Cutting Edge: T Cells from Aged Mice Are Resistant to Depletion Early During Virus Infection¹

Aging is associated with declines in both cell-mediated and humoral immunity. T cell function in both aged mice and humans has been characterized by decreased proliferation and altered cytokine production (1, 2, 3, 4). Although many studies of the changes in immune response with age have used mitogenic stimuli in vitro, we have demonstrated both declines and delays in specific T cell responses to infection with two different viruses. Aged mice infected with influenza virus demonstrate a diminished CD8 T cell response that occurs later, resulting in prolonged influenza virus expression (5). This age-related impairment of CD8 T cell function during infection is due to a defect in specific T cell expansion, rather than in effector activity (5). E55 + murine leukemia retrovirus (E55+MuLV)³ is a replication-competent murine retrovirus that establishes a low-level chronic infection after an initial acute infection. Initiation and maintenance of the low-level infection is due to an intact immune response (6, 7). Although young mice achieve this low level of chronic virus expression by 8 wk postinfection, the acute phase of E55+MuLV expression is prolonged in aged mice until 16–20 wk postinfection, even though there was no difference of virus burden in the spleen between young and aged mice up to 4 wk after infection (8). This delay in virus clearance was associated with the lack of a T cell proliferative response, a significantly lower cytotoxic T cell response, and significantly lower virus-neutralizing Ab levels in aged compared with young mice. The mechanisms for these alterations are still unknown.

Recently, it has been demonstrated in young mice that there is a depletion of nonspecific T cells early after infection (9). This depletion of nonresponding T cells may provide space for the proliferation and expansion of specific T cells that are required for clearance of intracellular microbes. The present studies were performed to examine whether or not altered depletion of T cells after infection contributes to the decreased specific T cell response of aged mice.

Six- and 22-mo-old C57BL/6 male mice were purchased from NIA colony of Charles River Laboratories (Wilmington, MA). All mice were maintained in American Association for the Accreditation of Laboratory Animal Care-approved facilities at the Drexel University College of Medicine. Mice were allowed to acclimate for at least 1 wk in our facilities before use. Mice demonstrating tumors were eliminated from the study. E55+MuLV was originally isolated from a spleen of BALB.K mice injected with cell-free culture supernatant from a T cell line derived from a leukemic mouse (10). The virus used in studies was propagated in vivo by i.p. injections of immunocompromised BALB/c mice (11). Mice were infected i.p. with 5 × 10⁴ focus-forming units (FFU) of E55+MuLV.

Splenocytes isolated from young and aged mice were purified by MACS using CD8a microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany), labeled with 5–10 μm CFSE (Molecular Probes, Eugene, OR) in PBS, and then quenched with 100% FCS. Cells were resuspended in PBS after washing with RPMI 1640/5% FCS, and were transferred i.v. into young or aged recipient mice (2–5 × 10⁶ cells/mouse, or 1–2 × 10⁷ cells/mouse for confocal microscopy). One day after transfer, recipients were infected i.p. with E55+MuLV at 5 × 10⁴ FFU.

Spleens and other tissues were removed from individual mice, and lymphocytes were prepared and resuspended in 1% BSA in PBS at a concentration of 1 × 10⁶ cells/well in a 96-well plate. Cells were stained for surface markers using mAbs purchased from BD PharMingen (San Diego, CA) and fixed with 1% paraformaldehyde in PBS. Flow cytometry was performed with a FACSCalibur (BD Biosciences, San Jose, CA), and data were analyzed with FlowJo software, version 4.2 (Tree Star, San Carlos, CA).

Frozen 5-μm-thick spleen and lymph node sections were fixed with 4% paraformaldehyde and permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate. In situ cell death detection kit/TMR red (Roche Dignostics, Indianapolis, IN) was used to detect TUNEL-positive cells in the tissues. Sections were mounted on SlowFade medium (Molecular Probes), covered with a glass slide, and then subjected to a confocal laser-scanning microscope system (TCS SP2; Leica Microsystems, Heidelberg, Germany).

To determine whether T cells of young and aged mice demonstrate differential sensitivity to depletion early during virus infection, we infected young and aged C57BL/6 mice with E55+MuLV. On days 1 to 3 postinfection, CD4 and CD8 T cells from spleens were quantitated by flow cytometry. In young mice, depletion of T cells was apparent on day 1 after infection, with 50% of CD8 T cells, as well as 25–50% of CD4 T cells, being depleted by day 3 after infection. There was no depletion of either CD8 or CD4 T cells in aged mice at these time points (Fig. 1 A).

CD44 is a cell adhesion receptor that is up-regulated on the surface of activated CD8 or CD4 T cells and is an indicator of the memory phenotype of T cells in mice (12). The percentage of T cells with the memory phenotype is increased with age (13). A possible explanation for the lack of depletion in aged mice is a decreased sensitivity of memory T cells to depletion. Examination of the memory phenotype of T cells after E55+MuLV infection indicated that, in young mice, both CD44^high and CD44^low T cells were depleted, indicating no preferential resistance to depletion. Interestingly, neither CD44^high nor CD44^low CD8 T cells were depleted in aged mice (Fig. 1 B). These results demonstrate that all T cells of aged mice are more resistant to depletion than T cells of young mice during the early phase of virus infection.

Two possibilities may account for the depletion of splenic T cells in young mice early during infection: altered trafficking or cell death. If altered trafficking was responsible for depletion in the spleen, the number of T cells in other lymphoid and peripheral organs should be increased. To examine this possibility, we transferred CFSE-labeled purified CD8 T cells from young or aged mice into young mice and then infected the recipients with E55+MuLV. On day 3 postinfection, the presence of the donor CD8 T cells was evaluated in various tissues, including spleen, liver, peripheral blood, and lung. For young donors, 90% of the transferred cells were depleted from all of the tissues examined, whereas there was no depletion of T cells from aged donors (Fig. 2). Similar decreases in donor T cells in all tissues examined indicated no altered trafficking. Importantly, depletion of host cells occurs in all E55+MuLV-infected mice regardless of age of donors (young→young, 25% depletion; aged→young, 22% depletion based on total CD8⁺ cells).

To examine whether cell death was occurring, we used confocal microscopy to visualize transferred CFSE-labeled CD8 T cells and TUNEL-positive apoptotic cells in sections of spleen and lymph nodes. Most donor CD8 T cells of young mice demonstrated apoptosis, whereas few donor CD8 T cells of aged mice were TUNEL positive on day 3 post-E55+MuLV infection in both spleen and inguinal lymph nodes (Fig. 3). However, apoptosis did occur in recipient cells, as indicated by the TUNEL⁺ (red) cells in mice receiving CD8⁺ T cells of either young or aged mice. These results suggest the following: 1) depletion in T cells of young mice early after infection with E55+MuLV is due to apoptosis rather than migration to other tissues; and 2) intrinsic cellular properties of the T cells are responsible for differential susceptibility of the aged T cells to depletion. These results of young mice with E55+MuLV infection are consistent with previous observations in Listeria monocytogenes and lymphocytic choriomeningitis infections (9).

Studies in vitro have shown that T cells of aged mice demonstrate limited proliferation and expansion compared with T cells of young mice after either specific or nonspecific stimulation (4, 5, 8, 13, 14). These results are consistent with our conclusion that intrinsic properties of aged T cells are involved in the limited T cell proliferation. However, in vivo, it is possible that the environment also influences the proliferative capacity of T cells. To further investigate whether age-associated alterations in T cell depletion is influenced by the environment or is limited to intrinsic characteristics of T cells, CFSE-labeled CD8 T cells of young and aged mice were adoptively transferred into aged recipients. Three days after E55+MuLV infection, no depletion of donor CD8 T cells of either young (Fig. 4) or aged mice (data not shown) in aged recipients was observed. Donor cells of neither young nor aged donor mice were TUNEL positive in the spleen and lymph node, indicating a lack of induction of apoptosis in the transferred T cells (data not shown). These results indicate that both environmental and intrinsic cellular properties influence depletion of T cells of aged mice.

Why is less depletion occurring in aged T cells early during virus infection? A number of studies have indicated that T cells of aged mice appear more resistant to apoptosis in various experimental conditions, such as spontaneous apoptosis in cultures (15), apoptosis leading to contraction after specific T cell expansion stimulated by Ags (13), or resistance to depletion by anti-CD4 Ab (16). Although more virus-specific CD8 T cells are generated during the peak of a primary response in young mice than in aged mice, there was minimal difference in the number of virus-specific memory CD8 T cells between young and aged mice, suggesting decreased apoptosis of virus-specific CD8 T cells in aged mice (13). It has also been reported that activated CD8 T cells from aged mice exhibit decreased activation-induced cell death in vitro (17).

A possible mechanism for decreased apoptosis in aged mice is altered expression of surface molecules involved in apoptosis. Fas is a bifunctional molecule that is critical for apoptosis and stimulation during T cell development. Fas expression and ligand-induced apoptosis is decreased on T cells from aged mice compared with young mice (14). Bcl-2 is an important antiapoptotic molecule that is expressed at higher levels in CD44^high CD8 T cells than in other T cell subsets of young mice, and at even higher levels on CD44^high CD8 T cells in aged mice. Although Bcl-2 is not elevated in CD44^high CD4 T cells, another antiapoptotic molecule, Bcl-x_L, is up-regulated in both CD44^high CD8 T cells and CD44^high CD4 T cells, and slightly increased expression of Bcl-x_L with aging has been reported (15). Although not all studies in aged mice demonstrate a decreased susceptibility to apoptosis (18, 19), there is considerable data that support our current results indicating that reduced susceptibility to apoptosis contributes to decreased response of T cells of aged mice to virus infection.

After virus infection, transient increases in IFNαβ induce the production of IL-15, which up-regulates Bcl-2 and Bcl-x_L expression (15). Depletion of T cells of young mice early during lymphocytic choriomeningitis infection correlated with the level of IFNαβ, because the IFN inducer poly(I:C) caused apoptosis of CD8 T cells in normal mice, but not in IFNαβR-deficient mice (20). In young mice, Bcl-x_L transgenic T cell depletion was indistinguishable from depletion of wild-type T cells early during Listeria infection (J. Jiang and H. Shen, unpublished data). These results indicate that additional environmental properties may limit the depletion of CD44^high T cells of aged mice, and also suggest that Bcl-2 and Bcl-x_L alone are not sufficient to prevent T cells of young mice from depletion early during infection. Due to conflicting data on T cells from aged mice to apoptosis and the data suggesting a role for cytokines in vivo to induce T cell apoptosis (20), future studies will focus on the effect of cytokines, such as TNF-α, IFNαβ, and IL-15, on the depletion of T cells early during infection.

What is the significance of the difference in depletion of T cells of young and aged mice early during infection? Recent studies have demonstrated that nonspecific T cells were depleted by apoptosis early during bacterial and viral infection. No specific T cells were depleted early during infection (9). This selective depletion of numerous nonspecific T cells may provide space, allowing maximal proliferation and expansion of Ag-specific T cells, which is critical for clearance of intracellular pathogens. It is known that age-related impairment of CD8 T cell function during virus infection is due to a defect in specific T cell expansion, rather than in effector activity (4, 5). Our results suggest that one reason for impairment of the T cell response to pathogens in aged individuals is that nonspecific T cells are not depleted early during infection, which may prevent efficient proliferation and expansion of specific T cells. These data provide a new explanation for the immunological alterations associated with aging, especially at the level of T cells, and may have important implications for the design of vaccines for elderly individuals.

We thank L. Bertrand for technical assistance, and H. Shen, L. L. Lau, and E. M. Gardner for critical review of this manuscript.