CD95 Expression in Aged Mice Infected with Tuberculosis

Abstract

The interaction between CD95 and its ligand is an important homeostatic mechanism that leads to the induction of apoptosis in activated T cells. In view of recent evidence that this pathway might be defective in aged mice, this study investigated CD95 expression on T cells in old mice activated by infection with Mycobacterium tuberculosis. The results of the study do not support the hypothesis that CD95 is poorly expressed on CD4 T cells from old mice; instead, it was found that similar numbers of T cells from young and old mice expressed CD95, with the intensity of expression if anything higher on the cells from the old mice. In addition, the study demonstrated that changes in CD44 and CD45RB expression previously observed in young infected mice proceeded in a similar fashion in old animals and, as would be predicted, that CD95^hi expression was primarily associated with CD4 T cells expressing the activated CD44^hi CD45RB^hi phenotype.

Like other physiological systems, the clonal expansion of antigen-specific T cells is controlled by homeostatic mechanisms that down-regulate this event by inducing a state of programmed cell death (apoptosis). A primary mechanism involved in this process is mediated by the interaction between the molecule CD95 (Fas/APO-1) and its ligand (CD95L). The CD95L molecule is not found on resting cells but increases in expression during T-cell activation, whereupon it can cross-link to CD95, leading to apoptosis (3, 5, 7, 8, 20, 21).

It has recently been suggested, however, that the CD95 homeostatic pathway becomes defective as an animal ages, with a reduction in the expression of this molecule (10, 22). If so, then it is possible that this dysfunction will contribute to a number of changes that occur within the immune system during senescence (13, 14) and might potentially reduce the capacity of the animal to respond to an infection.

We have studied these changes in the context of a realistic model of aging and susceptibility to disease. Tuberculosis is more common in the elderly than in other segments of the population (12, 19), and aged-mouse models of pulmonary infection have revealed a number of subtle defects in these animals that may pertain to this susceptibility (15). Hence, while a needed CD4-protective T-cell population can be induced in old mice, the kinetics of emergence of this population appears to be slowed by a diminished interleukin-12 response, needed to drive gamma interferon (IFN-γ) production by these cells (1). In addition, the accumulation of such cells in inflammatory sites in the lungs is further hampered by poor expression of the adhesion molecules needed for correct cell trafficking (16).

In the current study, we investigated the possibility that poor CD95 expression would be associated with poor regulation of activated T cells acquired in response to tuberculosis infection. However, contrary to the hypothesis and to previous data (22), we found that CD95 was in fact strongly expressed on many T cells from old mice, as assessed by flow cytometric analysis. Moreover, expression of CD95 correlated well with the known phenotype of activated T cells acquired in response to the infection. These data therefore seem to indicate that the CD95 pathway remains intact in old mice.

Young (3-month-old) and old (24-month-old) B6D2F1 female mice were purchased from the Trudeau Institute animal breeding facility, Saranac Lake, N.Y. They were kept under barrier conditions throughout the experiments. Upon sacrifice, each animal was checked carefully for tumors or other pathology and excluded from the study if any were noted. The Erdman strain of Mycobacterium tuberculosis was grown to mid-log phase in Proskauer-Beck medium containing 0.02% Tween 80 and then bottled in 1-ml aliquots and frozen at −70°C until used. The mice were inoculated via a lateral tail vein with 200 μl of sterile saline containing 10⁵ viable bacilli. Spleens were harvested 8 days later (when protective immunity can first be detected in this model) and 20 days later (when immunity first peaks).

Single-cell suspensions from spleens harvested from euthanized animals were prepared in RPMI 1640 medium lacking biotin and phenol red (DRPMI; Irvine Scientific, Santa Anna, Calif.) and supplemented with 1% l-glutamine, 1% HEPES, and 1% antibiotics. The cells were washed by centrifugation at 200 × g for 5 min. The cell pellet was then resuspended in 20 ml of DRPMI, and the suspension was pipetted onto sterile 150-mm-diameter tissue culture petri dishes. The cells were incubated for 1 h at 37°C and 6% CO₂ to allow macrophages to adhere. The nonadherent cell population was then carefully removed and washed as described above. The cell pellet was treated with ACK lysing buffer (0.15 M NH₄Cl, 1.0 mM KHCO₃, 0.1 mM Na₂EDTA, pH 7.2 to 7.4) for 5 min at 21°C to lyse erythrocytes. The cells were then washed twice and resuspended in 10 ml of DRPMI medium containing 0.1% NaN₃. The suspension was passed through a nylon mesh cell strainer and counted with a Coulter (Hialeah, Fla.) Counter ZF particle counter. The cells were adjusted to a final concentration of 2 × 10⁷ per ml in DRPMI. Replicate wells in 96-well microtiter plates were seeded with 100 μl of this suspension for each staining group for flow cytometric analysis. The 96-well plates containing the cells were then centrifuged (200 × g for 5 min) to form cell pellets at the bottom of the wells. The supernatant was gently shaken out of the wells, and 25 μl of conjugated antibody (20 to 25 μg/ml) specific for the phenotypic markers of interest was added. The antibodies used were as follows: CD3-phycoerythrin (PE) (145-2C11); CD4-fluorescein isothiocyanate (FITC) (R-M4-5); CD8-biotin (53-6.7); CD44-FITC (IM7); CD4 5RB-biotin (16A); CD95-biotin or CD95-PE (Jo2). Control wells containing no stain or isotype-matched antibody controls, as well as gating control wells (i.e., anti-CD4-FITC and anti-CD3-PE), were included. To blank wells, 50 μl of staining media was added. After the wells were washed, 50 μl of Streptavidin-RED670 (Gibco-BRL catalog no. 19543-024) was added to each well. The plates were then incubated for 15 min in the dark at 4°C. The plates were washed three times as described above and then resuspended with 100 μl of DRPMI. Identical samples were pooled into Eppendorf tubes containing 600 μl of DRPMI for analysis on the flow cytometer. Analysis was performed with a Coulter Epics flow cytometer; after gating on cell populations of interest and excluding background staining, we analyzed the list mode data with contouring and smoothing computer programs.

Nonadherent spleen cells from infected mice were incubated with a mixture of anti-CD8 (Lyt-2.43), anti-B-cell (J11d.2), anti-γδ (GL3), and anti-NK (PK136) monoclonal antibodies plus low-toxicity rabbit complement for 1 h at 37°C and then washed three times. Surviving CD4 T cells had a purity of >96% as assessed by flow cytometry.

Intravenous infection of young or old mice did not cause any overt changes in the relative numbers of CD4 and CD8 T cells in the spleen over the first 20 days (representative results are shown in Fig. 1). Moreover, in keeping with previous knowledge (6), many cells within the CD4 population (within which protective immunity to the tuberculosis infection is initially generated) in the old mice exhibited high expression of CD44 and low expression of CD45RB (Fig. 2). Following infection, the cytometric analysis provided evidence for an early expansion of CD4 T cells expressing the CD44^hi CD45RB^hi phenotype, followed, by 20 days post-challenge, by evidence of an increase in the CD44^hi CD45RB^lo population (Fig. 2).

FIG. 1.

Evidence that infection with M. tuberculosis did not cause any overt changes in the relative numbers of CD4 and CD8 T cells from young (left) or old (right) mice harvested 8 or 20 days into the infection. Uninfected age-matched controls gave similar patterns (data not shown). The cells were gated on stained CD3-positive cells and then analyzed for CD4 (LFL3) and CD8 (LFL1). The data were analyzed from a total of 20,000 events.

FIG. 2.

Flow cytometric profiles of CD44 (LFL2) and CD45RB (LFL3) expression on gated CD4 T cells harvested from old mice on day 0, 8, or 20 of the infection. Note the increasing density of cells expressing the CD44^hi CD45RB^mod/lo phenotype, which is similar to previous observations made in young infected mice. The data were analyzed from 10,000 total events.

After gating on CD4 T cells, one-parameter analysis indicated that CD95 was expressed on similar numbers of cells (based upon area under the curve) in young and old mice, but with about half of those from the old animals staining more brightly. This observation, which is contrary to reports in recent literature (22), was seen in three separate experiments (Fig. 3). After infection, the brightness of expression of CD95 increased slightly on cells from young mice but declined overall on cells from old animals (Fig. 3).

FIG. 3.

One-parameter flow cytometric analysis of CD95 expression on gated CD4 T cells. The two peaks on the left (one shaded for contrast) in each chart are from isogenic control samples. The data were analyzed from 10,000 total events.

In a final series of experiments, CD4⁺ T cells were harvested and gated on CD95^hi T cells, which were costained for CD44 and CD45RB. In these experiments, CD95^hi cells from old mice were found to be CD44^hi and within both the CD45RB^hi and CD45RB^lo populations (Fig. 4) prior to infection. Virtually no CD4 T cells from young mice were found within this gate prior to infection. Following infection, analysis of these cells for CD44 and CD45RB expression showed that the great majority were of the “blast/activated” phenotype, CD44^hi CD45^hi (Fig. 4).

FIG. 4.

Evidence that CD95 expression was primarily associated with CD4 T cells of the CD44^hi CD45RB^hi phenotype. CD4 T cells were isolated and gated for bright expression of CD95 and then analyzed for CD44 (LFL2) and CD45RB (LFL3) expression. Data for young (left) and old (right) mice on days 0, 8, and 20 are shown. The data were analyzed from 10,000 total events.

Previous analysis of young mice has clearly shown that the course of M. tuberculosis infection is associated with the emergence of an IFN-γ-secreting CD4 T-cell population (2, 4, 17, 18). About a week into the infection a distinct CD44^hi CD45RB^hi subset can be seen by flow cytometric analysis, which we have previously demonstrated consists of cells of increased physical size, presumably blasts (6). After approximately 3 weeks of infection, a second distinct population, CD44^hi CD45RB^lo, emerges, which may consist of a memory T-cell population (6).

The results of the current study indicate that a similar series of phenotypic changes are occurring in the CD4 population in the old mice. If correct, then this observation dissociates activation of this population by the infection from adequate IFN-γ production by these cells, which we have previously shown is delayed (16), and from the capacity of CD4 T cells to enter sites of inflammation in target organs, which we have previously hypothesized is a consequence of reduced adhesion molecule expression (16).

As would be predicted, expression of CD95^hi correlated mainly with the CD44^hi CD45^hi-activated CD4 T-cell phenotype. This phenotype emerges with the onset of IFN-γ cytokine secretion, and hence CD95-mediated apoptosis may act as a dampening mechanism given the tissue-damaging molecules (oxygen and nitrogen radicals, nitric oxide, peroxynitrite, etc.) that activated macrophages then proceed to elaborate (9).

It remains unclear why the CD95^hi CD44^hi CD45RB^lo population seen in resting old mice had apparently disappeared by day 20 of the infection. The most reasonable speculation is that these cells down-regulated CD95 expression and hence moved outside the preset CD95^hi gate; given the putative identification of CD44^hi CD45RB^lo cells as possible memory cells (6, 11), this would seem to be a necessary step for this population.

In summary, the results of this study indicate that T cells in old mice are capable of expressing CD95 in response to an infectious disease and that high expression of this molecule associates with T cells with an activation phenotype, similar to that seen in younger animals. This study provides no evidence to support the hypothesis that CD95 expression per se is dysfunctional in old mice, although we acknowledge that differences between our results and those of others (22) could simply reflect technical conditions, such as the condition of aged animals, the antibodies used, and so forth.

Despite our conclusions, however, it should be emphasized that the current findings do not provide information as to the actual functional role of CD95 on T cells from old mice, and this will bear investigation in a larger study. Such studies should include the expression of CD95 ligand (CD95L) on activated T cells given the interaction of CD95 and CD95L in apoptosis. Currently we view this possible interaction as a means to remove activated cytokine-secreting protective T cells if their prolonged presence results in local tissue damage, but whether this mechanism is dysfunctional in old mice and contributes to their increased susceptibility to tuberculosis or other infectious diseases common in the elderly remains unknown.