CCAAT/enhancer binding proteins alpha and epsilon cooperate with all-trans retinoic acid in therapy but differ in their antileukemic activities

CCAAT/enhancer binding proteins (C/EBPs) are a family of transcription factors that regulate cell growth and differentiation. Two members of this family, C/EBPα and C/EBPϵ, are of critical importance in granulopoiesis. Disruption of the C/EBPα gene in mice results in the loss of production of neutrophils and eosinophils,¹  whereas mice that lack C/EBPϵ generate neutrophils and eosinophils with abnormal function, gene regulation, and morphology.^2-4 

C/EBPα is the founding member of the bZIP class of DNA-binding proteins.⁵  Members of this family contain distinct N-terminal transactivation domains, C-terminal leucine-zipper dimerization domains, and basic DNA-binding regions.^6,7  C/EBPα's basic region confers not only its ability to bind DNA but also its inhibition of E2F pathways.^8,9  Previous studies have shown that the integrity of DNA binding, transactivation, and E2F inhibition is required for C/EBPα-dependent granulocytic differentiation.^9-12  Although the functional domains required for C/EBPϵ activity have not been well characterized, C/EBPϵ's role in directing expression of myeloid-specific genes associated with terminal differentiation of granulocytes has been clearly demonstrated.^2,13-15 

Because C/EBPα and C/EBPϵ are required for normal granulocytic differentiation, alterations in expression or function of these proteins likely contribute to the pathogenesis of acute myeloid leukemia (AML), a disease characterized by an early block in granulopoiesis. Prior studies^16,17  provide evidence that C/EBPα and C/EBPϵ may play a role in the pathogenesis of acute promyelocytic leukemia (APL), a subtype of AML in which a t(15;17) chromosomal translocation juxtaposes the promyelocytic (PML) gene to the retinoic acid receptor α (RARA) gene, creating an aberrant PML-RARα fusion protein.¹⁸ 

A unique characteristic of PML-RARα leukemic cells is their sensitivity to all-trans retinoic acid (ATRA).¹⁹  Treatment with ATRA induces remissions in patients with APL by causing the leukemic cells to differentiate into mature neutrophils.²⁰  While the mechanism underlying the sensitivity of promyelocytes to ATRA is not completely understood, we and others^21,22  have suggested that C/EBPs mediate the ATRA-induced maturation of APL cells.

In the present study, we explore the mechanism by which C/EBPs prolong survival in a murine model of APL. We also assess the potential for cooperativity between increased C/EBP activity and ATRA therapy. We demonstrate that both C/EBPα and C/EBPϵ significantly prolong survival; however, they are not functionally equivalent in this capacity. We also show that forced expression of C/EBPα or C/EBPϵ in combination with ATRA treatment has a synergistic effect on survival of leukemic mice compared with either therapy alone.

A rat C/EBPα cDNA (rC/EBPα) was generated by polymerase chain reaction (PCR) and cloned into the tamoxifen-inducible pBabepuro3:hb estrogen receptor* (pBP3:hbER*) to generate pBP3:rC/EBPα-ER. For generation of MIG-rC/EBPα-ER, the rC/EBPα-ER fragment was excised from pBP3:rC/EBPα-ER and cloned into the mouse stem cell virus–internal ribosomal entry site–green fluorescent protein (MSCV-IRES-GFP [MIG]) retroviral vector as a BamHI-HincII fragment into the BglII-HpaI sites in MIG. Construction of MIG-hC/EBPϵ-ER has been previously described.¹⁶  Briefly, human C/EBPϵ-ER was cloned into the EcoRI-HpaII sites in MIG as an EcoRI-SacI fragment. Point mutations were made in the DNA-binding domains of rC/EBPRα and hC/EBPϵ at positions R289 and R211, respectively, using the QuikChange XL site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to the manufacturer's instructions. Position R211 in C/EBPϵ is equivalent to C/EBPRα289. The resulting mutants, C/EBPϵR211A and C/EBPαR289A, were confirmed by DNA sequencing.

The 32Dcl3 cell line was modified to express high levels of the ecotropic receptor (32Dcl3-eco R), thereby facilitating retroviral transduction. These cells were maintained in Dulbecco modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS), 100 U/mL penicillin G, 100 μg/mL streptomycin, and 5% X63Ag8–mouse interleukin-3 (mIL-3)–conditioned media. BOSC23 cells were maintained in DMEM supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin. Freshly harvested leukemic no. 1111 cells²³  from bone marrow and spleen of leukemic mice were cultured in stem cell media (Myelocult M5300; StemCell Technologies, Vancouver, BC, Canada) with 15% FBS, 5% IL-3–conditioned media, 0.4 mM glutamine, 100 U/mL penicillin G, 100 μg/mL streptomycin, 10 ng/mL rIL-3, 10 ng/mL IL-6, and 10 ng/mL stem cell factor.

BOSC23 cells were transfected with retroviral constructs as previously described.²⁴  Retroviral supernatants were collected and used to transduce leukemic bone marrow (#1111; 2 × 10⁶ cells per well) and 32Dcl3 cells (100 000 cells per well) as previously described.¹⁶  After transduction, leukemic #1111 and 32Dc13 cells were sorted using the fluorescence-activated cell sorter (FACS) Vantage (Becton Dickinson, San Jose, CA). GFP-positive bone marrow cells were injected into sublethally irradiated mice, and GFP-positive 32Dcl3-eco R cells were maintained in culture in the presence or absence of 20 nM 4-hydroxytamoxifen (4-HT; Sigma, St Louis, MO). At 48-hour intervals, live 32Dc13 cells were counted using trypan blue exclusion, and the percentage of GFP-positive cells was assessed by flow cytometry.

Mice were bred and maintained at the University of California at San Francisco, and their care was in accordance with Institutional Animal Care and Use Committee guidelines. To examine C/EBPs in vivo, retrovirally transduced GFP-positive leukemic no. 1111 cells (50 000 cells per mouse) were injected into the lateral tail vein of sublethally irradiated (4.5 Gy) 6- to 8-week-old female FVB/N mice. Each group (C/EBPα, C/EBPϵ, or the mutant version of each) contained 20 mice, with 5 animals in each subgroup (placebo, 4-HT, ATRA, and 4-HT plus ATRA). Treatments (placebo, 25 mg 4-HT, and 10 mg ATRA) were administered to each subgroup as 21-day release, subcutaneous pellets (Innovative Research of America, Sarasota, FL).

Statistical analyses were performed in GraphPad Prism using the log-rank test or in Excel 2000 using the Student t test with 2-tailed distribution and unequal variance as appropriate.

C/EBPα and C/EBPϵ play central roles in normal myelopoiesis; therefore, it is likely that altered function of these proteins contributes to the pathogenesis of APL. In a previous study,¹⁶  we showed that expression of a tamoxifen-inducible form of C/EBPϵ, hC/EBPϵ-ER, in leukemic cells caused them to differentiate into mature neutrophils in vivo. In the present study, we expand this approach to C/EBPα and assess the abilities of both C/EBPα and C/EBPϵ to prolong survival in a mouse model of APL. To assess the antileukemic effect of C/EBPs in this system, we transduced PML-RARα leukemic cells²³  with either C/EBPα-ER or C/EBPϵ-ER retrovirus and transplanted them into sublethally irradiated histocompatible mice. After leukemias developed in the recipient animals, the mice were treated with either a placebo or 4-HT to induce C/EBP activity. In mice receiving transplants with C/EBPϵ-ER–transduced leukemias, treatment with 4-HT prolonged mean survival by 7 days compared with animals given a placebo (Figure 1A). Animals that received C/EBPα-ER–transduced leukemias demonstrated a more robust response following treatment with 4-HT (Figure 1B), exhibiting a mean increase in survival of 11 days compared with the placebo group. Therefore, tamoxifeninducible forms of both C/EBPα and C/EBPϵ significantly prolong survival in a mouse model of APL. Of note, 4-HT treatment of mice receiving untransduced cells had no effect on survival (data not shown).

Because C/EBPs belong to the bZIP family of DNA-binding proteins, we asked whether the ability to bind DNA was required for the antileukemic activities of C/EBPα and C/EBPϵ. To address this, we mutated R289 in C/EBPα and the corresponding amino acid (R211) in C/EBPϵ to alanine-generating mutants (C/EBPαR289A and C/EBPϵR211A) that were devoid of DNA-binding activity (data not shown). We selected these residues for substitution based on the observation that R289 is an integral component of the protein-DNA interface in the crystal structure of a C/EBPα bZIP polypeptide bound to its cognate DNA site.²⁵  We then transduced PML-RARα leukemic cells with C/EBPαR289A or C/EBPϵR211A retrovirus and transplanted them as before. In animals receiving transplants with C/EBPϵR211A-transduced leukemias, we found that treatment with 4-HT had no effect on survival (Figure 1C), thus indicating that the DNA-binding activity of C/EBPϵ is required for its antileukemic effect. We observed a similar phenomenon in the factor-dependent myeloid progenitor cell line 32Dcl3 (Figure 1D). While the wild-type C/EBPϵ suppressed growth of 32Dc13 cells in the presence of 4-HT, the R211A mutant was unable to repress proliferation, suggesting that the DNA-binding activity of C/EBPϵ is required for its function in immature myeloid cells.

Interestingly, mice receiving transplants with C/EBPαR289A-transduced leukemias exhibited a mean increase in survival of 8 days compared with the placebo group following treatment with 4-HT (Figure 1E). This finding indicates that, unlike C/EBPϵ, C/EBPα can suppress leukemia without binding DNA. 32Dc13 cells transduced with either wild-type or mutant C/EBPα fail to proliferate following treatment with 4-HT, suggesting that C/EBPα has antiproliferative effects that do not require DNA binding (Figure 1F). Together, the results presented in Figure 1 imply that C/EBPα and C/EBPϵ are not functionally equivalent in their ability to suppress growth and inhibit leukemogenesis.

The results shown in Figure 1, in conjunction with the idea that C/EBPs mediate the antileukemic effects of ATRA, led us to assess the potential for cooperativity between treatment with ATRA and forced expression of C/EBPs. To address this relationship, we transduced PML-RARα leukemic cells with either C/EBPα-ER or C/EBPϵ-ER retrovirus and transplanted them as before. After leukemias developed in recipient animals, we treated them with either a placebo, 4-HT, ATRA, or a combination of 4-HT and ATRA. In mice receiving transplants with C/EBPϵ-ER–transduced leukemias, ATRA treatment resulted in a more substantial increase in survival (mean increase of 23 days compared with the placebotreated group) than did treatment with 4-HT (Figure 2A). However, treatment with a combination of 4-HT and ATRA had the greatest impact on survival, with a mean increase of 37 days compared with the placebo group.

Animals that received C/EBPα-ER–transduced leukemias demonstrated a similar pattern of response; however, the magnitude of each response was greater than that seen in C/EBPϵ-ER mice (Figure 1B). Treatment with 4-HT and ATRA prolonged the survival of C/EBPα-ER animals by a mean increase of 11 and 38 days, respectively, but the greatest impact on survival was seen in mice given the combined treatment (mean increase in survival was 67 days). The results presented in Figure 2 demonstrate that forced expression of C/EBPα or C/EBPϵ in combination with ATRA treatment has a synergistic effect on survival of leukemic mice compared with either therapy alone. These findings suggest that the ability of ATRA to prolong the survival of leukemic mice includes mechanisms beyond simple induction of C/EBP activity. Furthermore, and of potential clinical importance, our results indicate that agents that increase C/EBP activity might function cooperatively with ATRA to exert antileukemic activity in AML.