K-Ras is essential for normal fetal liver erythropoiesis

Ras activation following binding of erythropoietin (EPO) and kit-ligand (KitL) to their receptors is essential for the differentiation, proliferation, and survival of erythroid progenitors in vitro.^1-10  In mammals there exist 3 homologous Ras proteins, H-Ras, N-Ras, and K-Ras (reviewed in Scheele et al,¹¹  Ehrhardt et al,¹²  and Rebollo and Martinez¹³ ). While H-Ras^-/-, N-Ras^-/-, and doubly mutant H-Ras^-/-; N-Ras^-/- knockout mice have no overt hematopoietic cell phenotypes,^14,15 K-Ras^-/- mice die between days 12.5 and 14 of gestation with fetal liver defects and evidence of anemia.^16,17  However, when a heterogenous population of K-Ras^-/- fetal liver cells was plated in methylcellulose assays, only a slight decrease in erythroid progenitor cell frequency was observed.¹⁶  Based on these results, the anemia observed in K-Ras^-/- embryos was attributed to a defect in the fetal liver microenvironment. Thus, despite in vitro studies supporting a role for Ras in regulating erythroid cell function, in vivo studies have not identified a Ras isoform essential for erythropoiesis. Given these discrepancies, we performed a more detailed analysis of erythropoiesis in K-Ras^-/- embryos by testing erythroid cell differentiation with a newly developed flow cytometric assay¹⁸  and the erythroid colony-forming ability of a purified population of hematopoietic progenitor cells.

K-Ras^-/- mice were obtained from Jackson Laboratories (Bar Harbor, ME) in a 129/SV background. Studies were conducted with a protocol approved by the Indiana University Animal Care and Use Committee. The K-Ras allele was genotyped by polymerase chain reaction as previously described.¹⁶ K-Ras^+/- mice were mated to produce day-13.5 K-Ras^-/- and wild-type (WT) embryos. Fetal livers were isolated as previously described.¹⁹  Single-cell suspensions were prepared by pushing hepatic tissues through a 23-gauge needle.

Fetal liver touch preparations were performed as previously described¹⁹  from day-13.5 WT and K-Ras^-/- embryos and stained with Wright Giemsa (Dade Behring, Newark, DE). Photomicrographs of touch preparations were taken with an Olympus DP11 microscope (Melville, NY).

Recombinant murine KitL and granulocyte macrophage-colony stimulating factor (GM-CSF) were obtained from Peprotech (Rocky Hill, NJ), and EPO was obtained from Amgen (Thousand Oaks, CA). Erythroid burst-forming unit (BFU-E), erythroid colony-forming unit (CFU-E), and granulocyte-macrophage colony-forming unit (CFU-GM) assays were performed exactly as previously described.¹⁹ 

Fetal liver cells were incubated with 5 μg of a fluorescein isothiocyanate (FITC)–conjugated c-kit monoclonal antibody (Pharmingen, San Diego, CA) per 10⁶ cells, placed on ice for 30 minutes, pelleted, washed, and resuspended in phosphate-buffered saline. C-kit⁺ cells were purified by immunomagnetic bead enrichment as previously described.¹⁹ 

Sorted c-kit⁺ fetal liver cells were stained with annexin V–FITC (Pharmingen) exactly per manufacturer's protocol followed by flow cytometric analysis (FACS) as previously described.¹⁹  For survival assays, 10 000 c-kit⁺ fetal liver cells were plated in 96-well plates with no serum in the presence of varying concentrations of either EPO or KitL alone or in combination. After 48 hours in culture, cells were stained with annexin V–FITC followed by FACS analysis. For thymidine incorporation assays, 10 000 c-kit⁺ fetal liver cells were plated in 96-well plates in serum-free medium (X-VIVO 10; CAMBREX, Walkersville, MD) in the presence of varying concentrations of either EPO or KitL alone or in combination. After 48 hours in culture, cells were pulsed with tritiated thymidine (New Life Sciences, Boston, MA) for 16 to 24 hours, harvested on glass-fiber filters, and β emission was measured. In some experiments, freshly isolated c-kit⁺ cells were cultured in the presence of both KitL and EPO for 24, 48, and 72 hours to differentiate the cells to later stages of erythroid cell maturation. After culture at each time point, survival and proliferation assays were performed exactly as described in this section.

For the differentiation assay, fetal liver cells were stained with anti-CD71–FITC (Pharmingen), anti–Ter-119–phycoerythrin (PE; Pharmingen), and anti–c-kit–allophycocyanin (APC; Pharmingen) antibodies followed by FACS analysis as previously established.¹⁸ 

To determine the effect of K-Ras deficiency on fetal liver erythropoiesis, K-Ras^+/- mice were intercrossed and day-13.5 embryos were harvested. As previously shown,¹⁶ K-Ras^-/- embryos and fetal livers were small and pale compared with WT controls (data not shown). K-Ras^-/- day-13.5 fetal liver touch preparations demonstrated an increase in immature proerythroblasts and early basophilic erythroblasts and a striking decrease in late basophilic, chromatophilic, and orthochromatophilic erythroblasts (Figure 1A). Further, there were very few mature enucleated erythrocytes in K-Ras^-/- fetal livers (Figure 1A). In contrast, there was a continuum of erythropoiesis in WT fetal livers with mature enucleated erythrocytes identified throughout the fetal liver (Figure 1A).

Based on this observation, we next compared erythroid cell differentiation in vivo between K-Ras^-/- and WT fetal livers. We used a recently developed flow cytometric assay, which readily distinguishes various stages of erythroid cell differentiation by analyzing the cell surface expression of CD71 and Ter-119.¹⁸  Consistent with our phenotypic observations (Figure 1A), we observed a delay in terminal erythroid differentiation in K-Ras^-/- fetal livers compared with WT controls (Figure 1B). Specifically, K-Ras^-/- fetal livers had a marked increase in early progenitor and proerythroblasts (region 1 [R1]; 12% vs 6%) and in early basophilic erythroblasts (R2; 18% vs 6%) with a decrease in more mature late basophilic erythroblasts (R3; 41% vs 65%) and chromatophilic and orthochromatophilic erythroblasts (R4; 4% vs 12%; Figure 1B). To further characterize the delay in erythroid cell maturation observed in K-Ras^-/- fetal livers, we costained fetal liver cells with antibodies directed against CD71, Ter-119, and c-kit and analyzed the cells by FACS. Interestingly, c-kit⁺ cells were observed mostly in the R1 and R2 populations, which define early progenitors, proerythroblasts, and early basophilic erythroblasts, but there was no significant difference in the percentage of c-kit⁺ cells in the different gated populations (R1-R5) between the 2 genotypes (Figure 1C). However, since there is an increased percentage of cells in R1 and R2 in K-Ras^-/- fetal livers (Figure 1B), these data indicate that there is an overall increase in the percentage of c-kit⁺ cells in K-Ras^-/- fetal livers compared with WT controls, which is consistent with an increased percentage of early erythroid progenitors and a delay in erythroid maturation observed in K-Ras^-/- day-13.5 embryos. Collectively, these results argue that K-Ras is important for normal fetal liver erythroid cell differentiation in vivo.

We next assayed for early (BFU-E) and late (CFU-E) erythroid progenitors in WT and K-Ras^-/- fetal livers. The frequency of BFU-Es and CFU-Es in day-13.5 K-Ras^-/- livers were reduced compared with WT controls (Figure 2A). To directly test the effect of K-Ras deficiency on erythroid colony formation independent of the effects of the microenvironment, we performed BFU-E and CFU-E assays with equal numbers of sorted WT and K-Ras^-/- c-kit⁺ cells. To confirm that there were no differences in cell viability, an aliquot of sorted c-kit⁺ cells was stained with annexin V to identify apoptotic cells. K-Ras^-/- and WT c-kit⁺ cells had a similar percentage of annexin V–positive cells (data not shown). A marked reduction in the frequency of both BFU-Es and CFU-Es was observed in K-Ras^-/- c-kit⁺ cells compared with WT controls (Figure 2B). Since the number of both BFU-Es and CFU-Es generated from K-Ras^-/- c-kit⁺ cells were reduced, these data argue that K-Ras activation is important for the proliferation and/or survival of both early and late erythroid progenitors in response to EPO and KitL. Interestingly, the number of myeloid colonies (CFU-GMs) generated in response to GM-CSF from K-Ras^-/- c-kit⁺ fetal liver cells was reduced compared with WT controls (30.17 ± 2.29 vs 184.13 ± 13.08; n = 5; P < .001), indicating that K-Ras activation may also be essential for the development of other hematopoietic cell lineages. Nevertheless, these results argue that an intrinsic K-Ras^-/- hematopoietic progenitor cell defect contributes at least in part to reduction of erythroid progenitors and mature erythrocytes in day-13.5 K-Ras^-/- fetal livers.

Since these data establish that K-Ras is essential for erythropoiesis in vivo, we next tested whether the K-Ras^-/- erythroid phenotype was also linked to either a decrease in proliferation and/or survival of hematopoietic progenitors in response to KitL or EPO. We performed thymidine incorporation assays using K-Ras^-/- and WT c-kit⁺ cells. Remarkably, K-Ras^-/- c-kit⁺ cells displayed a 50% to 60% decrease in proliferation in response to either EPO or KitL alone and EPO and KitL in combination (Figure 2C). Similar differences in proliferation between the 2 experimental genotypes were obtained when WT and K-Ras^-/- c-kit⁺ cells were differentiated to later stages of erythroid cell maturation in vitro and thymidine incorporation assays were performed with the different populations of erythroid progenitors in response to either EPO or KitL alone or in combination (data not shown). One explanation for this result is that K-Ras^-/- c-kit⁺ cells were undergoing an accelerated rate of apoptosis. To test this possibility, we compared the survival of K-Ras^-/- and WT c-kit⁺ cells in response to either EPO and/or KitL in the absence of serum. Both K-Ras^-/- and WT c-kit⁺ cells had a similar percentage of annexin V–positive cells at the initiation of the experiment (data not shown). After 24, 48, and 72 hours in culture, no differences were detected in the percentage of K-Ras^-/- or WT c-kit⁺ apoptotic cells (annexin V–positive) in response to either EPO or KitL alone or in combination (data not shown). Thus, these data indicate that K-Ras activation is essential for the proliferation of erythroid progenitor cells in response to both EPO and KitL.

These studies establish a previously unrecognized role for K-Ras in regulating fetal erythropoiesis in vivo. Specifically, K-Ras is essential for the differentiation of erythroid progenitor cells to late basophilic erythroblasts and the proliferation of hematopoietic progenitors in response to EPO and KitL. Recent data demonstrate that Ras isoforms undergo differential posttranslational modification (reviewed in Ehrhardt et al,¹²  Bar-Sagi,²⁰  and Hancock et al²¹ ) and compartmentalization within the cell (reviewed in Prior and Hancock,²²  Hancock,²³  and Chiu et al²⁴ ) and that individual Ras isoforms can differentially activate discrete signaling pathways (reviewed in Ehrhardt et al¹² ). Our data provide evidence that single Ras isoforms may also provide unique signals in hematopoietic cells in vivo.