Misfolded Superoxide Dismutase-1Increases Amyloid β Aggregation

Amyloid plaque is a key pathological feature of Alzheimer disease (AD) which ismainly composed of amyloid beta (Aβ) fibril. Aβ oligomer is wildly accepted as the mosttoxic form of Aβ and is a plausible cause for dementia via attacking synapse. The fact thatmore Aβ is expressed in the familial AD patients’ brain suggests the correlation betweenAβand AD. Aβ has a trend to aggregate and the molecular structure of the aggregation hasbeen studied. The “in-register”, β-sheet stack is called “cross-β” structure. Aggregated Aβsheets form a ribbon in which the long axis being perpendicular to β-sheet. This ribbon iscalled protofilament and several of them can twist together to form Aβ fibril and eventuallyAβ plaque. A relatively slow “nucleation” phase is required in the formation ofprotofilament. Once the “nucleus” is formed, protofilament can be elongated by Aβ rapidly.It’s obvious that more “nucleus” would increase Aβ aggregation.On the other hand, ageing is the most dangerous risk factor of AD but the mechanismhas not been fully understood yet. It is widely accepted that the accumulated damage byendogenous and exogenous reactive oxygen species leads to ageing. Superoxide dismutaseis a defensive protein to protect cells from being oxidized, however, itself can easily beingoxidized due to its frequent reaction with ROS. Oxidation and mutation of SOD1do notaffect its normal antioxidant function but bear a toxic function. Both oxidative SOD1andmutant SOD1are prone to be misfolded and aggregate. SOD1aggregation has a similarcross-β structure, and the co-localization and protein-protein reaction between SOD1andAβ has already been demonstrated.Based on the knowledge mentioned above, we propose the misfolded SOD1as the keyconnection between Aβ and ageing. Our hypothesis is that misfolded SOD1, either mutativeor oxidative, is the reason that increase Aβ aggregation and thus leads to AD and theformation of amyloid plaque. Furthermore we propose one possible mechanism in whichmisfolded SOD1as the additional “nucleus”. Materials and Methods1. Animal model and genotyping: G37R/APP double transgenic mice were generatedby crossing single transgenic mice expressing SOD/G37R mutation and mice expressingAPP Swedish mutation. Genotyping process included (Deoxyribonucleic Acid) DNAextraction, PCR and gel electrophoresis.2. Preparation of mice brain tissue section: Mice were sacrificed at their scheduledtime point. Left hemispheres were used for morphological study and right hemisphereswere used for protein quantitative study.3. Congo red staining of brain section: Congo red staining kit from SIGMA-ALDRICH was used for Congo red staining of brain section.4. Culture of N2a/APPsw cell line.5. Recombinant plasmid transformation, amplification and extraction: pcDNA3.1-G37R plasmid was offered by Dr. Xueping Chen (Department of Human Anatomy and CellScience University of Manitoba, Winnipeg, Canada). The transformation followed themanufacturer’s instructions of MAX Efficiency DH5α Competent Cells kit from Invitrogen.6. Plasmid transfection: to study the role of mutant SOD1in Aβ aggregation, we usedthe plasmid containing mutant SOD1gene to transfect N2a/APPsw cells. Lipofectamin2000(invitrogen) was used for the transfection.7. Double immunefluorescence for Aβ and SOD1.8. Brain cortex Aβ42/40enzyme linked immunosorbent assay.9. Statistics: results were expressed as the Mean±SD. SPSS20.0was used foranalyzing. Statistical significance of differences among the means of which variables weredetermined by analysis of variance (ANOVA). Values of P<0.05were consideredstatistically significant.Results1. Soluble and insoluble Aβ42/40increase in the cortex of double transgenic mice.Results suggest a similar increase feature of Aβ42and Aβ40concentration in thecortex of double transgenic mice brain. For the soluble Aβ, there are significant differencesin9m and12m group between double and single transgenic mice, indicating that mutantSOD1plays a role in increasing soluble Aβ. No significant difference was detected in the3m and6m group, indicating that Aβ was increased mainly after6m.For the insoluble Aβ42and Aβ40, there was significant difference between double and single transgenic mice in6m,9m and12m group. This results indicates that both Aβ42and Aβ40increased in the insoluble portion. In addition, the significant increase ofinsoluble Aβ，which shows difference as early as6m, is earlier than that of soluble A.2. Amyloid burden has no significant increase in12m APP+/G37R mice brain.The results suggest no significant increase of amyloid burden in term of size anddensity at the time point of12m. We either sporadically found very limited numbers ofplaques or found no plaque in double transgenic mice. In single transgenic mice, wesporadically found several small size plaque in3of them and no plaque in the other1mice.In addition, we didn’t find any plaque in mice brain of8m group. Anaylsis of amyloidburden and number of plaque shows no significant difference between the two genotypegroups. These results suggest no significant increase acceleration of amyloid formation inthis double transgenic mouse model.3. Aβ aggregation increased in12m double transgenic mice neuron.Aβ levels in double transgenic mice were much higher than those in single transgenicmice. The intracellular amyloid burden was calculated by measuring the percentage ofcondensed Aβ area in cells and average flourensece intensity was measured by the ratio offlourensece intensity and area of cytoplasma. The results showed intracellular amyloidburden increased almost8folds in the APP+/G37R+mice. In addition to the increase ofcondensed Aβ，the diffused forms of Aβ signal were also increased. These results suggest inthe double transgenic mouse model, the mutant SOD1increase the expression andaggregation of Aβ, which is consistent with Aβ ELISA results. Soluble and insoluble Aβrespectively corresponded to the aggregated and diffused forms of Aβ.4. Aβ expression and aggregation increased in N2a/APPsw/G37R cells.Cells successfully transfected with human SOD1gene can be detected by humanSOD1antibody thus show green fluorescence. The mutant SOD1were expressed in adiffused form in the cytoplasm but only few condensed particle were found. This may relateto a short time period after transfection. In the successfully transfected cells, Aβ levels werehigher than those in unsuccessful ones and condensed particles and plaques were detectedin cytoplasm. This result is consistent with ELISA and IF of brain tissue, indicating thatmutant SOD1can increase the expression and aggregation of Aβ. In addition, this resultsalso demonstrated that SOD1can increase Aβ expression and aggregation even when SOD1 is still in a diffused form. Different from IF in the brain tissue, no obvious co-localizationof SOD1and Aβ was detected by, which is possibly caused by the short time period aftertransfection.Discussion and ConclusionIn this research we studied the forms of Aβ aggregation and effects of misfolded SOD1in Aβ aggregation in a mouse model. We observed both soluble and insoluble Aβ form inthe cortex of transgenic mice. The results demonstrated that misfolded SOD1increased Aβaggregation in both soluble and insoluble forms, which is not exactly consistent with ourhypothesis. We propose that misfolded SOD1increase Aβ aggregation by acting asadditional “nucleus” of Aβ filament, not by increasing the total amounts. This resultsuggests that amount of total Aβ is increased in the double transgenic mice and misfoldedSOD1might affect Aβ aggregation in some other ways than we proposed, such as by thetoxicity SOD1obtained after misfolding. In addition, the results demonstrated an earliertime point of the increase for insoluble Aβ, suggesting a potential negative feedback ofsoluble Aβ. Results from Bradley J. Turner et al. demonstrated no obvious increase of APPand Aβ in G93A transfected mice as compared to widetype mice, which is inconsistent withour observation.Double immunofluorescence was used to observe intracellular SOD1and Aβ. Theresults met our anticipation that total Aβ expression, amount and size of aggregation wereincreased in double transgenic mice, which is consistent with ELISA analyzed results.Diffused and aggregated forms of Aβ were respectively corresponding to soluble andinsoluble Aβ. According to this result and results from other groups we propose that Aβincreases intracellularly first and will not be transported to extracellular matrix untilreaching a certain threshold.We used a corresponding cell model to comfirm our observation in vitro and yieldedsimilar results. Interestingly, SOD1in N2a cells did not form aggregated plaques orco-localized with Aβ, that suggest aggregation of SOD1is not necessary to increase Aβaggregation. This again suggests additional potential mechanisms of Aβ aggregation otherthan SOD1playing as additional “core”.In conclusion, our results suggest that misfolded SOD1improves Aβexpression andintracellular aggregation in vitro and in vivo. This indicates misfolded SOD1can potentially be an initial reason for AD.