Introduction
Chronic lymphocytic leukaemia (CLL) is the most common form of leukaemia in adults in Western Europe and North America. It has been described as a progressive accumulation of malignant, morphologically mature CD19
+CD5
+ cells in the peripheral blood, bone marrow, and lymphoid organs [
1‐
3]. The disease is characterised by a highly variable clinical presentation and evolution [
4,
5]. Some patients exhibit stable disease over years and never require a therapeutic intervention, whereas others progress rapidly towards more advanced stages and succumb relatively soon despite aggressive treatment. Based on genetic, phenotypic, and molecular characteristics of CLL, several prognostic markers have emerged in the past decade [
6,
7]. It has been well documented that the high-risk phenotype is typically associated with unmutated immunoglobulin heavy variable genes (IgV
H) [
8], high expression of the CD38 surface marker [
9], or the ζ-chain-associated protein 70 kDa (ZAP-70) [
10], as well as with chromosomal aberrations such as 17p (the site of tumour protein p53) or 11q23 deletions (the site of ataxia telangiectasia mutated ATM) [
11]. Recently, it has been postulated that the difference in clinical course among CLL patients may, at least in part, be determined by the proliferative capacity of CLL cells [
12,
13]. It is becoming increasingly evident that the cells located in the proliferation centres (PCs) of lymph nodes and bone marrow proliferate more than previously anticipated [
13], and CLL patients with higher birth rates are much more likely to exhibit active or to develop progressive disease than those with lower birth rates [
13]. Moreover, the proliferating cells in PCs represent the CLL proliferating reservoir that replenishes the downstream accumulation compartment.
In contrast to the proliferating cells in PCs, the vast majority of CLL cells circulating in the peripheral blood are arrested in the G0/early G1 phase of the cell cycle. The key regulators of G1 phase progression in human T and B lymphocytes are the D-type cyclins cyclin D2 and D3, and p27
KIP1
(cyclin-dependent kinase inhibitory protein 1). The D-type cyclins positively regulate passage through the G1 phase by binding to and stimulating the activities of their catalytic partners, the cyclin-dependent kinases (CDKs) cdk4 and cdk6 [
14‐
17], while p27
KIP1
exerts an inhibitory effect on the kinases cdk4 and cdk6 or their complexes with cyclin D2 or cyclin D3 [
17‐
19]. Moreover, one of the factors involved in regulation of cell cycle progression of T cells is cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4; CD152) [
20,
21]. It has been well documented that CTLA-4 prolongs the progression of T cells through the G1 phase by influencing the expression of the major regulators of this cell cycle phase [
20,
21]. CTLA-4 up-regulates the expression of cyclin D2 and inhibits cyclin D3, cdk4, and cdk6 production in these cells. Furthermore, CTLA-4 affects the degradation of p27
KIP1
protein and contributes to its earlier and stronger re-expression during the late stages of T cell activation [
20,
21]. In contrast to the well-documented involvement of CTLA-4 in the regulation of cell cycle progression in T cells [
20,
21], only limited information is known about the role of this protein in cell cycle progression in normal B cells and malignant B lymphocytes. Our previous study indicated that CTLA-4 is overexpressed in freshly drawn CLL cells and it may be involved in the regulation of G1 phase progression in these cells [
22]. We found that CTLA-4 expression positively correlated with both cyclin D2 and p27
KIP1
expression and negatively with cyclin D3 level. Moreover, CTLA-4 expression positively correlated with the percentage of leukaemic cells in G0/G1 phase. Here, we have extended our previous study to examine whether stimulation with DSP30, a CpG oligodeoxynucleotide (ODN), and rIL-2 influences CTLA-4 expression in CLL cells. The main aim of this study was to investigate whether the CTLA-4 molecule affects the expression of cell cycle regulators of G0/G1 phase. For that purpose, we blocked CTLA-4 on the surface of CLL cells using monoclonal anti-CTLA-4 antibodies to assess the expression of cyclins D2 and D3, and p27
KIP1
protein. To the best of our knowledge, such studies are lacking so far.
Materials and methods
Patients and healthy donors
The study design was approved by the local Bioethical Committee at the Medical University of Wroclaw, Poland, and is in accordance with the Helsinki Declaration of 1975. All participants gave written informed consent after the purpose of the study was explained to them. Thirty-eight previously untreated CLL patients of the Clinic of Haematology, Blood Neoplasms, and Bone Marrow Transplantation, Wroclaw Medical University, Poland, were enrolled in this study. In each of them, the diagnosis was established according to generally accepted criteria including absolute peripheral blood lymphocytosis ≥5 × 10
9/L and the co-expression of CD5, CD19, and CD23 antigens on malignant cells. The disease stages were determined according to the Rai classification. Clinical and laboratory features are presented in Table
1.
Table 1
Clinical characteristics of CLL patients
Number of patients | 38 |
Gender (female/male) | 15/23 |
Age (years, mean and SD) | 69.0 ± 11.2 |
Rai stage | |
0 | 14 |
I | 7 |
II | 6 |
III | 4 |
IV | 7 |
WBC count (1 × 109/l) | 72.2 ± 98.4 |
Lymphocyte count (1 × 109/l) | 66.5 ± 96.8 |
Hb level (g/dl) | 12.9 ± 1.9 |
Platelet count (1 × 109/l) | 135.2 ± 55.1 |
LDH (U/l) | 206.8 ± 55.3 |
β2-microglobulin (mg/l) | 3.3 ± 1.4 |
Leucocyte-enriched fractions of peripheral blood donated by 15 healthy volunteers matched for age and sex with the CLL patients were purchased from the Regional Centre of Blood Donation and Treatment in Wroclaw, Poland.
Cell isolation and separation procedures
Peripheral blood mononuclear cells (PBMCs) were separated from heparinised freshly drawn peripheral venous blood of CLL patients and healthy controls by buoyant density-gradient centrifugation on Lymphoflot (Bio-Rad Medical Diagnostics GmbH, Dreieich, Germany) and washed three times in phosphate-buffered saline (PBS) (without Ca2+ and Mg2+). The PBMCs were suspended in 95 % foetal calf serum (CytoGen GmbH, Sinn, Germany) containing 5 % DMSO (Sigma-Aldrich, St. Gallen, Switzerland) and stored in liquid nitrogen until used.
CLL cells were isolated from PBMCs by negative selection using EasySep Human B Cell Enrichment Kit without CD43 Depletion (STEMCELL Technologies Inc, Vancouver, Canada) according to the manufacturer’s instructions. Following this separation procedure, more than 98 % of the resulting cell population was CD19+CD5+ as assessed by flow cytometry using anti-CD19 and anti-CD5 monoclonal antibodies (mAbs) (Becton–Dickinson, BD Biosciences, San Diego, USA). Normal B cells from healthy individuals were isolated from PBMCs by negative selection using EasySep Human B Cell Enrichment Kit (STEMCELL Technologies Inc, Vancouver, Canada) according to the manufacturer’s instructions, achieving above 98 % purity as assessed by flow cytometry using anti-CD19 mAb.
Culture conditions
Purified normal CD19
+ lymphocytes or CLL cells were suspended at 1 × 10
6 cells/ml in RPMI-1640 medium (Gibco, Paisley, UK) supplemented with 10 % foetal calf serum (CytoGen GmbH, Sinn, Germany), 2 mmol/l
l-glutamine and 50 μg/ml gentamycin (KRKA-Poland, Warsaw, Poland), and cultured using 24-well U-bottom culture plates (Nunc GmbH & Co. KG, Langenselbold, Germany) at 37 °C in a 5 % CO
2 humidified atmosphere for 24 and 72 h either in medium alone or together with 1 µM DSP30 (5′-TCGTCGCTGTCTCCGCTTCTTCTTGCC-3′) (TIB MOLBIOL, Berlin, Germany) [
23] and 100 U/ml rIL-2 (Eurocetus, Amsterdam, The Netherlands) [
24]. For the blocking experiment, purified CLL cells and normal CD19
+ lymphocytes were cultured with 1 µM DSP30 and 100 U/ml IL-2 with the blocking anti-CTLA-4 mAbs (50 µg/ml) (BD Pharmingen, BD Biosciences, San Diego, USA) [
25] or control IgG2 (50 µg/ml) (BD Pharmingen, BD Biosciences, San Diego, USA).
Immunostaining of CTLA-4 and cell cycle regulators of G0/G1 phase, and flow cytometric analysis
The expression of these molecules was studied in purified CLL cells and normal CD19+ lymphocytes before and after 24 and 72 h culture by a single immunostaining method.
Briefly, for detection of surface expression of the CTLA-4 molecule, the cells were washed twice in PBS (without Ca2+ and Mg2+), divided into tubes at a concentration of 5 × 105 cells per tube and incubated with anti-CTLA-4 (CD152)/RPE mAbs (BD Pharmingen, BD Biosciences, San Diego, USA) for 30 min at 4 °C in the dark. Excess unbound antibodies were removed by two washes with PBS. Following these washes, the cells were resuspended in PBS and analysed by flow cytometry using a FACSCalibur flow cytometer (Becton–Dickinson, BD Biosciences, San Diego, USA). For determination of intracellular CTLA-4 expression, the cells were first fixed for 10 min at room temperature in 2 % paraformaldehyde (Fluka, Sigma-Aldrich, Buchs, Germany), washed in PBS, and incubated for 10 min at room temperature in BD Permeabilizing Solution 2 (Becton–Dickinson, BD Biosciences, San Diego, USA) according to the manufacturer’s instructions. Then, the cells were incubated with anti-CTLA-4 (CD152)/RPE mAbs for 30 min at 37 °C in the dark.
For detection of cyclins D2 and D3, and p27
KIP1
protein, the cells were fixed, permeabilised, and stained with anti-cyclin D2/FITC, anti-p27KIP1/FITC mAbs (Santa Cruz Biotechnology, Inc, Heidelberg, Germany), and anti-cyclin D3/FITC mAb (BD Pharmingen, BD Biosciences, San Diego, USA) according to the manufacturer’s instructions.
Negative controls were always done by omitting the mAbs and by incubating the cells with mouse Ig of the same isotype as the mAbs conjugated with RPE or FITC. At least, 10,000 events per sample were analysed. The results were expressed as the proportion of CTLA-4-, cyclin D2-, cyclin D3-, or p27
KIP1
-positive cells. The CellQuest program was used for statistical analysis of the acquired data.
Statistical analysis
Statistical analyses of the clinical data and laboratory findings were conducted using Statistica 10.0 or PQStat software. For clinical parameters, the mean values and standard deviation were calculated. For all other analysed variables, the median values and 25th and 75th interquartile range were calculated. All collected data were examined for normal distribution and for homogeneity of variances using the Shapiro–Wilk test and Levene’s test, respectively. If data were normally distributed and had homogeneous variances, the comparisons between the studied groups were performed using the Student t test for independent samples. If data were not normally distributed and/or had heterogeneous variances, the nonparametric Mann–Whitney U test was used. To test the effects of culture and CTLA-4 blockade on analysed variables, the repeated measures ANOVA and the Student t test for dependent samples were used. If data were not normally distributed and/or had heterogeneous variances, the Friedman ANOVA test followed by a post hoc test (Dunn test), and the nonparametric Wilcoxon signed-rank test were applied. The relationship between the clinical parameters and CTLA-4 expression was tested with Spearman’ rank correlation coefficient. In all analyses, differences were considered significant when P ≤ 0.05.
Discussion
In the present study, we confirmed our earlier observations [
22] that both the surface and cytoplasmic expression of the CTLA-4 molecule is overexpressed in freshly isolated peripheral blood CLL cells. This finding raises a question about the mechanisms underlying the elevated levels of CTLA-4 expression in CLL cells. Since CTLA-4 expression is transiently inducible during T and B activation [
27,
28] and it is considered to be an activation marker, the higher expression of CTLA-4 in CLL cells may reflect the systemic activation status in the periphery in CLL patients. It has been well documented that CLL cells receive many growth-promoting signals [
29‐
31], which are mediated by receptors constitutively expressed on the subset of neoplastic B cells [
32,
33]. In addition, literature data confirm that CLL cells bear the phenotype of activated B cells based on overexpression of the classical activation markers CD23, CD25, CD69 and CD71, and under-expression of CD22, Fcγ IIb, CD79b, and immunoglobulin D, which are down-regulated by cell triggering and activation [
34]. The suggestion that CLL cells may be in a partial state of activation also results from the observation of cyclin D2 expression, which is induced during cell stimulation by growth factors in early G0 phase. Accordingly, in the current study, we confirmed our previous observations [
22,
35,
36] that the median proportion of freshly isolated CLL cells co-expressing cyclin D2 was markedly higher than that observed in normal B lymphocytes. Furthermore, based on the fact that
CTLA-
4 gene polymorphisms may influence the CTLA-4 expression level in CLL cells [
22,
37], we cannot exclude that the strong divergence of CTLA-4 expression observed among the patients might result, at least in part, from the presence of specific alleles predisposing to overexpression of the CTLA-4 molecule.
We also found a negative correlation between CTLA-4 expression in CLL cells and the clinical parameters such as Rai stage and leucocyte as well as lymphocyte count. This finding is consistent with the study of Joshi et al. [
38] showing that the level of
CTLA-
4 gene expression in CLL cells predicts clinical outcome of CLL patients; lower expression of CTLA-4 is associated with significantly shorter time to treatment and poor prognosis compared with high CTLA-4 expression associated with good clinical outcome.
In the present study, we showed for the first time the impact of ex vivo specific stimulation of CLL cells with DSP30 (CpG ODN)+rIL-2 on CTLA-4 expression. An assessment of CTLA-4 expression under this stimulation condition seems to have clinical relevance in the context of the discussion about using CpG ODNs as therapeutic agents in CLL patients [
39]. We found that ex vivo stimulation led to a significant decrease in the median proportion of CLL cells expressing the CTLA-4 molecule on their surface, which became comparable to the value of the corresponding cells in healthy individuals. In contrast, the intracellular expression of this protein in leukaemic cells markedly increased after 72 h of cell culture in the presence of DSP30 and rIL-2 and remained significantly higher than in healthy volunteers. This observation seems to suggest that the down-regulation of sCTLA-4 expression upon stimulation may result from disturbed recycling of the CTLA-4 molecule to the cell surface in CLL. Moreover, as higher expression of the
CTLA-
4 gene is associated with good clinical outcome [
38], application of CpG ODN that reduces the surface expression of CTLA-4 in leukaemic cells as a therapeutic agent may not be beneficial for CLL patients.
Literature data show that ex vivo stimulation with CpG ODN results in down-regulation of p27
KIP1
expression and up-regulated expression of cyclins D2 and D3 in CLL cells and normal B lymphocytes [
40‐
43]. In the present study, we also observed the down-regulation of p27
KIP1
expression following ex vivo stimulation in CLL patients as well as in healthy volunteers. Of note, in spite of the decrease in p27
KIP1
expression in both studied groups, the median frequencies of p27
KIP1
-positive cells in CLL patients remained significantly higher compared with healthy individuals at each time point tested. It is noteworthy that high expression of this protein is associated with poorer outcome of CLL [
44‐
46]. As regards cyclin D2, in contrast to others [
40‐
42], after 24 h of stimulating culture, we did not observe up-regulation of cyclin D2 expression in CLL cells. This discrepancy might result from the different experimental procedures used. We estimated the expression of these proteins as the proportions of CLL cells co-expressing several proteins using flow cytometry, whereas the authors of the contradictory reports examined the protein expression by western blot analysis. Due to cyclin D2 being overexpressed in freshly drawn CLL cells compared to normal B lymphocytes [
22,
35,
36,
47] and maintained at similar high levels in CLL cells after 24 h of ex vivo stimulation, the median percentage of cyclin D2-positive cells in CLL patients was comparable to the corresponding cells in healthy individuals, in which up-regulation of cyclin D2 expression was observed. Moreover, in agreement with literature data [
40‐
43], we found a significant increase in the median percentages of cyclin D3-positive cells after 24 h of ex vivo stimulation in CLL patients as well as in healthy volunteers. Of note, the median proportions of CLL cells co-expressing cyclin D3 remained markedly lower at each time point tested compared to corresponding cells in healthy individuals. To elucidate this phenomenon, further studies are necessary.
One of the main aims of this study was to examine whether the CTLA-4 molecule is able to influence the expression of the key regulators of G0/G1 phase in CLL cells as well as normal B lymphocytes. Our previous data [
22] indicated that expression of the CTLA-4 molecule in CLL cells correlated with the expression of the G0/G1 phase regulators cyclins D2 and D3, and p27
KIP1
, suggesting the impact of CTLA-4 on prolonging the G1 phase in leukaemic lymphocytes. In the present study, we examined the expression of the mentioned cell cycle regulators following blockade of CTLA-4 on the surface of CLL cells as well as normal B lymphocytes. Blocking CTLA-4 on the surface of the studied cells is a useful method in research to elucidate the function of this protein. We found that CTLA-4 blockade did affect the expression of studied regulators of G0/G1 phase. In particular, we observed for the first time a significant decrease in the median proportions of both cyclin D2- and cyclin D3-positive cells in CLL cells as well as in normal B lymphocytes following 24 h of blocking culture. Moreover, we noted that CTLA-4 blockade led to an increase in the median proportions of p27
KIP1
-positive cells after 24 h of culture in CLL cells as well as in normal B lymphocytes, although this increase was statistically significant only in the group of CLL patients. What are the possible mechanisms responsible for this phenomenon? It has been shown that the expression of D-type cyclins and p27
KIP1
protein stability is controlled by PI3 K (phosphatidylinositol 3-kinase)-related pathways [
48,
49]. Activation of PI3 K-mediated pathways leads to transcription of D-type cyclins and to reduction of the p27
KIP1
protein level [
48,
49]. In some experimental systems, it has been shown that CTLA-4 engagement on T cells activates PI3K [
50]. As CTLA-4 blockade has an opposite effect, it seems highly probable that CTLA-4 blockade may lead to inactivation of PI3K-related pathways. Based on this hypothesis, we can speculate that blockade of CTLA-4 on CLL cells as well as on normal B lymphocytes may lead to inactivation of PI3K-related pathways, resulting in decreased expression of cyclins D2 and D3, and accumulation of p27
KIP1
protein, in consequence. Further studies are required to confirm a direct relationship between CTLA-4, PI3K-related pathways, and the expression of the mentioned cell cycle regulators of G0/G1 phase.
Furthermore, an interesting finding was that the median proportions of cyclin D2-positive cells in CLL patients and healthy volunteers were comparable following CTLA-4 blockade, whereas the median percentages of cyclin D3+ and of p27
KIP1+ leukaemic cells remained lower and higher, respectively, than in healthy individuals. Further investigations are needed to explain this phenomenon.
Extending our previous study, we have demonstrated by blocking experiments that the CTLA-4 molecule influences the expression of the key regulators of G0/G1 phase—cyclins D2 and D3, and p27
KIP1
—in both malignant and normal B lymphocytes. We also confirmed the recent observation that CTLA-4 expression in peripheral blood CLL cells negatively correlates with disease progression, suggesting its clinical relevance as a prognostic marker. Since CTLA-4, cyclin D2 and p27
KIP1
seem to have prognostic significance in CLL, further studies are necessary to elucidate the molecular mechanisms of the relationship between these proteins.