MXD3 regulation of DAOY cell proliferation dictated by time course of activation
© Ngo et al.; licensee BioMed Central Ltd. 2014
Received: 3 December 2013
Accepted: 2 July 2014
Published: 23 July 2014
Skip to main content
© Ngo et al.; licensee BioMed Central Ltd. 2014
Received: 3 December 2013
Accepted: 2 July 2014
Published: 23 July 2014
MXD3 is a basic-helix-loop-helix-leucine-zipper transcription factor involved in cellular proliferation. In previous studies we demonstrated that knock-down of MXD3 in the human medulloblastoma cell line DAOY resulted in decreased proliferation. Surprisingly, overexpression of MXD3 in DAOY cells also decreased proliferation and increased cell death, suggesting that persistent expression of MXD3 triggers an apoptotic response, perhaps as a fail-safe mechanism. To investigate this apparent paradox in detail we developed a tamoxifen inducible system to analyze the temporal effects of MXD3 in the proliferation and transcriptional response of DAOY cells upon acute induction compared with long-term expression of MXD3.
We find that acute induction of MXD3 initially promotes cell cycle progression as assessed by a transient increase in bromodeoxyuridine incorporation. However, persistent induction of MXD3 ultimately results in decreased proliferation based on cell counts. Finally, with microarray expression profiling and gene ontology analysis we identify several major pathways enriched in response to acute (immune response, apoptosis, cell cycle) versus persistent (cell adhesion) MXD3 activation.
In this study, we demonstrate that acute MXD3 activation results in a transient increase in cell proliferation while persistent activation of MXD3 eventually results in an overall decrease in cell number, suggesting that the time course of MXD3 expression dictates the cellular outcome. Microarray expression profiling and gene ontology analysis indicate that MXD3 regulates distinct genes and pathways upon acute induction compared with persistent expression, suggesting that the cellular outcome is specified by changes in MXD3 transcriptional program in a time-dependent manner.
Medulloblastoma, the most common brain tumor in children , develops due to uncontrolled proliferation of cerebellar granule neuron precursors (GNPs) . A large body of literature exists regarding the molecular mechanisms of medulloblastoma formation and progression. Thus far, four subtypes of medulloblastomas have been identified including the Wnt and Sonic hedgehog (Shh) subgroups [3, 4]. Medulloblastomas of the Shh subgroup have mutations in upstream components of the Shh pathway, including the receptors PTCH and SMO . PTCH, when bound by Shh, relieves its inhibition of SMO  which then initiates a complex cascade of events leading to cell cycle progression. One example of an established mechanism for Shh pathway-dependent cell cycle progression is through the upregulation of cyclins by the proto-oncogene MYCN . Mutation of downstream targets of Shh such as GLI1, GLI2 , and MYCN is a characteristic of medulloblastomas within the Shh subtype . However, the molecular mechanisms behind medulloblastoma formation and progression are not completely understood. Indeed, recent evidence suggests that a subset of cerebellar granule neurons originate not from GNPs but from a population of Nestin-expressing progenitors (NEPs) in the deep external germinal layer and that these NEPs are more susceptible to Shh-dependent tumor formation .
Previously, our lab identified the transcription factor MXD3 as a critical regulator of GNP proliferation during normal cerebellar development as a downstream component of the Shh pathway . Interestingly, we found that MXD3 is overexpressed in tumor tissue from the PTCH deficient heterozygote mouse model of medulloblastoma . Moreover, recently we showed that MXD3 is upregulated in human medulloblastomas and is required for the proliferation of the human medulloblastoma cell line DAOY . These results suggest a role for MXD3 in medulloblastoma in humans. The DAOY cell line was established from a medulloblastoma tumor mass obtained from a 4 year old patient . Tissue from this tumor had evidence of both neural and glial differentiation; however, these characteristics were lost during the establishment of DAOY as a cancer cell line .
MXD3 is a basic-helix-loop-helix-leucine-zipper (bHLHZ) transcription factor that is part of the MYC/MAX/MXD transcriptional network . Within this network, MYC and MXD family members compete with each other for MAX heterodimerization to invoke opposing transcriptional regulation of target genes [13, 14]. Specifically, MYC and MAX heterodimers recruit transcriptional activators  while MXD and MAX heterodimers recruit transcriptional repressors [13, 16, 17]. MYC family members have been shown to promote while MXD family members have been shown to repress cell cycle progression . MXD3, however, is an atypical member of the MXD family as it has been found to be expressed during the S-phase of the cell cycle [9, 19, 20] while other MXD family members are expressed in differentiated cells . Knockdown of MXD3 leads to a reduction in cell number suggesting that MXD3 is required for cell cycle progression [9, 10]. On the other hand, overexpression of MXD3 is sufficient to promote proliferation in mouse cerebellar GNPs . Consistent with these results, overexpression of MXD3 negatively regulates differentiation in mouse B cells derived from the spleen . Persistent overexpression of MXD3, however, in mouse GNPs and in human medulloblastoma cells results in decreased proliferation due to the activation of apoptosis [9, 10].
To characterize MXD3 overexpression in a time dependent manner, we engineered the DAOY cell line to express stably a fusion protein between the truncated Estrogen Receptor and MXD3 (ER-MXD3). In contrast to endogenous MXD3, which is localized to the nucleus , under baseline conditions the ER-MXD3 fusion protein is enriched in the cytoplasm. Upon treatment of 4-hydroxytamoxifen (4-OHT) the ER-MXD3 fusion protein translocates into the nucleus allowing for the timed activation of MXD3. Here we show that the nuclear translocation of ER-MXD3 initially leads to a transient increase in cell proliferation based on bromodeoxyuridine (BrdU) incorporation but ultimately results in an overall decrease in cell number. Furthermore, we identify candidate MXD3 regulated genes upon acute induction and long-term expression to investigate the opposing activities of MXD3 in the regulation of cellular proliferation in DAOY cells.
We have previously shown that MXD3 knock-down reduced proliferation of DAOY medulloblastoma cells, while persistent overexpression also decreased cellular proliferation . To distinguish between MXD3’s acute versus long-term effects in DAOY medulloblastoma cells, we developed 4-OHT inducible cell lines that express MXD3 as a fusion to a portion of the mouse estrogen receptor (254 C-terminal amino acid residues, lacking its DNA-binding domain). Furthermore, we have previously found that MXD3 activity is abolished upon mutation of a single amino acid at the 66th position in the basic domain of MXD3 (MXD3.E66D) . The E66D mutation has been shown to disrupt the basic domain binding of other bHLH proteins to the E-box DNA sequence in gel shift assays . Therefore, as a negative control for subsequent experiments, we developed cell lines stably expressing the inducible fusion ER-MXD3.E66D for comparison.
The MXD3 promoter region has been shown to be regulated by the transcription factor E2F1 , a critical transcriptional activator for the transition from the G1 to the S phase of the cell cycle (reviewed in ). In agreement, several groups have shown that MXD3 is expressed specifically in the S phase of the cell cycle [9, 19, 20]. The timing of MXD3 expression suggests that it may play a role in cell cycle progression through the S phase. In support of this possibility, it has been observed that a transient increase in proliferation occurs in both normal mouse GNPs  and now in this study of human medulloblastoma cells (Figure 2). The fact that this observation was made in both models suggests that the phenotype observed in response to MXD3 overexpression is a conserved aspect of MXD3 function.
Additionally, the overall decrease in cellular proliferation in response to persistent MXD3 overexpression is also true in both normal  and diseased  (Figure 3) models. This decrease in proliferation can be explained by the activation of apoptosis as a fail-safe mechanism upon persistent overexpression beyond the S phase of the cell cycle. Such a mechanism exists for oncogenes such as the transcriptional network relative MYC [26, 27]. In support of this possibility, both transient overexpression of MXD3 in GNPs  and stable MXD3 overexpression in human medulloblastoma cell lines  show an increase in apoptosis. When we examined apoptosis activity at 72 hours in our inducible cell lines, we were unable to find any significant difference in caspase 3/7 activity in response to MXD3 activation (Additional file 6: Figure S6); however, there is a trend towards increased caspase 3/7 activity at 72 hours for ER-MXD3, consistent with the decreased cell number observed. Interestingly, the maximal response to hydrogen peroxide treatment (used as a positive control in this assay) in the ER-MXD3 line was 0.608-fold less when compared to the control ER-MXD3.E66D cell line (Additional file 6: Figure S6), suggesting that the capacity of the ER-MXD3 cell line to undergo apoptosis is reduced even in the absence of MXD3 induction. The inherent leakiness in our inducible system could account for this difference between the cell lines. If this is true then it would suggest that at low nuclear concentrations MXD3 functions as an anti-apoptotic factor leading to the subsequent difference in the response to hydrogen peroxide under baseline conditions. Further experimentation will be necessary to test this intriguing possibility.
The results presented thus far indicate that MXD3 has a dual role in DAOY cell proliferation, as we suggested before , and that its role is dependent on how long MXD3 is active or present in the nucleus. We report here an initial burst in proliferation (as measured by BrdU incorporation within 12–24 hours) followed by decreased cell counts (at 72 hour) upon MXD3 translocation to the nucleus. Some remaining questions, then, are whether the observed phenotypes due to MXD3 activation are the result of two distinct mechanisms and/or whether the observed phenotypes are the direct result of MXD3 function. These questions remain for both normal and diseased models. To begin to address these questions in the context of human medulloblastoma, we examined the pathways changed in response to MXD3 activation in our inducible cell lines.
To this end, we conducted microarray experiments to compare gene expression between the “early” (increased BrdU incorporation) and “late” (decreased cell counts) effects of MXD3 overexpression. Samples were taken from 12 and 72 hours post treatment from both ER-MXD3 and ER-MXD3.E66D cell lines, in order to define acute and long-term changes in the pattern of gene expression elicited by MXD3. Differentially expressed genes were defined as those genes that showed greater than 2-fold changes in ER-MXD3 (4-OHT/vehicle) over ER-MXD3.E66D (4-OHT/vehicle). It should be noted that this approach aimed at identifying changes that require MXD3 binding to the DNA through its basic domain, as this interaction has been reported to be disrupted in the E66D mutation [10, 23, 28–30]. A complete list of differentially expressed genes is presented in Additional file 7: Table S1.
In this study, we use 4-OHT inducible human medulloblastoma cell lines to show that MXD3 activation results in a transient increase in cell proliferation. In corroboration with previous studies, persistent activation of MXD3 eventually results in an overall decrease in cell number. With microarray expression profiling we report candidate downstream targets of MXD3 differentially regulated upon acute versus persistent expression of MXD3. Lastly, with gene ontology analysis we identify several major pathways enriched in response to acute (immune response, apoptosis, cell cycle) versus persistent (cell adhesion) MXD3 activation that provide insight into the opposing roles of MXD3 in medulloblastoma proliferation in a time dependent manner.
pCMV-HAER (gift from Peggy Farnham) was the backbone vector for the constructs used in this study. HA-MXD3  was subcloned into pCMV-HAER using BamHI restriction sites to produce pCMV-HA-ER-HA-MXD3 (HA = hemagglutinin tag, ER = truncated estrogen receptor). The ER-MXD3.E66D mutant construct was produced using the Quikchange II Site-Directed Mutagenesis Kit (Agilent) according to manufacturer’s instructions. All constructs were verified by sequencing.
DAOY cells were acquired from ATCC and were cultured in a standard humidified incubator (5% CO2, 37°C). Culture media for DAOY (DAOY media) consisted of minimum essential media (Invitrogen) supplemented with 10% Fetal Bovine Serum (Invitrogen), 1 mM Sodium Pyruvate (Invitrogen), 100 μg/ml Penicillin/Streptomycin (Invitrogen). Stable cell lines were maintained with DAOY media supplemented with 800 μg/ml of G418 Sulfate (Cellgro).
Transfections were performed with Fugene HD transfection reagent (Roche) according to the manufacturer’s instruction. A 5:2 ratio of DNA to transfection reagent was used in the initial transfection for the production of stable cell lines. Briefly, 48 hours after initial seeding, 7 μg of respective constructs was transfected into DAOY cells. Stable cell lines were established using G418 selection (800 μg/ml) 48 hours post-transfection of respective constructs. After one week, cells were passaged to 96-well plates at 1 cell/well. Clones were expanded and subsequently stable expression was confirmed via immunoblotting and immunocytochemistry.
The following primary antibodies were used in this study: rat anti-HA (Roche), mouse anti-MXD3 (NeuroMab), mouse anti-β-tubulin (Millipore). Secondary antibodies were as follows: goat anti-rat-Cy3 (Jackson ImmunoResearch), donkey anti-mouse-Cy5 (Jackson ImmunoResearch), anti-goat-horse radish peroxidase (HRP) (Vector Labs), and anti-mouse-HRP (MP Biomedical).
Cell extracts were prepared with lysis buffer at pH 7.4 consisting of 150 mM NaCl, 50 mM Tris, 1% Triton-X100, 23.4 μM Leupeptin (Roche), 6.1 μM Aproptinin (Roche), 14.5 μM Pepstatin A (Roche), and 0.1 mM PMSF (Millipore). Protein concentration was determined with a Micro BCA Protein assay kit (Thermo Scientific). Extracts were separated on 12% acrylamide gels under denaturing and reducing conditions and transferred to nitrocellulose membranes. Blots were developed using standard film methods with HRP conjugated secondary antibodies in conjunction with Luminata Crescendo HRP substrate (Millipore).
Cells were grown on poly (L)-lysine coated glass coverslips in 6-well plates. Upon collection, cells were washed once with phosphate buffered saline (PBS) and then fixed with 4% paraformaldehyde (Millipore) and permeabilized with 0.01% Triton-X100 in PBS. Coverslips were then incubated overnight at 4°C with primary antibodies in 5% bovine serum albumin in PBS. Secondary antibody incubations were performed in 5% bovine serum albumin in PBS at room temperature for two hours. Subsequently, DAPI staining was performed in PBS at room temperature for 10 minutes. Lastly, coverslips were mounted onto Superfrost Plus microscope slides (Fisher Scientific) using Fluoromount-G (SouthernBiotech) and subsequently imaged with a LSM 710 confocal microscope (Carl Zeiss).
Cells were seeded at 15,000 cells/well in DAOY media supplemented with ethanol (vehicle) at 1:1000; we found that ethanol has an initial positive effect on cellular proliferation (data not shown) and thus we seeded cells in culture media with vehicle in order to account for this initial effect. After 48 hours, cells were treated with 1 μM 4-OHT (Sigma Aldrich) dissolved in ethanol and allowed to proliferate. Cells were counted in triplicate with a Coulter Counter Z1 (Beckman Coulter) after trypsinization. For Bromodeoxyuridine (BrdU) incorporation measurements, BrdU Cell Proliferation Kit (Millipore) was used according to the manufacturer’s instructions with cells seeded at 2×103 cells/well in a 96-well plate. 4-OHT treatment times were staggered such that all samples were processed and assayed simultaneously for BrdU incorporation after a two hour incubation with BrdU.
RNA samples were purified with an RNeasy kit (Qiagen) and quality checked using an Agilent Bioanalyzer. Subsequent reverse transcription and labeling reactions were conducted with Amino Allyl MessageAmp II aRNA Kit (Ambion). cDNA from each time point was collected and labeled with either Cy3 or Cy5 (source). Two control samples were generated by pooling all Cy3 or all Cy5-labeled samples at each time point. Subsequently, samples were hybridized to Whole Human Genome 4×44K microarrays (Agilent) with their associated opposing dye labeled control (e.g. ER-MXD3.12 hours.of.4-OHT.Cy3 + Pool.of.all.samples.Cy5). Microarrays were imaged using an Axon GenePix 4000B microarray scanner (Molecular Devices) and feature extraction was conducted using GenePix Pro 6.0. Fluorescence intensity of features of each sample was normalized to a dye swap pool. Subsequently, 4-OHT time points were normalized to the initial time point or vehicle control producing a ratio of (4-OHT/ethanol). Ratios, then, between ER-MXD3 and ER-MXD3.E66D cell lines were compared to identify MXD3 candidate regulated genes. We defined hits as genes that were up/down-regulated more than two times in ER-MXD3 (4-OHT/ethanol) compared to ER-MXD3.E66D (4-OHT/ethanol). Gene ontology analysis was performed with MetaCore (Thomson Reuters) using the “compare experiments” module with default parameters.
Statistical significance of the BrdU and cell proliferation assays was analyzed via two-way ANOVA with Bonferroni post-tests using GraphPad Prism (GraphPad Software).
The microarray gene expression dataset supporting the results of this article is available in the Gene Expression Omnibus (GEO) repository, [GSE5903; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE58903].
Granule Neuron Precursors
Truncated Estrogen Receptor and full-length MXD3
The authors would like to thank the labs of Drs. Donald Bers, Angela Gelli, and Heike Wulff for sharing of lab space, equipment, and reagents. Lastly, the authors would like to thank the members of the Diaz lab for all their advice, help, and support throughout the course of the study.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.