UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A study of auto-anti-idiotypes to BSA Forsyth, Robert Bruce 1989

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1989_A6_7 F67.pdf [ 2.34MB ]
JSON: 831-1.0097465.json
JSON-LD: 831-1.0097465-ld.json
RDF/XML (Pretty): 831-1.0097465-rdf.xml
RDF/JSON: 831-1.0097465-rdf.json
Turtle: 831-1.0097465-turtle.txt
N-Triples: 831-1.0097465-rdf-ntriples.txt
Original Record: 831-1.0097465-source.json
Full Text

Full Text

A S T U D Y O F A U T O - A N T I - I D I O T Y P E S T O B S A B y R O B E R T B R U C E F O R S Y T H B.Sc, The University of British Columbia, 1985 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F T H E REQUIREMENTS FOR T H E D E G R E E O F M A S T E R O F SCIENCE in T H E F A C U L T Y O F G R A D U A T E STUDIES (Microbiology) We accept this thesis as conforming to the required standard T H E U N I V E R S I T Y O F BRITISH C O L U M B I A March 1989 © Robert Bruce Forsyth, 1989 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) i i Abstract In order to study the idiotypic relationships between the antibody populations produced in different species during normal immune responses to ordinary protein antigens, we raised immune sera in mice and chickens using three protein antigens: Bovine serum albumin ( B S A ) , keyhole limpet hemocyanin ( K L H ) and diphtheria toxoid (DT). A n avidin-biotin E L I S A was used to measure idiotypic binding between antibody populations from these sera. We found that the chicken sera contained auto-anti-idiotypes ( A A I ) against antigen specific antibodies which were present in the same serum and which co-purified with those antibodies on antigen-sepharose columns. These A A I were present in secondary response chicken a n t i - B S A serum at levels comparable to that of the an t i -BSA antibody. The chicken A A I also react specifically with Ids in mouse ant i -BSA serum. Mouse ant i -BSA serum completely inhibited the binding between the chicken Id and A A I . This similarity between the idiotypes of whole populations of antibodies produced in two distantly related species, in the absence of any manipulation with idiotypic or anti-idiotypic reagents, suggests that the A A I detected in this way are internal image antibodies. It indicates there is positive selection for such auto-anti-idiotypes to be internal images. i i i Table of Contents Abstract ii. Table of Contents iii. List of Figures iv. List of Abbreviations v. A c k n o w l e d g e m e n t vii. Introduct ion 1. Materials and Methods 4. Results 9. Discussion 36. B ib l iography 42. i v Lis t of Figures Figure T i t l e Page 1. Specific self binding properties of affinity purified chicken antibodies 11. 2. Purity of the affinity purified chicken anti -BSA 13. 3. Molecular size of the chicken auto-anti-idiotypic antibodies 16. 4. Idiotypic binding is due mainly to interactions between IgG species 19. 5. Specific binding of mouse serum antibodies to affinity purified chicken anti-BSA measured in an E L I S A assay 21. 6. Molecular size of the mouse antibodies which bind to the chicken auto-anti-idiotype 24. 7. The effect of removing anti-BSA antibody from mouse anti-BSA serum on binding to chicken ant i -BSA auto-anti-idiotype 26. 8. Time course of induction of anti-idiotype and anti-BSA antibody in the immune response to B S A 29. 9. Inhibition by antigen of idiotypic binding to affinity purified chicken antibodies 33. 10. Inhibition of binding between chicken idiotype and auto-anti-idiotype using mouse and chicken anti -BSA serum antibodies 35. V L i s t of Abbrevia t ions A b b r e v i a t i o n Meaning . A 405 nm absorbence at 405 nm. A A I auto-anti-idiotype(s). B6 C57BL/6 mouse strain. Bis N,N'-methy lene-bis -acrylamide. B S A bovine serum albumin. C F A complete Freund's adjuvant. D B A D B A / 2 mouse strain. DT diphtheria toxoid. E D T A ethylenediamine-tetraacetic acid. E L I S A enzyme linked immunosorbent assay. h hour(s). Id(s) idiotype(s). IF A , incomplete Freund's adjuvant. I g immunoglobuUn. KD kilodalton. K L H keyhole limpet hemocyanin. L f limit of floculation unit. P A G E polyacrylamide gel electrophoresis. PBS phosphate buffered saline. vi L i s t of Abbrevia t ions (continued) A b b r e v i a t i o n Meaning . S A S saturated ammonium sulphate. SDS sodium dodecyl-sulphate. T N P trinitrophenyl. T r i s tris(hydroxymethyl)aminomethane. w / v weight to volume. vii Acknowledgements I owe a debt of gratitude to you my instructors; Drs. Geoffrey Hoffmann, Julia Levy and Hung-Sia Teh; who have illuminated the doorway before me and introduced me to a world of mysteries. I would especially like to thank Dr Geoffrey Hoffmann for giving me the opportunity to explore. Thank you for your patience and your passionate excitement for simple essence behind intricacies. I hope that my work may honor all of you. Here is my drop, a libation to the sea. "I devoted myself to study and to explore by wisdom all that is done under heaven. What a heavy burden God has laid on men!" Proverbs 1:13 1 Introduction The overall behaviour of the immune system is, in principle, relatively simple. Antigen induces the expression of specific effector functions which remove the antigen. Answering the questions of why and how the immune system manages to produce this desired effect is far more complicated, and so far no widely accepted model of overall immune regulation has been formulated. One popular line of investigation which interests us is the study of idiotypic interactions between specific receptor molecules and between immune cells via their receptor molecules The usual approach to studying idiotypes has been to use anti-idiotypic antibodies that have been produced by direct immunization with Ig bearing the idiotype (Id) of interest. This often involves the use of simple antigens (haptens) to produce monoclonals bearing dominant idiotypes. Though this approach has yielded much information about idiotypic interactions (1,2), it has been suggested that these idiotypic phenomena are artificial and unimportant to overall regulation (3). Since immune responses to complex antigens contain many different Ids, we took a different approach. We immunized two species (mice and chickens) with conventional antigens: B S A , K L H , or D T (diphtheria toxoid) and studied polyclonal auto-anti-idiotypic antibodies in the resulting immune sera. Auto-anti-idiotypes ( A A I ) are so called because they are produced by the same individual as the idiotype. In many systems they are produced as a natural part of the immune response to antigens (4) and there is evidence that they are involved in B cell 2 tolerance. A A I can be detected bound to spleen cells from immunized mice. These antibodies inhibit some of the idiotype positive B-cells from secreting Ig in a plaque assay (5). In humans A A I are produced in normal immune responses (6,7) but they may be involved in the pathology of some immunological disorders such as acquired C l inhibitor deficiency (8). There is evidence that A A I play a role in preventing auto-immune disease. M i c e of strains prone to auto-immunity are deficient in their production of A A I to trinitrophenyl-Ficoll (9). In N Z B strain mice the F i of crosses with normal strains have milder symptoms. These F i mice all produce A A I directed against the auto-antibody (10). Furthermore, in humans A A I directed against the auto-antibody (ant i -DNA) of systemic lupus erythematosus are present at increased levels during remission (11,12) and are normally present in healthy individuals (13). Thus auto-anti-idiotypic antibodies are proving to be very interesting whether they are generally involved in mediating regulation themselves or are by-products of the regulatory process. More needs to be known about their properties and their regulatory roles in the immune system. In view of the symmetry in idiotypic interactions one might imagine that when an immune response occurs, there could be a "battle" between Ids and anti-Ids specific for the antigen. We reasoned that in such a battle it is quite likely that there should be some survivors on "both sides" (14). In order for a clone to survive complementary clones should have been more or less eliminated or suppressed. Ids that are produced in large amounts would 3 effectively eliminate their complementary anti-Ids. Anti- idiotypic clones that managed to overcome any complementary Ids would also survive. In different species we would expect different Ids to be survivors since different species should have different repertoires. However, in both species we would expect that the anti-Ids present would bear internal images of epitopes on the antigen (15) because they were all selected by stimulation to be complementary to Ids of binding sites for those epitopes. It follows that surviving anti-idiotypic antibodies to a particular antigen in chicken antisera might bind to Ids in mouse antisera raised against the same antigen. It was this expectation that we set out to test. In this study we found chicken A A I against antibodies present in the same serum. These chicken antibodies (in animals immunized with B S A ) also bind to Id produced by mice. Antigen inhibits both forms of idiotypic binding and mouse ant i -BSA serum specifically inhibits the binding between chicken Id and A A I . Thus specific A A I from one species recognize the idiotypes from another species, though neither has been immunized with the other's antibodies. We interpret this cross species recognition of Ids as being the result of selective pressures favouring anti-Ids which are internal images, in preference to anti-Ids which merely recognize one or other private Id. 4 Mater ia l s and Methods Animals. Inbred D B A / 2 and C 5 7 B L / 6 strain mice were obtained from Jackson Laboratories (Bar Harbour, M E ) or raised in our breeding colony from stocks obtained from Jackson and used between 8 and 12 weeks of age. Dekalb breed chickens were obtained from the Poultry Divis ion of the Department of Animal Science at the University of British Columbia. Preparation of Immune Sera. Chickens were immunized with B S A (Sigma Chemical Company, St. Louis, M O . , 25 mg per injection) or K L H (Calbiochem, L a Jolla, C A . , 5 mg per injection): week 1 in a 50% (v/v) emulsion with C F A s.c, week 2 in PBS i.v., week 3 in IFA s.c, and week 8 in C F A s.c. Serum was obtained 12 days after the last injection. Another group of chickens were injected s.c. at three week intervals with diphtheria toxoid (DT) (Connaught Labs Toronto, Canada). Injections 1 and 2 were 20 L f and 10 L f respectively in aluminum hydroxide gel (precipitated form Y , see ref 16) containing 20 Lf /ml and 1 mg/ml Aluminum (as Aluminum Chloride). The third injection was 5 L f in PBS. Serum was obtained 11 days after the last injection. In addition small samples of serum were obtained from all of the chickens at different times during the immunization schedule. In all cases chicken blood was clotted 2h at 3 7 ° C in glass tubes and then refrigerated overnight at 4 ° C to allow for clot retract ion. Groups of 5 D B A / 2 and C 5 7 B L / 6 mice were immunized intraperitoneally with B S A (7 mg per injection) or K L H (100 jig per injection). The mice were injected with antigen in 1:1 C F A on days 0 5 and 14, rested for 1 month, injected with antigen in P B S , rested for 3 months, injected with antigen in 1:1 C F A and again 14 days later with antigen in P B S . Serum was obtained 7-10 days after each injection of antigen in saline. The pooled mouse sera obtained after the second soluble boost were used for all experiments unless otherwise stated. In all cases mouse blood was clotted l h at 3 7 ° C and refrigerated overnight at 4 ° C . Affinity purification of Chicken Antibodies. Aff in i ty columns were made using antigen coupled to C N B r activated (17) sepharose C L - 4 B (5 mg B S A , 5 mg K L H or 5000 L f D T per ml of beads). In each case 70-90% of the protein was coupled and residual reactive sites on the beads were blocked with 0.16 M ethanolamine. Immune serum diluted 1 in 3 with PBS and adjusted to 25 m M in E D T A was then passed one or more times over the appropriate antigen column at 4 ° C . The column was then washed with P B S - E D T A followed by PBS and eluted with 0.1 N H C l in 0.15 M N a C l . The column was then washed and neutralized and the depleted serum was passed over the column a second time. The column was washed and eluted as before. The fractions were monitored by absorbance at 280 nm ( D M S 90 spectrophotometer, Varian). The peak fractions of eluted material were neutralized with 1 M Tris Base. The eluted material from both runs was pooled, concentrated by dialysis on a bed of polyethylene glycol, dialysed against P B S , adjusted to p H 5-6 with 1 N H C l and stored at - 2 0 ° C . 6 Adsorption of mouse serum antibodies on antigen columns. Samples (100 of biotinylated serum diluted 1 in 20 were passed over 0.5 ml columns of B S A or K L H sepharose. The samples were washed through over 1 h at 4 ° C with PBS (10 ml) and eluted with 0.1 N H C l in 0.15 M N a C l (8 ml) while collecting 2 ml fractions. The acid fractions were neutralized with 1 M Tris-Base and all fractions were adjusted to 2.5 ml with PBS. Preparation of gamma globulin. Chicken sera were precipitated twice (18) with 45% saturated ammonium sulphate ( S A S ) , resuspending with 0.85% (w/v) N a C l . Mouse sera were precipitated with 50% S A S . The final precipitates were washed with 45% S A S , resuspended and dialysed against PBS. Protein determination. The concentration of protein was estimated where necessary by measuring U . V . absorbance [assuming (18) that 1 A 278 nm = 0.89 mg/ml K L H , 1 A 279 nm = 1.5 mg/ml B S A , and 1 A 280 nm = 0.74 mg/ml fowl y-globulin or 0.70 mg/ml mouse y-globulin (as IgG)] or by using the bicinchoninic acid protein assay at 6 0 ° for 15 minutes (kit B C A - 1 Sigma) Biotinylation of Proteins and Antibodies. Protein solutions in P B S were biotinylated (19) by adding 0.1 volume of biotinylation reagent [5 m M biotinamidocaproyl N-hydroxysuccinimide ester (S igma) in d imethy l - formamide] and incubat ing at room temperature. Serum was diluted 1 in 10 (protein concentration of 10-20 mg/ml) and biotinylated for 4 hours. Antigen and purified antibodies were diluted to 0.2 mg/ml and biotinylated; the reaction was stopped after 2 hours by adding 0.4 volume of 100 m M glycine in 25 m M succinate (pH 5.0). The dimethyl-formamide and unused 7 reagent were then removed by dialysis against PBS . Avidin-biotin ELISA for binding to specific antigen or anti-idiotypes. The E L I S A assay used was a modification of the standard micro E L I S A using the reagents as described by Voller et al (20). For convenience the PBS-Tween was prepared from a 5 fold concentrated stock which contained 0.4 g/1 NaN3 as a preservative. Microtitre plates (Immulon II, Dynatech) were coated with affinity purified antibodies, whole y-globul in (prepared by ammonium sulphate fractionation) or the protein antigens (100 u.1 at 10 (ig/ml unless otherwise stated) in carbonate buffer p H 9.6 for 3 hours at 3 7 ° C . The plates were washed with PBS-Tween, blocked with 5% casein (w/v in PBS-Tween) for 1 hour at 3 7 ° C and washed. The biotinylated primary reagent (diluted in 5% casein) was added and allowed to bind for 2 hours at 3 7 ° C . The plates were washed and avidin-alkaline phosphatase conjugate (Sigma, 1/400 dilution of a 0.5 mg/ml stock) in 5% casein was added for 1 hour at 3 7 ° C . The plates were washed and substrate ( S i g m a 104: 1 m g / m l p-nitrophenylphosphate in 10% diethanolamine p H 9.8) was added. The plates were incubated at 3 7 ° C until sufficient colour development had occurred. Absorbances at 405 nm (A 405 nm) were measured using an E L I S A plate reader (Titertec Multiskan). Inhibition of binding in ELISA. C o m p e t i t i v e inh ib i tors (antigen or serum antibodies) diluted in 5% casein were mixed with an equal volume of the primary reagent (biotinylated antibody or antigen) at twice the desired final concentration. The final concentrations of the biotinylated reagents were: B S A at 50 ng/ml, affinity purified chicken ant i -BSA at 200 ng/ml, chicken ant i -BSA 8 serum at a 1/4,000 dilution and mouse anti -BSA serum at 1/30,000. The mixtures were then used in the E L I S A assay. The inhibition was calculated from the A 405 nm values according to the equation: ( \ A j - M e a n b a c k g r o u n d % Inhibition= 1-A n - M e a n b a c k g r o u n d V ) 100 A j is the A 405 nm for samples with inhibitor and A Q is the A 405 nm for the samples without inhibitor. The background is the A 405 nm with uncoated wells (blocked with 5% casein). Non reducing polyacrylamide gel electrophoresis: separation, preparative fractionation of antibodies. Samples were heated to 6 0 ° C for 10 minutes in non-reducing sample buffer (0.0625 M Tris H C l (pH 6.8), 2% SDS and 10% glycerol) and subjected to P A G E (21,22) on linear gradients (23) of 3% to 10% or 3% to 20% T acrylamide (4% C) with stacking gels containing 3% T (5% C) acrylamide [(24) % T is the total % (w/v) of monomer (acrylamide and Bis) and % C is the % of the total monomer contributed by the crosslinking monomer (Bis)]. The gels were then cut into pieces and some sections were stained with Coomassie blue. Unstained sections containing the chicken anti-B S A antibodies to be isolated were sliced into horizontal strips which were ind iv idual ly chopped into 5 mm squares, placed in polypropylene tubes with about 1 gel volume of 0.05 M ammonium bicarbonate and rolled end over end for 3 days at 4 ° C . The buffer was changed and the gel slices were rolled for an additional 3 days with fresh buffer. The two buffer samples from each slice were pooled and this eluted material was stored at - 2 0 ° C . 9 R e s u l t s Auto-anti-idiotypic antibodies in affinity purified chicken antibody populations. A f f i n i t y p u r i f i e d c h i c k e n antibody populations specific for B S A , K L H or D T were biotinylated and an avidin-biotin E L I S A was used to measure their binding to plates coated with each of the same affinity purified antibody preparations in non-biotinylated form. The antibodies in each affinity purified preparation react preferentially with antibodies within the same preparation. That is, chicken ant i -BSA reacts specifically with chicken ant i -BSA, a n t i - K L H reacts with a n t i - K L H and anti-DT reacts with ant i -DT (Fig. 1). Antibodies in each preparation react only weakly with antibodies specific for other antigens. This specificity indicates that the binding is mediated by the V-region. That is, the purified preparations contain auto-anti-ids (AAI) against the antigen specific antibodies. Since the chicken anti-BSA preparation gave the strongest effect, it was used to further characterize the phenomenon. Characterization of the antibodies involved in binding of chicken anti-BSA to chicken anti-BSA. We did experiments to show that the specific binding of chicken anti-BSA to chicken anti-BSA is not due to residual antigen in the purified antibodies. Chicken anti-B S A was separated by non-reducing SDS P A G E . The major bands in the chicken anti-BSA migrate at the same positions as a purified yolk IgG standard (Fig . 2). Al though the gel was intentionally overloaded no albumin band nor lower molecular weight (degraded albumin) bands could be detected in the chicken ant i -BSA, either 1 0 Fig. 1. Specific self binding properties of affinity purified chicken antibodies. The results represent the binding in an E L I S A of biotin labeled, purified chicken antibodies ( • , a n t i : K L H . • , anti-BSA. • , anti-DT) to plates coated with purified chicken a n t i - K L H (a), anti-B S A (b) and anti-DT (c). The error bar is the standard deviation of duplicates. In panel (b) the two control curves (those for a n t i - K L H and anti-DT) overlap, and in some cases the error bar is smaller than the symbol. Biotinylated antibody (jig/mL) 1 2 Fig. 2. Purity of the affinity purified chicken anti-BSA. Samples of proteins were separated by non reducing SDS polyacrylamide gel electrophoresis on a linear 3 to 20% gradient gel and stained with Coomassie Blue. Lane 3 is purified chicken anti-BSA, approximately 30 ng and lane 4 is B S A , approximately 0.3 (ig. Lane 2 is chicken egg yolk IgG (25) as a standard. Lane 1 is molecular weight standards (Sigma SDS-6H: myosin (band not visible) P-galactosidase, phosphorylase b, bovine serum albumin, ovalbumin and carbonic anhydrase) with molecular weights as indicated. Molecular weight (KD) ro cn co co oi cn o> 1 4 visually or by scanning densitometer. A B S A standard equivalent toas little as 1% B S A contamination is clearly visible (Fig. 2, 3a). The minimum B S A contamination that would have been needed to account for the E L I S A signal was estimated from mixing experiments to be about 5%. Varying amounts of B S A were mixed with irrelevant chicken antibody (ant i -KLH) so as to maintain a constant protein concentration and the mixtures were used to coat E L I S A plates. The binding of chicken anti B S A to these plates increases linearly up to about 10% B S A . Plates coated at mixtures containing 5% of the protein as B S A bind the same amount of chicken anti-BSA as plates coated with purified chicken anti-BSA. The assay is even less sensitive to the effect of B S A contamination of the biotinylated reagent. The binding to plates coated with chicken ant i -BSA of varying quantities of biotinylated B S A or biotinylated chicken anti-B S A was compared. In the approximately linear portion of the binding curves 0.06 ng / m l of biotinylated B S A produces a signal equal to that of about 0.4 | ig /ml of biotinylated chicken ant i -BSA. Thus the biotinylated chicken anti-BSA would have to contain 15 % contamination with B S A to account for the binding observed. Fractionation with preparative gel electrophoresis was used to demonstrate that the specific binding is due to molecules the size of chicken IgG. The affinity purified chicken anti-BSA was eluted with 0 . 0 5 M ammonium bicarbonate from transverse slices of the unstained portion of the non-reducing SDS P A G E gel. Each eluted fraction was used to coat wells of an E L I S A plate to yield a profile of the gel from top to bottom. Biotinylated chicken anti-BSA bound in the E L I S A (Fig. 3b) mainly to material from gel slices corresponding 15 Fig. 3. Molecular size of the chicken auto-anti-idiotypic antibodies. Affinity purified chicken anti-BSA was separated by non reducing SDS polyacrylamide gel electrophoresis and eluted from gel slices as described in the Materials and Methods section. Separate sets of wells on E L I S A plates were coated with the material from each of these gel fractions diluted 1/5 with carbonate buffer. The top panel (a) shows densitometer scans of lanes 3 (chicken anti-B S A ) and 4 (BSA) of the gel (which is shown in Fig. 2). The bottom panels are E L I S A binding of biotinylated affinity purified chicken anti-BSA (b) and mouse anti-(chicken y globulin) (c) to each gel fraction. The arrows indicate the approximate positions of the pre-stained molecular weight markers (84 K D , fructose-6-phosphate kinase Sigma F-0387; 27 K D , triosphosphate isomerase Sigma T -9400). The asterisks in panel (c) indicate the fractions which were biotinylated and assayed (Fig.4) to determine which biotinylated Ig species bind to the chicken IgG on the E L I S A plate. Migration mm 1 7 to the position (Fig. 3a - lane 3) of the main protein band. There was no binding to any material from the slices corresponding to the position of the B S A band (Fig. 3a - lane 4). There was, however, some binding to higher molecular weight material which may be IgM or dimeric IgA (23) since mouse anti-serum against whole chicken y-globulin binds to some higher molecular weight material in that region of the profile (Fig. 3c). Material from the gel fractions corresponding to the peaks marked with asterisks in F i g . 3c was biotinylated and binding to chicken anti -BSA was measured (Fig. 4). These fractions represent the higher molecular weight (IgM or IgA) material and the main IgG peak. Binding of material in the peak corresponding to IgG accounts for the vast majority of the signal. Thus the binding between members of the affinity purified chicken ant i -BSA population is mainly due to binding among members of the IgG populations rather than binding between IgG and some minor component such as IgM or IgA. Idiotypic complementarity between antibodies from mouse and chicken immune responses to BSA. Biotinylated mouse ant i -BSA sera bind in the E L I S A to affinity purified chicken ant i -BSA antibodies but not to chicken a n t i - K L H antibodies (Fig 5d-i). Furthermore neither normal mouse serum nor hyperimmune mouse a n t i - K L H sera bind to chicken anti-BSA even though the latter have high titers of a n t i - K L H antibody (Fig. 5a- c). 1 8 Fig. 4. Idiotypic binding is due mainly to interactions between IgG species. Samples (200 |il) of the Ig containing gel fraction extracts marked with asterisks in Fig . 3c were dialyzed against PBS and biotinylated by adding 20 of biotinylation reagent, incubated 2 hours and dialyzed against PBS . The biotinylated samples were adjusted to 300 jil and used at 1/50. Binding to plates coated with purified chicken anti-BSA was measured. The error bar is the standard deviation of duplicates. Slices 8 and 9 give the strongest signal and are IgG (compare Fig. 3c). Relative binding (A 405 nm) 2 0 Fig. 5. Specific binding of mouse serum antibodies to affinity purified chicken anti-BSA measured in an E L I S A assay. The results show the binding of various biotin labeled normal or immune mouse sera to plates coated with affinity purified chicken anti -BSA(#) , affinity purified chicken a n t i - K L H (O), B S A (•) , and K L H (•) . The sera used in the various panels were obtained from: unimmunized mice (a); groups of mice after the second boost with soluble antigen (b, c, f, and i); two individual mice after the first boost with soluble antigen (d and e); a group of mice seven days and ten days respectively after the first boost with soluble antigen (g and h). The data points are means of duplicates. B i o t i n y l a t e d m o u s e s e r u m ( log of d i l u t i on f ac to r ) 2 2 In pooled sera of both D B A / 2 and B6 mice and in the sera of individual mice (Fig 5d,e) a large proportion of the mouse antibody is able to bind the chicken anti-Id. In all except one (Fig. 5f) of the mouse anti -BSA sera tested the titration curves for binding to excess B S A and excess chicken anti-BSA are separated by about a two fold dilution or less. Thus, assuming that equal signals are produced by equal amounts of bound antibody, the sera contain idiotype binding antibody at approximately half or more of the level of ant i -BSA ant ibody. T o rule out antigen as the explanation for this specific binding as well, biotinylated mouse ant i -BSA serum was fractionated using preparative gel electrophoresis (Fig 6). A l l of the material which binds to chicken ant i -BSA antibodies (top panels) migrates at the same position as the antibodies which bind to B S A (centre panels) and the main peak of the immunoglobulins as detected by binding to rabbit anti-mouse Ig antibodies (bottom panels). This peak is well resolved from the position of the B S A band (66 K D ) where no binding was observed. Biotinylated mouse ant i -BSA sera were also fractionated into antigen binding and non-binding fractions by adsorbing with B S A sepharose columns. More than 90% of the mouse antibodies which bind to chicken ant i -BSA are removed by adsorbtion with B S A sepharose but not K L H sepharose (Fig 7). Antibodies which bind to chicken anti -BSA were recovered in the B S A binding fraction eluted from the B S A column. The simplest explanation is that mouse antibodies against B S A bind to the chicken A A I contained in the purified chicken anti -BSA antibodies. 23 Fig. 6. Molecular size of the mouse antibodies which bind to the chicken auto-anti-idiotype. The protein components of whole biotinylated mouse antisera (B6 anti-BSA on the left, D B A anti-BSA in the centre and D B A a n t i - K L H on the right) were separated by non-reducing SDS polyacrylamide gel electrophoresis and eluted from gel slices as described in the Materials and Methods section. Each of these fractions was assayed in the E L I S A for binding to a plate coated with purified chicken anti-BSA (O), purified chicken anti-K L H (• ) , B S A (• ) , and the y globulin fraction of rabbit antisera against mouse Ig ( • ) . The arrows indicate the approximate positions in the stained portion of the gel of the molecular weight markers (97 K D , phosphorylase b Sigma S D S - 6 H ; 66 K D , bovine serum albumin and 14 K D , a-lactalbumin Sigma SDS-7) . B6 anti B S A o DBA anti BSA j DBA anti KLH I I I I l I I I I I . t t t : ha a a H * * * 1 1 a i | i i i i | t t t t 1 • i i • i i i i i i t t t Q Q Q / \ t~- to o- " j 1 <J> to T -1 |v, IO <tf O) <o i -to *t o> to T -| | 1 H i JD JL' I I I I I I I I I I I I I I I I I I I I I I I I I I I 0 1 0 20 0 10 20 0 10 20 Mig ra t i o n in P A G E (gel s l i c e numbe r ) 25 Fig. 7. The effect of removing anti-BSA antibody from mouse anti -BSA serum on binding to chicken anti-BSA auto-anti-idiotype. Samples of biotinylated D B A (panel a) and B6 (panel b) mouse anti-B S A sera were passed over small columns of K L H or B S A sepharose. The data shown represents the binding in the E L I S A of the antibodies in the first fraction (~2 ml) of the runthrough and of the eluate. The error bar is the standard deviation of duplicates and is negligible in some cases. Practically all of the anti-idiotypic activity is adsorbed to the B S A column. 26 E c m o O) c •5 c 0> rr 0.6-0.5-0.4-0.3-0.2-0.1 -0.0-0.6-0.5-0.4-0.3-0.2-0.1 -0.0 1 | | | 0 KLH column r Run-through. Eluate KLH column BSA column Run-through Eluate Column fraction 27 The time course of the level of anti-BSA and AAI in chickens after immunization with BSA. C h i c k e n y-globulin fractions were prepared by ammonium sulphate precipitation of small samples of serum obtained throughout the immunization period. The y -g lobul in samples were used to coat E L I S A plates and the binding of biotinylated B S A , mouse a n t i - B S A and chicken a n t i - B S A was measured (Fig. 8). The conditions were adjusted so that each biotinylated reagent was in excess and the signal varied linearly with the proportion of specific antibody in the y-globulin used to coat the plate. Some B S A binding but very little Id binding antibody is present by day 7 in the primary response after the first immunization. The main peak of both anti-BSA and anti-Id activity is produced during the secondary response. A n estimate was made for this assay of the ratio of molar signals for binding of biotinylated antibody as compared to B S A . For the biotinylated proteins coated directly onto E L I S A plates biotinylated chicken ant i -BSA antibody produces 1.3 ± 0.2 (mean ± SE) times the signal obtained with an equal molar quantity of biotinylated B S A . Since the peak signals (Fig. 8a and c) are roughly the same for binding of biotinylated B S A and for binding of biotinylated affinity purified chicken ant i -BSA, this ratio indicates that the concentration of idiotype binding antibodies is about 50-100% of that of the anti-BSA antibodies at this point in the time course. 28 Fig. 8. Time course of induction of anti-idiotype and anti-B S A antibody in the immune response to B S A . Gamma globulin was prepared from small samples of chicken serum taken at various times through the course of immunization with K L H ( • ) or B S A (• ) and used to coat separate wells of E L I S A plates. The data represent the binding of excess biotinylated B S A (a), mouse ant i -BSA serum (b), and affinity purified chicken anti -BSA (c) to antibodies in the gamma globulin fraction at each time during the immune response. The error bar (in most cases smaller than the symbol) is the standard deviation of duplicates. For B S A the values were obtained using sera from two chickens. Antigen injections were given at the times indicated by the arrows (in complete ( C F A ) or incomplete (IFA) Freund's adjuvant or as a so lut ion (Sol ) in sa l ine) . Relative binding (A 405 nm units per hour) to vo 3 0 Antigen (BSA) as an inhibitor of idiotypic binding to chicken anti-BSA. B S A was included as an inhibitor at various concentrations with constant amounts of various biotinylated reagents binding in the E L I S A to plates coated with affinity purified chicken antibodies. The binding of both biotinylated mouse anti-B S A serum and affinity purified chicken anti-BSA to affinity purified chicken ant i -BSA is completely inhibited by high concentrations of B S A (Fig. 9). This inhibition is specific since K L H does not block the binding of any of the reagents. Higher concentrations of B S A are needed to block the binding of the various anti-BSA antibodies than to block the binding of biotinylated B S A to the purified chicken anti-B S A . This may reflect a greater intrinsic affinity of the Ids and anti-Ids for each other than for B S A or a greater effective affinity due to the multivalence of antibodies. Specific antiserum as an inhibitor of idiotypic binding to chicken anti-BSA. Specific antisera were tested as inhibitors in the same assay as for inhibition by B S A . The binding of biotinylated affinity purified chicken ant i -BSA to affinity purified chicken anti-B S A can be completely inhibited by high concentrations of the chicken ant i -BSA serum from either the same chicken or another immune chicken (Fig. 10a). Normal chicken serum and immune serum against another antigen ( K L H ) give much less inhibition. More importantly, binding of Ids and anti-Ids in the affinity purified chicken ant i -BSA can be completely blocked by mouse ant i -BSA serum or y-globulin (Fig. 10b and c). This indicates that a large proportion of the chicken anti -BSA antibodies involved in auto-anti-3 1 idiotypic binding have Ids similar to the mouse anti-BSA antibodies. This inhibition by mouse serum is not due to mouse serum albumin since non immune and irrelevantly immune mouse sera are more than 10 fold less effective as inhibitors. Furthermore, y-globulin from mouse anti-BSA serum gives 50% inhibition at 10-20 | i g / m l whereas about 1000 u,g/ml of B S A is required for 50% inhibition (Fig. 9). 32 Fig. 9. Inhibition by antigen of idiotypic binding to affinity purified chicken antibodies. B S A (solid symbols) was added at various concentrations to block binding of biotinylated reagents ( • : B S A , • : chicken anti-BSA serum, A : mouse anti-BSA serum and • : affinity purified chicken anti-BSA) to affinity purified chicken anti-B S A on E L I S A plates. As a control, K L H (hollow symbols) was added to the same biotinylated reagents ( L 7 J , 0 , A , 0 ) . The error bar is the standard deviation of four replicates and is plotted where it is larger than 2.5% inhibition. 33 3 4 Fig . 10. Inhibition of binding between chicken idiotype and auto-anti-idiotype using mouse and chicken ant i -BSA serum antibodies. Normal ( • ) , a n t i - K L H (O) or anti-BSA ( • ) antibodies (as chicken sera in panel a, as mouse sera in panel b and as y g lobul in from mouse sera in panel c) were added at various dilutions to biotinylated affinity purified chicken ant i -BSA antibodies in an E L I S A with chicken a n t i - B S A on the plate. The error bar is the standard deviation of four replicates and is plotted where it is larger than 2.5% inhibition. In panel a the two curves [both marked ( • ) ] for blocking by chicken ant i -BSA serum are for two sera from individual chickens, one of which is the source of the purified chicken ant i -BSA. Gamma globulin (log Lig/mL) 36 Discussion We find that Id and A A I are present simultaneously in the sera of chickens immunized with conventional protein antigens. In the secondary response to one antigen ( B S A ) the A A I appears to be present at a concentration of the order of 50-100% of the concentration of the antibodies against the antigen. This indicates that it may be common for A A I to coexist at high levels with the corresponding Ids in immune responses to protein antigens. Furthermore, binding and inhibition experiments show that the chicken A A I also recognize antibodies produced in mice immunized with the same antigen. Nearly all of the chicken A A I appears to bear internal images of the antigen. This confirms for a normal immune response the validity of the conclusion (15) that antibodies which are complementary to an antigen bear "negative images" of epitopes on that antigen. Furthermore, it indicates that the anti-Ids were selected to recognize these negative images (that is, to be internal images). The presence of A A I in chicken antisera is not very surpr i s ing since their production during other normal immune responses is well documented (4). If the anti-Ids are induced by Id they must both be present together at some time during the response. However, the presence of large quantities of these anti-Ids in chicken serum and their ability to copurify with the antigen specific antibodies is curious. They could perhaps be present in sera together with antigen specific antibodies as soluble immune complexes. Such 37 complexes could bind to an antigen column via the antigen specific antibodies and both components would elute from the column together. Others have reported that Id and A A I can coexist and participate in soluble immune complex formation (26), albeit as part of a pathological condition ( selective IgA deficiency). A second possible explanation for the presence of anti-idiotypic antibodies in the column purified material is that these antibodies may have dual specificity; they may bind both to the antigen and to Ids on antigen specific antibodies. Clones that are both antigen-specific and anti-idiotypic could be expected to have a selective advantage in the immune response. They would receive stimulation from both antigen and antigen-specific idiotypes, and they could continue to be stimulated by idiotypes even after the antigen is e l iminated. The idea of these being dual specific antibodies may not be too far-fetched, since multispecific reactivity between antibodies is common (27,28). Furthermore, Bona et al. (29) have shown that certain anti-idiotypic antibodies raised in A / J mice against the V regions of I g M K m G l (a human monoclonal rheumatoid factor specific for IgG) also bind to IgG Fc fragments. They gave the name "epibody" to any antibody which binds both to an antigen and to the V regions of other antibodies against the same antigen. Related observations have been made by Kang and Kohler (30) who reported a monoclonal which binds to itself. They raised hybridoma antibodies against T15 (a myeloma which bears the dominant Id of the Balb/c anti-phosphorylcholine (PC) response). One hybridoma was found to react with both P C and T15. This monoclonal binds to 3 8 its own V-regions and so they called it an "autobody". Subsequently a V H region peptide involved in the binding has been identified ( 3 1 ) . Consistent with a dual specificity interpretation of our results is the fact that, to within the time resolution of the study, B S A binding and idiotype binding activity occur in the chicken immune response with identical kinetics (Fig. 8 ) . If the binding between antibodies in these purified chicken antibody preparations is indeed due to the presence of "epibodies" or "autobodies," this observation could begin to address the important question posed by Kang et al. ( 3 1 ) as to whether autobodies are biological relevant . From this perspective it would be worthwhile to explore more closely the Id and A A I clones in the immune response to B S A . Panels of anti-BSA monoclonal antibodies could be raised and screened for binding to one another and could be used to screen for the A A I producing clones. If the population of anti-BSA antibodies is in fact selected with time to be self-recognizing, this population may become largely self-regulating. This would then be a relatively simple system of idiotypic regulation, that could be more readily amenable to quantitative analysis than models in which two or more levels of idiotypes are involved. In contrast to our results it has been reported that A A I produced in human responses to tetanus toxoid do not bind to idiotypes present at that same point in the immune response (6). This is probably because in that report the assay required that the serum be carefully absorbed to remove any antibodies against the antigen. The absorption would also remove any A A I of the kind described here and could seriously affect the results. Others have reported simultaneous presence of Id and A A I in a rabbit immunized with bacterial polysaccharide (32) and in patients with systemic lupus erythematosus (11,12) selective IgA deficiency (26) and acquired C - l inhibitor deficiency (8). We see idiotypic binding of antibodies in chicken serum to the affinity purified chicken antibodies, and we also see mouse anti-idiotypes binding specifically to the chicken idiotypes. The specific interaction between chicken antibodies can be understood in terms of direct stimulation of auto-anti-idiotypic antibodies (4) by Ids present during the immune response. Mouse serum could also contain A A I against mouse Ids. However, this alone would not explain the ability of immune mouse serum to bind to these chicken antibodies and to inhibit the binding of chicken Id to A A I . Any anti-Ids in the mouse serum were not stimulated by the chicken Ids which they recognize and vice versa. The simplest explanation is that a selection process leads to the production in chickens of A A I , the majority of which bear internal images of antigen epitopes that are recognized by mouse antibodies against those epitopes. This is remarkable in that most anti-Ids described in the literature (which are mostly not auto-anti-ids) are not internal images; see the discussion by Pontillon et al. (33). However, the auto-anti-ids formed in human responses to house dust mite allergens (7) are also internal image anti-Ids. This suggests that auto-anti-idiotypes in general may have a tendency to bear internal images. The sharing of idiotypic specificity by antibodies from two different species is in contrast to the behaviour of Ids detected by 4 0 anti-Ids raised against purified antibodies. In their classic pioneering studies, Oudin and Miche l (34,35) produced anti-idiotypic antibodies by immuniz ing rabbits with bacterial cells agglutinated with antiserum from individual rabbits. The resulting anti-idiotypic sera precipitated specific antibodies from the original serum but not from the sera of other rabbits immunized with the same antigen. Thus the idiotypic specificities detected by Oudin's anti-idiotypic reagents are not shared even between members of the same species. In general immunization with an antibody leads to production of a range of anti-idiotypic antibodies. Some of these are against Ids which are shared with other (genetically distinct) individuals. A fraction of those which recognize shared Ids do so because they are internal images of epitopes on the antigen and as such could be recognized by any antibody against those epitopes (15,36). Some can even mimic some of the effects of biologically active antigens such as hormones (37). In chickens immunized with B S A almost all (rather than only a few) of the A A I are internal image. This observation suggests that the A A I selected in the course of an immune response to antigen are qualitatively different anti-Ids from those selected by active immunization with Id. One plausible speculation as to why A A I might tend to be internal image is suggested by the observation that anti-idiotypic T -cell clones can be activated by antibody (38). In vitro cultured human anti-idiotypic T cell lines have been produced which are activated by immune serum Ig when presented in the context of self class II ( H L A - D Q ) . If we speculate that B-cells which present antigen to T-cells get help (lymphokines for example) from those T -41 cells then any anti-idiotypic B-ce l l clone could get help by specifically picking up and presenting antibody molecules to anti-idiotypic T-cells in the same way that antigen specific B-cells pick up antigen at very low concentrations, accumulate it and present it to antigen specific T-cells (39). In this situation B cells producing an internal image would get more stimulation since they would recognize a broad range of idiotypes and thus be able to interact with more anti-idiotypic T-cells. Conversely B cells against antigen epitopes could possibly be stimulated by picking up internal image antibody and presenting it to T-cells. Thus the antigen recognizing clones and the internal image clones could serve to stabilize each other. 4 2 Bibliography 1. Rajewsky, K . , and T . Takemori. 1983. Genetics, expression and function of idiotypes. Ann. Rev. Immunol. 1:569. 2. Slaoui, M . , G . Urbain-Vansanten, C . Demeur, O . Leo, J. Marvel, M . Moser, J . Tassignon, M . I. Greene, and J . Urbain. 1986. Idiotypic games within the immune network. Immunol. Rev. 90:73. 3. Cohn, M . 1986. Idiotype network views of immune regulation: For whom the bell tolls, in "Idiotypes" M . Reichlin, and J. D . Capra, Eds. Academic Press, London and New York p. 321. 4. Thorbecke, G . J . and G . W . Siskind. 1984. Auto-anti-idiotype production during the response to antigen. In "The Biology of Idiotypes", M . I. Greene and A . Nisonoff, eds. Plenum Press, New York and London, p.417. 5. Bhoghal, B.S. , Y . D . Karkhanis, M . K . Bel l , P. Sanchez, B . Zemcik, G . W Siskind, and G.J . Thorbecke. 1986. Production of auto-anti-idiotypic antibody during the normal immune response. XII. E n h a n c e d auto-ant i - idiotypic antibody product ion as a mechanism for apparent B - c e l l tolerance in rabbits after feeding antigen. Cellular Immunol. 101:93. 6. Geha, R. S. 1982. Presence of auto-anti-idiotypic antibody during the normal immune response to tetanus toxoid antigen. J . Immunol. 129:139. 7. Saint-Remy, J . - M . R. , S J . Lebecque, P . M . Lebrun and M . G . Jacquemin. 1988. Human immune response to allergens of house dust mite, Dermatophagoides pteronyssinus. I V . Occurrence of natural autologous anti-idiotypic antibodies. Eur. J . Immunol. 18:83. 8. Geha, R.S. , I. Quinti, K . F . Austen, M . Cicardi, A . Sheffer and F.S. Rosen. 1985. Aquired Cl - inhib i tor deficiency associated with antiidiotypic antibody to monoclonal immunoglobulins. New England J. Med. 312:534. 9. Cohen, P . L . and R . A . Eisenberg. 1982. Anti-idiotypic antibodies to the Coombs antibody in N Z B F i mice. J. Exp. Med. 156:173. 10. Goidl , E . A . , R. A . Good, G . W . Siskind, M . E . Weksler, and G . Fernandes. 1986. Studies of immune responses in mice prone to autoimmune disorders. II. Decreased down-regulation by auto-anti-idiotype antibody in autoimmune-prone mice. Cellular Immunol. 101:281. 11. Abdou, N.I., H . Wal l , H . B . Lindsley, J .F . Halsey and T . Suzuki. 1981. Network theory in autoimmunity. In vitro suppression of the serum ant i -DNA antibody binding to D N A by antiidiotypic antibody in systemic lupus erythematosus. J . c l in . Invest. 67:1297. 12. Z o u a l i , M . and A . E y q u e m . 1983. Idiotypic-anti idiotypic interactions in systemic lupus erythematosus: Demonstration of oscil lary levels of a n t i - D N A autoantibodies and reciprocal antiidiotypic activity in a single patient. A n n . Immunol. Inst. Pasteur. 134 C:377. 4 4 13. Zoual i , M . and A . Eyquem. 1983. Expression of anti-idiotypic clones against auto-anti-DNA antibodies in normal individuals. Cellular Immunol. 76:137. 14. Hoffmann, G . W . , T . A . K i o n , R . B . Forsyth, K . G . Soga, and A . C o o p e r - W i l l i s . 1 9 8 8 . T h e N - d i m e n s i o n a l network. In "Theoretical Immunology", A . Perelson, ed., Addison Wesley Co. , p. 291. 15. Jerne, N . K . , J . Roland, and P . - A . Cazenave. 1982. Recurrent idiotypes and internal images. E M B O J. 1:243. 16. Yamamoto, A . , S. Kondo, S. Kameyama, and R. Murata. 1978. Studies of adjuvants for human prophylactics. I. Comparison of efficiencies of different adjuvants at various stages of immunization with tetanus and diphtheria toxoids. Japan. J . Med. Sci. Biol . 31:263. 17. Cuatrecasas, Pedro. 1970. Protein Purification by Affinity Chromatography. J . Biol . Chem. 245:3059. 18. Hudson, L . and F . C . Hay. 1980. "Practical Immunology", Blackwell Scientific Publications, Oxford and Boston. 19. Bayer, E . A . , and M . Wilchek.1980. The use of the avidin-biotin complex as a tool in molecular biology. In "Methods of Biochemical Analysis", vol. 26. D. Glick, E d . John Wiley and Sons, New York, p. 1. 20. Vol ler , A . , D . Bidwel l , and A . Bartlett. 1980. Enzyme-l inked immunosorbent Assay. In "Manual of Cl in ica l Immunology", 2nd E d . , N . R. Rose and H . Friedmann Eds. American Society For Microbiology, Washington, D C , p. 359. 45 21. Laemmli , U . K . 1970. Cleavage of structural proteins during the assembly if the head of Bacteriophage T4. Nature 227:680. 22. O'Farrel l , P. H . 1975. High resolution two-dimensional electrophoresis of proteins. J. Bio l . Chem. 250:4007. 23. Schranner, I., and U . Losch. 1986. Immunological identification of avian monomeric and polymeric immunoglobulin M and immunoglobulin A after fractionation on sodium dodecylsulfate pore gradient polyacrylamide gels. Poultry Science 65:360. 24. Hjerten, S. 1962. Molecular sieve chromatography on polyacrylamide gels, prepared according to a s impli f ied method. Arch. Biochem. Biophys., Suppl. 1:147. 25. Jensenius, J. C , I. Andersen, J. Hau, M . Crone, and C . Koch. 1981. Eggs : conveniently packaged antibodies. Methods for purification of yolk IgG. J . Immunol. Methods 46:63. 26. Cunningham-Rundles, C . 1982. Naturally occurring autologous anti-idiotypic antibodies. Participation in immune complex formation in selective IgA deficiency. J . Exp. Med. 155:711. 27. Holmberg, D . , S. Forsgren, F . Ivars, and A . Coutinho. 1984. Reactionsamong IgM antibodies derived from normal, neonatal mice. Eur. J . Immunol. 14:435. 28. Araujo, P. M . F . , D . Holmberg, C . Martinez-A., and A . Coutinho. 1987. Idiotypic multireactivity of 'natural' antibodies: 'Natural' anti-idiotypes also inhibit helper cells with cross-reactive clonotypes. Scand. J . Immunol. 25:497. 4 6 29. Bona, C . A . , S. Finley, S. Waters, and H . G . Kunkel. 1982. Anti-immunoglobulin antibodies. III. Properties of sequential anti-idiotypic antibodies to heterologous anti-y globulins. Detection of reactivity of anti-idiotype antibodies with epitopes of Fc fragments (homobodies) and with epitopes and idiotopes (epibodies) J . exp. Med. 156:986. 30. K a n g , C . - Y . and H . Kohler . 1986. Immunoglobulin with complimentary paratope and idiotope. J. Exp. Med. 163:787 31. Kang, C . - Y . , T . K . Brunck, T . Kieber-Emmons, J . E . Blalock, and H . Kohler. 1988. Inhibition of self-binding antibodies (autobodies) by a Vn-derived peptide. Science 240:1034. 32. Brown, J . C . and L . S. Rodkey. 1979. Autoregulation of an antibody response via network-induced auto-anti-idiotype. J . Exp. Med. 150:67. 33. Pontillon, Francoise, P. Moullier, P. M . Rorico and P. J . Verroust. 1988. Angiotens in II (Al l ) -re lated idiotypic network. I. Common occurrence of A l l internal image on rabbit anti-idiotypic antibodies. J . Immunol. 140:3052. 34. Oudin, J . , and M . Michel. 1969. Idiotypy of rabbit antibodies. I. Comparison of the idiotypy of antibodies against salmonella typhi with that of antibodies against other bacteria in the same rabbits, or of antibodies against Salmonella typhi in various rabbits. J . exp. Med. 130:595. 35. Oudin, J . , and M . Michel . 1969. Idiotypy of rabbit antibodies. II. Comparison of the idiotypy of various kinds of antibodies formed in the same rabbits against Salmonella typhi. J. Exp. M e d . 130:619. 36. Kieber-Emmons, T . , and H . Kohler. 1986. Towards a unified theory of i m m u n o g l o b u l i n s tructure-funct ion relat ions. Immunol. Rev. 90:29. 37. Linthicum, D . S., N . R. Farid. 1988. "Anti-idiotypes, receptors and molecular mimicry". Springer-Verlag, New York. 38. UytdeHaag, F . , I. Claassen, H . Bunschoten, H . Loggen, T . Ottenhoff, V . Teeuwsen, and A . Osterhaus. 1987. Human anti-idiotypic T lymphocyte clones are activated by autologous anti-rabies virus antibodies presented in association with H L A - D Q molecules. J . M o l . Cell . Immunol. 3:145. 39 . Lanzavecchia, A . 1985. Antigen-specific interaction between T and B cells. Nature 314:537. 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items