Abelson virus transformation prevents TRAIL expression by inhibiting FoxO3a and NF-kappaB

Abstract

The Abelson Murine Leukemia Virus (A-MuLV) encodes v-Abl, an oncogenic form of the ubiquitous cellular non-receptor tyrosine kinase, c-Abl. A-MuLV specifically transforms murine B cell precursors both in vivo and in vitro. Inhibition of v-Abl by addition of the small molecule inhibitor STI-571 causes these cells to arrest in the G1 phase of the cell cycle prior to undergoing apoptosis. We found that inhibition of v-Abl activity results in upregulation of transcription of the pro-apoptotic TNF-family ligand tumor-necrosis factor-related apoptosis-inducing ligand (TRAIL). Similarly to BCR-Abl-transformed human cells, activation of the transcription factor Foxo3a led to increased TRAIL transcription and induction of a G1 arrest in the absence of v-Abl inhibition, and this effect could be inhibited by the expression of a constitutively active AKT mutant. Multiple pathways act to inhibit FoxO3a activity within Abelson cells. In addition to diminishing transcription factor activity via inhibitory phosphorylation by AKT family members, we found that inhibition of IKKbeta activity results in an increase in the total protein level of FoxO3a. Furthermore overexpression of the p65 subunit of NF-kappaB results in an increase in TRAIL transcription and in apoptosis and deletion of IKKalpha and beta diminishes TRAIL expression and induction. We conclude that in Abelson cells, the inhibition of both NF-kappaB and FoxO3a activity is required for suppression of TRAIL transcription and maintenance of the transformed state.

Figure 1. TRAIL upregulation in response to STI treatment plays a role in G1 arrest and cell death

a) Two representative Abelson virus transformed pro-B cell lines were cultured for 16 hours in the presence or absence of 5 μM STI-571. Graph shows fold-induction of TRAIL transcript levels normalized to HPRT as measured by real time RT-PCR. Error bars are standard deviation of triplicate samples. The data is representative of at least three similar experiments. b) PD31 cells were retrovirally transduced with a vector expressing FADD-DN or stably transfected with a Bcl-xL expression vector or with both and selected for drug resistance. Cells were cultured in the presence (upper panels) or absence (lower panels) of 2.5 μM STI-571 for 36 hours, then stained with Annexin V-FITC and 7-AAD. The numbers indicate percentage of total cells in the gate. Results represent one of 3 similar experiments b) Cells in (b) were cultured in the presence or absence of 2.5 μM STI for 17 hours, ethanol fixed, and then stained with PI in the presence of RNase. Cells were analyzed for cell cycle status by flow cytometry, excluding cells with a sub-G1 DNA content. Numbers indicate the percentage of cells in G1 as determined with the Watson (pragmatic) protocol +/- the standard deviation of 3 replicates. Shaded histogram = untreated; solid line = STI treated.

Figure 2. Expression of a constitutively active FoxO3a mutant increases TRAIL transcription and induces G1 arrest of Abelson-transformed cells

a) PD31 cells were infected with a FoxO3a-GFP expressing retrovirus. Cells were sorted for GFP, grown for 3 days in culture, and then analyzed by real-time PCR for TRAIL and HPRT transcript levels. Graph represents fold TRAIL induction normalized for HPRT expression. Error bars are the standard deviation of triplicate samples. Results are representative of at least 3 similar experiments. b) 7G-S cells were infected with a retroviral FoxO3a(A3)-ER construct, sorted for a marker of infection, and cultured in the presence of 4 μM tamoxifen (OHT) or 10 μM STI for 17 hours and compared to uninfected parental cells treated the same way. TRAIL and HPRT transcript levels were determined by real-time PCR. Graph represents fold TRAIL induction normalized for HPRT expression. Error bars are the standard deviation of triplicate samples. Results are representative of 3 similar experiments. c) 7G-S cells were infected with a FoxO1(A3)-ER construct, and analyzed as in b. d) 7G-S cells were infected with FoxO3a(A3)-ER, FoxO1(A3)-ER, or FoxO3a(A3)H215R-ER expressing retroviruses. Cells were cultured in the absence (thin line), or presence of 10 μM STI (filled region), or 4 μM OHT (thick line) for 20 hours before being analyzed for DNA content as in 1b. Numbers indicate the percentage of cells in G1 +/- the standard deviation of 3 replicates.

Figure 3. FoxO3a is dephosphorylated upon v-Abl inhibition and TRAIL transcription is diminished in the presence of a constitutively active AKT

a) Lysates from 220-8 Abelson cells cultured for 16 hours in the presence (+) or absence (-) of 10 μM STI were analyzed by Western blot for T32-phosphorylated or total FoxO3a as indicated. b) 7G-S cells were infected with a retrovirus expressing ER-myr-Akt, and sorted. Cells were treated with either 10 μM STI, 4 μM OHT, or both for 17 hours. TRAIL transcription was analyzed by real time PCR and normalized to HPRT. Graph presents fold TRAIL induction.

Figure 4. IKKβ is active in v-Abl-transformed cells and inhibits FoxO3a activity

a) 7G-S cells were cultured in the presence of 2 μM AKT inhibitor VIII, 10 μM STI571, or 5 μM IKKβ inhibitor III for 20 hours at which time TRAIL and HPRT transcripts were analyzed by real-time PCR. TRAIL transcripts were normalized to HPRT. Graph presents fold-induction of TRAIL. b) myr-Akt-ER-transduced cells were cultured in the presence of 4 μM OHT, 5 μM IKKβ inhibitor III, or both for 17 hours at which time transcripts levels of TRAIL and HPRT were determined by real time PCR. Results indicate fold TRAIL induction. c) Total cell lysates from 7G-S Abelson cells cultured in the presence of 10 μM STI or 10 μM IKKβ inhibitor III, and analyzed by Western blot for FoxO3a and actin. d) The graph represents the intensity of the FoxO3a band normalized to the intensity of the actin band from two independent repetitions of the experiment shown part c, above.

Figure 5. RelA expression activates TRAIL transcription

a) PD31 cells were infected with a retrovirus expressing RelA and sorted based on a marker of infection. The infected and uninfected cells were grown in culture for 3 days and then analyzed for relative levels of TRAIL transcripts by real time PCR. The graph presents fold TRAIL induction upon RelA expression normalized for HPRT levels. Error bars are the standard deviation of triplicate samples. Results are representative of 3 similar experiments. b) 7G-S cells were infected with IκB-ΔN and sorted. Cells were treated with 10 μM STI for 17 hours and then analyzed for TRAIL and HPRT transcription via real time PCR. Graph represents the fold induction of STI treated infected and uninfected cells as compared to untreated cells. Error bars are the standard deviation of triplicate samples. Results representative of 3 similar experiments. c) Wildtype or IKKα/β knockout cell lines were analyzed for TRAIL and HPRT transcription via real time PCR. Graph represents the level of TRAIl transcription normalized to HPRT transcripts. Results are representative of three out of four independently derived cell lines. d) Wildtype or IKKα/β knockout cell lines were treated with 2.5 μM STI571 for 24 hours and then analyzed for TRAIL and HPRT transcription via real time PCR. Graph represents the fold induction of STI treated infected and uninfected cells as compared to untreated cells. Error bars are the standard deviation of triplicate samples. Results representative of 3 out of 4 cell lines tested.

Figure 6. A Model for TRAIL activation in v-Abl transformed cells

When v-Abl is active (left side), low levels of NF-κB activity exist. FoxO3a is phosphorylated by IKKβ, leading to degradation, as well as by AKT/SGK, preventing nuclear translocation of the basal level of the protein. The inactivation of both NF-κB and FoxO3a results in low TRAIL transcription. Upon v-Abl inhibition (right side), NF-κB is activated downstream of the IKK complex, leading to TRAIL activation. Simultaneously, though still being kept at a low total protein level via IKKβ phosphorylation, cellular Foxo3a is activated by the removal of the inhibitory phosphorylation and translocates to the nucleus, activating TRAIL transcription.