Plasma levels of GM-CSF, IL-3, IL-6 and IL-1 (alpha,beta) were investigated in steady-state sickle cell patients. We find that SS patients with low levels of HbF (< 9% [LFSS]) are characterized by elevated plasma GM-CSF, whereas in patients with high levels of HbF (HFSS), GM-CSF is not detectable. In contrast, HFSS patients exhibited increased plasma IL-3 (m = 84.8 +/- 57 pg/ml), whereas LFSS patients had lower or no detectable plasma IL-3, IL-1(alpha,beta) and IL-6 were also detected in plasma from some SS patients, but there was no correlation with HbF levels. Normal controls tested negative for all cytokines, except for one individual positive for IL-3. These results are compatible with a model in which the level of haemopoietic stress determines the level of participation of GM-CSF or IL-3 in the regulation of SS circulating BFU-E.

A hemoglobin F (HbF) level between eight and nine percent divides sickle cell anemia (SS) patients into two populations, according to the kinetics of circulating burst forming units-erythroid (BFU-E), long term culture-initialing cells (LTC-IC), and cytokine plasma concentrations. The SS patients with HbF levels lower than 8-9% are more anemic (LFSS patients) than those with HbF levels higher than 8-9% who have less severe anemia (HFSS patients). We report here that the level of erythropoiesis [evaluated by the levels of soluble transferin receptors (sTfR)] is not identical in these two patient populations, supporting the idea that a different set of regulatory mechanisms might be required to maintain the two levels of increased hematopoiesis. The plasma sTfR concentration was increased in all SS samples compared with controls (P < 0.002) and sTfR levels were negatively correlated with peripheral HbF%. (r = -0.574, P < 0.002). Furthermore, sTfR levels were higher in LFSS than in HFSS patients. Erythropoietin (Epo) levels were increased in the plasma of LFSS individuals (range = 34-215 ml U/ml), while the values in HFSS patients were in the normal range (3-20 ml U/ml). Furthermore, we identify here stem cell factor (SCF) and transforming growth factor-beta (TGF-beta) as regulatory factors specifically affected by the presence of SS genotype and its level of severity. The plasma concentrations of SCF and TGF-beta were increased compared with normal controls and high levels of SCF (up to 7,000 pg/ml) were detected in LFSS patients. The latter also showed increased proportion of SCF+ CD34 enriched circulating cells (49%). Low SCF in HFSS patients is associated with elevated TGF-beta, suggesting a regulatory role of the latter on either SCF release or c-kit expression in progenitor cells. Occasional elevation of granulocyte macrophage-colony stimulating factor (G-CSF), interleukin (IL)-7, and macrophage inflammatory protein (MIP)-1alpha in plasma of SS patients is not specific because no relation to HbF could be demonstrated. All plasma tested for leukemia inhibitory factor (LIF) were negative. Data presented here, complementing previously published information, supports a model in which HFSS patients achieve a balance between inhibitory (TGF-beta) and stimulatory (SCF, IL-3) factors, resulting in moderate erythropoietic response. In contrast, in LFSS patients, low levels of TGF-beta and the increased release of GM-CSF and SCF maintain the intense erythropoiesis in response to higher erythropoietic stress, in these more severe patients.

Our data suggest that PTE is associated with increased erythroid progenitor sensitivity to Ep. The effect of ACEI to decrease hematocrit in patients with PTE may be due to inhibition of red cell precursor growth.

No abstract

We have previously determined that in African sickle cell anemia (SS) patients three different beta-like globin gene cluster haplotypes are associated with different percent G gamma (one of the two types of non- alpha chains comprising hemoglobin F [HbF]), mean percent HbF, and percent dense cells. We report now that in adult New York SS patients, the presence of at least one chromosome with the Senegal haplotype is associated with higher Hb levels (1.2 g/dL higher) than is found for any other non-Senegal haplotype (P less than .004). The percent reticulocytes and the serum bilirubin levels were lower in these patients. When the effect of alpha-gene number was analyzed by examining a sample of SS patients with concomitant alpha-thalassemia, the same results were obtained. Because the HbF level is significantly higher among the Senegal haplotype carriers in this sample, the inhibitory effect on sickling of this Hb variant may be one of the reasons for the haplotype effect. We conclude that the Senegal beta- like globin gene cluster haplotype is associated with an amelioration of the hemolytic anemia that characterizes sickle cell disease.