whiereas non-allergic people may not develop IgE antibody (77). There were notable differences between individual dogs in terms of their IgE response and in contrast to the IgG response there was no consistent pattern within litters. It is not clear if this failure to see similar patterns within a litter in the IgE level reflects the antigen chosen to study IgE in these dogs or if there are multiple genes that govern IgE levels in dogs. By having such a small sample size, a consistent pattern might not be observed for IgE levels. one of the dogs failed initially to develop an IgE titer. The IgE antibody response started after this dog -s vaccinated with a modified-live canine distemper/hepatitis vaccine at four weeks of age. The immunization of dogs with this vaccine has been shown to enhance antigenspecific IgE response to an antigen administered at the same time (78). It has been hypothesized that this effect







57

is the result of a suppressi on of T-suppressor cells.

Because the mechanism that would normally suppress IgE synthesis is altered, IgE antibody response will develop. This alteration in the suppressive network and subsequent IgE antibody synthesis has been called the "allergic breakthrough" (79).







Twenty-eight dogs immunized to DNP-ASC at birth and

then three times at two week intervals produced serum antiDNP antibody. The IgM response was detected one week after primary immunization and lasted for up to five weeks. The IgE and IgG antibody response in general es not present until week three but persisted through the immunization schedule. Although variation in the level and duration of the antibody response was detected between individual dogs, each dog did have a response that included all three isotypes examined.



Conclusions



(1) Dogs immunized with DN'P-ASC develop a high

level, long term IgE and IgG antibody response but the IgM response followed a different kinetic pattern in that it did not persist after week five.







58

(2) There was a difference in the responsiveness to this antigen seen between individual dogs for all antibody isotypes. This %as imst probably a reflection of the genetic heterogeneity between these dogs.











CHAPTR THREE
ATTEMPTS TO REGULATE AN ANTIBODY RESPONSE WITH AUTOfJOGOUS ANTIBODY

Introduction



The mechanisms by which antibody responses are regulated have been studied extensively. Many experiments have shown that antibody can be self-regulating (80,81). There are at least two different ways that this can occur: 1) If antibody is present at the time of immunization, antibody can bind to and remove antigen. Therefore, the result would be a decrease or a failure to mount the response. 2) Antibody can induce an anti-idiotypic immune response which would regulate the subsequent expression of the antibody through id/anti-id interactions (80-82).

If the synthesis of IgE antibody could be suppressed with antibody, such therapy mray be very beneficial in controlling IgE mediated allergic disease. As discussed in Chapter one, passively administered antibo'dy in mice and rabbits has been shown to suppress IgE antibody (26,27). The purpose of the experiments in this chapter is to determine if the administration of autologous antibody has any effect on the ongoing antibody response in dogs.


59






60


Materials and Methods



Affinity Chromatography

Anti-DNP antibody as produced by immunizing dogs to DNP-ASC and vas purified from serum by chromatography through a DNP-HSA affinity column as described in Chapter two. The bound antibody was eluted with 0.1 M glycine HCl, pH 2.5.



RIA

The RIA for detection of anti-DNP antibody was described in Chapter two.



Animals and ILnunization Schedule

The same dogs that were described in Chapter two were used in these experiments. These dogs had received 100 pg of aluminum hydroxide precipitated DNP-ASC by the intraperitoneal route on the day of birth at two week intervals on three further occasions. Fifteen milliliters of serum were obtained from each dog at the time of final antigenic challenge. Anti-DNP antibody was purified from this serum by DNP-HSA affinity chromatography, concentrated to about 3 mg/ml by negative pressure dialysis and rendered bacterially sterile by passing through a filter having 0.2 micron sized pores. Dogs received either 10 or 100 pg of their own antibody enulsified in 2 ml of either complete







61

(CFA) or incomplete Freund's adjuvant (IFA). Dogs designated as controls received 2 ml of either CFA or IFA. In all cases, the injections were given at four sites subcutaneously seven and nine weeks after the first immunization with DNP/ASC. Animals were given DNP/ASC booster injections eight and ten weeks after the primary immunization

(Fig.10).








0 1 2 3 4 5 6 7 8 9 10 11

0 0




a = Administration of antigen

o = Administration of adjuvant with or without

autologous antibody



Figure 10
Time Schedule for Immunizations






62



Results



DNP Affinity Column

There was no detectable anti-DNP antibody in the serum of any dog after passage through the DNP affinity column. On the other hand, the glycine HCI eluate contained high levels of anti-DNP IgG but no detectable

anti-DNP IgM or IgE. Because anti-DNP IgE was not detected in either the effluent or the eluent from the affinity column but was detectable in the serum prior to such treatment, an aliquot of serum containing anti-DNP IgE was dialyzed against glycine HC1, pH 2.5 followed by dialysis against PBS, pH 7.2 to determine what effects glycine HC1 had on canine IgE. There was no detectable anti-DNP IgE in this serum after such treatment as assayed

by RIA.



Immunization with Autologous Antibody

As noted in Chapter two, there was considerable

variation in the anti-DNP antibody respone between dogs. Tables 7 and 8 show the mean relative concentration of anti-DNP IgG and IgE respectively in each group of dogs prior to the autologous antibody administration and thereafter. These data are presented graphically in figures 11 and 12. The individual relative antibody concentrations











63

Table 7
The Mean Relative Anti-DNP IgG Concentration


Group 1 a) Group 2 Group 3
N=8 N=4 N=8
Weeks
0 0 0 0
1 0 0 0
2 .2 + .3 .5+ .9 6.6 + 1.7
3 8 .7 + 7.6 4.4 + 3.6 13.7 + 7.5
4 12.1 + 7.2 14.3 + 6.3 13.2 + 7.6
5 12.5 + 8.4 13.3 + 6.5 12.2 T 5.6
6 13.7 + 8.7 12.5 T 6.0 7.8 T 2.1
7 11.4 + 7.7 13.4 + 8.3 10.6 + 5.9
8 12.6 + 8.5 14.2 + 8.6 12.1 T 4.3
9 13.2 + 7.9 14.5 + 7.6 13.8 + 5.9
10 13.2 + 7.9 14.5 + 7.6 13.8 + 5.9
11 15.2 + 9.2 17.8 + 6.9 15.1 T 7.4


Group 4 Group 5 Control
N=2 N=6 N=8
Weeks
0 0 0 0
1 0 .2 + .3 .1 + .3
2 6.4 + 1.1 8.1 + 4.4 7.6 + 3.7
3 8.2 + 3.5 12.7 + 3.7 11.5 + 4.1
4 11.2 + 3.7 13.7 + 5.2 13.1 ; 4.8
5 14.9 + 6.0 15.1 + 7.0 15.1 + 6.4
6 12.9 + 7.5 13.4 + 8.9 13.6 + 8.1
7 20.9 + 6.4 16.4 + 7.8 17.5 + 7.3
8 20.1 + 4.9 17.3 + 8.5 17.8 + 7.5
9 22.7 + 8.5 18.9 + 7.5 19.8 + 7.3
10 23.5 + 9.1 19.4 + 8.0 20.5 + 8.2
11 23.8 + 10.7 17.7 + 6.4 19.1 + 7.0

a) All dogs received DNP/ASC immunization at 0,2,4,6,8 & 10 weeks. At 7 and 9 weeks: Group 1 received 10 pg autologous anti-DNP antibody in CFA, Group 2 received 100 pg autologous anti-DNP antibody in IFA, Group' 3 received 100 pg autologous anti-DNP antibody in CFA. Group 4 received IFA alone, Group 5 received CFA
only. Control values were the mean of groups 4 and 5.











64

Table 8
The Mean Relative Anti-IgE Antibody Concentration


Group 1 a) Group 2 Group 3

Weeks
0 0 0 0
1 1 + .2 0 .4 + .85
2 3.5 + 3.2 2.1+ 3.5 3.8 + 3.4
3 8.7 + 4.3 4.9 + 4.4 10.7 + 6.5
4 6.3+ 2.6 5.2 3.6 7.7 + 1.6
5 9.6 + 4.5 5.8 + 4.2 11.4 + 10.5
6 6.7 4.9 3.8 T 3.0 3.7+ 4.5
7 8.0 T 5.3 7.8 T 6.3 10.4 T 4.6
8 7.2 3.9 3.7+ 2.7 7.8 + 2.5
9 8.3 3.2 5.3 + 3.1 8.8 1.0
10 7.6 + 3.3 4.9 + 2.6 7.3 + 2.4
11 8.3 + 3.8 5.4 + 3.1 7.0 + 2.7


Group 4 Group 5 Control
Weeks
0 0 0 0
1 0 0 0
2 2.3 + 1.1 1.1 + .8 1.4 + 1.0
3 3.6 + 1.3 2.0 + .9 2.4 + 1.2
4 3.5 + .7 2.9 T 2.3 3.0 T 2.0
5 4.1+ 2.1 2.6 3.1 3.0 + 2.8
6 2.6 + .4 1.9 + 1.5 2.1 T 1.3
7 3.7 + 1.6 2.4 + 3.0 2.7 + 2.6
8 6.6 + 2.0 2.8 + 1.9 3.8 + 2.5
9 8.3 + 1.0 3.2 + 2.2 4.4 + 3.0
10 7.1+ .1 2.7 + 2.2 3.8 + 2.7
11 6.9 + 2.3 3.0 + 2.3 4.0 + 2.7


a) All dogs received DNP/ASC immunization at 0,2,4,6,8 and 10 weeks. At 7 and 9 weeks: Group 1 received 10 pg autologous anti-DNP antibody in CFA, Group 2 received 100 pg autologous anti-DNP antibody in IFA, Group 3 received 100 pg autologous anti-DNP antibody in CFA, Group 4 received IFA alone, Group 5 received CFA alone. Control values were the mean of groups 4 and
5.



























C:
(a r_ w C C) 0 a)
0 < 8.11 -r"
r.4 4J 4-3
4
(z CN 41
"A
r- ::" 8 -P w
rT 0 0, 4J

4-J
4 U
4-)
CD $.4
5 6 41
(n U) Q)
-4 84
4-4 0 r--4 >
0 0 >1 0 I-q
-4 W 0 4
4J 4-) 4-) -4
a .41
0 tv
41 U 4 (1)
r--q 0 --q 41 -4 0)
-4 ri) .. c 3 c
-W c < Q)
8
.0 1 a 8 0' .01
a --i ro -4 a
>1 4-) 0 >4 u
'-f 4-J ro
T 41
a
.,4 4j Q) 4 0 4-) r- U
) )4 (t a 4i (1) 8 z
0 1 04 >
-1-4 z .14
> ILJ
a)
41 --q rc 0) .4.1 (1) 0
C =L 4 4J rO
r. 4
0
$-4 4 81
ul U) 0
-Li
r8 004 r-4 (1) 4-) (n
0 r_ (Z
0 4-)
m






66










C\j n
E 03 Q 0. 0Q- .4.- :3 =3 :3
c 0 0 0
Z L- UOD


Cl)
LO




C\j


0
C\j 0 W w lq- C\j 0 co QO CNN\\Q
C\j C\J
uoijDj ueouoo Apoq! uV 9AIJOI;98





























C) C) 0 z
-q (= I r)
't 4.J .1 4
a -4 4j
CN

0
-,q o:: tr
W 1: 81
4-) W T
IT
2
CNJ 0
< -q ,c Q) cn r-.4
41
aj 44 -4 rTl r--i 41
.-f 04 H 4-4
0 0 >1 0 4 r-I
41 41 ro .0 (3)
m m (1)
5-4 4j 4J C 11) CN r- 0 Iq
T -1-4 r3
ri 41
8 0 0 4J W
4 --4
8 4J 404
C4 4-J 4J
FX4 ty, 8 4w x
r Q)
4-J
41
C IL4
8,0U-)
M a4 C 4-J a4 0 Z -14
-4 m -1 5

S (3)
-U 4-) tO
4-) -4
a)
a 0
4 4 t7) ri) .-j
Q 'ZS 4-3 to r i o m 7 4M)
CY) r-I rd) r-q M c
Q) 0 Q) w 8
r8 94 r 4 4 4-)
M 0 -4
0 4-) (1) -0 C4
4-3 ::5 4
F4






68









0
c CL
< o =3
E 0 0 0 c:
%- U- %- 0 z (71 cr 0


0


co



LO



C\j


C\i 0 co w C\j 0

UOIIDjlueouoo Apoq!IuV OAIIDI@8












69

Table 9
The Relative Anti-DNP IgE Concentration in 28 Dogs a)

Groul1 )j (10 pg Anti-DNP Antibody in CFA)

Animal
Number 1 2 3 4
weeks
8 2.0+.17 6.4+.29 6.1+.49 8.1+.36
9 2.6+.26 9.3+.84 6.5+.25 9.4:+.36
10 3.d+.18 8.1T.31 3.4+.30 7.1+.79
11 2.8+.31 9.3+1.26 4.1+.19 8.2+.81
Animl
Number 5 6 7 8
Weeks
8 5.2+.57 15.6+.70 6.6+.68 7.2+.30
9 8.9+.63 14.+.69 6.8+.78 9.2+1.18
10 9.3+.72 13.2+.36 8.2+.14 8.2T.07
11 10.2+.64 15.1+.47 7.4+.66 8.6+.42

Group 2. (100 pg Anti-DNP Antibody in IEA) Animal
Nubr9 10 11 12
Weeks
8 9.8+.47 10.0+.22 5.7+.67 5.5+.86
9 9.5+1.89 9.1+.46 7.5+.83 8.5.33
10 9.8+.67 8.3+.57 4.i1+.65 6.+.41
11 9.4-+.59 4.1+T.12 5.3+.41 9.3+.17

Group 3 (100 pg Anti-D.NP Antibody in OFA) Animl
Number 13 14 15 16
Weeks
8 2.4+.02 3.3+.30 0 4.2+.21
9 4.i1+.62 4.+.50 0 2.9+.14
10 2.+.39 6.3+.35 0 7J7+.79
11 3.0+.48 9.6+.47 0 2.9+.29
Animal
Number 17 18 19 20
Weeks
8 6. 1+.22 5.6+.49 .6+.01 7.6+ .51
9 9.2+.35 8.9+.46 6.0+.43 7.0+.88
10 6.4+.57 5.2+.34 4.5+.06 6.8+.50
11 7.3+.87 6.8+.65 6.4+.06 7.2+1.01










70
Table 9 Continued

Group 4 (IFA Alone) Group 5 (CFA Alone)
Animal
Number 21 22 23 24
Weeks
8 8.0+.69 5.2+.48 2.2+.29 .3+.02
9 9.0+1.22 7.6+.71 2.2+.20 0
10 7.1+.51 7.0+1.06 1.8+.36 0
11 8.5+.34 5.3+.44 2.6+.09 0

Group 5 Continued
Animal
Number 25 26 27 28
Weeks
8 1.0+.02 4.3+.34 4.2+.44 5.0+.28
9 1.9+.18 4.3+.31 4.2+.35 6.3+.06
10 1..16 3.1+.41 5.7+.26 4.7T.71
11 1.47+.21 2.9+.24 5.2+.38 5.9+.29


a) The level of anti-DNP IgE in these dogs from 0 to week 7 is found
in Table 2.
b) Each dog received DNP/ASC irrnunization at weeks 0,2,4,6,8,10. At weeks 7 and 9: Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100 pg autologous anti-DNP antibody in IFA; Group 3, 100 pg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.











71

Table 10
The Relative Anti-DNP IgG Concentration in 28 Dogs a)


Group 1 b) (10 pg Anti-DNP Antibody in CFA)
Animal
Number 1 2 3 4
Weeks
8 9.0+.11 24.7+.45 6.6+.86 8.3+.75
9 9.0+.66 27.4+2.50 10.0+.52 7.5-+.78
10 10.2+.93 28.5+.76 10.A+.77 6.2+1.05
11 9.1+1.04 26.8+.39 6.9+.30 9.8+.76

Group 1 Continued
Animal
Number 5 6 7 8
Weeks
8 5.5+.23 24.7+1.73 9.4+1.37 17.3+.75
9 4.6+.68 27.0+.39 14.0+.38 20.3+.72
10 5.8+.89 29.1+2.13 14.7+87 17.6+.48
11 8.9+.24 26.6+.90 12.3+.99 19.7+1.29

Group 2 (100 pg Anti-DNP Antibody in IFA)
Animal
Number 9 10 11 12
Weeks
8 17.1+1.37 8.9+.76 8.1+.62 14.2+.83
9 20.9+1.47 8.8+.99 9.2+.50 16.4+.38
10 22.9+1.10 11.9+.26 8.7+.61 15.4+1.12
11 25.4+.43 11.3+.69 8.3+.90 15.0+.32

Group 3 (100 pg Anti-DNP Antibody in CFA) Animal
Number 13 14 15 16
Weeks
8 5.6+.17 13.9+.83 7.9+.67 16.7+.99
9 10.3+.62 19.0+.65 11.8+1.12 16.0+.87
10 12.7+.41 20.4+.1l 14.3+1.57 15.0+1.01
11 12.2+.73 20.3+2.24 14.7+.59 17.4+2.56










72

Table 10 Continued

Group 3 Continued
Animal
Number 17 18 19 20
Weeks
8 7.3+.51 25.5+3.14 24.8+.64 14.6+1.73
9 6.5+.60 21.6+1.19 29.0+2.62 16.8+.96
10 8.4+.62 24.3+.29 30.1+4.71 17.2+2.33
11 9.1+.76 23.1+1.79 30.8+1.06 22.9+1.87

Group 4 (IFA Alone) Group 5 (CFA Alone) Animal
NumJber 21 22 23 24
Weeks
8 15.6+.93 23.3+1.87 30.1+1.74 8.1+.93
9 16.7+.81 28.7+1.61 27.7+.50 9.5+.62
10 16.2+1.3 31.4+1.05 31.9+3.81 10.7+1.08
11 17.1+.43 30.0+2.62 26.5+2.62 9.6+.27

Group 5 Continued
Animal
Number 25 26 27 28
Weeks
8 9.3+.81 13.6+.55 20.2+.96 22.2+1.43
9 10.4+.65 18.7+1.17 22.4+.68 23.4+.93
10 11.7+.88 20.5+1.48 17.3+1.24 24.3+.74
11 11.2+.76 18.0+1.42 18.7+.48 22.0+2.38


b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At
weeks 7 and 9: Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100 pg autologous anti-DNP antibody in IFA; Group 3, 100
pg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.







73

for IgE and IgG for each dog is given in table 9 and 10. There was a gradual increase in the mean antibody concentration in general for both IgE and IgG antibody whereas IgM was not detected after week 6. In two dogs (15, 24), after the seventh and eighth weeks respectively, there was a cessation of the IgE response. Although the me~an IgE antibody response increased for the groups in general, individual dogs varied considerably. For example, dogs 14 and 18 had a peak IgE antibody response at week 7 and thereafter the response diminished, whereas the peak response for dogs 19 and 27 occurred at the end of the schedule.

There -was no marked difference in the antibody

response between the different groups of dogs. To determine if there were any patterns in the antibody response between these groups, an analysis of variance comparing

time by group was calculated for each antibody class with the assistance of the Department of Biostatistics, College of medicine, University of Florida. There was no significant difference between these groups at any given time by this analysis (p greater than .05).

Because of the large variation between dogs, an

analysis of variance was calculated comparing dogs within a single litter in one group to dogs from the same litter in other groups as a function of time. This analysis was used to determine if there was variation between one treatment in the IgE or IgG antibody response as compared to a second







74

treatment within a single litter. In no case wAas a signif icant difference observed.



Discussion



The fact that anti-DNP antibody could not be detected in the effluent from the affinity column indicates that the column was effective in removing all anti-DNP antibody. The inability to detect IgE in the glycine eluate was expected because canine IgE is not stable at low pH. Halliwell (7) has shown that at a pH of 2.5 for 30 minutes there is greater than a tenfold decrease in detectable IgE antibody.

There are at least four possible reasons why autologous antibody administration failed to regulate the antibody response in these dogs as had been achieved in laboratory animals. Firstly, in these experiments the dogs had an established antibody response whereas in many of the experimental systems where passive antibody showed regulatory effects on antibody production, a primary or early secondary response was manipulated. It has been shown that it is much more difficult to manipulate a preexisting and established response than to alter a developing one. Secondly, there may be something unique about the regulatory effects of passive antibody on an immune response in young animals. Antibody is transferred from mother to







75

young both before and shortly after birth. If this passive antibody would result in long term suppression, then it might result in a negative selection process in those animals by rendering them immunologically non-responsive to pathogenic agents. Therefore, very young animals may be

less susceptable to the regulatory action of passive antibody. In fact, in a very recent report by Jarrett and Hall (83), they demonstrated that 7Taternal antibody or passively administered antibody given to newborn rats resulted in an enhanced IgG antibody response when challenged with antigen at six weeks of life. However, not every rat so treated had this enhanced IgG response and sote rats had a decrease in their IgE response. Thirdly it has been hypothesized that one way in which passive antibody administration could regulate antibody was through the generation of an anti-id response (30). It is possible that any anti-id produced by these dogs was not sufficient to regulate antibody. Lastly, such regulation may result in a clonal escape mechanism. Pawlak et al. (84) has shown that the administration of anti-id to A/J mice against the major cross reactive idiotype produced in these mice irimunized to p-azobenzenearsonate would suppress this idiotype and other cross reactive idiotypes, but there was a compensatory increase in other idiotypes not ordinarily expressed in these mice. If the administration of autologous antibody had regulated a subpopulation of antibody







76

molecules but in response other antibody molecules were expressed, the net effect mray not be observable if the entire class specific response were to be measured as was the case in these experiments.



Sumiary and Conclusions



The administration of autologous anti-DNP antibody in adjuvant to dogs which had an ongoing anti-DNP antibody response did not have a significant effect on this antibody response as compared to control dogs who received adjuvant without autologous antibody. This would suggest either that the regulation by passive antibodies, as seen in laboratory animals, does not operate in this species or that any regulation that occurred by such treatment could not be detected by the methods used in this study.











CHAPTER FOUR
THE IDENTIFICATION OF ANrI-IDIOTYPIC ANTIBODY


Introduction


Anti-idiotypic antibody has been produced by immunizing an animal with isologous or autologous antibody in adjuvant (40-42). The use of isologous or autologous antibody rather than homologous antibody eliminates the potential that allotypic determinants might be recognized rather than idiotypic determinants. The immunization schedule used in the previous experiments included the administration of autologous antibody in adjuvant. It was hoped that this treatment would regulate IgE antibody, but unfortunately, it did not. It was not known if this failure was because of a lack of an anti-id response or for other reasons. This treatment nay have induced anti-id. It is also possible that anti-id may have been produced during the immunization with antigen. The purpose of the experiments in this chapter was to determine if, at any time during the immunization schedule, anti-id ws detectable.




77







78

Materials and Methods



Antisera

The antisera used in these experiments were described in Chapter two. Any cross reactive anti-mouse immunoglobulin activity that was present in the anti-canine IgG, IgM or IgE antisera used in the anti-id RIA was removed by passage through an affinity column which had bound to it a 40 percent saturated ammnium sulfate precipitate of normal mouse serum.



Animals and Lmunization

The animals and immunization schedules have been described in Chapter two and three except that in the experiment designed to determine if the specificity of the antibody was important in the induction of an anti-id response, a different imunization protocol was used. Eight mature dogs were injected with 100 pg of aluminum hydroxide precipitated DNP-ASC by the intraperitoneal route on the day of arrival. At the same time, these dogs received a second injection of 100 pg of aluminum hydroxide precipitated ABA-KLH by the same route at a different site. These dogs were immunized three times at two week intervals. At the time of the last immunization, 30 ml of blood were obtained from each dog. The serum from this blood was used to puriEy antibody. The dogs were







79

arbitrarily placed into one of three groups. Group 1 (N=3) received 100 ig of autologous anti-DNP antibody in CFA by the subcutaneous route. This antibody was purified from serum by adsorption to and elution from a DNP coupled affinity column followed by passage through an ABA coupled affinity column to ensure that the purified anti-DNP antibody had no cross reactive anti-ABA antibody. Conversely, the autologous anti-ABA antibody in CFA for Group 2 (N=3) as purified from serum by adsorption to and elution from an ABA affinity column followed by passage through a DNP column. The control group, Group 3 (N=2) received CFA without autologous antibody. This later immunization was administered six weeks after the primary injection of antigen. Two weeks after this last injection, serum from each animal was assayed for anti-idiotypic antibody.



Anti-id RIA

The RIA used to detect anti-id was performed essentially as described for the antigen specific RIA using a number of muse monoclonal anti-DNP antibodies as the antigen. These were

a) Anti-DNP IgG, (a gift from Dr. A.P. Lopes, University of Pennsylvania) and as a control antibody, anti-H2K IgG (a gift from Dr. P. Klein, University of Florida).

b) Anti-DNP and IgE as a control antibody, anti-OVA IgE.







80

c) Anti-DNP IgM and as a control antibody, anti-SRBC

IgM.

The antibodies b) and c) were obtained from SeraLabs, Accurate Chemical and Scientific Corp., Westbury, N.Y.

d) Anti-DNP IgM (a gift from Dr. C.W. Clmn, Mississippi State University. This antibody will be designated anti-DNP IgM-2), and as a control antibody, anti-SR3C IgM (a gift from Dr. W.C. Raschke, La Jolla Cancer Research Foundation, Ca.).

The wells were coated with 10 pg antibody in 50 4i of Tris buffer (0.1 M, pH 8.0). A radiolabelled anti-canine IgG antibody was used as the radiolabelled probe unless otherwise stated.



Hapten Inhibition of Id/anti-id Interaction

The RIA using muse anti-DNP IgG monoclonal antibody or control IgG monoclonal antibody was performed as previously described except that after blocking any remaining active sites by the addition of HSA to the plates, various amounts of 2,4-dinitrophenol glycine (Sigma Chemical Co., St. Louis, Mo.) ranging from 0.001 to 0.1 mg in PBS were incubated for three hours at 40C. Serum samples were

then added to the wells, incubated for three hours at 40C and washed to remove unbound protein. A radiolabelled anti-canine IgG antibody was added, incubated for three






81


hours at 40C and the wells were washed to remove unbound radiolabelled antibody. The amount of radioactivity associated with the well was determined in a gamma counter.



cpm sample in the presence of hapten
% Inhibition = cpm sample in the absence of hapten X 100 Inhibition of Antigen Antibody Interactions by

Anti-Idiotypic Antibodies

The RIA using mouse anti-DNP IgG monoclonal antibody or a subtype and allotype matched control mouse IgG monoclonal antibody was performed except that after blocking remaining active sites by the addition of HSA to the plates, serum with or without anti-id was added, incubated for three hours at 40C and washed three times with RASTbuffer. Antigen (125 I DNP-HSA, approximately 20,000 cpm) was added to the wells and incubated for three hours at 4C and each well was washed five times to remove unbound antigen. Any anti-id that was bound to the anti-DNP antibody may inhibit this antigen-anti-DNP interaction. The amount of antigen bound to the wells was determined in a Packard gamma counter. The percent inhibition by anti-id was calculated by





% Inhibition =
cwm bound to plates after serum incubation
1- cpm bound to plates after PBS incubation X 100






82



Results


Identification of Canine Anti-Iditoypic Antibody

A screening procedure was used to assay for the presence of anti-id in serum obtained during the immunization schedule. Those dogs that received autologous antibody produced an antibody which would bind to mouse monoclonal anti-DNP IgG as seen in table 11. This binding was not detectable prior to such treatment in any dog nor could it be detected in control dogs at any time. There was no detectable binding to the anti-H2K IgG muse antibody in serum from any dog.

The experiment was repeated with another group of 12 dogs and the anti-id activity was converted to an arbitrary relative antibody concentration by interpolation from a scale derived froa the titration of a positive high titer sample identified in the screening procedure (table 12). This standard serum as given a relative antibody concentration of 10. Some dogs produced detectable levels of this anti-idiotypic antibody within one week after autologous antibody administration whereas other dogs took three weeks to develop such a response. There was also variation in the magnitude of the response observed. This antiidiotypic antibody could not be detected if anti-canine IgE or IgM antisera ues used as the radiolabelled probe rather









83
Table 11
Screening for Canine IgG Anti-Idiotypic Antibody by RIA
Using Mouse Monoclonal Anti-DNP IgG as the Antigen

Animal # a) 2 3 4 6 14
Week
0 476+16 331+19 448+7 397+14 422+21
1 453+7 590+1 481+11 421+10 470+47
2 412+21 347+3 274+3 335+6 443+15
3 317+9 277+11 352+9 371+-3 36740
4 305+15 358+3 367+27 401+16 457+12
5 378+31 376+7 358+18 421+11 352+25
6 396+14 417+37 284+10 379+23 318+14
7 421+27 335+10 409+15 522+27 314+24
8 1357+31 1815+26 433+21 2173+68 277+43
9 2140+150 2027+36 1533+117 2250+137 1420+48
10 1862+17 2208+55 2092+130 2297+100 2135+186
11 2174+79 3058+121 2107+41 3375+46 5087+161

Animal # 17 19 20 23 24
Week
0 476+1 367+40 307+22 422+9 483+19
1 453+17 351+26 406+19 470+16 517+42
2 418+5 481+56 464+3 394+7 347+5
3 519+43 435+13 253+13 367+14 318+23
4 442+16 467+24 373+20 334+17 351+1
5 376+18 430+45 481+6 442+26 235+7
6 340+28 398+1 295+30 462+11 434+16
7 315+7 384+93 563+38 371+31 346+34
8 798+13 1937+179 721+37 529+21 471+18
9 973+10 2476+88 936+68 315+8 512+46
10 1611+86 2720+9 2437+177 447+60 341+28
11 1735+122 2925+66 2026+69 473+3 429+27

a) All dogs were immunized with DNP-ASC in adjuvant at weeks 0,2,4,6,8 and 10. At weeks 7 and 9, dog 2,3,4 and 6 received 10 pg autologous anti-DNP antibody in CFA; dogs 14,17,19 and 20 received 100 pg autologous anti-DNP antibody in CFA; dogs 23 and 24 received CFA alone.
b) This represents the mean + standard deviation of a triplicate sample. All samples were assayed at a serum dilution of 1/5 in PBS.
The mean + standard deviation for all samples assayed using mouse anti-H 2K IgG1 was 477+97.










84

Table 12
The Relative Antibody Concentration of Canine IgG Anti-Idiotypic Antibody as Measured in a RIA using
Mouse Monoclonal Anti-DNP IgG as the Antigen a)


Animal
Number 1 5 7 8 13
Weeks
0 0 0 0 0 0
1 0 0 0 0 0
2 0 0 0 0 0
3 0 0 0 0 0
4 0 0 0 0 0
5 0 0 0 0 0
6 0 0 0 0 0
7 0 0 0 0 0
8 3.1+.2 4.0+.2 0 0 0
9 5.3+.4 5.4+.4 3.1+.3 3.1+.4 0
10 4.3+.3 5.1+.4 3.8+.2 5.4+.6 3.4+.2
11 5.2+.4 8.0+.6 2.9+.l 5.5+.2 6.6+.4

Animal
Number 15 16 18 26 27
Weeks
0 0 0 0 0 0
1 0 0 0 0 0
2 0 0 0 0 0
3 0 0 0 0 0
4 0 0 0 0 0
5 0 0 0 0 0
6 0 0 0 0 0
7 0 0 0 0 0
8 0 2.4+.1 1.2+.1 0 0
9 3.7+.2 2.2+.2 2.8+.1 0 0
10 10.0+1.3 4.3+.3 5.2+.4 0 0
11 9.2+.7 3.3+.1 7.1+.3 0 0

a) The relative antibody concentration + range was determined by interpolating from the titration of a serum sample containing high levels of anti-id. A value of zero indicates no detectable anti-idiotypic antibody.
b) All animals received DNP-ASC adjuvant at weeks 0,2,4,6,8,10. At
weeks 7 and 9 all animals received autologous antibody in adjuvant except 26 and 27 who received adjuvant alone.






85

than the anti-IgG antisera, indicating that it was of the IgG class.

When three different anti-DNP monoclonal antibodies

and three control monoclonal antibodies were used, the binding activity was detected only with the original antiDNP IgG (table 13), and to a lesser extent, to anti-DNP IgM-2 (table 14). In those animals in which anti-DNP IgM-2 binding activity was detected, a comparison was made between the serum from a point in time prior to autologous antibody administration and serum obtained after such

treatment. A minimum value that was two standard deviations above the mean of the control was considered indicative of anti-id activity. As was the case with the IgG

antibody, only those animals which received autologous antibody, showed binding activity and only after administration of autologous antibody (table 14).


The Role of Antibody in the Specificity of the
Anti-Idiotypic Production

To determine if immunization with an antibody whose specificity was other than anti-DNP would result in antiDNP/anti-id, eight dogs were given both DNP/ASC and ABA/KLH three times at two week intervals. Six weeks after the

primary injection of antigen, three dogs received 100 pg of autologous anti-DNP antibody in CFA, and a different three dogs received 100 pg autologous anti-ABA antibody in









86

Table 13
Detection of Canine IgG Anti-Idiotypic Antibody by RIA With Various Mouse Monoclonal Antibodies

Anti-DNP Anti-OVA Anti-DNP
(IgE) (IgE) (IgM)
Animal #
26 234+7 267+25 264+4
27 281+1 335+89 151+6
16 171+3 286+48 286+3
15 321+31 261+31 301+29
8 261+13 284+14 322+17
7 287+69 221+8 361+37
13 389+11 205+34 257+28
108+45 105+34

Anti SRBC Anti-DNP Anti-ABA
(IgM) (IgG) (IgG)

Animal #
26 182+36 321+95 209+47
27 197+12 354+29 218+38
16 176+42 4777+65 207+7
15 106+7 7571+246 176+12
8 106+7 1273+93 178+31
7 114+8 980+47 168+14
13 189+15 1144+68 221+8

Serum was diluted 1:2 in PBS. All samples were from week 10 of the immunization schedule. All animals had received DNP-ASC in adjuvant at week 0,2,4,6,8,10 and at weeks 7 and 9 received autologous antibody in CFA except 26 and 27 who received CFA alone.









87

Table 14
Detection of Canine IgG Anti-Idiotypic Antibody by RIA with Mouse Monoclonal
Dilution of Anti-DNP IgM2 Antibody Dilution of
Serum Anti-DNP Ig2Antibody
a)
26-E 26-L 27-E 27-L 16-E 16-L

1/5 162+11 286+21 135+14 248+45 90+7 643+31
1/10 82+14 154+15 97+26 152+29 86+ 296+19
1/20 71+21 135+9 126+36 181T+11I 103+11 156+12

15-E 15-L 8-E 8-L 7-E 7-L

1/5 225+31 912+25 216+47 570+46 249+1 672+6
1/10 184+8 442+27 128+6 373+31 125+8 343+37
1/20 115+12 247+6 102-+25 270+12 86+7 236+24

5-E 5-L

1/5 128+33 741+54
1/10 106+7 358+54
1/20 134+8 229+11

a) The E indicates serum obtained prior to the administration of autologous antibody administraion (16,15,8,7) or adjuvant alone (26,27), at week 6.
The L indicates serum obtained after such treatment from week 10. The mean + standard deviation for all samples assayed using anti-SRBC IgM was 186+53.






88

CFA. Two dogs received CFA alone. Serum before and two weeks after this treatment was screened for anti-id activity. As seen in table 15, the dogs that received autologous anti-DNP antibody produced an anti-id which was detected by the anti-DNP IgG mouse monoclonal antibody and failed to bind to the anti-ABA IgG. Dogs that received anti-ABA antibody in adjuvant developed anti-id which bound to anti-ABA mouse monoclonal IgG but failed to bind to the anti-DNP IgG muse monoclonal antibody. The two control dogs produced no detectable anti-id that uas reactive with

either m=use monoclonal antibody.


Hapten Inhibition and Elution Studies of the Id/anti-id

Interaction

The anti-idiotypic RIA wes used to determine if

hapten could inhibit the binding of anti-id to the muse anti-DNP antibody. In two of the six cases (5,15), 10 pg of hapten ;as able to inhibit id/anti-id interaction (table 16) as shown by a slight decrease in the cm bound to the wells as compared to the same sample incubated with PBS (22 percent and 26 percent inhibition respectively). As the concentration of hapten decreased, so did the percent inhibition, (17 percent and 5 percent at a fivefold decrease in hapten concentration). However, it is unclear how significant this inhibition %as because of the extremely large amounts of hapten required to obtain these






89

results. Similarly, hapten was unable to elute anti-id from id when a concentration of DNP-glycine of up to 20 pg was used.


Inhibition of Antigen Binding to Antibody by
Anti-Idiotypic Antibody

Since hapten could not consistently interfere with id/anti-id, it was reasoned that perhaps a large hapten coupled molecule might interfere with this interaction.

Although the anti-id was not binding to id determinants within the antigen binding site, it ray have bound to determinants close enough to the hypervariable region of the antibody molecule to sterically hinder antigen binding to antibody. Serum containing anti-id was preincubated with anti-DNP mouse monoclonal antibody prior to the addition of a radiolabelled DNP-HSA antigen to determine if the presence of anti-id could inhibit the anti-DNP antibody DNP-HSA antigen interaction. As seen in table 17, anti-id

was able to inhibit the binding of the radiolabelled antigen to the mouse monoclonal antibody. Animal 26 had no detectable anti-id, and serum from this dog failed to interfere with antigen binding to antibody. In contrast, the other dogs had detectable anti-id and inhibited this interaction from 21 to 50 percent of the maximum cpm bound.









90
Table 15
Specificity of Canine IgG Anti-Idiotypic Antibody
After the Administration of
Autologous Antigen-Specific Antibody
Anti-ABA Anti-DNP
Group 1 a) Group 2 b)

Antibody in well Antibody in well

Animal Serum Anti-DNP Anti-ABA Animal Anti-DNP Anti-ABA # Dil. IgG i # IgG1 I l

29 1/5 391+36 1901+101 33 1243+37 134+53
1/10 236+13 1675+118 996+115 142+13
1/20 153+14 1145+141 703+10 249+86

30 1/5 293+114 1867+157 34 1221+101 1183+65
1/10 186+26 1383+57 683+6 261+69
1/20 85+37 1061:46 545+59 143+19

31 1/5 383+35 1070+96 35 902+36 317+29
1/10 248+89 899;18 531+106 131+16
1/20 103+15 494+86 331+21 129+21

Control c)
32 1/5 172+16 246+42 36 238+33 186+26
1/10 158+47 117+28 186+32 133+58
1/20 98+8 88+14 121+13 94+17

Each dog received three immunizations with ABA-KLH and DNP-ASC at two week intervals. Six weeks after primary immunization, dogs received either.
a) Group 1 received 100 pg autologous anti-ABA antibody in CFA b) Group 2 received 100 pg autologous anti-DNP antibody in CFA c) Control received CFA alone
d) Mean cpm of the sample assayed in triplicate + standard deviation. The serum was diluted 1/5 with PBS. The serum used in this assay was obtained 2 weeks after this later
immunization.











91
Table 16
The Inhibition of Canine IgG Anti-id Binding to
Mouse Monoclonal Anti-DNP IgG by Hapten as Measured by RIA

Animal Number a)

2, 4 DNP Glycine 1 5 7

0 pg 3943 + 73 b) 4322 + 138 1775 +34
10 pg 3802 + 100 3533 + 69 1704 T 91
2 pg 3983 + 280 3783+136 1675 + 46
1 pg 3807 + 214 4268 + 56 1614 T 101
0.1 pg 3804 + 12 4623 + 219 1734 T 73


8 15 16
0 pg 2281 + 162 2702 + 131 1643 + 52
10 pg 2203 + 200 2026 + 168 1454 + 15
2 pg 2012 + 115 2593 + 173 1691 T 14
1 pg 2148 + 83 2738 + 57 1377 + 34
0.1 pg 2213 + 129 2694 + 89 1526 + 31


a) Each dog was immunized with autologous anti-DNP antibody in CFA at weeks 7 and 9. All samples were from week 9 in the schedule except sample 15 which was from week 11. b) This is the mean c=n of triplicate samples + standard
deviation. All samples were assayed with serum-diluted one to four with PBS.
A control well with muse anti-H2K IgG1 rather than anti-DNP IgG was assayed for each sample. The mean + standard deviations for all samples 283+47.










92

Table 17
Inhibition of binding of 125 I-DNP/HSA to
Mouse Monoclonal Anti-DNP Antibody
by Canine Anti-Idiotypic Antibody

Animal a) C.P.M. Bound + S.D. b) % Inhibition c)

26 994 + 43 0
8 493 + 38 50
18 544 + 25 45
13 783 + 41 21
15 611 + 13 39
1 521 + 18 48


a) Animal 8, 18, 13, 15 and 1 received autologous antibody in CFA and had detectable levels of anti-id; animal 26 received CFA alone and did not have detectable anti-id. all serum was obtained at week 11.
b) This is the mean cpm + standard deviation of a sample assayed in triplicate
c)
% Inhibition = c.p.m. control c.p.m. sample c.p.m. control

The control was a set of wells coated with mouse anti-DNP IgG and incubated with PBS rather than serum. The value for this was 1026+19.







93



Discussion



Dogs immunized with autologous antibody produced an antibody which bound to one.anti-DNP mouse monoclonal IgG. This antibody was present only after such treatment and not present prior to the administration of autologous antibody. The specificity was limited to id determinants present on some but not all anti-DNP mouse monoclonal antibodies. It could not be detected using mouse monoclonal antibodies whose specificity was other than DNP such as anti-H2K IgG or anti-ABA IgG. This putative anti-id had no specfificity for mouse immunoglobulin heavy or light chain constant region determinants as indicated by the failure to detect any activity when an allotype and isotype match non-anti-DNP antibody was used in the assay as antigen. These findings suggested that this muse binding protein was anti-idiotypic in nature.

When the serum which contained this anti-id was

assayed to determine if this antibody could bind to other

mouse anti-DNP monoclonal antibodies, there was no detectable binding to two of the muse anti-DNP monoclonal antibodies and a limited binding to the monoclonal anti-DNP IgM antibodies. The difference in the level of anti-id detected when either the anti-DNP IgG or the anti-IgM antibody was used as the antigen indicates the difference in




Full Text
119


o
2
4 6
Weeks
MeanS.D. (range)
O O (O-O)
1 .3 .6 (0-2.6)
2 7 O 5.0 (0-25.4)
3 13 2 6.5 (37-270)
4 13.5 7.2 (56-30.0)
5 13.6 6.8 (4.8-26.9)
6 122 7.6 (50-28.7)
7 14.7 8.1 (3.1-26.3)


5
IgG antibody can have a role in regulating allergic symptoms
(22). An IgG response can be induced by administration of
allergen either by the normal route of exposure or by a
route other than for normal exposure (i.e. subcutaneous
versus inhalation). This IgG antibody presumably completes
the allergen-IgE antibody interactions (21). However, this
therapy is not without side effects (20). Furthermore, only
about 65 per cent of patients treated with hyposensitization
have clinical improvement (20).
An alternate type of hyposensitization involves
modifying the allergen, usually by mild denaturation.
Studies in mice with urea denatured ragweed showed that such
treatment reduced allergenicity while maintaining immuno-
genicity of the allergen. If large doses of urea denatured
ragweed were given to mice previously sensitized to
unmodified ragweed, such therapy resulted in antigen-
specific T suppressor cell induction without the development
of anaphylaxis (12). These cells suppressed the anti
ragweed IgE response. A controlled study is underway to
determine if this form of immunotherapy is any more
effective in controlling allergic symptoms than conventional
hyposensitization.
Another approach is to regulate the response with
products of the immune system. Smith (23), in 1909, was the
first person to recognize that antibody could suppress the


52
Table 4
The Relative Anti-DNP IgG Concentration
In Two Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M +S,
.D b)
Weeks
0
0
0
0
0
0
0
0
1
.8
0
2.6
0
0
0
.6
+ 1
2
2.4
3.7
1.7
11.7
3.6
10.9
5.7
+ 4.
3
8.0
7.4
4.8
12.8
8.8
10.0
8.6
+ 2.
4
4.8
7.4
6.2
8.9
9.7
8.2
7.5
+ 1.
5
8.6
4.8
6.9
7.6
9.0
7.8
7.5
+ 1.
6
6.7
6.1
5.3
9.8
6.6
5.0
6.6
+ 1.
7
5.4
6.0
3.1
11.9
7.2
6.4
6.7
+ 2.
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
.3
0
.1
0
2
9.1
9.1
5.3
4.8
12.9
X
8.2 + 3.
3
15.5
15.3
16.7
16.9
18.0
12.9
15.5+ 1.
4
16.4
17.0
14.3
21.3
13.9
22.8
17.6+ 3.
5
26.0
25.6
14.7
20.8
13.5
26.9
21.3+ 5.
6
23.3
24.1
10.4
21.9
12.7
28.7
20.2+ 7.
7
26.8
25.8
18.9
24.8
23.0
24.7
24.0+ 2.
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
DNP/ASC in adjuvant at weeks 0,2f4 and 6.
b) Mean + standard deviation
3
7
8
5
7
9
3
8
7
9
1
8


131
reciprocal id, then it would be expected that id would be
present in the serum before the appearance of anti-id, but
after the appearance of anti-id the reciprocal id would
disappear. Since a number of different ids were used as
the probe, the corresponding anti-id may be detected either
slightly before the point in time the id came from, coin
cident with the anti-id, or considerably later in time than
the id. Alternatively, if id/anti-id were complexed then
the disruption of these complexes might allow anti-id to be
detected.
When autologous anti-DMP Fiab)'^ was used to assay
for anti-id, three different patterns in the detection of
anti-id were evident. In the first pattern, anti-id was
detected after the appearance of id, but not coincident
with nor before its appearance. These results suggest that
there vas a lag phase between the appearance of id and the
corresponding anti-id in the serum. In the second pattern,
id and anti-id are present within a single serum sample.
The maximum level of anti-id vas detected in serum at the
same time that the id appeared in two cases. Furthermore,
anti-id were present very early in the response. Because
the dogs in these experiments would still be expected to
have colostrum-derived antibody, these anti-id may
represent maternal Immunoglobulin. Unfortunately, it was
not possible to obtain serum from these bitches to
determine if they had anti-id present.


Figure 16.
The identification of anti-id in various serum samples, in dog 1,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous antibody in adjuvant at weeks 7
and 9. The arrow, marked idiotype probe, indicates the time from
which the anti-DNP F(ab)'2 came. These were used as antigens to
detect anti-id. The bars represent the standard deviations of the
mean.


2L
solution of protein (70). The pH vas then adjusted with 0.1
N NaOH to pH 6.3 to ensure adequate precipitate.
Affinity Chromatography
Sepharose 4B beads (Pharmacia Fine Chemicals) were
activated using cyanogen bromide (CnBr) by adding 1.5 grams
CnBr in 20 ml distilled vater to 10 ml of washed Sepharose
4B beads and adjusted to pH 11 with I N NaOH. This mixture
was maintained on ice at pH 11 for 6 minutes after which
the beads vare vashed with 100 volumes of iced oold water.
Ninety milligrams of protein in 6 ml PBS, pH 7.2 were added
and incubated for 12 hours at 4C. Alternatively pre
activated Sepharose 4B beads were obtained (Pharmacia Fine
Chemicals) and protein was bound to these beads as
described by the manufacturer. To remove unbound protein
in both cases, the beads were washed with five alternate
cycles of 0.1 M Tris buffer, pH 8.3 containing 0.5 M NaCl
followed by 0.1 M glycine HC1, pH 2.8. Any retaining sites
were blocked by incubating the beads in 0.1 M Tris buffer,
pH 8.3 for four hours at room temperature. The column was
then flushed with normal canine serum and washed as
described above.


REFERENCES
1. Patterson, R. (1969) Laboratory Models of Reaginic
Allergy. Prog. .Allergy. 13:332-407.
2. Patterson, R., and Kelly, J.F. (1974) Animal Models of
the Asthmatic State. Ann. Rev. Med. 25:53-58.
3. Pepys, J. (1975) Atopy in Clinical Aspects of
Immunology, third ed. Edited by Gell P.G.H., Coombs,
R.R.A., Lachmann, P.J. Blackwell Scientific Publications,
Oxford, England.
4. Schwartzman, R.M. and Rockey, J.H. (1972) Atopy in the
Dog. Arch. Derm. 96:413-422
5. Halliwell, R.E.W., Schwartzman, R.M., and Rockey, J.H.
(1972) Antigenic Relationship between Human IgE and Canine
IgE. Clin. Exp. Immunol. 10:399-407.
6. Halliwell, R.E.W., Schwartzman, R.M., Montgomery, P.C.,
and Rockey, J.H. (1975) Physicochemical Properties of
Canine IgE. Transplant Proc. 8:537-543.
7. Halliwell, R.E.W. (1972) Studies on Canine IgE.
Ph.D dissertation. Cambridge University, Cambridge,
England.
8. Johansson, S.G.O. and Mel1bin, T., Vahlquist, B. (1963)
Immunoglobulin Levels in Ethiopian Preschool Children with
Special Reference to High Concentrations of Immunoglobulin
E. (IgND). Lancet 1:1118-1121.
9. Rosenberg, E.B., Whalen, G.E., Bennich, H. and
Johansson, S.G.O. (1971) Increased Circulating IgE in a
New Parasitic DiseaseHunan Intestinal Capillariasis. N.
Eng. J. Med. 283:1148-1149.
10. Tada, T., Ishizaka, K. (1970) Distribution of Gamma E
Forming Cells in Lymphoid Tissues of the Human and Monkey.
J. Immunol. 104:377-387.
11. Jarrett, E.E.E., and Miller, H.R.P. (1982) Production
and Activities of IgE in Helminth Infection. Prog Allergy
31:178-233.


10
inhibit the interaction by 100 percent, maximum inhibition
was only 68 percent (35). The most extreme example in which
hapten cannot inhibit id/anti-id interactions are in those
studies in which cross reactive ids are present on antibody
molecules of widely different specificity. For example,
Eichmann et al. (36) showed that one half of the A5A id
producing clones in A/J mice immunized with a streptococcal
carbohydrate lacked the ability to bind this antigen.
Obviously then, antigen would not be expected to inhibit
this id-anti-id interaction. In other studies, Bona et al.
(37) showed that not all the id positive antibody following
immunization with inulin could be removed with an inulin
immunoabsorbent. In these experiments, the anti-inulin anti
body produced following antigenic stimulation bears a
predominant id. However, some immunoglobulin following
antigenic stimulation had this id but lacked specificity for
inulin. These experiments therefore suggest that some
mechanism exists naturally in which id positive clones of
immunoglobulin producing cells are expanded following
antigen stimulation but that not all the id positive immuno
globulin is specific for the immunizing antigen. These
experiments clearly show that although id/anti-id can
usually be hapten inhibited, this property is not a require
ment for an antibody to be anti-idiotypic.


100
immunizing dogs with autologous anti-DNP antibody, anti-id
was detected. However, this anti-id was not detected
during the response to DNP-ASC.
The failure to detect anti-id prior to the immuni
zation with antibody in adjuvant could be because; 1)
anti-id was rot present or, 2) the method used to detect it
was wrong. The purpose of the experiments in this chapter
were to determine if anti-id could be detected at any time
during a DNP specific antibody response using autologous
anti-DNP FCab)^ antibody fragments as the source of id.
Materials and Methods
Preparation and Immobilization of Anti-DNP P(ab)'^
Fragments to a solid Matrix
Anti-DNP Fiab)^ fragments were prepared from anti
body purified from a single serum sample by affinity chroma
tography. The protein was digested with pepsin and the
Fiab)^ was separated from intact antibody and Fc frag
ments as previously described in Chapter two. The
F(ab)'2 from each sample was handled separately and
F(ab)'2 from a single sample will be referred to as a
set. Each set of antibody fragments was bound to a .solid
support matrix (Immunobeadr. BioRad Laboratories,
Richmond, CA) as described by the rranufacturer. Briefly, a
given quantity of anti-DNP F(ab)'2
in 0.003 M


8
assumptions are crucial premises to this theory. Firstly,
most idiotypes exist at a level too low to induce tolerance.
Thus, antigenic stimulation and expansion of these id will
stimulate the production of a reciprocal set of anti-id.
The id is then regulated directly by the anti-id, indirectly
by the anti-id on T-cellsor or by anti-id acting on T-cells.
As the concentration of anti-id reaches sane critical
threshold, a second anti-id response develops which is
specific for the id of the anti-id. This anti-id would,
therefore, be an anti-(anti-id) and would then stimulate a
fourth response and so on, thereby resulting in an inter
related network of regulation between antibody molecules.
Jerne also stated that id determinants can be present not
only on antibody molecules of one specificity, but may be
present on unrelated antibody molecules. Thus, antibody
against antigen x might share some ids with antibody against
antigen y. Lastly, although anti-id usually suppresses the
corresponding id, it can be stimulating for the id as well.
The anti-id would be expected to have a three dimensional
structure similar or identical to the specific antigenic
determinant. This type of anti-id is termed an internal
image of antigen.
.The-characteristics of idiotypes of antibody molecules
have been described (31-35). In many instances, idiotypes
are located in or very near to the antigen binding site.


the assay and 50 pi of the first- No immuno- react i ve Fc
125
material was detectable on any bead set vhen I
anti-canine heavy chain specific IgG was incubated with an
aliquot of each-bead set.
Detection of Natural Occurinq Anti-id
Anti-DNP F(ab)*2 fragments from a single serum
sample immobilized as described above vas used as an
antigen to detect anti-id. Various serum samples frcm the
same dog were assayed for anti-id after being chromato
graphed through a DNP-affinity column to remove anti-DNP
antibody which theoretically could compete by binding
anti-id. Each sample was concentrated by negative pressure
dialysis to approximately the starting volume of serum.
The samples were assayed by incubating an undilute, a 1/2
and a 1/4 dilution in PBS of each sample with a
standardized amount of autologous anti-DNP Ftab)^ bound
beads for three hours at room temperature. The beads were
then centrifuged and the supernatant removed and washed
with RAST+ three times to renove unbound antibody. Radio-
labelled heavy chain specific anti-canine IgG was added to
each set of beads (approximately 30,000 cpm/sample),
incubated for three hours at roan temperature and vashed to
remove unbound radiolabelled antibody. The radioactivity
of each sample vas determined in a Packard gamma counter.
Included in each assay at all sample dilutions were beads


22
Pepsin Digestion and Purification of F(ab)'^ Antibody
Fragments.
The usual procedure for F(ab)'^ digestion of immuno
globulin vas to digest the antibody with 6 percent pepsin
(w/v) in 0.2 M acetate buffer, pH 4.5 for 13 hours at 37
C. However, this process resulted in some loss of
antigen binding of the F(ab)'2 presumably from the
prolonged incubation time at pH 4.5. Where maintenance of
this activity vas critical, protein vas digested with 20
percent pepsin w/v in 0.2 M acetate buffer pH 4.5 for five
hours at 37C. The digested protein vas separated from
Fc pieces and intact antibody by passage through a cyanogen
bromide-activated heavy chain specific immunoabsorbent
column followed by passage through a Staphylococcus
protein A affinity column (Pharmacia Fine Chemicals). The
effluent vas concentrated by negative pressure dialysis and
dialyzed against PBS, pH 7.2.
Antisera
a) Preparation and purification of anti-IgG. Normal
canine serum (NCS) was precipitated with a 40 percent
saturated solution of ammonium sulfate. The precipitate
was dialyzed against 0.035 M phosphate buffer, pH 8.0 and
applied to a DEAE ion exchange column equilibrated with
this same buffer. The effluent protein was concentrated by


Figure 22.
The identification of anti-id in various serum samples, in dog 1,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous anti-DNP antibody at weeks 7 and
9. The arrow, marked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used as antigens
to detect anti-id. The bars represent the standard deviation of
the mean.


This dissertation is dedicated to my wife Nancy.
Without her love and support (and typing), it
would never have been.


75
young both before and shortly after birth. If this passive
antibody would result in long term suppression, then it
might result in a negative selection process in those
animals by rendering them immunologically non-responsive to
pathogenic agents. Therefore, very young animals may be
less susceptable to the regulatory action of passive
antibody. In fact, in a very recent report by Jarrett and
Hall (83), they demonstrated that maternal antibody or
passively administered antibody given to newborn rats
resulted in an enhanced IgG antibody response when
challenged with antigen at six weeks of life. However, not
every rat so treated had this enhanced IgG response and
sane rats had a decrease in their IgE response. Thirdly it
has been hypothesized that one way in which passive anti
body administration could regulate antibody '.vas through the
generation of an anti-id response (30). It is possible
that any anti-id produced by these dogs was not sufficient
to regulate antibody. Lastly, such regulation may result
in a clonal escape mechanism. Pawlak et al. (84) has shown
that the administration of anti-id to A/J mice against the
major cross reactive idiotype produced in these mice immu
nized to p-azobenzenearsonate would suppress this idiotype
and other cross reactive idiotypes, but there was a compens
atory increase in other idiotypes not ordinarily expressed
in these mice. If the administration of autologous
antibody had regulated a subpopulation of antibody


THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
DURING AN IMMUNE RESPONSE IN DOGS
BY
KEVIN T. SCHULTZ
A DISSERTATION PRESENTED TO 'THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1983


CHAPTER SIX CONCLUSIONS
137
REFERENCES 140
BIOGRAPHICAL SKETCH 149
v


acknowledgements
I wish to express my deep appreciation and thanks to
some very important people in my life. First of all, to ray
parents, my wife and ray family for their constant love and
support.
Very special thanks are extended to Dr. Richard
Halliwell for his help, guidance and friendship throughout
my graduate experience.
To Drs. Gail Kunkle, Robert Mason and Bill Hollaran,
my deep appreciation for their unwavering friendship (and
occasional needed prods) along my journey.
I also wish to thank Drs. K.I. Berns, M.D.P. Boyle,
R.B. Crandall, A.P. Gee, G.E. Gifford, M.J.P. Lawman, and
P.A. Small, Jr., for their suggestions, help and
encouragement.
Last (but not least), I would also like to thank my
fellow graduate students and all my friends for making this
experience a memorable one.
ill


Figure 12.
The mean relative anti-DNP IgE concentration as measured by RIA.
All dogs received DNP-ASC immunization at weeks 0,2,4,6,8 and 10.
At weeks 7 and 9, the treatment consisted of immunization with:
Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100
pg autologous anti-DNP antibody in IFA; Group 3, 100 pg
autologous antibody in CFA; Control, either IFA alone or CFA alone.
The relative antibody concentration was calculated from an
arbitrary scale derived Eran the titration of a standard anti-DNP
IgE containing serum (see text for further details).


6
development of an immune response. In these experiments he
showed that certain mixtures of diphtheria toxin and anti
toxin oould be very immunogenic in guinea pigs, but if there
was a large excess of antitoxin, the immunized guinea pig
would fail to mount an immune response against the toxin.
Numerous studies in the 1950's and 1960's verified this
observation and also demonstrated that the isotype, amount,
affinity and time of administration were important variables
in determining the degree of suppression float passive anti
body had on the immune response (reviewed in 24). For
example, IgG antibody given after antigenic exposure was
more effective in inducing antibody suppression than IgiM
antibody. Further, the suppressed state was longer lived
using IgG than IgM antibody. An interesting report by Chan
and Sinclair (25) stated that the administration of anti-
SR3C antibody given to mice after antigenic challenge led to
a suppression of this response and this tolerant state could
be transferred from one mouse to another with T-cells from
the tolerized mouse. They suggested that the regulatory
action of antibody operated through sane sort of "induced
pathway or secondary immune response" (25 p. 977).
In the early 1970's it was likewise shown that IgE
antibody could be regulated by passively administered
antibody (26-28). Rabbits were immunized to produce high
titer IgE antibody and were given passive antibody 24 hours


96
the antigen combining site and other anti-ids bind to ids
that are more distant. This failure to observe complete
inhibition could have been for at least two reasons. There
was not sufficient anti-id in the serum to block all the
antigen binding sites. Alternatively, the anti-id bound to
id determinants located on the molecule in such a way that
complete inhibition of the antigen binding site was not
possible.
The results in this chapter show that those animals
given autologous antibody in adjuvant produced an anti-id.
This anti-id response is not a function of the adminis
tration of adjuvant because control dogs given adjuvant
without antibody failed to produce detectable levels of
anti-id. In those dogs that produced anti-id the autol
ogous antibody that was used for immunization had been
subjected to harsh treatment (e.g. glycine HCl elution from
an affinity column) during the purification process. This
treatment could possibly alter the ids present in the anti
body. Therefore it might be possible that the antigen
specificity of the antibody has little or nothing to do
with the anti-id that is produced.
To address this question, dogs were immunized to two
different antigens, DNP-ASC and ABA-KLH. They were then
given either autologous anti-DNP antibody or autologous
anti-ABA antibody emulsified in adjuvant and subsequently
produced an anti-id which bound to mouse monoclonal


2
determinants of canine IgE (7). The level of anti-DNP
antibody increased both when the serum was inactivated and
when non-labelled anti-IgE antiserum was added to the
sample. This indicates that this anti-IgG antiserum has
minimal, if any, anti-IgE activity.
b) Preparation and purification of anti-IgE. A 40 per
cent saturated ammonium sulfate precipitate of serum
obtained, from a dog that was heavily parasitized and
presumed to have high levels of IgE, was dialyzed against
0.035 M phosphate buffer pH 8.0 and applied to a DEAE
cellulose column equilibrated with this same buffer. The
effluent protein was concentrated by negative pressure
dialysis and applied to a set of three in series Sephacryl
S-200 columns (Pharmacia Fine Chemicals, Piscataway, N.J.).
The first one-third of the second protein peak, which was
the IgE-rich fraction as determined by agar-gel immuno-
precipitation, was collected, concentrated by pressure
dialysis and reapplied to these columns. The resulting
IgE-rich fraction was collected and used to immunize
rabbits as described previously. The rabbits were bled as
described above. Serum that produced visible precipitation
lines against the immunizing antigen in an agar gel immuno
diffusion were pooled. The resulting antiserum detected
both IgE and IgG by immunoelectrophoresis. It was rendered
specific for the former protein by passage through an


23
negative pressure dialysis. One milligram of this material
was emulsified in complete Freund's adjuvant (CFA) and
administered intramuscularly to rabbits at two week
intervals four times. Fifty milliliters of blood were
obtained from the rabbit by ear vein venapuncture every two
weeks starting after the second immunization. All serum
which gave visible precipitation reactions by agar-gel
diffusion against canine IgG was pooled. This antisera
was passed through a cyanogen bromide-activated sepharose
4B F(ab)'2 affinity column, to remove light chain
activity, followed by adsorption to and elution with
alycine HC1 (Osl M), pH 2.8 from a canine IgG bound
affinity column. This anti-IgG detected three subclasses
of canine IgG (IgG^ IgG2ab, IgG2jJ but no other
protein as measured in an immunoelectrophoresis (70) of NCS
(figure 1). To determine if this antiserum detected IgE,
the antiserum was radiolabelled and used in a radio
immunoassay. The serum sample tested contained both anti-
DNP IgG and anti-DNP IgE. Therefore, an aliquot of this
serum was heat inactivated and the level of anti-DNP IgG
was compared in this aliquot to a second aliquot of this
serum that was not heat inactivated. Additionally, anti
canine IgE was added to an aliquot of this sample to
determine if this unlabelled anti-IgE might compete with
the anti-IgG for Fc binding sites. Heating serum for four
hours at 56C destroys the heavy chain antigenic


27
affinity column made with the heat inactivated immunogen
which removed all antibody except anti-IgE antibody.
Purified antibody was then prepared by adsorption to and
elution from an IgE-rich affinity column. This purified
antiserum detected a single heat-labile protein by immuno-
electrophoresis (figure 2), produced reverse cutaneous
anaphylaxis in dogs at a high dilution of serum (10 )
and was unable to detect canine anti-DNP IgG in a RIA
indicating that it had no specificity for this antibody.
c) Preparation and purification of anti-IgM. Canine IgM
myeloma serum, which contained approximately 58 mg/ml IgM
was chromatographed on Sepnacryl 3-200 and the void volume
was collected to obtain IgM. Two milligrams of this
material was emulsified in CFA and injected intra
muscularly at four sites into sheep. This was repeated at
two week intervals five times. Five hundred milliliters of
blood were collected by jugular vein venapuncture every two
weeks. Sera that produced precipitation lines against the
immunizing antigen, in an agar-gel diffusion against the
immunizing antigen, were pooled. Light chain activity was
removed from the antiserum by passage through a canine IgG
affinity column. Antibody was then purified by adsorption
to and elution from an IgM affinity column. The eluted
proteins produced two bands on Immunoelectrophoresis of
NCS, one of which was IgM and the other an unknown protein.


89
results. Similarly, hapten was unable to elute anti-id
from id when a concentration of DNP-glycine of up to 20 (jg
vas used.
Inhibition of Antigen Binding to Antibody by
Anti-Idiotypic Antibody
Since hapten could not consistently interfere with
id/anti-id, it vas reasoned that perhaps a large hapten
coupled molecule might interfere with this interaction.
Although the anti-id vas not binding to id determinants
within the antigen binding site, it may have bound to deter
minants close enough to the hypervariable region of the
antibody molecule to sterically hinder antigen binding to
antibody. Serum containing anti-id was preincubated with
anti-DNP mouse monoclonal antibody prior to the addition of
a radiolabelled DNP-HSA antigen to determine if the
presence of anti-id could inhibit the anti-DNP antibody
DNP-HSA antigen interaction. As seen in table 17, anti-id
was able to inhibit the binding of the radiolabelled
antigen to the mouse monoclonal antibody. Animal 26 had no
detectable anti-id, and serum from this dog failed to inter
fere with antigen binding to antibody. In contrast, the
other dogs had detectable anti-id and inhibited this inter
action from 21 to 50 percent of the maximum cpm bound.


90
Table 15
Specificity of Canine IgG Anti-Idiotypic Antibody
After the Administration of
Autologous Antigen-Specific Antibody
Anti-ABA
Anti-DNP
Group 1 a)
Group 2 b)
Antibody in well
Antibody
in well
Animal Serum Anti-DNP Anti-ABA
Aina 1
Anti-DNP
Anti-ABA
#
Dil.
m1
I2GX
#
laGg
12%
29
1/5
391+36
1901+101
33
1243+37
134+53
1/10
236+13
1675+118
996+115
.142+13
1/20
153+14
1145+141
703+10
249+86
30
1/5
293+114
1867+157
34
1221+101
113+65
1/10
186+26
1383+57
683+6
261+69
1/20
85+37
1061+46
545+59
143+19
31
1/5
383+35
1070+96
35
902+36
317+29
1/10
248+89
899+18
531+106
131+16
1/20
103+15
494+86
331+21
129+21
Control c)
32
1/5
172+16
246+42
36
238+33
186+26
1/10
158+47
117+28
186+32
133+58
1/20
98+8
88+14
121+13
94+17
Each
. dog received three
immunizations
with
ABA-KLH and DNP-ASC at two
week
: intervals. Six weeks after primary immunization, dogs received
either.
a)
Group 1
received 100
pg autologous anti
-ABA antibody
in CFA
b)
Group 2
received 100
pg autologous anti
-DNP antibody
in CFA
c)
Control
received CFA
alone
d) Mean cpm of the sample assayed in triplicate + standard
deviation. The serum was diluted 1/5 with PBS.
The serum used in this assay was obtained 2 weeks after this later
immunization.


Figure 18.
The identification of anti-id in various serum samples, as measured
in dog 24, by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and CFA at weeks 7 and 9. The arrow, marked
idiotype probe, indicates the time from which the anti-DNP F(ab)'2
fragments came. These were used as antigens to detect anti-id.
The bars represent the standard deviation of the mean.


81
hours at 4C and the wells were washed to remove unbound
radiolabelled antibody. The amount of radioactivity asso
ciated with the well was determined in a gamma counter.
cpm sample in the presence of hapten
% Inhibition = cpm sample in the absence of hapten X 100
Inhibition of Antigen Antibody Interactions by
Anti-Idiotypic Antibodies
The RIA using mouse anti-DNP IgG monoclonal antibody
or a subtype and allotype matched control mouse IgG mono
clonal antibody was performed except that after blocking
remaining active sites by the addition of HSA to the
plates, serum with or without anti-id was added, incubated
for three hours at 4C and washed three times with RAST-
buffer. Antigen (125 I DNP-HSA, approximately 20,000 cpm)
was added to the wells and incubated for three hours at
o
4 C and each well was washed five times to remove unbound
antigen. Any anti-id that ivas bound to the anti-DNP anti
body may inhibit this antigen-anti-DNP interaction. The
amount of antigen bound to the wells was determined in a
Packard gumma counter. The percent inhibition by anti-id
was calculated by
% Inhibition =
cpm bound to plates after serum incubation
cpm bound to plates after PBS incubation
1-
X 100


54
Table 6
The Relative Anti-DNP IgM Concentration
in 2 Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M +S.D. b)
Weeks
0
0
0
0
0
0
0
0
1
6.5
4.5
1.1
2.7
5.1
4.0
4.0 + 1.9
2
3.4
4.4
4.4
2.1
.6
6.2
3.5 + 2.0
3
0
.6
6.0
0
0
2.5
1.5 + 2.4
4
0
.1
1.7
0
0
0
.3 + .7
5
0
0
.8
0
0
0
.1 + .3
6
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
4.4
6.4
1.8
3.9
8.1
10.6
5.9 + 3.2
2
3.2
5.6
5.2
2.1
12.6
X
5.7 + 4.1
3
1.4
5.1
2.3
.8
2.7
4.2
2.8 + 1.6
4
.2
3.9
.5
0
0
.8
.9 + 1.5
5
0
0
0
0
0
.1
.01 + .04
6
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
immunization with DNP-ASC in adjuvant at weeks 0,2,4 and 6.
b) Mean + Standard Deviation


134
was that the antibody stimulated the subsequent production
of identical or very similar antibody through an id/anti-id
interaction Therefore, this unusual pattern in the
appearance of anti-id could be the result of autologous
antibody administration. However, the presence of this
anti-id my be the result of factors governing the
production and detection of anti-id which are unforseen at
this time.
In two of the five animals there is complete failure
to detect anti-id using autologous antibody in any serum
obtained throughout the immunization schedule. There are a
number of different possible explanations for this result.
Firstly, only anti-id of the IgG class was measured and it
is possible that other anti-id isotypes were produced in
these two animals. In fact, in a recent study it was
observed that in man there was an isotypic shift over time
of the anti-id specific for a given set of auto-antibodies
(53). Alternately, there my be certain ids which favor
the production of the reciprocal anti-ids. In an experi
ment in outbred rabbits, anti-id production seemed to be
associated with the presence of a few ids that favor
anti-id production. Those rabbits not expressing such ids
failed to produce detectable anti-id antibodies (57).
.Based on this, it is possible that, in dogs, certain id are
especially important for anti-id production and in those
dogs not expressing such ids there is a failure to produce


106
Table 19
Detection of Canine Anti-Iditoypic Antibody by RIA
Using Autologous Anti-DNP Fiab)^ as the Id
.Dog Number 14
Source of Anti-Id (Week) a)
2 4 7 11
Source of
id
(Week) b)
Effluent
Dilution
0
137+31
65+21
644+68
705+18
1/2
36+18
7+7
152+43
351+47
1/4
0
5+7
12+11
90+15
0
216+39
318+49
381+68
848+9
1/2
135+23
37+31
150+62
677+34
1/4
77+64
0
47+15
300+23
0
624+5
778+18
838+52
405+77
1/2
309+7
653+28
786+38
233+16
1/4
99+29
386+16
493+14
83+11
Control c)
0 321+37
1/2 186+7
1/4 128+29
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP F(ab)' immobilized to a solid
matrix. The dog was immunized with DNP-ASC in adjuvant at weeks
0,2,4,6,8, and 10 and received 100 gg autologous anti-DNP antibody
in CFA at weeks 7 and 9.
c) Control id was normal canine IgG F(ab)' immobilized to a solid
matrix.


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Richard E.W. Halliwell, Chairman
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
JL p A
George E. Gifford
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosopj;
//,
8*,
Parker A/ Small-/ Jr //?'
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
\
Richard B. Crandall
Professor of Immunology and
Medical Microbiology


28
This second activity was removed by adsorption with the
supernatant of a 50 percent saturated ammonium sulfate
precipitation of ISJCS (figure 3). This antiserum vas
assayed for anti-DNP IgE and IgG activity by RIA. Serum
that was used contained both of these antibody isotypes.
No antibody vas detected indicating the antiserum did not
have activity for IgE or IgG.
Isotope Labelling of Protein
Two methods vare used to label proteins with radio
active iodine. In the first method, between 1 and 2 mg of
protein in 0.1 ml PBS pH 7.0 without azide and .5 mCi
125
I (Amersham, Chicago, II.) was incubated on ice with
15 pi KI(0.1 mM) and 30 pi chloramine T (10 mM) for 15
minutes. After this incubation, 25 pi sodium metabi
sulphate (10 mM) and 50 pi KH100 mM) were added to stop
tiie reaction. To separate bound and free iodine, the mat
erial vas chromatographed through a G-25 Sephadex column.
The first peak containing radiolabel was pooled, concen
trated and dialysed against PBS, pH 7.2. Alternatively,
p
one lodobead (Pierce Chemical Co., Rockford, II.) was
added to 100 pg of protein in PBS, pH 7.0 and 0.5 mCi
125
I. After a fifteen minute incubation, the bound and
195
unbound I vas separated as described earlier. The
specific activity of the radiolabelled antibody was usually
about 300 pCi/mg protein (range 212-496).


46
Table 1 Continued
Animal
Number
17
18
19
20
Weeks
0
0
0
0
0
1
1.1+.03
14.8+2.01
3.9+.39
2.7+.22
2
4.4+.17
5.6+.81
2.1+.09
2.1+.40
3
6.0+.31
3.1+.23
.8+.02
0
4
1.7+.15
2.1+.10
0
0
5
. 8+.0
0
0
0
6
0
0
0
0
7
0
0
0
0
Animal
Number
21
22
23
24
Weeks
0
0
0
0
0
1
8.7+.83
.7+.15
8.1+.44
5.1+.46
2
7.9+.09
4.5+.18
12.6+.99
.6+.06
3
3.2+.48
6.2+.46
2.7+.23
0
4
1.6+.06
2.1+.05
0
0
5
.1+0
0
0
0
6
0
0
0
0
7
0
0
0
0
Animal
Number
25
26
27
28
Weeks
0
0
0
0
0
1
4.0+.09
13.8+.1.35
5.2+. 29
10.6+.87
2
6.2+.21
9.1+1.37
.7+0
X
3
2.5+.30
3.4+.40
1.0+.16
4.2+.36
4
0
2.1+.11
0
.8+.023
5
0
0
0
0
6
0
0
0
0
7
0
0
0
0
a) Each dog was immunized with DNP-ASC
in adjuvant at
weeks 0,2,
and 6.
b) The
units were
calculated from a
relative antibody
concentration scale derived from the
titration of
a serum sample
containing anti-DNP IgM.
A value of
zero indicates no
detectable
anti-DNP IgG. The
data
was the mean
antibody concentration of a
sample run in triplicate + the standard
deviation
from
the mean.
This was
calculated by adding and subtracting the
standard
deviation to the mean com and calculating the relative antibody
concentration for these numbers. These numbers were then
subtracted from the mean concentration.


50
Table 3
The Relative anti-DNP IgG Concentration in 28 Dogs
Following Immunization with DNP-ASC a)
Animal
Number
1
2
3
4
Weeks
1
0 b)
0
0
0
2
0
0
.8+.02
0
3
4.1+.21
9.1+1.4
2.4+.21
3.7+.11
4
3.7+.20
15.5+.20
8.0+.10
7.4+.66
5
6.0+.64
16.4+1.26
4.8+.26
7.4+.14
6
8.1+.33
26.0+1.33
8.6+.23
4.8+.26
7
7.9+1.05
23.3+.14
6.7+.16
6.1+.88
Animal
Number
5
6
7
8
Weeks
1
0
0
0
0
2
0.3+.02
0
0
0.1+0
3
3.7+.12
9.1+1.21
12.2+.77
25.4+1.31
4
8.6+.17
15.3+.86
11.1+.39
27.0+1.92
5
9.1+.34
17.0+1.41
9.2+.64
30.0+3.21
6
7.4+.23
25.6+.73
9.3+1.07
20.1+.49
7
8.6+.61
24.1+1.72
5.9+.26
8.7+1.26
Animal
Number
9
10
11
12
Weeks
1
0
0
0
0
2
0
0
0
0
3
7.3+.S9
4.3+.05
8.3+.80
6.3+.29
4
15.4+1.71
4.3+.12
12.4+.64
22.5+1.10
5
18.5+.86
5.6+.09
7.9+.32
20.8+.93
6
16.0+.46
7.1+.51
7.5+.93
17.8+1.79
7
8.3+.62
6.8+.19
5.5+.68
10.4+.23
Animal
Number
13
14
15
16
Weeks
1
0
0
0
0
2
0
0
0
.3+.02
3
0
5.3+.31
3.8+.43
6.0+.11
4
10.1+.61
16.7+1.1
8.8+.16
' 4.4+.70
5
9.3+.75
14.3+.42
9.9+.63
11.6+1.60
6
7.8+.26
14.7+.32
8.9+.91
11.1+1.13
7
5.0+.16
10.4+.6
9.7+.46
17.2+.32


79
arbitrarily placed into one of three groups. Group 1 (N=3)
received 100 pg of autologous anti-DNP antibody in CFA by
the subcutaneous route. This antibody was purified from
serum by adsorption to and elution from a DNP coupled
affinity column followed by passage through an ABA coupled
affinity column to ensure that the purified anti-DNP anti
body had no cross reactive anti-ABA antibody. Conversely,
the autologous anti-ABA antibody in CFA for Group 2 (N=3)
was purified from serum by adsorption to and elution from
an ABA affinity column followed by passage through a DNP
column. The control group, Group 3 (N=2) received CFA
without autologous antibody. This later immunization was
administered six weeks after the primary injection of
antigen. Twd weeks after this last injection, serum from
each animal was assayed for anti-idiotypic antibody.
Anti-id RIA
The RIA used to detect anti-id was performed essen
tially as described for the antigen specific RIA using a
number of mouse monoclonal anti-DNP antibodies as the
antigen. These were
a) Anti-DNP IgG, (a gift from Dr. A.P. Lopes, Univer
sity of Pennsylvania) and as a control antibody, anti-H2K
IgG (a gift from Dr. P. Klein, University of Florida).
b) Anti-DNP and IgE as a control antibody, anti-OVA
IgE.


12
There are numerous reports that have shown that the
passive administration of anti-id or the active induction of
anti-id results in the suppression of the corresponding id
(reviewed in 40-46). This modulation acts directly on
B-cells or indirectly through T-cells. For example, in a
B-cell tumor model, Balb/c mice irrcnunized with MOPC 315
myeloma protein produced antibody with specificity for the
id of MOPC 315. Subsequently these mice were injected with
a MOCP 315 bearing plasmacytoma and the tumor growth was
inhibited. It has also been shown that the immunization of
MOPC 315 protein also induces idiotype specific T-suppressor
cells that inhibit the MOPC 315 tumors secretion in vivo
(47). Cosenza and Kohler (48) demonstrated that anti-id can
act as an anti-antigen receptor antibody and specificially
inhibit the induction of a primary immune response. In
other studies by this same group, anti-id, which was
specific for anti-phosyphorylcholine (PC) antibody, signif
icantly inhibited anti-PC plaque forming cells to a degree
similar to the inhibition seen with antigen (49).
These studies show that experimentally, the admin
istration of anti-id or immunization with id to induce
anti-id can result in id suppression. However, if anti-id
regulates id during a normal immune response, auto-anti-id
should be part of the response.


83
Table 11
Screening for Canine IgG Anti-Idiotypic Antibody by RIA
Using Mouse Monoclonal Anti-DNP IgG as the Antigen
Animal
# a) 2
3
4
6
14
Week
0
476+16
331+19
448+7
397+14
422+21
1
453+7
590+1
481+11
421+10
470+47
2
412+21
347+3
274+3
335+6
443+15
3
317+9
277+11
352+9
371+3
367+40
4
305+15
358+3
367+27
401+16
457+12
5
378+31
376+7
358+18
421+11
352+25
6
396+14
417+37
284+10
379+23
318+14
7
421+27
335+10
409+15
522+27
314+24
8
1357+31
1815+26
433+21
2173+68
277+43
9
2140+150
2027+36
1533+117
2250+137
1420+48
10
1862+17
2208+55
2092+130
2297+100
2135+186
11
2174+79
3058+121
2107+41
3375+46
5087+161
Animal #
17
19
20
23
24
Week
0
476+1
367+40
307+22
422+9
483+19
1
453+17
351+26
406+19
470+16
517+42
2
418+5
481+56
464+3
394+7
347+5
3
519+43
435+13
253+13
367+14
318+23
4
442+16
467+24
373+20
334+17
351+1
5
376+18
430+45
481+6
442+26
235+7
6
340+28
398+1
295+30
462+11
434+16
7
315+7
384+93
563+38
371+31
346+34
8
793+13
1937+179
721+37
529+21
471+18
9
973+10
2476+88
936+68
315+8
512+46
10
1611+86
2720+9
2437+177
447+60
341+28
11
1735+122
2925+66
2026+69
473+3
429+27
a) All dogs were immunized with DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10. At weeks 7 and 9, dog 2,3,4 and 6 received 10 pg
autologous anti-DNP antibody in CFA; dogs 14,17,19 and 20 received
100 pg autologous anti-DNP antibody in CFA; dogs 23 and 24 received
CFA alone.
b) This represents the mean + standard deviation of a triplicate
sample. All samples were'assayed at a serum dilution of 1/5 in PBS.
The mean + standard deviation for all samples assayed using mouse
anti-t^K IgG^ was 477+97.


Figure 17.
The identification of anti-id in various serum samples over time,
in dog 14, as measured by RIA. The dog received DNP-ASC in
adjuvant at weeks 0,2,4,6,8 and 10, and autologous anti-DNP
antibody in adjuvant at weeks 7 and 9. The arrow, marked idiotype
probe, indicates the time from which the anti-DNP F(ab)'2 fragments
came. These were used as antigens to detect anti-id. The bars
represent the standard deviation of the mean.


14
erythematosus (54) and in seme IgA-deficient people in terms
of anti-casein antibody and its reciprocal anti-id (55).
In these later experiments the anti-id vas detected using
homologous antibody as the id probe.
These experiments suggest that because anti-id is
present during a normal immune response and regulates the
expression of ids, anti-id may be an important part of the
regulation of the immune response.
In reference to IgE, Geczy and his associates (58) have
shown that in guinea pigs, the administration of synge-
neically derived antibody led to a marked suppression in the
IgE level as measured by passive cutaneous anaphylaxis.
This treatment also resulted in the production of anti-id
and if this anti-id was given to a guinea pig followed by
antigen stimulation, there was a marked suppression in the
subsequent response. This group has shown that in the
mouse, the preexisting anti-hapten IgE and IgG antibody
response could be suppressed with either anti-hapten or
anti-carrier anti-idiotypic antibody (59-61).
These experiments and others like them show that
id/anti-id interaction results usually in suppression of the
immune response. However, this is not always the case. For
example, Eichmann and Rajewsky (62) showed that the
injection of guinea pig IgG^ anti-id would enhance the
expression of id designated A5A when stimulated with


53
Table 5
The Relative Anti-DNP IgE Concentration
In 2 Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M + S.D. b)
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
3.7
3.2
.2
3.3
.1
1.0
1.9 + 1.7
3
10.0
4.8
11.0
4.2
1.0
2.7
5.6 + 4.0
4
3.5
4.6
7.7
4.8
1.0
2.0
4.8 + 3.0
5
10.1
5.1
9.8
5.2
1.6
1.8
5.6 + 3.7
6
3.3
3.3
2.2
5.8
.9
.7
2.7 + 1.9
7
7.0
4.9
6.7
7.1
1.3
1.2
4.7 + 2.8
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
10.8
1.9
1.8
0
2.1
X
3.3 + 4.3
3
11.3
13.4
9.5
1.4
1.7
1.1
6.5 + 5.6
4
9.2
8.0
8.4
.3
1.8
7.3
5.8 + 3.8
5
11.6
12.7
12.0
1.0
1.0
8.8
7.9 + 5.6
6
5.6
16.1
8.5
1.5
.8
3.3
6.0 + 5.7
7
7.7
17.8
19.3
2.3
.9
8.4
9.4 + 7.7
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
DNP-ASC in adjuvant at week 0,2,4 and 6.
b) Mean + Standard Deviation


26


THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
DURING AN IMMUNE RESPONSE IN DOGS
BY
KEVIN T. SCHULTZ
A DISSERTATION PRESENTED TO 'THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1983

This dissertation is dedicated to my wife Nancy.
Without her love and support (and typing), it
would never have been.

acknowledgements
I wish to express my deep appreciation and thanks to
some very important people in my life. First of all, to ray
parents, my wife and ray family for their constant love and
support.
Very special thanks are extended to Dr. Richard
Halliwell for his help, guidance and friendship throughout
my graduate experience.
To Drs. Gail Kunkle, Robert Mason and Bill Hollaran,
my deep appreciation for their unwavering friendship (and
occasional needed prods) along my journey.
I also wish to thank Drs. K.I. Berns, M.D.P. Boyle,
R.B. Crandall, A.P. Gee, G.E. Gifford, M.J.P. Lawman, and
P.A. Small, Jr., for their suggestions, help and
encouragement.
Last (but not least), I would also like to thank my
fellow graduate students and all my friends for making this
experience a memorable one.
ill

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS iii
ABSTRACT vi
CHAPTER ONE INTRODUCTION 1
CHAPTER TWO THE INDUCTION AND KINETICS OF AN
ANTI-DNP IGE RESPONSE
Introduction 17
Materials and Methods 18
Results 31
Discussion 55
Suntnary 57
Conclusions 57
CHAPTER THREE ATTEMPTS TO REGULATE AN ANTIBODY
RESPONSE WITH AUTOLOGOUS ANTIBODY
Introduction 59
Materials and Methods 60
Results 62
Discussion 74
Summary and Conclusions 76
CHAPTER FOUR THE IDENTIFICATION OF ANTI-
IDIOTYPIC ANTIBODY
Introduction 77
Materials and Methods 78
Results 82
Discussion 94
Summary 99
Conclusions 99
CHAPTER FIVE DETECTION OF ANTI-IDIOTYPIC
ANTIBODY USING AUTOLOGOUS ANTI-DNP
F(AB)' FRAGMENTS AS THE IDIOTYPIC
ANTIGEN
Introduction 99
Material and Methods 100
Results 103
Discussion 130
Summary and Conclusions 135
IV

CHAPTER SIX CONCLUSIONS
137
REFERENCES 140
BIOGRAPHICAL SKETCH 149
v

Abstract of Dissertation Presented to the Graduate
Council of the University of Florida
In Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
DURING AN IMMUNE RESPONSE IN THE DOG
By
Kevin T. Schultz
August, 1983
Chairman: Richard E. Halliwell
Major Department: Immunology and Medical Microbiology
This investigation cornmensed with the development of
an animal model to study the synthesis of IgE antibody.
Repeated immunizations with a haptenated parasite extract
(dinitrophenol coupled asearis) in young dogs resulted in
the production of anti-DNP antibody of the IgE, IgG and IgM
class.
Although attempts to regulate this anti-hapten anti
body response by administration of autologous anti-DNP anti
body were unsuccessful, such therapy did result in the pro
duction of anti-idiotypic antibodies. These anti-idiotypic
antibodies were demonstrable using mouse hybridoma-derived
anti-DNP antibodies.
vi

This antibody was shown to be anti-idiotypic rather
than an internal image of antigen because it bound to only
two of four monoclonal anti-DNP antibodies and failed to
inhibit the id anti-id interaction with hapten. Anti-
idiotypic antibodies were detected during the immunization
schedule in three of five dogs using autologous anti-DNP
F(ab)'2 fragments as the source of idiotypes.
The anti-idiotypic antibodies identified using the
mouse monoclonal antibody were the result of the immuni
zation procedure and did not appear to be physiologically
relevant to regulation of the immune response. On the other
hand, the anti-idiotypic antibodies identified with the
autologous source of idiotypes appear to be produced during
tile DNP-ASC immune response and were detected before autol
ogous antibody immunization. The antigens that induced the
anti-idiotypic response appeared to be, in this case, the
idiotypes on the anti-DNP antibody that were produced from
tiie DNP-ASC immunization.
vi 1

CHAPTER ONE
INTRODUCTION
Allergic diseases of the immediate type are very
important pathologic disorders in both man and dogs.
Clinical signs are initiated by an interaction of antigen
and IgE antibody with resultant mediator release from mast
cells and basophils. In man and dogs, allergic reactions
cause considerable morbidity and can be fatal (1,2).
Anaphylaxis from a bee sting is a classic example for both
species.
Atopy in man is an inherited disease which is
associated with an antigen specific IgE response against
environmental allergens. This disease is expressed
clinically as asthma, hay fever, atopic dermatitis or any
combination of these three (3). The dog is an excellent
experimental animal model to study IgE mediated hyper
sensitivity for atopic diseases of man because of the
similarity of the allergic reaction in both species (1,2,4).
Canine IgE shares many physicochemical properties with
hunan IgE (5,6,7). Dogs like man, develop spontaneous
disease associated with increased synthesis of IgE antibody
(3,4) and in dogs the disease is also familial (4).
1

2
There are a number of unique features of IgE antibody-
synthesis. Firstly, IgE circulates in very small amounts as
compared to other antibody classes. In man, the serum level
of this antibody is about 1/10,000 the level of serum IgG
(3), and in dogs serum IgE is about 1/100 the level of serum
IgG (7). Serum IgE levels of internally parasitized people
and dogs are elevated as compared to non-parasitized indivi
duals (7,8,9). 'The higher IgE level in dogs is felt to be
the result of a greater parasite burden in this species (7).
Secondly, IgE is produced predominantly locally by
lymph nodes in the respiratory and gastrointestinal tracts
as well as in regional lymph nodes (10). These observa
tions have led to the suggestion that this immunoglobulin is
important in host defense of mucosal surfaces and partic
ularly against parasites. Furthermore, IgE has been shown
to participate in parasite killing through antibody depen
dent cell mediated cytotoxicity (11). Thirdly, the antigens
that stimulate IgE antibody are usually very complex and
heterogenous substances such as allergens or parasites and
their extracts. When an animal is exposed to these anti
gens, the antibody response usually includes high titer IgE
antibody whereas bacteria and viruses usually do not induce
IgE antibody in spite of being very immunogenic (12).

3
The induction of IgE antibody experimentally requires
special conditions. For example, high doses of antigen and
strong adjuvants such as complete Freund's adjuvant are
unfavorable to the development of an IgE response whereas
low doses of antigen in an adjuvant such as aluminum
hydroxide tend to favor IgE production (13). Furthermore,
if haptens are coupled to parasite extracts, high titer
anti-hapten IgE antibody responses will frequently develop.
However, if the same hapten is coupled to a T-independent
antigen or a different T-dependent carrier there is usually
no IgE response (13,14). 'This suggests that IgE production
is dependent on both the carrier and T-cells. The reason
why parasites and their extracts are efficient inducers of
IgE antibody is not fully understood. It is known that this
enhancing effect is modulated through factors produced by T
cells. Ishizaka's group (15) have shown that T cells
derived from (N. brasiliensis) parasitized rats produce
an IgE-potentiating factor which selectively potentiates a
non-specific IgE antibody response. This factor has
affinity for IgE, binds to IgE-bearing B cells through
surface IgE and enhances the differentiation of these cells
into IgE forming cells. A factor with similar properties
has been produced from T-cells obtained from patients with
hyper-IgE syndrome, suggesting that the regulatory factors

4
and pathways for enhancing IgE antibody production for
parasites and IgE in general might be similar (16).
A number of approaches have been used in an attempt to
control allergic disease. Avoidance of the antigen is one
approach, but this is rarely possible. Drugs that inhibit
mediator release or control the effects of that release are
also employed, but they have side effects and often require
continual therapy. However, the ideal approach would be to
regulate the production of the unwanted IgE antibody.
The mechanisms used to regulate IgE antibody responses
involve either the inactivation of B cell precursors or the
manipulation of T cell populations. To this end, primary
and ongoing antibody responses, including IgE antibody, have
been suppressed in mice by antigen coupled to non-immuno
genic carriers such as d-glutamine-d-lysine (dGL) or poly
vinyl alcohol (17,18). Both of these carriers inactivate
hapten-specific B cells and can induce hapten-specific
suppressor T cells. Moreover, when dGL is coupled to
proteins rather than haptens, the resultant suppression in
mice is isotype specific (i.e. suppresses IgE alone) (19).
Unfortunately, there are no published results of the use of
this compound in dogs or man.
Hyposensitization has also teen used in an effort to
control allergies in both man and dogs (20-22). The
mechanism by which it works is not clear. It is known that

5
IgG antibody can have a role in regulating allergic symptoms
(22). An IgG response can be induced by administration of
allergen either by the normal route of exposure or by a
route other than for normal exposure (i.e. subcutaneous
versus inhalation). This IgG antibody presumably completes
the allergen-IgE antibody interactions (21). However, this
therapy is not without side effects (20). Furthermore, only
about 65 per cent of patients treated with hyposensitization
have clinical improvement (20).
An alternate type of hyposensitization involves
modifying the allergen, usually by mild denaturation.
Studies in mice with urea denatured ragweed showed that such
treatment reduced allergenicity while maintaining immuno-
genicity of the allergen. If large doses of urea denatured
ragweed were given to mice previously sensitized to
unmodified ragweed, such therapy resulted in antigen-
specific T suppressor cell induction without the development
of anaphylaxis (12). These cells suppressed the anti
ragweed IgE response. A controlled study is underway to
determine if this form of immunotherapy is any more
effective in controlling allergic symptoms than conventional
hyposensitization.
Another approach is to regulate the response with
products of the immune system. Smith (23), in 1909, was the
first person to recognize that antibody could suppress the

6
development of an immune response. In these experiments he
showed that certain mixtures of diphtheria toxin and anti
toxin oould be very immunogenic in guinea pigs, but if there
was a large excess of antitoxin, the immunized guinea pig
would fail to mount an immune response against the toxin.
Numerous studies in the 1950's and 1960's verified this
observation and also demonstrated that the isotype, amount,
affinity and time of administration were important variables
in determining the degree of suppression float passive anti
body had on the immune response (reviewed in 24). For
example, IgG antibody given after antigenic exposure was
more effective in inducing antibody suppression than IgiM
antibody. Further, the suppressed state was longer lived
using IgG than IgM antibody. An interesting report by Chan
and Sinclair (25) stated that the administration of anti-
SR3C antibody given to mice after antigenic challenge led to
a suppression of this response and this tolerant state could
be transferred from one mouse to another with T-cells from
the tolerized mouse. They suggested that the regulatory
action of antibody operated through sane sort of "induced
pathway or secondary immune response" (25 p. 977).
In the early 1970's it was likewise shown that IgE
antibody could be regulated by passively administered
antibody (26-28). Rabbits were immunized to produce high
titer IgE antibody and were given passive antibody 24 hours

7
after antigenic challenge. A complete inhibition of the
passive cutaneous anaphylaxis titer and a narked decrease in
the hemagglutination titer of these rabbits resulted as
compared to controls (26). It vas shown by- Tada and Okumura
(27) that, in the rat, the administration of anti-DNP
ascaris antibody resulted in marked suppression of a
preexisting IgE antibody response and this suppression was
maintained for an extended period of time. This was in
contrast to studies in the mouse in which administration of
anti-ovalbumin IgG had little effect on the preexisting
anti-ovalbumin IgE response (28). These differences were
explained as species variation. Alternatively, they may be
due to the difference in the antigenic system employed.
One explanation for the mechanism of regulation by
passive antibody is that the administration of this antibody
acted as an antigen and stimulated an anti-antibody
response. Lahss et al. (29) were the first to show that
sane anti-antibodies would bind to structures on antibody
close to or within the antigen combining site. These deter
minants have been named idiotypes (id) and the immune
response directed to them is termed an anti-idiotypic
(anti-id) response. In 1974, Jerne (30) proposed his network
hypothesis of antibody regulation. The basic premise of
this theory is that tine immune system is regulated by a
network of interactions between id and anti-id. A number of

8
assumptions are crucial premises to this theory. Firstly,
most idiotypes exist at a level too low to induce tolerance.
Thus, antigenic stimulation and expansion of these id will
stimulate the production of a reciprocal set of anti-id.
The id is then regulated directly by the anti-id, indirectly
by the anti-id on T-cellsor or by anti-id acting on T-cells.
As the concentration of anti-id reaches sane critical
threshold, a second anti-id response develops which is
specific for the id of the anti-id. This anti-id would,
therefore, be an anti-(anti-id) and would then stimulate a
fourth response and so on, thereby resulting in an inter
related network of regulation between antibody molecules.
Jerne also stated that id determinants can be present not
only on antibody molecules of one specificity, but may be
present on unrelated antibody molecules. Thus, antibody
against antigen x might share some ids with antibody against
antigen y. Lastly, although anti-id usually suppresses the
corresponding id, it can be stimulating for the id as well.
The anti-id would be expected to have a three dimensional
structure similar or identical to the specific antigenic
determinant. This type of anti-id is termed an internal
image of antigen.
.The-characteristics of idiotypes of antibody molecules
have been described (31-35). In many instances, idiotypes
are located in or very near to the antigen binding site.

9
This has been demonstrated by hapten inhibition studies.
Brient and Nisonoff (31) induced anti-p-azobenzoate anti
bodies in rabbits. These antibodies were purified and
injected into allotypically matched rabbits and the resul
tant antiserum bound to determinants present on some rabbit
anti-p-azobenzoate antibodies. They then studied the
effects that adding increasing concentration of hapten would
have on the reaction between radiolabelled anti-azobenzoate
antibodies and the anti-idiotypic antiserum. They found
that the binding affinities of the benzoate derivatives
correlated closely with their ability to inhibit the
antibody/anti-id interaction. In many other studies
(32-34), anti-id was induced in animals immunized with an
anti-hapten antibody. This anti-id was purified from the
sera by initial adsorption to an affinity column having the
immunizing antibody bound to it and was then eluted with the
appropriate hapten. This purification process then would
select for anti-idiotypic antibodies which were directed to
those idiotypic determinants very close to or within the
antigen binding site and it would be expected that hapten
could inhibit the id/anti-id interaction.
On the other hand, it is not always possible for hapten
to inhibit id/anti-id. For example, Sher and Cohn (35)
showed that there was variation in the ability of hapten to
inhibit id/anti-id interaction. Hapten was not able to

10
inhibit the interaction by 100 percent, maximum inhibition
was only 68 percent (35). The most extreme example in which
hapten cannot inhibit id/anti-id interactions are in those
studies in which cross reactive ids are present on antibody
molecules of widely different specificity. For example,
Eichmann et al. (36) showed that one half of the A5A id
producing clones in A/J mice immunized with a streptococcal
carbohydrate lacked the ability to bind this antigen.
Obviously then, antigen would not be expected to inhibit
this id-anti-id interaction. In other studies, Bona et al.
(37) showed that not all the id positive antibody following
immunization with inulin could be removed with an inulin
immunoabsorbent. In these experiments, the anti-inulin anti
body produced following antigenic stimulation bears a
predominant id. However, some immunoglobulin following
antigenic stimulation had this id but lacked specificity for
inulin. These experiments therefore suggest that some
mechanism exists naturally in which id positive clones of
immunoglobulin producing cells are expanded following
antigen stimulation but that not all the id positive immuno
globulin is specific for the immunizing antigen. These
experiments clearly show that although id/anti-id can
usually be hapten inhibited, this property is not a require
ment for an antibody to be anti-idiotypic.

11
Identical ids have been found irrespective of the
isotype of the antibody. The mechanism by Which IgE and IgG
antibody can have identical idiotypes relates to the gene
rearrangement that occurs during differential expression of
heavy chain genes (38). As a single clone of cells goes
through isotypic shift, a single variable region of the
genes which includes the idiotype, will become linked to
various heavy chain gene fragments. A single cell will
differentiate into plasma cells which express different
heavy chain genes but the same variable gene sequence (39).
Therefore, it is possible for ids to be shared between anti
bodies of the sane binding ability irrespective of the
isotype. This implies that regulation of IgG antibody by
anti-id networks may also result in IgE antibody regulation.
Idiotypic determinants are usually defined sero
logically. There are a number of different ways to produce
anti-id (reviewed in 40-42). Anti-id can be produced across
tine species barrier, within the same species, within the
same strain, or more importantly, even within the same
individual that produced the id. Anti-id have been used to
determine if the id of the antibody molecule may have a
function other than to bind antigen. This has been done by
examining yiiat functional significance the presence of
anti-id had on the corresponding id in vivo.

12
There are numerous reports that have shown that the
passive administration of anti-id or the active induction of
anti-id results in the suppression of the corresponding id
(reviewed in 40-46). This modulation acts directly on
B-cells or indirectly through T-cells. For example, in a
B-cell tumor model, Balb/c mice irrcnunized with MOPC 315
myeloma protein produced antibody with specificity for the
id of MOPC 315. Subsequently these mice were injected with
a MOCP 315 bearing plasmacytoma and the tumor growth was
inhibited. It has also been shown that the immunization of
MOPC 315 protein also induces idiotype specific T-suppressor
cells that inhibit the MOPC 315 tumors secretion in vivo
(47). Cosenza and Kohler (48) demonstrated that anti-id can
act as an anti-antigen receptor antibody and specificially
inhibit the induction of a primary immune response. In
other studies by this same group, anti-id, which was
specific for anti-phosyphorylcholine (PC) antibody, signif
icantly inhibited anti-PC plaque forming cells to a degree
similar to the inhibition seen with antigen (49).
These studies show that experimentally, the admin
istration of anti-id or immunization with id to induce
anti-id can result in id suppression. However, if anti-id
regulates id during a normal immune response, auto-anti-id
should be part of the response.

13
A number of studies have shown the presence of auto-
anti-id during a normal immune response to an antigen
(50-57). Bankert and Pressman (50) showed that an antibody
with auto-anti-id activity could be detected in rabbits
during primary and secondary immune response to both sheep
red blood cells and to the hapten, 3-iodo-4-hydroxy-5-
nitrophenyl-acetic acid. Kelsoe and Cerny (51) have demon
strated a reciprocal expansion of antigen activated idiotype
bearing clones of lymphocytes followed by expansion of
clones which bear anti-id receptors in Balb/c mice immunized
to Streptococcus pneumonia. They hypothesized that the
out of phase expansion of the reciprocal cell sets was the
result of interactions of id and anti-id. The production of
auto-anti-id in man has been demonstrated to occur during
tiie immune response against tetanus toxoid. The presence of
this anti-id was associated with the loss of some of the
anti-tetanus toxoid idiotypes (52). Naturally occurring
anti-id has also been demonstrated in myasthenia gravis
patients using, as the idiotype probe, a mouse monoclonal
antibody. Those patients with the highest titer of anti
receptor antibody had the lowest level of anti-id, while in
patients with the lowest titer of anti-receptor antibody
(id), the highest titer of anti-id was detected (53).
Comparable findings have been reported in patients with
anti-DNA. antibody and reciprocal anti-id in systemic lupus

14
erythematosus (54) and in seme IgA-deficient people in terms
of anti-casein antibody and its reciprocal anti-id (55).
In these later experiments the anti-id vas detected using
homologous antibody as the id probe.
These experiments suggest that because anti-id is
present during a normal immune response and regulates the
expression of ids, anti-id may be an important part of the
regulation of the immune response.
In reference to IgE, Geczy and his associates (58) have
shown that in guinea pigs, the administration of synge-
neically derived antibody led to a marked suppression in the
IgE level as measured by passive cutaneous anaphylaxis.
This treatment also resulted in the production of anti-id
and if this anti-id was given to a guinea pig followed by
antigen stimulation, there was a marked suppression in the
subsequent response. This group has shown that in the
mouse, the preexisting anti-hapten IgE and IgG antibody
response could be suppressed with either anti-hapten or
anti-carrier anti-idiotypic antibody (59-61).
These experiments and others like them show that
id/anti-id interaction results usually in suppression of the
immune response. However, this is not always the case. For
example, Eichmann and Rajewsky (62) showed that the
injection of guinea pig IgG^ anti-id would enhance the
expression of id designated A5A when stimulated with

15
Streptococcus whereas if the anti-id was an IgC^' the
expression of A5A id was suppressed. Recently, Forni et al.
(63) showed that the injection of anti-SRBC IgM into normal
mice induced plaque forming cells of the same specificity as
the injected antibody. Further analysis established that the
mechanism of this enhanced responsiveness was based on
id/anti-id interactions (63,64). The authors state that
"these results support network concepts. Thus if an antigen
specific response can be induced solely by using components
of the immune system itself, it follows that, in its basic
economy, this system is autonomous and does not depend on
the introduction of antigen to adjust to new dynamic states"
(63 p. 1127). In this case anti-id most probably acted as
an internal image of antigen. There have been other
examples that demonstrated the mimicry of antigen by anti
body. For example, Sege and Peterson (65) showed that
anti-id prepared against antibody to insulin could mimic the
action of insulin in cells. Schreiber et al. (66) showed
that anti-id against rabbit antibodies to alprenolol would
compete with alprenolol for the binding site on turkey red
blood cells. This anti-id could also stimulate adenylate
cyclase activity in the cells.
This discussion raises the possibility that the admin
istration of autologous antibody might regulate antigen
specific IgE response in the dog through id/anti-id

16
networks. Therefore the objectives of the work presented
here were
1) To develop a consistent IgE antibody response in
the dog and to study the kinetics of this response.
2) To examine the effects that autologous antibody
administration had on an ongoing IgE response.
3) To determine if an anti-id response occured at any
point during the experiment and if so, to examine the
relationship between ids and anti-ids.

CHAPrER 'EvO
THE I INDUCTION AND KINETICS
OF AN ANTI-DNP IGE RESPONSE
Introduction
The value of the dog as an experimental model to
study atopy has been described. However, the expense and
difficulty of obtaining atopic dogs necessitated the
development of a system in which antigen-specific IgE could
be consistently induced. The use of a hapten-coupled
carrier as an antigen was felt to be more convenient than a
more complex, heterogenous substance such as an allergen to
study the synthesis and regulation of IgE antibody.
Furthermore, Halliwell (7) and Schwartzman et al. (67) have
shown that two dogs immunized with dinitrophenol coupled to
ascaris antigen and administered in aluminum hydroxide as
tlie adjuvant, developed anti-DNP IgE antibody. However, it
is not known a) if all dogs so immunized produce IgE anti
body, b) how long the detectable IgE response remains, and
c) what the immune response in terms of other isotypes
might be. The purpose of the following experiments, then,
was to induce a consistent anti-hapten IgE antibody
17

18
response and to examine the kinetics of the IgE, IgG and
IgM anti-hapten antibody response.
Materials and Methods
Protein Concentration Determination
The concentration of immunoglobulin was determined
from known molar extinction coefficients and by its ability
to absorb light at 280 nm. Alternatively, the protein
concentration was determined at 595 nm using Bradford's
reagent (68) and interpolated from a standard curve derived
from the absorption values of a series of dilutions of a
similar freeze dried purified protein of known concen
tration. The measurements with both techniques gave concor
dant results.
Antigens
Azobenzenarsonate coupled to keyhole limpet hemo-
cyanin (ABA-KLH) was a gift from Dr. Mark Greene, Harvard
University. Ascaris antigen vas prepared from adult
Toxocara canis by the method of Strejan and Campbell
(69) and modified as follows: Fifty adult T. canis
were obtained from the gastrointestinal tract of euthanized
dogs. The worms were washed with phosphate buffered saline
(PBS), pH 7.2, containing 0.02 percent sodium azide, ground
with a mortar and pestle and incubated for 48 hours at 4
C. Large particulate natter was removed by centrifugation

19
at 1000 x g for ten minutes in an IEC centra-7R centrifuge
(International Equipment Co.) The supernatant was then
centrifuged at 49000 x g for one hour in an L8-70 ultra
centrifuge (Beckman Instrument Co., Norcross, Ga.) to
remove fine particles and was then chromatographed through
a Sephadex G-100 column (Pharmacia Fine Chemicals,
Piscataway, N.J.). The first peak was pooled, concentrated
by negative pressure dialysis, dialyzed against PBS, pH
7.2, passed through a filter having 0.2 micron pores filter
(Acrodisc, Gelman Co., Ann Arbor, Mi.) and used as the
ascaris antigen (ASC). Human serum albumin (HSA) fraction
V was obtained from Sigma Chemical Co. (St. Louis, Mo.).
Bovine gamma globulin (BGG) was prepared from serum of an
adult cow by precipitation with 40 percent saturated
ammonium sulfate. The precipitate was dialyzed against
0.035 M phosphate buffer, pH 8.0 and was then chromato
graphed through a diethylaminoethyl cellulose (DEAE) ion
exchange column (DEA, DE52, Whatman Chemicals, Kent,
England) equilibrated with this same buffer. The effluent
protein was concentrated by negative pressure dialysis and
dialyzed against PBS, pH 7.2.
Dinitrophenylation of Proteins
Dinitrophenylation of protein was performed by mixing
equal weights of protein, potassium carbonate (Fisher
Scientific Co., St. Louis, Mo.) and

2D
2,4-dinitrobenzenesulphonic acid (DNP) (Eastman Kodak Co.,
Rochester, N.Y.) were mixed in distilled water (70). This
was then incubated while gently stirring for 18 hours at
room temperature. The solution was chromatographed through
a Sephadex G-25 column (Pharmacia Fine Chemicals,
Pi seataway, N.J.) to separate bound from free DNP. The
dinitrophenylated protein was concentrated by negative
pressure dialysis and extensively dialyzed against PBS, pH
7.2. The extent of substitution was estimated by measuring
light adsorption at 360 nm and assuming a molar extinction
coefficient oE 1.75 x 10^ for the dinitrophenyl group.
The average epitope density expressed as molecules of DNP
per molecule carrier was DNP^-HSA, DNP14 g-BGG. Since
ASC extract vas a complex mixture of proteins, the extent
of substitution was expressed as moles DNP/mg ASC and was
6.32 x 10 ^ DNP/ASC. A single batch of each of these
antigens was prepared and used throughout the experiment.
These antigens, when not in use, were stored at -70C.
The degree of substitution did not change due to storage.
Aluminum hydroxide Precipitation of Protein
Aluminum hydroxide precipitation of protein was
performed by mixing one part of a 5 percent sterile
solution of aluminum potassium sulfate (AIK (30^)^),
Mallincrodt, Paris, Kentucky) with five parts of 1 mg/ml

2L
solution of protein (70). The pH vas then adjusted with 0.1
N NaOH to pH 6.3 to ensure adequate precipitate.
Affinity Chromatography
Sepharose 4B beads (Pharmacia Fine Chemicals) were
activated using cyanogen bromide (CnBr) by adding 1.5 grams
CnBr in 20 ml distilled vater to 10 ml of washed Sepharose
4B beads and adjusted to pH 11 with I N NaOH. This mixture
was maintained on ice at pH 11 for 6 minutes after which
the beads vare vashed with 100 volumes of iced oold water.
Ninety milligrams of protein in 6 ml PBS, pH 7.2 were added
and incubated for 12 hours at 4C. Alternatively pre
activated Sepharose 4B beads were obtained (Pharmacia Fine
Chemicals) and protein was bound to these beads as
described by the manufacturer. To remove unbound protein
in both cases, the beads were washed with five alternate
cycles of 0.1 M Tris buffer, pH 8.3 containing 0.5 M NaCl
followed by 0.1 M glycine HC1, pH 2.8. Any retaining sites
were blocked by incubating the beads in 0.1 M Tris buffer,
pH 8.3 for four hours at room temperature. The column was
then flushed with normal canine serum and washed as
described above.

22
Pepsin Digestion and Purification of F(ab)'^ Antibody
Fragments.
The usual procedure for F(ab)'^ digestion of immuno
globulin vas to digest the antibody with 6 percent pepsin
(w/v) in 0.2 M acetate buffer, pH 4.5 for 13 hours at 37
C. However, this process resulted in some loss of
antigen binding of the F(ab)'2 presumably from the
prolonged incubation time at pH 4.5. Where maintenance of
this activity vas critical, protein vas digested with 20
percent pepsin w/v in 0.2 M acetate buffer pH 4.5 for five
hours at 37C. The digested protein vas separated from
Fc pieces and intact antibody by passage through a cyanogen
bromide-activated heavy chain specific immunoabsorbent
column followed by passage through a Staphylococcus
protein A affinity column (Pharmacia Fine Chemicals). The
effluent vas concentrated by negative pressure dialysis and
dialyzed against PBS, pH 7.2.
Antisera
a) Preparation and purification of anti-IgG. Normal
canine serum (NCS) was precipitated with a 40 percent
saturated solution of ammonium sulfate. The precipitate
was dialyzed against 0.035 M phosphate buffer, pH 8.0 and
applied to a DEAE ion exchange column equilibrated with
this same buffer. The effluent protein was concentrated by

23
negative pressure dialysis. One milligram of this material
was emulsified in complete Freund's adjuvant (CFA) and
administered intramuscularly to rabbits at two week
intervals four times. Fifty milliliters of blood were
obtained from the rabbit by ear vein venapuncture every two
weeks starting after the second immunization. All serum
which gave visible precipitation reactions by agar-gel
diffusion against canine IgG was pooled. This antisera
was passed through a cyanogen bromide-activated sepharose
4B F(ab)'2 affinity column, to remove light chain
activity, followed by adsorption to and elution with
alycine HC1 (Osl M), pH 2.8 from a canine IgG bound
affinity column. This anti-IgG detected three subclasses
of canine IgG (IgG^ IgG2ab, IgG2jJ but no other
protein as measured in an immunoelectrophoresis (70) of NCS
(figure 1). To determine if this antiserum detected IgE,
the antiserum was radiolabelled and used in a radio
immunoassay. The serum sample tested contained both anti-
DNP IgG and anti-DNP IgE. Therefore, an aliquot of this
serum was heat inactivated and the level of anti-DNP IgG
was compared in this aliquot to a second aliquot of this
serum that was not heat inactivated. Additionally, anti
canine IgE was added to an aliquot of this sample to
determine if this unlabelled anti-IgE might compete with
the anti-IgG for Fc binding sites. Heating serum for four
hours at 56C destroys the heavy chain antigenic

2
determinants of canine IgE (7). The level of anti-DNP
antibody increased both when the serum was inactivated and
when non-labelled anti-IgE antiserum was added to the
sample. This indicates that this anti-IgG antiserum has
minimal, if any, anti-IgE activity.
b) Preparation and purification of anti-IgE. A 40 per
cent saturated ammonium sulfate precipitate of serum
obtained, from a dog that was heavily parasitized and
presumed to have high levels of IgE, was dialyzed against
0.035 M phosphate buffer pH 8.0 and applied to a DEAE
cellulose column equilibrated with this same buffer. The
effluent protein was concentrated by negative pressure
dialysis and applied to a set of three in series Sephacryl
S-200 columns (Pharmacia Fine Chemicals, Piscataway, N.J.).
The first one-third of the second protein peak, which was
the IgE-rich fraction as determined by agar-gel immuno-
precipitation, was collected, concentrated by pressure
dialysis and reapplied to these columns. The resulting
IgE-rich fraction was collected and used to immunize
rabbits as described previously. The rabbits were bled as
described above. Serum that produced visible precipitation
lines against the immunizing antigen in an agar gel immuno
diffusion were pooled. The resulting antiserum detected
both IgE and IgG by immunoelectrophoresis. It was rendered
specific for the former protein by passage through an

Figure 1.
The specificity of anti-canine IgG as assayed in an
immunoelectorphoresis against normal canine serum.
Figure 2.
The specificity of anti-canine IgE as assayed in an
immunoelectrophoresis against normal canine serum
(bottom well) and this same serum after heat
inactivation (top well).
Figure 3.
The specificity of anti-canine IgM as assayed in an
immunoelectrophoresis against normal canine serum.
The anti-canine IgM in the bottom through is before
adsorption with the supernatant of a 50 percent
saturated ammonium sulfate precipitate of normal
canine serum. The top trough has the anti-canine IgM
antiserum after this treatment.

26

27
affinity column made with the heat inactivated immunogen
which removed all antibody except anti-IgE antibody.
Purified antibody was then prepared by adsorption to and
elution from an IgE-rich affinity column. This purified
antiserum detected a single heat-labile protein by immuno-
electrophoresis (figure 2), produced reverse cutaneous
anaphylaxis in dogs at a high dilution of serum (10 )
and was unable to detect canine anti-DNP IgG in a RIA
indicating that it had no specificity for this antibody.
c) Preparation and purification of anti-IgM. Canine IgM
myeloma serum, which contained approximately 58 mg/ml IgM
was chromatographed on Sepnacryl 3-200 and the void volume
was collected to obtain IgM. Two milligrams of this
material was emulsified in CFA and injected intra
muscularly at four sites into sheep. This was repeated at
two week intervals five times. Five hundred milliliters of
blood were collected by jugular vein venapuncture every two
weeks. Sera that produced precipitation lines against the
immunizing antigen, in an agar-gel diffusion against the
immunizing antigen, were pooled. Light chain activity was
removed from the antiserum by passage through a canine IgG
affinity column. Antibody was then purified by adsorption
to and elution from an IgM affinity column. The eluted
proteins produced two bands on Immunoelectrophoresis of
NCS, one of which was IgM and the other an unknown protein.

28
This second activity was removed by adsorption with the
supernatant of a 50 percent saturated ammonium sulfate
precipitation of ISJCS (figure 3). This antiserum vas
assayed for anti-DNP IgE and IgG activity by RIA. Serum
that was used contained both of these antibody isotypes.
No antibody vas detected indicating the antiserum did not
have activity for IgE or IgG.
Isotope Labelling of Protein
Two methods vare used to label proteins with radio
active iodine. In the first method, between 1 and 2 mg of
protein in 0.1 ml PBS pH 7.0 without azide and .5 mCi
125
I (Amersham, Chicago, II.) was incubated on ice with
15 pi KI(0.1 mM) and 30 pi chloramine T (10 mM) for 15
minutes. After this incubation, 25 pi sodium metabi
sulphate (10 mM) and 50 pi KH100 mM) were added to stop
tiie reaction. To separate bound and free iodine, the mat
erial vas chromatographed through a G-25 Sephadex column.
The first peak containing radiolabel was pooled, concen
trated and dialysed against PBS, pH 7.2. Alternatively,
p
one lodobead (Pierce Chemical Co., Rockford, II.) was
added to 100 pg of protein in PBS, pH 7.0 and 0.5 mCi
125
I. After a fifteen minute incubation, the bound and
195
unbound I vas separated as described earlier. The
specific activity of the radiolabelled antibody was usually
about 300 pCi/mg protein (range 212-496).

29
Radioimmunoassay for the Detection of DNP Specific
Antibody
Microtiter wells (Iirmulon Ramov-a-well Strips ,
Dynateck, Richmond, Va.) were coated with 50 pi of 20
pg/ml dinitrophenylated bovine gamma globulin (DNP-BGG) in
Tris buffer, pH 8 (0.1 M Tris, 0.15 M NaCl). After incu
bating for 12 hours at 4C the wells were then washed
three times with this buffer. Any remaining sites were
blocked with 2.0 percent HSA in PBS containing 0.5 percent
Tween 20 for three hours at room temperature. Phosphate
buffered saline, pH 7.2 containing 0.5 percent Tween 20 and
2.0 percent HSA is referred to as RAST+ and this same
buffer without HSA is called RAST-. Serum samples were
diluted in PBS, pH 7.2 added to the appropriate wells,
incubated for three hours at 4C followed by five washes
with RAST-. Approximately 50,000 counts per minute (cpm)
of radiolabelled antiserum in RAST- was added to the well,
incubated for three hours at 4C and washed three times
with RAST-. The radioactivity associated with each well
was determined in a Searle-Packard gamma counter (Chicago,
II.). Each sample was assayed in triplicate and each
sample was counted for one minute. The maximum number of
cpm bound was about 20% of the amount added. The back
ground activity was determined by including in each assay
the following controls: 1) A set of triplicate wells in
which BGG rather than DNP-BGG was used as the antigen, 2) A

30
triplicate set of wells in which PBS rather than serum was
added. The mean cpm from these controls were subtracted
from the cpm of the test sample. Although the values
varied from experiment to experiment, the maximum cpm of
these controls were consistently lower than the lowest
values obtained for test samples.
A standard serum sample was included with each
assay as an internal reference. An arbitrary antibody
concentration was determined by assigning a value of 64
units to the undiluted standard IgG and IgE sample and 32
units to the undilute IgM standard. By interpolating from
the linear portion of the standard curve, the relative
units of antibody for test samples were calculated.
Animals and Immunization Schedule
Outbred pregnant female dogs were obtained from the
Division of Animal Resources, University of Florida. Serum
from these dogs was screened by RIA to ensure that they did
not have anti-DNP antibody at the time of whelping.
The puppies of these bitches were used as experimental
animals. Serum samples were obtained on the day of birth
and weekly therafter. Each puppy received 100 gg aluminum
hydroxide precipitated dinitrophenol coupled ascaris
antigen by the intraperitbheal route on the day of birth
and at two week intervals on three further occasions. Each

dog received a distemper-hepatitis modified live virus
vaccination at week four and eight.
31
Results
Standard Curve
The relative antigen-specific antibody concentration
was determined by interpolation from the linear portion of
the standard curve included with each assay. This serum
sample contained high levels of the isotype under inves
tigation. An example of a standard curve for anti-DNP IgE,
IgG and IgM is given in figures 4, 5 and 6.
Antibody Response
Twenty-eight dogs immunized with 100 pg DNP-ASC in
aluminum hydroxide developed an IgE, IgG and IgM serum anti
body response. The mean relative antibody concentration
for the three isotypes is depicted in figure 7, 8 and 9.
The IgM response usually was highest in samples taken seven
days after the first injection of antigen. However, as
seen in table 1, eight of the dogs (5,7,8,12,14,16,23,25)
had anti-DNP IgM concentrations that were greatest in
samples obtained at two weeks and three dogs (9,17,22)
after three weeks. Four weeks after the first antigenic
challenge, five dogs had no detectable IgM antibody and

Figure 4.
Dilutions of tlie standard anti-DNP IgE serum sample assayed by
RIA. The bars represent the standard deviation of the mean.

o
o
35
30
25
20
Cl 15
O
10
5
5 10 20 40
Reciprocal of
80 160
Dilution
OJ
U>

Figure 5.
Dilutions of the standard anti-DNP IgG serum sample assayed by RIA.
The bars represent the standard deviation of the mean.

8
o
o
o
E
Q_
O
5 10 20 40
Reciprocal
80 160 320640 1280
of Dilution

Figure 6.
Dilutions of the standard anti-DNP IgM serum sample
assayed by RIA. The bars represent the standard
deviation of the mean.

37
Reciprocal of Dilution

38
after six weeks the levels of IgM antibody fell to back
ground despite maintenance of the immunizing protocol.
The anti-DNP IgE and IgG antibody responses followed
similar kinetics to each other. There was an initial lag
of two weeks before antibody of these classes was detected
(figures 8 and 9). At the time of the second immunization
(two weeks after the primary immunization), there was a
sharp rise in the antibody levels which continued for one
additional week. Thereafter, the antibody concentration
was maintained at that level or started to gradually
decline. As was the case in the IgM antibody response,
sane dogs deviated from the general trend. Two dogs had
detectable IgE antibody levels one week after primary
immunization (Table 2) whereas three dogs failed to
develop a detectable response until after the third week
and, in the case of one dog, IgE antibody was not detected
until the fifth week iron primary immunization. The IgE
antibody response persisted through the seven week course
of the experiment in all dogs. Anti-DNP IgG antibody vas
detected in nine dogs one week after primary immunization
(table 3) and by the fourth week, all dogs had an IgG anti
body response. Detectable IgG persisted throughout the
immunization schedule but there vas a gradual decline in
IgG antibody levels towards the end of the immunizing
schedule (See figure 9, table 3).

Figure 7.
The mean relative anti-DNP IgM concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC (see text for further details). The antibody
concentration was calculated from an arbitrary antibody
concentration scale derived from the titration of a standard
anti-DNP IgM containing serum.

o
8
Weeks
Mean S.D.
(range)
0 0
(0-0)
1 6.1 4.3
(0-13.8)
2 5.1 3.3
(.7-10.6)
3 23 1.9
(0-42)
4 8 1.0
(0-3.9)
5 .03 .1
(0-.8)
6 0
(0-0)
7 0
(0-0)

Figure 8.
The mean relative anti-DNP IgE concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC. The antibody concentration was calculated
from an arbitrary antibody concentration scale derived from the
titration of a standard anti-DNP IgE containing serum. (See text
for further details.)

12
MeanS.D
(range)
0
0
(0-0)
1
.1 .4
(0*1.7)
2
26 2.9
(0-10.8)
3
6.1 5.0
(0-18.2)
4
5.2 3
(0-92)
5
6.9 5.9
(.3-26.8)
6
4.3 3.6
(.6-16.1)
7
6.7 5.4
(.8-19.3)

Figure 9.
The mean relative anti-DNP IgG concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC. The antibody concentration was calculated
from an arbitrary antibody concentration scale derived from the
titration of a standard anti-DNP IgG containing serum (see text for
further details).

o
2
4 6
Weeks
MeanS.D. (range)
O O (O-O)
1 .3 .6 (0-2.6)
2 7 O 5.0 (0-25.4)
3 13 2 6.5 (37-270)
4 13.5 7.2 (56-30.0)
5 13.6 6.8 (4.8-26.9)
6 122 7.6 (50-28.7)
7 14.7 8.1 (3.1-26.3)

45
Table 1
The Relative Anti-DNP IgM Concentration in 28 Dogs
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
1
0
16.6+.87
10.6+1.23
.3+0
.2+.01
0
0
0
5
0
2.9+.25
5.8+.90
0
0
0
0
0
9
0
0
2.0+.14
3.3+.29
0
0
0
0
13
0
5.5+.16
5.0+.23
4.8+.07
.9+.13
0
0
0
0
4.4+.39
3.2+.23
1.4+.56
.2+0
0
0
0
3
0
6.5+.47
3.4+.24
0
0
0
0
0
0
4.5+.06
4.4+.27
.6+.14
.1+.03
0
0
0
6
0
6.4+.82
5.6+.39
5.1+.07
3.9+.24
0
0
0
10
0
10.9+1.03
3.0+.41
.8+.01
0
0
0
0
14
0
1.8+.18
5.2+.33
2.3+.16
.5+.22
0
0
0
7
0
6.8+1.01
7.S+.32
3.4+.31
0
0
0
0
n
0
7.0+.02
2.8+.46
1.8+.19
1.0+.07
0
0
0
15
0
10.1+.79
6.1+.43
1.8+109
1.6+.17
0
0
0
0
3.1+.28
5.2+.17
0
0
0
0
0
12
0
5.0+.25
5.1+.37
2.0+.21
1.8+.27
0
0
0
16
0
1.3+.12
2.6+.37
2.5+.14
1.2+.20
0
0
0

46
Table 1 Continued
Animal
Number
17
18
19
20
Weeks
0
0
0
0
0
1
1.1+.03
14.8+2.01
3.9+.39
2.7+.22
2
4.4+.17
5.6+.81
2.1+.09
2.1+.40
3
6.0+.31
3.1+.23
.8+.02
0
4
1.7+.15
2.1+.10
0
0
5
. 8+.0
0
0
0
6
0
0
0
0
7
0
0
0
0
Animal
Number
21
22
23
24
Weeks
0
0
0
0
0
1
8.7+.83
.7+.15
8.1+.44
5.1+.46
2
7.9+.09
4.5+.18
12.6+.99
.6+.06
3
3.2+.48
6.2+.46
2.7+.23
0
4
1.6+.06
2.1+.05
0
0
5
.1+0
0
0
0
6
0
0
0
0
7
0
0
0
0
Animal
Number
25
26
27
28
Weeks
0
0
0
0
0
1
4.0+.09
13.8+.1.35
5.2+. 29
10.6+.87
2
6.2+.21
9.1+1.37
.7+0
X
3
2.5+.30
3.4+.40
1.0+.16
4.2+.36
4
0
2.1+.11
0
.8+.023
5
0
0
0
0
6
0
0
0
0
7
0
0
0
0
a) Each dog was immunized with DNP-ASC
in adjuvant at
weeks 0,2,
and 6.
b) The
units were
calculated from a
relative antibody
concentration scale derived from the
titration of
a serum sample
containing anti-DNP IgM.
A value of
zero indicates no
detectable
anti-DNP IgG. The
data
was the mean
antibody concentration of a
sample run in triplicate + the standard
deviation
from
the mean.
This was
calculated by adding and subtracting the
standard
deviation to the mean com and calculating the relative antibody
concentration for these numbers. These numbers were then
subtracted from the mean concentration.

47
Table 2
The Relative Anti-DNP IgE Concentration In 28 Dogs
Following
Immunization with
DNP-ASC a)
Animal
Number
1
2
3
4
Weeks
0
0 b)
0
0
0
1
0
0
0
0
2
1.2+.31
10.8+1.90
3.2+.36
3.2+.21
3
1.5+.27
11.3+.27
10.0+.76
4.8+.07
4
1.7+.16
9.2+.36
8.5+.44
4.6+.29
5
1.3+.49
11.6+.24
10.1+.83
5.1+.36
6
1.0+.22
5.6+.70
3.3+.12
3.3+.16
7
3.0+.96
7.7+1.01
7.0+.41
4.9+.37
Animal
Number
5
6
7
8
Weeks
0
0
0
0
0
1
.6+.03
0
0
0
2
3.6+.46
1.9+.17
3.2+.10
.6+.07
3
11.9+.99
13.4+1.35
11.8+.77
5.0+1.00
4
7.8+1.04
8.0+.61
6.7+.83
3.9+.43
5
9.0+.63
12.7+.62
11.8+2.45
15.2+.18
6
11.0+.71
16.1+.90
5.5+.69
7.6+.45
7
14.1+1.42
12.8+.88
2.8+.25
6.6+.51
Animal
Number
9
10
11
12
Weeks
0
0
0
0
0
1
1.7+.39
0
0
0
2
5.6+.26
0
2.1+.24
7.5+.88
3
18.2+2.73
3.1+.69
8.3+.74
13.2+1.86
4
8.4+.97
5.5+.43
7.8+.79
9.1+1.02
5
10.0+.64
5.2+.47
3.9+.81
26.8+1.30
6
7.1+.31
3.0+.30
2.4+.65
7.0+1.51
7
8.7+.51
12.6+.76
4.8+2.38
15.4+.36
Animal
Number
13
14
15
16
Weeks
0
0
0
0
0
1
0
0
0
0
2
0
1.8+.36
1.0+.11
0
3
0
9.5+.37
3.4+.48
.7+. 12
4
0
8.4+.67
7.5+1.06
3.9+.61
5
.3+.11
12.0+.87
5.1+.56
3.9+.14
6
.8+.23
8.5+.21
.6+.01
3.7+.37
7
1.6+.52
19.3+1.98
5.2+.50
4.5+.66

48
Table 2 Continued
Animal
Number
17
18
19
20
Weeks
0
0
0
0
0
1
0
0
0
0
2
.2+.07
10.1+.21
0
3.3+.09
3
11.0+.96
9.3+.49
1.4+.08
4.2+.15
4
7.7+.14
S.9+.47
.3+.04
4.S+.33
5
9.8+.34
9.4+.67
1.0+.20
5.2+.40
6
2.2+.26
6.9+.52
1.5+.08
5.8+.26
7
6.7+.31
15.3+.89
2.3+.21
7.1+.43
Animal
Number
21
22
23
24
Weeks
0
0
0
0
0
1
0
0
0
0
2
3.0+.10
1.5+.17
2.1+.13
.1+.04
3
4.5+.11
2.6+.12
1.7+.09
1.0+.06
4
4.0+.27
3.0+.31
1.8+.14
1.0+.17
5
5.6+.36
2.6+.21
1.0+.13
1.6+.31
6
2.3+.12
2.8+.32
0.8+.07
0.9+.07
7
4.8+.22
2.6+.60
0.9+.21
1.3+.29
Animal
Number
25
26
27
28
Weeks
0
0
0
0
0
1
0
0
0
0
2
1.0+.29
1.7+.37
.6+.05
X
3
2.7+.31
2.1+.05
3.2+.22
1.1+.06
4
2.0+.26
3.0+.42
2.0+.23
7.3+.26
5
1.8+.19
1.3+.12
1.0+.15
8.8+.51
6
.7+.09
4.3+.27
1.5+.13
3.3+.16
7
1.2+.10
1.7+.28
0.8+.06
8.4+.42
a) Each dog was immunized with DNP-ASC in adjuvant at weeks 0,2,4 and 6
b) The units were calculated from relative antibody concentration scale
derived from the titration of a serum sample containing anti-DNP IgE. A
value of zero indicates no detectable anti-DNP IgE. The data was the
mean antibody concentration of a sample run in triplicate + the
standard deviation from, the mean. This was calculated by adding and
subtracting the standard deviation to the mean and calculating the
relative antibody concentration for this number. The relative
concentration for this number was subtracted from the mean concentration

49
As TOuld be expected for outbred animals, there was
considerable variation in the immune response between dogs.
If however, the IgG antibody concentration of dogs within
single litters are examined, a more homogeneous trend is
observed (table 4). There was, however, considerable
animal-bo-animal variation within a litter in the level of
antigen specific IgE and IgM (tables 5 and 6). If these
two litters are compared statistically, at each time point,
using a student T test, there is a significant difference
in the mean antibody level between the two groups in the
IgG antibody after the first week (P is less than 0.001 in
all instances).
When all the animals are considered, it appears that
some are generally high responders to the antigenic stimu
lation whereas the response in other dogs is low. The high
response or low response is seen for both IgG and IgE
antibody classes in a single animal. For example, dogs 2
and 6 have a strong IgE and IgG response whereas dogs 15
and 24 have very weak responses. Although this trend
predominates, this association of high responses or low
responses is not always consistent, and a regression
analysis comparing the level of anti-DNP IgG to the
anti-DNP IgE failed to show a statistically significant
correlation (p greater than 0.05).

50
Table 3
The Relative anti-DNP IgG Concentration in 28 Dogs
Following Immunization with DNP-ASC a)
Animal
Number
1
2
3
4
Weeks
1
0 b)
0
0
0
2
0
0
.8+.02
0
3
4.1+.21
9.1+1.4
2.4+.21
3.7+.11
4
3.7+.20
15.5+.20
8.0+.10
7.4+.66
5
6.0+.64
16.4+1.26
4.8+.26
7.4+.14
6
8.1+.33
26.0+1.33
8.6+.23
4.8+.26
7
7.9+1.05
23.3+.14
6.7+.16
6.1+.88
Animal
Number
5
6
7
8
Weeks
1
0
0
0
0
2
0.3+.02
0
0
0.1+0
3
3.7+.12
9.1+1.21
12.2+.77
25.4+1.31
4
8.6+.17
15.3+.86
11.1+.39
27.0+1.92
5
9.1+.34
17.0+1.41
9.2+.64
30.0+3.21
6
7.4+.23
25.6+.73
9.3+1.07
20.1+.49
7
8.6+.61
24.1+1.72
5.9+.26
8.7+1.26
Animal
Number
9
10
11
12
Weeks
1
0
0
0
0
2
0
0
0
0
3
7.3+.S9
4.3+.05
8.3+.80
6.3+.29
4
15.4+1.71
4.3+.12
12.4+.64
22.5+1.10
5
18.5+.86
5.6+.09
7.9+.32
20.8+.93
6
16.0+.46
7.1+.51
7.5+.93
17.8+1.79
7
8.3+.62
6.8+.19
5.5+.68
10.4+.23
Animal
Number
13
14
15
16
Weeks
1
0
0
0
0
2
0
0
0
.3+.02
3
0
5.3+.31
3.8+.43
6.0+.11
4
10.1+.61
16.7+1.1
8.8+.16
' 4.4+.70
5
9.3+.75
14.3+.42
9.9+.63
11.6+1.60
6
7.8+.26
14.7+.32
8.9+.91
11.1+1.13
7
5.0+.16
10.4+.6
9.7+.46
17.2+.32

51
Table 3 Continued
Animal
Number
17
18
19
20
Weeks
1
0
0
0
0
2
2.6+.13
.9+.04
.3+0
0
3
1.7+.14
1.7+.13
4.8+.17
11.7+.11
4
4.8+.36
22.7+.81
16.9+.65
12.8+.96
5
6.2+.58
24.9+1.65
21.3+.17
8.9+.04
6
6.9+.31
21.8+.77
20.8+1.16
7.6+.47
7
5.3+.40
28.2+1.24
21.9+.84
9.8+.36
Animal
Number
21
22
23
24
Weeks
1
0
0
0
0
2
0
7.1+.33
12.9+.92
3.6+.24
3
5.6+.46
7.1+.33
12.9+.92
3.6+.24
4
5.4+.27
10.7+.75
18.0+.75
S.8+.43
5
8.5+.12
13.8+.99
13.9+.51
9.7+.81
6
10.6+.93
19.1+1.53
13.5+.87
9.0+.29
7
7.6+.46
18.2+.75
12.7+.60
6.6+.64
Animal
Number
25
26
27
28
Weeks
1
0
0
0
0
2
0
0
.8+.07
.4+0
3
10.9f.47
3.3+.33
9.6+.4S
X
4
10.0+1.79
16.3+.84
10.3+.68
12.9+.85
5
S.2+.97
15.4+.93
12.1+.06
22.8+1.68
6
7.8+.11
14.7+.43
18.7+1.78
26.9+1.41
7
5.0+.36
10.5+.70
19.6+.96
28.7+1.92
a) Each dog received immunization with DNP-ASC in adjuvant at
weeks 0,2,4 and 6.
b) The units were calculated from a relative antibody-
concentration scale derived from the titration of a serum sample
containing anti-DNP IgG. A value of zero indicates no detectable
anti-DNP IgG. The data was the mean antibody concentration of a
sample run in triplicate + the standard deviation from the mean.
This was calculated by adding and subtracting the standard
deviation to the mean cpm and calculating the relative antibody
concentration for these numbers. This number vas then subtracted
from the mean concentration.

52
Table 4
The Relative Anti-DNP IgG Concentration
In Two Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M +S,
.D b)
Weeks
0
0
0
0
0
0
0
0
1
.8
0
2.6
0
0
0
.6
+ 1
2
2.4
3.7
1.7
11.7
3.6
10.9
5.7
+ 4.
3
8.0
7.4
4.8
12.8
8.8
10.0
8.6
+ 2.
4
4.8
7.4
6.2
8.9
9.7
8.2
7.5
+ 1.
5
8.6
4.8
6.9
7.6
9.0
7.8
7.5
+ 1.
6
6.7
6.1
5.3
9.8
6.6
5.0
6.6
+ 1.
7
5.4
6.0
3.1
11.9
7.2
6.4
6.7
+ 2.
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
.3
0
.1
0
2
9.1
9.1
5.3
4.8
12.9
X
8.2 + 3.
3
15.5
15.3
16.7
16.9
18.0
12.9
15.5+ 1.
4
16.4
17.0
14.3
21.3
13.9
22.8
17.6+ 3.
5
26.0
25.6
14.7
20.8
13.5
26.9
21.3+ 5.
6
23.3
24.1
10.4
21.9
12.7
28.7
20.2+ 7.
7
26.8
25.8
18.9
24.8
23.0
24.7
24.0+ 2.
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
DNP/ASC in adjuvant at weeks 0,2f4 and 6.
b) Mean + standard deviation
3
7
8
5
7
9
3
8
7
9
1
8

53
Table 5
The Relative Anti-DNP IgE Concentration
In 2 Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M + S.D. b)
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
3.7
3.2
.2
3.3
.1
1.0
1.9 + 1.7
3
10.0
4.8
11.0
4.2
1.0
2.7
5.6 + 4.0
4
3.5
4.6
7.7
4.8
1.0
2.0
4.8 + 3.0
5
10.1
5.1
9.8
5.2
1.6
1.8
5.6 + 3.7
6
3.3
3.3
2.2
5.8
.9
.7
2.7 + 1.9
7
7.0
4.9
6.7
7.1
1.3
1.2
4.7 + 2.8
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
0
0
0
0
0
0
0
2
10.8
1.9
1.8
0
2.1
X
3.3 + 4.3
3
11.3
13.4
9.5
1.4
1.7
1.1
6.5 + 5.6
4
9.2
8.0
8.4
.3
1.8
7.3
5.8 + 3.8
5
11.6
12.7
12.0
1.0
1.0
8.8
7.9 + 5.6
6
5.6
16.1
8.5
1.5
.8
3.3
6.0 + 5.7
7
7.7
17.8
19.3
2.3
.9
8.4
9.4 + 7.7
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
DNP-ASC in adjuvant at week 0,2,4 and 6.
b) Mean + Standard Deviation

54
Table 6
The Relative Anti-DNP IgM Concentration
in 2 Litters of Dogs
Following Immunization with DNP-ASC a)
Litter 1
3
4
17
20
24
25
M +S.D. b)
Weeks
0
0
0
0
0
0
0
0
1
6.5
4.5
1.1
2.7
5.1
4.0
4.0 + 1.9
2
3.4
4.4
4.4
2.1
.6
6.2
3.5 + 2.0
3
0
.6
6.0
0
0
2.5
1.5 + 2.4
4
0
.1
1.7
0
0
0
.3 + .7
5
0
0
.8
0
0
0
.1 + .3
6
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
Litter 2
2
6
14
19
23
28
M + S.D.
Weeks
0
0
0
0
0
0
0
0
1
4.4
6.4
1.8
3.9
8.1
10.6
5.9 + 3.2
2
3.2
5.6
5.2
2.1
12.6
X
5.7 + 4.1
3
1.4
5.1
2.3
.8
2.7
4.2
2.8 + 1.6
4
.2
3.9
.5
0
0
.8
.9 + 1.5
5
0
0
0
0
0
.1
.01 + .04
6
0
0
0
0
0
0
0
7
0
0
0
0
0
0
0
a) The relative antibody concentration was determined by
extrapolation of a standard serum sample. A value of zero
indicates no detectable antibody activity. Each dog received
immunization with DNP-ASC in adjuvant at weeks 0,2,4 and 6.
b) Mean + Standard Deviation

55
Discussion
The purpose of the experiments in this chapter vas to
induce an anti-DNP antibody response which included IgE and
to examine the kinetics of this response. As described in
the results section of this chapter, the anti-DNP IgE, IgG
and IgM antibody response followed expected kinetics
(71-73). The IgM response was present before IgE or IgG
antibody was detected and disappeared after the sixth week
in spite of continued antigenic challenge. The IgE and IgG
production had a two week lag period in general, but once
they developed, they were maintained throughout the immuni
zation course.
If inbred laboratory animals such as mice are
immunized with an antigen, a homogeneous response
ordinarily results (74). However, in species that are
genetically heterogeneous, such as man and dogs, the immune
response to the antigen would be expected to be more highly
variable (74). In these outbred dogs there was marked
variation in the kinetics and magnitude of the antibody
responses. The marked variations seen in these dogs are
most probably the result of the genetic differences between
than. In this context, it was noteworthy that the IgG
response was more homogeneous within the same litter than
between litters.

56
The genetic makeup of the animal also influences the
class of antibody produced following antigen challange. In
certain inbred animal strains, it is very difficult to
mount an IgE antibody response without some type of manipu
lative process to eliminate T-suppressor cells (75,76).
Furthermore, if a comparison is .made between allergic and
non-ailergic people, a marked difference in antigen-
specific IgE responsiveness to certain antigens is seen.
Those individuals with allergic tendencies will have an
enhanced IgE response to allergens whereas non-allergic
people may not develop IgE antibody (17). There were
notable differences between individual dogs in terms of
their IgE response and in contrast to the IgG response
there was no consistent pattern within litters. It is not
clear if this failure to see similar patterns within a
litter in the IgE level reflects the antigen chosen to
study IgE in these dogs or if there are multiple genes that
govern IgE levels in dogs. By having such a small sample
size, a consistent pattern might not be observed for IgE
levels. One of the dogs failed initially to develop an IgE
titer. The IgE antibody response started after this dog was
vaccinated with a modified-live canine distemper/hepatitis
vaccine at four weeks of age. The immunization of dogs
with this vaccine has been shown to enhance antigen-
specific IgE response to an antigen administered at the
same time (78). It has been hypothesized that this effect

57
is the result of a suppression of T-suppressor cells.
Because the mechanism that would normally suppress IgE
synthesis is altered, IgE antibody response will develop.
This alteration in the suppressive network and subsequent
IgE antibody synthesis has been called the "allergic break
through" (79).
Summary
Twenty-eight dogs immunized to DNP-ASC at birth and
then three times at two week intervals produced serum anti-
DNP antibody. The IgM response was detected one week after
primary immunization and lasted for up to five weeks. The
IgE and IgG antibody response in general was not present
until week three but persisted through the immunization
schedule. Although variation in the level and duration of
the antibody response was detected between individual dogs,
each dog did have a response that included all three
isotypes examined.
Conclusions
(1) Dogs immunized with DNP-ASC develop a high
level, long term IgE and IgG antibody response but the IgM
response followed a different kinetic pattern in that it
did not persist after week five.

58
(2) There was a difference in the responsiveness to
this antigen seen between individual dogs for all antibody
isotypes. This vas most probably a reflection of the
genetic heterogeneity between these dogs.

CHAPTER THREE
ATTEMPTS TO REGULATE AN ANTIBODY RESPONSE
WITH AUTOLOGOUS ANTIBODY
Introduction
The mechanisms by which antibody responses are regu
lated have been studied extensively. Many experiments have
shown that antibody can be self-regulating (30,81). There
are at least two different ways that this can occur: 1)
If antibody is present at the time of immunization, anti
body can bind to and remove antigen. Therefore, the result
would be a decrease or a failure to mount the response. 2)
Antibody can induce an anti-idiotypic immune response which
would regulate the subsequent expression of the antibody
through id/anti-id interactions (80-82).
If the synthesis of IgE antibody could be suppressed
with antibody, such therapy may be very beneficial in con
trolling IgE mediated allergic disease. As discussed in
Chapter one, passively administered antibody in mice and
rabbits has been shown to suppress IgE antibody (26,27).
The purpose of the experiments in this chapter is to
determine if the administration of autologous antibody has
any effect on the ongoing antibody response in dogs.
59

60
Materials and Methods
Affinity Chromatography
Anti-DNP antibody was produced by immunizing dogs to
DNP-ASC and was purified from serum by chromatography
through a DNP-HSA affinity column as described in Chapter
two. The bound antibody was eluted with 0.1 M glycine HCl,
pH 2.5.
RIA
The RIA for detection of anti-DNP antibody was
described in Chapter two.
Animals and Immunization Schedule
The same dogs that were described in Chapter two were
used in these experiments. These dogs had received 100 pg
of aluminum hydroxide precipitated DNP-ASC by the intra-
peritoneal route on the day of birth at two week intervals
on three further occasions. Fifteen milliliters of serum
were obtained from each dog at the time of final antigenic
challenge. Anti-DNP antibody was purified from this serum
by DNP-HSA affinity chromatography, concentrated to about 3
mg/ml by negative pressure dialysis and rendered
bacterially sterile by passing through a filter having 0.2
micron sized pores. Dogs received either 10 or 100 gg of
their own antibody emulsified in 2 ml of either complete

61
(CFA) or incomplete Freund's adjuvant (IFA). Dogs
designated as controls received 2 ml of either CFA or IFA.
In all cases, the injections were given at four sites subcu
taneously seven and nine weeks after the first immunization
with DNP/ASC. Animals were given DNP/ASC booster injec
tions eight and ten weeks after the primary immunization
(Fig.10).
0123456789 10 11
= Administration of antigen
o = Administration of adjuvant with or without
autologous antibody
Figure 10
Time Schedule for Immunizations

62
Results
DNP Affinity Column
There was no detectable anti-DNP antibody in the
serum of any dog after passage through the DNP affinity
column. On the other hand, the glycine HCl eluate
contained high levels of anti-DNP IgG but no detectable
anti-DNP IgM or IgE. Because anti-DNP IgE was not
detected in either the effluent or the eluent from the
affinity column but was detectable in the serum prior to
such treatment, an aliquot of serum containing anti-DNP IgE
was dialyzed against glycine HCl, pH 2.5 followed by
dialysis against PBS, pH 7.2 to determine what effects
glycine HCl had on canine IgE. There was no detectable
anti-DNP IgE in this serum after such treatment as assayed
by RIA.
Immunization with Autologous Antibody
As noted in Chapter two, there was considerable
variation in the anti-DNP antibody respone between dogs.
Tables 7 and 8 show the mean relative concentration of
anti-DNP IgG and IgE respectively in each group of dogs
prior to the autologous antibody administration and there
after. These data are presented graphically in figures 11
and 12. The individual relative antibody concentrations

Table 7
The Mean Relative Anti-DNP IgG Concentration
63
Group 1 a)
N = 8
Weeks
0 0
1 0
2 .2 + .3
3 8.7+ 7.6
4 12.1 + 7.2
5 12.5+8.4
6 13.7 + 8.7
7 11.4 + 7.7
8 12.6 + 8.5
9 13.2 + 7.9
10 13.2 + 7.9
11 15.2 + 9.2
Group 4
N = 2
Weeks
0 0
1 0
2
6,
.4
+
1.1
3
8,
.2
+
3.5
4
11,
.2
+
3.7
5
14,
.9
+
6.0
6
12,
.9
+
7.5
7
20,
.9
+
6.4
8
20,
.1
+
4.9
9
22,
.7
+
8.5
10
23,
.5
+
9.1
11
23,
.8
+
10.7
Group 2
N = 4
0
0
.5+ .9
4.4+ 3.6
14.3 + 6.3
13.3 + 6.5
12.5 + 6.0
13.4 + 8.3
14.2 + 8.6
14.5 + 7.6
14.5 + 7.6
17.8+ 6.9
Group 5
N = 6
0
.2 + .3
8.1 + 4.4
12.7 + 3.7
13.7 + 5.2
15.1 + 7.0
13.4 + 8.9
16.4 + 7.8
17.3 + 8.5
18.9 + 7.5
19.4 + 8.0
17.7 + 6.4
Group 3
N = 8
0
0
6.6 + 1.7
13.7+7.5
13.2 + 7.6
12.2 + 5.6
7.8 + 2.1
10.6 + 5.9
12.1 + 4.3
13.8 + 5.9
13.8 + 5.9
15.1 + 7.4
Control
N = 8
0
.1 + .3
7.6 + 3.7
11.5 + 4.1
13.1 + 4.8
15.1 + 6.4
13.6 + 8.1
17.5 + 7.3
17.8 + 7.5
19.8 + 7.3
20.5 + 8.2
19.1 + 7.0
a) All dogs received DNP/ASC iirimunization at 0,2,4,6,8 & 10 weeks.
At 7 and 9 weeks: Group 1 received 10 gg autologous anti-DNP
antibody in CFA, Group 2 received 100 gg autologous anti-DNP
antibody in IFA, Group 3 received 100 gg autologous anti-DNP
antibody in CFA. Group 4 received IFA alone, Group 5 received CFA
only. Control values were the mean of groups 4 and 5.

Table 8
The Mean Relative Anti-IgE Antibody Concentration
64
Group 1 a)
Group 2
Group 3
seks
0
0
0
0
1
. 1
+ .2
0
.4 + .85
2
3.5
+ 3.2
2.1
+ 3.5
3.8 + 3.4
3
8.7
+ 4.3
4.9
+ 4.4
10.7 + 6.5
4
6.3
+ 2.6
5.2
+ 3.6
7.7 + 1.6
5
9.6
+ 4.5
5.8
+ 4.2
11.4 + 10.5
6
6.7
+ 4.9
3.8
+ 3.0
3.7 + 4.5
7
8.0
+ 5.3
7.8
+ 6.3
10.4 + 4.6
8
7.2
+ 3.9
3.7
+ 2.7
7.8 + 2.5
9
8.3
+ 3.2
5.3
+ 3.1
8.8 + 1.0
10
7.6
+ 3.3
4.9
+ 2.6
7.3 + 2.4
11
8.3
+ 3.8
5.4
+ 3.1
7.0 + 2.7
Group 4
Group 5
Control
jeks
0
0
0
0
1
0
0
0
2
2.3
+ 1.1
1.1
+ .8
1.4 + 1.0
3
3.6
+ 1.3
2.0
+ .9
2.4 + 1.2
4
3.5
+ .7
2.9
+ 2.3
3.0 + 2.0
5
4.1
+ 2.1
2.6
+ 3.1
3.0 + 2.8
6
2.6
+ .4
1.9
+ 1.5
2.1 + 1.3
7
3.7
+ 1.6
2.4
+ 3.0
2.7 + 2.6
8
6.6
+ 2.0
2.8
+ 1.9
3.8 + 2.5
9
8.3
+ 1.0
3.2
+ 2.2
4.4 + 3.0
10
7.1
+ .1
2.7
+ 2.2
3.8 + 2.7
11
6.9
+ 2.3
3.0
+ 2.3
4.0 + 2.7
a) All dogs received DNP/ASC immunization at 0,2,4,6,8 and 10
weeks. At 7 and 9 weeks: Group 1 received 10 gg autologous
anti-DNP antibody in CFA, Group 2 received 100 gg autologous
anti-DNP antibody in IFA, Group 3 received 100 gg autologous
anti-DNP antibody in CFA, Group 4 received IFA alone, Group 5
received CFA alone. Control values were the mean of groups 4 and
5.

Figure 11.
The mean relative anti-DNP IgG concentration as measured by RIA.
All dogs received DNP-ASC immunization at weeks 0,2,4,6,8 and 10.
At weeks 7 and 9, the treatment consisted of immunization with:
Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100
pg autologous anti-DNP antibody in IFA; Group 3, 100 pg
autologous anti-DNP antibody in CFA; Control, either IFA or CFA
alone. The relative antibody concentration was calculated from an
arbitrary antobody concentration scale derived from the titration
of a standard anti-DNP IgG containing serum (see text for further
details).

u 0 I
or
234 56789
Weeks
I DNP/ASC
Treatment
10 II
control
group 2
group 3
group I
O

Figure 12.
The mean relative anti-DNP IgE concentration as measured by RIA.
All dogs received DNP-ASC immunization at weeks 0,2,4,6,8 and 10.
At weeks 7 and 9, the treatment consisted of immunization with:
Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100
pg autologous anti-DNP antibody in IFA; Group 3, 100 pg
autologous antibody in CFA; Control, either IFA alone or CFA alone.
The relative antibody concentration was calculated from an
arbitrary scale derived Eran the titration of a standard anti-DNP
IgE containing serum (see text for further details).

I 23456789
Weeks
| DNP/ASC
T reatment
group
group 3
group 2
control
10 II

Table 9
The Relative Anti-DNP IgE Concentration in 28 Dogs a)
69
Group
lb) (10
gg Anti-DNP Antibody
in CFA)
Animal
Number
1
2
3
4
Weeks
3
2.0+.17
6.4+.29
6.1+.49
8.1+.36
9
2.6+.26
9.3+.84
6.5+.25
9.4+.36
10
3.0+.18
8.7+.31
3.4+.30
7.1+.79
11
2.8+.31
9.3+1.26
4.1+.19
8.2+.81
Animal
Number
5
6
7
8
Weeks
8
5.2+.57
15.6+.70
6.6+.6S
7.2+.30
9
8.9+.63
14.0+.69
6.8+.78
9.2+1.18
10
9.3+.72
13.2+.36
8.2+.14
8.2+.07
11
10.2+.64
15.1+.47
7.4+.66
8.6+.42
Group 2 (100 gg
Anti-DNP Antibody in
IFA)
Animal
Number
9
10
11
12
Weeks
8
9.8+.47
10.0+.22
5.7+.67
5.S+.86
9
9.5+1.89
9.7+.46
7.5+.83
S.5+.33
10
9.8+.67
8.3+.57
4.1+.65
6.8+.41
11
9.4+.59
4.1+.12
5.3+.41
9.3+.17
Group
3 (100 gg Anti-DNP Antibody in CFA)
Animal
Numbsr
13
14
15
16
Weeks
8
2.4+.02
3.3+.30
0
4.2+.21
9
4.1+.62
4.0+.50
0
2.9+.14
10
2.0+.39
6.3+.35
0
7.7+.79
11
3.0+.48
9.6+.47
0
2.9+.29
Animal
Number
17
18
19
20
Weeks
8
6.1+.22
5.6+.49
.6+.01
7.6+.51
9
9.2+.35
8.9+.46
6.0+.43
7.0+.88
10
6.4+.57
5.2+.34
4.5+.06
6.8+.50
11
7.3+.87
6.8+.65
6.4+.06
7.2+1.01

70
Table 9 Continued
Animal
Group 4 (IFA Alone) Group 5 (CFA Alone)
Number
21
22
23
24
Weeks
8
8.0+.69
5.2+.48
2.2+.29
.3+.02
9
9.0+1.22
7.6+.71
2.2+.20
0
10
7.1+.51
7.0+1.06
1.8+.36
0
11
8.5+.34
5.3+.44
2.6+.09
0
Group 5
Continued
Animal
Number
25
26
27
28
Weeks
8
1.0+.02
4.3+.34
4.2+.44
5.0+.28
9
1.9+.18
4.3+.31
4.2+.35
6.3+.06
10
1.1+.16
3.1+.41
5.7+.26
4.7+.71
11
1.4+.21
2.9+.24
S.2+.38
5.9+.29
a) The level of anti-DNP IgE in these dogs from 0 to week 7 is found
in Table 2.
b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At
weeks 7 and 9: Group 1, 10 pg autologous anti-DNP antibody in CFA;
Group 2, 100 pg autologous anti-DNP antibody in IFA; Group 3, 100
pg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.

Table 10
The Relative Anti-DNP IgG Concentration in 28 Dogs a)
71
Group 1 b) (10 pg Anti-DNP Antibody in
CFA)
Animal
Number
1
2
3
4
Weeks
8
9.0+.11
24.7+.45
6.6+.86
8.3+.75
9
9.0+.66
27.4+2.50
10.0+.52
7.5+.78
10
10.2+.93
28.5+.76
10.1+.77
6.2+1.05
11
9.1+1.04
26.8+.39
6.9+.30
9.S+.76
Group 1 Continued
Anina 1
Number
5
6
7
8
Weeks
8
5.5+.23
24.7+1.73
9.4+1.37
17.3+.75
9
4.6+.68
27.0+.39
14.0+.38
20.3+.72
10
5.S+.89
29.1+2.13
14.7+87
17.6+.48
11
8.9+.24
26.6+.90
Group 2 (100 pg
12.3+.99
Anti-DNP Antibody
19.7+1.29
in IFA)
Animal
Number
9
10
11
12
Weeks
8
17.1+1.37
8.9+.76
8.1+.62
14.2+.83
9
20.9+1.47
8.8+.99
9.2+.50
16.4+.38
10
22.9+1.10
11.9+.26
8.7+.61
15.4+1.12
11
25.4+.43
11.3+.69
Group 3 (100 pg
8.3+.90
Anti-DNP Antibody
15.0+.32
in CFA)
Animal
Number
13
14
15
16
Weeks
8
5.6+.17
13.9+.83
7.9+.67
16.7+.99
9
10.3+.62
19.0+.65
11.8+1.12
16.0+.87
10
12.7+.41
20.4+.11
14.3+1.57
15.0+1.01
11
12.2+.73
20.3+2.24
14.7+.59
17.4+2.56

72
Table 10 Continued
Animal
Number
Weeks
8
9
10
11
Group 3 Continued
17 18 19
7.3+.51
6.5+.60
8.4+.62
9.1+.76
25.5+3.14
21.6+1.19
24.3+.29
23.1+1.79
24.8+.64
29.0+2.62
30.1+4.71
30.8+1.06
20
14.6+1.73
16.8+.96
17.2+2.33
22.9+1.87
Animal
Number
Weeks
8
9
10
11
Group 4 (IFA Alone) Group 5 (CFA Alone)
21
15.6+.93
16.7+.81
16.2+1.3
17.1+.43
22
23.3+1.87
28.7+1.61
31.4+1.05
30.0+2.62
23
30.1+1.74
27.7+.50
31.9+3.81
26.5+2.62
24
8.1+.93
9.5+.62
10.7+1.08
9.6+.27
Animal
Number
Weeks
8
9
10
11
Group 5 Continued
25 26 27
9.3+.81
10.4+.65
11.7+.88
11.2+.76
13.6+.55
18.7+1.17
20.5+1.48
18.0+1.42
20.2+.96
22.4+.68
17.3+1.24
18.7+.48
28
22.2+1.43
23.4+.93
24.3+.74
22.0+2.38
b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At
weeks 7 and 9: Group 1, 10 gg autologous anti-DNP antibody in CFA;
Group 2, 100 gg autologous anti-DNP antibody in IFA; Group 3, 100
gg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.

73
for IgE and IgG for each dog is given in table 9 and 10.
There was a gradual increase in the mean antibody concen
tration in general for both IgE and IgG antibody whereas
IgM was not detected after week. 6. In two dogs (15, 24),
after the seventh and eighth weeks respectively, there was
a cessation of the IgE response. Although the mean IgE
antibody response increased for the groups in general,
individual dogs varied considerably. For example, dogs 14
and 18 had a peak IgE antibody response at week 7 and there
after the response diminished, whereas the peak response
for dogs 19 and 27 occurred at the end of the schedule.
There was no marked difference in the antibody
response between the different groups of dogs. To deter
mine if there were any patterns in the antibody response
between these groups, an analysis of variance comparing
time by group vras calculated for each antibody class with
the assistance of the Department of Biostatistics, College
of Medicine, University of Florida. There was no signif
icant difference between these groups at any given time by
this analysis (p greater than .05).
Because of the large variation between dogs, an
analysis of variance was calculated comparing dogs within a
single litter in one group to dogs from the same litter in
other groups as a function of time. This analysis was used
to determine if there was variation between one treatment
in the IgE or IgG antibody response as compared to a second

74
treatment within a single litter. In no case was a signif
icant difference observed.
Discussion
The fact that anti-DNP antibody could not be detected
in the effluent from the affinity column indicates that the
column was effective in removing all anti-DNP antibody.
The inability to detect IgE in the glycine eluate was
expected because canine IgE is not stable at low pH.
Halliwell (7) has shown that at a pH of 2.5 for 30 minutes
there is greater than a tenfold decrease in detectable IgE
antibody.
There are at least four possible reasons why autol
ogous antibody administration failed to regulate the anti
body response in these dogs as had been achieved in labora
tory animals. Firstly, in these experiments the dogs had
an established antibody response whereas in many of the
experimental systems where passive antibody showed regu
latory effects on antibody production, a primary or early
secondary response was manipulated. It has been shown that
it is much more difficult to manipulate a preexisting and
established response than to alter a developing one.
Secondly, there may be something unique about the regu
latory effects of passive antibody on an immune response in
young animals. Antibody is transferred from mother to

75
young both before and shortly after birth. If this passive
antibody would result in long term suppression, then it
might result in a negative selection process in those
animals by rendering them immunologically non-responsive to
pathogenic agents. Therefore, very young animals may be
less susceptable to the regulatory action of passive
antibody. In fact, in a very recent report by Jarrett and
Hall (83), they demonstrated that maternal antibody or
passively administered antibody given to newborn rats
resulted in an enhanced IgG antibody response when
challenged with antigen at six weeks of life. However, not
every rat so treated had this enhanced IgG response and
sane rats had a decrease in their IgE response. Thirdly it
has been hypothesized that one way in which passive anti
body administration could regulate antibody '.vas through the
generation of an anti-id response (30). It is possible
that any anti-id produced by these dogs was not sufficient
to regulate antibody. Lastly, such regulation may result
in a clonal escape mechanism. Pawlak et al. (84) has shown
that the administration of anti-id to A/J mice against the
major cross reactive idiotype produced in these mice immu
nized to p-azobenzenearsonate would suppress this idiotype
and other cross reactive idiotypes, but there was a compens
atory increase in other idiotypes not ordinarily expressed
in these mice. If the administration of autologous
antibody had regulated a subpopulation of antibody

76
molecules but in response other antibody molecules were
expressed, the net effect may not be observable if the
entire class specific response were to be measured as was
the case in these experiments.
Summary and Conclusions
The administration of autologous anti-DNP antibody in
adjuvant to dogs which had an ongoing anti-DNP antibody
response did not have a significant effect on this antibody
response as compared to control dogs who received adjuvant
without autologous antibody. This would suggest either
that the regulation by passive antibodies, as seen in
laboratory animals, does not operate in this species or
that any regulation that occurred by such treatment could
not be detected by the methods used in this study.

CHAPTER FOUR
THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
Introduction
Anti-idiotypic antibody has been produced by immu
nizing an animal with isologous or autologous antibody in
adjuvant (40-42). The use of: isologous or autologous anti
body rather than homologous antibody eliminates the poten
tial that allotypic determinants might be recognized rather
than idiotypic determinants. The immunization schedule
used in the previous experiments included the admini
stration of autologous antibody in adjuvant. It was hoped
that this treatment would regulate IgE antibody, but unfor
tunately, it did not. It was not known if this failure was
because of a lack of an anti-id response or for other
reasons. This treatment may have induced anti-id. It is
also possible that anti-id may have been produced during
the immunization with antigen. The purpose of the experi
ments in this chapter was to determine if, at any time
during the immunization schedule, anti-id was detectable.
77

78
Materials and Methods
Antisera
The antisera used in these experiments were described
in Chapter two. Any cross reactive anti-mouse
immunoglobulin activity that was present in the anti-canine
IgG, IgM or IgE antisera used in the anti-id RIA was
removed by passage through an affinity column which had
bound to it a 40 percent saturated ammonium sulfate
precipitate of normal mouse serum.
Animals and Immunization
The animals and immunization schedules have been
described in Chapter two and three except that in the
experiment designed to determine if the specificity of the
antibody was important in the induction of an anti-id
response, a different immunization protocol was used.
Eight mature dogs were injected with 100 pg of aluminum
hydroxide precipitated DMP-ASC by the intraperitoneal route
on the day of arrival. At the same time, these dogs
received a second injection of 100 pg of aluminum
hydroxide precipitated ABA-KLH by the same route at a
different site. These dogs were immunized three times at
two week intervals. At the time of the last immunization,
30 ml of blood were obtained from each dog. The serum from
this blood was used to purify antibody. The dogs were

79
arbitrarily placed into one of three groups. Group 1 (N=3)
received 100 pg of autologous anti-DNP antibody in CFA by
the subcutaneous route. This antibody was purified from
serum by adsorption to and elution from a DNP coupled
affinity column followed by passage through an ABA coupled
affinity column to ensure that the purified anti-DNP anti
body had no cross reactive anti-ABA antibody. Conversely,
the autologous anti-ABA antibody in CFA for Group 2 (N=3)
was purified from serum by adsorption to and elution from
an ABA affinity column followed by passage through a DNP
column. The control group, Group 3 (N=2) received CFA
without autologous antibody. This later immunization was
administered six weeks after the primary injection of
antigen. Twd weeks after this last injection, serum from
each animal was assayed for anti-idiotypic antibody.
Anti-id RIA
The RIA used to detect anti-id was performed essen
tially as described for the antigen specific RIA using a
number of mouse monoclonal anti-DNP antibodies as the
antigen. These were
a) Anti-DNP IgG, (a gift from Dr. A.P. Lopes, Univer
sity of Pennsylvania) and as a control antibody, anti-H2K
IgG (a gift from Dr. P. Klein, University of Florida).
b) Anti-DNP and IgE as a control antibody, anti-OVA
IgE.

80
c) Anti-DNP IgM and as a control antibody, anti-SRBC
IgM.
The antibodies b) and c) were obtained from Sera-
Labs, Accurate Chemical and Scientific Corp., Westbury,
N.Y.
d) Anti-DNP IgM (a gift from Dr. C.W. Clem, Miss
issippi State University. This antibody will be designated
anti-DNP IgM-^), and as a control antibody, anti-SRBC IgM
(a gift from Dr. W.C. Raschke, La Jolla Cancer Research
Foundation, Ca.).
The wells were coated with 10 pg antibody in 50 pi
of Tris buffer (0.1 M, pH 8.0). A radiolabelled
anti-canine IgG antibody was used as the radiolabelled
probe unless otherwise stated.
Hapten Inhibition of Id/anti-id Interaction
The RIA using mouse anti-DNP IgG monoclonal antibody
or control IgG monoclonal antibody was performed as prev
iously described except that after blocking any remaining
active sites by the addition of HSA to the plates, various
amounts of 2,4-dinitrophenol glycine (Sigma Chemical Co.,
St. Louis, Mo.) ranging from 0.001 to 0.1 mg in PBS were
incubated for three hours at 4C. Serum samples were
then added to the wells, incubated for three hours at 4C
and cashed to renove unbound protein. A radiolabelled
anti-canine IgG antibody was added, incubated for three

81
hours at 4C and the wells were washed to remove unbound
radiolabelled antibody. The amount of radioactivity asso
ciated with the well was determined in a gamma counter.
cpm sample in the presence of hapten
% Inhibition = cpm sample in the absence of hapten X 100
Inhibition of Antigen Antibody Interactions by
Anti-Idiotypic Antibodies
The RIA using mouse anti-DNP IgG monoclonal antibody
or a subtype and allotype matched control mouse IgG mono
clonal antibody was performed except that after blocking
remaining active sites by the addition of HSA to the
plates, serum with or without anti-id was added, incubated
for three hours at 4C and washed three times with RAST-
buffer. Antigen (125 I DNP-HSA, approximately 20,000 cpm)
was added to the wells and incubated for three hours at
o
4 C and each well was washed five times to remove unbound
antigen. Any anti-id that ivas bound to the anti-DNP anti
body may inhibit this antigen-anti-DNP interaction. The
amount of antigen bound to the wells was determined in a
Packard gumma counter. The percent inhibition by anti-id
was calculated by
% Inhibition =
cpm bound to plates after serum incubation
cpm bound to plates after PBS incubation
1-
X 100

82
Results
Identification of Canine Anti-Iditoypic Antibody
A screening procedure was used to assay for the
presence of anti-id in serum obtained during the immun
ization schedule. Those dogs that received autologous
antibody produced an antibody which would bind to mouse
monoclonal anti-DNP IgG as seen in table 11. This binding
was not detectable prior to such treatment in any dog nor
could it be detected in control dogs at any time. There
was no detectable binding to tine anti-H^K IgG mouse anti
body in serum from any dog.
The experiment was repeated with another group of 12
dogs and the anti-id activity was converted to an arbitrary
relative antibody concentration by interpolation from a
scale derived from the titration of a positive high titer
sample identified in the screening procedure (table 12).
This standard serum was given a relative antibody concen
tration of 10. Some dogs produced detectable levels of
this anti-idiotypic antibody within one week after autol
ogous antibody administration whereas other dogs took three
weeks to develop such a response. There was also variation
in the magnitude of the response observed. This anti-
idiotypic antibody could not be detected if anti-canine IgE
or IgM antisera was used as the radiolabelled probe rather

83
Table 11
Screening for Canine IgG Anti-Idiotypic Antibody by RIA
Using Mouse Monoclonal Anti-DNP IgG as the Antigen
Animal
# a) 2
3
4
6
14
Week
0
476+16
331+19
448+7
397+14
422+21
1
453+7
590+1
481+11
421+10
470+47
2
412+21
347+3
274+3
335+6
443+15
3
317+9
277+11
352+9
371+3
367+40
4
305+15
358+3
367+27
401+16
457+12
5
378+31
376+7
358+18
421+11
352+25
6
396+14
417+37
284+10
379+23
318+14
7
421+27
335+10
409+15
522+27
314+24
8
1357+31
1815+26
433+21
2173+68
277+43
9
2140+150
2027+36
1533+117
2250+137
1420+48
10
1862+17
2208+55
2092+130
2297+100
2135+186
11
2174+79
3058+121
2107+41
3375+46
5087+161
Animal #
17
19
20
23
24
Week
0
476+1
367+40
307+22
422+9
483+19
1
453+17
351+26
406+19
470+16
517+42
2
418+5
481+56
464+3
394+7
347+5
3
519+43
435+13
253+13
367+14
318+23
4
442+16
467+24
373+20
334+17
351+1
5
376+18
430+45
481+6
442+26
235+7
6
340+28
398+1
295+30
462+11
434+16
7
315+7
384+93
563+38
371+31
346+34
8
793+13
1937+179
721+37
529+21
471+18
9
973+10
2476+88
936+68
315+8
512+46
10
1611+86
2720+9
2437+177
447+60
341+28
11
1735+122
2925+66
2026+69
473+3
429+27
a) All dogs were immunized with DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10. At weeks 7 and 9, dog 2,3,4 and 6 received 10 pg
autologous anti-DNP antibody in CFA; dogs 14,17,19 and 20 received
100 pg autologous anti-DNP antibody in CFA; dogs 23 and 24 received
CFA alone.
b) This represents the mean + standard deviation of a triplicate
sample. All samples were'assayed at a serum dilution of 1/5 in PBS.
The mean + standard deviation for all samples assayed using mouse
anti-t^K IgG^ was 477+97.

84
Table 12
The Relative Antibody Concentration of Canine IgG
Anti-Idiotypic Antibody as Measured in a RIA using
Mouse Monoclonal Anti-DNP IgG as the Antigen a)
Animal
Number
1
5
7
8
13
Weeks
0
0
0
0
0
0
1
0
0
0
0
0
2
0
0
0
0
0
3
0
0
0
0
0
4
0
0
0
0
0
5
0
0
0
0
0
6
0
0
0
0
0
7
0
0
0
0
0
8
3.1+.2
4.0+.2
0
0
0
9
5.3+.4
5.4+.4
3.1+.3
3.1+.4
0
10
4.3+.3
5.1+.4
3.S+.2
5.4+.6
3.4+.2
11
5.2+.4
8.0+.6
2.9+.1
5.5+.2
6.6+.4
Animal
Number
15
16
18
26
27
Weeks
0
0
0
0
0
0
1
0
0
0
0
0
2
0
0
0
0
0
3
0
0
0
0
0
4
0
0
0
0
0
5
0
0
0
0
0
6
0
0
0
0
0
7
0
0
0
0
0
8
0
2.4+.1
1.2+.1
0
0
9
3.7+.2
2.2+.2
2.8+.1
0
0
10
10.0+1.3
4.3+.3
5.2+.4
0
0
11
9.2+.7
3.3+.1
7.1+.3
0
0
a) The relative antibody concentration + range was determined by
interpolating from the titration of a serum sample containing high
levels of anti-id. A value of zero indicates no detectable
anti-idiotypic antibody.
b) All animals received DNP-ASC adjuvant at weeks 0,2,4,6,8,10. At
weeks 7 and 9 all animals received autologous antibody in adjuvant
except 26 and 27 who received adjuvant alone.

85
than the anti-IgG antisera, indicating that it was of the
IgG class.
When three different anti-DNP monoclonal antibodies
and three control monoclonal antibodies were used, the
binding activity was detected only with the original anti-
DNP IgG (table 13), and to a lesser extent, to anti-DNP
IgM~2 (table 14). In those animals in which anti-DNP
IgM~2 binding activity was detected, a comparison was
made between the serum from a point in time prior to autol
ogous antibody administration and serum obtained after such
treatment. A minimum value that was two standard devi
ations above the mean of the control was considered indi
cative of anti-id activity. As was the case with the IgG
antibody, only those animals which received autologous anti
body, showed binding activity and only after administration
of autologous antibody (table 14).
The Role of Antibody in the Specificity of the
Anti-Idiotypic Production
To determine if immunization with an antibody whose
specificity was other than anti-DNP would result in anti-
DNP/anti-id, eight dogs were given both DNP/ASC and ABA/KLH
three times at two week intervals. Six weeks after the
primary injection of antigen, three dogs received 100 gg
of autologous anti-DNP antibody in CFA, and a different
three dogs received 100 gg autologous anti-ABA antibody in

86
Table 13
Detection of Canine IgG Anti-Idiotypic Antibody
by RIA With Various
Mouse Monoclonal
Antibodies
Anti-DNP
Anti-OVA
Anti-DNP
(IgE)
(IgE)
(IgM)
Animal
#
26
234+7
267+25
264+4
27
281+1
335+89
151+6
16
171+3
286+48
286+3
15
321+31
261+31
301+29
8
261+13
284+14
322+17
7
287+69
221+8
361+37
13
389+11
205+34
257+28
108+45
105+34
Anti SRBC
Anti-DNP
Anti-ABA
dgM)
(IgG)
(IgG)
Animal
#
26
182+36
321+95
209+47
27
197+12
354+29
218+38
16
176+42
4777+65
207+7
15
106+7
7571+246
176+12
8
106+7
1273+93
178+31
7
114+8
980+47
168+14
13
189+15
1144+68
221+8
Serum was diluted 1:2 in PBS. All samples were from week 10 of the
immunization schedule. All animals had received DNP-ASC in adjuvant
at week 0,2,4,6,8,10 and at weeks 7 and 9 received autologous
antibody in CFA except 26 and 27 who received CFA alone.

87
Table 14
Detection of Canine IgG Anti-Idiotypic
Antibody by RIA with Mouse Monoclonal
Anti-DNP IgM Antibody
Dilution of
Serum Anti-DNP IgM^Antibody
a) -
26-E
26-L
27-E
27-L
16-E
16-L
1/5
162+11
286+21
135+14
248+45
90+7
643+31
1/10
82+14
154+15
97+26
152+29
86+
296+19
1/20
71+21
135+9
126+36
181+11
103+11
156+12
15-E
15-L
8-E
8-L
7-E
7-L
1/5
225+31
912+25
215+47
570+46
249+1
672+6
1/10
184+8
442+27
128+6
373+31
125+8
343+37
1/20
115+12
247+6
102+25
270+12
86+7
236+24
5-E
5-L
1/5
128+33
741+54
1/10
106+7
358+54
1/20
134+8
229+11
a) The
E indicates serum
obtained
prior to the
administration of
autologous antibody administraion (16,15,8,7) or adjuvant alone
(26,27), at week 6.
The L indicates serum obtained after such treatment from week 10.
The mean + standard deviation for all samples assayed using
anti-SRBC IgM was 136+53.

88
CFA. 'Two dogs received CFA alone. Serum before and two
weeks after this treatment was screened for anti-id
activity. As seen in table 15, the dogs that received
autologous anti-DNP antibody produced an anti-id which was
detected by the anti-DNP IgG mouse monoclonal antibody and
failed to bind to the anti-ABA IgG. Dogs that received
anti-ABA antibody in adjuvant developed anti-id which bound
to anti-ABA mouse monoclonal IgG but failed to bind to the
anti-DNP IgG mouse monoclonal antibody. The two control
dogs produced no detectable anti-id that was reactive with
either mouse monoclonal antibody.
Hapten Inhibition and Elution Studies of the Id/anti-id
Interaction
The anti-idiotypic RIA was used to determine if
hapten could inhibit the binding of anti-id to the mouse
anti-DNP antibody. In two of the six cases (5,15), 10 gg
of hapten was able to inhibit id/anti-id interaction (table
16) as shown by a slight decrease in the cpm bound to the
wells as compared to the same sample incubated with PBS (22
percent and 26 percent inhibition respectively). As the
concentration of hapten decreased, so did the percent
inhibition, (17 percent and 5 percent at a fivefold
decrease in hapten concentration). However, it is unclear .
how significant this inhibition was because of the
extremely large amounts of hapten required to obtain these

89
results. Similarly, hapten was unable to elute anti-id
from id when a concentration of DNP-glycine of up to 20 (jg
vas used.
Inhibition of Antigen Binding to Antibody by
Anti-Idiotypic Antibody
Since hapten could not consistently interfere with
id/anti-id, it vas reasoned that perhaps a large hapten
coupled molecule might interfere with this interaction.
Although the anti-id vas not binding to id determinants
within the antigen binding site, it may have bound to deter
minants close enough to the hypervariable region of the
antibody molecule to sterically hinder antigen binding to
antibody. Serum containing anti-id was preincubated with
anti-DNP mouse monoclonal antibody prior to the addition of
a radiolabelled DNP-HSA antigen to determine if the
presence of anti-id could inhibit the anti-DNP antibody
DNP-HSA antigen interaction. As seen in table 17, anti-id
was able to inhibit the binding of the radiolabelled
antigen to the mouse monoclonal antibody. Animal 26 had no
detectable anti-id, and serum from this dog failed to inter
fere with antigen binding to antibody. In contrast, the
other dogs had detectable anti-id and inhibited this inter
action from 21 to 50 percent of the maximum cpm bound.

90
Table 15
Specificity of Canine IgG Anti-Idiotypic Antibody
After the Administration of
Autologous Antigen-Specific Antibody
Anti-ABA
Anti-DNP
Group 1 a)
Group 2 b)
Antibody in well
Antibody
in well
Animal Serum Anti-DNP Anti-ABA
Aina 1
Anti-DNP
Anti-ABA
#
Dil.
m1
I2GX
#
laGg
12%
29
1/5
391+36
1901+101
33
1243+37
134+53
1/10
236+13
1675+118
996+115
.142+13
1/20
153+14
1145+141
703+10
249+86
30
1/5
293+114
1867+157
34
1221+101
113+65
1/10
186+26
1383+57
683+6
261+69
1/20
85+37
1061+46
545+59
143+19
31
1/5
383+35
1070+96
35
902+36
317+29
1/10
248+89
899+18
531+106
131+16
1/20
103+15
494+86
331+21
129+21
Control c)
32
1/5
172+16
246+42
36
238+33
186+26
1/10
158+47
117+28
186+32
133+58
1/20
98+8
88+14
121+13
94+17
Each
. dog received three
immunizations
with
ABA-KLH and DNP-ASC at two
week
: intervals. Six weeks after primary immunization, dogs received
either.
a)
Group 1
received 100
pg autologous anti
-ABA antibody
in CFA
b)
Group 2
received 100
pg autologous anti
-DNP antibody
in CFA
c)
Control
received CFA
alone
d) Mean cpm of the sample assayed in triplicate + standard
deviation. The serum was diluted 1/5 with PBS.
The serum used in this assay was obtained 2 weeks after this later
immunization.

91
Table 16
The Inhibition of Canine IgG Anti-id Binding to
Mouse Monoclonal Anti-DNP IgG by Hapten
as Measured by RIA
Animal Number a)
2, 4 DNP Glycine 1
5
7
0 pg
3943
+
73 b)
4522 + 138
1775
+34
10 pg
3802
+
100
3533 + 69
1704
91
2 pg
3983
+
280
3783+136
1675
+ 46
1 pg
3807
+
214
4268 + 56
1614
+ 101
0.1 pg
3804
+
12
4623 + 219
1734
+ 73
8
15
16
0 pg
2281
+
162
2702
+ 131
1643
+ 52
10 pg
2203
+
200
2026
+ 168
1454
+ 15
2 pg
2012
+
115
2593
+ 173
1691
+ 14
1 pg
2148
+
83
2738
+ 57
1377
+ 34
0.1 pg
2213
+
129
2694
+ 89
1526
+ 31
a) Each dog was immunized with autologous anti-DNP antibody in CFA
at weeks 7 and 9. All samples were from weak 9 in the schedule
except sample 15 which was from week 11.
b) This is tlie mean cpm of triplicate samples + standard
deviation. All samples were assayed with serum diluted one to four
with PBS.
A control well with mouse anti-^K IgG. rather than anti-DNP
IgG was assayed for each sample. The mean + standard deviations
for all samples 283+47.

92
Table 17
Inhibition of binding of 125 I-DNP/HSA to
Mouse Monoclonal Anti-DNP Antibody
by Canine Anti-Idiotypic Antibody
Animal a) C.P.M. Bound + S.D. b) % Inhibition c)
26
994
+
43
0
8
493
+
38
50
18
544
+
25
45
13
783
+
41
21
15
611
+
13
39
1
521
+
18
48
a) Animal 8, 18, 13, 15 and 1 received autologous antibody in CFA
and had detectable levels of anti-id; animal 26 received CFA alone
and did not have detectable anti-id. all serum was obtained at
week 11.
b) This is the mean cpm + standard deviation of a sample assayed
in triplicate
c)
% Inhibition = c.p.m. control c.p.m. sample
c.p.m. control
The control was a set of wells coated with mouse anti-DNP IgG and
incubated with PBS rather than serum. The value for this was
1026+19.

93
Discussion
Dogs immunized with autologous antibody produced an
antibody which bound to one- anti-DNP mouse monoclonal IgG.
This antibody was present only after such treatment and not
present prior to the administration of autologous antibody.
The specificity was limited to id determinants present on
some but not all anti-DNP mouse monoclonal antibodies. It
could not be detected using mouse monoclonal antibodies
whose specificity vas other than DNP such as anti-H^K IgG
or anti-AM IgG. This putative anti-id had no specfificity
for mouse immunoglobulin heavy or light chain constant
region determinants as indicated by the failure to detect
any activity when an allotype and isotype match
non-anti-DNP antibody was used in the assay as antigen.
These findings suggested that this mouse binding protein
was anti-idiotypic in nature.
When the serum which contained this anti-id vas
assayed to determine if this antibody could bind to other
mouse anti-DNP monoclonal antibodies, there was no detect
able binding to two of the mouse anti-DNP monoclonal anti
bodies and a limited binding to the monoclonal anti-DNP IgM
antibodies. The difference in the level of anti-id
detected when either the anti-DNP IgG or the anti-IgM anti
body was used as the antigen indicates the difference in

94
the ability of the anti-id to bind to these two antigens.
This difference could be a function of 1) different idio
typic determinants present on the two antibodies, or, 2)
the difference in the accessability of tine id to the
anti-id or 3) a combination of both of these.
Id/anti-id interactions can, in many cases, be inhi
bited by hapten. If tine interaction is hapten inhibitable,
it suggests that anti-id binds to id determinants within
the antigen combining portion of an antibody molecule or to
idiotypes intimately associated with this region. In those
instances where hapten is unable to inhibit this inter
action, it can be concluded that anti-id is binding to
those ids not within the antigen binding site. It also
indicates that anti-id does not act as an internal image of
antigen. High concentrations of hapten relative to the
amount of antibody on the plate were preincubated with
mouse anti-DNP IgG antibody. The presence of hapten did
not consistently inhibit canine anti-id/mouse id inter
action. Only two of the six samples tested showed inhi
bition, with 27 percent being the maximum inhibition. In
other experiments, hapten could not displace the anti-id
from the mouse anti-DNP antibody. These results suggest
that the majority of anti-id is not binding to structures
within the antigen binding site of the mouse monoclonal
anti-DNP IgG. An anti-id and an internal image of antigen
both bind to structures within the variable regions of an

95
antibody molecule. However, an internal image of antigen
binds to the hypervariable regions associated with the
antigen combining site.
If this antibody was an internal image of antigen
then the activity should have been detected when each anti-
DNP antibody was used as the antigen. Also, a hapten
should inhibit the binding of an internal image of antigen
to the respective antibody. Since anti-id bound to only
two anti-DNP mouse monoclonal antibodies and the id-anti-id
interactions ware not consistently inhibited by hapten, it
can be concluded that this anti-id is not an internal image
of antigen.
Although hapten could not inhibit the id/anti-id
interactions in all cases, indicating that the recognized
idiotopes were not within the antigen combining sites,
these id may be very close to the antigen combining site.
Serum which contained anti-id was assayed to determine if
the sample could interfere with the interaction between the
mouse anti-DNP antibody and a radiolabelled dinitro-
phenylated antigen. In all samples containing anti-id, the
level of antigen bound to the mouse antibody was decreased,
although complete inhibition of this binding was not
observed. Anti-id could consistently inhibit antibody/-
antigen interactions. However, the id/anti-id interactions
were not hapten inhibitable. Therefore, some of these
anti-ids must bind to id determinants which are close to

96
the antigen combining site and other anti-ids bind to ids
that are more distant. This failure to observe complete
inhibition could have been for at least two reasons. There
was not sufficient anti-id in the serum to block all the
antigen binding sites. Alternatively, the anti-id bound to
id determinants located on the molecule in such a way that
complete inhibition of the antigen binding site was not
possible.
The results in this chapter show that those animals
given autologous antibody in adjuvant produced an anti-id.
This anti-id response is not a function of the adminis
tration of adjuvant because control dogs given adjuvant
without antibody failed to produce detectable levels of
anti-id. In those dogs that produced anti-id the autol
ogous antibody that was used for immunization had been
subjected to harsh treatment (e.g. glycine HCl elution from
an affinity column) during the purification process. This
treatment could possibly alter the ids present in the anti
body. Therefore it might be possible that the antigen
specificity of the antibody has little or nothing to do
with the anti-id that is produced.
To address this question, dogs were immunized to two
different antigens, DNP-ASC and ABA-KLH. They were then
given either autologous anti-DNP antibody or autologous
anti-ABA antibody emulsified in adjuvant and subsequently
produced an anti-id which bound to mouse monoclonal

97
antibody of the same specificity as the immunizing
antibody. Prior to such treatment this autologous antibody
is present in the dog but does not induce detectable levels
of anti-id. However, after the administration of this same
antibody, anti-id is detected. Therefore, these results
indicate that during the purification procedure and/or the
immunization procedure, anti-id determinants present on the
autologous antibody are immunogenically enhanced. However,
any changes that occur in the protein molecule must be
subtle because the specificity of the anti-id response was
determined by the specificity of the immunizing antibody.
If there was narked change in the id determinants, then the
anti-id might not be expected to maintain its specificity
for the immunizing antibody. However, it is not possible
from these results to determine if the immunogenic
enhancement of the id was a result of the purification
process, the route of immunization or a combination of both
these things.
It was very fortunate that the mouse antibody used as
the antigen in these assays had id determinants which could
bind to the canine anti-id, although shared idiotypy
between animals has been reported (53,57,85-87).

98
Summary
The results in this chapter suggest the following:
Anti-id can be induced by the administration of autologous
antibody. The majority of this anti-id is not hapten
inhibitable and is detectable with only a few monoclonal
antibodies of the same specificity which presumably bear
tiie same or a similar set of cross reactive idiotypes.
Furthermore, this anti-id is not an internal image of
antigen.
Conclusions
1) The administration of autologous antibody in
adjuvant induces a reciprocal anti-idiotypic antibody
response. 2) The identification of these anti-id anti
bodies was achieved by the use of monoclonal anti-DNP
antibody from another species as a source of idiotype.

CHAPTER FIVE
DETECTION OF ANTI-IDIOTYPIC ANTIBODY
USING AUTOLOGOUS ANTI-DNP F(ab)'2 FRAGMENTS
AS THE IDIOTYPIC ANTIGEN
Introduction
Anti-idiotypic antibody, as noted in Chapter one, can
have a regulatory function during an inmune response. It
can either enhance (62) or suppress (34,88) the level of
the corresponding idiotype. Even if anti-id stimulates
only a limited number of ids, the overall result is an
enhancement in the total antigen-specific antibody response
(63). Anti-id that acts as an internal image of antigen
can stimulate or enhance an immune response in a way anala-
gous to antigen (89). On the other hand, anti-id has been
shown to suppress an entire antigen-specific isotype
(60-61). If anti-id is important as a natural means to
regulate antigen-specific antibody, it would seem logical
that anti-id would be detectable during a normal immune
response. The results of Chapter four indicated that after
99

100
immunizing dogs with autologous anti-DNP antibody, anti-id
was detected. However, this anti-id was not detected
during the response to DNP-ASC.
The failure to detect anti-id prior to the immuni
zation with antibody in adjuvant could be because; 1)
anti-id was rot present or, 2) the method used to detect it
was wrong. The purpose of the experiments in this chapter
were to determine if anti-id could be detected at any time
during a DNP specific antibody response using autologous
anti-DNP FCab)^ antibody fragments as the source of id.
Materials and Methods
Preparation and Immobilization of Anti-DNP P(ab)'^
Fragments to a solid Matrix
Anti-DNP Fiab)^ fragments were prepared from anti
body purified from a single serum sample by affinity chroma
tography. The protein was digested with pepsin and the
Fiab)^ was separated from intact antibody and Fc frag
ments as previously described in Chapter two. The
F(ab)'2 from each sample was handled separately and
F(ab)'2 from a single sample will be referred to as a
set. Each set of antibody fragments was bound to a .solid
support matrix (Immunobeadr. BioRad Laboratories,
Richmond, CA) as described by the rranufacturer. Briefly, a
given quantity of anti-DNP F(ab)'2
in 0.003 M

101
KH2(P04) buffer, pH 6.3 and a proportionate amount of
beads were incubated together for one hour at 4C
followed by the addition of l-ethyl-3 (3-dimethylamino
propyl) carbodiimide HC1 (EDAC) with an additional incu
bation at 4C for one hour. Any remaining active sites
were blocked with 1 percent HSA in 0.005 phosphate buffer,
pH 7.2 by incubating this with the beads for one hour at
roan temperature. The beads were pelleted by centri
fugation at 1,000 x g for 10 minutes at 4C and alter
nately washed with PBS, pH 7.2 followed by 1.4 M NaCl-PBS,
pH 7.2 three times to ramove unbound protein. After the
final wash the beads were suspended in RAST+ buffer. The
beads used in a single experiment were standardized for
both DNP-HSA binding and total F(ab)1^ content by incu
bating an aliquot from each bead set with various dilutions
of I DNP-HSA or I anti-canine light chain anti
body. For example, 50 pg of one bead set bound 5,656 +
61 cpm radiolabelled anti-canine light chain specific
antibody (this number of cpm is approximately 140 ng of
anti-canine light chain antibody) and 1,040 + 53 cprn
radiolabelled DNP-HSA. A second set bound 4,690 + 79 cpm
anti-light chain antibody (this is approximately 110 ng of
anti-canine light chain antibody), and 747 + 21 cpm
antigen. The second set of beads had approximately 75
percent of the binding capacity of the first set.
Therefore 63 pi of beads from the second set were used in

the assay and 50 pi of the first- No immuno- react i ve Fc
125
material was detectable on any bead set vhen I
anti-canine heavy chain specific IgG was incubated with an
aliquot of each-bead set.
Detection of Natural Occurinq Anti-id
Anti-DNP F(ab)*2 fragments from a single serum
sample immobilized as described above vas used as an
antigen to detect anti-id. Various serum samples frcm the
same dog were assayed for anti-id after being chromato
graphed through a DNP-affinity column to remove anti-DNP
antibody which theoretically could compete by binding
anti-id. Each sample was concentrated by negative pressure
dialysis to approximately the starting volume of serum.
The samples were assayed by incubating an undilute, a 1/2
and a 1/4 dilution in PBS of each sample with a
standardized amount of autologous anti-DNP Ftab)^ bound
beads for three hours at room temperature. The beads were
then centrifuged and the supernatant removed and washed
with RAST+ three times to renove unbound antibody. Radio-
labelled heavy chain specific anti-canine IgG was added to
each set of beads (approximately 30,000 cpm/sample),
incubated for three hours at roan temperature and vashed to
remove unbound radiolabelled antibody. The radioactivity
of each sample vas determined in a Packard gamma counter.
Included in each assay at all sample dilutions were beads

103
bound with normal canine IgG F(ab)'^ (with no detectable
anti-DNP activity) and with HSA bound beads. Specific
binding was calculated by using the following formula:
specific binding = cpm bound to autologous anti-DNP
Fiab)^ beads of a sample at a given dilution cpm bound
to NCS IgG Fiab)^ beads of the sample at the same
dilution. Since beads were standardized for an amount of
antibody in each experiment, the volume of beads used
ranged from 50 pi to 78 pi per sample. When 50 or 100
pi of non-specific Fiab)^ was incubated with the
sample, there was less than a 15 percent difference in the
cpm indicating that the increase in bead volume had little
influence on the background activity.
Results
Identification of Anti-Idiotypic Antibody Using
Autologous Idiotypes
The purpose of these experiments was to determine if
autologous id could be used to detect anti-id. Anti-DNP
FCab)^ fragments from various time points in the immuni
zation schedule were used as antigens to detect anti-id in
autologous serum. The autologous serum used in these
experiments were first chromatographed through a DNP
affinity column bo remove anti-DNP antibody. This was done
to eliminate any possible interference the presence of this

104
antibody might have. Anti-id was detected during the
DNP-ASC immunization schedule in three of the five dogs
tested (tables 18-20) but no anti-id was evident in the
other two dogs. The kinetics and the amount of anti-id
varied depending upon what set of ids were used an antigens.
and which serum sample was tested. Three different
patterns in the appearance of anti-id are seen: Pattern
one: Anti-id could not be detected before or coincident
with the id but could be detected later, as was seen with
two samples in two dogs (figure 13,14). In dog 14, the ids
used to detect anti-id were from week two, anti-id was not
detected until week seven (figure 13). Similarly, in dog
1, when the ids from week four were used as antigens,
anti-id was rot detected until week six (figure 14).
Pattern two w/as seen in three dogs using six serum samples.
In a single sample, id and anti-id were both present
(figure 15-20). For example, when the antibody obtained at
week six was used as an id antigen, anti-id was detected at
week six but the maximum level of anti-id was later than
week six (figure 15). In two samples assayed, the highest
level of anti-id was detected from the same samples that
were used to obtain the antibody which was used as the id
antigen (figures 19 and 20). Pattern three: In dog 14,
when the ids which were used as antigens to detect anti-id
w?ere from blood obtained at week 11, anti-id was detected
with each sample tested (figure 21). Similarly, by using

105
Table 18
The Detection of Canine Anti-Idiotypic Antibody
by RIA Using Autologous Anti-DNP Fiab)'^ as the Id
Dog Number 1
Source of Anti-id (Week) a)
2 4 6 8 11
Source of
id
(Week) b)
Effluent
Dilution
4
0
82+64
12+12
1332+112
2035+279
150+48
1/2
41+70
51+38
939+99
1519+199
286+45
1/4
63+49
0
171+34
1069+170
14+18
6
0
37+31
193+39
611+50
1241+179
1586+42
1/2
14+13
86+55
534+37
370+27
1333+26
1/4
10+21
3+5
5835
12+25
413+53
7
0
3+5
253+26
455+37
1496+14
1300+17
1/2
43+12
279+121
397+61
534+29
714+51
1/4
38+27
21+28
179+41
30+21
346+15
10
0
51+25
426+41
556+47
595+257
459+101
1/2
14+23
349+153
438+32
437+89
257+57
1/4
150+63
79+71
214+83
139+81
179+68
Control c)
0 186+9
1/2 149+31
1/4 138+46
101+7
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP F(ab)' immobilized to a solid
matrix. The dog was immunized with DNP-ASC in adjuvant at week
0,2,4,6,8 and 10 and received 10 gg autologous anti-DNP antibody in
CFA at weeks 7 and 9.
c) Control id was normal canine IgG Ftab)^ immobilized to a solid
matrix.

106
Table 19
Detection of Canine Anti-Iditoypic Antibody by RIA
Using Autologous Anti-DNP Fiab)^ as the Id
.Dog Number 14
Source of Anti-Id (Week) a)
2 4 7 11
Source of
id
(Week) b)
Effluent
Dilution
0
137+31
65+21
644+68
705+18
1/2
36+18
7+7
152+43
351+47
1/4
0
5+7
12+11
90+15
0
216+39
318+49
381+68
848+9
1/2
135+23
37+31
150+62
677+34
1/4
77+64
0
47+15
300+23
0
624+5
778+18
838+52
405+77
1/2
309+7
653+28
786+38
233+16
1/4
99+29
386+16
493+14
83+11
Control c)
0 321+37
1/2 186+7
1/4 128+29
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP F(ab)' immobilized to a solid
matrix. The dog was immunized with DNP-ASC in adjuvant at weeks
0,2,4,6,8, and 10 and received 100 gg autologous anti-DNP antibody
in CFA at weeks 7 and 9.
c) Control id was normal canine IgG F(ab)' immobilized to a solid
matrix.

107
Table 20
The Detection of Canine Anti-Iditoypic Antibody by RIA
Using Autologous Anti-DNP F(ab'^ as the Id
Dog Number 21
Source of Anti-Id (Week) a)
1 3 5 7 10
Source of
id
(Week) b)
Effluent
Dilution
1
0
358+94
1000+49
1/2
247+37
539+97
1/4
88+23
102+61
3
0
253+63
1532+88
1/2
77+31
865+109
1/4
83+71
276+123
7
0
104+60
0
1/2
0
0
1/4
46+29
0
11
0
49+41
0
1/2
0
65+58
1/4
108+39
17+11
Control
c)
0
238+83
1/2
211+61
1/4
103+48
307+69
435+146
684+33
204+39
197+55
377+36
35+7
27+26
246+29
1031+63
299+27
31+28
940+39
5+3
150+123
220+47
2+4
121+38
0
1466+229
1077+10
0
364+97
591+101
75+50
169+11
359+66
89+31
0
0
28+49
69+63
64+57
21+43
101+34
106+29
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP Ftab)^ immobilized to a solid
matrix. The dog was immunized to DNP-ASC in adjuvant at weeks
0,2,4,6,8, and 10 and received CFA alone at weeks 7 and 9.
c) Control id was normal canine IgG F(ab)' immobilized to a solid
matrix.

Figure 13.
The identification of anti-id in various serum samples over time in
dog 14 as measured by RIA. The dog received DNP-ASC in adjuvant at
weeks 0,2,4,6,8 and 10, and autologous anti-DNP antibody in
adjuvant at weeks 7 and 9. The arrow, marked idiotype probe,
indicates the time from which the anti-DNP F(ab)'2 fragments cane.
These were used as antigens to detect anti-id. The bars represent
the standard deviation of the mean.

10
o 8

Figure 14.
The identification of anti-id in various serum samples over time,
in dag 1, as measured by RIA. The dog received DNP-ASC in adjuvant
at weeks 0,2,4,6,8 and 10 and autologous antibody in adjuvant at
weeks 7 and 9. The arrow, narked idiotype probe, indicates the
time from which the anti-DNP F(ab)'2 fragments came. These were
used as antigen to detect anti-id. The bars represent the standard
deviation of the mean.

Ill

Figure 15.
The identification of anti-id in various samples over time, in dog
1, as measured by RIA. The dog received DNP-ASC in adjuvant at
weeks 0,2,4,6,8 and 10, and autologous anti-DNP antibody in
adjuvant at weeks 7 and 9. The arrow, marked idiotype probe,
indicates the time from which the anti-DNP F(ab)'2 fragments came.
These were used as antigens to detect anti-id. The bars represent
the standard deviation of the mean.

113

Figure 16.
The identification of anti-id in various serum samples, in dog 1,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous antibody in adjuvant at weeks 7
and 9. The arrow, marked idiotype probe, indicates the time from
which the anti-DNP F(ab)'2 came. These were used as antigens to
detect anti-id. The bars represent the standard deviations of the
mean.

0 2 4 6 8 10
Weeks
115

Figure 17.
The identification of anti-id in various serum samples over time,
in dog 14, as measured by RIA. The dog received DNP-ASC in
adjuvant at weeks 0,2,4,6,8 and 10, and autologous anti-DNP
antibody in adjuvant at weeks 7 and 9. The arrow, marked idiotype
probe, indicates the time from which the anti-DNP F(ab)'2 fragments
came. These were used as antigens to detect anti-id. The bars
represent the standard deviation of the mean.

cpm x 100
0 2 4 6 8 10 12
Weeks

Figure 18.
The identification of anti-id in various serum samples, as measured
in dog 24, by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and CFA at weeks 7 and 9. The arrow, marked
idiotype probe, indicates the time from which the anti-DNP F(ab)'2
fragments came. These were used as antigens to detect anti-id.
The bars represent the standard deviation of the mean.

119

Figure 19.
The identification of anti-id in various serum
samples over time, in dog 24, as measured by RIA.
The dogs received DNP-ASC in adjuvant at weeks
0,2,4,6,8,10, and CFA at weeks 7 and 9. The arrow,
narked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used
as antigens to detect anti-id. The bars represent
the standard deviation of the mean.

121

Figure 20.
The identification of anti-id in various serum
samples, in dog 24, as measured by RIA. The dog
received DNP-ASC in adjuvant at weeks 0,2,4,6,8,10,
and CFA at weeks 7 and 9. The arrow, marked idiotype
probe, indicates the time from which the anti-DNP
F(ab)'2 fragments come. These were used as antigens
to detect anti-id. The bars represent the standard
deviation of the mean.

123

Figure 21.
The identification of anti-id in various serum samples, in dog 14,
as measured by RIA. THe dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous anti-DNP antibody at weeks 7 and
9. The arrow, marked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used as antigens
to detect anti-id. The bars represent the standard deviation of
the mean.

10
o
o
x
8
6
2
0 2 4 6 8 10
Weeks

Figure 22.
The identification of anti-id in various serum samples, in dog 1,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous anti-DNP antibody at weeks 7 and
9. The arrow, marked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used as antigens
to detect anti-id. The bars represent the standard deviation of
the mean.

127

Figure 23.
The identification of anti-id in various serum samples, in dog 24,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and CFA at weeks 7 and 9. The arrow, marked
idiotype probe, indicates the time from which the anti-DNP F(ab)'2
fragments came. These were used as antigens to detect anti-id.
The bars represent the standard deviation of the mean.

10
Weeks
¡diotype
probe
8 10
129

130
ids obtained from dog 1 at week 10, anti-id was detected in
all but the first sample tested (figure 22). Both of these
dogs had received autologous antibody in adjuvant. Eoth
these ids were obtained from the dogs after this treatment.
In contrast to the anti-id detected in these two dogs, id
was obtained from dog 24 at a similar time in the immuni
zation schedule (week 11), but revealed no anti-id using
these id antigens. This dog received adjuvant without
autologous antibody. In two dogs there was no detectable
anti-id at any point in the immunization schedule using
autologous id as the probe. This is in spite of the fact
that one of the dogs was immunized with autologous antibody
and did have detectable levels of anti-id using the mouse
monoclonal antibody as the .id probe.
Discussion
In the experiments in Chapter four, anti-id was only
detected in serum after autologous antibody administration.
If anti-id does serve in a regulatory fashion, it is not
clear why its presence is not detected during the course of
the anti-DNP antibody response. This inability to detect
anti-id throughout the immunization schedule itay.be a
function of the xenogeneic probe used to detect anti-id or
it may be because anti-id is present only after artificial
manipulation. Furthermore, if anti-id does regulate the

131
reciprocal id, then it would be expected that id would be
present in the serum before the appearance of anti-id, but
after the appearance of anti-id the reciprocal id would
disappear. Since a number of different ids were used as
the probe, the corresponding anti-id may be detected either
slightly before the point in time the id came from, coin
cident with the anti-id, or considerably later in time than
the id. Alternatively, if id/anti-id were complexed then
the disruption of these complexes might allow anti-id to be
detected.
When autologous anti-DMP Fiab)'^ was used to assay
for anti-id, three different patterns in the detection of
anti-id were evident. In the first pattern, anti-id was
detected after the appearance of id, but not coincident
with nor before its appearance. These results suggest that
there vas a lag phase between the appearance of id and the
corresponding anti-id in the serum. In the second pattern,
id and anti-id are present within a single serum sample.
The maximum level of anti-id vas detected in serum at the
same time that the id appeared in two cases. Furthermore,
anti-id were present very early in the response. Because
the dogs in these experiments would still be expected to
have colostrum-derived antibody, these anti-id may
represent maternal Immunoglobulin. Unfortunately, it was
not possible to obtain serum from these bitches to
determine if they had anti-id present.

132
Both the first and second pattern of anti-id are
consistent with the hypothesis that id is acting as an
antigen to induce anti-id. These data are raniniscent of
the type of curves seen when one plots the disappearance of
antigen as a function of time and superimposes on that
curve the appearance of antibody that is specific for the
antigen (90). When antigen is first introduced into an
animal there is initially a very slow loss of this antigen
from the circulation. After a few days, however, there is
a precipitous drop in the level of antigens which is the
result of antibody production and is called immune elimi
nation. Antibody when first produced is not detected
because it is cornplexed with antigen and removed from the
circulation. At a certain point, however, both antibody
and antigen will become apparent in a cornplexed form. The
variables that determine this point include the valence of
the antigen and its size, the isotype of the antibody, the
affinities between the antibody and the antigen, and the
efficiency of the reticuloendothelial system in removing
these complexes (90,91). The similarity between immune
elimination of antigen and the experimental results
obtained suggest that id is removed in a fashion analagous
with antigen removal. The appearance of anti-id in the
third pattern is difficult to explain. In this case,
anti-id was present at a time considerably before the
appearance of id. That is, in dog 14, anti-id was

133
detectable in every serum sample checked (figure 21), and
in dog 1, anti-id was detectable in all but the very
earliest serum sample (figure 21). Both of these dogs had
received autologous antibody in adjuvant and the idiotype
used as the antigen was obtained from serum after such
treatment. In contrast, when idiotypes from a comparable
time in the immunization schedule were used as antigen
from, dog 24, which received adjuvant without autologous
antibody, the unusual appearance of anti-id was not
observed. A possible explanation would be that the
administration of autologous antibody in adjuvant
stimulated the production of antibody having similar
idiotypes. That is, anti-DNP antibodies used for
immunization were from week six. The ids on this
immunizing antibody may have stimulated additional antibody
with the same id. Any anti-id which would be produced
because of the id of week six might also bind to ids
produced from the immunization of the antibody from week
six. Therefore, if this were the case, anti-id could be
present in serum prior to the sample from which the id was
derived. There is experimental precedence for this
suggestion (63,64). Forney et al. (63) have shown that
mice given hybridoma-derived anti-sheep red blood cell anti
body without stimulation with antigen will subsequently
produce anti-SRBC antibody of a similar idiotypic speci
ficity to the immunizing antibody. Their interpretation

134
was that the antibody stimulated the subsequent production
of identical or very similar antibody through an id/anti-id
interaction Therefore, this unusual pattern in the
appearance of anti-id could be the result of autologous
antibody administration. However, the presence of this
anti-id my be the result of factors governing the
production and detection of anti-id which are unforseen at
this time.
In two of the five animals there is complete failure
to detect anti-id using autologous antibody in any serum
obtained throughout the immunization schedule. There are a
number of different possible explanations for this result.
Firstly, only anti-id of the IgG class was measured and it
is possible that other anti-id isotypes were produced in
these two animals. In fact, in a recent study it was
observed that in man there was an isotypic shift over time
of the anti-id specific for a given set of auto-antibodies
(53). Alternately, there my be certain ids which favor
the production of the reciprocal anti-ids. In an experi
ment in outbred rabbits, anti-id production seemed to be
associated with the presence of a few ids that favor
anti-id production. Those rabbits not expressing such ids
failed to produce detectable anti-id antibodies (57).
.Based on this, it is possible that, in dogs, certain id are
especially important for anti-id production and in those
dogs not expressing such ids there is a failure to produce

135
reciprocal anti-id. This lack of anti-id would not
necessarily result in abnormal antibody regulation if these
animals had an alternate means to accomplish this, such as
a T-suppressor cell pathway.
One of these animals had no recognizable anti-id
using autologous antibody as the id probe, but did have
recognizable anti-id when mouse monoclonal antibody was
used as the id probe. There are several possible reasons
for this. 1) There are only a few idiotopes present on
monoclonal antibodies while in a heterogeneous population
of molecules there would be expected to be many idiotypes
and therefore even more idiotopes. Therefore, this
negative result may be a function of the concentration of
id present in the assay system. 2) The purification
process may have altered the id on the antibody molecules
just enough so that when this antibody was used to immunize
a dog, these altered id were able to induce an anti-id
response specific for the mouse id. Or 3) the ids
detected with the mouse antibody were different than the
ids on the canine anti-DNP Fiab)^.
Sunznary and Conclusions
In three of five dogs immunized with DNP-ASC, anti-id
could be detected using autologous anti-DNP Fiab)^

fragments as the id. The kinetics in the appearance of
this anti-id in relationship to the id suggest that id is
acting as an antigen to stimulate a corresponding anti-id
response.

CHAPrER SIX
CONCLUSION
Dogs were chosen to study IgE antibody synthesis and
regulation because they develop an IgE mediated disease
that is very similar to atopic disease of nan.
When dogs were immunized with 100 gg of aluminum
hydroxide precipitated DNP-ASC by the intraperitoneal
route, each dog synthesized anti-DNP IgG, IgE and IgM anti
body. There ^as a difference in the responsiveness to this
antigen between individual dogs Which most probably
reflected the genetic heterogeneity between them. In an
attempt to regulate anti-DNP antibody, autologous anti-DNP
antibody in adjuvant was administered to these dogs during
the ongoing response. Although dogs so treated did not
have a difference in the magnitude of the anti-DNP IgE or
IgG response, as compared to control dogs who received
adjuvant without antibody, these dogs did produce an
anti-id which was detected using mouse monoclonal anti-DNP
IgG and IgM.
This anti-id was not an internal image of antigen as
shown by 1) a failure of these antibodies to bind to each
137

138
mouse monoclonal anti-DNP antibody tested and 2) a failure
of hapten to inhibit the id/anti-id interaction. An
internal image of antigen should bind to corresponding anti
body to an extent similar to antigen and should also be
hapten inhibitable (91).
Anti-idiotypic antibody could be detected during the
DNP-ASC immune response when the ids used to assay for anti
body were autologous anti-DNP F(ab)'2 fragments. These
anti-ids can be considered natural in that their appearance
was associated with anti-DNP antibody produced during an
immune response. This is in contrast to the anti-ids
detected with the mouse antibodies. These latter anti-ids
were detected subsequent to the immunization with anti-DNP
antibody in complete Freund's adjuvant. Three of the five
dogs assayed for anti-id using autologous ids had detect
able amounts of anti-id whereas two dogs had no detectable
levels. It was concluded that anti-id can be detected
during a DNP-ASC response in dogs.
This method of detecting anti-id would allow for the
identification of a different anti-id response during other
immune responses. For example, the aim of hyposensi
tization for allergic disease is felt to be the production
of a blocking antibody. However, this treatment is not
always effective, even if blocking antibody is produced.
An additional possible reason why hyposensitization works

139
is because IgE antibody is regulated by id/anti-id inter
action. If this is the case, the identification of anti-id
during hyposensitization may help in establishing more
effective immunotherapy for allergies.

REFERENCES
1. Patterson, R. (1969) Laboratory Models of Reaginic
Allergy. Prog. .Allergy. 13:332-407.
2. Patterson, R., and Kelly, J.F. (1974) Animal Models of
the Asthmatic State. Ann. Rev. Med. 25:53-58.
3. Pepys, J. (1975) Atopy in Clinical Aspects of
Immunology, third ed. Edited by Gell P.G.H., Coombs,
R.R.A., Lachmann, P.J. Blackwell Scientific Publications,
Oxford, England.
4. Schwartzman, R.M. and Rockey, J.H. (1972) Atopy in the
Dog. Arch. Derm. 96:413-422
5. Halliwell, R.E.W., Schwartzman, R.M., and Rockey, J.H.
(1972) Antigenic Relationship between Human IgE and Canine
IgE. Clin. Exp. Immunol. 10:399-407.
6. Halliwell, R.E.W., Schwartzman, R.M., Montgomery, P.C.,
and Rockey, J.H. (1975) Physicochemical Properties of
Canine IgE. Transplant Proc. 8:537-543.
7. Halliwell, R.E.W. (1972) Studies on Canine IgE.
Ph.D dissertation. Cambridge University, Cambridge,
England.
8. Johansson, S.G.O. and Mel1bin, T., Vahlquist, B. (1963)
Immunoglobulin Levels in Ethiopian Preschool Children with
Special Reference to High Concentrations of Immunoglobulin
E. (IgND). Lancet 1:1118-1121.
9. Rosenberg, E.B., Whalen, G.E., Bennich, H. and
Johansson, S.G.O. (1971) Increased Circulating IgE in a
New Parasitic DiseaseHunan Intestinal Capillariasis. N.
Eng. J. Med. 283:1148-1149.
10. Tada, T., Ishizaka, K. (1970) Distribution of Gamma E
Forming Cells in Lymphoid Tissues of the Human and Monkey.
J. Immunol. 104:377-387.
11. Jarrett, E.E.E., and Miller, H.R.P. (1982) Production
and Activities of IgE in Helminth Infection. Prog Allergy
31:178-233.

141
12. Katz, D.H. (1980) Recent Studies on the Regulation of
IgE Antibody Synthesis in Experimental Animals and Man.
Immunol. 41:1-24 .
13. Ishizaka, K. (1976) Cellular Events in the IgE
Antibody Response. Adv. Immunol. 23:1-75.
14. Shinohara, N. and Tada, T. (1974) Hapten-Specific IgM
and IgG Antibody Responses in Mice against a
Thymus-Independent Antigen (DNP-Salmonella). Int. Arch
Allergy Appl. Immunol. 47:762-776.
15. Suemura, M., Yodai, J., Hirashima, M. and Ishizaka, K.
(1980) Regulatory Role of IgE binding Factors from Rat
T-lymphocytes. 1. Mechanism of Enhancement of IgE Response
by IgE-Potentiating Factor. J. Inmunol. 125:148-154.
16. Saryan, J.A., Leung, D.Y.M., and Geha, R.S. (1983)
Induction of Human IgE Synthesis by a Factor Derived from
T-cells of Patients with Hyper-IgE States. J. Immunol.
130:242-247.
17. Katz, D.H., Davie, J.M., Paul, W.E., and Benacerraf, B.
(1971) Carrier Function in Anti-Hapten Antibody Responses.
IV Experimental Conditions for the Induction of
Hapten-specific Tolerance or for the Stimulations of
Anti-Hapten Anmnestic Responses by 'non-immunogenic'
Hapten-Polypeptide Conjugates. J. Exp. Med. 134:201-224.
18. Sehon, A.H. and, Lee, W.Y. (1981) Specific
Downregulation of IgE Antibodies by Tolerogenic Conjugates
of Haptens and Allergens with Synthetic Polymers. Int. Arch
Allergy Appl. Immunol. 66 (suppl. l):39-42.
19. Liu, F.T., Bogowitz, C.A., Bargatze, R.F., Zinnecher,
M., Katz, L.R., and Katz, D.H. (1979) Immunologic
Tolerance to Allergenic Protein Determinants: Properties of
Tolerance Induced in Mice Treated with Conjugates of Protein
and a Synthetic Copolymer of D-Glutamic Acid and D-Lysine.
J. Immunol. 123:2456-2465.
20. Zeiss, C.R. (1979) Immunotherapy in Allergic Disease.
In Allergy and Clinical Immunology, Medical Examination
Publ. Co., Inc., Garden City, N.Y.
21. Halliwell, R.E.W. (1977) Hyposensitization in the
Treatment of Atopic Disease. In Current Veterinary Therapy
VI. W.B. Saunders Co., Phila, Pa.

Int.
22. Faith, R.E., Hessler, J.R., Small, P.A. (1977)
Respiratory Allergy in the Dog: Induction by the
Respiratory Route and the Effect of Passive Antibody.
Arch. Allergy Appl. Immunol. 53:530-544.
23. Smith, T. (1909) Active Immunity Produced by
So-Called Balanced or Neutral Mixtures of Diphtheria Toxin
and Antitoxin. J. Exp. Med. 11:241-256.
24. Uhr, J.W., and Moller, G. (1968) Regulatory Effect of
Antibody on the Immune Response. Adv. Immunol. 8:81-127.
25. Chan, P.L., and Sinclair N.R. St. C. (1971)
Regulation of the Immune Response. V. An Analysis of the
Function of the Fc Portion of Antibody in Suppression of an
Immune Response with Respect to Interaction with Components
of the Lymphoid System. Immunol. 21:967-981.
26. Strannegard, 0., and Belin, L. (1970) Suppression of
Reagin Synthesis in Rabbits by Passively Administered
Antibody. Immunol. 18:775-785.
27. Tada, T., and Okumura, K. (1971) Regulation of
Hcmocytotropic Antibody Formation in the Rat. 1. Feed-Back
Regulation by Passively Administered Antibody. J. Immunol.
106:1002-1011.
28. Ishizaka, K., and Okudaira, H. (1972) Reaginic
Antibody Formation in the Mouse. 1. Antibody-Mediated
Suppression of Reaginic Formation. J. Immunol. 109:84-89.
29. Lahss, F., Weiler, E., and Hillmann, G. (1953)
Myelom-Plasma-Proteine. Ill Mitteilung zur Immunochemie der
Gaima-Myelom-Proteine. Z. Naturforsch, Teil., 8B:625-631.
30. Jerne, N.K. (1974) Towards a Network Theory of the
Immune System. Ann. Immunol. 125(c):373-389 .
31. Brient, B.W. and Nisonoff, A. (1970) Quantitative
Investigations of Idiotypic Antibodies. IV. Inhibition by
Specific Haptens of the Reaction of Anti-Hapten Antibody
with its Anti-Idiotypic Antibody. J. Exp. Med.
132:951-963.
32. Claflin, J.L., and Davie, J.M. (1975) Specific
Isolation and Characterization of Antibody Directed to
Binding Site Antigenic Determinants. J. Immunol. 114:70-75.

143
33. Brient, B.W., Haimovich, J., and Nisonoff, A. (1971)
Reaction of Anti-Idiotypic Antibody with the Hapten-binding
Site of a Myeloma Protein. Proc. Natl. Acad. Sci. U.S.A.
68:3136-3139.
34. Briles, D.E., and Krause, R.M. (1974) Meuse
Strain-Specific Idiotypy and Intrastrain Idiotypic Cross
Reactions. J. Immunol. 113:522-530.
35. Sher, A., and Cohn, M. (1972) Effect of Haptens on
the Reaction of Anti-Idiotype Antibody with a Mouse
Anti-Phosphorylcholine Plasmocytoma Protein. J. Immunol.
109:176-178.
36. Eichmann, K., Coutinho, A., and Melchers, F. (1977)
Absolute Frequencies of Lipopolysaccharide-Reactive B Cells
Producing A5A Idiotype in Unprimed, Streptococcal A
Carbohydrate-Primed Anti-A5A Idiotype-Sensitized and
Anti-A5A Idiotype-Suppressed A/J Mice. J. Exp. Med.
146:1436-1449.
37. Bona, C., Mond, J.M., Stein, K.E., House, S., Liberman,
R., and Paul, W.E. (1979) Immune Response to Levan. Ill
The Capacity to Produce Anti-Inulin Antibodies and
Cross-Reactive Idiotypes Appears Late in Ontogeny. J.
Immunol. 123:1484-1490.
38. Marcu, K.B. (1982) Immunoglobulin Heavy-Chain
Constant-Region Genes. Cell 29:719-721.
39. Kolb, H., and Bosma, M.J. (1977) Clones Producing
Antibody of More than One Class. Immunol. 33:461-469.
40. Hopper, J.E., and Nisonoff, A. (1971) Individual
Antigenic Specifity of Immunoglobulins. Adv. Immunol.
13:57-99.
41. Capra, J.D., and Kohoe, J.M. (1975) Hypervariable
Regions, Idiotypy, and the Antibody-Combining Site. Adv.
Immunol. 20:1-40.
42. Eichmann, K. (1978) Expression and Function of
Idiotypes on Lymphocytes. Adv. Immunol. 26:195-254.
43. Janeway, C.A. (1980) Idiotypes, T-Cell Receptors and
T-B Cooperation. Contemp. Top. Immunobiol. 9:171-203.

44. Rodkey, L.S. (1980) Autoregulation of Inmune
Responses via Idiotype Network Interaction. Microbiol. Rev.
44:631-659.
45. Bankert, R.B., Bloor, A.G., and Jou, Y.H. (1982)
Idiotypes: Their Presence on B and T Lymphocytes and their
Role in the Regulation of the Immune Response. Vet.
Immunol. Immunopath. 3:147-184.
46. Greene, M.I., Nelles, M.J., and Man-Sun, S. (1982)
Regulation of Immunity to the Azobenzenearsonate Hapten.
Adv. Immunol. 32:254-300.
47. Lynch, R.G., Rohrer, J.W., Odermatt, B., Gebel, H.M.,
Autry, J.R., and Hoover, R.G. (1979) Immunoregulation of
Murine Myeloma Cell Growth and Differentiation: A
Monoclonal Model of B Cell Differentiation. Immunol. Rev.
48:45-80.
48. Cosenza, H., and Kohler, H. (1972) Specific
Inhibition of Plaque Formation to Phosphorylcholine by
Antibody Against Antibody. Science 176:1027-1029.
49. Cosenza, H., and Kohler, H. (1972) Specific
Suppression of the Antibody Response by Antibodies to
Receptors. Proc. Natl. Acad. Sci. U.S.A. 69:2701-2705.
50. Bankert, R.B., and Pressman, D. (1976)
Receptor-Blocking Factors Present in Immune Serum Resembling
Auto-Anti-Idiotype Antibody. J. Immunol. 117:457-462.
51. Kelsoe, G., and Cerny, J. (1979) Reciprocal
Expansions of Idiotypic and Anti-Idiotypic Clones Following
Antigen Stimulation. Nature. 279:333-334.
52. Geha, R.S. (1983) Presence of Circulating
Anti-Idiotype Bearing Cells after Booster Immunization with
Tetanus Toxoid (TT) and Inhibition of Anti-TT Antibody
Synthesis by Auto-Anti-Idiotypic Antibody. J. Immunol.
130-1634-1639.
53. Dwyer, D.S., Bradley, R.J., Urquhart, C.K., and
Kearney, J.F. (1983) Naturally Occurring Anti-Idiotypic
Antibodies in Myasthenia Gravis Patients. Nature
301:611-614.

54. Abdou, N.I., Wall, H., Lindsley, H.B., Halsey, J.N.,
and Suzuki, T. (1981) Network Theory in Autoimmunity. In
Vitro Suppression of Serum Anti-DNA Antibody Binding DNA
by Anti-Idiotypic Antibody in Systemic Lupus Erythematosus.
J. Clin. Invest. 67:1297-1304.
55. Cunningham-Rundles, C. (1982) Naturally Occuring
Autologous Anti-Idiotypic Antibodies: Participation in
Immune Complex Formation in Selective IgA Deficiency. J.
Exp. Med. 155:711-719.
56. Schrater, A.F., Goidl, E.A., Thorbecke, G.J., and
Siskind, G.W. (1979) Production of Auto-Anti-Idiotypic
Antibody during the Normal Immune Response to TNP-Ficoll. I
Occurrence in AKR/J and BALB/C Mice of Hapten-Augmentable,
Anti-TNP Plaque-Forming Cells and their Accelerated
Appearance in Recipients of Immune Spleen Cells. J. Exp.
Med. 150:138-153.
57. Binion, S.B., and Rodkey, L.S. (1982) Naturally
Induced Auto-Anti-Idiotypic Antibodies. Induction by
Identical Idiotypes in some Members of an Outbred Rabbit
Family. J. Exp. Med. 156:860-872.
58. Geczy, A.F., de Week, A.L., Geczy, C.L., and Toffler,
0. (1978) Suppression of Reaginic Antibody Formation in
Guinea Pigs by Anti-Idiotypic Antibodies. J. Allergy Clin.
Iirmunol. 62:261-270.
59. Blaser, K., Nakagawa, T., and de Week, A.L. (1981)
Suppression of Anti-Hapten IgE and IgG Antibody Responses by
Isologous Anti-Idiotypic Antibodies Against Purified
Anti-Carrier (Ovalbumin) Antibodies in BALB/C Mice. J.
Immunol. 126:1180-1184.
60. Blaser, K., Nakagawa, T., and de Week, A.L. (1980)
Suppression of the Benzylpenicelloyl (BPO) Specific IgE
Formation with Isologous Anti-Idiotypic Antibodies in BALB/C
Mice. J. Inrrtunol. 125:24-30.
61. Blaser, K., Geiser, M., and de Week, A.L. (1979)
Suppression of Phosphorylcholine Specific IgE Antibody
Formation in Balb/C Mice by Isologous Anti-T15 Antiserum.
Eur. J. Immunol. 9:1017-1020.
62. Eichmann, K., and Rajewsky, K. (1975) Induction of T
and B Cell Immunity by Anti-Idiotypic Antibody. Eur. J.
Immunol. 5:661-666.

63. Forni, L., Coutinho, A., Kohler, G., and Jerne, N.K.
(1980) IgM Antibody Induces the Production of Antibodies of
the Same Specificity. Proc. Natl. Acad Sci. U.S.A.
77:1125-1128.
64. Ivars, F., Holmberg, D., Forni, L., Cazenave, P.A., and
Coutinko, A. (1982) Antigen-Independent, IgM-Induced
Antibody Responses: Requirement for "Recurrent" Idiotypes.
Eur. J. Immunol. 12:146-151.
65. Sege, K., and Peterson P.A. (1978) Anti-Idiotypic
Antibodies Against Anti-Vitamin A Transporting Protein React
with Prealbumin. Nature. 271:167-168.
66. Schreiber, A., Andre, C., Vray, B., and Strosberg, A.D.
(1980) Anti-Alprenolol Anti-Idiotypic Antibodies Bind to
B-Adrenergic Receptors and Modulate Catecholamine Sensitive
Adenylate Cyclase. Proc. Natl. Acad. Sci. U.S.A.
77:7385-7389.
67. Schwartzman, R.M., Rockey, J.H., and Halliwell, R.E.W.
(1971) Canine Reaginic Antibody. Characterization of the
Spontaneous Anti-Ragweed and Induced Anti-Dinitrophenyl
Reaginic Antibodies of the Atopic Dog. Clin. Exp. Immunol.
9:549-569.
68. Bradford, M.M. (1976) A Rapid and Sensitive Method
for the Quantification of Microgram Quantities of Protein
Utilizing the Principle of Protein-Dye Binding. Anal
Biochem. 72:248-254.
69. Strejan, G., and Campbell, D.H. (1963)
Hypersensitivity to Ascaris Antigen. IV Production of
Hanocytotropic Antibodies in the Rat. J. Irrmunol.
101:628-637.
70. Williams, C.A., and Chase, M.W. (1967) Methods in
Immunology and Iirmunochemistry. Vol. 1, Academic Press,
New York, New York.
71. Imada, Y., Namoto, K., Yamada, H., and Takeya, K.
(1977) Relationships between IgE Antibody Production and
other Immune Responses. 1. Ontogenic Development of IgE
Antibody-Producing Capacity in Mice. Int. Arch Allergy
Appl. Immunol. 53:50-55.
72. Vaz, E.M., Vaz, N.M., and Levine B.B. (1971)
Persistent Formation of Reagins in Mice Injected with Low
Doses of Ovalbumin. Immunol. 21:11-15.

73.Revoltella, R., and Ovary, Z. (1971) Reagenic
Antibody Production in Different Mouse Strains. Immunol.
17:45-54.
14 7
74. Geha, R.S. (1981) Regulation of the Immune Response
by Idiotype-Antiidiotype Interactions. N. Eng. J. Med.
305:25-28.
75. Jarrett, E.E.E., and Ferguson, A. (1974) Effect of
T-Cell Depletion on the Potentiated Reagin Response. Nature
250:420-422.
76. Chiorazzi, N., Fox, D.A., and Katz, D.H. (1976)
Hapten-Specific Enhancement of IgE Antibody Production by
Low Doses of X-Irradiation and by Cyclophosphamide. J.
Immunol. 117:1629-1637.
77. Nagel, J.E., White, C., Lin, M.S., and Fireman, P.
(1979) IgE Synthesis in Man. II. Comparison of Tetanus and
Diptheria IgE Antibody in Allergic and Non-Allergic
Children. J. Allergy Clin. Immunol. 63:308-314.
78. Frich, O.L., and Brooks, D.L. (1983) Immunoglobulin E
Antibodies to Pollens Augmented in Dogs by Virus Vaccines.
Am. J. Vet. Res. 44:440-445.
79. Katz, D. (1978) The Allergic Phenotype.
Manifestation of "allergic breakthrough" and imbalance in
normal "Damping" of IgE Antibody Production. Immunol. Rev.
41:77-108.
80. Rowley, D.A., Fitch, F.W., Stuart, F.P., Kohler, H.,
and Cosenza, H. (1973) Specific Suppression of Immune
Responses. Science 181:1133-1141.
81. Voisin, G.A. (1980) Role of Antibody Classes in the
Regulatory Facilitation Reaction. Immunol. Rev. 49:3-59.
82. Hart, D.A., Pawlak, L.L. and Nisonoff, A. (1973)
Nature of Anti-Hapten Antibodies Arising after Immune
Suppression of a Set of Cross-Reactive Idiotypic
Specificities. Eur. J. Immunol. 3:44-48.
83. Jarrett, E.E.E., and Hall, E. (1983) IgE Suppression
by Maternal IgG. Immunol. 48:49-59.
84. Pawlak, L.L., Hart, D.A., and Nisonoff, A. (1973)
Requirements for Prolonged Suppression of an Idiotypic
Specificity in Adult Mice. J. Exp. Med. 137:1442-1458.

85. Claflin, J.L., and Davies, J.M. (1975) Clonal Nature
of the Inmune Response to Phosphorylcholine (PC). V.
Cross-Idiotypic Specificity among Heavy Chains of Murine
Anti-PC Antibodies and PC-Binding Myeloma Proteins. J. Exp.
Med. 141:1073-1083.
86. Karol, R.A., Reichlin, M., and Noble, R.W. (1977)
Evolution of an Idiotypic Determinant: Anti-Val. J. Exp.
Med. 146:435-444.
87. Schwartz, M., Norvick, D., Givol, D., and Fuchs, S.
(1978) Induction of Anti-Idiotypic Antibodies by
Inmunization with Syngeneic Spleen Cells Educated with
Acetylcholine Receptor. Nature 273:543-545.
88. Eichmann, K. (1974) Idiotype Suppression: I
Influence of the Effector Functions of Anti-Idiotypic
Antibody on the Production of an Idiotyoe. Eur. J. Inmunol.
4:296-302.
89. Nisonoff, A. and Lamoyi, E. (1981) Implications of
the Presence of an Internal Inage of the Antigen in
Anti-Idiotypic Antibodies: Possible Application to Vaccine
Production. Clin. Immunol, and Inmunopath. 21:397-406.
90. Eisen, H.N. (1980) Immunology. Second Edition.
Harper and Row Publishers, Hagerstown, Md.
91. Theofilopoulos, A.N., and Dixon, F.J. (1979) The
Biology and Detection of Inmune Complexes. Adv. Immunol.
28:39-220.

BIOGRAPHICAL SKETCH
My narre is Tippy Schultz. I am a border collie, spitz
dog who was an abused puppy scheduled to be euthanized on
the day I met my future owner, Kevin Schultz. He rescued me
from my fate and since he was in his first year of
veterinary school at Purdue University, I became his living
anatomical model to study dog topography. I learned that he
had been born in Chicago in 1951, and, because his family
moved alot, he attended many different grade and high
schools before finally moving to Fort Wayne, Indiana, where
he graduated from high school. We studied hard in
veterinary school and I sent Kevin to work at B.F. Goodrich
Tire Company during summers to keep me in food and shelter
which provided motivation for him to continue his education.
We graduated in 1976 and since that time, I have
gotten to do a great deal of traveling with him. In Dodge
City, Kansas, while he was in practice, he met a very pretty
and very nice woman. He saved her cat from dying after
being hit by a car, and since she could also type, they got
married and we all moved to Chicago so Kevin could practice

150
for a year We then hit the road for Philadelphia, Pa.,
where Kevin studied comparative dermatology at the
University of Pennsylvania, and he put Nancy to work typing.
In September, 1979, we packed up and moved again, this time
to Gainesville, Florida. Hera both Kevin and Nancy attended
the University of Florida and I didn't see much of either of
them while they were busy getting their degrees.
During our treks across the country, we have adopted
two permanent house guests, Fanny and Pippin, who are OK for
being just cats. In July, 1983, we will all be moving
again, this time to Boston, Massachusetts. There Kevin will
be a post-doctoral fellow in the Department of Pathology at
Harvard Medical School which is fine with me as long as he
continues to keep tie warm and fed.

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Richard E.W. Halliwell, Chairman
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
JL p A
George E. Gifford
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality, as
a dissertation for the degree of Doctor of Philosopj;
//,
8*,
Parker A/ Small-/ Jr //?'
Professor of Immunology and
Medical Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
\
Richard B. Crandall
Professor of Immunology and
Medical Microbiology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Michael D.P. dfoyle
Associate Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Michael J.P. Lawman
Assistant Professor of
Veterinary Medicine
This dissertation was submitted to the Graduate
Faculty of the College of Medicine and to the Graduate
Council, and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
Dean, College of Medicine
Dean, Gradate School
August, 1983

3/7y
/9f3



93
Discussion
Dogs immunized with autologous antibody produced an
antibody which bound to one- anti-DNP mouse monoclonal IgG.
This antibody was present only after such treatment and not
present prior to the administration of autologous antibody.
The specificity was limited to id determinants present on
some but not all anti-DNP mouse monoclonal antibodies. It
could not be detected using mouse monoclonal antibodies
whose specificity vas other than DNP such as anti-H^K IgG
or anti-AM IgG. This putative anti-id had no specfificity
for mouse immunoglobulin heavy or light chain constant
region determinants as indicated by the failure to detect
any activity when an allotype and isotype match
non-anti-DNP antibody was used in the assay as antigen.
These findings suggested that this mouse binding protein
was anti-idiotypic in nature.
When the serum which contained this anti-id vas
assayed to determine if this antibody could bind to other
mouse anti-DNP monoclonal antibodies, there was no detect
able binding to two of the mouse anti-DNP monoclonal anti
bodies and a limited binding to the monoclonal anti-DNP IgM
antibodies. The difference in the level of anti-id
detected when either the anti-DNP IgG or the anti-IgM anti
body was used as the antigen indicates the difference in


dog received a distemper-hepatitis modified live virus
vaccination at week four and eight.
31
Results
Standard Curve
The relative antigen-specific antibody concentration
was determined by interpolation from the linear portion of
the standard curve included with each assay. This serum
sample contained high levels of the isotype under inves
tigation. An example of a standard curve for anti-DNP IgE,
IgG and IgM is given in figures 4, 5 and 6.
Antibody Response
Twenty-eight dogs immunized with 100 pg DNP-ASC in
aluminum hydroxide developed an IgE, IgG and IgM serum anti
body response. The mean relative antibody concentration
for the three isotypes is depicted in figure 7, 8 and 9.
The IgM response usually was highest in samples taken seven
days after the first injection of antigen. However, as
seen in table 1, eight of the dogs (5,7,8,12,14,16,23,25)
had anti-DNP IgM concentrations that were greatest in
samples obtained at two weeks and three dogs (9,17,22)
after three weeks. Four weeks after the first antigenic
challenge, five dogs had no detectable IgM antibody and


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Michael D.P. dfoyle
Associate Professor of
Immunology and Medical
Microbiology
I certify that I have read this study and that in my opinion
it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality as
a dissertation for the degree of Doctor of Philosophy.
Michael J.P. Lawman
Assistant Professor of
Veterinary Medicine
This dissertation was submitted to the Graduate
Faculty of the College of Medicine and to the Graduate
Council, and was accepted as partial fulfillment of the
requirements for the degree of Doctor of Philosophy.
Dean, College of Medicine
Dean, Gradate School
August, 1983


BIOGRAPHICAL SKETCH
My narre is Tippy Schultz. I am a border collie, spitz
dog who was an abused puppy scheduled to be euthanized on
the day I met my future owner, Kevin Schultz. He rescued me
from my fate and since he was in his first year of
veterinary school at Purdue University, I became his living
anatomical model to study dog topography. I learned that he
had been born in Chicago in 1951, and, because his family
moved alot, he attended many different grade and high
schools before finally moving to Fort Wayne, Indiana, where
he graduated from high school. We studied hard in
veterinary school and I sent Kevin to work at B.F. Goodrich
Tire Company during summers to keep me in food and shelter
which provided motivation for him to continue his education.
We graduated in 1976 and since that time, I have
gotten to do a great deal of traveling with him. In Dodge
City, Kansas, while he was in practice, he met a very pretty
and very nice woman. He saved her cat from dying after
being hit by a car, and since she could also type, they got
married and we all moved to Chicago so Kevin could practice


91
Table 16
The Inhibition of Canine IgG Anti-id Binding to
Mouse Monoclonal Anti-DNP IgG by Hapten
as Measured by RIA
Animal Number a)
2, 4 DNP Glycine 1
5
7
0 pg
3943
+
73 b)
4522 + 138
1775
+34
10 pg
3802
+
100
3533 + 69
1704
91
2 pg
3983
+
280
3783+136
1675
+ 46
1 pg
3807
+
214
4268 + 56
1614
+ 101
0.1 pg
3804
+
12
4623 + 219
1734
+ 73
8
15
16
0 pg
2281
+
162
2702
+ 131
1643
+ 52
10 pg
2203
+
200
2026
+ 168
1454
+ 15
2 pg
2012
+
115
2593
+ 173
1691
+ 14
1 pg
2148
+
83
2738
+ 57
1377
+ 34
0.1 pg
2213
+
129
2694
+ 89
1526
+ 31
a) Each dog was immunized with autologous anti-DNP antibody in CFA
at weeks 7 and 9. All samples were from weak 9 in the schedule
except sample 15 which was from week 11.
b) This is tlie mean cpm of triplicate samples + standard
deviation. All samples were assayed with serum diluted one to four
with PBS.
A control well with mouse anti-^K IgG. rather than anti-DNP
IgG was assayed for each sample. The mean + standard deviations
for all samples 283+47.


141
12. Katz, D.H. (1980) Recent Studies on the Regulation of
IgE Antibody Synthesis in Experimental Animals and Man.
Immunol. 41:1-24 .
13. Ishizaka, K. (1976) Cellular Events in the IgE
Antibody Response. Adv. Immunol. 23:1-75.
14. Shinohara, N. and Tada, T. (1974) Hapten-Specific IgM
and IgG Antibody Responses in Mice against a
Thymus-Independent Antigen (DNP-Salmonella). Int. Arch
Allergy Appl. Immunol. 47:762-776.
15. Suemura, M., Yodai, J., Hirashima, M. and Ishizaka, K.
(1980) Regulatory Role of IgE binding Factors from Rat
T-lymphocytes. 1. Mechanism of Enhancement of IgE Response
by IgE-Potentiating Factor. J. Inmunol. 125:148-154.
16. Saryan, J.A., Leung, D.Y.M., and Geha, R.S. (1983)
Induction of Human IgE Synthesis by a Factor Derived from
T-cells of Patients with Hyper-IgE States. J. Immunol.
130:242-247.
17. Katz, D.H., Davie, J.M., Paul, W.E., and Benacerraf, B.
(1971) Carrier Function in Anti-Hapten Antibody Responses.
IV Experimental Conditions for the Induction of
Hapten-specific Tolerance or for the Stimulations of
Anti-Hapten Anmnestic Responses by 'non-immunogenic'
Hapten-Polypeptide Conjugates. J. Exp. Med. 134:201-224.
18. Sehon, A.H. and, Lee, W.Y. (1981) Specific
Downregulation of IgE Antibodies by Tolerogenic Conjugates
of Haptens and Allergens with Synthetic Polymers. Int. Arch
Allergy Appl. Immunol. 66 (suppl. l):39-42.
19. Liu, F.T., Bogowitz, C.A., Bargatze, R.F., Zinnecher,
M., Katz, L.R., and Katz, D.H. (1979) Immunologic
Tolerance to Allergenic Protein Determinants: Properties of
Tolerance Induced in Mice Treated with Conjugates of Protein
and a Synthetic Copolymer of D-Glutamic Acid and D-Lysine.
J. Immunol. 123:2456-2465.
20. Zeiss, C.R. (1979) Immunotherapy in Allergic Disease.
In Allergy and Clinical Immunology, Medical Examination
Publ. Co., Inc., Garden City, N.Y.
21. Halliwell, R.E.W. (1977) Hyposensitization in the
Treatment of Atopic Disease. In Current Veterinary Therapy
VI. W.B. Saunders Co., Phila, Pa.


139
is because IgE antibody is regulated by id/anti-id inter
action. If this is the case, the identification of anti-id
during hyposensitization may help in establishing more
effective immunotherapy for allergies.


29
Radioimmunoassay for the Detection of DNP Specific
Antibody
Microtiter wells (Iirmulon Ramov-a-well Strips ,
Dynateck, Richmond, Va.) were coated with 50 pi of 20
pg/ml dinitrophenylated bovine gamma globulin (DNP-BGG) in
Tris buffer, pH 8 (0.1 M Tris, 0.15 M NaCl). After incu
bating for 12 hours at 4C the wells were then washed
three times with this buffer. Any remaining sites were
blocked with 2.0 percent HSA in PBS containing 0.5 percent
Tween 20 for three hours at room temperature. Phosphate
buffered saline, pH 7.2 containing 0.5 percent Tween 20 and
2.0 percent HSA is referred to as RAST+ and this same
buffer without HSA is called RAST-. Serum samples were
diluted in PBS, pH 7.2 added to the appropriate wells,
incubated for three hours at 4C followed by five washes
with RAST-. Approximately 50,000 counts per minute (cpm)
of radiolabelled antiserum in RAST- was added to the well,
incubated for three hours at 4C and washed three times
with RAST-. The radioactivity associated with each well
was determined in a Searle-Packard gamma counter (Chicago,
II.). Each sample was assayed in triplicate and each
sample was counted for one minute. The maximum number of
cpm bound was about 20% of the amount added. The back
ground activity was determined by including in each assay
the following controls: 1) A set of triplicate wells in
which BGG rather than DNP-BGG was used as the antigen, 2) A


82
Results
Identification of Canine Anti-Iditoypic Antibody
A screening procedure was used to assay for the
presence of anti-id in serum obtained during the immun
ization schedule. Those dogs that received autologous
antibody produced an antibody which would bind to mouse
monoclonal anti-DNP IgG as seen in table 11. This binding
was not detectable prior to such treatment in any dog nor
could it be detected in control dogs at any time. There
was no detectable binding to tine anti-H^K IgG mouse anti
body in serum from any dog.
The experiment was repeated with another group of 12
dogs and the anti-id activity was converted to an arbitrary
relative antibody concentration by interpolation from a
scale derived from the titration of a positive high titer
sample identified in the screening procedure (table 12).
This standard serum was given a relative antibody concen
tration of 10. Some dogs produced detectable levels of
this anti-idiotypic antibody within one week after autol
ogous antibody administration whereas other dogs took three
weeks to develop such a response. There was also variation
in the magnitude of the response observed. This anti-
idiotypic antibody could not be detected if anti-canine IgE
or IgM antisera was used as the radiolabelled probe rather


Figure 1.
The specificity of anti-canine IgG as assayed in an
immunoelectorphoresis against normal canine serum.
Figure 2.
The specificity of anti-canine IgE as assayed in an
immunoelectrophoresis against normal canine serum
(bottom well) and this same serum after heat
inactivation (top well).
Figure 3.
The specificity of anti-canine IgM as assayed in an
immunoelectrophoresis against normal canine serum.
The anti-canine IgM in the bottom through is before
adsorption with the supernatant of a 50 percent
saturated ammonium sulfate precipitate of normal
canine serum. The top trough has the anti-canine IgM
antiserum after this treatment.


103
bound with normal canine IgG F(ab)'^ (with no detectable
anti-DNP activity) and with HSA bound beads. Specific
binding was calculated by using the following formula:
specific binding = cpm bound to autologous anti-DNP
Fiab)^ beads of a sample at a given dilution cpm bound
to NCS IgG Fiab)^ beads of the sample at the same
dilution. Since beads were standardized for an amount of
antibody in each experiment, the volume of beads used
ranged from 50 pi to 78 pi per sample. When 50 or 100
pi of non-specific Fiab)^ was incubated with the
sample, there was less than a 15 percent difference in the
cpm indicating that the increase in bead volume had little
influence on the background activity.
Results
Identification of Anti-Idiotypic Antibody Using
Autologous Idiotypes
The purpose of these experiments was to determine if
autologous id could be used to detect anti-id. Anti-DNP
FCab)^ fragments from various time points in the immuni
zation schedule were used as antigens to detect anti-id in
autologous serum. The autologous serum used in these
experiments were first chromatographed through a DNP
affinity column bo remove anti-DNP antibody. This was done
to eliminate any possible interference the presence of this


107
Table 20
The Detection of Canine Anti-Iditoypic Antibody by RIA
Using Autologous Anti-DNP F(ab'^ as the Id
Dog Number 21
Source of Anti-Id (Week) a)
1 3 5 7 10
Source of
id
(Week) b)
Effluent
Dilution
1
0
358+94
1000+49
1/2
247+37
539+97
1/4
88+23
102+61
3
0
253+63
1532+88
1/2
77+31
865+109
1/4
83+71
276+123
7
0
104+60
0
1/2
0
0
1/4
46+29
0
11
0
49+41
0
1/2
0
65+58
1/4
108+39
17+11
Control
c)
0
238+83
1/2
211+61
1/4
103+48
307+69
435+146
684+33
204+39
197+55
377+36
35+7
27+26
246+29
1031+63
299+27
31+28
940+39
5+3
150+123
220+47
2+4
121+38
0
1466+229
1077+10
0
364+97
591+101
75+50
169+11
359+66
89+31
0
0
28+49
69+63
64+57
21+43
101+34
106+29
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP Ftab)^ immobilized to a solid
matrix. The dog was immunized to DNP-ASC in adjuvant at weeks
0,2,4,6,8, and 10 and received CFA alone at weeks 7 and 9.
c) Control id was normal canine IgG F(ab)' immobilized to a solid
matrix.


138
mouse monoclonal anti-DNP antibody tested and 2) a failure
of hapten to inhibit the id/anti-id interaction. An
internal image of antigen should bind to corresponding anti
body to an extent similar to antigen and should also be
hapten inhibitable (91).
Anti-idiotypic antibody could be detected during the
DNP-ASC immune response when the ids used to assay for anti
body were autologous anti-DNP F(ab)'2 fragments. These
anti-ids can be considered natural in that their appearance
was associated with anti-DNP antibody produced during an
immune response. This is in contrast to the anti-ids
detected with the mouse antibodies. These latter anti-ids
were detected subsequent to the immunization with anti-DNP
antibody in complete Freund's adjuvant. Three of the five
dogs assayed for anti-id using autologous ids had detect
able amounts of anti-id whereas two dogs had no detectable
levels. It was concluded that anti-id can be detected
during a DNP-ASC response in dogs.
This method of detecting anti-id would allow for the
identification of a different anti-id response during other
immune responses. For example, the aim of hyposensi
tization for allergic disease is felt to be the production
of a blocking antibody. However, this treatment is not
always effective, even if blocking antibody is produced.
An additional possible reason why hyposensitization works


97
antibody of the same specificity as the immunizing
antibody. Prior to such treatment this autologous antibody
is present in the dog but does not induce detectable levels
of anti-id. However, after the administration of this same
antibody, anti-id is detected. Therefore, these results
indicate that during the purification procedure and/or the
immunization procedure, anti-id determinants present on the
autologous antibody are immunogenically enhanced. However,
any changes that occur in the protein molecule must be
subtle because the specificity of the anti-id response was
determined by the specificity of the immunizing antibody.
If there was narked change in the id determinants, then the
anti-id might not be expected to maintain its specificity
for the immunizing antibody. However, it is not possible
from these results to determine if the immunogenic
enhancement of the id was a result of the purification
process, the route of immunization or a combination of both
these things.
It was very fortunate that the mouse antibody used as
the antigen in these assays had id determinants which could
bind to the canine anti-id, although shared idiotypy
between animals has been reported (53,57,85-87).


78
Materials and Methods
Antisera
The antisera used in these experiments were described
in Chapter two. Any cross reactive anti-mouse
immunoglobulin activity that was present in the anti-canine
IgG, IgM or IgE antisera used in the anti-id RIA was
removed by passage through an affinity column which had
bound to it a 40 percent saturated ammonium sulfate
precipitate of normal mouse serum.
Animals and Immunization
The animals and immunization schedules have been
described in Chapter two and three except that in the
experiment designed to determine if the specificity of the
antibody was important in the induction of an anti-id
response, a different immunization protocol was used.
Eight mature dogs were injected with 100 pg of aluminum
hydroxide precipitated DMP-ASC by the intraperitoneal route
on the day of arrival. At the same time, these dogs
received a second injection of 100 pg of aluminum
hydroxide precipitated ABA-KLH by the same route at a
different site. These dogs were immunized three times at
two week intervals. At the time of the last immunization,
30 ml of blood were obtained from each dog. The serum from
this blood was used to purify antibody. The dogs were


0 2 4 6 8 10
Weeks
115


3/7y
/9f3


63. Forni, L., Coutinho, A., Kohler, G., and Jerne, N.K.
(1980) IgM Antibody Induces the Production of Antibodies of
the Same Specificity. Proc. Natl. Acad Sci. U.S.A.
77:1125-1128.
64. Ivars, F., Holmberg, D., Forni, L., Cazenave, P.A., and
Coutinko, A. (1982) Antigen-Independent, IgM-Induced
Antibody Responses: Requirement for "Recurrent" Idiotypes.
Eur. J. Immunol. 12:146-151.
65. Sege, K., and Peterson P.A. (1978) Anti-Idiotypic
Antibodies Against Anti-Vitamin A Transporting Protein React
with Prealbumin. Nature. 271:167-168.
66. Schreiber, A., Andre, C., Vray, B., and Strosberg, A.D.
(1980) Anti-Alprenolol Anti-Idiotypic Antibodies Bind to
B-Adrenergic Receptors and Modulate Catecholamine Sensitive
Adenylate Cyclase. Proc. Natl. Acad. Sci. U.S.A.
77:7385-7389.
67. Schwartzman, R.M., Rockey, J.H., and Halliwell, R.E.W.
(1971) Canine Reaginic Antibody. Characterization of the
Spontaneous Anti-Ragweed and Induced Anti-Dinitrophenyl
Reaginic Antibodies of the Atopic Dog. Clin. Exp. Immunol.
9:549-569.
68. Bradford, M.M. (1976) A Rapid and Sensitive Method
for the Quantification of Microgram Quantities of Protein
Utilizing the Principle of Protein-Dye Binding. Anal
Biochem. 72:248-254.
69. Strejan, G., and Campbell, D.H. (1963)
Hypersensitivity to Ascaris Antigen. IV Production of
Hanocytotropic Antibodies in the Rat. J. Irrmunol.
101:628-637.
70. Williams, C.A., and Chase, M.W. (1967) Methods in
Immunology and Iirmunochemistry. Vol. 1, Academic Press,
New York, New York.
71. Imada, Y., Namoto, K., Yamada, H., and Takeya, K.
(1977) Relationships between IgE Antibody Production and
other Immune Responses. 1. Ontogenic Development of IgE
Antibody-Producing Capacity in Mice. Int. Arch Allergy
Appl. Immunol. 53:50-55.
72. Vaz, E.M., Vaz, N.M., and Levine B.B. (1971)
Persistent Formation of Reagins in Mice Injected with Low
Doses of Ovalbumin. Immunol. 21:11-15.


I 23456789
Weeks
| DNP/ASC
T reatment
group
group 3
group 2
control
10 II


150
for a year We then hit the road for Philadelphia, Pa.,
where Kevin studied comparative dermatology at the
University of Pennsylvania, and he put Nancy to work typing.
In September, 1979, we packed up and moved again, this time
to Gainesville, Florida. Hera both Kevin and Nancy attended
the University of Florida and I didn't see much of either of
them while they were busy getting their degrees.
During our treks across the country, we have adopted
two permanent house guests, Fanny and Pippin, who are OK for
being just cats. In July, 1983, we will all be moving
again, this time to Boston, Massachusetts. There Kevin will
be a post-doctoral fellow in the Department of Pathology at
Harvard Medical School which is fine with me as long as he
continues to keep tie warm and fed.


Figure 19.
The identification of anti-id in various serum
samples over time, in dog 24, as measured by RIA.
The dogs received DNP-ASC in adjuvant at weeks
0,2,4,6,8,10, and CFA at weeks 7 and 9. The arrow,
narked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used
as antigens to detect anti-id. The bars represent
the standard deviation of the mean.


113


CHAPrER 'EvO
THE I INDUCTION AND KINETICS
OF AN ANTI-DNP IGE RESPONSE
Introduction
The value of the dog as an experimental model to
study atopy has been described. However, the expense and
difficulty of obtaining atopic dogs necessitated the
development of a system in which antigen-specific IgE could
be consistently induced. The use of a hapten-coupled
carrier as an antigen was felt to be more convenient than a
more complex, heterogenous substance such as an allergen to
study the synthesis and regulation of IgE antibody.
Furthermore, Halliwell (7) and Schwartzman et al. (67) have
shown that two dogs immunized with dinitrophenol coupled to
ascaris antigen and administered in aluminum hydroxide as
tlie adjuvant, developed anti-DNP IgE antibody. However, it
is not known a) if all dogs so immunized produce IgE anti
body, b) how long the detectable IgE response remains, and
c) what the immune response in terms of other isotypes
might be. The purpose of the following experiments, then,
was to induce a consistent anti-hapten IgE antibody
17


18
response and to examine the kinetics of the IgE, IgG and
IgM anti-hapten antibody response.
Materials and Methods
Protein Concentration Determination
The concentration of immunoglobulin was determined
from known molar extinction coefficients and by its ability
to absorb light at 280 nm. Alternatively, the protein
concentration was determined at 595 nm using Bradford's
reagent (68) and interpolated from a standard curve derived
from the absorption values of a series of dilutions of a
similar freeze dried purified protein of known concen
tration. The measurements with both techniques gave concor
dant results.
Antigens
Azobenzenarsonate coupled to keyhole limpet hemo-
cyanin (ABA-KLH) was a gift from Dr. Mark Greene, Harvard
University. Ascaris antigen vas prepared from adult
Toxocara canis by the method of Strejan and Campbell
(69) and modified as follows: Fifty adult T. canis
were obtained from the gastrointestinal tract of euthanized
dogs. The worms were washed with phosphate buffered saline
(PBS), pH 7.2, containing 0.02 percent sodium azide, ground
with a mortar and pestle and incubated for 48 hours at 4
C. Large particulate natter was removed by centrifugation


73
for IgE and IgG for each dog is given in table 9 and 10.
There was a gradual increase in the mean antibody concen
tration in general for both IgE and IgG antibody whereas
IgM was not detected after week. 6. In two dogs (15, 24),
after the seventh and eighth weeks respectively, there was
a cessation of the IgE response. Although the mean IgE
antibody response increased for the groups in general,
individual dogs varied considerably. For example, dogs 14
and 18 had a peak IgE antibody response at week 7 and there
after the response diminished, whereas the peak response
for dogs 19 and 27 occurred at the end of the schedule.
There was no marked difference in the antibody
response between the different groups of dogs. To deter
mine if there were any patterns in the antibody response
between these groups, an analysis of variance comparing
time by group vras calculated for each antibody class with
the assistance of the Department of Biostatistics, College
of Medicine, University of Florida. There was no signif
icant difference between these groups at any given time by
this analysis (p greater than .05).
Because of the large variation between dogs, an
analysis of variance was calculated comparing dogs within a
single litter in one group to dogs from the same litter in
other groups as a function of time. This analysis was used
to determine if there was variation between one treatment
in the IgE or IgG antibody response as compared to a second


76
molecules but in response other antibody molecules were
expressed, the net effect may not be observable if the
entire class specific response were to be measured as was
the case in these experiments.
Summary and Conclusions
The administration of autologous anti-DNP antibody in
adjuvant to dogs which had an ongoing anti-DNP antibody
response did not have a significant effect on this antibody
response as compared to control dogs who received adjuvant
without autologous antibody. This would suggest either
that the regulation by passive antibodies, as seen in
laboratory animals, does not operate in this species or
that any regulation that occurred by such treatment could
not be detected by the methods used in this study.


13
A number of studies have shown the presence of auto-
anti-id during a normal immune response to an antigen
(50-57). Bankert and Pressman (50) showed that an antibody
with auto-anti-id activity could be detected in rabbits
during primary and secondary immune response to both sheep
red blood cells and to the hapten, 3-iodo-4-hydroxy-5-
nitrophenyl-acetic acid. Kelsoe and Cerny (51) have demon
strated a reciprocal expansion of antigen activated idiotype
bearing clones of lymphocytes followed by expansion of
clones which bear anti-id receptors in Balb/c mice immunized
to Streptococcus pneumonia. They hypothesized that the
out of phase expansion of the reciprocal cell sets was the
result of interactions of id and anti-id. The production of
auto-anti-id in man has been demonstrated to occur during
tiie immune response against tetanus toxoid. The presence of
this anti-id was associated with the loss of some of the
anti-tetanus toxoid idiotypes (52). Naturally occurring
anti-id has also been demonstrated in myasthenia gravis
patients using, as the idiotype probe, a mouse monoclonal
antibody. Those patients with the highest titer of anti
receptor antibody had the lowest level of anti-id, while in
patients with the lowest titer of anti-receptor antibody
(id), the highest titer of anti-id was detected (53).
Comparable findings have been reported in patients with
anti-DNA. antibody and reciprocal anti-id in systemic lupus


143
33. Brient, B.W., Haimovich, J., and Nisonoff, A. (1971)
Reaction of Anti-Idiotypic Antibody with the Hapten-binding
Site of a Myeloma Protein. Proc. Natl. Acad. Sci. U.S.A.
68:3136-3139.
34. Briles, D.E., and Krause, R.M. (1974) Meuse
Strain-Specific Idiotypy and Intrastrain Idiotypic Cross
Reactions. J. Immunol. 113:522-530.
35. Sher, A., and Cohn, M. (1972) Effect of Haptens on
the Reaction of Anti-Idiotype Antibody with a Mouse
Anti-Phosphorylcholine Plasmocytoma Protein. J. Immunol.
109:176-178.
36. Eichmann, K., Coutinho, A., and Melchers, F. (1977)
Absolute Frequencies of Lipopolysaccharide-Reactive B Cells
Producing A5A Idiotype in Unprimed, Streptococcal A
Carbohydrate-Primed Anti-A5A Idiotype-Sensitized and
Anti-A5A Idiotype-Suppressed A/J Mice. J. Exp. Med.
146:1436-1449.
37. Bona, C., Mond, J.M., Stein, K.E., House, S., Liberman,
R., and Paul, W.E. (1979) Immune Response to Levan. Ill
The Capacity to Produce Anti-Inulin Antibodies and
Cross-Reactive Idiotypes Appears Late in Ontogeny. J.
Immunol. 123:1484-1490.
38. Marcu, K.B. (1982) Immunoglobulin Heavy-Chain
Constant-Region Genes. Cell 29:719-721.
39. Kolb, H., and Bosma, M.J. (1977) Clones Producing
Antibody of More than One Class. Immunol. 33:461-469.
40. Hopper, J.E., and Nisonoff, A. (1971) Individual
Antigenic Specifity of Immunoglobulins. Adv. Immunol.
13:57-99.
41. Capra, J.D., and Kohoe, J.M. (1975) Hypervariable
Regions, Idiotypy, and the Antibody-Combining Site. Adv.
Immunol. 20:1-40.
42. Eichmann, K. (1978) Expression and Function of
Idiotypes on Lymphocytes. Adv. Immunol. 26:195-254.
43. Janeway, C.A. (1980) Idiotypes, T-Cell Receptors and
T-B Cooperation. Contemp. Top. Immunobiol. 9:171-203.


121


54. Abdou, N.I., Wall, H., Lindsley, H.B., Halsey, J.N.,
and Suzuki, T. (1981) Network Theory in Autoimmunity. In
Vitro Suppression of Serum Anti-DNA Antibody Binding DNA
by Anti-Idiotypic Antibody in Systemic Lupus Erythematosus.
J. Clin. Invest. 67:1297-1304.
55. Cunningham-Rundles, C. (1982) Naturally Occuring
Autologous Anti-Idiotypic Antibodies: Participation in
Immune Complex Formation in Selective IgA Deficiency. J.
Exp. Med. 155:711-719.
56. Schrater, A.F., Goidl, E.A., Thorbecke, G.J., and
Siskind, G.W. (1979) Production of Auto-Anti-Idiotypic
Antibody during the Normal Immune Response to TNP-Ficoll. I
Occurrence in AKR/J and BALB/C Mice of Hapten-Augmentable,
Anti-TNP Plaque-Forming Cells and their Accelerated
Appearance in Recipients of Immune Spleen Cells. J. Exp.
Med. 150:138-153.
57. Binion, S.B., and Rodkey, L.S. (1982) Naturally
Induced Auto-Anti-Idiotypic Antibodies. Induction by
Identical Idiotypes in some Members of an Outbred Rabbit
Family. J. Exp. Med. 156:860-872.
58. Geczy, A.F., de Week, A.L., Geczy, C.L., and Toffler,
0. (1978) Suppression of Reaginic Antibody Formation in
Guinea Pigs by Anti-Idiotypic Antibodies. J. Allergy Clin.
Iirmunol. 62:261-270.
59. Blaser, K., Nakagawa, T., and de Week, A.L. (1981)
Suppression of Anti-Hapten IgE and IgG Antibody Responses by
Isologous Anti-Idiotypic Antibodies Against Purified
Anti-Carrier (Ovalbumin) Antibodies in BALB/C Mice. J.
Immunol. 126:1180-1184.
60. Blaser, K., Nakagawa, T., and de Week, A.L. (1980)
Suppression of the Benzylpenicelloyl (BPO) Specific IgE
Formation with Isologous Anti-Idiotypic Antibodies in BALB/C
Mice. J. Inrrtunol. 125:24-30.
61. Blaser, K., Geiser, M., and de Week, A.L. (1979)
Suppression of Phosphorylcholine Specific IgE Antibody
Formation in Balb/C Mice by Isologous Anti-T15 Antiserum.
Eur. J. Immunol. 9:1017-1020.
62. Eichmann, K., and Rajewsky, K. (1975) Induction of T
and B Cell Immunity by Anti-Idiotypic Antibody. Eur. J.
Immunol. 5:661-666.


49
As TOuld be expected for outbred animals, there was
considerable variation in the immune response between dogs.
If however, the IgG antibody concentration of dogs within
single litters are examined, a more homogeneous trend is
observed (table 4). There was, however, considerable
animal-bo-animal variation within a litter in the level of
antigen specific IgE and IgM (tables 5 and 6). If these
two litters are compared statistically, at each time point,
using a student T test, there is a significant difference
in the mean antibody level between the two groups in the
IgG antibody after the first week (P is less than 0.001 in
all instances).
When all the animals are considered, it appears that
some are generally high responders to the antigenic stimu
lation whereas the response in other dogs is low. The high
response or low response is seen for both IgG and IgE
antibody classes in a single animal. For example, dogs 2
and 6 have a strong IgE and IgG response whereas dogs 15
and 24 have very weak responses. Although this trend
predominates, this association of high responses or low
responses is not always consistent, and a regression
analysis comparing the level of anti-DNP IgG to the
anti-DNP IgE failed to show a statistically significant
correlation (p greater than 0.05).


130
ids obtained from dog 1 at week 10, anti-id was detected in
all but the first sample tested (figure 22). Both of these
dogs had received autologous antibody in adjuvant. Eoth
these ids were obtained from the dogs after this treatment.
In contrast to the anti-id detected in these two dogs, id
was obtained from dog 24 at a similar time in the immuni
zation schedule (week 11), but revealed no anti-id using
these id antigens. This dog received adjuvant without
autologous antibody. In two dogs there was no detectable
anti-id at any point in the immunization schedule using
autologous id as the probe. This is in spite of the fact
that one of the dogs was immunized with autologous antibody
and did have detectable levels of anti-id using the mouse
monoclonal antibody as the .id probe.
Discussion
In the experiments in Chapter four, anti-id was only
detected in serum after autologous antibody administration.
If anti-id does serve in a regulatory fashion, it is not
clear why its presence is not detected during the course of
the anti-DNP antibody response. This inability to detect
anti-id throughout the immunization schedule itay.be a
function of the xenogeneic probe used to detect anti-id or
it may be because anti-id is present only after artificial
manipulation. Furthermore, if anti-id does regulate the


10
Weeks
¡diotype
probe
8 10
129


73.Revoltella, R., and Ovary, Z. (1971) Reagenic
Antibody Production in Different Mouse Strains. Immunol.
17:45-54.
14 7
74. Geha, R.S. (1981) Regulation of the Immune Response
by Idiotype-Antiidiotype Interactions. N. Eng. J. Med.
305:25-28.
75. Jarrett, E.E.E., and Ferguson, A. (1974) Effect of
T-Cell Depletion on the Potentiated Reagin Response. Nature
250:420-422.
76. Chiorazzi, N., Fox, D.A., and Katz, D.H. (1976)
Hapten-Specific Enhancement of IgE Antibody Production by
Low Doses of X-Irradiation and by Cyclophosphamide. J.
Immunol. 117:1629-1637.
77. Nagel, J.E., White, C., Lin, M.S., and Fireman, P.
(1979) IgE Synthesis in Man. II. Comparison of Tetanus and
Diptheria IgE Antibody in Allergic and Non-Allergic
Children. J. Allergy Clin. Immunol. 63:308-314.
78. Frich, O.L., and Brooks, D.L. (1983) Immunoglobulin E
Antibodies to Pollens Augmented in Dogs by Virus Vaccines.
Am. J. Vet. Res. 44:440-445.
79. Katz, D. (1978) The Allergic Phenotype.
Manifestation of "allergic breakthrough" and imbalance in
normal "Damping" of IgE Antibody Production. Immunol. Rev.
41:77-108.
80. Rowley, D.A., Fitch, F.W., Stuart, F.P., Kohler, H.,
and Cosenza, H. (1973) Specific Suppression of Immune
Responses. Science 181:1133-1141.
81. Voisin, G.A. (1980) Role of Antibody Classes in the
Regulatory Facilitation Reaction. Immunol. Rev. 49:3-59.
82. Hart, D.A., Pawlak, L.L. and Nisonoff, A. (1973)
Nature of Anti-Hapten Antibodies Arising after Immune
Suppression of a Set of Cross-Reactive Idiotypic
Specificities. Eur. J. Immunol. 3:44-48.
83. Jarrett, E.E.E., and Hall, E. (1983) IgE Suppression
by Maternal IgG. Immunol. 48:49-59.
84. Pawlak, L.L., Hart, D.A., and Nisonoff, A. (1973)
Requirements for Prolonged Suppression of an Idiotypic
Specificity in Adult Mice. J. Exp. Med. 137:1442-1458.


CHAPrER SIX
CONCLUSION
Dogs were chosen to study IgE antibody synthesis and
regulation because they develop an IgE mediated disease
that is very similar to atopic disease of nan.
When dogs were immunized with 100 gg of aluminum
hydroxide precipitated DNP-ASC by the intraperitoneal
route, each dog synthesized anti-DNP IgG, IgE and IgM anti
body. There ^as a difference in the responsiveness to this
antigen between individual dogs Which most probably
reflected the genetic heterogeneity between them. In an
attempt to regulate anti-DNP antibody, autologous anti-DNP
antibody in adjuvant was administered to these dogs during
the ongoing response. Although dogs so treated did not
have a difference in the magnitude of the anti-DNP IgE or
IgG response, as compared to control dogs who received
adjuvant without antibody, these dogs did produce an
anti-id which was detected using mouse monoclonal anti-DNP
IgG and IgM.
This anti-id was not an internal image of antigen as
shown by 1) a failure of these antibodies to bind to each
137


Abstract of Dissertation Presented to the Graduate
Council of the University of Florida
In Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
DURING AN IMMUNE RESPONSE IN THE DOG
By
Kevin T. Schultz
August, 1983
Chairman: Richard E. Halliwell
Major Department: Immunology and Medical Microbiology
This investigation cornmensed with the development of
an animal model to study the synthesis of IgE antibody.
Repeated immunizations with a haptenated parasite extract
(dinitrophenol coupled asearis) in young dogs resulted in
the production of anti-DNP antibody of the IgE, IgG and IgM
class.
Although attempts to regulate this anti-hapten anti
body response by administration of autologous anti-DNP anti
body were unsuccessful, such therapy did result in the pro
duction of anti-idiotypic antibodies. These anti-idiotypic
antibodies were demonstrable using mouse hybridoma-derived
anti-DNP antibodies.
vi


This antibody was shown to be anti-idiotypic rather
than an internal image of antigen because it bound to only
two of four monoclonal anti-DNP antibodies and failed to
inhibit the id anti-id interaction with hapten. Anti-
idiotypic antibodies were detected during the immunization
schedule in three of five dogs using autologous anti-DNP
F(ab)'2 fragments as the source of idiotypes.
The anti-idiotypic antibodies identified using the
mouse monoclonal antibody were the result of the immuni
zation procedure and did not appear to be physiologically
relevant to regulation of the immune response. On the other
hand, the anti-idiotypic antibodies identified with the
autologous source of idiotypes appear to be produced during
tile DNP-ASC immune response and were detected before autol
ogous antibody immunization. The antigens that induced the
anti-idiotypic response appeared to be, in this case, the
idiotypes on the anti-DNP antibody that were produced from
tiie DNP-ASC immunization.
vi 1


70
Table 9 Continued
Animal
Group 4 (IFA Alone) Group 5 (CFA Alone)
Number
21
22
23
24
Weeks
8
8.0+.69
5.2+.48
2.2+.29
.3+.02
9
9.0+1.22
7.6+.71
2.2+.20
0
10
7.1+.51
7.0+1.06
1.8+.36
0
11
8.5+.34
5.3+.44
2.6+.09
0
Group 5
Continued
Animal
Number
25
26
27
28
Weeks
8
1.0+.02
4.3+.34
4.2+.44
5.0+.28
9
1.9+.18
4.3+.31
4.2+.35
6.3+.06
10
1.1+.16
3.1+.41
5.7+.26
4.7+.71
11
1.4+.21
2.9+.24
S.2+.38
5.9+.29
a) The level of anti-DNP IgE in these dogs from 0 to week 7 is found
in Table 2.
b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At
weeks 7 and 9: Group 1, 10 pg autologous anti-DNP antibody in CFA;
Group 2, 100 pg autologous anti-DNP antibody in IFA; Group 3, 100
pg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.


Figure 5.
Dilutions of the standard anti-DNP IgG serum sample assayed by RIA.
The bars represent the standard deviation of the mean.


Figure 20.
The identification of anti-id in various serum
samples, in dog 24, as measured by RIA. The dog
received DNP-ASC in adjuvant at weeks 0,2,4,6,8,10,
and CFA at weeks 7 and 9. The arrow, marked idiotype
probe, indicates the time from which the anti-DNP
F(ab)'2 fragments come. These were used as antigens
to detect anti-id. The bars represent the standard
deviation of the mean.


72
Table 10 Continued
Animal
Number
Weeks
8
9
10
11
Group 3 Continued
17 18 19
7.3+.51
6.5+.60
8.4+.62
9.1+.76
25.5+3.14
21.6+1.19
24.3+.29
23.1+1.79
24.8+.64
29.0+2.62
30.1+4.71
30.8+1.06
20
14.6+1.73
16.8+.96
17.2+2.33
22.9+1.87
Animal
Number
Weeks
8
9
10
11
Group 4 (IFA Alone) Group 5 (CFA Alone)
21
15.6+.93
16.7+.81
16.2+1.3
17.1+.43
22
23.3+1.87
28.7+1.61
31.4+1.05
30.0+2.62
23
30.1+1.74
27.7+.50
31.9+3.81
26.5+2.62
24
8.1+.93
9.5+.62
10.7+1.08
9.6+.27
Animal
Number
Weeks
8
9
10
11
Group 5 Continued
25 26 27
9.3+.81
10.4+.65
11.7+.88
11.2+.76
13.6+.55
18.7+1.17
20.5+1.48
18.0+1.42
20.2+.96
22.4+.68
17.3+1.24
18.7+.48
28
22.2+1.43
23.4+.93
24.3+.74
22.0+2.38
b) Each dog received DNP/ASC immunization at weeks 0,2,4,6,8,10. At
weeks 7 and 9: Group 1, 10 gg autologous anti-DNP antibody in CFA;
Group 2, 100 gg autologous anti-DNP antibody in IFA; Group 3, 100
gg autologous anti-DNP antibody in CFA; Group 4, IFA alone; Group
5, CFA alone.


133
detectable in every serum sample checked (figure 21), and
in dog 1, anti-id was detectable in all but the very
earliest serum sample (figure 21). Both of these dogs had
received autologous antibody in adjuvant and the idiotype
used as the antigen was obtained from serum after such
treatment. In contrast, when idiotypes from a comparable
time in the immunization schedule were used as antigen
from, dog 24, which received adjuvant without autologous
antibody, the unusual appearance of anti-id was not
observed. A possible explanation would be that the
administration of autologous antibody in adjuvant
stimulated the production of antibody having similar
idiotypes. That is, anti-DNP antibodies used for
immunization were from week six. The ids on this
immunizing antibody may have stimulated additional antibody
with the same id. Any anti-id which would be produced
because of the id of week six might also bind to ids
produced from the immunization of the antibody from week
six. Therefore, if this were the case, anti-id could be
present in serum prior to the sample from which the id was
derived. There is experimental precedence for this
suggestion (63,64). Forney et al. (63) have shown that
mice given hybridoma-derived anti-sheep red blood cell anti
body without stimulation with antigen will subsequently
produce anti-SRBC antibody of a similar idiotypic speci
ficity to the immunizing antibody. Their interpretation


86
Table 13
Detection of Canine IgG Anti-Idiotypic Antibody
by RIA With Various
Mouse Monoclonal
Antibodies
Anti-DNP
Anti-OVA
Anti-DNP
(IgE)
(IgE)
(IgM)
Animal
#
26
234+7
267+25
264+4
27
281+1
335+89
151+6
16
171+3
286+48
286+3
15
321+31
261+31
301+29
8
261+13
284+14
322+17
7
287+69
221+8
361+37
13
389+11
205+34
257+28
108+45
105+34
Anti SRBC
Anti-DNP
Anti-ABA
dgM)
(IgG)
(IgG)
Animal
#
26
182+36
321+95
209+47
27
197+12
354+29
218+38
16
176+42
4777+65
207+7
15
106+7
7571+246
176+12
8
106+7
1273+93
178+31
7
114+8
980+47
168+14
13
189+15
1144+68
221+8
Serum was diluted 1:2 in PBS. All samples were from week 10 of the
immunization schedule. All animals had received DNP-ASC in adjuvant
at week 0,2,4,6,8,10 and at weeks 7 and 9 received autologous
antibody in CFA except 26 and 27 who received CFA alone.


u 0 I
or
234 56789
Weeks
I DNP/ASC
Treatment
10 II
control
group 2
group 3
group I
O


80
c) Anti-DNP IgM and as a control antibody, anti-SRBC
IgM.
The antibodies b) and c) were obtained from Sera-
Labs, Accurate Chemical and Scientific Corp., Westbury,
N.Y.
d) Anti-DNP IgM (a gift from Dr. C.W. Clem, Miss
issippi State University. This antibody will be designated
anti-DNP IgM-^), and as a control antibody, anti-SRBC IgM
(a gift from Dr. W.C. Raschke, La Jolla Cancer Research
Foundation, Ca.).
The wells were coated with 10 pg antibody in 50 pi
of Tris buffer (0.1 M, pH 8.0). A radiolabelled
anti-canine IgG antibody was used as the radiolabelled
probe unless otherwise stated.
Hapten Inhibition of Id/anti-id Interaction
The RIA using mouse anti-DNP IgG monoclonal antibody
or control IgG monoclonal antibody was performed as prev
iously described except that after blocking any remaining
active sites by the addition of HSA to the plates, various
amounts of 2,4-dinitrophenol glycine (Sigma Chemical Co.,
St. Louis, Mo.) ranging from 0.001 to 0.1 mg in PBS were
incubated for three hours at 4C. Serum samples were
then added to the wells, incubated for three hours at 4C
and cashed to renove unbound protein. A radiolabelled
anti-canine IgG antibody was added, incubated for three


4
and pathways for enhancing IgE antibody production for
parasites and IgE in general might be similar (16).
A number of approaches have been used in an attempt to
control allergic disease. Avoidance of the antigen is one
approach, but this is rarely possible. Drugs that inhibit
mediator release or control the effects of that release are
also employed, but they have side effects and often require
continual therapy. However, the ideal approach would be to
regulate the production of the unwanted IgE antibody.
The mechanisms used to regulate IgE antibody responses
involve either the inactivation of B cell precursors or the
manipulation of T cell populations. To this end, primary
and ongoing antibody responses, including IgE antibody, have
been suppressed in mice by antigen coupled to non-immuno
genic carriers such as d-glutamine-d-lysine (dGL) or poly
vinyl alcohol (17,18). Both of these carriers inactivate
hapten-specific B cells and can induce hapten-specific
suppressor T cells. Moreover, when dGL is coupled to
proteins rather than haptens, the resultant suppression in
mice is isotype specific (i.e. suppresses IgE alone) (19).
Unfortunately, there are no published results of the use of
this compound in dogs or man.
Hyposensitization has also teen used in an effort to
control allergies in both man and dogs (20-22). The
mechanism by which it works is not clear. It is known that


19
at 1000 x g for ten minutes in an IEC centra-7R centrifuge
(International Equipment Co.) The supernatant was then
centrifuged at 49000 x g for one hour in an L8-70 ultra
centrifuge (Beckman Instrument Co., Norcross, Ga.) to
remove fine particles and was then chromatographed through
a Sephadex G-100 column (Pharmacia Fine Chemicals,
Piscataway, N.J.). The first peak was pooled, concentrated
by negative pressure dialysis, dialyzed against PBS, pH
7.2, passed through a filter having 0.2 micron pores filter
(Acrodisc, Gelman Co., Ann Arbor, Mi.) and used as the
ascaris antigen (ASC). Human serum albumin (HSA) fraction
V was obtained from Sigma Chemical Co. (St. Louis, Mo.).
Bovine gamma globulin (BGG) was prepared from serum of an
adult cow by precipitation with 40 percent saturated
ammonium sulfate. The precipitate was dialyzed against
0.035 M phosphate buffer, pH 8.0 and was then chromato
graphed through a diethylaminoethyl cellulose (DEAE) ion
exchange column (DEA, DE52, Whatman Chemicals, Kent,
England) equilibrated with this same buffer. The effluent
protein was concentrated by negative pressure dialysis and
dialyzed against PBS, pH 7.2.
Dinitrophenylation of Proteins
Dinitrophenylation of protein was performed by mixing
equal weights of protein, potassium carbonate (Fisher
Scientific Co., St. Louis, Mo.) and


92
Table 17
Inhibition of binding of 125 I-DNP/HSA to
Mouse Monoclonal Anti-DNP Antibody
by Canine Anti-Idiotypic Antibody
Animal a) C.P.M. Bound + S.D. b) % Inhibition c)
26
994
+
43
0
8
493
+
38
50
18
544
+
25
45
13
783
+
41
21
15
611
+
13
39
1
521
+
18
48
a) Animal 8, 18, 13, 15 and 1 received autologous antibody in CFA
and had detectable levels of anti-id; animal 26 received CFA alone
and did not have detectable anti-id. all serum was obtained at
week 11.
b) This is the mean cpm + standard deviation of a sample assayed
in triplicate
c)
% Inhibition = c.p.m. control c.p.m. sample
c.p.m. control
The control was a set of wells coated with mouse anti-DNP IgG and
incubated with PBS rather than serum. The value for this was
1026+19.


fragments as the id. The kinetics in the appearance of
this anti-id in relationship to the id suggest that id is
acting as an antigen to stimulate a corresponding anti-id
response.


Figure 15.
The identification of anti-id in various samples over time, in dog
1, as measured by RIA. The dog received DNP-ASC in adjuvant at
weeks 0,2,4,6,8 and 10, and autologous anti-DNP antibody in
adjuvant at weeks 7 and 9. The arrow, marked idiotype probe,
indicates the time from which the anti-DNP F(ab)'2 fragments came.
These were used as antigens to detect anti-id. The bars represent
the standard deviation of the mean.


Ill


CHAPTER FIVE
DETECTION OF ANTI-IDIOTYPIC ANTIBODY
USING AUTOLOGOUS ANTI-DNP F(ab)'2 FRAGMENTS
AS THE IDIOTYPIC ANTIGEN
Introduction
Anti-idiotypic antibody, as noted in Chapter one, can
have a regulatory function during an inmune response. It
can either enhance (62) or suppress (34,88) the level of
the corresponding idiotype. Even if anti-id stimulates
only a limited number of ids, the overall result is an
enhancement in the total antigen-specific antibody response
(63). Anti-id that acts as an internal image of antigen
can stimulate or enhance an immune response in a way anala-
gous to antigen (89). On the other hand, anti-id has been
shown to suppress an entire antigen-specific isotype
(60-61). If anti-id is important as a natural means to
regulate antigen-specific antibody, it would seem logical
that anti-id would be detectable during a normal immune
response. The results of Chapter four indicated that after
99


Figure 8.
The mean relative anti-DNP IgE concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC. The antibody concentration was calculated
from an arbitrary antibody concentration scale derived from the
titration of a standard anti-DNP IgE containing serum. (See text
for further details.)


87
Table 14
Detection of Canine IgG Anti-Idiotypic
Antibody by RIA with Mouse Monoclonal
Anti-DNP IgM Antibody
Dilution of
Serum Anti-DNP IgM^Antibody
a) -
26-E
26-L
27-E
27-L
16-E
16-L
1/5
162+11
286+21
135+14
248+45
90+7
643+31
1/10
82+14
154+15
97+26
152+29
86+
296+19
1/20
71+21
135+9
126+36
181+11
103+11
156+12
15-E
15-L
8-E
8-L
7-E
7-L
1/5
225+31
912+25
215+47
570+46
249+1
672+6
1/10
184+8
442+27
128+6
373+31
125+8
343+37
1/20
115+12
247+6
102+25
270+12
86+7
236+24
5-E
5-L
1/5
128+33
741+54
1/10
106+7
358+54
1/20
134+8
229+11
a) The
E indicates serum
obtained
prior to the
administration of
autologous antibody administraion (16,15,8,7) or adjuvant alone
(26,27), at week 6.
The L indicates serum obtained after such treatment from week 10.
The mean + standard deviation for all samples assayed using
anti-SRBC IgM was 136+53.


16
networks. Therefore the objectives of the work presented
here were
1) To develop a consistent IgE antibody response in
the dog and to study the kinetics of this response.
2) To examine the effects that autologous antibody
administration had on an ongoing IgE response.
3) To determine if an anti-id response occured at any
point during the experiment and if so, to examine the
relationship between ids and anti-ids.


95
antibody molecule. However, an internal image of antigen
binds to the hypervariable regions associated with the
antigen combining site.
If this antibody was an internal image of antigen
then the activity should have been detected when each anti-
DNP antibody was used as the antigen. Also, a hapten
should inhibit the binding of an internal image of antigen
to the respective antibody. Since anti-id bound to only
two anti-DNP mouse monoclonal antibodies and the id-anti-id
interactions ware not consistently inhibited by hapten, it
can be concluded that this anti-id is not an internal image
of antigen.
Although hapten could not inhibit the id/anti-id
interactions in all cases, indicating that the recognized
idiotopes were not within the antigen combining sites,
these id may be very close to the antigen combining site.
Serum which contained anti-id was assayed to determine if
the sample could interfere with the interaction between the
mouse anti-DNP antibody and a radiolabelled dinitro-
phenylated antigen. In all samples containing anti-id, the
level of antigen bound to the mouse antibody was decreased,
although complete inhibition of this binding was not
observed. Anti-id could consistently inhibit antibody/-
antigen interactions. However, the id/anti-id interactions
were not hapten inhibitable. Therefore, some of these
anti-ids must bind to id determinants which are close to


85. Claflin, J.L., and Davies, J.M. (1975) Clonal Nature
of the Inmune Response to Phosphorylcholine (PC). V.
Cross-Idiotypic Specificity among Heavy Chains of Murine
Anti-PC Antibodies and PC-Binding Myeloma Proteins. J. Exp.
Med. 141:1073-1083.
86. Karol, R.A., Reichlin, M., and Noble, R.W. (1977)
Evolution of an Idiotypic Determinant: Anti-Val. J. Exp.
Med. 146:435-444.
87. Schwartz, M., Norvick, D., Givol, D., and Fuchs, S.
(1978) Induction of Anti-Idiotypic Antibodies by
Inmunization with Syngeneic Spleen Cells Educated with
Acetylcholine Receptor. Nature 273:543-545.
88. Eichmann, K. (1974) Idiotype Suppression: I
Influence of the Effector Functions of Anti-Idiotypic
Antibody on the Production of an Idiotyoe. Eur. J. Inmunol.
4:296-302.
89. Nisonoff, A. and Lamoyi, E. (1981) Implications of
the Presence of an Internal Inage of the Antigen in
Anti-Idiotypic Antibodies: Possible Application to Vaccine
Production. Clin. Immunol, and Inmunopath. 21:397-406.
90. Eisen, H.N. (1980) Immunology. Second Edition.
Harper and Row Publishers, Hagerstown, Md.
91. Theofilopoulos, A.N., and Dixon, F.J. (1979) The
Biology and Detection of Inmune Complexes. Adv. Immunol.
28:39-220.


135
reciprocal anti-id. This lack of anti-id would not
necessarily result in abnormal antibody regulation if these
animals had an alternate means to accomplish this, such as
a T-suppressor cell pathway.
One of these animals had no recognizable anti-id
using autologous antibody as the id probe, but did have
recognizable anti-id when mouse monoclonal antibody was
used as the id probe. There are several possible reasons
for this. 1) There are only a few idiotopes present on
monoclonal antibodies while in a heterogeneous population
of molecules there would be expected to be many idiotypes
and therefore even more idiotopes. Therefore, this
negative result may be a function of the concentration of
id present in the assay system. 2) The purification
process may have altered the id on the antibody molecules
just enough so that when this antibody was used to immunize
a dog, these altered id were able to induce an anti-id
response specific for the mouse id. Or 3) the ids
detected with the mouse antibody were different than the
ids on the canine anti-DNP Fiab)^.
Sunznary and Conclusions
In three of five dogs immunized with DNP-ASC, anti-id
could be detected using autologous anti-DNP Fiab)^


48
Table 2 Continued
Animal
Number
17
18
19
20
Weeks
0
0
0
0
0
1
0
0
0
0
2
.2+.07
10.1+.21
0
3.3+.09
3
11.0+.96
9.3+.49
1.4+.08
4.2+.15
4
7.7+.14
S.9+.47
.3+.04
4.S+.33
5
9.8+.34
9.4+.67
1.0+.20
5.2+.40
6
2.2+.26
6.9+.52
1.5+.08
5.8+.26
7
6.7+.31
15.3+.89
2.3+.21
7.1+.43
Animal
Number
21
22
23
24
Weeks
0
0
0
0
0
1
0
0
0
0
2
3.0+.10
1.5+.17
2.1+.13
.1+.04
3
4.5+.11
2.6+.12
1.7+.09
1.0+.06
4
4.0+.27
3.0+.31
1.8+.14
1.0+.17
5
5.6+.36
2.6+.21
1.0+.13
1.6+.31
6
2.3+.12
2.8+.32
0.8+.07
0.9+.07
7
4.8+.22
2.6+.60
0.9+.21
1.3+.29
Animal
Number
25
26
27
28
Weeks
0
0
0
0
0
1
0
0
0
0
2
1.0+.29
1.7+.37
.6+.05
X
3
2.7+.31
2.1+.05
3.2+.22
1.1+.06
4
2.0+.26
3.0+.42
2.0+.23
7.3+.26
5
1.8+.19
1.3+.12
1.0+.15
8.8+.51
6
.7+.09
4.3+.27
1.5+.13
3.3+.16
7
1.2+.10
1.7+.28
0.8+.06
8.4+.42
a) Each dog was immunized with DNP-ASC in adjuvant at weeks 0,2,4 and 6
b) The units were calculated from relative antibody concentration scale
derived from the titration of a serum sample containing anti-DNP IgE. A
value of zero indicates no detectable anti-DNP IgE. The data was the
mean antibody concentration of a sample run in triplicate + the
standard deviation from, the mean. This was calculated by adding and
subtracting the standard deviation to the mean and calculating the
relative antibody concentration for this number. The relative
concentration for this number was subtracted from the mean concentration


Table 7
The Mean Relative Anti-DNP IgG Concentration
63
Group 1 a)
N = 8
Weeks
0 0
1 0
2 .2 + .3
3 8.7+ 7.6
4 12.1 + 7.2
5 12.5+8.4
6 13.7 + 8.7
7 11.4 + 7.7
8 12.6 + 8.5
9 13.2 + 7.9
10 13.2 + 7.9
11 15.2 + 9.2
Group 4
N = 2
Weeks
0 0
1 0
2
6,
.4
+
1.1
3
8,
.2
+
3.5
4
11,
.2
+
3.7
5
14,
.9
+
6.0
6
12,
.9
+
7.5
7
20,
.9
+
6.4
8
20,
.1
+
4.9
9
22,
.7
+
8.5
10
23,
.5
+
9.1
11
23,
.8
+
10.7
Group 2
N = 4
0
0
.5+ .9
4.4+ 3.6
14.3 + 6.3
13.3 + 6.5
12.5 + 6.0
13.4 + 8.3
14.2 + 8.6
14.5 + 7.6
14.5 + 7.6
17.8+ 6.9
Group 5
N = 6
0
.2 + .3
8.1 + 4.4
12.7 + 3.7
13.7 + 5.2
15.1 + 7.0
13.4 + 8.9
16.4 + 7.8
17.3 + 8.5
18.9 + 7.5
19.4 + 8.0
17.7 + 6.4
Group 3
N = 8
0
0
6.6 + 1.7
13.7+7.5
13.2 + 7.6
12.2 + 5.6
7.8 + 2.1
10.6 + 5.9
12.1 + 4.3
13.8 + 5.9
13.8 + 5.9
15.1 + 7.4
Control
N = 8
0
.1 + .3
7.6 + 3.7
11.5 + 4.1
13.1 + 4.8
15.1 + 6.4
13.6 + 8.1
17.5 + 7.3
17.8 + 7.5
19.8 + 7.3
20.5 + 8.2
19.1 + 7.0
a) All dogs received DNP/ASC iirimunization at 0,2,4,6,8 & 10 weeks.
At 7 and 9 weeks: Group 1 received 10 gg autologous anti-DNP
antibody in CFA, Group 2 received 100 gg autologous anti-DNP
antibody in IFA, Group 3 received 100 gg autologous anti-DNP
antibody in CFA. Group 4 received IFA alone, Group 5 received CFA
only. Control values were the mean of groups 4 and 5.


12
MeanS.D
(range)
0
0
(0-0)
1
.1 .4
(0*1.7)
2
26 2.9
(0-10.8)
3
6.1 5.0
(0-18.2)
4
5.2 3
(0-92)
5
6.9 5.9
(.3-26.8)
6
4.3 3.6
(.6-16.1)
7
6.7 5.4
(.8-19.3)


Figure 9.
The mean relative anti-DNP IgG concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC. The antibody concentration was calculated
from an arbitrary antibody concentration scale derived from the
titration of a standard anti-DNP IgG containing serum (see text for
further details).


58
(2) There was a difference in the responsiveness to
this antigen seen between individual dogs for all antibody
isotypes. This vas most probably a reflection of the
genetic heterogeneity between these dogs.


38
after six weeks the levels of IgM antibody fell to back
ground despite maintenance of the immunizing protocol.
The anti-DNP IgE and IgG antibody responses followed
similar kinetics to each other. There was an initial lag
of two weeks before antibody of these classes was detected
(figures 8 and 9). At the time of the second immunization
(two weeks after the primary immunization), there was a
sharp rise in the antibody levels which continued for one
additional week. Thereafter, the antibody concentration
was maintained at that level or started to gradually
decline. As was the case in the IgM antibody response,
sane dogs deviated from the general trend. Two dogs had
detectable IgE antibody levels one week after primary
immunization (Table 2) whereas three dogs failed to
develop a detectable response until after the third week
and, in the case of one dog, IgE antibody was not detected
until the fifth week iron primary immunization. The IgE
antibody response persisted through the seven week course
of the experiment in all dogs. Anti-DNP IgG antibody vas
detected in nine dogs one week after primary immunization
(table 3) and by the fourth week, all dogs had an IgG anti
body response. Detectable IgG persisted throughout the
immunization schedule but there vas a gradual decline in
IgG antibody levels towards the end of the immunizing
schedule (See figure 9, table 3).


CHAPTER THREE
ATTEMPTS TO REGULATE AN ANTIBODY RESPONSE
WITH AUTOLOGOUS ANTIBODY
Introduction
The mechanisms by which antibody responses are regu
lated have been studied extensively. Many experiments have
shown that antibody can be self-regulating (30,81). There
are at least two different ways that this can occur: 1)
If antibody is present at the time of immunization, anti
body can bind to and remove antigen. Therefore, the result
would be a decrease or a failure to mount the response. 2)
Antibody can induce an anti-idiotypic immune response which
would regulate the subsequent expression of the antibody
through id/anti-id interactions (80-82).
If the synthesis of IgE antibody could be suppressed
with antibody, such therapy may be very beneficial in con
trolling IgE mediated allergic disease. As discussed in
Chapter one, passively administered antibody in mice and
rabbits has been shown to suppress IgE antibody (26,27).
The purpose of the experiments in this chapter is to
determine if the administration of autologous antibody has
any effect on the ongoing antibody response in dogs.
59


7
after antigenic challenge. A complete inhibition of the
passive cutaneous anaphylaxis titer and a narked decrease in
the hemagglutination titer of these rabbits resulted as
compared to controls (26). It vas shown by- Tada and Okumura
(27) that, in the rat, the administration of anti-DNP
ascaris antibody resulted in marked suppression of a
preexisting IgE antibody response and this suppression was
maintained for an extended period of time. This was in
contrast to studies in the mouse in which administration of
anti-ovalbumin IgG had little effect on the preexisting
anti-ovalbumin IgE response (28). These differences were
explained as species variation. Alternatively, they may be
due to the difference in the antigenic system employed.
One explanation for the mechanism of regulation by
passive antibody is that the administration of this antibody
acted as an antigen and stimulated an anti-antibody
response. Lahss et al. (29) were the first to show that
sane anti-antibodies would bind to structures on antibody
close to or within the antigen combining site. These deter
minants have been named idiotypes (id) and the immune
response directed to them is termed an anti-idiotypic
(anti-id) response. In 1974, Jerne (30) proposed his network
hypothesis of antibody regulation. The basic premise of
this theory is that tine immune system is regulated by a
network of interactions between id and anti-id. A number of


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS iii
ABSTRACT vi
CHAPTER ONE INTRODUCTION 1
CHAPTER TWO THE INDUCTION AND KINETICS OF AN
ANTI-DNP IGE RESPONSE
Introduction 17
Materials and Methods 18
Results 31
Discussion 55
Suntnary 57
Conclusions 57
CHAPTER THREE ATTEMPTS TO REGULATE AN ANTIBODY
RESPONSE WITH AUTOLOGOUS ANTIBODY
Introduction 59
Materials and Methods 60
Results 62
Discussion 74
Summary and Conclusions 76
CHAPTER FOUR THE IDENTIFICATION OF ANTI-
IDIOTYPIC ANTIBODY
Introduction 77
Materials and Methods 78
Results 82
Discussion 94
Summary 99
Conclusions 99
CHAPTER FIVE DETECTION OF ANTI-IDIOTYPIC
ANTIBODY USING AUTOLOGOUS ANTI-DNP
F(AB)' FRAGMENTS AS THE IDIOTYPIC
ANTIGEN
Introduction 99
Material and Methods 100
Results 103
Discussion 130
Summary and Conclusions 135
IV


61
(CFA) or incomplete Freund's adjuvant (IFA). Dogs
designated as controls received 2 ml of either CFA or IFA.
In all cases, the injections were given at four sites subcu
taneously seven and nine weeks after the first immunization
with DNP/ASC. Animals were given DNP/ASC booster injec
tions eight and ten weeks after the primary immunization
(Fig.10).
0123456789 10 11
= Administration of antigen
o = Administration of adjuvant with or without
autologous antibody
Figure 10
Time Schedule for Immunizations


CHAPTER ONE
INTRODUCTION
Allergic diseases of the immediate type are very
important pathologic disorders in both man and dogs.
Clinical signs are initiated by an interaction of antigen
and IgE antibody with resultant mediator release from mast
cells and basophils. In man and dogs, allergic reactions
cause considerable morbidity and can be fatal (1,2).
Anaphylaxis from a bee sting is a classic example for both
species.
Atopy in man is an inherited disease which is
associated with an antigen specific IgE response against
environmental allergens. This disease is expressed
clinically as asthma, hay fever, atopic dermatitis or any
combination of these three (3). The dog is an excellent
experimental animal model to study IgE mediated hyper
sensitivity for atopic diseases of man because of the
similarity of the allergic reaction in both species (1,2,4).
Canine IgE shares many physicochemical properties with
hunan IgE (5,6,7). Dogs like man, develop spontaneous
disease associated with increased synthesis of IgE antibody
(3,4) and in dogs the disease is also familial (4).
1


15
Streptococcus whereas if the anti-id was an IgC^' the
expression of A5A id was suppressed. Recently, Forni et al.
(63) showed that the injection of anti-SRBC IgM into normal
mice induced plaque forming cells of the same specificity as
the injected antibody. Further analysis established that the
mechanism of this enhanced responsiveness was based on
id/anti-id interactions (63,64). The authors state that
"these results support network concepts. Thus if an antigen
specific response can be induced solely by using components
of the immune system itself, it follows that, in its basic
economy, this system is autonomous and does not depend on
the introduction of antigen to adjust to new dynamic states"
(63 p. 1127). In this case anti-id most probably acted as
an internal image of antigen. There have been other
examples that demonstrated the mimicry of antigen by anti
body. For example, Sege and Peterson (65) showed that
anti-id prepared against antibody to insulin could mimic the
action of insulin in cells. Schreiber et al. (66) showed
that anti-id against rabbit antibodies to alprenolol would
compete with alprenolol for the binding site on turkey red
blood cells. This anti-id could also stimulate adenylate
cyclase activity in the cells.
This discussion raises the possibility that the admin
istration of autologous antibody might regulate antigen
specific IgE response in the dog through id/anti-id


11
Identical ids have been found irrespective of the
isotype of the antibody. The mechanism by Which IgE and IgG
antibody can have identical idiotypes relates to the gene
rearrangement that occurs during differential expression of
heavy chain genes (38). As a single clone of cells goes
through isotypic shift, a single variable region of the
genes which includes the idiotype, will become linked to
various heavy chain gene fragments. A single cell will
differentiate into plasma cells which express different
heavy chain genes but the same variable gene sequence (39).
Therefore, it is possible for ids to be shared between anti
bodies of the sane binding ability irrespective of the
isotype. This implies that regulation of IgG antibody by
anti-id networks may also result in IgE antibody regulation.
Idiotypic determinants are usually defined sero
logically. There are a number of different ways to produce
anti-id (reviewed in 40-42). Anti-id can be produced across
tine species barrier, within the same species, within the
same strain, or more importantly, even within the same
individual that produced the id. Anti-id have been used to
determine if the id of the antibody molecule may have a
function other than to bind antigen. This has been done by
examining yiiat functional significance the presence of
anti-id had on the corresponding id in vivo.


84
Table 12
The Relative Antibody Concentration of Canine IgG
Anti-Idiotypic Antibody as Measured in a RIA using
Mouse Monoclonal Anti-DNP IgG as the Antigen a)
Animal
Number
1
5
7
8
13
Weeks
0
0
0
0
0
0
1
0
0
0
0
0
2
0
0
0
0
0
3
0
0
0
0
0
4
0
0
0
0
0
5
0
0
0
0
0
6
0
0
0
0
0
7
0
0
0
0
0
8
3.1+.2
4.0+.2
0
0
0
9
5.3+.4
5.4+.4
3.1+.3
3.1+.4
0
10
4.3+.3
5.1+.4
3.S+.2
5.4+.6
3.4+.2
11
5.2+.4
8.0+.6
2.9+.1
5.5+.2
6.6+.4
Animal
Number
15
16
18
26
27
Weeks
0
0
0
0
0
0
1
0
0
0
0
0
2
0
0
0
0
0
3
0
0
0
0
0
4
0
0
0
0
0
5
0
0
0
0
0
6
0
0
0
0
0
7
0
0
0
0
0
8
0
2.4+.1
1.2+.1
0
0
9
3.7+.2
2.2+.2
2.8+.1
0
0
10
10.0+1.3
4.3+.3
5.2+.4
0
0
11
9.2+.7
3.3+.1
7.1+.3
0
0
a) The relative antibody concentration + range was determined by
interpolating from the titration of a serum sample containing high
levels of anti-id. A value of zero indicates no detectable
anti-idiotypic antibody.
b) All animals received DNP-ASC adjuvant at weeks 0,2,4,6,8,10. At
weeks 7 and 9 all animals received autologous antibody in adjuvant
except 26 and 27 who received adjuvant alone.


o
o
35
30
25
20
Cl 15
O
10
5
5 10 20 40
Reciprocal of
80 160
Dilution
OJ
U>


8
o
o
o
E
Q_
O
5 10 20 40
Reciprocal
80 160 320640 1280
of Dilution


CHAPTER FOUR
THE IDENTIFICATION OF ANTI-IDIOTYPIC ANTIBODY
Introduction
Anti-idiotypic antibody has been produced by immu
nizing an animal with isologous or autologous antibody in
adjuvant (40-42). The use of: isologous or autologous anti
body rather than homologous antibody eliminates the poten
tial that allotypic determinants might be recognized rather
than idiotypic determinants. The immunization schedule
used in the previous experiments included the admini
stration of autologous antibody in adjuvant. It was hoped
that this treatment would regulate IgE antibody, but unfor
tunately, it did not. It was not known if this failure was
because of a lack of an anti-id response or for other
reasons. This treatment may have induced anti-id. It is
also possible that anti-id may have been produced during
the immunization with antigen. The purpose of the experi
ments in this chapter was to determine if, at any time
during the immunization schedule, anti-id was detectable.
77


Figure 23.
The identification of anti-id in various serum samples, in dog 24,
as measured by RIA. The dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and CFA at weeks 7 and 9. The arrow, marked
idiotype probe, indicates the time from which the anti-DNP F(ab)'2
fragments came. These were used as antigens to detect anti-id.
The bars represent the standard deviation of the mean.


10
o 8


Figure 13.
The identification of anti-id in various serum samples over time in
dog 14 as measured by RIA. The dog received DNP-ASC in adjuvant at
weeks 0,2,4,6,8 and 10, and autologous anti-DNP antibody in
adjuvant at weeks 7 and 9. The arrow, marked idiotype probe,
indicates the time from which the anti-DNP F(ab)'2 fragments cane.
These were used as antigens to detect anti-id. The bars represent
the standard deviation of the mean.


Table 9
The Relative Anti-DNP IgE Concentration in 28 Dogs a)
69
Group
lb) (10
gg Anti-DNP Antibody
in CFA)
Animal
Number
1
2
3
4
Weeks
3
2.0+.17
6.4+.29
6.1+.49
8.1+.36
9
2.6+.26
9.3+.84
6.5+.25
9.4+.36
10
3.0+.18
8.7+.31
3.4+.30
7.1+.79
11
2.8+.31
9.3+1.26
4.1+.19
8.2+.81
Animal
Number
5
6
7
8
Weeks
8
5.2+.57
15.6+.70
6.6+.6S
7.2+.30
9
8.9+.63
14.0+.69
6.8+.78
9.2+1.18
10
9.3+.72
13.2+.36
8.2+.14
8.2+.07
11
10.2+.64
15.1+.47
7.4+.66
8.6+.42
Group 2 (100 gg
Anti-DNP Antibody in
IFA)
Animal
Number
9
10
11
12
Weeks
8
9.8+.47
10.0+.22
5.7+.67
5.S+.86
9
9.5+1.89
9.7+.46
7.5+.83
S.5+.33
10
9.8+.67
8.3+.57
4.1+.65
6.8+.41
11
9.4+.59
4.1+.12
5.3+.41
9.3+.17
Group
3 (100 gg Anti-DNP Antibody in CFA)
Animal
Numbsr
13
14
15
16
Weeks
8
2.4+.02
3.3+.30
0
4.2+.21
9
4.1+.62
4.0+.50
0
2.9+.14
10
2.0+.39
6.3+.35
0
7.7+.79
11
3.0+.48
9.6+.47
0
2.9+.29
Animal
Number
17
18
19
20
Weeks
8
6.1+.22
5.6+.49
.6+.01
7.6+.51
9
9.2+.35
8.9+.46
6.0+.43
7.0+.88
10
6.4+.57
5.2+.34
4.5+.06
6.8+.50
11
7.3+.87
6.8+.65
6.4+.06
7.2+1.01


37
Reciprocal of Dilution


3
The induction of IgE antibody experimentally requires
special conditions. For example, high doses of antigen and
strong adjuvants such as complete Freund's adjuvant are
unfavorable to the development of an IgE response whereas
low doses of antigen in an adjuvant such as aluminum
hydroxide tend to favor IgE production (13). Furthermore,
if haptens are coupled to parasite extracts, high titer
anti-hapten IgE antibody responses will frequently develop.
However, if the same hapten is coupled to a T-independent
antigen or a different T-dependent carrier there is usually
no IgE response (13,14). 'This suggests that IgE production
is dependent on both the carrier and T-cells. The reason
why parasites and their extracts are efficient inducers of
IgE antibody is not fully understood. It is known that this
enhancing effect is modulated through factors produced by T
cells. Ishizaka's group (15) have shown that T cells
derived from (N. brasiliensis) parasitized rats produce
an IgE-potentiating factor which selectively potentiates a
non-specific IgE antibody response. This factor has
affinity for IgE, binds to IgE-bearing B cells through
surface IgE and enhances the differentiation of these cells
into IgE forming cells. A factor with similar properties
has been produced from T-cells obtained from patients with
hyper-IgE syndrome, suggesting that the regulatory factors


45
Table 1
The Relative Anti-DNP IgM Concentration in 28 Dogs
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
Animal
Number
Weeks
0
1
2
3
4
5
6
7
1
0
16.6+.87
10.6+1.23
.3+0
.2+.01
0
0
0
5
0
2.9+.25
5.8+.90
0
0
0
0
0
9
0
0
2.0+.14
3.3+.29
0
0
0
0
13
0
5.5+.16
5.0+.23
4.8+.07
.9+.13
0
0
0
0
4.4+.39
3.2+.23
1.4+.56
.2+0
0
0
0
3
0
6.5+.47
3.4+.24
0
0
0
0
0
0
4.5+.06
4.4+.27
.6+.14
.1+.03
0
0
0
6
0
6.4+.82
5.6+.39
5.1+.07
3.9+.24
0
0
0
10
0
10.9+1.03
3.0+.41
.8+.01
0
0
0
0
14
0
1.8+.18
5.2+.33
2.3+.16
.5+.22
0
0
0
7
0
6.8+1.01
7.S+.32
3.4+.31
0
0
0
0
n
0
7.0+.02
2.8+.46
1.8+.19
1.0+.07
0
0
0
15
0
10.1+.79
6.1+.43
1.8+109
1.6+.17
0
0
0
0
3.1+.28
5.2+.17
0
0
0
0
0
12
0
5.0+.25
5.1+.37
2.0+.21
1.8+.27
0
0
0
16
0
1.3+.12
2.6+.37
2.5+.14
1.2+.20
0
0
0


o
8
Weeks
Mean S.D.
(range)
0 0
(0-0)
1 6.1 4.3
(0-13.8)
2 5.1 3.3
(.7-10.6)
3 23 1.9
(0-42)
4 8 1.0
(0-3.9)
5 .03 .1
(0-.8)
6 0
(0-0)
7 0
(0-0)


Figure 21.
The identification of anti-id in various serum samples, in dog 14,
as measured by RIA. THe dog received DNP-ASC in adjuvant at weeks
0,2,4,6,8 and 10, and autologous anti-DNP antibody at weeks 7 and
9. The arrow, marked idiotype probe, indicates the time from which
the anti-DNP F(ab)'2 fragments came. These were used as antigens
to detect anti-id. The bars represent the standard deviation of
the mean.


85
than the anti-IgG antisera, indicating that it was of the
IgG class.
When three different anti-DNP monoclonal antibodies
and three control monoclonal antibodies were used, the
binding activity was detected only with the original anti-
DNP IgG (table 13), and to a lesser extent, to anti-DNP
IgM~2 (table 14). In those animals in which anti-DNP
IgM~2 binding activity was detected, a comparison was
made between the serum from a point in time prior to autol
ogous antibody administration and serum obtained after such
treatment. A minimum value that was two standard devi
ations above the mean of the control was considered indi
cative of anti-id activity. As was the case with the IgG
antibody, only those animals which received autologous anti
body, showed binding activity and only after administration
of autologous antibody (table 14).
The Role of Antibody in the Specificity of the
Anti-Idiotypic Production
To determine if immunization with an antibody whose
specificity was other than anti-DNP would result in anti-
DNP/anti-id, eight dogs were given both DNP/ASC and ABA/KLH
three times at two week intervals. Six weeks after the
primary injection of antigen, three dogs received 100 gg
of autologous anti-DNP antibody in CFA, and a different
three dogs received 100 gg autologous anti-ABA antibody in


105
Table 18
The Detection of Canine Anti-Idiotypic Antibody
by RIA Using Autologous Anti-DNP Fiab)'^ as the Id
Dog Number 1
Source of Anti-id (Week) a)
2 4 6 8 11
Source of
id
(Week) b)
Effluent
Dilution
4
0
82+64
12+12
1332+112
2035+279
150+48
1/2
41+70
51+38
939+99
1519+199
286+45
1/4
63+49
0
171+34
1069+170
14+18
6
0
37+31
193+39
611+50
1241+179
1586+42
1/2
14+13
86+55
534+37
370+27
1333+26
1/4
10+21
3+5
5835
12+25
413+53
7
0
3+5
253+26
455+37
1496+14
1300+17
1/2
43+12
279+121
397+61
534+29
714+51
1/4
38+27
21+28
179+41
30+21
346+15
10
0
51+25
426+41
556+47
595+257
459+101
1/2
14+23
349+153
438+32
437+89
257+57
1/4
150+63
79+71
214+83
139+81
179+68
Control c)
0 186+9
1/2 149+31
1/4 138+46
101+7
a) Serum from different times during the immunization schedule.
b) The id was autologous anti-DNP F(ab)' immobilized to a solid
matrix. The dog was immunized with DNP-ASC in adjuvant at week
0,2,4,6,8 and 10 and received 10 gg autologous anti-DNP antibody in
CFA at weeks 7 and 9.
c) Control id was normal canine IgG Ftab)^ immobilized to a solid
matrix.


98
Summary
The results in this chapter suggest the following:
Anti-id can be induced by the administration of autologous
antibody. The majority of this anti-id is not hapten
inhibitable and is detectable with only a few monoclonal
antibodies of the same specificity which presumably bear
tiie same or a similar set of cross reactive idiotypes.
Furthermore, this anti-id is not an internal image of
antigen.
Conclusions
1) The administration of autologous antibody in
adjuvant induces a reciprocal anti-idiotypic antibody
response. 2) The identification of these anti-id anti
bodies was achieved by the use of monoclonal anti-DNP
antibody from another species as a source of idiotype.


Figure 4.
Dilutions of tlie standard anti-DNP IgE serum sample assayed by
RIA. The bars represent the standard deviation of the mean.


Figure 6.
Dilutions of the standard anti-DNP IgM serum sample
assayed by RIA. The bars represent the standard
deviation of the mean.


51
Table 3 Continued
Animal
Number
17
18
19
20
Weeks
1
0
0
0
0
2
2.6+.13
.9+.04
.3+0
0
3
1.7+.14
1.7+.13
4.8+.17
11.7+.11
4
4.8+.36
22.7+.81
16.9+.65
12.8+.96
5
6.2+.58
24.9+1.65
21.3+.17
8.9+.04
6
6.9+.31
21.8+.77
20.8+1.16
7.6+.47
7
5.3+.40
28.2+1.24
21.9+.84
9.8+.36
Animal
Number
21
22
23
24
Weeks
1
0
0
0
0
2
0
7.1+.33
12.9+.92
3.6+.24
3
5.6+.46
7.1+.33
12.9+.92
3.6+.24
4
5.4+.27
10.7+.75
18.0+.75
S.8+.43
5
8.5+.12
13.8+.99
13.9+.51
9.7+.81
6
10.6+.93
19.1+1.53
13.5+.87
9.0+.29
7
7.6+.46
18.2+.75
12.7+.60
6.6+.64
Animal
Number
25
26
27
28
Weeks
1
0
0
0
0
2
0
0
.8+.07
.4+0
3
10.9f.47
3.3+.33
9.6+.4S
X
4
10.0+1.79
16.3+.84
10.3+.68
12.9+.85
5
S.2+.97
15.4+.93
12.1+.06
22.8+1.68
6
7.8+.11
14.7+.43
18.7+1.78
26.9+1.41
7
5.0+.36
10.5+.70
19.6+.96
28.7+1.92
a) Each dog received immunization with DNP-ASC in adjuvant at
weeks 0,2,4 and 6.
b) The units were calculated from a relative antibody-
concentration scale derived from the titration of a serum sample
containing anti-DNP IgG. A value of zero indicates no detectable
anti-DNP IgG. The data was the mean antibody concentration of a
sample run in triplicate + the standard deviation from the mean.
This was calculated by adding and subtracting the standard
deviation to the mean cpm and calculating the relative antibody
concentration for these numbers. This number vas then subtracted
from the mean concentration.


Int.
22. Faith, R.E., Hessler, J.R., Small, P.A. (1977)
Respiratory Allergy in the Dog: Induction by the
Respiratory Route and the Effect of Passive Antibody.
Arch. Allergy Appl. Immunol. 53:530-544.
23. Smith, T. (1909) Active Immunity Produced by
So-Called Balanced or Neutral Mixtures of Diphtheria Toxin
and Antitoxin. J. Exp. Med. 11:241-256.
24. Uhr, J.W., and Moller, G. (1968) Regulatory Effect of
Antibody on the Immune Response. Adv. Immunol. 8:81-127.
25. Chan, P.L., and Sinclair N.R. St. C. (1971)
Regulation of the Immune Response. V. An Analysis of the
Function of the Fc Portion of Antibody in Suppression of an
Immune Response with Respect to Interaction with Components
of the Lymphoid System. Immunol. 21:967-981.
26. Strannegard, 0., and Belin, L. (1970) Suppression of
Reagin Synthesis in Rabbits by Passively Administered
Antibody. Immunol. 18:775-785.
27. Tada, T., and Okumura, K. (1971) Regulation of
Hcmocytotropic Antibody Formation in the Rat. 1. Feed-Back
Regulation by Passively Administered Antibody. J. Immunol.
106:1002-1011.
28. Ishizaka, K., and Okudaira, H. (1972) Reaginic
Antibody Formation in the Mouse. 1. Antibody-Mediated
Suppression of Reaginic Formation. J. Immunol. 109:84-89.
29. Lahss, F., Weiler, E., and Hillmann, G. (1953)
Myelom-Plasma-Proteine. Ill Mitteilung zur Immunochemie der
Gaima-Myelom-Proteine. Z. Naturforsch, Teil., 8B:625-631.
30. Jerne, N.K. (1974) Towards a Network Theory of the
Immune System. Ann. Immunol. 125(c):373-389 .
31. Brient, B.W. and Nisonoff, A. (1970) Quantitative
Investigations of Idiotypic Antibodies. IV. Inhibition by
Specific Haptens of the Reaction of Anti-Hapten Antibody
with its Anti-Idiotypic Antibody. J. Exp. Med.
132:951-963.
32. Claflin, J.L., and Davie, J.M. (1975) Specific
Isolation and Characterization of Antibody Directed to
Binding Site Antigenic Determinants. J. Immunol. 114:70-75.


127


cpm x 100
0 2 4 6 8 10 12
Weeks


Figure 11.
The mean relative anti-DNP IgG concentration as measured by RIA.
All dogs received DNP-ASC immunization at weeks 0,2,4,6,8 and 10.
At weeks 7 and 9, the treatment consisted of immunization with:
Group 1, 10 pg autologous anti-DNP antibody in CFA; Group 2, 100
pg autologous anti-DNP antibody in IFA; Group 3, 100 pg
autologous anti-DNP antibody in CFA; Control, either IFA or CFA
alone. The relative antibody concentration was calculated from an
arbitrary antobody concentration scale derived from the titration
of a standard anti-DNP IgG containing serum (see text for further
details).


88
CFA. 'Two dogs received CFA alone. Serum before and two
weeks after this treatment was screened for anti-id
activity. As seen in table 15, the dogs that received
autologous anti-DNP antibody produced an anti-id which was
detected by the anti-DNP IgG mouse monoclonal antibody and
failed to bind to the anti-ABA IgG. Dogs that received
anti-ABA antibody in adjuvant developed anti-id which bound
to anti-ABA mouse monoclonal IgG but failed to bind to the
anti-DNP IgG mouse monoclonal antibody. The two control
dogs produced no detectable anti-id that was reactive with
either mouse monoclonal antibody.
Hapten Inhibition and Elution Studies of the Id/anti-id
Interaction
The anti-idiotypic RIA was used to determine if
hapten could inhibit the binding of anti-id to the mouse
anti-DNP antibody. In two of the six cases (5,15), 10 gg
of hapten was able to inhibit id/anti-id interaction (table
16) as shown by a slight decrease in the cpm bound to the
wells as compared to the same sample incubated with PBS (22
percent and 26 percent inhibition respectively). As the
concentration of hapten decreased, so did the percent
inhibition, (17 percent and 5 percent at a fivefold
decrease in hapten concentration). However, it is unclear .
how significant this inhibition was because of the
extremely large amounts of hapten required to obtain these


60
Materials and Methods
Affinity Chromatography
Anti-DNP antibody was produced by immunizing dogs to
DNP-ASC and was purified from serum by chromatography
through a DNP-HSA affinity column as described in Chapter
two. The bound antibody was eluted with 0.1 M glycine HCl,
pH 2.5.
RIA
The RIA for detection of anti-DNP antibody was
described in Chapter two.
Animals and Immunization Schedule
The same dogs that were described in Chapter two were
used in these experiments. These dogs had received 100 pg
of aluminum hydroxide precipitated DNP-ASC by the intra-
peritoneal route on the day of birth at two week intervals
on three further occasions. Fifteen milliliters of serum
were obtained from each dog at the time of final antigenic
challenge. Anti-DNP antibody was purified from this serum
by DNP-HSA affinity chromatography, concentrated to about 3
mg/ml by negative pressure dialysis and rendered
bacterially sterile by passing through a filter having 0.2
micron sized pores. Dogs received either 10 or 100 gg of
their own antibody emulsified in 2 ml of either complete


2
There are a number of unique features of IgE antibody-
synthesis. Firstly, IgE circulates in very small amounts as
compared to other antibody classes. In man, the serum level
of this antibody is about 1/10,000 the level of serum IgG
(3), and in dogs serum IgE is about 1/100 the level of serum
IgG (7). Serum IgE levels of internally parasitized people
and dogs are elevated as compared to non-parasitized indivi
duals (7,8,9). 'The higher IgE level in dogs is felt to be
the result of a greater parasite burden in this species (7).
Secondly, IgE is produced predominantly locally by
lymph nodes in the respiratory and gastrointestinal tracts
as well as in regional lymph nodes (10). These observa
tions have led to the suggestion that this immunoglobulin is
important in host defense of mucosal surfaces and partic
ularly against parasites. Furthermore, IgE has been shown
to participate in parasite killing through antibody depen
dent cell mediated cytotoxicity (11). Thirdly, the antigens
that stimulate IgE antibody are usually very complex and
heterogenous substances such as allergens or parasites and
their extracts. When an animal is exposed to these anti
gens, the antibody response usually includes high titer IgE
antibody whereas bacteria and viruses usually do not induce
IgE antibody in spite of being very immunogenic (12).


55
Discussion
The purpose of the experiments in this chapter vas to
induce an anti-DNP antibody response which included IgE and
to examine the kinetics of this response. As described in
the results section of this chapter, the anti-DNP IgE, IgG
and IgM antibody response followed expected kinetics
(71-73). The IgM response was present before IgE or IgG
antibody was detected and disappeared after the sixth week
in spite of continued antigenic challenge. The IgE and IgG
production had a two week lag period in general, but once
they developed, they were maintained throughout the immuni
zation course.
If inbred laboratory animals such as mice are
immunized with an antigen, a homogeneous response
ordinarily results (74). However, in species that are
genetically heterogeneous, such as man and dogs, the immune
response to the antigen would be expected to be more highly
variable (74). In these outbred dogs there was marked
variation in the kinetics and magnitude of the antibody
responses. The marked variations seen in these dogs are
most probably the result of the genetic differences between
than. In this context, it was noteworthy that the IgG
response was more homogeneous within the same litter than
between litters.


Table 8
The Mean Relative Anti-IgE Antibody Concentration
64
Group 1 a)
Group 2
Group 3
seks
0
0
0
0
1
. 1
+ .2
0
.4 + .85
2
3.5
+ 3.2
2.1
+ 3.5
3.8 + 3.4
3
8.7
+ 4.3
4.9
+ 4.4
10.7 + 6.5
4
6.3
+ 2.6
5.2
+ 3.6
7.7 + 1.6
5
9.6
+ 4.5
5.8
+ 4.2
11.4 + 10.5
6
6.7
+ 4.9
3.8
+ 3.0
3.7 + 4.5
7
8.0
+ 5.3
7.8
+ 6.3
10.4 + 4.6
8
7.2
+ 3.9
3.7
+ 2.7
7.8 + 2.5
9
8.3
+ 3.2
5.3
+ 3.1
8.8 + 1.0
10
7.6
+ 3.3
4.9
+ 2.6
7.3 + 2.4
11
8.3
+ 3.8
5.4
+ 3.1
7.0 + 2.7
Group 4
Group 5
Control
jeks
0
0
0
0
1
0
0
0
2
2.3
+ 1.1
1.1
+ .8
1.4 + 1.0
3
3.6
+ 1.3
2.0
+ .9
2.4 + 1.2
4
3.5
+ .7
2.9
+ 2.3
3.0 + 2.0
5
4.1
+ 2.1
2.6
+ 3.1
3.0 + 2.8
6
2.6
+ .4
1.9
+ 1.5
2.1 + 1.3
7
3.7
+ 1.6
2.4
+ 3.0
2.7 + 2.6
8
6.6
+ 2.0
2.8
+ 1.9
3.8 + 2.5
9
8.3
+ 1.0
3.2
+ 2.2
4.4 + 3.0
10
7.1
+ .1
2.7
+ 2.2
3.8 + 2.7
11
6.9
+ 2.3
3.0
+ 2.3
4.0 + 2.7
a) All dogs received DNP/ASC immunization at 0,2,4,6,8 and 10
weeks. At 7 and 9 weeks: Group 1 received 10 gg autologous
anti-DNP antibody in CFA, Group 2 received 100 gg autologous
anti-DNP antibody in IFA, Group 3 received 100 gg autologous
anti-DNP antibody in CFA, Group 4 received IFA alone, Group 5
received CFA alone. Control values were the mean of groups 4 and
5.


9
This has been demonstrated by hapten inhibition studies.
Brient and Nisonoff (31) induced anti-p-azobenzoate anti
bodies in rabbits. These antibodies were purified and
injected into allotypically matched rabbits and the resul
tant antiserum bound to determinants present on some rabbit
anti-p-azobenzoate antibodies. They then studied the
effects that adding increasing concentration of hapten would
have on the reaction between radiolabelled anti-azobenzoate
antibodies and the anti-idiotypic antiserum. They found
that the binding affinities of the benzoate derivatives
correlated closely with their ability to inhibit the
antibody/anti-id interaction. In many other studies
(32-34), anti-id was induced in animals immunized with an
anti-hapten antibody. This anti-id was purified from the
sera by initial adsorption to an affinity column having the
immunizing antibody bound to it and was then eluted with the
appropriate hapten. This purification process then would
select for anti-idiotypic antibodies which were directed to
those idiotypic determinants very close to or within the
antigen binding site and it would be expected that hapten
could inhibit the id/anti-id interaction.
On the other hand, it is not always possible for hapten
to inhibit id/anti-id. For example, Sher and Cohn (35)
showed that there was variation in the ability of hapten to
inhibit id/anti-id interaction. Hapten was not able to


123


94
the ability of the anti-id to bind to these two antigens.
This difference could be a function of 1) different idio
typic determinants present on the two antibodies, or, 2)
the difference in the accessability of tine id to the
anti-id or 3) a combination of both of these.
Id/anti-id interactions can, in many cases, be inhi
bited by hapten. If tine interaction is hapten inhibitable,
it suggests that anti-id binds to id determinants within
the antigen combining portion of an antibody molecule or to
idiotypes intimately associated with this region. In those
instances where hapten is unable to inhibit this inter
action, it can be concluded that anti-id is binding to
those ids not within the antigen binding site. It also
indicates that anti-id does not act as an internal image of
antigen. High concentrations of hapten relative to the
amount of antibody on the plate were preincubated with
mouse anti-DNP IgG antibody. The presence of hapten did
not consistently inhibit canine anti-id/mouse id inter
action. Only two of the six samples tested showed inhi
bition, with 27 percent being the maximum inhibition. In
other experiments, hapten could not displace the anti-id
from the mouse anti-DNP antibody. These results suggest
that the majority of anti-id is not binding to structures
within the antigen binding site of the mouse monoclonal
anti-DNP IgG. An anti-id and an internal image of antigen
both bind to structures within the variable regions of an


57
is the result of a suppression of T-suppressor cells.
Because the mechanism that would normally suppress IgE
synthesis is altered, IgE antibody response will develop.
This alteration in the suppressive network and subsequent
IgE antibody synthesis has been called the "allergic break
through" (79).
Summary
Twenty-eight dogs immunized to DNP-ASC at birth and
then three times at two week intervals produced serum anti-
DNP antibody. The IgM response was detected one week after
primary immunization and lasted for up to five weeks. The
IgE and IgG antibody response in general was not present
until week three but persisted through the immunization
schedule. Although variation in the level and duration of
the antibody response was detected between individual dogs,
each dog did have a response that included all three
isotypes examined.
Conclusions
(1) Dogs immunized with DNP-ASC develop a high
level, long term IgE and IgG antibody response but the IgM
response followed a different kinetic pattern in that it
did not persist after week five.


Figure 7.
The mean relative anti-DNP IgM concentration in 28 dogs as measured
by RIA. Each dog was immunized biweekly four times starting at
week 0 with DNP-ASC (see text for further details). The antibody
concentration was calculated from an arbitrary antibody
concentration scale derived from the titration of a standard
anti-DNP IgM containing serum.


Table 10
The Relative Anti-DNP IgG Concentration in 28 Dogs a)
71
Group 1 b) (10 pg Anti-DNP Antibody in
CFA)
Animal
Number
1
2
3
4
Weeks
8
9.0+.11
24.7+.45
6.6+.86
8.3+.75
9
9.0+.66
27.4+2.50
10.0+.52
7.5+.78
10
10.2+.93
28.5+.76
10.1+.77
6.2+1.05
11
9.1+1.04
26.8+.39
6.9+.30
9.S+.76
Group 1 Continued
Anina 1
Number
5
6
7
8
Weeks
8
5.5+.23
24.7+1.73
9.4+1.37
17.3+.75
9
4.6+.68
27.0+.39
14.0+.38
20.3+.72
10
5.S+.89
29.1+2.13
14.7+87
17.6+.48
11
8.9+.24
26.6+.90
Group 2 (100 pg
12.3+.99
Anti-DNP Antibody
19.7+1.29
in IFA)
Animal
Number
9
10
11
12
Weeks
8
17.1+1.37
8.9+.76
8.1+.62
14.2+.83
9
20.9+1.47
8.8+.99
9.2+.50
16.4+.38
10
22.9+1.10
11.9+.26
8.7+.61
15.4+1.12
11
25.4+.43
11.3+.69
Group 3 (100 pg
8.3+.90
Anti-DNP Antibody
15.0+.32
in CFA)
Animal
Number
13
14
15
16
Weeks
8
5.6+.17
13.9+.83
7.9+.67
16.7+.99
9
10.3+.62
19.0+.65
11.8+1.12
16.0+.87
10
12.7+.41
20.4+.11
14.3+1.57
15.0+1.01
11
12.2+.73
20.3+2.24
14.7+.59
17.4+2.56


132
Both the first and second pattern of anti-id are
consistent with the hypothesis that id is acting as an
antigen to induce anti-id. These data are raniniscent of
the type of curves seen when one plots the disappearance of
antigen as a function of time and superimposes on that
curve the appearance of antibody that is specific for the
antigen (90). When antigen is first introduced into an
animal there is initially a very slow loss of this antigen
from the circulation. After a few days, however, there is
a precipitous drop in the level of antigens which is the
result of antibody production and is called immune elimi
nation. Antibody when first produced is not detected
because it is cornplexed with antigen and removed from the
circulation. At a certain point, however, both antibody
and antigen will become apparent in a cornplexed form. The
variables that determine this point include the valence of
the antigen and its size, the isotype of the antibody, the
affinities between the antibody and the antigen, and the
efficiency of the reticuloendothelial system in removing
these complexes (90,91). The similarity between immune
elimination of antigen and the experimental results
obtained suggest that id is removed in a fashion analagous
with antigen removal. The appearance of anti-id in the
third pattern is difficult to explain. In this case,
anti-id was present at a time considerably before the
appearance of id. That is, in dog 14, anti-id was


Figure 14.
The identification of anti-id in various serum samples over time,
in dag 1, as measured by RIA. The dog received DNP-ASC in adjuvant
at weeks 0,2,4,6,8 and 10 and autologous antibody in adjuvant at
weeks 7 and 9. The arrow, narked idiotype probe, indicates the
time from which the anti-DNP F(ab)'2 fragments came. These were
used as antigen to detect anti-id. The bars represent the standard
deviation of the mean.


74
treatment within a single litter. In no case was a signif
icant difference observed.
Discussion
The fact that anti-DNP antibody could not be detected
in the effluent from the affinity column indicates that the
column was effective in removing all anti-DNP antibody.
The inability to detect IgE in the glycine eluate was
expected because canine IgE is not stable at low pH.
Halliwell (7) has shown that at a pH of 2.5 for 30 minutes
there is greater than a tenfold decrease in detectable IgE
antibody.
There are at least four possible reasons why autol
ogous antibody administration failed to regulate the anti
body response in these dogs as had been achieved in labora
tory animals. Firstly, in these experiments the dogs had
an established antibody response whereas in many of the
experimental systems where passive antibody showed regu
latory effects on antibody production, a primary or early
secondary response was manipulated. It has been shown that
it is much more difficult to manipulate a preexisting and
established response than to alter a developing one.
Secondly, there may be something unique about the regu
latory effects of passive antibody on an immune response in
young animals. Antibody is transferred from mother to


30
triplicate set of wells in which PBS rather than serum was
added. The mean cpm from these controls were subtracted
from the cpm of the test sample. Although the values
varied from experiment to experiment, the maximum cpm of
these controls were consistently lower than the lowest
values obtained for test samples.
A standard serum sample was included with each
assay as an internal reference. An arbitrary antibody
concentration was determined by assigning a value of 64
units to the undiluted standard IgG and IgE sample and 32
units to the undilute IgM standard. By interpolating from
the linear portion of the standard curve, the relative
units of antibody for test samples were calculated.
Animals and Immunization Schedule
Outbred pregnant female dogs were obtained from the
Division of Animal Resources, University of Florida. Serum
from these dogs was screened by RIA to ensure that they did
not have anti-DNP antibody at the time of whelping.
The puppies of these bitches were used as experimental
animals. Serum samples were obtained on the day of birth
and weekly therafter. Each puppy received 100 gg aluminum
hydroxide precipitated dinitrophenol coupled ascaris
antigen by the intraperitbheal route on the day of birth
and at two week intervals on three further occasions. Each


10
o
o
x
8
6
2
0 2 4 6 8 10
Weeks


44. Rodkey, L.S. (1980) Autoregulation of Inmune
Responses via Idiotype Network Interaction. Microbiol. Rev.
44:631-659.
45. Bankert, R.B., Bloor, A.G., and Jou, Y.H. (1982)
Idiotypes: Their Presence on B and T Lymphocytes and their
Role in the Regulation of the Immune Response. Vet.
Immunol. Immunopath. 3:147-184.
46. Greene, M.I., Nelles, M.J., and Man-Sun, S. (1982)
Regulation of Immunity to the Azobenzenearsonate Hapten.
Adv. Immunol. 32:254-300.
47. Lynch, R.G., Rohrer, J.W., Odermatt, B., Gebel, H.M.,
Autry, J.R., and Hoover, R.G. (1979) Immunoregulation of
Murine Myeloma Cell Growth and Differentiation: A
Monoclonal Model of B Cell Differentiation. Immunol. Rev.
48:45-80.
48. Cosenza, H., and Kohler, H. (1972) Specific
Inhibition of Plaque Formation to Phosphorylcholine by
Antibody Against Antibody. Science 176:1027-1029.
49. Cosenza, H., and Kohler, H. (1972) Specific
Suppression of the Antibody Response by Antibodies to
Receptors. Proc. Natl. Acad. Sci. U.S.A. 69:2701-2705.
50. Bankert, R.B., and Pressman, D. (1976)
Receptor-Blocking Factors Present in Immune Serum Resembling
Auto-Anti-Idiotype Antibody. J. Immunol. 117:457-462.
51. Kelsoe, G., and Cerny, J. (1979) Reciprocal
Expansions of Idiotypic and Anti-Idiotypic Clones Following
Antigen Stimulation. Nature. 279:333-334.
52. Geha, R.S. (1983) Presence of Circulating
Anti-Idiotype Bearing Cells after Booster Immunization with
Tetanus Toxoid (TT) and Inhibition of Anti-TT Antibody
Synthesis by Auto-Anti-Idiotypic Antibody. J. Immunol.
130-1634-1639.
53. Dwyer, D.S., Bradley, R.J., Urquhart, C.K., and
Kearney, J.F. (1983) Naturally Occurring Anti-Idiotypic
Antibodies in Myasthenia Gravis Patients. Nature
301:611-614.


2D
2,4-dinitrobenzenesulphonic acid (DNP) (Eastman Kodak Co.,
Rochester, N.Y.) were mixed in distilled water (70). This
was then incubated while gently stirring for 18 hours at
room temperature. The solution was chromatographed through
a Sephadex G-25 column (Pharmacia Fine Chemicals,
Pi seataway, N.J.) to separate bound from free DNP. The
dinitrophenylated protein was concentrated by negative
pressure dialysis and extensively dialyzed against PBS, pH
7.2. The extent of substitution was estimated by measuring
light adsorption at 360 nm and assuming a molar extinction
coefficient oE 1.75 x 10^ for the dinitrophenyl group.
The average epitope density expressed as molecules of DNP
per molecule carrier was DNP^-HSA, DNP14 g-BGG. Since
ASC extract vas a complex mixture of proteins, the extent
of substitution was expressed as moles DNP/mg ASC and was
6.32 x 10 ^ DNP/ASC. A single batch of each of these
antigens was prepared and used throughout the experiment.
These antigens, when not in use, were stored at -70C.
The degree of substitution did not change due to storage.
Aluminum hydroxide Precipitation of Protein
Aluminum hydroxide precipitation of protein was
performed by mixing one part of a 5 percent sterile
solution of aluminum potassium sulfate (AIK (30^)^),
Mallincrodt, Paris, Kentucky) with five parts of 1 mg/ml


56
The genetic makeup of the animal also influences the
class of antibody produced following antigen challange. In
certain inbred animal strains, it is very difficult to
mount an IgE antibody response without some type of manipu
lative process to eliminate T-suppressor cells (75,76).
Furthermore, if a comparison is .made between allergic and
non-ailergic people, a marked difference in antigen-
specific IgE responsiveness to certain antigens is seen.
Those individuals with allergic tendencies will have an
enhanced IgE response to allergens whereas non-allergic
people may not develop IgE antibody (17). There were
notable differences between individual dogs in terms of
their IgE response and in contrast to the IgG response
there was no consistent pattern within litters. It is not
clear if this failure to see similar patterns within a
litter in the IgE level reflects the antigen chosen to
study IgE in these dogs or if there are multiple genes that
govern IgE levels in dogs. By having such a small sample
size, a consistent pattern might not be observed for IgE
levels. One of the dogs failed initially to develop an IgE
titer. The IgE antibody response started after this dog was
vaccinated with a modified-live canine distemper/hepatitis
vaccine at four weeks of age. The immunization of dogs
with this vaccine has been shown to enhance antigen-
specific IgE response to an antigen administered at the
same time (78). It has been hypothesized that this effect


101
KH2(P04) buffer, pH 6.3 and a proportionate amount of
beads were incubated together for one hour at 4C
followed by the addition of l-ethyl-3 (3-dimethylamino
propyl) carbodiimide HC1 (EDAC) with an additional incu
bation at 4C for one hour. Any remaining active sites
were blocked with 1 percent HSA in 0.005 phosphate buffer,
pH 7.2 by incubating this with the beads for one hour at
roan temperature. The beads were pelleted by centri
fugation at 1,000 x g for 10 minutes at 4C and alter
nately washed with PBS, pH 7.2 followed by 1.4 M NaCl-PBS,
pH 7.2 three times to ramove unbound protein. After the
final wash the beads were suspended in RAST+ buffer. The
beads used in a single experiment were standardized for
both DNP-HSA binding and total F(ab)1^ content by incu
bating an aliquot from each bead set with various dilutions
of I DNP-HSA or I anti-canine light chain anti
body. For example, 50 pg of one bead set bound 5,656 +
61 cpm radiolabelled anti-canine light chain specific
antibody (this number of cpm is approximately 140 ng of
anti-canine light chain antibody) and 1,040 + 53 cprn
radiolabelled DNP-HSA. A second set bound 4,690 + 79 cpm
anti-light chain antibody (this is approximately 110 ng of
anti-canine light chain antibody), and 747 + 21 cpm
antigen. The second set of beads had approximately 75
percent of the binding capacity of the first set.
Therefore 63 pi of beads from the second set were used in


104
antibody might have. Anti-id was detected during the
DNP-ASC immunization schedule in three of the five dogs
tested (tables 18-20) but no anti-id was evident in the
other two dogs. The kinetics and the amount of anti-id
varied depending upon what set of ids were used an antigens.
and which serum sample was tested. Three different
patterns in the appearance of anti-id are seen: Pattern
one: Anti-id could not be detected before or coincident
with the id but could be detected later, as was seen with
two samples in two dogs (figure 13,14). In dog 14, the ids
used to detect anti-id were from week two, anti-id was not
detected until week seven (figure 13). Similarly, in dog
1, when the ids from week four were used as antigens,
anti-id was rot detected until week six (figure 14).
Pattern two w/as seen in three dogs using six serum samples.
In a single sample, id and anti-id were both present
(figure 15-20). For example, when the antibody obtained at
week six was used as an id antigen, anti-id was detected at
week six but the maximum level of anti-id was later than
week six (figure 15). In two samples assayed, the highest
level of anti-id was detected from the same samples that
were used to obtain the antibody which was used as the id
antigen (figures 19 and 20). Pattern three: In dog 14,
when the ids which were used as antigens to detect anti-id
w?ere from blood obtained at week 11, anti-id was detected
with each sample tested (figure 21). Similarly, by using


62
Results
DNP Affinity Column
There was no detectable anti-DNP antibody in the
serum of any dog after passage through the DNP affinity
column. On the other hand, the glycine HCl eluate
contained high levels of anti-DNP IgG but no detectable
anti-DNP IgM or IgE. Because anti-DNP IgE was not
detected in either the effluent or the eluent from the
affinity column but was detectable in the serum prior to
such treatment, an aliquot of serum containing anti-DNP IgE
was dialyzed against glycine HCl, pH 2.5 followed by
dialysis against PBS, pH 7.2 to determine what effects
glycine HCl had on canine IgE. There was no detectable
anti-DNP IgE in this serum after such treatment as assayed
by RIA.
Immunization with Autologous Antibody
As noted in Chapter two, there was considerable
variation in the anti-DNP antibody respone between dogs.
Tables 7 and 8 show the mean relative concentration of
anti-DNP IgG and IgE respectively in each group of dogs
prior to the autologous antibody administration and there
after. These data are presented graphically in figures 11
and 12. The individual relative antibody concentrations


47
Table 2
The Relative Anti-DNP IgE Concentration In 28 Dogs
Following
Immunization with
DNP-ASC a)
Animal
Number
1
2
3
4
Weeks
0
0 b)
0
0
0
1
0
0
0
0
2
1.2+.31
10.8+1.90
3.2+.36
3.2+.21
3
1.5+.27
11.3+.27
10.0+.76
4.8+.07
4
1.7+.16
9.2+.36
8.5+.44
4.6+.29
5
1.3+.49
11.6+.24
10.1+.83
5.1+.36
6
1.0+.22
5.6+.70
3.3+.12
3.3+.16
7
3.0+.96
7.7+1.01
7.0+.41
4.9+.37
Animal
Number
5
6
7
8
Weeks
0
0
0
0
0
1
.6+.03
0
0
0
2
3.6+.46
1.9+.17
3.2+.10
.6+.07
3
11.9+.99
13.4+1.35
11.8+.77
5.0+1.00
4
7.8+1.04
8.0+.61
6.7+.83
3.9+.43
5
9.0+.63
12.7+.62
11.8+2.45
15.2+.18
6
11.0+.71
16.1+.90
5.5+.69
7.6+.45
7
14.1+1.42
12.8+.88
2.8+.25
6.6+.51
Animal
Number
9
10
11
12
Weeks
0
0
0
0
0
1
1.7+.39
0
0
0
2
5.6+.26
0
2.1+.24
7.5+.88
3
18.2+2.73
3.1+.69
8.3+.74
13.2+1.86
4
8.4+.97
5.5+.43
7.8+.79
9.1+1.02
5
10.0+.64
5.2+.47
3.9+.81
26.8+1.30
6
7.1+.31
3.0+.30
2.4+.65
7.0+1.51
7
8.7+.51
12.6+.76
4.8+2.38
15.4+.36
Animal
Number
13
14
15
16
Weeks
0
0
0
0
0
1
0
0
0
0
2
0
1.8+.36
1.0+.11
0
3
0
9.5+.37
3.4+.48
.7+. 12
4
0
8.4+.67
7.5+1.06
3.9+.61
5
.3+.11
12.0+.87
5.1+.56
3.9+.14
6
.8+.23
8.5+.21
.6+.01
3.7+.37
7
1.6+.52
19.3+1.98
5.2+.50
4.5+.66