Abstract. The results reported establish a cellular basis for the inverse relationship between the net electrical charge of immunogens and of the antibodies provoked by them. Glass bead columns were used to reduce the number of immunocompetent spleen cells preferentially reactive with more acidic immunogens. After a single immunization, titers of antibodies to an acidic dinitrophenylated copolymer of tyrosine, glutamic acid, and lysine (DNP-901) elicited in recipient mice by filtered spleen cells were significantly lower than those generated by unfiltered cells. After secondary stimulation, the major portion of the antibodies provoked by the acidic antigen was found in the more acidic fraction eluted from DEAE-Sephadex, in contrast to the more basic antibodies, of the same specificity, generated by unfiltered spleen cells. Results obtained by transplanting a limiting number of spleen cells indicate a depletion in the number of precursor cells reactive with dinitrophenyl on DNP-901 after glass bead chromatography, whereas there was no change in the response to dinitrophenyl on a basic copolymer, DNP-912, containing the same amino acids in different molar ratios.The inverse relationship between the net electric charge of immunogenic macromolecules and the net charge of antibodies provoked by them has been demonstrated for antibodies produced in rabbits1'2, mice3, goats4, and humans5, and is valid for antibodies of IgG6, IgM7, and, probably, IgEl class. Thus, e.g., antibodies to the acidic diphtheria toxoid and to negatively-charged synthetic copolymers of amino acids appear mainly under the first peak upon fractionation on DEAE-Sephadex 8, whereas antibodies to the basic lysozyme and positivelycharged amino acid copolymers appear predominantly, or exclusively, under the second peak 3.Haptens such as the dinitrophenyl group1 2 or a peptide of L-alanine (unpublished data of Licht, A., B. Schechter, I. Schechter, and M. Sela) will lead to antibodies fractionating under the first or the second DEAE-Sephadex peak, depending on whether they are attached to a negatively or positively charged carrier. The anti-hapten antibodies produced upon immunization with such conjugates differ in their net electric charge, but not in their specificity or affinity, as measured by means of small molecules related chemically to the original haptens. The inverse correlation described depends on the net electric charge of the intact immunogen, and not on the net charge within a limited area around the anti-1288

An inverse relationship exists between the net electrical charge of immunogens and the antibodies they elicit (1). Results of an earlier study have demonstrated that the net charge phenomenon has a cellular basis, since the immune response potential of murine spleen cells to 2,4-dinitrophenyl (DNP) on a negatively charged synthetic polypeptide carrier was reduced by cell fractionation over negatively charged glass beads, whereas the response to the same hapten on a positively charged carrier was unaffected (14). To verify that the net charge correlation is expressed at the cellular level, spleen cells were fractionated over positively charged poly-L-lysine-coated glass bead columns, and their immunocompetence to DNP on positively and negatively charged carriers was tested by cell transfers in irradiated recipient mice. In this case, the fractionated cells showed reduced response potential to DNP on the positively charged carrier only. Thus, the cellular basis of the net charge phenomenon has been demonstrated for both positively and negatively charged immunogens (for the same specificity) by cell separation techniques over columns of opposite charge. In order to establish whether the cell population relevant for the charge properties of immunogens was of thymus or marrow origin, thymocytes and bone marrow cells were selectively passed over positively or negatively charged columns and mixed with unfractionated cells of the complementary type. Transfers of the filtered and unfiltered cell mixtures in irradiated recipient mice immunized with DNP on either a positive or a negative synthetic polypeptide carrier indicated that fractionation of thymocytes, but not of marrow cells, correlated with the spleen population. Thus, thymocytes fractionated over negatively charged columns and mixed with unfractionated marrow cells exhibited reduced response to DNP on the negative carrier, but normal responses to DNP on the positive carrier. The opposite result was obtained when thymocytes were passed over positively charged columns. No effect on the anti-DNP response was detected by filtration of bone marrow cells over columns of either charge. These findings indicate that it is possible to distinguish between thymocytes on the basis of their capacity to react with more acidic or more basic surfaces and that a population of thymus-derived cells may recognize immunogens on the basis of their overall electrical charge. No evidence was found by these techniques that marrow-derived cells contribute to the net charge phenomenon.

An inverse relationship has been demonstrated between the net electrical charge of immunogens and the charge of the antibodies elicited by them (1-3). IgG antibodies to natural and synthetic negatively charged immunogens were found in the first, more basic, fraction of immunoglobulin eluted from diethylaminoethyl Sephadex A-50 . In contrast, the antibodies elicited by basic immunogens appeared in the second, more acidic, eluted fraction (1-5). This observation has been extended to include antihapten antibody responses generated by groups such as 2,4-dinitrophenyl and a tetrapeptide Of D-alanine, attached to positively and negatively charged carriers (1, 2, 6) . These hapten-specific antibodies differed in net electrical charge, but were indistinguishable with respect to their specificity and affinity (6). It has been previously established that the inverse net charge phenomenon has a cellular basis (7,8) . Thus, the immune response potential ofmouse spleen to the 2,4-dinitrophenyl hapten attached to a negatively charged synthetic polypeptide carrier was reduced by cell fractionation over negatively charged glass bead columns, whereas the response to the same hapten on a positively charged carrier was unaffected (7). Furthermore, spleen cells fractionated over positively charged poly-L-lysine-coated glass bead columns showed reduced response potential to the dinitrophenyl hapten on the positively charged carrier only.It was established that the cell population relevant for the charge properties of immunogens was of thymus and not of marrow origin, by experiments in which thymocytes and bone marrow cells were selectively passed over positively or negatively charged columns and mixed with unfractionated cells of the complementary type (8). Thus, thymocytes fractionated over negatively charged columns and mixed with unfractionated marrow cells exhibited reduced antibody response to the hapten on the negative carrier but normal responses to hapten

Previous studies have demonstrated an inverse relationship between the net electrical charge of immunogens and the antibodies they elicit. This correlation was found to be expressed at the cellular level. It has been shown that thymus-derived cells may recognize immunogens on the basis of their overall electrical charge. In this study, charged T-independent copolymers composed of tyrosine, glutamic acid, and lysine of the D configuration were prepared in order to find out whether this net charge phenomenon holds also for immunogens which do not require helper T cells for generation of immune response. Spleen cells were fractionated over negatively charged glass bead columns, and their immunocompetence was tested by transferring them into irradiated recipient mice which were immunized with the dinitrophenylated acidic or basic T-independent carrier. No differences in the responsiveness to the dinitrophenyl group on either carrier could be detected in recipients of unfractionated or fractionated cells on the charged columns. Similar results were obtained with the negatively charged T-independent branched polypeptide poly(DTyr,DGlu)-poly(DPro)--poly(DLys). Filtration of spleen cells through glass bead columns did not affect the immune response potential of the recipient mice. In contrast, a significant reduction was observed in the frequency of positive responses in recipients of filtered cells, which were immunized with the negatively charged T-dependent poly(DTyr,DGlu)-poly(DPro)--poly(DLys). Thus, the inverse net charge phenomenon holds only for T-dependent antigens.