Reduced p53 in peripheral blood mononuclear cells from patients with rheumatoid arthritis is associated with loss of radiation-induced apoptosis.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 52, No. 4, April 2005, pp 1047–1057 DOI 10.1002/art.20931 © 2005, American College of Rheumatology Reduced p53 in Peripheral Blood Mononuclear Cells From Patients With Rheumatoid Arthritis Is Associated With Loss of Radiation-Induced Apoptosis Kevin Maas, Matthew Westfall, Jennifer Pietenpol, Nancy J. Olsen, and Thomas Aune apoptosis in patients with RA. We hypothesize that this liability may contribute to autoimmunity. Objective. Patients with autoimmune disorders exhibit highly reproducible gene expression profiles in their peripheral blood mononuclear cells. This profile includes, at least in part, a collection of underexpressed genes that encode proteins that inhibit cell cycle progression and stimulate apoptosis. We aimed to determine whether this gene expression profile confers functional liability on lymphocytes from patients with rheumatoid arthritis (RA). Methods. Viability studies in response to a panel of proapoptotic stimuli revealed that T lymphocytes from patients with RA were resistant to gamma radiation–induced apoptosis, a process known to be dependent on p53. To assess p53 function in RA peripheral blood mononuclear cells, baseline levels of p53 protein and TP53 transcript were measured in patients with RA and controls. The cellular p53 response to gamma radiation was also assessed by immunoblotting. Results. Lymphocytes from patients with RA had lower baseline levels of TP53 messenger RNA (mRNA) and p53 protein than did those from control subjects and were deficient in their ability to increase p53 after exposure to gamma radiation. A subgroup of patients with RA had a second biochemical defect characterized by expression of very low baseline levels of checkpoint kinase 2 mRNA and protein. Conclusion. We conclude that defects in the expression of TP53 mRNA and, in a subgroup, defects in expression of CHK2 mRNA, lead to severe defects in Autoimmune diseases, which are characterized by the site(s) of immune-mediated destruction, affect 3–5% of the population. Autoimmune disorders range from those associated with organ-specific attacks, as in multiple sclerosis (MS) and type I diabetes (insulindependent diabetes mellitus [IDDM]), to those with more systemic manifestations, as in rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE). Despite the prevalence of these disorders, the mechanisms underlying their pathogenesis remain poorly understood (1). There is also significant clinical heterogeneity within patient groups, with variation in disease activity, time of onset, and response to therapy (2). Because of this clinical heterogeneity, finding common underlying defects in human autoimmune disease has proven difficult (3). There are numerous ways to induce clinical manifestations of autoimmunity in animal models (4). This raises the following question: are there also numerous underlying mechanisms that cause human autoimmune disease? Recent studies examining gene expression profiles in peripheral blood mononuclear cells (PBMCs) revealed identifiable similarities among patients with autoimmune disease. For example, interferon-␣ (IFN␣)– and IFN␤-responsive gene expression signatures are present in PBMCs from both children (5) and adults (6) with severe lupus, supporting a potential role of these cytokines in SLE pathogenesis. IFN signatures are linked to disease activity. Common genes encoding proteins that regulate cell cycle progression and apoptosis are differentially expressed in many autoimmune disorders (7–9). In particular, our previous study revealed a common gene expression profile among patients with RA, SLE, IDDM, or MS (8). This expression signature was ob- Supported by grants from the NIH (T32-GM-07347, RR00095, DK-58765, AI-44924, AI-53984, and CA-070856). Kevin Maas, BS, Matthew Westfall, BS, PhD, Jennifer Pietenpol, BS, PhD, Nancy J. Olsen, MS, MD, Thomas Aune, BS, PhD: Vanderbilt University, Nashville, Tennessee. Address correspondence and reprint requests to Thomas Aune, BS, PhD, Vanderbilt University Medical Center, MCN T3219, 1161 21st Avenue South, Nashville, TN 37232. E-mail: thomas. firstname.lastname@example.org. Submitted for publication April 6, 2004; accepted in revised form December 13, 2004. 1047 1048 MAAS ET AL served in all of the patient populations studied and perfectly discriminated autoimmune patients from controls. In contrast to the expression patterns described in the lupus studies cited above, this expression pattern does not correlate with disease severity and is present in unaffected first-degree relatives, suggesting that there is a heritable component. To gain insight into the significance of these altered gene expression profiles, we examined lymphocytes obtained from patients with RA for defects in apoptosis. We observed a marked resistance to gamma radiation–induced apoptosis in patients with RA. The tumor suppressor protein p53 plays a key role in the gamma radiation–induced apoptotic response by T cell receptor ␣/␤ T cells and B lymphocytes in the peripheral blood of both murine models and humans (10,11). PBMCs obtained from patients with RA had lower baseline levels of TP53 messenger RNA (mRNA) and p53 protein and were deficient in their ability to increase p53 after exposure to gamma radiation. PBMCs from approximately half of the RA patients had a second defect, which was markedly reduced expression of checkpoint kinase 2 (CHK2) mRNA and protein. Chk2 is one of the upstream regulators of p53. We conclude that lymphocytes from patients with RA have at least 2 prominent defects in the p53 damageresponse pathway. One of these defects is underexpression of TP53 mRNA and p53 protein (which was observed in all RA patients), and the second is underexpression of CHK2 mRNA and Chk2 protein (which was observed in ⬃50% of RA patients). We hypothesize that these defects represent a cellular liability that may contribute to development of autoimmune disease. PATIENTS AND METHODS Patient populations. The study group comprised 43 control subjects who had no current chronic or acute infection and no family history of autoimmunity, and 52 patients meeting the American College of Rheumatology (ACR; formerly, the American Rheumatism Association) clinical criteria for RA (12). In both the control and patient populations, the approximate female-to-male ratio was 3:1. The age ranges (22–55 years) and racial distributions in both groups were similar. Human subject studies were approved by the Committee for the Protection of Human Subjects of the Vanderbilt University Institutional Review Board. Sample processing and PBMC isolation. PBMCs were isolated from heparinized blood by centrifugation on a Histopaque gradient (Sigma, St. Louis, MO). Isolated PBMCs were washed twice in HEPES buffered saline. Analysis of cell viability by flow cytometry. PBMCs were suspended at 1 ⫻ 106 cells/ml in complete media (RPMI 1640 medium, 10% fetal calf serum, glutamine, and penicillin/ streptomycin). Pilot experiments with control PBMCs were used to establish optimal doses and time points for each apoptosis-inducing agent. Based on results from these pilot experiments, cells were left untreated or were treated with different apoptosis-inducing agents, as follows: 10-Gy gamma radiation, 5 M dexamethasone, 1 M staurosporine, or 100 J/m2 ultraviolet (UV) radiation. At appropriate time points, cells were harvested and washed with fluorescence-activated cell sorting (FACS) buffer (10% bovine serum albumin in phosphate buffered saline [PBS] with 0.2% sodium azide) and incubated with fluorescein isothiocyanate– and phycoerythrinlabeled antibodies against CD3, CD4, CD14, CD19, and CD45RO (Becton Dickinson, San Jose, CA). Cells were washed and suspended in 500 l FACS buffer supplemented with 2 l of 0.5 M 7-aminoactinomycin D (7-AAD; Molecular Probes, Eugene, OR) as a viability marker prior to flow cytometric analysis. Samples were analyzed for 1 minute at high flow rate. Cells positive for 7-AAD were excluded from analysis, and the total numbers of remaining lymphocytes were tabulated to determine viability. RNA isolation and quantitative polymerase chain reaction (PCR). TRI Reagent (Molecular Research Center, Cincinnati, OH) was used to isolate total RNA from PBMCs. Five micrograms of total RNA was reverse transcribed with SuperScript II reverse transcriptase (Gibco BRL Life Technologies, Rockville, MD) to prepare complementary DNA (cDNA). Complementary DNA was also prepared from the HCT116 cell line in order to construct relative standard curves. TP53- and GAPDH-specific primers were used to amplify cDNA samples with SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA). Fluorescence was monitored using an ABI PRISM 7000 detector (Applied Biosystems). Relative quantities of TP53 and GAPDH transcripts in control and patient samples were calculated using a standard curve derived from the HCT116 cell line. Western blot analysis. Whole cell lysates were prepared in 1⫻ PBS, 1% Nonidet P40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, plus a protease inhibitor cocktail (Sigma). For each sample, equal amounts of total protein were electrophoresed and transferred to Immobilon-P membranes (Millipore, Bedford, MA). Membranes were blocked in a 5% nonfat milk, 0.1% Tween 20 in PBS, and probed with combinations of the following primary antibodies diluted in 1% nonfat milk with PBS plus 0.1% Tween 20: p53, Chk2, ␤-actin (Santa Cruz Biotechnology, Santa Cruz, CA), p21 (Oncogene Research Products, San Diego, CA), p45 up-regulated modulator of apoptosis (PUMA; Abcam, Cambridge, MA), poly(ADP-ribose) polymerase 1 (PARP-1; Cell Signaling Technology, Beverly, MA), and proliferating cell nuclear antigen (PCNA; Calbiochem, San Diego, CA). Membranes were washed 3 times with PBS–Tween (PBST), and probed with goat anti-mouse horseradish peroxidase–conjugated secondary antibodies (Santa Cruz Biotechnology) in 1% nonfat milk in PBST. Membranes were washed 3 times for 20 minutes with PBST. The ECL Plus Chemiluminescent Immunodetection Kit (Applied Biosystems) was used to visualize bands. Western blot analysis of luminescence intensity. Multiple exposures of films were captured using the Fluor-S MAX imaging system (Bio-Rad, Hercules, CA). Both background chemiluminescence and chemiluminescence intensities for individual bands were measured. Background chemilumines- TP53 AND APOPTOSIS DEFECTS IN RA 1049 Figure 1. Defects in gamma radiation–mediated apoptosis in rheumatoid arthritis (RA) peripheral blood mononuclear cells (PBMCs). A, Cell viability as assessed by flow cytometry of PBMCs from control (ctrl) subjects (n ⫽ 5) and RA patients (n ⫽ 6) 1–3 days after gamma radiation (10 Gy). B, PBMCs from control subjects (n ⫽ 10) or RA patients (n ⫽ 12) were treated with different apoptosis-inducing agents. Relative cell viability for whole PBMCs and for CD4⫹, CD8⫹, and CD19⫹ lymphocytes was determined 3 days (dexamethasone, ultraviolet [UV] radiation, and gamma radiation) or 1 day (staurosporine) after treatment, by flow cytometry. C, Scatter plot of CD4⫹ T lymphocyte viability in controls (n ⫽ 6) and RA patients (n ⫽ 12) 3 days after exposure to 10-Gy gamma radiation. D, Viability of CD3⫹,CD45ROhigh and CD3⫹,CD45ROlow T lymphocytes in controls (n ⫽ 6) and RA patients (n ⫽ 12) 3 days after exposure to 10-Gy gamma radiation. Bars show the mean ⫾ SD. cence was subtracted from all band intensities. Images of bands that were overexposed (as detected in the Fluor-S MAX software suite) were excluded from analysis. Intensities of samples were normalized relative to baseline HCT116 levels in order to make interblot comparisons and compensate for differences in exposure time. Reverse transcription–PCR (RT-PCR) and semiquantitative PCR analysis. Complementary DNA prepared from PBMC RNA from control subjects or patients with RA was used for Chk2 and GAPDH amplifications. Thirty cycles were used for initial RT-PCR amplification of samples. Semiquantitative PCR was performed on control and RA samples. PCRs were carried out as described above, with the exception that cDNA was serially diluted, and 35 cycles were used for amplification. Statistical analysis. Results are expressed as the mean ⫾ SD. Statistically significant differences between groups were determined by Student’s t-test. P values less than 0.05 were considered significant. RESULTS Defects in gamma radiation–induced apoptosis in RA PBMCs. A unique gene expression profile in PBMCs obtained from patients with autoimmune disease suggests that these cells may exhibit defects in apoptosis (7–9). For example, one of the most underexpressed genes in RA PBMCs is TP53, the gene that encodes the tumor suppressor protein p53. To test the hypothesis that p53 function is compromised in PBMCs obtained from patients with RA, PBMCs from 6 patients with RA and from 5 control subjects were exposed to 10-Gy gamma radiation (a known p53-dependent proapoptotic stimulus), and cellular viability was determined over a 3-day time course. RA PBMCs exhibited resistance to gamma radiation–induced cell death com- 1050 pared with control PBMCs (Figure 1A) (P ⬍ 0.005 for all time points). To determine whether this defect in apoptosis reflected a generalized defect or a selective defect in p53-dependent apoptosis, we performed additional studies using a panel of additional agents that induce apoptosis, as follows: dexamethasone, staurosporine, UV radiation, and gamma radiation. PBMCs from 10 control subjects and from 12 patients with RA were treated with these agents and analyzed by flow cytometry for viability of whole PBMCs and CD4⫹, CD8⫹, and CD19⫹ lymphocytes. Each apoptosis-inducing agent caused apoptosis in both PBMCs and lymphocytes (Figure 1B). We found no difference in the level of apoptosis between RA and control PBMCs or lymphocytes after treatment with dexamethasone and staurosporine. Treatment with UV radiation revealed no significant difference in overall PBMC or T lymphocyte viability; however, B lymphocytes from control subjects were more resistant to UV-mediated apoptosis compared with those from RA patients. Consistent with our initial observations, both PBMCs and lymphocytes from patients with RA were markedly resistant to gamma radiation–induced apoptosis relative to controls. This defect in apoptosis was most pronounced in CD4⫹ and CD8⫹ T lymphocytes (P ⬍ 0.002 and P ⬍ 0.001, respectively) and insignificant in CD19⫹ lymphocytes (Figure 1B). A scatter plot of CD4⫹ T cell viability after exposure to gamma radiation in different individuals demonstrated a highly homogeneous gamma radiation–induced apoptosis response in the control population, while there was substantial variability among lymphocytes from patients with RA (Figure 1C). Quantitative differences were observed in the response to gamma radiation by CD45ROlow and CD45ROhigh T cells. CD45ROlow T cells exhibited greater cell death compared with CD45ROhigh cells (Figure 1D). Both populations of RA lymphocytes were more resistant to apoptosis induced by gamma radiation than were control lymphocytes. These results revealed a significant defect in gamma radiation–induced apoptosis in T lymphocytes from patients with RA. Underexpression of TP53 mRNA and p53 protein in RA PBMCs. Because gamma radiation–induced apoptosis is dependent on functional p53 (10,13), low baseline levels of TP53 mRNA and p53 protein could contribute to defective apoptosis in RA. Results from previous microarray studies (8) revealed that TP53 transcript levels were lower in RA patients than in controls. We included additional microarray data from 12 previously unanalyzed RA samples, in order to MAAS ET AL Figure 2. Baseline TP53 transcript and p53 protein levels in RA PBMCs. A, Scatter plot displaying microarray results for TP53 transcript levels in control PBMCs (n ⫽ 9), PBMCs from patients with early RA (mean ⫾ SD disease duration 1 ⫾ 0.2 years; n ⫽ 16), and PBMCs from patients with established RA (mean ⫾ SD disease duration 10 ⫾ 1 years; n ⫽ 9). B, Normalized TP53 transcript levels in control (n ⫽ 4) and RA PBMCs (n ⫽ 5) as determined by quantitative polymerase chain reaction (PCR). C, Representative immunoblot comparing baseline p53 levels in PBMCs from a control cell line (HCT116), control subjects (C1–3), and RA patients (RA1–3). ␤-actin was included as an internal loading control. D, Normalized luminescence data for all individuals (control, n ⫽ 9; RA, n ⫽ 10) expressed in relative units of p53 protein. Bars in B and D show the mean and SD. See Figure 1 for other definitions. compare TP53 expression levels with our previously reported findings in control subjects and RA patients. TP53 mRNA levels in control subjects were relatively heterogeneous but were uniformly higher than the levels in patients with RA (Figure 2A). Real-time RT-PCR analysis confirmed that TP53 message levels were lower in patients with RA than in controls (Figure 2B). These TP53 AND APOPTOSIS DEFECTS IN RA results revealed that TP53 mRNA was consistently underexpressed in the RA patient population relative to the control population. To further characterize p53 in the RA patient population, we used Western blotting techniques to measure baseline p53 levels. Whole cell extracts were prepared from PBMCs obtained from 10 previously unanalyzed control subjects and from 10 patients with RA. Luminescence intensities were measured using a Fluor-S MAX imaging system. A representative blot using samples from human cell line HCT116 (as a positive control), 3 control subjects, and 3 patients with RA clearly demonstrated a marked reduction in p53 protein levels in RA PBMCs in the resting state (Figure 2C). Compiled luminescence data revealed that baseline p53 levels were significantly lower (⬎10-fold) in RA PBMCs compared with those in control PBMCs (P ⬍ 0.001) (Figure 2D). Levels of p53 after exposure to gamma radiation. In addition to measuring basal levels of p53 in the RA patient population, we examined p53 protein levels after exposure to gamma radiation. Under normal circumstances, p53 levels are relatively low in the resting state due to rapid turnover through the ubiquitin–proteasome pathway (14). In response to DNA-damaging agents (i.e., gamma radiation), the N-terminal domain of p53 is phosphorylated, blocking protein turnover (13,15). As a result, levels of p53 increase markedly after exposure to gamma radiation. Protein p53 acts as a transcription factor to induce damage-response target genes such as CDKN1A (p21) (16), GADD45A (17), NOXA1 (18), and BBC3 (PUMA) (19). We measured increases in steady-state levels of p53 and p21 induction in response to gamma radiation in previously unanalyzed samples. PBMCs from control subjects (n ⫽ 12) and RA patients (n ⫽ 12) were challenged with exposure to 10-Gy gamma radiation or remained untreated. Twenty-four hours after challenge, whole cell extracts were prepared and immunoblotted for p53, p21, and ␤-actin. The high intensity of ␤-actin on the immunoblots made obtaining reproducible exposures between blots difficult, and thus it served as a control for equal protein loading within a blot rather than as a loading control among blots. Instead, the HCT116 cell line was included as a positive control for p53 responsiveness to gamma irradiation (20) and to serve as an indicator for exposure time and normalization in later quantitative analyses. Representative immunoblots for control subjects and RA patients are presented in Figure 3. As expected, levels of p53 in PBMCs from con- 1051 Figure 3. Defects in p53 protein stabilization after exposure to gamma radiation. Representative immunoblots of peripheral blood mononuclear cells (PBMCs) cultured for 24 hours (control, n ⫽ 12; rheumatoid arthritis [RA], n ⫽ 12) and either left untreated (⫺) or treated with 10-Gy gamma irradiation (⫹). Samples were blotted for p53 and p21, with ␤-actin indicating equal protein loading within a blot. Based on steady-state p53 protein levels with and without radiation, PBMCs from patients with rheumatoid arthritis (RA) were segregated into gamma radiation nonresponders (n ⫽ 5) or gamma radiation partial responders (n ⫽ 5). trol subjects (n ⫽ 5) increased markedly in response to gamma irradiation (Figure 3A). Increased levels of p21 confirmed downstream p53 transcriptional activity in response to gamma irradiation. The p53-dependent response to gamma irradiation in PBMCs from patients with RA (n ⫽ 10) differed from that in controls. The most striking difference was that approximately half of the patients failed to increase p53 steady-state levels following exposure to gamma radiation (Figure 3). Based on these results, RA patients were organized into 2 distinct groups. PBMCs from RA patients who were 1052 gamma radiation nonresponders (GNRs) failed to increase p53 levels in response to irradiation (Figure 3B). PBMCs from RA patients who were gamma radiation partial responders (GPRs) displayed a modest increase in p53 levels, albeit to lower levels than those observed in controls (Figure 3C). In addition to their failure to increase levels of p53, PBMCs from GNRs did not increase p21 levels after exposure to gamma radiation (Figure 3B). In contrast, PBMCs from GPRs had increased p53 and p21 levels in response to gamma irradiation, albeit to lower levels than those in controls (Figure 3C). Two distinct defects in the p53 damage response in RA patients. In order to quantitate the relative protein levels for all of the control, RA GNR, and RA GPR samples examined in the gamma radiation experiment, we calculated luminescence intensities from the immunoblots (Figure 4) (control subjects, n ⫽ 10; RA GNRs, n ⫽ 7; RA GPRs, n ⫽ 5) using the Fluor-S MAX imaging system. Luminescence intensities were normalized using the untreated HCT116 cell line, present on all blots, in order to make comparisons among blots. Initially, we examined protein levels for the untreated and irradiated conditions in control, GPR, and GNR groups. To accomplish this, we normalized luminescence intensities using the untreated HCT116 cell line present on all blots to make comparisons among blots. These results revealed that p53 levels in untreated and gammairradiated cells from both the RA GNR and GPR groups were significantly lower than those in controls (P ⬍ 0.001 for both groups) (Figure 4A). We believe that these lower p53 levels arise from lower TP53 transcript levels in the RA patient population. We also calculated the gamma radiation–induced fold induction of p53 and p21 for each individual in the control, GPR, and GNR groups. Fold induction was calculated as the ratio of the luminescence intensity of protein bands between the gamma-irradiated group and the untreated group. Fold induction was averaged for all individuals within the control, GPR, and GNR groups. Both controls and RA GPRs demonstrated increased induction of p53, in comparable magnitude, in response to gamma irradiation (Figure 4B). This implies that, despite low baseline p53 levels, the signaling events needed to increase p53 levels after exposure to gamma radiation are intact in GPRs. In contrast, the GNR group showed undetectable induction of p53 after exposure to gamma radiation, suggesting that additional defects are present in the p53-dependent damageresponse pathway in this group of RA patients. To a MAAS ET AL Figure 4. Quantitative analysis of response to irradiation in controls, patients with rheumatoid arthritis (RA) who were gamma radiation partial responders (GPRs), and RA gamma radiation nonresponders (GNRs). Normalized luminescence intensities were compiled to permit quantitative comparisons among the groups described in Figure 3. A, Untreated (Unt) and postradiation (IR) p53 levels were determined in controls (n ⫽ 10), RA GPRs (n ⫽ 5), and RA GNRs (n ⫽ 7). B, Average fold induction of protein in response to exposure to gamma radiation was calculated for p53, p21, and ␤-actin among the different groups. Fold induction is defined as the ratio of gamma-irradiated luminescence intensity versus the untreated luminescence intensity. Bars show the mean and SD. MDM2 ⫽ mouse double minute 2. large degree, induction of p21 after gamma irradiation mirrored the p53 results for the control, GPR, and GNR groups. Defective induction of effectors of apoptosis in RA PBMCs. Protein p53 that accumulates after exposure to gamma radiation acts as a transcription factor to induce proapoptotic target genes (21,22). Because we observed defects in gamma radiation–induced apoptosis, we investigated whether this was accompanied by defective induction of proapoptotic target genes. This was accomplished by immunoblotting representative untreated and gamma-irradiated extracts from previously analyzed controls (n ⫽ 2), RA GPRs (n ⫽ 2), and RA TP53 AND APOPTOSIS DEFECTS IN RA Figure 5. Defective induction of p53 downstream effectors in peripheral blood mononuclear cells from patients with RA. Whole cell extracts from representative controls (n ⫽ 2), GPRs (n ⫽ 2), and GNRs (n ⫽ 2). Untreated (⫺) and gamma-irradiated (⫹) samples were immunoblotted 24 hours after challenge. Lysates were probed for poly(ADP-ribose) polymerase 1 (PARP-1), proliferating cell nuclear antigen (PCNA), p45 up-regulated modulator of apoptosis (PUMA), and ␤-actin. See Figure 4 for other definitions. GNRs (n ⫽ 2). Extracts were analyzed for p53 levels to confirm their partial responder/nonresponder status. Extracts were also probed for PUMA (19,23), PARP-1 cleavage, a biochemical marker for apoptosis (24), PCNA as a measure of cellular proliferative status, and ␤-actin as a control. PCNA and ␤-actin levels were relatively consistent among samples and thus also served as protein-loading controls (Figure 5). PBMCs from RA GNRs and, to a more limited extent, those from RA GPRs demonstrated marked differences when compared with controls. Control PBMCs exhibited increased p53 levels in response to gamma irradiation, with corresponding increases in PUMA and PARP-1 cleavage compared with untreated samples (Figure 5). In contrast, PBMCs from RA GNRs exhibited negligible increases in p53 levels and PUMA after exposure to gamma radiation. RA GNRs had lower levels of both full-length and cleavage forms of PARP-1 and did not display increased PARP-1 cleavage after exposure to gamma radiation. The RA GPRs exhibited variability in p53 levels and downstream effector function in response to gamma radiation. One patient had low p53 levels as well as negligible PUMA induction and PARP-1 cleavage. The other RA patient had a somewhat higher response. These results provide further evidence that downstream p53 apoptotic effector function is compromised in PBMCs from patients with RA. 1053 Correlation of decreased Chk2 expression with RA GNR status. Our quantitative analysis of relative protein levels and protein induction revealed that PBMCs obtained from patients with RA contain lower levels of p53 both before and after exposure to gamma radiation. However, the GNR group appeared to have additional defects in the p53-dependent damage response that prevented them from increasing p53 levels after irradiation. To explore potential causes for this additional defect, we examined Chk2, an upstream kinase that can phosphorylate p53 (25) and prevent its ubiquitinmediated degradation (26). Extracts from previously identified RA GNRs (n ⫽ 5), RA GPRs (n ⫽ 5), and control subjects (n ⫽ 5) were reanalyzed for Chk2. All control PBMCs contained comparable levels of Chk2 (Figure 6A). Levels of Chk2 in control PBMCs decreased after gamma radiation. RA GNRs had uniformly low levels of Chk2 (Figure 6B), while PBMCs from RA GPRs contained levels of Chk2 equivalent to those of control PBMCs (Figure 6C). To determine whether low Chk2 levels in patients with RA correlated with low transcript levels of CHK2, we analyzed PBMC RNA derived from previously unexamined RA patients and controls (Figure 6D). A portion of the RA patients had CHK2 transcript levels similar to those of controls, and a portion had very low CHK2 transcript levels. Semiquantitative RT-PCR analysis was used to more accurately determine relative levels of transcript in a representative control sample and RA sample with very low CHK2 transcript abundance. This confirmed that CHK2 transcript levels were substantially lower in a subset of RA patients compared with controls (Figure 6D). We believe that underexpression of CHK2 mRNA may account, at least in part, for the heterogeneity in the response of RA PBMCs to gamma radiation. Differential expression of many p53-regulated genes in PBMCs from patients with RA. The abovementioned studies focused on p53 protein, TP53 transcript levels, and a few well-characterized downstream transcriptional targets of p53. We also wanted to determine whether additional known p53-regulated genes were differentially expressed in RA compared with control PBMCs. To do so, we compared results from our microarray analysis of differential gene expression between control and RA PBMCs (8) to microarray analysis of differential gene expression between cell lines that have or do not have functional p53 (27,28). Genes identified in these studies include both known direct transcriptional targets of p53 (p21, PUMA) as well as genes that may be direct transcriptional targets of p53 or may be differentially expressed as a result of secondary 1054 effects of the presence or absence of p53. We examined our microarray data to determine expression levels of genes (identified in cell line studies) that have altered MAAS ET AL Table 1. Expression levels of p53-regulated genes* Fold reduction or increase Gene title p53-inducible genes Transglutaminase E-cadherin CDKN1B, p27 Caspase 6 Myosin 1B Epoxide hydrolase RAD52 homolog Ubiquinone APAF1 GADD45A PIG11 c-fos Endoglin BTG family, 2 Tyrosinase p53-repressible genes Adenosine deaminase DAG kinase Fibronectin 1 IL-8 EGF receptor Cyclin A2 Stathmin NF-IL3 RAD51 homolog Carboxypeptidase CDC28 PK2 COP9 Cyclin E1 CDC6 homolog Galectin 3 SGK ⫺11.1 ⫺6.3 ⫺5.0 ⫺5.3 ⫺5.6 ⫺8.3 ⫺4.8 ⫺4.5 ⫺4.3 ⫺3.2 ⫺2.6 ⫺1.8 ⫺2.5 ⫺2.1 ⫺2.1 3.8 4.1 4.8 5.4 3.6 3.4 2.9 3.1 2.9 3.5 2.4 1.9 2.5 1.9 2.5 2.1 * Values for p53-inducible genes are the fold reduction; values for p53-repressible genes are the fold increase. Data were compiled from previous microarray experiments (8). Genes that were p53 responsive (26,27) were selected from our microarray data. Figure 6. Low levels of Chk2 protein and CHK2 transcript in peripheral blood mononuclear cells (PBMCs) from patients with rheumatoid arthritis (RA). Whole cell extracts from the same patient lysates (untreated [⫺], gamma-irradiated [⫹]) described in Figure 4 were immunoblotted for Chk2 protein. A–C, Immunoblots of controls (n ⫽ 5), RA gamma radiation nonresponders (GNRs) (n ⫽ 4), and RA gamma radiation partial responders (GPRs) (n ⫽ 5). D, Reverse transcription–polymerase chain reaction (PCR) performed on cDNA from total RNA in previously unexamined control (n ⫽ 3) and RA (n ⫽ 6) PBMCs. Top panel shows amplification results after 30 cycles. Thirty-five–cycle semiquantitative PCR was performed on serial dilutions of cDNA. Results for control (C8) and RA PBMCs with low CHK2 transcript levels (RA16) are shown in the paired figures in the lower panel. HCT ⫽ HCT116 cell line. expression profiles in the presence or absence of p53 (Table 1). All genes that were overexpressed in p53positive cell lines were underexpressed in RA PBMCs. Conversely, all genes that were underexpressed in p53positive cell lines were overexpressed in RA PBMCs. These results further support our hypothesis that defects in p53 expression and induction in PBMCs from patients with RA may account for a significant portion of the unique gene expression profile observed in these affected individuals. DISCUSSION In this study, we found uniform defects in the p53 damage-response pathway in PBMCs from patients with RA. T lymphocytes from patients with RA were significantly more resistant to gamma radiation–induced cell death than were control T lymphocytes. In contrast, TP53 AND APOPTOSIS DEFECTS IN RA lymphocytes from RA patients did not display defects in p53-independent modes of apoptosis. Both p53 protein and TP53 mRNA baseline levels were substantially reduced in RA PBMCs, providing a possible explanation for defects in gamma radiation–induced apoptosis. Levels of p21 (a cyclin-dependent kinase inhibitor) were also substantially lower in RA patients. Furthermore, half of the patients with RA failed to exhibit increased steady-state levels of p53 after exposure to gamma radiation. PBMCs from these RA GNRs also contained negligible levels of Chk2, an upstream kinase that phosphorylates p53 (25) and may prevent ubiquitin-mediated degradation of p53 (29) after DNA damage. Our previous work revealed the presence of a gene expression signature in patients with a range of autoimmune disorders (8). Many underexpressed genes present in this signature encode proteins necessary for apoptosis. One of the most consistently underexpressed genes is TP53, a central mediator of cellular responses to stress that induces cell cycle arrest or apoptosis (30,31). This series of biochemical studies support our initial observations of significantly reduced expression of TP53 mRNA in autoimmunity and demonstrate that lymphocytes from patients with RA have a defective p53mediated damage-response pathway. Defects in lymphocyte apoptosis are hypothesized to contribute to development of autoimmunity. Some of the best support for this theory comes from observations of lpr or gld mutations (mutations in Fas or Fas ligand [FasL], respectively) on the MRL murine background (32). These mice develop autoantibodies and succumb to fatal glomerulonephritis. With the exception of the rare autoimmune lymphoproliferative syndrome (33), efforts to identify defects in Fas or FasL in more common human autoimmune disorders have been relatively unsuccessful (34). Our results clearly demonstrate that there are uniform defects in apoptosis in PBMCs from patients with RA, but that these defects are present in the p53 damage-response pathway. Other investigations have addressed the role of p53 in RA. The majority of these studies have focused on the synovium. It has been proposed that high levels of oxidative stress in rheumatoid synovium may cause somatic mutations in the TP53 gene (35). Presumably, mutations in synovial p53 may allow pathologic proliferation of synovial cells that may lead to joint destruction and other clinical manifestations of RA. Alternatively, it has been proposed that the cytokine, macrophage migratory inhibitory factor (MIF), may cause decreased cellular p53 levels (36,37). MIF lowers endogenous p53 levels both in vivo and in vitro, and high levels of MIF could contribute to synovial proliferation and pannus 1055 formation. Although these results specifically address the impact of MIF on the synovium, serum levels of MIF are also elevated in RA patients compared with controls (38). Elevated MIF levels may contribute to the underexpression of p53 in PBMCs from RA patients. However, our results clearly show that T lymphocytes, representing 80% of our PBMC preparations, are defective in p53-mediated apoptosis, and T lymphocytes are not known to respond to MIF. There is also evidence that p53 maintains tolerance in lymphocytes by regulating cell cycle progression. Human T lymphocytes from peripheral blood or intestinal lamina propria show an inverse relationship between p53 levels and the rate of progression through the cell cycle (39). Cell cycle delays mediated by elevated levels of p53 in lamina propria T lymphocytes may be a mechanism that maintains tolerance against environmental antigens. Preliminary studies by Leech et al, using an antigen-induced arthritis model on a p53⫺/⫺ background, revealed that T lymphocytes proliferate more readily and produce more IFN␥ in the absence of p53 (40). Similar results in models of collagen-induced arthritis (41) suggest that inflammatory responses may be exacerbated in the absence of p53. Lymphocytes from healthy individuals with the HLA–DRB1*04 allele (42) and patients with RA (43) show signs of inappropriate aging as measured by telomeric shortening. T lymphocytes from RA patients also proliferate less readily (compared with controls) in response to stimulation with anti-CD3 or recall antigens. Similar observations have been made in p53⫺/⫺ mice (44). These studies revealed no defects in lymphocyte development. Rather, lymphocytes from p53⫺/⫺ mice exhibited signs of accelerated aging and nonresponsiveness to T cell receptor (TCR) stimuli. Recent studies also demonstrated that p53 is a necessary element of cellular senescence (45–47). Therefore, we speculate that persistent underexpression of TP53 mRNA and p53 protein, and perhaps CHK2 mRNA and Chk2 protein, may explain the accelerated aging and lack of lymphocyte responsiveness to TCR signals observed in patients with RA. We observed that approximately half of the patients with RA failed to demonstrate increased steadystate levels of p53 in response to exposure to gamma radiation, and that this correlated with depressed levels of Chk2. CHK2 mRNA levels are almost absent in this subset of RA patients. Chk2 is believed to be an upstream regulator of p53 stability in response to certain types of DNA damage (15,29), although there is now debate in the field about its absolute necessity for p53 stabilization (48,49). Studies in Chk2⫺/⫺ mice reveal no 1056 MAAS ET AL abnormalities in lymphocyte development; however, nothing more is known about the role of Chk2 in the immune system. Defects in downstream p53 target genes may also play a role in promoting systemic autoimmunity. Microarray results and the results of our studies presented here demonstrate that many p53 downstream gene targets are dysregulated in a manner consistent with p53 dysfunction. For example, p21, a downstream cyclindependent kinase inhibitor and transcriptional target of p53, is also underexpressed in patients with RA. Although controversial (50), studies have documented increased autoantibody production, glomerulonephritis, and mortality in p21⫺/⫺ female mice (51). T lymphocytes from these mice are hyperproliferative when cultured with interleukin-2 after activation compared with wildtype littermates. Additional studies examining GADD45A, another downstream target of p53 effector function, link this gene to systemic autoimmunity and abnormalities in T lymphocyte function (52). GADD45A⫺/⫺, p21⫺/⫺ mice exhibit aggressive autoimmunity comparable with that observed in MRL/lpr mice. These observations, combined with our results in the RA patient population, indicate that defects in the expression of molecules in the DNA damage-response pathway might play a role in autoimmune pathogenesis. These proteins may also represent new targets for therapeutic approaches. It might be possible to design therapies to either correct defects in the p53 damage response pathway or inhibit downstream effectors that are normally inhibited by damage-response proteins such as p53, p21, or GADD45A. ACKNOWLEDGMENTS We thank Drs. Theodore Pincus, Howard Fuchs, Victor Byrd, Tuulikki Sokka, and their patients for access to their clinics and for providing blood samples. We would also like to thank Bo Yelverton, Sukumar Narasimhulu, and Xuan Li for technical assistance with experiments. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. REFERENCES 21. 1. Marrack P, Kappler J, Kotzin BL. Autoimmune disease: why and where it occurs. Nat Med 2001;7:899–905. 2. Moxley G, Cohen HJ. Genetic studies, clinical heterogeneity, and disease outcome studies in rheumatoid arthritis. Rheum Dis Clin North Am 2002;28:39–58. 3. Gregersen PK. Genetics of rheumatoid arthritis: confronting complexity. Arthritis Res 1999;1:37–44. 4. Wakeland EK, Liu K, Graham RR, Behrens TW. Delineating the genetic basis of systemic lupus erythematosus. Immunity 2001;15: 397–408. 5. Bennett L, Palucka AK, Arce E, Cantrell V, Borvak J, Banchereau 22. 23. 24. 25. J, et al. Interferon and granulopoiesis signatures in systemic lupus erythematosus blood. J Exp Med 2003;197:711–23. Baechler EC, Batliwalla FM, Karypis G, Gaffney PM, Ortmann WA, Espe KJ, et al. Interferon-inducible gene expression signature in peripheral blood cells of patients with severe lupus. Proc Natl Acad Sci U S A 2003;100:2610–5. Bomprezzi R, Ringner M, Kim S, Bittner ML, Khan J, Chen Y, et al. Gene expression profile in multiple sclerosis patients and healthy controls: identifying pathways relevant to disease. Hum Mol Genet 2003;12:2191–9. Maas K, Chan S, Parker J, Slater A, Moore J, Olsen N, et al. Cutting edge: molecular portrait of human autoimmune disease. J Immunol 2002;169:5–9. Ramanathan M, Weinstock-Guttman B, Nguyen LT, Badgett D, Miller C, Patrick K, et al. In vivo gene expression revealed by cDNA arrays: the pattern in relapsing-remitting multiple sclerosis patients compared with normal subjects. J Neuroimmunol 2001; 116:213–9. Clarke AR, Purdie CA, Harrison DJ, Morris RG, Bird CC, Hooper, ML, et al. Thymocyte apoptosis induced by p53-dependent and independent pathways. Nature 1993;362:849–52. Seki H, Kanegane H, Iwai K, Konno A, Ohta K, Yachie A, et al. Ionizing radiation induces apoptotic cell death in human TcR ␥/␦⫹ T and natural killer cells without detectable p53 protein. Eur J Immunol 1994;24:2914–7. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. Kastan MB, Onyekwere O, Sidransky D, Vogelstein B, Craig RW. Participation of p53 protein in the cellular response to DNA damage. Cancer Res 1991;51:6304–11. Haupt Y, Maya R, Kazaz A, Oren M. Mdm2 promotes the rapid degradation of p53. Nature 1997;387:296–9. Hirao A, Kong YY, Matsuoka S, Wakeham A, Ruland J, Yoshida H, et al. DNA damage-induced activation of p53 by the checkpoint kinase Chk2. Science 2000;287:1824–7. El-Deiry WS, Tokino T, Velculescu VE, Levy DB, Parsons R, Trent JM, et al. WAF1, a potential mediator of p53 tumor suppression. Cell 1993;75:817–25. Kastan MB, Zhan Q, el-Deiry WS, Carrier F, Jacks T, Walsh WV, et al. A mammalian cell cycle checkpoint pathway utilizing p53 and GADD45 is defective in ataxia-telangiectasia. Cell 1992;71: 587–97. Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T, et al. Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 2000;288:1053–8. Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA induces the rapid apoptosis of colorectal cancer cells. Mol Cell 2001;7:673–82. Kennedy AS, Harrison GH, Mansfield CM, Zhou XJ, Xu JF, Balcer-Kubiczek EK. Survival of colorectal cancer cell lines treated with paclitaxel, radiation, and 5-FU: effect of TP53 or hMLH1 deficiency. Int J Cancer 2000;90:175–85. Wu X, Deng Y. Bax and BH3-domain-only proteins in p53mediated apoptosis. Front Biosci 2002;7:d151–6. Yu J, Zhang L. No PUMA, no death: implications for p53dependent apoptosis. Cancer Cell 2003;4:248–9. Nakano K, Vousden KH. PUMA, a novel proapoptotic gene, is induced by p53. Mol Cell 2001;7:683–94. Lazebnik YA, Kaufmann SH, Desnoyers S, Poirier GG, Earnshaw WC. Cleavage of poly(ADP-ribose) polymerase by a proteinase with properties like ICE. Nature 1994;371:346–7. Shieh SY, Ahn J, Tamai K, Taya Y, Prives C. The human homologs of checkpoint kinases Chk1 and Cds1 (Chk2) phosphorylate p53 at multiple DNA damage-inducible sites. Genes Dev 2000;14:289–300. TP53 AND APOPTOSIS DEFECTS IN RA 26. Chehab NH, Malikzay A, Stavridi ES, Halazonetis TD. Phosphorylation of Ser-20 mediates stabilization of human p53 in response to DNA damage. Proc Natl Acad Sci U S A 1999;96:13777–82. 27. Zhao R, Gish K, Murphy M, Yin Y, Notterman D, Hoffman WH, et al. Analysis of p53-regulated gene expression patterns using oligonucleotide arrays. Genes Dev 2000;14:981–93. 28. Kannan K, Amariglio N, Rechavi G, Jakob-Hirsch J, Kela I, Kaminski N, et al. DNA microarrays identification of primary and secondary target genes regulated by p53. Oncogene 2001;20: 2225–34. 29. Chehab NH, Malikzay A, Appel M, Halazonetis TD. Chk2/hCds1 functions as a DNA damage checkpoint in G(1) by stabilizing p53. Genes Dev 2000;14:278–88. 30. Kuerbitz SJ, Plunkett BS, Walsh WV, Kastan MB. Wild-type p53 is a cell cycle checkpoint determinant following irradiation. Proc Natl Acad Sci U S A 1992;89:7491–5. 31. Stewart ZA, Pietenpol JA. p53 signaling and cell cycle checkpoints. Chem Res Toxicol 2001;14:243–63. 32. Nagata S, Suda T. Fas and Fas ligand: lpr and gld mutations. Immunol Today 1995;16:39–43. 33. Fisher GH, Rosenberg FJ, Straus SE, Dale JK, Middleton LA, Lin AY, et al. Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 1995;81:935–46. 34. McNally J, Yoo DH, Drappa J, Chu JL, Yagita H, Friedman SM, et al. Fas ligand expression and function in systemic lupus erythematosus. J Immunol 1997;159:4628–36. 35. Yamanishi Y, Boyle DL, Rosengren S, Green DR, Zvaifler NJ, Firestein GS. Regional analysis of p53 mutations in rheumatoid arthritis synovium. Proc Natl Acad Sci U S A 2002;99:10025–30. 36. Hudson JD, Shoaibi MA, Maestro R, Carnero A, Hannon GJ, Beach DH. A proinflammatory cytokine inhibits p53 tumor suppressor activity. J Exp Med 1999;190:1375–82. 37. Leech M, Lacey D, Xue JR, Santos L, Hutchinson P, Wolvetang E, et al. Regulation of p53 by macrophage migration inhibitory factor in inflammatory arthritis. Arthritis Rheum 2003;48:1881–9. 38. Leech M, Metz C, Hall P, Hutchinson P, Giantis K, Smith M, et al. Macrophage migration inhibitory factor in rheumatoid arthritis: evidence of proinflammatory function and regulation by glucocorticoids. Arthritis Rheum 1999;42:1601–8. 39. Sturm A, Itoh J, Jacobberger JW, Fiocchi C. p53 negatively regulates intestinal immunity by delaying mucosal T cell cycling. J Clin Invest 2002;109:1481–92. 1057 40. Leech M, Xue JR, Poulos G, Hall P, Morand E. p53 modulates the systemic immune response and arthritis severity in antigeninduced arthritis [abstract]. Arthritis Rheum 2003; 48 Suppl 9:S348. 41. Yamanishi Y, Boyle DL, Pinkoski MJ, Mahboubi A, Lin T, Han Z, et al. Regulation of joint destruction and inflammation by p53 in collagen-induced arthritis. Am J Pathol 2002;160:123–30. 42. Schonland SO, Lopez C, Widmann T, Zimmer J, Bryl E, Goronzy JJ, et al. Premature telomeric loss in rheumatoid arthritis is genetically determined and involves both myeloid and lymphoid cell lineages. Proc Natl Acad Sci U S A 2003;100:13471–6. 43. Koetz K, Bryl E, Spickschen K, O’Fallon WM, Goronzy JJ, Weyand CM. T cell homeostasis in patients with rheumatoid arthritis. Proc Natl Acad Sci U S A 2000;97:9203–8. 44. Ohkusu-Tsukada K, Tsukada T, Isobe K. Accelerated development and aging of the immune system in p53-deficient mice. J Immunol 1999;163:1966–72. 45. Sharpless NE, DePinho RA. p53: good cop/bad cop. Cell 2002; 110:9–12. 46. Campisi J. Cancer and ageing: rival demons? Nat Rev Cancer 2003;3:339–49. 47. D’Adda di Fagagna F, Reaper PM, Clay-Farrace L, Fiegler H, Carr P, von Zglinicki T, et al. A DNA damage checkpoint response in telomere-initiated senescence. Nature 2003;426:194–8. 48. Ahn J, Urist M, Prives C. Questioning the role of checkpoint kinase 2 in the p53 DNA damage response. J Biol Chem 2003; 278:20480–9. 49. Keramaris E, Hirao A, Slack RS, Mak TW, Park DS. Ataxia telangiectasia-mutated protein can regulate p53 and neuronal death independent of Chk2 in response to DNA damage. J Biol Chem 2003;278:37782–9. 50. Lawson BR, Kono DH, Theofilopoulos AN. Deletion of p21 (WAF-1/Cip1) does not induce systemic autoimmunity in female BXSB mice. J Immunol 2002;168:5928–32. 51. Balomenos D, Martin-Caballero J, Garcia MI, Prieto I, Flores JM, Serrano M, et al. The cell cycle inhibitor p21 controls T-cell proliferation and sex-linked lupus development. Nat Med 2000;6: 171–6. 52. Salvador JM, Hollander MC, Nguyen AT, Kopp JB, Barisoni L, Moore JK, et al. Mice lacking the p53-effector gene Gadd45a develop a lupus-like syndrome. Immunity 2002;16:499–508.