Search

The conformation of a peptide defines its antigenic specificity. In most cases, a free peptide is in a random form, whereas the same peptide within a protein has a defined conformation. Immunological cross-reactions between the two are rare. Nevertheless, there are cases where an anticonformational antibody may transconform (change the conformation of) the free peptide, allowing the reaction to occur. Based on such considerations, synthetic vaccines against infectious diseases may be constructed. The same is true for autoimmune diseases, where, at least in one case--that of the exacerbating-remitting type of multiple sclerosis--we have developed a polypeptidic drug-vaccine, copolymer 1 (denoted Copaxone), capable of suppressing the autoimmune phenomena in a specific way. This drug-vaccine has been approved for treatment by the US Food and Drug Administration (FDA).

Several inbred mouse strains were screened for their ability to respond to the ordered periodic collagen-like polymer (Pro-Gly-Pro)(n), to the random copolymer (Pro(66), Gly(34))(n), to the protein conjugate Pro-Gly-Pro-ovalbumin, to rat tail tendon collagen, rat tail tendon gelatin, and to Ascaris cuticle collagen. Differences were obtained in the magnitude of the antibody titers towards the above immunogens among the strains tested. The level of the response to the ordered polymer (Pro-Gly-Pro)(n) was not similar to that towards the random (Pro(66), Gly(34))(n), confirming differences in the antigenic determinants of the two immunogens. The role of the thymus in the immune response to (Pro-Gly-Pro)(n) and (Pro(66), Gly(34))(n) as well as to two collagens and gelatin, was studied in order to find out a possible correlation with the structural features of the immunogens. Heavily irradiated recipients were injected with syngeneic thymocytes, marrow cells, or a mixture of both cell populations and were immunized with the above-mentioned antigens. An efficient immune response to the ordered collagen-like (Pro-Gly-Pro)(n) was obtained in the absence of transferred thymocytes. The thymus independence of (Pro-Gly-Pro)(n) was confirmed when thymectomized irradiated mice were used as recipients. In contrast with these results, cooperation between thymus and marrow cells was necessary in order to elicit an immune response to (Pro(56), Gly(34))(n). Similarly, the immune response to the triple helical collagen was found to be independent of the thymus, whereas for an effective response to its denatured product, gelatin, thymus cells were required. These findings indicate that a unique three-dimensional structure of immunogens possessing repeating antigenic determinants plays an important role in determining the need for cell to cell interaction in order to elicit an antibody response.

An inverse relationship exists between the net-electrical charge of immunogens and the antibodies elicited (1). The cellular basis of the net charge phenomenon has been established for both positively and negatively charged immunogens, by cell separation techniques over columns of opposite charge (7, 8). To establish whether this phenomenon can be extended to include cell-mediated immunity, the response to basic encephalitogenic protein (BE) which induces experimental allergic encephalomyelitis (EAE) was now investigated. Lymph node cells from sensitized strain 13 guinea pigs were fractionated over positively and negatively charged columns and compared to unfractionated cell populations in two assay systems: (a) in vitro response to BE in terms of lymphocyte transformation and (b) the passive transfer of EAE to unsensitized syngeneic recipients. The response was found to be confined to the fraction of cells eluted from glass bead columns, namely, the more negative cells. Cells eluted from poly-L-lysine-coated glass bead columns (i.e., positive cells) were devoid of the capacity to respond to this antigen either in vivo or in vitro. It was previously established that thymocytes rather than bone marrow cells account for the inverse charge phenomenon as assayed by T-helper-cell function in in vivo antibody production (8). We have now extended the inverse charge effect to include cell-mediated immune response of the delayed hypersensitivity type.

The synthetic antigen denoted P2-A--L, comprising the fragment P2 of the coat protein of MS-2 coliphage attached to multichain poly-DL-alanine, served for the immunization of guinea-pigs. Immunization was carried out either in phosphate buffered saline (PBS) or in Freund's incomplete adjuvant (FIA) in the presence or absence of a small molecular weight peptidoglycan prepared from Bacillus megaterium, which was checked for its adjuvant effect. The various antisera were assessed by their capacity to neutralize MS-2 bacteriophage viability. When injected in PBS or FIA, P2-A--L did not elicit any measurable anti-phage activity. Addition of the peptidoglycan by simple mixing did not bring about a significant increase in antibody production. However, when the peptidoglycan was chemically linked to the P2-A--L conjugate, it had a marked adjuvant effect when the material was administered in FIA, almost identical to the extent of the effect of Freund's complete adjuvant.

Some synthetic peptides derived from the B subunit of cholera toxin, conjugated to tetanus toxoid have been shown by us to be efficient when administered in complete Freund's adjuvant (CFA), in induction of anticholera toxin response with neutralizing capacity. When the peptide CTP3, corresponding to residues 50-64 within the sequence of the B subunit, was attached to the hardly immunogenic A-L (multichain poly-DL-alanine) the resulting conjugate did not lead to a significant anti-peptide response even when given in CFA. On the other hand, when CTP3 was coupled to the immunogenic (T,G)-A-L, the resulting CTP3-(T,G)-A-L led, in CFA, to protective anti-cholera antibodies. Moreover, covalent attachment of the synthetic adjuvant N-acetylmuramyl-L-alanyl-D-isoglutamine (MDP) to the latter immunogen led to MDP-CTP3-(T,G)-A-L which, when administered in aqueous solution, elicited in rabbits antibodies with neutralizing capacity.

Previous studies have demonstrated an inverse charge relationship between the net electrical charge of antigen and the responding cells. In the present study we attempted to establish whether this phenomenon holds also for the primary recognition phase of cell-mediated immunity, when the involvement of the macrophage in presenting antigen is obligatory. Normal spleen cells were fractionated over negatively and positively charged columns. The fractionated cell populations, as well as the original cells, were sensitized in vitro on macrophage monolayers that were pulsed either with the basic encephalitogenic protein of myelin or with the acidic copolymer poly(Glu50,Tyr50). Cells eluted from glass bead columns(i.e., the more negative cells) could be sensitized only with the basic antigen, while cells eluted from poly(L-lysine)-glass bead columns (i.e., more positive cells) could be sensitized only to the acidic antigen. Thus, in delayed type hypersensitivity the inverse charge relationship prevails also for the primary immunological recognition of antigen bound to macrophages.

The crossreactivity between the random synthetic polypeptide antigen poly(Tyr,Glu)-poly(DLAla)--poly(Lys) [(T,G)-A--L] and its ordered-sequence analogs (Tyr-Tyr-Glu-Glu)-poly(DLAla)--poly(Lys) [(T-T-G-G)-A--L] and (Tyr-Glu-Tyr-Glu)-poly(DLAla)--poly(Lys) [(T-G-T-G)-A--L] at the level of humoral and cellular responses was studied. For delayed type hypersensitivity responses, (T,G)-A--L-activated T cells could be challenged with the homologous antigen as well as with the ordered analogs. T cells activated by (T-T-G-G)-A--L could be challenged with either the homologous antigen or with (T,G)-A--L but not with (T-G-T-G)-A--L. Similarly, no cross stimulation was observed between (T-G-T-G)-A--L-activated cells and (T-T-G-G)-A--L, whereas (T,G)-A--L could challenge the latter cells to mediate significant responses. Similar but not identical cross reactions were observed when primed spleen cells or lymph nodes were transferred to irradiated recipients that were boosted for the production of antibodies. In contrast to observations at the level of cellular responses, (T-G-T-G)-A--L-primed spleen or lymph node cells could not be boosted with (T,G)-A--L for the production of detectable amounts of antibodies, although boosting with the homologous antigen resulted in significant levels of (T-G-T-G)-A--L-specific antibodies. Transfer experiments in which mixtures of T and B cells, each primed to a different ordered polypeptide antigen, were injected into irradiated recipients showed that successful cooperation occurs provided that the boost is given with the T-cell-specific antigen. The antibodies produced were specific to the antigen used for B-cell priming. The T-cell-B-cell collaboration probably occurs through specific determinants that are shared between the two antigens in which the ordered peptides are attached to the same multichain polymer and that are recognized by both the T and the B cells.

Two synthetic polypeptides which differ only in the order of amino acids in their NH2-terminal side chains, namely, (Tyr-Tyr-Glu-Glu)-poly(DLAla)- -poly(LLys) and (Tyr-Glu-Try-Glu)-poly(DLAla)- -poly(LLys), were found to be under different genetic control. By three different in vivo systems for thymus-derived cell depletion, it was demonstrated that (Tyr-Tyr-Glu-Glu)-poly(DLAla)- -poly(LLys), which represents the random poly(Tyr,Glu)-poly(DLAla)- -poly(Lys) in the pattern of immune responses and in the quality of antibodies they elicit, is thymus-dependent whereas (Tyr-Glu-Tyr-Glu)-poly(DLAla)-poly(LLys) does not require thymus-derived cell help for efficient antibody production. Therefore, the two ordered polypeptides which are similar chemically differ in parameters, not yet determined, which affect their capability to trigger bone marrow-derived cells.

The injection of two synthetic polypeptide antigens into adult rabbits treated with 6-mercaptopurine, or into newborn rabbits, resulted in immunological "unresponsiveness" to subsequent immunizations with these antigens. The "tolerant" animals were shown to be reactive toward the non-related antigen ovalbumin.

SJL mice are high responders to the synthetic multichain polypeptide antigen (T,G)-Pro--L, whereas DBA/1 mice are low responders (10, 11). In order to determine whether the genetic control of immune response can be correlated with the number of antigen-sensitive precursor cells, spleen cell suspensions from normal and immunized SJL and DBA/1 donor mice were transplanted into lethally X-irradiated syngeneic recipients (incapable of immune response) along with (T, G)-Pro--L. Anti-(T, G)-Pro--L responses (donor-derived) were assayed in the sera of the hosts 12-16 days later. By transplanting graded and limiting numbers of spleen cells, inocula were found which contained one or a few antigen-sensitive precursors reactive with the immunogen. Using this method to estimate the relative numbers of such cells for the high responder SJL strain, one precursor was detected in approximately 1.3 x 10(6) and approximately 7.2 x 10(6) spleen cells from immunized and normal donors, respectively. In contrast, one precursor was detected in about 30 x 10(6) spleen cells from low responder DBA/1 mice, irrespective of whether the donors had been immunized. These results indicate that the genetic control of immunity to the synthetic polypeptide antigen investigated is directly correlated to the relative number of precursor cells reactive with the immunogen in high and low responder strains.

The role of thymus in antibody responses to a series of four synthetic polypeptide immunogens of the general formula multi-copoly(Tyr,Glu)-poly(Pro)-poly(Lys) was investigated as a function of the optical activity of the amino acids composing their structure. Irradiated nonthymectomized and thymectomized SJL mice were injected with thymocytes, marrow cells, or a mixture of both. Each group of recipients was immunized with the following copolymer enantiomorphs: all L-amino acids; L-amino acids outside and D inside; D-amino acids outside and L inside; or all D-amino acids. The antibody response to the immunogen composed of all L-amino acids was thymus-dependent, whereas the responses to the other three copolymers were all independent of the thymus. Similar cell transfers were performed in DBA/1 mice immunized with multi-copoly(L-Phe,L-Glu)-poly(D-Pro)-poly(D-Lys). This mouse strain produces specific antibodies against the (Phe,Glu) region and against the poly(D-prolyl) region. The immune response to the determinant with only L-amino acids on the outside was thymusdependent, whereas the response to the inside immunopotent region with only D-amino acids was thymusindependent. Since earlier studies have demonstrated that synthetic polypeptide antigens that contain D-amino acids are poorly metabolized, the thymus-independence of the antibody responses to these multichain synthetic polypeptides that possess repeating antigenic determinants was correlated with the metabolizability of the immunogens or their component determinants.

1. Three random linear copolymers composed of two or three of the amino acids d-tyrosine, d-glutamic acid, d-alanine and d-lysine, and a branched multichain copolymer with a poly-d-lysine backbone and polymeric side chains of d-tyrosine and d-glutamic acid, were found to be non-antigenic in rabbits, by precipitin and passive cutaneous anaphylaxis, and in guinea pigs, by delayed hypersensitivity tests. The corresponding four copolymers of l-amino acids were shown to be antigenic by all the three criteria. 2. No immunological cross-reactions were observed between the polypeptides composed of d-amino acids and the corresponding l-amino acid copolymers. 3. Similarly, an azobenzenearsonic acid conjugate of poly-d-tyrosine was shown to be non-antigenic in guinea pigs, in contrast with an analogous conjugate of poly-l-tyrosine. Animals sensitized with the conjugate of poly-l-tyrosine did not exhibit delayed skin reactions, when cross-tested with the d-conjugate. 4. A linear polymer composed of d-tyrosine, l-glutamic acid and l-alanine was found to be immunogenic and to cross-react with the corresponding polymer composed exclusively of d-amino acids.

Two series of experiments were performed, utilizing a modification of the hemolysin plaque technique which registers 19S antibody, in an attempt to determine the frequency of cells capable of simultaneously producing antibody to two non-cross-reacting antigens. Mice were immunized i.v. with rabbit and camel RBC and their spleens assayed for cells producing antibody against both antigens. 16,904 cells producing antibody of one or the other specificity, from 26 mice, were counted. Not one cell was detected which produced antibody of two specificities. Rabbits were immunized intradermally with HSA to which polyalanyl and p-azobenzenearsonate groups were chemically attached. The individual haptens, polyalanyl, and p-azobenzenearsonate groups were coupled to separate aliquots of SRBC, and the lymph nodes of immunized rabbits were assayed for cells releasing antibody against both haptens. In a study of 11 rabbits, after counting 27,845 cells producing antibody, we detected no "double" plaques.