Nephrol Dial Transplant (2003) 18: 133-140
© 2003 European Renal Association-European Dialysis and Transplant Association
Poor response to recombinant erythropoietin is associated with loss of T-lymphocyte CD28 expression and altered interleukin-10 production
1 Department of Immunology and 2 Department of Renal Medicine, Guy's, King's and St Thomas' School of Medicine, King's College Hospital, London, UK
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Background. Recombinant erythropoietin (Epo) therapy is well established as an effective treatment for the anaemia of end-stage renal disease. However, 5–10% of such patients do not respond adequately and an important contributory factor to this is chronic inflammation.
Methods. The present study compares the circulating T-cell phenotypes of haemodialysis patients who respond poorly to Epo with those who respond well, along with normal controls. Isolated peripheral blood mononuclear cells were labelled with immunofluorescent monoclonal antibodies to surface antigens and analysed by flow cytometry. In vitro mononuclear cell cytokine secretion was also studied in the three subject groups. The cells were cultured for 48 h either without stimulus, with lipopolysaccharide or with monoclonal antibodies to CD3 and CD28.
Results. C-reactive protein levels were increased in poor responders to Epo (18.6±20.7 mg/l) compared with good responders (8.7±8.0 mg/l) and normal controls (3.8±1.1 mg/l). Patients responding poorly to Epo had increased circulating levels of CD4+/CD28- and CD8+/CD28- T-cells compared with patients responding well to Epo and normal controls. Unstimulated mononuclear cells from poor responders showed increased in vitro generation of interleukin-10 (IL-10) compared with both patients responding well to Epo and normal controls. Additionally, IL-10 generation stimulated by monoclonal antibodies to CD3 and CD28 was increased in poor responders compared with normal controls.
Conclusions. These findings suggest that patients responding poorly to Epo may show enhanced immune activation as manifest by changes in both T-cell function and phenotype.
Keywords: anaemia; CD28; cytokines; erythropoietin; haemodialysis; T-cells
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Recombinant human erythropoietin (Epo) has transformed the management of anaemia in patients with end-stage renal disease (ESRD) over the last decade. The vast majority (90–95%) of patients respond with an increase in haemoglobin (Hb) concentration and this is associated with considerable benefits, including enhanced quality of life, increased exercise capacity and improved cardiac function. However, the dose requirements of Epo to achieve correction of anaemia vary enormously, with some patients requiring as much as 10 times the dose of other patients [1]. Many factors can affect the response to Epo [1], but even accounting for these, the marked variability in sensitivity to the drug is still unexplained. Furthermore, around 5–10% of patients show a marked resistance to Epo therapy and again the reasons for this are unclear. It is, however, recognized that one of the most potent causes of a suboptimal response to Epo is chronic inflammation [2] and this may be mediated via pro-inflammatory cytokines.
Cytokines are critical effectors in immune-cell function and it is therefore likely that the uraemic state leads to complex alterations in the cytokine network. The haemodialysis (HD) procedure itself may cause further changes in cytokine generation [3]. Haemodialysis patients have increased circulating levels of cytokines compared with healthy controls [4] and monocytes from HD patients have been shown to produce high levels of interleukin-6 (IL-6), IL-10 and tumour necrosis factor-
(TNF-) in vitro [5,6].
One of the clinical consequences of aberrant cytokine production may be impaired erythropoiesis. In this regard, TNF- and interferon-
(IFN-) have been shown to suppress erythroid colony formation in vitro [7], by promoting apoptosis of these cells [8]. These findings may explain, in part, the observation that HD patients with inflammation or infection are resistant to Epo therapy. Production of IL-6 and TNF- from cultured peripheral blood mononuclear cells (PBMC) has been shown to correlate with Epo dose [9].
The aim of this study was therefore to investigate the relationship between T-cell function and responsiveness to Epo in HD patients and to compare the results with those from healthy controls. The patients selected as poor responders for the study had all the usual causes of Epo resistance excluded, such as iron deficiency, underdialysis and severe hyperparathyroidism. T-cell phenotypes were assessed using flow cytometry and T-cell function was studied by measuring in vitro cytokine production.
|
Subjects and methods |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Patients
Thirty-two patients and 17 normal controls were included in this study. Two distinct groups of HD patients were evaluated: (i) HD patients not responding to Epo with Hb <10 g/dl, Epo/Hb ratio >15 (defined as the dose of Epo (International Units/kg body weight) divided by the Hb concentration); and (ii) HD patients responding well to Epo with Hb >10 g/dl and Epo/Hb ratio <10. The mean time on HD did not vary between the two groups (29±20 months for good responders vs 39±38 months for poor responders; P=0.41). Similarly, the iron status of the good responders (ferritin 622±287 µg/ml) was equivalent to that of the poor responders (ferritin 636±224 µg/ml; P=0.44) and all the HD patients in both groups received intravenous iron as required for maintenance of serum ferritin >100 µg/l. The serum albumin levels were comparable in the two groups (37±6 ng/l for good responders vs 37±7 ng/l for poor responders; P=0.68). The urea reduction ratio was not significantly different between the good and poor responder groups (67±60% and 65±88%, respectively; P=0.45). Likewise, there was no significant difference in parathyroid hormone levels between the two groups (391±438 ng/l for good responders vs 357±371 ng/l for poor responders; P=0.94). The patients were all maintained on the same dose of subcutaneous Epo for at least 3 months prior to investigation.
The causes of renal failure in the patient population were as follows: hypertensive nephropathy (n=6), mesangiocapillary glomerulonephritis (n=2), IgA nephropathy (n=1), mesangioproliferative glomerulonephritis (n=1), focal segmental glomerulosclerosis (n=1), ischaemic nephropathy (n=1), medullary sponge kidneys (n=1), nephrocalcinosis (n=1), obstructive nephropathy (n=1), polycystic kidney disease (n=1), reflux nephropathy (n=1), renal amyloid (n=1), renovascular disease (n=1), cortical necrosis (n=1) and unknown (n=12). Of the good responders, 4/16 were taking ramipril and 2/16 were taking enalapril. For the poor responders, 4/16 were taking ramipril, 3/16 were taking enalapril and 1/16 was taking perindopril. Patients with autoimmune disease, malignancy, haematological disorders, diabetes mellitus, systemic vasculitis, iron deficiency, overt acute and chronic infection or those taking immunosuppressive therapy were all excluded from the study.
The patients were dialysed three times per week for 3–5 h (mean dialysis time: 3.7±0.6 h), using either a cellulose diacetate (good responders, n=10; poor responders, n=8), haemophan (good responders, n=5; poor responders, n=5) or polyacrylonitrile (good responders, n=2; poor responders, n=2) dialyser. A total of 26 patients had an arteriovenous fistula as vascular access, three patients had a polytetrafluoroethylene graft and three patients had a tunnelled dialysis catheter as vascular access. The bacterial count of the dialysate solution was within British Renal Association guidelines. Healthy volunteers were used as normal controls and, as far as possible, the three groups were age- and sex-matched. The study was approved by the Research Ethics Committee of King's College Hospital and informed consent was obtained from all subjects.
For flow cytometry experiments, the following were studied: 12 HD patients responding poorly to Epo, 12 HD patients responding well to Epo and 14 normal controls. For the cytokine studies, there were 10 subjects studied in each of the three groups.
Cell preparation and culture conditions
Blood samples (25 ml) were drawn into plastic tubes containing 1000 U sodium heparin (Leo Laboratories Ltd, Princes Risborough, UK). Blood from dialysis patients was taken immediately prior to a dialysis session. Peripheral blood mononuclear cells were isolated from whole blood on a Lymphoprep® (Robbins Scientific Europe Ltd, Solihull, UK) density gradient at 400xg for 30 min. The upper plasma layer was taken for cytokine measurement. The mononuclear cell layer was harvested and washed three times using phosphate-buffered saline (PBS) (Gibco BRL, Life Technologies Ltd, Paisley, UK) at 300xg for 20 min. Cell viability was assessed using trypan blue and samples that were <99% viable were discarded. The PBMC were cultured at 2x106/ml in Iscoves-modified buffer containing Glutamax (Gibco BRL, Life Technologies Ltd, Paisley, UK), supplemented with 10% foetal calf serum (FCS) (Gibco BRL, Life Technologies Ltd, Paisley, UK), 100 U/ml penicillin (Gibco BRL, Life Technologies Ltd, Paisley, UK) and 50 µg/ml streptomycin (Gibco BRL, Life Technologies Ltd, Paisley, UK). The cultures were set up either in the absence of stimulus, in the presence of 10 µg/ml lipopolysaccharide (LPS) (Sigma Chemical Co., Poole, UK) or in the presence of monoclonal antibodies (mAbs) to CD3 (5 µg/ml) (Clone HIT3a, Becton Dickinson (UK) Ltd, Oxford, UK) and CD28 (2 µg/ml) (Clone M1456, Eurogenetics UK Ltd, Hampton, UK). Lipopolysaccharide activates monocytes while the combination of mAbs to CD3 and to CD28 activates T-lymphocytes. Each experimental condition was performed in duplicate. The PBMC were incubated for 48 h (37°C, 5%, CO2) and the cells were then pelleted (30 s, 15 000xg). The supernatants were then stored at -80°C until analysed.
Cytokine measurement in culture supernatants and plasma samples
Cytokine levels were determined in PBMC culture supernatants and plasma samples using commercially available enzyme-linked immunosorbent assay (ELISA) kits (BioSource International, Camarillo, CA, USA) according to the manufacturer's instructions. The assay ranges for IFN-, IL-10 and sIL-2R were 0–1000, 0–500 and 0–800 pg/ml, respectively, and the lower detection limits were 4, 5 and 16 pg/ml, respectively. An ultrasensitive ELISA kit was used for IL-4 determination, with a range of 0–25 pg/ml and a lower detection limit of 0.27 pg/ml. Analysis of plasma IFN-, IFN- and TNF- was carried out using ELISA, developed in-house in the Department of Clinical Immunology, King's College Hospital. The assay ranges for IFN-, IFN- and TNF- were 0–500, 0–1000 and 0–10 000 pg/ml, respectively, and the lower detection limits were 8, 1 and 10 pg/ml, respectively. Plasma IL-6 and sIL-2R levels were determined using commercially available ELISA kits (BioSource International, Camarillo, CA, USA). The IL-6 assay range was 0–500 pg/ml with a lower detection limit of 2 pg/ml.
Flow cytometry
Dual-colour immunofluorescence was performed by labelling cells with fluorescein isothiocyanate-conjugated and phycoerythrin-conjugated mAbs (Becton Dickinson (UK) Ltd, Oxford, UK). Peripheral blood mononuclear cells were aliquoted, then incubated with the optimal concentration of appropriate mAb (together with the appropriate isotype matched controls) for 30 min at 4°C in the dark. The samples were washed with PBS containing 1% FCS and centrifuged at 200xg for 10 min and resuspended in PBS containing 1% paraformaldehyde (Sigma Chemical Co., Poole, UK). The labelled PBMC were stored at 4°C in the dark until analysed within 1 week. Measurements were performed on a Becton Dickinson FACScan flow cytometer and the Cellquest software 3.1. Initial gating was performed using forward and side scatter to identify T-lymphocytes. A total of 20 000 events was collected. The percentage of cells with a fluorescence intensity 
95% of control events (per cent positive) was determined.
C-reactive protein
C-reactive protein (CRP) levels were measured on a Behring Nephelometer Analyser II, using a mAb to CRP and the latex enhanced nephelometric reaction. The detection range was 0–10 mg/ml.
Statistical analysis
Statistical analysis was carried out using Prism V3.0 statistical software (Graphpad, San Diego, USA). Results are expressed as means±SD. Differences between the means were analysed using the Mann–Whitney test for unpaired data and were considered significant at P<0.05. Differences between survival curves were analysed using the Mantel–Haenszel test and were considered significant at P<0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
Demographic and haematological characteristics
Table 1
summarizes the demographic and haematological characteristics of the three groups studied. By definition, the mean Epo dose in the poor responders to Epo was significantly higher than the mean Epo dose of the HD patients responding well to Epo. The poor responders were significantly more anaemic than the good responders. Thus, the Epo/Hb ratio was significantly different between the two HD groups. The good responders and the poor responders were age-matched (Table 1). Although age matching was attempted, the mean age of the normal control group was found to be slightly lower than both the HD patient groups. C-reactive protein levels were higher in the poor responders (18.6±20.7 mg/l) compared with the good responders (8.7±8.0 mg/l; P<0.05), which in turn were higher than in the normal controls (3.8±1.1 mg/l; P<0.002).
|
The white-cell count in the three subject groups did not differ significantly (Table 1
). However, lymphocyte numbers were significantly lower in both the poor responders and the good responders compared with the normal controls. There was no significant difference in the lymphocyte count between good and poor responders. Monocyte levels were significantly elevated in the good responders compared with the normal controls, but there was no such increase in the poor responders. Similarly, both the neutrophil count and the basophil numbers were significantly increased in the good responders compared with the normal controls. Eosinophil numbers were not significantly different among the three subject groups.
Flow cytometric immunofluorescence analysis
Immunofluorescence and flow cytometry were used to compare the surface antigen expression on freshly isolated CD4+ and CD8+ T-cells in the three subject groups. The results are summarized in Table 2. Low levels of CD25 antigen expression (which comprises the alpha chain of the IL-2 receptor), was found on CD4+ T helper cells isolated from all three subject groups.
|
Figure 1A
shows the percentage of CD4+ cells that expressed the T-cell co-stimulatory molecule, CD28. Both T-cell proliferation and cytokine production require two signals: one involving the T-cell receptor–CD3 complex and a second co-stimulatory signal through the B7-1 or B7-2 molecule on antigen-presenting cells and the CD28 antigen on T-cells. Interestingly, CD28 expression in the poor responders was significantly lower at 88±11% compared with both the normal controls (99±1%; P<0.001) and the good responders (97±4%; P<0.005). Thus, the poor responders had a mean number of 12±11% CD4+/CD28- cells (range: 58–97%).
|
Isoforms of the antigen CD45 (a phosphotyrosine phosphatase involved in T-cell signalling) have been used to distinguish between naive and memory T-cells. CD45RO+ T-cells are considered to be memory cells due to their ability to respond vigorously in vitro to previously encountered antigen and they migrate throughout the body. CD45RA+ cells are regarded as naive since they do not respond well to antigen and they migrate almost exclusively throughout lymphoid tissue. CD45RA expression on CD4+ T-cells was not significantly different in the three groups (Table 2
). Similarly, CD45RO expression on CD4+ cells did not significantly differ between the three subject groups.
For CD8+ T cytotoxic cells, CD25 antigen expression was found to be negligible in all three groups (Table 2). As with the CD4+ T-cells, CD28 expression in the CD8+ cells was reduced in the poor responder group compared with both the normal controls and good responders (Figure 1B). CD45RA and CD45RO expression on CD8+ cells did not differ significantly among each of the three groups (Table 2).
PBMC generation of cytokines
In vitro PBMC generation of cytokines was studied in order to compare immune reactivity among the three subject groups. Table 3 summarizes the results obtained for cytokine detection in PBMC culture supernatants. Significant IL-4 was only detected in PBMC that were stimulated with mAbs to CD3/CD28 and the levels were similar in all three subject groups (Table 3).
|
A high level of IFN-
was only detected in the unstimulated supernatant of one out of 10 of the poor responders (Table 3). Little IFN- was detected in supernatants of either the normal controls or the good responders (Table 3). Similarly, LPS-stimulation caused high IFN- generation in only one of the poor responders and negligible generation in the other two groups (Table 3). CD3/CD28 mAbs challenge of PBMC generated similar amounts of IFN- in all three groups (Table 3).
The membrane-bound CD25 antigen (alpha chain of IL-2 receptor) can be proteolytically cleaved to generate a soluble form that is present in human plasma (abbreviated to sIL-2R). In PBMC cultures, low levels of sIL-2R were detected in unstimulated and LPS-stimulated cultures. Higher levels were detected when PBMC were activated with mAbs to CD3/CD28 (Table 3). For all three culture conditions, there was no significant difference in sIL-2R levels among any of the subject groups.
Figure 2A shows IL-10 generation in unstimulated PBMC. Poor responders had significantly higher levels of IL-10 compared with both good responders and normal controls. Interleukin-10 production in unstimulated PBMC from good responders did not differ statistically from normal controls. Lipopolysaccharide challenge resulted in elevated IL-10 levels in both poor responders and good responders compared with normal controls (Figure 2B and Table 3). In contrast, mAbs to CD3/CD28-stimulation resulted in elevated IL-10 levels in the poor responders alone, compared with normal controls (Figure 2C and Table 3), while IL-10 generation in the good responder group was not statistically different from the other two groups.
|
Plasma IFN-, IFN-, TNF-, IL-6 and sIL-2R
Plasma levels of IFN-, IFN- and TNF- were undetectable in both HD patients and normal controls (Table 4). IL-6 was significantly elevated in both HD patient groups compared with the normal controls although there was no significant difference between the two patient groups (Table 4). Likewise, plasma sIL-2R was found to be significantly elevated in both good and poor responders compared with controls and again there was no statistical difference between the HD patient groups.
|
Mortality
Figure 3 shows the Kaplan–Meier survival curves for the good responders to Epo (n=17) and the poor responders to Epo (n=15). During the 24 month study period, seven poor responders died due to the following reasons: sepsis (n=4), post-transplant failure (n=1), cardiovascular death (n=1) and access failure (n=1). In the good responder group, two patients died due to cardiovascular death. Using the Mantel-Haenszel log-rank test, the survival curve for the poor responder group was significantly different from the good responder group (P<0.05).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
This study demonstrates for the first time that HD patients responding poorly to Epo with no obvious underlying cause have different circulating T-cell immunophenotypes compared with good responders and normal controls. Peripheral blood T-cells from poor responders have increased proportions of both CD4+/CD28- cells and CD8+/CD28- cells compared with both good responders to Epo and normal controls.
CD4+/CD28- T-cells are rarely found in healthy individuals (the proportion is usually<1%), but they have been described in a number of other chronic inflammatory states. Thus, increased numbers of CD4+/CD28- cells have been found in patients with rheumatoid arthritis [10], unstable angina [11], HIV infection [12] and Wegener's granulomatosis [13]. Interestingly, this is the first time that CD4+/CD28- cells have been found in HD patients that are responding poorly to Epo. Compared with their CD28+ counterparts, CD4+/CD28- T-cells produce significantly more IFN- [10], enabling them to function as pro-inflammatory cells. CD4+/CD28- cells persist in the circulation for years and this has recently been attributed to a defect in their apoptotic pathway [14], Hence, CD4+/CD28- T-cells are resistant to apoptosis, accounting for their clonal outgrowth and maintenance in vivo [14].
It was important that the two HD patient groups in this study were age-matched because increased proportions of CD8+/CD28- T-cells have been observed during ageing [15]. Healthy adults usually have between 25 and 50% of CD8+ T-cells that are CD28- [16]. Viral infections such as HIV-1 infection [12] and Epstein–Bar virus-induced mononucleosis [17] are associated with higher populations of CD8+/CD28- T-lymphocytes. A higher incidence of these cells has also been found in Wegener's granulomatosis [13]. The present study found that poor responders to Epo also had increased proportions of CD8+/CD28- compared with good responders to Epo and normal controls. The pathological relevance of CD8+/CD28- and CD4+/CD28- to the poor response to Epo is unclear. The accumulation of these unusual cell types may be an epiphenomenon, without actually contributing to the anaemia seen in poor responders to Epo. It is interesting, however, that CD8+/CD28- and CD4+/CD28- T-cells secrete high levels of IFN- [11] since this pro-inflammatory cytokine inhibits erythroid colony growth in vitro [7,8] and is implicated as one of the major factors causing impaired erythropoiesis and anaemia in chronic inflammatory states.
This study did not find any difference in secretion of IFN- or IL-4 in PBMC cultures obtained from HD patients and normal controls. Furthermore, there was no difference in secretion of these cytokines between good and poor responders. A previous study has suggested that PBMC isolated from poor responders secreted lower levels of IFN- compared with good responders [9]. Our clinical data indicated that HD patients have reduced circulating levels of lymphocytes in peripheral blood. Therefore, the PBMC culture data do not reflect cytokine production per absolute cell number. Monocyte IL-6 and IL-10 production assessed by flow cytometry has been shown to be increased in HD patients compared with normal controls [5]. However, the same study found that there was no difference in cultured PBMC generation of the same cytokines. The authors concluded that the intracellular cytokine staining method could delineate differences in cytokine production between subject groups more precisely than ELISA methods. Hence, the reduced relative numbers of lymphocytes in our mixed leukocyte cultures may have biased the cytokine secretion experiments against detecting differences among the three subject groups.
We did, however, find that unstimulated PBMC generation of IL-10 was significantly increased in poor responders compared with both normal controls and good responders. Monoclonal antibodies to CD3/CD28-stimulation also caused increased IL-10 production in poor responders compared with normal controls. Interleukin-10 is generated by monocytes, natural killer cells and T-lymphocytes [18]. The source of IL-10 in our unstimulated cultures could be from any or all of these cell types. The CD3/CD28 mAbs exclusively stimulate T-cells and it is therefore likely that the IL-10 in these cultures is derived from T-cells alone, although it is possible that a T-cell-derived factor may have activated monocytes to produce IL-10 in our mixed leukocyte cultures.
Interleukin-10 has been shown to down-regulate inflammatory reactions, predominantly by limiting monocyte function. This cytokine potently inhibits both cytokine and chemokine secretion from monocytes [18]. IL-10 also interferes with the antigen presenting capacity of monocytes by inhibiting their surface expression of MHC class II, B7-1 and B7-2, causing the inhibition of CD4+ T-cell proliferation and cytokine production [18]. In contrast, IL-10 has a stimulatory action on CD8+ T-cells and induces their recruitment, cytotoxicity and proliferation [18]. Activation of T-cells in the presence of IL-10 can cause irreversible non-responsiveness known as T-cell anergy and such anergy can be induced in vivo by continuous antigenic challenge [19].
The observation that poor responders to Epo have increased levels of IL-10 in vitro compared with both good responders and normal controls may indicate that leukocytes from these HD patients are primed to secrete higher levels of this anti-inflammatory cytokine. This effect may be ‘adaptive’, occurring as a result of increased underlying inflammation. In this context, the increased production of IL-10 would down-regulate the inflammatory process that may be associated with the poor response to Epo. It is interesting to note that we have found that poor responders to Epo have increased plasma levels of IL-12 compared with good responders and normal controls (unpublished data). A previous study has shown that spontaneous PBMC secretion of IL-12 is increased in HD patients dialysed with cuprophan membranes [20]; IL-10 is known to inhibit the production of IL-12 from activated macrophages and monocytes [18]. T-cells from poor responders to Epo may therefore secrete more IL-10 to compensate for the elevated IL-12 production.
Although not a primary objective of the study, it became apparent on following up the HD patients that more deaths were occurring among the poor responders than the good responders. This was therefore assessed using the Mantel–Haenszel log-rank test, and the survival curves were indeed different, with seven of the 15 poor responder patients dying compared with only two of the 17 good responders. This is consistent with the observation that chronic inflammation and high levels of inflammatory markers are predictive of mortality in a dialysis population [4].
In summary, this study shows that poor response to Epo is associated with an increase in IL-10 generation in vitro and this may be indicative of enhanced immune activity. The observation that poor responders have raised CRP levels compared with both good responders and normal controls, despite the absence of any overt infection or inflammation, supports this theory. It was interesting that poor responders had increased numbers of both CD4+/CD28- and CD8+/CD28- because these T-cell phenotypes occur in chronic inflammatory diseases, again implying that inflammation is linked to poor response to Epo. CD4+/CD28- and CD8+/CD28- T-cells produce large amounts of the pro-inflammatory cytokine IFN-, which has been shown to inhibit erythropoiesis. The overall results from this study suggest that, in the absence of any obvious cause, poor response to Epo may be mediated by generation of pro-inflammatory cytokines from a subpopulation of activated T-cells, which then promote apoptosis in erythroid progenitor cells in the bone marrow [8].
|
Notes |
|---|
Correspondence and offprint requests to: Angela Cooper, Department of Immunology, Rayne Institute, 123 Coldharbour Lane, London SE5 9NU, UK. Email: angela.redrup{at}kcl.ac.uk
|
References |
|---|
|
TopAbstractIntroductionSubjects and methodsResultsDiscussionReferences |
|---|
- Macdougall IC. Poor response to erythropoietin: practical guidelines on investigation and management. Nephrol Dial Transplant1995; 10:607–614
[Free Full Text] - Kaysen GA. The microinflammatory state in uremia: causes and potential consequences. J Am Soc Nephrol2001; 12:1549–1557
[Abstract/Free Full Text] - Sitter T, Bergner A, Schiffl H. Dialysate related cytokine induction and response to recombinant human erythropoietin in haemodialysis patients. Nephrol Dial Transplant2000; 15:1207–1211
[Abstract/Free Full Text] - Kimmel PL, Phillips TM, Simmens SJ et al. Immunological function and survival in hemodialysis patients. Kidney Int1998; 54:236–244[CrossRef][Web of Science][Medline]
- Girndt M, Sester U, Kaul H, Kohler H. Production of proinflammatory cytokines and regulatory monokines in hemodialysis patients shown at a single cell level. Am Soc Nephrol1998; 9:1689–1696[Abstract]
- Girndt M, Kohler H, Schiedhelm Weick E, Schlaak JF, Meyer zum Buschenfelde KH, Fleischer B. Production of interleukin-6, tumour necrosis factor alpha and interleukin-10 in vitro correlates with the clinical immune defect in hemodialysis patients. Kidney Int1995; 47:559–565[Web of Science][Medline]
- Allen DA, Breen C, Yaqoob MM, Macdougall IC. Inhibition of CFU-E colony formation in uremic patients with inflammatory disease: role of IFN-gamma and TNF-alpha. J Investig Med1999; 47:204–211[Web of Science][Medline]
- Dai C, Krantz SB. Interferon gamma induces upregulation and activation of caspases 1, 3, and 8 to produce apoptosis in human erythroid progenitor cells. Blood1999; 93:3309–3316
[Abstract/Free Full Text] - Goicoechea M, Martin J, de Sequera P et al. Role of cytokines in the response to erythropoietin in haemodialysis patients. Kidney Int1998; 54:1337–1343[CrossRef][Web of Science][Medline]
- Martens PB, Goronzy JJ, Schaid D, Weyand CM. Expansion of unusual CD4+ T cells in severe rheumatoid arthritis. Arthritis Rheum1997; 40:1106–1114[Web of Science][Medline]
- Liuzzo G, Kopecky SL, Frye RL et al. Perturbation of the T-cell repertoire in patients with unstable angina. Circulation1999; 100:2135–2139
[Abstract/Free Full Text] - Choremi-Papadopoulou H, Viglis V, Gargalianos P, Kordossis T, Iniotaki-Theodoraki A, Kosmidis J. Downregulation of CD28 surface antigen on CD4+ and CD8+ T lymphocytes during HIV-1 infection. J Acquir Immune Defic Syndr1994; 7:245–253
- Moosig F, Csernok E, Wang G, Gross WL. Costimulatory molecules in Wegener's granulomatosis (WG): lack of expression of CD28 and preferential up-regulation of its ligands B7-1 (CD80) and B7-2 (CD86) on T cells. Clin Exp Immunol1998; 114:113–118[CrossRef][Web of Science][Medline]
- Vallejo AN, Schirmer M, Weyand CM, Goronzy JJ. Clonality and longevity of CD4+CD28null T cells are associated with defects in apoptotic pathways. J Immunol2000; 165:6301–6307
[Abstract/Free Full Text] - Fagnoni FF, Vescovini R, Mazzola M et al. Expansion of cytotoxic CD8+ CD28- T cells in healthy ageing people, including centenarians. Immunology1996; 88:501–507[CrossRef][Web of Science][Medline]
- Azuma M, Phillips JH, Lanier LL. CD28- T lymphocytes. Antigenic and functional properties. J Immunol1998; 150:1147–1159
- Borthwick NJ, Bofill M, Hassan I et al. Factors that influence activated CD8+ T-cell apoptosis in patients with acute herpes virus infections: loss of costimulatory molecules CD28, CD5 and CD6 but relative maintenance of Bax and Bcl-X expression. Immunology1996; 88:508–515[Web of Science][Medline]
- Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A. Interleukin-10 and the interleukin-10 receptor. Ann Rev Immunol2001; 19:683–765[CrossRef][Web of Science][Medline]
- Buer J, Lanoue A, Franzke A, Garcia C, von Boehmer H, Sarukan A. Interleukin 10 secretion and impaired effector function of major histocompatibility complex class II-restricted T cells anergized in vivo. J Exp Med1998; 187:177–183
[Abstract/Free Full Text] - Memoli B, Marzano L, Bisesti V, Andreucci M, Guida B. Hemodialysis related lymphonuclear release of interleukin-12 in patients with end-stage renal disease. J Am Soc Nephrol1999; 10:2171–2176
[Abstract/Free Full Text]
Accepted in revised form: 21. 8.02
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. K. Hendrikx, E. A. F. J. van Gurp, W. M. Mol, W. Schoordijk, V. D. K. D. Sewgobind, J. N. M. IJzermans, W. Weimar, and C. C. Baan End-stage renal failure and regulatory activities of CD4+CD25bright+FoxP3+ T-cells Nephrol. Dial. Transplant., June 1, 2009; 24(6): 1969 - 1978. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Cooper, A. Mikhail, M. W. Lethbridge, D. M. Kemeny, and I. C. Macdougall Pentoxifylline Improves Hemoglobin Levels in Patients with Erythropoietin-resistant Anemia in Renal Failure J. Am. Soc. Nephrol., July 1, 2004; 15(7): 1877 - 1882. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||