Regulation of CD23 expression by Notch2 in B-cell chronic lymphocytic leukemia.

The original observation that sera from patients with chronic B-cell lymphocytic leukemia (B-CLL) contain high amounts of soluble CD23 (sCD23), which reflect disease activity and tumor load has been confirmed by numerous reports and serial determinations of sCD23 are now recognized as important indicators of disease progression. The reason why the leukemic cells over express CD23 and subsequently release large quantities of sCD23 as compared to healthy persons or patients with other lymphoproliferative disorders is still not clear. However, progress has been made in understanding the mechanism leading to the upregulation of CD23 in the leukemic cells. Following is an update on clinical data and a short review on the potential functions of CD23 as well as its regulation by Notch2 in B-CLL.

Induction of apoptosis by proteasome inhibitors in B-CLL cells is associated with downregulation of CD23 and inactivation of Notch2.

Recently, proteasome inhibitors (PI) have attracted interest as novel anticancer agents in B-cell chronic lymphocytic leukemia (B-CLL). A prominent feature of B-CLL cells is the high expression of CD23, which is closely related to cell survival and is regulated by Notch2. Since several components of the Notch signaling cascade are tightly regulated by proteasomal degradation, we studied the effect of PI on Notch2 activity and CD23 expression. Exposure of B-CLL cells to PI led to induction of apoptosis, a time- and dose-dependent downregulation of CD23 expression and a decline in DNA binding of transcriptionally active Notch2. In contrast, the transcription factor NF-AT and its putative target gene CD5, which is highly expressed in B-CLL cells, were unaffected. When the late phase of PI-induced apoptosis was arrested by inhibition of caspase 3, the reduction of Notch2 activity was still observed, indicating that reduction of active Notch2 took place already during an earlier phase of apoptosis. Enforced expression of constitutively active Notch2 decreased PI-mediated apoptosis in a human B-cell line. These data indicate that downregulation of CD23 and loss of Notch2 activity are early steps in PI-induced apoptosis of B-CLL lymphocytes and may be part of the full apoptotic response.

Selection of autologous CD4+ T-cells for adoptive T-cell substitution in patients with CD23+ B-cell CLL.

BACKGROUND: B-cell CLL (B-CLL) is accompanied by a progressive decrease in cellular immune functions, and treatment-related immunosuppression can further aggravate T-cell immunodeficiency. To reduce the risks of T-cell depletion, it seems feasible to collect autologous CD4+ cells at an early disease stage and subsequently reinfuse them during periods of profound T-cell depletion. METHOD: We describe a two-step cell-selection method to obtain highly enriched CD4+ T-cells from leukapheresis compounds of patients with CD23+ B-CLL. The double selection procedure was performed using the CellPro Ceperate device, and consisted of CD4+ selection followed by CD23 purging to further remove contaminating CD23+ B-cells from the CD4+ cell fraction. The results of eight runs performed with leukapheresis material obtained from eight patients with CD23+ B-CLL at different disease stages are presented. RESULTS: The CD4/CD23 double cell-selection procedure resulted in the purification of > 90% CD4+ cells. Median recovery of CD4+ T lymphocytes after selection was 46%, and was negatively affected by the initial tumor cell load. The final T-cell fraction still contained lymphocytes of the B-CLL clone, as determined by FACS and PCR. The cell-processing procedure had no impact on T-cell function, as assessed by the in vitro production of the cytokine interferon-gamma. Moreover, the selected CD4+ cells retained their capacity to co-stimulate mitogen-induced B-cell IgG production in vitro. CONCLUSION: The described CD4 selection/CD23 depletion strategy is a suitable approach to obtaining high numbers of functional active autologous CD4+ T cells for adoptive T-cell transfer in patients with CD23+ BCLL.

Reconstitution of endogenous interferon a by recombinant interferon in hairy cell leukemia.

Recombinant human IFN alpha (rhIFN-alpha) plays an important role in the treatment of hairy cell leukemia (HCL). However, the mechanisms leading to its beneficial effect are not completely clarified, and there is no information on IFN-alpha gene expression in this disease. Therefore, we investigated the pattern of IFN-alpha gene expression and protein production in HCL and their potential regulation by rhIFN-alpha. Blood samples from 10 patients with HCL and 8 healthy donors (HD) were investigated. Expression of IFN-alpha mRNA was assessed by reverse transcription-PCR analysis in peripheral blood mononuclear cells (PBMCs) under basal conditions and on induction with rhIFN-alpha and polyionosinic-polycytidylic acid [poly(I.C)]. IFN-alpha concentrations in plasma and culture supernatants were measured by immunoassays, and intracellular IFN-alpha was evaluated by fluorescence-activated cell sorting analysis. Results showed that, in contrast to blood samples from HDs, freshly isolated PBMCs from un treated HCL patients did not express IFN-alpha mRNA, whereas IFN-alpha transcripts were found in patients who were under rhIFN-alpha therapy Plasma of untreated patients contained no, or extremely low levels of IFN-alpha as compared with plasma of treated patients and HDs. Ex vivo treatment of PBMCs with rhIFN-alpha or poly(I.C) resulted in a remarkable up-regulation of IFN-alpha at the mRNA and protein level. In HCL, however the amounts of IFN-alpha protein remained less than in HD. Inhibition of IFN-alpha transcription was found after exposure of PBMCs to serum fron untreated patients. Finally, a reduced capacity to produce IFN-alpha was found within B- cell, T-cell, and monocyte compartments in HCL patients which could be enhanced by rhIFN-alpha. The results demonstrate the ability, of rhIFN-alpha to up-regulate the expression of IFN-alpha gene and protein production and suggest that priming the production of endogenous IFN-alpha is a critical step in the mechanism of action of rhIFN-alpha in HCL.

Basic fibroblast growth factor is expressed by CD19/CD11c-positive cells in hairy cell leukemia.

Several features are characteristic for hairy cell leukemia (HCL). Among those are pancytopenia, bone marrow fibrosis, and the appearance of a defined tumor cell phenotype in peripheral blood (PB), bone marrow (BM), and spleen. Hairy cells (HC) coexpress antigens specific for B lymphocytes and monocytes/macrophages and thus the malignant cell does not seem to be restricted to a defined lineage. When serum or bone marrow aspirate was screened by enzyme-linked immunosorbent assay (ELISA) for basic fibroblast growth factor (bFGF), specimen derived from HCL (serum: mean value, 29 pg/mL; BM aspirate: mean value, 641 pg/mL) contained significantly higher levels than those from healthy subjects. To study whether peripheral blood mononuclear cells (PBMC) derived from patients suffering from HCL and healthy donors (HD) were capable of producing bFGF, culture supernatant (conditioned medium, [CM]) was tested for the presence of this cytokine. While bFGF was not detectable in cell cultures from HD, HCL-derived CM contained relatively high levels of bFGF. CM was successfully used for stimulation of mesenchymal cell proliferation, which could be inhibited by a neutralizing anti-bFGF antibody. Cellular activation by pokeweed mitogen (PWM) or the combination of 12-o-tetradecanoyl-phorbol-13-acetate (TPA) plus calcium ionophore (Ca-Ip) led to an enhanced mRNA expression. Results of Western blot experiments showed that HC synthesize at least three isoforms (approximately 18, 23, and 25 kD), but only the 23-kD isoform is exported. To assess the nature of the producer cell, double immunofluorescence analysis using a bFGF-specific and an anti-CD11c monoclonal antibody (MoAb) was undertaken. The majority of cells scoring positive for CD11c were also reactive with the anti-bFGF MoAb. Furthermore, enrichment of CD19/CD11c-positive cells correlated with enhanced bFGF levels, thereby supporting the argument for HC being the producer cells of bFGF. A biological function of bFGF in HCL might be mediation of chemoresistance, as 2-chlorodeoxyadenosine (2-CdA)-induced inhibition of cell proliferation can be reversed by bFGF. Endogenous bFGF production by HC is not affected by this purine analogue and 2-CdA-induced apoptosis is diminished in bFGF-producing HC as compared with normal PBMC. Therefore, bFGF expression by HC might be important for resistance to chemotherapy and survival of the malignant cells.

Inadequate production of hematopoietic growth factors in hairy cell leukemia: up-regulation of interleukin 6 by recombinant IFN-alpha in vitro.

The course of hairy cell leukemia (HCL) is characterized by progressive pancytopenia. The pathogenesis of this phenomenon is still not fully understood. To study if the decrease in hematopoiesis in HCL is accompanied by abnormal concentrations of growth factors, we investigated the production of granulocyte colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, interleukin 3 (IL-3), interleukin 6 (IL-6), and tumor necrosis factor alpha by peripheral blood mononuclear cells (PBMCs) of eight patients with HCL. The results point to a severe deficiency of production of all cytokines tested as compared to healthy donors. However, enrichment of autologous monocytes by counterflow centrifugation resulted in a marked increase of the levels of granulocyte colony-stimulating factor, granulocyte-macrophage colony-stimulating factor, IL-6, and tumor necrosis factor alpha. The most pronounced effects were seen with IL-6. Reverse transcription-PCR analysis indicated that pokeweed mitogen, IFN-alpha, and poly(I:C) are capable of inducing the expression of IL-6-specific mRNA in HCL cells. These findings are substantiated on the protein level by immunofluorescence analysis. Incubation of PBMCs with IFN-alpha resulted in a significant increase of intracellular IL-6 in HCL but not in healthy donors. This increase was also seen in hairy cells positive for CD19 and CDllc. Furthermore, IFN-alpha induced the secretion of IL-6 from PBMCs of HCL patients but not healthy donors. In conclusion, our studies with PBMCs from patients with HCL revealed an inadequate supply of hematopoietic growth factors that might, in part, be due to the monocytopenia characteristic for this disease. The findings also indicate that IFN-alpha is capable of inducing the production of IL-6 in the patients' PBMCs as well as in their hairy cells. These data from our in vitro studies support the clinical observation that treatment with IFN-alpha leads to reconstitution of hematopoiesis.

The role of cytokines in haematopoiesis.

Cytokines, by definition, exert an effect on haematopoiesis. Diseases characterized by haematopoietic insufficiency, such as aplastic anaemia, should therefore be investigated for abnormal expression of these regulatory proteins. In studies on hairy cell leukaemia, a severe deficiency was found in the production of interleukin-3 (IL-3), granulocyte-macrophage colony stimulating factor (GM-CSF), granulocyte CSF, IL-6 and tumour necrosis factor alpha (TNF alpha). Further studies on IL-6 at the mRNA and protein levels revealed that peripheral blood mononuclear cells and even hairy cells could be stimulated by interferon alpha (IFN alpha) to produce IL-6. It is interesting to speculate on the beneficial effects of IFN alpha therapy on the expansion of normal haematopoiesis and suppression or even elimination of malignant cells. Studies on a patient with angio-immunoblastic lymphadenopathy, another disease showing haematopoietic insufficiency, who developed severe aplastic anaemia, showed massive increases in IFN gamma and TNF alpha levels in serum; IL-6 and GM-CSF levels were below the limit of detection. These results correlated with an abnormal distribution of CD4+ and CD8+ T lymphocytes in the patient's blood and were compatible with the suppressive effects of IFN gamma and TNF alpha on haematopoiesis.

Myelosuppression in HCL: role of hairy cells, T cells and haematopoietic growth factors.

To elucidate mechanisms which may be responsible for the haematopoietic insufficiency in hairy cell leukaemia (HCL), we investigated in an autologous in vitro system the influence of haematopoietic growth factors (CSFs) and the effects of hairy cells (HCs) as well as T cells on the formation of haematopoietic colonies (CFU). Colony forming assays were performed using peripheral blood mononuclear cells (PBMC) of 6 HCL patients. To remove HCs, PBMCs were subjected to complement-mediated lysis, T cells were removed by E-rosette formation. Assays were done with and without recombinant human (rh) interleukin-3 (IL-3) and rh granulocyte-macrophage-colony-stimulating factor (GM-CSF). All 6 patients exhibited a severe reduction of their circulating progenitor cell (CPC) compartment. There was no correlation between the degree of colony reduction and the number of HCs. However, a correlation was found between the numbers of CPCs of HCL patients and healthy donors and the monocyte counts in these groups (r = 0.8573, p < 0.001). The removal of autologous HCs, but also of T cells, resulted in a significant increase in colony formation (BFU-E, CFU-GM, CFU-mix). In none of the experiments, however, did colony numbers come close to the normal range. This was only achieved by supplementation of the culture medium with rh IL-3 and rh GM-CSF. The results suggest that the haematopoietic failure observed in HCL patients is probably due to an inadequate supply of CSFs as well as to an inhibitory activity of HCs and T cells which might exert their effects in a synergistic fashion. There is also evidence that the lack of monocytes plays a role in the development of the haematopoietic insufficiency in HCL.

Evidence of colony suppressor activity and deficiency of hematopoietic growth factors in hairy cell leukemia.

The cause of myelosuppression in hairy cell leukemia (HCL) has been ascribed to a reduction of the circulating progenitor cell (CPC) compartment and to suppression of hematopoiesis by TNF-alpha. The present study was performed to evaluate the inhibitory effect of hairy cells (HCs) and a possible lack of hematopoietic growth factors on the number of autologous CPCs in vitro. In initial experiments the number of circulating BFU-E, CFU-GM and CFU-mix in HCL patients was found decreased. Monocytopenia but not the number of circulating HCs correlated to the degree of colony reduction in our patients. This pointed to a lack of colony stimulating factors (CSFs) in HCL. Actually, the growth of BFU-E, CFU-GM, and CFU-mix improved upon the addition of IL-3 and GM-CSF in HCL patients but not in healthy donors. To test the suppressive role of HCs in our assay system, cultures were performed after removal of autologous HCs. The results showed that in HC-depleted cultures the numbers of BFU-E, CFU-GM, and CFU-mix were significantly higher. This inhibitory effect of HCs could partially be neutralized by the addition of monoclonal antibodies against TNF-alpha. When the assays were performed with the removal of HCs and the addition of CSFs normal progenitor cell counts were detected in most patients. We conclude that HCs mediate the inhibition of colony growth in part by TNF-alpha. Monocytopenia is related with a deficiency of CSFs in this disease. The reduced colony growth in HCL, therefore, is due to both the inhibitory effects of HCs and the deficiency of CSFs. We suppose that the CPC-compartment is actually preserved in this disease.

[The therapeutic effect of interferon-alpha exemplified by hairy cell leukemia]. / Die therapeutische Wirkung von Interferon-alpha am Beispiel der Haarzelleukämie.

Hairy cell leukemia (HCL) was one of the first malignancies in which therapy with natural interferon-alpha achieved complete remissions. In an early multicenter trial we could confirm the efficacy of IFN, especially of recombinant preparations. In more than 14 clinical studies using all types of IFN-alpha in dose ranges of 2.000,000 to 3.000,000 remission rates of 80 to 90% have been documented. The observation that interruption of IFN therapy may result in a relapse which responds again to IFN implies the need for continuous therapy with IFN at a low dose level in patients with HCL. Refractory relapses might respond to adenosine-desaminase-inhibitors. Own investigations show that the pancytopenia in HCL is a result of a lack of hematopoietic growth factors with a dominance of inhibitory factors such as TNF-alpha. Thus IFN therapy does not directly act on hairy cells but rather via influences on other cellular regulatory mechanisms.

Beta 2-microglobulin. A reliable parameter for differentiating between graft rejection and severe infection after cardiac transplantation.

We investigated the role of beta 2-microglobulin as a noninvasive parameter to monitor acute rejection and severe infection in 45 consecutive heart transplant recipients. Endomyocardial biopsy revealed moderate (41 patients) or severe (three patients) rejection in 44 patients. Severe infections of bacterial septicemia (11 patients), bronchopneumonia (two patients), and viral infection (seven patients) were detected by a meticulous schedule of various clinical and laboratory tests. beta 2-Microglobulin levels in serum, generally corrected for serum creatinine, were significantly elevated in patients with infections (median, 6.3 mg/l; range Q10-Q90, 3.47-10.27 mg/l) compared with levels in patients with rejection (p less than 0.0001) or in patients in obviously good condition (p less than 0.0001). At the onset of acute rejection, the median corrected beta 2-microglobulin serum level was 1.56 mg/l (range Q10-Q90, -0.05-3.46 mg/l) and was significantly different from the control group (p less than 0.01). In addition, density function and empirical quantile analyses allowed us to define ranges of beta 2-microglobulin levels that would differentiate between rejection (2.05-3.46 mg/l) and infection (greater than 3.46 mg/l). With these values, sensitivity and specificity were 0.9 and 0.938 for detection of infection and 0.23 and 0.925 for detection of rejection, respectively. By means of beta 2-microglobulin, two cases of infection were misinterpreted as rejection (10%), and four of 44 rejections were mistaken for infections (9%). We conclude that measurements of beta 2-microglobulin may improve the management of heart transplant patients.

Long-term remission in a patient with erythroleukemia following interferon-alpha treatment.

A patient in second relapse of acute erythroleukemia (AEL) was treated with 10 MU of recombinant interferon (IFN)-alpha subcutaneously for 24 days after he had failed standard chemotherapy. Besides fever up to 38.9 degrees C and a transient raise in liver function parameters, treatment was well tolerated but had to be discontinued because of a severe decrease in white blood cell and platelet count. After termination of IFN treatment, both cell populations showed a continuous rise during the following weeks and hemoglobin increased concomitantly. Partial remission could be reached as was demonstrated by a decrease in PAS-positive erythroblasts in bone marrow biopsy. Further studies with IFN in patients with AEL refractory to cytostatic chemotherapy are recommended.

Duration of first remission as an indicator of long-term survival in chronic myelogenous leukaemia.

Approximately 31 patients with chronic myelogenous leukaemia (CML) are documented in the literature who survived more than 10 years after diagnosis. We present a CML-patient whose survival of 27 years is probably the longest reported so far. The analysis of the course of disease in these patients revealed that the duration of unmaintained first remission after chemotherapy is of high prognostic significance. In 17 of 24 evaluable patients the remission lasted more than 1 year and in another five at least 6 months (mean 73.8 months, range 0-240 months). In most patients busulfan was used as initial therapy. There was no correlation between the amount of drug given and the duration of remission or survival. Other parameters such as sex, age, initial leucocyte counts, differential count, haemoglobin, platelet count or spleen size seemed to have no prognostic relevance. While approximately 25% of CML patients with typical duration of survival exhibit a Ph1 chromosome mosaicism only, this finding was present in nearly half of the long-term survivers.

[Mechanism of action of interferon-alpha 2 in hairy cells and lymphoblastoid cell lines: the significance of type I interferon receptors and RNA synthesis]. / Zum Wirkungsmechanismus von Interferon-alpha 2 bei Haarzellen und lymphoblastoiden Zellinien: Die Bedeutung von Typ-I-Interferon-Rezeptoren und RNS-Synthese.

Lymphoblastoid cells were subjected to Western Blot analysis for detection of Interferon-alpha-binding proteins. JOK-1 cells--a human hairy cell leukemia line--revealed three proteins with apparent molecular weights of 120, 100 and 32 kD, respectively. Down-regulation of the receptor was observed. A differential binding pattern of two proteins (100 and 85 kD) was observed in T-CLL, whereas no signal detection was achieved in B-CLL. Mononuclear cells from 6 patients with hairy cell leukemia and 6 patients with CLL were found to differ significantly in terms of nucleic acid precursor incorporation.

Interferon alpha-2 for hairy cell leukemia: evidence for induction of RNA synthesis in hairy cells and failure to correlate enhancement of natural killer cells with elimination of hairy cells.

The effect of human recombinant interferon alpha 2 (IFN alpha 2) on hairy cells obtained from 16 patients was evaluated. All patients promptly responded to induction of remission with 2 X 10(6) U/m2 interferon alpha 2 b, three times a week, sc. In order to achieve a more detailed insight into the mode of action of interferon in this disease, we determined the influence of IFN alpha 2 on the incorporation of radiolabeled thymidine and uridine into hairy cells. While both 3H-thymidine and 3H-uridine incorporation were unaffected by IFN alpha 2 in a 3-hour incubation period, a significant increase in uridine incorporation into hairy cells, but not CLL cells, was observed after 24 h. Cell surface marker analysis performed with monoclonal antibodies did not reveal a quantitative alteration of the immunophenotype of hairy cells in vitro. In addition, natural killer cells, assessed by monoclonal antibodies and a cytotoxicity assay against K 562 cells, were found to be decreased in 9 out of 10 patients prior to therapy. Although IFN alpha 2 could stimulate natural killer cells in vivo, we did not find a consistent correlation between the activation of these cells and the response to therapy. We conclude, therefore, that NK cells play no major role in the regression of hairy cells. Furthermore, IFN alpha 2 does not alter antigenic determinants in vitro, but leads to an enhanced incorporation of 3H-uridine into hairy cells in vitro, thus indicating a possible role for the induction of RNA synthesis in vivo.

[Recombinant (IFN-alpha 2b) therapy in hairy cell leukemia]. / Rekombinante (IFN-alpha 2b)-Therapie der Haarzell-Leukämie.

Eighty-five patients with hairy-cell leukemia were treated in a multicentric "open label" study with IFN-alpha 2b and evaluated. Induction therapy was 2 X 10(6) U IFN-alpha 2b/m2, 3 times a week, s.c. The results show this regimen to be highly effective with only a few tolerable and transient side effects consisting mainly of flu-like symptoms. After 6 months of therapy 4% CR, 69% PR, and 16% MR, were noted. In a small group of four patients who had achieved CR or PR, we tested the effect of varying doses for maintenance therapy. Our preliminary results indicate that a relapse caused by interruption of IFN therapy or dose reduction to 3 X 10(6) U given once a week, o.c. could be successfully treated by readministration, or escalating the dosage of IFN. It seems that remission maintenance requires long-term treatment with IFN. In a short-term in vitro test we studied the effect of IFN-alpha 2 on the incorporation of 3H-thymidine and 3H-uridine into hairy cells of five patients. Fort both precursors no appreciable effect was detected. However, after prolonged incubation for 48 h, a significant enhancement of 3H-uridine incorporation was observed, while 3H-thymidine incorporation remained unaffected. Cell marker analysis performed with monoclonal antibodies before and after incubation of hairy cells with IFN-alpha 2 for up to 7 days did not reveal any change of the phenotype of hairy cells.