N‐Myc is a downstream target of RET signaling and is required for transcriptional regulation of p18^Ink4c by the transforming mutant RET^C634R

2.1. Search for a putative transcription factor target of RET2A that mediates p18 transcriptional regulation

To understand how RET2A regulates p18S transcription, we searched for putative transcription factor (TF) targets of RET2A signaling that could bind to the p18 promoter. A detailed search for highly conserved cis‐acting elements across human, mouse and rat genomes in a 5500 basepair region upstream of the p18 translation start site (TSS, set to +1) uncovered the presence of several consensus TF binding sites (method presented in Supplementary methods and data shown in Supplementary Table 1). We used three criteria that would qualify a TF as a likely candidate for RET2A‐mediated p18 transcriptional regulation. First, we predicted that a bona fide downstream target of RET2A signaling capable of regulating p18S transcription would be expressed in adrenal chromaffin and thyroid parafollicular C cells or in endocrine organs. Developmentally, adrenal chromaffin and thyroid parafollicular C cells derive from Neuronal Progenitor Cells (NPCs) and it is known that during murine development, N‐Myc is expressed exclusively in NPCs and other neuroendocrine cells (Downs et al., 1989). Second, N‐Myc is regulated by the MAPK pathway which is required for p18 regulation by RET2A. Treatment of KP‐N‐RT human neuroblastoma cells with IGF‐I results in a MAPK dependent N‐Myc induction (Misawa et al., 2000). Thirdly, in developing cerebella, loss of N‐Myc results in accumulation of p18 and p27, correlate with growth arrest in cerebellar Granule Neuron Progenitors (GNPs) (Zindy et al., 2006). In early neural development, expression of N‐Myc in NPCs is required to maintain low p18 and p27 levels correlate with rapid NPC proliferation (Knoepfler et al., 2002). Loss of both p18 and p27 expression in an N‐Myc null background restores normal proliferation of the GNPs, suggesting that during embryonic development, both p18 and p27 are regulated by N‐Myc to promote proliferation. Taken together, these studies make N‐Myc a likely candidate to study as a downstream intermediate in RET2A/MAPK‐dependent p18 transcriptional repression. N‐Myc has been shown to act as a transcriptional inhibitor by binding to an inr consensus sequence, although the commonly accepted role of N‐Myc is enhancer box (E‐box)–mediated transcriptional activation of target genes (Blackwell et al., 1993; Eberhardy et al., 2000). Our sequence analysis of the p18 promoter region revealed presence of two loosely conserved palindromic inr elements at positions −166 to −161 and −160 to −155 relative to the p18S TSS (Figure 1A). Typically, the sequence of an inr element resembles 5′‐YYAN(A/T)YY‐3′, with considerable variation in conservation as well as identity. The 5′‐GGCAGGCCGACC‐3′ (−167 to −156) sequence of the p18 promoter represents two weak palindromic consensus inr sequences. Both the inr elements, CCGTCC (complimentary strand) and CCGACC, fit the YYAN(A/T)YY consensus sequence, except both are missing the A in the 3rd position. Therefore, we hypothesized that RET2A signaling induces N‐Myc expression, which in turn prevents transcription of p18S.

2.2. RET2A expression leads to a MAPK pathway‐dependent induction of N‐Myc at the mRNA and protein levels

In our previously described tetracycline (Tet) inducible NIH3T3 cell based expression system, in order for p18 and p27 levels to remain low in post‐confluent culture conditions, RET2A must continually be expressed starting in sub‐confluent culture conditions (Joshi et al., 2007). If RET2A expression is initiated after post‐confluent growth arrest is established, p18 and p27 accumulate, remain high, and the levels of neither CDKI can be lowered by RET2A signaling. This suggests two important steps in RET2A‐dependent p18 regulation. First, RET2A signaling prevents induction of p18 rather than actively repressing accumulated p18. Second, the p18 regulatory mechanism leading to high p18 levels in growth arrested cells is active in sub‐confluent conditions, prior to initiation of growth arrest. The latter is supported by similar observations in another type of growth arrest, which is mediated by serum starvation. RET2A expression in serum starvation‐mediated growth arrested cells fails to reduce p18 and p27 levels unless RET2A expression is induced prior to serum starvation.

To determine whether RET2A expression can induce N‐Myc levels, we assayed N‐Myc protein levels starting from sub‐confluent growth conditions in our Tet‐inducible RET2A expression system. This system is particularly suited for such analysis because other available anterior pituitary adenoma cell lines or pheochromocytoma cell lines have impaired p18 expression which makes them ineffective for p18 regulation studies (Acton et al., 2004; Joshi et al., 2007). Total protein lysates from RET2A cells treated with Tet (+Tet), beginning at 80% confluency (Figure 1B 0 h), were harvested at 12 h intervals until the cells reached a 36 h post‐confluency timepoint (48 h). Compared to the −Tet controls, N‐Myc protein levels are increased approximately six‐fold at 12 h and drop to undetectable levels by 24 h (12 h post‐confluency) (Figure 1B). The 12 h timepoint is when cells become 100% confluent and expression of N‐Myc at 12 h is timed perfectly if N‐Myc prevents p18S mRNA accumulation in a RET2A‐dependent manner. Any later would be at a post‐confluency timepoint when RET2A cannot prevent p18 induction. In the absence of RET2A (−Tet), p18 (and p27) protein levels increase at 24, 36 and 48 h, while RET2A expression leads to a three‐fold decrease in both p18 and p27 at the same timepoints. Consistent with previous observations, Cyclin D1/D2 protein levels were approximately three‐fold increased at 12, 24, 36 and 48 h.

Being an immediate early gene, it is possible that N‐Myc is induced prior to 12 h in response to RET2A expression. To test this, we performed a shortened timecourse experiment where we harvested protein lysates at 80% confluency (0 h) and at 3, 6, 12 (100% confluency) and 24 h. We detected a four‐fold induction of N‐Myc protein at 6 h, and a six‐fold induction at 12 h (Figure 1C, quantitated in 1D). Consistent with Figure 1B, N‐Myc was not induced at 24 h. Again expression of N‐Myc prior to 100% confluency is appropriate timing for N‐Myc to prevent the beginning of p18 (and p27) mRNA and/or protein accumulation. Activation of the MAPK pathway in RET2A‐expressing cells, detected by a phospho‐specific ERK antibody, can be seen as early as 3 h and lasts throughout the timecourse. At post‐confluent timepoints, compared to −Tet samples, we detected approximately three‐fold lower levels of p18 and p27, which were comparable to those in sub‐confluent growth conditions. There was a steady increase in Cyc D1/D2 levels at 3, 6, 12 and 24 h consistent with previously published results.

Using Quantitative Reverse Transcriptase‐Polymerase Chain Reaction (QRT‐PCR) analysis, we next examined whether RET2A‐mediated induction of N‐Myc protein was due to an increase in N‐Myc mRNA expression. Indeed, there is a two‐ and three‐fold increase in N‐Myc mRNA levels at 6 and 12 h respectively (Figure 1E). At 24 h N‐Myc mRNA and protein levels return to basal levels. We also observed that the three‐fold increase in p27 and p18 mRNA levels at 24 h is prevented by RET2A. These data establish a sequential timeline of events leading to regulation of p18 and p27 in response to RET2A expression. Within 3 h of RET2A expression, there is activation of the MAPK pathway that results in induction of N‐Myc at 6 h, lasting through 12 h and dropping back to basal levels by 24 h. Low levels of p18 and p27, comparable to those in sub‐confluent cells, persist as cells continue into post‐confluency despite the later absence of N‐Myc expression. It is possible that N‐Myc acts as a “molecular switch” that prevents inr‐mediated p18 transcription in sub‐confluent cells which would otherwise allow p18 accumulation and growth arrest due to confluency. Once this point is past, inr‐mediated p18 transcription cannot occur under post‐confluent conditions, and the requirement for N‐Myc expression lessens.

We have previously shown that, RET2A‐mediated maintenance of low p18S transcription in post‐confluent conditions requires a functional MAPK pathway therefore; we tested if the transient induction of N‐Myc is MAPK dependent. We repeated the short timecourse in the presence of a chemical inhibitor of the MAPK pathway, PD98059. Cells treated with PD98059 for 1 h to inhibit MAPK signaling were then treated with Tet to express RET2A. PD98059 efficiently inhibited phosphorylation of ERK throughout the course of the experiment (Figure 2A). N‐Myc induction, previously observed at 6 and 12 h, was not seen (Figure 2A), suggesting that N‐Myc induction is MAPK dependent. Both p18 and p27 accumulate at 24 h in PD98059‐treated cells. These results show that activation of the MAPK pathway by RET2A is required for induction of N‐Myc expression and maintenance of low CDKI levels.

2.3. N‐Myc induction is required for RET2A‐mediated p18 regulation

We next assessed whether N‐Myc induction is required to maintain low p18 and p27 levels as cells become growth arrested. We used siRNA directed against N‐Myc to inhibit its expression. In negative control (non‐specific siRNA) cells treated with Tet, we detected an approximately six‐fold increase in N‐Myc expression at 12 h. We also detected a four‐fold induction of D‐type cyclins (at 12 and 48 h) and four‐ and 2.5‐fold decrease in levels of p18 and p27 respectively (at 48 h, Figure 2B, quantitated in C). On the other hand, in N‐Myc siRNA transfected cells, expression of N‐Myc was abolished, but activation of the MAPK pathway was still seen at 12 and 48 h. Most importantly, p18 and p27 were no longer maintained at low levels at 48 h. In other words, loss of N‐Myc induction at 12 h prevents RET2A‐mediated maintenance of low p18 and p27 levels under growth arrest conditions at 48 h. A third set of samples treated with PD98059 were included in this analysis, because such treatment inhibits MAPK activation and N‐Myc induction. Similar results were seen in PD98059‐treated samples, which indicates that N‐Myc is downstream of the RET2A‐MAPK signaling pathway and that MAPK dependent N‐Myc induction is essential for RET2A‐mediated prevention in accumulation of p18 and p27. Consistent with our previous report where MAPK activity is not involved in RET2A‐mediated induction of Cyclin D1/D2, their levels remained unaffected in all of the +Tet samples, irrespective of the N‐Myc status.

We also confirmed that protein levels correlate with mRNA expression; total RNA samples from parallel experiments were analyzed by QRT‐PCR for indicated genes. Consistent with protein expression results there was a three‐fold increase in N‐Myc mRNA levels in +Tet and control siRNA samples at 12 h, as compared to −Tet cells (Figure 2D) which was not observed in N‐Myc siRNA or PD98059‐treated cells. Although p18 and p27 mRNAs were unaffected at 12 h, the three‐fold increase at 48 h in p18 and p27 mRNA levels in −Tet samples was absent in +Tet or control siRNA samples, suggesting that they were maintained from sub‐confluent growth conditions. At 48 h, p18 and p27 mRNA levels were reciprocally higher in cells that lack N‐Myc (N‐Myc siRNA and PD98059‐treated cells). Additionally, RET2A‐mediated induction of Cyclin D1/D2 mRNA remained unaffected in the absence of N‐Myc (data not shown). Together, these results suggest that RET2A‐mediated MAPK pathway activation and N‐Myc induction are required to maintain low levels of p18 and p27, but not for the induction of Cyclin D1/D2.

2.4. Promoter sequence containing the N‐Myc consensus binding site is required for RET2A‐mediated effects on p18 transcription

We first determined if the N‐Myc binding site (inr element) resides within the minimal RET2A responsive region of the p18 promoter. We subcloned various fragments of the p18 promoter into the pGL2‐basic firefly luciferase reporter gene vector. In response to RET2A expression, measured luciferase activity of a full‐length (−1 to −3496) and that of a 5′ truncated (−1 to −992, LucA) construct was found to be identical (data not shown), suggesting that there are no RET2A responsive elements in the −992 to −3496 region of the p18 promoter. Therefore, this region was excluded from subsequent promoter analyses. We next determined if the inr element is required for RETA‐mediated prevention of p18 accumulation. LucA and LucB contain 992 bp (−1 to −992) and 365 bp (−1 to −365) immediately 5′ to the p18 translation start site (designated as +1, ATG) respectively, while LucC is 499 bp (−367 to −868) and lacks 367 bp immediately 5′ to the p18 translation start site (i.e. lacks LucB sequence) (Figure 3A). LucA and LucB contain the N‐Myc binding site (−156 to −167, inr element), whereas LucC does not. RET2A cells transfected with LucA, B or C, and negative control or N‐Myc siRNA at 60–80% confluency became ∼90% confluent (0 h timepoint) at 24 h post‐transfection when they were treated with Tet and harvested for protein lysates. The luciferase activities driven by the LucA, B and C fragments at 0 h were ∼3000, ∼4500 and ∼12,500 Relative Luciferase Units (RLU), respectively (Figure 3B). At this timepoint, as compared to −Tet samples, neither Tet treatment nor transfection of control or N‐Myc siRNA had any effect on luciferase activity of any of the three constructs. A parallel set of treated cells were harvested after 48 h of Tet treatment (36 h post‐confluent). At this post‐confluent timepoint, p18 transcription is maintained at low levels by RET2A therefore, we predict lower luciferase activity in +Tet samples. The overall luciferase activity in –Tet cells decreased for all three constructs. Regardless, expression of RET2A leads to a further, statistically significant, three‐fold decrease in luciferase activity of LucA and LucB. This suggests that LucA and LucB contain RET2A responsive transcriptional regulatory elements that could cause lower transcription of p18S. Similar results were obtained in control siRNA transfected RET2A‐expressing cells however, loss of N‐Myc by siRNA leads to restoration of luciferase activity driven by both, LucA and LucB (Figure 3B). This indicates that N‐Myc is required for RET2A‐mediated repression of luciferase activity. The activity of LucC remained unaffected at an average of ∼1900 RLU under all four conditions (−Tet, +Tet, +Tet + control siRNA or +Tet + N‐Myc siRNA). These data suggest that the −1 to −365 region of the p18 locus, containing the inr N‐Myc binding site, is required for RET2A‐mediated repression of luciferase reporter gene activity and for the restoration of luciferase activity in response to N‐Myc siRNA treatment.

Several other TFs known to be regulated by MAPK, and whose binding sites were present within the −1 to −365 region of the p18 promoter, were unaffected at the mRNA level by RET2A. One exception was the TF c‐Fos, which was induced two‐ and three‐fold at the mRNA level at 3 and 6 h respectively (see Supplementary Figure 1). The timing of c‐Fos induction at 3 and 6 h precedes that of N‐Myc which occurs at 6 and 12 h. Therefore, it is possible that c‐Fos may act as an intermediate TF that induces N‐Myc transcription in response to RET2A expression and MAPK activation. While we have not pursued involvement of c‐Fos in this study, and cannot completely eliminate involvement of other TFs whose putative binding sites are lost due to the DNA sequences deleted in the LucC construct, the requirement of N‐Myc expression for RET2A‐mediated effects on luciferase activity (and p18 transcription, Figure 2B–D) led us to conclude that the inr element within this 365 basepair region appears to be critical for RET2A and N‐Myc dependent transcriptional regulation of p18.

2.5. RET2A expression leads to increased binding of N‐Myc to its target inr site within the p18S promoter

Next, we sought a more direct approach to determine direct binding of N‐Myc to the inr element in the p18 promoter. We used a combination of Chromatin Immuno‐Precipitation (ChIP) followed by quantitative‐PCR (qPCR) analysis of precipitated DNA fragments to determine N‐Myc binding to its consensus inr binding site within the p18 promoter in a RET2A‐dependent manner. RET2A cells transfected with negative control or N‐Myc siRNA or treated with PD98059 were induced with Tet. Nuclear lysates were harvested at 12 h post‐induction, when N‐Myc expression is maximal. Crosslinked and fragmented chromatin was immunoprecipitated with an N‐Myc antibody and the DNA bound to N‐Myc was then used as a template in qPCR reactions to amplify non‐specific or N‐Myc binding site specific amplicons (Figure 4A).

Expression of RET2A (+Tet) results in a six‐ to seven‐fold enrichment in the N‐Myc consensus binding site specific amplicon relative to the non‐specific amplicon suggesting that there is increased binding of N‐Myc to inr element (Figure 4B). In the presence of N‐Myc siRNA (or PD98059 MAPK inhibitor), we did not detect any discernible N‐Myc expression (Figure 4C, consistent with Figure 2A and B). Hence, it is not surprising that we do not detect any binding of N‐Myc to the N‐Myc binding site under such conditions. The enrichment remained unaffected in negative control siRNA samples. N‐Myc protein expression and phosphorylation of ERK was confirmed by western analysis using 10% of pre‐cleared nuclear extracts (Figure 4C). ERK phosphorylation was inhibited efficiently in PD98059‐treated cells and N‐Myc was induced only in +Tet and + Tet + control siRNA samples. These results demonstrate that N‐Myc binds to its genomic target site within the p18 promoter in response to RET2A expression. Taken together, the luciferase and ChIP assays suggest that the N‐Myc inr binding site within the p18S promoter is critical for RET2A‐mediated regulation of p18 expression and that N‐Myc binds to its consensus binding site in response to RET2A expression in a MAPK‐dependent manner

2.6. N‐Myc induction is required for the RET2A‐mediated increase in proliferation

Continued RET2A expression starting from sub‐confluent through post‐confluent growth arrest conditions confers a proliferative advantage. In other words, RET2A‐expressing cells continue to proliferate instead of arresting under post‐confluent growth conditions. Cells seeded at 60% confluency and RET2A expression induced at 90% confluency (0 h timepoint) were fixed, propidium iodide (PI) stained and analyzed for DNA content at 48 h (36 h post‐confluent). The percentage of cells in various phases of the cell cycle (G₀/G₁, S and G₂/M) was determined. Uninduced 90% sub‐confluent cells (−Tet 0 h) are cycling, as indicated by a proliferative index (S + G₂/M phases) of 28.1 + 2.4% (Figure 5A). Cells treated with Tet at 0 h exhibit a similar proliferative index of 28.6 + 2.3%. This is expected since RET2A has no effect on proliferation under sub‐confluent growth conditions (Joshi et al., 2007). As uninduced cells become post‐confluent (−Tet 48 h) they become growth arrested due to contact inhibition, reflected by a reduced proliferative index of 6.11 + 1.52%. RET2A expression (+Tet 48 h) allows post‐confluent cells to continue proliferation (S + G₂ M = 17.8 + 4.2%) at a three‐fold higher rate relative to −Tet samples (48 h). Therefore, ethanol‐treated cells growth arrest due to contact inhibition by 48 h while, RET2A‐expressing cells continue to proliferate at a higher rate.

Since RET2A‐dependent induction of N‐Myc is required to maintain low levels of p18, and low levels of p18 are critical for RET2A‐mediated hyperproliferation, it is possible that N‐Myc is required for RET2A‐mediated hyperproliferation. Using FACS analysis we assessed whether RET2A‐expressing cells could hyperproliferate under growth arrest conditions in the absence of N‐Myc. RET2A cells transfected at 60–80% confluency with no DNA (Mock), control siRNA or N‐Myc siRNA were induced 24 h post‐transfection when cells reached 90% confluency (0 h timepoint) with Tet or ethanol for 48 h (approximately 24 h post‐confluent). RET2A‐expressing (+Tet), mock or control siRNA transfected post‐confluent cells continue to proliferate at an approximately two‐fold higher rate, as compared to ethanol‐treated cells (Figure 5B). However, cells lacking N‐Myc fail to gain a growth advantage, reflected by no change in the proliferative index relative to the ethanol‐treated N‐Myc siRNA sample. Together these data suggest that the proliferative advantage conferred by RET2A expression requires expression of N‐Myc.

4.1. Plasmids and cell culture

LucA (−1 to −992) and LucB (−1 to −365) fragments were amplified by PCR from wild type genomic mouse tail DNA using LucAF (5′‐ATGGTACCAAGTTACTGGTGAGCTGGCCGGGTGCA‐3′) and LucBF (5′‐ACGGTACCACGTGACAGCTCTGCCTACA‐3′), respectively as forward primers with a common reverse primer LucR (5′‐CGCTCGAGTCTTTAGGGTCCTGGCGATCGG‐3′). The PCR conditions used were 95 °C/5 min–one cycle; 95 °C/1 min, 60 °C/1 min, 72 °C/2 min–40 cycles; 72 °C/10 min–one cycle, 4 °C hold. KpnI and XhoI linkers were added to the forward and reverse primers respectively. LucA and LucB PCR products digested with KpnI/XhoI were ligated with KpnI/XhoI digested pGL2‐Basic firefly luciferase reporter gene vector. The pGL2‐Basic and pGL2‐LucC (designated as pGL2‐Luc17, kindly provided by Dr. Yue Xiong (Phelps et al., 1998). The N‐Myc siRNA was purchased (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and used according to manufacturer's recommendations. The control siRNA was kindly provided by Dr. Elizabeth Taparowsky.

NIH3T3 fibroblasts harboring both pcDNA6‐TR and pcDNA4‐RET2A have been described previously (hereafter referred to as RET2A cells) (Joshi et al., 2007). Cells were grown in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) (Mediatech, Inc., Herndon, VA, USA) and 500 U/ml penicillin ‐ 500 μg/ml streptomycin (Invitrogen, Carlsbad, CA, USA) and incubated at 37 °C and 5% CO₂. Transfection was carried out using Polyfect transfection reagent according to the manufacturer's protocol (Qiagen, Valencia, CA, USA). Expression of RET2A was induced by addition of 2 μg/ml tetracycline (Sigma, St. Louis, MO, USA) (+Tet). As a negative vehicle control, ethanol was added (−Tet) for all experiments. Ethanol or tetracycline was boosted every 12 h. The MEK inhibitor PD98059 (100 μM, final concentration) (Calbiochem, San Diego, CA, USA) was added to culture medium 1 h prior to addition of tetracycline and continued for the course of the experiment.

4.2. Antibodies and immunoblotting

For immunoblotting, the following commercially available antibodies were purchased: rabbit anti‐ERK‐MAPK (C‐16), rabbit anti‐phospho‐ERK‐MAPK antisera (Santa Cruz Biotechnology, Santa Cruz, CA, USA), rabbit anti‐N‐Myc (M‐50) (Santa Cruz Biotechnology, Santa Cruz, CA, USA), mouse anti‐tubulin (Ab‐4, NeoMarkers, Fremont, CA, USA). Polyclonal rabbit anti‐RET (03‐20), rabbit anti‐p27 (03‐22) (Joshi et al., 2007), rabbit anti‐cyclin D1 and rabbit anti‐p18 (Franklin and Xiong, 1996; Phelps and Xiong, 1998) polyclonal antibodies have been described previously. All procedures for immunoblotting have been described previously (Franklin et al., 1998, 2000, 1996, 1995, 1998). Total quantity of proteins in lysates was determined using Bradford assay. Individual bands on the immunoblots were quantified using Licor Odyssey v1.2. and normalized to tubulin (with the exception of Figure 4C). In Figure 4C, total ERK was used as a loading control. Fold change was calculated as the ratio of induced (+Tet) to uninduced (−Tet) values. The average relative fold change in proteins at the 0 h timepoint was set to +1 for proteins that are induced by RET2A or to −1 for proteins that are repressed by RET2A. P‐values were obtained by performing Student's t‐test.

4.3. QRT‐PCR

Total RNA isolation, synthesis of cDNA, SYBR green based QRT‐PCR and data analysis has been previously described (Joshi et al., 2007). Each QRT‐PCR reaction was carried out in the ABI prism 7300 sequence detection system (Applied Biosystems, Foster City, CA, USA) Fold induction or repression of target gene transcripts relative to 18S rRNA were obtained and subjected to statistical analysis using Student's t‐test. The Primer Express software provided by Applied Biosystems was used to design primers for each target gene. Sequences for each primer are as follows:

18S rRNA forward primer (5′‐CACGGCCGGTACAGTGAAA‐3′),18S rRNA reverse primer (5′‐AGAGGAGCGAGCGACCAA‐3′);

p18S CDKI forward primer (5′‐ACCATCCCAGTCCTTCTGTCA‐3′),p18S CDKI reverse primer (5′‐GTTGCTTCACTTTTTCCCCTTTC‐3′);

p27 CDKI forward primer (5′‐CTTCCGCCTGCAGAAATCTCT‐3′),p27 CDKI reverse primer (5′‐CGGCAGTGCTTCTCCAAGTC‐3′);

N‐Myc forward primer (5′‐CTGAGCTGGTGAAGAACGAGAA‐3′),N‐Myc reverse primer (5′‐CTCGGTGGCCTTTTTCAAGA‐3′);

4.4. Luciferase assay

RET2A cells seeded in 12‐well plates were grown to 60–80% confluency at the time of transfection. Polyfect (Qiagen, Valencia, CA, USA) transfection reagent was used according to manufacturer's protocol for transfection. For each transfection, 1 μg of the pGL2‐LucA, B or C plasmid along with 0.1 μg of pRL‐null plasmid was used. 1 μg of control or N‐Myc siRNA plasmid was included in the transfection where indicated. After 23 h transfection, PD98059 (Calbiochem, San Diego, CA, USA) was added to a final concentration of 100 μM. At 24 h post‐transfection, tetracycline (2 μg/ml) or 100% ethanol (vehicle) were added. The 0 h samples were harvested at this timepoint. The 48 h samples were further incubated by boosting RET2A expression with tetracycline at 12 h intervals. Cells were washed twice with cold PBS and lysed with 100 μl of passive lysis buffer (Promega, Madison, WI, USA). 10 μl of cell extract was used to determine Firefly and Renilla luciferase activity by using firefly luciferase substrate (Promega, Madison, WI, USA) and Coelenterazine (Sigma, St. Louis, MO, USA) respectively, according to the manufacturer's protocol in the lumat LB9501 luminometer (Berthold Technologies, Oak Ridge, TN, USA). The firefly luciferase activity was normalized to Renilla luciferase activity and represented as relative luciferase units (RLU).

4.5. Chromatin Immunoprecipitation (ChIP) assay

RET2A cells seeded in 100 cm plates were grown to 60–80% confluency for transfections using the Polyfect (Qiagen, Valencia, CA, USA) transfection reagent according to manufacturer's protocol. For each transfection, 10 μg of the control or N‐Myc siRNA plasmid was used. One hour prior to tetracycline addition, PD98059 (Calbiochem, San Diego, CA, USA) was added to a final concentration of 100 μM where indicated. At 24 h post‐transfection, tetracycline (2 μg/ml) or 100% ethanol (vehicle) were added. 12 h after treatment with Tet (or ethanol) cells were treated with 0.25% formaldehyde (Sigma, St. Louis, MO, USA) followed by 1 M glycine (Sigma, St. Louis, MO, USA). Cells were then washed with PBS + protease inhibitor cocktail (100 mM aprotinin, 100 mM Laupeptin, 100 mM trypsin inhibitor and 100 mM benzamidine). Cytoplasmic lysis was accomplished by using the ChIP lysis buffer (10 mM Hepes pH 7.9, 10 mM KCl, 1.5 mM MgCl2, 0.01% 2‐mercaptoethanol, 0.1% NP‐40, 1 mM protease inhibitor cocktail) followed by nuclear lysis using the ChIP dilution buffer (0.1% sodium dodecoyl sulfate, 1% triton X‐100, 16 mM Tris–HCl pH 8.1, 165 mM sodium chloride). Chromatin was fragmented by sonicating three times for 15 s pulses at an interval of 5 min. Fragmented chromatin was then pre‐cleared using protein A sepharose beads (Pierce Biotechnologies, Rockford, IL, USA) with rocking for 1 h. Cleared lysates were then used for overnight immunoprecipitation with 10 μg of an N‐Myc antibody (M − 50) (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Precipitates were reverse‐cross‐linked (250 mM NaCl) and treated with Proteinase K (Promega, Madison, WI, USA). DNA was purified using a PCR purification kit (Qiagen, Valencia, CA, USA) according to manufacturer's protocol. Equal amounts of DNA and SYBR Green PCR Master Mix (Applied Biosystems, Foster City, CA, USA) were used in each reaction of qPCR in the ABI prism 7300 sequence detection system (Applied Biosystems, CA, USA). The qPCR parameters were: 50 °C/2 min–one cycle; 95 °C/10 min–one cycle and 95 °C/15 s, 60 °C/1 min–40 cycles. Data were analyzed, according to the manufacturer's recommendations to obtain fold induction of N‐Myc amplicon relative to a non‐specific amplicon, and subjected to statistical analysis using Student's t‐test. Primers specific for the N‐Myc amplicon, p18ChIP‐NMyc forward primer (5′‐TGGGCTCACTTTTGCTGAATAA‐3′) and p18ChIP‐NMyc reverse primer (5′‐ GTCGGCCTGCCAAAAGC‐3′) or non‐specific amplicon, p18ChIP‐control forward primer (5′‐ CCCGGCGGTTTTGGTT‐3′) and p18ChIP‐control reverse primer (5′‐ CAGGACAGCCAGGGCTATACA‐3′) were used (Figure 4A).

4.6. Fluorescence activated cell sorting (FACS) analysis

Cells were harvested and fixed as described previously (Juan and Darzynkiewicz, 2001). Fixed cells were washed with phosphate buffered saline and treated with 100 mg/ml RNAse for 5 min and stained with 50 mg/ml propidium iodide for 30 min in dark. 675 nm emission from the DNA bound propidium iodide was measured with a Coulter FC‐500 flow cytometer. Approximately 10,000 cells from each sample were analyzed. The computer program ModFit was used to determine the percentages of cells within the G₁, S and G₂/M phases of the cell cycle. The proliferative index was calculated by obtaining the sum of all S and G₂/M‐phase populations in any given sample. Then, relative proliferative index was calculated by taking the ratio of the induced (+Tet) to uninduced (−Tet) samples for a given sample. Student's t‐test was performed to obtain P‐values.