Oxidized LDL modulates apoptosis of regulatory T cells in patients with ESRD

Abstract

Increased levels of oxidized low-density lipoproteins (oxLDL) contribute to the increased risk for atherosclerosis, which persists even after adjusting for traditional risk factors, among patients with ESRD. Regulatory T cells (CD4+/CD25+ Tregs), which down-regulate T cell responses to foreign and self-antigens, are protective in murine atherogenesis, but whether similar immunoregulation occurs in humans with ESRD is unknown. Because cellular defense systems against oxLDL involve proteolytic degradation, the authors investigated the role of oxLDL on proteasome activity of CD4+/CD25+ Tregs in patients with ESRD. CD4+/CD25+ Tregs isolated from uremic patients' peripheral blood, especially that of chronically hemodialyzed patients, failed to suppress cell proliferation, exhibited cell-cycle arrest, and entered apoptosis by altering proteasome activity. Treating CD4+/CD25+ Tregs with oxLDL or uremic serum ex vivo decreased the number and suppressive capacity of CD4+/CD25+ Tregs. In vitro, oxLDL promoted the accumulation of p27Kip1, the cyclin-dependent kinase inhibitor responsible for G1 cell cycle arrest, and increased apoptosis in a time- and concentration-dependent manner. In summary, proteasome inhibition by oxLDL leads to cell cycle arrest and apoptosis, dramatically affecting the number and suppressive capacity of CD4+/CD25+ Tregs in chronically hemodialyzed patients. This response may contribute to the immune dysfunction, microinflammation, and atherogenesis observed in patients with ESRD.

Figure 1.

Plasma concentrations of oxidized low-density lipoproteins (oxLDLs) in the study groups of patients and control subjects. Plasma oxLDL concentration in U/L in hemodialysis (HD) patients (n = 15) ▪, peritoneal dialysis (PD) patients (n = 15), chronic kidney disease (CKD) patients (n = 15), and in control subjects (n = 15) □. Box-and-whisker plots are used to represent the distributions. The bottom and the top of a box represent the 25th and 75th percentiles, and the line in the box shows the median (50^th percentile). Whiskers extend on either side of the box and aim to cover all observations, but they never exceed 1.5 times the height of the box (i.e., the interquartile range). Any values outside the whiskers are plotted individually and are considered possible outliers. On a normal distribution, the box and whiskers cover roughly 99% of the population. *P < 0.01 (ANOVA)

Figure 2.

Frequency of circulating CD4⁺/CD25⁺ T cells in the study groups of patients and control subjects ex vivo, and after phytohemagglutinin (PHA) stimulation in vitro. (A) CD4⁺ T cells were separated into CD25^high, CD25^low and CD25⁻ T cell subsets, defined by the fluorescence intensity of CD25 obtained using flow cytometry as described (see Concise Methods for details). Flow cytometry analysis from a hemodialyzed (HD) patient (I.), a peritoneal dialysis (PD) patient (II.), a chronic kidney disease (CKD) patient (III.), and a control subject (IV.), all representative. PE, phycoerythrin. (B) Frequency of CD4⁺/CD25^high population as percentage of CD4⁺ T cells in hemodialyzed (HD) patients (n = 15) ▪, peritoneal dialysis (PD) patients (n = 15), chronic kidney disease (CKD) patients (n = 15), and in control subjects (n = 15) □. CD4⁺ T cells were cultured in vitro after 24-h PHA stimulation. *P < 0.01 (ANOVA). Data are expressed as mean ± SEM. (C) The frequency of CD4⁺/CD25^high T cells from control subjects (n = 5) was examined after incubation in 10% human uremic serum (three sera from HD patients [C + 10% HD] ▪, three from PD patients [C + 10% PD], or three from CKD patients [C + 10% CKD]) after 24-h PHA stimulation. Data are expressed as box-and-whisker plots. The bottom and the top of a box represent the 25th and 75th percentiles, and the line in the box shows the median (50th percentile). Whiskers extend on either side of the box and aim to cover all observations, but they never exceed 1.5 times the height of the box (i.e., the interquartile range). On a normal distribution, the box and whiskers cover roughly 99% of the population. *P < 0.01 (ANOVA).

Figure 3.

Forkhead family transcription factor 3 (FOXP3) expression in stimulated CD4⁺/CD25⁺ T cells. (A) FOXP3 mRNA levels in phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁺ T cells from the study groups of patients and control subjects are presented as the cycle threshold value ratio of FOXP3/β-actin, in hemodialyzed (HD) patients (n = 15) ▪, peritoneal dialysis (PD) patients (n = 15), chronic kidney disease (CKD) patients (n = 15), and control subjects (n = 15) □. *P < 0.01 (ANOVA). Data are expressed as mean ± SEM. (B) The frequency of CD4⁺, CD25⁺, and FOXP3⁺ T cells was determined by flow cytometry in hemodialyzed (HD) patients (n = 15) ▪, peritoneal dialysis (PD) patients (n = 15), chronic kidney disease (CKD) patients (n = 15), and control subjects (n = 15) □ Percentage of FOXP3⁺ cells within the CD4⁺/CD25^high, CD4⁺/CD25^low and CD4⁺/CD25⁻ T cell populations after PHA stimulation. *ANOVA. Data are expressed as mean ± SEM. (C) Mean fluorescence intensity (MFI) of FOXP3 from PHA-stimulated CD4⁺/CD25^high T cells in patients and controls. HD patients (n = 15) ▪, PD patients (n = 15), CKD patients (n = 15), and control subjects (n = 15) □. Indicated values (shown as box and whiskers) represent MFI signals obtained from FOXP3 stainings. Mean fluorescence intensity values of cells stained by isotype-matched control antibody (background MFI) were subtracted from FOXP3 MFI values. *P < 0.01 (ANOVA). AU, arbitrary unit

Figure 4.

Oxidized LDL (oxLDL) down-regulation of stimulated CD4⁺/CD25^high/FOXP3⁺ T cells (CD4⁺/CD25⁺ Tregs) and FOXP3 expression in stimulated CD4⁺/CD25^high T cells. (A) Percentage of phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁻ T cells from control subjects (n = 15) observed by flow cytometry after 48 h in presence of native LDL □ or oxidized LDL (oxLDL) ▪. Percentage of PHA-stimulated CD4⁺/CD25⁻ T cells from control subjects (n = 5) in presence of 10% uremic serum from hemodialyzed (HD) patients and from HD patients in culture medium (HD medium; n = 5) was also analyzed for comparison. *P = nonsignificant, compared with native LDL-treated T cells for the same concentration of lipoproteins; **P = nonsignificant, compared with oxLDL-treated T cells (200 μg/ml). Data are expressed as mean ± SEM. (B) Same as Figure A with CD4⁺/CD25^high/FOXP3⁺ T cells (CD4⁺/CD25⁺ Tregs). *P = 0.01; **P < 0.001, compared with native LDL-treated T cells for the same concentration of lipoproteins; ***P = nonsignificant; ^#P = 0.01, compared with oxLDL-treated T cells (200 μg/ml). Data are expressed as mean ± SEM. (C) Expression of FOXP3 mRNA in PHA-stimulated CD4⁺/CD25^high T cells from control subjects. Electrophoresis of FOXP3 RT-PCR products in 1.5% agarose gel after competitive PCR. The bands corresponding to FOXP3 and β-actin mRNAs were densitometrically scanned. The results are expressed as a full-length FOXP3/β-actin mRNAs ratio. Total mRNA was isolated from purified CD4⁺/CD25^high T cells of control subjects (n = 5) after 48 h in presence of native LDL □ or oxidized LDL (oxLDL) ▪, reverse transcribed (as described in the Concise Methods section) and assayed for the FOXP3 mRNA expression levels. PHA-stimulated CD4⁺/CD25^high T cells from control subjects (n = 5) in presence of 10% uremic serum from hemodialyzed (HD) patients and from HD patients in culture medium (n = 5) were also analyzed for comparison. *P < 0.001, compared with native LDL-treated T cells for the same concentration of lipoproteins; **P = 0.01; ***P = nonsignificant, compared with oxLDL-treated T cells (200 μg/ml). Data are expressed as mean ± SEM. (D) Western blot analyses of FOXP3 in CD4⁺/CD25^high T cells after PHA stimulation. CD4⁺/CD25^high T cells from one control subject were cultured in culture medium (C medium) and in presence of 100 μg/ml and 200 μg/ml oxLDL (C + oxLDL 100 and C + oxLDL 200, respectively) for 48 h (see Concise Methods for details). CD4⁺/CD25^high T cells from one HD patient were cultured in culture medium (HD medium). Values obtained by densitometric analysis of Western blots for FOXP3 were expressed relative to the control values. Expression of β-actin was used to control equal protein loading. *P = 0.01, compared with control values (C medium); **P = nonsignificant, compared with HD medium. Data are expressed as mean ± SEM of three independent experiments.

Figure 5.

CD4⁺/CD25⁺ Tregs suppression capacity in coculture with CD4⁺/CD25⁻ T cells. (A) CD4⁺/CD25⁺ Tregs from patients (five patients in each group) and control subjects (n = 5) were analyzed in coculture to assess their suppression capacity. CD4⁺/CD25⁺ Tregs (2 × 10⁴) and CD4⁺/CD25⁻ T cells (2 × 10⁴) were phytohemagglutinin (PHA) stimulated alone and in coculture (1:1 ratio) with 2 × 10⁴ CD4⁺/CD25⁻ responder T cells in the presence of 1 × 10⁵ antigen-presenting cells. The CD4⁺/CD25⁻ T cells were also stimulated alone. Data are representative of three independent experiments and are presented as the mean of proliferation at day 5 ± SE. The stimulus used for activation was PHA at the concentration of 2.5 μg/ml. CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were cultured in 10% uremic serum from HD patients (C + 10% HD). The control cells (C) were also incubated with 200 μg/ml native LDL (C + LDL 200) or 200 μg/ml oxidized LDL (oxLDL) for 48 h (C + oxLDL 200). Cell proliferation was measured by [³H]-thymidine-uptake assay. The mean radioactivity (counts per minute × 10³) was used for calculations. Data are expressed as mean ± SEM. (B) In the cocultured wells, the number of CD4⁺/CD25⁻ responder T cells was constant, whereas the number of CD4⁺/CD25⁺ Tregs varied by serial threefold dilution. IFN-γ levels were assayed by ELISA from supernatants removed from the cultures just before [³H] thymidine addition.

Figure 6.

20 S and 26 S proteasome activity in stimulated CD4⁺/CD25⁺ Tregs in vivo and in vitro. (A) Phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁺ Tregs from hemodialysis (HD) patients (n = 5) ▪, peritoneal dialysis (PD) patients (n = 5), chronic kidney disease (CKD) patients (n = 5), and control subjects (n = 5) □ were analyzed for proteasome activity. Purified 20 S and 26 S proteasome were incubated with fluorogenic peptide substrates for chymotrypsin-like activity. After incubation, the fluorescence intensity of AMCs was determined as described in the Concise Methods section.*P = 0.003 (20 S proteasome activity); *P = 0.001 (26 S proteasome activity). Data are expressed as mean ± SEM. (B) Inhibitory effect of oxLDL on chymotrypsin-like activity of the 26 S proteasome in intact but PHA-stimulated CD4⁺/CD25⁺ Tregs. CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were incubated with various concentrations of native LDL □, oxidized LDL (oxLDL) ▪ for 48 h, or lactacystin for 12 h, and then PHA stimulated and coincubated for 2 h with fluorogenic peptide substrate for chymotrypsin-like proteasome activity. After incubation, the medium was collected and subsequently subjected to a proteasome activity assay as described in the Concise Methods section. The chymotrypsin-like activity was expressed as the percentage of the control (defined as 100%). LDL and oxLDL concentrations were in μg/ml, and lactacystin concentrations were in μM. *P = non significant. **P < 0.001, compared with LDL-treated T cells for the same concentration of lipoproteins or for lactacystin as mentioned. Data are expressed as mean ± SEM.

Figure 7.

Cell cycle arrest at G₁ phase and accumulation of proteasomal degradation-related protein p27^Kip1 in stimulated CD4⁺/CD25⁺ Tregs. Cells were divided into two ex vivo portions: one for measurement of cell cycle distribution by flow cytometry (A) and the other for total protein extraction and quantitation of p27^Kip1 protein by Western blotting (B). CD4⁺/CD25⁺ Tregs from a control subject were analyzed in vitro (C). (A) Cell cycle analysis of phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁺ Tregs was done by flow cytometry for the indicated times. Data shown are expressed from one representative hemodialysis (HD) patient and one control subject out of five analyzed. (B) PHA-stimulated CD4⁺/CD25⁺ Tregs from HD patients (n = 5) ▪, peritoneal dialysis (PD) patients (n = 5), chronic kidney disease (CKD) patients (n = 5), and control subjects (n = 5) □ were analyzed for Western blotting analysis of p27^Kip1. Simultaneous immunoblotting of β-actin was used as an internal control for equivalent protein loading. Data are expressed as mean ± SEM. (C) PHA-stimulated CD4⁺/CD25⁺ Tregs from one HD patient in culture medium ▪ were analyzed for proteasomal degradation-related protein p27^Kip1. Stimulated CD4⁺/CD25⁺ Tregs from one control subject in presence of 10% uremic serum (HD patients) (C + 10% HD), with oxLDL 200 μg/ml (C + oxLDL 200) ▪, with native LDL 200 μg/ml (C + LDL 200) □, and with the proteasome inhibitor lactacystin (C + lactacystin) were also determined for Western blotting analysis to evaluate proteasomal degradation-related protein p27^Kip1. Simultaneous immunoblotting of β-actin was used as an internal control for equivalent protein loading. *P = non significant. **P < 0.001, compared with oxLDL-treated T cells (200 μg/ml). Data are expressed as mean ± SEM from three separate experiments.

Figure 8.

Correlation of growth inhibition induced by oxidized LDL (oxLDL), corresponding high level and half-life increase of p27^Kip1 protein. (A) Phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁺ Tregs from a control subject were treated with oxLDL or native LDL at the concentrations indicated. Anchorage-dependent growth assay was performed in 96-well microplates, and measurement of p27^Kip1 protein was determined by Western blotting analysis as described in the Concise Methods section. CD4⁺/CD25⁺ Tregs at similar cell density to the microplate growth assay were plated into 100-mm culture dishes and treated for 48 h with oxLDL or native LDL at similar conditions. (B) PHA-stimulated CD4⁺/CD25⁺ Tregs from a control subjects observed after 48 h in presence of oxLDL or native LDL at the same concentrations (200 μg/ml). Cells were then washed, harvested and subjected to total protein preparation and Western blotting analysis with anti-p27^Kip1. Semiquantitation of p27^Kip1 expression is represented. Results were normalized according to the level of β-actin control. The p27^Kip1 expression of both oxLDL and native LDL group at the 0-h interval was set as 1 unit. (C) CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were cultured in culture medium (C medium) or in 10% uremic serum from HD patients (C + 10% HD). Cells were harvested and subjected to total protein preparation and Western blotting analysis with anti-p27^Kip1 and anti-β-actin antibodies. Western blot shown is representative of three replicate analyses.

Figure 9.

(A) Accumulation of proteasomal degradation-related protein Bax but not Bcl-x_L in stimulated CD4⁺/CD25⁺ Tregs. Western blot and corresponding mean fluorescence intensity (MFI) were calculated for Bax ▪ and Bcl-x_L □ in phytohemagglutinin (PHA)-stimulated CD4⁺/CD25⁺ Tregs from hemodialyzed (HD) patients (n = 5) in culture medium (HD medium), in presence of 10% human normal serum (C) (HD + 10% C), in PHA-stimulated CD4⁺/CD25⁺ Tregs from control subjects (n = 5), in culture medium (C medium), and after incubation in 10% human uremic serum (HD patients) (C + 10% HD). Furthermore, PHA-stimulated CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were cultured in the presence of oxidized LDL (oxLDL; 200 μg/ml) (C + oxLDL 200) and native LDL at the same concentration (C + LDL 200) for 48 h. Then, the cells were analyzed as described above for proteasomal degradation-related protein Bax and Bcl-x_L. Box-and-whisker plots are used to represent the distributions. The bottom and the top of a box represent the 25th and 75th percentiles, and the line in the box shows the median (50th percentile). Whiskers extend on either side of the box and aim to cover all observations, but they never exceed 1.5 times the height of the box (i.e., the interquartile range). Any values outside the whiskers are plotted individually and are considered possible outliers. On a normal distribution, the box and whiskers cover roughly 99% of the population. (B) Fas expression in stimulated CD4⁺/CD25⁺ Tregs in vitro. PHA-stimulated CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were cultured for 48 h in presence of native LDL □ or oxidized LDL (oxLDL) ▪ at various concentrations. The cells were then Western blotted for Fas protein determination. Simultaneous immunoblotting of α-tubulin was used as an internal control for equivalent protein loading. Fas protein determination in PHA-stimulated CD4⁺/CD25⁺ Tregs from control subjects (n = 5) in presence of 10% uremic serum from hemodialyzed (HD) patients (C + 10% HD) ▒ and from HD patients in culture medium (n = 5) (HD medium) ▪ was also analyzed for comparison. *P = 0.01; **P < 0.001, compared with LDL-treated T cells for the same concentration of lipoproteins; ***P = 0.01; ^#P = nonsignificant, compared with oxLDL-treated T cells (200 μg/ml). Data are expressed as mean ± SEM. (C) Agarose gel analysis of cell DNA fragmentation in stimulated CD4⁺/CD25⁺/Fas⁺ Tregs. PHA-stimulated CD4⁺/CD25⁺ Tregs from control subjects (n = 5) were cultured for 48 h in presence of native LDL or oxLDL at various concentrations. Lanes represent Fas-sorted CD4⁺/CD25⁺ Tregs DNA isolated from a control subject cultured with native LDL or oxLDL as mentioned. DNA fragmentation was also determined in PHA-stimulated CD4⁺/CD25⁺/Fas⁺ Tregs from control subjects (n = 5) in presence of 10% uremic serum (HD patients) and from HD patients in culture medium (n = 5) for comparison.