| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
* Academic Unit of Ophthalmology, Department of Clinical Science at South Bristol, University of Bristol;
Department of Gastroenterology, United Bristol Healthcare Trust;
Department of Cellular and Molecular Medicine, University of Bristol; and
Henry Wellcome Laboratories for Integrative Neuroscience and Endocrinology, Department of Clinical Science at South Bristol, University of Bristol, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Inhibition of lymphocyte proliferation by GCs was first demonstrated in 1961(4) and later reports established this was due to suppression of IL-2 production and mRNA expression (5, 6). Impaired in vitro responses to the synthetic GC dexamethasone (Dex) have been observed in PHA stimulated PBMCs from patients with steroid resistant (SR) asthma, SR UC, and SR rheumatoid arthritis, as well as SR renal transplant recipients (7, 8, 9, 10). There is now accumulating evidence that disparities in IL-2 responses to GCs are central to the SR phenotype. IL-2, in combination with IL-4, reduces both glucocorticoid receptor (GR) binding affinity and Dex inhibition of proliferation in PBMCs from normal volunteers (NVs) to levels seen in PBMCs from SR patients (11). It also sustains reduced GR affinity in PBMCs from SR patients, which is reversed on transfer to cytokine-free cell culture medium (11). Furthermore, IL-2 mRNA expression is significantly greater in bronchoalveolar lavage samples from patients with SR asthma than steroid sensitive (SS) controls (12).
IL-2’s influence on steroid sensitivity is not limited to modulation of GR binding, as it also inhibits STAT-5-dependent GR translocation into the nucleus in the murine HT-2 cell line (13). In human PBMCs, it has been shown to induce expression of the transcription factor AP-1 (which suppresses GR binding to DNA) as well as the nonbinding GRβ isoform (which suppresses GR
-mediated GC action) (14, 15). Both AP-1 and GRβ are reportedly increased in patients with SR asthma, (15, 16) and GRβ expression is also increased in PBMCs from SR patients with UC (17). Other potential mediators of clinically significant IL-2/GR interactions include FoxO3 (a forkhead transcription factor responsible for IL-2 inhibition of the GC-inducible gene, gilz), histone deacetylase 2 (via the effect of deacetylated GR on NF-
B-dependent gene expression), and GR gene polymorphisms (which may alter GR-induced regulation of IB). (18, 19, 20) Although, disparities in these between SS and SR patients have not yet been demonstrated.
The experimental evidence for the role of IL-2 in SR disease has been further supported by the results of a recent phase II trial of a mAb to the -subunit of the IL-2 receptor (CD25), which demonstrated clinical rescue in SR UC (21). We therefore hypothesized that the SR phenotype is not a feature of all PBMCs in SR patients, but rather is limited to CD4+ T cells that express CD25. A proposal that is also consistent with recent murine evidence that CD4+CD25+ T cells are preferentially spared after systemic GC exposure (22).
In humans, the T regulatory (Treg) cells responsible for inhibiting immune responses and maintaining self-tolerance predominantly reside in the 2% of CD4+ cells with greatest CD25 expression (CD25high) (23). However, it is unlikely that these CD25high Treg cells mediate the benefits of anti-CD25 mAb therapy observed in SR UC as their regulatory function is unaffected by anti-CD25 mAbs in vitro (24). We therefore propose that the influence exerted by CD25 on steroid sensitivity resides in the remaining CD4+ cells with lower levels of CD25 expression (CD25int). We postulate that in SR patients, CD4+CD25int cells are less responsive to the inhibitory effects of GCs, enabling them to continue proliferating and perpetuating disease.
The current study uses flow cytometric techniques to identify a subpopulation of CD4+ T cells that are resistant to the anti-proliferative effects of GCs and are prevalent in patients with clinically defined SR UC. We further demonstrate that CD4+CD25int cells are resistant to GC inhibition in contrast to CD4+CD25– cells, which are exquisitely sensitive, and show that anti-CD25 mAbs abrogate sustained CD4+ T cell proliferation in the presence of GCs.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
Twenty-three NVs and six patients with a previous history of SR UC participated. A previous episode of SR UC was defined as either (i) ongoing disease activity despite corticosteroid treatment requiring colectomy, (ii) active colitis (UC Symptom Score 
6) (25) despite at least 2 wk of prednisone 30 mg per day, or (iii) Truelove and Witts measures of disease severity (26) with poor indicators of outcome (27) despite at least 3 days of i.v. steroids. All SR patients had been off systemic GC treatment for at least 3 mo before enrollment. Three had previously participated in a randomized control trial of anti-CD25 mAb therapy in SR UC (a minimum of 1 year before participation in this study), in which they received a single i.v. infusion of 40 mg Basiliximab (Bas) (21).
This study was reviewed and approved by the Central and South Bristol Local Research Ethics Committee (Reference no. 04/Q2006/163).
PBMC preparation and culture
PBMCs were isolated from whole blood on a density gradient (Ficoll-Paque PLUS, GE Healthcare) and stained with CFSE (Vybrant CFDA SE cell tracer kit (V-12883); Molecular Probes) before culture in RPMI 1640 with 10% heat-inactivated FBS (Invitrogen Life Technologies), 100 IU/ml penicillin G, 100 mg/ml streptomycin, and 20 mM HEPES buffer (Invitrogen Life Technologies). Cells were seeded in 96-well U-bottom plates (1 x 105 per well) and incubated at 37°C in humidified air with 5% CO2 in the presence of 10–10M-10–6M Dex (Sigma-Aldrich), 1 µg/ml Bas (a chimeric anti-CD25 mAb; Novartis Pharmaceuticals), or 100 ng/ml Tacrolimus (a calcineurin inhibitor; Astellas Pharma). CD3CD28 coated Dynabeads (Dynal Biotech) were then added (2 µl per well) to stimulate T cell proliferation.
Preculture CD3CD25 phenotyping
A separate sample of fresh PBMCs was stained with PE-conjugated anti-CD25 (Miltenyi Biotec, 1/10 dilution) and PE-alexa fluor 610-conjugated anti-CD3 (Caltag Laboratories, 1/40 dilution). Preculture CD3+CD25 expression was then determined by flow cytometry (FACSCalibur; BD Biosciences) and analyzed using CellQuest software.
CD3+CD4+ cell proliferation studies
PBMCs were washed after 5 days culture and stained with PE-conjugated anti-CD4 (BD Pharmingen, 1/100 dilution) and allophycocyanin-conjugated anti-CD3 (BD Pharmingen, 1/100 dilution). CFSE dilution in CD3+CD4+ cells was then quantified by flow cytometry (28).
Quantification of apoptosis and cell death
The forward and side-scatter characteristics of apoptosing and dead cells, after 5 days culture in the presence of 10–6M Dex, were determined in four NVs using annexin V-FITC (BD Biosciences) and 7-aminoactinomycin (7-AAD; BD Biosciences). These forward and side-scatter profiles were then used as a proxy measure of the proportion of viable and nonviable cells in subsequent cultures.
Comparison of CD4+CD25– and CD4+CD25int cell responses to Dex
PBMCs were labeled with PE-conjugated anti-CD25 and PE-Alexa Fluor 610-conjugated anti-CD3 and sorted on a FACSVantage to isolate CD3+CD25– and CD3+CD25int T cells, achieving >80% purity (the 2.5% of cells with greatest CD25 expression were designated CD25high, and the remaining CD25+ cells were defined as CD25int). Each population was then separately stained with CFSE and cultured in the same conditions described above for PBMCs. After 5 days, cultured cells were washed and labeled with PE-conjugated anti-CD4. CFSE dilution in CD4+ cells was then determined by flow cytometry.
Determination of the effect of inducing "intermediary" expression of CD25 on CD4+ cell responses to Dex
CD4+CD25– cells were isolated using magnetic beads (achieving >90% purity; CD4+ T cells were first positively selected and then depleted of CD25+ cells (Miltenyi Biotec, CD4+ T cell isolation kit II and CD25 MicroBeads II kit; both used in accordance with the manufacturer’s instructions) and then either stained with CFSE and cultured in the same conditions described above for PBMCs, or seeded into a 96-well U-bottom plate (1 x 105 cells per well) and activated with 0.25 µl per well CD3CD28 coated Dynabeads to induce "intermediate" level CD25 expression in >75% of cells. After 24 h, the activated cells were washed free of Dynabeads, stained with CFSE, and cultured for 5 days in the same conditions. CFSE dilution was then determined by flow cytometry and the proliferation of the unactivated vs activated cells was compared.
Preculture CD4+CD25int cell activation status and Treg marker expression studies
PBMCs were labeled with selected combinations of the following Abs to determine the preculture activation status and Treg marker expression of CD4+CD25int cells using four-color flow cytometry (FACSCalibur; BD Biosciences): PE-conjugated anti-CD25 (Miltenyi Biotec; 1/10 dilution), PE-Alexa Fluor 610-conjugated anti-CD3 (Caltag Laboratories; 1/40 dilution), FITC-conjugated anti-CD69 (BD Pharmingen; 1/50 dilution), allophycocyanin-conjugated anti-HLA-DR (BD Pharmingen; 1/100 dilution), allophycocyanin-conjugated anti-CD4 (BD Pharmingen; 1/100 dilution), biotinylated anti-CD44 (BD Pharmingen; 1/100 dilution), FITC-conjugated anti-CD62L (BD Pharmingen, 1/200 dilution), FITC-conjugated anti-CD45RO (Caltag Laboratories; 1/100 dilution), biotinylated anti-CD127 (BD Pharmingen; 1/10 dilution), Alexa Fluor 750-R-PE streptavidin (Caltag Laboratories; 1/200 dilution), and FITC-conjugated anti-FoxP3 staining set (eBioscience, used in accordance with manufacturers instructions).
NVs whose CD4+ cells were SS (>70% nonproliferating cells) on exposure to 10–6M Dex (see Results below) and SR patients with a clearly defined subpopulation of dividing (CFSElow) CD4+ cells despite exposure to 10–6M Dex were selected for case-control comparisons.
Flow cytometric and statistical analyses
All flow-cytometric analyses (including proliferation studies) were conducted using FlowJo software (Tree Star). Cells exposed to 100 ng/ml Tacrolimus did not proliferate and a marker set on this population was then applied to all other samples to determine their proportion of nonproliferating cells. All statistical comparisons were calculated using Mann-Whitney tests in GraphPad Prism v. 4.00 for Windows (GraphPad Software) unless otherwise stated.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The mean age of SR patients and NVs used for case-control comparisons was 40 years (range, 21–58 years) and 38 years (range, 23–63 years), respectively. Sex distribution was equal between groups (2/3 male, 1/3 female), and all study participants were Caucasian. All the NVs and three of the SR patients, who had colectomies, were not taking any form of immunosuppressive therapy. The remaining SR patients were all in disease remission, maintained with 5-aminosalicylic acid (5-ASA) alone (Table I).
|
Fig. 1 illustrates the different CFSE profiles of Dex-treated CD3+CD4+ cells from NVs (Fig. 1A) and SR patients (Fig. 1B). All cultures had a proportion of weakly fluorescent (CFSElow) cells that had undergone multiple divisions, but the size of this proliferating subpopulation was greater in SR patients. Cell division was inversely quantified by calculating the "percentage of nonproliferating cells," which is the proportion of cells with the same CFSE fluorescence as 95% of cells in parallel cultures maximally suppressed with 100 ng/ml Tacrolimus (Fig. 1C). This revealed two categories of CFSE profile: those in which >70% of cells were undivided and those in which <70% of cells were undivided. Only one (16.7%) SR patient had greater than 70% inhibition of proliferation vs 13 (86.7%) NVs (median 47.3% undivided cells in SR patients and 76.9% in NVs, p = 0.03, Fig. 1D). Furthermore, all cultures with <70% undivided cells had a characteristic, prominent CFSElow peak. This was neither an artifact of variable preculture CFSE staining (as undivided control cells exposed to Tacrolimus had uniformly high fluorescence, Fig. 1C) nor variable T cell activation (as CFSE dilution in uninhibited cells was equivalent in all study subjects, Fig. 1E). In SR patients, the presence of a prominent CFSElow peak was also unrelated to 5-ASA use (p = 1.0, Fisher’s exact test). The number of dead and apoptosing cells (identified by their characteristic forward and side-scatter profile on flow cytometry, Fig. 2, A and B) after 5 days culture in the presence of 10–6M Dex was less in SR patients compared with NVs. However, this difference did not reach statistical significance (Fig. 2C).
|
|
A typical preculture CD25 expression profile is shown in Fig. 3A (see figure legend for definition of CD25–, CD25int, and CD25high cells). Fig. 3B shows the CFSE proliferation profiles of CD4+CD25– and CD4+CD25int cells that had been isolated by flow cytometry and exposed to increasing concentrations of Dex. Both CD4+CD25– and CD4+CD25int cells proliferated equally well in the presence of subtherapeutic (10 –10M) Dex, but CD4+CD25– cells were very sensitive to increasing GC concentrations with complete abrogation of CFSE dilution at 10–7M Dex. In contrast, CD4+CD25int cells continued to proliferate even in the presence of 10–6M Dex (Fig. 3B). The same result was replicated on samples from three different NVs (median 86.0% undivided CD4+CD25– cells vs 35.6% undivided CD4+CD25int cells, Fig. 2C). CFSE dilution in uninhibited CD4+CD25– and CD4+CD25int cells was equal (Fig. 3D), confirming that the difference in these two subpopulations is not secondary to a disparity in the degree of T cell activation. Although the median proportion of CD4+CD25int cells in fresh PBMCs from SR patients was greater than in PBMCs from NVs (34.6% vs 27.4%), this was not statistically significant (Fig. 3E).
|
To determine whether the level of CD25 expression in CD25int cells was functional, we explored the effects of CD25 inhibition in combination with 10–8M Dex. CFSE dilution in CD3+CD4+ cells from NVs was only partially suppressed on exposure to either 10–8M Dex or 1 µg/ml Bas (equivalent to therapeutic serum levels) alone (Fig. 4A). However, when Dex and Bas were combined the remaining peak of rapidly proliferating CFSElow cells was lost (Fig. 4A), significantly reducing the average number of cell divisions (p = 0.02, Fig. 4B).
|
Despite their CD25 expression, fresh, unstimulated CD4+CD25int cells did not express either the early activation marker CD69 or the late activation marker HLA-DR. Representative FACS profiles are shown in Fig. 5A, together with the CD69 and HLA-DR profiles of control PBMCs 48 h after activation with CD3CD28 coated Dynabeads (Fig. 5B). The proportion of CD62L– and CD45RO+CD44high cells was also equivalent in CD4+CD25int cells from NV and SR patients (Fig. 5C). To further determine whether the SR phenotype of CD4+CD25int was secondary to activation, we induced an "intermediate" level of CD25 expression by activating CD4+CD25– cells with CD3CD28 coated Dynabeads before culture. These cells were equally susceptible to Dex inhibition of proliferation as unactivated CD4+CD25– cells (Fig. 6). Furthermore, there was no significant difference in FoxP3 or CD127 expression in CD4+ cells from NVs and SR patients, either overall or in the CD25int subset (Fig. 7).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
The ability of Dex-exposed CD4+ cells from SR patients to proliferate when in disease remission implies that their SR-phenotype is stable over time and corroborates previous clinical evidence that SR is independent of disease severity (5). Indeed, SR may be unrelated to the presence of disease at all. Seventeen percent of our NVs had a SR phenotype, and this is consistent with previous reports of varying in vitro steroid sensitivity in healthy individuals, adding weight to the theory that SR is an inherent and stable characteristic of an individual which is only manifest in the context of GC therapy (5, 29).
The interpretation of our data is potentially confounded by 5-ASA treatment in the SR group. However, only three (50%) patients were taking 5-ASA, and this did not correlate with their proportion of CFSElow CD4+ cells. It is also counterintuitive to suggest that prior in vivo T cell exposure to 5-ASA would promote greater proliferation. In addition, three (50%) SR patients had received Bas as part of a previous clinical trial, but as Bas has been shown to increase steroid sensitivity in vitro (21) and the patients’ last Bas dose was at least 12 mo before participation in this study, it is unlikely that this has influenced our findings. Furthermore, there was no age, sex, or racial discrepancy between the SR or control study groups. Hence, it appears justifiable to attribute their different CFSE profiles to differences in their Dex response.
It is also important to emphasize that the apparently large populations of CFSElow cells in cultures from SR patients may have arisen from a tiny subgroup of preculture CD4+ cells. By basing our statistical comparisons on the proportion of nonproliferating cells, we are highlighting differences in the relative size of proliferating and nonproliferating populations. Consequently, the number of dividing cells may appear greater as a simple consequence of increased apoptosis and cell death in the undivided group. The trend toward lower numbers of dead cells in PBMC cultures from our SR patients (Fig. 2C) suggests that this is not the case. However, more accurate quantification of the dynamics of cell death and apoptosis over the full culture period is needed to properly validate this assertion, as well as to establish whether SR CD4+ cells from SR patients are also resistant to GC-induced apoptosis.
The continued proliferation of CD4+CD25int cells in the presence of 10–6M Dex, combined with the exquisite sensitivity of CD4+CD25– cells to 10–7M Dex (Fig. 3B), suggests that the SR CD4+ cells seen in our PBMC cultures arose from the CD4+CD25int population. This is further corroborated by the complete suppression of proliferation in cells escaping low-dose (10–8M) Dex inhibition in the presence of CD25 blockade. However, the similar number of CD25int cells in NVs and SR subjects implies that CD25int expression alone does not account for the SR phenotype (Fig. 3E), and we therefore speculate that the CD4+ SR cells we have demonstrated represent only a fraction of the CD25int group.
One potential explanation for continued cell division in CD4+CD25int cells despite exposure to GCs is that they are simply activated and already committed to proliferate. However, this is not substantiated by their CD69 and HLA-DR expression (Fig. 5A) or the fact that CD25int cells account for 29% (mean) of circulating CD3+ cells. Furthermore, the proportion of CD62L– and CD45RO+CD44high cells was equivalent in CD4+CD25int cells from NV and SR patients (Fig. 5C). The failure to induce SR in T cells activated to express an intermediary level of CD25 before culture (Fig. 6) also mitigates against activation, per se, being the cause of the CD25int cell SR we observed.
The association we have made between T cell CD25 expression and an SR phenotype is consistent with previous reports of IL-2-enhanced GR binding affinity and increased IL-2 mRNA expression in SR patients (11, 12). This suggests that SR is secondary to an up-regulation of IL-2, which in turn may reflect altered earlier signaling. For example, induction of the IKBA gene by GCs has been shown to lead to an increased rate of IB protein synthesis, hence blocking translocation of NF-B to the nucleus and consequent IL-2 transcription (33). Such upstream events are therefore candidates for altered GC action in SR T cells, which merit future investigation, as does the potential for their restriction to the CD4+CD25int cell subpopulation. It is also possible that the SR phenotype is influenced by Treg cell function, but we did not find a significant difference in CD4+ cell FoxP3 or CD127 expression in our NVs and SR patients (Fig. 7). However, the association between SR disease and a failure of induction of Treg cells has recently been identified in a cohort of SR asthma patients (34).
In summary, we have identified a subpopulation of SR CD4+ cells which are more prevalent in patients with a previous history of SR disease and have shown that CD4+CD25int cells replicate this SR phenotype. We therefore propose a model of SR disease based on the selection of such SR CD4+ T cells following GC treatment. In addition, we have provided further in vitro evidence to support the rationale for using anti-CD25 therapies in conjunction with GCs to treat SR disease.
|
Acknowledgments |
|---|
|
Disclosures |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
|
Footnotes |
|---|
1 This work was supported by Grants RJ4740 and RJ4750 (to R.W.J.L. and A.D.D.) from the National Eye Research Centre. R.W.J.L.’s research fellowship has been jointly funded by the National Eye Research Centre and Astellas Pharma. 
2 Address correspondence and reprint requests to Prof. Andrew Dick, Bristol Eye Hospital, Lower Maudlin Street, Bristol, BS1 2LX, U.K. E-mail address: a.dick{at}bristol.ac.uk
3 Abbreviations used in this paper: GC, glucocorticoid; SR, steroid resistant; GR, glucocorticoid receptor; SS, steroid sensitive; Dex, dexamethasone; UC, ulcerative colitis; Treg, T regulatory cell; NV, normal volunteer; Bas, basiliximab; 7-AAD, 7-aminoactinomycin; 5-ASA, 5-aminosalicylic acid; int, intermediate.
Received for publication March 28, 2007. Accepted for publication September 30, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionDisclosuresReferences |
|---|
-interferon messenger RNA. J. Immunol. 133: 273-276. [Abstract] gene expression in steroid-resistant asthma. J. Exp. Med. 181: 33-40. B suppression. J. Exp. Med. 203: 7-13. production. J. Immunol. 140: 1401-1407. [Abstract]B activity through induction of IB synthesis. Science 270: 286-290. This article has been cited by other articles:
|
|
 |
|
  R. W. J. Lee, L. P. Schewitz, L. B. Nicholson, C. M. Dayan, and A. D. Dick Steroid Refractory CD4+ T Cells in Patients with Sight-Threatening Uveitis Invest. Ophthalmol. Vis. Sci., September 1, 2009; 50(9): 4273 - 4278. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Creed, R. W. Lee, P. V. Newcomb, A. J. di Mambro, M. Raju, and C. M. Dayan The Effects of Cytokines on Suppression of Lymphocyte Proliferation by Dexamethasone J. Immunol., July 1, 2009; 183(1): 164 - 171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Ysrraelit, M. I. Gaitan, A. S. Lopez, and J. Correale Impaired hypothalamic-pituitary-adrenal axis activity in patients with multiple sclerosis Neurology, December 9, 2008; 71(24): 1948 - 1954. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
