水上 浩明

A 65-year-old man with marked leukocytosis was admitted for diagnosis and treatment. His peripheral blood leukocyte count was 37,500/mu-l and the leukocytes consisted of mature neutrophil-like cells. A high neutrophil alkaline phosphatase score and a normal bone marrow cell karyotype suggested that the patient had chronic neutrophilic leukemia rather than chronic myeloid leukemia. Several neutrophil functions, such as superoxide production, nitroblue tetrazolium reduction activity, and phagocytosis, were elevated. These data and the morphological features (toxic granules and Dohle bodies) indicated that the patient's neutrophils were in an activated stage.

Granulocyte colony-stimulating factor (G-CSF) is known to act on the neutrophilic granulocytes from chronic myelogenous leukemia (CML) patients to induce neutrophil alkaline phosphatase (NAP) activity. Gamma-interferon (IFN-gamma) and granulocyte-macrophage colony-stimulating factor (GM-CSF) have been reported to suppress NAP induction with G-CSF. We confirmed that this inhibitory effect of GM-CSF is accompanied by the decrease of the NAP mRNA level. Moreover, we found that the simultaneous addition of retinoic acid completely neutralized this inhibitory effect of GM-CSF. Recovery of the NAP activity brought about by the retinoic acid was also accompanied by the increase of NAP mRNA level. These results indicate that retinoic acid neutralizes the inhibitory effect of GM-CSF on the induction of NAP activity through the change of the NAP mRNA level.

To investigate the physiologic role of macrophage colony-stimulating factor (M-CSF) in hematological recovery from bone marrow hypoplasia, we used an enzyme-linked immunosorbent assay to measure serial changes of the serum M-CSF level during 25 intensification chemotherapy courses given to seven patients with acute non-lymphocytic leukemia who were in complete remission. Three M-CSF peaks were observed during therapy: the first peak was during or just after chemotherapy, the second peak was around the leukocyte nadir, and the third peak coincided with a rapid increase in the monocyte count. We could find no significant correlation between the height of the second peak and the time from the initiation of therapy to hematological recovery. On the other hand, there was a significant positive correlation between the height of the second peak and the interval from the last day of chemotherapy to the peak (r = 0.62, p = 0.001), and there was a significant negative correlation between the peak height and the time from the peak until hematological recovery (defined as a neutrophil count of over 500/mu-l (r = -0.63, p = 0.001) and a leukocyte count of over 1,000/mu-l (r = 0.55, p = 0.008). However, we found only a weak correlation between the peak height and monocyte recovery. These data suggest that increased M-CSF levels lead to the stimulation of granulocyte progenitors, and that we can predict the time of neutrophil recovery by monitoring the serum M-CSF level and finding its peak.

A 71-year-old woman with leukocytosis was admitted for treatment of malignant lymphoma. During the clinical course, neutrophilia of unknown origin occurred in parallel with the progression of the malignant lymphoma. The supernatant of lymphoma tissue culture contained a high titer of granulocyte colony-stimulating factor (G-CSF), and lymphoma cells were positive when immunohistochemically stained by anti-G-CSF antibody. Western blot analysis and mouse colony assay of the supernatant also confirmed that the lymphoma produced G-CSF.

We examined steady-state levels of mRNA for alkaline phosphatase in neutrophils (NAP) treated with granulocyte colony-stimulating factor (G-CSF). The amount of mRNA for NAP was shown to increase after 6 hours of culture with G-CSF when no increase in NAP activity was yet observed, and the transcript was the greatest after 20-24 h of culture with G-CSF. Treatment of neutrophils with both G-CSF and retinoic acid augmented the amount of mRNA for NAP over the amount obtained by G-CSF alone, which was most marked at 24 h. These results show that both G-CSF-mediated NAP induction and its potentiation by retinoic acid are due to the increased levels of mRNA for NAP.