Electron microscopic observations on antibody-producing lymph node cells.

Lymph node cells of rabbits injected with sheep erythrocytes, identified as antibody-producing by their ability to produce plaques of hemolysis in erythrocyte-containing agar layers, have been examined by electron microscopy, by the use of a procedure devised for subjecting single cells to such examination. The antibody-producing cells thus examined were found to fall into two classes, according to the current terminology: some were in the category of lymphocytes, and others, in the category of plasma cells. Within each class, cells were found to vary in certain characteristics, especially in the degree of development of such organelles as the nucleolus, Golgi apparatus, and the endoplasmic reticulum. In the case of the endoplasmic reticulum especially, it could be seen that a series of these plaque-producing cells, ranked in order of increasing size and development of the endoplasmic reticulum, would extend over a considerable range from those lymphocytes with the least developed organelles to the mature plasma cells with the greatest development of these structures.

Studies on antibody-producing cells. II. Appearance of 3 H-thymidine-labeled rosette-forming cells.

A study of the kinetics of antibody-producing cells has been carried out by the use of rosette formation for detection of individual antibody-producing cells, and labeling with tritiated thymidine, in cells obtained from mouse spleens at intervals after injection of SRBC. Following a primary injection of the antigen, the number of RFC per million cells was found to increase to a peak at 5 days, then, after a decrease, to a second peak at about the 10th day. The curve of tritium labeling of RFC was also biphasic, with peaks on the 3rd and 7th day. The second increase in rosette-forming cells could be shown to involve, especially between the 7th and 9th day, a second increase in lymphoid cell RFC and, among these, 7S antibody-producing cells. When the population examined was restricted to large lymphocytes, two peaks of RFC per million cells and two peaks of labeling were again found. In this case, however, the peaks of RFC and of labeling were reached on the same day in each instance, rather than with the 2 day difference found in the entire spleen cell suspension or the entire lymphoid cell population. Electron microscopic examination of labeled rosette-forming cells showed these to be largely lymphocytes, but to include rather well differentiated plasmablasts as well. No macrophages were found among labeled RFC in the primary response. A substantial number of labeled lymphocytes were found in close contiguity with rosette-forming macrophages. The percentage of labeling in such lymphocytes was as high, on the respective days, as the percentage of labeled cells among the RFC of the entire suspension.

Studies on antibody-producing cells. I. Ultrastructure of 19S and 7S antibody-producing cells.

Antibody-bearing cells of spleen and lymph node of the mouse and rabbit detected by rosette formation with the antigenic red blood cells were collected by micropipet and studied by electron microscopy. More than 300 such cells were examined. In the lymph nodes, rosette-forming cells were all in the lymphocytic and plasmacytic categories. In cells of the mouse spleen, macrophages were also found among the RFC, especially in the later days after immunization. The great majority of the RFC, 70-100%, were of the lymphocytic category. These included small, medium, and large lymphocytes with fine gradations of differentiation, and blast forms with little heterochromatin. The endoplasmic reticulum of these cells occurred in short, very narrow pieces, usually in contact with a mitochondrion. The cells of the plasmacytic category also showed fine gradations from plasmablasts to typical mature plasma cells. Plaque-forming cells of mouse and rabbit were also collected by micropipet. Of 162 such cells, fine gradations were also found throughout the lymphocytic and plasmacytic categories, but in this case the great majority were in the plasmacytic group, and more plasma cells showed amorphous nuclear chromatin. Among antibody-forming cells detected by both reactions, some of the more highly differentiated large lymphocytes contained ER which differed from that in the other large lymphocytes in that the channels were slightly and variably distended, with deposition of some precipitate, and with some tendency to a more nearly parallel orientation of the few channels seen. These were considered transitional cells. Of 10 RFC found in mitosis, all were in the lymphocytic category, in various stages of differentiation, the most advanced of which (in 2 of the 10 cells) was that of the transitional lymphocyte described here. Cells producing plaques facilitated by antisera vs. IgG of the mouse or rabbit (7S) showed the same distribution between cell categories and the same fine gradations as the direct (19S) PFC. Cells producing rosettes which were resistant to lysis in the presence of complement, and were thus presumably producing 7S antibody, showed a distribution similar to that found generally with rosette-forming cells, approximately 80-90% in the lymphocytic category.

Studies on antibody-producing cells. 3. Identification of young plaque-forming cells by thymidine- 3 H labeling.

Mice injected with sheep RBC and then, 4 days later with thymidine-(3)H, were sacrificed on the day of thymidine-(3)H injection or 1 or 2 days later. Hemolytic antibody plaque preparations were made of cells from the draining lymph nodes by a thin-plating procedure permitting collection of isolated PFC for electron microscopic examination and radioautography. Of cells obtained on the day of thymidine-(3)H injections, 65% of the labeled PFC were in the lymphocytic category, in comparison with 13% found previously in the entire population of such cells. The remaining 35% were plasmablasts in early stages of differentiation. Cells obtained 1 day after the thymidine-(3)H injections showed a shift to a majority of labeled cells in the plasmacytic category. Also, the plasmablasts were substantially more differentiated than those of the previous day, and some mature plasma cells were now seen. The labeled PFC obtained on day 2 gave no indication of further differentiation. Cells of rabbit lymph nodes labeled in vitro with thymidine-(3)H showed a range of labeled PFC. The majority were in the plasmacytic category, including some mature plasma cells. The data from the experiments with in vivo labeling suggest a direct differentiation from antibody-synthesizing lymphocytes to plasma cells. Further, the in vivo experiments indicated that differentiation from nascent lymphocyte to plasma cell could be essentially completed within 1 day.

Studies on antibody-producing cells. IV. Ultrastructue of plaque-forming cells of rabbit lymph.

Efferent lymph of the popliteal lymph nodes of rabbits was collected 4 days after a single footpad injection of SRBC. Thin-layer agar plating was done to isolate plaque-forming cells of the lymph for electron microscope examination, and the numbers of plaque-forming cells (PFC) in cells from the lymph and lymph nodes were determined. Of 71 PFC of lymph isolated and examined, 93% were lymphocytes, most of them with signs of substantial levels of physiologic activity. The cytoplasm showed an abundance of free ribosomes and many finger-like projections. The endoplasmic reticulum (ER) was barely detectable in most of the active lymphocytic PFC, and in some, a few short narrow channels of ER could be seen. Approximately one-fifth of the lymphocytic PFC presented an appearance of senescence, with signs of degeneration: rounded cells, with amorphous nuclear chromatin, and very few microvilli. The remaining 7% of the PFC of the lymph showed an unusual combination of features: small round cells with a narrow ring of cytoplasm which, however, contained well-organized channels of ER. Such cells had been found only among PFC of peripheral blood of the rabbit. The number of PFC per million cells was higher in the lymph than in the suspensions of lymph node cells. In both the contralateral lymph node and its efferent lymph, the number of PFC was less than 1% that of the injected side.

Electron microscopic observations on antibody-producing cells in lymph and blood.

Antibody-producing cells have been identified among cells obtained from efferent lymphatic vessels, the thoracic duct, and peripheral blood. These cells, which produced plaques of hemolysis and which were quite rare (20 to 50 per million), due in most instances to 19S antibody, were located and studied by electron microscopy. Of the antibody-producing cells found in these three sites there were several features common to all: small size (5 to 8 micro), generally spherical shape, approximately central position of the nucleus, and retention in the nucleus of the condensations of chromatin characteristic of the lymphocyte. The differences among the cells of these sources were largely in the relative amount and state of organization of the organelles of the cytoplasm. In cells of the efferent lymphatic vessel and the thoracic duct, the endoplasmic reticulum showed a range from relative scarcity to considerable numbers of well organized channels. Between these extremes were cells with a considerable amount of endoplasmic reticulum, the channels disorganized and sectioned at various angles. The cytoplasmic matrix of all of these contained a profusion of polyribosomes. Antibody-producing cells obtained from peripheral blood showed, around the roughly spherical nucleus, a ring of cytoplasm which was narrow, but wholly organized into parallel lamellae of endoplasmic reticulum, with many polyribosomes between these, and a large Golgi body. Some similarities and some differences of these cells, in comparison with antibody-producing cells obtained from lymph nodes, have been indicated.

Enhancement of growth of allogeneic mouse tumor by the IgG1 fraction of alloantibody preparations.

Because staphylococcal protein A binds all the known subclasses of mouse IgG except IgG1, ethanol-fixed staphylococci were used as an adsorbent to prepare IgG1 fractions of anti-BALB/c alloantibody-containing globulins and normal globulins of the same strains. The loss of more than 99% of the IgG2 as a result of this adsorption was demonstrated by immunodiffusion. The IgG1 fractions of C3H and CBA anti-BALB/c globulins were tested for their effect on growth of the BALB/c plasmacytomas MOPC-315 and MOPC-460 in C3H and CBA mice by incubation with the tumor cells before transplantation and by injection periodically thereafter into the hosts. With alloantibody-containing globulins that showed slight enhancement of growth of these tumors, or none, the IgG1 preparations caused considerable enhancement of tumor growth. Control preparations of normal C3H or CBA globulins, or IgG1 fractions similarly prepared from the normal globulins, showed no enhancing effect on the growth of these tumors.

A preparation of Fc fragment of normal mouse IgG1 for production of rabbit anti mouse IgG1 serum.

Ascitic fluid globulins obtained by peritoneal irritation from normal CBA mice were digested with papain, and the digests were examined by gradient chromatography on CM cellulose. The classical three peaks of OD280 found by Porter with papain digests of rabbit IgG were obtained. Fraction II contained material which reacted in immunodiffusion against mouse anti-IgG. The ascending part of this peak reacted against anti-IgG1 and not against anti-TgG2 reacting material appearing at detectable levels in the descending part of this peak. A pool from the ascending part of this peak, containing Fc of IgG1 and not visibly contaminated with Fc of IgG2, was injected into rabbits. The resulting antisera were free enough of anti-IgG2 to be effective in removing IgG1 from alloantibody-containing globulins without appreciable loss of antibodies of IgG2 class, thus allowing the complement-dependent IgG2-class antibodies full expression of their titer, without competition by Ig1 class antibodies. Chromatography of native mouse globulin did not produce a similar degree of separation of Ig1 from IgG2.

An improved plaque assay for mouse myeloma (MOPC 315) cells for use in studies of humoral and cell-mediated immunity.

Dinitrophenyl-bovine albumin was coupled at room temperature to sheep red blood cells in a procedure which minimized spontaneous lysis and allowed the preparation of large batches and their use for at least 3 weeks. The modified erythrocytes were used as a substrate for detecting local hemolytic plaques in agar by myeloma MOPC 315 cells, which secrete a paraprotein IgA with high affinity for dinitrophenyl ligand. Conditions maximizing the number of plaques formed by a given number of tumor cells were found to include coupling the erythrocytes at 1 mg/ml dinitrophenyl-bovine albumin with a molar ratio of about 50, and incubation with an amino-to-carboxy cross-linking agent, 1-ethyl-3(3 dimethyl aminopropyl) carbodiimide, at 2 mg/ml for 50 min. The method thus developed was employed to measure cellular and antibody-dependent immune reactions against the MOPC 315 cells. The experimental results show comparisons of the plaque technique with other measurements of tumor cell injury. The nature of the assay, which requires only 500 cells per plating, and which tests the synthetic capacity of single cells, suggests its use in experiments which limit the number of target cells, and in immune reactions causing injury, but not necessarily lysis, of the target cells.

Shifts from IgG-2 class to IgG-1 class in CBA and C3H anti-BALB/c antibody.

With use of a recently developed method for determining relative levels of IgG-1 and IgG-2 class antibodies of a given specificity within an unfractionated serum, it has been possible to examine anti-BALB/c antibodies in the early and late part of an immunization with allogeneic spleen cells. At about 6 days after primary immunization of CBA or C3H mice with BALB/c spleen cells, suppressive antibodies can be measured in the sera of the animals. About half of these are attributable to IgM class, and this contribution decreases to zero by the 12th day. The remaining suppressive antibodies are of IgG-2 class and these increase in concentration until day 8 or 12, or begin to decline between day 8 and day 12. Anti-BALB/c antibodies of IgG-1 class have not yet appeared on day 6, but thereafter appear and increase in concentration. Thus, antibodies of IgG-1 class begin to appear after those of the IgG-2 class and may still be increasing after the IgG-2 class has stopped to increase in concentration, antibody of IgG-1 class is continuing to increase and may even continue to increase after IgF-2 class antibody has begun to decrease in concentration. Thus, the synthesis of IgG-1 class antibody begins later and continues later than that of IgG-2 class. The implications of this sequence for our data on various effects of anti-H-2 antibodies on retention of skin allografts are discussed.