The Journal of Experimental Medicine

O'Toole, C.; Stejskal, V.; Perlmann, P.; Karlsson, M.

doi: 10.1084/jem.139.3.457pmid: 4591169

Peripheral lymphocytes from patients with urinary bladder carcinoma and controls have been separated on the basis of rosette formation with sheep erythrocytes. The fractions were tested for tumor-specific cytotoxicity. The E rosette-forming cells of purity ⩾ 90% respond well in PHA-induced cytotoxicity but are totally inactive in the tumor assay. The non-E rosette-forming cells (purity ⩾ 91%) give enhanced activity in the tumor-specific cytotoxicity as well as in antibody-mediated target cell lysis in a model system. These data support the notion that the effector cells in cell-mediated immunity to carcinoma of the urinary bladder are members of the nonthymus-derived population of peripheral lymphocytes. Footnotes Submitted: 24 October 1973

Kolb, Cornelia; Di Pauli, Raimund; Weiler, Eberhardt

doi: 10.1084/jem.139.3.467pmid: 4544246

IgG of paternal allotype first becomes detectable in the serum of (BALB/c x C57BL/6)F 1 mice between day 12 and 14 after birth and reaches adult levels at an age of 5 wk. Since in mice there is a transfer of maternal IgG molecules through the placenta and via milk, F 1 heterozygous at the allotype locus were used and the concentrations of IgG with paternal allotype were measured. This was done by a sensitive method capable of detecting IgG concentrations as low as 5 x 10 –4 of normal adult serum levels. It is based on the quantitative inhibition of allotype-specific facilitation of hemolysis. When lipid A or Salmonella bacteria were injected into neonatal mice, a stimulation of IgG synthesis was observed. Thus IgG levels were enhanced 10-30-fold compared to the nontreated mice. No increase in IgG levels was obtained in adult mice after treatment with lipid A. Whether the newborns were injected at birth, on day 2, 4, or 7, IgG was first demonstrable in the treated mice at an age of 6–11 days. The increase in IgG levels was not paralleled by a demonstrable antibody activity against lipid A, SRBC, and LPS. Thus the bulk of newly induced IgG is probably a statistical distribution of different specificities. Footnotes Submitted: 19 October 1973

Hunt, Simon V.; Williams, Alan F.

doi: 10.1084/jem.139.3.479pmid: 4544247

The origin of immunoglobulin on the surface of TDL in the rat has been studied by comparing the binding of purified alloantibodies recognizing the Ig-1a allotype of rat light chain, with that of rabbit antirat Fab antibodies. Both reagents labeled all TDL from rats of the DA strain (Ig-1a) with two categories of cells being detected; one binding 100–2,000 molecules of antibody, the other 10,000–100,000 molecules. These categories were likely to be synonomous with T and B cells, respectively. The 125 Iantiallotype antibodies did not bind to TDL from rats of the PVG/c strain (Ig-1b). When the binding to TDL from (PVG/c x DA)F 1 animals was studied it was found that allelic exclusion occurred in the heavily labeled cells, but not in the lightly labeled ones. Furthermore, when lymphocytes of one allotype were transferred to irradiated recipients of the opposite allotype and recovered from the TDL or spleen of the recipient 20–30 h later, the immunoglobulin on heavily labeled cells was of the donor type, while that of lightly labeled ones bore the recipient marker. Thus heavily labeled cells (B lymphocytes) had synthesized their own immunoglobulin while lightly labeled cells (T lymphocytes) had acquired theirs passively by adsorption. The class of immunoglobulin on lightly labeled cells was also studied and it was found that 125 Ianti-IgM antibodies bound to an extent approaching the 125 Ianti-Fab binding, while 125 Ianti-IgG 2a+2b antibodies gave much less binding.

Cappel, Roger; Schluederberg, Ann; Gifford, Robert H.; Horstmann, Dorothy M.

doi: 10.1084/jem.139.3.497pmid: 4204728

A precipitating antigen, rho , was first detected in the blood of persons with rubella and in rubella virus-infected cell culture fluids (1). Partially purified antigens from both sources were examined and shown to have similar properties, although antigen from serum sedimented more heterogeneously, with estimated coefficients from 15 to 21 S, while that from culture fluids sedimented in the 11–14 S region. In each case, antigen was located in the ß-1 zone after electrophoresis in agarose, and at a density of 1.305 g/ml after centrifugation in CsCl. Stability characteristics were typical of protein antigens. Immunofluorescent microscopy revealed that rubella virus induced the appearance of rho antigen scattered throughout the cytoplasm of infected cells. When cells containing antigen were exposed for 24 h to 5 µg/ml actinomycin D rho was no longer detectable, indicating the probable cellular origin of the antigen. Also, titers in medium of infected cultures showed a reduction after actinomycin treatment, but levels of the virus-specified antigen, iota , were relatively unaffected. Rho appears to be a protein common to man and many animals. In vitro, it was induced by rubella virus and by adenovirus. In vivo, rho titers were shown to be elevated after rubella virus infection and, to a lesser extent, after infection with certain other viruses. High titers were also demonstrated in women late in pregnancy and in patients with rheumatoid arthritis. In man and the chimpanzee, the appearance and decline of rho in the blood after rubella virus infection were temporally similar to the patterns of CRP, although rho seemed to be a more sensitive indicator of infection. The data presented indicate that rho is a newly recognized acute phase protein inducible by certain virus infections and by other unidentified stimuli present prominently in pregnancy and rheumatoid arthritis. Footnotes Submitted: 11 November 1973

Bosma, Melvin J.; Bosma, Gayle C.

doi: 10.1084/jem.139.3.512pmid: 4130242

This paper derives from the unexpected observations of the "wrong immunoglobulin allotype" in a congenic partner strain of BALB/c mice from the Institute of Cancer Research (ICR CB-17). These mice were specially bred so as not to differ from BALB/c mice in any known way except to carry immunoglobulin structural genes of the C57BL/Ka allotype. In this respect, ICR CB-17 mice were defined as allotypically homozygous according to the Mendelian inheritance of mouse allotype markers. The homozygosity of these mice was challenged, however, when in certain instances immunoglobulins of the BALB/c allotype appeared in the serum of some ICR CB-17 mice. The appearance of this hidden allotype was usually transient and only associated with immunoglobulins of the IgG (IgG2a) class. The implications of these findings for the inheritance and expression of immunoglobulin structural genes are discussed. Footnotes Submitted: 28 September 1973

Lagrange, P. H.; Mackaness, G. B.; Miller, T. E.

doi: 10.1084/jem.139.3.528pmid: 4591170

Delayed-type hypersensitivity (DTH) develops in the absence of an adjuvant when mice are injected intravenously or subcutaneously with an appropriate dose of sheep red blood cells (SRBC). The optimal intravenous dose of 10 5 SRBC (in CD-1 mice) produces maximum DTH which decays exponentially from its peak on day 4. Increasing the dose of SRBC reduces and eventually abolishes all evidence of DTH. DTH fails to reappear in respose to secondary stimulation except in splenectomized mice in whom the development of DTH is not suppressed, even by massive doses of SRBC. Hence the suppression cannot be due to antigen as such. The optimal dose of SRBC for sensitization by footpad inoculation is 100-fold higher (10 7 SRBC in CD-1 mice), but even 10 9 SRBC do not block the induction of DTH by this route of immunization. A blocking dose of SRBC, given intravenously 1 day before footpad inoculation, completely suppresses cell proliferation in the draining lymph node, prevents PFC production there, and blocks the induction of DTH by a sensitizing dose of SRBC. If given 1 day after footpad sensitization, intravenous antigen has little effect on the cellular response in the regional node but DTH is still completely suppressed. Blocking of induction and expression may depend, therefore, on different mechanisms. Footnotes Submitted: 25 October 1973

Mackaness, G. B.; Lagrange, P. H.; Miller, T. E.; Ishibashi, T.

doi: 10.1084/jem.139.3.543pmid: 4591171

An explanation was sought for the fact that delayed-type hypersensitivity (DTH) does not normally occur in response to T-cell-dependent antigens unless an adjuvant is used. But when sheep red blood cells (SRBC) were administered intravenously DTH did appear, provided that the dose of antigen was less than that required to give a maximum antibody response. Animals in which T-cell activity had been blocked by a large dose of antigen could not be sensitized adoptively, and their spleen cells failed to transfer DTH to normal recipients. The serum of blocked animals partially inhibited the induction of DTH, and after absorption with SRBC its blocking activity increased substantially. Moreover, absorbed serum inhibited DTH in previously sensitized animals, but it did not inhibit the proliferative response to SRBC in peripheral lymph nodes or reduce the number of plaque-forming cells produced therein. On the contrary, the hemagglutinating titer was actually increased by blocking serum even though DTH was totally suppressed. It is concluded that a product of the interaction between antigens and antibody blocks the activated T cells which mediate DTH without interfering with helper cells. Footnotes Submitted: 25 October 1973

Hoff, Richard L.; Frenkel, J. K.

doi: 10.1084/jem.139.3.560pmid: 4812629

The capacity of hamster peritoneal cell populations to control viability and growth of Besnoitia and Toxoplasma organisms was assessed in vivo and in vitro. Immunized hamsters reduced the homologous organisms 100- to 10,000-fold over a 5-day period, but the heterologous infection increased 100- to 1,000-fold in numbers, similar as in the nonimmune controls. Passively administered antibody was ineffective although lytic cofactors were supplied by hamsters. In cultures, peritoneal cells from Besnoitia-immune hamsters delayed the growth of homologous parasites to an average of 38.5 h per division; however, in Toxoplasma-immune and nonimmune cells, Besnoitia divided every 12.8 h. Specificity of immunity was pronounced against both infections. With cross-infections, Toxoplasma-immune cultures did not effectively delay Besnoitia growth; however, Besnoitia-immune cultures reduced Toxoplasma growth by one-half. Co-cultivation experiments demonstrated that specifically committed lymphocytes could instruct macrophages to reduce the homologous organism 10-fold, whereas heterologous organisms were reduced only 2-fold. Lymphocyte supernatants initiated hypersensitivity as indicated by macrophage activation and giant cell formation in culture. However, these supernatants did not transfer infection immunity. Lymphokines could account for the hypersensitivity phenomena, but cell-mediated infection immunity in this model required close lymphocyte-macrophage proximity. These studies indicate that a number of distinct processes including delayed hypersensitivity, macrophage activation, and specific cellular immunity are acting simultaneously during latent Besnoitia infection of hamsters. All three processes are mediated by lymphoid cells and appear to be specifically induced. Although activated macrophages develop some heightened nonspecific capabilities, these were several orders of magnitude below the specific effects. Footnotes Submitted: 2 October 1973

Jones, Patricia P.; Cebra, John J.; Herzenberg, Leonard A.

doi: 10.1084/jem.139.3.581pmid: 4591172

Lymphocytes from b 5 / b 9 rabbits were stained in suspension with fluorescent antiallotype antibody reagents to selectively label with fluorescent molecules those cells bearing membrane immunoglobulin (Ig) of the b5 or b9 allotype. After staining, the cells were separated by the fluorescence-activated cell sorter into populations markedly enriched in cells bearing b5 or b9 membrane Ig or totally depleted of cells with detectable membrane Ig. The potential of these separated cells to give rise to Ig-synthesizing plasma cells either in vivo after transfer into irradiated recipients or in vitro during culture in the presence of phytohemagglutinin or pokeweed mitogen was assessed by immunofluorescence. The relative proportion of b5 and b9 cytoplasmic Ig-stained cells (CSC) arising from the separated cells was determined to test directly whether B lymphocytes and their progeny are committed to the synthesis of Ig of one allotype. It was found that b5- and b9-bearing cells gave rise almost exclusively to b5- and b9-producing plasma cells, respectively, in both the in vivo and in vitro assay systems. Most of these CSC were probably not derived from previously existing CSC but arose as the result of the differentiation of lymphocytes with membrane Ig. When cell populations totally depleted of Ig-bearing lymphocytes were cultured, very few CSC were found, indicating that the majority of immediate precursors of CSC have membrane Ig. These results suggest that individual B cell clones are phenotypically restricted to the expression of immunoglobulin genes on one chromosome; the significance of this clonal allelic exclusion is discussed. Footnotes Submitted: 3 December 1973

Kulczycki, Anthony; Isersky, Chaviva; Metzger, Henry

doi: 10.1084/jem.139.3.600pmid: 4812630

A rat basophilic leukemia cell line originally described by Eccleston et al. (3) has been successfully transplanted into eight Wistar strains and adapted to suspension cell culture without noticeable morphological changes. The cells deplete the reaginic activity of mouse and rat immune sera to an extent roughly equivalent to that reported for normal rat mast cells. Studies utilizing radioiodinated antilight-chain antibodies and radioiodinated partially purified rat IgE indicate that the cells bind IgE to their surface membrane with high specificity. Heating or mildly reducing the rat IgE destroyed its binding activity. The binding is unaffected by the presence or absence of Ca ++ and Mg ++ , and is markedly inhibited by reaginic but not by normal rat or mouse serums. The affinity of these cells for human IgE, if present at all, is substantially lower than for rat IgE. Footnotes Submitted: 11 November 1973