Abstract
The AMP-activated protein kinase (AMPK) is a molecular sensor to help maintain cellular energy homeostasis. AMPK is a heterotrimeric complex and its enzymatic catalytic subunit includes two isoforms: α1 and α2. Dysregulation of AMPK signaling is linked to neuronal diseases characterized with cognitive impairments. Emerging evidence also suggest isoform-specific roles of AMPK in the brain. AMPK regulates protein synthesis, which is critical for memory formation and neuronal plasticity. However, the consequence of altering AMPK activity on the translation of specific proteins in the brain is unknown. Here we use unbiased mass spectrometry-based proteomics approach to analyze protein profile alterations in hippocampus and prefrontal cortex of transgenic mice in which the genes for the two AMPKα isoforms are conditionally deleted. The study revealed identities of proteins whose expression is sensitive to suppression of AMPKα1 and/or α2 isoform. These data may serve as a basis for future in-depth study. Elucidation of the functional relevance of the alteration of specific proteins could provide insights into identification of novel therapeutic targets for neuronal disorders characterized with AMPK signaling dysregulation and impaired cellular energy metabolism.
Keywords: AMPK, proteomics, protein synthesis, signal transduction, hippocampus
INTRODUCTION
Brain energy metabolism is central to neuronal plasticity and cognitive function (J.Magistretti & Allaman, 2015). Dysregulation of neuronal energy metabolism has been linked to multiple neurological disorders characterized with cognitive impairments such as Alzheimer’s disease (AD) (Frisardi et al., 2010; Lin & Beal, 2006). AMP-activated protein kinase (AMPK) is a sensor at the molecular level in response to energy and nutrient status to maintain energy homeostasis. Ubiquitously expressed in all eukaryotes, AMPK exists as a heterotrimeric complex consisting of catalytic α subunit, and regulatory β and γ subunits. For mammals, each subunit has multiple isoforms encoded by distinct genes. The catalytic subunit has two isoforms AMPKα1 and α2, which are encoded by PRKAA1 on chromosome 5 and PRKAA2 on chromosome 1, respectively (Hardie, Ross, & Hawley, 2012; Steinberg & Kemp, 2009; Viollet et al., 2010). Isoform-specific roles of AMPK subunits in the central nervous system remain unclear. Recently emerging evidence reveals potential roles of AMPKα isoforms in brain physiology and pathophysiology (Wang, Zimmermann, & Ma, 2019). For instance, suppression of AMPKα2 but not AMPKα1 isoform in mice results in cognitive impairment and synaptic failure (W. Yang, Zhou, Zimmermann, & Ma, 2020). Moreover, expression of AMPKα isoforms is dysregulated in brains of AD patients and AD animal models, and genetic restoration of brain AMPKα1 level alleviates multiple aspects of pathophysiology in AD model mice (Zimmermann et al., 2020).
One of the established downstream effects of AMPK is regulation of protein synthesis or mRNA translation. Substantial research demonstrates that de novo protein synthesis is required for maintenance of long-term forms of synaptic plasticity and memory (Alberini, 2008; Klann & Dever, 2004). AMPK can regulate proteins synthesis through at least two canonical mechanisms: the eukaryotic elongation factor 2 kinase (eEF2K) and the mammalian target of rapamycin complex 1 (mTORC1) signaling cascades. Briefly, AMPK phosphorylates and activates eEF2K, leading to phosphorylation of translational factor eEF2 and consequently inhibition of general protein synthesis. Additionally, AMPK inhibits (directly or indirectly) the mTORC1 signaling pathway, activation of which promotes synthesis of translational apparatus and cap-dependent translation initiation (Steinberg & Kemp, 2009). A most recent study also indicates a link between AMPKα2 and protein synthesis through mRNA translational initiation factor eIF2 and its kinase PERK (W. Yang et al., 2020). However, how AMPK isoform modulation influences translation of specific proteins in the brain is unknown.
In the current study, we take advantage of two lines of transgenic mice that we recently developed in which the genes for the two AMPKα isoforms are conditionally knocked out in forebrain areas (W. Yang et al., 2020). Using unbiased mass spectrometry-based proteomics approach, the study is aimed to identify proteins whose expression pattern is sensitive to suppression of AMPKα isoforms in the brain.
RESULTS AND DISCUSSION
As described, mice expressing a forebrain-specific Cre recombinase (Camk2a-cre) (Trinh et al., 2012) were bred with transgenic mice harboring either loxP-flanked Prkaa1 or Prkaa2 (genes encoding AMPKα1 or α2) to generate conditional AMPKα1 or α2 isoform knockout mice: AMPKα1 cKO or AMPKα2 cKO (Figure 1a). We previously demonstrated in these mutant mice that expression of the two AMPKα isoforms are selectively reduced i.e. significantly reduced in hippocampus and prefrontal cortex, but unaltered in cerebellum or peripheral organs (e.g. liver, heart) (W. Yang et al., 2020). Importantly, we did not observe compensated alterations in protein expression levels of the other isoform in the mutant mice (Figure 1b).
FIGURE 1.
Characterization of protein synthesis in conditionally AMPKα knockout mice. (a) Schematic illustration of generation of AMPKα1 cKO and α2 cKO mice. The presence of the Cre, Prkaa1loxP, and Prkaa2loxP transgene was determined using PCR-specific primers. (b) Biochemical analysis confirmed corresponding reduction of AMPKα1 or AMPKα2 protein levels in area CA1 of hippocampus. Representative Western blot gels and cumulative data for quantification presented in bar graphs are shown (n=5, 4, 4 for Cre +/−, α1 cKO and α2 cKO respectively, *p<0.05, **p<0.01, ***p<0.0005, ****p<0.0001, one-way ANOVA, followed by Tukey’s post hoc test). (c) Representative Western blot images and quantification for puromycin staining from the SUnSET assay. n=3 mice. One-way ANOVA. (d) Heat map showed the patterns of protein expression in Hippocampus. The number of peptide spectrum matches was normalized by that of the control (Cre +/−) group. The proteins (accessary on the left) with significant changes compared to control were selected for heat map plotting using R software (https://www.r-project.org/).
To assess effects of AMPKα isoform suppression on overall de novo protein synthesis, we performed SUnSET experiments with hippocampal slices (Beckelman et al., 2019). Hippocampal CA1 areas were further micro-dissected for biochemical assay. In agreement with previous findings in whole hippocampus (W. Yang et al., 2020), we did not detect significant alterations of puromycin incorporation (indicator of de novo protein synthesis) in either AMPKα1 cKO or α2 cKO mice, compared to the control group (Figure 1c).
We next performed mass spectrometry-based proteomic analysis and identified in total 2406 proteins (average) in hippocampus (supplementary excel file). Proteins that are significantly regulated (up- or down-regulation) upon suppression of AMPKα isoforms were entered into the “PANTHER” (Protein ANalysis THrough Evolutionary Relationships) protein classification system (http://www.pantherdb.org/) to uncover their functional classification based on broad categories. A heat map is generated to summarize all the altered proteins (Figure 1d). The pattern shown in the heat map indicates alteration of protein expression in both directions (up or down) upon AMPKα isoform suppression, consistent with the results from the SUnSET experiment showing unaltered overall protein synthesis.
Compared to the control (Cre+/−) group, there were 16 up-regulated proteins in AMPKα1 cKO, and 10 proteins were upregulated in AMPKα2 cKO mice (Table 1 and 2, Figure 2). Functionally, they all belong to four broad categories: binding, catalytic activity, structural molecule activity, and transporter activity. Moreover, there are only 2 proteins whose synthesis was commonly increased in both AMPKα1 cKO and α2 cKO mice: protein kinase C/casein kinase substrate in neurons protein 2 (Q9WVE8), EF-hand domain-containing protein D2 (Q9D8Y0). Notably, one of the top proteins that was uniquely up-regulated in hippocampi of the AMPKα1 cKO mice is NEDD8-activating enzyme E1 regulatory subunit (Q8VBW6), which can bind to the amyloid precursor protein (APP) and is considered to play a role in AD pathogenesis (Chen, McPhie, Hirschberg, & Neve, 2000; Gong & Yeh, 1999). Such finding is interesting in light of our recent studies suggesting a critical role of AMPKα1 dysregulation in AD pathophysiology (Zimmermann et al., 2020). It is also worth mentioning another upregulated protein in the AMPKα1 cKO condition: Polyadenylate-binding protein 1 (PABP1, P29341). PABP1 is one of the “TOP” (terminal oligopyrimidine) class mRNA-encoded proteins, which includes components of translational machinery and may play important roles in synaptic plasticity and cognitive function (Tsokas, Ma, Iyengar, Landau, & Blitzer, 2007; Wells et al., 2001).
Table 1.
List of up-regulated proteins in hippocampus of AMPKα1 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| Neutral amino acid transporter A | O35874 | 3.333 |
| Kininogen-1 | O08677 | 3 |
| NEDD8-activating enzyme E1 regulatory subunit | Q8VBW6 | 3 |
| Protein kinase C and casein kinase substrate in neurons protein 2+ | Q9WVE8 | 2.5 |
| Polyadenylate-binding protein 1 | P29341 | 2.333 |
| Trifunctional enzyme subunit alpha, mitochondrial | Q8BMS1 | 2.25 |
| 3-ketoacyl-CoA thiolase, mitochondrial | Q8BWT1 | 2.032 |
| EF-hand domain-containing protein D2+ | Q9D8Y0 | 1.909 |
| Vinculin | Q64727 | 1.833 |
| Neuronal calcium sensor 1 | Q8BNY6 | 1.767 |
| Cathepsin B | P10605 | 1.722 |
| Transcriptional activator protein Pur-beta | O35295 | 1.7 |
| Mitochondrial carrier homolog 2 | Q791V5 | 1.644 |
| Protein phosphatase 1 regulatory subunit 7 | Q3UM45 | 1.643 |
| 4-trimethylaminobutyraldehyde dehydrogenase | Q9JLJ2 | 1.556 |
| Glycerol-3-phosphate dehydrogenase 1-like protein | Q3ULJ0 | 1.548 |
Up-regulated also in AMPKα2 cKO mice
Table 2.
List of up-regulated proteins in hippocampus of AMPKα2 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| Plasminogen activator inhibitor 1 RNA-binding protein | Q9CY58 | 3.833 |
| Proteasome subunit alpha type-7 | Q9Z2U0 | 2.833 |
| Amine oxidase [flavin-containing] B | Q8BW75 | 2.167 |
| Protein kinase C and casein kinase substrate in neurons protein 2+ | Q9WVE8 | 2.167 |
| Voltage-gated potassium channel subunit beta-2 | P62482 | 1.81 |
| Secretory carrier-associated membrane protein 3 | O35609 | 1.667 |
| Nuclear migration protein nudC | O35685 | 1.667 |
| Sorting and assembly machinery component 50 homolog | Q8BGH2 | 1.644 |
| EF-hand domain-containing protein D2+ | Q9D8Y0 | 1.589 |
| NADH dehydrogenase [ubiquinone] flavoprotein 1, mitochondrial | Q91YT0 | 1.556 |
Up-regulated also in AMPKα1 cKO mice
FIGURE 2.
Identification of proteins that were up-regulated in hippocampus of AMPKα1 cKO and α2 cKO mice. (a) Pie charts showed functional classification of significantly regulated proteins identified from proteomics analysis of AMPKα1 cKO mice. (b) Pie charts showed functional classification of significantly regulated proteins identified from proteomics analysis of AMPKα2 cKO mice. Functional classification based on “PANTHER” protein classification system (http://www.pantherdb.org/) was listed below the pie charts. GO: Gene Ontology. n=3 mice. (c) Venn diagram showed numbers of proteins that were up-regulated in hippocampus of AMPKα1 cKO or α2 cKO mice, compared to Cre+/− group. Two proteins (as indicated with overlap) were commonly up-regulated: protein kinase C/casein kinase substrate in neurons protein 2 (Q9WVE8), EF-hand domain-containing protein D2 (Q9D8Y0).
Compared to the up-regulated proteins, there are more hippocampal proteins (in number) that are down-regulated with AMPKα isoform suppression: 11 in AMPKα1 cKO, and 21 in AMPK α2 cKO mice (Table 3 and 4, Figure 3). Proteins whose synthesis was down-regulated in AMPKα1 cKO belong to three categories: binding, catalytic activity and translation regulator activity. In contrast, proteins whose synthesis was downregulated upon AMPKα2 inhibition belong to five categories: binding, catalytic activity, structural molecule activity, translation regulator activity, and transporter activity. Down-regulation of 4 proteins is overlapped in AMPKα1 cKO and α2 cKO mice: 116 kDa U5 small nuclear ribonucleoprotein component (O08810), Cullin-3 (Q9JLV5), Sorcin (Q6P069), and Monoglyceride lipase (Q35678). One protein of interest from the neuroscience perspective is Sorcin (Soluble resistance-related calcium binding protein), which is a calcium-binding protein highly expressed in both brain and heart (Ilari et al., 2015; Mao et al., 2020). Previous studies indicate that Sorcin increases Ca2+ accumulation in the endoplasmic (ER) and mitochondria. Further, Sorcin functions to prevent ER stress and suppression of Sorcin facilitates apoptosis (Hu et al., 2013; Ilari et al., 2015). Thus, suppression of either AMPKα isoforms may cause disruption of Ca2+ homeostasis (via Sorcin), and consequently elevation of ER stress and/or apoptosis, which could contribute to the behavioral and synaptic plasticity phenotypes in the AMPKα1 cKO and α2 cKO mice (W. Yang et al., 2020).
Table 3.
List of down-regulated proteins in hippocampus of AMPKα1 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| 116 kDa U5 small nuclear ribonucleoprotein component+ | O08810 | 0.167 |
| Fatty-acid amide hydrolase 1 | O08914 | 0.167 |
| Core histone macro-H2A.1 | Q9QZQ8 | 0.167 |
| Prefoldin subunit 5 | Q9WU28 | 0.167 |
| Cullin-3+ | Q9JLV5 | 0.250 |
| Hepatocyte growth factor-regulated tyrosine kinase substrate | Q99LI8 | 0.250 |
| NADH dehydrogenase [ubiquinone] 1 beta subcomplex subunit 5, mitochondrial | Q9CQH3 | 0.296 |
| Sorcin+ | Q6P069 | 0.533 |
| Rap guanine nucleotide exchange factor 2 | Q8CHG7 | 0.536 |
| Monoglyceride lipase+ | O35678 | 0.548 |
| MAGUK p55 subfamily member 6 | Q9JLB0 | 0.611 |
Down-regulated also in AMPKα2 cKO mice
Table 4.
List of down-regulated proteins in hippocampus of AMPKα2 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| 116 kDa U5 small nuclear ribonucleoprotein component+ | O08810 | 0.167 |
| Cholesterol 24-hydroxylase | Q9WVK8 | 0.167 |
| Heterogeneous nuclear ribonucleoprotein U-like protein 1 | Q8VDM6 | 0.222 |
| Unconventional myosin-Id | Q5SYD0 | 0.222 |
| Emerin | O08579 | 0.222 |
| Guanine nucleotide-binding protein G(I)/G(S)/G(O) subunit gamma-4 | P50153 | 0.250 |
| Small nuclear ribonucleoprotein Sm D2 | P62317 | 0.278 |
| Branched-chain-amino-acid aminotransferase, cytosolic | P24288 | 0.278 |
| Calcium uptake protein 3, mitochondrial | Q9CTY5 | 0.333 |
| Cullin-3+ | Q9JLV5 | 0.417 |
| GTPase HRas | Q61411 | 0.418 |
| Sorcin+ | Q6P069 | 0.467 |
| 60S ribosomal protein L27 | P61358 | 0.524 |
| Myotubularin-related protein 5 | Q6ZPE2 | 0.556 |
| Adenosine kinase | P55264 | 0.567 |
| Calcium uniporter protein, mitochondrial | Q3UMR5 | 0.600 |
| T-complex protein 1 subunit alpha | P11983 | 0.600 |
| Vesicle-associated membrane protein 2 | P63044 | 0.608 |
| Septin-6 | Q9R1T4 | 0.609 |
| Neural Wiskott-Aldrich syndrome protein | Q91YD9 | 0.619 |
| Monoglyceride lipase+ | O35678 | 0.643 |
Down-regulated also in AMPKα1 cKO mice
FIGURE 3.
Identification of proteins that were down-regulated in hippocampus of AMPKα1 cKO and α2 cKO mice. (a) Pie charts showed functional classification of significantly regulated proteins identified from proteomics analysis of AMPKα1 cKO mice. (b) Pie charts showed functional classification of significantly regulated proteins identified from proteomics analysis of AMPKα2 cKO mice. n=3 mice. (c) Venn diagram showed numbers of proteins that were down-regulated in hippocampus of AMPKα1 cKO or α2 cKO mice, compared to Cre+/− group. Four proteins (as indicated with overlap) were down-regulated in both AMPKα1 cKO and α2 cKO mice: 116 kDa U5 small nuclear ribonucleoprotein component (O08810), Cullin-3 (Q9JLV5), Sorcin (Q6P069), Monoglyceride lipase (Q35678)
It is worth mentioning that among the top regulated proteins in both mutant mice, several proteins (e.g. Proteasome subunit alpha type-7 and Cullin-3) are known to be involved in the process of protein degradation/clearance including the lysosomal proteolysis and ubiquitin-proteasome pathway, dysregulation of both pathways is indicated in multiple neurodegenerative diseases (Boland et al., 2018; Nixon, 2020). Degradation of specific proteins mediated through the aforementioned pathways in AMPKα mutant mice and the functional relevance in the central nervous system would be an interesting topic for future study.
Moreover, we performed Quantitative Real-time PCR analyses to assess the effects of AMPKα isoform reduction on mRNAs of 5 selected proteins mentioned above: Polyadenylate-binding protein 1 (pabpn1), Sorcin (sri), Cullin 3 (cul3), Monoglyceride lipase (mgll), and 116 KDa U5 Small Nuclear Ribonucleoprotein Component (eftud2). Interestingly, we observed alterations of hippocampal mRNA levels of 2 proteins: Monoglyceride lipase and 116 KDa U5 Small Nuclear Ribonucleoprotein Component upon genetic suppression of AMPKα isoform (Figure S1). Such findings indicate profound effects of brain AMPKα isoform reduction at the gene expression level.
In parallel, we analyzed protein expression alteration in prefrontal cortex with AMPKα isoform suppression, and the results on PFC experiments are included in the supplementary data (Table 5–8, Figure S2–S4). In total, there were 2455 proteins identified in PFC (supplementary excel file), and the number is similar to the experiments in hippocampus (2406 proteins). It is interesting to compare the protein profiling alteration between hippocampus and PFC upon AMPKα isoform suppression. First, compared to the results from hippocampus assays, there are overall more proteins identified in PFC whose synthesis is significantly altered with knockout of either AMPKα1 or α2 isoform. For instance, there are 40 proteins whose synthesis is significantly reduced in AMPKα2 cKO mice (Table 8, Figure S3c). Moreover, the top-regulated (up- or down-regulated) proteins in PFC are different from those in hippocampus of the same mutant mice (for details, see Table 5–8, Figure S2–S4). We speculate that disruption of AMPKα isoform homeostasis may exert more profound and distinct effects on mRNA translation (or protein degradation/clearance) in PFC. The indication of such changes (e.g. effects on hippocampus- or PFC-dependent cognitive function) would be interesting topics for future studies. Second, similar to the results from hippocampus, most altered proteins in PFC are uniquely regulated by suppression of either AMPKα1 or α2 (but not both). The results are consistent with recent studies indicating isoform-specific roles of AMPKα in neuronal plasticity, cognitive function, and neurodegenerative diseases (W. Yang et al., 2020; Zimmermann et al., 2020). As discussed above, AMPK plays a central role in regulating cellular energy homeostasis, disruption of which is indicated many neuronal disorders. Therefore findings from the current study can be used as a basis for future in-depth research. Elucidation of how changes of specific proteins (in response to AMPKα isoform dysregulation) in PFC or hippocampus contribute brain physiology and pathophysiology would provide insights into identification of potential therapeutic targets for neuronal diseases characterized with AMPK dysregulation and impaired cellular energy metabolism.
Table 5.
List of up-regulated proteins in PFC of AMPKα1 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| Prolyl endopeptidase | Q9QUR6 | 4.5 |
| Long-chain fatty acid transport protein 4 | Q91VE0 | 3.333 |
| Unconventional myosin-XVIIIa+ | Q9JMH9 | 2.5 |
| Arf-GAP with GTPase, ANK repeat and PH domain-containing protein 2+ | Q3UHD9 | 2.333 |
| Transmembrane emp24 domain-containing protein 10 | Q9D1D4 | 2.167 |
| Rho-related GTP-binding protein RhoB+ | P62746 | 2.156 |
| CUGBP Elav-like family member 2 | Q9Z0H4 | 1.833 |
| IQ motif and SEC7 domain-containing protein 1 | Q8R0S2 | 1.833 |
| Programmed cell death 6-interacting protein+ | Q9WU78 | 1.778 |
| 60S ribosomal protein L14 | Q9CR57 | 1.778 |
| 60S ribosomal protein L30 | P62889 | 1.733 |
| Lipid phosphate phosphohydrolase 3+ | Q99JY8 | 1.733 |
| Sodium/potassium-transporting ATPase subunit beta-3 | P97370 | 1.722 |
| Calcium/calmodulin-dependent protein kinase type 1D | Q8BW96 | 1.667 |
| AFG3-like protein 2 | Q8JZQ2 | 1.667 |
| Cell adhesion molecule 4+ | Q8R464 | 1.533 |
| AP-3 complex subunit beta-2 | Q9JME5 | 1.5 |
Up-regulated also in AMPKα2 cKO mice
Table 8.
List of down-regulated proteins in PFC of AMPKα2 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| 40S ribosomal protein S2 | 0.259 | |
| Glia maturation factor beta | P25444 | 0.267 |
| Histone H1.3+ | Q9CQI3 | 0.285 |
| Ribonuclease inhibitor+ | P43277 | 0.289 |
| Serine/threonine-protein phosphatase PGAM5, mitochondrial | Q91VI7 | 0.289 |
| 5'-nucleotidase domain-containing protein 3 | Q8BX10 | 0.306 |
| Synaptogyrin-3+ | Q3UHB1 | 0.308 |
| Destrin | Q8R191 | 0.322 |
| Protein canopy homolog 2+ | Q9R0P5 | 0.333 |
| Kinesin-like protein KIF2A | Q9QXT0 | 0.333 |
| Glutaredoxin-3+ | P28740 | 0.350 |
| Cation-dependent mannose-6-phosphate receptor | Q9CQM9 | 0.356 |
| Aldose reductase+ | P24668 | 0.389 |
| 60S ribosomal protein L4 | P45376 | 0.394 |
| Clathrin interactor 1 | Q9D8E6 | 0.400 |
| Neurogranin | Q99KN9 | 0.400 |
| Stromal membrane-associated protein 1 | P60761 | 0.433 |
| Trifunctional enzyme subunit beta, mitochondrial | Q91VZ6 | 0.442 |
| 60S ribosomal protein L23a | Q99JY0 | 0.444 |
| Protein SOGA3 | P62751 | 0.444 |
| Sepiapterin reductase | Q6NZL0 | 0.483 |
| Eukaryotic initiation factor 4A-I | Q64105 | 0.488 |
| Spliceosome RNA helicase Ddx39b | P60843 | 0.511 |
| UMP-CMP kinase | Q9Z1N5 | 0.524 |
| Neurochondrin | Q9DBP5 | 0.525 |
| Succinate dehydrogenase [ubiquinone] iron-sulfur subunit, mitochondrial | Q9Z0E0 | 0.539 |
| Histone H1.4+ | Q9CQA3 | 0.558 |
| 39S ribosomal protein L12, mitochondrial | P43274 | 0.567 |
| Adenylate kinase 4, mitochondrial | Q9DB15 | 0.567 |
| Eukaryotic initiation factor 4A-II | Q9WUR9 | 0.572 |
| Peptidyl-prolyl cis-trans isomerase A | P10630 | 0.584 |
| Growth factor receptor-bound protein 2 | P17742 | 0.593 |
| Vacuolar protein sorting-associated protein 35 | Q60631 | 0.593 |
| DNA damage-binding protein 1+ | Q9EQH3 | 0.600 |
| WD repeat-containing protein 1 | Q3U1J4 | 0.618 |
| Thioredoxin-like protein 1 | O88342 | 0.625 |
| Thioredoxin-dependent peroxide reductase, mitochondrial | Q8CDN6 | 0.627 |
| Pyruvate dehydrogenase E1 component subunit beta, mitochondrial | P20108 | 0.646 |
| D-3-phosphoglycerate dehydrogenase | Q9D051 | 0.656 |
| ATP-citrate synthase | Q61753 | 0.658 |
| Q91V92 |
Down-regulated also in AMPKα1 cKO mice
DETAILED METHODS
Mice of 3–6 month old were used for all experiments. For Western blotting, tissues were removed from appropriate structures and flash frozen on dry ice. Membranes were blocked and then probed overnight at 4°C using following primary antibodies: AMPKα1 (1:1000, Abcam, Cat# ab3759), AMPKα2 (1:1000, Abcam, Cat# ab3760), GAPDH (1:10,000, Cell Signaling, Cat# 2118S). For SUnSET protein synthesis assay, acute 400 μm transverse hippocampal slices were prepared as described (W Yang, Zhou, Cavener, Klann, & Ma, 2016). Slices were maintained before experimentation at room temperature for at least 2 h in artificial cerebrospinal fluid (ACSF). Slices were incubated in puromycin (1 μg/ml) for 1 h at 32 °C in bubbling ACSF. Slices were then flash frozen on dry ice, and area CA1 was micro-dissected for Western blot analysis. Puromycin-labeled proteins were identified using the mouse monoclonal antibody 12D10 (1:5000; EMD Millipore, Burlington, MA, catalog MABE343). Protein synthesis levels were determined by analyzing total lane density from 10 kDa to 250 kDa. Densitometric analysis was performed using ImageJ software.
For proteomics analysis, brain tissues were lysed in radioimmunoprecipitation (RIPA) buffer containing protease inhibitor. 100 μg of protein was reduced and alkylated in the presence of 10mM dithiothreitol and 30mM iodoacetamide. Purified protein pellet was obtained from cold acetone precipitation, which was then suspended in 50 mM ammonium bicarbonate. 2 μg of sequencing grade modified trypsin was added and the mixture was incubated at 37°C overnight. Peptides were purified using a C18 spin column and prepared in 5% (v/v) ACN containing 1% (v/v) formic acid for LC-MS/MS analysis. Samples were analyzed on a LC-MS/MS system consisted of a Q Exactive HF Hybrid Quadrupole-Orbitrap Mass Spectrometer (Thermo Scientific, Rockford, IL) and a Dionex Ultimate-3000 nano-UPLC system (Thermo Scientific, Waltham, MA). An Acclaim PepMap 100 (C18, 5 μm, 100 Å, 100 μm x 2 cm) trap column and an Acclaim PepMap RSLC (C18, 2 μm, 100 Å, 75 μm x 50 cm) analytical column were employed for peptide separation. MS spectra were acquired by data dependent scans consisting of MS/MS scans of the twenty most intense ions from the full MS scan with dynamic exclusion option which was 10 seconds. Spectra were searched using Sequest HT algorithm within the Proteome Discoverer v2.1 (Thermo Scientific, Waltham, MA) in combination with the mouse protein FASTA database (16,747 annotated entries, Swiss-Prot, December 2015). Search parameters were as follow; FT-trap instrument, parent mass error tolerance of 10 ppm, fragment mass error tolerance of 0.02 Da (monoisotopic), variable modifications of 16 Da (oxidation) on methionine and fixed modification of 57 Da (carbamidomethylation) on cysteine. The number of peptide spectrum matches was normalized by that of Cre +/− group, and the fold change was calculated as fold change (n) = mean of AMPK α1( or α2) cKO / mean of Cre +/−. The probability p value was also calculated using the Student’s t-test between AMPK α1 cKO and Cre +/− or between AMPK α2 cKO and Cre +/− groups. The proteins with p ≤ 0.08 and n ≥ 1.5 were considered as up regulated, and those with p ≤ 0.08 and n ≤ 0.667 as down regulated. The accessions of the proteins matching the criteria were loaded into the PANTHER classification system (Gene Ontology (GO) Reference Genome Project, http://www.pantherdb.org/). The functional classification pie charts and Venn diagrams were generated by PANTHER and Inkscape software (The Inkscape Project), respectively.
For Real-time PCR experiment, total RNA was extracted from mouse brain tissue using TRIzol (Thermo Fisher Scientific) and subjected to reverse transcription using iScriptTM Reverse Transcription Supermix (Bio-Rad). Quantitative real-time PCR was performed in triplicates with gene-specific primers and SYBR Supermix (Quantabio) in a Bio-Rad CFX96 REAL TIME SYSTEM following manufacturer’s protocols. GADPH was used as internal control.
For statistical analysis of non-proteomics experiments, data were presented as mean ± SEM. For comparisons between two groups, a two-tailed unpaired Student’s t-test was used. For comparisons between multiple groups, one-way ANOVA was used followed by individual post hoc tests when applicable. Error probabilities of p < 0.05 were considered statistically significant. Data were analyzed using GraphPad Prism software.
Supplementary Material
Table 6.
List of up-regulated proteins in PFC of AMPKα2 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| Uncharacterized protein KIAA1671 | Q8BRV5 | 3.333 |
| Lipid phosphate phosphohydrolase 3+ | Q99JY8 | 2.867 |
| Arf-GAP with GTPase, ANK repeat and PH domain-containing protein 2+ | Q3UHD9 | 2.833 |
| NADH-ubiquinone oxidoreductase chain 4 | P03911 | 2.556 |
| Rho-related GTP-binding protein RhoB+ | P62746 | 2.511 |
| Branched-chain-amino-acid aminotransferase, cytosolic | P24288 | 2.333 |
| Calcium-activated potassium channel subunit alpha-1 | Q08460 | 2.167 |
| Cytochrome c oxidase subunit 5A, mitochondrial | P12787 | 2.088 |
| Programmed cell death protein 5 | P56812 | 1.944 |
| Uncharacterized protein KIAA1107 | Q80TK0 | 1.933 |
| Programmed cell death 6-interacting protein+ | Q9WU78 | 1.889 |
| Unconventional myosin-XVIIIa+ | Q9JMH9 | 1.833 |
| Early endosome antigen 1 | Q8BL66 | 1.704 |
| Cell adhesion molecule 4+ | Q8R464 | 1.7 |
| Serine/threonine-protein kinase PAK 1 | O88643 | 1.537 |
| Hippocalcin-like protein 1 | P62748 | 1.525 |
Up-regulated also in AMPKα1 cKO mice
Table 7.
List of down-regulated proteins in PFC of AMPKα1 cKO mice.
| Description | Accession | Fold change |
|---|---|---|
| Apolipoprotein A-I | Q00623 | 0.250 |
| Syntaxin-binding protein 5-like | Q5DQR4 | 0.258 |
| Histone H1.3+ | P43277 | 0.315 |
| Synaptogyrin-3+ | Q8R191 | 0.350 |
| Glutaredoxin-3+ | Q9CQM9 | 0.350 |
| Ribonuclease inhibitor+ | Q91VI7 | 0.356 |
| Phospholemman | Q9Z239 | 0.400 |
| NADH dehydrogenase [ubiquinone] 1 alpha subcomplex subunit 13 | Q9ERS2 | 0.415 |
| Protein canopy homolog 2+ | Q9QXT0 | 0.467 |
| Peptidyl-prolyl cis-trans isomerase NIMA-interacting 1 | Q9QUR7 | 0.500 |
| Aldose reductase+ | P45376 | 0.505 |
| Transcription elongation factor A protein-like 5 | Q8CCT4 | 0.564 |
| Transcription elongation factor B polypeptide 2 | P62869 | 0.567 |
| Tumor protein D54 | Q9CYZ2 | 0.583 |
| Contactin-associated protein-like 2 | Q9CPW0 | 0.589 |
| Histone H1.4+ | P43274 | 0.592 |
| Major prion protein | P04925 | 0.600 |
| DNA damage-binding protein 1+ | Q3U1J4 | 0.600 |
| Sodium- and chloride-dependent GABA transporter 3 | P31650 | 0.663 |
| Disks large homolog 2 | Q91XM9 | 0.667 |
Down-regulated also in AMPKα2 cKO mice
Acknowledgements
This work was supported by National Institutes of Health grants R01 AG055581 and R01 AG056622 (T.M.), the BrightFocus Foundation grant A2017457S (T.M.). The mass spectrometry analysis was performed by the Proteomics and Metabolomics Shared Resource at Wake Forest School of Medicine, which is partly supported by the Comprehensive Cancer Support Grant (P30CA012197).
Footnotes
Conflict of Interest
The authors declare they have no conflict of interest.
REFERENCES
- Alberini C (2008). The role of protein synthesis during the labile phases of memory: revisiting the skepticism. Neurobiol Learn Mem, 89(3), 234–246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Beckelman BC, Yang W, Kasica NP, Zimmermann HR, Zhou X, Keene CD, . . . Ma T (2019). Genetic reduction of eEF2 kinase alleviates pathophysiology in Alzheimer’s disease model mice. J Clin Invest, 129(2), 820–833. doi: 10.1172/jci122954 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Boland B, Yu WH, Corti O, Mollereau B, Henriques A, Bezard E, . . . Millan MJ (2018). Promoting the clearance of neurotoxic proteins in neurodegenerative disorders of ageing. Nat Rev Drug Discov, 17(9), 660–688. doi: 10.1038/nrd.2018.109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chen Y, McPhie DL, Hirschberg J, & Neve RL (2000). The amyloid precursor protein-binding protein APP-BP1 drives the cell cycle through the S-M checkpoint and causes apoptosis in neurons. J Biol Chem, 275(12), 8929–8935. doi: 10.1074/jbc.275.12.8929 [DOI] [PubMed] [Google Scholar]
- Frisardi V, Solfrizzi V, Seripa D, Capurso C, Santamato A, Sancarlo D, . . . Panza F (2010). Metabolic-cognitive syndrome: a cross-talk between metabolic syndrome and Alzheimer’s disease. Ageing Res Rev, 9(4), 399–417. doi: 10.1016/j.arr.2010.04.007 [DOI] [PubMed] [Google Scholar]
- Gong L, & Yeh ET (1999). Identification of the activating and conjugating enzymes of the NEDD8 conjugation pathway. J Biol Chem, 274(17), 12036–12042. doi: 10.1074/jbc.274.17.12036 [DOI] [PubMed] [Google Scholar]
- Hardie DG, Ross FA, & Hawley SA (2012). AMPK: a nutrient and energy sensor that maintains energy homeostasis. Nat Rev Mol Cell Biol, 13(4), 251–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hu Y, Cheng X, Li S, Zhou Y, Wang J, Cheng T, . . . Xiong D (2013). Inhibition of sorcin reverses multidrug resistance of K562/A02 cells and MCF-7/A02 cells via regulating apoptosis-related proteins. Cancer Chemother Pharmacol, 72(4), 789–798. doi: 10.1007/s00280-013-2254-2 [DOI] [PubMed] [Google Scholar]
- Ilari A, Fiorillo A, Poser E, Lalioti VS, Sundell GN, Ivarsson Y, . . . Colotti G (2015). Structural basis of Sorcin-mediated calcium-dependent signal transduction. Sci Rep, 5, 16828. doi: 10.1038/srep16828 [DOI] [PMC free article] [PubMed] [Google Scholar]
- J.Magistretti P, & Allaman I (2015). A cellular perspective on brain energy metabolism and functional imaging. Neuron, 86(4), 883–901. [DOI] [PubMed] [Google Scholar]
- Klann E, & Dever TE (2004). Biochemical mechanisms for translational regulation in synaptic plasticity. Nat Rev Neurosci, 5(12), 931–942. doi: 10.1038/nrn1557 [DOI] [PubMed] [Google Scholar]
- Lin MT, & Beal MF (2006). Mitochondrial dysfunction and oxidative stress in neurodegenerative diseases. Nature, 443(7113), 787–795. [DOI] [PubMed] [Google Scholar]
- Mao J, Ling F, Gislaine Pires Sanches J, Yu X, Wei Y, & Zhang J (2020). The potential mechanism of action of Sorcin and its interacting proteins. Clin Chim Acta. doi: 10.1016/j.cca.2020.09.011 [DOI] [PubMed] [Google Scholar]
- Nixon RA (2020). The aging lysosome: An essential catalyst for late-onset neurodegenerative diseases. Biochim Biophys Acta Proteins Proteom, 1868(9), 140443. doi: 10.1016/j.bbapap.2020.140443 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Steinberg GR, & Kemp BE (2009). AMPK in Health and Disease. Physiol Rev, 89(3), 1025–1078. [DOI] [PubMed] [Google Scholar]
- Trinh MA, Kaphzan H, Wek RC, Pierre P, Cavener DR, & Klann E (2012). Brain-Specific Disruption of the eIF2α Kinase PERK Decreases ATF4 Expression and Impairs Behavioral Flexibility. Cell Rep, 1(6), 676–688. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tsokas P, Ma T, Iyengar R, Landau EM, & Blitzer RD (2007). Mitogen-activated protein kinase upregulates the dendritic translation machinery in long-term potentiation by controlling the mammalian target of rapamycin pathway. J Neurosci, 27, 5885–5894. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Viollet B, Horman S, Leclerc J, Lantier L, Foretz M, Billaud M, . . . Andreelli F (2010). AMPK inhibition in health and disease. Crit Rev Biochem Mol Biol, 45(4), 276–295. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang X, Zimmermann HR, & Ma T (2019). Therapeutic Potential of AMP-Activated Protein Kinase in Alzheimer’s Disease. J Alzheimers Dis, 68(1), 33–38. doi: 10.3233/jad-181043 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wells DG, Dong X, Quinlan EM, Huang YS, Bear MF, Richter JD, & Fallon JR (2001). A role for the cytoplasmic polyadenylation element in NMDA receptor-regulated mRNA translation in neurons. J Neurosci, 21(24), 9541–9548. doi: 10.1523/jneurosci.21-24-09541.2001 [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yang W, Zhou X, Cavener HZD, Klann E, & Ma T (2016). Repression of the eIF2α kinase PERK alleviates mGluR-LTD impairments in a mouse model of Alzheimer’s disease. Neurobiol Aging, 41, 19–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yang W, Zhou X, Zimmermann HR, & Ma T (2020). Brain-specific suppression of AMPKα2 isoform impairs cognition and hippocampal LTP by PERK-mediated eIF2α phosphorylation. Mol Psychiatry. doi: 10.1038/s41380-020-0739-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zimmermann HR, Yang W, Kasica NP, Zhou X, Wang X, Beckelman BC, . . . Ma T (2020). Brain-specific repression of AMPKα1 alleviates pathophysiology in Alzheimer’s model mice. J Clin Invest, 130(7), 3511–3527. doi: 10.1172/jci133982 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.