The Expression of CNS-Specific PPARGC1A Transcripts Is Regulated by Hypoxia and a Variable GT Repeat Polymorphism
PPARGC1A encodes a transcriptional co-activator also termed peroxisome proliferator-activated receptor (PPAR) gamma coactivator 1-alpha (PGC-1α) which orchestrates multiple transcriptional programs. We have recently identified CNS-specific transcripts that are initiated far upstream of the reference gene (RG) promoter. The regulation of these isoforms may be relevant, as experimental and genetic studies implicated the PPARGC1A locus in neurodegenerative diseases. We therefore studied cis- and trans-regulatory elements activating the CNS promoter in comparison to the RG promoter in human neuronal cell lines. A naturally occurring variable guanidine thymidine (GT) repeat polymorphism within a microsatellite region in the proximal CNS promoter increases promoter activity in neuronal cell lines. Both the RG and the CNS promoters are activated by ESRRA, and the PGC-1α isoforms co-activate ESRRA on their own promoters suggesting an autoregulatory feedback loop. The proximal CNS, but not the RG, promoter is induced by FOXA2 and co-activated by PGC-1α resulting in robust activation. Furthermore, the CNS, but not the RG, promoter is targeted by the canonical hypoxia response involving HIF1A. Importantly, the transactivation by HIF1A is modulated by the size of the GT polymorphism. Increased expression of CNS-specific transcripts in response to hypoxia was observed in an established rat model, while RG transcripts encoding the full-length reference protein were not increased. These results suggest a role of the CNS region of the PPARGC1A locus in ischemia and warrant further studies in humans as the activity of the CNS promoter as well as its induction by hypoxia is subject to inter-individual variability due to the GT polymorphism.
KeywordsPPARGC1A PGC-1α CNS-specific isoforms FOXA2 ESRRA Hypoxia Microsatellite
Peroxisome proliferator–activated receptor gamma co-activator 1A (PGC-1α) encoded by PPARGC1A, is a versatile transcriptional co-activator involved in multiple transcriptional programs [1, 2, 3]. A key issue to understand the diverse functions of PGC-1α is to gain insight into the mechanisms that confer specificity to its interactions with numerous transcription factors. The regulation of PGC-1α function occurs at several levels. Expression of PGC-1α and the transcription factors co-activated by it are critical for some pathways . Furthermore, PGC-1α is targeted by several signaling pathways at the posttranslational level, as site-specific phosphorylation [5, 6], acetylation , ubiquitination [8, 9], and methylation  have been shown to alter the stability and/or activity of PGC-1α. Alternative splicing and/or transcriptional initiation represent another level of control that results in gain or loss of domains interacting with signaling targets. The truncated NT-PGC-1α resulting from alternative splicing contains the N-terminal activation and nuclear interaction domains and displays functional differences in comparison to the full-length protein [11, 12]. In skeletal muscle, two novel promoters have been described 14 kbp upstream of the reference promoter that are differentially regulated in response to metabolic cues such as fasting and exercise [13, 14]. In the human liver, an alternative promoter in intron 2 of the reference gene gives rise to a 75-kDa protein that showed subtle differences in co-activation selectivity compared to the reference protein . We recently described several distinct CNS-specific PPARGC1A mRNAs that are transcribed from a novel promoter located ~ 586 kbp upstream of the reference promoter. These transcripts are more abundant than reference gene (RG) mRNAs in human post-mortem brain samples and are partially conserved in rodents .
PGC-1α activates transcriptional programs that are relevant for neurodegenerative diseases such as mitochondrial biogenesis and function, the defense against reactive oxygen species, and autophagy [17, 18, 19, 20, 21]. Indeed, several functional studies in animal or cell culture models suggest that reduced PGC-1α function plays a role in neurodegenerative diseases [22, 23, 24, 25, 26, 27]. In addition, PGC-1α has been implicated in multiple sclerosis [28, 29]. Genetic studies in humans have suggested associations of the PPARGC1A locus with Alzheimer’s, Huntington’s, and Parkinson’s disease and amyotrophic lateral sclerosis [16, 30, 31, 32, 33, 34, 35, 36]. As some of these associations involved the CNS-specific region of the PPARGC1A gene, it is reasonable to suspect that the CNS-specific isoforms and/or their regulation play a role in these disorders. While the signaling pathways that converge at the RG promoter have been identified in mouse muscle and liver [reviewed in 37] as well as in human tissues and cell lines [15, 38], virtually nothing is known about the regulation of the CNS-specific promoter with the exception of its activation by USF1 . We therefore have compared the regulation of the CNS-specific and RG PPARGC1A promoters in human neuronal cell lines and a rat model of ischemia. We identified not only similarities but also substantial differences in the transcriptional regulation of the two promoters.
Materials and Methods
Expression plasmids pPGC-1α , B4-PGC-1α, B5-PGC-1α, B4-7a-PGC-1α, B5-7a PGC-1α , and FOXA2  were described previously. The human ESRRA (NM_001282450) ORF Clone in pcDNA 3.1+/C-(K)-DYK was purchased from GenScript (Piscataway, NJ, USA). HA-HIF-1α-pcDNA3 and HA-HIF-2α-pcDNA3 were gifts from William Kaelin (Addgene plasmids #18949 and #18950 respectively ). Long and short promoters of both the RG  and CNS-specific PPARGC1A  have been described. Short CNS-specific promoters with increasing guanidine thymidine (GT) sizes (602 bp, − 539 to + 63 relative to the transcription start site in promoters containing 11 GT) were cloned into pGL4.11[luc2P] (Promega, Madison, WI, USA) from DNA of subjects that harbored identical CNS promoter sequences but differed in their GT repeat size. A very short promoter luciferase construct (− 84 to + 63 bp relative to the transcription start site) was produced by cloning amplification products of the proximal promoter into pGL4.11[luc2P] (Promega). Primers used are shown in Supplementary Material, Table S1.
Cell Culture and Transfection Experiments
SH-SY5Y and NTERA-2cl.D1 (NT2/D1) cells were obtained from ATCC and cultured in DME (D6046)/F12 1:1 (Sigma-Aldrich, St. Louis, MO, USA) and DME (D0547)/F12 (Sigma-Aldrich) supplemented with 0.21% NaHCO3, respectively. The murine hippocampal cell line HT22 was a kind gift from D. Schubert, Salk Institute, La Jolla, CA, USA and cultured in DME (D0547)/F12 1:1 supplemented with 0.21% NaHCO3. Media for all cells were supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Invitrogen, Carlsbad, CA, USA). All cells were cultured in a humidified 5% CO2 atmosphere at 37 °C. Hypoxia culture experiments were performed in an O2-regulatable incubator (Binder GmbH, Tuttlingen, Germany) or in a H135 HEPA hypoxystation (Don Whitley Scientific Limited, West Yorkshire, UK) in a humidified 5% CO2 and 1% O2 atmosphere. Ciclopirox olamine and deferoxamine were obtained from Sigma-Aldrich. Cells were plated in 24-well plates 1 day before transient transfection with plasmid constructs for 24–48 h using Lipofectamine 3000 (Invitrogen) or DNAfectinTM Plus (ABM Inc., Vancouver, Canada). The pRL-TK plasmid (Promega) was used as a transfection control. Luciferase activities were measured using the Dual Luciferase Reporter Assay System (Promega) as described . Results are representative of two or more experiments, each performed in quadruplicate, and are shown as means + SD.
RNA and Protein Extraction
Total RNA from SH-SY5Y, NT2/D1 cells, and rat brain regions was extracted using the QIAzol Lysis Reagent (Qiagen, Hilden, Germany) and the RNeasy Lipid Tissue Mini kit (Qiagen). Concentrations of RNA were measured using the Nanovue Plus spectrophotometer (GE Healthcare, Chicago, IL, USA). In addition, the QIAxcel Advanced Instrument (Qiagen) was used to ascertain RNA quality. RNA integrity scores were between 8 and 9 for all samples used. For total protein preparation of whole cultured cells, cell pellets were solubilized in RIPA buffer containing protease inhibitors (HaltTM Protease Inhibitor Cocktail, Thermo Fisher Scientific, Waltham, MA, USA). Clear supernatants were obtained after centrifugation (10,000×g for 20 min at 4 °C) of suspensions that were intermittently incubated on ice and heated at 95 °C and sonicated (Sonoplus HB 27, Bandelin, Berlin, Germany). Nuclear and cytoplasmic extracts were prepared using the NE-PER Nuclear and Cytoplasmic Extraction Reagents Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol. Protein concentrations were determined by the RCDC Protein Quantification Assay (Bio-Rad Laboratories, Inc. Hercules, CA, USA).
Western Blot Analysis
Immunoblotting was performed as described . In brief, equal amounts of NT2/D1 cell extracts (20 or 40 μg) were subjected to SDS electrophoresis in 7–10% polyacrylamide gels. Separated proteins were transferred to polyvinylidene fluoride membranes (Bio-Rad Laboratories, Inc.) that were blocked with Tris-buffered saline (TBS), pH 7.6, containing 10% Blotting Grade Blocker (Bio-Rad Laboratories, Inc.) for 2 h at room temperature. Membranes were incubated with the primary antibodies directed against HIF1A (HIF-1α, NB100-105, Novus Biologicals, Centennial, CO, USA), diluted 1:1000 in 0.1% Tween 20 in 1× TBS overnight at 4 °C. Membranes were then incubated for 1 h at room temperature with the respective secondary antibody (IRDye® 800CW Goat anti-Rabbit IgG(H + L) or IRDye® 800CW Goat anti-Mouse IgG(H + L) from LI-COR, Lincoln, NE, USA) and analyzed using the ODYSSEY infrared imager (LI-COR). β-Actin (#4967, Cell Signaling technology, Danvers, MA, USA) or Histone H3 (NB500-171, Novus Biologicals) antibodies were used as loading controls. For densitometric analysis of Western blots, Image Studio™ software (LI-COR) was used.
Real Time PCR
DNase I-treated total RNA from cell cultures (1 μg/reaction) was reverse transcribed with the QuantiTect Reverse Transcription kit (Qiagen), using a mix of random hexamers and oligo-dT primers. cDNAs were amplified in duplicate by real-time PCR using Maxima SYBR Green (Thermo Scientific) or GoTaq qPCR Master Mix (Promega) and primers targeting exons B1 and B4 or B5 and exon 2 to quantify the two main CNS-specific PPARGC1A transcripts. Primers targeting exon 1 and exon 2 were used to target RG transcripts. The transcripts encoding the class of NT-PGC-1α isoforms were quantified by targeting exon 5 and exon 6A (Supplementary Material, Table S1 for primer sequences). To directly compare the amounts of PPARGC1A transcripts, gene segments containing the sequences targeted by the respective transcript-specific assays were cloned and used for the construction of standard curves to normalize for the efficiency of assays. Primers to quantify VEGF transcripts are also shown in Table S1. Total RNA from rat brain regions was reverse transcribed as described above. As the B5 encoded isoform is not present in rats, we measured B1b4, exon 1, exon 2, exon 5, and exon 6A containing transcripts as well as transcripts encoding VEGF and GLUT1. The respective primers are shown in Supplementary Table S1. For transcript amplification, the Rotor-Gene TM Q (Qiagen) instrument or the LightCycler TM 480 Instrument (Roche, Basel, Switzerland) was used. Relative mRNA levels were calculated using the comparative threshold cycle method (ΔCr). Constitutively expressed RPLP0 (Ribosomal Protein, large, P0) mRNA was used for normalization of mRNA abundance.
The chromatin immunoprecipitation (ChIP)-IT Express Enzymatic kit (Active Motif Europe, La Hulpe, Belgium) was used. Cross-linking of 106 NT2/D1 cells, enzymatic shearing and preparation of input DNA were performed as described . After an overnight incubation with 4 μg of HIF1A-specific monoclonal antibodies at 4 °C, chromatin/DNA complexes were immunoprecipitated using protein G magnetic beads. After DNA purification, primers described in Supplementary Material, Table S1 were used to amplify fragments of the CNS-specific promoter. As positive control, we designed the human equivalent primers of those described  spanning an HRE in the mouse Ntrk2 (TrkB) promoter known to be activated by HIF1A. As a negative control, primers spanning Exon B5 of CNS-PPARGC1A that did not contain a predicted HRE were used. IgG instead of the HIF1A antibody was used to confirm the absence of non-specific bands.
The experiments were carried out in accordance with the protocol for animal care approved by European Communities Council Directive (2010/63/EU) with permission of The State Veterinary and Food Administration of the Slovak Republic (4451/14-221 and 4247/15-221) under the supervision of Ethical Council of the Institute of Neurobiology, Bratislava Slovak Academy of Sciences. Every effort was made to minimize animal suffering and reduce the number of animals used. Adult male albino Wistar rats weighing 300–500 g obtained from Velaz (Czech Republic) were housed in the certified vivarium of the Institute of Neurobiology, Biomedical Research Center of the Slovak Academy of Sciences, maintained on a 12-h light/dark cycle and given food and water ad libitum. Food was withdrawn 1 day before surgery.
Four groups of animals were used: two ischemic groups of 3 rats each with different post-ischemic reperfusion intervals (ischemia 5 min followed by 1 h or 3 h of reperfusion) and two sham control animal groups (n = 3 each) sacrificed at the time points of the experimental ischemic animals.
Model of ischemia
Transient forebrain global ischemia was induced  with modifications . The rats were anesthetized with 4% halothane in anesthetic cages and maintained with 1.5% halothane during surgery. On the first day, both vertebral arteries were irreversibly occluded by coagulation through the alar foramina. The next day, rats were re-anesthetized and the common carotid arteries were dissected free and clamped just before awakening by small atraumatic clips to induce forebrain ischemia. After 5 min of ischemia, blood flow was restored by releasing the clips and verified visually. Normothermic conditions (approximately 37 °C) were maintained by a feedback-controlled heating lamp and pad during all surgical procedures. Animals that became unresponsive within 30 s after clip tightening, lost the righting reflex during bilateral carotid artery occlusion, and showed no seizures during and after ischemia were used for the experiments. Sham-operated controls were surgically treated the same as the ischemic group, but the common carotid arteries were not clamped on the second day.
Tissue sample collection
Animals were sacrificed by decapitation under halothane anesthesia; the brains were quickly removed and maintained at 0 °C. The fronto-parietal and occipital cortex, striatum (dorsolateral part and rest of the striatum, respectively), hippocampus, thalamus, and cerebellum were separated. To compare the selectively vulnerable CA1 versus the relatively resistant dentate gyrus and CA3 regions, the hippocampus was divided into the CA1 region and the rest of the hippocampus (DG) under a dissecting microscope. Collected tissue was weighed, divided into halves, and frozen at − 20 °C in 5 volumes of RNAlater (Sigma-Aldrich) until RNA analysis.
In transient transfection studies, ANOVA was used for comparisons of means and the Tukey honest significance test for post-hoc comparisons. Two-way ANOVA was used to ascertain effects of transcriptional activators, co-activators, and their interactions. ANOVA was used to compare effects of reduced oxygen tension and CPX. Three-way ANOVA was used to analyze effects on transcript levels in the in vivo model of transient ischemia. Treatment (sham/transient ischemia), post-ischemic reperfusion time (1 h/3 h) and brain regions sampled (1–7) were used as independent variables. Reported p values are 2-tailed.
Robust expression of PGC-1α isoforms may be maintained by positive feedback loops. As demonstrated in mouse skeletal muscle, Mef2c binds and activates the reference promoter and co-activation of Mefc2 by PGC-1α further increases promoter activity resulting in sustained PGC-1α expression . To determine whether the CNS-specific promoter is subject to autoregulation in a human neuronal cell model, SH-SY5Y cells were transiently co-transfected with expression plasmids encoding full-length (FL) reference or CNS-specific proteins and reporter constructs harboring the sequences of the reference (2.6 kbp) or CNS-specific (2 kbp) promoters cloned upstream of the luciferase gene. All promoter constructs were comparably activated by FL reference and the FL-B4 and FL-B5-PGC-1α proteins, even though minor differences in the extent of co-activation were apparent (Fig. 1b). Thus, reference and CNS-specific promoters are co- and cross-regulated by their encoded proteins via co-activation of transcription factors not further identified in these experiments.
We next compared the role of ESRRA in the transcriptional regulation of the reference and CNS-specific promoters. SH-SY5Y cells were transiently co-transfected with reporter constructs harboring the reference promoter (2.6 kbp) and the CNS-specific promoter (2 kbp) along with plasmids expressing ESRRA and FL reference PGC-1α or FL-B4-PGC-1α. Both promoters were significantly induced by ESRRA, the reference FL-PGC-1α, or FL-B4-PGC-1α. A strong synergistic effect of ESRRA and PGC-1α at both promoters was observed (Fig. 1c).
FOXA2, a member of the forkhead family of winged-helix transcription factors, plays an essential role in the development and maintenance of mid-brain neuronal function [50, 51] and has been shown to influence Ppargc1a expression in mice . We therefore compared the transactivation of the reference and CNS promoter by FOXA2 in SH-SY5Y cells in transient co-transfection assays. The CNS-specific promoter, but not the reference promoter, was induced by FOXA2, and synergistic activation of the CNS-specific promoter by FOXA2 and FL-PGC-1α was noted (p < 0.001). As we identified putative FOXA2 binding sites in the distal (− 1175 bp) and proximal (− 218 bp) CNS promoter using the PROMO 3.0 transcription factor binding site tool/ALGGEN Research Software (http://alggen.lsi.upc.edu, [53, 54]), we repeated the experiment with a truncated promoter construct comprising − 539 to + 63 bp relative to the transcription start site. As the activation pattern was comparable between the full and truncated CNS-specific promoters, the putative FOXA2 binding site located at (− 218 bp) was likely the cis-functional site (Fig. 1d).
Alternative promoter usage and/or splicing occur in the great majority of mammalian genes and are key mechanisms in transcriptional regulation and generation of protein diversity [58, 59, 60]. Several tissue-selective promoters have been described for PPARGC1A, but the CNS-specific promoter is unique in that it is localized far upstream of the reference gene promoter and is responsible for the majority of PPARGC1A transcripts in human brain, especially in neuronal cells. To this end, we compared some relevant regulations of the CNS-specific and the RG promoter and report not only similar but also distinct factors controlling their activities.
Transcriptional activity of the CNS promoter was modulated by a variable GT tandem repeat region in that a larger GT size enhanced promoter activity. As such, a result was obtained in three neuronal cell types; it is likely to occur also in vivo. Hence, the GT size is a likely source of inter-individual variability in human CNS transcript expression. This conclusion is in keeping with a recent study that highlighted the importance of short tandem repeats in the genetic architecture of quantitative human traits. In a survey, more than 2000 tandem repeats were identified that influenced transcript expression. Such repeats were enriched in conserved regions, co-localized with regulatory elements and modulated certain histone marks . In addition, the instable GT microsatellite is part of a highly predicted Z-DNA region. Sequences transiently forming Z-DNAs are non-randomly distributed, occur more frequently proximal to transcription start sites, and influence promoter activity [46, 47, 48].
Self- and cross-co-activation of both the RG and CNS promoters by the reference protein and CNS-specific isoforms were consistent findings in our transient transfection studies of SH-SY5Y cells. Even though we did not further identify the transcription factors co-activated in the initial experiments, a synergistic co-activation of ESRRA on both the RG and CNS promoters was clearly demonstrated. Expression profiles of ESRRA and PPARGC1A transcripts are similar in various tissues and both genes are induced by cold-exposure [62, 63]. An autoregulatory feedback loop consisting of ESRRA response elements in the ESRRA and the reference PPARGC1A promoters has been discovered previously. The evolutionary conserved ESRRA binding site in the human RG promoter is located 1978-bp upstream of the transcription start site , while a 23-bp sequence that contains a functional ERR response element and is present in 1 to 4 copies has been mapped within − 700 bp of the transcription start site of the ESRRA promoter . In embryonic fibroblasts of Esrra null mice, the transcriptional activity of Pgc-1α was reduced in comparison to cells from wild-type mice but could be restored by the ectopic expression of Esrra. . We also identified two putative ESRRA binding sites in the CNS promoter, suggesting that a similar feedback loop augments the expression of CNS transcripts. This regulatory loop may be of clinical relevance, as loss of ESRRA function in brain regions of mice or loss-of-function mutations in humans have been associated with eating disorders [67, 68].
Unlike ESRRA, FOXA2 selectively activated the CNS-specific promoter in a positive feedback loop with FL-PGC1α. This result is supported by data showing that adenoviral overexpression of murine Foxa2 in Neuro-2a cells resulted in a dramatic increase in CNS-specific transcripts. In addition, murine Foxa1, another member of the forkhead family of winged-helix transcription factors, induced the transcriptional activity of the CNS-specific promoter in HEK293T cells . FOXA1 and FOXA2 share over 95% homology between mice and humans and near perfect sequence homology in their DNA binding and transactivation domains [69, 70]. The two transcription factors play important, but redundant roles in the development and maintenance of dopaminergic midbrain neurons [51, 71], arguing for the involvement of CNS-specific PGC-1α isoforms. Interestingly, genetic studies implicated the CNS-specific PPARGC1A region in the pathogenesis of Parkinson’s disease .
PGC-1α, specifically the truncated NT-isoform, has been shown to be upregulated in myotubes by hypoxia, but the canonical hypoxia response pathway characterized by HIF-1α stabilization was not involved [72, 73]. The increase in NT-PGC strongly induced Vegf expression by co-activation of Esrra via conserved binding sites in the Vegf promoter and in the first intron. The increased levels of NT-PGC-1α likely result from enhanced alternative splicing and perhaps increased transcription, but the exact mechanisms have not been elucidated. Our results are consistent with these findings as hypoxia, but not HIF1A stabilization, increased the levels of E5E6A transcripts. Furthermore, the E5E6A level was also increased in the transient ischemia rat model. As demonstrated in the current and previous studies [72, 73], HIF1A does not activate transcription from the reference promoter, but the autoregulatory loop with ESSRA may play a role, as treatment of cardiomyocytes with an ESRRA inverse agonist caused knockdown of Esrra protein expression and blocked the increase in Ppargc1a mRNA levels in response to hypoxia . Importantly, the activation of CNS-specific transcripts in hypoxia does occur via the canonical HIF-1α pathway. This conclusion is based on effects of both hypoxia and non-hypoxic HIF1A stabilizing stimuli on CNS-specific PPARGC1A transcripts, the activation of CNS-specific promoter reporter constructs by HIF1A expression plasmids, and the data from ChIP assays. In addition, the in vivo results in an established transient ischemia rat model also are consistent with this conclusion. Our data also showed that the average values of Vegf and CNS-specific, but not RG transcripts increased after ischemia in all brain regions sampled, except in the cerebellum, where the levels did not change (data not shown). Due to the sample number, the data on individual brain regions must be viewed as preliminary. Clearly, the effects of ischemia require further studies on the time course of response and effects in specific brain regions, as their vulnerability and response may differ. A link between hypoxia and CNS-specific PPARGC1A expression is supported by studies in the subterranean rodent Nannospalax galili. In this blind mole that survives at 3% O2 for up to 11 h as compared to 2–4 h for rats , CNS-specific transcripts are upregulated .
HIFs interact with HREs with the core sequence 5′-RCGTG-3′ , but only a small proportion of the potential binding sites are bound by HIFs across the genome . Several sequence preferences outside the core binding motif have been described but are not absolute. In addition, two or more HREs and/or binding sites for other transcription factors are often found near functional HIF sites and interactions of HIF1A with USF1 and FOXA2 (both located near HREs in the CNS-specific promoter) have been reported [77, 78]. Furthermore, various epigenetic marks are likely playing an important role in HIF binding [76, 79]. An interesting finding of our studies is the interaction of GT repeat size with HIF1A on the transcriptional activity of the CNS promoter. A similar interaction has been described at the NRAMP1 (SLCA11A1) promoter, as an ACGTG site next to a GT microsatellite with high Z-DNA potential affected transcriptional activity and susceptibility to inflammatory and infectious diseases . HIF2A also has been implicated in the hypoxic response, has been shown to trans-activate HRE containing target genes , and may therefore influence the expression of the CNS-specific promoter. Indeed, our preliminary studies suggest that HIF2A is expressed in nuclei of CPX-treated cultures and activates the CNS promoter in transient transfection studies (Soyal SM, Patsch W, unpublished results), but additional studies will be required to gain more insight into the regulation of the CNS-specific promoter by HIF2A.
Upregulation of CNS-specific PGC-1α could have major implications for the pathology of cerebral ischemia as well as neuronal physiology. Our results on the interaction of HIF1A and the GT repeat size would imply inter-individual differences in the response to hypoxia. CNS-specific transcripts encode for proteins that differ from the reference protein only at the N-terminus. The N-terminal activation and repression domains appeared to be responsible for gene programs and splicing events that are specific for some PGC-1α isoforms . Our preliminary data in SH-SY5Y in whom the CNS-specific or the RG promoters were selectively upregulated using the CRISPR-dCAS9 approach show overlapping as well as distinct targets for the CNS isoforms and the reference protein. Importantly, CNS-specific forms appear to participate in pathways supporting neural and axonal growth, while the reference protein was mainly related to mitochondrial biogenesis and function (Kwik M, Patsch W, Soyal SM, unpublished results). The generation of new mitochondria, specifically at the peripheral axonal tree, has been shown to be induced by reference PGC-1α and is essential for the axonal growth . Thus, the functions of the CNS isoforms and the reference protein may be intertwined by cross-talk between the CNS and reference proteins as demonstrated by the co-activation of each other’s promoters as shown in the current study.
In conclusion, the CNS-specific, but not the RG, promoter is activated by FOXA2 via an autoregulatory loop, suggesting a role of CNS-specific transcripts in the development and maintenance of dopaminergic mid-brain neurons. The activation of the CNS-specific promoter by HIF1A supports a role of CNS isoforms in ischemia. The hypoxic/ischemic activation may differ between individuals, as an interaction is likely to occur between a variable GT repeat and HIF1A. Finally, cross-talk between the CNS and RG promoters via transcriptional co- and cross-activation may support a balanced expression of the reference protein and the CNS-specific isoforms.
This study was supported from the Austrian Science Fund (FWF): Project V344-B24, Elise Richter Program to S.S., MJFOX foundation (RRIA 2012) to W.P., the Paracelsus Medical University Salzburg (Projects E-13/18/094_PAT) to W.P., and (PMU-FFA-14/01/011-SOY) to S.S.
Open access funding provided by Austrian Science Fund (FWF).
Compliance with Ethical Standards
The experiments were carried out in accordance with the protocol for animal care approved by European Communities Council Directive (2010/63/EU) with permission of The State Veterinary and Food Administration of the Slovak Republic (4451/14-221 and 4247/15-221) under the supervision of Ethical Council of the Institute of Neurobiology, Bratislava Slovak Academy of Sciences.
Conflict of Interest
The authors declare that they have no conflict of interest.
- 8.Sano M, Tokudome S, Shimizu N, Yoshikawa N, Ogawa C, Shirakawa K, Endo J, Katayama T et al (2007) Intramolecular control of protein stability, subnuclear compartmentalization, and coactivator function of peroxisome proliferator-activated receptor gamma coactivator 1alpha. J Biol Chem 282(35):25970–25980CrossRefPubMedGoogle Scholar
- 13.Miura S, Kai Y, Kamei Y, Ezaki O (2008) Isoform-specific increases in murine skeletal muscle peroxisome proliferator-activated receptor-gamma coactivator-1alpha (PGC-1alpha) mRNA in response to beta2-adrenergic receptor activation and exercise. Endocrinology 149(9):4527–4533CrossRefPubMedGoogle Scholar
- 15.Felder TK, Soyal SM, Oberkofler H, Hahne P, Auer S, Weiss R, Gadermaier G, Miller K et al (2011) Characterization of novel peroxisome proliferator-activated receptor gamma coactivator-1alpha (PGC-1alpha) isoform in human liver. J Biol Chem 286(50):42923–42936CrossRefPubMedPubMedCentralGoogle Scholar
- 23.Weydt P, Pineda VV, Torrence AE, Libby RT, Satterfield TF, Lazarowski ER, Gilbert ML, Morton GJ et al (2006) Thermoregulatory and metabolic defects in Huntington’s disease transgenic mice implicate PGC-1alpha in Huntington’s disease neurodegeneration. Cell Metab 4(5):349–362CrossRefPubMedGoogle Scholar
- 26.Katsouri L, Lim YM, Blondrath K, Eleftheriadou I, Lombardero L, Birch AM, Mirzaei N, Irvine EE et al (2016) PPARgamma-coactivator-1alpha gene transfer reduces neuronal loss and amyloid-beta generation by reducing beta-secretase in an Alzheimer’s disease model. Proc Natl Acad Sci U S A 113(43):12292–12297CrossRefPubMedPubMedCentralGoogle Scholar
- 27.Zhao W, Varghese M, Yemul S, Pan Y, Cheng A, Marano P, Hassan S, Vempati P et al (2011) Peroxisome proliferator activator receptor gamma coactivator-1alpha (PGC-1alpha) improves motor performance and survival in a mouse model of amyotrophic lateral sclerosis. Mol Neurodegener 6(1):51CrossRefPubMedPubMedCentralGoogle Scholar
- 28.Nijland PG, Witte ME, van het Hof B, van der Pol S, Bauer J, Lassmann H, van der Valk P, de Vries HE et al (2014) Astroglial PGC-1alpha increases mitochondrial antioxidant capacity and suppresses inflammation: implications for multiple sclerosis. Acta Neuropathol Commun 2:170CrossRefPubMedPubMedCentralGoogle Scholar
- 29.Witte ME, Nijland PG, Drexhage JAR, Gerritsen W, Geerts D, van het Hof B, Reijerkerk A, de Vries HE et al (2013) Reduced expression of PGC-1α partly underlies mitochondrial changes and correlates with neuronal loss in multiple sclerosis cortex. Acta Neuropathol 125(2):231–243CrossRefPubMedGoogle Scholar
- 36.Baker E, Sims R, Leonenko G, Frizzati A, Harwood J, Grozeva D, et al. Gene-based analysis in HRC imputed genome wide association data identifies three novel genes for Alzheimer’s disease. www.biorxiv.org/content/early/2018/07/23/374876.
- 39.Oberkofler H, Hafner M, Felder T, Krempler F, Patsch W (2009) Transcriptional co-activator peroxisome proliferator-activated receptor (PPAR)gamma co-activator-1beta is involved in the regulation of glucose-stimulated insulin secretion in INS-1E cells. J Mol Med (Berl) 87(3):299–306CrossRefGoogle Scholar
- 50.Lin W, Metzakopian E, Mavromatakis YE, Gao N, Balaskas N, Sasaki H, Briscoe J, Whitsett JA et al (2009) Foxa1 and Foxa2 function both upstream of and cooperatively with Lmx1a and Lmx1b in a feedforward loop promoting mesodiencephalic dopaminergic neuron development. Dev Biol 333(2):386–396CrossRefPubMedGoogle Scholar
- 51.Pristera A, Lin W, Kaufmann AK, Brimblecombe KR, Threlfell S, Dodson PD, Magill PJ, Fernandes C et al (2015) Transcription factors FOXA1 and FOXA2 maintain dopaminergic neuronal properties and control feeding behavior in adult mice. Proc Natl Acad Sci U S A 112(35):E4929–E4938CrossRefPubMedPubMedCentralGoogle Scholar
- 57.Semenza GL, Jiang BH, Leung SW, Passantino R, Concordet JP, Maire P, Giallongo A (1996) Hypoxia response elements in the aldolase A, enolase 1, and lactate dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. J Biol Chem 271(51):32529–32537CrossRefPubMedGoogle Scholar
- 65.Laganiere J, Tremblay GB, Dufour CR, Giroux S, Rousseau F, Giguere V (2004) A polymorphic autoregulatory hormone response element in the human estrogen-related receptor alpha (ERRalpha) promoter dictates peroxisome proliferator-activated receptor gamma coactivator-1alpha control of ERRalpha expression. J Biol Chem 279(18):18504–18510CrossRefPubMedGoogle Scholar
- 66.Cunningham KF, Beeson GC, Beeson CC, Baicu CF, Zile MR, McDermott PJ (2015) Estrogen-related receptor alpha (ERRalpha) is required for adaptive increases in PGC-1 isoform expression during electrically stimulated contraction of adult cardiomyocytes in sustained hypoxic conditions. Int J Cardiol 187:393–400CrossRefPubMedGoogle Scholar
- 80.Bayele HK, Peyssonnaux C, Giatromanolaki A, Arrais-Silva WW, Mohamed HS, Collins H, Giorgio S, Koukourakis M et al (2007) HIF-1 regulates heritable variation and allele expression phenotypes of the macrophage immune response gene SLC11A1 from a Z-DNA forming microsatellite. Blood 110(8):3039–3048CrossRefPubMedGoogle Scholar
- 81.Carroll VA, Ashcroft M (2006) Role of hypoxiy factor (HIF)-1α versus HIF-2α in the regulation of HIF target genes in response to hypoxia, insulin-like growth factor –I, or loss of von Hippel-Lindau function: implications for targeting the HIF pathway. Cancer Res 66(12):6264–6270CrossRefPubMedGoogle Scholar
- 82.Martinez-Redondo V, Jannig PR, Correia JC, Ferreira DM, Cervenka I, Lindvall JM, Sinha I, Izadi M et al (2016) Peroxisome proliferator-activated receptor gamma coactivator-1 alpha isoforms selectively regulate multiple splicing events on target genes. J Biol Chem 291(29):15169–15184CrossRefPubMedPubMedCentralGoogle Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.