Deletions in chromosome bands 11q22-q23 were recently shown to be one of the most frequent chromosome aberrations in B-cell chronic lymphocytic leukemia (B-CLL). Patients suffering from B-CLL with 11q deletion are characterized by extensive lymphadenopathy, rapid disease progression, and short survival times. Phenotypic and functional characteristics of B-CLL cells with 11q deletion that may help to explain the pathophysiology of this entity are yet unknown. In the present study, B-CLL cells with (n = 19) and without (n = 19) 11q deletion were analyzed for their expression of functionally relevant cell surface molecules (n = 57). B-CLL cells with 11q deletion carried significantly lower levels of the adhesion molecules CD11a/CD18 (integrin L/β2), CD11c/CD18 (integrin X/β2), CD31 (PECAM-1), CD48, and CD58 (LFA-3). Furthermore, B-CLL cells with 11q deletion expressed less the cell signaling receptors CD45 (leukocyte common antigen [LCA]), CD6, CD35 (complement receptor 1), and CD39. Reduced CD45 levels and low-level expression of CD49d correlated with decreased overall survival. B-CLL cells with or without 11q deletion did not differ in their growth fractions, expression levels of transcription factor NF-κB, or their response to mitogenic stimuli. Decreased levels of functionally relevant adhesion molecules and of cell signaling receptors may contribute to the pathogenesis of the subgroup of B-CLL characterized by 11q22-q23 deletion.
DELETIONS IN chromosome bands 11q22-q23 were recently shown to be one of the most frequent chromosome aberrations in B-cell chronic lymphocytic leukemia (B-CLL).1,2 In an interphase cytogenetic study of 214 patients with B-CLL, such 11q deletions were found in 20% of the cases and were the second most common chromosome aberration after 13q14 deletions.2 Patients with 11q deletion had a characteristic clinical picture. These patients were younger than other B-CLL patients, they had extensive lymphadenopathy, and they suffered from B-symptoms more frequently. Importantly, the presence of 11q deletion was an independent prognostic factor predicting rapid disease progression and short survival times. The critical region of these deletions has been delineated to a 2- to 3-Mb sized genomic segment in bands 11q22.3-q23.1.3 This genomic region likely contains a novel tumor-suppressor gene that is important for the development and progression of this clinically relevant subgroup of B-CLL. Known candidate genes within this region include radixin (RDX), which has homology to the neurofibromatosis-type 2 (NF2) tumor-suppressor gene,4 and the ataxia telangiectasia mutated (ATM) gene.5 Evidence that the ATM gene functions as a tumor-suppressor gene comes from murine knock-out models and from the recent observation of biallelic mutations of the gene in T-cell prolymphocytic leukemia.6-8
The characteristic extensive lymphadenopathy of B-CLL cases with 11q deletions points towards functional aberrations of these leukemic cells. Massive infiltration of secondary lymphoid organs likely involves the action of multiple adhesion molecules regulating homophilic and heterophilic binding processes. It has recently been shown that B-CLL cells posses several different adhesion pathways and these appear to vary with the stage of the disease.9 We therefore investigated whether B-CLL cells with or without 11q deletion differ in their expression pattern of functionally relevant adhesion molecules and of other cell signaling receptors.
The present study demonstrates that B-CLL cells with 11q deletion express significantly lower levels of several adhesion molecules and proteins regulating relevant cell functions.
MATERIALS AND METHODS
Fluorescein isothiocyanate (FITC)-conjugated monoclonal antibodies (MoAbs) against CD5 (BL1a), CD11a (integrin αL, LFA-1; 25.3.1), CD21 (CR2, EBVR; BL13), CD23 (9P.25), CD29 (integrin β1; K20), CD49d (integrin α4, VLAα4; HP2/1), CD54 (ICAM-1; 84H10), CD80 (B7.1; MAB104), FMC7 (FMC7), phycoerythrin (PE)-conjugated anti-CD19 (J4.119), and PE-conjugated mouse IgG1 (isotype control; 679.1Mc7) were purchased from Immunotech (Hamburg, Germany). Unconjugated MoAbs against the cell surface markers CD6 (SPVL 14), CD10 (CALLA; ALB1), CD11b (integrin αM, MAC-1; BEAR1), CD11c (integrin αX, p150, 95; BU15), CD18 (integrin β2; 7E4), CD20 (B9E9), CD22 (SJ.10.1H11), CD24 (ALB9), CD26 (BA5), CD27 (LT27), CD30 (HRS-4), CD31 (PECAM-1; 5.6E), CD32 (2E1), CD35 (CR1; J3.D3), CD37 (BL14), CD38 (T16), CD39 (AC2), CD40 (MAB89), CD40L (TRAP1), CD44 (J-173), CD46 (J4-48), CD48 (J4-57), CD49b (integrin α2, VLAα2; Gi9), CD49c (integrin α3, VLAα3; M-KID2), CD49e (integrin α5, VLAα5; SAM), CD49f (integrin α6, VLAα6; GoH3), CD50 (ICAM-3; HP2/19), CD51 (integrin αV; AMF7), CD58 (LFA-3; AICD58), CD61 (integrin β3; SZ.21), CD62L (L-selectin; DREG65), CD69 (TP1/55.3.1), CD70 (HNE51), CD71 (YDJ.1.2.2.), CD72 (J3.109), CD77 (38-13), CD79b (CB3-1), CD81 (TAPA-1; JS64), CD95 (APO-1, Fas; CH11), CD102 (ICAM-2; B-T1), CD103 (HML-1; 2G5.1), surface IgM (sIgM; AF6), and surface IgD (sIgD; JA11) were acquired from Immunotech. Anti-CD86 [2331 (FUN-1)], anti-κ (G20-193), and anti-λ (JDC-12) were obtained from Pharmingen (San Diego, CA). The isotype control mouse IgM (R4A3-22-12) was purchased from Coulter Clone (Coulter Corp, Miami, FL). Anti-CD43 (DF-T1) and anti-CD45 [LCA; T29/33.(1)] were obtained from Dako (Glostrup, Denmark). FITC-conjugated rabbit-antimouse F(ab′)2 fragments (Dako) served as the secondary antibody for unconjugated MoAbs. Contaminating T cells, monocytes, and natural killer (NK) cells were quantified using MoAbs against CD3 (FITC-conjugated; UCHT-1), CD14 (RMO52), and CD56 (T199), respectively (all purchased from Immunotech). Expression of the nuclear proliferation antigen Ki-67 was determined using a FITC-conjugated MoAb (MIB-1) from Dako. IgG1-FITC (Pharmingen) served as isotype control.
Proper function of the MoAbs was verified using defined cell preparations as positive controls for each antibody used (Fig 1): Burkitt lymphoma cell line Daudi (CD10, CD24, CD37, CD71, CD79b, CD102, surface IgM, and κ light chain); Burkitt lymphoma cell line Raji (CD20, CD21, CD22, CD40, CD45, CD54, CD80, CD81, CD86, and Ki-67); CLL cell line EHEB (obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany; CD23, CD30, CD39, CD43, CD44, CD46, CD48, CD50, CD58, CD70, CD77, and FMC7); normal activated peripheral blood T lymphocytes at day 14 of culture with 300 U/mL interleukin-2 (IL-210; CD5, CD6, CD11a, CD18, CD26, CD27, CD29, CD38, CD40L, CD49b, CD49d, CD49e, CD69, CD95, and CD103); normal platelets (CD51 and CD61), normal granulocytes (CD62L), and normal monocytes (CD11b, CD11c, CD31, and CD32) were analyzed in peripheral blood by setting the gates according to the characteristic pattern of these cells in forward and light scatter analysis; normal peripheral blood B lymphocytes (CD35, CD72, and surface IgD) were identified using a PE-conjugated anti-CD19 MoAb; leukemic cells expressing lambda light chain were isolated from a patient with prolymphocytic leukemia (λ light chain); a greater than 97% cytokeratin-positive primary colon carcinoma cell line (CD49c and anti-CD49f) was generously donated by B. Trojaneck (Charité, Berlin, Germany).
To compare the survival times of patients who differ in the CD45 and CD49d intensities on the B-CLL cells, discerning levels for these antigens were defined. Cut-off level of CD49d was the mean fluorescence that discriminated negative from positive antigen expression (mean fluorescence of 2.8; Fig 1). Cut-off level of CD45 was defined as the mean fluorescence that discerned patients who were alive from those who died during follow-up (mean fluorescence of 400; Fig 2A). Thus, B-CLL cells could be defined as CD49dlow-positive (>2.9 mean fluorescence) and CD49dnegative (≤2.9 mean fluorescence), CD45high-positive (>400 mean fluorescence), and CD45low-positive (≤400 mean fluorescence).
Patients and cells.
Mononuclear cells from 38 patients with B-CLL were analyzed. At the time of analysis for the present study, 5 patients had Rai stage 0, 4 stage 1, 20 stage 2, 3 stage 3, and 6 stage 4 disease. Diagnosis of B-CLL was confirmed by flow cytometric assessment of coexpression of CD5, CD19, CD23, and surface Ig (low) on the B-CLL cells. Interphase cytogenetic analysis was performed as described previously.2 Nineteen cases had 11q deletion, whereas the remaining 19 cases had a normal karyotype or other chromosome aberrations. As of June 1998, of the 38 patients whose B-CLL cells were examined, 25 patients are alive and 12 patients have died. No survival data of 1 patient could be obtained due to loss to follow-up. Mean follow-up was 74 months (range, 12 to 216 months).
Expression of cell surface antigens (n = 57) by B-CLL cells was determined using a FACScan (Becton Dickinson, Heidelberg, Germany). B-CLL cells were specifically detected and phenotyped by examining coexpression of CD19 and the antigen under investigation with two-color flow cytometry.
Detection of nuclear proliferation antigen Ki-67.
For detection of Ki-67, B-CLL cells were fixed at room temperature using Reagent A (Fix & Perm Cell Permeabilisation Kit; An der Grub GmbH, Kaumberg, Austria) and fixed at 4°C in precooled absolute methanol. Cells were washed once in cold phosphate-buffered saline (PBS), permeabilized with Reagent B (Fix & Perm Cell Permeabilisation Kit), and incubated with FITC-labeled MIB-1 MoAbs or the isotype control, respectively. Cells were washed in cold PBS and analyzed for expression of Ki-67 by flow cytometry.
Analysis of DNA content.
DNA content of B-CLL cells was determined by fixing the cells for 2 hours at −20°C in cold 75% ethanol. Raji cells were used as positive controls. Cells were washed in PBS (0.1% NaN3, 1% heat-inactivated fetal calf serum [FCS; GIBCO BRL, Eggenstein, Germany]) and treated with cold 0.25% Triton X-100 in PBS for 5 minutes on ice. Subsequently, cells were resuspended in PBS (10 μg/mL propidium iodide [PI], 0.1% RNase A), incubated at 4°C for 20 minutes, and analyzed by flow cytometry.
Electrophoretic mobility shift assays.
Activity of transcription factor NF-κB in B-CLL cells was determined by a previously described assay using a 32P-labeled oligonucleotide probe containing a high-affinity NF-κB binding site.11,12 The identity of the specific NF-κB-DNA complex in this assay has been previously determined both by antibody supershift and by competition assay.11,12 Jurkat cells expressing high levels of active NF-κB were prepared as positive controls by treatment for 15 minutes with 200 U/mL tumor necrosis factor-α (TNF-α). Briefly, total cell extracts were prepared using a high-salt detergent buffer (Totex; 20 mmol/L HEPES, pH 7.9, 350 mmol/L NaCl, 20% [wt/vol] glycerol, 1% [wt/vol] NP-40, 1 mmol/L MgCl2, 0.5 mmol/L EDTA, 0.1 mmol/L EGTA, 0.5 mmol/L dithiothreitol [DTT], 0.1% phenylmethyl sulfonyl fluoride [PMSF], and 1% aprotinin). Cells were harvested by centrifugation, washed once in ice-cold PBS (Sigma, Deisendorf, Germany) and resuspended in four cell volumes of Totex buffer. The cell lysate was incubated on ice for 30 minutes and then centrifuged for 5 minutes at 13,000g at 4°C. The protein content of the supernatant was determined and equal amounts of protein (10 to 20 μg) were added to a reaction mixture containing 20 μg bovine serum albumin (BSA; Sigma), 2 μg poly(dI-dC) (Boehringer Mannheim, Mannheim, Germany), 2 μL buffer D+ (20 mmol/L HEPES, pH 7.9, 20% glycerin, 100 mmol/L KCl, 0.5 mmol/L EDTA, 0.25% NP-40, 2 mmol/L DTT, 0.1% PMSF), 4 μL buffer F (20% Ficoll 400, 100 mmol/L HEPES, 300 mmol/L KCl, 10 mmol/L DTT, 0.1% PMSF), and 100,000 cpm (Cerenkov) of a32P-labeled oligonucleotide in a final volume of 20 μL. Samples were incubated at room temperature for 25 minutes. NF-κB oligonucleotides (Promega, Madison, WI) were labeled using γ-[32P]-ATP (3,000 Ci/mmol; Amersham, Arlington Heights, IL) and T4 polynucleotide kinase (New England Biolabs, Beverly, MA). The film was scanned and the relative amounts of the NF-κB DNA complexes were quantified using the NIH-Image software. Based on these analyses, a negative or a positive score for NF-κB-activity could be assigned to each B-CLL sample.
B-CLL cells (2 × 105/well), with (n = 9) and without (n = 10) 11q deletion, were cultured in 200 μL RPMI 1640 medium (10% FCS) in a 96-well plate. Cells were stimulated with 50 ng/mL phorbol myristate acetate (PMA; Sigma), 200 U/mL recombinant IL-2 (rIL-2; PromoCell GmbH, Heidelberg, Germany), 10 mg/mL antihuman IgM F(ab)2 fragments (Dako), or antihuman IgM F(ab)2 fragments in combination with rIL-2. B-CLL cells stimulated with PMA served as positive control and cells cultured in medium witout mitogens were used as negative control. B-CLL cells were incubated in triplicate for 48 hours at 37°C. Response to the stimuli was quantified using a colorimetric proliferation assay (EZ4U; Biomedica GmbH, Wien, Austria) based on MTT (tetrazolium13). This assay was performed according to the instructions of the manufacturer. Absorbance was measured at 450 nm using a MRX microplate reader (Dynatech, Denkendorf, Germany).
Significance of the differences of the mean fluorescence between B-CLL cells was determined using the Mann-Whitney-U test. Overall survival was measured from the time of diagnosis until death from any cause. Significance of the difference of the overall survival of patients was determined using the log-rank test. P values less than .05 were considered significant.
Differential expression pattern of cell surface antigens by B-CLL cells with or without 11q deletion.
To define the expression pattern of functionally relevant cell surface antigens on B-CLL cells, two-color flow cytometry was performed with a large panel of MoAbs (n = 57) and an anti-CD19 MoAb. Differential staining results obtained with each antibody were validated using defined positive control cell preparations. The mean fluorescence of each marker was determined as a measurement of the relative amount of the antigen on the cell surface (Fig 1). All cells had an antigen pattern characteristic for B-CLL cells, ie, they coexpressed CD19, CD5, and CD23 and had low levels of cell surface Igs. Markers could be divided into groups of high (++), low (+), and no (−) expression on B-CLL cells (Fig 1). B-CLL cells were most intensively stained by MoAbs against the B-cell antigens CD20 and CD24 and by antibodies against CD37, CD43, CD44, CD45, and CD50. In contrast, a panel of functionally important molecules, such as CD40L, B7.1 (CD80), B7.2 (CD86), and Fas/Apo-1 (CD95), was not found on B-CLL cells (Fig 1). B-CLL cells of most patients expressed the integrins α3 (CD49c) and α5 (CD49e) but lacked the corresponding integrin β1 chain (CD29).
Among all 57 antigens examined, 12 markers showed a differential expression pattern on B-CLL cells with or without 11q deletion (Fig 1). CD6, the integrins αL (CD11a), αX (CD11c), and β2 (CD18), PECAM-1 (CD31), the complement receptor 1 (CD35), CD39, the leukocyte common antigen (CD45), CD48, and LFA-3 (CD58) were expressed at significantly lower levels on 11q-deleted B-CLL cells than on B-CLL cells that lacked a 11q deletion. As shown in Table 1, the differences of the expression levels of these 12 antigens were significant. CD62L and CD71 were also differentially expressed but were found only on a few of the B-CLL cases examined.
|Antigen .||P Value .||n* .||Antigen .||P Value .||n* .|
|Antigen .||P Value .||n* .||Antigen .||P Value .||n* .|
Abbreviation: NS, not significant (P ≥ .05).
Total number of cases examined. Equal numbers of B-CLL cases with/without 11q deletion were investigated.
CD62L was expressed in 1 of 9 cases with and in 3 of 9 cases without 11q deletion.
CD71 was expressed in 2 of 9 cases with and in 5 of 9 cases without 11q deletion.
Flow cytometric analyses also showed morphological differences between B-CLL cells with and without 11q deletion. As determined by the forward scatter (FSC), B-CLL cells with 11q deletion were on the average 11% smaller than B-CLL cells without 11q deletion (P < .05).
Correlation of levels of antigen expression with survival of the patients.
To determine the possible relevance of the differential antigen expression of B-CLL cells for the clinical course of the disease, the overall survival times of the patients were determined. Patients with 11q deletion had a significantly reduced survival compared with patients without 11q deletion (P = .0038; data not shown). Among all antigens examined, expression levels of CD45 and CD49d correlated significantly with overall survival of the patients (Fig 2A and B). B-CLL cells of patients who died during follow-up had lower levels of CD45 (P = .017) and higher levels of CD49d (P= .0051) than the leukemic cells of patients who were still alive (Fig2A). When only cases with 11q deletion were analyzed, CD45 levels and survival no longer correlated, whereas CD49d was still higher expressed on cells of patients who died (P = .0055; Fig 2A).
To further evaluate the clinical relevance of the CD45 and CD49d levels, all patients were divided into groups depending on the expression intensity of these antigens on the B-CLL cells (see also Materials and Methods). As shown in Fig 2B, patients with CD45low-positive B-CLL cells had a significantly reduced overall survival compared with patients with CD45high-positive leukemic cells (P = .032). Furthermore, patients with CD49dlow-positive B-CLL cells had shorter survival times than patients with CD49dnegative B-CLL cells (P = .0012).
Growth fraction, cell cycle position, and NF-κB expression of B-CLL cells with or without 11q deletion.
B-CLL cells with or without 11q deletion did not differ in their growth fraction or cell cycle position. A total of 0.7% of B-CLL cells with (n = 9) and 1.3% of B-CLL cells without 11q deletion (n = 8) expressed the nuclear proliferation antigen Ki-67. Accordingly, 98.9% of B-CLL cells with and 98.5% without 11q deletion were found to be in G0/G1-phase of the cell cycle. In contrast, of the Raji cells serving as positive control, 97% were positive for Ki-67, 50% in G0/G1-phase, 30% in S-phase, and 20% in G2/M-phase of the cell cycle.
We further investigated whether 11q deletion in B-CLL cells was correlated with a differential expression of the transcription factor NF-κB. NF-κB is a central mediator of the human immune response regulating the expression of various immune-modulatory molecules, among them adhesion receptors such as intercellular adhesion molecule (ICAM) 1, vascular adhesion molecule (VCAM) 1, and endothelial adhesion molecule (ELAM).13 Using a 32P-labeled oligonucleotide probe containing a high-affinity NF-κB binding site, B-CLL cells of different patients were shown to contain various expression levels of NF-κB (Fig 3). However, no correlation was observed between the level of NF-κB-expression and the presence or absence of a 11q deletion.
Stimulatory response of B-CLL cells with or without 11q deletion.
We investigated the response of B-CLL cells with or without 11q deletion to various mitogenic stimuli (Fig4). All B-CLL cells were stimulated most strongly by PMA, whereas anti-IgM alone or in combination with rIL-2 exerted an intermediate stimulation of the B-CLL cells. rIL-2 alone resulted in a low-grade stimulation. As shown in Fig 4, no significant difference was noted in the cellular response of B-CLL cells with or without 11q deletion to these mitogenic substances.
Recent cytogenetic studies have suggested that deletions in chromosome bands 11q22-q23 identify a new subset of B-CLL, characterized by extensive lymph node involvement, rapid disease progression, and short survival times.2,3 The present study demonstrates a differential expression pattern of functionally relevant adhesion molecules and cell signaling receptors on B-CLL cells with or without 11q deletion.
Coexpression of CD19, CD5, CD23, and cell surface Igs confirmed that all leukemic cells were truly B-CLL cells. The finding that the integrins αL/β2 (CD11a/CD18) and αX/β2 (CD11c/CD18) are relatively weakly expressed by all B-CLL cells is consistent with earlier reports.14-16 Furthermore, expression of CD43,17 CD44,15,16 and CD5018 and lack of CD51 and CD61 expression19by B-CLL cells is in agreement with published findings. Therefore, results obtained by the technique used here to analyze antigen expression levels correlate well with findings of other investigators.
The underlying cause for the observation that B-CLL cells expressed CD49c and CD49e (integrin α3 and α5, respectively) but lacked the corresponding CD29 (integrin β1 chain) is unclear. Yet, this finding is consistent with earlier reports. Overexpression of integrin-α-chains contrasted to low or negative expression levels of integrin-β1-chains in CLL, whereas acute lymphocytic leukemia and multiple myeloma cells had a balanced expression pattern of these adhesion molecules.20 Furthermore, lymphoid cells within the mantle zone of the lymphoid follicle that may represent normal counterparts of B-CLL cells were found to have increased CD49d levels and decreased expression of CD29.21 As suggested by Möller et al,20 these findings may indicate that on these lymphoid cells integrin α4 chains may associate with other β-integrins than β1-integrins.
Phenotypic analyses confirmed the working hypothesis of the present study that B-CLL cells with or without 11q deletion differentially express functionally relevant cell surface molecules. All of these markers CD6, CD11a, CD11c, CD18, CD31, CD35, CD39, CD45, CD48, and CD58 were found at lower levels on cells with 11q deletion than on leukemic cells that lacked this chromosomal aberration. These antigens exert a broad variety of cellular functions. Yet, a property common to all of these molecules—except for the complement receptor CD3522—is that they are able to mediate cell adhesion.
CD6 belongs, together with CD5, to the scavenger-receptor-cystein-rich (SRCR) super family23,24 and acts as a receptor for the activated leukocyte cell adhesion molecule (ALCAM). Binding via CD6 appears to influence the fate of B-CLL cells, because ligation of CD6 protects these cells from anti-IgM–induced apoptosis by increasing the Bcl-2/Bax ratio.25
The integrins αL/β2 (CD11a/CD18) and αX/β2 (CD11c/CD18) are essential for mediating leukocyte migration.26,27 The importance of these adhesion molecules for the clinical course of the B-CLL disease has not been clarified unambiguously. Lack of integrin β2 chains was shown to be associated with more favorable clinical features.28,29Consistently, expression of integrin β215 and integrin αX30 correlated with more advanced disease stages and with a diffuse bone marrow infiltration, respectively. In contrast, other publications demonstrated that low levels of integrin β216,31 and integrin αX32 on B-CLL cells were associated with higher mortality. Data presented in the current study are in agreement with these latter reports. We detected lower expression levels of integrin αL/β2 within the group of B-CLL cells with 11q deletion that previously was shown to correlate with extensive lymphadenopathy and early disease progression.2
PECAM-1 (CD31) is a broadly expressed adhesion molecule mediating homophilic and heterophilic intercellular binding.33,34Similarly, CD39 is an activation-dependent molecule that mediates rapid integrin αL/β2-dependent and -independent homotypic adhesion.35,36 So far, the relevance of PECAM-1 and CD39 for the clinical presentation of a B-CLL has not been determined.
It was interesting to note that two ligands for the T-cell adhesion molecule CD2, ie, CD48 and CD58, were expressed at lower levels on B-CLL cells with 11q deletion. These receptors are essential for inducing antigen-dependent and antigen-independent immune and inflammatory T-cell responses.26,27
In addition, CD45, the transmembrane protein tyrosine phosphatase expressed on nucleated hematopoietic cells, was reported to initiate cell adhesion. Triggering via CD45 induces an integrin-αLβ2/ICAM-1– and -2–mediated cell aggregation.37 This homotypic adhesion leads also to a coclustering of CD45 and integrin αLβ2.37
It is not clear whether the decreased expression levels of adhesion molecules on B-CLL cells with 11q deletion can help to explain the pathophysiology of this subentity. Possibly, the altered antigen pattern of B-CLL cells with 11q deletion could influence the migratory properties of these cells within lymphoid organs and in peripheral blood. This might participate in the development of the increased lymphadenopathy that is typical for B-CLL cases with 11q deletion.
In addition, the differential expression of antigens by 11q-deleted B-CLL cells may impair the cellular immune defense, because some of these molecules are relevant for the recognition of target cells by cytolytic T cells.26 Downregulation of CD58 by T-cell leukemia38,39 and Burkitt lymphoma cells40 has been correlated with unsusceptibility to killing by cytotoxic lymphocytes. B-CLL cells with or without 11q deletion did not express B7.1 (CD80), B7.2 (CD86), and Fas/Apo-1 (CD95). B7.1 and B7.2 are costimulatory molecules essential for inducing a T-cell response and Fas/Apo-1 is an important receptor for cytotoxic T cells leading to programmed cell death (apoptosis) of the target cells.41,42In addition, even in the presence of B7-1 or B7-2 on tumor cells, CD48 appears to be required for the generation of T-cell–specific antitumor immunity.43 Taken together, absence or decreased expression of molecules mediating T-cell cytolysis on B-CLL cells with 11q deletion might contribute to the impairment of the cellular immune defense against these leukemic cells.
Our observation that patients with 11q deletion had a reduced overall survival has been previously reported.2 Data of the present study support the notion that differential expression levels of CD45 and CD49d on B-CLL cells may also influence the clinical outcome of the disease. Patients with CD45low-positive or CD49dlow-positive B-CLL cells had a significantly reduced overall survival compared with patients with CD45high-positive and CD49dnegative leukemic cells, respectively. When only B-CLL cases with 11q deletion were analyzed, levels of CD49d but not those of CD45 were still relevant for poor survival. Most likely, lower CD45 levels correlated with reduced survival of all patients examined, because this antigen alteration was the most significant marker of B-CLL cells with 11q deletion. Therefore, reduction of CD45 levels does not appear to be a marker for poor survival independently of 11q deletion. In contrast, differential expression of CD49d further divided the B-CLL cases with 11q deletion into groups with differing overall survival. Thus, low-level CD49d expression may represent a factor of poor survival that is independent of 11q deletion. Support for the latter interpretation comes from the observation that expression of CD49d correlated with advanced disease stages.9 In addition, expression of β1, β2, and β3 integrins on B-CLL cells was shown to be associated with poor prognosis and with splenomegaly.28
We did not observe significant differences in the growth fraction, cell cycle position, expression of the transcription factor NF-κB, or stimulatory response of B-CLL cells with or without 11q deletion. The current study sets the basis for further investigations aiming to define which in vitro detectable function of B-CLL cells with 11q deletion is altered and is responsible for the poor clinical outcome of this entity.
It is yet unclear if there is a unifying cause for the decreased expression levels of the functionally diverse receptors on B-CLL cells. Conceivably, loss of a putative tumor-suppressor gene by the 11q deletion could represent a common defect impairing several cellular pathways in B-CLL cells. Further identification of these cell signaling aberrations in B-CLL cells with 11q deletion might help to explain the pathogenesis of this clinically relevant B-CLL subentity.
Supported by grants from the Wilhelm Sander Stiftung (94.042.1), the Tumorzentrum Heidelberg/Mannheim (I/I.1), the Deutsche Forschungsgemeinschaft (DFG Pa 611/1-2 and SFB 364 project C2), and the Deutsche Krebshilfe (10-0917-DöI). F.S. is supported by a research fellowship of the DFG (Schr 318/3-1).
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. section 1734 solely to indicate this fact.
Address reprint requests to Folke Schriever, MD, Virchow-University Hospital, Department of Hematology and Oncology, Augustenburger Platz 1, 13353 Berlin, Germany; e-mail: email@example.com.