Conserved anti-proliferative effect and poor inhibitionof TNFα secretion by regulatory CD4+CD25+ Tcells inpatients with systemic lupus erythematosusJesús Gómez a,⁎, Catuxa Prado b, Patricia López b,Ana Suárez b, Carmen Gutiérrez a,b
a Immunology Department, Hospital Universitario Central de Asturias, C/ Julian Clavería s/n, 33006 Oviedo, Spainb Department of Functional Biology, Immunology Area, Faculty of Medicine, University of Oviedo,C/ Julian Clavería s/n, 33006 Oviedo, Spain
Received 29 October 2008; accepted with revision 23 May 2009Available online 21 June 2009
KEYWORDSSystemic lupuserythematosus;Natural regulatory T cells;Cell proliferation;TNFα secretion
Abstract A role of natural regulatory CD4+CD25+ T cells in peripheral tolerance has beenestablished, but a causative role of these cells has not been proved in human autoimmunediseases. Functional alterations have been reported in arthritis and contradictory results havebeen published in systemic lupus erythematosus (SLE). In this study, the CD4+CD25+ Tcell functionwas studied in 12 SLE patients (6 with only antimalarials and 6 also with corticoids treatment) inparallel to 12 healthy controls. CD4+CD25− T cells were cocultured with allogenic monocytederived dendritic cells (MDDC) and CD4+CD25+ Tcells were added. Proliferation was measured by
Introduction peripheral tolerance controlling the autoreactive cells that
Regulatory mechanisms are essential for the healthy func-tioning of the immune system. Central and peripheraltolerance usually avoids or controls potential autoimmuneresponses [1]. Regulatory cells seem to play a critical role in
ed.
have escaped from deletion. Regulatory Tcells (Treg) are thebest known and different types have been reported: inducedTreg: Tr1 and Th3, which acquire their functions in theperiphery and act through IL-10 and TGF-β secretion, andnaturally occurring regulatory T cells CD4+CD25high (nTreg)that acquire their regulatory functions in the thymus andperform suppression in a cell contact dependent manner[2–4]. Experiments in murine models have shown that theremoval of these cells leads to spontaneous inflammatoryautoimmune diseases [5]. In humans, the existence ofCD4+CD25high cells with regulatory functions has also been
demonstrated [6] and their involvement in the pathogenesisof systemic and organ specific autoimmune diseases has beensuggested [7–11].
Natural regulatory T cells constitutively express a widevariety of surface markers as CD25, CTLA4, GITR and CD62L[6]. However, intracellular expression of FoxP3 has been themost common marker used to identify this cellular subset inmice [12,13]. Nevertheless, in humans, a number of workshave demonstrated that activated T cells express transitorilythis transcription factor [14]. This, together with otherdifficulties, has called the attention of researchers toreconsider the suitability of FoxP3 as regulatory cell marker.Currently, there is no feasible specific marker for nTreg cellsapart from the functional studies of their inhibitive properties.
Systemic lupus erythematosus (SLE) is the most systemicof the human autoimmune diseases. A variable myriad ofsymptoms characterise it, from a relatively limited diseaseto a very florid and severe condition [15]. Many componentsof the immune system are known to be altered in lupuspatients. Antibodies, immune complexes, activation of TandB lymphocytes are among the effector mechanisms foundoccasioning the tissue damage [16]. Cytokine imbalance andclearance disturbances have also been confirmed [17,18]. Apossible role for CD4+CD25high cells has been proposed inlupus pathogenesis. However, there are contradictory resultsabout the possible role that an aberrant number and/or animpair function of natural T regulatory cells could play in SLEpathogenesis [19–24]. Valencia et al. [23] found a deficientnTreg function in patients with flares while no functionalabnormalities were detected in patients with quiescentdisease. Alvarado-Sanchez et al. [21] found similar cellnumbers in SLE patients and controls and could showimpairment function only in some patients (4/10 of patientswith active disease). In contrast, Miyara et al. [20] showedconserved nTreg cell function in active and inactive SLEpatients, but a diminished number of CD4+CD25+ nTreg cellsin active patients. In all above mentioned works, resultswere from pools of patients under different regimens oftreatment. In only one report [24], the role of nTreg cells inSLE was analysed in untreated, new-onset SLE patients,finding no alterations in either number or function of thesecells disregarding the condition of active or inactive disease.
In a previous study we found high levels of CD4+CD25high
cells in SLE patients in association with corticoids treatment[22]. Other authors [25,26] have also reported high numbersof regulatory Tcells in asthmatic patients receiving steroids.Thus, corticoid treatment could also increase or modify theregulatory properties of these cells. In this work, we studiedthe regulatory function of CD4+CD25high cells in healthycontrols and SLE patients under two different regimens oftreatment: steroids or just antimalarial drugs. Suppressionwas evaluated in the anti-proliferative effect and in theinhibition of TNFα secretion.
Materials and methods
Subjects
The present study was approved by the regional ethicscommittee for clinical investigation in compliance with theHelsinki Declaration. A database was constructed for a
previous study [27,28] with the clinical and immunologicalfeatures of virtually all patients of Asturias (northern Spanishautonomy) that fulfilled the American College of Rheuma-tology (ACR) criteria for diagnosis of SLE. From thisperiodically updated database, twelve patients wereselected and enrolled in this study based on their willingnessand the treatment received in the six previous months: 6patients receiving only antimalarial drugs and 6 withcorticosteroid treatment (all but one receiving also anti-malarial drugs). All were females between 24 and 59 yearsold (46.9±10.3). Disease activity was scored based on theSLE disease activity index (SLEDAI) and all patients wereconsidered with inactive disease with SLEDAI values lowerthan 10. Signed informed consent was always obtained andkept on records. Twelve adult voluntary blood donors roughly(±5 years) matched for age (44.5±9.8, ns.) were taken ascontrols in parallel experiments. Patient's clinical data areshown in Table 1.
Cell culture reagents
RMPI medium 1640 with 25 mM Hepes and L-glutamine(BioWhitaker, Verviers, Belgium) supplemented with 10%heat-inactivated foetal bovine serum (HyClone, Logan, Utah,USA), 20 U/mL of penicillin and 20 μg/mL of streptomycinwas used in proliferation assays.
Monocyte derived dendritic cells (MDDC)
Peripheral blood mononuclear cells (PBMNC) were separatedfrom a healthy voluntary blood donor buffy coat bylymphoprep™ density gradient (Axis-Shield PoC AS, Oslo,Norway). Monocytes were obtained from these PBMNC bypositive selection using anti-CD14 microbeads and magneticcolumns (Miltenyi Biotec, Bergisch Gladbach, Germany) andderived to MDDC in 24-well flat bottom plates for 7 days with70 ng/mL of GM-CSF and 35 ng/mL of IL-4 (R&D SystemsEurope, Oxon, UK). MDDC were matured for two more days inthe presence of 50 ng/mL of TNFα (R&D Systems Europe,Oxon, UK). Mature MDDC were then stored in frozen aliquotsuntil required.
CD4+CD25+ and CD4+CD25− T cell isolation
CD4+CD25+ and CD4+CD25− Tcells were enriched from PBMNCobtained from peripheral blood or buffy coats by lympho-prep™ density gradient using a CD4+CD25+ regulatory T cellisolation kit (Miltenyi Biotec, Bergisch Gladbach, Germany).Briefly, CD4+ Tcells are obtained from a firstmagnetic columnby negative selection with a monoclonal antibody cocktail(CD8, CD14, CD16, CD19, CD36, CD56, CD123, TCRγ/δ andCD235a), incubated with anti-CD25 microbeads and thenpassed through a second column. The negative fraction of thesecond column is an enriched population of CD4+CD25− Tcellsand the positive fraction is an enriched population ofCD4+CD25+ T cells (mainly CD25high).
Proliferation assays
To assess proliferation, 50×103 CD4+CD25− cells (used asresponders) were cultured for 5 days in complete RPMI
Table 1 Lupus patients characteristics.
Pt Age(years)
Age at diagnosis(years)
Sex Treatment Clinical and immunological features SLEDAI
387Downregulation of TNFα secretion and cell proliferation by regulatory CD4+CD25+ T cells in SLE
medium in 96-well U-bottom plates with 3500 allogenicMDDC. For all these experiments, a concentration of 3500MDDC/well was used and always these MDDC were derivedfrom the same donor. In previous dose/effect experiments,percentages of MDDC lower than 10% (out of the number ofeffector cells) were optimum for determining the anti-proliferative effect of CD4+CD25+ cells. In our hands, highernumbers of MDDCmasked the regulatory effect of these cells.For assessment of regulatory properties, CD4+CD25+ cellswere added to cultures at 1/2 and 1/4 ratio (CD4+CD25+/CD4+CD25−). Quadruplicates were performed for all cellsubset combinations. After 4 days of culture, 30 μL ofsupernatant from each well was removed for cytokinedetection. Medium was replaced and 1 μCi of [3H] Thymidine(37 kBq/well) was added to each well. Plates were culturedfor 16 additional hours, and [3H]-thymidine incorporation wasmeasured using a liquid scintillation counter.
Immunostaining and flow cytometric analysis
To measure frequencies of CD4+CD25+ T cells, 100 μL ofperipheral whole blood was incubated for 10 min withmonoclonal antibodies at room temperature, erythrocyteswere lysed for 10 min, the preparation was washed twicewith PBS and then analysed in the flow cytometer. Theanalysis was based on cells of the living region definedusing forward and side scatter. Cells were further gatedaccording the CD3 expression. A fixed region in a CD4 vs.CD25 dot plot was used to determine CD4+CD25high cells
and the frequency was expressed as percentage out ofCD4+ cells.
To assess purity of cell separations, cells were labelledsimilarly and analysed by flow cytometry. Cells weresuspended in PBS containing 0.5% bovine serum albumin(BSA), incubated with the monoclonal antibodies for 10 minat room temperature and washed twice. All samples weregated using forward and side scatter to exclude dead cells.Further analysis was carried out on the living and CD3+ cellgates.
Murine monoclonal antibodies of the following specifi-cities were used: CD4 fluorescein isothiocyanate (FITC), CD3peridinin chlorophyll protein (PerCP) (BD Pharmingen, SanJose, CA, USA), CD25 phycoerythrin (PE) (Miltenyi Biotec,Bergisch Gladbach, Germany). Corresponding isotype con-trols (all obtained from BD Pharmingen) were used. Flowcytometric data was acquired and analysed on a FACScancytometer using the CellQuest Pro software (BD, FranklinLakes, New Jersey, USA).
Cytokine determination
Supernatants were obtained from each cell culture for everyexperimental condition and then stored frozen until use. Allaliquots were thawed and processed in one run by an “inhouse” sandwich enzyme immunoassay (ELISA) for TNFαquantification. Briefly, plates were coated overnight withanti-TNFα (R&D Systems Europe, Oxon, UK) at a concentra-tion of 8 pg/mL, 50 μL of supernatant was incubated for 3 h in
388 J. Gómez et al.
the anti-TNFα coated plates. Plates were washed extensivelyand then incubated with a second biotinylated TNFαantibody (R&D Systems Europe, Oxon, UK) for 2 h. Afterwashing, streptavidine–alkaline phosphatase (Caltag Lab,Burlingame, USA) and p-nitrophenol phosphate (Sigma,Berlin, Germany) as substrate were used as detectionsystem. Detection limit of this assay was 7.5 pg/mL.
Statistical analysis
The mean±SD thymidine uptake was calculated for eachexperimental condition. Percentage of proliferation wasdetermined as: (cpm incorporated in the coculture tocalculate−cpm incorporated in CD4+CD25− cells withoutMDDC stimulation)/(cpm incorporated in CD4+CD25− cellswith MDDC stimulation−cpm incorporated in CD4+CD25−
cells without MDDC stimulation)×100. After the demonstra-tion of normality of the series of values by Kolmogorov–Smirnov test in both controls and patients, a Student T testfor independent samples was used to evaluate possibledifferences in the CD4+CD25+ cell function (on proliferationand TNFα secretion) between patient groups and controls.Conversely, as SLEDAI did not show normal distribution, thenon parametric Spearman correlation coefficient was used toevaluate possible correlations between SLEDAI andCD4+CD25+ cell number in peripheral blood, cell proliferationand TNFα secretion. All statistical tests were preformedusing the SSPS 12.0 statistical software package (SPSS Inc.)
Figure 1 Inhibitive effect of CD4+CD25+ cells on proliferation of CMDDC. Graphs represent the cell proliferation observed with the addand (B) line graphs with individual values expressed as cpm. (C) mexpressed as percentage of proliferation. The significant differenc(+25,000 cells) is signalled in F (⁎).
with the exception of statistical power that was calculatedusing the Power and Sample Size Calculation program (PS,version 2.1.31).
Results
Effect of CD4+CD25+ T cells on cell proliferation
To assess the anti-proliferative effect of CD4+CD25+ T cells,blood samples for patients and controls were processedsimultaneously in each experiment in the same experimentalconditions. CD4+CD25− and CD4+CD25+ Tcell separation frompatients and controls achieved purities higher than 98% forCD4+ T cells and higher than 85% for CD25+cells. Cocultureswith allogenic MDDC were preformed and proliferation incrude counters per minute was converted to percentages. Acurve was drawn per each patient and control with cellproliferation values at the three concentrations of addedCD4+CD25+ T cells: none, +12,500 (1/4 related to respon-ders), +25,000 (1/2 related to responders). Always, inpatients and controls, cell proliferation values were loweras more CD4+CD25+ Tcells were added to culture. In addition,CD4+CD25+ cells did not proliferate upon allogenic stimula-tion, since cultured with MDDC in the absence of respondercells, cpm values were similar to negative controls (data notshown). Then, anti-proliferative effect of regulatoryCD4+CD25+ Tcells was analysed in the entire group of patientsand compared with healthy controls. Lower proliferation
D4+CD25− effector cells in response to allogenic stimulation withition of CD4+CD25+ cells in SLE patients and healthy controls. (A)eans and standard deviations of the series. (D–F) same data
e (p=0.009) between patient and control means found for 1/2
389Downregulation of TNFα secretion and cell proliferation by regulatory CD4+CD25+ T cells in SLE
percentages were found in patients than in controls at 1/2ratio (+25,000 CD4+CD25+ T cells) (p=0.009) and no sig-nificant differences were found at 1/4 (+12,500 CD4+CD25+ Tcells) (Fig. 1).
Effect of CD4+CD25+ T cells on TNFα secretion
Supernatants of each cell culture combination were keptfrozen until use in an ELISA that was performed at once todetermine the effect of CD4+CD25+ Tcells on TNFα secretion.Similar curves to cell proliferation evaluation were drawnexpressing TNFα concentration in picograms and in percen-tages using the zero added CD4+CD25+ Tcells as 100% to makerelative values as in cell proliferation assessments. Lowerpercentages of inhibition of TNFα secretion were found inpatients than in controls at 1/2 ratio (+25,000 CD4+CD25+ Tcells) (p=0.02) and no significant differences were found at1/4 ratio (+12,500 CD4+CD25+ T cells). Much higher varia-bility was found in TNFα secretion in patients than in controls(SD=56.5 at 1/2 for patients and SD=16.4 for controls)(Fig. 2).
Influence of SLEDAI on cell number, proliferationand TNFα secretion
Although all our patients were selected on the condition ofnon-flare stage, we also tested possible influences of disease
Figure 2 Inhibitive effect of CD4+CD25+ cells on TNFα secretiostimulation with MDDC). Graphs represent the TNFα secretion obsehealthy controls. (A and B) line graphs with individual values expresse(D–F) same data expressed as percentage out of the secretion of CD(p=0.028) between patient and control means found for 1/2 (+25,0
activity on cell number, proliferation and TNFα secretion.The percentages of CD4+CD25+ Tcells were calculated out ofCD4+ T cells in peripheral whole blood and no significantdifferences were found between patients and correspondingcontrols and no correlation was found with the SLEDAI(Spearman's rho: −0.10, ns.). Also, we did not find correla-tion between cell proliferation (Spearman's rho: −0.19, ns.)or TNFα secretion (Spearman's rho: −0.25, ns.) and SLEDAIscores. Finally, when patients were classified according to acommonly used cut-off of 5 (non active: ≤5 and active: N5),no significant differences were found neither in cellproliferation nor TNFα secretion (data not shown).
Influence of treatment on CD4+CD25+ T cellsinhibition effects
Patients were originally selected on the base of thetreatment received in the previous six months. Six patientswere receiving antimalarials and the other six patients werealso treated with corticoids. On proliferation assays, nosignificant differences were found between cell proliferationpercentages in patients under different treatments at anyCD4+CD25+ T cell concentration. There were also nosignificant differences between any patient group andcontrols. In regard with TNFα, similarly to proliferation, nosignificant differences were found when both kinds oftreatment were compared (Fig. 3). Although no significant
n in MLR (CD4+CD25− effector cells in response to allogenicrved with the addition of CD4+CD25+ cells in SLE patients andd in picograms. (C) means and standard deviations of the series.4+CD25− cells stimulated with MDDC. The significant difference00 cells) is signalled in F (⁎).
Figure 3 Effect of CD4+CD25+ cells in MLR according to treatments received by patients during the previous six months. Line graphsrepresent the effect of CD4+CD25+ cells on cell proliferation (A–C) and on TNFα secretion (D–F). Shown values have been corrected bysubtracting basal levels (negative control). No significant differences were found between the two kinds of treatments.
390 J. Gómez et al.
differences were found, the CD4+CD25− Tcells from patientstreated with just antimalarials stimulated with MDDC (with-out CD4+CD25+ T cells) secreted less TNFα than controls(106.3 vs. 149.5 pg/mL); while in patients treated withcorticoids secreted similar quantities of TNFα as controls(149.5 vs. 140.2 pg/mL). Remarkably, a very high variability(SD=106.7) in TNFα secretion was noted in CD4+CD25− Tcellsstimulated with MDDC (without CD4+CD25+ Tcells) in patientswith corticoids while this variability was lower in healthycontrols (SD=85.3) and patients receiving antimalarials(SD=28.9). Another factor that limits this analysis is thatthe number of patients in each treatment group is not bigenough to detect relative differences between 10% and 20%with a statistical power of 80% or higher.
Discussion
In SLE, conflicting results have not allowed to achievedefinitive conclusions concerning nTreg cell function.Valencia et al. [23] and Alvarado-Sanchez et al. [21] havefound deficient anti-proliferative function in some of thepatients with flares, while Miyara et al. [20] and Zang et al.[24] have found conserved nTreg cell function in disregardwith the condition of active or inactive disease. Recently,Yates et al. [29] have also reported a nTreg cell function inthe majority of patients with lupus nephritis comparable tothat with healthy controls. We more agree with these laterauthors, as no single patient showed a notable impairednTreg anti-proliferative function and when mean values of
patients and controls were analysed no qualitativelyimpaired anti-proliferative function was found. Moreover,surprisingly it was significantly higher in patients than incontrols. Similarly to Miyara et al. [20] and differently toZang et al. [24], our patients were under treatment. All ourpatients were considered with mild or moderate diseaseactivity (SLEDAI≤10) and all except one (with only corti-coids) were receiving antimalarial drugs that are known todownregulate TNFα secretion. It is also known that anti-proliferative function of nTreg is damaged by TNFα. Thus,this enhanced anti-proliferative function found in our SLEpatients could be attributed to a low TNFα production invivo, consequence of treatment. An alternative to thishypothesis aroused when we analysed proliferation valuesexpressed in cpm. Responder CD4+CD25− T cells stimulatedwith MDDC in the absence of CD4+CD25+ Tcells showed lowercounts in patients than in controls. This could be due to themigration of the most active CD4+CD25− Tcells from blood totissue in SLE patients, while the less compromised orimmature cells stay in intravascular space or, alternatively,to impairment in CD4+CD25− T cells. In case of confirmationof the last hypothesis this impairment could be an inherentfeature of the disease or consequence of the treatment. Itwas notable that isolated CD4+CD25− T cell numbers wereparticularly low in lymphopenic patients and CD4+CD25+ Tcell numbers were unaffected (data not shown). All thesemean that starting from lower initial values of CD4+CD25−
cell proliferation could facilitate the anti-proliferativeaction of nTreg cells or, alternatively, CD4+CD25− cellsfrom patients under treatment could be more easily down
391Downregulation of TNFα secretion and cell proliferation by regulatory CD4+CD25+ T cells in SLE
regulated. Other aspects to take into account whendifferent studies are compared is that different kind andintensity of stimulus (allogenic, anti-CD3, PHA) are used tostimulate effector cells and it could also affect the results.Also different stimulus intensities can also affect the results.In our hands, MDDC relative ratios equal or higher than 10%out of effector cells diminish up to mask the inhibition effectof nTreg cells.
Other groups have also analysed the inhibition of cytokinesecretion by nTreg in SLE patients. Valencia et al. [23] havefound a conserved nTreg regulatory function on INFγsecretion in inactive SLE patients as well as Miyara et al.[20] have studied a panel of cytokines that included IL-2,IL-4, IL-5, IL-10, TNFα and INFγ and did not find impairedregulatory function on cytokine secretion. A significant lowerinhibition of TNFα secretion by nTreg cells was found in thepatients of our series but it is difficult to make comparisonsas, among other aspects, cytokines were measured bydifferent methods. Other authors have also found discor-dance between both functions of regulatory cells. Ehrensteinet al. [9] have reported that nTreg cells in rheumatoidarthritis patients have a conserved anti-proliferative func-tion while they could be defective in controlling proinflam-matory cytokine production. Nevertheless, variation oflevels of supernatant TNFα was very high in our patients; inparticular in those that were receiving treatment withcorticoids and have also a more variable and worse clinicalcondition. Although differences were not significant, inthese patients with corticoids the CD4+CD25− T cellsstimulated with MDDC and in the absence of CD4+CD25+ Tcells secreted higher levels of TNFα than in patients withonly antimalarials. As patients under treatment with onlyantimalarials showed less variation in TNFα secretion thanthose with corticoids and healthy controls, it is possible thatTNFα secretion of genetically low and high TNFα producers isdownregulated [30] and made patients with antimalarialsshow similar low levels of TNFα.
Absolute absence or damage of nTreg cells is associatedwith very severe conditions. Scurfy mice, mutants forFoxP3, die three weeks after birth. In humans, a very floridand severe disease, the IPEX (immune dysfunction poly-endocrinopathy enteropathy X-linked) syndrome, has beenobserved in patients with FOXP3 mutations [31,32]. Thus, itis not surprising that no major role had been demonstratedin common autoimmune diseases in spite of having beensignalled to this respect. A very important defect in nTregcells would lead to more dramatic consequences. Inrheumatoid arthritis, there is consensus about decreasednTreg function, but it seems to be a consequence of theexcess of TNFα secretion and the restoration of thefunction with anti-TNFα treatment is a convincing evidence[9,33].
In summary, we could not find any relevant damage innTreg cell anti-proliferative function. According to ourresults and other authors', no SLE causative role can beargued for CD4+CD25+ T cells, which are mainly functionallyundamaged. In our view, alterations in anti-proliferativefunction that can be found in certain patients could beinterpreted as secondary or circumstantial. Nevertheless,inhibition of TNFα secretion was lower in SLE patients than incontrols. This finding could have some implications in lupuspathogenic mechanisms.
Acknowledgments
This study was supported by the Asociación Lúpicos deAsturias (Association of Lupus Patients of Asturias, Spain) andby the Grants PI052409 from the Fondo de InvestigaciónSanitaria. Patricia López was supported by a Grant from theFundación Eugenio Rodríguez Pascual. We are especiallygrateful to the Asociación Lúpicos de Asturias for itscontinuous encouragement and support.
References
[1] L. Van Parijs, A.K. Abbas, Homeostasis and self-tolerance in theimmune system: turning lymphocytes off, Science 280 (1998)243–248.
[2] S. Sakaguchi, Regulatory Tcells: key controllers of immunologicself-tolerance, Cell 101 (2000) 455–458.
[3] H. Groux, A. O'Garra, M. Bigler, M. Rouleau, S. Antonenko, J.E.de Vries, M.G. Roncarolo, A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis, Nature 389 (1997)737–742.
[4] H.L. Weiner, Induction and mechanism of action of transform-ing growth factor-β-secreting Th3 regulatory cells, Immunol.Rev. 182 (2001) 207–214.
[5] S. Sakagushi, N. Sakagushi, M. Asano, M. Itoh, M. Toda,Immunologic self-tolerance maintained by activated T cellsexpressing IL-2 receptor α-chains (CD25): breakdown of a singlemechanism of self-tolerance causes various autoimmunediseases, J. Immunol. 155 (1995) 1151–1164.
[6] D. Dieckmann, H. Plottner, S. Berchtold, T. Berger, G. Schuler,Ex vivo isolation and characterization of CD4+CD25+ Tcells withregulatory properties from human blood, J. Exp. Med. 193(2001) 1303–1310.
[7] S. Lindley, C.M. Dayan, A. Bishop, B.O. Roep, M. Peakman, T.I.Tree, Defective suppressor function in CD4+CD25+ T-cells frompatients with type 1 diabetes, Diabetes 54 (2005) 92–99.
[8] V. Viglietta, C. Baecher-Allan, H.L. Weiner, D.A. Hafler, Lossof functional suppression by CD4+CD25+ regulatory T cellsin patients with multiple sclerosis, J. Exp. Med. 199 (2004)971–999.
[9] M.R. Ehrenstein, J.G. Evans, A. Singh, S. Moore, G. Warnes,D.A. Isenberg, C. Mauri, Compromised function of regulatoryT cells in rheumatoid arthritis and reversal by anti-TNFαtherapy, J. Exp. Med. 200 (2004) 277–285.
[10] M.A. Kriegel, T. Lohmann, C. Gabler, N. Blank, J.R. Kalden,H.M. Lorenz, Defective suppressor function of humanCD4+CD25+ regulatory T cells in autoimmune polyglandularsyndrome type II, J. Exp. Med. 199 (2004) 1285–1291.
[11] H. Sugiyama, R. Gyulai, E. Toichi, E. Garaczi, S. Shimada, S.R.Stevens, T.S. McCormick, K.D. Cooper, Dysfunctional blood andtarget tissue CD4+CD25high regulatory T cells in psoriasis:mechanism underlying unrestrained pathogenic effector T cellproliferation, J. Immunol. 174 (2005) 164–173.
[12] J.D. Fontenot, M.A. Davin, A.Y. Rudensky, Foxp3 programs thedevelopment and function of CD4+CD25+ T regulatory cells,Nat. Immunol. 4 (2003) 330–336.
[13] S. Hori, T. Nomura, S. Sakaguchi, Control of regulatory T celldevelopment by the transcription factor Foxp3, Science 299(2003) 1057–1061.
[14] J. Wang, A. Ioan-Facsinay, E.I. van der Voort, T.W. Huizinga,R.E. Toes, Transient expression of FOXP3 in human activatednonregulatory CD4+ T cells, Eur. J. Immunol. 37 (2007)129–138.
[15] J.A. Mills, Systemic lupus erythematosus, N. Engl. J. Med. 330(1994) 1871–1879.
[16] P.E. Lispky, Systemic lupus erythematosus: an autoimmune
392 J. Gómez et al.
disease of B cell hyperactivity, Nat. Immunol. 2 (2001)764–766.
[17] L. Rönnblom, G.V. Alm, A pivotal role for the natural interferonα-producing cells (plasmacytoid dendritic cells) in the patho-genesis of lupus, J. Exp. Med. 194 (2001) F59–63.
[18] M.F. Denny, P. Chandaroy, P.D. Killen, R. Caricchio, E.E. Lewis,B.C. Richardson, K.D. Lee, J. Gavalchin, M.J. Kaplan, Acceler-ated macrophage apoptosis induces autoantibody formationand organ damage in systemic lupus erythematosus,J. Immunol. 176 (2006) 2095–2104.
[19] J.C. Crispin, A. Martinez, J. Alcocer-Varela, Quantification ofregulatory Tcells in patientswith systemic lupus erythematosus,J. Autoimmun. 21 (2003) 273–276.
[20] M. Miyara, Z. Amoura, C. Parizot, C. Badoual, K. Dorgham, S.Trad, D. Nochy, P. Debre, J.C. Piette, G. Gorochov, Globalnatural regulatory T cell depletion in active systemic lupuserythematosus, J. Immunol. 175 (2005) 8392–8400.
[21] B. Alvarado-Sanchez, B. Hernandez Castro, D. Portales-Perez,L. Baranda, E. Layseca-Espinosa, C. Abud-Mendoza, A.C.Cubillas-Tejeda, R. Gonzalez-Amaro, Regulatory T cells inpatients with systemic lupus erythematosus, J. Autoimmun.27 (2006) 110–118.
[22] A. Suarez, P. Lopez, J. Gomez, C. Gutierrez, Enrichment ofCD4+CD25high T cell population in patients with systemic lupuserythematosus treated with glucocorticoids, Ann. Rheum. Dis.65 (2006) 1512–1517.
[23] X. Valencia, C. Yarboro, G. Illei, P.E. Lipsky, DeficientCD4+CD25high T regulatory cell function in patients with activesystemic lupus erythematosus, J. Immunol. 178 (2007)2579–2588.
[24] B. Zhang, X. Zhang, F. Tang, L. Zhu, Y. Liu, W. Chen, P. Lipsky,Clinical significance of increased CD4+CD25−Foxp3+ T cells inpatients with new-onset systemic lupus erythematosus, Ann.Rheum. Dis. 67 (2008) 1037–1040.
[25] C. Karagiannidis, M. Akdis, P. Holopainen, N.J. Woolley, G.Hense, B. Rückert, P.Y. Mantel, G. Menz, C.A. Akdis, K. Blaser,
C.B. Schmidt-Weber, Glucocorticoids upregulate FOXP3 expres-sion and regulatory T cells in asthma, J. Allergy Clin. Immunol.114 (2004) 1425–1433.
[26] D. Hartl, B. Koller, A.T. Mehlhorn, D. Reinhardt, T. Nicolai, D.J.Schendel, M. Griese, S. Krauss-Etschmann, Quantitative andfunctional impairment of pulmonary CD4+CD25hi regulatory Tcells in pediatric asthma, J. Allergy Clin. Immunol. 119 (2007)1258–1266.
[27] J. Gomez, A. Suarez, P. Lopez, L. Mozo, J.B. Diaz, C. Gutierrez,Systemic lupus erythematosus in Asturias, Spain: clinical andserologic features, Medicine (Baltimore) 85 (2006) 157–168.
[28] P. Lopez, L. Mozo, C. Gutierrez, A. Suarez, Epidemiology ofsytemic lupus erythematosus in a northern Spanish population:gender and age influence on immunological features, Lupus 12(2003) 862–865.
[29] J. Yates, A. Whittington, P. Mitchell, R.I. Lechler, L. Lightstone,G. Lombardi, Natural regulatory T cells: number and functionare normal in the majority of patients with lupus nephritis,Clin. Exp. Immunol. 153 (2008) 44–55.
[30] P. Lopez, J. Gomez, L. Mozo, C. Gutierrez, A. Suarez, Cytokinepolymorphisms influence treatment outcomes in SLE patientstreated with antimalarial drugs, Arthritis Res. Ther. 8 (2006)R42.
[31] R. Khattri, T. Cox, S.A. Yasayko, F. Ramsdell, An essential rolefor Scurfin in CD4+CD25+ T regulatory cells, Nat. Immunol. 4(2003) 337–342.
[32] R. Bacchetta, L. Passerini, E. Gambineri, M. Dai, S.E. Allan, L.Perroni, F. Dagna-Bricarelli, C. Sartirana, S. Matthes-Martin, A.Lawitschka, C. Azzari, S.F. Ziegler, M.K. Levings, M.G.Roncarolo, Defective regulatory and effector T cell functionsin patients with FOXP3 mutations, J. Clin. Invest. 116 (2006)1713–1722.
[33] X. Valencia, G. Stephens, R. Goldbach-Mansky, M. Wilson, E.M.Shevach, P.E. Lipsky, TNF downmodulates the function ofhuman CD4+CD25hi T-regulatory cells, Blood 108 (2006)253–261.
