Chaperone-mediated autophagy degrades SERPINA1^E342K/α1-antitrypsin Z variant and alleviates cell stress

ABSTRACT

Chaperone-mediated autophagy (CMA) is a specific form of autophagy that selectively targets proteins containing a KFERQ-like motif and relies on the chaperone protein HSPA8/HSC70 for substrate recognition. In SERPINA1/a1-antitrypsin deficiency (AATD), a disease characterized by the hepatic buildup of the SERPINA1^E342K/ATZ, CMA’s role had been unclear. This work demonstrates the critical role that CMA plays in preventing SERPINA1^E342K/ATZ accumulation; suppressing CMA worsens SERPINA1^E342K/ATZ accumulation while activating it through chemical stimulation or LAMP2A overexpression promotes SERPINA1^E342K/ATZ breakdown. Specifically, SERPINA1^E342K/ATZ’s ₁₂₁QELLR₁₂₅ motif is critical for HSPA8/HSC70 recognition and LAMP2A’s charged C-terminal cytoplasmic tail is vital for substrate binding, facilitating CMA-mediated degradation of SERPINA1^E342K/ATZ. This selective activation of CMA operates independently of other autophagy pathways and alleviates SERPINA1^E342K/ATZ aggregate-induced cellular stress. In vivo administration of AR7 promotes hepatic SERPINA1^E342K/ATZ elimination and mitigates hepatic SERPINA1^E342K/ATZ aggregation pathology. These findings highlight CMA’s critical function in cellular protein quality control of SERPINA1^E342K/ATZ and place it as a novel target for AATD treatment.

Abbreviation: AR7: atypical retinoid 7; ATG16L1: autophagy related 16 like 1; AATD: SERPINA1/alpha-1 antitrypsin deficiency; CHX: cycloheximide; CMA: chaperone-mediated autophagy; CQ: chloroquine; ER: endoplasmic reticulum; GAPDH: glyceraldehyde-3-phosphate dehydrogenase; HSPA8/HSC70: heat shock protein family A (Hsp70) member 8; LAMP2A: lysosomal associated membrane protein 2A; LAMP2B: lysosomal associated membrane protein 2B; LAMP2C: lysosomal associated membrane protein 2C; MG132: carbobenzoxy-L-leucyl-L-leucyl-L-leucinal; PAS-D: periodic acid-Schiff plus diastase; SERPINA1/A1AT: serpin family A member 1; SERPINA1^E342K/ATZ: Z variant of SERPINA1; TMRE: tetramethyl rhodamine ethyl ester perchlorate.

Introduction

The autophagy-lysosome pathway is a major protein aggregate clearance mechanism that is conserved in all eukaryotic cells. This system can be classified into three main types based on their distinctive digestion processes: macroautophagy, microautophagy, and chaperone-mediated autophagy (CMA) [1]. Macroautophagy (referred to as autophagy) either selectively or non-selectively engulfs various substrates (including entire organelles and protein aggregates) within phagophores that mature into autophagosomes, whereas CMA specifically targets individual proteins containing the CMA-targeting KFERQ-like motif [2,3]. In the CMA pathway, HSPA8/HSC70 (heat shock protein family A (Hsp70) member 8) recognizes the KFERQ-like motif of a substrate to form a substrate-chaperone complex [4]; the complex is then delivered to the lysosome surface to interact with the CMA specific receptor LAMP2A (lysosomal associated membrane protein 2A). LAMP2A plays a critical role in delivering substrates directly to the lysosomal lumen for degradation [4,5]. Increasing evidence suggests that CMA plays a crucial role in various physiological processes; CMA malfunction has been linked to a variety of diseases including neurodegenerative illnesses, metabolic disorders, lysosomal storage disorders and cancer [3].

SERPINA1/Alpha one antitrypsin (serpin family A member 1) is a serum protease inhibitor primarily synthesized in the liver and serves as a pulmonary neutrophil elastase neutralizer to protect the lung from self-elastolytic digestion. Thus, it plays a critical role in maintaining proper pulmonary proteolytic-inhibitor balance [6,7]. SERPINA1 is a highly polymorphic protein, over 120 mutations in the SERPINA1 gene have been identified and most mutation caused disease pathogenesis manifests as pulmonary emphysema because the loss of function as elastase neutralizer. These mutated SERPINA1-driven manifestations are called alpha one antitrypsin deficiency (AATD). One genetic mutation results in a mutant protein, SERPINA1^E342K/ATZ, with an E342K substitution that exerts the most severe phenotype.

SERPINA1^E342K/ATZ adopts a misfolded conformation and is prone to aggregate into insoluble polymers; the abnormal SERPINA1^E342K/ATZ polymerization could result in liver cirrhosis [7,8]. Lung diseases resulting from AATD are treated with intravenous infusions of SERPINA1/A1AT from donated human plasma, but this therapy does not mitigate liver damage caused by hepatic SERPINA1^E342K/ATZ aggregates [9].

It has been established that, the ubiquitin-proteasome system and autophagy-lysosomal pathway represent the two major routes for SERPINA1^E342K/ATZ’s bio-digestion [10,11]. Enhancing autophagy using chemical compounds have all been shown to prohibit SERPINA1/A1AT protein accumulation and alleviate cell injury [12]. To date, there have been no studies exploring the role of CMA in the regulation of SERPINA1^E342K/ATZ catabolism. Theoretically, proteins with vKFERQ-like motif could be recognized by HSPA8/HSC70 and delivered for CMA degradation. Searching the sequence of SERPINA1^E342K/ATZ, there is a typical pentapeptide motif within the SERPINA1^E342K/ATZ sequence and those amino acids are exposed on the surface of the protein molecule. This observation leads us to speculate that SERPINA1^E342K/ATZ could be targeted for CMA, and manipulating CMA might offer a solution for SERPINA1^E342K/ATZ digestion, preventing substantial polymerization.

Results

CMA influence homeostasis of wild-type and Z variant of SERPINA1

In CMA, its activity is primarily modulated by changes in the levels of LAMP2A at the lysosomal membrane [3]. To evaluate the influence of CMA on SERPINA1^E342K/ATZ, we generated a stable cell line expressing GFP-SERPINA1^E342K/ATZ-FLAG. Consistent with previous studies, we observed that GFP-SERPINA1^E342K/ATZ-FLAG colocalizes with the endoplasmic reticulum (ER) in HEK293 cells (Fig. S1). These cells were subsequently transfected with incremental quantities of the LAMP2A-encoding plasmid. This resulted in a dose-dependent diminution of GFP-SERPINA1^E342K/ATZ-FLAG protein levels corresponding to the augmented expression of exogenous LAMP2A (Figure 1A, B). Simultaneously, a reduction in the standard CMA substrate GAPDH was observed, suggestive of enhanced CMA activity (Figure 1B).

Figure 1.

LAMP2A influence SERPINA1/α1-antitrypsin variant Z steady-state protein level. (A) GFP-SERPINA1^E342K/ATZ-FLAG-expressing HEK 293 cells were transfected with increasing amounts (1, 2, 3, 4, or 5 μg/mL of cell medium) of a vector expressing LAMP2A for 48 h. The fluorescence of GFP-SERPINA1^E342K/ATZ-FLAG was analyzed by flow cytometry. (B) HEK 293 cells stably expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with increasing concentrations (1, 2, 3, 4, or 5 μg/mL of cell medium) of vectors expressing LAMP2A for 48 h. The protein levels of GFP-SERPINA1^E342K/ATZ-FLAG relative to ACTB in cell lysates were quantitatively analyzed by measuring densitometry values from the blots. (C) HEK 293 cells were co-transfected with LAMP2A and either SERPINA1^E342K/ATZ or SERPINA1/A1AT. After 48 h, the cells were lysed using 0.1% triton X-100 as a detergent, and the lysates were divided into triton X-100 soluble cell lysates and triton X-100 insoluble cell pellet fractions. Equal aliquots of cell lysates (30 μg) were loaded onto SDS-PAGE and immunoblotted with antibodies against SERPINA1/A1AT and LAMP2A. Ctrl: pCDH empty plasmid. (D) HepG2 cells were transfected with empty or LAMP2A-expressing vectors. After 48 h of transfection, cell lysates were harvested and immunoblotted with the indicated antibodies. (E) HEK 293 cells stably expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with shRNA1 or shRNA2 to knock down LAMP2A protein expression for 48 h. The levels of GFP-SERPINA1^E342K/ATZ-FLAG protein in cell lysates were quantitatively analyzed. Cells transfected with non-targeted scramble-shRNA were used as the control. Ctrl: scramble shRNA, sh1: LAMP2A-shRNA1, sh2: LAMP2A-shRNA2. (F) HEK 293 cells stably expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with shRNA1 or shRNA2 to knock down LAMP2A expression for 48 h. The fluorescence of GFP-SERPINA1^E342K/ATZ was analyzed by flow cytometry. (G) HepG2 cells were transfected with shLAMP2A (shRNA2) to knock down LAMP2A expression for 48 h. The levels of SERPINA1/A1AT protein in cell lysates were quantitatively analyzed. (H) GFP-SERPINA1^E342K/ATZ-FLAG-expressing HEK 293 cells were transfected with empty or LAMP2A-expressing vectors. Cells were treated with cycloheximide (CHX; 10 μg/mL) for the indicated times, starting 24 h after transfection. The graph illustrates the quantitative changes in steady-state levels of GFP-SERPINA1^E342K/ATZ-FLAG relative to ACTB during the CHX treatment period. Data were presented as the mean ± SEM of 3 or more independent experiments. p<0.05.*; p<0.01.**; p<0.001.***; p<0.0001.****.

Due to its aggregation-prone properties, the mutant protein SERPINA1^E342K/ATZ leads to the formation of insoluble protein inclusions [13]. We thus resuspended the cell pellet in cell lysis buffer containing 2% SDS and dissect the CMA’s effect on insoluble SERPINA1^E342K/ATZ. Unlike SERPINA1/A1AT, we observed an aberrant amount of SERPINA1^E342K/ATZ in the insoluble fraction. However, with the expression of LAMP2A, the insoluble SERPINA1^E342K/ATZ level also significantly reduced (Figure 1C). Additionally, transfecting LAMP2A into the HepG2 hepatic cell line, which naturally expresses SERPINA1/A1AT, resulted in a significant decline in endogenous SERPINA1/A1AT levels upon LAMP2A overexpression (Figure 1D). This indicates that LAMP2A overexpression disrupts the homeostasis of both wild-type SERPINA1/A1AT and SERPINA1^E342K/ATZ.

To further validate the regulatory effect of LAMP2A on SERPINA1^E342K/ATZ, we attempted to manipulate the endogenous levels of LAMP2A. We designed two short hairpin RNAs (shRNAs) that specifically target the cytosolic tail of the LAMP2A transcript, a region that differs between LAMP2 gene isoforms A and B, C (Fig. S2A,B). After transfecting the cells expressing GFP-SERPINA1^E342K/ATZ with the two shRNA-coded vectors for 48 h, a significant (>70%) reduction in LAMP2A mRNA levels was observed compared to cells treated with a scrambled shRNA, while no noticeable changes were detected in the transcripts of LAMP2B and LAMP2C (Figure S2C, D). Moreover, in the same samples, we detected a notable increase in GFP-SERPINA1^E342K/ATZ protein levels (Figure 1E, F). Additionally, we found that shLAMP2A could also effectively downregulate LAMP2A expression in HepG2 cells, leading to a significant increase of SERPINA1/A1AT level (Figure 1G).

As the modulation of LAMP2A did not alter SERPINA1^E342K/ATZ transcription (Fig. S2E,F), we speculate that the changes in SERPINA1^E342K/ATZ levels may result from altered protein degradation dynamics. To confirm this, we blocked protein synthesis in the SERPINA1^E342K/ATZ expression cells with cycloheximide (CHX) 24 h after LAMP2A-coding plasmid transfection and monitored the remaining SERPINA1^E342K/ATZ at different time points post the drug treatment. The results indicated an accelerated degradation of SERPINA1^E342K/ATZ in cells with LAMP2A overexpression, as evidenced by a decrease in half-life from 8 to 6 h (Figure 1H). Additionally, overexpression of HSPA8/HSC70, another key component of CMA [4], also heightened the degradation of GFP-SERPINA1^E342K/ATZ in a dose-dependent manner, without affecting SERPINA1^E342K/ATZ transcription (Fig. S3A,B). These experiments collectively suggest that CMA plays a significant role in the proteostasis of both SERPINA1^E342K/ATZ and SERPINA1/A1AT.

SERPINA1^E342K/ATZ is degraded through the CMA pathway mediated by LAMP2A

The above assays have provided evidence supporting CMA’s participation in regulating SERPINA1^E342K/ATZ’s degradation, and we hypothesize that CMA directly targets SERPINA1^E342K/ATZ for clearance. To corroborate this hypothesis, we conducted an affinity-isolation assay of SERPINA1^E342K/ATZ-FLAG protein and confirmed the association with LAMP2A (Figure 2A). Additionally, we stained endogenous LAMP2A and examined its colocalization with SERPINA1^E342K/ATZ. We observed distinct CMA puncta, as indicated by LAMP2A, and a significant overlap between SERPINA1^E342K/ATZ and LAMP2A (Figure 2B). we also observed the colocalization of GFP-SERPINA1^E342K/ATZ with HSPA8/HSC70 (Fig. S3C).

Figure 2.

SERPINA1^E342K/ATZ undergoes degradation via the CMA pathway facilitated by LAMP2A. (A and C) HEK 293 stably expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with an empty vector and LAMP2A or HSPA8/HSC70 for 48 h, cells were lysed and subjected to immunoprecipitation with antibodies against IgG or flag. Aliquots of IP proteins and input proteins were analyzed by immunoblot. (B) HEK 293 cells with were transfected with LAMP2A and SERPINA1^E342K/ATZ-FLAG for 48 h. Immunofluorescence microscopy images of SERPINA1^E342K/ATZ-FLAG (red) and LAMP2A (green) in HEK 293 cells were obtained after staining with anti-LAMP2A and FLAG antibodies. Each nucleus was stained with DAPI (blue). The colocalization of LAMP2A and FLAG-SERPINA1^E342K/ATZ is indicated by the yellow coloration resulting from the overlap of red and green signals. Pixel intensity plots are displayed along the white lines in the adjacent images, with colors corresponding to those in the merged images. Scale bars: 10 μm. (D) GFP-SERPINA1^E342K/ATZ-FLAG-expressing HEK 293 cells were transfected with an empty vector or a vector encoding LAMP2A for 36 h. Subsequently, they were treated with CQ (10 μM) or DMSO for an additional duration of 4, 5, or 6 h. Following the treatment, the cells were lysed and underwent western blot analysis. (E) HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were treated with DMSO or QX77(10 μM) for 48 h. Cell lysates were immunoblotted with the indicated antibodies. (F) schematic depiction of a full-length LAMP2A construct and LAMP2A-mutant construct. TS: transmembrane structure. (G) GFP-SERPINA1^E342K/ATZ-FLAG-expressing HEK 293 cells were transfected with full-length LAMP2A, LAMP2A-mutant, or LAMP2A-truncate constructs for 48 h. The protein levels of GFP-SERPINA1^E342K/ATZ-FLAG in different transfections were quantitatively analyzed. Tr: LAMP2A truncate, mu: LAMP2A mutant. (H) HEK 293 cells were transfected with LAMP2A, LAMP2A-mutant, or LAMP2A-truncate along with FLAG-SERPINA1^E342K/ATZ for 48 h. The cell lysates were then subjected to immunoprecipitation (IP) using an anti-flag antibody, and the precipitates were analyzed by western blotting. Data were presented as the mean ± SEM of 3 or more independent experiments. p<0.05.*; p<0.01.**; p<0.001.***; ns, no significance. vs. indicated group.

The degradation of substrates by CMA relies on the acidic environment within lysosomes [14]. To investigate whether SERPINA1^E342K/ATZ degradation is dependent on the lysosome, we treated cells overexpressing LAMP2A with the lysosomal inhibitor chloroquine (CQ). We observed that CQ could block SERPINA1^E342K/ATZ degradation even in the presence of LAMP2A overexpression (Figure 2D). These data thus indicate the directed accessibility of CMA to SERPINA1^E342K/ATZ and its mediation of degradation depending on lysosomal hydrolysis activity. Consistent with this, we found that QX77 [15], a CMA activator, was able to boost LAMP2A’s level and eliminate the typical CMA substrate GAPDH, as well as GFP-SERPINA1^E342K/ATZ (Figure 2E).

The cytosolic tail of LAMP2A, comprising 12 amino acids, is essential for the recruitment of HSPA8/HSC70-substrate complexes to lysosomes, facilitating their interaction via positively charged residues [16,17]. To elucidate how LAMP2A recognizes SERPINA1^E342K/ATZ, we engineered plasmids for two LAMP2A variants: one encoding a truncated version lacking the C-terminal tail (LAMP2A-truncate), and another encoding a mutant wherein positively charged amino acids at positions 401 to 404 (₄₀₁KHHH₄₀₄) were substituted with alanines (₄₀₁AAAA₄₀₄) (Figure 2F and Fig. S4). These plasmids were introduced into cells expressing GFP-SERPINA1^E342K/ATZ. Our findings revealed that only the wild-type LAMP2A significantly reduced SERPINA1^E342K/ATZ levels, whereas both the LAMP2A-truncate and LAMP2A-mutant variants were ineffective (Figure 2G). Additionally, immunoprecipitation assays with Flag-tagged SERPINA1^E342K/ATZ demonstrated markedly reduced co-immunoprecipitation with the truncated and mutant forms of LAMP2A (Figure 2H). These data further affirm that CMA facilitates SERPINA1^E342K/ATZ degradation in a LAMP2A-dependent manner, highlighting the critical role of the C-terminal cytosolic tail in mediating this interaction.

The ₁₂₁QELLR₁₂₅ pentapeptide motif is crucial in CMA-mediated SERPINA1^E342K/ATZ degradation

The CMA pathway differs from autophagy in its distinct substrate preference. Substrates degraded through the CMA pathway typically contain the classical “KFERQ” pentapeptide sequence, which serves as a recognized site by HSPA8/HSC70 [3]. Using the “KFERQ finder V0.8” web server (https://rshine.einsteinmed.edu/), we identified a typical KFERQ-like motif (₁₂₁QELLR₁₂₅) within the SERPINA1^E342K/ATZ protein. To investigate the role of this specific motif, we engineered a mutant variant, AA-SERPINA1^E342K/ATZ (₁₂₁AALLR₁₂₅) (Figure 3A) and evaluated its binding affinity to HSPA8/HSC70 and LAMP2A through a co-IP assay. Remarkably, the mutant demonstrated a marked decrease in binding to both LAMP2A and HSPA8/HSC70 compared to the SERPINA1^E342K/ATZ protein (Figure 3B, C). Immunofluorescence assays further substantiated the diminished interaction between the mutant SERPINA1^E342K/ATZ protein and LAMP2A (Figure 3D).

Figure 3.

Pentapeptide ₁₂₁QELLR₁₂₅ on SERPINA1^E342K/ATZ is critical in CMA mediated degradation. (A) surface representation of SERPINA1/A1AT with the red-highlighted KFERQ motif (₁₂₁QELLR₁₂₅) exposed for HSPA8/HSC70 accessibility. This model was generated using PyMOLTM based on the protein data bank code 3CWM. (B and C) HEK 293 cells transfected with LAMP2A or HSPA8/HSC70 and SERPINA1^E342K/ATZ-FLAG or AA-SERPINA1^E342K/ATZ-FLAG for 48 h. Lysates were immunoprecipitated (IP) with either IgG, anti-LAMP2A antibody and probed for SERPINA1^E342K/ATZ-FLAG, AA-SERPINA1^E342K/ATZ-FLAG and LAMP2A or HSPA8/HSC70. (D) HEK 293 cells were con-transfected for 48 h with LAMP2A expressing vectors or vector coding SERPINA1^E342K/ATZ-FLAG or AA-SERPINA1^E342K/ATZ-FLAG mutant. Representative confocal images showing the colocalization of SERPINA1^E342K/ATZ-FLAG (red) or AA-SERPINA1^E342K/ATZ-FLAG (red) and LAMP2A (green) in HEK 293 cells. Scale bars: 10 μm. (E) HEK 293 cells were co-transfected with LAMP2A expressing vectors or vectors encoding SERPINA1^E342K/ATZ or AA-SERPINA1^E342K/ATZ mutant for 48 h. The cells were lysed, and western blot analysis was conducted to assess the expression of GFP-SERPINA1^E342K/ATZ. (F) HEK 293 stably expressing LAMP2A cells were transfected with SERPINA1^E342K/ATZ-FLAG or AA-SERPINA1^E342K/ATZ-FLAG expressing vectors. The transfected cells were cultured for 24 h before being further incubated with cyclohexamide (CHX; 10 μg/ml) for the indicated time. The levels of SERPINA1^E342K/ATZ-FLAG or AA-SERPINA1^E342K/ATZ-FLAG at different time points were detected by western blot. Data were presented as the mean ± SEM of 3 or more independent experiments. p<0.05.*; p<0.01.**; p<0.0001.****.

In addition, we observed that the basal protein level of AA-SERPINA1^E342K/ATZ was significantly higher than that of SERPINA1^E342K/ATZ under the condition of equal transfection of SERPINA1^E342K/ATZ or AA-SERPINA1^E342K/ATZ plasmids, regardless of whether LAMP2A was overexpressed or not (Figure 3E). We hypothesized that this discrepancy was caused by the inefficient degradation of mutant AA-SERPINA1^E342K/ATZ by CMA. To prove this, we compared the degradation rate of SERPINA1^E342K/ATZ with that of AA-SERPINA1^E342K/ATZ under conditions of high CMA activity with a CHX chase experiment. The results showed that, indeed, with the activation of CMA, the degradation rate of AA-SERPINA1^E342K/ATZ was slower than SERPINA1^E342K/ATZ, with a protein half-life reduced from 5.2 h to 3.5 h (Figure 3F). These results underscore the importance of the CMA pentapeptide ₁₂₁QELLR₁₂₅ in SERPINA1^E342K/ATZ for its degradation through the CMA pathway. They further demonstrate that SERPINA1^E342K/ATZ is bona fide substrate of CMA.

CMA degrades SERPINA1^E342K/ATZ independent of macroautophagy

With above assays, we confirmed that SERPINA1^E342K/ATZ could be degraded via the CMA pathway. However, there are complicated compensatory regulatory mechanisms between macroautophagy and CMA [18–20]. To mitigate the influence of macroautophagy, we employed autophagy-deficient HeLa ATG16L1^-/- cells [21] to reassess our findings. In wild-type HeLa cells, we observed that serum starvation not only activates CMA but also macroautophagy, as indicated by increased levels of LAMP2A and MAP1LC3/LC3-II, alongside enhanced SERPINA1^E342K/ATZ degradation (Figure 4A). Conversely, in HeLa ATG16L1^-/- cells subjected to identical starvation conditions, the absence of LC3-II formation confirmed the disruption of autophagy, yet SERPINA1^E342K/ATZ protein levels still diminished (Figure 4B). Furthermore, in HeLa ATG16L1^−/− cells, the overexpression of exogenous LAMP2A resulted in dose-dependent degradation of SERPINA1^E342K/ATZ, which was similarly effective in wild-type SERPINA1/A1AT and insoluble SERPINA1^E342K/ATZ aggregates (Figure 4C, D). By employing CMA-specific activators, AR7 and QX77 [22,23], we noted identical degradation patterns for the prototypical CMA substrates GAPDH and SERPINA1^E342K/ATZ (Figure 4E). Further investigations into the pentapeptide’s influence in autophagy-deficient cells showed that AA-SERPINA1^E342K/ATZ exhibited a lower degradation efficiency than SERPINA1^E342K/ATZ (Figure 4F). These data suggesting that CMA is capable of degrading SERPINA1^E342K/ATZ independently of macroautophagy.

Figure 4.

SERPINA1^E342K/ATZ was a bona fide substrate of CMA. Ectopic expression of LAMP2A in macro-autophagy deficient cell causes reduction of SERPINA1^E342K/ATZ. (A and B) transfection of the SERPINA1^E342K/ATZ-FLAG plasmid was performed in HeLa, HeLa ATG16L1^-/- cells. After 48 h, the cells were cultured in fbs-free DMEM for 24 h for starvation treatment, followed by cell lysis for immunoblot analysis. ST+: starvation treatment (fbs-free DMEM culture conditions), ST-: normal culture conditions (DMEM culture conditions with 10% FBS). (C) HeLa ATG16L1^−/− cells were transfected with GFP-SERPINA1^E342K/ATZ-FLAG and increasing amounts of a LAMP2A-expressing vector. Cells were lysed for western analysis. (D) HeLa ATG16L1^-/- cells were transfected with LAMP2A and SERPINA1^E342K/ATZ-FLAG or SERPINA1/A1AT-FLAG vectors. After 48 h, cells were lysed using 0.1% triton X-100 as a detergent, and the lysates were fractionated into triton X-100 soluble cell lysates and triton X-100 insoluble cell pellet fractions. Equal amounts of cell lysates (30 g) were loaded onto SDS-PAGE and subjected to immunoblotting using antibodies against SERPINA1^E342K/ATZ and SERPINA1/A1AT and LAMP2A. (E) HeLa ATG16L1^-/- transfected with SERPINA1^E342K/ATZ-FLAG for 36 h, treated with QX77(10 μM) or AR7 (10 μM) for 24 h, cell lysates were harvested and immunoblot with the indicated antibodies. (F) immunoblotted assay of HeLa ATG16L1^-/- cells transfected with LAMP2A and SERPINA1^E342K/ATZ-FLAG or AA-SERPINA1^E342K/ATZ-FLAG. (G) transfection of the SERPINA1^E342K/ATZ-FLAG plasmid was performed in HeLa BECN1^-/- cells. After 48 h, the cells were cultured in fbs-free DMEM for 24 h for starvation treatment, followed by cell lysis for immunoblot analysis. (H) HeLa BECN1^-/- cells were transfected with SERPINA1^E342K/ATZ-FLAG and an empty vector or LAMP2A for 36 h and then treated with MG132 (20 μM) for an additional 4 h. Cells were lysed and subjected to western blot analysis. Data were presented as the mean ± SEM of 3 or more independent experiments. p<0.05.*; p<0.01.**; p<0.001.***; p<0.0001.****; ns, no significance. vs. indicated group.

In addition to its role in autophagy, ATG16L1 has been implicated in anti-inflammatory responses and the regulation of hormone secretion [24,25]. To rule out the influence of ATG16L1‘s non-autophagic functions, we utilized another autophagy deficient model, the HeLa BECN1^-/- cell line [26,27]. In the HeLa BECN1^−/− cells, a similar decrease in SERPINA1^E342K/ATZ protein levels was observed after inducing CMA through serum starvation and LAMP2A overexpression (Figure 4G, H). Notably, this decline in SERPINA1^E342K/ATZ persisted despite the inhibition of the proteasome pathway when LAMP2A was overexpressed (Figure 4H). These findings lend further support to the notion that CMA can independently facilitate the degradation of SERPINA1^E342K/ATZ.

CMA-mediatedSERPINA1^E342K/ATZ degradation mitigates cellular stress

Prior research indicates that SERPINA1^E342K/ATZ aggregates lead to the cellular toxicity including mitochondrial damage [28,29]. We conducted a comparison of cell survival rates under normal conditions and after activating CMA. Overexpression of LAMP2A and HSPA8/HSC70 in cell lines stably expressing GFP-SERPINA1^E342K/ATZ-FLAG significantly improved cell viability (Figure 5A). Mitochondrial membrane potential alterations serve as a critical metric for mitochondrial damage assessment [30]. Using the TMRE probe to measure mitochondrial inner membrane potential [31], a notable increase in fluorescence intensity was detected upon overexpression of LAMP2A and HSPA8/HSC70, as compared to controls (Figure 5B, C). This enhancement suggests a reestablishment of mitochondrial integrity, implicating the potential of CMA activation in mitigating mitochondrial damage.

Figure 5.

The degradation of SERPINA1^E342K/ATZ facilitated by CMA alleviates cellular distress and ameliorates cellular states. (A) HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with LAMP2A or HSPA8/HSC70 for 48 h. Cell viability analysis was performed using the CellTiter-glo luminescent cell viability assay. (B) TMRE signals in HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were detected by fluorescence microscopy. Scale bars: 50 μm. CCCP: negative control. Ctrl: pCDH empty plasmid. (C) HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with either LAMP2A or HSPA8/HSC70, then stained with TMRE or CCCP for flow cytometry analysis. CCCP: negative control. Ctrl: pCDH empty plasmid. (D) A volcano plot was generated to compare the log2 fold change (logFC) versus the p-value between control cells and LAMP2A-expressing cells. Gray circles represent RNAs with a p-value >0.05, blue circles represent downregulated RNAs with a p-value <0.05 and fold change greater than 2, while red circles represent upregulated RNAs with a p-value <0.05 and fold change greater than 2, n = 3. (E and F) after 48 h of LAMP2A transfection in HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells, gene ontology (GO) (E) and KEGG enrichment analyses (F) were conducted on the differentially expressed genes identified from transcriptomic sequencing. Control: empty vector, LAMP2A OE: LAMP2A overexpression. (G) the volcano plot shows the proteins obtained from HEK293 cells and HEK293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG by proteomic profiling (log2 fold change [logFC] vs. p-value). In the plot, proteins are represented as circles categorized by color. Gray circles represent proteins with p-values greater than 0.05, indicating statistical insignificance; blue circles denote downregulated proteins with p-values less than 0.05 and a fold change exceeding 1.5; red circles signify upregulated proteins that meet the criteria of p-values less than 0.05 and a fold change greater than 1.5, n = 6. (H) the volcano plot illustrates the differential proteins in HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells 48 h after LAMP2A transfection, identified through proteomic analysis. (I) the protein level of DNAJB1, UGT2B28, CYB5R4, and LAMP2A in HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells after LAMP2A transfection. (J) gene ontology (GO) and Kyoto Encyclopedia of genes and genomes) KEGG pathway analyses demonstrate the enrichment of functional annotations for down-regulated proteins in the lysates of HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells transfected with LAMP2A. (K) the volcano plot illustrates the differential proteins in HEK293 cells 48 h after LAMP2A transfection, identified through proteomic analysis. Control: empty vector, LAMP2A OE: LAMP2A overexpression. (L) the protein level of DNAJB1, CYB5R4, and LAMP2A in HEK293 cells after LAMP2A transfection. Data were presented as the mean ± SEM of 3 or more independent experiments. p<0.05.*; p<0.001.***; p<0.0001.****.

To investigate how CMA-mediated SERPINA1^E342K/ATZ degradation impacts cellular stress, we used HEK293 cells stably expressing GFP-SERPINA1^E342K/ATZ-FLAG and induced CMA either by LAMP2A overexpression (Figure 5D, F) or AR7 treatment (Fig. S5B), followed by RNA sequencing. Transcriptome analysis revealed that, compared to the empty vector control, CMA activation through LAMP2 overexpression resulted in minimal transcriptional changes, affecting only a small subset of genes (n = 19) out of the 15,284 genes detected (Figure 5D). As expected from our previous findings, LAMP2A overexpression did not significantly alter SERPINA1^E342K/ATZ transcription levels (Fig. S5A and Table S1). Gene ontology enrichment analysis revealed a significant downregulation of several molecular chaperones and genes related to protein misfolding, including DNAJB1 (DnaJ heat shock protein family (Hsp40) member B1), HSPA6 (heat shock protein family A (Hsp70) member 6) and EDEM2 (ER degradation enhancing alpha-mannosidase like protein 2) (Figure 5E). KEGG enrichment analysis further indicated that most of these downregulated genes were involved in protein processing in the ER (Figure 5F). Similar results were observed in AR7-treated HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells, where the transcription of 230 genes significantly changed compared to DMSO-treated controls (Fig. S5B and Table S2). Notably, AR7 treatment, like LAMP2A overexpression, also led to a statistically significant downregulation of both HSPA6 and DNAJB1, though to a lesser extent than with LAMP2A overexpression. The reduced effects of AR7 might be attributed to its overall lower efficiency and selectivity in activating CMA (Figure S5). These findings suggest that both genetic and pharmacological induction of CMA promote SERPINA1^E342K/ATZ degradation and alleviate intracellular ER stress.

To further explore how CMA-mediated SERPINA1^E342K/ATZ degradation alleviates cellular stress, we conducted a proteomic analysis of HEK293 cells with and without SERPINA1^E342K/ATZ overexpression. Among the 6,623 detected proteins, SERPINA1^E342K/ATZ overexpression significantly altered the levels of 30 proteins, with 27 being upregulated and 3 downregulated (Figure 5G and Table S3). Several ER stress-related proteins, including DNAJB1, UGT2B28, and CYB5R4, were among those upregulated by SERPINA1^E342K/ATZ overexpression. These findings are consistent with previous reports that SERPINA1^E342K/ATZ accumulation induces ER stress [32]. We then tested whether activating the CMA pathway could lower the ER-stress-associated proteins upregulated by SERPINA1^E342K/ATZ accumulation. To this end, we induced CMA by LAMP2A overexpression in HEK293 cells stably expressing SERPINA1^E342K/ATZ and compared the proteomes of these cells with vector-transfected control cells. Indeed, CMA induction significantly downregulated the ER-stress-associated proteins along with SERPINA1^E342K/ATZ (Figure 5H and Table S4). GO enrichment analysis showed that following CMA-mediated degradation of SERPINA1^E342K/ATZ, the levels of proteins related to “mitochondrial electron transport activity” and “mitochondrial ATP synthesis coupled electron transport” were downregulated (Figure 5J), which may indicate reduced mitochondrial toxicity. KEGG analysis further revealed that differentially expressed proteins were mainly involved in oxidative phosphorylation and nonalcoholic fatty liver disease (Figure 5J). Finally, to ensure these effects were specific to SERPINA1^E342K/ATZ, we compared the proteomes of HEK293 cells with and without LAMP2A overexpression in the absence of SERPINA1^E342K/ATZ. Although LAMP2A overexpression altered the proteomic landscape as expected, there was no selective enrichment of ER-stress-related proteins (Figure 5 K, L and Table S5). These results suggest that CMA alleviates cellular stress mainly through SERPINA1^E342K/ATZ degradation.

Activation of CMA by AR7 reduces the levels of SERPINA1^E342K/ATZ protein in the liver of PiZ mice

Since CMA has been demonstrated to facilitate the degradation of SERPINA1^E342K/ATZ in vitro, we aim to investigate whether a similar effect can be observed in vivo. To achieve this, we utilized a transgenic mouse model, PiZ [33], which harbors a mutant human PiZ SERPINA1 gene integrated into its genome (Fig. S6A). These PiZ mice recapitulate the liver disease observed in individuals with AATD, characterized by the expression of misfolded SERPINA1^E342K/ATZ and forming insoluble inclusion bodies which could be identified by periodic acid Schiff plus diastase (PAS-D) staining as small puncta [33]. AR7 has also been proved activate CMA in vivo [34]. In mice, the half-life of this drug is 9.2 h, with a mean residence time of 9.9 h when administered intraperitoneally at a dosage of 10 mg/kg body weight (Fig. S7). We administered AR7 to 4-week-old PiZ mice for four weeks (Fig. S6B). Compared to the control group treated with DMSO, we observed that AR7 treatment did not cause significant pathological changes in the mice (Fig. S6C). However, there was a significant increase in hepatic LAMP2A protein levels following AR7 treatment, with a notable decrease in the classical CMA substrate GAPDH, confirming the activation of CMA (Figure 6A). Additionally, there was a concurrent reduction in SERPINA1^E342K/ATZ protein levels, but not in mRNA levels (Fig. S6D) indicating post-translational reduction of SERPINA1^E342K/ATZ. A systematic analysis of the PAS-D staining of liver samples revealed that AR7-treated mice exhibited fewer PAS-D-stained globules, indicating a reduction in the formation of insoluble protein aggregates (Figure 6B). These findings provide compelling evidence for the effectiveness of CMA activation in promoting SERPINA1^E342K/ATZ elimination in vivo.

Figure 6.

AR7 treatment reduce SERPINA1^E342K/ATZ protein levels in PiZ mouse livers. (A) Western blot analysis was performed on protein samples obtained from the livers of experimental PiZ mice that received subcutaneous injections of 10 mg/kg AR7 or control DMSO (n = 6). The protein levels were quantitatively analyzed and presented as mean ± SD. (B) Representative PAS-D-stained images from normal and PiZ mice were captured at a magnification of 20 × . Scale bar: 50 μm. The PAS-D positively stained area was quantitatively analyzed by examining at least 3 random visual fields of each analyzed liver tissue. In the graphs, each dot, triangle, represent the analyzed mean value from each visual field. Statistical significance between samples was determined using unpaired two-tailed Student’s t-test, P < 0.01 * *.

Discussion

A comprehensive understanding of the catabolic mechanism responsible for the degradation of SERPINA1^E342K/ATZ is essential for advancing therapeutic strategies for AATD. To date, the degradation of SERPINA1^E342K/ATZ has been associated with both the proteasome and autophagy-lysosome pathways [35]. Multiple evidences indicated upregulated autophagic pathway in the AATD pathologic cells, and augmentation of autophagy has been demonstrated to facilitate the clearance of SERPINA1^E342K/ATZ aggregates [11,36]. Furthermore, recent evidence has identified a noncanonical ER-to-lysosome-associated degradation pathway (ERLAD) that transports SERPINA1^E342K/ATZ directly from the ER to the lysosome [37]. Despite this, the specific role of CMA in SERPINA1^E342K/ATZ catabolism remains to be fully elucidated. This study presents biochemical evidence confirming CMA’s role in SERPINA1^E342K/ATZ degradation and elucidates its underlying mechanisms.

CMA is a selective process where the chaperone protein HSPA8/HSC70 recognizes misfolded proteins, delivering them to LAMP2A for internalization [38]. Our first piece of evidence supporting the involvement of CMA in the regulation of SERPINA1^E342K/ATZ comes from the finding that LAMP2A directly interacts with SERPINA1^E342K/ATZ and promotes its degradation (Figure 1 and 2). LAMP2A, produced by alternative splicing of the LAMP2 gene, is the only isoform implicated in CMA [39,40]. The selective silencing of LAMP2A leads to increased SERPINA1^E342K/ATZ levels, underscoring CMA’s unique role in managing SERPINA1^E342K/ATZ (Figure 1).

The structural distinctions between LAMP2 isoforms are localized to their C-terminal transmembrane region and cytoplasmic tail. The cytoplasmic tail’s positively charged residues are vital for substrate recognition in CMA [40]. Our research demonstrates that alterations to the cytoplasmic tail, either through deletion or modification of these key amino acids, compromise substrate binding and consequently hinder degradation (Figure 2G, H). This corroborates previous research underscoring the critical role of the cytoplasmic tail in the recognition of substrates by the CMA pathway.

CMA specific pentapeptide KFERQ-like motif in substrate is absolutely necessary for CMA mediated degradation, because it serves as a recognition site for HSPA8/HSC70 binding [41]. In this study, we identified a pentapeptide sequence, ₁₂₁QELLR₁₂₅, on SERPINA1^E342K/ATZ and demonstrated that modifications to this sequence significantly weaken the interactions among SERPINA1^E342K/ATZ, HSPA8/HSC70, and LAMP2A (Figure 3B, C). Variants of SERPINA1^E342K/ATZ lacking this motif displayed a marked decrease in degradation rate, resulting in significant accumulation, even under normal physiological conditions (Figure 3E, F). The identification of ₁₂₁QELLR₁₂₅ motif further confirm SERPINA1^E342K/ATZ as a bona fide substrate for CMA. It is important to note that this motif is also present in the wild-type SERPINA1/A1AT molecule. This is evidenced by changes in endogenous SERPINA1/A1AT levels in hepatic cells, which exhibit an inverse correlation with CMA activity (Figure 1D, G), supporting the idea that native SERPINA1/A1AT is also a substrate of CMA.

Macroautophagy and CMA, both lysosomal-dependent digestion pathways, are coordinately regulated in response to stress, with complex interplay and compensatory mechanisms. For example, the activation of CMA has been observed to suppress macroautophagy, while blocking CMA triggers an upregulation of macroautophagy [18,20]. Conversely, the induction of macroautophagy has been shown to diminish CMA activity [19]. In our study, we utilized two different autophagy-deficient cell lines to confirm that CMA acts independently to degrade SERPINA1^E342K/ATZ (Figure 4). Conventionally, it is believed that in AATD affected cells, soluble SERPINA1^E342K/ATZ is degraded by the proteasome system [10], whereas insoluble aggregates are cleared by autophagy-lysosome pathways and the ERLAD pathway [11,36,37]. The identification of CMA as a novel pathway for SERPINA1^E342K/ATZ clearance brings us closer to achieving a comprehensive understanding of the catabolic processes involved in SERPINA1^E342K/ATZ metabolism.

Microautophagy is another cellular degradation pathway conserved in mammalian cells, characterized by the direct engulfment of substrates by invagination of the lysosomal membrane [38]. The available data allow us to conclude that SERPINA1^E342K/ATZ is a substrate of CMA, as it meets all the criteria for CMA substrates, including possessing the KFERQ-like pentapeptide, interacting with HSPA8/HSC70 and LAMP2A, and undergoing digestion within the lysosome. But it also raises another question: what is microautophagy’s role in SERPINA1^E342K/ATZ’s catabolic regulation? While the precise involvement of microautophagy remains to be conclusively determined, emerging evidence suggests it could play a role in modulating SERPINA1^E342K/ATZ levels. For example, overexpression of HSPA8/HSC70 exerts a greater effect on the clearance of SERPINA1^E342K/ATZ and overall cellular recovery than LAMP2A overexpression does (Figure 5A). Such an effect is consistent with the process of endosomal microautophagy/e-MI, which selectively engulfs soluble proteins that present a KFERQ-like pattern, with HSPA8/HSC70 facilitation [42]. We postulate in that case overexpressed HSPA8/HSC70 may redirect a portion of SERPINA1^E342K/ATZ to endosomal microautophagy for degradation, offering additional cellular protection. Extensive research is necessary to thoroughly elucidating these mechanisms in this regard.

ER stress and mitochondrial damage are two typical characteristics of SERPINA1^E342K/ATZ-induced cellular malfunctions [29,43]. In our study, we observed simultaneous alleviation of mitochondrial damage and restoration of cellular growth status upon the effective activation of CMA. Transcriptomic analysis revealed that CMA activation marginally altered global gene expression, four were intimately linked to protein folding, suggests a rebalancing of proteostasis due to the reduced presence of intracellular misfolded proteins [44]. The diminished expression of EDEM2, a gene implicated in the clearance of misfolded proteins in ER [45], hints successful removing of SERPINA1^E342K/ATZ leads to the reduction of ER stress. Moreover, notable decreases in the expression of DNAJB1 [46] and ANKRD1 (ankyrin repeat domain 1) both associated with apoptotic regulation [47], suggesting reduced cell apoptosis further confirming the positive role played by CMA toward maintaining cellular viability (Figure 5).

Proteomics analyze also indicated a clear recovery of ER function with CMA activation by identify lots of protein chaperones associated with ER stress (Figure 5E). Besides with that CMA activation release the mitochondrial damage as indication of the down regulation of oxidative reactions enzymes. Lipid metabolic malfunction is another typical phenotype recently discovered associated with AATD, in our study we also found the recovery of gene relative with lipid metabolic dysfunction (Fig. S6C) which is concomitated with the recent finding that SERPINA1^E342K/ATZ accumulation could also cause hepatic alteration of lipid metabolism, and we speculated the down regulation of lipid malfunction proteins were also the direct result of the CMA caused degradation of SERPINA1^E342K/ATZ.

However, it is important to note that CMA serves as an essential intracellular protein degradation pathway. In theory, CMA targets proteins containing the KFERQ-like motif, which accounts for approximately 30% of soluble cytosolic proteins. Enhancing CMA activity would facilitate the removal of misfolded and oxidized proteins, and ultimately enhancing cellular viability [18]. Thus, the observed cellular recovery should not be exclusively credited to SERPINA1^E342K/ATZ aggregate removal but also acknowledged as a broader impact of CMA in reducing general cellular stress by targeting various oxidized and misfolded proteins.

The aggregation of SERPINA1^E342K/ATZ is a major contributor to the pathogenesis of AATD liver disease, prompting various attempts to halt or reverse the aggregation process and mitigate AATD progression. These strategies include inhibiting SERPINA1^E342K/ATZ synthesis [48–50], enhancing secretion [51,52], blocking polymerization [53,54] and promoting degradation [36,55]. The identification of macroautophagy activation as a means to accelerate SERPINA1^E342K/ATZ degradation has led to the investigation of autophagy-boosting drugs such as carbamazepine, rapamycin, and fluphenazine [12]. Our discovery of CMA involvement in the regulation of SERPINA1^E342K/ATZ suggests that specific enhancers of CMA may offer new therapeutic options for AATD. Since the discovery of CMA pathway, there have been many attempts to discover small molecules for CMA manipulation. In recent years, a series of small molecular drugs that can activate CMA have been found. However, most CMA activators interfere with other cellular pathways, often causing disturbance in essential cellular processes, including biomaterial metabolism or protein synthesis, which hinders their application [56]. AR7 and QX77both derivatives of retinoic acid, stands out as the most effective CMA inducer developed thus far [15,23].

In our model, we found the treatment with AR7 and QX77, enhancing CMA activity and clearing SERPINA1^E342K/ATZ in both normal and autophagy-deficient cells (Figure 1E and Figure 4E). This finding suggests that CMA-specific activators could serve as potential small-molecule drug therapies for AATD. However, it is important to note that QX77also induces the upregulation of RAB11 expression, which plays a crucial role in regulating intracellular membrane transport, the vesicular system, and autophagosome maturation [34]. Nevertheless, Previous studies conducted with Alzheimer disease mice have also demonstrated the clearance of pathological proteins when treated with AR7 derivatives, thereby providing support for their potential therapeutic implications [34]. Interestingly, when we treated PiZ mice with AR7, we observed a significant activation of CMA, which led to an increased elimination of SERPINA1^E342K/ATZ (Figure 6). These findings not only provide evidence for the effectiveness of CMA activation in promoting SERPINA1^E342K/ATZ elimination in vivo but also demonstrate the immense potential of CMA-boosting drugs for the future treatment of AATD liver disease.

In conclusion, our study establishes CMA as a previously neglected protein degradation pathway involved in regulating the catabolism of SERPINA1^E342K/ATZ. As depicted in Figure 7, the SERPINA1^E342K/ATZ protein is recognized by HSPA8/HSC70 through the pentapeptide ₁₂₁QELLR₁₂₅ and delivered to LAMP2A to interact with its C terminal cytosolic tail on the lysosomal membrane, the LAMP2A then transporting the substrate into the lysosome for digestion. Activation of CMA, either through the overexpression of LAMP2A or the use of small molecular activators AR7, enhances the digestion of SERPINA1^E342K/ATZ, thereby slowing down its aggregation process and alleviated it caused liver pathology.

Figure 7.

The schematic diagram illustrates the degradation mechanism of SERPINA1^E342K/ATZ through the CMA pathway. Besides proteasome and autophagy pathways, CMA also plays a role in SERPINA1^E342K/ATZ degradation. Specifically, the HSPA8/HSC70 protein recognizes the SERPINA1^E342K/ATZ protein through the pentapeptide sequence ₁₂₁QELLR₁₂₅ and transports it to LAMP2A (lysosomal associated membrane protein 2A) on the lysosomal membrane. Augmenting CMA activity through LAMP2A overexpression or AR7 utilization enhances SERPINA1^E342K/ATZ digestion via this pathway, thereby alleviating cellular stress, including ER stress and mitochondrial impairment.

Materials and methods

Cell culture and plasmids transfection

HEK 293 and HepG2 cells were obtained from the National Collection of Authenticated Cell Cultures (SCSP-5425, SCSP-510). The HeLa ATG16L1^-/- and HeLa BECN1^-/- cell lines were provided by Dr. Yue Xu at Shanghai Jiao Tong University, China. All cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM; TransGen Biotech, FI101–01) supplemented with 10% fetal bovine serum (FBS; TransGen Biotech, FS301–02), 50 IU penicillin, and 50 µg/ml streptomycin under conditions of 37°C with a CO₂ concentration of 5%. Following the manufacturer’s instructions for transfection reagents (Lipo8000; Beyotime, C0533), cells at a confluence level of approximately 70% to 80% underwent plasmid transfection. Cells were then harvested for further analysis after 48 to 72 h post-transfection.

Antibodies and chemicals

SERPINA1/A1AT (Thermo Fisher Scientific,702047; 1:3000), MAP1LC3/LC3-II (Beyotime, AL221; 1:3000), Flag (ABclonal, AE005; 1:5000), ATG16L1(ABclonal, A3637; 1:3000), BECN1 (ABclonal, A21191; 1:3000), ACTB/β-Actin (ABclonal, AC026; 1:5000), LAMP2A (ABclonal, A1961;WB: 1:3000; IF: 1:200), HSPA8/HSC70 (ABclonal, A0415,WB: 1:3000; IF:1:100), HRP-conjugated Goat Anti-Rabbit IgG (H+L) (ABclonal, AS014; 1:3000), HRP-conjugated Goat Anti-Rabbit IgG (H+L) (ABclonal, AS003; 1:3000), ABflo® 594-conjugated Goat Anti-Mouse IgG (H+L) (ABclonal, AS054; 1:300), ABflo® 594-conjugated Goat Anti-rabbit IgG (H+L) (ABclonal, AS039; 1:300), ABflo® 488-conjugated Goat Anti-Rabbit IgG (H+L) (ABclonal, AS073; 1:300), cycloheximide (Sigma-Aldrich 01,810), DMSO (Solarbo, D8371), puromycin (Solarbo, P8230), streptomycin (Solarbo, P1400), AR7 (MedChemExpress, HY-101106), QX77(MedChemExpress, HY-112483), MG-132 (MedChemExpress, HY-13259), chloroquine (MedChemExpress, HY-17589A).

Molecular cloning and ShRNA knockdown

LAMP2A and HSC70 were cloned into pCDH-CMV (Addgene 72,265; deposited by Kazuhiro Oka) by PCR amplifying the ORFs from cDNA templates of HSC70 (Addgene 19,514; deposited by Harm Kampinga) and LAMP2A (Addgene 86,146; deposited by Janice Blum), followed by digestion with BamHI and EcoRI restriction enzymes. The molecular cloning was performed using T4 DNA ligase and the resulting constructs were transformed into DH5α E. coli cells. Plasmid purification and extraction steps were carried out using the Endo Free Mini Plasmid Kit II (TIANGEN, DP118). ShRNA sequences are detailed in Fig. S2, and the corresponding shRNA plasmids were transfected into cells using Lipo8000 transfection reagent (Beyotime, C0533FT), following the manufacturer’s protocol.

Cloning of lentiviral expression constructs

To develop GFP-SERPINA1^E342K/ATZ and LAMP2A-expressing HEK 293 cells in a stable manner within the pCDH-CMV vector using lentivirus, the LAMP2A and SERPINA1 genes were amplified by PCR, specifically targeting the ORFs from the cDNA templates of LAMP2A or SERPINA1. The resulting cDNA insert of pCDH-CMV was then sub-cloned into the lentiviral backbone plasmid pCHD-CMV-IRES-pur, thus constructing the pCDH-CMV-GFP-SERPINA1^E342K/ATZ or pCDH-CMV-LAMP2A plasmids. To generate lentiviral particles expressing the pCDH-CMV-GFP-SERPINA1^E342K/ATZ or pCDH-CMV-LAMP2A plasmids, the corresponding constructs were co-transfected with the VSV-G and PVSV-2 helper plasmids into HEK 293 cells. The transfected cells were cultured for 4 days to ensure sufficient expression time. Subsequently, the culture medium containing viral particles was collected, filtered to remove debris, and used to infect HEK 293 cells, inducing efficient expression of the target proteins within the host cells.

Discovery and mutations of the CMA recognition motif “KFERQ-like pentapeptide”

To demonstrate the involvement of CMA recognition motif, the complete amino acid sequence of the SERPINA1 gene was obtained by accessing the UniProt database (P01009). A search was performed using the web server “KFERQ finder V0.8” (https://rshine.einsteinmed.edu/) to identify pentapeptide sequences associated with the CMA recognition motif. Within the amino acid sequence of SERPINA1/A1AT, a canonical pentapeptide fragment, QELLR, was found at positions 121–125. To modify the physical properties of amino acids within the targeting motif, as described by Orenstein and Cuervo [57], we introduced mutations into the CMA recognition motif (5”-CAGGAACTCCTCCGT-3‘) resulting in 5’-GCGGCACTCCTCCGT-3”. The mutated motif was then sub-cloned into a pCDH-CMV plasmid.

Fractionation of cytoplasmic soluble proteins and aggregates

The cells were lysed using RIPA lysis buffer (Solarbio, R0010). After centrifugation at 13,000 g for 10 min at 4°C, the supernatant was collected as the cytoplasmic soluble fraction, and the pellet was resuspended in lysis buffer containing an additional 2% SDS. The resuspended pellet was further homogenized by passing through a 28-gauge needle ten times.

Western blot analysis

Cell lysates were prepared in RIPA buffer to extract total protein. The obtained lysates were then subjected to centrifugation at 17,000 g for 20 min at 4°C to separate the soluble lysates from the insoluble microspheres. Protein concentrations were determined using the Bicinchoninic acid method, as directed by the manufacturer’s instructions using the Pierce BCA protein assay kit (Thermo Fisher Scientific 23,225). Equal amounts of protein were transferred to a polyvinylidene fluoride (PVDF) membrane through sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE). The primary antibody was added and incubated overnight at 4°C, followed by incubation with the secondary antibody at room temperature for 45 min. Blot was performed using a chemiluminescence kit (BeyoECL Plus; Beyotime, P0018S) and visual detection was performed using a western blotting detection system (Tanon, 5200). The gray values of protein bands in each group were subsequently quantified using ImageJ software, and the ratio of target bands to ACTB/β-actin bands was utilized as the measure of protein expression level. The experiment was repeated three times with distinct protein samples.

Co-immunoprecipitation assay

To obtain the whole cell extract, the cells were treated with lysis buffer containing protease inhibitor, and the resulting supernatant was collected. Magnetic beads (ABclonal, AC047, AE037) were pre-treated with antibody binding and subsequently mixed thoroughly with the extracted total cell protein samples. The mixture was then subjected to rotational agitation and incubated overnight at 4°C to facilitate the formation of immune complexes. The protein-bound magnetic beads were isolated using a magnetic separation rack, and the supernatant from the heated beads was analyzed through SDS-PAGE and western blot.

Immunofluorescence staining

The cells were washed three times with pre-chilled PBS (Solarbio, P1020) (136 mm NaCl, 2.6 mm KCl, 8 mm Na₂HPO₄, 2 mm KH₂PO₄), for 1 min each time. Subsequently, the cells were incubated at 37°C for 20 min in an immunostaining fixation solution (Beyotime, P0098), which contained 4% paraformaldehyde (Sigma-Aldrich 441,244) dissolved in 1× PBS and then underwent three additional washes with PBS, each lasting 5 min. Following a 25-min incubation at room temperature in an immunostaining permeabilization solution (Triton X-100, Beyotime, P0096), consisting of 0.1% Triton X-100 in 1× PBS, the cells were again washed three times with PBS for 5 min each. Next, the cells were permeabilized, and nonspecific sites were blocked using an immunostaining blocking solution (Beyotime, P0102) containing 5% BSA (Thermo Fisher Scientific 11,020,021) and 0.3% Triton X-100 at room temperature for 1 h. The primary antibodies, such as rabbit anti-LAMP2A (ABclonal, A1961, 1:200) and mouse anti-FLAG (ABclonal, AE024, 1:300) and rabbit anti-HSPA8/HSC70 (ABclonal, A0415, 1:100), diluted in an immunol staining primary antibody dilution solution (Beyotime, P0103), were then incubated overnight at 4°C. On the following day, the primary antibodies were removed, and a wash with immunol staining wash buffer (Beyotime, P0106) was performed on the cells. Subsequently, the cells were incubated at room temperature for 1 h with secondary antibodies, such as ABflo® 488-conjugated goat anti-rabbit IgG (H+L) (ABclonal, AS073, 1:300) and ABflo® 594-conjugated goat anti-mouse IgG (H+L) (ABclonal, AS054, 1:300) and ABflo® 594-conjugated goat anti-rabbit IgG (H+L) (ABclonal, AS039, 1:300), diluted in an immunol fluorescence staining secondary antibody dilution buffer (Beyotime, P0108). The cells were washed three times with PBS.

Finally, the cells were mounted onto slides using a fluorescence decay mounting medium containing DAPI (Solarbio, S2110). Confocal microscopy images were captured, processed using a laser scanning confocal microscope (Leica, TCS SP5II), with a 63× oil objective lens, using excitation wavelengths of 488 nm for Alexa Fluor 488, 594 nm for Alexa Fluor 594, and 405 nm for DAPI and analyzed using ImageJ.

Colocalization of SERPINA1^E342K/ATZ with the endoplasmic reticulum

To examine the colocalization of SERPINA1^E342K/ATZ protein with the endoplasmic reticulum in HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells, the cells were incubated with ER-tracker Red (Beyotime, C1041S, 1:1000) at 37°C in a 5% CO₂ atmosphere for 30 min. After incubation, the cells were washed with PBS, and the stained HEK293-GFP-SERPINA1^E342K/ATZ-FLAG cells were imaged using a laser scanning confocal microscope according to the previously described protocols.

Cellular GFP-SERPINA1^E342K/ATZ protein clearance rate assessed by flow cytometry

HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were transfected with a plasmid for 48 h. After centrifugation, the cells were resuspended in phosphate-buffered saline and analyzed using FACScan flow cytometry (BD biosciences, LSRFortessa^TM X-20) for GFP fluorescence (488-nm laser excitation, 530/30 filter for detection). A total of 3 × 10⁵ cells underwent scanning to exclude cellular debris based on forward and side-angle scatters (FSC and SSC). Cells expressing GFP were gated, and the percentage depletion of GFP-positive cells was calculated relative to the control group. The clearance rate of GFP-SERPINA1^E342K/ATZ protein is assessed by measuring the average fluorescence intensity. The clearance rate of GFP-SERPINA1^E342K/ATZ-FLAG protein is assessed by measuring the average fluorescence intensity.

Cycloheximide Chase Analysis

The CHX chase experiments involved a 48-h cell culture period following plasmid transfection. Protein synthesis was blocked using CHX treatment at a concentration of 100 μg/ml. Cells were collected at various time points, lysed with RIPA buffer, and subjected to SDS-PAGE and immunoblotting analysis with specific antibodies for protein detection.

Real-time quantitative PCR assay

Total RNA was extracted from GFP-SERPINA1^E342K/ATZ expression HEK 293 stable cell line using TRIZOL (Thermo Fisher Scientific, 15596026CN) according to the manufacturer’s instructions. A total of 1 μg RNA was used for reverse transcription (QIAGEN 205,311). qRT-PCR was performed using gene-specific primers and SYBR Green Real-time PCR Master Mix (Toyobo, QPK-201). The gene expression levels were normalized to ACTB. Real-time PCR and data collection were conducted on a Bio-Rad CFX96 Connect instrument. The primer sequences used for PCR were as follows: LAMP2A, LAMP2B and LAMP2C forward oligo: 5′-GAA GGA AGT GAA CAT CAG CATG-3′, LAMP2A reverse oligo: 5′-CTC GAG CTA AAA TTG CTC ATA TCC AGC-3′, LAMP2B reverse oligo: 5′-CAA GCC TGA AAG ACC AGC ACC-3′, LAMP2C reverse oligo: 5′-CTC GAG TTA CAC AGA CTG ATA ACC AGT AC-3′, HSPA8 Forward oligo: 5′- CAC TTG GGT GGA GAA GAT TTTG-3′, HSPA8 reverse oligo: 5′- CTG ATG TCC TTC TTA TGC TTG C-3′, SERPINA1 forward oligo: 5′-GGC TGA CAC TCA CGA TGAAA-3′, SERPINA1 reverse oligo: 5′- GTG TCC CCG AAG TTG ACA GT-3′, ACTB Forward oligo: 5′-TCA CCC ACA CTG TGC CCA TCT ACG A-3′, ACTB reverse oligo: 5′- CAG CGG AAA ATG G-3′.

Transcriptional analysis

HEK 293 cells expressing GFP-SERPINA1^E342K/ATZ-FLAG were cultured in the presence of 10% DMSO. The cells were transfected with LAMP2A or treated with 10 μM AR7 for 48 h, total RNA was extracted from the cells using TRIzol reagent. According to the manufacturer’s instructions, genomic DNA was eliminated from the samples using RNeasy mini kit (Qiagen 74,104). The concentration and integrity of RNA were determined by Nanodrop™ 2000 spectrophotometry (Thermo Fisher, ND-2000) and agarose gel electrophoresis. Complementary DNA libraries were prepared following Illumina’s standard protocol and sequenced on an Illumina platform. Quality assessment of raw reads was conducted using FastQC tool, followed by removal of low-quality data with TrimGalore software. Alignment to human reference genome GRCh38 was performed using HISAT2 software. HTseq software along with DESeq software quantified transcript abundances and differential gene expression levels based on mapping read positions onto genes’ locations. Sample-to-sample correlation analysis was conducted between the LAMP2A-transfected and AR7-treated samples of the HEK293-GFP-SERPINA1^E342K/ATZ-FLAG stable cell line and the control group. Genes exhibiting |log2 Fold Change|≥2 and DESeq P-adj value < 0.05 between these six sample groups were considered significantly differentially expressed. The KFEE pathway and GO term enrichment were perform using the Hiplot Pro, a comprehensive web service for biomedical data analysis and visualization. It involved utilizing the clusterProfiler package (v.4.5.0) in R software (v.4.2.2), as described by Wu et al [58]. Hiplot Pro served as the platform for importing and analyzing the data, while clusterProfiler facilitated various bioinformatics analyzes. The analysis was carried out by using R software (v.4.2.2) package clusterProfiler (v.4.5.0) through Hiplot Pro (https://hiplot.com.cn/). Enrichment items with corrected p<0.05 were considered significantly enriched.

Proteomics profiling analysis

LC/MS-MS experiments were conducted at the Proteomics Mass Spectrometry Core Facility, Tianjin Institute of Industrial Biotechnology, Chinese Academy of Sciences. The proteome of HEK 293 cells, HEK-GFP-SERPINA1^E342K/ATZ-FLAG and HEK-SERPINA1^E342K/ATZ-FLAG with LAMP2A overexpression was extracted using 8 M urea and 50 mm Tris-HCl (pH 8.0) with ultrasonication. The proteins were then treated with 10 mm DTT at 37°C for 4 h and 40 mm 2-chloroacetamide at room temperature for 30 min. The solution was diluted to 1 M urea with 50 mm NH₄HCO₃ and 1 mm CaCl₂, followed by trypsin digestion at a 1:100 ratio for 4 h. Peptides were desalted, dried, and reconstituted for analysis. Approximately 200 ng of peptides were separated over 60 min on a homemade column (25 cm × 75 μm, 1.9 μm C18-AQ particles; Dr. Maisch, Parker Hannifin Corporation, r119.ap.003) using a 5–95% gradient of 0.1% (v:v) formic acid in water to 100% acetonitrile. All samples were analyzed using a Bruker timsTOF Pro2 hybrid TIMS quadrupole time-of-flight mass spectrometer with a Captive Spray nanoelectrospray ion source. The mass spectrometer was operated in DIA-PASEF [59] mode. The Spectronaut software was used to process DIA-PASEF data in direct-DIA mode with the same Homo sapiens uniport database to generate the predicted spectrum library. Q-value of 0.01 was set as the criteria at precursor and gene group levels for all identified protein groups. To explore significantly enriched pathways and gene functions, proteomics data analysis was performed using the Hiplot Pro platform. KEGG pathway and GO term enrichment analyses were conducted with the clusterProfiler package (v.4.5.0) in R software (v.4.2.2). Initially, proteomics data, normalized and filtered for fold change greater than 1.5 or less than 1.5, were uploaded to Hiplot Pro for visualization. Subsequent bioinformatics analysis was performed using the clusterProfiler package. Enrichment was evaluated using the Benjamini-Hochberg method for p-value correction, with items showing a corrected p-value of less than 0.05 considered statistically significant.

Mitochondrial condition measurements

Cellular mitochondrial membrane potential was assessed using the Mitochondrial Membrane Potential Assay Kit with TMRE (Beyotime, C2001S) with the tetramethylrhodamine, ethyl ester (TMRE) dye. TMRE, a mitochondria-specific fluorescent probe with red fluorescence, exhibits distinct fluorescence signals upon binding to ATP within the mitochondria [31]. Cells were washed three times with chilled PBS and then incubated with TMRE fluorescent dye at 37°C for 30 min. Fluorescence intensity was measured using fluorescence microscopy and flow cytometry, with an excitation wavelength of 550 nm and an emission wavelength of 575 nm.

Cell viability

To assess cell viability, the Cell Titer-Glo® luminescent cell viability assay kit (Promega, G7572) was used. The amount of ATP inside the cells was measured to indicate cellular metabolic activity. The assay measures the luminescent signal produced by the conversion of ATP to adenosine monophosphate (AMP) by the enzyme luciferase [60]. The treated cells were mixed with the reagent and agitated to induce cell lysis. The mixture was then incubated at room temperature to stabilize the luminescent signal. The luminescence intensity was recorded using the BioTek Synergy Neo2 multimode microplate reader.

Animal experiments

Transgenic PiZ mice (Tg [SERPINA1*E342K]#Slcw; JAX 035,411) [33] expressing human SERPINA1^E342K/ATZ protein were purchase from Jackson Laboratory. The experimental animals were heterozygous offspring resulting from breeding heterozygous PiZ mice with C57BL/6 mice. The mice were housed in an air-conditioned room at a temperature of 24°C and subjected to a 12-h light/dark cycle. They had free access to water and food for one week before the experiment began. At four weeks old, the male mice were randomly divided into two groups, each consisting of six mice. Group 1 received vehicle treatment by injecting 100 μL of saline with 0.5% DMSO, while Group 2 received AR7 (10 mg/kg, intraperitoneally, 100 μL per mouse) six days a week for four weeks. One day after the last injection, the mice were humanely sacrificed using CO₂, and their livers were harvested. Mouse liver cell lysis samples prepared for western blot analysis were adapted using a method previously described with slight modifications [61]. In brief, liver samples were initially cut into small pieces and then mixed with RIPA Lysis Buffer. Subsequently, they were homogenized using a prechilled Dounce homogenizer for 30 repetitions and then passing them through a 28-gauge needle 10 times. All animal procedures adhered to protocols approved by the review board of Institutional Animal Care and Use Committee at Tianjin University of Science & Technology in accordance with established guidelines.

Mouse liver histology analysis

Livers from all experimental mice were fixed in 10% neutral-buffered formalin (Thermo Fisher Scientific 6,764,254) and embedded in paraffin. The sections (5 µm) were deparaffinized and rehydrated, and then treated with diastase (Solarbio, G-1283) for 20 min at room temperature, followed by rinsing with water. Next, the sections were incubated with 0.5% Periodic acid (PAS Diastase Staining kit; Solarbio, G-1282) for 10 min and rinsed with water. The slides were scanned and digitized using photographic file scanner and the areas of PAS-D staining for each liver sample were calculated using ImageJ software based on three randomly chosen views.

Pharmacokinetic analysis of AR7

AR7 was administered intraperitoneally at an equimolar dose of 10 mg/kg. Blood samples (0.3 mL) were collected from the retro-orbital venous plexus at 0.25, 0.5, 1, 2, 4, 6, 12, 24, and 48 h post-administration, with at least four mice sampled at each time point. The blood samples were placed in centrifuge tubes containing heparin (MedChemExpress, HY-17567B) and centrifuged at 5000 g for 3 min. Plasma samples aliquots (20 μL) were treated with four volumes of acetonitrile, containing diclofenac as analytical internal standard (IS). The proteins were removed by centrifugation and the supernatant was lyophilized under vacuum. The AR-7 and IS were re-dissolved with Methanol-Water (50:50, v:v) and were analyzed by Thermo Orbitrap Eclipse UPLC-MS/MS system. The analytes were separated on the ODS column (BEH-C18, 2.1 mm x 100 mm, 1.7 μm; Waters, Sigma-Aldrich, W4602) and were detected under the Parallel Reaction Monitoring mode. Pharmacokinetic parameters T1/2, Tmax, Cmax, AUC, MRT were determined by non-compartmental analysis using DAS 3.2.8 (the Mathematical Pharmacology Committee, Chinese Pharmacological Society, China).

Statistical analysis

The statistical analysis of the data was performed using GraphPad Prism 9 (GraphPad Software). The experimental data was presented as mean ± SEM. Comparisons between two groups were conducted by unpaired two-tailed Student’s t-test for parametric data, and comparisons between multiple groups were conducted by one-way ANOVA. Statistical significance was set at p<0.05.*; p<0.01.**; p<0.001.***; p<0.0001.****; NC, no significance.

Funding Statement

This study was funded by: Tianjin Synthetic Biotechnology Innovation Capacity Improvement Project [TSBICIP-CXRC-048]; Public hospital reform and high-quality development demonstration project research fund, gastrointestinal tumors [2023SGGZ114]; Major Project of Inner Mongolia Medical University [YKD2022ZD002].

Disclosure statement

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

Data availability statement

The data relevant to this study are included in the article or uploaded as supplementary materials. Data and materials can be provided upon reasonable request.

Ethics approval and consent to participate

The experimental protocol was established, according to the ethical guidelines of the Helsinki Declaration and was approved by the Human Ethics Committee of Tianjin University of Science & Technology. Written informed consent was obtained from individual or guardian participants.

supplementary materials

Supplemental data for this article can be accessed online at https://doi.org/10.1080/15548627.2025.2480037