Abstract

Mitochondrial dysfunction, which is consistently observed in Down syndrome (DS) cells and tissues, might contribute to the severity of the DS phenotype. Our recent studies on DS fetal hearts and fibroblasts have suggested that one of the possible causes of mitochondrial dysfunction is the downregulation of peroxisome proliferator-activated receptor gamma, coactivator 1 alpha (PGC-1α or PPARGC1A)—a key modulator of mitochondrial function—and of several nuclear-encoded mitochondrial genes (NEMGs). Re-analysis of publicly available expression data related to manipulation of chromosome 21 (Hsa21) genes suggested the nuclear receptor interacting protein 1 (NRIP1 or RIP140) as a good candidate Hsa21 gene for NEMG downregulation. Indeed, NRIP1 is known to affect oxidative metabolism and mitochondrial biogenesis by negatively controlling mitochondrial pathways regulated by PGC-1α. To establish whether NRIP1 overexpression in DS downregulates both PGC-1α and NEMGs, thereby causing mitochondrial dysfunction, we used siRNAs to decrease NRIP1 expression in trisomic human fetal fibroblasts. Levels of PGC-1α and NEMGs were increased and mitochondrial function was restored, as shown by reactive oxygen species decrease, adenosine 5′-triphosphate (ATP) production and mitochondrial activity increase. These findings indicate that the Hsa21 gene NRIP1 contributes to the mitochondrial dysfunction observed in DS. Furthermore, they suggest that the NRIP1-PGC-1α axe might represent a potential therapeutic target for restoring altered mitochondrial function in DS.

INTRODUCTION

Data from several studies show that trisomy of chromosome 21 (TS21) affects both mitochondrial function and reactive oxygen species (ROS) production. Lower levels of the mitochondrial complexes I, III and V have been observed in the cerebellar and brain regions of subjects affected by Down syndrome (DS) (1). Moreover, reduced mitochondrial redox activity and membrane potential have been observed in DS astrocytes and neuronal cultures (2,3). Further evidence for mitochondrial dysfunction was found in the Ts1Cje mouse model for DS that shows decreased levels of ATP production (4). Similarly, fetal DS fibroblasts show both a decreased efficiency of the mitochondrial energy production apparatus, involving adenine nucleotide translocators, ATP synthase, and adenylate kinase and a selective deficit of complex I, which might contribute to ROS overproduction by DS mitochondria. These events were correlated with changes in the cAMP/PKA signaling pathway (5,6). Similar research conducted on human primary lines of fibroblasts (HFFs) from TS21 fetuses has revealed that TS21 disrupts mitochondrial morphology, decreases oxygen consumption, increases mtCa2+ load and ROS production (7). Moreover, by analyzing mitochondrial defects according to the cardiac phenotype, a more severe mitochondrial dysfunction was evidenced in cardiopathic-derived TS21 fibroblasts (7). A possible interpretation of these results is that a more pronounced pro-oxidative state might contribute to generating a more severe cardiac phenotype—a concept that might be extended to other phenotypic traits. Studies of genome-wide expression analysis in DS have demonstrated that nuclear-encoded mitochondrial genes (NEMGs) represent the main downregulated category in fetal TS21 heart samples (8). Downregulation is also manifest in DS fetal brains (9). These observations led us to hypothesize that NEMG dysregulation is likely a cause of mitochondrial dysfunction in DS (8). Among the dysregulated genes, the peroxisome proliferator-activated receptor gamma, coactivator 1 alpha (PGC-1α/PPARGC1A) has been found hypo-expressed at the transcriptional and protein levels in TS21 HFFs (7). PGC-1α is indeed known to play a central role in regulating mitochondrial biogenesis and respiratory function through the interaction with transcriptional partners, such as NRF1, ERRα, PPARs and YY1 (10).

A known repressor of PGC-1α activity is the nuclear receptor interacting protein 1 (NRIP1/RIP140) (11–13). This protein is coded by a highly conserved chromosome 21 (Hsa21) gene with consistent dosage effect in many studies on DS samples (14). NRIP1 acts as a scaffold protein recruiting regulatory factors, such as histone deacetylases (15), COOH terminal binding protein (16) and histone metyltransferases (17), to exert its co-repressive function. Furthermore, NRIP1 directly interacts with some nuclear receptors including PPARs, ERRs and ERs (18–20). More specifically, NRIP1 negatively controls the expression and the activity of PGC-1α as well as the expression of its targets. Indeed, in PGC-1α null mice (21,22), as well as in knock-in NRIP1 mice (23), NEMG expression is decreased. Likewise, NRIP1 is always upregulated by 1.5- to 4-fold in the heart (8) and fibroblasts (7) from DS fetuses. NRIP1 protein is also increased in the hippocampal tissue from DS subjects (24).

Thus, building upon previous research, in this study, we endeavored to provide new insights into the transcriptional changes influencing the molecular mechanisms associated with mitochondrial dysfunction in DS. In particular, we first investigated whether Hsa21 gene overexpression causes NEMG downregulation by focusing on those Hsa21 genes, transcription factors, and kinases that have the highest probability of affecting the expression of many genes. To this aim, we analyzed the public expression data related to the manipulation of Hsa21 genes to investigate their effects on NEMG downregulation. These analyses led us to the identification of NRIP1 as a good candidate for the downregulation of mitochondria-related genes in DS.

Then, to determine whether NRIP1 downregulation can effectively counteract mitochondrial dysfunction and some of its pathophysiological effects, we attenuated NRIP1 expression in human fibroblasts from DS fetuses (DS-HFFs).

Understanding the molecular correlation between NRIP1 expression levels, NEMG regulation, and mitochondrial function could lay the basis for the development of new therapeutic protocols for DS.

RESULTS

Analysis of public expression data suggests that NRIP1 affects NEMG expression

Several Hsa21 genes can possibly interfere with NEMG expression. For instance, DYRK1A and DSCR1/RCAN1 play key roles in the calcineurin/NFAT pathway, which affects mitochondrial activity and morphology during heart development (25). Two more genes, NRIP1 (23) and GABPA/NRF2 (26) are also involved in mitochondrial pathways.

To identify which Hsa21 gene might possibly downregulate NEMG expression, we screened the Gene Expression Omnibus (27) repository (http://www.ncbi.nlm.nih.gov/geo) for gene expression data related to the modulation of Hsa21 genes. We selected the GEO GSE 19836 experiment (28), a set of data derived from a mouse embryonic stem cell (ESC) bank in which several orthologs of Hsa21 genes, with potential regulatory role, are individually overexpressed in an inducible manner. Expression data were available for 13 transcription factors (including NRIP1, RCAN1 and GABPA), the transcriptional activator RCAN1 and six protein kinases (including DYRK1A) (for details see Materials and Methods). We re-analyzed this series of data by focusing on the mitochondria-related categories and pathways dysregulated by the overexpression of each gene looking for Hsa21 genes that when overexpressed would induce NEMG downregulation. Among the 20 analyzed Hsa21 genes, only NRIP1, one of the seven genes that are considered ‘effective’ for the expression perturbation in the manipulated cells (28), was able to cause NEMG downregulation when overexpressed. Our analysis showed that NRIP1 overexpression caused a significant enrichment of NEMGs among 298 downregulated genes. The ‘Mitochondrion’ was the most affected Cell Component Gene Ontology (GO) category (P < 0.0001) (Table 1 and Supplementary Material, Fig. S1), with a cluster of 37 downregulated genes. Motif enrichment analysis, by clustering downregulated genes on the basis of their promoter regions, revealed a significant enrichment (P < 0.005) in genes with the ERRα motif. Twenty-five downregulated genes, instead of the expected 10, showed promoter regions around the transcription start site containing the ERRα motif.

Table 1.

Gene Ontology categories affected by NRIP1 overexpression in GSE 19836 series with a P-value <0.05

Biological process GO category Genes in category Observed Expected O/R ratio P-value 
Carboxylic acid metabolic process GO:0019752 630 20 7.51 2.66 0.0089 
Cellular carbohydrate metabolic process GO:0044262 198 11 2.36 4.66 0.0089 
Monocarboxylic acid metabolic process GO:0032787 339 14 4.04 3.47 0.0089 
Carboxylic acid catabolic process GO:0046395 139 1.66 4.83 0.0100 
Organic acid catabolic process GO:0016054 139 1.66 4.83 0.0100 
Regulation of transmembrane receptor protein
serine/threonine kinase signaling pathway 
GO:0090092 137 1.63 4.9 0.0100 
Carbohydrate metabolic process GO:0005975 527 17 6.28 2.71 0.0100 
Lipid metabolic process GO:0006629 881 23 10.5 2.19 0.0100 
Oxoacid metabolic process GO:0043436 667 20 7.95 2.52 0.0100 
Organic acid metabolic process GO:0006082 680 20 8.1 2.47 0.0100 
Molecular function 
 Kinase activity GO:0016301 736 21 9.07 2.32 0.0219 
 Transferase activity GO:0016740 1574 34 19.4 1.75 0.0328 
 Transferase activity. transferring
phosphorus-containing groups 
GO:0016772 857 21 10.56 1.99 0.0511 
 Cofactor binding GO:0048037 244 3.01 2.99 0.0602 
Cellular component 
 Mitochondrion GO:0005739 1480 37 17.7 2.09 0.0009 
 Mitochondrial envelope GO:0005740 446 17 5.34 3.19 0.0010 
 Mitochondrial membrane GO:0031966 424 16 5.07 3.15 0.0013 
 Organelle inner membrane GO:0019866 330 13 3.95 3.29 0.0036 
 Mitochondrial part GO:0044429 548 17 6.56 2.59 0.0043 
 Mitochondrial inner membrane GO:0005743 312 12 3.73 3.22 0.0047 
 Organelle envelope GO:0031967 671 18 8.03 2.24 0.0122 
 Envelope GO:0031975 683 18 8.17 2.2 0.0133 
 Mitochondrial outer membrane GO:0005741 111 1.33 4.52 0.0174 
 Organelle outer membrane GO:0031968 125 1.5 4.01 0.0277 
Biological process GO category Genes in category Observed Expected O/R ratio P-value 
Carboxylic acid metabolic process GO:0019752 630 20 7.51 2.66 0.0089 
Cellular carbohydrate metabolic process GO:0044262 198 11 2.36 4.66 0.0089 
Monocarboxylic acid metabolic process GO:0032787 339 14 4.04 3.47 0.0089 
Carboxylic acid catabolic process GO:0046395 139 1.66 4.83 0.0100 
Organic acid catabolic process GO:0016054 139 1.66 4.83 0.0100 
Regulation of transmembrane receptor protein
serine/threonine kinase signaling pathway 
GO:0090092 137 1.63 4.9 0.0100 
Carbohydrate metabolic process GO:0005975 527 17 6.28 2.71 0.0100 
Lipid metabolic process GO:0006629 881 23 10.5 2.19 0.0100 
Oxoacid metabolic process GO:0043436 667 20 7.95 2.52 0.0100 
Organic acid metabolic process GO:0006082 680 20 8.1 2.47 0.0100 
Molecular function 
 Kinase activity GO:0016301 736 21 9.07 2.32 0.0219 
 Transferase activity GO:0016740 1574 34 19.4 1.75 0.0328 
 Transferase activity. transferring
phosphorus-containing groups 
GO:0016772 857 21 10.56 1.99 0.0511 
 Cofactor binding GO:0048037 244 3.01 2.99 0.0602 
Cellular component 
 Mitochondrion GO:0005739 1480 37 17.7 2.09 0.0009 
 Mitochondrial envelope GO:0005740 446 17 5.34 3.19 0.0010 
 Mitochondrial membrane GO:0031966 424 16 5.07 3.15 0.0013 
 Organelle inner membrane GO:0019866 330 13 3.95 3.29 0.0036 
 Mitochondrial part GO:0044429 548 17 6.56 2.59 0.0043 
 Mitochondrial inner membrane GO:0005743 312 12 3.73 3.22 0.0047 
 Organelle envelope GO:0031967 671 18 8.03 2.24 0.0122 
 Envelope GO:0031975 683 18 8.17 2.2 0.0133 
 Mitochondrial outer membrane GO:0005741 111 1.33 4.52 0.0174 
 Organelle outer membrane GO:0031968 125 1.5 4.01 0.0277 

The ‘Mitochondrion’ is the category most affected by NRIP1 upregulation (enrichment = 37 observed genes instead of 17.7 expected genes with P < 0.001).

Neither DYRK1A, nor RCAN1, nor GABPA, all considered ‘silent’ genes (28), caused NEMG downregulation when overexpressed.

Modulation of NRIP1 and PGC-1α expression dysregulates the same NEMGs downregulated in DS fetal hearts

To investigate whether the sets of genes regulated by NRIP1 and/or PGC-1α showed any overlapping to the NEMGs downregulated in DS fetal hearts (8), we performed a meta-analysis comparing three sets of gene expression data, SET1, SET2 and SET3. SET1 included 123 genes which were both upregulated after NRIP1 silencing and downregulated after NRIP1 re-expression in mouse adipocytes (29). SET2 included 129 genes which were upregulated after PGC-1α induction in SAOS2 cells (human osteoblast-like cells) (30). SET3 included the 70 genes downregulated in DS fetal heart tissues (8) belonging to the ‘mitochondrion’ GO category (Supplementary Material, Table S1). The comparison was aimed at identifying genes consistently dysregulated across these studies.

The Venn Diagram shows that NEMGs in SET3, which were downregulated in DS fetal hearts, overlap with both SET1 and SET2 (Fig. 1A). The three sets of genes overlap each other for at least 25 genes. Fifteen genes are consistently dysregulated across all three experiments (Fig. 1B). Most of these genes are included in the electron transport chain, mainly in complex I, and in oxidative phosphorylation pathways. It is also interesting to note that 42 genes overlap between the sets of genes inversely regulated by NRIP1 and PGC-1α (SET1 and SET2), in agreement with the antagonistic functions of the two coregulators (19).

Figure 1.

Comparison of NEMGs downregulated in DS fetal hearts with those dysregulated by NRIP1 and/or PGC-1α. (A). Venn diagram showing overlapping among the three sets of data. Out of the 70 mitochondrial genes that are downregulated in DS fetal hearts (SET3) (8), 25 overlap the list of NRIP1 regulated genes (SET1) (29) and 29 overlap the list of PGC-1α regulated genes (SET2) (30). (B) List of mitochondria-related genes overlapping in the three sets of data. The complete lists of genes are in Supplementary Material, Table S1.

Figure 1.

Comparison of NEMGs downregulated in DS fetal hearts with those dysregulated by NRIP1 and/or PGC-1α. (A). Venn diagram showing overlapping among the three sets of data. Out of the 70 mitochondrial genes that are downregulated in DS fetal hearts (SET3) (8), 25 overlap the list of NRIP1 regulated genes (SET1) (29) and 29 overlap the list of PGC-1α regulated genes (SET2) (30). (B) List of mitochondria-related genes overlapping in the three sets of data. The complete lists of genes are in Supplementary Material, Table S1.

NRIP1 attenuation by siRNA affects NEMG expression in DS-HFFs

We previously demonstrated that NRIP1 is upregulated in DS-HFFs in which Hsa21 trisomy negatively regulates NEMGs and impairs mitochondrial function (7).

To test the hypothesis that NRIP1 overexpression perturbs mitochondrial function and that this effect is associated with PGC-1α downregulation, we performed silencing experiments of NRIP1 gene in DS-HFFs. In brief, after re-analyzing all DS-HFF lines used for silencing experiments, we demonstrated that NRIP1 is significantly upregulated in all trisomic samples if compared with euploid controls (Fig. 2A). Seventy-two hours after transfection of a specific SMART pool of siRNAs in DS-HFFs, an inverse correlation between NRIP1 and PGC-1α expression, in a siRNA dosage-dependent way, was demonstrated by qRT–PCR (Fig. 2B). By immunofluorescence analysis, we demonstrated that the NRIP1 protein localizes to the cell nucleus, as expected for a corepressor protein, both in euploid and in trisomic fibroblasts (Fig. 3). Fluorescent signal was more intense over nuclei of DS-HFFs (Fig. 3B) with respect to euploid HFFs (Fig. 3A), indicating a higher concentration of the NRIP1 protein in trisomic cells. In these cells, some fluorescent signal was also present over the cytoplasm (Fig. 3B) likely due to the overexpression of the NRIP1 protein. In DS-HFFs treated with siRNAs to attenuate NRIP1 mRNA expression, NRIP1 fluorescent signal was significantly decreased in a siRNA dosage-dependent way (Fig. 3). Quantitative evaluation of fluorescence intensities in euploid, trisomic and siRNA transfected cells (Fig. 3F) indicated that siRNA transfection reduces NRIP1 protein levels of trisomic cells down to the range of diploid cells or even lower (Fig. 3D and E).

Figure 2.

NRIP1 modulates PGC-1α expression in HFFs. (A) NRIP1 mRNA expression level in euploid cells (N1-N5), and in trisomic (DS1-DS8) HFF lines used for silencing experiments. For each sample, values represent the average determination ± SEM for three qRT–PCR experiments. A pool of euploid cells was used as a calibrator. **P < 10−4. P-values express statistical significance for euploid versus trisomic comparisons. (B) NRIP1 and PGC-1α expression levels in trisomic cells transfected with a scrambled siRNA and with a NRIP1-specific SMART pool of siRNAs. A decrease in the NRIP1 expression level corresponds to an increase of PGC-1α expression level in a siRNA-dependent way. Values represent the average determination ± SEM for eight NRIP1-silenced DS-HFFs carried out in triplicate. *P < 0.05, **P < 0.01. P-values express statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 2.

NRIP1 modulates PGC-1α expression in HFFs. (A) NRIP1 mRNA expression level in euploid cells (N1-N5), and in trisomic (DS1-DS8) HFF lines used for silencing experiments. For each sample, values represent the average determination ± SEM for three qRT–PCR experiments. A pool of euploid cells was used as a calibrator. **P < 10−4. P-values express statistical significance for euploid versus trisomic comparisons. (B) NRIP1 and PGC-1α expression levels in trisomic cells transfected with a scrambled siRNA and with a NRIP1-specific SMART pool of siRNAs. A decrease in the NRIP1 expression level corresponds to an increase of PGC-1α expression level in a siRNA-dependent way. Values represent the average determination ± SEM for eight NRIP1-silenced DS-HFFs carried out in triplicate. *P < 0.05, **P < 0.01. P-values express statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 3.

NRIP1 immunofluorescence in NRIP1-silenced DS-HFFs. Representative images of NRIP1 immunofluorescence analysis in (A) euploid cells, (B) trisomic cells and trisomic cells transfected (C) with a scrambled siRNA, (D) with 5 nmNRIP1 siRNA and (E) 20 nmNRIP1 siRNA. (F) Semi-quantitative analysis of the immunodetected signals, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A decrease of the fluorescent signal is observed in silenced versus scrambled DS-HFFs. Signal from 5 nmNRIP1 siRNA transfected cells is comparable with euploid HFFs. Statistical significance: **P < 0.01 for trisomic versus euploid comparisons; #P < 0.05 for NRIP1-silenced versus scrambled comparisons.

Figure 3.

NRIP1 immunofluorescence in NRIP1-silenced DS-HFFs. Representative images of NRIP1 immunofluorescence analysis in (A) euploid cells, (B) trisomic cells and trisomic cells transfected (C) with a scrambled siRNA, (D) with 5 nmNRIP1 siRNA and (E) 20 nmNRIP1 siRNA. (F) Semi-quantitative analysis of the immunodetected signals, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A decrease of the fluorescent signal is observed in silenced versus scrambled DS-HFFs. Signal from 5 nmNRIP1 siRNA transfected cells is comparable with euploid HFFs. Statistical significance: **P < 0.01 for trisomic versus euploid comparisons; #P < 0.05 for NRIP1-silenced versus scrambled comparisons.

To determine the effects of NRIP1 attenuation by siRNA on other mitochondria-related genes, we compared the expression of seven genes in silenced versus scrambled cells using qRT–PCR. Three genes, i.e. COX5A, NDUFA1 and NDUFS3, were chosen from the list of 15 genes that resulted consistently dysregulated across the three sets compared in the meta-analysis (Fig. 1). The fourth gene, NRF1, which is downregulated in DS hearts (8) and fibroblasts (7), was chosen because of its role both as a PGC-1α partner and as its target. Finally, three other genes, i.e. ANT1/SLC25A4, ANT2/SLC25A5 and ANT3/SLC25A6, which are downregulated in DS fetal fibroblasts (our unpublished data), were also chosen as PGC-1α targets (SET2, Supplementary Material, Table S1) (30). ANT1/SLC25A4 is downregulated in SET3 and after NRIP1 overexpression in the De Cegli's dataset (28). The expression ratio of these genes in NRIP1-silenced DS-HFFs versus scrambled transfected DS-HFFs demonstrated that five out of seven analyzed genes were significantly upregulated after NRIP1 attenuation by siRNA (Fig. 4).

Figure 4.

Mitochondria-related gene expression in NRIP1-silenced DS-HFFs. Relative mRNA expression of seven mitochondria-related genes was measured in NRIP1-silenced DS-HFFs versus scrambled transfected DS-HFFs. Five out of the seven genes show a significant increase in their expression level. Values represent the average determination ± SEM for three DS-HFF samples carried out in triplicate. A pool of scrambled transfected euploid cells was used as calibrator. *P < 0.05. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 4.

Mitochondria-related gene expression in NRIP1-silenced DS-HFFs. Relative mRNA expression of seven mitochondria-related genes was measured in NRIP1-silenced DS-HFFs versus scrambled transfected DS-HFFs. Five out of the seven genes show a significant increase in their expression level. Values represent the average determination ± SEM for three DS-HFF samples carried out in triplicate. A pool of scrambled transfected euploid cells was used as calibrator. *P < 0.05. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Furthermore, to verify the effect of NRIP1 attenuation and consequent PGC-1α upregulation on the mitochondrial biogenesis, we used qRT–PCR to quantify D-LOOP and ACTIN gene expression in scrambled and silenced cells, as mitochondrial and nuclear markers, respectively. The average of the D-LOOP/ACTIN ratio increased by 2.5-fold in silenced trisomic cells (Fig. 5), thereby suggesting that mtDNA content does increase after NRIP1 attenuation by siRNA and consequent PGC-1α overexpression.

Figure 5.

mtDNA content in NRIP1-silenced DS-HFFs. Ratio between the mtDNA marker D-LOOP and the nuclear DNA marker ACTIN indicates an increase after NRIP1 attenuation by siRNA. The ratio was calculated upon normalization to a reference gene (ABELSON) by qRT–PCR. Values represent the average determination ± SEM for three NRIP1-silenced trisomic samples carried out in triplicate. *P < 0.05. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 5.

mtDNA content in NRIP1-silenced DS-HFFs. Ratio between the mtDNA marker D-LOOP and the nuclear DNA marker ACTIN indicates an increase after NRIP1 attenuation by siRNA. The ratio was calculated upon normalization to a reference gene (ABELSON) by qRT–PCR. Values represent the average determination ± SEM for three NRIP1-silenced trisomic samples carried out in triplicate. *P < 0.05. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Mitochondrial function is improved in DS-HFFs after NRIP1 attenuation by siRNA

We verified whether NRIP1 attenuation by siRNA, along with the consequent increases in PGC-1α and other mitochondrial genes, might counteract the mitochondrial dysfunction in trisomic cells. For this study, intracellular ROS production, mitochondrial activity, mitochondrial calcium and ATP content were evaluated in DS-HFFs after transient NRIP1 siRNA-mediated attenuation.

ROS production was measured by confocal microscopy imaging of cells treated with the redox-sensitive fluorescent probe dichlorofluorescein (DCF). Seventy-two hours after transfection with NRIP1 siRNA, DCF-related fluorescence was lower with respect to scrambled DS-HFFs. Semi-quantitative analysis of fluorescent signals demonstrated that, on an average basis, the ROS-related DCF fluorescence decreased up to 50% in a siRNA dosage-dependent manner (Fig. 6).

Figure 6.

ROS decrease in NRIP1-silenced DS-HFFs. A confocal microscopy live cell imaging of the DCF fluorescence in transfected DS-HFFs: (A) scrambled, (B) 5 nmNRIP1 siRNA and (C) 20 nmNRIP1 siRNA. (D) Semi-quantitative analysis of the DCF-related fluorescence, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A significant decrease of DCF-related fluorescence is observed after NRIP1 attenuation in a siRNA-dependent way. **P < 10−4. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 6.

ROS decrease in NRIP1-silenced DS-HFFs. A confocal microscopy live cell imaging of the DCF fluorescence in transfected DS-HFFs: (A) scrambled, (B) 5 nmNRIP1 siRNA and (C) 20 nmNRIP1 siRNA. (D) Semi-quantitative analysis of the DCF-related fluorescence, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A significant decrease of DCF-related fluorescence is observed after NRIP1 attenuation in a siRNA-dependent way. **P < 10−4. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

We then established whether decreases in ROS could depend on a rescue of respiratory chain complex activities. To this aim, we incubated silenced DS-HFFs with the specific mitochondrial superoxide indicator, MitoSOX Red. This reagent is a live-cell permeant that is rapidly and selectively targeted to mitochondria. Once in the mitochondria, MitoSOX Red reagent is oxidized by superoxide and exhibits red fluorescence. In experiments performed in NRIP1-silenced DS-HFFs, a reduction of the MitoSOX Red signal was demonstrated thus suggesting that the decrease in ROS was partially associated with mitochondrial activity (Fig. 7).

Figure 7.

Intra-mitochondrial superoxide decrease in NRIP1-silenced DS-HFFs. Confocal microscopy live cell imaging of the MitoSOX Red fluorescence in transfected DS-HFFs: (A) scrambled and (B) 20 nmNRIP1 siRNA. Note that the distribution of MitoSOX Red signal resembles the mitochondrial network. (C) Semi-quantitative analysis of the MitoSOX-related fluorescence, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A reduction of the fluorescent signal over the mitochondrial network is detected. **P < 0.01. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 7.

Intra-mitochondrial superoxide decrease in NRIP1-silenced DS-HFFs. Confocal microscopy live cell imaging of the MitoSOX Red fluorescence in transfected DS-HFFs: (A) scrambled and (B) 20 nmNRIP1 siRNA. Note that the distribution of MitoSOX Red signal resembles the mitochondrial network. (C) Semi-quantitative analysis of the MitoSOX-related fluorescence, by the ImageJ software (means ± SEM of three assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. A reduction of the fluorescent signal over the mitochondrial network is detected. **P < 0.01. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Then, to confirm even further that NRIP1 attenuation by siRNA improves mitochondrial function, we incubated trisomic silenced cells with the MitoTracker Red dye, a reagent that stains mitochondria in live cells and whose accumulation is dependent upon membrane potential. A significant 50% increase of the MitoTracker Red-related fluorescence was observed in NRIP1-silenced cells when compared with scrambled controls, thus indicating an increase in respiratory activity (Fig. 8).

Figure 8.

Mitochondrial activity in DS NRIP1-silenced DS-HFFs. Confocal microscopy live cell imaging of the Mitotracker fluorescence in transfected DS-HFFs: (A) scrambled and (B) with 20 nmNRIP1 siRNA. (C) Semi-quantitative analysis of the Mitotracker-related fluorescence, by the ImageJ software (means ± SEM of five assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. An increase of Mitotracker-related fluorescence is observed in NRIP1-silenced DS-HFFs. **P < 0.005. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 8.

Mitochondrial activity in DS NRIP1-silenced DS-HFFs. Confocal microscopy live cell imaging of the Mitotracker fluorescence in transfected DS-HFFs: (A) scrambled and (B) with 20 nmNRIP1 siRNA. (C) Semi-quantitative analysis of the Mitotracker-related fluorescence, by the ImageJ software (means ± SEM of five assayed samples). Fifty randomly selected, different cells for each sample/experimental condition were analyzed. An increase of Mitotracker-related fluorescence is observed in NRIP1-silenced DS-HFFs. **P < 0.005. P-value expresses statistical significance for NRIP1-silenced versus scrambled comparisons.

NRIP1 attenuation by siRNA does not affect mitochondrial Ca2+ homeostasis

In DS-HFFs, the mitochondrial Ca2+ concentration is significantly greater than that of euploid fetal fibroblasts (7). Many extracellular stimuli exert their effect through an increase in cytosolic Ca2+ concentration ([Ca2+]c) mediated by the influx of extracellular Ca2+ and/or the release of Ca2+ from intracellular stores, predominantly the endoplasmic reticulum (ER). When [Ca2+]c increases, mitochondria undergo a major rise in the matrix Ca2+ concentration ([Ca2+]m). The amplitude of this rise largely exceeds that observed in the cytosol thanks to electrochemical potential across the cation-impermeant inner mitochondrial membrane that provides the driving force for mitochondrial Ca2+ accumulation (31).

Variations in [Ca2+]m were measured as previously described (7). In brief, DS-HFFs were transfected with a mitochondrially targeted aequorin (32) and then stimulated with histamine. This agonist elicited the production of inositol 1,4,5 trisphosphate (IP3) and the consequent release of Ca2+ from the ER, through the IP3 receptor.

We found no significant differences in the mitochondrial [Ca2+]m uptake in NRIP1-silenced DS-HFFs compared with control cells transfected with the non-targeting scrambled siRNA (42.0 ± 2.6 versus. 38.7 ± 3.9 µm, P = 0.5) (Fig. 9).

Figure 9.

Mitochondrial calcium measurement. (A) The barplot of the [Ca2+]m in scrambled and NRIP1 siRNA transfected DS-HFFs. The light signal was collected and calibrated into [Ca2+] values, as described in Materials and Methods. Results are shown as the average of measurements from four different NRIP1-silenced DS-HFFs ± SEM. (B) Effect of histamine on [Ca2+]m. The traces show the average [Ca2+]m in DS-HFFs transfected with the mitochondrially targeted aequorin. Where indicated, the cells were treated with 100 µm histamine added to KRB. No significant variation in [Ca2+]m is observed in the two conditions.

Figure 9.

Mitochondrial calcium measurement. (A) The barplot of the [Ca2+]m in scrambled and NRIP1 siRNA transfected DS-HFFs. The light signal was collected and calibrated into [Ca2+] values, as described in Materials and Methods. Results are shown as the average of measurements from four different NRIP1-silenced DS-HFFs ± SEM. (B) Effect of histamine on [Ca2+]m. The traces show the average [Ca2+]m in DS-HFFs transfected with the mitochondrially targeted aequorin. Where indicated, the cells were treated with 100 µm histamine added to KRB. No significant variation in [Ca2+]m is observed in the two conditions.

NRIP1 attenuation by siRNA strongly increases cellular ATP content

Levels of phosphorylated adenosine nucleotides, including the universal energy carrier ATP, define the energy state in living cells and depend mainly on mitochondrial function (33). In NRIP1-silenced DS-HFFs, we investigated the intramitochondrial ATP concentration ([ATP]m). For this purpose, we used a chimera of the ATP-sensitive photoprotein luciferase specifically targeted to mitochondria (mtLuc) to obtain a dynamic monitoring of [ATP]m. Luciferase has been widely employed to measure ATP content both in isolated mitochondria and in intact cells; its reaction with luciferin produces a flash of yellow-green light with a peak emission at 560 nm, the intensity of which is proportional to the amount of substrates in the reaction mixture.

We found that silenced DS-HFFs showed a very strong increase (+50%, P = 10−4) in basal ATP content, calculated by the luminescence values of the plateau generated after the addition of luciferin (Fig. 10). Since basal ATP content is highly dependent on the abundance of transfected luciferase, we determined the exact amount of the luciferase transduced under our experimental conditions through an immunoblot assay. We found that the levels of luciferase protein transduced in NRIP1-silenced DS-HFFs were comparable with those detected in control cells transfected with the non-targeting scrambled siRNA (Fig. 11).

Figure 10.

Mitochondrial ATP measurement. (A) The barplot of the mitochondrial ATP content and of the basal ATP content in scrambled and NRIP1 siRNA-transfected DS-HFFs. (B) The traces show mitochondrial [ATP]m changes elicited by mitochondrial [Ca2+] increase in cells perfused with 100 µm histamine as agonist. mtLuc luminescence data are expressed as a percentage of the initial value ± SEM (n = 4). The traces are representative of four independent experiments. *P = 0.05, **P = 10−4. P-values express statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 10.

Mitochondrial ATP measurement. (A) The barplot of the mitochondrial ATP content and of the basal ATP content in scrambled and NRIP1 siRNA-transfected DS-HFFs. (B) The traces show mitochondrial [ATP]m changes elicited by mitochondrial [Ca2+] increase in cells perfused with 100 µm histamine as agonist. mtLuc luminescence data are expressed as a percentage of the initial value ± SEM (n = 4). The traces are representative of four independent experiments. *P = 0.05, **P = 10−4. P-values express statistical significance for NRIP1-silenced versus scrambled comparisons.

Figure 11.

Luciferase expression following NRIP1 attenuation by siRNA. (A) Representative immunoblot of luciferase protein in three NRIP1-silenced or scrambled DS-HFFs transfected with a luciferase-encoding plasmid specifically targeted to mitochondria (mtLuc) and cultured in complete medium for 72 h. (B) Quantification of luciferase accumulation by the LUCIFERASE/GAPDH ratio.

Figure 11.

Luciferase expression following NRIP1 attenuation by siRNA. (A) Representative immunoblot of luciferase protein in three NRIP1-silenced or scrambled DS-HFFs transfected with a luciferase-encoding plasmid specifically targeted to mitochondria (mtLuc) and cultured in complete medium for 72 h. (B) Quantification of luciferase accumulation by the LUCIFERASE/GAPDH ratio.

In parallel, NRIP1-silenced cells were slightly decreased in mitochondrial ATP production 72 h after transfection. This was calculated by subtracting the basal cellular luminescence plateau, generated after the addition of luciferin, from the luminescence values of the second plateau, generated after the addition of the Ca2+ mobilizing agent histamine (Fig. 10).

DISCUSSION

This study originates from previous analyses demonstrating a global mitochondrial dysfunction in several DS models (1–4) and a significant dysregulation of NEMGs in the heart (8), brain (9) and fibroblasts (7) from human fetuses with DS. From these studies, it emerged that genes and transcription factors responsible for the activity of respiratory complexes and mitochondrial biogenesis are globally repressed. Thus, we speculated that most of the underexpressed NEMGs might be under the same regulatory control and that this control might be affected by the trisomy of Hsa21. In the present study, we looked for a regulator of NEMGs that maps to Hsa21 and that is upregulated in DS samples, by the virtue of a gene dosage effect. To this aim, we re-evaluated the expression data from the GEO repository (http://www.ncbi.nlm.nih.gov/geo) by focusing on an experiment in which regulatory genes mapping to Hsa21 were individually overexpressed in mouse ESCs (28). Our analysis demonstrated that only one gene is able to cause NEMG downregulation and that no other Hsa21 tested gene exerts such an effect. This gene is NRIP1 which encodes for a corepressor protein. Although the mean dysregulation of each NEMG elicited by NRIP1 overexpression was not very strong, the number of affected genes was significantly enriched (P < 0.001). The role of NRIP1 in mitochondrial dysfunction is supported by previous findings demonstrating that both in cellular and in animal models NRIP1 silencing upregulates the expression of genes responsible for mitochondrial biogenesis and oxidative phosphorylation, whereas NRIP1 re-expression downregulates them (29,23). Experiments of NRIP1 manipulation, performed in transgenic mice and human cells, have actually demonstrated that even mild variations in NRIP1 expression can significantly affect oxidative metabolism and mitochondrial biogenesis (11,23,29,34).

We also considered the possible effects of the overexpression of other Hsa21 genes that have previously been implicated in the regulation of mitochondrial function such as DYRK1A, DSCR1 and GABPA, but none of these genes turned out to regulate per se NEMG expression. GABPA, in particular, is a nuclear respiratory factor that would be expected to downregulate mitochondria-related genes when it is downregulated. However, GABPA is never downregulated in DS samples. Indeed it was normoregulated in DS fetal hearts (8), upregulated in DS fetal fibroblasts (7) and inconsistently dysregulated in Vilardell's meta-analysis (14).

On the other hand, the effect of NRIP1 on NEMG expression could be further reinforced by another Hsa21 gene, SUMO3, as sumoylation modulates NRIP1 activity (35). We thus speculate that the simultaneous upregulation of both NRIP1 and SUMO3 exerts a synergistic effect on mitochondrial dysfunction.

NRIP1 is supposed to exert a repression of mitochondrial biogenesis by either interacting with nuclear receptors (19,18) or regulating PGC-1α activity (10,12,13). PGC-1α knockout mice show not only a decreased number of mitochondria but also a decreased respiratory capacity in skeletal muscle (21). In particular, under physiological conditions, PGC-1α, by coactivating several transcription factors, including nuclear receptors such as PPARγ, PPARα and ERRα, promotes mitochondrial biogenesis and regulates mitochondrial respiratory efficiency (10,21,36). Interestingly, among the 37 NEMGs downregulated after NRIP1 induction in the GEO GSE 19836 experiment (28), we observed an enrichment both of genes involved in PPARs pathways (8 genes) and of genes containing the ERRα motif in their promoter regions (25 genes) (P < 0.0005). Notably, the known targets of PGC-1α, namely, CIDEA (12) and ANT1/SLC25A4 (30), are included in the list of genes that are downregulated following NRIP1 overexpression (28).

Moreover, to investigate whether the NEMGs repressed by NRIP1 and induced by PGC-1α corresponded to the NEMGs downregulated genes in DS fetal hearts (8), we performed a meta-analysis by comparing our microarray data with the results of two experiments in which the gene expression of NRIP1 or PGC-1α was modulated. We found that the correspondence between the three sets of genes was remarkably high, considering that they all derived from different species, tissues and experimental approaches. The high number of overlapping genes in SET1 and SET2 is in agreement with previous research indicating an interrelationship between PGC-1α and NRIP1 activity on mitochondrial pathways (11).

These results, combined with the data from previous research, finally led us to verify the potential role of NRIP1 in mitochondrial dysfunction in DS. When we transiently attenuated NRIP1 in trisomic fibroblasts, we demonstrated an inverse correlation between NRIP1 and PGC-1α expression. Accordingly, we found that this attenuation induced the upregulation of five out of seven genes randomly chosen in SET3, all of which overlapped with the lists of genes regulated by NRIP1 (SET1) and/or PGC-1α (SET2). Moreover, NRIP1 siRNA-mediated attenuation in DS-HFFs, and the consequent PGC-1α and NRF1 upregulation, elicited a significant increase in mtDNA. This result fully corroborates similar experiments performed in cardiomyocytes (11).

In the same trisomic fibroblasts, ROS production was decreased and mitochondrial activity was increased, demonstrating that the induction of NEMG expression in silenced DS-HFFs counteracts mitochondrial impairment and partially rescues mitochondrial function. However, no significant alterations of mitochondrial [Ca2+] were observed after NRIP1 attenuation by siRNA. A possible explanation to this phenomenon is either that 72 h is not a sufficient time to determine detectable differences in Ca2+ uptake or that many other mechanisms affect calcium uptake in TS21 cells, e.g. the trisomy of genes involved in the calcineurin pathway (DYRK1A and DSCR1) (25). Other Ca2+ regulators could play a role.

Interestingly, in NRIP1-silenced trisomic cells, we found a significant 50% increase in the basal ATP content. These results, together with the finding that NRIP1 attenuation by siRNA leads to an increase in the adenine nucleotide translocators ANT1/SLC25A4 and ANT2/SLC25A5 (Fig. 4), suggest that a more efficient exchange of ATP is induced, thus benefitting the mitochondrial activity and function of these cells, as demonstrated by the reduction in ROS production at the mitochondrial level (Fig. 7).

Supporting evidence for the opposite effects of NRIP1 and PGC-1α on mitochondrial function and NEMG regulation is that in neonatal rat cardiomyocytes NRIP1 mediates an antagonistic role versus PGC-1α in the regulation of mitochondrial energy metabolism (11). Indeed, overexpressed NRIP1 abrogates PGC-1α-mediated induction of mitochondrial membrane potential and mitochondrial biogenesis (11). Furthermore, the NRIP1-dependent repression of genes involved in mitochondrial function is closely linked with post-natal impaired cardiac function as a result of reduced mitochondrial electron-transport chain activity and oxygen consumption. NRIP1 hyperexpressing mice are indeed affected by cardiac hypertrophy (34).

NRIP1 and PGC-1α are also involved in glucose uptake and therefore in the physiopathology of diabetes through the regulation of the insulin sensitive glucose transporter GLUT4 expression and its sub-cellular localization (37). These findings correlate with the fact that cardiac hypertrophy and diabetes are two important post-natal complications of DS.

Mitochondrial dysfunction might also contribute to determining DS mental retardation and other DS-associated post-natal pathologies, such as Alzheimer's disease (AD) and obesity. It is known that mitochondria also play a central role in many neurodegenerative diseases such as AD, Parkinson's disease, Huntington's disease and amyotrophic lateral sclerosis. Impaired energy metabolism, defective mitochondrial enzymatic activity, abnormal mitochondrial respiration, mutated mtDNAs and oxidative stress are all common features of these neurodegenerative conditions (38).

It is interesting to note that the bioinformatic functional analysis of the 25 genes overlapping SET1 (genes regulated by NRIP1) and SET3 (genes downregulated in DS fetal hearts) showed that 16 out of 25 genes characterized the mitochondrial dysfunction pathways described in neurodegenerative diseases such as AD and Parkinson's disease [KEGG Pathways http://www.genome.jp/kegg/ (39)]. However, given that there is a high prevalence of AD in DS patients, we cannot neglect the possibility that the overexpression of the Hsa21 gene APP might have a main role in the development of AD in DS patients.

Taken all together our study indicates that NRIP1 is a key gene in the regulation of the mitochondrial pathways and that it is linked to mitochondrial dysfunction in DS. Accordingly, this should be taken into account when planning therapeutic approaches aimed at improving functions and cognitive performance in DS mouse models. Many of these models are indeed inadequate since they are not trisomic for all Hsa21 genes and may also have duplications of regions non-syntenic to Hsa21. This is true, for instance, for the very popular Ts65Dn mouse that, among other Hsa21 genes, is not trisomic for either NRIP1 or SUMO3. Thus, preclinical models of Ts65Dn will be unable to address all phenotypic problems and we speculate that clinical trial oftentimes fails because the overexpression of important genes such as NRIP1 is not taken into account.

Our results do highlight that NRIP1 plays a relevant role in DS mitochondrial dysfunction, as evidenced by the ability of NRIP1 inhibition to counteract mitochondrial dysfunction. However, we cannot rule out the likelihood that other genes may, in fact, be involved. A case in point is that the DS mouse model Ts1Cje manifests mitochondrial dysfunction even though it is not trisomic for either NRIP1 or for SUMO3. Thus, further studies are indeed warranted to identify additional genes possibly responsible for DS mitochondrial alterations.

Finally, these results provide the basis for clinical trials aimed at restoring mitochondrial function in DS subjects to counteract specific phenotypic features such as neurodegeneration, cardiac hypertrophy, diabetes and obesity. Such therapeutic approach would be highly desirable considering that the very few therapeutic approaches undertaken so far in this direction using antioxidants and nutraceutics have yielded either poor or discordant outcomes (40,41).

Thus, we speculate that a possible therapeutic approach in DS could be based either on PGC-1α activators, which have been tested in other disease mouse models (42–45), or on PPARγ agonists, which attenuate mitochondrial dysfunction in AD mouse models (46–51). Such drugs are already routinely used in clinical practice for the treatment of metabolic syndromes, type 2 diabetes, and neurodegenerative diseases such as AD (52–54).

In conclusion, our study has provided further insights into the transcription factors that influence mitochondrial dysfunction in DS. Our findings could indeed pave the way for the development of new and more effective drugs capable of selectively targeting the intricate set of molecular mechanisms underlying the pathogenesis of this disease.

MATERIALS AND METHODS

Analysis of public expression data

A set of expression data from GSE 19836 series (28) was obtained from Gene Expression Omnibus repository GEO (http://www.ncbi.nlm.nih.gov/geo). This set of data was derived from the analysis of a mouse ESC bank in which 32 orthologs of human chromosome 21 genes, including transcription factors and protein kinases, were individually overexpressed in an inducible manner. A set of clones individually overexpressing 20 of the 32 genes, namely 13 transcription factors (Aire, Bach1, Erg, Ets2, Gabpa, Nrip1, Olig1, Olig2, Pknox1, Runx1, Sim2, ZFP295, 1810007M14Rik), one transcriptional activator (Dscr1-Rcan1) and six protein kinases (DYRK1A, SNF1LK, Hunk, Pdxk, Pfkl, Ripk4), was transcriptionally profiled under inducing and non-inducing conditions with Affymetrix Gene Chip Mouse 430_2. Specifically, RNAs from three induced mouse ESCs and three controls were profiled for each inducible Hsa21 gene (28). In our analysis, we used GeneSpring software vers. 11.5 Multi–Omic Analysis (Agilent technologies, Inc.) for data interpretation; however, our criteria were different from those used by the authors of the gene expression dataset, focusing on downregulated genes. We considered genes differentially expressed with a Fold change (LogFC) >0.3 and <−0.3 with P < 0.05, thus producing two lists of dysregulated genes: 511 upregulated genes and 298 downregulated genes. Gene ontology (GO) functional class scoring of all the lists of significantly upregulated or downregulated genes was performed using the Web-based Gene Set Analysis Toolkit V2 (http://bioinfo.vanderbilt.edu/webgestalt/) (55,56). Special attention was given to mitochondria-related categories and pathways.

Meta-analysis

We compared three sets of gene expression data from different experiments, to identify genes consistently dysregulated across the three studies. The first set, SET1, included genes dysregulated by Nrip1 modulation in mouse adipocytes (29). The second set, SET2, included genes upregulated after PGC-1α induction in SAOS2 cells (human osteoblast like cells) (30). The third set included mitochondria-related genes, downregulated in DS fetal heart tissue (8). The three sets were filtered according to the GO cell component category ‘mitochondrion’ with the above-mentioned Web-based Gene Set Analysis Toolkit V2. The resulting genes—123 genes in SET1, 129 in SET2 and 70 in SET3 (Supplementary Material, Table S1)—were intersected using the R software (http://www.R-project.org/). A Venn diagram was built, which shows overlapping genes across the three sets.

Ethics statement

HFFs were obtained from the ‘Telethon Bank of Fetal Biological Samples’ at the University of Naples. All experimental protocols were approved by the local Institutional Ethics Committee.

Samples

Eight skin biopsies were explanted from human fetuses with trisomy of Hsa21 (DS-HFF) after therapeutic abortion at 18–22 gestational weeks. Fibroblasts from biopsies were cultured in T25 flasks (BD Falcon) with Chang medium B + C (Irvine Scientific) supplemented with 1% penicillin/streptomycin (Gibco) at 37°C in 5% CO2 atmosphere; all the analyses described throughout this study were carried out at cell culture passages 4–5.

Transfection protocol

NRIP1 was transiently silenced in eight DS-HFF lines using a pool of specific NRIP1-siRNAs (ON-TARGETplus SMARTpool, Dharmacon), with negative (ON-TARGETplus SMARTpool Non-targeting siRNAs control, Dharmacon) and positive controls (ON-TARGETplus SMARTpool, GAPDH siRNAs, Dharmacon). Interferin transfection reagent (Polyplus transfection) was used. Cells were plated on 12-well plates (50 000 cells/well) for RNA collection, on 35 mm diameter plates with 20 mm slides (Delchimica) (50 000 cells/well) for ROS production analysis and on 24-well plates (30 000 cells/well) (BD Falcon) for immunofluorescence and mitochondrial activity assays. DS-HFFs were transfected with 5 and 20 nm siRNA according to the manufacturer's protocol (Polyplus transfection). Seventy-two hours after transfection, the effects of NRIP1 siRNA-mediated attenuation were evaluated.

NRIP1 immunofluorescence

For the evaluation of NRIP1 protein by immunofluorescence, 30 000 cells were plated in 24-well plates on 12-mm diameter round glass coverslips. Cells were fixed in 3 : 1 methanol: acetic acid for 15 min, washed twice with PBS and then incubated twice in 0.1 m borate buffer pH 8.5 for 10 min to neutralize the pH. After two washes with PBS, the cells were incubated with DNase 1 : 10 in RDD buffer (Qiagen) at 37°C for 1 h and then treated with 2% BSA in PBS to block non-specific protein–protein interactions. The cells were then incubated with the antibody anti-NRIP1 (30 μg/ml, ab42126 Abcam, Cambridge Science Park, Cambridge, UK) overnight at +4°C. The secondary antibody (green) was Alexa Fluor® 488 goat anti-rabbit IgG (H + L) used at a 1/200 dilution for 1 h (57). Cells were finally mounted in 50% glycerol in PBS. Immunofluorescence analysis was performed at a confocal laser scanning microscope LSM 510 (Zeiss, Gottingen, Germany) equipped with an Argon ionic laser whose λ was set at 488 nm, and an HeNe laser whose λ was set at 633 nm. Emission of fluorescence was revealed by a BP 505–530 band pass filter for Alexa Fluor 488 and by a 615 long pass filter for DRAQ5. Images were acquired at a resolution of 1024 × 1024 pixels. Analysis of data was performed with the ImageJ software, version 1.37 (58). Fifty random single cells were analyzed for each imaging analysis.

Laser scanning confocal microscopy live cell imaging of ROS production

For the evaluation of ROS production after NRIP1 siRNA transfection, 50 000 cells were plated on 25-mm diameter round glass coverslips in an Attofluor cell chamber (Molecular Probe, Leiden, the Netherlands). Seventy-two hours later, the cells were incubated for 15 min at 37°C with 10 µm of 2,7-dichlorofluorescin diacetate (DCF-DA) which is converted to DCF by intracellular esterases, for detection of H2O2, or with 5 µm of MitoSOX™ Red reagent (Life Technologies, Molecular Probes), which is a live-cell permeant and is rapidly and selectively targeted to the mitochondria. Once in the mitochondria, MitoSOX™ Red reagent is oxidized by superoxide and exhibits red fluorescence. After incubation, cells were washed three times with medium without serum. To maintain the cells alive during observation and to create the proper environmental conditions, the specimen was placed in an Oko Lab (Na, Italy) Water Jacket Top Stage Incubator, kept at 37°C, under humidified condition of 5% CO2 and 95% air by means of temperature controllers, gas mixers and humidifiers right on the microscope.

The analysis of immunofluorescence was performed with a confocal laser scanner microscopy Zeiss LSM 510 (Carl Zeiss, Gottingen, Germany), equipped with Argon ionic laser whose λ was set at 488 nm, an HeNe laser whose λ was set at 546 nm and an immersion oil objective 63×/1.4f. Emission of fluorescence was revealed by BP 505–530 band pass filter for DCF and 560 Long Pass for MitoSOX Red. Images were acquired in the green or in the red channels and then saved in TIFF format to prevent the loss of information. They were acquired with a resolution of 1024 × 1024 pixel with the confocal pinhole set to one Airy unit.

Analysis of data was performed with the ImageJ software, version 1.37. Fifty random single cells were analyzed for each imaging analysis.

Mitotracker immunofluorescence

For the evaluation of mitochondrial activity, MitoTracker® Red CMXRos (Molecular Probes) was chosen. MitoTracker® probes passively diffuse across the plasma membrane and accumulate in actively respiring mitochondria. Thirty-thousand cells were plated on 24-well plates on 12 mm diameter round glass coverslips and then incubated with 150 nm of Mitotracker Red for 30 min. After incubation cells were fixed for 20 min in PBS containing 4% paraformaldehyde (Sigma) and then washed once with PBS 1×. Nuclei were stained with the DNA intercalant DRAQ5 (Bio status, Alexis Corporation). Cells were finally mounted in 50% glycerol in PBS. Immunofluorescence analysis was performed with a confocal laser scanning microscope LSM 510 (Zeiss, Gottingen, Germany). The lambda of the two HeNe lasers was set at 546 and at 633 nm. Fluorescence emission was revealed by BP 560–615 band pass filter for Mitotracker Red and by a 615-long pass filter for DRAQ5. Double staining immunofluorescence images were acquired separately in the red and infrared channels at a resolution of 1024 × 1024 pixels, with the confocal pinhole set to one Airy unit, and then saved in TIFF format. Fifty random single cells were analyzed for each imaging analysis using the ImageJ version 1.37.

RNA extraction and quantitative real-time PCR

Total RNA from each sample was extracted using TRIzol reagent (Gibco/BRL Life Technologies, Inc., Gaithersburg, MD) and was reverse transcribed using the iScript cDNA Synthesis kit (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

Real-time PCR was performed using iQ Supermix SYBR Green 2× on a Bio-Rad iCycler according to the manufacturer's protocols. PCR reactions were performed in triplicate. Primer pairs (MWG Biotech, Ebersberg, Germany) were designed using the Primer 3 software (http://frodo.wi.mit.edu/primer3) to obtain amplicons ranging from 100 to 150 base pairs. Expression values were normalized either versus scrambled transfected cells or versus scrambled transfected euploid cells. ABELSON and GAPDH housekeeping genes were chosen as reference genes.

mtDNA quantification

To quantify the mtDNA content, we selected two genes: D-LOOP as the mitochondrial target and ACTIN as the nuclear target. Both targets were quantified by qRT–PCR using cDNA reverse-transcribed from RNA of three NRIP1-silenced trisomic samples and scrambled control. Normalization of gene expression was obtained using the ABELSON gene as housekeeping. The ratio between D-LOOP and ACTIN expression under each condition (NRIP1-silenced or scrambled trisomic cells) was calculated.

Aequorin measurement

A chimeric aequorin targeted to the mitochondria (mtAEQmut) was used as a probe. For the experiments with mtAEQmut, cells were incubated with 5 mm coelenterazine (Fluka, 7372) for 1–2 h in DMEM supplemented with 1% FBS. A coverslip with transfected cells was placed in a perfused thermostated chamber located in close proximity to a low-noise photomultiplier with a built-in amplifier/discriminator. All aequorin measurements were performed in KRB supplemented with 1 mm CaCl2. Agonist was added to the same medium as specified in figure legends. The experiments were terminated by lysing cells with 100 mm digitonin in a hypotonic Ca2+-containing solution (10 mm CaCl2 in H2O), thus discharging the remaining aequorin pool. The output of the discriminator was captured by a Thorn EMI photon-counting board and stored in an IBM-compatible computer for further analyses. The aequorin luminescence data were calibrated offline into [Ca2+] values using a computer algorithm based on the Ca2+ response curve of mutant aequorins.

Immunoblotting

For immunoblotting, cells were scraped into ice-cold phosphate-buffered saline and lysed in a modified 10 mm Tris buffer, pH 7.4, containing 150 mm NaCl, 1% Triton X-100, 10% glycerol, 10 mm EDTA and protease inhibitor cocktail. After 30 min of incubation on ice, the lysates were cleared via centrifugation at 12 000g at 4°C for 10 min. Protein concentrations were determined by the Lowry procedure. Protein extracts (18 μg) were separated on 4–12% Bis-Tris acrylamide Gel (Life Technologies, NP0323) and electron-transferred to PVDF or nitrocellulose membrane according to standard procedures. Unspecific binding sites were saturated by incubating membranes with TBS-Tween 20 (0.05%) supplemented with 5% non-fat powdered milk for 1 h. Next, the membranes were incubated overnight with primary antibodies [GAPDH (Cell Signaling, 2118); LUCIFERASE (Invitrogen, 356700)] and the detection was assessed by appropriate HRP-labeled secondary antibodies [Santa Cruz, sc-2004 (goat anti-rabbit) and sc-2005 (goat anti-mouse)] plus a chemiluminescent substrate (Thermo Scientific, 34080). Equal loading of lanes was confirmed by incubation with an anti-GAPDH antibody.

Luciferase measurements

Cells were seeded on glass coverslips (13 mm in diameter) for single sample luminescence measurements and allowed to grow until 50% confluence. The cells were then transfected with a cytosolic (untargeted) firefly luciferase and a mtLuc (59,60).

Cell luminescence was measured in the same purpose-built luminometer used for the aequorin measurements, constantly perfused with KRB, supplemented with 1 mm CaCl2 and 20 mm luciferin. The light output of a coverslip of infected cells was in the range of 1000–10 000 counts per second (cps) versus a background <10 cps. All compounds employed in the experiments were tested for non-specific effects on the luminescence, but none was observed.

Statistics

The ANOVA test, with a Bonferroni post hoc correction in case of multiple comparisons, was applied to evaluate the statistical significance of differences measured throughout the datasets presented. The threshold for statistical significance (P-value) was set at 0.05.

SUPPLEMENTARY MATERIAL

Supplementary Material is available at HMG online.

FUNDING

This work has been supported by POR Campania FSE 2007-2013, Project CREME from Campania Region to L.N.

ACKNOWLEDGEMENTS

We thank Paola Merolla for language revision and editing.

Conflict of Interest statement. None declared.

REFERENCES

1
Kim
S.H.
Vlkolinsky
R.
Cairns
N.
Fountoulakis
M.
Lubec
G.
The reduction of NADH ubiquinone oxidoreductase 24- and 75-kDa subunits in brains of patients with Down syndrome and Alzheimer's disease
Life Sci.
 , 
2001
, vol. 
68
 (pg. 
2741
-
2750
)
2
Busciglio
J.
Pelsman
A.
Wong
C.
Pigino
G.
Yuan
M.
Mori
H.
Yankner
B.A.
Altered metabolism of the amyloid beta precursor protein is associated with mitochondrial dysfunction in Down's syndrome
Neuron
 , 
2002
, vol. 
33
 (pg. 
677
-
688
)
3
Helguera
P.
Seiglie
J.
Rodriguez
J.
Hanna
M.
Helguera
G.
Busciglio
J.
Adaptive downregulation of mitochondrial function in down syndrome
Cell. Metab.
 , 
2013
, vol. 
17
 (pg. 
132
-
140
)
4
Shukkur
E.A.
Shimohata
A.
Akagi
T.
Yu
W.
Yamaguchi
M.
Murayama
M.
Chui
D.
Takeuchi
T.
Amano
K.
Subramhanya
K.H.
, et al.  . 
Mitochondrial dysfunction and tau hyperphosphorylation in Ts1Cje, a mouse model for Down syndrome
Hum. Mol. Genet.
 , 
2006
, vol. 
15
 (pg. 
2752
-
2762
)
5
Valenti
D.
Tullo
A.
Caratozzolo
M.F.
Merafina
R.S.
Scartezzini
P.
Marra
E.
Vacca
R.A.
Impairment of F1F0-ATPase, adenine nucleotide translocator and adenylate kinase causes mitochondrial energy deficit in human skin fibroblasts with chromosome 21 trisomy
Biochem. J.
 , 
2010
, vol. 
431
 (pg. 
299
-
310
)
6
Valenti
D.
Manente
G.A.
Moro
L.
Marra
E.
Vacca
R.A.
Deficit of complex I activity in human skin fibroblasts with chromosome 21 trisomy and overproduction of reactive oxygen species by mitochondria: involvement of the cAMP/PKA signalling pathway
Biochem. J.
 , 
2011
, vol. 
435
 (pg. 
679
-
688
)
7
Piccoli
C.
Izzo
A.
Scrima
R.
Bonfiglio
F.
Manco
R.
Negri
R.
Quarato
G.
Cela
O.
Ripoli
M.
Prisco
M.
, et al.  . 
Chronic pro-oxidative state and mitochondrial dysfunctions are more pronounced in fibroblasts from Down syndrome foeti with congenital heart defects
Hum. Mol. Genet.
 , 
2013
, vol. 
22
 (pg. 
1218
-
1232
)
8
Conti
A.
Fabbrini
F.
D'Agostino
P.
Negri
R.
Greco
D.
Genesio
R.
D'Armiento
M.
Olla
C.
Paladini
D.
Zannini
M.
, et al.  . 
Altered expression of mitochondrial and extracellular matrix genes in the heart of human fetuses with chromosome 21 trisomy
BMC Genomics
 , 
2007
, vol. 
8
 pg. 
268
 
9
Mao
R.
Wang
X.
Spitznagel
E.L.
Jr
Frelin
L.P.
Ting
J.C.
Ding
H.
Kim
J.W.
Ruczinski
I.
Downey
T.J.
Pevsner
J.
Primary and secondary transcriptional effects in the developing human Down syndrome brain and heart
Genome Biol.
 , 
2005
, vol. 
6
 pg. 
R107
 
10
Scarpulla
R.C.
Metabolic control of mitochondrial biogenesis through the PGC-1 family regulatory network
Biochim. Biophys. Acta
 , 
2011
, vol. 
1813
 (pg. 
1269
-
1278
)
11
Chen
Y.
Wang
Y.
Chen
J.
Chen
X.
Cao
W.
Chen
S.
Xu
S.
Huang
H.
Liu
P.
Roles of transcriptional corepressor RIP140 and coactivator PGC-1alpha in energy state of chronically infarcted rat hearts and mitochondrial function of cardiomyocytes
Mol. Cell Endocrinol.
 , 
2012
, vol. 
362
 (pg. 
11
-
18
)
12
Hallberg
M.
Morganstein
D.L.
Kiskinis
E.
Shah
K.
Kralli
A.
Dilworth
S.M.
White
R.
Parker
M.G.
Christian
M.
A functional interaction between RIP140 and PGC-1alpha regulates the expression of the lipid droplet protein CIDEA
Mol. Cell Biol.
 , 
2008
, vol. 
28
 (pg. 
6785
-
6795
)
13
Rytinki
M.M.
Palvimo
J.J.
SUMOylation attenuates the function of PGC-1alpha
J. Biol. Chem.
 , 
2009
, vol. 
284
 (pg. 
26184
-
26193
)
14
Vilardell
M.
Rasche
A.
Thormann
A.
Maschke-Dutz
E.
Perez-Jurado
L.A.
Lehrach
H.
Herwig
R.
Meta-analysis of heterogeneous Down syndrome data reveals consistent genome-wide dosage effects related to neurological processes
BMC Genomics
 , 
2011
, vol. 
12
 pg. 
229
 
15
Wei
L.N.
Hu
X.
Chandra
D.
Seto
E.
Farooqui
M.
Receptor-interacting protein 140 directly recruits histone deacetylases for gene silencing
J. Biol. Chem.
 , 
2000
, vol. 
275
 (pg. 
40782
-
40787
)
16
Vo
N.
Fjeld
C.
Goodman
R.H.
Acetylation of nuclear hormone receptor-interacting protein RIP140 regulates binding of the transcriptional corepressor CtBP
Mol. Cell Biol.
 , 
2001
, vol. 
21
 (pg. 
6181
-
6188
)
17
Kiskinis
E.
Hallberg
M.
Christian
M.
Olofsson
M.
Dilworth
S.M.
White
R.
Parker
M.G.
RIP140 directs histone and DNA methylation to silence Ucp1 expression in white adipocytes
Embo J.
 , 
2007
, vol. 
26
 (pg. 
4831
-
4840
)
18
Nautiyal
J.
Christian
M.
Parker
M.G.
Distinct functions for RIP140 in development, inflammation, and metabolism
Trends Endocrinol. Metab.
 , 
2013
, vol. 
24
 (pg. 
451
-
459
)
19
Fritah
A.
Christian
M.
Parker
M.G.
The metabolic coregulator RIP140: an update
Am. J. Physiol. Endocrinol. Metab.
 , 
2010
, vol. 
299
 (pg. 
E335
-
E340
)
20
Christian
M.
White
R.
Parker
M.G.
Metabolic regulation by the nuclear receptor corepressor RIP140
Trends Endocrinol. Metab.
 , 
2006
, vol. 
17
 (pg. 
243
-
250
)
21
Leone
T.C.
Lehman
J.J.
Finck
B.N.
Schaeffer
P.J.
Wende
A.R.
Boudina
S.
Courtois
M.
Wozniak
D.F.
Sambandam
N.
Bernal-Mizrachi
C.
, et al.  . 
PGC-1alpha deficiency causes multi-system energy metabolic derangements: muscle dysfunction, abnormal weight control and hepatic steatosis
PLoS Biol.
 , 
2005
, vol. 
3
 pg. 
e101
 
22
Mitra
R.
Nogee
D.P.
Zechner
J.F.
Yea
K.
Gierasch
C.M.
Kovacs
A.
Medeiros
D.M.
Kelly
D.P.
Duncan
J.G.
The transcriptional coactivators, PGC-1alpha and beta, cooperate to maintain cardiac mitochondrial function during the early stages of insulin resistance
J. Mol. Cell Cardiol.
 , 
2012
, vol. 
52
 (pg. 
701
-
710
)
23
Seth
A.
Steel
J.H.
Nichol
D.
Pocock
V.
Kumaran
M.K.
Fritah
A.
Mobberley
M.
Ryder
T.A.
Rowlerson
A.
Scott
J.
, et al.  . 
The transcriptional corepressor RIP140 regulates oxidative metabolism in skeletal muscle
Cell Metab.
 , 
2007
, vol. 
6
 (pg. 
236
-
245
)
24
Gardiner
K.
Transcriptional dysregulation in Down syndrome: predictions for altered protein complex stoichiometries and post-translational modifications, and consequences for learning/behavior genes ELK, CREB, and the estrogen and glucocorticoid receptors
Behav. Genet.
 , 
2006
, vol. 
36
 (pg. 
439
-
453
)
25
Arron
J.R.
Winslow
M.M.
Polleri
A.
Chang
C.P.
Wu
H.
Gao
X.
Neilson
J.R.
Chen
L.
Heit
J.J.
Kim
S.K.
, et al.  . 
NFAT dysregulation by increased dosage of DSCR1 and DYRK1A on chromosome 21
Nature
 , 
2006
, vol. 
441
 (pg. 
595
-
600
)
26
O'Leary
D.A.
Pritchard
M.A.
Xu
D.
Kola
I.
Hertzog
P.J.
Ristevski
S.
Tissue-specific overexpression of the HSA21 gene GABPalpha: implications for DS
Biochim. Biophys. Acta
 , 
2004
, vol. 
1739
 (pg. 
81
-
87
)
27
Sturgeon
X.
Le
T.
Ahmed
M.M.
Gardiner
K.J.
Pathways to cognitive deficits in Down syndrome
Prog. Brain Res.
 , 
2012
, vol. 
197
 (pg. 
73
-
100
)
28
De Cegli
R.
Romito
A.
Iacobacci
S.
Mao
L.
Lauria
M.
Fedele
A.O.
Klose
J.
Borel
C.
Descombes
P.
Antonarakis
S.E.
, et al.  . 
A mouse embryonic stem cell bank for inducible overexpression of human chromosome 21 genes
Genome Biol.
 , 
2010
, vol. 
11
 pg. 
R64
 
29
Powelka
A.M.
Seth
A.
Virbasius
J.V.
Kiskinis
E.
Nicoloro
S.M.
Guilherme
A.
Tang
X.
Straubhaar
J.
Cherniack
A.D.
Parker
M.G.
, et al.  . 
Suppression of oxidative metabolism and mitochondrial biogenesis by the transcriptional corepressor RIP140 in mouse adipocytes
J. Clin. Invest.
 , 
2006
, vol. 
116
 (pg. 
125
-
136
)
30
Schreiber
S.N.
Emter
R.
Hock
M.B.
Knutti
D.
Cardenas
J.
Podvinec
M.
Oakeley
E.J.
Kralli
A.
The estrogen-related receptor alpha (ERRalpha) functions in PPARgamma coactivator 1alpha (PGC-1alpha)-induced mitochondrial biogenesis
Proc. Natl Acad. Sci. USA
 , 
2004
, vol. 
101
 (pg. 
6472
-
6477
)
31
Patergnani
S.
Suski
J.M.
Agnoletto
C.
Bononi
A.
Bonora
M.
De Marchi
E.
Giorgi
C.
Marchi
S.
Missiroli
S.
Poletti
F.
, et al.  . 
Calcium signaling around Mitochondria Associated Membranes (MAMs)
Cell Commun. Signal
 , 
2011
, vol. 
9
 pg. 
19
 
32
Bonora
M.
Giorgi
C.
Bononi
A.
Marchi
S.
Patergnani
S.
Rimessi
A.
Rizzuto
R.
Pinton
P.
Subcellular calcium measurements in mammalian cells using jellyfish photoprotein aequorin-based probes
Nat. Protoc.
 , 
2013
, vol. 
8
 (pg. 
2105
-
2118
)
33
Bonora
M.
Patergnani
S.
Rimessi
A.
De Marchi
E.
Suski
J.M.
Bononi
A.
Giorgi
C.
Marchi
S.
Missiroli
S.
Poletti
F.
, et al.  . 
ATP synthesis and storage
Purinergic Signal.
 , 
2012
, vol. 
8
 (pg. 
343
-
357
)
34
Fritah
A.
Steel
J.H.
Nichol
D.
Parker
N.
Williams
S.
Price
A.
Strauss
L.
Ryder
T.A.
Mobberley
M.A.
Poutanen
M.
, et al.  . 
Elevated expression of the metabolic regulator receptor-interacting protein 140 results in cardiac hypertrophy and impaired cardiac function
Cardiovasc. Res.
 , 
2010
, vol. 
86
 (pg. 
443
-
451
)
35
Rytinki
M.M.
Palvimo
J.J.
SUMOylation modulates the transcription repressor function of RIP140
J. Biol. Chem.
 , 
2008
, vol. 
283
 (pg. 
11586
-
11595
)
36
Scarpulla
R.C.
Vega
R.B.
Kelly
D.P.
Transcriptional integration of mitochondrial biogenesis
Trends Endocrinol. Metab.
 , 
2012
, vol. 
23
 (pg. 
459
-
466
)
37
Fritah
A.
Steel
J.H.
Parker
N.
Nikolopoulou
E.
Christian
M.
Carling
D.
Parker
M.G.
Absence of RIP140 reveals a pathway regulating glut4-dependent glucose uptake in oxidative skeletal muscle through UCP1-mediated activation of AMPK
PLoS One
 , 
2012
, vol. 
7
 pg. 
e32520
 
38
Petrozzi
L.
Ricci
G.
Giglioli
N.J.
Siciliano
G.
Mancuso
M.
Mitochondria and neurodegeneration
Biosci. Rep.
 , 
2007
, vol. 
27
 (pg. 
87
-
104
)
39
Kanehisa
M.
Goto
S.
KEGG: kyoto encyclopedia of genes and genomes
Nucleic Acids Res.
 , 
2000
, vol. 
28
 (pg. 
27
-
30
)
40
de la Torre
R.
Dierssen
M.
Therapeutic approaches in the improvement of cognitive performance in Down syndrome: past, present, and future
Prog. Brain Res.
 , 
2012
, vol. 
197
 (pg. 
1
-
14
)
41
Costa
A.C.
Scott-McKean
J.J.
Prospects for improving brain function in individuals with Down syndrome
CNS Drugs
 , 
2013
, vol. 
27
 (pg. 
679
-
702
)
42
Rodgers
J.T.
Lerin
C.
Haas
W.
Gygi
S.P.
Spiegelman
B.M.
Puigserver
P.
Nutrient control of glucose homeostasis through a complex of PGC-1alpha and SIRT1
Nature
 , 
2005
, vol. 
434
 (pg. 
113
-
118
)
43
Dong
W.
Gao
D.
Zhang
X.
Mitochondria biogenesis induced by resveratrol against brain ischemic stroke
Med. Hypotheses
 , 
2007
, vol. 
69
 (pg. 
700
-
701
)
44
Lagouge
M.
Argmann
C.
Gerhart-Hines
Z.
Meziane
H.
Lerin
C.
Daussin
F.
Messadeq
N.
Milne
J.
Lambert
P.
Elliott
P.
, et al.  . 
Resveratrol improves mitochondrial function and protects against metabolic disease by activating SIRT1 and PGC-1alpha
Cell
 , 
2006
, vol. 
127
 (pg. 
1109
-
1122
)
45
Jager
S.
Handschin
C.
St-Pierre
J.
Spiegelman
B.M.
AMP-activated protein kinase (AMPK) action in skeletal muscle via direct phosphorylation of PGC-1alpha
Proc. Natl Acad. Sci. USA
 , 
2007
, vol. 
104
 (pg. 
12017
-
12022
)
46
Bastin
J.
Aubey
F.
Rotig
A.
Munnich
A.
Djouadi
F.
Activation of peroxisome proliferator-activated receptor pathway stimulates the mitochondrial respiratory chain and can correct deficiencies in patients’ cells lacking its components
J. Clin. Endocrinol. Metab.
 , 
2008
, vol. 
93
 (pg. 
1433
-
1441
)
47
Nicolakakis
N.
Aboulkassim
T.
Ongali
B.
Lecrux
C.
Fernandes
P.
Rosa-Neto
P.
Tong
X.K.
Hamel
E.
Complete rescue of cerebrovascular function in aged Alzheimer's disease transgenic mice by antioxidants and pioglitazone, a peroxisome proliferator-activated receptor gamma agonist
J. Neurosci.
 , 
2008
, vol. 
28
 (pg. 
9287
-
9296
)
48
Escribano
L.
Simon
A.M.
Perez-Mediavilla
A.
Salazar-Colocho
P.
Del Rio
J.
Frechilla
D.
Rosiglitazone reverses memory decline and hippocampal glucocorticoid receptor down-regulation in an Alzheimer's disease mouse model
Biochem. Biophys. Res. Commun.
 , 
2009
, vol. 
379
 (pg. 
406
-
410
)
49
Johri
A.
Calingasan
N.Y.
Hennessey
T.M.
Sharma
A.
Yang
L.
Wille
E.
Chandra
A.
Beal
M.F.
Pharmacologic activation of mitochondrial biogenesis exerts widespread beneficial effects in a transgenic mouse model of Huntington's disease
Hum. Mol. Genet.
 , 
2012
, vol. 
21
 (pg. 
1124
-
1137
)
50
Sauerbeck
A.
Gao
J.
Readnower
R.
Liu
M.
Pauly
J.R.
Bing
G.
Sullivan
P.G.
Pioglitazone attenuates mitochondrial dysfunction, cognitive impairment, cortical tissue loss, and inflammation following traumatic brain injury
Exp. Neurol.
 , 
2011
, vol. 
227
 (pg. 
128
-
135
)
51
Yamaguchi
S.
Li
H.
Purevsuren
J.
Yamada
K.
Furui
M.
Takahashi
T.
Mushimoto
Y.
Kobayashi
H.
Hasegawa
Y.
Taketani
T.
, et al.  . 
Bezafibrate can be a new treatment option for mitochondrial fatty acid oxidation disorders: evaluation by in vitro probe acylcarnitine assay
Mol. Genet. Metab.
 , 
2012
, vol. 
107
 (pg. 
87
-
91
)
52
Watson
G.S.
Cholerton
B.A.
Reger
M.A.
Baker
L.D.
Plymate
S.R.
Asthana
S.
Fishel
M.A.
Kulstad
J.J.
Green
P.S.
Cook
D.G.
, et al.  . 
Preserved cognition in patients with early Alzheimer disease and amnestic mild cognitive impairment during treatment with rosiglitazone: a preliminary study
Am. J. Geriatr. Psychiatry
 , 
2005
, vol. 
13
 (pg. 
950
-
958
)
53
Sato
T.
Hanyu
H.
Hirao
K.
Kanetaka
H.
Sakurai
H.
Iwamoto
T.
Efficacy of PPAR-gamma agonist pioglitazone in mild Alzheimer disease
Neurobiol. Aging
 , 
2011
, vol. 
32
 (pg. 
1626
-
1633
)
54
Marciano
D.P.
Chang
M.R.
Corzo
C.A.
Goswami
D.
Lam
V.Q.
Pascal
B.D.
Griffin
P.R.
The therapeutic potential of nuclear receptor modulators for treatment of metabolic disorders: PPARgamma, RORs, and Rev-erbs
Cell Metab.
 , 
2014
, vol. 
19
 (pg. 
193
-
208
)
55
Zhang
B.
Kirov
S.
Snoddy
J.
WebGestalt: an integrated system for exploring gene sets in various biological contexts
Nucleic Acids Res.
 , 
2005
, vol. 
33
 (pg. 
W741
-
W748
)
56
Wang
J.
Duncan
D.
Shi
Z.
Zhang
B.
WEB-based GEne SeT AnaLysis Toolkit (WebGestalt): update 2013
Nucleic Acids Res.
 , 
2013
, vol. 
41
 (pg. 
W77
-
W83
)
57
Cali
G.
Gentile
F.
Mogavero
S.
Pallante
P.
Nitsch
R.
Ciancia
G.
Ferraro
A.
Fusco
A.
Nitsch
L.
CDH16/Ksp-cadherin is expressed in the developing thyroid gland and is strongly down-regulated in thyroid carcinomas
Endocrinology
 , 
2012
, vol. 
153
 (pg. 
522
-
534
)
58
Cali
G.
Zannini
M.
Rubini
P.
Tacchetti
C.
D'Andrea
B.
Affuso
A.
Wintermantel
T.
Boussadia
O.
Terracciano
D.
Silberschmidt
D.
, et al.  . 
Conditional inactivation of the E-cadherin gene in thyroid follicular cells affects gland development but does not impair junction formation
Endocrinology
 , 
2007
, vol. 
148
 (pg. 
2737
-
2746
)
59
Jouaville
L.S.
Pinton
P.
Bastianutto
C.
Rutter
G.A.
Rizzuto
R.
Regulation of mitochondrial ATP synthesis by calcium: evidence for a long-term metabolic priming
Proc. Natl Acad. Sci. USA
 , 
1999
, vol. 
96
 (pg. 
13807
-
13812
)
60
Porcelli
A.M.
Pinton
P.
Ainscow
E.K.
Chiesa
A.
Rugolo
M.
Rutter
G.A.
Rizzuto
R.
Targeting of reporter molecules to mitochondria to measure calcium, ATP, and pH
Methods Cell Biol.
 , 
2001
, vol. 
65
 (pg. 
353
-
380
)

Author notes

The authors wish to be known that, in their opinion, the first three authors should be regarded as joint First Authors.

Supplementary data