RSK2 is a new Pim2 target with pro-survival functions in FLT3-ITD-positive acute myeloid leukemia.

Acute myeloid leukemia (AML) with the FLT3 internal tandem duplication (FLT3-ITD AML) accounts for 20-30% of AML cases. This subtype usually responds poorly to conventional therapies, and might become resistant to FLT3 tyrosine kinase inhibitors (TKIs) due to molecular bypass mechanisms. New therapeutic strategies focusing on resistance mechanisms are therefore urgently needed. Pim kinases are FLT3-ITD oncogenic targets that have been implicated in FLT3 TKI resistance. However, their precise biological function downstream of FLT3-ITD requires further investigation. We performed high-throughput transcriptomic and proteomic analyses in Pim2-depleted FLT3-ITD AML cells and found that Pim2 predominantly controlled apoptosis through Bax expression and mitochondria disruption. We identified ribosomal protein S6 kinase A3 (RSK2), a 90 kDa serine/threonine kinase involved in the mitogen-activated protein kinase cascade encoded by the RPS6KA3 gene, as a novel Pim2 target. Ectopic expression of an RPS6KA3 allele rescued the viability of Pim2-depleted cells, supporting the involvement of RSK2 in AML cell survival downstream of Pim2. Finally, we showed that RPS6KA3 knockdown reduced the propagation of human AML cells in vivo in mice. Our results point to RSK2 as a novel Pim2 target with translational therapeutic potential in FLT3-ITD AML.

A sequence of the CIS gene promoter interacts preferentially with two associated STAT5A dimers: a distinct biochemical difference between STAT5A and STAT5B.

Two distinct genes encode the closely related signal transducer and activator of transcription proteins STAT5A and STAT5B. The molecular mechanisms of gene regulation by STAT5 and, particularly, the requirement for both STAT5 isoforms are still undetermined. Only a few STAT5 target genes, among them the CIS (cytokine-inducible SH2-containing protein) gene, have been identified. We cloned the human CIS gene and studied the human CIS gene promoter. This promoter contains four STAT binding elements organized in two pairs. By electrophoretic mobility shift assay studies using nuclear extracts of UT7 cells stimulated with erythropoietin, we showed that these four sequences bound to STAT5-containing complexes that exhibited different patterns and affinities: the three upstream STAT binding sequences bound to two distinct STAT5-containing complexes (C0 and C1) and the downstream STAT box bound only to the slower-migrating C1 band. Using nuclear extracts from COS-7 cells transfected with expression vectors for the prolactin receptor, STAT5A, and/or STAT5B, we showed that the C1 complex was composed of a STAT5 tetramer and was dependent on the presence of STAT5A. STAT5B lacked this property and bound with a stronger affinity than did STAT5A to the four STAT sequences as a homodimer (C0 complex). This distinct biochemical difference between STAT5A and STAT5B was confirmed with purified activated STAT5 recombinant proteins. Moreover, we showed that the presence on the same side of the DNA helix of a second STAT sequence increased STAT5 binding and that only half of the palindromic STAT binding sequence was sufficient for the formation of a STAT5 tetramer. Again, STAT5A was essential for this cooperative tetrameric association. This property distinguishes STAT5A from STAT5B and could be essential to explain the transcriptional regulation diversity of STAT5.

High mTORC1 activity drives glycolysis addiction and sensitivity to G6PD inhibition in acute myeloid leukemia cells.

Alterations in metabolic activities are cancer hallmarks that offer a wide range of new therapeutic opportunities. Here we decipher the interplay between mTORC1 activity and glucose metabolism in acute myeloid leukemia (AML). We show that mTORC1 signaling that is constantly overactivated in AML cells promotes glycolysis and leads to glucose addiction. The level of mTORC1 activity determines the sensitivity of AML cells to glycolysis inhibition as switch-off mTORC1 activity leads to glucose-independent cell survival that is sustained by an increase in mitochondrial oxidative phosphorylation. Metabolic analysis identified the pentose phosphate pathway (PPP) as an important pro-survival pathway for glucose metabolism in AML cells with high mTORC1 activity and provided a clear rational for targeting glucose-6-phosphate dehydrogenase (G6PD) in AML. Indeed, our analysis of the cancer genome atlas AML database pinpointed G6PD as a new biomarker in AML, as its overexpression correlated with an adverse prognosis in this cohort. Targeting the PPP using the G6PD inhibitor 6-aminonicotinamide induces in vitro and in vivo cytotoxicity against AML cells and synergistically sensitizes leukemic cells to chemotherapy. Our results demonstrate that high mTORC1 activity creates a specific vulnerability to G6PD inhibition that may work as a new AML therapy.

Renal Mitochondrial Lipid Peroxidation during Sepsis.

Sepsis can provoke kidney injury, which increases mortality. Human and animal studies have documented increased renal oxidative injury and mitochondrial damage during sepsis. However, few studies have attempted to dissect specific renal targets and/or types of oxidative injury using the cecal ligation and puncture (CLP) murine model of sepsis. The purpose of this short communication is to examine the extent of lipid peroxidation within renal mitochondria using CLP and blue native gel electrophoresis which separates intact mitochondrial respiratory complexes. Our results show that CLP induced increased 4-hydroxy-nonenal protein adduction (marker of lipid peroxidation) in renal homogenates and mitochondrial fractions. Blue native gel electrophoresis revealed that respiratory complex III was selectively targeted within mitochondrial fractions. This supports our prior report showing renal complex III inactivation following CLP. Future studies will identify specific renal proteins within complex III that are modified during sepsis to provide mechanistic insight on how mitochondrial respiration is inhibited during sepsis.

Abnormal erythropoietin (Epo) gene expression in the murine erythroleukemia IW32 cells results from a rearrangement between the G-protein beta2 subunit gene and the Epo gene.

Abnormal production of erythropoietin (Epo) has been described in several human and murine erythroleukemia. The murine IW32 cell line is derived from an F-MuLV-induced erythroleukemia. An autocrine Epo production due to the rearrangement of one Epo allele has been previously described (Beru et al., 1989). However, the exact mechanism leading to the transcriptional activation of the abnormal Epo gene was unknown. In this study, we show that this deregulated expression results from a deletion within chromosome 5. The Epo gene in the abnormal allele is under the control of the G-protein beta2 subunit gene promoter and the expressed mRNA results from the fusion of the non coding exon 1 of the G-protein beta2 subunit gene to a truncated Epo exon 1 gene. This resulting abnormal cDNA allows the expression of a normal Epo protein.

Cooperation between STAT5 and phosphatidylinositol 3-kinase in the IL-3-dependent survival of a bone marrow derived cell line.

Cytokine-dependent activation of distinct signaling pathways is a common scheme thought to be required for the subsequent programmation into cell proliferation and survival. The PI 3-kinase/Akt, Ras/MAP kinase, Ras/NFIL3 and JAK/STAT pathways have been shown to participate in cytokine mediated suppression of apoptosis in various cell types. However the relative importance of these signaling pathways seems to depend on the cellular context. In several cases, individual inhibition of each pathway is not sufficient to completely abrogate cytokine mediated cell survival suggesting that cooperation between these pathways is required. Here we showed that individual inhibition of STAT5, PI 3-kinase or MEK activities did not or weakly affected the IL-3 dependent survival of the bone marrow derived Ba/F3 cell line. However, the simultaneous inhibition of STAT5 and PI 3-kinase activities but not that of STAT5 and MEK reduced the IL-3 dependent survival of Ba/F3. Analysis of the expression of the Bcl-2 members indicated that phosphorylation of Bad and Bcl-x expression which are respectively regulated by the PI 3-kinase/Akt pathway and STAT5 probably explain this cooperation. Furthermore, we showed by co-immunoprecipitation studies and pull down experiments with fusion proteins encoding the GST-SH2 domains of p85 that STAT5 in its phosphorylated form interacts with the p85 subunit of the PI 3-kinase. These results indicate that the activations of STAT5 and the PI 3-kinase by IL-3 in Ba/F3 cells are tightly connected and cooperate to mediate IL-3-dependent suppression of apoptosis by modulating Bad phosphorylation and Bcl-x expression.

Role of Gab proteins in phosphatidylinositol 3-kinase activation by thrombopoietin (Tpo).

In this study, we show that upon thrombopoietin (Tpo) stimulation the two adapter proteins Gab1 and Gab2 are strongly tyrosine phosphorylated and associated with Shc, SHP2, PI 3-kinase and Grb2 in mpl-expressing UT7 cells. Although Gab1 and Gab2 seem to mediate overlapping biological signals in many cells, only Gab1 is expressed and phosphorylated in response to Tpo in primary human megakaryocytic progenitors; furthermore, it associates with the same proteins. Although a low level of tyrosine phosphorylated IRS-2 protein is also detected in PI 3-kinase immunoprecipitates, Gab proteins are the essential proteins associated with PI 3-kinase after Tpo stimulation. We demonstrate that, albeit no association is detected between the Tpo receptor mpl and Gab proteins, Y112 located in the C-terminal cytoplasmic domain of mpl is required for Gab1/2 tyrosine phosphorylation. Gab proteins are not tyrosine phosphorylated after Tpo stimulation of UT-7 and Ba/F3 cells expressing a mpl mutant lacking Y112. Moreover, no activation of the PI 3-kinase/Akt pathway is observed in cells expressing this mpl mutant. Finally, we show that this mutant does not allow cell proliferation, thereby confirming that PI 3-kinase activation is required for Tpo-induced cell proliferation.

BCR-ABL and constitutively active erythropoietin receptor (cEpoR) activate distinct mechanisms for growth factor-independence and inhibition of apoptosis in Ba/F3 cell line.

The interleukin-3 dependent murine Ba/F3 cell line has been widely used as an experimental model of cell transformation by BCR-ABL oncogenes as assessed by induction of growth-factor-independence and inhibition of apoptosis in vitro. The signaling pathways used by BCR-ABL oncogenes to exert these effects are unknown. To gain insights into this phenomenon, we have introduced the p190- and p210-encoding BCR-ABL oncogenes as well as the constitutively activated oncogenic murine erythropoietin receptor (cEpoR) into Ba/F3 and compared the behavior of individual clones in response to apoptotic stimuli. Both p210 and p190 BCR-ABL vectors induced IL-3-independent growth and the same result was obtained with the cEpo-R vector. Individual clones of Ba/F3 cells expressing BCR-ABL exhibited significant resistance to apoptosis induced by either etoposide, serum deprivation or growth-factor withdrawal. In contrast, Ba/F3 cells expressing the constitutively active cEpoR behaved like parental Ba/F3 cells undergoing apoptosis when similarly treated with etoposide or upon serum deprivation. Bc12 and Bax levels were similar in all BCR-ABL and cEpoR-transfected clones. However, in band-shift assays, nuclear extracts from growth-factor-independent Ba/F3 clones expressing cEpoR had no detectable STAT activity as opposed to the constitutive STAT activation detected in all Ba/F3 clones expressing p210 or p190 BCR-ABL. Our results indicate that although both constitutively activated cEpoR and BCR-ABL oncogenes induce growth-factor independence in Ba/F3 cells, only BCR-ABL is able to protect cells from etoposide and serum-deprivation-induced apoptosis and induce a strong constitutive activation of STAT factors, suggesting a role for these molecules in the anti-apoptotic activity of BCR-ABL.

Glomerular thromboxane A2/prostaglandin H2 receptors: characterization and effect of adriamycin-induced nephrotic syndrome.

We have characterized the thromboxane (TX) A2/prostaglandin (PG) H2 receptor in glomeruli isolated from the rat using the agonist radioligand [125I]-BOP. Binding of [125I]-BOP was highly specific, stereoselective, and to a single class of high affinity binding sites (Kd = 1.16 +/- 0.22 nM and Bmax = 348 +/- 32 fmol/mg protein; n = 6). Binding of [125I]-BOP was competed for by the agonist ONO11113 (Kd = 50.8 +/- 8.0 nM; n = 4) and the antagonists SQ29548 (Kd = 15.8 +/- 1.0 nM; n = 3), L657925 (Kd = 12.1 +/- 2.2 nM; n = 3) and L657926 (Kd = 1642 +/- 135 nM; n = 3). I-BOP also produced a TXA2/PGH2 receptor-mediated rise in [Ca2+]i in isolated glomeruli In adriamycin-induced nephrotic syndrome in the rat, the development of proteinuria is reported to be dependent on increased renal TXA2 production. We therefore examined whether or not changes in glomerular TXA2/PGH2 receptors occur between control and nephrotic rats. No changes in expression or affinity of either glomerular or platelet TXA2/PGH2 receptors were observed. Kd and Bmax values for isolated glomeruli were 1.45 +/- 0.24 nM and 406 +/- 72 fmol/mg for controls and 1.22 +/- 0.25 nM and 321 +/- 62 fmol/mg for nephrotic rats (n = 6).

Evidence for a bidirectional prostaglandin endoperoxide shunt between platelets and the bovine coronary artery.

While platelets have been shown to be capable of supplying prostaglandin (PG) H2 to endothelial cells in culture for PGI2 synthesis, endothelial cells have been shown unable to supply PGH2 to platelets for thromboxane (TX) A2 synthesis. We incubated rings of the bovine coronary artery (BCAR) with human platelets treated with aspirin (to inhibit cyclooxygenase) or CGS 13080 (to inhibit TXA2 synthase) in the presence of 20 microM arachidonic acid. BCAR, with damaged endothelium, produced significantly less PGI2 than that with intact endothelium. However, co-incubation with CGS 13080-treated platelets resulted in an increase in PGI2 independent of endothelium, demonstrating a shunt of PGH2 from platelets to BCAR. Co-incubation of BCAR with aspirin-treated platelets resulted in a net increase in TXA2 demonstrating a shunt of PGH2 from BCAR to platelets. Employing [14C]PGH2 as substrate, BCAR with and without intact endothelium produced similar amounts of 6-keto-[14C]PGF1 alpha. Likewise, homogenates (50 micrograms protein) of intimal and subintimal regions of BCAR and BCAR converted similar amounts of PGH2 to 6-keto-PGF1 alpha. These data suggest that vascular production of PGH2 is more dependent on an intact endothelium than is the conversion of PGH2 to PGI2. These data also suggest a potential for a bidirectional exchange of PGH2 between platelets and vascular wall during platelet-vascular wall interactions.

Effect of the antiglucocorticoid agent RU 38486 on the dexamethasone inhibition of Friend cell differentiation.

Glucocorticoid hormones are known to inhibit the erythroid differentiation of Friend cells. The mechanism of action of these hormones has been questioned, and results suggesting an action not involving the nuclear binding of the receptors have been published. We have used the antiglucocorticoid RU 38486 to block the inhibitory effect of dexamethasone on the induced differentiation of Friend cells. Our results strongly suggest a glucocorticoid action involving the binding of classical receptors to the cell nucleus.

Kinetics of dexamethasone binding to the glucocorticoid receptor of intact erythroid cells from rat fetal liver.

The erythroid cells from the rat fetal liver have been shown to possess a receptor for glucocorticoids. In the present work, the characteristics of [3H]dexamethasone binding have been studied on intact cells, in order to minimize receptor degradation, and at 4 degrees C, in order to prevent the activation of the hormone-receptor complex. Dissociation kinetics were those of a first-order reaction and the value of the rate constant of dissociation was similar to the values available in the literature. When studied at low concentrations of the ligand and using short-term incubations, association kinetics were apparently those of a simple bimolecular reaction. But at high ligand concentrations and/or using long-term incubations, association kinetics indicated a more complex reaction. Our results were compatible with the model proposed by Pratt W.B., Kaine J.L. and Pratt V.D. (J. Biol. Chem. 250 (1975) 4584-4591) for cytosolic preparations. This model implies the rapid formation of a transient unstable form of the complex, further converted into a stable form with slower kinetics. Equilibrium dissociation constant of the first (rapid) reaction was 80 microM and the rate constant of 'stabilization' was of the order of 70 X 10(-3) min-1. These values agree with the results of Pratt et al. relative to a cytosolic preparation from rat thymocytes.

Differentiating activity of adriamycin in human erythroleukemic cells: effect on globin and heme synthesis.

The human leukemic cell line K 562 can be induced to differentiate along the erythroid lineage by various chemical compounds and particularly by the anthracyclic antitumor drug, adriamycin (ADR). In this study, we show that, in the presence of a subtoxic concentration of ADR (30 nM), the appearance of hemoglobin-producing K 562 cells is associated with a specific increase in globin mRNA accumulation corresponding to epsilon-, zeta-, gamma-, alpha-globin chains. At the translational level, bulk protein synthesis is strongly decreased following ADR treatment, whereas globin chain synthesis is specifically enhanced. Globin chains represent about 20% of total proteins in ADR-treated cells, versus about 3.5% in controls on day 3. Similarly, on day 3, heme synthesis (55Fe incorporation) is about 10-times higher in ADR-treated cells than in control cells (20,888 dpm/10(5) cells versus 1693 dpm/10(5) cells) which confirms the increase in heme content (420 pM/10(6) treated cells versus 100 pM/10(6) control cells). In the presence of succinylacetone, a heme synthesis inhibitor which prevented the differentiating effects of ADR, the globin mRNA accumulation was not affected. This suggests that heme did not play a regulatory role in globin mRNA transcription, a result at variance with observations published by others. Such results strongly support the notion that in addition to cytostatic properties, ADR stimulates specifically globin and heme synthesis.

Induction of erythropoietin receptors during aclacinomycin-mediated erythroid differentiation of K562 leukemia cells.

Human K562 leukemia cells have been induced to differentiate along the erythroid lineage by aclacinomycin (ACM), an anthracyclic antitumor drug. During differentiation over 3 days in culture, the expression and the nature of erythropoietin (EPO) receptors have been analyzed using 125I-labeled bioactive recombinant human EPO. Aclacinomycin at 20 nM, the concentration inducing optimum differentiation, progressively increased EPO-specific binding. On day 3, EPO binding was nine-fold higher than that of the controls (1031 +/- 101 cpm/5 x 10(6) cells versus 112 +/- 15 cpm); with various concentrations of ACM, the increase in EPO binding appeared to parallel the recruitment of hemoglobin-producing cells. However, at 95% of growth inhibition, EPO binding remained constant while the percentage of differentiated cells decreased. Specific binding was reversible, saturable, and proportional to cell number; bound EPO was displaced by unlabeled EPO. Scatchard analysis of the equilibrium binding data suggested the existence of a single class of EPO receptors with an apparent Kd of 867 +/- 458 pM, corresponding to 400 +/- 142 receptors per cell. Affinity cross-linking of 125I-EPO using disuccinimidyl suberate followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis under reducing conditions demonstrated that EPO was preferentially cross-linked to a protein of approximately 116 kD. Taken together, these results demonstrate that, in addition to cytostatic properties, antitumor drugs such as ACM can modulate the expression of differentiation factor receptors on the surface of leukemic cells.

Induction by hemin of proliferation and of differentiation of progenitor erythroid cells responsible for erythropoietin.

Erythroid progenitor cells were obtained from rat fetal liver by immunolysis of the whole erythroid population with an antiserum directed against adult rat erythrocytes, followed by separation on a density gradient. Immediately after their isolation, these cells contained only minute amounts of globin mRNAs and their heme synthesis was negligible. In the absence of erythropoietin (Epo), they did not proliferate or differentiate. In the presence of Epo, they proliferated, synthetized heme and globins actively, accumulated large amounts of globin mRNAs, and developed hemoglobinized colonies in methylcellulose. Hemin, in concentrations of 5-100 microM, induced, in the absence of Epo, the proliferation and differentiation of these cells (e.g., accumulation of globin mRNAs, synthesis of heme and globins, and increased density of membrane antigens characteristic of the erythrocyte). Nevertheless, Epo and hemin actions were not superimposable: in methylcellulose, Epo induced the appearance of large (greater than or equal to 32 cells) hemoglobinized colonies in 48 h, whereas hemin induced smaller and fewer colonies in only 24 h. Succinylacetone (SA, inhibitor of heme synthesis) mostly prevented the effects of Epo on cell proliferation and differentiation; SA inhibition was relieved by hemin. Thus, hemin seems to intervene in erythroid differentiation as a factor of both proliferation and maturation.

Modulation of histidine decarboxylase activity and cytokine synthesis in human leukemic cell lines: relationship with basophilic and/or megakaryocytic differentiation.

In the present study, we show that UT7D1 cells, derived from the pluripotent cell line UT7, express high levels of histidine decarboxylase (HDC) mRNA spontaneously. These cells conserve the ability to differentiate into megakaryocytes upon stimulation with PMA, while greatly increasing their HDC activity. We provide evidence that enhanced HDC activity reflects the basophil rather than the megakaryocytic differentiation potential of UT7DI cells. Indeed, in addition to HDC mRNA, they express spontaneously several other mRNA coding for molecules present in basophils (FcepsilonRI, CCR3, IL-4Ralpha, IL-5Ralpha). Furthermore, the basophil antigen Bsp-1 is displayed on the surface of some UT7D1 cells in response to PMA concomitantly with increased histamine synthesis and mRNA expression of typical basophil-derived cytokines (IL-6, IL-4, and IL-13). Nevertheless, PMA cannot sustain the differentiation of this lineage, because mRNAs for basophil markers gradually diminish during long-term culture, whereas molecules associated with the megakaryocytic lineage remain prominent. In support of the notion that HDC activity is not related with megakaryopoiesis, we show that PMA-induced CD41 expression and PDGF transcription occurs in the K562 cells, though neither HDC mRNA nor any known basophil marker are expressed in these conditions. In contrast, all these markers are expressed in the basophilic leukemia cell line KU812F. Interestingly, the megakaryocytic cell line HEL produces also substantial amounts of histamine and expresses FcepsilonRI, thus revealing its basophil differentiation potential. HEL as well as KU812F need not be stimulated with PMA to react with Bsp-1 mAb, suggesting that they are more engaged into the basophil differentiation scheme than UT7D1. Other leukemic cell lines unrelated to the megakaryocyte or basophil lineage, like HL60 and U937 do neither synthesize histamine nor express basophil markers before or after PMA stimulation. To our knowledge, this is the first evidence for a factor-dependent cell line with megakaryocyte/basophil bipotentiality with which early stages of basophil commitment can be analyzed.

Murine erythroleukaemia cells (Friend cells) possess high-affinity binding sites for erythropoietin.

Murine erythroleukaemia cells represent erythroid precursors blocked near the CFU-E or proerythroblast stage. In contrast to their non-leukaemic equivalents, neither their proliferation nor their differentiation seems to be affected by erythropoietin. However, we show in this paper that both uncommitted and committed, benzidine-positive, cells bind iodinated erythropoietin. The binding is of high affinity (Kd = 490 +/- 160 pM) and reversible with a half-life of the complex of 77 +/- 19 min. The number of binding sites is low (300-600 per cell). In contrast the haematopoietic non-erythroid cell lines HL 60 and L 1210 and the myeloid-erythroid human cell line K 562 do not exhibit specific binding. If these binding sites represent true hormone receptors, their presence on a permanent cell line should facilitate erythropoietin receptor purification.

Glycosylation of the murine erythropoietin receptor.

Murine erythropoietin-responsive Rauscher Red 5-1.5 cells were used to determine the contribution of glycosylation to the size and function of the erythropoietin receptor. The half life of the receptors was determined to be 4 h. The number of receptors was not significantly decreased in cells treated for 48 h with inhibitors of glycosylation (tunicamycin, glucosamine or swainsonine) and their affinity was slightly enhanced in tunicamycin- or glucosamine-treated cells. Erythropoietin was cross-linked with two proteins of 104 and 86 kDa. Their molecular masses were not significantly reduced in cells treated with the glycosylation inhibitors. When immunoprecipitated cross-linked receptors were digested with endoglycosidases, the molecular masses of both proteins were only slightly modified giving values of 100 and 82 kDa. Thus we can conclude that the proteins cross-linked to erythropoietin are very weakly glycosylated.

Involvement of the p38 mitogen-activated protein kinase pathway in tissue inhibitor of metalloproteinases-1-induced erythroid differentiation.

We examined the role of the mitogen-activated protein (MAP) kinase pathway in tissue inhibitor of metalloproteinases-1 (TIMP-1)-mediated cellular effects in a human erythroleukemic cell line UT-7. We show that TIMP-1 induced both UT-7 cell erythroid differentiation and proliferation and tyrosine phosphorylation of many intracellular proteins. Using a panel of phosphospecific antibodies, we also demonstrate that phosphorylation of the p38 and c-Jun N-terminal kinases is increased by TIMP-1 whereas phosphorylation of extracellular signal-regulated kinase 1/2 is not induced. Moreover, inhibition of the p38 activity by SB203580 significantly reduces erythroid differentiation induced by TIMP-1, suggesting that the p38 MAP kinase pathway is involved in TIMP-1-induced erythroid differentiation.

Binding of glucocorticosteroids to hepatic erythropoietic cells of the rat fetus.

A role for glucocorticosteroids in the evolution of the fetal liver erythron in vivo has been proposed. If direct, such an intervention implies the presence of receptor sites in the cells of the erythroid cell lines. To test this hypothesis cell suspensions were prepared from liver haematopoietic tissue obtained from rat fetuses aged 13, 14, 17 and 20 days. Their composition was determined by light microscopy. The populations consisted essentially of cells of the erythroid line: juvenile cell types (progenitor and basophilic cells) predominated in tissue collected at 13 and 14 days of gestation, more mature types (polychromatic and acidophilic cells) became more abundant at 17 and 20 days of gestation. Suppressible binding of glucocorticosteroid at 4 degrees C was studied on these suspensions, using [3H]dexamethasone. Binding sites were found at all stages of gestation studied. The mean number per cell (for the entire erythroid population) was roughly 3600, 3500, 2300 and 1700 at 13, 14, 17 and 20 days of gestation respectively, without any change in the apparent equilibrium dissociation constant (Kd:5.5 X 10(-9) to 6.5 X 10(-9) mol/l). Suspensions containing essentially progenitor cells were prepared from erythropoietic cells from livers obtained at 14 and 16 days of gestation, using a rabbit immune serum against rat erythrocytes in the presence of an excess of complement. These cells had 1800 and 1600 receptor sites per cell respectively (same Kd as above). No receptors were found on circulating adult or fetal erythrocytes. The results strongly suggest that there is an uneven distribution of the number of the glucocorticosteroid binding sites per cell amongst the different cell types of the erythroid line. These sites were present in progenitors, increased in number in the basophilic cells and then progressively disappeared as the cells matured.