Aging beyond menopause selectively decreases CD8+ T cell numbers but enhances cytotoxic activity in the human endometrium

CD8+ T cell numbers decrease with age in the endometrium

Previous studies from our laboratory have demonstrated that EM CD8+ T cell cytotoxicity varies with the stage of the menstrual cycle, is markedly increased following menopause, and is suppressed both directly and indirectly by sex hormones [12, 14, 17]. In the present study, we investigated the underlying mechanisms that influence CD8+ T protection in the EM as women age following menopause. First, we measured the number of EM CD8+ T cells per gram of tissue in each patient. Consistent with our past findings [11], the number of EM CD8+ T cells per gram of tissue was significantly lower in post-menopausal versus pre-menopausal women (Supplementary Fig. 1A). In both populations there was a wide range of total numbers of EM CD8+ T cells, ranging from 1.02 × 104 to 3.64 × 105 cells/g and 1 × 104 to 2.49 × 105 cells/g in pre- and post-menopausal women respectively. We then stratified CD8+ T cell numbers as a function of age across the entire study population. As seen in Fig. 1A, we found that EM CD8+ T cell numbers decreased significantly with increasing age in the entire population. However, this overall decrease in CD8+ T cell number masked two distinct profiles in the pre- and post-menopausal populations. In pre-menopausal women, the number of EM CD8+ T cells/g did not significantly change with increasing age prior to menopause (range 26–53 years; n = 35) (Fig. 1B, left). In contrast, in post-menopausal women EM CD8+ T cell numbers declined significantly with increasing age (range 48–81 years; n = 33) (Fig. 1B, right).

Fig. 1figure 1

CD8+ T cell numbers decrease with age in the EM. A Correlation between number of CD8+ T cells recovered per gram of EM tissue after magnetic bead isolation and age from the entire study population (n = 68). B Correlation between number of CD8+ T cells recovered per gram of EM tissue after separating pre- (left; n = 35) and post-menopausal cells (right; n = 33) women. C Correlation between number of CD103 + CD8+ T cells recovered per gram of EM tissue and age from pre- (left; n = 21) or post-menopausal (right; n = 17) women. D Correlation between number of CD103-CD8+ T cells recovered per gram of EM tissue and age from pre- (left; n = 21) or post-menopausal (right; n = 17) women. Each dot represents a single patient. Spearman test

In previous studies, we found that EM CD8+ T cells consist of both tissue resident cells and non-resident CD8+ T cells that are CD103+ and CD103-, respectively [13, 14]. To determine the extent to which age influences their presence in the EM, CD8+ T cells were divided into CD103+ and CD103- cells. As seen in Supplementary Fig. 1B, the number of CD103+ (left) or CD103- (right) CD8+ T cells per gram of tissue indicated no differences between pre- and post-menopausal women. As a percentage of total CD8+ T cells, the percent of CD103+ from pre- and post-menopause were 41 and 49%, respectively. However, when stratified by age, as shown in Fig. 1C and D, both CD103+ and CD103-CD8+ T cell numbers decreased significantly with increasing age in post-menopausal women (range 49–81 years; n = 17). In contrast, there were no differences in CD103+ and CD103-CD8+ T cell numbers with increasing age in pre-menopausal women (range 28–51 years; n = 21).

CD8+ T cells cytotoxic activity increases with age in the endometrium

We have demonstrated previously that EM CD8+ T cell cytotoxic activity varies with stage of the menstrual cycle and increases following menopause [12, 14]. To determine the impact of age on cytotoxicity, cytotoxic killing by EM CD8+ T cells from pre- and post-menopausal women was analyzed and correlated with age. To measure cytotoxic activity, CD8+ T cells were incubated with CFSE-labelled allogeneic target cells at an Effector:Target ratio of 1:1, and the number of dead cells quantified by time-lapse imaging as described before [14, 17, 18]. Cytotoxicity was calculated by measuring the average number of dead target cells over the first 4 h and data was normalized as the fold change in number of dead target cells to be able to compare between experiments with different background mortality in the target cell alone control as previously described [14, 17, 18]. Consistent with our previous findings, the number of dead target cells in the presence of EM CD8+ T cells was significantly increased relative to target cells alone over a period of 4 h (Supplementary Fig. 2A), and EM CD8+ T cell cytotoxic killing was significantly higher in post-menopausal compared to pre-menopausal women (Supplementary Fig. 2B). When we stratified CD8+ T cell cytotoxicity as a function of age across the entire study population, we found that EM CD8+ T cell cytotoxicity increased significantly with increasing age (Fig. 2A). However, this increase in cytotoxic killing was due to the enhanced activity in CD8+ T cells from post-menopausal women with increasing age (P = 0.01) (Fig. 2B), while there was no effect in the pre-menopausal group (P = 0.18).

Fig. 2figure 2

CD8+ T cells cytotoxic activity increases with age in the EM. Purified EM CD8+ T cells (or CD103+ or CD103− as indicated) were co-cultured with CFSE-stained allogeneic blood CD4+ T cells as target cells (ratio 1:1). Cytotoxicity was calculated by measuring the average number of dead target cells over the first 4 h. Graph represents the fold change in number of dead target cells in CD8+ T cells + target cells cultures compared to target cells alone. A Correlation between CD8+ T cells cytotoxicity and age from the entire study population (n = 53). B Correlation between CD8+ T cells cytotoxicity and age after separating pre- (left; n = 22) and post-menopausal cells (right; n = 31) women. C Correlation between CD103 + CD8+ T cells (left) or CD103-CD8+ T cells (right) cytotoxicity and age from the entire study population (n = 18). D Correlation between the difference of matched CD103+ and CD103-CD8+ T cells cytotoxicity and age (n = 18). Each dot represents a single patient. Spearman test

We then compared cytotoxic killing between EM CD103+ and CD103-CD8+ T cells and evaluated correlation with age across our entire patient population. As seen in Fig. 2C, cytotoxic killing in both the CD103+ and CD103-CD8+ T cell population significantly increased with increasing age. We have previously reported that cytotoxic killing by CD103-CD8+ T cells is significantly higher than that of CD103 + CD8+ T cells [14]. Therefore, here we investigated whether this difference was maintained with aging, and found that the difference of cytotoxic killing between CD103- and CD103+ CD8+ T cells was significantly increased with increasing age (Fig. 2D). These findings indicate that EM CD103- CD8+ T cell activity was preferentially and progressively enhanced relative to that seen with CD103 + CD8+ T cells in postmenopausal women as they age. Overall, these studies indicated that under conditions in which CD103+ and CD103-CD8+ T cell numbers decline with aging following menopause, there is a compensatory increase in CD103-CD8+ T cell cytotoxic activity.

Aging and menopause differentially regulate intracellular cytotoxic molecules in endometrial CD8+ T cells

To understand the mechanisms involved in the observed age-induced changes in CD8+ T cell cytotoxicity, CD8+ T cells in mixed cell suspensions from EM tissues were analyzed by flow cytometry for intracellular expression of the cytotoxic molecules perforin (PRF), granzyme A (GZMA) and granzyme B (GZMB) under resting conditions as described before [14, 17, 18]. As seen in Fig. 3A, overall, 80 and 70% of EM CD8 + T cells were GZMA+ and GZMB+ respectively. The percentage of GZMA+ and GZMB+CD8+ T cells were significantly greater than PRF + CD8+ T cells, which accounted for less than 10% of the total EM CD8+ T cell population. The percentage of GZMA+CD8+ T cells was significantly higher than the percentage of GZMB. Considering these observations, we then calculated the ratio between GZMA and GZMB expressing CD8+ T cells and detected a significant increase in GZMA/GZMB ratio with increasing age (Fig. 3B). Analysis of menopausal status demonstrated a significant increase in the percentage of GZMA+ and GZMB+CD8+ T cells (Fig. 3C) and GZMA/GZMB ratio (Fig. 3D) when post-menopausal women were compared to pre-menopausal women, with no changes detected in the percentage of PRF+ CD8+ T cells. Interestingly, when stratified by age (Fig. 3E), the percentage of GZMA+ and GZMB+CD8+ T cells increased significantly with increasing age in pre-menopausal women prior to menopause. In contrast, the percentage of GZMA+ and GZMB+CD8+ T cells in post-menopausal women remained elevated and constant with increasing age. In contrast no changes were seen in the percentage of PRF + CD8+ T cells when pre- or post-menopausal populations were analyzed with increasing age. Changes in the percentage of GZMA+ and GZMB+ cells following menopause suggest an explanation for increased cytotoxic activity of the total CD8+ T cells as women age.

Fig. 3figure 3

Aging and menopause differentially regulate intracellular cytotoxic molecules in EM CD8+ T cells. Mixed cell suspensions from EM tissues were stained for the intracellular cytotoxic molecules perforin, granzyme A and granzyme B for analysis by flow cytometry. A Percentage positive cells of perforin, granzyme A and granzyme B in EM CD8+ T cells (n = 27). B Correlation between age and the ratio of Granzyme A/Granzyme B in EM CD8+ T cells (n = 27). C Comparison of pre- vs. post-menopausal women percentage positive cells of perforin, granzyme A and granzyme B in EM CD8+ T cells. D Comparison of pre- vs. post-menopausal women the ratio of Granzyme A/Granzyme B from EM CD8+ T cells. E Correlation between age and the percentage positive cells of perforin (left), granzyme A (right) and granzyme B (middle) in EM CD8+ T cells from pre- and post-menopausal women. Pre-menopausal women (black circle; n = 10), post-menopausal women (white circle; n = 17). Each dot represents a different patient. Mean ± SEM are shown. *P < 0.05, **P < 0.01, ***P < 0.001; Friedman test followed by Dunns post-test (A), Spearman test (B, E). Kruskal-Wallis test followed by Dunns post-test or Friedman test followed by Dunns post-test (C), Mann–Whitney U-test (D)

Aging differentially regulates intracellular cytotoxic molecules in endometrial CD103+ and CD103-CD8+ T cells

Since we observed differential aging effects on the cytotoxic activity of CD103+ and CD103-CD8+ T cells, we next analyzed cytotoxic molecules in these populations by flow cytometry as previously described [14, 17, 18]. As seen in Fig. 4A, GZMA+ and GZMB+ were significantly more abundant in CD103- than CD103+ cells. GZMA expression was significantly greater than GZMB in both CD103+ and CD103-CD8+ T cells. The ratio of GZMA/GZMB in both CD103+ and CD103-CD8+ T cells significant increased with increasing age (Fig. 4B). When analyzed according to menopausal status (Supplementary Fig. 3A), we found that the percentage of GZMA+ cells in both the CD103+ and CD103-CD8+ T cell population increased significantly in post-menopausal women compared to pre-menopausal women. The ratio of GZMA/GZMB in CD103-CD8+ T cells, but not in CD103 + CD8+ T cells, was significantly higher in post-menopausal women compared to pre-menopausal women (Supplementary Fig. 3B).

Fig. 4figure 4

Aging differentially regulates intracellular cytotoxic molecules in EM CD103+ and CD103-CD8+ T cells. Mixed cell suspensions from EM tissues were stained for the intracellular cytotoxic molecules perforin, granzyme A and granzyme B prior to analysis by flow cytometry. A Percentage positive cells expressing perforin, granzyme A and granzyme B in EM CD103+ (grey circle) and CD103- (black square) CD8+ T cells (n = 27). B Correlation between age and the ratio of Granzyme A/Granzyme B in EM CD103+ (left) or CD103- (right) CD8+ T cells (n = 27). C, D Correlation between age and the percentage positive cells of perforin (left), granzyme A (right) and granzyme B (middle) in EM CD103 + CD8+ T cells (C) or in CD103-CD8+ T cells (D) from pre- (black circle; n = 10) and post-menopausal (white circle; n = 17) women. Each dot represents a different patient. Mean ± SEM are shown. *P < 0.05, **P < 0.01, ***P < 0.001; Friedman test followed by Dunns post-test (A), Spearman test (B-D)

Next, we analyzed the percent of cells positive for cytotoxic molecules as a function of age in CD103+ and CD103-CD8+ T cells. As seen in Fig. 4C and D, the percentage of GZMA+ and GZMB+ cells in both the CD103+ and CD103-CD8+ T cell population increased significantly with age in pre-menopausal women and plateaued at a high level after menopause irrespective of increasing age. In contrast, there was no effect of age on the percentage of PRF+ in either the CD103+ and CD103-CD8+ T cells from pre- or post-menopausal women. Overall, these studies demonstrate that the percentage of GZMA+ and GZMB+ EM CD8+ T cells progressively increases with increasing pre-menopausal age, reaching their peak prior to menopause, and maintaining these levels in the post-menopausal period.

Aging and menopause differentially regulate the production of TNFα, IL-6 and IFNγ by endometrial CD8+ T cells

In addition to direct killing, CD8+ T cells are known to exert their actions through the secretion of cytokines and chemokines [19]. To test the hypothesis that CD8+ T cells secretion varies with age, studies were undertaken to measure TNFα, IL-6 and IFNγ production by resting CD8+ T cells as a function of aging. Purified EM CD8+ T cells were incubated for 48 h after which supernatants were collected and assayed for TNFα, IL-6 and IFNγ by Luminex assay. Data was calculated according to the cell number to compare baseline production or normalized as the fold change in baseline production to determine the effect of sex hormones. As seen in Fig. 5A, CD8+ T cells constitutively produced TNFα (23.1 ± 2.5 pg/million cells), IL-6 (421.3 ± 75.4 pg/million cells), and IFNγ (14.2 ± 2.6 pg/million cells) under resting conditions. When secretion of all three cytokines was analyzed based on menopausal status (Fig. 5B), we found significantly higher secretion of TNFα by post-menopausal CD8+ T cells compared to pre-menopausal CD8+ T cells. In contrast, there were no differences in IL-6 and IFNγ CD8+ T cell secretion between pre- and post-menopausal women. When CD8+ T cells were stratified by age, the production of TNFα (Fig. 5C) but not IL-6 (Fig. 5D) and IFNγ (Fig. 5E) increased significantly with increasing age in the post-menopausal population. There was no effect of age on the secretion of the three cytokines by pre-menopausal women.

Fig. 5figure 5

Aging and menopause differentially regulate the production of TNFα, IL-6 and IFNγ by EM CD8+ T cells. Resting purified EM CD8+ T cells were cultured with or without E2 (5 × 10− 8 M) or P (1 × 10− 7 M) for 48 h. The culture supernatants were collected and assayed for TNFα, IL-6 and IFNγ by Luminex assay. A Constitutive production of TNFα, IL-6 and IFNγ by EM CD8+ T cells after 48 h in the culture (n = 38). B Comparison of pre- (black circle; n = 17) vs. post-menopausal (white circle; n = 21) women constitutive production of TNFα, IL-6 and IFNγ by EM CD8+ T cells after 48 h in the culture. C-E Correlation between age and the production of TNFα (C), IL-6 (D) and IFNγ (E) by EM CD8+ T cells from pre- (n = 17) and post-menopausal (n = 21) women. F Effect of E2 or P on the production of TNFα, IL-6 and IFNγ by EM CD8+ T cells after 48 h in the culture. Graph represents the fold change in TNFα, IL-6 and IFNγ in culture supernatant following E2 or P treatment compared with untreated controls (E2: n = 27; P: n = 21). Each dot represents a single patient. Mean ± SEM are shown. **P < 0.01; Mann–Whitney U-test (B), Spearman test (C-E), Kruskal-Wallis test followed by Dunns post-test or Friedman test followed by Dunns post-test (F)

In prior studies [17], we discovered that estradiol (E2) and progesterone (P) act both directly and indirectly to suppress CD8+ T cell cytotoxicity. To determine if sex hormones regulate the secretion of cytokines and chemokines, purified EM CD8+ T cells were incubated with E2 (5 × 10− 8 M) or P (1 × 10− 7 M) for 48 h prior to Luminex analysis. As shown in Fig. 5F, E2 but not P significantly suppressed TNFα secretion by EM CD8+ T cells, with no effect of either sex hormone on IL-6 or IFNγ secretion. Overall, these studies indicate that following menopause and with increasing post-menopausal age, constitutive secretion of TNFα by CD8+ T cells increases in response to the absence of E2, possibly to enhance immune protection in the EM, at a time when protection is declining.

Aging and menopause differentially regulate the production of TNFα, IL-6 and IFNγ by endometrial CD103+ and CD103-CD8+ T cells after activation

In previous studies, we identified differences after activation of resident and non-resident CD8+ T cells and found that EM CD103 + CD8+ T cells had significantly higher degranulation following activation compared to CD103-CD8+ T cells, but that higher CD103+ CD8+ T cell degranulation did not result in the increased release of cytotoxic molecules [14]. Whether aging affects the secretion of TNFα, IL-6, or IFNγ by CD103+ and CD103-CD8+ T cells remains unknown. We therefore analyzed the secretions of all three cytokines following activation and stratified these results according to menopausal status and age. Purified EM CD103+ and CD103-CD8+ T cells were activated with PMA (100 ng/ml) and ionomycin (2 μM) for 24 h as previously described [14]. The culture supernatants were collected and assayed for TNFα, IL-6 and IFNγ by Luminex assay. Data were calculated according to the cell number and normalized as the fold change in baseline production to determine response to PMA stimulation. As shown in Fig. 6A, secretion of TNFα and IFNγ increased significantly in both CD103+ and CD103-CD8+ T cells after 24 h stimulation with PMA, with no change in secretion of IL-6. When stratified by menopausal status (Fig. 6B), secretion of TNFα by both CD103+ and CD103-CD8+ T cells significantly increased in post-menopausal women. In contrast, IL-6 and IFNγ secretion by CD103-CD8+ T cells increased in post-menopausal but not in CD103 + CD8+ T cells.

Fig. 6figure 6

Aging and menopause differentially regulate the production of TNFα, IL-6 and IFNγ by EM CD103+ and CD103-CD8+ T cells after activation. Purified EM CD103+ and CD103-CD8+ T cells were stimulated with PMA (100 ng/ml) and ionomycin (2 μM) for 24 h. The culture supernatants were collected and assayed for TNFα, IL-6 and IFNγ by Luminex assay. A Production of TNFα, IL-6 and IFNγ by EM CD103+ (grey circle) and CD103- (black square) CD8+ T cells after 24 h stimulation (n = 20). C represents untreated control. S represents stimulation with PMA/ionomycin. B Comparison of pre- (black circle; n = 13) vs. post-menopausal (white circle; n = 7) women production of TNFα, IL-6 and IFNγ by EM CD103+ (left) and CD103- (right) CD8+ T cells after 24 h stimulation. C, D Correlation between age and the production of TNFα, IL-6 and IFNγ by EM CD103 + CD8+ T cells (C) and CD103-CD8+ T cells (D) after 24 h stimulation. Graphs of (B-D) represent the fold change in TNFα, IL-6 and IFNγ in culture supernatant following stimulation compared with untreated controls (n = 20). Each dot represents a single patient. Mean ± SEM are shown. *P < 0.05, **P < 0.01, ****P < 0.0001; Friedman test followed by Dunns post-test (A), Kruskal-Wallis test followed by Dunns post-test (B), Spearman test (C, D)

We then stratified secretion of TNFα, IL-6, and IFNγ by both activated CD103+ and CD103-CD8+ T cells with increasing age. As seen in Fig. 6C and D, TNFα, IL-6 and IFNγ secretion by CD103-CD8+ T cells significantly increased with increasing age. In contrast, only TNFα secretion increased with increasing age in CD103 + CD8+ T cells. Overall, these studies indicate that with age, CD8+ T cells are able to respond to potential challenges with increased secretion of pro-inflammatory mediators that are capable of mounting immune protection.

留言 (0)

沒有登入
gif