In order to study immune responses in mice to individual Msgs, we expressed 2 recombinant Msg variants, 107 and 119, which differed from each other by 11%, in E. coli. Because of difficulties in expressing the entire Msg as a single protein, we expressed each variant in two overlapping fragments, corresponding approximately to the amino and carboxyl halves of the full-length protein. In addition, we expressed each fragment in multiple vectors, one of which was used for immunization studies, and another for in vitro assays. This eliminated cross-reactive immune responses to the vector-encoded portion of the recombinant protein which we identified in preliminary studies.
We utilized proliferation to examine cellular T cell responses to Pneumocystis antigens; this will primarily represent CD4+ T cell responses to protein (vs. peptide) antigens. In a subset of animals we also examined culture supernatant cytokine levels following antigenic stimulation. Attempts to detect intracellular cytokine responses, including interferon-gamma, interleukin-4, interleukin-17, and tumor necrosis factor-alpha, by flow cytometry were unsuccessful despite numerous attempts using a variety of Pneumocystis antigens.
We initially immunized separate sets of 4 mice 1, 2, or 4 times with the amino fragment of each of the two recombinant Msg variants 107 and 119 (20
μg/immunization), because there was less conservation in the amino (86%) compared to the carboxyl (91%) fragments. We examined proliferative and antibody responses to both variants; results are shown in Figure . Splenocytes from one of 4 mice immunized with a single dose of Msg119Am proliferated in response only to the immunizing antigen. No proliferation was seen to either the immunizing or the non-immunizing variant fragment following 2 immunizations with either amino fragment. Following the 4th
immunization, proliferative responses to the immunizing antigen were seen in splenocytes from 6 of 8 animals, while proliferation to the non-immunizing Msg amino fragment was seen in splenocytes from 5 of these 8 animals. In contrast, antibody responses as measured by ELISA were seen in 5 of the 8 animals to both fragments following the first immunization, and in 16 of 16 animals to both fragments following 2 or 4 immunizations; optical densities were similar to both fragments regardless of whether 107 or 119 was used for immunization.
Antibody responses to crude P. murina
antigens were seen only in animals with antibody responses to the Msg variants (20 of 24 animals). Thus, cross-reactive proliferative responses were seen only following multiple immunizations, while cross-reactive antibody responses were seen after a single immunization. No immune responses were seen to the recombinant Msg fragments or crude P. murina
antigens in control animals immunized with adjuvant alone (data not shown).
Figure 2 A. Cell proliferation following immunization with recombinantP. murinaMsg proteins MSG107Am and MSG119Am. Splenocytes from animals that were immunized with these recombinant Msg proteins were cultured in triplicate with each of the recombinant Msg proteins (more ...)
To verify that the antibody responses to recombinant proteins recognized native Msg, we utilized western blots with either crude Pneumocystis antigens or purified native Msg as the antigen. Using sera from animals immunized with the amino recombinant fragments, we were able to demonstrate reactivity with Msg in both antigen preparations in 2 of 2 animals tested. Figure shows results using sera from animals immunized with 107Am or 119Am.
Figure 3 Western blot analysis of antibody responses generated following immunization with recombinant Msg. Panel A shows a Coomassie blue-stained gel and panel B shows western blots using sera from 2 immunized mice and 1 control mouse. For both the gel and the (more ...)
We next examined proliferative and antibody responses following immunization with crude P. murina
antigens. Msg is the most abundant protein in Pneumocystis
extracts, and in immunosuppressed animals, which are the source of this Pneumocystis
, multiple Msg variants are expressed [9
]. Splenocytes from immunized animals (n
5) did not proliferate in response to challenge with either the amino or the carboxyl fragment of either MSG107 or MSG119, but did proliferate (5 of 5 animals) to crude P. murina
antigens, which were used to immunize the animals (Figure ). Proliferative responses were also seen to purified native Msg in splenocytes from immunized animals (2 of 2) but not in a control animal. In contrast to this, antibody responses were seen to both amino (4 of 5 for 107Am and 3 of 3 for 119Am) and carboxyl fragments of MSG107 and MSG119, as well as to crude P. murina
antigens, in all 5 animals. Antibody responses were also seen to purified native Msg (2 of 2). It is noteworthy that the antibody responses were greater (higher OD) to both carboxyl fragments, which are more highly conserved, than to the amino fragments. The antibody responses demonstrate that the animals were exposed to, and were able to mount an immune response to, Msg in the crude antigen preparations.
Figure 4 A. Cell proliferation following immunization with crudePneumocystisantigens. Splenocytes from animals that were immunized four to five times with crude antigens were cultured in triplicate with recombinant P. murina Msg proteins (n=5 (more ...)
Because immunization with either recombinant or crude antigen preparations presents an artificial encounter with Pneumocystis
antigens, we wanted to examine immune responses following natural infection. Since immunosuppressed animals have inadequate immune responses to Pneumocystis
, which allow the organism to replicate in an uncontrolled or poorly controlled manner, we utilized a model in which healthy animals develop Pneumocystis
infection following exposure to an already infected seeder animal. In this model, infection peaks at ~5-6
weeks following exposure, and is typically cleared by 10
]. Splenocytes from exposed mice proliferated in response to crude Pneumocystis
antigens in 7 of 9 animals and to native Msg in 2 of 2 animals, but none of the animals showed proliferation to any of the 4 recombinant Msg protein fragments (Figure ). Antibody responses were again seen to the carboxyl recombinant Msg fragments (9 of 9 and 8 of 9 for MSG107 and MSG119, respectively) as well as to crude Pneumocystis
antigen and native Msg (8 of 9 for each), but not to the amino fragments (0 of 9 and 1 of 7 for MSG107 and MSG119, respectively), again indicating that these naturally infected animals were exposed to Msg.
Figure 5 A. Cell proliferation followingPneumocystisinfection. Blood and spleen cells were obtained 10 to 12weeks after initial exposure to Pneumocystis-infected seeders, a time by which Pneumocystis infection has typically developed and been cleared (more ...)
As an additional marker of CD4 responses, we examined in vitro cytokine production. Among the cytokines examined, IL-17 and MCP-3 were consistently secreted by splenocytes from animals infected with P. murina when cultured with P. murina Ag or native Msg antigen, but not with the recombinant Msg variants (Figure ). These responses represented memory responses, as no cytokine production was seen in cells from uninfected control animals cultured with the same antigens. Among the other cytokines, results with crude antigen or native Msg were variably positive (data not shown) but in no case were they produced in response to the recombinant antigens.
Figure 6 Analysis of cytokines secreted by splenocytes fromPneumocystis-infected animals. Splenocytes were obtained 10 to 12weeks after exposure, by which time Pneumocystis infection had typically cleared, and were cultured with recombinant Msg proteins, (more ...)
Serologic studies have found that most humans have been exposed to Pneumocystis
at a young age and have detectable antibodies to Msg. To see if PBMCs from healthy humans can proliferate to recombinant Msg, cells from 5 volunteers were cultured with HuMSG14Am, HuMSG14Ca, concanavalin A or tetanus toxoid. No patient demonstrated proliferation to either HuMSG fragment after 5
days, while all samples proliferated in response to concanavalin A or tetanus toxoid. Moreover, sera from 4 of the 5 volunteers were positive in an ELISA utilizing HuMSG14Ca, while the 5th
volunteer had high non-specific reactivity.