Sex and region-specific disruption of autophagy and mitophagy in Alzheimer's disease: linking cellular dysfunction to cognitive decline

Abstract

Macroautophagy and mitophagy are critical processes in Alzheimer's disease (AD), yet their links to behavioral outcomes, particularly sex-specific differences, are not fully understood. This study investigates autophagic (LC3B-II, SQSTM1) and mitophagic (BNIP3L, BNIP3, BCL2L13) markers in the cortex and hippocampus of male and female 3xTg-AD mice, using western blotting, transmission electron microscopy (TEM), and behavioral tests (novel object recognition and novel object placement). Significant sex-specific differences emerged: female 3xTg-AD mice exhibited autophagosome accumulation due to impaired degradation in the cortex, while males showed fewer autophagosomes, especially in the hippocampus, without significant degradation changes. TEM analyses demonstrated variations in mitochondrial and mitophagosome numbers correlated with memory outcomes. Females had enhanced mitophagy, with higher BNIP3L and BCL2L13 levels, whereas males showed elevated BNIP3 dimers. Cognitive deficits in females correlated with mitochondrial dysfunction in the cortex, while in males, higher LC3B-II levels associated positively with cognitive performance, suggesting protective autophagy effects. Using machine learning, we predicted mitophagosome and mitochondrial numbers based on behavioral data, pioneering a predictive approach to cellular outcomes in AD. These findings underscore the importance of sex-specific regulation of autophagy and mitophagy in AD and support personalized therapeutic approaches targeting these pathways. Integrating machine learning emphasizes its potential to advance neurodegenerative research. Sex-specific differences in autophagy and mitophagy regulation in Alzheimer's disease (AD) are highlighted. Female 3xTg-AD mice show autophagosome accumulation and cognitive deficits, while males exhibit variations in mitophagy markers and behavior.

Sex-specific differences in autophagy and mitophagy regulation in Alzheimer’s disease (AD) are highlighted. Female 3xTg-AD mice show autophagosome accumulation and cognitive deficits, while males exhibit variations in mitophagy markers and behavior.

Fig. 1. Autophagosome status in control and 3xTg-AD mice.

A, D Representative western blot bands showing LC3 and SQSTM1 protein levels in male and female mice. B, C Quantification of LC3B-II protein levels in the cortex and hippocampus of female and male mice. E, F Quantification of SQSTM1 protein levels in female and male mice. Results show significantly lower autophagosome degradation (SQSTM1) and higher autophagosome marker (LC3B-II) in the cortex of 3xTg-AD females (P < 0.05). In 3xTg-AD males, there was a significant reduction in the autophagosome marker (LC3B-II) in the hippocampus (P < 0.05). In contrast, no significant differences were found in autophagosome degradation (SQSTM1) in both the cortex and hippocampus (P > 0.05). Statistical analyses were performed using a two-way ANOVA test. *P < 0.05; **P < 0.01. Data are presented as mean ± SD, with n = 5 per group (n = 5 females and n = 5 males). Panels A, D - LC3B-II and SQSTM are run on the same gel. In other words, the loading control is exactly the same.

Fig. 2. Mitophagy-related BCL2-family protein status ((BNIP3L) in control and 3xTg-AD mice.

A, D western blot analysis of BNIP3L monomer and dimer in the cortex and hippocampus of male and female 3xTg-AD and control mice. B, E Quantification of BNIP3L monomer expression levels in females and males, respectively. In female 3xTg-AD mice, a significant increase in BNIP3L monomer expression was observed in the cortex (P < 0.0001) and hippocampus (P < 0.001). C, F Quantification of BNIP3L dimer levels in female and male mice, respectively. Comparing the control and 3xTg-AD groups, a significantly higher BNIP3L dimer level was found in the cortex (P < 0.01), while a lower level was observed in the hippocampus of 3xTg-AD females (P < 0.01). No significant differences in BNIP3L monomer or dimer levels were found between control and AD males.

Fig. 3. Mitophagy-related BCL2-family protein status (BNIP3) in control and 3xTg-AD mice.

A, B Western blot analysis of BNIP3 monomer and dimer in female and male mice. C, D Quantification of BNIP3 monomer expression levels in females and male mice, respectively. BNIP3 monomer expression was significantly lower in the cortex (P < 0.01) and hippocampus (P < 0.001) of 3xTg-AD females than in control, while no significant differences were found in male groups. E, F Quantification of BNIP3 dimer levels in females and males, respectively. Our investigation showed higher BNIP3 dimer level in the cortex (P < 0.0001) and the hippocampus (P < 0.05) of the 3xTg-AD male group and also in the cortex (P < 0.05) of 3xTg-AD female compared to the control. Statistical analyses were performed using a two-way ANOVA test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. The results are presented as the mean ± SD, n = 5 per group (n = 5 females and n = 5 males).

Fig. 4. Mitophagy-related BCL2-family protein status (BCL2L13) in control and 3xTg-AD mice.

A, B Western blot analysis of BCL2L13 in female and male mice. C Quantification of the expression levels of BCL2L13 in the cortex and hippocampus of female 3xTg-AD and control mice. BCL2L13 expression in the cortex (P < 0.0001) and the hippocampus (P < 0.01) of 3xTg-AD females is higher than in control. D Quantification of the expression levels of BCL2L13 in the cortex and hippocampus of male 3xTg-AD and control mice. No significant differences were found in male groups. Statistical analyses were performed using a two-way ANOVA test. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. The results are presented as the mean ± SD, n = 5 per group (n = 5 females and n = 5 males).

Fig. 5. Transmission electron microscopy reveals mitochondrial and mitophagosome changes in 3xTg-AD mice in the cortex and hippocampus.

A Transmission electron microscopy images of the cortex and hippocampus in male and female control and 3xTg-AD mice. Arrows indicate mitophagosomes and mitochondria. B, C Quantification of mitochondria in the cortex of female and male mice, respectively. The lower number of mitochondria in the cortex (P < 0.001) and hippocampus (P < 0.0001) of 3xTg-AD females indicates potentially higher mitophagy rates in the AD compared to the control female group. In males, the potential mitophagy rate in the cortex of the 3xTg-AD group is significantly higher than the control (P < 0.01). D, E Quantification of mitophagosomes in the cortex of female and male mice, respectively. A significant increase in the mitophagosome number of the cortex in male 3xTg-AD compared to the control (P < 0.01) indicates a potentially higher mitophagy rate (or lower mitophagosome degradation) in male 3xTg-AD compared to the control. Similarly, in females, the number of mitophagosomes is significantly increased in the cortex (P < 0.0001) and the hippocampus (P < 0.0001) of 3xTg-AD compared to the control group, indicating a potentially higher mitophagy rate (or lower mitophagosome degradation) in the cortex and hippocampus of AD compared to control. Statistical analyses were performed using a two-way ANOVA test. *P < 0.05; **P < 0.01; ***P < 0.001; ns, non-significant. The results are presented as the means ± SD.

Fig. 6. Behavioral assessment of cognitive performance in 3xTg-AD mice using novel object recognition (NOR) and novel object placement (NOP) tests.

A Memory impairment (NOR test) comparison between female and male control and 3xTg-AD mice. Females exhibited significant non-spatial memory deficits compared to 3xTg-AD males (P < 0.01) and the female control group (P < 0.0001). B Spatial memory impairment (NOP test) comparison between female and male control and 3xTg-AD mice. Our analysis showed no significant differences in spatial memory between male and female 3xTg-AD mice and controls. Statistical analyses were performed using a two-way ANOVA test. **P < 0.01; ****P < 0.0001; ns, non-significant. Data are presented as mean ± SD, n = 5 per group (n = 5 females, n = 5 males).

Fig. 7. Association of autophagosomes with memory impairment.

A, B Association of LC3B-II levels (autophagosomes) with the NOP and NOR test in males, respectively. In females, no association was found between autophagosome and spatial memory impairment (P ˃ 0.05; see the supplement). In male mice, a negative correlation was observed between LC3B-II levels in the cortex and the NOP test, suggesting lower autophagosome levels are associated with better memory performance (NOP) (P = 0.010). Conversely, higher LC3B-II levels in the cortex of male 3xTg-AD are associated with greater memory (P = 0.010). Spearman’s correlation test was used for statistical analysis, with P-value < 0.05 considered significant. LC3B-II, microtubule-associated protein 1 light chain 3 beta-II; NOR, novel object recognition; NOP, novel object placement.

Fig. 8. Association of BCL2 family proteins expression involved in mitophagy and mitochondria and mitophagosome numbers with memory impairment.

A Correlation of NOR test with the levels of BCL2L13 in the cortex and hippocampus of females. Our results show that higher hippocampal expression of BCL2L13 is correlated with greater non-spatial memory in 3xTg-AD females (P = 0.037). B Correlation of NOP test with the levels of BNIP3L dimer in the cortex and hippocampus of females. In female 3xTg-AD mice, the NOP test showed a significant positive correlation with BNIP3L dimer levels in the cortex (P = 0.037). C Correlation of NOR test with mitochondria number in the cortex and hippocampus of female 3xTg-AD mice. Analysis shows a lower mitochondria number in the cortex correlated with greater non-spatial memory in female 3xTg-AD mice (P = 0.010). D Correlation of NOR test with mitophagosome number in the cortex and hippocampus of female 3xTg-AD mice. The number of mitophagosomes in the cortex was significantly positively correlated with the NOR behavior of females (P = 0.041). Spearman’s correlation test was used for statistical analysis, with a P-value < 0.05 considered statistically significant. BNIP3L, BCL2/adenovirus E1B interacting protein 3-like; BCL2L13, BCL2 like 13; NOR, novel object recognition; NOP, novel object placement.

Fig. 9. Scatter plots of predicted cortex and hippocampus values are shown as a function of their respective actual values, based on behavioral features coupled with machine learning.

A, B Prediction of mitophagosomes in the cortex and hippocampus. C, D Prediction of mitochondria in the cortex and hippocampus. Favorable prediction accuracy for the number of mitophagosomes in the cortex and hippocampus, as well as for the number of mitochondria in the cortex.

Scheme 1

Summary of the method: In this study, 11-month-old male and female 3xTg-AD and C57BL/6 mice were subjected to behavioral tests, including Novel Object Recognition and Object Placement tests, to assess cognitive performance. Brain tissues were extracted for protein analysis using western blotting and for the quantification of mitochondria and mitophagosomes through transmission electron microscopy. These steps aim to examine autophagy and mitophagy in relation to AD progression. The entire experimental workflow is summarized in Scheme 1 for a comprehensive visual overview.