It is known that GC therapy is the treatment of choice for patients with SLE. However some patients fail to respond to the treatment even when given high-dose GCs. For those patients, the treatment should be bolstered by synergising GCs with other immunosuppressants. If clinical response was predicted before therapy, synergised treatment might be performed at the beginning of the treatment to avoid side effects of chronic high-dose hormone therapy, which could improve the individual response to GC therapy and benefit more patients. GR seems to be related to the pathogenesis of steroid resistance. Thus, we hypothesised that detection of GR by a suitable method before treatment might be used to predict steroid response.
In the present study, we detected GR in PBMCs in SLE patients by FCM with GR-mAb and FITC-Dex probes, combined with cell surface staining for CD3 and CD14. Characteristics of patients, including age, male/female ratio, doses of prednisolone, disease history, etc, between SR and SS groups were coincident. Our results show that by FCM, the expression and binding of GR in SR patients with SLE were lower than those in both SS and control groups. However, there was no difference in the expression and binding of GRs in SS patients compared with those in controls except for the binding of GR in T lymphocytes. Similar results were also found in patients with nephrotic syndrome (data not shown). A previous study found decreases in the number and affinity of GR in SR patients with idiopathic nephritic syndrome and purported that the altered secretion of cytokines may be involved in tissue sensitivity to GCs [10
]. Other studies have shown no change in GR of SR patients compared with those of SS patients and controls [7
]. Different age distribution and pathological states may account for these conflicting results. In our study, the subjects were mainly adult patients with similar age distribution. We determined GR in CD3+
subset rather than the heterogeneous aggregate, which may play a role in these differences. Alternatively, GR was detected by FCM with different probes for the expression and binding at the same time. Moreover, the administration of GR agonists results in a significant down-regulation of the expression of GR in vitro
and in vivo
, so we detected GR of SLE patients in the active phase before the treatment rather than during the treatment. In summary, these data suggest that down-regulation of GR expression and binding might mediate steroid resistance in some patients with SLE.
We also evaluated the percentage of positive lymphocytes and monocytes presenting GR in SLE patients and normal controls. No difference was found between the results of patients and controls. Similar results were obtained by Wasilewska and colleagues [33
]. From these results, we infer that the expression and binding of GR on a per cell basis, but not the overall positive rate of GR, might be correlated with steroid response.
In addition, we performed an in vitro
corticosteroid sensitivity assay in patients and normal controls. In vitro
Dex inhibition of Con-A stimulated cytokine release has been used to examine the effect of corticosteroids in idiopathic nephritic syndrome [10
]. In our study, the control and SS groups presented higher inhibition of cytokine secretion than SR group, which confirms patient classification as SR and SS by SLEDAI score. As a result, the detection of GR by FCM before steroid therapy as well as in vitro
corticosteroid sensitivity assay may predict clinical response to the treatment more accurately. As to the relation of GR and inhibition of cytokine secretion, there was a positive relation between the GR binding in two subsets (T lymphocytes and monocytes) and the inhibition rate of cytokine secretion, which was different from the GR expression. However, whether the low binding capacity of GR is contributing to the reduced inhibition rate of cytokines or the systemic inflammation induced GC resistance [34
] need to be further discussed.
GR binding and expression have been detected recently by RLBA and western blot, separately. However, the two methods could not evaluate individual cells or specific cell types in a given tissue sample. Moreover, it is challenging to use the two methods in a clinic for they are time consuming, expensive and radioactive materials are unavoidably involved.
FCM is a convenient tool for determining protein expression of cell surface receptors and has also been shown to be useful in identifying expression of intracellular proteins in permeabilised cells [36
]. Especially when complex mixtures of cells are present, FCM can be used to sort subpopulations of cells and therefore identify GR expressed in specific cell types. In a recent study, a method for the quantification of GR by FCM was developed in different cell lines [37
]. Although it established the method which could detect GR, even quantifying the number of receptor molecules per cell, it seems that they did not reflect GR from various points of view. A few studies in the literature indicated that a high level of GR did not always relate to a satisfying response to GC therapy [38
]. Similar observations have been made in studies of patients with ulcerative colitis [39
]. As a result, the analysis of GR expression and binding was important to the evaluation of the response to GC therapy.
Detection of GR expression and binding by FCM using mAb and fluorescein ligand probes at the same time was first reported in clones of CCRF-CEM human leukaemic cells [27
]. In the current study, it was first used to identify biomarkers that clinically predict steroid response. To confirm the GR measurements by FCM, RLBA, which had been used for GR binding detection, and western blot, which has been used for GR expression detection, were performed at the same time and the results were consistent between FCM and western blot or BLRA.