Functional but abnormal adult erythropoiesis in the absence of the stem cell leukemia gene

Abstract

Previous studies have indicated that the stem cell leukemia gene (SCL) is essential for both embryonic and adult erythropoiesis. We have examined erythropoiesis in conditional SCL knockout mice for at least 6 months after loss of SCL function and report that SCL was important but not essential for the generation of mature red blood cells. Although SCL-deleted mice were mildly anemic with increased splenic erythropoiesis, they responded appropriately to endogenous erythropoietin and hemolytic stress, a measure of late erythroid progenitors. However, SCL was more important for the proliferation of early erythroid progenitors because the predominant defects in SCL-deleted erythropoiesis were loss of in vitro growth of the burst-forming erythroid unit and an in vivo growth defect revealed by transplant assays. With respect to erythroid maturation, SCL-deleted proerythroblasts could generate more mature erythroblasts and circulating red blood cells. However, SCL was required for normal expression of TER119, one of the few proposed target genes of SCL. The unexpected finding that SCL-independent erythropoiesis can proceed in the adult suggests that alternate factors can replace the essential functions of SCL and raises the possibility that similar mechanisms also explain the relatively minor defects previously observed in SCL-null hematopoietic stem cells.

FIG. 1.

SCL is not required for adult erythropoiesis. (A) Peripheral blood counts of control (+/Δ) and SCL-deleted (−/Δ) mice following administration of poly(I:C). A minimum of six mice for each time point were used to calculate the means and standard deviations. (B) Spleens from mice 4 weeks after administration of poly(I:C) were analyzed by Southern blotting for the presence of wild-type (+), null (−), loxP-flanked (loxP), and loxP-deleted (Δ) SCL alleles. (C) Southern blot of purified Mac-1⁺ (M) and TER119⁺ (E) cells from SCL-deleted mice. (D) Expression of SCL mRNA measured by real-time PCR in bone marrow cells from control (+/Δ) and SCL-deleted (−/Δ) mice harvested 4 weeks after administration of poly(I:C) compared with bone marrow cells from wild-type mice. The means and standard deviations were calculated for three mice of each genotype. ND, not detected. (E) Comparison of cellular content of spleens from control (+/Δ) and SCL-deleted (−/Δ) mice at least 4 weeks after poly(I:C). Cell types were counted for four mice of each genotype. ProE, proerythroblast; E, normoblast; My, granulocyte; Ly, lymphocyte. *, P < 0.05; **, P < 0.01.

FIG. 2.

SCL is not required for an erythroid stress response. (A) Control mice (+/Δ) and SCL-deleted mice (−/Δ) were injected with saline (−) or a single dose of darbepoetin alfa (+), and hematocrit (HCT) was measured on day 8. The means and SD of samples from six mice for each group are shown. (B) Mice injected with darbepoetin alfa were bled each day to assess the reticulocyte response. (C) Southern blot of bone marrow (B) and spleen (S) from darbepoetin alfa-treated mice probed for wild type (+), null (−), loxP-flanked (loxP), and loxP-deleted (Δ) SCL alleles. (D) Hematocrit levels in mice (n = 3) injected with phenylhydrazine on days 0 and 1.

FIG. 3.

SCL is not required for maturation of proerythroblasts. (A) Expression of CD71 and TER119 on bone marrow cells from control (+/Δ) and SCL-deleted (−/Δ) mice. The regions corresponding to erythroid maturation (regions I to IV) according to Socolovsky et al. (32) are shown. Note the approximate fourfold decrease in intensity of staining with TER119 compared with control cells and the appearance of a CD71^pos TER119^neg population (region 0). The proportion of red blood cells in each region is shown in parentheses and was calculated from six mice of each genotype. (B) Cytospins of sorted CD71^pos TER119^neg cells (region 0) and CD71^pos TER119^pos (region II) were stained with May-Grunwald-Giemsa. (C) Annexin V staining of control erythroid cells (+/Δ) and SCL-deleted erythroid cells (−/Δ) from region 0 (SCL-deleted only) and region II. The proportion of annexin-positive cells is shown. (D) Gene expression of purified cells from regions 0 and II of SCL-deleted bone marrow compared with control erythroid cells from region II. (E) Peripheral blood stained with TER119 demonstrating the presence of TER119^neg erythrocytes in SCL-deleted mice.

FIG. 4.

Absence of BFU-E in SCL-deleted mice. (A) Bone marrow cells from control (+/Δ; solid bars) and SCL-deleted mice (−/Δ; open bars) at least 28 days after poly(I:C) were grown in methylcellulose to determine erythroid progenitor cell numbers (CFU-E and BFU-E). The mean numbers and SD for three independent experiments are shown. (B) CFU-E numbers from control and SCL-deleted mice 5 days after administration of phenylhydrazine. (C) CD34 and FcγR expression on Lin⁻ Sca-1⁻ IL-7Rα⁻ c-kit⁺ bone marrow cells from control (+/Δ) and SCL-deleted (−/Δ) mice. The regions used to calculate the number of common myeloid progenitors (CMP), granulocyte-macrophage progenitors (GMP), and MEPs relative to the total number of bone marrow mononuclear cells are indicated in the dot plots, and the mean percentage of BMMC and standard deviation were calculated from results from eight mice in each genotype.

FIG. 5.

Erythropoiesis generated from SCL-deleted CFU-S12. (A) Representative CFU-S12 generated from control (+/Δ) and SCL-deleted (−/Δ) bone marrow cells 6 days or 28 days after poly(I:C) treatment. (B) Individual CFU-S12 from control and SCL-deleted mice were picked and analyzed by Southern blot for the presence of wild-type (+), null (−), loxP-flanked (loxP), and loxP-deleted (Δ) SCL alleles. (C) Individual CFU-S12 colonies from control and SCL-deleted bone marrow cells were analyzed by flow cytometry for the expression of CD71 and TER119. Regions 0 to III are shown. (D) Expression of annexin V on cells in regions I and II. The mean percentage of positive cells is shown (three CFU-S12 colonies for each genotype).

FIG. 6.

Transplantation of SCL-deleted erythropoiesis. (A) Schematic of transplant assay. Donor cells (SCL deleted; −/Δ) were distinguished from competitor and recipient cells (SCL wild type; +/+) by single (β^s) and diffuse (β^d) hemoglobin alleles. (B) Expression of CD71 and TER119 on bone marrow cells from recipient mice 16 weeks after transplantation with SCL-deleted bone marrow cells alone (100%) or with a mixture of 95% SCL-deleted and 5% competitor bone marrow cells. (C) Hb electrophoresis of recipient mice 16 weeks after transplant with 100% or 95% SCL-deleted bone marrow cells. (D) Hb electrophoresis of recipient mice 4 and 8 weeks after transplant with 95% SCL-deleted bone marrow cells.