Author + information
- Received September 11, 2014
- Revision received January 7, 2015
- Accepted January 14, 2015
- Published online March 31, 2015.
- Davide Flego, PhD,
- Anna Severino, PhD,
- Francesco Trotta, MD,
- Marco Previtero, MD,
- Sarassunta Ucci, PhD,
- Chiara Zara, PhD,
- Gianluca Massaro, MD,
- Daniela Pedicino, MD,
- Luigi M. Biasucci, MD,
- Giovanna Liuzzo, MD, PhD∗ ( and )
- Filippo Crea, MD
- ↵∗Reprint requests and correspondence:
Dr. Giovanna Liuzzo, Institute of Cardiology, Catholic University, Largo A. Gemelli, 8, 00168 Rome, Italy.
Background Critical impairment of adaptive immune response has been observed in patients with acute coronary syndromes (ACS) with reduced expansion of regulatory T cells (Treg) and enhanced effector T-cell responsiveness, both associated with poorer outcomes.
Objectives This study investigated the mechanisms underlying T-cell dysregulation in ACS.
Methods We evaluated both early and downstream T-cell receptor activation pathways after ex vivo stimulation with anti-CD3 and anti-CD28 crosslink in CD4+ T cells from 20 patients with non-ST-segment elevation myocardial infarction (NSTEMI), 20 with stable angina (SA), and 20 controls. We reassessed 10 NSTEMI and 10 SA patients after 1 year.
Results Phospho-flow analysis revealed reduced phosphorylation of the zeta-chain–associated protein kinase of 70 kDa at the inhibitory residue tyrosine 292, enhancing T-cell activation, in NSTEMI helper T cells versus SA and controls (each, p < 0.001), resulting from increased expression of the protein tyrosine phosphatase, nonreceptor type, 22 (PTPN22) (p < 0.001 for both comparisons), persisting at follow-up. We also observed reduced phosphorylation (p < 0.001 versus controls) and lower levels of binding to interleukins 2 and 10 core promoter regions of the transcription factor cyclic adenosine monophosphate response element-binding protein (CREB) in NSTEMI (p < 0.05 vs. controls), which recovered at 1 year. Finally, in NSTEMI patients, helper T cells had a reduced ability in T-cell receptor–induced Treg generation (p = 0.002 vs. SA; p = 0.001 vs. controls), partially recovered at 1 year. Restoring CREB activity and silencing PTPN22 enhanced NSTEMI patients’ ability to generate Treg.
Conclusions The persistent overexpression of PTPN22 and the transient reduction of CREB activity, associated with impaired Treg differentiation, might play a role in ACS.
Although the outcomes of acute coronary syndromes (ACS) have considerably improved in the past decade, ACS persists as the main cause of morbidity and mortality worldwide. Inflammation plays a key role in atherosclerotic plaque initiation and progression (1,2). In experimental models of atherogenesis, reduced frequency of regulatory T cells (Treg), the helper T-cell compartment involved in peripheral tolerance, promotes atherosclerosis, while increased Treg response induces regression of atherosclerosis and stabilization of plaque (3).
The transition from coronary stability to instability, however, is less well understood as we lack animal models. The abrupt clinical presentation of ACS gives a strong signal of discontinuity in the natural history of atherothrombosis. The causes of such discontinuity are complex, probably multiple, and still largely unknown (4).
In ACS, helper T-cell (CD4+ lymphocyte) subpopulations are dysregulated, and their abnormalities are associated with poorer outcomes (5–11). In particular, ACS patients have a reduced expansion of Treg (8,11–13), and lymphocytes in ACS patients show T-cell receptor (TCR) signaling abnormalities leading to enhanced effector T-cell responsiveness (14,15). The putative molecular mechanisms involved in these TCR signaling abnormalities are also unclear.
Beyond the cytokine environment, TCR signal strength is important in generating different T-cell subsets (16–18). In particular, strong upstream TCR-signaling activation seems to negatively modulate Treg differentiation (18).
Protein phosphatases, signal-transducing enzymes that dephosphorylate cellular phosphoproteins, play a key role in controlling the intensity of TCR activation. Protein tyrosine phosphatase, nonreceptor type, 22 (PTPN22) controls early TCR-signal transduction acting on lymphocyte-specific protein tyrosine kinase and on zeta-chain–associated protein kinase of 70 kDa (ZAP-70) (19,20). A recent study showed a crucial role of PTPN22 in the regulation of murine Treg function, expression, and survival (21).
TCR triggering initiates a complex network of signaling events that activate several downstream pathways, including cyclic adenosine monophosphate response element–binding protein (CREB) (22), a molecule that plays different roles in immune function, particularly Treg generation and maintenance, and the production of interleukin (IL)-2 (23), a key cytokine essential for Treg function (24). In experimental models of cardiovascular disease, CREB is down-regulated by risk factors (25).
In this study, we compared early TCR-activation pathways in patients with non-ST-segment elevation myocardial infarction (NSTEMI) to those in patients with stable angina (SA) and in controls. We studied the role of PTPN22 in modulating ZAP-70 activation and Treg differentiation, plus we assessed downstream TCR-activation pathways, in particular CREB activity. We reanalyzed NSTEMI and SA patients at 1-year follow-up, during a stable phase of coronary artery disease. Additionally, to dissect the role of a general inflammatory response (e.g., that characterizing NSTEMI patients) in TCR abnormalities, we performed additional experiments in healthy controls in the presence of the proinflammatory cytokine IL-6.
We enrolled 20 patients admitted to our coronary care unit with a diagnosis of NSTEMI, defined as a rise and fall of cardiac troponin T and at least one of the following: angina, ST-segment depression, and T-wave inversion. We also enrolled 20 patients with chronic SA admitted to our cardiovascular ward to undergo coronary angiography because of severe symptoms and/or high-risk abnormalities on noninvasive testing, and 20 patients >50 years of age at intermediate risk for cardiovascular disease and without a history of and/or current symptoms or signs of ischemic heart disease (controls).
We drew venous blood samples at patient enrollment. In ACS, we collected venous blood samples within 24 h of symptom onset (mean: 11.5 ± 5.7 h). We obtained peripheral blood mononuclear cells from whole blood samples by standard-gradient centrifugation over Ficoll-Hypaque (GE Healthcare Bio-Sciences, Piscataway, New Jersey).
The continuous variables that were normally distributed, as assessed by the Shapiro-Wilk test, were described as mean ± SEM and analyzed with parametric tests. For comparisons among the 3 groups, we used 1-way analysis of variance (ANOVA) with Bonferroni correction. For multiple pairwise comparisons, we used 2-way ANOVA for repeated measures, with Bonferroni correction. For between-group comparisons, we used an unpaired Student t test. To compare the means of 2 related samples within groups, we used a paired-samples Student t test. For correlations, we used the Pearson's test. Values of high-sensitivity C-reactive protein that were non-normally distributed were described as median (range) and compared using the Kruskal-Wallis nonparametric ANOVA, with the Dunn’s test for comparisons among groups. Proportions were compared using the chi-square test. A 2-tailed p value <0.05 was considered statistically significant. Statistical analysis was performed with GraphPad Prism version 5.00 (GraphPad Software, San Diego, California) and SPSS version 18.0 (SPSS Inc., Chicago, Illinois).
For a detailed description of all methods, see the Online Appendix.
We performed TCR-signaling experiments in all study patients. After stimulation with anti-CD3 and anti-CD28 crosslink, phosphorylation of ZAP-70 at regulatory residue tyrosine 292 (Y-292) in helper T cells was reduced in the NSTEMI group compared with that in SA patients and controls at all time points of TCR stimulation (2, 5, and 10 min) (Figure 1A). Notably, we did not find between-group differences in phosphorylation at ZAP-70 Y-319 (Figure 1B). In NSTEMI patients, early tyrosine phosphorylation events were enhanced compared with those in SA patients and controls (both, p < 0.001) (Figure 1C), confirming the TCR hyper-reactivity previously described in ACS (15).
ZAP-70 is a substrate of phosphatase PTPN22. In helper T cells in NSTEMI patients, we observed a higher expression of both PTPN22 messenger ribonucleic acid (mRNA) (p = 0.001 vs. SA; p < 0.001 vs. controls) and PTPN22 protein (both, p < 0.001) (Figure 1D). In the overall study population, PTPN22 expression was positively correlated with phosphotyrosine levels (r = 0.377; p = 0.003) and negatively with ZAP-70 Y-292 phosphorylation (r = –0.476; p = 0.001) (Figure 1E). Online Table 4 shows the predictors of PTPN22 at multivariate logistic regression analysis.
To investigate TCR activation during a stable phase of the disease, we reassessed 10 NSTEMI patients and 10 SA patients after 1-year follow-up. None of the patients had experienced an event during the follow-up period. As observed at the time of the index event, ZAP-70 Y-292 phosphorylation was reduced in the helper T cells of NSTEMI compared with SA patients, 2 and 5 min after TCR stimulation (p = 0.007 and p = 0.004, respectively) (Figure 2A, Online Figure 1). Accordingly, at 1-year follow-up, both mRNA and protein expression levels of PTPN22 were higher in NSTEMI patients compared with those in SA patients (both, p < 0.001) (Figure 2B, Online Figure 1). Moreover, at 5 min after TCR stimulation, we observed an enhancement of early tyrosine phosphorylation events in the helper T cells of NSTEMI patients compared with those in SA patients (p < 0.001) (Figure 2C), confirming that TCR hyper-reactivity persisted after the acute phase of ACS. PTPN22 expression was positively correlated with phosphotyrosine levels (r = 0.712; p < 0.001) and negatively with ZAP-70 Y-292 phosphorylation (r = –0.602; p < 0.005) (Figure 2D). Changes over time in each group are shown in Online Figure 1.
To investigate downstream TCR-signaling pathway activation, we assessed the phosphorylation of the well-characterized transcription factor CREB in the same 20 patients in each group.
NSTEMI patients exhibited a lower phosphorylation of CREB at residue serine 133 at 2 and 5 min after TCR stimulation compared with that in controls (p < 0.001) (Figure 3A). We also analyzed the biological activity of CREB by assessing its intranuclear localization and its binding levels to the IL-2, IL-10, Foxp3, and c-fos promoter regions in five patients in each group. CD4+ T cells from NSTEMI patients exhibited reduced intranuclear localization of CREB after TCR stimulation (p = 0.045 vs. controls) (Figure 3B) and reduced binding levels of CREB to IL-2 and IL-10 core promoter regions compared with controls (IL-2: p = 0.024; IL-10: p = 0.001); we did not find statistical differences in CREB binding to Foxp3 and c-fos promoters (Figure 3C).
CREB is required for generating and maintaining Tregs (23); therefore, we evaluated TCR-induced Treg generation. After 6 days of TCR stimulation, differentiation of Treg (identified as CD4+CD25+CD127lowFoxp3+) was reduced in NSTEMI patients compared with that in SA patients and controls (p < 0.001) (Figure 3D).
At 1-year follow-up, stimulated helper T-cell CREB phosphorylation was significantly increased from baseline in the NSTEMI and SA groups (p < 0.01 for all comparisons) (Figure 3E, Online Figure 1). Moreover, in NSTEMI patients, the proportion TCR-induced Tregs in the helper T cells was greater at follow-up compared with that observed at the time of the index event (p < 0.05), although less than those observed at follow-up in SA patients and in controls (p = 0.03 and p = 0.01, respectively) (Figure 3F, Online Figure 1). Changes over time in each group are shown in Online Figure 1.
To confirm whether restoring CREB activity results in increased Treg induction in ACS, we treated helper T cells from NSTEMI patients with okadaic acid (OA) (2 nM), an inhibitor of the catalytic subunit of protein phosphatase 2A involved in CREB dephosphorylation. With OA treatment, CREB phosphorylation in the TCR-stimulated helper T cells was increased in NSTEMI patients (n = 5) (Figure 4A). Moreover, with OA treatment, Foxp3, IL-10, and IL-2 mRNA expression were increased (although not significantly so for IL-2), but transforming growth factor-β mRNA expression was not (Figure 4B). Finally, we evaluated the TCR-induced Treg generation in ACS, SA, and controls with or without OA (n = 10 per group). After 6 days of TCR stimulation, OA treatment was associated with significantly increased Treg frequency in the three groups (Figure 4C).
To determine the role of PTPN22 in TCR signaling and in Treg differentiation in ACS, we transfected the cells of 5 NSTEMI patients with PTPN22-targeting small interfering RNA (siRNA). After 48 h of transfection, we observed reduced PTPN22 expression (Figure 5A) and increased ZAP-70 Y-292 phosphorylation (Figure 5B). We analyzed the expression of Treg after 48 h of TCR stimulation. Helper T cells in NSTEMI patients transfected with PTPN22-targeting siRNA showed an increased frequency of Treg (identified as CD4+CD25+CD127lowFoxp3+) (p = 0.001) (Figure 5C).
To evaluate the role of an inflammatory environment in the observed TCR pathway abnormalities, we stimulated peripheral blood mononuclear cells from 5 controls with αCD3/αCD28 in the presence of IL-6. As shown in Figure 6A, IL-6 treatment was associated with increased CREB and ZAP-70 Y-292 phosphorylation after 24 and 72 h of TCR stimulation, but no difference was observed in ZAP-70 Y-319 activation. Furthermore, after 24 h of stimulation, down-regulation of PTPN22 expression was induced with both IL-6 alone and TCR stimulation alone (Figure 6B), whereas TCR activation with αCD3/αCD28 in the presence of IL-6 did not affect PTPN22 expression. No differences in Treg generation after IL-6 treatment were observed (Figure 6C).
In our study, a double-impairment of TCR signaling in NSTEMI patients resulted in enhanced effector T-cell responsiveness and reduced Treg generation. These TCR-signaling abnormalities consist of a persistent increased expression of the protein tyrosine phosphatase PTPN22 that modulates early TCR activation, and a transient reduced activity, during the acute phase of the disease, of the transcription factor CREB (Central Illustration).
Several studies have highlighted the role of PTPN22 in T-cell activation and autoimmunity, particularly the association of the PTPN22 R620W allelic variant with autoimmune disease development (20). Knowledge of PTPN22 derives mainly from studies of its mouse orthologue, hence the precise mechanism of how PTPN22 regulates the immune cell function in humans is not entirely defined and remains controversial. Nonetheless, this phosphatase plays an important role in immune homeostasis, regulating both effector and regulatory T-cell function in vivo.
A substrate of PTPN22 is ZAP-70, a protein kinase key in the early steps of T-cell activation involved in the fine regulation of TCR-signal transduction (26). Among several tyrosine residues of ZAP-70, Y-292 is involved in down-modulation of the TCR/CD3 complex during the immunological synapse formation and of the proximal events in TCR signaling (26–28). The reduced ZAP-70 Y-292 activation that we observed in NSTEMI patients is consistent with the increased TCR/CD3 recruitment in T cells in ACS and with the enhanced early TCR-induced phosphorylation events demonstrated here and in previous studies (14,15). PTPN22 acts on several substrates, including the activating Y-394 on lymphocyte-specific protein tyrosine kinase and Y-493 on ZAP-70 (19). The phosphorylation of these residues enhances T-cell activation; accordingly, the increased PTPN22 expression in NSTEMI patients that dephosphorylates these residues should be expected to reduce, not enhance, T-cell activation. We propose that in NSTEMI patients, the higher expression of PTPN22, increasing ZAP-70 Y-292 dephosphorylation, reduces the negative-feedback loop controlled by this inhibitory residue, thus enhancing TCR signaling. This working hypothesis also might explain why the polymorphic gain-of-function variant of PTPN22, R620W, is associated with several autoimmune diseases.
The importance of the TCR signal strength in differentiating helper T-cell subsets has increasingly been defined; in particular, strong TCR signaling in early activation stages negatively regulates Treg differentiation (16–18). ACS patients have low levels of circulating Treg (8,11–13), and here we have demonstrated a reduced ability of helper T cells in TCR-induced Treg generation in NSTEMI patients. Hence, the persistent higher expression of PTPN22 in NSTEMI patients, altering the threshold of TCR signaling, might well play a role also in reduced Treg expansion. Of note, a lack of PTPN22 increases both the induction and the immunosuppressive function of extrathymic murine Treg (21). Our experiments in CD4+ T cells from NSTEMI patients transfected with PTPN22-targeting siRNA seem to confirm the role of PTPN22 in ZAP-70 Y-292 dephosphorylation and Treg induction. Moreover, our findings in healthy controls, after in vitro stimulation with the proinflammatory cytokine IL-6, suggest that the effects observed in NSTEMI patients are not secondary to the inflammatory status in these patients and are rather primary and/or “intrinsic” T-cell abnormalities. In particular, IL-6 stimulation was associated with reduced PTPN22 expression and increased ZAP-70 phosphorylation at its inhibitory residue, Y-292.
PTPN22-enhanced expression and the associated early TCR signaling abnormalities in NSTEMI patients were still present at 1-year follow-up, yet the ability of helper T cells from NSTEMI patients to generate Treg was partially recovered at 1 year after the index event. This partial recovery might have been related to a transient defective phosphorylation of CREB, a transcription factor that plays a key role in Treg generation (23). Indeed, this defective phosphorylation, confined to the acute phase of ACS, substantially improves at 1-year follow-up. Phosphorylation of CREB results in an increased expression of the anti-inflammatory cytokine IL-10 and Treg generation, as observed in our study.
Phospho(p)-ZAP-70 Y-292 (pY292) is a crucial residue that negatively regulates TCR activation. The reduced pY292 (probably due to increased expression of PTPN22) in NSTEMI should be expected to enhance, not reduce, CREB phosphorylation. Indeed, during the acute settings of NSTEMI, we observed strong upstream TCR activation and paradoxically weak downstream CREB activation. Thus, the increased CREB activation in NSTEMI at 1-year follow-up, associated with the persistent reduction of pY292 and increased expression of PTPN22, seems reasonable. A possible explanation might come from the recent observation that the expression of the immunomodulatory molecule CD31 is transiently reduced in helper T cells in ACS patients, limited to the acute phase of the disease (14). This molecule interferes with the mitogen-activated protein kinases involved in CREB pathways. Thus, the reduced expression of CD31 in acute-phase ACS might contribute to the reduced CREB activation. However, CREB phosphorylation is regulated by several stimuli beyond TCR triggering (29); recent evidence in experimental models of cardiovascular disease suggests that CREB phosphorylation is down-regulated by risk factors (25). Therefore, a prolonged exposure to environmental stressors, such as oxidized low-density lipoprotein, might lead to altered CREB pathways.
Finally, the reduced levels of binding to IL-2 and IL-10, but not to the Foxp3 promoter regions, observed in the NSTEMI patients underline the complexity of genetic and epigenetic mechanisms involved in Treg differentiation. Indeed, Foxp3 is under the control of transcription factors other than CREB. In particular, the Janus kinase/signal transducers and activators of transcription pathway, triggered by both IL-2 and IL-10, is a principal player in Foxp3 induction and in Treg function and generation (24,30). Thus, the reduced CREB activation might influence Treg induction by reducing the production of these cytokines.
Our study was a prospective observational analysis that included a limited number of patients. No power calculation could be performed because of a lack of previous studies in this setting; thus, the enrollment of 20 patients in each group was arbitrary. NSTEMI and SA patients and controls were not properly matched for age, sex, and risk factors; however, no significant differences were observed in these regards between the NSTEMI and SA groups, and none of these variables were independently associated with PTPN22 expression and ZAP-70 Y-292 and CREB phosphorylation on univariate analysis. These limitations imply two dominant methodological issues that cannot be eluded. First, several variables other than the coronary disease state might explain the differences observed across these three populations. Second, it is impossible in this type of study to determine a cause–effect relationship. Although we performed in vitro experiments using IL-6 stimulation to simulate a proinflammatory environment and exclude potential upstream determinants, we cannot rule out that the reduced CREB activity might have been a part of the general stress response in acute phase ACS. Moreover, in our experimental conditions (cell transfection with targeting siRNA), PTPN22 silencing induced only a modest, although significant, increase in Treg frequency, so further studies are needed. Finally, we focused our attention on TCR-mediated signal transduction pathways only, excluding the role of cytokine signaling. A shift in the TCR threshold setting does not eliminate the role of the cytokine environment as a key factor in controlling T-cell activity and differentiation. Further and deeper studies will help to determine the role of cytokine signaling, in particular IL-2 and IL-10, in ACS. However, a strong upstream TCR activation associated with weak downstream CREB activation may set the stage for an enhanced immune responsiveness unhampered by immune tolerance.
Our study adds novel pieces of information to the important role of adaptive immunity alterations in ACS. PTPN22 and CREB might represent novel potential therapeutic targets as well as biomarkers in the subset of ACS patients in whom an inflammatory outburst is the likely cause of coronary instability.
COMPETENCY IN MEDICAL KNOWLEDGE: Elevated soluble markers of inflammation in patients with ACS are associated with worse outcomes, and T-cell dysregulation is pivotal to the immune response.
TRANSLATIONAL OUTLOOK: Markers of adaptive immunity, such as PTPN22 and CREB, may be more accurate guides to prognosis in patients with ACS and could be useful in developing specific anti-inflammatory treatments.
The authors thank Dr. Manuela Bianco for technical and scientific support.
Dr. Biasucci has received a research grant (Protocollo n. 70501473) from the Ministero dell'Istruzione, dell'Universitè e della Ricerca. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Liuzzo and Crea contributed equally to this work.
- Abbreviations and Acronyms
- acute coronary syndrome(s)
- cyclic adenosine monophosphate response element–binding protein
- messenger ribonucleic acid
- non-ST-segment elevation myocardial infarction
- okadaic acid
- protein tyrosine phosphatase, nonreceptor type, 22
- stable angina
- T-cell receptor
- regulatory T cell
- zeta-chain–associated protein kinase of 70 kDa
- Received September 11, 2014.
- Revision received January 7, 2015.
- Accepted January 14, 2015.
- 2015 American College of Cardiology Foundation
- Crea F.,
- Liuzzo G.
- Liuzzo G.,
- Goronzy J.J.,
- Yang H.,
- et al.
- Caligiuri G.,
- Paulsson G.,
- Nicoletti A.,
- Maseri A.,
- Hansson G.K.
- De Palma R.,
- Del Galdo F.,
- Abbate G.,
- et al.
- Liuzzo G.,
- Biasucci L.M.,
- Trotta G.,
- et al.
- Wigren M.,
- Björkbacka H.,
- Andersson L.,
- et al.
- Zhang W.C.,
- Wang J.,
- Shu W.,
- et al.
- Pryshchep S.,
- Goronzy J.J.,
- Parashar S.,
- Weyand C.M.
- Molinero L.L.,
- Miller M.L.,
- Evaristo C.,
- Alegre M.L.
- Wu J.,
- Katrekar A.,
- Honigberg L.A.,
- Smith A.M.,
- et al.
- Brownlie R.J.,
- Miosge L.A.,
- Vassilakos D.,
- Svensson L.M.,
- Cope A.,
- Zamoyska R.
- Yu C.T.,
- Shih H.M.,
- Lai M.Z.
- Wen A.Y.,
- Sakamoto K.M.,
- Miller L.S.
- Schauer I.E.,
- Knaub L.A.,
- Lloyd M.,
- et al.
- Davanture S.,
- Leignadier J.,
- Milani P.,
- et al.
- Magnan A.,
- Di Bartolo V.,
- Mura A.M.,
- et al.
- Sabat R.,
- Grütz G.,
- Warszawska K.,
- et al.