MECHANISM OF ACTION OF MIGRATION INHIBITORY FACTOR (MIF) : I. EVIDENCE FOR A RECEPTOR FOR MIF PRESENT ON THE PERITONEAL MACROPHAGE BUT NOT ON THE ALVEOLAR MACROPHAGE

The initial interaction between migration inhibitory factor (MIF) and the guinea pig alveolar and peritoneal macrophage was studied. MIF-containing supernatants were generated from sensitized lymph node lymphocytes obtained from guinea pigs immunized with bovine gamma globulin in complete Freund's adjuvant. MIF-containing supernatants were markedly inhibitory for the migration of the peritoneal macrophage but had no effect on the alveolar macrophage. A linear relationship was observed between per cent inhibition of migration and serial twofold dilution of supernatant. Reexpressed in arbitrary MIF units, this relationship reflects a dose-response relationship with saturation characteristics. Pulse exposure of peritoneal macrophages to MIF resulted in adsorption of MIF onto both viable and nonviable cells with corresponding depletion of supernatant MIF. The alveolar macrophage did not adsorb MIF. Pulse adsorption of MIF onto the peritoneal macrophage is dependent on time, temperature, and cell number. Pretreatment of the cells with proteolytic enzyme prevents the adsorption of MIF while leaving migration unaffected. These observations support the existence of a specific cell surface receptor for MIF. The existence of such a receptor provides selectivity of immune modulation of macrophage populations by lymphocytes in delayed hypersensitivity reactions.     

脂多糖对腹膜和骨髓巨噬细胞产生红系造血调控因子的影响

腹膜和骨髓巨噬细胞培养中含红系造血祖细胞增殖的刺激因子,它不但能提高红系祖细胞的产率,而且还可增加每个集落中细胞数。大肠杆菌脂多糖(lipopolysaccharide,LPS)可提高巨噬细胞产生红系增殖刺激因子的能力。经过LPS激活的巨噬细胞培养液对晚期红系造血祖细胞增殖表现出双向调控作用,低浓度促进,高浓度抑制,其中LPS激活腹膜贴壁巨噬细胞的活性高于骨髓巨噬细胞。     

CELLULAR IMMUNITY IN VITRO : I. IMMUNOLOGICALLY MEDIATED ENHANCEMENT OF MACROPHAGE BACTERICIDAL CAPACITY

An in vitro model of cellular immunity in the guinea pig was established. Animals were immunized with tubercle bacilli, bovine gamma globulin, or picrylated human serum albumin in complete Freund's adjuvant. Oil-induced peritoneal exudates from immune and control animals were cultured overnight with and without specific antigen. The cultures were washed and the macrophage monolayers were infected with Listeria monocytogenes. At intervals the monolayers were lysed and the numbers of viable intracellular bacteria were quantitated by pour plate cultures. Random monolayers were also evaluated in sequence by visually counting the intracellular bacteria on Gram-stained plates. Both methods demonstrated that the macrophages from immune animals had markedly enhanced listericidal activity when the peritoneal exudates were cultured with antigen before infection. Macrophage migration inhibition was also demonstrated under these conditions. The experiments reported here describe an in vitro model of cellular immunity which will allow separation and recombination of cell types and direct assay of cell products in efforts to elucidate further the mechanisms of the immunologically mediated enhancement of macrophage bactericidal capacity.     

EFFECT OF NORMAL AND ACTIVATED HUMAN MACROPHAGES ON TOXOPLASMA GONDII

Human macrophages derived from in vitro culture of peripheral blood monocytes were studied under a variety of conditions to determine their microbicidal capacity for the obligate intracellular protozoan, Toxoplasma gondii. The effect of macrophages on intracellular Toxoplasma was evaluated morphologically by light and phase microscopy and by autoradiography. When macrophages from dye test (DT)-negative or DT-positive individuals were infected with Toxoplasma in the presence of normal human serum, the organisms were able to multiply intracellularly with resultant destruction of the monolayer. Once organisms were intracellular, the presence of antibody-containing serum in the medium did not alter this inability of the macrophages to kill Toxoplasma. However, when Toxoplasma were incubated in the presence of heat-inactivated DT-positive serum just before infection of the monolayers, the intracellular organisms were inhibited or killed by normal macrophages. Attempts were made to activate macrophages in vitro to kill Toxoplasma. Macrophages incubated in the presence of sensitized lymphocytes and Streptokinase-Streptodornase (SK-SD) or Toxoplasma lysate antigen (TLA) were found to kill Toxoplasma when compared to macrophages incubated in the presence of lymphocytes from DT-negative individuals and TLA or lymphocytes alone. Thus, in vitro induction of resistance (both specifically and nonspecifically) in human macrophages was accomplished by culturing these cells in the presence of specifically sensitized lymphocytes and antigen. These results suggest that, as in the mouse model, activated human macrophages have the ability to inhibit or kill intracellular Toxoplasma and that these cells may be important as effector cells in cell-mediated immunity (CMI) to toxoplasmosis in man.     

Assessment in vitro of immunity against Toxoplasma gondii

Studies have been made on humoral and cellular immune respones in mice immunized with an attenuated strain of Toxoplasma gondii. Heat- inactivated antitoxoplasma serum did not cause morphologic changes in the organisms, but did markedly influence their interactions with host cells. Toxoplasma exposed to antibody were no longer capable of entering fibroblasts or HeLa cells. They were readily engulfed by macrophages, but the antibody treatment strikingly altered the intracellular fate of the parasites leading to killing and digestion of the toxoplasmas in phagolysosomes. Addition of antitoxoplasma antibody immediately after infection of macrophages in vitro had no effect on intracellular multiplication of the organism. The division time of virulent toxoplasmas in mouse peritoneal macrophages in vitro was markedly prolonged in cells from immunized mice. During the first 2-3 mo after immunization, the macrophages harvested from the peritoneal cavity demonstrated this cellular immunity directly; thereafter exposure of the macrophages to immune lymphocytes and toxoplasma antigen, or to supernates from such an interaction was required for induction of the maximal capacity to inhibit growth of toxoplasmas. Induction of the alternation in macrophages by the lymphocyte product was detectable in 6 h and maximal at 18-24 h. Cultivation in vitro of macrophages from immunized animals for periods longer than 48 h rendered the cells nonresponsive to the immune lymphocyte-toxoplasma product. Macrophages from the peritoneal cavities of normal, nonimmunized mice were also incapable of developing the capacity to inhibit growth of toxoplasmas in response to this product. The nonresponsiveness of normal macrophages, or of macrophages cultured for several days in vitro was not changed by exposure of the cells to antitoxoplasma serum.     

ALTERATIONS OF MACROPHAGE FUNCTIONS BY MEDIATORS FROM LYMPHOCYTES

Sensitized lymphocytes were incubated in vitro with the specific antigen Supernatants from these cultures were chromatographed on Sephadex G-100 columns. Supernatant fractions containing MIF, chemotactic factor, and lymphotoxin, but free of antigen and antibody, were incubated with normal peritoneal exudate macrophages. Macrophage adherence, phagocytosis, spreading, motility, and direct hexose monophosphate oxidation were enhanced, while protein synthesis was unaffected. Thus, antigen-stimulated lymphocytes secrete a factor or factors which enhance certain macrophage functions. Implications for models of cellular immunity and cellular hypersensitivity are discussed.     

THE INTERACTION IN VITRO OF MYCOPLASMA PULMONIS WITH MOUSE PERITONEAL MACROPHAGES AND L-CELLS

Methods have been devised for establishing infection in vitro of mouse macrophages and fibroblasts with Mycoplasma pulmonis. The mycoplasmas attached to the cells and under appropriate cultural conditions grew into a lawn of microorganisms covering most of the cell surface. The mycoplasmas grew abundantly on fibroblasts cultured in minimal essential medium containing 20% fetal calf serum; supplementation of this medium with heart infusion broth was necessary to obtain similar growth on macrophages. The infection of these cells appeared to be essentially an extracellular process; only rarely were partially degraded mycoplasmas seen with phagocytic vacuoles. The addition to heavily infected macrophage cultures of low concentrations of anti-mycoplasma antibody stimulated rapid, massive phagocytosis of the surface microorganisms. In sharp contrast, the same antiserum had no discernable effect on the mycoplasma-fibroblast relationship. The antibody effect in the macrophage system was apparently a direct opsonic one rather than an indirect result of microbial killing, since the mycoplasmas in macrophage or fibroblast cultures incorporated labelled thymidine into DNA after the addition of antiserum to the medium. The phagocytic event and the subsequent fate of the mycoplasmas were studied in detail after the addition of antibody to the macrophage cultures. Phase-contrast cinemicrophotography revealed membrane ruffles surrounding the surface mycoplasmas and disappearance from view of the organisms; 10–30 min later translucent grapelike clusters were seen in large phagocytic vacuoles. On electronmicroscopic study the surface mycoplasmas were surrounded by pincers-like projections of the macrophage. Numerous mycoplasmas were seen in phagocytic vacuoles; in the early minutes after the addition of antibody the intracellular mycoplasmas appeared normal, but within 2 hr they appeared partially degraded with a central electron-lucent area and electron-opaque deposits at the microbial cell margin. 24 hr after the addition of antiserum, digestion of the mycoplasmas was nearly complete; the cells appeared normal except for large residual bodies composed of amorphous moderately dense material and increased lipid deposits. Degradation of mycoplasmas within macrophages was also studied using infected cultures in which the mycoplasmas, but not the macrophages, had incorporated tritiated thymidine into DNA. The appearance of large amounts of acid-soluble radiolabel after phagocytosis stimulated by antibody confirmed the degradation of the intracellular mycoplasmas.     

SYNERGISTIC INTERACTION OF MACROPHAGES AND LYMPHOCYTES IN ANTIGEN-INDUCED TRANSFORMATION OF LYMPHOCYTES

The role of macrophages and lymphocytes in antigen-induced transformation of lymphocytes has been investigated. Lymphocytes and macrophages were obtained from inbred strain 13 guinea pigs which were either unimmunized or immunized with complete Freund's adjuvant (CFA) or tetanus toxoid in CFA. The transformation response to PPD or tetanus toxoid was assayed by tritiated thymidine incorporation. Addition of macrophages to immune lymphocytes significantly increased their response to purified protein derivative (PPD) or tetanus toxoid. This was observed if the macrophages were (a) "immune" or "nonimmune", (b) unirradiated or irradiated (3000 R), (c) 99% pure, and (d) peritoneal or alveolar in origin. Neither immune nor nonimmune macrophages were able to induce nonimmune lymphocytes to respond to PPD or tetanus toxoid. When macrophages were incubated with PPD or tetanus toxoid and then washed, they stimulated immune lymphocytes to transform. An incubation time of ½ hr was adequate, however, 2–4: hr was optimal. These studies indicate (a) that antigen-induced transformation of lymphocytes is greatly enhanced by macrophages; (b) that macrophage-antigen interaction can antecede lymphocyte-antigen interaction and results in macrophages which are able to stimulate lymphocyte transformation; and (c) that the immunological memory requisite to elicit specific transformation responses is a property of the lymphocyte and not the macrophage.     

ANTIGENIC DIFFERENCES BETWEEN MATT HEMOLYTIC STREPTOCOCCI AND THEIR GLOSSY VARIANTS

The matt and the glossy forms of four strains of hemolytic streptococci were used to immunize rabbits. Precipitin tests showed that rabbit sera prepared against matt organisms, whether virulent or avirulent for mice, contained type-specific antibody while sera prepared against completely degraded glossy organisms contained no type-specific antibody. Type-specific antibody was removed from the sera by absorption with homologous matt organisms but was unaffected by absorption with homologous glossy organisms. Passive protection experiments on mice showed that anti-matt sera were protective and anti-glossy sera non-protective against infection with homologous virulent organisms. Vaccination of mice with matt organisms rendered them immune to subsequent infection with homologous virulent cultures; but vaccination with glossy organisms established no active immunity.     

Bacille Calmette-Guerin infection in the mouse. Regulation of macrophage plasminogen activator by T lymphocytes and specific antigen

High levels of plasminogen activator (PA) were induced in mouse peritoneal macrophages by infection with BCG, 2-6 X 10(7) viable organisms intravenously, followed 3-4 wk later by intraperitoneal challenge with purified protein derivative (PPD) 2 days before harvest. Macrophages obtained from infected animal without boosting showed little fibrinolytic activity, but challenge of Bacille-Calmette-Guèrin (BCG)-primed peritoneal cells with PPD in culture also enhanced macrophage PA 4- to 10-fold. Stimulation of macrophage PA by PPD depended on specifically sensitized thymus-derived (T) lymphocytes because it was abolished by pretreatment of BCG-primed peritoneal cells with anti-thy 1.2 antiserum and complement. A direct assay was developed in which nylon wool separated sensitized lymphocytes and PPD induced PA in macrophages from uninfected animals under defined conditions on 125I-fibrin. Enhanced macrophage fibrinolysis was proportional to concentration of PPD and the number of sensitized lymphocytes transferred. An indirect two-stage assay was also used to show that BCG-sensitized peritoneal cells released a soluble inducer of macrophage PA into the culture medium, after challenge with PPD. Induction of macrophage PA by PPD challenge in vitro made it possible to study the generation and activity of sensitized peritoneal lymphocytes at different stages of infection. Our results show that nonadherent peritoneal cells of BCG-infected mice provide a rich source of specifically sensitized lymphocytes and that macrophage activation is limited by continued availability of antigen, as well as sensitized lymphocytes. Induction of macrophage PA provides a sensitive, versatile, and rapid in vitro assay to study the role of lymphocytes and specific antigen in macrophage activation by BCG.     

IN VITRO STUDIES OF THE SUPPRESSION OF DELAYED HYPERSENSITIVITY BY THE INDUCTION OF PARTIAL TOLERANCE

Suppression of delayed hypersensitivity in vivo is correlated in vitro with the absence of macrophage migration inhibition in the presence of the antigen used to induce partial tolerance. The suppression of delayed hypersensitivity is antigen-specific in vivo as well as in vitro. The lymphocytes, and not the macrophages, are the cells involved in the induction of tolerance in terms of delayed hypersensitivity which is characterized by an absence of migratory factor activity.     

THE SPECIFICITY OF ALLERGIC REACTIONS : I. DELAYED VERSUS ARTHUS HYPERSENSITIVITY

Guinea pigs sensitized with either hen, duck, or goose egg albumin showed delayed hypersensitivity followed by Arthus reactions to the homologous antigen, but tended to have much weaker delayed responses and slower antibody formation to heterologous antigens. Guinea pigs with delayed hypersensitivity to one of the avian antigens had a slower antibody response to a secondary injection of heterologous antigen than to one of the homologous antigen. Sensitization with a protein conjugated with a hapten such as picryl chloride (Pi) or dinitrofluorobenzene (DFB) resulted in delayed hypersensitivity to the homologous conjugate, the homologous protein, and the homologous protein with a heterologous hapten. Circulating antibody and Arthus reactions occurred subsequently to the homologous conjugate, as well as to the homologous hapten attached to a heterologous protein. Delayed hypersensitivity thus seemed associated with the protein moiety, and Arthus responses with the hapten. Anamnestic responses followed injection of an antigen causing delayed hypersensitivity, but not of a hapten not causing delayed reactions. Thus, animals sensitized initially with Pi·HEA, DFB·HEA, or HEA produced antibodies sooner after a secondary injection of Pi·HEA than did unsensitized animals. No anamnestic response resulted when animals sensitized to Pi·BGG were injected with Pi·HEA. Thus, delayed hypersensitivity is indicated to be a preliminary and immature step in the immune process, with specificity directed against broad, more general features of the protein antigen. This intermediate step is followed by production of circulating antibody to any antigen having a similar basic structure, with the specificity of the antibody also directed against smaller immunologically active sites on the antigen molecule.     

Mode of immunopotentiating action of BCG: macrophage activation produced by BCG-infection.

Macrophage activation as measured by increased rate of carbon clearance and spreading of peritoneal macrophage was studied in mice infected with BCG, strain Japan. BCG caused marked increase of the numbers of peritoneal cells and spread macrophages. The increases of spread macrophages reached a peak at the 3rd week of BCG infection introduced by the both routes of intravenous(i.v.) injection and foot pad(f.p.) injection. BCG also enhanced the clearance of carbon. In the case of BCG given i.v., the increase of the rate of carbon clearance was biphasic : an early increase reaching maximum at the 1st week and a late increase reaching maximum at the 3rd week of BCG infection. When BCG given into one foot pad, peak increase was reached at the 5th week. The activation of macrophages as measured by increased levels of carbon clearance and increased numbers of spread macrophages in the mice receiving BCG i.v. was approximately two fold greater than that in the mice receiving BCG by f.p. route. When sheep red blood cells (SRBC) as antigen were injected i.v. into the mice primed with BCG i.v., the optimal interval between BCG priming and subsequent antigen injection varied with the dose of antigen for the induction of the highest level of delayed type hypersensitivity (DTH) to SRBC, but not with the degree of macrophage activation.     

HEMATOPOIETIC ORIGIN OF MACROPHAGES AS STUDIED BY CHROMOSOME MARKERS IN MICE

The origin of macrophages was studied in mouse radiation chimeras by chromosome marker technique. Macrophage cultures were established from peritoneal exudate, from lung washings, and from organ cultures of bone marrow, spleen, lymph node, thymus, and lung. Cultured macrophages were induced to divide by adding conditioned medium from L cell cultures. In chimeras which were lethally irradiated and given injections of bone marrow or spleen cells, dividing macrophages were of donor type, independent of the source of the macrophages. When chimeras were established by injections of a mixture of bone marrow cells and cells from other hematopoietic tissues of two genetically different donors, the ratio of cells with different genotypes was approximately the same in bone marrow cells and in macrophage cultures. Thymus, lymph node, and peritoneal exudate cells were not found to contain precursor cells for macrophages. Precursor cells for macrophages and for bone marrow cells appeared to be equally sensitive to sublethal irradiation. The results indicate that macrophages from different sources can all be derived from hematopoietic tissues, and suggest that only hematopoietic tissues contain precursor cells for macrophages which are capable of in vitro division. The close relationship between the source of cells in bone marrow and in macrophage cultures suggests that, at the maturation level at which the irradiated host is repopulated, the precursor or stem cells for macrophages may be identical with those for myeloid and erythroid series of cells.     

Antimicrobial activities of benzoxazinorifamycin (KRM-1648) and clarithromycin against Mycobacterium avium-intracellulare complex within murine peritoneal macrophages, human macrophage-like cells and human alveolar epithelial cells

Profiles of the in-vitro antimicrobial effects of KRM-1648 and clarithromycin against Mycobacterium avium-intracellulare complex (MAC) within murine peritoneal macrophages, the THP-1 human macrophage cell line and the A-549 type II alveolar epithelial cell line were determined. MAC organisms grew more rapidly in A-549 and THP-1 cells than in murine peritoneal macrophages. Peritoneal macrophages produced significant amounts of reactive nitrogen intermediates in response to MAC infection, but A-549 and THP-1 cells did not. KRM-1648 progressively killed M. intracellulare residing in macrophages, but did not completely eliminate M. intracellulare from A-549 and THP-1 cells. Moreover, in the case of M. intracellulare-infected A-549 and THP-1 cells, bacterial regrowth was observed during the middle to late phase of cultivation. Clarithromycin exhibited moderate levels of microbicidal activity against M. intracellulare residing in peritoneal macrophages, THP-1 cells and A-549 cells. The profiles of clarithromycin-mediated killing or inhibition of the intracellular organisms in A-549 and THP-1 cells were similar to those observed for organisms within peritoneal macrophages.     

The influence of erythrophagocytosis on the interaction of macrophages and salmonella in vitro

Phagocytosis and killing of Salmonella typhimurium by mouse peritoneal macrophages was inhibited when the bacteria and antibody-coated homologous erythrocytes or heterologous erythrocytes were simultaneously exposed to macrophages in vitro. No inhibition of phagocytosis or killing was observed in experiments employing uncoated or disrupted antibody-coated homologous erythrocytes. Degradation of S. typhimurium as measured by the loss of fluorescence from intracellular salmonella coated with fluorescein-labeled antibody was inhibited in macrophages which had previously ingested antibody-coated homologous erythrocytes. Anti-mouse-erythrocyte serum was found to have a cytotoxic action on mouse macrophages. However, the viability of macrophages was not altered by phagocytosis of antibody-coated homologous erythrocytes or uncoated heterologous erythrocytes.     

Contact sensitivity in the mouse. IV. The role of lymphocytes and macrophages in passive transfer and the mechanism of their interaction

Contact sensitivity skin reactions were produced in mice by immunization with 2-phenyl-4-ethoxymethylene oxazolone (oxazolone) and detected by the increase in ear thickness after challenging the ears with 2% oxazolone. These skin reactions can be transferred from immunized donors to irradiated recipients by peritoneal exudate cells induced by thioglycollate. The peritoneal exudate cells were separated into purified macrophage and purified lymphocyte populations. Both cell populations transferred skin reactions. However, their time course was different. The reactions produced by lymphocytes were greater at 24 hr than at 12 hr while the reactions produced by macrophages declined slightly between 12 and 24 hr. The working hypothesis was formed that the peritoneal lymphocytes conveyed a factor (presumptive cytophilic antibody) to peritoneal macrophages which enabled them to transfer ear reactions. Experiment showed that peritoneal and lymph node lymphocytes from sensitized donors within a Millipore chamber conveyed a factor to macrophages outside the chamber which enabled them to transfer ear reactions. In contrast, peritoneal macrophages (from sensitized donors) within the chamber and peritoneal lymphocytes outside the chamber were inactive. These findings suggested that there are three modes of immunological tissue damage: hypersensitivity mediated by lymphocytes (classical delayed hypersensitivity), hypersensitivity mediated by circulating antibody (classical immediate type hypersensitivity), and hypersensitivity mediated by macrophages which have passively acquired a factor (macrophage-mediated hypersensitivity).     

Immunization of mice against African trypanosomiasis using anti- idiotypic antibodies

Anti-idiotypic (anti-Id) antibodies were raised against three protective monoclonal antibodies, each with specificity for the variable antigen type (VAT) of a clone of Trypanosoma rhodesiense. The IgG1 fractions of each were pooled and administered to BALB/c mice 3-4 wk before homologous challenge. The course of primary parasitemia was altered in 19 of 30 anti-Id-treated animals. The immunity was manifested as either: (a) complete protection, (b) reduced parasitemia, or (c) selection against parasites bearing the original VAT. The three idiotypes (Id) were found in variable levels in serum during the course of infection in control animals. However, in all anti-Id-treated mice that displayed immunity, one Id in particular (7H11) was detectable much earlier in infection and in higher levels than in control mice or anti-Id-treated, nonimmune mice. Six of nine mice treated with the anti- 7H11 Id alone also displayed immunity, manifested in this case exclusively as selection against parasites bearing the original VAT. The effect was again associated with the more rapid appearance of the Id after infection. Specificity of the anti-Id-induced immunity was supported by the failure of anti-7H11 Id treatment to alter the course of infection with a heterologous clone of T. rhodesiense. To our knowledge, this is the first report of the antigen-independent induction of antimicrobial immunity using anti-Id antibodies.     

Species-related innate resistance to Schistosoma mansoni. Role of mononuclear phagocytes in schistosomula killing in vitro.

Resistance to infection with the multicellular parasite Schistosoma mansoni has been previously demonstrated to vary among several host species. The current investigation was designed to examine the basis for this species-related resistance in vitro. Adherent peritoneal macrophages or peripheral blood mononuclear cells from several species of host animals were incubated with S. mansoni schistosomula for 18-24 h; parasite viability was then assayed by methylene blue exclusion. Peritoneal exudate macrophages from susceptible species, such as mice (C57Bl/6) and hamsters killed, respectively, 6.6 +/- 2 and 8.0 +/- 2% of incubated schistosomula. In contrast, cells from resistant species: rats, guinea pigs, and rabbits, killed 21 +/- 2.3, 15 +/- 4.6, and 17 +/- 5.5%, respectively. Furthermore, blood monocytes from rabbits resulted in a mean of 25.9 +/- 2.8% dead organisms. Schistosomula killing by mononuclear phagocytes obtained from resistant species (rats or rabbits) was dependent on the cell/parasite ratio. Significant schistosomula mortality resulted from culture supernatants of rat macrophages or rabbit monocytes. Killing by cells from both species was significantly reduced upon addition of L-arginine, while catalase reduced killing only by rat macrophages. We conclude that mononuclear phagocytes may play a key role in species-related innate resistance to schistosomiasis; their in vitro schistosomulicidal activity parallels the known in vivo susceptibility of the donor species. Killing is mediated by lysosomal enzymes (arginase) and by products of oxidative metabolism, the predominant mechanism depends on the specific animal species.     

A COMPARATIVE STUDY OF EXPERIMENTAL PROPHYLACTIC INOCULATION AGAINST LEPTOSPIRA ICTEROH?MORRHAGI?

Guinea pigs were inoculated with suspensions of Leptospira icterohæmorrhagiæ obtained from pure cultures of several different strains, in order to determine whether or not an active immunity against a subsequent infection with virulent organisms would develop in the vaccinated animals. The experiments were so arranged as to make possible a determination of the existence of immunity against homologous strains as well as against the strains not employed as vaccine, and a brief quantitive estimation of the degree and duration of the immunity in relation to the quantities of the vaccines inoculated. Following the general rule of prophylactic inoculations with various pathogenic organisms, the inoculations were repeated subcutaneously on three consecutive occasions at intervals of 5 days. With respect to the amounts of vaccine, the experiments were divided into three groups for each vaccine, one group receiving three doses of 0.5 cc., the second three of 0.05 cc., and the third three of 0.005 cc. Four different strains were employed as vaccines, American Strain 1, American Strain 2, and one each of the Japanese and the European strains. The determination of the development, degree, and duration of the immunity was made by inoculating intraperitoneally several minimum lethal doses of each of the five following strains: American Strains 1, 2, and 3, the Japanese, and the European strains. The virulence of the different strains varied considerably, the strongest being the Japanese strain, which killed the guinea pig in a dose of 0.00001 cc., and the weakest American Strain 3, the minimum lethal dose of which was as large as 0.01 cc. The vaccinated guinea pigs were tested for immunity at the end of 2, 4, and 8 weeks after the last inoculation. The results obtained show that three successive inoculations of 0.5 cc. of the emulsions of killed cultures of Leptospira icterohæmorrhagiæ into guinea pigs rendered them completely resistant to a subsequent infection with the virulent cultures of both homologous and heterologous strains. With 0.05 cc. the protection was not so general, the animals succumbing to an experimental infection with some heterologous strains while resisting the homologous and other heterologous strains. The animals which were vaccinated with 0.005 cc. survived the infection experiments with the homologous strains in the case of American Strain 1 and the Japanese strain, but they were not protected against any other strains. The vaccines of other strains were unable to immunize the guinea pigs so highly even against their homologous strains, when the amount of each inoculation was only 0.005 cc., but 0.05 cc. conferred complete protection against the same strains. It may be concluded, therefore, that when a sufficient quantity of killed cultures of Leptospira icterohæmorrhagiæ is given, the guinea pigs will become immune to all strains of the same organism, but that smaller quantities may protect them against homologous but not against heterologous strains. A close analysis reveals the existence of group or type affinities among different strains which can be brought '' out by immunizing the animals with smaller quantities of killed cultures. In the present series of experiments American Strains 1 and 3 form one group, American Strain 2 and the European strain another, and the Japanese strain a third, which is also closely allied to the first group. In order to insure universal immunity it is wise to employ as many group or type cultures as possible in the preparation of vaccines, a polyvalent vaccine being recommended. It is not improbable that the strain recently encountered in Lorient, France, is an unusually deviated type of Leptospira icterohæmorrhagiæ, and that if successfully cultivated and used as vaccine in sufficient amount it might protect the animals against other strains of the same organism. The active immunity induced in the vaccinated guinea pigs was found to persist for at least 8 weeks after the last inoculation. It will no doubt last for a much longer period.