Specificity of Acquired Resistance Produced by Immunization with Mycobacterial Cells and Mycobacterial Fractions

Mice were immunized intraperitoneally with 5.0 mg of living and 5.0 mg of heat-killed H37Ra cells of the attenuated strain Mycobacterium tuberculosis and challenged intraperitoneally with Listeria monocytogenes and Klebsiella pneumoniae. The period of protection provided by the living and heat-killed H37Ra cells against both heterologous infections was the same. When mice were immunized intraperitoneally with graded doses of living and heat-killed H37Ra and challenged intraperitoneally with listeria or klebsiella, the lowest immunizing dose providing protection against both klebsiella and listeria challenge was the same for living and heat-killed cells. Living and heat-killed cells also immunized equally effectively when the routes of immunization and challenge were different. Mice also were immunized intraperitoneally with mycobacterial ribosomal fraction, mycobacterial cell walls, and several nonspecific agents (Escherichia coli endotoxin, mineral oil emulsion, and Freund's incomplete adjuvant). The mice were challenged intraperitoneally with listeria or klebsiella at varying times after immunization. The mycobacterial components and all the nonspecific agents provided transitory protection lasting no longer than 4 days after immunization. Only the mycobacterial cell walls and the endotoxin provided protection against listeria challenge. It was concluded that the protection provided by the mycobacterial ribosomal fraction is specific for tuberculosis infection, since this fraction provided no protection against listeria infection and only transitory protection against klebsiella. It was also concluded that the mycobacterial component providing protection against heterologous infections is heat stable and probably is found in the cell wall.     

Specificity of Acquired Resistance Produced by Immunization with Listeria monocytogenes and Listeria Fractions

Mice were immunized with 1.0 mg of an attenuated strain of Listeria monocytogenes to determine the period of protection afforded by this strain when the mice were challenged intravenously with 5 MLD of listeria. Protection appeared 2 days after immunization and was still apparent 4 weeks after immunization. If the challenge dose was decreased to 1 MLD, protection was apparent at 10 weeks. Mice immunized with a comparable dose of mycobacterial cells and challenged intravenously with 1 MLD of listeria showed no protection at 10 weeks. The magnitude of the immune response to listeria challenge was not increased in mice immunized with the same virulent strain as that used for challenge. It was also found that resistance to listeria challenge appeared early after listeria immunization if the immunizing dose was large. As the immunizing dose was decreased and the challenge dose increased, resistance appeared later. Listeria killed by heat or ultraviolet irradiation, living but nonmultiplying streptomycin-dependent listeria, or listeria ribosomal fraction gave no protection against listeria challenge. The magnitude of the immune responses after listeria immunization to listeria challenge and to mycobacteria challenge were compared. It was found that protection after listeria challenge was of longer duration. In addition, a 100-fold larger vaccinating dose was required to give comparable protection against tuberculous infection.     

Interference with tolerance induction in vivo by inhibitors of prostaglandin synthesis

Prostaglandins (PG) have been implicated as modulators of both humoral and cellular immune responses. In order to evaluate a possible role for PG in tolerance, the effect of inhibitors of prostaglandin synthesis on tolerance induction and circumvention has been investigated. Injection of deaggregated human gamma-globulin (DHGG) into A/J mice leads to unresponsiveness to a subsequent challenge with immunogenic aggregated human gamma-globulin (AHGG). Administration of indomethacin (IM) or acetylsalicylic acid (ASA) shortly before and after DHGG injection prevents tolerance induction. PGE2 reverses the tolerance overriding effect provided by IM. IM is not able to overcome unresponsiveness when given 10 and 20 days after tolerance induction, at a time point when both T and B lymphocytes are tolerant. As previously shown, lipopolysaccharide (LPS) both inhibits the induction of tolerance to HGG and circumvents tolerant T helper cells late in tolerance when competent B cells are present. In contrast, IM is unable to circumvent T-helper cell tolerance when given at Day 60 after tolerogen, when B cells (but not T cells) are responsive. Furthermore, LPS acts as an adjuvant, B-cell mitogens, inducer of polyclonal Ig secretion, and primes mice when given with tolerogen, while IM has none of these properties. These results indicate a difference between the effects of IM and LPS on tolerance and a possible role of PG in DHGG-mediated tolerance induction.     

Enhancement of the in vivo antibody response by an 8-derivatized guanine nucleoside

The capacity of the C8-substituted guanine ribonucleosides to enhance the in vivo humoral immune response to the protein antigen, human gamma globulin (HGG), in A/J mice was evaluated. It has been shown recently that the C8-substituted guanine ribonucleosides are a new class of potent adjuvant for humoral immune responses to the sheep erythrocyte antigen. The current studies extend these findings to HGG with 8-bromoguanosine (8BrGuo), a representative of this group of nucleosides. The adjuvant activity of 8BrGuo in this system is highly dose and time dependent. Although 8BrGuo enhanced responses when injected either early (Day 0) or late (Day 4 or 5) after immunization, its administration on Day 1 or 2 most often led to no enhancement, suggesting that 8BrGuo may act on two events separated by a resistant stage in an ongoing immune response. The plaque-forming cell (PFC) response to HGG was enhanced optimally at doses as low as 1 mg 8BrGuo/mouse administered either on the day of immunization or 4 days thereafter. In contrast, however, serum anti-HGG antibody concentration assayed by enzyme-linked immunosorbent assay (ELISA) was enhanced only at doses of 10 mg or more, injected on the day of immunization, but doses as low as 1 mg were effective on Day 4. 8BrGuo was also an effective adjuvant when injected after antigen administration in incomplete Freund's adjuvant or when administered by several different routes (intraperitoneal, subcutaneous, oral).     

The induction of peripheral T cell unresponsiveness in adult mice by monomeric human gamma-globulin

Monomeric human gamma-globulin (HGG), when injected into adult mice, induces a state of specific immunologic unresponsiveness to further challenge with immunogenic forms of HGG. In this report we have directly determined the role of the thymus in the induction of HGG tolerance and the proliferative responsiveness of T cells from normal and HGG-tolerant mice. Draining lymph node T cells were isolated from HGG-tolerized and -challenged mice, and tested for their proliferative response to HGG in vitro. T cells from untreated but challenged adult CBA/CaJ and A/J mice proliferate in response to HGG, whereas such mice given monomeric HGG before challenge fail to show an HGG-specific proliferative response. APC from tolerant or nontolerant mice were equally effective in the support of Ag-specific proliferation of primed T cells. The influence of the thymus gland on HGG-induced T cell unresponsiveness was assessed by determining whether thymectomized mice could be tolerized to HGG. The results suggest that the generation of T cell tolerance to HGG is independent of thymic function as assayed by both antibody production in vivo and T cell proliferation in vitro. Unresponsiveness of T cells from tolerant mice was not a result of the presence of CD8+ cells since removal of CD8+ cells from lymph node T cells did not alter unresponsiveness to HGG in vitro. Further, mixing tolerant T cells with normal HGG-primed T lymphocytes did not inhibit proliferation of the HGG-primed cells. The results of this investigation suggest that this mouse model of tolerance to HGG represents a thymus-independent unresponsiveness of mature peripheral T cells to a nonself-Ag. Understanding the regulation of tolerance to HGG may give additional insight into the mechanisms required for the maintenance and possibly the induction of tolerance to certain self-Ag in peripheral lymphoid organs.     

Single immunization with MF59-adjuvanted inactivated whole-virion H7N9 influenza vaccine provides early protection against H7N9 virus challenge in mice

H7N9 influenza infection in humans would result in severe respiratory illness. Vaccination is the best way to prevent influenza virus. In this paper, we investigated the effect of early protection provided by inactivated whole-virion H7N9 influenza vaccine in a mouse model.Mice were immunized intramuscularly once with different doses of inactivated whole-virion H7N9 influenza vaccine alone or in combination with MF59 adjuvant. Specific IgM and IgG antibody titers in sera of mice were detected by ELISA 3, 5 and 7days after immunization. To evaluate the early protection provided by the vaccine, mice were challenged with lethal dose (40LD₅₀) of homologous virus 3, 5 and 7 days after immunization respectively. The survival rate and body weight change of mice during 21 days after challenge and the residual lung virus titer on 3rd day after challenge were determined. The results demonstrated that mice could obtain effective protection 3 days after immunization with the vaccine at a high dose, and 5–7 days after immunization even at a low dose. Thus early immune responses induced by inactivated whole-virion H7N9 vaccine could provide effective protection.     

Suppressor cells in tolerance to HGG: kinetics and cross-suppression in high dose tolerance--absence in low dose tolerance.

Spleen cells from mice made tolerant with high doses of human gamma-globulin (HGG) specifically suppress the immune response of normal, syngeneic, spleen cells. These suppressor cells were found to be cross-reactive in that they would suppress the immune response of normal spleen cells to bovine gamma-globulin (BGG) as well as to HGG. In contrast, suppressor cells could not be demonstrated in spleens of mice made tolerant with low doses of HGG (i.e., T-cell tolerance), nor could they be found in high dose tolerant mice following a second injection of DHGG at a time when the initial suppressor activity had waned. The role of suppressor cells in the induction, maintenance, and loss of tolerance is discussed.     

Studies on the resistance to tolerance induction against human IgG in DDD mice. I. Organ differences of tolerogen susceptibility and cellular sites responsible for the resistance.

Cellular sites of the tolerogen resistance in DDD mice against human IgG (HGG) were examined by reconstitution experiments in which cells of various lymphoid organs from tolerized mice were transferred into lethally irradiated syngeneic recipients with or without the supplement of an excess number of untreated T or B cells. It was shown that T cells but not B cells in the spleen and bone marrow-locating B cells were tolerogen resistant. Kinetic profiles of tolerance induction were compared among thymus, lymph node, and spleen T cells. Thymus cells fall into unresponsive state as early as 2 days after the tolerogen (tHGG) injection when only partial tolerance was observed in lymph node T cells. By 1 week of tolerogen treatment, the tolerant state was completed in both thymus cells and lymph node T cells, while spleen T cells showed marked resistance. Tolerance induced in thymus cells and spleen T cells was of relatively short duration and responsiveness was completely recovered by 5 weeks after the injection of tHGG. At this time lymph node T cells still showed hyporesponsiveness. The differences in tolerance inducibility were also shown among different lymphoid organs in tolerogen dose response. Lymph node T cells were very sensitive to tolerance induction, giving no response even by the injection of 0.01 mg of tHGG. Thymus cells were much less sensitive with the gradual loss of responsiveness by increasing the amount of tHGG. In contrast, spleen T cells showed gradual resistance with increasing amount of tHGG, indicating that some positive response was evoked in spleen T cells by a relatively high dose of tHGG. These results seem to suggest that the tolerogen resistance of spleen T cells may be due to their capability of showing positive response against the tolerogenic material. This was also suggested by the fact that the treatment with cyclophosphamide following the tolerogen injection diminished completely the responsiveness against the subsequent challenge immunization.     

Immunizing Capacity of Viable and Killed Attenuated Mycobacterial Cells Against Experimental Tuberculous Infection

The relationship of the dose of vaccine to the immune response was determined in CF-1 mice vaccinated intraperitoneally with viable cells of the attenuated H37Ra strain of Mycobacterium tuberculosis and in mice vaccinated with cells of the same strain killed by autoclaving at 121 C for 15 min. The results showed, in terms of increased resistance to tuberculous infection, that the immune response with both living and killed cells was dependent upon the dose of vaccine, whereas only the living cells were dependent upon the time of challenge after vaccination. The dose response curves show dramatically that viable cells, which do not multiply in vivo, are several hundred times more effective immunizing agents against tuberculous infection than are autoclaved cells. Viable 2-week-old H37Ra cells were far more immunogenic than viable 4-week-old cells. Autoclaved 2-week-old cells, however, were no more immunogenic than autoclaved 4-week-old cells. H37Ra cells killed by boiling (98 C), exposure to 65 C for 30 min, treating with 2% phenol, or by being dried with acetone also lost most of their capacity to immunize mice. The effect of adjuvant on the immune response of mice to tuberculous infection was tested by incorporating both viable and autoclaved cells in Freund's incomplete adjuvant. We found that this vehicle had little or no effect on the immunizing capacity of either viable or heat-killed mycobacterial cells. The relationship of all the findings to the specificity of the immune response to tuberculosis is discussed.     

Antibody formation in young rabbits immunized with heterologous serum proteins

Rabbits were immunized with human or bovine albumin at different intervals after birth and antibody formation was studied by haemagglutination of red cells sensitized with the relevant antigen. The intraperitoneal injection of antigen in amounts of 5 mg. induced antibody formation in some litters 16–20 days after immunization, if the animals were over three days old when immunized. In younger rabbits the same dose induced tolerance. Even when different methods of enhancing the effect of the antigen (Freund’s adjuvant, Al (OH)₃, antigen-conjugated red cells, immune precipitates) or very small doses of antigen were used, antibody formation was still not detected before the 20th day of life. The use of¹³¹I-BSA did not demonstrate the immune phase of elimination of the antigen during 17 days after administration of the antigen, even in rabbits immunized 14 days after birth. The relationship of antibody formation to the induction of tolerance and the difference in the response of newborn rabbits to immunization with the different types of antigen is discussed.     

Specific suppression of the immune response by HGG tolerant spleen cells. I. Parameters affecting the level of suppression.

After adoptive transfer, the spleen cells from mice made tolerant to human gamma-globulin (HGG) specifically suppress the immune response of normal spleen cells. However, this suppressive activity in the spleen cells of tolerant mice is only present for a brief period after treatment with tolerogen. Spleen cells from animals injected 10 days earlier with tolerogen reduce the immune response of an equal number of normal spleen cells by 75%. Spleen cells from mice made tolerant 40 days previously are, however, no longer suppressive, even though they remain completely unresponsive. These data suggest that active suppression of antigen-reactive cells is not the mechanism responsible for maintaining tolerance to HGG, but rather is only transiently associated with the tolerant state. Further evidence in favor of the separation of the tolerant state from suppressive activity is that complete suppression of the normal spleen cell response requires either a high ratio of tolerant to immune competent cells or a delay in the antigenic challenge of the reconstituted recipients. By contrast, such manipulations are not required to demonstrate the complete unresponsiveness of the tolerant cells after adoptive transfer.     

Kinetics of the antibody response to type III pneumococcal polysaccharide. II. Factors influencing the serum antibody levels after immunization with an optimally immunogenic dose of antigen.

When the number of PFC present in the spleen was measured at 24-hr intervals after immunizing with an optimally immunogenic dose of type III pneumococcal polysaccharide (SSS-III), maximal numbers of PFC were attained 4 days after immunization; thereafter, the number of PFC decreased rapidly. By contrast, serum antibody levels, which were measured in the same mice using a Farr test, reached peak values 5 days after immunization and then declined much more slowly than did the number of PFC. Two factors were found to contribute to this disparity. First, experiments conducted with splenectomized mice showed that extrasplenic antibody synthesis, which began between days 3 and 4 after immunization and peaked on days 6 to 7, accounted for nearly one-third of the total amount of serum antibody produced. Second, the average rate of antibody synthesis by PFC increased through day 6 after immunization and then declined. Antigen-antibody dissociation tests showed that the avidity of the serum antibody obtained 4 to 7 days after immunization was the same. Moreover, during the same interval, all the antibody detected by the Farr test was of the IgM class. Thus, a change in avidity or class of immunoglobulin after day 5 did not account for the disparity observed. The clearance rate of antibody injected i.v. into nonimmune and immunized mice was studied. The data obtained indicated that accelerated clearance of antibody was occurring prior to day 3 after immunization; however, after day 3 the antibody clearance rate was constant and was the same as that found when antibody was injected into nonimmune mice. These findings affirmed the results of previous studies showing that treadmill neutralization was not important in determining the serum antibody levels present after immunization with an optimally immunogenic dose of SSS-III.     

Specific, transient suppression of the immune response by HGG tolerant spleen cells. II. Effector cells and target cells.

In a previous report, it was shown that spleen cells from mice made tolerant to human gamma-globulin (HGG)5 could specifically inhibit the immune response of normal spleen cells after adoptive transfer to lethally irradiated recipients. However, that report also showed that the suppressive activity was only transiently associated with tolerant spleen cell populations. It was concluded from those experiments that while suppressive activity could be demonstrated in tolerant spleen cells under certain conditions, such activity was not obligatory for the maintainance of the tolerant state. The experiments presented here were performed to determine the nature of the effector cell(s) and the target cell(s) involved in this system of suppression of the immune response. Treatment of cells from tolerant animals with anti-thymocyte serum and complement to remove thymus-derived (T) cells completely abrogated suppresive activity. Removal of adherent cells from tolerant spleen cells by passage over glass wool columns resulted in partial loss of the suppression. The inhibitory activity of the suppressor cells was resistant to 900 R irradiation regardless of whether the tolerant spleen cells were irradiated before or after adoptive transfer. The cellular target(s) for the supprssor cells was examined by using lipopolysaccharide (LPS) as an alternative source of helper activity for the response to HGG. LPS, injected at the time of the initial antigenic challenge of mice that had been reconstituted with tolerant and normal spleen cells, prevented the expression of suppression against bone marrow-derived (B) cells. However, when LPS was presented only at the time of secondary antigenic challenge, it was unable to overcome suppression of the immune response of reconstituted recipients. Thus, LPS could produce a state where the B cells were resistant to suppression, but LPS could not rescue the responsiveness of B cells once the cells in the reconstituted recipient had been suppressed. In addition, the immune response to both the hapten dinitrophenol (DNP) and the carrier (HGG) were suppressed when recipients of tolerant and normal spleen cells were challenged with DNP6HGG. This indicates that T helper cells are also a target for suppression. The results presented in this paper are discussed in relation to a possible mechanism of suppression which proposes that suppressive activity represents the induction of tolerance in immunologically competent cells by HCG which is closely associated with the tolerant spleen cells.     

Immunologic unresponsiveness after gastric administration of human gamma-globulin: antigen requirements and cellular parameters

Gastric administration of human gamma-globulin (HGG) into adult A/J mice leads to the establishment of an antigen-specific unresponsive state to subsequent parenteral challenge with HGG. An unresponsive state is induced in both helper T and B lymphocyte populations. Unresponsiveness in helper T cells is of longer duration than in B cells, lasting at least 9 wk after intragastric intubation. Adoptive cell transfer of spleen cells from gastrically inoculated mice into healthy irradiated, syngeneic recipients revealed that the unresponsive state is stable upon cell transfer and that suppressor cells are present in the spleens of gastrically tolerized mice. The establishment of HGG-specific unresponsiveness is dependent upon both the dose and the form of the antigen adminstered. Soluble and deaggregated HGG are both more efficient than is heat-aggregated HGG in inducing unresponsiveness gastrically. The administered HGG is rapidly eliminated from the animal and only a small fraction reaches the circulation as immunoreactive protein. Although the cellular parameters of the systemic unresponsiveness induced by intragastric intubation with HGG appear similar to the parameters of parenterally induced unresponsiveness, the precise mechanisms by which gastric unresponsive states are established remain to be resolved.     

Prevention by a platelet-derived factor (platelet factor 4) of induction of low dose tolerance to pneumococcal polysaccharides

It was previously shown that human or mouse serum, and platelet factor 4 (PF4) prepared from human platelet releasate, counteracts nonspecific immunosuppression induced in mice by injection of concanavalin A or syngeneic gamma-irradiated lymphoma cells. The present studies show that PF4 prepared from normal mouse or human serum by absorption to heparin-agarose and elution between 0.5 and 1.5 M NaCl is also active in this respect. The ability of PF4 to counteract antigen-specific suppression of the antibody response to pneumococcal polysaccharide (pps) was now studied. PF4 derived from human or mouse serum as well as recombinant PF4 interferes with induction of antigen-specific low dose tolerance when they are injected at the same time as a low dose (0.2 microgram) of type 14 pps 3 days before an optimal immunizing dose (25 micrograms). Furthermore, injection of platelet releasate at the time of an optimal primary immunizing dose of pps type 14 enhances the secondary response to killed bacteria injected 2 weeks later, but not the primary response itself. Both effects are interpreted as due to interference with antigen-specific suppressor cell induction during primary immunization. Injection of PF4 is much less effective in reversing low dose tolerance to an optimal immunizing dose (0.1 microgram) of type 3 pps induced by injection of 0.005 microgram of this antigen. Differences in the mechanism of tolerance induction for the two pps types that might be responsible for this are discussed.     

T cell tolerance in the chicken. I. Parameters affecting tolerance induction to human gamma-globulin in agammaglobulinemic and normal chickens.

T cell-mediated delayed hypersensitivity (DH) to human gamma-globulin (HGG) can be induced in chickens by subcutaneous injection of the antigen in complete Freund's adjuvant (CFA). In the present work, it has been demonstrated that specific tolerance of the cells mediating this DH can readily be induced in both normal and bursectomized (BX) FP strain chickens by simple i.v. injection of soluble antigen, regardless of the presence of antibody production to the tolerogen. A significant degree of tolerance at the DH and helper T cell levels could be generated in BX birds by injection of as little as 0.5 mg of HGG; such a dose only induced tolerance in normal birds when it had been previously deagregated by ultracentrifugation. Regular, nondeaggregated antigen could produce tolerance in normal animals, but only at doses of greater than 5 mg. The tolerizing injection induced a primary antibody response in normal birds in all cases, but a secondary response could not be obtained in animals rendered tolerant at the T cell level. Establishment of tolerance appeared to be very rapid, and animals remained refractory to induction of DH for at least 3 weeks after the tolerizing injection. The mode in which the antigen was presented to the animals appeared to be crucial in determining whether tolerance or sensitivity would be established.     

Genetic control of the immune response of mice to highly dinitrophenylated human gamma-globulin (DNP56HGG)

The immune response to highly dinitrophenylated human gamma-globulin (DNP56HGG) was tested in inbred strains of mice. Significant differences in the anti-DNP response among inbred strains were found, including the magnitude of serum antibody and the location of plaque-forming cells (spleen or lymph nodes). The strain differences persisted when the dose and adjuvant were changed. The genetic control of the anti-DNP response to DNP56HGG was investigated. The analysis of the response of congenic and F1 hybrid mice to DNP56HGG suggests that at least two genes are involved in the control of the anti-DNP response. The two genes are demonstrated by complementation in the F1 generation, and show no correlation with H-2 haplotype or IgG2a allotype. A third gene may be implicated by differences in response observed between male and female mice.     

Dual effect of meningococcal antigens on a T cell dependent immune response

Meningococcal antigens (MA) showed adjuvant activity when administered to mice at the same time as antigen (sheep erythrocyte (SE], by increasing the splenocyte plaque-forming response in a dose-related manner. However, when SE were given 1 day after MA administration, the subsequent plaque formation was diminished from normal in proportion to the dose of MA injected. Splenocytes taken from mice up to 5 days after MA injection actively inhibited plaque formation when mixed with splenocytes immunized with SE 4 days earlier. Two days after MA injection the nonspecific inhibition of plaque formation was mainly due to adherent spleen cells, while at 5 days nonadherent cells had acquired the inhibitory activity. It appears that it is the degree of activation of adherent cells resulting from the timing and dosage of MA which modulates the subsequent development and secretion of antibody-forming cells.     

Effect of neonatal androgenization on positive feedback in female mice

Exposure of female mice to androgens within 5 days of birth impairs fertility. Such treatment in rats results in a post-pubertal acyclic state of persistent vaginal cornification and in an inability, when ovariectomized, to show normal positive feedback on luteinizing hormone (LH) release in response to steroid challenge. In the present study, we explored whether neonatally androgenized mice demonstrate positive feedback. Female mice were administered 100 micrograms of testosterone propionate (TP) on either Day 1 (TP1) or Day 5 (TP5) after birth, or vehicle on Day 1 (SO1). Androgen-treated mice had a statistically significant advance in onset of vaginal opening as compared with vehicle-treated mice. All mice that received TP entered constant vaginal estrus, whereas those given vehicle showed variable cytology. All mice were ovariectomized at 7 wk of age and received Silastic capsules containing a priming dose of 17 beta-estradiol. When all mice were challenged 1 wk later with sequential administration of estradiol benzoate and progesterone, a significant increase in plasma LH level was present only in the vehicle-treated mice. We conclude that neonatal androgenization defeminizes the neuroendocrine mechanisms controlling gonadotropin release.     

Intramammary immunization with live-attenuated Staphylococcus aureus: microbiological and immunological studies in a mouse mastitis model

Abstract Mammary infection was induced in lactating mice by intramammary injection of Staphylococcus aureus . Histopathological analysis revealed infiltration and lesions of varying magnitude that were still apparent 21 days after the challenge. Concomitantly, viable S. aureus was recovered from infected mammary glands. Mice were immunized by the intramammary route with 5 × 10⁶ colony forming units of a temperature-sensitive mutant of S. aureus and subsequently received a boosting injection seven days later. On day 14 mice were challenged by the intramammary route with the wild-type strain. Intramammary immunization induced a significant increase in milk IgA ( P < 0.05), serum IgG ( P < 0.05) and serum IgA ( P < 0.05) on the day of the challenge, when compared with non-immunized mice. Immunization decreased significantly ( P < 0.01) the number of S. aureus colony forming units recovered 96 h after intramammary challenge. In conclusion, the feasibility of immunizing locally with temperature-sensitive S. aureus to induce immunity in the mouse mammary gland was demonstrated. The mouse model of mastitis is proposed as a useful system for screening temperature-sensitive S. aureus strains to be utilized in the development of a vaccine.