Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Mar 20.
Published in final edited form as: Br J Haematol. 2014 Sep 24;168(2):274–283. doi: 10.1111/bjh.13126

In vitro TNF blockade enhances ex vivo expansion of regulatory T cells in patients with immune thrombocytopenia

Hui Zhong 1, James Bussel 2, Karina Yazdanbakhsh 1
PMCID: PMC5358513  NIHMSID: NIHMS850421  PMID: 25252160

Summary

Tumour necrosis factor-α (TNF) is an inflammatory cytokine that is elevated in a number of autoimmune diseases including immune thrombocytopenia (ITP), a bleeding disorder characterized by low platelet counts. In vitro TNF blockade increases expansion of the regulatory T cell (Treg) IKZF2 (also termed Helios) subset in T cell-monocyte cocultures from healthy donors, but its role on proliferative responses of Tregs in ITP patients, who have altered immunoregulatory compartment, remains unclear. TNF in CD4+ T cells from patients with chronic ITP were elevated and negatively correlated with peripheral Treg frequencies, suggesting a possible inhibitory effect of TNF on ITP Tregs. In vitro antibody neutralization with anti-TNF in T cell-monocyte cocultures resulted in a robust expansion of pre-existing ITP Tregs, higher than in healthy controls. Similar to the effects of anti-TNF antibodies, TNF blockade with antibodies against TNFRSF1B (anti-TNFRSF1B, previously termed anti-TNFRII) almost doubled ITP Treg expansion whereas neutralization with anti-TNFRSF1A (anti-TNFRI) antibodies had no effect on proliferative responses of Tregs. In addition, TNFRSF1B levels on ITP Tregs were significantly elevated, which may explain the increased susceptibility of patient Tregs to the actions of TNF blockade. Altogether, these data raise the possibility that TNF blockers, through their ability to increase Treg proliferation, may be efficacious in ITP patients.

Keywords: immune thrombocytopenia, regulatory T cells, IKZF2 (Helios), Tumour necrosis factor, TNFRSF1B (TNFRII)

Immune thrombocytopenia (ITP) is an autoimmune bleeding disorder resulting from the immune destruction of platelets and insufficient platelet production. Anti-platelet autoantibodies are responsible for increased platelet clearance by the reticuloendothelial system and probably for inhibition of megakaryopoiesis (Gernsheimer, 2009). In addition, a role for cytotoxic T cells in the direct lysis of platelets and megakaryocytes in the bone marrow has been proposed.(Olsson et al, 2003; Zhang et al, 2006; Li et al, 2007) ITP patients harbour platelet-autoreactive T helper (Th) cells with increasing cytokine imbalance towards interleukin 2 (IL2) and γ-interferon (IFNG, IFN-γ) (Semple & Freedman, 1991; Kuwana et al, 1998; Ogawara et al, 2003), as well as altered regulatory T cell (Treg) and B cell (Breg) numbers and function (Ling et al, 2007; Liu et al, 2007; Sakakura et al, 2007; Stasi et al, 2008; Yu et al, 2008; Audia et al, 2011; Li et al, 2012; Yazdanbakhsh et al, 2013). More recently, we found that Treg dysregulation was partly due to increased inhibition of Treg proliferation by ITP CD16+ monocytes through IL12 (Zhong et al, 2012; Yazdanbakhsh et al, 2013). However, the possibility still remains that other cells or inflammatory cytokines may also contribute to altered Treg compartment in ITP patients.

Tumour necrosis factor-α (TNF) is an inflammatory cytokine that plays a pivotal role in infectious and autoimmune diseases, as evidenced by the successful treatment of these diseases with anti-TNF drugs. Treatment with TNF blockers has only been reported in a small group of patients with ITP, albeit efficacious in all 4 treated patients (McMinn et al, 2003; Litton, 2008). Higher levels of TNF have been reported in the plasma/serum from ITP patients,(Culic et al, 2013; Jernas et al, 2013; Talaat et al, 2014) although TNF expression levels in specific ITP cell types, such as CD4+ T lymphocytes, has not been examined. Interestingly, ITP patients have an increased frequency of TNF −308 A polymorphic allele associated with higher TNF-α production (Pehlivan et al, 2011; Sarpatwari et al, 2011; El Sissy et al, 2014). TNF binds to two distinct receptors: TNFRSF1A (TNFRI, p55) and TNFRSF1B (TNFRII; p75) (Locksley et al, 2001). TNFRSF1A is ubiquitously expressed on almost all cell types while TNFRSF1B is typically expressed on cells of the immune system including Tregs (Faustman & Davis, 2013). Surface expression of TNFRSF1B can be induced by IL1B, IL2 and TNF (Carpentier et al, 2004), cytokines reported to be increased in ITP patients (Rocha et al, 2010; Culic et al, 2013; Jernas et al, 2013; Talaat et al, 2014), although TNFRSF1B levels on Tregs in ITP patients have not been measured. It is generally accepted TNF can regulate Treg differentiation and proliferation, but the exact outcome appears to depend on Treg subset, the disease setting and the cytokine microenvironment (Biton et al, 2012). For example, in a murine autoimmune diabetes model, effector T cell-derived TNF was shown to be essential for Treg development (Grinberg-Bleyer et al, 2010). In addition, in vitro studies have found that addition of TNFRSF1B agonists accelerate IL2-driven human Treg expansion, consistent with the positive effects of TNF on Treg development (Okubo et al, 2013). However, contrary to these data are reports suggesting a suppressive role of TNF on Treg compartment. These include in vitro data showing that addition of TNF or agonistic antibody to TNFRSF1B can reverse the suppressive function of Tregs by downmodulating FOXP3 expression (Valencia et al, 2006; Nie et al, 2013). In addition, increased frequency of circulating of Treg subset lacking expression of IKZF2 (also termed Helios), a transcription factor that was originally reported as a marker that could distinguish natural from induced Tregs (Shevach & Thornton, 2014) have been described in patients with rheumatoid arthritis (RA) responsive to anti-TNF treatment (McGovern et al, 2012). Consistent with these latter data, we recently showed that addition of TNF to in vitro cultures from healthy donors suppressed IKZF2-Treg proliferation whereas blocking anti-TNF antibodies preferentially increased proliferation of the IKZF2-Treg subset (Zhong & Yazdanbakhsh, 2013). It is not currently known whether TNF blockade has a positive or negative effect on Treg compartment, and its IKZF2+ and IKZF2− subsets in ITP patients. There are currently several TNF blockers in the clinic for treatment of various disease indications (Palladino et al, 2003). Since their use was efficacious in four treated ITP patients (McMinn et al, 2003; Litton, 2008), a detailed examination of the effect of TNF on Tregs in ITP patients may not only help to dissect the mechanism of Treg dysfunction in ITP patients but also provide a potential therapeutic target for ITP patients. In the current study, we measured TNF levels in circulating CD4+ T cells and monocytes from ITP patients and compared the effect of TNF blockade on Treg subsets between ITP patients and healthy controls using an in vitro T cell-monocyte co-culture system.

Materials and methods

Human samples

All the studies were approved by the Institutional Review Boards of the New York Blood Center and Weill Cornell Medical School. Peripheral blood was obtained after consent from 40 patients with ITP (Patients 1–40, Table I). All of the patients had chronic ITP (defined >1 year since diagnosis, age range 15–60 years old, Table I), except three (Patients 20, 22 and 23) who were in remission having platelet counts >100 × 109/l for several months at the time of the blood draw. In some of the assays, we sampled blood from patients who had not received any ITP treatments for various lengths of time (Table I). Patients who were on ITP treatment were exclusively on US Food and Drug Administration (FDA)-approved thrombopoietic agents (Romiplostim or Promacta) at the time of blood sampling (Table I). Not all patients were analysed in every assay (Table I). As a control, peripheral blood samples were obtained from closely age-matched healthy volunteer donors of the New York Blood Center.

Table I.

Clinical characteristics of ITP patients at the time of blood sampling for the study.

Patient Plt count (× 109/l) ITP Treatment Duration off treatment Dose Duration of treatment Age (years) Sex Splene ctomy % CD4+ TNF+ % Tregs
CD4+TNF levels in patients off treatment (Fig 1B, C, F)
 1 10 Off 1 month 54 M No 22 N/A
 2 13 Off 1 month 17 M No 23 3·0
 3 18 Off 2 months 15 M Yes 17 3·2
 4 18 Off 1 month 56 M No 19 4·1
 5 21 Off 2 months 45 M Yes 49 2·6
 6 33 Off 2 weeks 29 M Yes 6 3·0
 7 36 Off 9 months 56 M Yes 20 3·2
 8 40 Off 2 weeks 36 M Yes 47 N/A
 9 42 Off 1 month 18 M No 22 3·5
 10 47 Off 3 months 50 M Yes 33 N/A
 11 198 Off 3 weeks 20 F Yes 17 3·5
 12 8 Off 1 month 59 F No 55 3·7
 13 9 Off 3 weeks 18 F Yes 65 N/A
 14 13 Off 1 month 16 M No 11 3·3
 15 13 Off 3 weeks 29 F Yes 22 4·1
 16 15 Off 1 month 37 M No 39 3·2
 17 17 Off 5 years 16 M No 13 N/A
 18 17 Off 3 months 18 F Yes 25 N/A
 19 54 Off 15 months 60 M Yes 26 4·1
 20 132 Off 10 months 48 F Yes 61 N/A
 21 236 Off 2 weeks 48 F Yes 21 3·6
 22 242 Off 6 months 57 F No 49 2·2
 23 466 Off 2 years 35 M Yes 31 3·2
CD4+TNF levels in patients on treatment (Fig 1D)
 1 22 Romiplostim 7 mg/kg 6 years 54 M Yes 33 4·4
 2 52 Promacta 50 mg 2 months 17 M Yes 18 3·6
 3 45 Romiplostim 5 mg/kg 1 month 15 M Yes 16 3·3
 4 43 Promacta 75 mg 8 months 56 M Yes 35 2·7
 5 49 Promacta 75 mg 3 months 45 M Yes 52 5·0
 6 91 Promacta 50 mg 1 month 29 M No 24 3·3
 7 64 Promacta 25 mg 2 months 56 M No 23 4·9
 8 68 Romiplostim 8 mg/kg 3 months 36 M No 25 4·7
 9 191 Romiplostim 10 mg/kg 2 years 18 M No 15 4·8
 10 233 Promacta 75 mg 3 years 50 M No 17 3·5
 11 381 Romiplostim 7 mg/kg 2 years 20 F Yes 34 3·8
CD4+TNF levels in patients on treatment (Fig 1E, F)
 24 35 Promacta 50 mg 9 months 19 F No 24 2·7
 25 36 Promacta 25 mg 2 years 50 F Yes 45 2·5
 26 45 Promacta 75 mg 4 months 19 M No 17 3·6
 27 48 Romiplostim 15 mg/kg 5 years 36 M Yes 39 2·4
 28 87 Promacta 75 mg 1 month 49 F No 60 N/A
 29 132 Romiplostim 10 mg/kg 5 years 56 F Yes 66 2·8
 30 148 Promacta 50 mg 2 years 54 F No 15 3·8
 31 154 Romiplostim 5 mg/kg 3 months 48 F Yes 43 3·1
T cell-monocyte coculture studies (Fig 1H)
 1 22 Romiplostim 7 mg/kg 6 years 54 M Yes 33 4·4
 3 45 Romiplostim 5 mg/kg 1 month 15 M Yes 16 3·3
 4 43 Promacta 75 mg 8 months 56 M Yes 35 2·7
 8 68 Romiplostim 8 mg/kg 3 months 36 M No 25 4·7
 27 48 Romiplostim 15 mg/kg 5 years 36 M Yes 39 2·4
 29 132 Romiplostim 10 mg/kg 5 years 56 F Yes 66 2·8
 33 127 Off 2 weeks 20 M Yes N/A N/A
 34 354 Off 1 month 27 F Yes N/A N/A
TNF blockade studies (Fig 2 and 3)
 1 22 Romiplostim 7 mg/kg 6 years 54 M Yes 33 4·4
 3 45 Romiplostim 5 mg/kg 1 month 15 M Yes 16 3·3
 8 68 Romiplostim 8 mg/kg 3 months 36 M No 25 4·7
 27 48 Romiplostim 15 mg/kg 5 years 36 M Yes 39 2·4
 29 132 Romiplostim 10 mg/kg 5 years 56 F Yes 66 2·8
 32 9 Promacta 25 mg 1 month 21 F Yes N/A N/A
 33 127 Off 2 weeks 20 M Yes N/A N/A
 34 354 Off 1 month 27 F Yes N/A N/A
TNFRSF1B levels (Fig 4)
 7 36 Off 9 months 56 M Yes 20 3·2
 35 17 Off 2 weeks 60 F Yes N/A N/A
 36 32 Off 2 weeks 20 F No N/A N/A
 37 34 Off 1 year 54 F No N/A N/A
 38 45 Off 2 months 60 F No N/A N/A
 39 65 Off 6 months 59 F Yes N/A N/A
 40 100 Off 3 months 27 M No N/A N/A
Open in a new tab

The list of patients analysed for each study is indicated. Clinical characteristics, percent CD4+TNF+ and circulating Tregs are given for each patient whenever the data was available. ‘Off’ ITP treatment indicates that patients had been without ITP medication for at least 2 weeks prior to the study. Romiplostim and Promacta are FDA-approved thrombopoietic agents. Patients 1–11 were studied while off treatment and on treatment with thrombopoietic agents. Patients 20, 22 and 23 were in remission at the time of blood sampling.

ITP, immune thrombocytopenia; Plt, platelet; TNF, tumour necrosis factor α; Tregs, T regulatory cells; N/A, not available.

Cell isolation and purification

Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll (GE Healthcare, Port Washington, NY, USA) density centrifugation and subjected to cell subset purification by magnetic beads (all from Miltenyi Biotec, Auburn, CA, USA) for isolation of total T cells and monocytes using the PAN T cell isolation kit and CD14 microbeads, respectively (purity>95% for both). For some studies, CD4+ cells were first enriched from PBMCs using magnetic beads, stained with anti-CD25-allophyocyanin (APC) (BD Biosciences, San Jose, CA, USA) and subjected to cell sorting to purify CD4+CD25− T cells (>95%, data not shown).

T cell and monocyte stimulation assays

Purified T cells or sort-purified CD4+CD25− T cells were stained with Carboxyfluoresceindiacetatesuccinimidyl ester (CFSE, Invitrogen, Grand Island, NY, USA) according to manufacturer’s instructions. CFSE-labelled T cells (1·25 × 105 cells/ml) were mixed with autologous purified total monocytes at a ratio of 2:1 in culture medium containing RPMI 1640 (Invitrogen) supplemented with 5% human AB serum (Valley Biomedical, Winchester, VA, USA), 2 mmol/l glutamine (Invitrogen), 100 units penicillin and streptomycin (Invitrogen) and 0·055 mmol/l 2-mercaptoethanol (Invitrogen) and stimulated with 1 μg/ml soluble anti-CD3 antibody (1 μg/ml, clone HIT3α, BD Biosciences) in U-bottomed 96-well plates for 7 days. For antibody neutralization studies, anti-TNF (1 μg/ml, clone MABTNFA5, TNF Blockade Increases ITP Tregs BD Biosciences), anti-TNFRSF1A (4 μg/ml, clone 16805, R&D Systems, Minneapolis, MN, USA), anti-TNFRSF1B (4 μg/ml, clone 22221, R&D Systems) or isotope matched controls (2 μg/ml, R&D Systems) were added at the start of the cocultures.

Intracellular FOXP3, IKZF2 and TNF expression analysis

All expression analyses were performed by flow cytometry using BD FACSCanto with Diva software (BD Biosciences). For analysis of peripheral circulating Tregs, whole blood (150 μl) was first incubated with CD4-peridinin chlorphyll (PerCP) and CD25-APC (BD Biosciences) and then stained with FOXP3 (clone PCH101; eBiosciences, San Diego, CA, USA) and an isotype control following manufacturer’s instructions. To analyse Tregs in the anti-CD3 stimulated proliferation studies, cells were first fixed and permeabilized with FOXP3 fixation/permeabilization solution (eBioscience, San Jose, CA, USA), and incubated with anti-CD4-PerCP (clone RPA-T4, eBioscience), anti-CD3 Alexa700 (clone UCHT1, eBioscience), anti-FOXP3-phycoerythrin (PE) (clone PCH101, eBioscience) and anti-IKZF2-APC (clone 22F6; Biolegend, Inc, San Diego, CA, USA) following manufacturer’s instructions. CD4+ T cells that had divided were defined as the CFSEloCD3+CD4+ population. Within the CFSEloCD3+CD4+ cells, Tregs were identified as FOXP3 hiCD4+ T cells, based on previous studies showing that only the high level expression of FOXP3 correlates with Treg population in anti-CD3 stimulated PBMCs (Gavin et al, 2006; Zhong et al, 2012). To measure the proliferative responses of Treg susbets, the percentage of IKZF2+ or IKZF2− Tregs in divided CD4+ T cells was also analysed.

Intracellular TNF+ expression in sort-purified CD4+ CD25− cells was measured following stimulation with phorbol myristate acetate (PMA, 0·25 μg/ml; Sigma-Aldrich, St. Louis, MO, USA) and ionomycin (0·375 μg/ml, Sigma-Aldrich) in the presence of GolgiPlug at 37°C for 5 h, after which the cells were fixed and permeabilized following manufacturer’s instructions followed by intracellular staining with anti-TNF antibody (BD Biosciences). To measure TNFRSF1B expression levels on Tregs, PBMCs were first stained with anti-CD4, anti-CD25 and anti-TNFRSF1B-biotin (clone hTNFR-M1, BD Biosciences) for 15 min, washed several times followed by incubation with Streptavidin-PE-Cy7 (eBioscience) for another 15 min. After that, the cells were subjected to intracellular staining for FOXP3 and IKZF2 as described above.

Statistical analysis

Data are expressed as mean values ± standard error of the mean (SEM). Statistical significance of differences between groups was determined by Mann–Whitney test and statistical significance of differences of paired data were determined by paired t test: P-values ≤0·05 were considered statistically significant. Association studies were performed using the Spearman Rank-Order Correlation analysis. Statistical analyses were performed using PASW Statistics Version 18 software (SPSS Inc. Chicago, IL, USA).

Results

ITP CD4+ T cells express higher levels of TNF

Circulating TNF levels are elevated in ITP patients (Culic et al, 2013; Jernas et al, 2013; Talaat et al, 2014), but their levels in TNF expressing cells, such as CD4+ cells, has not been previously described in this patient population. We therefore measured TNF levels in CD4+ T cells from a group of patients with chronic ITP (n = 23; median age 36, range 15–60 years; 15 males and 8 females with median platelet count 27 × 109/l, range 8–466 × 109/l; 14 were splenectomized), who had been without any ITP treatment for at least 2 weeks prior to the blood draw (Table I). Included in this analysis were five patients with platelet counts >100 × 109/l; Patients 11 and 21 had stopped treatment with thrombopoietic agents two and three weeks, respectively, prior to the blood draw and patients 20, 22 and 23 were in remission. Analysis was restricted to using sort-purified CD4+CD25− cells to avoid contamination with potentially anergic, recently-activated (CD4+CD25int) or Treg (CD4+CD25hiFOXP3+) cell populations that may express less TNF upon restimulation. Following PMA and ionomycin stimulation, the overall frequency of TNF+ cells was higher in sorted CD4+CD25− cells (Fig 1A–B) from ITP patients than from healthy controls (n = 20, P = 0·02). Sub-analysis of CD4+TNF+ percentages in patients who had been off ITP treatment for ≥2 months (32·4 ± 5·0%, n = 10, Patients 3, 5, 7, 10, 17, 18, 19, 20, 22 and 23) as compared to those in healthy donors (19·1 ± 2·4%, n = 10) also revealed significant differences between the two groups (P = 0·02). In addition, the cohort that had been off ITP treatment for ≥2 months had similar levels of CD4+TNF-α+ as patients without ITP treatment for ≤1 month (28·4 ± 4·9%, n = 13, P = 0·4) or ≤3 weeks (29·7 ± 9·0%, n = 6, Patients 6, 8, 11, 13, 15, 21, P = 0·9), suggesting that the length of time off ITP treatment did not affect CD4+TNF+ levels. Moreover, CD4+TNF+ cell frequencies were comparable, regardless of whether the patients had undergone splenectomy or not (Fig 1C). Similarly, analysis of a subset of patients (n = 11, Patients 1–11) did not reveal any significant difference in the levels of CD4+TNF levels before and during treatment with thrombopoietic agents (Fig 1D). In addition, comparison of our ‘off treatment’ patient cohort (n = 23) and a separate group of patients with chronic ITP on thrombopoietic agents (n = 8; Patients 24–31: median age 41, range 19–56 years; 2 males and 6 females with median platelet counts of 85 × 109/l, range 35–154 × 109/l; 4 were splenectomized, Table I) also indicated similar overall frequency of TNF+ cells in sorted CD4+CD25− cells as patients without ITP treatment (Fig 1E), suggesting that these drugs may not affect TNF levels in ITP CD4+ T cells. In some patients, we also measured circulating Treg frequencies, classically defined as CD25hiFOXP3+ cells in CD4+ T lymphocytes (Table I, % Tregs) and found a trend toward negative correlation between frequencies of CD4+TNF+ cells and Tregs (Fig 1F, P = 0·03), consistent with the reported inhibitory effect of TNF on FOXP3 expression (Valencia et al, 2006; Nie et al, 2013). Given that PMA/ionomycin induces T cell receptor (TCR)-independent activation of T cells, we also checked TNF levels in CD4+ T cells stimulated using anti-CD3 antibodies (i.e., TCR-dependent activation) for 7 days in the presence of autologous monocytes (Fig 1G) in a smaller group of patients (n = 8, Table I) as well as in healthy controls (n = 9). Similar to the PMA/ionomycin data, we found a higher frequency of TNF+CD4+ T cells following anti-CD3/monocyte stimulation in ITP patients as compared to healthy controls (Fig 1H, ‘Total CD4+’). Given that T cells proliferate when activated through anti-CD3 and monocytes, we also compared the frequency of divided CD4+ T cells (CFSElo) that express TNF and found a higher frequency of these divided cells in the ITP patient group (Fig 1G–H, ‘CD4+CFSElo’).

Fig 1.

Fig 1

Open in a new tab

Elevated frequency of TNF+ cells in ITP patients. (A) Representative histogram showing the gating strategy based on isotype control to measure intracellular TNF+ expression levels in CD4+ T cells following 5-h stimulation with phorbol myristate acetate (PMA)/ionomycin. (B) Percentage of CD4+ cells expressing cytoplasmic TNF + cells in PMA/ionomycin-stimulated, sorted CD4+CD25− T cells from healthy controls (n = 20) and patients with chronic ITP who had been off ITP treatment for at least 2 weeks prior to the blood draw (n = 23, Patients 1 23, Table I). (C) Frequency of CD4+TNF+ cells in the same cohort of patients as in B but grouped based on whether they had not or had undergone splenectomy. (D) To determine the effect of thrombopoietin (THPO) agent treatment, frequency of CD4+TNF+ cells in 11 patients (Patients1–11, Table I) off treatment (‘No’) and on treatment with THPO agents for at least 2 weeks at the time of the blood draw (‘Yes’) were compared. (E) Comparison of CD4+TNF+ cell frequencies in the same cohort of patients as in B who were all off treatment (‘No’) vs a separate group of patients (Patients 24–31, Table I) who were on treatment with THPO agents for at least 2 weeks at the time of the blood draw (‘Yes’). (F) In a subset of patients (n = 23), circulating Treg frequencies, as defined by percentages of CD25hiFOXP3+ in CD4+ T cells, were measured. Correlation between Treg and CD4+TNF+ cell frequency shows a negative relationship (r = −0·46, P = 0·03) between these two cell populations in ITP patients, as determined by Spearman Rank-Order Correlation analysis. (G) Representative dot plot of TNF expression in CD4+ T cells with gating strategy to measure cytoplasmic TNF+ expression in divided (CFSElo) CD4+ T cells in T cell-monocyte cocultures stimulated with soluble anti-CD3 antibody for 7 days. (H) Frequency of TNF+ cells in total CD4+ T cells and divided CD4+ T cells (CD4+CFSElo) in healthy controls (n = 9) and chronic ITP patients (n = 8).

Anti-TNF antibodies induce a strong Treg expansion in ITP patients

We have previously shown that in anti-CD3 stimulated T cell cultures from healthy donors, addition of TNF cytokine inhibits proliferation of the IKZF2-Tregs subset, which was originally reported to represent inducible Tregs (Shevach & Thornton, 2014) whereas antibody blocking with anti-TNF increases IKZF2-Treg expansion (Zhong & Yazdanbakhsh, 2013). Given that our above data indicate that CD4+TNF+ frequencies can be negatively associated with circulating Tregs in ITP patients, we hypothesized that the higher TNF levels may contribute to lower Tregs in ITP patients and that antibody neutralization with anti-TNF would have a more pronounced effect on proliferation of ITP Tregs. As a first step to test whether TNF blockade has an effect on Treg proliferation in patients with ITP, we randomly selected patients, regardless of their CD4+TNF+ levels (Table I). Purified CFSE-labelled total T cells from ITP patients (n = 8) were stimulated with anti-CD3 in the presence of purified, autologous peripheral monocytes without or with anti-TNF antibody for 7 days. Although CD4+ T cell proliferation was not affected after blocking with anti-TNF (P = 0·35), ITP Treg expansion, as measured by the frequency of CFSEloFOXP3+ in CD4+ T cells, was significantly increased (>2-fold of IgG control, P = 0·002, Fig 2A–C), specifically in the IKZF2− Treg subset (Fig 2D–E). Furthermore, expansion of Tregs, including both IKZF2+/− subsets, was significantly higher in coculture from ITP patients as compared to healthy controls (Fig 2F). As these studies were performed using total T cells in the T cell-1monocyte cocultures, it was unclear whether the anti-TNF treatment affected the expansion of pre-existing Tregs or their conversion from effector T cells. To distinguish between these two possibilities, we compared the effects of anti-TNF blockade in T cell-monocyte cocultures from four ITP patients (Patients 3, 29, 33 and 34, Table I) using total CD4+ T cell fraction vs sorted CD4+CD25− T cells (without Tregs). Unlike T cell-monocyte cocultures that included both Tregs and effector T cells, we did not detect any Tregs in the cocultures with sorted CD4+CD25− T cells in the absence or presence of anti-TNF, suggesting that anti-TNF blockade affects expansion of pre-existing Tregs rather than their induction (Fig 2G).

Fig 2.

Fig 2

Open in a new tab

In vitro proliferation of ITP Tregs is increased following anti-TNF blockade. (A) Representative dot plot showing gating strategy to measure FOXP3 expression in divided, CFSElo CD4+ T cells in T cell-monocyte cocultures stimulated with soluble anti-CD3 antibody for 7 days. Frequency of (B) CFSEloCD4+ T cells and (C) CFSElo Tregs in 7 day T cell-monocyte cocultures from ITP patients (n = 9) treated with isotype control (IgG) and anti-TNF antibodies are shown; P values calculated by paired t-test. (D) Representative dot plot of FOXP3 and IKZF2 expression CFSElo CD4+ T cells from 7-day T cell-monocyte cocultures stimulated with soluble anti-CD3 showing the gating strategy to measure IKZF2+ and IKZF2− Tregs in divided CD4+ T cells. (E) Frequency of IKZF2+/− Tregs in CFSEloCD4+ T cells in T cell-monocyte cocultures from ITP patients treated with isotype control (IgG) and anti-TNF antibodies are shown. (F) Fold change in total Tregs and IKZF2+/− Treg subsets after anti-TNF blockade in the T cell-monocyte cocultures was calculated (IgG group was set to 100%) in healthy controls and ITP patients; differences between groups indicated by P values determined by Mann–Whitney test. (G) Total T cells or sorted CD4+CD25− T cells were purified from 4 ITP patients, co-cultured with autologous total monocyte fraction, and stimulated with soluble anti-CD3 antibody for 7 days in the presence of IgG or neutralizing anti-TNF antibody. Frequencies of Tregs in CFSEloCD4+ cells are shown. Under these conditions, <1% Tregs were detected in co-cultures with sorted CD4+CD25− T cells. All data are expressed as mean values ± standard error of the mean.

TNF blockade with anti-TNFRSF1B, but not anti-TNFRSF1A affects Treg proliferation in ITP patients

To determine the relative role of each of the two TNF receptors, TNFRSF1A and TNFRSF1B, on Treg expansion in ITP patients, we used specific neutralization antibodies in our coculture assays. Whereas anti-TNFRSF1A antibody treatment did not alter Treg expansion (Fig 3A–B), anti-TNFRSF1B treatment induced a strong Treg expansion, almost doubling their proliferation (Fig 3C). These data indicate that TNF inhibits Treg expansion through TNFRSF1B but not TNFRSF1A. Although in healthy controls treatment with antibodies against TNFRSF1B increased IKZF2− Treg expansion with little effect on IKZF2+ Treg expansion, anti-TNFRSF1B treatment increased both IKZF2+/− subsets in ITP patients, suggesting that its effect is more pronounced in ITP patients as compared to healthy controls (Fig 3D). Given that ITP Tregs are more sensitive to blockade with anti-TNFRSF1B, we examined the relative expression levels of TNFRSF1B on Tregs, classically defined as CD25hiFOXP3+ in CD4+ T cells, from ITP patients vs healthy controls in unstimulated PBMC samples. Almost all CD25hiFOXP3+ Tregs were TNFRSF1B-positive with slightly higher TNFRSF1B levels on IKZF2-Tregs as compared to IKZF2+ Tregs (Fig 4A–B). However, ITP patients expressed higher levels of TNFRSF1B on Tregs including both IKZF2+/− subsets (Fig 4B), consistent with their higher sensitivity to TNF regulation as compared to Tregs from healthy controls.

Fig 3.

Fig 3

Open in a new tab

Anti-TNFRSF1B antibody increases ITP Treg expansion in T cell-monocyte cocultures. (A) Frequency of CFSElo total Tregs and IKZF2+/− subsets in 7 day T cell-monocyte cocultures from ITP patients (n = 8) treated with isotype control (IgG) and anti-TNFRSF1A; P values calculated by paired t-test. (B) Fold change in total Tregs and IKZF2+/− Treg subsets after anti-TNFRSF1A blockade in the T cell-monocyte cocultures was calculated (IgG group was set to 100%) in healthy controls (n = 12) and ITP patients from A; differences between groups indicated by P values determined by the Mann–Whitney test. (C) Frequency of CFSElo total Tregs and IKZF2+/− subsets in the T cell-monocyte cocultures from the same patient samples as in (A) were treated with anti-TNFRSF1B. The P values indicate the difference in the values as compared to the IgG control and were calculated by a Paired t-test. (D) Fold change in total Tregs and IKZF2+/− Treg subsets after anti-TNFRSF1B blockade (IgG group was set to 100%) are shown; differences between groups are indicated by P values determined by the Mann–Whitney test.

Fig 4.

Fig 4

Open in a new tab

TNFRSF1B expression levels on Tregs are elevated in ITP patients. (A) Representative dot plot of (left) CD25 and FOXP3 expression in CD4+ T cells in peripheral blood mononuclear cells (PBMCs) from an ITP patient to show gating strategy for CD25hiFOXP3+ (Tregs); (right, top) representative dot plot of TNFRSF1B and FOXP3 and (right, bottom) of TNFRSF1B and IKZF2 expression in gated CD25hiFOXP3+ (Tregs) in the same patient. (B) Surface TNFRSF1B expression levels as determined by mean fluorescent units in CD25hiFOXP3+ Tregs and IKZF2+/− subsets are shown in unstimulated PBMCs from healthy controls ‘HD’ (n = 6) and ITP patients (n = 7). Data are expressed as mean values ± standard error of the mean; P values were calculated by the Mann–Whitney test.

Discussion

In this study, overall higher levels of TNF expression were detected in CD4+ T cells from patients with chronic ITP and these levels correlated negatively with peripheral Treg frequencies. Furthermore, using an in vitro T cell proliferation assay with purified T cells and monocytes, a robust increase in Treg expansion, including both IKZF2+/− Treg subsets, was detected following TNF blockade in ITP patients as compared to healthy controls. The expanded cells were derived from the pre-existing pool of Tregs rather than de novo-induced Tregs. Antibody neutralization with anti-TNFRSF1B, but not anti-TNFRSF1A, increased Treg proliferation, suggesting that the effect of anti-TNF treatment on Tregs was mediated through TNFRSF1B. Interestingly, we found elevated levels of TNFRSF1B on Tregs from ITP patients, which may explain the increased susceptibility of patient Tregs to the actions of TNF blockade. As TNF can down-modulate the function of human Tregs (Valencia et al, 2006; Nie et al, 2013), our current data raises the possibility that heightened sensitivity of Tregs due to the combined effects of elevated receptor expression (TNFRSF1B) and ligand pool (TNF) may contribute to defective Treg compartment in ITP patients. Although the current study was performed using samples from patients with a range of platelet counts, including some patients who were in remission, and the TNF blockade experiments included analysis of patient samples with both high and low CD4+TNF-α levels, it is likely that patients with the most severe thrombocytopenia would probably have the most dysfunctional T cells and therefore such patients would most benefit from the use of TNF blockers. It should be noted that our previous data suggest that patients treated with thrombopoietin (THPO) agents have an improved Treg compartment (Bao et al, 2010). However, we did not find any difference in TNF+CD4+ T cells in our cohort of patients treated with THPO agents as compared to those off treatment. These data suggest that the effect of THPO agents on the Treg compartment is not mediated through TNF, raising the possibility that combining TNF blocker drugs with THPO agents might enable improvement of the Treg compartment through different synergistic mechanisms to restore the immune balance in these patients.

Currently, TNF blockade strategies involve antibodies against TNF, such as the mouse-human chimeric antibody (infliximab), fully humanized antibodies (Adalimumab and Golimumab), PEGylated Fab fragment (Certolizumab) or the use of soluble TNF receptor, including Etanercept, which is a fusion protein consisting of the extracellular portion of the TNFRSF1B linked to the Fc region of the human IgG1. With respect to mechanism of action, it should be noted that TNF receptor drugs, such as Etanercept, block the action of both TNF and LTA (TNFB, TNF-β), which share the same receptors, whereas antibodies against TNF are specific to TNF. In fact, treatment efficacy appears to differ depending on the TNF blocker used, which is partly explained by differences in their mechanism of action, although clinical outcomes are also associated with the underlying disease.(Mpofu et al, 2005) For example, all TNF blockers are highly effective in RA, but in Crohn disease there is a clear clinical benefit with infliximab but not etanercept (Targan et al, 1997; Sandborn et al, 2001) and the latter treatment exacerbates the pathology in multiple sclerosis.(Ghosh, 2012) A systematic characterization of Treg compartment in these patients has not been reported. Based on our data, we propose that the discrepancies in clinical outcomes with the TNF blockers may be partly due to differences in effectiveness of the drugs in improving the Treg compartment in different disease settings. For example, we found that anti-TNF treatment in cocultures from healthy donors reduced IKZF2+ Treg expansion, but increased IKZF2-Treg proliferation, such that the net effect of anti-TNF on overall Treg proliferation was minimal.( Zhong & Yazdanbakhsh, 2013) In contrast, anti-TNF antibodies doubled the total ITP Treg expansion. Moreover, anti-TNFRSF1B antibodies, which block the binding of TNF and LTA to their receptor, only targeted IKZF2− Treg subset proliferation in healthy donors with little effect on IKZF2+ Tregs, but increased the expansion of both IKZF2+ and IKZF2− Treg subsets in ITP patients. Interestingly, etanercept treatment was efficacious in four treated ITP patients (McMinn et al, 2003; Litton, 2008). Our data thus raise the possibility that responsiveness to TNF blockers in these ITP patients might have been due to an effective increase in Treg subsets.

In summary, we have shown that in vitro TNF blockade can induce much stronger Treg expansion ex vivo in ITP patients as compared to healthy controls, partly due to the potentially higher sensitivity of ITP Tregs to TNF mediated inhibition. These findings raise the possibility that TNF blockers, through their ability to increase Treg subset proliferation, may be efficacious in ITP patients.

Acknowledgments

This work was supported in part by NIH National Heart, Lung, and Blood Institute grant R01HL096497 (K. Yazdanbakhsh).

Footnotes

Authorship and disclosures

H.Z. designed and performed research, analysed data, drafted the manuscript, J.B. selected patients and coordinated consent and K.Y. designed, directed and wrote the paper.

Competing interests

The authors have no competing interests.

References

  1. Audia S, Samson M, Guy J, Janikashvili N, Fraszczak J, Trad M, Ciudad M, Leguy V, Berthier S, Petrella T, Aho-Glele S, Martin L, Maynadie M, Lorcerie B, Rat P, Cheynel N, Katsanis E, Larmonier N, Bonnotte B. Immunologic effects of rituximab on the human spleen in immune thrombocytopenia. Blood. 2011;118:4394–4400. doi: 10.1182/blood-2011-03-344051. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bao W, Bussel JB, Heck S, He W, Karpoff M, Boulad N, Yazdanbakhsh K. Improved regulatory T-cell activity in patients with chronic immune thrombocytopenia treated with thrombopoietic agents. Blood. 2010;116:4639–4645. doi: 10.1182/blood-2010-04-281717. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Biton J, Boissier MC, Bessis N. TNFalpha: activator or inhibitor of regulatory T cells? Joint Bone Spine. 2012;79:119–123. doi: 10.1016/j.jbspin.2011.09.017. [DOI] [PubMed] [Google Scholar]
  4. Carpentier I, Coornaert B, Beyaert R. Function and regulation of tumor necrosis factor receptor type 2. Current Medicinal Chemistry. 2004;11:2205–2212. doi: 10.2174/0929867043364694. [DOI] [PubMed] [Google Scholar]
  5. Culic S, Salamunic I, Konjevoda P, Dajak S, Pavelic J. Immune thrombocytopenia: serum cytokine levels in children and adults. Medical Science Monitor. 2013;19:797–801. doi: 10.12659/MSM.884017. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. El Sissy MH, El Sissy AH, Elanwary S. Tumor necrosis factor-alpha -308G/A gene polymorphism in Egyptian children with immune thrombocytopenic purpura. Blood Coagulation & Fibrinolysis. 2014;5:458–463. doi: 10.1097/MBC.0000000000000089. [DOI] [PubMed] [Google Scholar]
  7. Faustman DL, Davis M. TNF receptor 2 and disease: autoimmunity and regenerative medicine. Frontiers In Immunology. 2013;4:478. doi: 10.3389/fimmu.2013.00478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Gavin MA, Torgerson TR, Houston E, DeRoos P, Ho WY, Stray-Pedersen A, Ocheltree EL, Greenberg PD, Ochs HD, Rudensky AY. Single-cell analysis of normal and FOXP3-mutant human T cells: FOXP3 expression without regulatory T cell development. Proceedings of the National Academy of Sciences of the United States of America. 2006;103:6659–6664. doi: 10.1073/pnas.0509484103. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Gernsheimer T. Chronic idiopathic thrombocytopenic purpura: mechanisms of pathogenesis. The Oncologist. 2009;14:12–21. doi: 10.1634/theoncologist.2008-0132. [DOI] [PubMed] [Google Scholar]
  10. Ghosh S. Biologic therapies: lessons from multiple sclerosis. Digestive Diseases. 2012;30:383–386. doi: 10.1159/000338131. [DOI] [PubMed] [Google Scholar]
  11. Grinberg-Bleyer Y, Saadoun D, Baeyens A, Billiard F, Goldstein JD, Gregoire S, Martin GH, Elhage R, Derian N, Carpentier W, Marodon G, Klatzmann D, Piaggio E, Salomon BL. Pathogenic T cells have a paradoxical protective effect in murine autoimmune diabetes by boosting Tregs. The Journal of Clinical Investigation. 2010;120:4558–4568. doi: 10.1172/JCI42945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Jernas M, Hou Y, Stromberg CF, Shao L, Nookaew I, Wang Q, Ju X, Mellgren K, Wadenvik H, Hou M, Olsson B. Differences in gene expression and cytokine levels between newly diagnosed and chronic pediatric ITP. Blood. 2013;122:1789–1792. doi: 10.1182/blood-2013-05-502807. [DOI] [PubMed] [Google Scholar]
  13. Kuwana M, Kaburaki J, Ikeda Y. Autoreactive T cells to platelet GPIIb-IIIa in immune thrombocytopenic purpura. Role in production of anti-platelet autoantibody. The Journal of Clinical Investigation. 1998;102:1393–1402. doi: 10.1172/JCI4238. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Li S, Wang L, Zhao C, Li L, Peng J, Hou M. CD8+ T cells suppress autologous megakaryocyte apoptosis in idiopathic thrombocytopenic purpura. British Journal of Haematology. 2007;139:605–611. doi: 10.1111/j.1365-2141.2007.06737.x. [DOI] [PubMed] [Google Scholar]
  15. Li X, Zhong H, Bao W, Boulad N, Evangelista J, Haider MA, Bussel J, Yazdanbakhsh K. Defective regulatory B-cell compartment in patients with immune thrombocytopenia. Blood. 2012;120:3318–3325. doi: 10.1182/blood-2012-05-432575. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ling Y, Cao X, Yu Z, Ruan C. Circulating dendritic cells subsets and CD4+Foxp3+ regulatory T cells in adult patients with chronic ITP before and after treatment with high-dose dexamethasome. European Journal of Haematology. 2007;79:310–316. doi: 10.1111/j.1600-0609.2007.00917.x. [DOI] [PubMed] [Google Scholar]
  17. Litton G. Refractory idiopathic thrombocytopenic purpura treated with the soluble tumor necrosis factor receptor etanercept. American Journal of Hematology. 2008;83:344. doi: 10.1002/ajh.21135. [DOI] [PubMed] [Google Scholar]
  18. Liu B, Zhao H, Poon MC, Han Z, Gu D, Xu M, Jia H, Yang R, Han ZC. Abnormality of CD4 (+) CD25 (+) regulatory T cells in idiopathic thrombocytopenic purpura. European Journal of Haematology. 2007;78:139–143. doi: 10.1111/j.1600-0609.2006.00780.x. [DOI] [PubMed] [Google Scholar]
  19. Locksley RM, Killeen N, Lenardo MJ. The TNF and TNF receptor superfamilies: integrating mammalian biology. Cell. 2001;104:487–501. doi: 10.1016/s0092-8674(01)00237-9. [DOI] [PubMed] [Google Scholar]
  20. McGovern JL, Nguyen DX, Notley CA, Mauri C, Isenberg DA, Ehrenstein MR. Th17 cells are restrained by Treg cells via the inhibition of interleukin-6 in patients with rheumatoid arthritis responding to anti-tumor necrosis factor antibody therapy. Arthritis and Rheumatism. 2012;64:3129–3138. doi: 10.1002/art.34565. [DOI] [PubMed] [Google Scholar]
  21. McMinn JR, Jr, Cohen S, Moore J, Lilly S, Parkhurst J, Tarantino MD, Terrell DR, George JN. Complete recovery from refractory immune thrombocytopenic purpura in three patients treated with etanercept. American Journal of Hematology. 2003;73:135–140. doi: 10.1002/ajh.10331. [DOI] [PubMed] [Google Scholar]
  22. Mpofu S, Fatima F, Moots RJ. Anti-TNF-alpha therapies: they are all the same (aren’t they?) Rheumatology (Oxford) 2005;44:271–273. doi: 10.1093/rheumatology/keh483. [DOI] [PubMed] [Google Scholar]
  23. Nie H, Zheng Y, Li R, Guo TB, He D, Fang L, Liu X, Xiao L, Chen X, Wan B, Chin YE, Zhang JZ. Phosphorylation of FOXP3 controls regulatory T cell function and is inhibited by TNF-alpha in rheumatoid arthritis. Nature Medicine. 2013;19:322–328. doi: 10.1038/nm.3085. [DOI] [PubMed] [Google Scholar]
  24. Ogawara H, Handa H, Morita K, Hayakawa M, Kojima J, Amagai H, Tsumita Y, Kaneko Y, Tsukamoto N, Nojima Y, Murakami H. High Th1/Th2 ratio in patients with chronic idiopathic thrombocytopenic purpura. European Journal of Haematology. 2003;71:283–288. doi: 10.1034/j.1600-0609.2003.00138.x. [DOI] [PubMed] [Google Scholar]
  25. Okubo Y, Mera T, Wang L, Faustman DL. Homogeneous expansion of human T-regulatory cells via tumor necrosis factor receptor 2. Scientific Reports. 2013;3:3153. doi: 10.1038/srep03153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Olsson B, Andersson PO, Jernas M, Jacobsson S, Carlsson B, Carlsson LM, Wadenvik H. T-cell-mediated cytotoxicity toward platelets in chronic idiopathic thrombocytopenic purpura. Nature Medicine. 2003;9:1123–1124. doi: 10.1038/nm921. [DOI] [PubMed] [Google Scholar]
  27. Palladino MA, Bahjat FR, Theodorakis EA, Moldawer LL. Anti-TNF-alpha therapies: the next generation. Nature Reviews Drug discovery. 2003;2:736–746. doi: 10.1038/nrd1175. [DOI] [PubMed] [Google Scholar]
  28. Pehlivan M, Okan V, Sever T, Balci SO, Yilmaz M, Babacan T, Pehlivan S. Investigation of TNF-alpha, TGF-beta 1, IL-10, IL-6, IFNgamma, MBL, GPIA, and IL1A gene polymorphisms in patients with idiopathic thrombocytopenic purpura. Platelets. 2011;22:588–595. doi: 10.3109/09537104.2011.577255. [DOI] [PubMed] [Google Scholar]
  29. Rocha AM, De SC, Rocha GA, de Melo FF, Saraiva IS, Clementino NC, Marino MC, Queiroz DM. IL1RN VNTR and IL2-330 polymorphic genes are independently associated with chronic immune thrombocytopenia. British Journal of Haematology. 2010;150:679–684. doi: 10.1111/j.1365-2141.2010.08318.x. [DOI] [PubMed] [Google Scholar]
  30. Sakakura M, Wada H, Tawara I, Nobori T, Sugiyama T, Sagawa N, Shiku H. Reduced Cd4+Cd25+ T cells in patients with idiopathic thrombocytopenic purpura. Thrombosis Research. 2007;120:187–193. doi: 10.1016/j.thromres.2006.09.008. [DOI] [PubMed] [Google Scholar]
  31. Sandborn WJ, Hanauer SB, Katz S, Safdi M, Wolf DG, Baerg RD, Tremaine WJ, Johnson T, Diehl NN, Zinsmeister AR. Etanercept for active Crohn’s disease: a randomized, double-blind, placebo-controlled trial. Gastroenterology. 2001;121:1088–1094. doi: 10.1053/gast.2001.28674. [DOI] [PubMed] [Google Scholar]
  32. Sarpatwari A, Bussel JB, Ahmed M, Erqou S, Semple JW, Newland AC, Bennett D, Pharoah P, Provan D. Single nucleotide polymorphism (SNP) analysis demonstrates a significant association of tumour necrosis factor- alpha (TNFA) with primary immune thrombocytopenia among Caucasian adults. Hematology. 2011;16:243–248. doi: 10.1179/102453311X13025568941808. [DOI] [PubMed] [Google Scholar]
  33. Semple JW, Freedman J. Increased antiplatelet T helper lymphocyte reactivity in patients with autoimmune thrombocytopenia. Blood. 1991;78:2619–2625. [PubMed] [Google Scholar]
  34. Shevach EM, Thornton AM. tTregs, pTregs, and iTregs: similarities and differences. Immunological Reviews. 2014;259:88–102. doi: 10.1111/imr.12160. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Stasi R, Cooper N, Del Poeta G, Stipa E, Laura EM, Abruzzese E, Amadori S. Analysis of regulatory T-cell changes in patients with idiopathic thrombocytopenic purpura receiving B cell-depleting therapy with rituximab. Blood. 2008;112:1147–1150. doi: 10.1182/blood-2007-12-129262. [DOI] [PubMed] [Google Scholar]
  36. Talaat RM, Elmaghraby AM, Barakat SS, Ebeed ME. Alterations in immune cell subsets and their cytokine secretion profile in childhood idiopathic thrombocytopenic purpura (ITP) Clinical & Experimental Immunology. 2014;176:291–300. doi: 10.1111/cei.12279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Targan SR, Hanauer SB, van Deventer SJ, Mayer L, Present DH, Braakman T, DeWoody KL, Schaible TF, Rutgeerts PJ. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor alpha for Crohn’s disease. Crohn’s Disease cA2 Study Group. New England Journal of Medicine. 1997;337:1029–1035. doi: 10.1056/NEJM199710093371502. [DOI] [PubMed] [Google Scholar]
  38. Valencia X, Stephens G, Goldbach-Mansky R, Wilson M, Shevach EM, Lipsky PE. TNF downmodulates the function of human CD4+CD25hi T-regulatory cells. Blood. 2006;108:253–261. doi: 10.1182/blood-2005-11-4567. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Yazdanbakhsh K, Zhong H, Bao W. Immune dysregulation in immune thrombocytopenia. Seminars in Hematology. 2013;50:S63–S67. doi: 10.1053/j.seminhematol.2013.03.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Yu J, Heck S, Patel V, Levan J, Yu Y, Bussel JB, Yazdanbakhsh K. Defective circulating CD25 regulatory T cells in patients with chronic immune thrombocytopenic purpura. Blood. 2008;112:1325–1328. doi: 10.1182/blood-2008-01-135335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhang F, Chu X, Wang L, Zhu Y, Li L, Ma D, Peng J, Hou M. Cell-mediated lysis of autologous platelets in chronic idiopathic thrombocytopenic purpura. European Journal of Haematology. 2006;76:427–431. doi: 10.1111/j.1600-0609.2005.00622.x. [DOI] [PubMed] [Google Scholar]
  42. Zhong H, Yazdanbakhsh K. Differential control of Helios (+/−) Treg development by monocyte subsets through disparate inflammatory cytokines. Blood. 2013;121:2494–2502. doi: 10.1182/blood-2012-11-469122. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhong H, Bao W, Li X, Miller A, Seery C, Haq N, Bussel J, Yazdanbakhsh K. CD16+ monocytes control T-cell subset development in immune thrombocytopenia. Blood. 2012;120:3326–3335. doi: 10.1182/blood-2012-06-434605. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES