C/EBPalpha binds and activates the PU.1 distal enhancer to induce monocyte lineage commitment

Abstract

The PU.1 gene contains a 237-base pair distal enhancer located 14 kilobases upstream of its promoter. We have identified 2 sites within the PU.1 enhancer that strongly bind C/EBPalpha in a gel shift assay, and interaction with endogenous C/EBPalpha was confirmed by chromatin immunoprecipitation. Mutation of these DNA elements reduced activity of a distal enhancer-promoter construct 2- or 5-fold in a myeloid cell line, while mutation of a weaker C/EBPalpha-binding site located in the promoter minimally reduced activity in this context. These findings strengthen the link between C/EBPalpha and PU.1 expression. Reduction of C/EBPalpha activity in cases of acute myeloid leukemia may therefore contribute to transformation by reducing PU.1 levels. In addition, induction of PU.1 by C/EBPalpha during normal hematopoiesis may contribute to stem cell commitment to the myeloid lineages and further commitment to monopoiesis. Consistent with a requirement for C/EBPalpha induction of PU.1 during myeloid development, we demonstrate that C/EBPalpha induces monocytic development when expressed in PU.1(+/+), PU.1(+/-), or PU.1(+/kd) marrow myeloid progenitors but induces granulocyte lineage commitment in PU.1(kd/kd) cells lacking the PU.1 distal enhancer and does not induce either lineage in PU.1(-/-) cells.

Figure 1

C/EBP consensus binding sites in the PU.1 distal enhancer and interaction of endogenous C/EBPs with the enhancer. (A) Positions of the 5 potential C/EBP-binding sites, C1 to C5, and the PU.1 site in the distal enhancer. (B) Sequences of the C1 to C5 and NE C/EBP sites, with flanking sequences in lower case. (C) HF1 myeloid cells or 32Dcl3 myeloid cells cultured in G-CSF for one day (32D-G1) were subjected to the ChIP assay using C/EBPα or C/EBPβ antiserum or rabbit Ig followed by PCR for a 154–base pair PU.1 −14-kb distal enhancer fragment (PU.1 DE), a 138-bp β-actin promoter fragment, or a 329-bp fragment located at −20 kb in the PU.1 gene. Bands were visualized after agarose gel electrophoresis by ethidium bromide staining. I indicates input DNA.

Figure 2

Identification of C/EBPα-binding sites in the PU.1 distal enhancer. (A) Gel shift assay was carried out using 10 μg nuclear extract from 293T cells transfected with CMV-C/EBPα and 1 ng radiolabeled NE-C/EBP probe. Where indicated, 5 ng or 25 ng unlabeled NE or C1 to C5 double-stranded oligonucleotides were added as competitors (comp.) prior to addition of radiolabeled probe. Data with C5 are from a separate gel. (B) Gel shift assay was carried out using 10 μg nuclear extract from 293T cells transfected with CMV (−) or CMV-C/EBPα (α) and 1 ng of the indicated wild-type and mutant radiolabeled C/EBP site probes. See “Materials and methods” for a description of the point mutations in mC1, mC2, and mC5. (C) Gel shift assay was carried out using 12 μg nuclear extract from U937 cells and 1 ng radiolabeled C2 or C5 probes. Rabbit Ig (Ig), C/EBPα (α) or C/EBPβ (β) antiserum, or 25 ng unlabeled wild-type (WT) C2 or C5 or mutant (Mut) mC2 or mC5 competitor oligonucleotides were included in the gel shift reactions. The C/EBPα and C/EBPβ gel shift species and the supershifted species (*) are indicated.

Figure 3

Activity of the PU.1 distal enhancer C/EBP-binding sites in myeloid cells. (A) Diagram of reporter constructs containing point mutations in the C1, C2, C5, or promoter C/EBP-binding sites. DE indicates 237-bp PU.1 distal enhancer; Pr, PU.1 485-bp PU.1 promoter; and LUC, luciferase cDNA. (B) 32DPKCδ cells were transfected with 15 μg of the indicated reporters and 0.5 μg CMV-βGal. Two days later, cell extracts were analyzed for luciferase and β-galactosidase activities. The ratio of these activities for DE was set to 100% in each experiment. Results shown are mean and SE from 6 determinations. (C) The indicated constructs were assayed similarly, with DE linked to wild-type promoter (Pr) set to 100% activity in each experiment. Results are mean and SE from 6 determinations. P values versus DE-Luc (or versus Pr-Luc for mPr) are from the Student t test.

Figure 4

Induction of monocyte development by C/EBPα requires the distal enhancer. (A) Marrow mononuclear cells from PU.1^+/+ or PU.1^+/− mice exposed 5 days earlier to 5-fluorouracil were transduced with pBabePuro-C/EBPαER for 3 days and then exposed to 2 μg/mL puromycin for one day. Viable cells were then lineage-depleted and finally cultured with or without E2 for 2 days before FACS analysis. Percentage monocyte (% M) was calculated as (Mac-1⁺Gr-1⁻)/(Mac-1⁺Gr-1⁻ + Mac-1⁺Gr-1⁺) cells × 100%. (B) PU.1^+/kd or PU.1^kd/kd mice were analyzed similarly. (C) Morphology of PU.1^+/kd or PU.1^kd/kd cells transduced with C/EBPα-ER and cultured with or without E2 for 2 days. (D) Percentage of CFU-Gs or CFU-Ms compared with CFU-Gs + CFU-Ms obtained in methylcellulose culture with IL-3/IL-6/SCF without E2 after transduction of C/EBPα-ER into PU.1^+/+, PU.1^+/kd, or PU.1^kd/kd cells and cultured for 24 hours ± E2 prior to plating. Results shown are from 2 independent transduction experiments (top and bottom panels). The graphs represent the mean and SE from 4 determinations within each experiment (P values are from the Student t test).

Figure 5

Inability of C/EBPα to induce myeloid maturation of PU.1^−/− cells. (A) Western blots showing expression of C/EBPα-ER or PU.1-ER(T) in 503 PU.1^−/− cells. Lane 1 in each blot contains extracts from the parental cell line. The filters were probed with human (left) or murine (right) ERα antisera. (B) C/EBPα-ER–expressing cells exposed to E2 and PU.1-ER(T) cells exposed to 4HT for 2 days were subjected to FACS analysis. Percentage monocytes is indicated (% M). (C) C/EBPα-ER–expressing cells exposed to E2 and PU.1-ER(T) cells exposed to 4HT for 3 days were cytospun and visualized by Wright-Giemsa staining. See “Cell culture and transfection” for image acquisition information.

Figure 6

Model for the role of C/EBPα induction of PU.1 during hematopoiesis. PU.1 achieves sufficient levels in the absence of C/EBPα to help specify the CMPs. C/EBPα increases PU.1 to allow commitment of the CMP to the GMP stage. C/EBPα helps maintain PU.1 expression in the granulocyte lineage but further induces PU.1 in cooperation with cytokine signals and additional transcription factors to allow monocyte lineage specification.