Fig 2: Exogenous CART supplements reduce oxidative stress and mitochondrial dysfunction in the cerebral cortex of APP/PS1 mice. WT, APP/PS1 and CART-treated APP/PS1 mice (age, 8 months) were used. (A) 8-OHdG level in 10% cerebral cortex tissue homogenate. (B) The mitochondrial membrane potential (JC-1) in the cerebral cortex tissue. (C) ROS levels in cerebral cortex cells. (D) 3-NT levels in 10% cerebral cortex tissue homogenate. Relative protein expression levels of Bmi-1, SOD1, Prdx1 and ?-H2A.X in the cerebral cortex were (E) determined via western blotting and (F) semi-quantified. (G) Relative mRNA expression levels of SOD1, Sirt1, Bmi-1 and Nrf2 in cerebral cortex extracts were measured via reverse transcription-quantitative PCR. GAPDH was used as the internal control. Data are presented as the mean ± SEM (n=6 per group). *P<0.05, **P<0.01 and ***P<0.001 vs. control; #P<0.05, ##P<0.01 and ###P<0.001 vs. APP/PS1. CART, cocaine amphetamine regulated transcript; WT, wild-type; 8-OHdG, 8-hydroxy-2'-deoxyguanosine; 3-NT, neurotrophin-3; ROS, reactive oxygen species; Bmi-1, polycomb complex protein; SOD1, superoxide dismutase; Prdx1, peroxiredoxin 1; ?-H2A.X, ?-H2A histone family member X; Sirt1, sirtuin 1; Nrf2, nuclear factor erythroid 2 like 2.

Fig 3: Prdx1 interacts with ASK1-TBD in the presence of H2O2 through its peroxidatic cysteine via disulfide bond formation, while Prdx2 does not interact with ASK1-TBD. (A) BLI assay in the presence of 10 µM H2O2. (B) BLI assay in the absence of H2O2 (to guarantee no oxidation, 10 µM DTT was added to the buffer solution). The concentration of Prdx1 WT, C52A, C173A, and Prdx2 WT were fixed at 1 µM and MBP was used as reference. The y-axis indicates the wavelength shift (in nm) of the OctetRed 96 instrument. The vertical dashed line separates the association and dissociation phases.

Fig 4: The number of interacting proteins of Prdx1, Prdx2, and ASK1. The diagram was generated by Venny (version 2.1) (https://bioinfogp.cnb.csic.es/tools/venny/, accessed in 30 April 2021) and shows the number of common interactors in the cross sections of the diagram using the data from the PPI database (BioGrid version 4.3.196) (https://thebiogrid.org/, accessed in 30 March 2021) [62].

Fig 5: All ab initio Prdx1-mLumin models and the ASK1-LN model maintain the structural fold of Prdx1 and ASK1. Prdx1- and ASK1-mLumin ab initio models were superimposed with the human Prdx1 C83S mutant (PDB ID: 4XCS) andthe human ASK1 kinase domain (PDB ID: 2CLQ). (A) All Prdx1-mLumin ab initio models (blue) showed a low r.m.s.d. value with human Prdx1 C83S mutant (PDB: 4XCS) (yellow). (B) Only the ASK1-LN ab initio model (blue) showed a superimposition with a relatively low r.m.s.d. value with human ASK1 kinase domain (PDB ID: 2CLQ) (orange). The r.m.s.d. value is displayed for each superimposition. Superimpositions were done in PyMol (Version 2.4.1, Schrödinger, Inc., New York, USA) [40]. LN: large N-terminal half of mLumin. LC: small C-terminal half of mLumin.