BECN1^F121A mutation increases autophagic flux in aged mice and improves aging phenotypes in an organ-dependent manner

ABSTRACT

Macroautophagy/autophagy is necessary for lifespan extension in multiple model organisms and autophagy dysfunction impacts age-related phenotypes and diseases. Introduction of an F121A mutation into the essential autophagy protein BECN1 constitutively increases basal autophagy in young mice and reduces cardiac and renal age-related changes in longer lived Becn1^F121A mutant mice. However, both autophagic and lysosomal activities decline with age. Thus, whether autophagic flux is maintained during aging and whether it is enhanced in Becn1^F121A mice is unknown. Here, we demonstrate that old wild-type mice maintained functional autophagic flux in heart, kidney and skeletal muscle but not liver, and old Becn1^F121A mice had increased autophagic flux in those same organs compared to wild type. In parallel, Becn1^F121A mice were not protected against age-associated hepatic phenotypes but demonstrated reduced skeletal muscle fiber atrophy. These findings identify an organ-specific role for the ability of autophagy to impact organ aging phenotypes.

Introduction

Autophagy is an evolutionarily conserved lysosomal degradative process essential for the maintenance of cellular homeostasis and the promotion of cell survival under stress conditions. As a result, dysregulated or diminished autophagy activity impacts a wide variety of diseases, including age-related diseases, as well as aging [1,2]. Indeed, multiple lines of evidence over the past 30 years have demonstrated that autophagy activity declines with age in diverse organisms [3]. Lysosomal protease activity is reduced in aged C. elegans [4] and defective lysosomes and autophagic vacuoles accumulate with age in rodent liver [5,6]. Moreover, the expression of several macroautophagy/autophagy genes decreases over time in Drosophila [7–9]. Likewise in mammals, levels of the essential autophagy proteins LC3, ATG5, and ATG7 decline with age in mouse brain as well as both mouse and human muscle [10,11]. In addition to declining during normal aging, expression of ATG proteins is also reduced in the setting of age-related disorders such as cardiomyopathy, neurodegenerative diseases, and osteoarthritis [12,13]. Genetic studies in organismal models have provided more direct evidence of the importance of autophagy in longevity and have confirmed the original finding that knockdown of the autophagy gene bec-1 (encoding BECN1) abrogates the lifespan extension of long-lived mutant worms [14,15]. Similarly, loss-of-function genetic studies indicate that autophagy is essential for lifespan extension in long-lived flies [3] and reciprocally, induction of autophagy by the overexpression of ATG8 and ATG1 increases lifespan in flies [7,16]. Since the deletion of autophagy genes results in neonatal lethality in mice [17] and systemic deletion of autophagy genes in adult mice results in early lethality [18], genetic studies of autophagy and aging in mammals have taken advantage of mice with tissue-specific deletion of Atg genes [11,19–23]. In these models, decreased autophagy results in multiple defects including the accumulation of dysfunctional organelles and protein aggregates that are also found in aging tissues of otherwise non-genetically modified animals [24].

The mammalian protein BECN1, an ortholog of the yeast protein Vps30/Atg6, is an essential component of the class III phosphatidylinositol 3-kinase (PtdIns3K) complex that promotes initiation of autophagosome formation. The function of BECN1 in autophagy is negatively regulated by binding to BCL2. Disruption of the BECN1-BCL2 complex enhances the lipid kinase activity of the BECN1-PtdIns3K complex and subsequent induction of autophagy [25–27]. Transgenic mice bearing a Becn1^F121A knockin (KI) mutation that disrupts BECN1 binding to BCL2 represent a unique autophagy gain-of-function mouse model that provided genetic evidence that mice with constitutively increased autophagy have an extended lifespan and improved cardiac and renal aging [28,29]. In addition, constitutively increased autophagy in Becn1^F121A KI mice also prevented the age-related decline in neurogenesis and olfaction [30]. Whether autophagy declines in all mouse tissues equally during aging and whether increased autophagy could alleviate the age-associated dysfunction of all tissue/organ during aging is unknown. Moreover, a recent study monitoring autophagy in C. elegans during aging demonstrated that autophagic flux is reduced with age in all tissues/organs but lifespan extension by different interventions relies on autophagy in a tissue-specific manner [31]. Thus, whereas increased autophagy has been demonstrated to improve mouse lifespan and healthspan at a whole-body level, the tissue-specific regulation of autophagy during mammalian aging remains to be explored.

In this study, we determined the impact of the BECN1^F121A KI mutation on autophagic flux in the liver, heart, kidney, and skeletal muscle of old mice and investigated if age-related phenotypes are improved in tissues of longer lived Becn1^F121A KI mice. We find that old Becn1^F121A KI mice have increased autophagic flux in heart, kidney, and skeletal muscle but not in liver. Importantly, we show that liver aging phenotypes were not impacted by the Becn1^F121A KI allele, whereas increased autophagic flux in the skeletal muscle of Becn1^F121A KI mice is associated with improved skeletal muscle aging phenotypes. Taken together, we identify an organ-specific role for the Becn1^F121A KI allele in the context of mammalian tissue aging.

Results

We previously reported that the Becn1^F121A KI homozygous mice that have a constitutive increase of basal autophagy have extended lifespan and delayed age-related cardiac and renal pathological phenotypes [28]. To further investigate if the KI mutation could delay aging phenotypes of other organs, we measured lipid accumulation and fibrosis in aged KI and control (WT) mice. Using 20-months-old WT and KI mice, we analyzed H&E-stained tissue sections (Figure 1A) and quantified the percentage of liver area covered by lipid droplets. Contrary to our expectations, there was no statistically significant difference in lipid accumulation between old KI and WT mice, despite a slight trend toward a greater number of KI mice with less lipid content (Figure 1B). To confirm this result, we measured hepatic triglycerides and observed no difference between old KI and WT mice (Figure 1C). We also measured the serum level of GPT/ALT (glutamic pyruvic transaminase, soluble), a biomarker of liver injury, and again did not detect a significant difference between old KI and WT (Figure 1D). Next, we evaluated age-related hepatic fibrosis on sections stained with Masson’s trichrome (Figure 1E,F) and similarly, did not observe a statistically significant difference in liver fibrosis between old KI and WT mice. These results indicated that the KI mutation did not protect mice from age-related hepatic pathological changes and led us to ask if the level of autophagy was still higher in the livers of old KI mice compared to old WT mice. Indeed, we previously demonstrated that the KI mutation increased basal autophagy in mouse livers in 6-months-old young adult mice though it has been well characterized that hepatic autophagy declines with age [28,32]. To monitor autophagy in old mice, we studied KI and WT mice that had been crossed to transgenic mice expressing GFP tagged LC3 [33,34]. In the livers of old mice, we observed no difference in the number of GFP-LC3 puncta between KI and WT mice (Figure 1G,H). To assess autophagic flux, we treated mice with the autophagy inhibitor chloroquine (CQ). In CQ treated old mice, neither KI nor WT mice had a statistically significant increase in accumulation of GFP-LC3 puncta in the livers compared to untreated mice, and similarly, we did not observe a significant difference in flux when comparing the livers of old KI and WT mice. We also confirmed that hepatic autophagic flux was increased by the KI mutation in young mice, and that hepatic autophagic flux was reduced with age, as the number of GFP-LC3 puncta greatly declined in the liver of old mice compared to young ones in both WT and KI mice both in the presence and absence of CQ (Fig. S1). There was also no change in the protein expression of autophagy receptor and substrate SQSTM1/p62 in the livers of old KI and WT mice as evaluated by western blot, thus confirming similar autophagic flux between old KI and WT mice in the liver (Figure 1I). Taken together, these results demonstrate that in contrast to the livers of young mice, old Becn1^F121A KI mice did not display increased liver autophagy, which is correlated with the inability of the KI mutation to impact liver aging phenotypes.

Figure 1.

BECN1^F121A mutation does not improved liver age-related phenotype and does not increase autophagic flux in old mice. (A) Representative images of H&E-stained liver sections of old Becn1^+/+ (WT) and Becn1^F121A/F121A (KI) mice. Scale bars: 40 μm. (B) Percentage of old mice with lipid droplets covering liver section at the indicated percentage of covered liver section area (n = 19 for WT (8 Females (F) + 11 Males (M), n = 24 for KI (11 F +13 M)). (C) Quantification of triglycerides (TG) in the liver of old WT (n = 7 M) and KI (n = 7 M) mice. Data are mean ± s.e.m. (D) GPT/ALT enzyme values in the serum of old WT (n = 7 M) and KI (n = 9 M) mice. (E) Representative images of old WT and KI mice liver sections stained with Masson trichrome to quantify fibrosis visualized by the presence of collagen in light blue color. Scale bars: 40 μm. (F) Distribution of old mice according to their liver fibrosis score using the following score: 0 for absence of damage; 1 for ≤1% tissue area; 2 for 1–5% tissue area; 3 for ≥5% tissue area with fibrosis (n = 12 for WT and KI (4 F +8 M)). (G) Representative images of GFP-LC3 puncta indicative of autophagosomes in the liver of old WT and KI mice that transgenically express GFP-LC3, with or without chloroquine (CQ) for 6 h. Scale bars: 10 μm. Arrows indicate autophagosomes. (H) Quantification of GFP-LC3 puncta with or without CQ in old WT and KI mice. Data are mean ± s.e.m. (n = 7 for WT (4 F +3 M) and n = 9 (5 F +4 M) for KI without CQ and n = 10 for WT (6 F +4 M) and n = 8 for KI (4F +4 M) with CQ). P values were determined by a two-sided unpaired t-test. (I) Western blot analysis of SQSTM1/p62 autophagy marker and actin in the liver of old WT and KI mice. Shown are representative western blots of 3 independent experiments.

We next investigated whether in other organs, the increase in basal autophagy in Becn1^F121A KI mice was also impaired with age. As age-related cardiac and renal aging is prevented in the KI mice [28], we measured autophagic flux in the heart and kidneys of old KI and WT mice expressing GFP-LC3. Compared to old WT mice, old KI mice had significantly more GFP-LC3 puncta in the heart (Figure 2A,B), renal glomeruli (Figure 2D,E) and renal proximal convoluted tubules (PCT) (Figure 2F,G). We observed similar differences in GFP-LC3 puncta in the hearts and kidneys of old KI vs WT mice treated with CQ to block autophagic flux. In contrast to liver (Figure 1), CQ treatment further increased the number of puncta in the hearts and kidneys of old KI and WT mice, indicating that autophagic flux is intact in hearts and kidneys of 22-months-old mice. As in the liver, we confirmed that autophagic flux declined with age in the heart and kidney of WT and KI mice as the number of GFP-LC3 puncta was greatly reduced in old mice compared to young mice both in the presence and absence of CQ (Fig. S2). Finally, we found that the protein level of the autophagy substrate SQSTM1/p62 is decreased in the heart and kidney of old KI mice compared to old WT mice (Figure 2C,H). Thus, in contrast to liver, hearts and kidneys of old KI mice demonstrate greater autophagic flux compared to WT mice even at an advanced age.

Figure 2.

Autophagic flux is maintained in the heart and kidneys of old mice and is further increased by BECN1^F121A mutation. Representative images of GFP-LC3 puncta indicatives of autophagosomes in the heart (A), in the kidney’s glomeruli (D) and proximal convoluted tubules (PCT) (F) of old WT and KI mice that transgenically express GFP-LC3, with or without chloroquine (CQ) for 6 h. Scale bars: 10 μm. Quantification of GFP-LC3 puncta with or without CQ in the heart (B), in the kidney’s glomeruli (E) and PCT (G) of old WT and KI mice. Data are mean ± s.e.m. (n = 7 for WT (4 F +3 M) and n = 9 (5 F +4 M) for KI without CQ in all tissues and n = 9 (5 F +4 M) for WT and n = 8 for KI (4F +4 M) with CQ). P values were determined by a two-sided unpaired t-test. Western blot analysis of SQSTM1/p62 autophagy marker and actin, in the heart (C) and in the kidneys (H) of old WT and KI mice. Shown are representative western blots of 3 independent experiments.

To further investigate the potential of the BECN1^F121A KI mutation to increase autophagy during aging, we next focused on skeletal muscle. Using KI and WT mice expressing GFP-LC3, we observed that old KI mice had a statistically significant increased number of GFP-LC3 puncta in skeletal muscle compared to WT mice (Figure 3A,B). Old KI mice treated with CQ also had a higher number of GFP-LC3 puncta than WT mice, indicating that autophagic flux is increased in skeletal muscle of old KI mice (Figure 3B). It is worth noting that, like both heart and kidney, autophagic flux in the skeletal muscle of old mice remains intact as both WT and KI old mice treated with CQ had more GFP-LC3 puncta than untreated mice (Figure 3B). The expression level of SQSTM1/p62 is also lower in the muscle of old KI mice not treated with CQ compared to WT (Figure 3C). We also showed that autophagic flux declined with age in the skeletal muscle of both WT and KI mice but to a lesser extent than in the heart, kidney, and liver (Fig. S3A). Altogether, these data indicate that increased autophagic flux in skeletal muscle of KI mice is sustained in old age.

Figure 3.

BECN1^F121A mutation increases autophagic flux in the skeletal muscle of old mice and prevents age-related decrease muscle fiber size. (A) Representative images of GFP-LC3 puncta indicatives of autophagosomes in the vastus lateralis of old WT and KI mice that transgenically express GFP-LC3, with or without chloroquine (CQ) for 6 h. (B) Quantification of GFP-LC3 puncta with or without CQ in old WT and KI mice. Data are mean ± s.e.m. (n = 7 for WT (4 F +3 M) and n = 9 for KI (5 F +4 M) without CQ and n = 9 for WT (5 F +4 M) and n = 8 for KI (4F +4 M) with CQ). (C) Western blot analysis of SQSTM1/p62 autophagy marker and ACTB, in the liver of old WT and KI mice. Shown are representative western blots of 3 independent experiments. (D) Cross-sectional area (CSA) of skeletal muscle fibers of the vastus lateralis muscle of young (5-months-old) and old (20-months-old) WT and KI mice (n = 3 per group for young mice and n = 5 per group for old mice, all males). Graphs represents median CSA and interquartile and all myofiber CSA values are shown. (E) Representative images of vastus lateralis skeletal muscle sections of young and old WT and KI mice stained with a LAMA2/laminin antibody to outline the myofibers and DAPI. (F) Frequency of distribution of old WT and KI mice muscle fibers CSA (n = 5 mice per group). Data are presented as histograms of fiber size per CSA bin with 700 μm² width. Scale bars: 10 μm. P values were determined by a two-tailed ANOVA with correction for multiple comparisons.

As increased autophagic flux in the heart and kidneys of old KI mice (Figure 2) correlates with improved cardiac and renal aging phenotypes [28], we next investigated if old KI mice also had improved skeletal muscle aging phenotypes. One characteristic age-related change in the skeletal muscle is myofiber atrophy [35,36]. Indeed, when we measured the cross-sectional area of skeletal muscle fibers in young and old mice, we observed a clear age-related decrease in the muscle fiber size in WT mice (Figure 3D,E). While KI mice also showed an age-related decrease in myofiber size, the impact of aging was much less than in WT mice (Figures 3D,E and S3B). Median myofiber cross-sectional area was significantly higher in old KI compared to old WT mice, although median myofiber cross-sectional area was similar between young KI and WT mice (Figure 3D). Another characteristic of muscle aging is an increased heterogeneity in myofiber size, and this phenotype was more evident in old WT than in old KI mice (Figures 3D,F and S3B). When we analyzed the frequency distribution of cross-sectional area of myofibers, we observed a clear shift toward larger fiber sizes in old KI compared to WT mice (Figure 3F) suggesting that KI mice were protected from age-related myofiber atrophy. Indeed, whereas the frequency distribution of myofiber size was markedly shifted toward smaller fibers in old WT mice compared to young WT mice, this age-related difference in fiber size was less evident in KI mice (Fig. S3B). Thus, we conclude that old KI mice have increased autophagic flux and delayed muscle aging compared to old WT mice.

Discussion

Our study indicates that in old mice, BECN1^F121A KI mutation increases autophagic flux in a tissue-specific manner and reduces aging phenotypes in the corresponding tissues/organs. Disruption of BECN1 binding to BCL2 by the KI mutation results in constitutively increased autophagic flux in all young mouse tissues explored to date: heart, kidneys, liver, skeletal muscles, mammary gland, adipose tissue, pancreas, and brain [28,29,37,38]. However, as lysosomal function and autophagy activity have been described to decline with age, autophagic flux might also be inhibited. As a result, measuring autophagy by direct quantification of autophagy markers such as SQSTM1/p62 and LC3 by immunoblot or fluorescence imaging under basal conditions may not properly reflect the autophagy level in aged animals and tissues [3,4]. Here, we investigated if autophagic flux remains increased by the Becn^F121A KI mutation in aged animals transgenically expressing GFP-LC3 and treated or not with the lysosomal inhibitor, chloroquine. Our data indicate that the constitutive increase in autophagic flux is maintained throughout aging in some tissues such as heart, kidney, and skeletal muscle but not in others like the liver, thus revealing an unexpected tissue-specificity in the upregulation of autophagy during mouse aging.

Our results also demonstrate that although autophagy declines with age, old WT mice maintain an active autophagic flux in some organs like heart, kidney, and skeletal muscle but not liver and this autophagic flux can be increased further in old mice via expression of the BECN1^F121A mutant protein. In aged kidney, autophagic flux is functional and can be further increased by the BECN1^F121A KI mutation in both proximal convoluted tubules (PCT) and glomeruli. Our result is consistent with a previous study that described an active autophagic flux in kidney PCT of old mice [39]. However, we observed a decline in autophagic flux in the PCT of old mice compared to young mice which is in opposition to this study which found that old mice have higher autophagic flux than young mice in the PCT [39]. These conflicting results could potentially be explained by the techniques used to quantify autophagic flux. Where Yamamoto et al. used anti-LC3 staining, we used GFP-LC3 transgenic mice and also took great care to exclude from quantification the fluorescent lipofuscin aggregates that accumulate with age especially in the PCT, as discussed in greater detail below [39]. As the age-related renal phenotypes that are exacerbated in autophagy deficient mice are improved in old Becn1^F121A KI mice, increasing autophagy could represent a potential strategy to alleviate age-related kidney diseases [28,38].

Autophagic flux was also increased in the hearts of old Becn1^F121A KI mice, which is consistent with a recent observation that exercise increases autophagic flux in old mice [40]. However, in their study, the authors did not detect any flux under basal resting condition, whereas our data indicated intact autophagy flux in old WT mice [40]. This difference could be explained by the different assays used to measure flux. Although both studies used chloroquine to block lysosomal degradation, in the exercise study autophagy and flux were determined by immunoblot of LC3 and SQSTM1/p62, whereas we used quantification of GFP-LC3 puncta combined with SQSTM1/p62 analysis. Nevertheless, whether through genetic intervention via BECN1^F121A KI mutation, or a physiological intervention via exercise, increased autophagic flux in the heart correlates with decreased cardiac aging in mice [28,40]. Similarly, our data show that in skeletal muscle, autophagic flux is not only active in aged mice but also higher in aged Becn1^F121A KI mice and correlates with improved skeletal muscle aging. Previous studies have shown that autophagy deficiency in skeletal muscle leads to muscle loss and accumulation of protein aggregates which resemble accelerated muscle aging phenotypes in mice [11]. Likewise, caloric restriction improves skeletal muscle aging phenotypes and increases the number of autophagosomes, although flux was not assessed [41]. Our results demonstrate that increased autophagy in old Becn1^F121A KI mice reduces age-associated skeletal muscle fiber atrophy and provide supportive evidence for inducing autophagy as a potential therapeutic strategy to mitigate sarcopenia, the age-related decrease in skeletal muscle mass and strength.

In contrast to cardiac, renal and skeletal muscle aging, which were all improved in old Becn1^F121A KI mice, hepatic aging was not improved by the BECN1^F121A KI mutation. The inability of the KI mutation to affect age-related liver damage phenotypes correlates with the absence of functional autophagic flux in old mice and the loss of autophagy induction in old KI mice compared to WT mice. As opposed to the heart, kidney, and skeletal muscle, the constitutive increase in basal autophagy observed in all young KI mice tissues, was not observed in the liver of old KI mice. Of note, in contrast to our findings, caloric restriction was shown to increase autophagic flux in the liver of old mice though this observation was specific to female C57BL/6J mice and may reflect the different techniques used to measure flux as liver autophagy flux was evaluated ex vivo on hepatocytes, whereas we monitored liver autophagy flux in vivo on old GFP-LC3 mice treated or not with chloroquine [42].

That autophagy flux declines with age was highlighted by the observation of decreased lysosomal activity and accumulation of autophagic and lysosomal vesicles in the livers of old mice and rats [5,6]. Although we observed age-associated autophagy decline in all the organs we investigated, autophagy flux was abolished only in the liver of WT and KI mice. This result suggests that the inability of KI mutation to increase autophagy in the liver might be linked with a general block of autophagy flux in the liver of old mice. One possible explanation is that the dysfunctional autophagic flux in liver of old mice prevents further increase in autophagy by the KI mutation. Autophagy plays an essential role in the liver since autophagy deficiency in the liver leads to multiple pathologic outcomes such as hepatomegaly, mild injuries, and the accumulation of abnormal organelles and lipid droplets in young mice [22,43,44]. In addition, deficiency in chaperone mediated autophagy accelerates liver aging [45].

During aging, defects in the recruitment of motor proteins lead to the inefficient trafficking of lysosomes and autophagosomes in the liver of old mice [46]. Further studies are needed to investigate the extent to which this vesicular trafficking defect observed in the liver can be generalized to other tissues. The age-related decrease in vesicular trafficking in the liver could be one mechanism to explain the inability of KI mutation to increase autophagy flux in the liver of old mice and particularly if the liver is more sensitive to age-related motor protein defects than other tissues. In our study, the number of GFP-LC3 autophagic vesicles was greatly decreased in the liver of old mice compared to young mice, indicating that autophagosomes are not massively accumulating in old mice and suggesting that other mechanisms besides trafficking and lysosomal defects may account for the absence of autophagy flux in the liver of old KI and WT mice. Intriguingly, although we also observed the decline in autophagic flux with age in old mice, our results diverge from a comprehensive C. elegans study of autophagy flux during aging that observed both a decline in autophagic flux and an accumulation of autophagic vesicles in various tissues in worms during aging [31]. Further studies are needed to determine the relative contributions of mechanisms like lysosomal dysfunction and vesicular trafficking defects to the regulation of autophagy during aging and whether they are regulated differently in diverse species and tissues. To that end, signaling pathways that are independent of BECN1 function such as the mTOR pathway that is widely involved in the regulation of autophagy and longevity might be differentially regulated in each organ which would potentially explain the liver-specific phenotype observed here [47]. Alternatively, even if the BECN1^F121A KI mutation results in increased initiation and maturation of autophagosomes, if the age-related defects in autophagosome biogenesis and elongation that have been described in neurons also involve other cell types and certain organs in particular such as the liver, autophagy flux may not be enhanced by the KI mutation in the liver of old mice [48].

Our work demonstrates that autophagy is differentially regulated in each organ during aging. In the brain, although dysfunctions in autophagy flux have also been described during aging, our previous results indicate that age-related decline in autophagy is partially reversed in neural stem cells of 18-month-old Becn1^F121A KI mice [30,49]. Expanding studies of autophagic flux and lysosomal dysfunctions to different cell populations of the brain and other organs of old mice will improve our understanding of the tissue-specificity of autophagy regulation during mammalian aging. Interestingly, though the BECN1^F121A KI mutation did not improve autophagic flux or the age-associated changes in the liver of old mice, the benefit of increased autophagy on other organs is sufficient to extend their lifespan [28].

Of note, our previous work demonstrated that KI mice have decreased age-related spontaneous malignancies, however solid tumors such as liver and lung cancer were rarely observed in both WT and KI 20-month-old mice [28]. Thus, the organ-specific impact of autophagy dysregulation on spontaneous tumorigenesis during aging would be interesting to explore.

Finally, on a technical note, in our fluorescence imaging and analyses, we observed an age-dependent accumulation of lipofuscin aggregates in all tissues, some of which appeared punctate. Accumulation of such aggregates is a well-established hallmark of aging that accumulates over time in all tissues [50]. As lipofuscin fluorescence has a broad emission spectrum that overlaps with GFP, it is important to account for the presence of small, punctate lipofuscin aggregates when quantifying true GFP-LC3 puncta to avoid overestimating the number of GFP-LC3 autophagic vesicles in old tissues [50].

To our knowledge, our study is the first to evaluate autophagic flux in multiple tissues of old mice and highlights the importance of establishing a systemic evaluation of autophagic flux in aging mammals. We also demonstrate that increased autophagic flux in some old Becn1^F121A KI mouse tissues correlates with improved aging phenotypes in a tissue-specific manner. Overall, our data suggest that increasing autophagic flux during aging mitigates age-related phenotypes in multiple tissues.

Materials and methods

Mice

Becn1^F121A/F121A knock-in mice were generated in Beth Levine’s lab and backcrossed for more than 12 generations to C57BL/6J mice (Jackson Laboratories) as described [28,29]. Becn1^+/+ (WT) and Becn1^F121A/F121A (KI) littermate mice were crossed with GFP-LC3 transgenic C57BL/6J animals [34] and tissues of offspring were used for autophagic flux analyses. Old mice were 20- to 22-months-old littermates and young mice were 3- and 5-months-old. Both males and females were used for all analyses. All animal procedures were performed in accordance with institutional guidelines and with approval from the UT Southwestern Medical Center Institutional Animal Care and Use Committee.

Autophagy analyses

To assess the autophagic flux in aged mice, 22-months-old homozygous Becn1^+/+;GFP-LC3 or Becn1^F121A/F121A;GFP-LC3 mice were synchronized by a 16 h starvation followed by 3 h of feeding before treatment with either PBS (Sigma, D8537) or chloroquine (50 mg kg⁻¹; Sigma, C6628) for 6 h. Mice were then perfused with 4% paraformaldehyde (PFA) in PBS and tissues were collected and processed for frozen sectioning as described [28]. The mouse heart, liver, vastus lateralis skeletal muscle and kidney tissue sections were imaged using a 40× objective on a Zeiss AxioPlan 2 microscope. For each tissue, the total number of GFP-LC3 puncta was counted per 2,500 μm² area (more than 20 randomly chosen fields were used per mouse) and was determined by an observer blinded to genotype. The average value for each tissue for each mouse was then calculated and graphed. For western blot analysis, tissues were lysed in ice-cold lysis buffer (300 mM Tris-HCl, pH 8, 2% SDS) with complete mini protease (Roche, 11836153001) and Halt phosphatase (Thermo Scientific, 78420) inhibitor cocktails for 30 min at 4°C and the lysates were then centrifuged at 15,000g for 10 min. Cleared lysates were diluted in 2× SDS – PAGE loading buffer and submitted to western blotting using anti-SQSTM1/p62 (Progen, GP62-C; 1:1,000 dilution), anti-LC3B (Sigma, L7543; 1:10,000 dilution), anti-BECN1 (Santa Cruz Biotechnology, sc-7382; 1:500 dilution), anti-BCL2 (Santa Cruz Biotechnology, sc-7382; 1:200 dilution) and anti-ACTB/actin (Santa Cruz Biotechnology, sc -47778; 1:5,000 dilution) antibodies.

Histopathological analyses

Mice were perfused with 4% PFA in PBS before tissue collection, fixation, and preparation of paraffin-embedded sections for histopathological analyses. Liver sections were stained with Hematoxylin and Eosin (H&E) then scanned using NanoZoomer 2.0-HT and analyzed using free NDPView2 software. To determine the lipid accumulation in the liver, each field of H&E-stained liver sections was given a score using the following four categories: ≤5% tissue area; ≤25% tissue area; ≤50% tissue area; ≥50% tissue area with lipid droplets visualized as white empty vesicles on liver sections. For analyses of hepatic fibrosis, liver sections were stained with Masson’s trichrome according to the manufacturer’s instructions (Abcam, ab150686) and the sections were imaged using a 20× objective on a Zeiss AxioPlan 2 microscope. Ten random fields were evaluated per mouse and each field was given a fibrosis score using the following scale: 0, absence of damage; 1, ≤1% tissue area; 2, 1–5% tissue area; 3, ≥5% tissue area with fibrosis. The scores of each field were averaged to give a final fibrosis score for each mouse, ranging from 0 to 3. Quantification of all histopathological analyses was performed by an observer blinded to genotype.

Hepatic triglycerides and serum GPT/ALT measurements

Hepatic triglycerides quantification was performed on 100 mg of liver homogenate according to the manufacturer’s instructions (Sigma, MAK266). The GPT (glutamic pyruvic transaminase, soluble) activity assay was performed on 2 μl of serum according to the manufacturer’s instructions (Sigma, MAK052).

Muscle fiber size analyses

Skeletal muscle sections were staining with anti-LAMA2/laminin-2 (Sigma, L0663; 1:1000 dilution) to outline the muscle fibers. The average cross-sectional area of vastus lateralis muscles was determined using Myosight plugin [51] for FIJI (Just ImageJ) software. For old mice, 260 to 270 muscle fibers per mouse and 5 mice per genotype were analyzed. For young mice, 200 muscle fibers per mouse and 3 mice per genotype were analyzed.

Statistical analyses

Data were analyzed using the GraphPad Prism 9 software. Two-tailed unpaired Student’s t-tests were used for analyses of autophagy. For the analysis of myofibers CSA, data were analyzed by two-way ANOVA with correction for multiple comparisons.

Funding Statement

The work was supported by the Fondation Leducq [15CBD04]; National Institute of Allergy and Infectious Diseases [U19AI142784].

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2022.2111852.