The authors have declared that no competing interests exist.
Conceived and designed the experiments: CP BB SM CDH. Performed the experiments: CP WK. Analyzed the data: CP BB. Contributed reagents/materials/analysis tools: CP WK BB SM CDH. Wrote the manuscript: CP BB WK.
Patients with sepsis are often immune suppressed, and experimental mouse models of sepsis also display this feature. However, acute sepsis in mice is also characterized by a generalized B cell activation and plasma cell differentiation, resulting in a marked increase in serum antibody concentration. Its effects on humoral memory are not clearly defined. We measured the effects of experimental sepsis on long-term immunological memory for a defined antigen: we induced colon ascendens stent peritonitis (CASP) 8 weeks after 2 rounds of immunization with ovalbumin. Four weeks later, the antigen-specific bone marrow plasma cell count had doubled in immunized non-septic animals, but remained unchanged in immunized septic animals. Sepsis also caused a decrease in antigen-specific serum antibody concentration. We conclude that sepsis weakens humoral memory by impeding the antigen-specific plasma cell pool’s development, which is not complete 8 weeks after secondary immunization.
Sepsis is still associated with astoundingly high morbidity and mortality despite improvements in intensive care [
Humoral memory consists of resting memory B cells that are rapidly activated after repeated contact with antigen, and long-lived plasma cells that reside in survival niches of the bone marrow [
Sepsis suppresses the adaptive immune system. This has been demonstrated for the priming of B cell- and T cell responses and for T cell effector functions [
Female C57BL/6 wild type mice (Charles River, Sulzfeld, Germany) were housed in a conventional, temperature-controlled animal facility with a 12-hour light/12-hour dark cycle and provided with food and water ad libitum. All experiments were performed according to the German animal safety regulations and approved by the animal ethics committee of the local animal protection authority (Regional Authority for Agriculture, Food Safety and Fishery of Mecklenburg-Vorpommern). All efforts were made to minimize suffering. Mice were immunized i.p. at 6 weeks (wk) with 100 µg trinitrophenlyl-13-ovalbumin (TNP-13-OVA, Biosearch Technologies, Inc, CA) and 50 µg ovalbumin (OVA) in alum (Pierce, Rockford, IL) and boosted 3 wk later. Eight weeks later, CASP surgery was performed as described previously [
Total serum IgM and IgG concentrations were determined using the Milliplex Mouse Immunoglobulin Isotyping Immunoassay (Millipore, Billerica, MA) according to the manufacturer’s instructions. The samples were measured with the Luminex 200 System (Bio-Rad Laboratories, Munich, Germany).
OVA-specific antibodies in serum were determined by coating MaxiSorb-microtiterplates (Nunc, Roskilde, Denmark) with OVA (10 µg/ml) and blocking with 10% FCS/PBS. Diluted serum samples were incubated and bound antibodies were detected with peroxidase-conjugated goat-anti-mouse-IgM (Dianova, Hamburg, Germany, 1:30000 in 10% FCS/PBS) or -IgG (SouthernBiotech, Birmingham, AL, 1:30000 in 10% FCS/PBS) and OptEIA™ TMB Substrate Reagent (BD Biosciences, San Jose, CA). Optical density was measured at 450 nm. In the ELISA, anti-OVA and anti-TNP-BSA antibodies were measured against a standard serum.
MultiScreenHTS-IP Filterplates (Millipore, Billerica, MA) were either coated with 100 µl OVA (50 µg/ml) or 100 µl TNP-13-BSA (50 µg/ml). Free protein binding sites were blocked with 2% FCS/PBS. In all, 1×106 splenocytes or bone marrow cells were seeded into the cavities in serial dilution. After thorough washing, bound antibodies were detected with biotin labeled rat-anti-mouse-IgM (BD, Heidelberg, Germany 1 µg/ml) or goat-anti-mouse-IgG (Dianova, Hamburg, Germany 1 µg/ml), peroxidase-labeled Streptavidin (Dianova, Hamburg, Germany 5 µg/ml) and AEC-Substrate (Sigma-Aldrich, Steinheim). Spots were enumerated with the Immunospot Series 5 Versa ELISPOT Analyzer (CTL Europe GmbH, Bonn, Germany) and the ImmunoCaptureTM Image Acquisition software. Data were analyzed with ImmunoSpot® Analysis software V5.
Statistical analyses were performed using GraphPad Prism 5 for Windows (GraphPad software, San Diego, CA). We used one-way ANOVA with the Bonferroni’s posttest for selected pairs. P-values < 0.05 were considered to be significant. Analyses with a non-parametric two-tailed Man-Whitney U test gave similar results (
We established an experimental mouse model of humoral memory against which to measure the effects of sepsis. We immunized C57Bl/6 mice with OVA at 6 wk and performed a secondary OVA immunization at 9 wk. Ten and 12 week after the secondary immunization, we measured parameters of humoral memory. Against this background, we induced experimental sepsis by CASP 8 wk after the secondary immunization to assess humoral responses to sepsis. Thus the readout of memory took place two and 4 wk after sepsis induction. The CASP model has been developed to closely mimic the clinical situation of diffuse fecal peritonitis, a complication of abdominal surgery caused by intestinal leakage [
C57BL/6 mice were subjected to CASP (18G). At the indicated time points blood, spleen, peritoneal lavage, liver, lung, and kidney were harvested. Solid organs were homogenized with an ultrathorax. Liquids and organ suspensions were plated on agar-plates and incubated for 22 h, when cfus were enumerated. n=5-6 mice/group.
In our model of humoral memory OVA-specific bone marrow antibody-secreting cells (ASC) were abundant in the bone marrow at 10 wk after the secondary immunization and they further increased during the next two weeks (from 83.4±38.3 × 106 to 125±94.2 × 106; p = 0.09) This shows that the antigen-specific, long-lived bone marrow plasma cell population is not fully established by 10 wk after a secondary immunization (
OVA-specific IgG-ASCs in spleen and bone marrow 2 and 4 wk after CASP (colon ascendens stent peritonitis) are shown here. C57BL/6 mice were immunized and boosted with OVA at 6 and 9 wk, respectively, and CASP was induced 8 wk after the secondary immunization. Control animals were either untreated, only immunized or only CASP induced. A combination of two independent experiments with the same trend is shown here. A statistical analysis was performed with the one-way ANOVA and Bonferroni’s post test for selected pairs. N = 10-18/group.
*: p<0.05; **: p<0.01.
Using our model of humoral memory, we induced experimental sepsis by CASP 8 wk after the secondary immunization. Two and 4 wk later, the total serum IgG concentration tended to be higher in septic animals than in the control animals (
Serum IgM and IgG concentrations 2 or 4 wk after CASP (colon ascendens stent peritonitis) are shown here. C57BL/6 mice were immunized and boosted with OVA at 6 and 9 wk, respectively, and CASP was induced 8 wk after the secondary immunization. Control animals were either untreated, only immunized or only CASP induced. A statistical analysis was performed with the one-way ANOVA and Bonferroni’s post test for selected pairs. n=10-18/group. A single experiment is shown here.
In contrast to the spleen, there was no increase of OVA-specific ASCs in the bone marrow of immunized septic animals, where the majority of the ASCs was located. In fact, the number of bone marrow antigen-specific IgG-ASCs was significantly lower than in non-septic animals at 4 wk after sepsis (
OVA-specific serum-IgM and -IgG at 2 and 4 wk after CASP (colon ascendens stent peritonitis) are shown here (au = arbitrary units). C57BL/6 mice were immunized and boosted with OVA at 6 and 9 wk, respectively, and CASP was induced 8 wk after the secondary immunization. Control animals were either untreated, only immunized or only CASP induced. A combination of two independent experiments with the same trend is shown here. A statistical analysis was performed with the one-way ANOVA and Bonferroni’s post test for selected pairs.
n=10-18/group. *: p<0.05; **: p<0.01.
It is not known whether sepsis influences established humoral memory also in humans. However, since major surgery is also associated with suppression of the adaptive immune system [
The mechanism by which sepsis impedes the complete establishment of antigen-specific ASCs in the bone marrow at 12 wk after secondary immunization is unclear. Based on the current literature, sepsis would be expected to exert a number of stimulating effects on plasma cell survival. For example, eosinophil granulocytes are an important component of the survival niches [
A failure to increase OVA-specific plasma cells after sepsis could be the result of an impaired migration from the periphery into the bone marrow. However, the possibility cannot be excluded that long-lived plasma cells die and are replaced with migrating plasma cells. If this occurs in balanced proportions, a change in cell numbers cannot be measured.
Another explanation for the lack of plasma cell pool expansion in our model might be a competition for the bone marrow niches between OVA-specific plasma cells and plasma cells with other specificities that originate during sepsis. On one hand, Bortnick et al. showed in a mouse model, that T cell independent B cell reactions can produce long-lived plasma cells, thereby also inducing an immunological memory [
In conclusion, sepsis impairs the pre-existing humoral memory in mice. We observed a decrease in antigen-specific serum antibody concentration and the number of antigen-specific bone marrow plasma cells of septic animals remained lower than in control animals. Altogether 4 wk after the induction of sepsis, there were only half as many antigen-specific bone marrow plasma cells in septic mice compared with immunized non-septic animals. The mechanism by which sepsis impedes the antigen-specific plasma cell population remains to be elucidated. Our data demonstrate, however, that in borderline cases, sepsis could endanger protective immunity.
(DOCX)
We thank Dr Alysia Battersby, PhD from BioMedWord Ltd for her excellent editing work on the manuscript.