N-myc Regulates Parkin Expression*

Mutations in the parkin gene are common in early-onset and familial Parkinson’s disease (PD), and the parkin protein interacts in the ubiquitin-proteasome system as an E3 ligase. However, the regulatory pathways that govern parkin expression are unknown. In this study, we showed that a phylogenetically conserved N-myc binding site in the bi-directional parkin promoter interacted with myc-family transcription factors in reporter assays, and N-myc bound to the parkin promoter in chromatin immunoprecipitation assays and repressed transcription activity. Parkin expression was inversely correlated with N-myc levels in the developing mouse and human brain, in human neuroblastoma cell lines with various levels of n- myc amplification, and in an inducible N-myc cell line. Although parkin and N-myc expression were dramatically altered upon retinoic ac-id-induced differentiation of a human neuroblastoma cell line, modulation of parkin expression did not significantly affect either rates of cellular proliferation or levels of cyclin E. Analysis of additional genes associated with familial PD revealed a shared basis of diphenyltetrasodium bromide tetrazolium salt to measure metabolic activity, which correlates with the number of cells in a given sample.

neuroprotective effects of parkin overexpression in cells simultaneously challenged with various insults (reviewed by Feany and Pallanck (4)). Given the plethora of potential parkin substrates and numerous cellular pathways parkin might interact in, the timing and localization of parkin expression may be a critical feature of gene function.
Parkin expression in the mammalian brain is largely neuron-specific and developmentally regulated, perhaps implying a role for parkin in neuronal maturation (5,6). Evidence suggesting that the parkin gene is frequently mutated in breast and ovarian cancer (7,8) is noteworthy in this regard, because it raises the possibility that parkin may be intimately involved in cell cycle regulation, perhaps even functioning as an effective tumor suppressor in the periphery. Parkin protein may directly interact in the cell cycle as a component of a Skp1cullin-F-box-like ubiquitin ligase complex capable of targeting cyclin E for degradation (9). Disruption of cell cycle regulation could therefore be plausibly invoked as a mechanism leading to neuronal degeneration as a result of parkin dysfunction.
To explore the nature of parkin's developmentally regulated pattern of expression and potential interaction in the cell cycle, we analyzed the parkin promoter for regulatory elements known to interact with transcription factors involved in cell cycle control. We identified a transcription regulation domain, conserved across several species, containing a sequence similar to an E-box motif, a promoter element known for interacting with the myc family of proteins. We demonstrate that parkin expression inversely correlates with expression of N-myc, a transcription factor critically involved in both neuronal development (10) and tumorigenesis (11). N-myc also regulates transcription of other familial PD genes in an inducible cell line, thereby suggesting a common biological pathway among genes associated with PD.
Dual-luciferase Assay-Cells were plated 24 h before transfection into 24-well culture plates at ϳ80% confluence, and transfection was performed with Fugene 6 reagent (Roche Biochemicals), using 0.2 g of DNA per well in a 1:3 ratio of DNA/Fugene reagent, and added to cells in serum-free media for 12 h. Luciferase-containing constructs (pGL3) were co-transfected with pRL-SV40 renilla vector (Promega, Madison, WI) to control for transfection efficiency, in a molar ratio of 1:40 (pRL-SV40 versus pGL3 vectors). Forty hours after transfection, cells were rinsed with phosphate-buffered saline and then harvested with passive lysis buffer (Promega). The dual-luciferase system (Promega) was used according to the manufacturer's protocol, and experiments were repeated in six wells. Readings were taken in duplicate on a Turner Designs 20/20 single injector luminometer (Promega).
Chromatin Immunoprecipitation-The N-myc-inducible cell line was plated at 10 7 cells per 100-cm dish and cultured for 5 days in the presence or absence of 1 g/ml doxycycline. Cultures were rinsed with phosphate-buffered saline and cross-linked for 15 min at 37°C in serum-free medium containing 1% formaldehyde. Media was then removed, and un-reacted formaldehyde was quenched by incubation in 125 mM glycine for 10 min at room temperature. Cultures were washed, and cells were collected by scraping into phosphate-buffered saline containing protease inhibitors (0.2 mM phenylmethylsulfonyl fluoride and 1 g/ml each of aprotinin, pepstatin B, and leupeptin). Cell pellets were re-suspended in 1 ml of lysis buffer (1% SDS, 10 mM EDTA, and 50 mM Tris-HCl, pH 8.0) containing protease inhibitors and incubated on ice for 10 min. DNA was then sheared by sonication to an average size of 300 bp and samples were centrifuged. 0.2-ml supernatants from control and doxycycline-treated samples were diluted 1:10 with 0.01% SDS, 1.1% Triton X-100, 1.2 mM EDTA, 167 mM NaCl, and 16.7 mM Tris-HCl containing protease inhibitors, and 5% was set aside as input DNA. Samples were pre-cleared for immunoprecipitation by incubation with protein A-agarose blocked with 1 mg/ml of sonicated salmon sperm DNA, 1 mg/ml BSA, 1 mM EDTA, 10 mM Tris-HCl, pH 8.0. 5 g of either rabbit anti-N-myc (Santa Cruz Biotechnology, Santa Cruz, CA) or control IgG were added to each sample, and incubations were continued overnight at 4°C. Immune complexes were collected with protein Aagarose and washed with 0.1% SDS, 1% Triton X-100, 2 mM EDTA, 20 mM Tris-HCl, pH 8.0, containing 150 mM NaCl, and then 500 mM NaCl, 0.25 M LiCl, 1% Nonidet P-40, 1% deoxycholate, 1 mM EDTA, 10 mM Tris-HCl, pH 8.0, and finally Tris-EDTA. To simultaneously elute immune complexes and reverse cross-links, agarose-bound adducts were incubated overnight at 65°C in 1% SDS, 0.1 M NaHCO 3 , and 0.2 M NaCl. Samples of Input DNA were made 0.3 M in NaCl and also incubated overnight at 65°C. DNA was purified using the Qiagen PCR clean-up kit (Qiagen, Valencia, CA).
Real-time quantitative PCR analysis of the number of copies of parkin promoter DNA in input DNA, N-myc, or control IgG samples was performed on a Light Cycler system using Qiagen QuantiTect SYBR-Green PCR reagents and primers that flank the parkin promoter E-box sequence (forward, 5Ј-TAG AAC TAC GAC TCC CAG CAG GC-3Ј; reverse, 5Ј-CGG CTC TCC TGG GTT AAA TCC TC-3Ј). For each sample analyzing the parkin E-box sequence, copy number of parkin promoter DNA was determined by comparison to a standard curve composed of 100 -10,000 copies of linearized plasmid DNA containing parkin promoter sequence versus the crossing-over point. Control primers amplifying 4 kb upstream of the parkin E-box sequence were also designed (forward, 5Ј-ACC TGT CAG CCT CTC TTG CAA CTA G-3Ј; reverse, 5Ј-CCC AGA AAC AGC AAT CCT CAC TCC-3Ј) in addition to primers that span the E-box sequence in the telomerase promoter (forward, 5Ј-AAG GTG AAG GGG CAG GAC GGC-3Ј; reverse, 5Ј-GAG TGG ATT CGC GGG CAC AGA-3Ј). All samples were assayed in triplicate, and N-myc and control IgG copy number were normalized with respect to Input DNA.
Western and Northern Blot Analysis-Western blots were performed as described previously (13). Antibodies to parkin (Cell Signaling Technology, Beverly, MA), N-myc, and cyclin E (both from Santa Cruz Biotechnology) were used according to manufacturer's recommended conditions. In all experiments, a pre-stained molecular weight marker confirmed the expected size of the target proteins. For internal controls, the blots were stripped and reprobed with monoclonal antibodies against ␤-actin (Sigma-Aldrich). Northern blotting was performed as described previously (12). To generate cDNA probes, primers specific to mouse parkin (forward, 5Ј-CAA CTC CCT GAT TAA AGA GCT CCA TC-3Ј; reverse, 5Ј-GGC TGA GGA CAC TTC ATG TGC ATA C-3Ј) and N-myc transcript (forward, 5Ј-GCT AGA GCG CGC AGT GAA CG-3Ј; reverse, 5Ј-CTG AGT CGC TGA AGG TAT CCT CTC C-3Ј) were used in a PCR reaction with mouse brain cDNA (BD Biosciences Clontech). Probes were radiolabeled and hybridized to a mouse embryo MTN blot (BD Biosciences Clontech) as described previously.
Cell Proliferation Assay-To measure the rate of cellular proliferation, 3 ϫ 10 5 cells were plated in replicate in 6-well dishes, with or without dox (1 g/ml). At a given time interval, cells were harvested by trypsinization, collected by centrifugation, and then re-suspended in phosphate-buffered saline containing trypan blue (Sigma-Aldrich). Cells were manually counted on a hemocytometer by an investigator blinded to the identity of the culture, and cells that stained blue (Ͻ5% on average) were excluded. In addition, the One Solution cell proliferation assay was used according to manufacturer's recommended conditions (Promega). This assay uses the 3-[4,5-dimethylthiazol-2-yl]-2,5diphenyltetrasodium bromide tetrazolium salt to measure metabolic activity, which correlates with the number of cells in a given sample.

Identification of an Evolutionarily Conserved Potential Nmyc Binding
Sequence within the Parkin Promoter-We identified previously an E-box motif in the parkin/PACRG promoter 46 bp upstream of the parkin transcription initiation site that interacted with nuclear protein derived from human substantia nigra and overlapped with a region responsible for transcription regulation (12). To identify functional domains near the parkin E-box motif, we sequenced the parkin promoter in a number of evolutionarily divergent species using PCR primers placed in the more highly conserved exonic regions of the parkin and PACRG genes. As a whole, the parkin promoter displayed a conservation of 52% between human and mouse, in contrast to 83% conservation in the parkin open reading frame. We resolved a palindromic phyloprint (evolutionarily conserved promoter element) containing the E-box motif (CGCGTGCGC) 44 bp upstream of the parkin transcription initiation site (Fig. 1A).
Using the MatInspector protein binding prediction program (www.genomatix.de), a number of proteins from the myc family of transcription factors were predicted to interact with the parkin promoter E-box, with highest similarity to the N-myc consensus binding domain matrix (Fig. 1B). The CGCGTG motif in the Pax-3 promoter has previously been shown to bind both N-myc and max in gel-shift assays (14). The CGCGTG motif found in the parkin and pax-3 promoter differs from the canonical E-box sequence of CACGTG by 1 base pair, yet both sequences are known to strongly interact with myc-family proteins (15). In humans and mice, neither the CGCGTG nor CACGTG sequence motif is repeated within 20 kbp of genomic sequence about the parkin promoter E-box.
To determine the importance of the parkin promoter E-box and myc protein expression on transcription regulation, a DNA construct containing sequence from the first exon of parkin to the E-box sequence was inserted upstream of a promoter-less luciferase reporter gene in the pGL3-basic vector (pGL3-154). This plasmid was transiently transfected along with the control vector pRL-SV40 into the SK-N-BE (2)-M17 cell line, and strong transcription activation relative to control plasmid (pGL3-basic) was observed (Fig. 1C), as we have reported previously (12). To examine whether N-myc might affect this transcription activity, reporter assays were conducted in parallel with the overexpression of the N-myc transcription factor, max, or both N-myc and max. In all cases, the transcription activity of the parkin promoter construct was reduced; however, transcription activity of the SV40 promoter (pGL3-SV40) was not significantly affected, thereby suggesting activity in pRL-SV40, used to control for transfection efficiency, was also unaffected (data not shown). Taken together, these data suggest that N-myc repressed transcription activation in the parkin promoter reporter constructs.
Parkin Expression Is Down-regulated by N-myc-N-myc-induced down-regulation of transcription in reporter assays does not necessarily convey a physiologically significant interaction in which N-myc regulates parkin expression. To determine whether modulation of N-myc expression alters endogenous parkin levels, we first examined parkin expression in an Nmyc-inducible (tet-off) cell line. This cell line is of human SH-EP neuroblastoma origin and endogenously possesses a single copy of the N-myc gene, which expresses protein at very low levels (16). Promoter constructs containing either the pGL3-SV40 control vector or the pGL3-154 parkin promoter vector were transiently transfected into the N-myc-inducible cell line. Control transfections with GFP demonstrated a low level of transfection efficiency in these cells (Ͻ5%, data not shown), however an increase in the transcription activity of the parkin promoter was observed when dox was included in the culture media ( Fig. 2A). Next, parkin and N-myc mRNA was measured in the inducible N-myc cells grown for 1 week in the presence or absence of dox. Compared with levels of ␤-actin control mRNA, parkin mRNA was ϳ5-fold less abundant in the cells grown without dox, a condition under which N-myc mRNA was increased more than 10-fold (Fig. 2B). We further confirmed the inverse relationship in expression of N-myc and parkin by measuring protein derived from the N-myc-inducible cells (Fig. 2C).
N-myc Binds to the Parkin Promoter-To determine whether N-myc might be directly binding to the chromatin/nucleoprotein complex near the E-box sequence in the parkin promoter, we employed a chromatin immunoprecipitation strategy using the N-myc-inducible cell-line and a well characterized antibody directed against N-myc (Fig. 3A). PCR primers were designed to The position and orientation of the predicted interaction is given by the black arrows, relative to indicated sequence. C, luciferase activity (normalized to renilla activity) is given, where 1 unit equals the activity of the control vector, pGL3-SV40, driven by the SV40 early promoter, as determined using the dual-luciferase assay. Plasmids encoding either the promoterless pGL3-Basic or the parkin promoter (pGL3-154) were transiently co-transfected in addition to transcription factor(s) into BE (2)-M17 cells and harvested 40 h later. The molar ratios of vectors in each experiment were conserved by supplementing with no-insert plasmid DNA (pcDNA3.1) as necessary. Error bars represent 2ϫ S.E. of at least three independent experiments. *, p Ͻ 0.001; **, p Ͻ 0.0001, with respect to the transcription activity of pGL3-154.
flank the E-box sequence in the parkin promoter, and quantification of precipitated DNA was achieved using a SYBR green fluorescence PCR system. In cells grown in the presence of dox, insignificant amounts of the parkin promoter were precipitated by a polyclonal antibody specific to N-myc, relative to IgG controls (Fig. 3, B-D). On the other hand, in cells grown without dox (and therefore high levels of N-myc; see Fig. 2, B and C), ϳ18 times the copies of the parkin promoter were immunoprecipitated relative to cells grown with dox (Fig. 3C), thereby demonstrating that N-myc interacts with the chromatin complex at or near the E-box sequence in the parkin promoter.
To further validate this system, we analyzed a region 4 kb upstream of the parkin E-box sequence in the parkin promoter that does not contain sequence similar to an E-bo, and would not be expected to immunoprecipitate with the anti-N-myc antibody. Insignificant amounts of DNA immunoprecipitated in cells grown with or without dox when normalized to IgG controls (Fig. 3D). We also analyzed a known N-myc binding site by designing primers that flank an E-box motif in the human telomerase gene. In agreement with a previous study (17), approximately seven times the amount of DNA precipitated from cells grown without dox versus IgG or cells grown with dox (Fig. 3D).
Parkin Expression Inversely Correlates with N-myc during Development and in Human Neuroblastoma Cell Lines-Northern analysis of the amount of parkin and N-myc mRNA in poly-A purified RNA derived from mouse embryos showed parkin expression to be restricted to the latest stages of embryonic development (day 17), whereas N-myc expression was highest during mitotic stages of neuron growth (day 11; Fig. 4A), a

FIG. 3. Chromatin immunoprecipitation.
A, schematic of the chromatin immunoprecipitation (ChIP) procedure. DNA from N-myc-inducible cells, grown with or without doxycycline for 1 week, was cross-linked and sheared. Immunoprecipitation (IP) was carried out with anti-N-myc antibody or IgG control. Cross-links were reversed and DNA was PCR-amplified. B, representative amplification plots, generated by an ABI 7900HT. C, copies of the parkin promoter were immunoprecipitated with either IgG control or anti-N-myc antibody. Absolute quantitative PCR was achieved by comparison with a standard curve generated by using known amounts of the parkin promoter in linearized plasmid DNA (pGL3-154; data not shown). D, the amount of DNA precipitated with anti-N-myc was normalized to the amount of DNA precipitated with IgG (nonspecific binding). Amplification was performed with the indicated primer sets, including primers located 4 kbp upstream of the parkin E-box sequence (Up-stream), primers that flank the E-box in the parkin promoter (Parkin E-box) and primers that flank the E-box in the telomerase promoter (Telomerase E-box). Error bars represent 2ϫ S.E., and measurements were conducted in triplicate.
finding consistent with previously published results (18,19). Likewise, quantitative RT-PCR analysis of fetal and adult human brain tissue showed that roughly double the parkin mRNA is present in the adult brain versus that of the fetal brain, measured relative to control mRNA (Fig. 4B).
In addition to the developing mammalian embryo, N-myc expression is also highly variable in human neuroblastoma. Neuroblastoma accompanied by amplification of the N-myc gene is associated with enhanced tumor aggressiveness and a poor clinical outcome (20). Most neuroblastoma derived cell lines have N-myc amplification; however, both the SH-SY5Y line and the SK-N-F1 line possess only a single copy of N-myc. Western blotting was used to measure relative parkin and N-myc expression in various neuroblastoma cell lines. Again, parkin expression inversely correlated with N-myc amplification (Fig. 4C).
Retinoic Acid Up-regulates Parkin Expression-Retinoic acid is known to potently down-regulate N-myc expression and to induce growth arrest and differentiation in neuroblastoma cells (21). SK-N-BE(2)-M17 cells cultured for 10 days in media containing retinoic acid developed a complex network of cell processes indicative of differentiated and post-mitotic neurons (Fig. 5, A and B). Levels of N-myc and parkin were analyzed at days 3 and 10 of exposure to retinoic acid. N-myc mRNA decreased ϳ20-fold at the day 10 (data not shown), whereas parkin mRNA doubled relative to control mRNA (Fig. 5C). Western blot analysis demonstrated an increase of parkin protein in differentiated cells relative to control proliferating cells and a corresponding decrease in N-myc and cyclin E protein (Fig. 5 D).
Parkin Overexpression Does Not Affect Cellular Proliferation or Levels of Cyclin E-To determine whether parkin may play a role in the cessation of cellular proliferation, growth rates were measured, over the course of 96 h, in PC-12 cells that express inducible human parkin (tet-off system). Although the removal of dox in the N-myc-inducible cell line (thereby increasing N-myc expression) dramatically increased cellular proliferation (Fig. 6A, f versus Ⅺ), as has been previously reported (16), increasing parkin expression by removal of dox from the PC-12 cell media did not produce a significant change in cell growth rate (Fig. 6A, OE versus ‚). Cell growth rates were also measured using the 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrasodium bromide cell proliferation assay, and similar results were obtained (data not shown).
Parkin has been proposed to participate in the degradation of cyclin E, an important regulator of the cell cycle, via the ubiquitin proteasome system (9). To determine whether modulation of parkin expression might affect endogenous levels of cyclin E, protein levels were measured in lysates derived from the parkin-inducible cells grown with or without dox. Modulation of parkin expression had no significant effect on levels of endogenous cyclin E protein relative to ␤-actin protein in this cell line (Fig. 6B). DISCUSSION Soon after the parkin gene was identified, Northern analysis revealed a broad distribution of parkin expression in the brain (22), and parkin regulation was described as being akin to a class of genes with ubiquitous expression known as "cellular housekeepers." Contrary to these original speculations, indepth examination revealed a neuron-specific and developmentally regulated pattern of expression in the brain (5,6). Given this background, and in light of the current heterogeneity of hypothesized parkin functions, further investigation into the regulation of parkin expression may divulge details of how loss of parkin function might cause PD and whether parkin might be involved in cancer. The parkin gene contains massive introns and spans a common fragile site (7), making it one of the largest genes in the genome. Yet despite a first intron Ͼ280 kbp in size, another expanded gene (PACRG) lies a mere 204 bp  4. In vivo N-myc and parkin expression. A, Northern analysis using probes specific for mouse N-myc and parkin, hybridized to a mouse multiple-stage embryo blot. The expected size of the transcripts was confirmed by a corresponding marker. B, quantitative RT-PCR measurement of parkin and N-myc in the human fetal or adult brain, normalized to ␤-actin mRNA in each case. One unit is defined as the amount of transcript present in the fetal brain. Human fetal brain RNA used in this experiment was purchased from BD Biosciences Clontech and derived from a 30-week-old fetus. Error bars represent 2ϫ S.E., and measurements were conducted in triplicate. C, Western analysis of lysates derived from various human neuroblastoma cell lines. Blots were stripped and re-probed with the indicated antibody.
upstream of parkin, with domains of strong transcription activation in between (12).
The first domain of transcription regulation upstream of the parkin transcription initiation site within the shared parkin/ PACRG promoter contains sequence similar to an E-box motif (CGCGTG) that is electronically predicted to interact with myclike proteins and is evolutionarily conserved. Harris et al. demonstrated that N-myc and Max protein can directly bind to the CGCGTG sequence motif found in the pax-3 promoter in vitro using gel-shift assays (14). In this study, we showed that N-myc can bind to the endogenous chromatin/nucleoprotein complex at this site in the parkin promoter, thereby implying a cisacting role for N-myc-mediated parkin transcription. Whereas N-myc activated transcription in the pax-3 promoter in reporter assays (14), our data suggest that N-myc represses transcription in the parkin promoter. It is unclear at this time how N-myc might be repressing transcription; one possibility is that N-myc is displacing an activating transcription factor through competition for a shared binding site. The identification of activating elements within the parkin promoter should allow testing of this hypothesis.
Two instances in nature are known to highly modify N-myc expression and were employed in this study to demonstrate a physiologically significant interaction between N-myc and parkin expression. The first is the process of embryogenesis and cell differentiation, in which N-myc is abundantly expressed in rapidly dividing cells and down-regulated in postmitotic cells (11). Conversely, we show that parkin is highly expressed in postmitotic cells and down-regulated in rapidly dividing cells, a feature perhaps critical to gene function. Parkin expression in fruit flies mimics that of the human and mouse gene orthologs, where detectable levels of parkin mRNA were restricted to the latest stages of embryonic development and subsequently expressed at high levels throughout adulthood (23). These results demonstrate an extraordinarily conserved mechanism across divergent species, suggesting that the regulation of parkin transcription is indeed a critical component of gene function. The second model system modifying N-myc expression is in human neuroblastoma, where a proportion of tumors and cell lines demonstrate amplification of N-myc accompanied by a proportionate increase in N-myc expression. In the neuroblastoma cell lines we analyzed, parkin inversely correlated with N-myc expression. This finding was consistent with expression of N-myc and parkin in a well described N-myc-inducible neuroblastoma cell line. Thus, N-myc would seem to robustly down-regulate parkin expression and may account for the developmental regulation of parkin expression.
The targets of the N-myc transcription factor have been of interest because the discovery of N-myc amplification in human cancer. Those targets up-regulated are probably involved in the initiation and continuation of cell growth, and those down-regulated might be involved in differentiation. Parkin is known to possess E3-ligase activity, where target proteins are polyubiquitinated and degraded via the proteasome. Numerous tumor suppressor genes with E3-ligase activity and protein domains homologous to parkin protein have also been described, including MDM2 (24) and BRCA1 (25). It is intriguing that the Parc protein (Parkin-like ubiquitin ligase) might also function as a tumor suppressor gene by regulating p53 activity (26). Given parkin's relationship with N-myc and the implication of parkin as a tumor suppressor, we hypothesized that parkin might be a key component in mediating the transition from cell growth to cessation and differentiation. We performed basic experiments to determine whether parkin might affect cellular proliferation rates or levels of cyclin E but found no such evidence. However, additional studies that more directly measure parkin's effect on tumorigenesis now seem warranted. Another gene down-regulated by N-myc with a similar expression profile to parkin, NDRG1 (N-myc down-regulated gene 1), is involved in growth arrest and cell differentiation and is mutated in hereditary motor and sensory neuropathy-Lom (27). The involvement of autosomal-recessive loss of function mutations in NDRG1 in human peripheral neuropathy may be all the more relevant to parkin-linked disease, in that loss-offunction parkin mutations have also been identified in human patients with peripheral and sensory neuropathy (28,29). It is conceivable that the mechanisms of neurodegeneration resulting from mutations in the parkin or NDRG1 gene overlap, at least in the peripheral nervous system.
Given N-myc's profound effect on the cell cycle, microarray studies and SAGE analysis report a strikingly small number of genes affected by N-myc expression: 114 genes up-regulated, most of which deal with ribosome biogenesis and protein synthesis, and 70 down-regulated genes (30). It is therefore of interest that several of the genes currently associated with PD are affected by N-myc expression; ␣-synuclein and UCH-L1 were previously shown to be up-regulated by N-myc in the inducible N-myc cell line (30). We have duplicated these original findings in this cell line using quantitative PCR (data not shown), in addition to demonstrating that parkin is downregulated by N-myc. N-myc also seems to regulate select components of the UPS, including proteasome subunits and heat shock proteins, thereby exerting downstream control of protein degradation and further implicating the importance of the UPS on cell cycle control. Thus, a common basis of transcription regulation among the genes associated with PD may be suggesting a biological overlap in endogenous gene functions.
The parkin gene (park2) is located within the common fragile site Fra6E in a region of the genome thought to contain a tumor suppressor gene (7). Multiple loci for PD are now known to be nearby or overlap regions associated with multiple tumor suppressors on the end of chromosome arm 1p, including park6 (1p36-p35), park7 (1p36), park9 (1p36), and park10 (1p32). Perhaps the family-based genetic studies in PD and cancer cytogenetics are suggesting a connection between the causes of the two diseases, which might be further supported by functional studies of the associated genes. Indeed, DJ-1 (park7), another gene associated with early-onset recessive PD (31), was originally isolated as a c-Myc interacting protein and can transform cells in coordination with the oncogenes c-myc and/or c-ras (32) in addition to functioning as a circulating tumor antigen in human breast cancer (33). Likewise, UCH-L1 (Park5) demonstrates abnormal expression in cancer that correlates with tumor severity (34) and may contribute to the degradation of tumor suppressor Kip1 (35). Last of all, overexpression of ␣-synuclein (Park1) results in increased cellular proliferation rates and an increase in the number of cells in the S phase (36). Thus, a provocative link between PD, the UPS, and cell cycle control is demonstrated on both a genetic and functional level.
Breach of cell cycle checkpoints is probably one of the primary mechanisms by which postmitotic neurons undergo death (37). Although this mechanism of cell death may have obvious implications in the developing nervous system and cancer, the link to a progressive neurodegenerative disorder is less clear. Because the UPS is a central component of cell cycle control (38), neurons that have an attenuated UPS because of genetic abnormalities, inherent cellular properties, and/or environmental insults might be unable to maintain control over the cell cycle and ultimately succumb to death from inappropriate re-entry. Functional studies that bridge together the genes associated with PD into a common pathway may reconcile disease pathogenesis and provide rational approaches to therapeutic intervention.