Recombinant human erythropoietin induces proliferation and Ca(2+)-influx in specific leukocyte subpopulations of rainbow trout (Oncorhynchus mykiss) blood and head kidney cells.

In mammals, erythropoietin regulates the development and differentiation of erythrocytes. Although hematopoietic cells of bony fish correspond in their ontogeneic development, morphology, and function to their mammalian counterparts, an erythropoietin (EPO)-like molecule has not been identified. In this study we present evidence for a mitogenic response of blood and head kidney leukocytes of rainbow trout after stimulation by recombinant human EPO (rhu EPO). The modulation of cellular activities is accompanied by the induction of DNA-binding activities in nuclear extracts of these cells. In addition, flow cytometric analysis of intracellular Ca2+ concentrations revealed a long-lasting and rhu EPO dose-dependent increase, which was shown to be abrogated by cross-aggregation of surface IgM using anti-trout-IgM monoclonal antibodies (mabs). In flow cytometric dual-labeling experiments using rhu EPO/anti-EPO antiserum and mabs specific for trout leukocyte subpopulations, it was shown that a subpopulation of trout B-cells binds rhu EPO. Moreover, in a modified Ca2+ activation assay, it was demonstrated that this blood B-cell subpopulation is the rhu EPO responder population. In conclusion, the data suggest the existence of EPO-binding receptors in trout that are able to trigger Ca(2+)-independent intracellular signaling in hematopoietic cells of head kidney and Ca(2+)-dependent activation of a subpopulation of B-lymphocytes.

Activation of the transcription factor NF-kappaB by the erythropoietin receptor: structural requirements and biological significance.

The transcription factor nuclear factor kappa B (NF-kappaB) has been implicated in the regulation of genes mainly involved in inflammation and immune response. We analysed the role of NF-kappaB in signalling pathways induced by the hematopoietic growth factor erythropoietin (EPO). Our data, obtained by electrophoretic mobility shift assays (EMSA) and reporter gene assays, show that the intracellular domain of the EPO receptor (EPOR) transmits signals leading to the activation of NF-kappaB. Studies employing an inhibitor specific for the EPOR-associated tyrosine kinase JAK2 suggest that JAK2-dependent pathways are not involved. The induction of an NF-kappaB-triggered reporter gene construct was inhibited by cotransfection of dominant negative forms of the src kinase Lyn, but not by dominant negative JAK2. Using epidermal growth factor (EGF)/EPOR hybrids containing mutant forms of the EPOR intracellular domain, we were able to further define the critical structures for the induction of NF-kappaB. The data show that although the activity of JAK2 seems to be dispensable, its association to the receptor, as well as the phosphorylation of membrane proximal tyrosine residues, are essential. Furthermore, the functional analysis of different receptor forms revealed a correlation of the abilities to induce NF-kappaB activity and to generate antiapoptotic signals.

Maturation of erythroid cells and erythroleukemia development are affected by the kinase activity of Lyn.

This study examined the impact of the tyrosine kinase Lyn on erythropoietin-induced intracellular signaling in erythroid cells. In J2E erythroleukemic cells, Lyn coimmunoprecipitated with numerous proteins, including SHP-1, SHP-2, ras-GTPase-activating protein, signal transducers and activators of transcription (STAT) 5a, STAT5b, and mitogen-activated protein kinase; however, introduction of a dominant-negative Lyn (Y397F Lyn) inhibited the interaction of Lyn with all of these molecules except SHP-1. Cells containing the dominant-negative Lyn displayed altered intracellular phosphorylation patterns, including mitogen-actiated protein kinase, but not erythropoietin receptor, Janus-activated kinase (JAK) 2, or STAT5. As a consequence, erythropoietin-initiated differentiation and basal proliferation were severely impaired. Y397F Lyn reduced the protein levels of erythroid transcription factors erythroid Kruppel-like factor and GATA-1 up to 90%, which accounts for the inability of J2E cells expressing Y397F Lyn to synthesize hemoglobin. Although Lyn was shown to bind several sites on the cytoplasmic domain of the erythropoietin receptor, it was not activated when a receptor mutated at the JAK2 binding site was ectopically expressed in J2E cells indicating that JAK2 is the primary kinase in erythropoietin signaling and that Lyn is a secondary kinase. In normal erythroid progenitors, erythropoietin enhanced phosphorylation of Lyn; moreover, exogenous Lyn increased colony forming unit-erythroid, but not burst forming uniterythroid, colonies from normal progenitors, demonstrating a stage-specific effect of the kinase. Significantly, altering Lyn activity in J2E cells had a profound effect on the development of erythroleukemias in vivo: the mortality rate was markedly reduced and latent period extended when either wild-type Lyn or Y397F Lyn was introduced into these cells. Taken together, these data show that Lyn plays an important role in intracellular signaling in nontransformed and leukemic erythroid cells.

Induction of erythroid proliferation and differentiation by a trophoblast-specific cytokine involves activation of the JAK/STAT pathway.

Pregnancy is characterized by increased erythropoiesis within maternal and fetal compartments. The placenta has been shown to produce factors that stimulate erythropoiesis but convincing evidence for placental production of erythropoietin (EPO) is still lacking. Prolactin-like protein E (PLP-E) was recently found to stimulate expression of the adult beta major globin gene in mouse erythroleukemia cells. Here we demonstrate that PLP-E transiently expressed in COS-7 cells stimulates proliferation and erythroid differentiation of murine and human erythroid progenitor cell lines. Electrophoretic mobility shift assays were used to show the activation of STAT5 by PLP-E in the human erythroid cell line TF1. Furthermore, we compared the effects of PLP-E on murine myeloid FDCP1 cells which do not express EPO receptors (EPORs) with effects on cells genetically engineered to express functional EPORs. We provide evidence that PLP-E-dependent proliferation and STAT5 activation is independent of the expression of the EPOR. Taken together, these data suggest that PLP-E acts on specific receptors of erythroid-committed murine and human cells by the activation of intracellular signaling pathways promoting cell growth and differentiation.

Determination of cytokine mRNA-expression in term human placenta of patients with gestational hypertension, intrauterine growth retardation and gestational diabetes mellitus using polymerase chain reaction.

OBJECTIVE: Our objective was to test the hypothesis, that pregnancy-related diseases are going along with changes in cytokine mRNA-expression at the placental site, either as a part of a pathological process or in connection with regulatory mechanisms induced by disturbances at the feto-maternal interface resulting from previous pathological changes--in the sense of counterregulation. MATERIAL AND METHODS: The cytokines chosen for this investigation are known to 1.) be expressed in the human placental tissue, 2.) to be involved in immunological processes and 3.) the regulation of growth and differentiation processes of different cell types of the placenta or decidua, 4.) to play a role in the angiogenesis at the feto-placental interface and 5.) to be involved in pathological processes in other human diseases. 32 samples derived from term human placentas were examined for messenger RNA levels of interleukin 1 alpha (II-1 alpha), tumor necrosis factor-alpha (TNF-alpha), platelet derived growth factor-A chain (PDGF-A), platelet derived growth factor-B chain (PDGF-B), and platelet derived growth factor receptor (PDGF-R) using a semiquantitative reverse transcriptase (RT) polymerase chain reaction (PCR) protocol. To calibrate samples in our procedure, beta-actin mRNA (messenger ribonucleic acid) known as a "house keeping" gene was proven to be constantly expressed. The sample-groups consisted of normal pregnancies (n = 8), gestational hypertension (GH, n = 7), intrauterine growth retardation (IUGR, n = 6), gestational diabetes mellitus (GDM, n = 5), and gemini (n = 3 x 2). RESULTS: Throughout the 32 samples, a significant correlation between PDGF-A and PDGF-R expression, PDGF-A and TNF-alpha expression was stated (p = 0.007). Compared with the pattern of expression in normal placentas, placentas of growth retarded pregnancies had higher Il-1 alpha mRNA (p = 0.016), PDGF-A (p = 0.029) and PDGF-B (p = 0.001) levels. The samples of the gestational hypertension group and placentas of patients with gestational diabetes displayed a significantly stronger PDGF-R mRNA signal (p = 0.0029 and p = 0.008). CONCLUSIONS: Though these marked differences in cytokine mRNA levels between clinical groups were statistically proven, clear correlation of these differences with clinical data was not found.

Perinuclear localization of the protein-tyrosine phosphatase SHP-1 and inhibition of epidermal growth factor-stimulated STAT1/3 activation in A431 cells.

The SH2 domain protein-tyrosine phosphatase SHP-1 has been shown earlier to bind to the epidermal growth factor receptor and to have the capacity for receptor dephosphorylation. New bi- and tricistronic expression vectors (pNRTIS-21 and pNRTIS-33, respectively) based on the tetracycline system were constructed and employed to generate stable cell lines with inducible expression of SHP-1. Inducible overexpression of SHP-1 in A431 cells led to attenuation of epidermal growth factor (EGF) receptor autophosphorylation and of EGF-induced DNA binding of 'signal transducers and activators of transcription' (STAT) 1 and 3. SHP-1 was localized in the cytoplasm with an enrichment in the perinuclear compartment. Association of SHP-1 with perinuclear structures may form the basis for a partial cofractionation with nuclei observed in different types of transfected cells and also with endogenous SHP-1 in U-937 cells. Treatment of SHP-1-overexpressing A431 cells or of HaCaT human keratinocytes expressing SHP-1 endogenously with the Ca2+-ionophore A23187 resulted in partial nuclear accumulation of SHP-1. Thus, SHP-1 may interact with substrates or regulatory proteins in perinuclear or nuclear structures.

Activation of STAT5 during EPO-directed suppression of apoptosis.

The ligand-dependent activation of the JAK/STAT (Januskinase/Signal Transducer and Activator of Transcription) pathway has been implicated in the explanation of cytokine-specific regulation of gene expression. Previous studies have reported conflicting results on the role of the transcription factor STAT5 in erythropoietin (EPO)-induced cellular responses. In this study we focused on the functional importance of STAT5 docking sites in the intracellular EPO receptor (EPOR) domain for the mediation of antiapoptotic activities. We demonstrate that EPO-dependent survival of erythroleukemic cell lines is accompanied by sustained STAT5 DNA-binding activity. The role of single tyrosine residues was dissected by the analysis of myeloid FDCP-1 cells stably expressing mutant EPOR proteins. The data show that receptors having a high potential to mediate antiapoptotic signals also effectively activate STAT5, whereas receptors lacking STAT5 docking sites are diminished in both activities. We conclude that the transcription factor STAT5 is functionally implicated in the EPO-dependent survival of cells.

High concentration of soluble HLA-DR in the synovial fluid: generation and significance in "rheumatoid-like" inflammatory joint diseases.

In the search for its role in inflammatory joint diseases, soluble HLA-DR (sHLA-DR) was quantitated in 72 synovial fluids (SF) by a newly established immunoenzyme assay. Unlike other soluble receptors which accumulated only moderately (sCD25, sCD4) or negligibly (sHLA class I, sCD8) in the SF, SF sHLA-DR levels exceeded serum levels by up to 3 orders of magnitude and varied disease dependently from "control" values (traumatic synovitis and osteoarthritis: 9.9 +/- 6.1 ng/ml). Clear-cut different SF sHLA-DR values in HLA-DR-associated "rheumatoid-like" (136.5 +/- 130.0 ng/ml) vs HLA-B27-associated "spondylarthropathy-like" arthritic forms (28.4 +/- 29.1 ng/ml) were most significant comparing oligoarticular juvenile chronic arthritis type I (147.6 +/- 112.6 ng/ml) and type II (3.3 +/- 1.1 ng/ml), thus offering a new classification marker. Also ex vivo, large amounts of sHLA-DR were released spontaneously by SF mononuclear cells and found to be related to the T-cell activation state. SF sHLA-DR may be shed in large complexes or micelles, as it eluted mainly at >450 kDa on gel filtration. Western blotting revealed that the majority of SF sHLA-DR consisted of full-length alpha- and beta-chains. Minor fractions of smaller sized antigens seemed to be generated by proteolytic cleavage rather than by alternative splicing, since only minute amounts of HLA-DRB mRNA lacking the transmembrane exon could be amplified by RT-PCR. Distinct forms of high-dose sHLA-DR, able to provoke rather than to suppress T-cell responses, are discussed as contributing to some HLA-DR disease association.

cDNA cloning and functional analysis of a truncated STAT5a protein from autonomously growing FDCP-1 cells.

The transcription factor STAT5 is activated by multiple hematopoietic cytokine receptors and has been implicated in the induction of cellular processes such as differentiation, proliferation and antiapoptotic activities. Here, we report cloning of the cDNA and characterization of a mutant STAT5a protein that is expressed in interleukin-3 (IL-3)-independently growing FDCP-1 cells. Analysis of the cDNA revealed a deletion of both the transactivation and the SH2 domains. Stable expression of the protein in parental IL-3-dependent cells results in elevated DNA binding activity of wild type (WT)-STAT5 in the nucleus, enhanced growth rates and a reduced susceptibility to undergo apoptosis after withdrawal of IL-3. Although the protein is not present in DNA/protein complexes in the nucleus, we observed pronounced effects on IL-3-induced signal transduction. The results suggest competition of the mutant protein with cytosolic mechanisms regulating STAT5 activity. In conclusion, the data support the hypothesis of an involvement of STAT5 in mitogenic and antiapoptotic signaling.

Trypanosoma brucei brucei induces interferon-gamma expression in rat dorsal root ganglia cells via a tyrosine kinase-dependent pathway.

Dorsal root ganglia (DRG) were shown to express neuronal interferon (IFN)-gamma, which supports Trypanosoma brucei brucei growth. The ability of a trypanosome-derived factor (TLTF) to activate DRG to neuronal IFN-gamma secretion was investigated, together with the signaling pathway that might be involved during this process. Immunohistochemical staining revealed expression of neuronal IFN-gamma on stimulation with TLTF, which was blocked with the tyrosine protein-kinase inhibitor, tyrphostin A47. Western blot was used to analyze DRG lysates prepared at different time points after stimulation with TLTF. A tyrosine-phosphorylated protein induced at 15 min was seen as a band of 120-150 kDa, followed by a decrease to control levels after 30 min. A47 greatly suppressed the TLTF-induced tyrosine protein kinase activity. In addition, evidence suggesting that the transcription factor STAT-1 may play a key role in the TLTF signaling pathway was provided by the blocking effects of A47 on STAT-1 translocation to the nucleus.

SHP1 protein tyrosine phosphatase negatively modulates erythroid differentiation and suppression of apoptosis in J2E erythroleukemic cells.

The SH2 domain-containing tyrosine phosphatase SHP1 is known to play a crucial role in the regulation of hematopoiesis. It has been shown previously that SHP1 associates with the activated erythropoietin receptor (EPOR) and negatively regulates mitogenic signaling. To further elucidate the role of SHP1 in erythropoietin (EPO)-induced cellular responses we employed J2E erythroleukemic cells as a model for erythroid maturation and cytokine-triggered suppression of apoptosis. Our data indicate that overexpressed SHP1 inhibits both EPO-induced differentiation as well as prevention of apoptosis. The specific signaling pathways responsible are not unraveled so far. Therefore, we analyzed the involvement of SHP1 in two established EPO-stimulated pathways, the JAK/STAT and the MAP kinase cascades, by transient coexpression of reporter constructs containing binding sites for transcription factors targeted by these pathways and a SHP1 cDNA. Both pathways are inhibited by SHP1 as indicated by the lower induction of reporter gene activity. In conclusion, SHP1 regulates the transcriptional activity stimulated by the EPO-induced JAK/STAT and MAPK pathways and is involved in the signaling machinery responsible for erythroid differentiation and suppression of apoptosis.

Interferon-alpha inhibits proliferation of Ba/F3 cells by interfering with interleukin-3 action.

Interferons (IFNs) are potent inhibitors of cell proliferation that are used for the treatment of several haematological malignancies. The mechanisms through which IFNs exert their antiproliferative effects on target cells, however, are largely unknown. Here we show that IFN-alpha, in murine Ba/F3 cells, directly interferes with the action of the essential mitogen interleukin (IL)-3. In transiently transfected Ba/F3 cells, IFN-alpha efficiently inhibited the IL-3-stimulated expression of a luciferase reporter construct, GAS-luc, that is activated through the JAK2/STAT5 pathway. Electrophoretic mobility shift assays and Northern blot experiments, however, revealed that neither the IL-3-induced DNA binding of STAT5 nor the transcription of the STAT5-dependent genes oncostatin-M, pim-1 and c-fos were suppressed by IFN-alpha, suggesting that the diminished expression of the luciferase protein was due to a direct inhibition of IL-3-stimulated protein synthesis. This hypothesis was supported by the observation that IFN-alpha, even though it had no effect on the transcription of the c-fos gene, efficiently suppressed the IL-3-dependent expression of the c-Fos protein. Furthermore, our results indicate that IFN-alpha induced an overexpression of the double-stranded RNA-activated protein kinase (PKR), an enzyme that inhibits protein synthesis through the phosphorylation and inactivation of the eukaryotic initiation factor-2. Therefore, we hypothesize that IFN-alpha, in Ba/F3 cells, interrupts IL-3-dependent mitogenic signals, at least in part, through the suppression of protein synthesis and that induction of PKR activity may play a pivotal role in this process.

Role of STAT5 in interferon-alpha signal transduction in Ba/F3 cells.

In interferon-alpha (IFN-alpha) signalling, the essential role of the transcription factors STAT1 and STAT2 is well established. In contrast, the involvement of other STAT proteins, including STAT5, is much less well understood. Here we show that, in IFN-alpha-responsive Ba/F3 cells, this cytokine stimulates the DNA-binding of STAT5A and B but that IL-3 is a much more potent activator of both STAT5 isoforms. A stably expressed dominant-negative mutant of JAK2 suppressed the IL-3- but not the IFN-alpha-dependent DNA binding of STAT5, suggesting independent mechanisms of its activation. Northern blots revealed that IL-3 strongly induced the expression of two STAT5-regulated genes, pim-1 and oncostatin-M, whereas IFN-alpha had a weak stimulatory effect on pim-1 expression only. In summary our results suggest that, despite the capability of IFN-alpha to stimulate DNA binding of STAT5, this transcription factor does not play a pivotal role in IFN-alpha signalling in Ba/F3 cells.

Tyrosine kinases are required for interferon-gamma-stimulated proliferation of Trypanosoma brucei brucei.

The tyrosine kinase activity of Trypanosoma brucei brucei upon stimulation with interferon-gamma (IFN-gamma) was investigated. IFN-gamma induced a rapid and strong increase of tyrosine phosphorylation of several cellular proteins that reached maximum after 5 min and was followed by a decrease to control levels after 120 min. The tyrosine kinase-specific inhibitor tyrphostin A47 at a concentration of 10(-6) M reduced IFN-gamma-induced protein phosphorylation. In vitro application of 10(-6) M tyrphostin A47 to the trypanosome cultures caused a significant reduction of [3H]thymidine uptake by IFN-gamma-stimulated trypanosomes. In animals, 2 x 0.5 mg of tyrphostin A47 (injected intraperitoneally) caused a significant reduction of parasite growth compared with the vehicle dimethyl sulfoxide or the inactive compound tyrphostin A1. In conclusion, tyrosine kinases are strongly up-regulated in IFN-gamma-stimulated T. b. brucei, and specific tyrosine kinase inhibitors can prevent trypanosome growth in vitro and in vivo.

Involvement of phosphatidylinositol 3-kinase in the mediation of erythropoietin-induced activation of p70S6k.

We have previously shown that, in HCD-57 cells, erythropoietin (EPO) induces a biphasic activation of the ribosomal S6 kinase p70S6k, an enzyme playing a key role in the regulation of cell cycle progression. Here we present evidence that p70S6k is activated through both phosphatidylinositol (PI) 3-kinase-dependent and independent pathways: whereas the early phase of EPO-dependent stimulation of p70S6k activity was strongly suppressed by the potent PI 3-kinase inhibitor wortmannin, late phase was much less affected. The dose-dependent inhibition of cell growth by wortmannin indicates an important role of PI 3-kinase in the mediation of EPO-induced cell proliferation. Furthermore, our data suggest that the EPO-receptor-associated tyrosine kinase JAK2 is not essentially involved in the mediation of EPO-induced p70S6k activation.

Prevention of apoptosis in J2E erythroid cells by erythropoietin: involvement of JAK2 but not MAP kinases.

The J2E erythroid cell line, transformed by retroviral v-raf/v-myc oncogenes, proliferates and differentiates in response to erythropoietin. Here we show that J2E cells undergo apoptosis rapidly after serum withdrawal and that only erythropoietin of seven growth factors tested, could protect the cells from death. The role of JAK2 and MAP kinases in transmitting viability signals initiated by erythropoietin was examined in these cells. Despite constitutive raf kinase activity, phosphorylation of MAP kinases fell after serum withdrawal. However, an antisense oligonucleotide strategy revealed that JAK2, but not the MAP kinases, was involved in transmitting signals to maintain the viability of J2E cells. Several cell cycle proteins and transcription factors were also studied; although c-jun rose sharply during apoptosis, erythropoietin could not suppress this increase. It was concluded that erythropoietin-induced protection from apoptosis involved JAK2, but not MAP kinases or c-jun.

Requirement for JAK2 in erythropoietin-induced signalling pathways.

Erythropoietin (EPO) exerts its activities by the induction of multiple signalling pathways through interaction with the erythropoietin receptor (EPOR). Previous studies have suggested that the Ras/MAP kinase as well as the JAK/STAT signalling cascades play significant roles in the induction of EPO-responsive genes. Here we show that, in HCD-57 erythroleukemic cells, both pathways are activated by EPO in a dose-dependent manner with similar sensitivities and kinetics. The activation of signalling molecules is closely related to the proliferative status of the cells. Using an antisense strategy, we were able to show that the downregulation of the JAK2 protein level in HCD-57 cells results in a distinct reduction of the ability to induce not only STAT5 DNA-binding, but also MAP kinase activity. Our results thus provide evidence for a significant contribution of the cytosolic tyrosine kinase JAK2 to the EPO-induced activation of the Ras/MAP kinase cascade.

Inhibition of proliferation but not erythroid differentiation of J2E cells by rapamycin.

During erythropoiesis, replication and maturation are tightly coupled processes. Here, we show that the immunosuppressant rapamycin inhibited basal- as well as erythropoietin-stimulated proliferation of the erythroid cell line J2E. In addition, it enhanced the antiproliferative effect of sodium butyrate. Although rapamycin suppressed erythroid cell division, it did not affect terminal differentiation induced by erythropoietin or sodium butyrate. The proliferative status of J2E cells correlated well with the activity of the ribosomal S6 kinase p70S6k, an enzyme effectively blocked by rapamycin. It was concluded from this study that erythroid maturation proceeded normally despite the rapamycin-induced inhibition of mitosis and of p70S6k activity. These data provide further evidence that separate signalling pathways for proliferation and differentiation exist in erythroid cells.

Regulation of the erythropoietin receptor and involvement of JAK2 in differentiation of J2E erythroid cells.

In response to erythropoietin, J2E cells proliferate and differentiate into mature hemoglobin-producing erythroid cells. Here we show that following hormonal stimulation, between 10 and 17 proteins, including the erythropoietin receptor and JAK2, were tyrosine phosphorylated immediately after exposure to the hormone. Although the receptor was only phosphorylated to 15% of its maximum with 0.1 unit/ml erythropoietin, this was sufficient to induce peak hemoglobin synthesis. The importance of JAK2 to J2E cell maturation was demonstrated by inhibiting JAK2 protein synthesis with antisense oligonucleotides; not only was erythropoietin-stimulated mitogenesis inhibited by this procedure, but differentiation was also suppressed. In addition, the activation of STAT5 paralleled the kinetics of receptor phosphorylation. During differentiation, 94% decrease in surface erythropoietin receptors was detected 48 h after ligand binding, but transcription of the receptor gene, mRNA steady-state levels, protein content, and translation rates did not alter with hormonal stimulation. We concluded from these experiments that (a) sub-maximal receptor phosphorylation is sufficient for differentiation to proceed; (b) JAK2 is required for erythropoietin-induced cell division and maturation; and (c) post-translational processing, or translocation, play important roles in controlling surface erythropoietin receptor numbers.

Truncated erythropoietin receptor in a murine erythroleukemia cell line.

The Friend spleen focus forming virus produces a 55 kDa envelope glycoprotein which associates with the erythropoietin receptor. We compared the erythropoietin receptor in Friend virus transformed murine erythroleukemic F4N and 707 cell lines with the J2E erythroid line generated by the J2 retrovirus. Reverse transcriptase PCR was used to determine transcript size. Erythropoietin receptor cDNAs were then sequenced and protein products analysed by Western blotting and immunoprecipitation. We show here that the F4N murine erythroleukemic cell line had an enlarged erythropoietin receptor mRNA. In contrast, the 707 and J2E cell line had normal sized transcripts for the receptor. Sequence analysis of the receptor in F4N cells revealed that introns which separate the exons coding for the cytoplasmic domain of the receptor were retained in these transcripts. As a consequence, a premature stop codon had been introduced, leaving only four amino acids in the intracellular portion of the receptor molecule. The normal erythropoietin receptor is approx. 66-70 kDa, but immunoprecipitation of [35S]methionine/cysteine labelled cell lysates with an antibody to the amino-terminus of the erythropoietin receptor identified a truncated 37 kDa protein in F4N cells. Despite the severe carboxy-terminal truncation of the erythropoietin receptor, F4N cells continued to proliferate like the other murine erythroleukemia cell lines. This study shows that failure to remove introns from the erythropoietin receptor mRNA in F4N cells has resulted in the production of a smaller protein with virtually no cytoplasmic domain.