EHS
EHS

α-Synuclein Impacts on Intrinsic Neuronal Network Activity Through Reduced Levels of Cyclic AMP and Diminished Numbers of Active Presynaptic Terminals




doi: 10.3389/fnmol.2022.868790.


eCollection 2022.

Affiliations

Item in Clipboard

Kristian Leite et al.


Front Mol Neurosci.


.

Abstract

α-synuclein (α-Syn) is intimately linked to synucleinopathies like Parkinson’s disease and dementia with Lewy bodies. However, the pathophysiological mechanisms that are triggered by this protein are still largely enigmatic. α-Syn overabundance may cause neurodegeneration through protein accumulation and mitochondrial deterioration but may also result in pathomechanisms independent from neuronal cell death. One such proposed pathological mechanism is the influence of α-Syn on non-stimulated, intrinsic brain activity. This activity is responsible for more than 90% of the brain’s energyconsumption, and is thus thought to play an eminent role in basic brain functionality. Here we report that α-Syn substantially disrupts intrinsic neuronal network burst activity in a long-term neuronal cell culture model. Mechanistically, the impairment of network activity originates from reduced levels of cyclic AMP and cyclic AMP-mediated signaling as well as from diminished numbers of active presynaptic terminals. The profound reduction of network activity due to α-Syn was mediated only by intracellularly expressed α-Syn, but not by α-Syn that is naturally released by neurons. Conversely, extracellular pre-formed fibrils of α-Syn mimicked the effect of intracellular α-Syn, suggesting that they trigger an off-target mechanism that is not activated by naturally released α-Syn. A simulation-based model of the network activity in our cultures demonstrated that even subtle effect sizes in reducing outbound connectivity, i.e., loss of active synapses, can cause substantial global reductions in non-stimulated network activity. These results suggest that even low-level loss of synaptic output capabilities caused by α-Syn may result in significant functional impairments in terms of intrinsic neuronal network activity. Provided that our model holds true for the human brain, then α-Syn may cause significant functional lesions independent from neurodegeneration.


Keywords:

cAMP; intrinsic network activity; pre-formed fibrils; synapses; α-Synuclein.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures



Figure 1

AAV vectors as used in this study and characteristics of cAMP sensors. (A) Schematical depiction of AAV vector genomes: for most experiments the synucleins (α-Syn, β-Syn, or γ-Syn) were expressed from bi-cistronic vectors, which also express the fluorophore nuclear-targeted mCherry (Nuc-mCherry, NmC) from an independent transcription unit. As an additional control, or “empty vector”, NmC was expressed alone from a mono-cistronic vector. Bcl-xL, genetically encoded fluorescent sensors, and synucleins were expressed from mono-cistronic vectors in specific conditions. ITR, inverted terminal repeat of AAV2; hSyn, human synapsin 1 gene promoter; Int., Intron (splice sites); BGH, bovine growth hormone polyadenylation site; WPRE, woodchuck hepatitis virus posttranscriptional regulatory element. (B) Representative trace of cAMPr absolute fluorescence changes in one neuron, which was treated with compounds that alter cAMP level. The trace has been corrected for bleaching. Dashed line = time duration for applying linderane, 20 mM, and SQ-22536 = 20 μM; straight line = time duration for applying rolipram, 5 μM, and forskolin, 5 μM). A.U., arbitrary units; PDE, phosphodiesterase; AC, adenylate cyclase. (C) Representative trace of AKAR4-cAMPs FRET ratio changes in one neuron after reducing cAMP levels with linderane/SQ-22536 or enhancing cAMP levels with forskolin/rolipram.


Figure 2


Figure 2

α-Synreduces network burst activity in aged neurons. (A) Scatterplot showing the activity of 1,000 α-Syn expressing neuronswithin one minute at DIV 28. Every dot represents the calcium transient of one cell as a surrogate marker for a train of action potentials. Vertical accumulations of this activity represent synchronized network bursts. (B) Scatter plot showing activity of 1,000 γ-Syn expressing neurons at DIV 28. (C) Neuron numbers relative to DIV10 in α-Syn expressing cultures (white bars) and γ-Syn expressing cultures (gray bars). (D) Frequency of network bursts in α-Syn expressing neurons (white bars) and γ-Syn expressing neurons (gray bars). (E) Percentage of neurons contributing to network bursts at 16 and 28 DIV in α-Syn expressing neurons (white bars) and γ-Syn expressing neurons (gray bars). N = 18 neuronal cultures for (C,D) and 41 neuronal cultures for (E). Statistics by unpaired, two-tailed T-test; n.s. = not significant; ***p < 0.001; ****p < 0.0001. Statistical power (1 – ß error probability) > 97% for all conditions.


Figure 3


Figure 3

Quantification and characterization of natively released extracellular α-Syn. (A) Western blot after SDS-PAGE showing α-Syn from cell lysate (intracellular) and from cell culture supernatant (extracellular) of neurons overexpressing human α-Syn. Recombinantly produced peptide standards were used for quantification. Detection was performed with human-specific anti-α-Syn Ab, Syn211. (B) Quantitative analysis of intra- and extracellular amounts of α-Syn at different time points. Statistics by 1-way ANOVA with Tukey’s test for multiple comparisons. Notably, the 1-way ANOVA was done separately for the four values from cell lysates and the four values from culture mediums. N = 4 independent neuronal cultures transduced with 2 × 108 tu of AAV-α-Syn -NmC. n.s = not significant; *p < 0.05. (C) Western blot after SDS-PAGE showing separation of soluble and insoluble α-Syn by ultracentrifugation. PFFs generated from recombinant α-Syn (6× His tagged, resulting in higher molecular weight as compared to cell expressed α-Syn and the α-Syn used as peptide standards) were found only in the pellet (pel.) but not in the supernatant (sup.), while extracellular, natively released α-Syn was found almost exclusively in the supernatant (sup.). (D) Quantification of insoluble (pellet) and soluble (supernatant) natively released α-Syn after ultracentrifugation. N = 3 independently obtained cell culture supernatants. (E) Western blot after native PAGE of recombinant (rec.) α-Syn monomers and PFFs (left panel) and α-Syn obtained from cell culture supernatant. Cell culture supernatant was either loaded “as it is” (native) or after treatment with SDS and heating to 95°C. Detection was performed with human-specific anti-α-Syn Ab, Syn211. (F) Intracellular, but not natively secreted α-Syn is phosphorylated at S129. Western blot after SDS-PAGE showing α-Syn from cell lysate (intracellular) and from cell culture supernatant (extracellular) of neurons overexpressing human α-Syn. Recombinantly produced non-phosphorylated peptide standards served as controls for phospho-specific detection. Detection of total α-Syn was performed with human-specific anti-and α-Syn Ab, Syn211, detection of S129P α-Syn was performed with ab168381.


Figure 4


Figure 4

Impactof naturally secreted extracellular α-Syn on network burst activity of native neurons. (A) Schematic depiction of the experimental layout. α-Syn expressing/secreting neurons were cultured alongside native neurons, which only express NmC and GCaMP6f. At DIV 22, the cell culture supernatant of α-Syn expressing/secreting neurons was used to replace the cell culture supernatant of the native recipient neurons, which were then analyzed for their network activity up to DIV 31. (B) Network bursts per minute are shown for recipient neurons that received α-Syn-containing medium on DIV 22 (white bars) or were further maintained in native medium (dark gray bars). N = 12 independent neuronal cultures per condition. Statistics by unpaired, 2-tailed T-test. n.s. = not significant.


Figure 5


Figure 5

Impact of extracellular recombinant monomers and processed fibrils of α-Syn on network burst activity. (A) Neuronal network bursts per minute in cultures exposed to PBS (white bars), monomeric α-Syn (MM, gray bars), or ultrasound-processed preformed fibrils (pPFF) at the indicated times after addition of 0.4 μM of each α-Syn species. (B) Neuron numbers in cultures exposed to PBS (white bars), monomeric α-Syn (MM, gray bars), or ultrasound-processed preformed fibrils (pPFF) at the indicated times after the addition of 0.4 μM of each α-Syn species. Statistics by 1-way ANOVA. N = 22 independent cultures for PBS, 24 for MM and 26 for pPFFs (A,B). n.s. = not significant; *p < 0.05; **p < 0.01; Statistical power (1 – β error probability) = 99% (A), DIV 6, 78% (A), DIV 7, 88% (A), DIV 9.


Figure 6


Figure 6

α-Synexpression causes reductions of axon structures and active synapses. (A) Detection of dendrites, shown in the left panel as a fluorescent stain for microtubule-associated protein 2 (MAP2), and in the right panel as a binarized image used for quantification. The bar diagram shows the percent coverage of the binary with dendrite structures for neurons expressing α-Syn (white bars) or γ-Syn (dark bars), at either DIV 16 or DIV 28. (B) Detection of axons, shown in the left panel as a fluorescent stain for neurofilament L (NFL), and in the right panel as a binarized image used for quantification. The bar diagram shows the percent coverage of the binary with axon structures for neurons expressing α-Syn (white bars) or γ-Syn (dark bars), at either DIV 16 or DIV 28. (C) Detection of active presynaptic terminals. The fluorescent signal of SyGCaMP is shown in the resting state, during network activity, and as the resulting difference image, which identifies active synapses. The bar diagram shows the number of active synapses per mm2 at DIV 16 and DIV 28 for neurons expressing α-Syn (white bars) or γ-Syn (dark bars). N (independent cultures): (A) DIV 16 = 4; DIV 28 = 8; (B) DIV 16 = 6; DIV 28 = 8; (C) DIV 16 = 9; DIV 28 = 28. Statistics by two-tailed unpaired T-tests. n.s. = not significant; *p < 0.05; **p < 0.01. Statistical power (1 – ß error probability) = 76% (B), 81% (C).


Figure 7


Figure 7

α-Syncauses diminished levels of cAMP and PKA phosphorylation.(A) Schematic depiction of cAMP-mediated effects onnetwork activity. AC, adenylate cyclase; AMP, adenosine monophosphate; ATP, adenosine triphosphate; BDNF, brain-derived neurotrophic factor; cAMP, cyclic adenosine monophosphate; HCN, hyperpolarization-activated cyclic nucleotide-gated channel; LTP, long-term potentiation; PDE, phosphodiesterase; PKA, protein kinase A. (B) HCN channel inhibitor ZD-7288 reduces the frequency of network bursts. At DIV 28, network activity was determined before (white bars) and 30 min after (dark bars) addition of 10 μM ZD-7288 to neurons which do not express human synuclein. (C) BDNF stimulates network activity. At DIV 28, network activity was determined before (white bars) and 30 min after (dark bars) addition of 2 nM BDNF to neurons which do not express human synuclein. (D) Forskolin stimulates network activity only in γ-Syn expressing neurons. At DIV 28, network activity was determined before (white bars) and 30 min after (dark bars) addition of 1 μM forskolin in either α-Syn (left) orγ-Syn (right) expressing neurons. (E) Rolipram stimulates network activity only in γ-Syn expressing neurons. At DIV 28, network activity was determined before (white bars) and 30 mi after (dark bars) addition of 1 μM rolipram in either α-Syn (left) orγ-Syn (right) expressing neurons. (F) As a prerequisite for the determination of cAMP levels by ELISA, numbers of neurons and astrocytes were quantified at DIV 28. (G) Quantification of cAMP levels by ELISA at DIV 28 in cultures expressing α-Syn (white bar) or γ-Syn (gray bar). (H) Quantification of neuronal cAMP levels at DIV 16 and DIV 28, as determined by the fluorescent signals of the cAMPr sensor α-Syn expressing neurons (white bars) and in γ-Syn expressing neurons (dark bars). Quantification normalized to γ-Syn expressing neurons at DIV 16 = 100%. (I) Quantification of neuronal PKA phosphorylation at DIV 16 and DIV 28, as determined by the fluorescent signals of the AKAR4-cAMPs sensor in α-Syn expressing neurons (white bars) and in γ-Syn expressing neurons (dark bars). Quantification normalized to γ-Syn expressing neurons at DIV 16 = 100%. N (as independently recorded neuron cultures) = 6–8 in (B,C); 10 in (D,E); 12 in (F,G); 6 in (H), DIV 16; 36 in (H), DIV 28; 12 in (I), DIV 16; and 18 in (I), DIV 28. Statistics by two-tailed paired or unpaired T-test as appropriate. n.s. = not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.00001. Statistical power (1 – ß error probability) = 91% in (B); 78% in (C); 99% in (D); 96% in (E); 92% in (G); 93% in (H); 100% in (I).


Figure 8


Figure 8

Moderatereductions in outbound connectivity cause robust reductions in intrinsic network activity. (A) The frequency of network bursts decreases with the reduction of synapses (i.e., outbound connections), assuming mono-synaptic connectivity (single, black solid line) or multiple, redundant synaptic connectivity with compensation (redundant, dotted gray line). Raster plots show network activity for the mono-synaptic connectivity with different levels connectivity: 100% connectivity (B), 97.5% connectivity (C), 95% connectivity (D), 92.5% connectivity (E), and 90% connectivity (F). Dots represent single spikes, lines the network bursts.

References

    1. Ahn K. J., Paik S. R., Chung K. C., Kim J. (2006). Amino acid sequence motifs and mechanistic features of the membrane translocation of alpha-synuclein. J. Neurochem. 97, 265–279. 10.1111/j.1471-4159.2006.03731.x



      DOI



      PubMed

    1. Alvarez-Lacalle E., Moses E. (2009). Slow and fast pulses in 1-D cultures of excitatory neurons. J. Comput. Neurosci. 26, 475–493. 10.1007/s10827-008-0123-5



      DOI



      PubMed

    1. Andrade-Talavera Y., Duque-Feria P., Sihra T. S., Rodriguez-Moreno A. (2013). Pre-synaptic kainate receptor-mediated facilitation of glutamate release involves PKA and Ca2+-calmodulin at thalamocortical synapses. J. Neurochem. 126, 565–578. 10.1111/jnc.12310



      DOI



      PubMed

    1. Argyrousi E. K., Heckman P. R. A., Prickaerts J. (2020). Role of cyclic nucleotides and their downstream signaling cascades in memory function: being at the right time at the right spot. Neurosci. Biobehav. Rev. 113, 12–38. 10.1016/j.neubiorev.2020.02.004



      DOI



      PubMed

    1. Belgacem Y. H., Borodinsky L. N. (2017). CREB at the crossroads of activity-dependent regulation of nervous system development and function. Adv. Exp. Med. Biol. 1015, 19–39. 10.1007/978-3-319-62817-2_2



      DOI



      PubMed



Source link

EHS
Back to top button