Increased inducible apoptosis in CD4+ T lymphocytes during polymicrobial sepsis is mediated by Fas ligand and not endotoxin

Abstract

Recent studies suggest that increased lymphocyte apoptosis (Ao) detected in peripheral blood T cells from burn patients appears to contribute to decreased lymphocyte immunoresponsiveness. However, while it is known that sepsis induces a marked depression in the splenocyte immune response (i.e. decreased interleukin-2, interferon-gamma production and proliferation) in response to the T-cell mitogen concanavalin A (Con A), it is unknown whether this depression is associated with an increase in inducible Ao and if so, which mediators control this process. To assess this, splenocytes were harvested from mice at 24 hr (a period associated with decreased Con A response) after the onset of polymicrobial sepsis [caecal ligation and puncture (CLP)] or sham-CLP (Sham) and then stimulated with 2.5 microg Con A/ml (24 hr). Septic mouse splenocytes stimulated with Con A, while not showing a change in their phenotypic make-up, did exhibit a marked increase in the percentage of splenocyte that were Ao+ which was associated with altered cytokine release. This appears to be due to an increase in the percentage of Ao+ cells in the CD4+ CD8- population and was associated with enhanced Fas antigen expression as well as an increase in mRNA for the Fas-FasL gene family. To determine if the changes in Ao are due to either endotoxin (a product of Gram-negative bacteria seen in CLP mice) or the expression of Fas ligand (FasL; a mediator of activation-induced lymphocyte Ao), a second set of studies examining Con A-inducible Ao was performed with splenocytes harvested from septic endotoxin-tolerant C3H/HeJ and the FasL-deficient C3H/HeJ-Fasl gld mice. The results show that increased splenocyte Ao detected following CLP is due to a FasL-mediated process and not to endotoxin. Thus the inadvertent up-regulation of FasL-mediated splenocyte Ao may contribute to the depression of splenocyte immune responses seen during polymicrobial sepsis.

Figure 1

(a)–(c) Splenocytes harvested from C3H/HeN mice 24 hr following the onset of CLP release significantly less IL-2 and IFN-γ when stimulated with the non-specific T-cell mitogen Con A. Alternatively, this was associated with a marked increase in the capacity of the same cells to release IL-10 (see Fig. 1c). Although not shown, we detected none of these cytokines in the absence of Con A. Significance by Student’s t-test as indicated by * was P < 0·05 versus Sham; mean±SEM; n is five to six mice sampled/group.

Figure 2

(a)–(h) illustrates the results of typical three-colour flow cytometric analysis of a cell suspension obtained from spleen 24 hr following Sham-CLP (a,c,e,g) or CLP (b,d,f,h) and stimulated in vitro for 24 hr with 2·5 μg Con A/ml. The contour plot (a, b) of the total cell sample illustrates the typical cell cycle progression and the primary gate established on this population (R1). (c) and (d) are the cell cycle histograms of cell number versus DNA content generated from the R1 gated population in (a) and (b) respectively, which show the typical regions defined for these studies. With respect to the Sham, it can be seen that spleen cells extracted from a septic mouse at 24 hr exhibited a population of cells trailing off with lower DNA content, A_o⁺ (M1), from the G₀/G₁ peak (M2) as opposed to S/G₂/M phase cells (M3)(a, c versus b, d). However, no marked shift in the phenotypic make-up of the stimulated splenocytes is evident in septic animal cells at 24 hr (e versus f). Histograms of DNA content produced from each of these gated phenotypic populations illustrate that the majority of the increase in apoptotic cells is present in the single-positive CD4⁺ stained cell population (g versus h).

Figure 3

Neither (a) ex vivo (innate or unstimulated) nor (b) in vitro Con A-stimulated (24 hr) splenocytes harvested from mice 24 hr following CLP or Sham-CLP exhibit marked changes in their phenotypic make-up. (c) Neither the mixed (inset) nor those cells differentiated by their phenotypic (CD4 and/or CD8) expression from the ex vivo splenocytes harvested from mice 24 hr following CLP or Sham-CLP exhibited a marked increase in the percentage of A_o⁺ based on number of cells residing in the M1 region of the cell cycle figures depicted in Fig. 2(c) or (d). Phenotype and associated cell cycle analysis is gated as depicted in Fig. 2(e, g),(f, h). (d) Alternatively (inset) the mixed splenocytes harvested from septic mice show a significant increase (*P < 0·05 versus Sham group) in the percentage of cells which are A_o⁺ following 24 hr stimulation with Con A. Furthermore, when the CD4 and/or CD8 phenotypic expression was correlated with the extent of A_o⁺ expression, a marked increase (*P < 0·05 versus Sham group) in the percentage of single-positive CD4⁺, T helper cells, which were also apoptotic (A_o⁺) was observed. Mean ± SEM; n is six mice sampled/group.

Figure 4

(a) Ex vivo splenocytes exhibit no detectable genomic DNA fragmentation (as detected by TBE–polyacrylamide gel electrophoresis) at 24 hr following CLP. The gels are documented for splenocytes taken from three mice in each group (lanes marked 1–3 for each independent mouse cell sample) by the production of reverse image of the ethidium bromide-stained gel using a MOCHA Image Analysis System (Jandel Sci., San Rafael, CA). The Lambda-DNA HindIII digest standard is listed on the right hand side of the gel by base pair (bp) number. (b) Following 24 hr of in vitro stimulation with Con A there is a marked increase in the genomic DNA fragmentation in both Sham and CLP mouse splenocytes. However, stimulated septic mouse splenocyte DNA showed a consistently heavier (darker) staining pattern. (c) Densitometric measurement of the 200-bp multimere staining intensity present in the fragmentation gel (illustrated in Fig. 5B. as the broad bands between the dashed lines) indicates that stimulated splenocytes exhibit a significantly higher concentration of this nucleosomal fragment. (d) Mixed splenocytes harvested from septic mice also show a significant increase in the percentage of cells which are Annexin V⁺ following 24 hr stimulation with Con A. Significance indicated by * at P < 0·05 versus the Sham group; Mann–Whitney U-test; mean±SEM; n is six mice sampled/group.

Figure 5

Con A-stimulated splenocytes obtained from C3H/HeN mice 24 hr post-CLP demonstrate a marked increase in Fas receptor density per cell (mean fluorescent intensity) (a) and percentage of Fas⁺ cells (b). Significance indicated by * at P < 0·05versus the Sham group; Mann–Whitney U-test; mean±SEM; n is four mice sampled/group.

Figure 6

Following 24 hr of stimulation with Con A splenocytes from C3H/HeN mice subjected to CLP 24 hr prior exhibited an enhanced expression of the FasL, Fas antigen and its associated signalling components (FLICE, FADD, FAF) as well as in components of Fas–FasL-related apoptotic receptor complexes such as, Fas2L/TRAIL, TRADD, and RIP, as detected by RPA. The results of a typical gel are presented in (a). Summary data for three repeated experiments on the splenocytes from three different mice in each group are presented in (b) Significance indicated by * at P < 0·05 versus the equivalent Sham group.

Figure 7

Only Con A-stimulated splenocytes obtained from C3H/HeJ mice 24 hr post-CLP demonstrate a marked increase in the frequency of Fas-positive cells (c), Fas receptor density per cell (mean fluorescent intensity) (b), and percentage of cells which are apoptotic (a). No such increase was detected in the cells obtained from the FasL-negative C3H/HeJ-Fasl^gld mouse following CLP. Significance indicated by * at P < 0·05 versus the equivalent strain Sham group or # at P < 0·05 versus the C3H/HeJ strain Sham group; Mann–Whitney U-test; mean±SEM; n is five to seven mice sampled/group.

Figure 8

As was seen with splenocytes harvested from the endotoxin-sensitive C3H/HeN mice (Fig. 1a–c), cells obtained from septic endotoxin-tolerant C3H/HeJ mice showed a sharp decline in their capacity to liberate IL-2 (a) and IFN-γ (b) associated with a significant increase in the ability to secrete IL-10 (c). However, a divergent pattern was observed in the splenocytes obtained from C3H/HeJ-Fasl^gld (endotoxin-tolerant/FasL-deficient) mice. While a partial restoration of IFN-γ release was seen in the stimulated septic C3H/HeJ-Fasl^gld mouse cells, no improvement in IL-2 productive capacity was observed and no decline in IL-10 secretion was evident. However, with respect to IL-10 release capacity, this was consistently higher in both Sham and CLP mice in splenocytes from C3H/HeJ-Fasl^gld compared to the respective C3H/HeJ group. Significance indicated by * at P < 0·05 versus the equivalent strain Sham group or # at P < 0·05 versus the C3H/HeJ strain Sham group; Student’s t-test; mean±SEM; n is five or six mice sampled/group.