| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
-synuclein
Division of Basic Neurosciences, Foundation for Biomedical Research of the Academy of Athens (IIBEAA), Athens, Greece
1Correspondence: Division of Basic Neurosciences, Foundation for Biomedical Research of the Academy of Athens (IIBEAA), 11527 Athens, Greece. E-mail: lstefanis{at}bioacademy.gr
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Synuclein is an abundant neuronal protein that has been linked to both normal synaptic function and neurodegeneration—in particular, Parkinson’s disease (PD). Uncovering mechanisms that control -synuclein transcription is therefore critical for PD pathogenesis and synaptic function. We previously reported that in PC12 cells and primary neurons, -synuclein is transcriptionally up-regulated after application of growth factors. In the current work we have characterized the pathway involved in this regulation in PC12 cells. The MAP/ERK pathway, and in particular Ras, is both sufficient and necessary for the NGF and basic fibroblast growth factor (bFGF) -mediated response. Significantly, response elements for this pathway, including a putative occult promoter, lie within intron 1, a hitherto unappreciated regulatory region of the gene that may be utilized in this or other settings. The PI3 kinase pathway is also involved in -synuclein regulation, but response elements for this pathway appear to lie primarily outside of intron 1. These findings indicate that NGF- and bFGF-mediated signal transduction via the MAP/ERK and PI3 kinase pathways, and in part via regulatory regions within intron 1, may be involved in -synuclein transcriptional regulation. Targeting of these pathways may serve to modulate -synuclein so that it achieves desirable levels within neuronal cells.—Clough, R. L., Stefanis, L. A novel pathway for transcriptional regulation of -synuclein.
Key Words: Parkinson’s disease • PC12 cells • ERK • PI3 kinase • NGF
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-SYNUCLEIN (-SYN) IS AN ABUNDANT neuronal protein that has been linked both to normal synaptic function and to neurodegeneration. Most significantly, point mutations in the gene encoding -synuclein are responsible for Parkinson’s disease (PD) in rare familial cases with autosomal dominant inheritance, and the aggregated protein is a major component of Lewy bodies found in sporadic PD (reviewed in ref. 1
). Accumulating evidence implicates expression levels of -SYN in PD pathogenesis. Duplications and triplications of the -SYN locus have been reported in familial PD and various polymorphic alleles within and around the -SYN gene have been reported to correlate with the disease in sporadic cases (2
3
4
5
6
7)
. In some animal models, overexpression of wild-type (WT) -SYN leads to neuronal death or dysfunction and abnormal -SYN deposition (8
9
10)
. These studies indicate that overexpression of WT -SYN is sufficient to cause PD, and that its transcriptional regulation may have a profound effect on the development of sporadic cases of PD. Therefore, deciphering mechanisms governing -SYN transcriptional control is critical for elucidating PD pathogenesis.
A-SYN is a protein that is expressed abundantly in the nervous system. Its expression is developmentally regulated. In the rodent central nervous system (CNS), mRNA expression levels begin to rise at late embryonic stages and reach a peak in the first weeks of postnatal life, with mRNA levels decreasing during adulthood (11
, 12)
. The mechanisms involved in this developmental transcriptional regulation of -SYN are unknown. It could be that the same transcriptional control mechanisms may also play a role in the maintenance of -SYN levels in adulthood. Furthermore, various stimuli have been shown to regulate -SYN levels in adult rodents in vivo (12
13
14)
. Again, the mechanisms involved are unclear but may include an adaptive-plasticity response that recapitulates features of developmental regulation of -SYN. Aside from this developmental transcriptional regulation of -SYN, there is evidence for an increase in the stability of the -SYN protein associated with aging at the post-translational level (15)
.
Rat pheochromocytoma PC12 cells have been used as a model system to study the mechanisms of differentiation of proliferating neuronal precursors to postmitotic, differentiated neuron-like cells, which possess many characteristics of sympathetic neurons. Proliferating "naive" PC12 cells differentiate into "primed" neuronal-like PC12 cells after application of physiological concentrations of nerve growth factor (NGF) (16)
. We previously showed that -SYN is induced at the mRNA and protein level in PC12 cells after prolonged treatment with NGF, thus recapitulating in vivo developmental induction of -SYN during neuronal differentiation. Another growth factor, bFGF, which also causes PC12 cells to differentiate, induced -SYN expression (17)
. This growth factor-dependent regulation of -SYN was not limited to this cell system, as expression levels of -SYN were also regulated by NGF in primary postnatal rat sympathetic neurons (17)
. Furthermore, in embryonic cultures of rat ventral midbrain, bFGF accelerated the in vitro induction of -SYN levels in a manner that was sensitive to transcriptional inhibition. GABAergic neurons in the same cultures or cultured cortical neurons did not show this response to bFGF, suggesting that the mechanisms involved may be cell type-specific and may pertain only to catecholaminergic neurons (18)
.
Given the importance of the expression levels of -SYN as a risk factor for developing PD, we undertook the present study to further investigate the mechanisms involved in the transcriptional regulation of -SYN in the PC12 cell system. In these studies, we have deciphered the signaling pathways involved in regulating the -SYN in response to NGF- and bFGF-induced differentiation, and have uncovered a novel regulatory region for -SYN transcription located within the first intronic sequence, reported here for the first time.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 19)
for rat tail collagen-coated tissue culture dishes in RPMI 1640 medium (Sigma, St. Louis, MO, USA) supplemented with 5% heat-inactivated FBS, 10% heat-inactivated horse serum, 200 µM L-glutamine, and 1% pen/strep (complete medium). PC12 were differentiated in complete medium containing either 100 ng/ml recombinant human NGF (a kind gift from Genentech, San Francisco, CA, USA) or 50 ng/ml bFGF (Sigma). For experiments using the various kinase pathway inhibitors, PC12 cells were preincubated with the inhibitors for 45 min prior to the addition of the growth factors. U0126 (Sigma) was added to the complete media at a range of 1–50 µM. The following inhibitors were also used: LY294002 at 25 µM, SP600125 at 10 µM, and SB203580 at 30 µM (all from Sigma).
Cloning/subcloning
A range of different lengths of the 5' promoter sequence of -SYN cloned into the pGL3-basic luciferase vector (LUC; Promega, Madison, WI, USA) were generously provided by Drs. Nussbaum and Chiba-Falek, National Institutes of Health (NIH), Bethesda, MD, USA. The human intron 1 sequence of -synuclein was cloned from the 1.46 kb luciferase construct (5)
into the TOPO TA vector (Invitrogen, Carlsbad, CA, USA) (human-intron1-F-gtaagtacctgtggatctaaacgg, human-intron1-R-cctttacaccacactggaaaac). This fragment was subsequently cloned into the KpnI-XhoI restriction endonuclease sites of pGL3-basic (Promega). The 744 bp and 502 bp fragments of human intron 1 were also subcloned from the 1.46 kb sequence into TOPO-TA (744 bp-F, ccgctctcttggtttttctc, 502 bp-F, tagaaccatcacacttgggc and with the same reverse primer used for the full-length human intrinsic sequence). Both were subcloned into the KpnI-XhoI sites of pGL3-basic. The 308 bp fragment was cloned from the 1.46 kb sequence using a partial SmaI-XhoI digest and cloned into the corresponding sites of pGL3-basic. The rat -SYN intron 1 sequence was cloned with polymerase chain reaction (PCR) from genomic DNA extracted from the PC12 cell line (rat intron1-F, gtaagtacctgtagatgcagtagctc, rat intron1-R, tgaacacatccatggctaaag) into the TOPO TA vector (Invitrogen), then subcloned into the KpnI-XhoI sites of pGL3-basic. The 1 kb-738 subclone was generated using a PCR strategy from the 1.46 kb clone using the hum-intron1-F primer and the primer 738Rev: gcgggcagacagattttatg. The PCR product was cloned into TOPO-TA. The fragment was liberated using the KpnI-XhoI sites and cloned into pGL3 promoter (Promega). The 502–308 bp construct was generated by digesting the 502 bp clone in TOPO-TA with KpnI-SmaI and cloning it into the corresponding sites of pGL3 promoter.
Transfections and luciferase assays
All transfections were performed using lipofectamine 2000 (Invitrogen) according to the manufacturer’s instructions. Briefly, for the luciferase assay all transfections were performed in 6-well tissue culture dishes with 6 x 105 cells with 2 µg per well of target vector and 0.5 µg of TK-renilla (internal control). A 1:3 ratio of transfected DNA to lipofectamine was determined to give the best transfection efficiency. All transfections were performed overnight and complete medium was added the next day. Samples for Western immunoblotting were harvested after 48 h.
In all luciferase assays, except where indicated, cells were harvested on day 2.5 after addition of the complete media to the cells post-transfection. Luciferase assays were detected with the Dual Luciferase Assay (Promega) according the manufacturer’s instructions. Graphs show mean value of at least three experiments, with error bars depicting the SEM. All values were calculated against the screened vector in untreated PC12 cells.
RT-polymerase chain reaction (RT-PCR)
Total RNA was extracted from PC12 cells on day 3 after the start of treatment using Trizol (Invitrogen) and cDNA was generated with the Reverse Transcription System (Promega) according to the manufacturer’s instructions. We then performed RT-PCR using the cDNA as template. In preliminary experiments, all primer pairs were optimized to be in the log-exponential phase of amplification. The following primers were used to perform PCR: 1) -SYN-F, ttctgcggaagcctagagag, -SYN-R, tcctccaacatttgtcacttgc* (product size=253 bp); 2) ß-actin-F, tcaccatggatgatgatatcgcc,* ß-actin-R, ccacacgcagctcattgtagaagg* (product size=282 bp; primers marked with an asterisk were from ref. 20
). Products were subsequently resolved on agarose gels.
Western immunoblotting
PC12 cells were washed twice in PBS, then harvested in lysis buffer (150 mM NaCl, 50 mM Tris pH7.6, 0.1% SDS, 1% Nonidet P-40, 2 mM EDTA). Protein concentrations were determined using the Bradford method (Bio-Rad, Hercules, CA, USA). Fifty micrograms of lysates were mixed with 2x Laemmli buffer prior to running on 12% SDS-polyacrylamide gels. After transfer to a nitrocellulose membrane, blots was probed with antibodies directed against 1) -synuclein (1:1000; BD Biosciences, Bedford, MA, USA), 2) Erk (loading control; 1:5000; Santa Cruz Biotechnology, Santa Cruz, CA, USA), 3) phosphorylated Erk (1:2000; Ph-Erk; Santa Cruz Biotechnology), 4) Ras (1:2000; BD Biosciences). Blots were probed with HRP-conjugated secondary antibodies and visualized with Western Lightning® (Perkin Elmer, Norwalk, CT, USA) and exposed to Super RX film (Fuji Film). After scanning the images with Adobe Photoshop, Image J software (21)
was used to quantify the intensity of the bands.
Statistics
Statistical analysis was performed using the Student’s t test for single analyses. Where multiple testing was required, a one-way ANOVA test was utilized with a post hoc Tukey’s HSD test. P values of < 0.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-synuclein mRNA synthesis after NGF treatment in PC12 cells-SYN is induced at the transcriptional and translation levels after treatment with NGF or bFGF (17)
. We wanted to see whether we could observe a similar phenomenon in the PC12 cell clones currently available in our laboratory. We did indeed observe increases in the level of -SYN after 1 day of NGF treatment at both the mRNA and protein levels. By 3 days of treatment, high levels of expression were seen (Fig. 1
).
|
Identification of intron 1 of -synuclein as a region that confers transcriptional responsiveness to NGF or bFGF
Given that increased mRNA synthesis appeared to be involved in the induction of -SYN in this model, there should be identifiable DNA elements within the -synuclein gene responsible for these effects. We therefore screened a series of different constructs comprising up to 10.7 kb of the 5' promoter sequence of human -SYN (5)
cloned into the pGL3-basic luciferase vector in an attempt to identify elements in this sequence that are involved in the activation of -SYN by both NGF and bFGF. The longest of these clones contained the NACP-Rep1 element (22)
, which has been linked to expression levels of -SYN and to the risk of developing sporadic PD (4
, 23
24
25)
, although evidence for this remains inconclusive (26
27
28
29
30)
. For NGF (100 ng/ml) and bFGF (50 ng/ml), we observed a significant up-regulation in the luciferase assay only for the 1.46 kb clone (Fig. 2
A). Intriguingly, this construct is located 3' of the canonical transcriptional start site (see Fig. 2B
), and contained 283 bp of exon 1 and the entire intron 1 of -SYN, terminating 10 bp 5' of the ATG translational start site in exon 2.
|
We next subcloned a sequence for both human and rat intron 1 (with the same 3' end site) into LUC and assessed activation in response to NGF or bFGF in order to examine whether the putative response elements to this treatment existed within this region and to verify that induction also occurred with the rat sequence. We observed even higher levels of activation from the human intron 1 sequence when compared with the 1.46 kb construct for both treatments (Fig. 2C
). We also observed significantly strong levels of activation from the corresponding rat -SYN intron 1 sequence, although at lower levels than the human sequence (Fig. 2C
).
The luciferase assay was also used to assess whether NGF-induced activation of the intronic sequence followed a pattern similar to that observed for the mRNA and protein levels, as shown in Fig. 1
. We found there was already an induction by day 1. This was followed by another large increase on days 2 and 3 after addition of the growth factor (Fig. 2D
). bFGF treatment also led to a gradual increase in luciferase activity over the 3 days of sampling (Fig. 2D
). In conjunction with other data, these findings identify the intron 1 sequence of human and rat -SYN as responsive to NGF and bFGF transcriptional activation.
NGF- or bFGF-induced activation of -synuclein can be repressed with a pharmacological inhibitor of the MEK/ERK kinase pathway
It has been demonstrated before that in the PC12 cell line, signaling of NGF and bFGF is dependent in part on the MEK/ERK kinase pathway (31
32
33)
. To assess involvement of this signaling pathway in the induction of intron 1-driven transcription, we applied the selective inhibitor of this pathway, U0126, concurrently with NGF or bFGF treatment. We confirmed that this inhibitor inhibited NGF-induced ERK phosphorylation selectively without affecting Akt phosphorylation (data not shown). For NGF, there was a dose-dependent inhibition of human intron 1-driven transcription, with a plateau effect for inhibition after 10 µM of U0126; 25 µM of U0126 efficiently suppressed rat intron 1-driven transcription. Similar levels of repression were also observed for this pharmacological inhibitor for bFGF-induced activation for both the human and rat -SYN intron 1 sequences (Fig. 3
A).
|
To ascertain whether this inhibition was reflected at the transcriptional level of endogenous -SYN, we assessed mRNA by semiquantitative RT-PCR. U0126 repressed the induction of -SYN mRNA mediated by either NGF or bFGF (Fig. 3B
) in the PC12 cell line. Furthermore, U0126 dose-dependently inhibited the NGF-mediated induction of -SYN at the protein level, as assessed by Western immunoblotting (Fig. 3C, D
). The dose dependency of this phenomenon mirrored the effects observed on the intron 1 sequence. U0126 also inhibited the bFGF-induced induction of -SYN protein (Fig. 3E
). We conclude that in this system -synuclein transcription is regulated by the ERK/MAPK pathway.
NGF- and bFGF-mediated induction of -synuclein is repressed by LY294002 (LY2), a pharmacological inhibitor of the PI3 kinase pathway
We then assessed involvement of the PI3 kinase (PI3K) signaling pathway in the phenomenon of -SYN induction in this system, as it has been shown that many signaling cascades can act in parallel in propagating signals from the cell surface to the nucleus (33
34
35
36)
. We used LY294002 (LY2, 25 µM), a pharmacological inhibitor of the PI3K pathway. NGF-induced activation of the human or rat intron 1 sequence was reduced by 
2-fold when cells were treated with LY2 (Fig. 4
A, left panel). However, no inhibition was seen for bFGF-induced luciferase activity (Fig. 4A
, right panel), suggesting that the NGF and bFGF signaling pathways are propagated in part through different signaling pathways.
|
We next assessed the effects of this inhibitor on NGF- and bFGF-induced activation of -SYN mRNA levels, as detected by RT-PCR. This induction was clearly inhibited in both cases (Fig. 4B
; left panel for NGF, right panel for bFGF). Similar results were obtained at the protein level (Fig. 4C
, left panel for NGF, right panel for bFGF).
We conclude that the PI3 kinase pathway is also involved in transcriptional regulation of -SYN in this system. The dissociation observed between the luciferase assay and the mRNA and protein levels of -SYN, which is especially obvious for bFGF, is indicative of other response elements located outside of intron 1.
c-Jun NH2-terminal kinase (JNK) and p38 are not implicated in the NGF and bFGF signaling cascade that leads to -synuclein induction
We then turned our attention to two additional signaling cascades of JNK and p38-MAPK that are also known to mediate effects of NGF in PC12 cells (34
, 36
, 37)
. These pathways were inhibited using the pharmacological inhibitors SP600125 (SP6) and SB203580 (SB2), respectively.
In the luciferase assay we observed a reduction in activation for the human -SYN intron 1 sequence upon treatment with SB2 (30 µM) or SP6 (10 µm)compared with cells treated with NGF alone. This effect, however, was not observed with the rat intron 1 sequence (Fig. 5
A, left panel), and activation from this intronic sequence for bFGF, for both the human and rat sequences, was not inhibited in the presence of the pharmacological compounds (Fig. 5A
, right panel). We observed no reduction in NGF- and bFGF-induced activation of -SYN mRNA or protein levels in the presence of the inhibitors SB2 or SP6 (Fig. 5B, C
). We conclude that the signaling pathways mediated by JNK and p38-MAPK do not play a major role in transcriptional regulation of -SYN in this system.
|
The Ras pathway is involved in the transcriptional induction of -synuclein in PC12 cells
The data described up to this point regarding the signaling pathways involved are based on pharmacological approaches. To gain insight into the molecular mechanisms involved, we have overexpressed H-Ras and assessed induction of the intron 1-related sequence and endogenous -synuclein levels. We elected to focus on the Ras pathway because it is a major regulator of MAP-ERK signaling after NGF treatment in PC12 cells (31
32
33)
. We observed that cotransfection of the H-Ras expression vector with the human or rat -SYN intron 1 sequence was sufficient to drive expression from these constructs (in the absence of either NGF or bFGF) (Fig. 6
A, left panel). We also observed that the Ras-induced activation from human intron 1 could be repressed in the presence of the pharmacological inhibitor of MEK1/2, U0126 (Fig. 6A
, left panel). In contrast, overexpression of Ras dominant negative (Ras DN), which lacked Ras activity, did not induce intron 1-related activity (Fig. 6A
, right panel). Furthermore, overexpression of H-Ras, but not Ras DN, induced ERK phosphorylation and was sufficient to induce -SYN at the protein level (Fig. 6B
). We then examined whether the Ras pathway was involved in the NGF- or bFGF-mediated induction of intron 1 activity. We found that this was indeed the case, as overexpression of Ras DN inhibited the NGF- or bFGF-mediated induction of luciferase activity for the intron 1 sequences (Fig. 6C
). We conclude that the Ras pathway is sufficient and necessary for the induction of -SYN observed in this system.
|
The PI3 kinase/Akt pathway is partially involved in intron 1 responses
Our results with the pharmacological inhibitor of the PI3 kinase pathway, LY2, suggested that this pathway only partially impinged on intron 1 regulatory elements. To investigate this further, we overexpressed various forms of Akt, which is the major regulatory kinase in this pathway, together with the human -SYN intron 1 sequence, and assessed luciferase induction in the absence of growth factor treatment. We found that overexpression of Akt or of constitutively active Akt alone had no effect on human intron 1-mediated transcriptional activation (Fig. 7
, left panel). However, a DN form of Akt partially suppressed the NGF-mediated induction of intron 1 transcriptional activity (Fig. 7
, middle panel) but had no effect on bFGF-mediated responses (Fig. 7
, right panel).
|
We conclude that Akt activation is partially involved in intron 1-mediated transcriptional induction after NGF treatment, but that it is not sufficient for this induction. It is not involved in the bFGF-mediated effects on intron 1. These results are consistent with those obtained with the pharmacological inhibitor LY2, and indicate that the primary site of action of PI3 kinase signaling lies elsewhere within the -SYN gene, outside of intron 1.
Subcloning of the human -synuclein intronic region identifies areas that are responsive to NGF and bFGF treatment
In an attempt to further identify regions that may harbor response elements that mediate -SYN transcriptional activity, we subcloned a range of clones, of different lengths, of the human -SYN intron 1 and examined these clones for activation by NGF or bFGF. A diagram depicting these clones is shown in Fig. 2B
. For NGF treatment, a large drop of activation was observed between the full intronic sequence and the 744 bp clone, which lacked the 319 bp of the 5' sequence. A second drop of activation was observed between the 502 bp clone and the 308 bp clone (Fig. 8
A, left panel). This is indicative of at least two NGF response elements in these two areas of intron 1. For bFGF treatment (hatched bars), activation is lost for all subclones of this intronic sequence, with levels of activation falling to background (Fig. 8A
, right panel). This is indicative of a response element to bFGF activation within the first 319 bp of the 5' sequence of intron 1. To further define the molecular pathways to these response elements, we characterized the response element located in the 502 bp clone (Fig. 8B
). The NGF response of this element was shown to be inhibited by the pharmacological inhibitor U0126 and not by LY294002 (Fig. 8B
, left panel). We overexpressed H-Ras together with the 502 bp luciferase clone and observed no induction from the latter. Ras DN was unable to repress NGF-induced activation from the 502 bp clone (Fig. 8B
, right panel). These data indicate that only the 5' half of the intronic sequence contained response elements to Ras.
|
To verify that the two identified regions—the one containing the first and the other the third quadrant of the intron 1 sequence—indeed contained response elements, we next cloned these two regions into the pGL3 promoter vector (Promega), which contains an SV40 promoter upstream of the luciferase gene and allows for screening of putative enhancer elements (see Fig. 2B
). Despite high background levels of activation from the empty vector, in the presence of NGF we observed significantly higher levels of activation for both clones (Fig. 8C
, left panel). In the presence of bFGF, we observed an induction only for the 1 kb-738 clone and not for the 502–308 bp clone (Fig. 8C
, middle panel). Similarly, upon cotransfection with H-Ras, we observed high levels of activation from the 1 kb-738 bp clone but not the 502–308 bp clone (Fig. 8C
, right panel).
These data indicate there are at least two response elements with the human intron 1 sequence of -SYN. The first is located within the 5'-prime quadrant and is responsive to Ras, bFGF, and NGF; the second is located within the third quadrant and is responsive to NGF but not to Ras or bFGF.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-SYN transcriptional regulation. As the story of -SYN has unfolded, it has become increasingly obvious that its regulation may be critical for PD pathogenesis. Although transcriptional regulation of -SYN levels has been described in various in vivo and cell culture paradigms, the manner in which this regulation occurs, the signal transduction mechanisms, and the DNA response regions involved remain unclear. Here we present data indicating for the first time that -SYN is regulated at the transcriptional level via the MAP/ERK and PI3K pathways, and that the former may mediate its effects via a hitherto unappreciated regulatory region within the first intron of the -SYN gene.
Our findings pertain to the model of neuronal differentiation of PC12 cells after the application of growth factors and may be relevant to the neuronal differentiation that normally occurs during development. We previously reported that -synuclein levels are also regulated by NGF and bFGF in cultured sympathetic neurons and ventral midbrain dopaminergic neurons at defined developmental periods (17
, 18)
. -SYN is normally up-regulated at the transcriptional level during the late embryonic-early postnatal period in rodents (11
, 12)
. -SYN transcription also increases within the postnatal rodent substantia nigra pars compacta after various injurious stimuli, such as excitotoxic injury to the striatum or exposure to MPTP or paraquat (12
13
14)
, and is also reported to be up-regulated in cocaine abusers within the same region (38)
. Furthermore, -SYN mRNA was significantly down-regulated during a critical period of song model acquisition in nuclei controlling this function in zebra finch (39)
. In all these cases it appears that a common denominator may be neuronal plasticity and adaptive response rather than injury. Therefore, whether mediated by growth factors or not, the pathways we have uncovered may be relevant in in vivo settings, but this will require further study.
A model of the deciphered pathway is depicted in Fig. 9
. Based on results achieved with the specific inhibitor of Mek1/2/ERK, U0126 and the effects of Ras and Ras DN, we conclude that there are two MAP/ERK-dependent parallel pathways by which NGF mediates -SYN transcription. The first is dependent on a response element (NGF-RE1) present within the first quadrant of intron 1 and requires Ras; the second is dependent on a response element (NGF-RE2) present within the third quadrant and does not require Ras. The existence of two separate response elements was confirmed by the response of the 1 kb-738 bp and 502–308 bp subclones to NGF. In addition, transcriptional regulation via NGF is mediated by a third parallel pathway, which requires PI3K activity. This appears to depend only in part on intron 1 sequences, as the inhibition achieved in the luciferase assay with LY2, the pharmacological inhibitor of this pathway, was only partial, whereas at the mRNA and protein level, -synuclein expression reverted to background. We found that, as expected, LY2 did not inhibit NGF-induced ERK phosphorylation (data not shown), and so its effects are independent of this pathway. We suggest, therefore, that there may be additional sequences within the promoter sequence of -SYN, outside intron 1 that contain response elements to the PI3K pathway. These are likely to be the same response elements that are activated by bFGF, as the bFGF effects on intron 1-mediated transcription are not inhibited by LY2. Another difference in the bFGF pathway compared with the NGF pathway is that bFGF appears to induce only one arm of the ERK-related pathway: the one dependent on Ras and the first quadrant of intron 1 (Fig. 9)
.
|
Gomez-Santos et al. (40)
have reported that -SYN is up-regulated in neuroblastoma cells after dopamine administration and that the pathway involves JNK and p38. We did not find involvement of these pathways in our system, but it is worth noting that human, as opposed to rat, intron 1-mediated transcription did show partial dependence on these pathways. Therefore, it is possible that in human neuroblastoma cells, elements within intron 1 may mediate the observed effects of JNK and p38.
The discovery of regulatory regions and transcriptional capability of the intron 1 region was unexpected given that the accepted transcriptional start site lies further upstream (41)
. The observation that we get a significant induction of luciferase activity with this sequence cloned into the pGL3-basic vector, which does not contain a promoter, indicates that within intron 1 there lies a hitherto unappreciated putative promoter region. Whether this site is used in the context of the whole gene and whether it plays a role in -SYN induction in the current model are not known. Further studies are under way to examine the role of this potential promoter region in -SYN regulation in this and other models. As alternative transcriptional start sites are often associated with cell-specific expression and our previous findings indicate that growth factor regulation of -SYN levels may be specific to catecholaminergic neurons (17
, 18)
, it is possible that the potential occult promoter/transcription start site may mediate transcription specifically within catecholaminergic neurons. Alternatively, the response elements we have identified within intron 1 may mediate transcription by acting on the 5' consensus transcriptional start site. The existence of response elements within intronic sequences that act in this manner has been observed for other genes (42
, 43)
.
The sharp drop in growth factor-induced luciferase activity when sequences longer that 1.4 kb were cloned into the pGL3-basic vector indicates that, at least for the human sequence, there are repressor elements for this pathway within these larger DNA regions. It is likely that in the context of the whole gene, these repressor elements are inactive, thus permitting intron 1-mediated transcriptional regulation of -SYN. The excellent correlation between transcriptional and translation control and the intron 1 responses in the luciferase assay gives us a high degree of confidence that this region is indeed important for -SYN regulation in this cellular context. The only case where this correlation breaks down is in experiments with the PI3K inhibitor. Results with this pharmacological agent merely indicate the existence of additional response elements outside intron 1, as discussed above.
Our findings may be relevant to PD pathogenesis. We have found that the MAP/ERK pathway (in particular, Ras activation) is both sufficient and necessary for -SYN induction. Increasing data suggest that activation of the MAP/ERK pathway may play a detrimental role in certain settings, especially under conditions of oxidative stress (44
45
46)
. Our present data suggest that this pathway may have an additional role in intron 1-mediated transcriptional regulation of -SYN. If the regulatory pathway we have identified plays a role in the regulation of -SYN in the adult CNS, then modulation of the MAP/ERK or PI3K pathway may serve to control excessive induction of -SYN, which is increasingly associated with PD pathogenesis. We are now in the process of examining the specific response elements and transcription factors responsible for -SYN induction in this model. Identification of such elements could lead to an even more targeted approach to the modulation of -SYN levels.
Most of the attention regarding the promoter region of -synuclein has been focused on the NACP-Rep1 region (22)
. It appears that the length of tandem repeats in this region is associated with the risk of developing PD in some, but not all, populations (4
, 6
, 7
, 23
24
25
26
27
28
29
30)
. In other cases, unbiased approaches have failed to find significant association of PD with this region, and instead have found a correlation with the 3' and coding regions of the gene (7)
. If, indeed, intron 1 is important for human -SYN regulation in the adult human CNS, polymorphisms within this region may be related to the risk of developing PD.
|
ACKNOWLEDGMENTS |
|---|
-SYN; Dr. Sally Dawson (UCL Ear Institute, UK) for the TK-renilla construct; Dr. Dimitra Mangoura (Foundation for Biomedical Research of the Academy of Athens, Greece) for the Ras constructs; Dr. Thomas Franke (Columbia University, New York, NY, USA) for the Akt constructs; and Genentech for the gift of recombinant human NGF. This study was supported in part through an award from the Parkinson’s Disease Foundation (R.L.C. and L.S.) and a Marie Curie International Reintegration Grant (L.S.). Received for publication August 10, 2006. Accepted for publication September 14, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
|
