Poor prognosis acute myelogenous leukemia: 3--biological and molecular biological changes during remission induction therapy.

This is the third paper in a series which describes a new remission induction regimen for patients with 'poor prognosis' acute myelogenous leukemia (AML). Twenty-four patients were treated with two one day pulses of chemotherapy separated by 96 h. Each pulse consisted of two doses of cytarabine and a single dose of mitoxantrone. Amifostine was administered three times a week after the second pulse of chemotherapy until treatment outcome became known. The first paper described the outcome of treatment while the second described the relationship of treatment outcome to the pretherapy characteristics of the leukemia. This paper describes the changes in the leukemia cells which occur during remission induction therapy. While only a limited number of specimens were available for each post treatment study, the studies demonstrated a profound fall in blood counts, BM cellularity, and telomerase activity in leukemia cells after pulse #1 of treatment. This fall was usually accompanied by a coordinate rise in IL6, TNFalpha, and IL1beta transcripts within the AML cells which survived chemotherapy. High levels of telomerase activity in the day 5 marrow was correlated with high levels of IL1beta transcripts which in turn were associated with treatment failure ascribable to resistant disease.

Cytokine production by in vitro processed and unprocessed haematopoietic cells.

The studies described here demonstrate that in vitro processing of cells before extraction of RNA has a major effect on the number and type of cytokine transcripts present within MDS and leukemia cells. Transcripts for GM-CSF, a cytokine whose production by leukemia cells is believed to play an important role in the pathogenesis of leukemia, was not detectable in 12/13 unprocessed AML specimens, in 12/12 MDS specimens, or in 7/7 CML specimens but once detected in many specimens after processing. These data strongly suggest that leukemia cell production of GMCSF rarely occurs in vivo.

SCF, IL-1beta, IL-1ra and GM-CSF in the bone marrow and serum of normal individuals and of AML and CML patients.

This study compared cytokine transcript and protein levels in BM cells of normal individuals and leukemic patients. AML differed from normal in that: (1) AML marrow cells contain more IL-1beta protein than normal cells, (2) IL-1ra transcripts are absent from AML marrow cells, (3) AML marrow serum contains less IL-1ra protein than normal, (4) peripheral blood and marrow serum of AML patireents contains more SCF protein than normal serum, and (5) SCF transcripts have been detected in AML marrow biopsies and not in aspirate cells. These data suggest that unbalanced cytokine production may make a significant contribution to the abnormal behaviour of AML cells.

Suppression of telomerase activity and cytokine messenger RNA levels in acute myelogenous leukemia cells in vivo in patients by amifostine and interleukin 4.

High levels of telomerase activity and high rates of cell proliferation are associated with a poor prognosis in acute myelogenous leukemia. Furthermore, cytokine production by leukemia cells is believed to play an important role in determining the proliferative characteristics of leukemia. The in vivo effects of two noncytotoxic agents on these parameters were determined in 33 acute myelogenous leukemia patients. Three daily doses of interleukin (IL) 4 or a single dose of amifostine reduced telomerase activity in the leukemia marrow cells in 7 of 9 and 11 of 13 patients, respectively. The administration of a single dose of amifostine resulted in a reduction in tumor necrosis factor alpha and IL-6 transcript levels in the marrow cells of 10 of 13 and 12 of 13 patients in which these transcripts were present. The administration of only three doses of IL-4 or a single dose of amifostine has a significant effect on leukemia cell parameters, which are believed to have a significant impact on the in vivo biology of the disease and on its response to remission induction therapy.

Evidence of the role of protein kinase C during aclacinomycin induction of erythroid differentiation in K562 cells.

At subtoxic concentrations, aclacinomycin is effective in controlling erythroid differentiation of K562, a human erythroleukemic cell line. To better understand early events implicated in this process, we have used bisindolylmaleimide (GF109203X), an inhibitor with a high selectivity for protein kinase C (PKC). Our data show that GF109203X inhibits aclacinomycin effects on K562, evidenced by a strong reduction of hemoglobinized cells and a marked decrease of mRNA rates of erythroid genes. To establish firmly PKC involvement, we also verified that aclacinomycin stimulates its rapid translocation, from the cytosolic to the membrane compartment. By Western blot analysis, we also show that after short induction times, PKCalpha was the most implicated.

Marrow cytokine transcripts and the secondary hematologic disorders.

A comparison was made of the cytokine transcripts in normal, monoclonal, MDS, and AML marrow aspirates. While both normal and monoclonal marrow aspirates contain transcripts for SCF, few MDS or AML marrow aspirates contain these transcripts. Similarly, IL1ra transcripts are found with reduced frequency in MDS and AML marrow aspirates. The fall in SCF transcripts between monoclonal and MDS marrow aspirates parallels the appearance of apoptosis and the reduced in vitro proliferative ability which are characteristics of MDS marrow aspirate cells. The frequent IL1beta production by MDS and AML marrow aspirate cells, with few marrow aspirates producing IL1ra transcripts, suggests that unbalanced IL1beta effects may contribute to the proliferative advantage of MDS and AML cells over their normal counterparts.

Cytokine gene activity in AML cells in vivo in patients.

The proliferation of acute myelogenous leukemia cells is dependent upon cytokine stimulation. Additionally, there is a body of literature which reports that leukemia cells produce GMCSF, IL6, and other cytokines. The study reported here, using an rt-multiplex polymerase method, determined the presence or absence of transcripts in freshly obtained AML cells for the following cytokine or cytokine-related genes: IL 1beta, IL1ra, TNF alpha, GMCSF, IL6, flt 3, and hSCF. This demonstrated that leukemia cell populations usually contain transcripts for IL1beta, TNF alpha, flt 3 and flt 3 ligand in vivo and that transcripts for the other cytokines only appear after the leukemia cells are processed in vitro. The presence of TNF alpha transcripts appears to be associated with resistance to remission induction therapy. Furthermore, the transcript profile of the leukemia cells can change during remission induction therapy. The data also demonstrate the assessment of cytokine production by leukemia cells after in vitro manipulation should not be extrapolated to the in vivo situation.

Activation of Raf-1 and mitogen-activated protein kinases by erythropoietin and inositolphosphate-glycan in normal erythroid progenitor cells: involvement of protein kinase C.

In this work, we show that erythropoietin and inositolphosphate-glycan activate Raf-1 and the mitogen-activated protein kinases (MAP kinases) in normal erythropoietin-responsive cells. Using a protein kinase C (PKC) activator such as the phorbol ester, 12-O-tetradecanoyl-phorbol-13-acetate and the PKC inhibitor GF109203X, we investigated a possible involvement of PKC during activation of Raf-1 and MAP kinase by erythropoietin or inositolphosphate-glycan. We found that erythropoietin increased MAP kinase level with a maximum stimulation reached at 5-10 min. Inositolphosphate-glycan and 12-O-tetradecanoyl-phorbol-13-acetate increased MAP kinase activity in the same manner. This activity was inhibited by cell preincubation with GF109203X. Two MAP kinase isoforms were present in erythroid progenitor cells, the 44 and 42 kDa proteins. We report here that erythropoietin, inositolphosphate-glycan, and 12-O-tetradecanoyl-phorbol-13-acetate activated only the p44 form (erk-1) of MAP kinase and the Raf-1 protein. GF109203X was used at a concentration which inhibited by 50% erythroid colonie (CFU-E) proliferation and differentiation induced by erythropoietin or inositolphosphate-glycan. These results support the hypothesis that erythropoietin and inositolphosphate-glycan activate Raf-1 and MAP kinases in normal erythroid progenitor cells and suggest that this activation involves PKC.

Erythropoietin stimulates glycosylphosphatidylinositol hydrolysis in rat erythroid progenitor cells and inositolphosphate glycan modulates their proliferation.

The involvement of a glycosylphosphatidylinositol/inositolphosphate glycan (GPI/IPG) system in the erythropoietin (Epo) signal transduction was investigated. Endogenous GPI was evidenced in extracts of normal Epo-responsive cells after incorporation of [3H]glucosamine, [3H]inositol and [32P]orthophosphate. Incubation of these cells with Epo produced a rapid and transient hydrolysis of GPI with parallel release of IPG. IPG production was Epo dose dependent and the maximal effect was obtained with the same concentration of Epo which gave the maximal mitogenic effect, i.e. 1 U/ml. The number and size of erythroid colonies (CFU-E) were increased by the addition of purified rat erythroid IPG to the culture medium, but not to the same extent as with a maximal Epo treatment. Exogenous IPG effect was dose dependent. In the presence of suboptimal Epo concentrations, IPG has been found to potentiate Epo-induced CFU-E growth. These results support the hypothesis that a GPI/IPG based signal transduction system may be involved in Epo-induced cell proliferation.