AP39,A Mitochondria-targeted Hydrogen Sulfide Donor, Supports Cellular Bioenergetics And Protects Against Alzheimer’s Disease By Preserving Mitochondrial Function In APP/PS1 Mice And Neurons

ObjectiveIncreasing evidence suggests that mitochondrial bioenerge tics and function are altered in Alzheimer’s disease(AD) and play an important role in AD pathogenes is. It has been well established that hydrogen sulfide(H2S) homeostasis is balanced in AD. The emerging mitochondrial roles of H2 S include antioxidative effects, anti-apoptotic effects and the modulation of cellular bioenergetics. AP39 is a newly synthesized mitochondrially targeted H2 S donor and the protective effects of AP39 have been verified in the other models in vitro and in vivo. However, it is not clear that the possible protective effect and its mechanism of AP39 on neuronal mitochondrial dysfunction in AD. This study was designed specifically to study the effect of AP39 on mitochondrial dysfunction in APP/PS1 transgenic mice and its involved mechanism. MethodsFirstly, according to different months, the APP/PS1 transgenic mice were divided into 3 groups(3, 6, 12month). Then Morris water maze and novel object recognition task(NORT) would be used to observe the changes of cognitive function of APP/PS1 transgenic mice; methylene blue spectrophotometry method would be used to detect H2 S levels in the cortex and hippocampus and the changes of CBS, CSE and 3MST enzyme activity; Western blot would be used to detect the protein expressions of CBS, CSE and 3MST. APP/PS1 pregnant mice were sacrificed during pregnancy 15-16(E15-16) days, the embryonic brain were collected and cultured primary neurons. Neurons were identified by Cell Immunofluorescence technique. After neurons were treated with different concentrations of AP39 for 24 h, MTT assay was used to detect the neuronal survival; LDH assay was used to detect the damaged neurons; TUNEL assay was used to detect the apoptosis of neurons; methylene blue spectrophotometry was used to detect the H2 S levels in neurons; fluorescent probe was used to detect the production of H2 S in the mitochondria. The XF24 Extracellular Flux Analyzer was used to detect neuronal mitochondrial respiratory function. Control neurons were transfected in parallel with nontargeting s i RNA(Life Technologies). After 24h(s i RNA-SQR), 35,000cells/0.32cm2/well were plated into 24-well XF24 Seahorse plates, and bioenergetic measurements were conducted.The mitochondrial respiratory complex I-IV enzyme activity kit was used to detect their respective activity. The ATP kit was used to detect the ATP content in the neurons. Polymerase chain reaction(PCR) was used to detect the mt DNA integrity. Reactive Oxygen Species detection kit was used to evaluate the generation of ROS in neurons. Mitochondrial membrane potential kit was used to evaluate the changes of mitochondrial membrane potential in neurons. Western blot was used to detect the expressions of mitochondrial protein Mfn-1, Mfn-2, OPA-1, Fis-1and Drp-1. Secondly, 12-month-old APP/PS1 transgenic mice and wild-type(WT) mice were redmondly divided into four groups(each group 25): WT mice+H2O; WT mice+AP39; APP/PS1 mice+H2O; APP/PSl mice+AP39. The mice were treated with AP39 or H2 O once by intraperitoneal injection a day for six weeks, and then the Morris water maze and the NORT test were performed. After behavioral tests, the mice were done the MRI scans, and then were sacrificed to collect their blood and brain tissue. Blood was collected and then centrifuged at 4℃1000g 10 min. The supernatant was collected to detect Aβ40 and Aβ42 levels by ELISA. The ultrastructure of brain tissue was detected by electron microscopy and Aβ deposition was detected by immunohistochemistry. Results1. The results in water maze and NORT test showed that with increasing age, the time of APP/PS1 transgenic mice finding escape platform was more and more long, the number of shuttle target quadrant and the retention time was less and less, the cognition index to the new object B was lower and lower(p <0.01).2. With age growing, the H2 S levels in cerebral cortex and hippocampus of APP/PS1 transgenic mice gradually reduced. Especially, the H2 S level in the cortex and hippocampus of 12-months-old APP/PS1 transgenic mice significantly reduced(p<0.05).3. CSE enzyme activity in the liver was strongly expressed, but CSE enzyme activity in the brain cortex and hippocampus scarcely expressed. With months growing, the CBS and 3MST enzyme activity in cerebral cortex and hippocampus of APP/PS1 mice gradually decreased. However, the 3MST enzyme activity more obviously decreased than CBS enzyme activity. The protein level of CSE, CBS, and 3MST expressed the same as the enzyme activity. Moreover, we found that 3MST were located in the neuronal mitochondria.4. AP39 increased the H2 S generation and regulated the cellular bioenergetics: After AP39 treated WT neurons with different concentrations of AP39 for 2h, we found that(25-250 n M) AP39 increased the H2 S generation in a concentration dependent manner. In WT neurons, 100 n M AP39 significantly increased the based OCR, but 250 n M AP39 markedly reduced the based OCR. 25, 100 n M AP39 dose-dependently increased the maximal OCR, but 250 n M AP39 obviously decreased the maximal OCR.5. AP39 exerted cytoprotective effects on APP/PS1 neurons: Exposure of WT neurons to AP39 for 24 h had no effect on cell viability, but treating the APP/PS1 neurons with 25,100 n M AP39 for 24 h resulted in the increase in cell viability. Meanwhile, AP39 had no effect on basal LDH release in WT neurons, but there was a significant increase in the amount of LDH in the APP/PS1 neuron culture medium. In contrast, after treatment with AP39, particularly at 100 n M, the amount of LDH in the APP/PS1 neuron culture medium was decreased. To explore the apoptosis inhibition effect of AP39, neurons apoptosis was detected by TUNEL analysis. Fluorescence microscopy revealed that the apoptosis in APP/PS1 neurons was significantly increased compared to the WT neurons, but when these neurons were pretreated with 25, 100 n MAP39 for 24 h, the apoptosis obviously decreased. However, 250 n M AP39 further aggravated the apoptosis.6. AP39 attenuated the loss of cellular bioenergetics in APP/PS1 neurons and SQR was participated in the adjustment process: We measured cellular bioenergetics using the XF24 Extracellular Flux Analyzer. Cellular bioenergetic parameters were significantly decreased in the mitochondria of APP/PS1 neurons compared with those of WT neurons. Importantly, 100 n M AP39 s ignificantly increased the basal respiratory rate and the OCR-linked maximal respiratory capacity of the APP/PS1 neurons. After interference SQR, the based respiratory rate and the maximum respiration rate was reduced in APP/PS1 neurons(P<0.01).7. AP39 protected mitochondrial function in APP/PS1 neurons: Mitochondrial respiratory chain complex I-IV activity in APP/PS1 neurons were lower than those in WT neurons, while 100 n M AP39 can significantly increased the mitochondrial respiratory chain complex activity in APP/PS1 neurons(P<0.01). AP39 marked ly increased the ATP production in WT and APP/PS1 neurons. Next, mt DNA and nuclear genomic DNA integrity was assessed via PCR of long DNA fragments. We found that mt DNA but not nuclear DNA integrity was clearly reduced in APP/PS1 neurons, compared to WT neurons. However, AP39 significantly protected against mt DNA damage in APP/PS1 neurons by partially restoring mt DNA integrity. 100 n M AP39 effectively decreased ROS levels and reversed MMP in APP/PS1 neurons.8. AP39 shifted the mitochondrial dynamics toward fission in APP/PS1 neurons: The changes in the levels of proteins involved in mitochondrial dynamics were determined by Western blot analys is. The levels of the mitochondrial fus ion proteins Mfn1 and OPA1 were significantly reduced, and the level of the mitochondrial fission protein Fis1 was markedly increased in APP/PS1 neurons compared with WT neurons. OPA1, Mfn1 and Mfn2 catalyzed mitochondrial fusion. AP39 increased the levels of OPA1 and Mfn1 but not Mfn2. Moreover, AP39 decreased the levels of Fis1 but not Drp1.9. 25-250 n M AP39 dose-dependently increased the H2 S production of cortex and hippocampus in the WT and APP/PS1 mice.10. AP39 reversed the memory deficits of APP/PS1 transgenic mice: To test the ability of AP39 to ameliorate the spatial learning deficits of the AD model mice, we treated 12-month-old AD model mice and WT mice with 100 n M/kg AP39. Then, behavioral testing was initiated 6 weeks after the initiation of AP39 therapy using the Morris water maze and the NORT. In the training trials, treatment with AP39 for 6 weeks eliminated the pre-existing deficits of the transgenic mice, as demonstrated by a reduced latency to locate the hidden platform compared to the latency of the WT control with AP39. The WT mice with or without AP39 treatment preferred the target quadrant in the probe trial. In contrast, the AD model mice treated with H2 O spent much less time in the target quadrant. The AD model mice treated with AP39 showed the same strong preference for the target quadrant as the WT mice. This benefit was concentration-dependent, as 25 n M/kg AP39 treatment did not improve the water maze performance of the AD model mice. Next, AD mice treated with water performed poorly and presented an impairment compared to the WT mice, indicated by a significant reduction in the percentage of time exploring the novel object. In contrast, the AP39-treated AD mice showed significantly improved performance levels compared to the water-treated AD mice level. Thus, treatment of aged APP/PS1 mice with 100 n M/kg AP39 for 6 weeks reversed the age-dependent memory impairment.11. AP39 partly inhibited the brain atrophy of APP/PS1 transgenic mice: The mice were randomly selected for intravital scanning to non--invas ively study brain structure using MRI data. We studied the coronal and axial sections of the mouse brains. As expected, there were no apparent structural abnormalities in the WT mouse brains. However, we observed visible atrophy in the brains of the 12-month-old AD model mice treated with water. Their ventric les were asymmetric, with a much larger size on the right than the left s ide, and had unclear edges. The apparent diffus ion coefficient(ADC) value in parietal cortex and hippocampus of each group of mice were compared with diffusion weighted imaging(DWI) sequences. It showed the ADC value in parietal cortex and hippocampus of APP/PS1 mice significant declined compared to those in WT mice; whereas AP39 significantly increased the ADC values in parietal cortex and hippocampus of APP/PS1 mice(P<0.05).12. The effect of AP39 on the morphology of mitochondria in APP/PS1 mice: APP/PS1 mice neurons contained more mitochondria per cell body and exhibited more condensed chromatin, irregular nuclear outlines, fragmented endoplasmic reticulum, ruptured or engulfed mitochondria, and clumping lysosomes than WT mice neurons. Moreover, the cytoplasm of APP/PS1 mice neurons exhibited evident vacuolization and protrusions. In APP/PS1 mice neurons pretreated with AP39 the cellular ultrastructure was similar to that of the WT mice neurons. Ultimately, the observations from TEM demonstrated that AP39 preserved the ultrastructure of APP/PS1 mice neurons.13. AP39 reduced the levels of Aβ and Aβ deposition in APP/PS1 transgenic mice: After different treatments for 6 weeks, Aβ levels and plaque development were examined in vivo. The Aβ40 and Aβ42 levels in the 12-month-old AD transgenic model mouse brains were 925.59 and 532.15 pg/m L. Importantly, treatment with 100 n M/kg AP39 decreased the Aβ40 and Aβ42 levels in AD mice compared to water treatment. As expected, amyloid plaques were not observed in WT mice. However, intracellular amyloid in the hippocampal and cortical tissues of the AD model mice treated with water was remarkably increased. Importantly, treatment with AP39 significantly reduced both the number and the s ize of intracellular amyloid plaques in the cortical and hippocampal tissues compared to treatment with water in the AD mice. The amyloid plaque burden in the brains of AD mice treated with AP39 was decreased compared with the brains of AD mice treated with water. Conclusions1. With age increas ing, the cognitive function of APP/PS1 transgenic mice is getting worse; the levels of H2 S in cerebral cortex and hippocampus gradually decreased, while the CBS and 3MST enzyme activity and protein levels gradually decreased, but 3MST enzyme activity and protein levels more obviously reduced than CBS. Moreover, we observed that majority of 3MST located in the mitochondria. These results suggest that mitochondria H2 S is mainly produced by 3MST. 2. AP39 increased the H2 S generation in neurons, especially H2 S formation of mitochondria.The addition of AP39(25-250 n M) to WT neurons increased intracellular H2 S levels, mainly in mitochondrial regions. AP39 exerted dose-dependent effects on mitochondrial activity in APP/PS1 neurons, including increased cellular bioenergy metabolism and cell viability at low concentrations(25-100 n M) and decreased energy production and cell viability at a high concentration(250n M). Furthermore, 100 n M AP39 treatment potently increased ATP levels, protected mitochondrial DNA and decreased ROS generation. Moreover, AP39 treatment regulated mitochondrial dynamics, shifting from fission toward fusion, in APP/PS1 neurons. 3. After 6 weeks of treatment, AP39 administration to APP/PS1 mice significantly ameliorated their spatial memory defic its in the Morris water maze and NORT and reduced amyloid-β(Aβ) deposition in their brains. Additionally, AP39 inhibited brain atrophy in APP/PS1 mice. Based on these results, AP39 was proposed as a promis ing drug candidate for the treatment of AD, and its anti-AD mechanism may involve protection against mitochondrial damage.