Autophagy modulator scoring system: a user-friendly tool for quantitative analysis of methodological integrity of chemical autophagy modulator studies

ABSTRACT

Over the past 20 years (1999–2019), we have witnessed a rapid increase in publications involving chemical macroautophagy/autophagy modulators. However, an overview of the methodologies used in these studies is still lacking, and methodology flaws are frequently observed in some reports. To provide an objective and quantitative analysis of studies involving autophagy modulators, we present an Autophagy Modulator Scoring System (AMSS), which is designed to evaluate methodological integrity. AMSS-A includes the autophagy characterization by 4 aspects, namely, autophagosome quantification, autophagy-related biochemical changes, autophagy substrate degradation, and autophagic flux. AMSS-B contains the pharmacological and functional characteristics of chemical autophagy modulators, including lysosomal function, drug targets, autophagy-dependent pharmacological effects, and validation in multiple cell lines and in vivo models. Our analysis shows that of the 385 studies reporting chemical autophagy modulators, only 142 single studies had examined all 4 aspects of autophagy characterization in AMSS-A, and only 10 out of 142 studies had fulfilled all the AMSS criteria in a single study. A comprehensive analysis of the methodologies used in all the studies was made, along with a summary of studies that demonstrated the highest methodological integrity based on AMSS ranking. To test the reliability of the AMSS, a co-efficiency analysis of scores and co-citation values in the co-citation network was performed, and a significant co-efficiency was obtained. Collectively, AMSS provides insight into the methodological integrity of autophagy modulators studies and also offers a user-friendly toolkit to help choose appropriate assays to characterize autophagy modulators.

Abbreviations: 3-MA: 3-methyladenine; AMSS: Autophagy Modulator Scoring System; ATG: autophagy-related; BAF: bafilomycin A₁; BECN1: beclin 1; CQ: chloroquine; GFP: green fluorescent protein; LC3: microtubule associated protein 1 light chain 3; mRFP: monomeric red fluorescent protein; MTOR: mechanistic target of rapamycin kinase; PtdIns3K: phosphatidylinositol 3-kinase; PtdIns3P: phosphatidylinositol-3-phosphate

Introduction

Autophagy is a conserved cellular process of bulk self-digestion through the lysosome; it is maintained at a basal level under steady-state conditions and can be induced under various stress conditions, such as nutrient deprivation (starvation), inflammation, oxidative stress and pathogen infection [1]. Autophagy is essential for maintaining cellular homeostasis, and dysregulation of this process has been linked to a wide range of diseases [2]. During the past 20 years, modulation of autophagy has been considered a promising strategy for the treatments of certain human diseases, including neurodegenerative diseases, cancer, infection, inflammation and autoimmune diseases, among others [3–11]. Consequently, the publication number of the studies involving chemical autophagy modulators has increased dramatically from less than 5 to more than 25 publications each year, thus reflecting the growing interest in pharmacological modulation of autophagy for biomedical applications (Figure 1).

Figure 1.

Number of chemical autophagy modulator studies published from 1999 to 2019.

As autophagy is a complex and dynamic process, there is no simple method to determine the status of autophagy by a single experiment that can be applied to all the conditions. Multiple assays are often required to define the activity of autophagy modulators. The methods to monitor autophagy have been well described in the autophagy guidelines [1]. However, it is still a difficult task, especially for researchers new to the field, to select the assays that are both suitable and sufficient to define a chemical as an autophagy modulator and designate its activity as an autophagy inducer, inhibitor or blocker. When examining the methodology section of publications involving chemical autophagy modulators, the insufficiency of evidence to support autophagy induction or inhibition is sometimes noticed. To make a quantitative analysis of the methodology part of the studies that focus on chemical autophagy modulators, we here propose an Autophagy Modulator Scoring System (AMSS) (Table 1).

Table 1.

Autophagy modulator scoring system (AMSS).

AMSS-A: Autophagy characterization assays
1. Autophagosome quantification via microscopy analysis
□ Transmission electron microscopy
□ Endogenous LC3 puncta (ICC/IHC/IF)
□ Exogenous LC3 puncta (FM) (e.g., GFP-LC3)
2. Autophagosome formation-related biochemical changes
□ LC3-II level (WB)
□ ATG12–ATG5 conjugation (WB)
□ BECN1-BCL2 interaction (IP)
□ PtdIns3K activity and PtdIns3P generation (kinase assay, ICC/IHC/IF, FM) (e.g., FYVE-GFP)
□ ULK1 phosphorylation (WB)
3. Autophagy substrate degradation
□ SQSTM1/p62 degradation (WB, ICC/IHC/IF, FM)
□ Aggregation-prone protein degradation (WB, FM, PR) (e.g., HA-SNCA/α-syn or EGFP-HTT[Q74])
□ Long-lived protein degradation (e.g., L-azidohomoalanine or [3H]-leucine or [14C]-valine labeled amino acids degradation)
4. Autophagic flux
Autophagosome flux assay: □ Autophagosome biogenesis inhibition assay (LC3-II level with early-stage autophagy inhibitors) (WB, ICC/IHC/IF, FM) □ LC3-II turnover assay (LC3-II level with late-stage autophagy inhibitors) (WB, ICC/IHC/IF, FM) □ Tandem mRFP/mCherry-GFP-LC3 (mTagRFP-mWasabi-LC3 or GFP-LC3-RFP-LC3ΔG) (FM, PR, FC) □ GFP-LC3 lysosomal delivery and proteolysis (WB)
Autophagic cargo flux assay: □ Autophagy substrate level (e.g., SQSTM1/p62, SNCA/α-synuclein and HTT [huntingtin]) with autophagy inhibitors (WB, FM, PR)
AMSS-B: Pharmacological and functional characteristics
5. Lysosome function-related assays
□ Autophagosome-lysosome fusion process (ICC/IHC/IF, FM)
□ Lysosomal acidic environment and lysosomal enzymes activity (WB, ICC/IHC/IF, FM, kinase assay)
6. Chemical autophagy modulators targets identification
□ MTOR-dependent/independent pathways (WB, IP, ICC/IHC/IF, FM, kinase assay)
□ Specific target proteins of the autophagic pathway (WB, IP, ICC/IHC/IF, FM, kinase assay)
7. Autophagy-dependent pharmacological effects (cytoprotective or cytotoxic effects)
□ Pharmacological activity in the presence of autophagy inhibitors
□ Pharmacological activity in the presence of autophagy genes KD/KO
8. Autophagy modulation confirmed in at least two cell lines in vitro
□ Cell lines derived from multiple organs or species
□ Human pluripotent stem cell-derived relevant cell types
9. Autophagy modulation confirmed in vivo
□ C. elegans, Drosophila, zebrafish, mice, rat

Abbreviations in the Table: ATG, autophagy-related; BECN1, beclin 1; FC, flow cytometry; FM, fluorescent microscopy; GFP, green fluorescent protein; ICC, immunocytochemistry; IF, immunofluorescence; IHC, immunohistochemistry; IP, immunoprecipitation; KD, knockdown; KO, knockout; LC3, microtubule associated protein 1 light chain 3; mRFP, monomeric red fluorescent protein; MTOR, mechanistic target of rapamycin kinase; PtdIns3K, phosphatidylinositol 3-kinase; PtdIns3P, phosphatidylinositol-3-phosphate; PR, plate reader; WB, western blotting.

The AMSS contains 2 parts: A) the autophagy characterization assays; B) pharmacological and functional characteristics of autophagy modulators. AMSS-A comprises 4 categories of assays: 1. autophagosome quantification, which reflects the steady-state number of autophagosomes; 2. autophagy-related biochemical changes, which reflect the changes in autophagy signaling; 3. autophagy substrate degradation, which reflects the functional execution of autophagy; 4. autophagic flux, which includes 2 parts, namely an autophagosome flux assay and autophagic cargo flux assay. Autophagosome flux assay refers to a) the level of autophagosome marker LC3-II under the inhibition of autophagosome biogenesis or maturation, b) the process of the quench or degradation of LC3-linked green fluorescence protein. Autophagic cargo flux assay refers to the degradation of the autophagy substrates, e.g., SQSTM1/p62 and the aggregation-prone proteins (SNCA/α-synuclein and HTT [huntingtin]), under normal conditions versus under autophagy inhibition conditions. AMSS-B explores the pharmacological and functional characteristics of autophagy modulators and contains 5 categories of assays: 5. lysosome functional measurement, which reflects the efficacy of autophagy pathway endpoint; 6. drug target exploration, which reflects the molecular mechanisms for autophagy modulation; 7. confirmation of autophagy-dependent pharmacological effects (via autophagy inhibitors or gene knockdown/knockout), which reflect the logical relationship between autophagy modulation and pharmacological activity; 8. autophagy modulation (induction or inhibition) confirmed in at least two cell lines; 9. autophagy modulation (induction or inhibition) confirmed in vivo. This system is designed to score each study in a binary fashion as follows: 1 point is given if at least one experiment in a category is performed; even if multiple experiments are performed in the same category, only 1 point is given. In the AMSS, we list only the representative assays in each category and other assays that fall within the 9 categories can also be accepted.

Results

A total of 19,011 publications were identified through the initial literature search; from these publications, 385 studies were included in the final analysis based on the inclusion criteria. The flowchart illustrates the process of the study selection (Figure 2), and a full list of 385 studies and their autophagy-monitoring methods have been summarized (Table S1). During the past 20 years, the number of chemical autophagy modulator studies showed a progressively increasing trend and reached more than 25 studies per year in the past 5 years (Figures 1 and 3A). Overall, the number of single studies that examined all 4 aspects of autophagy characterization (earning 4 points) in AMSS-A was 142 of 385, indicating that approximately 1/3 of these studies met the highest methodological standards for defining the chemical autophagy modulators (Figure 3A). Between 1999 and 2004, no study gained 4 points in AMSS-A. Since 2005, the number of studies earning 4 points in AMSS-A had gradually increased but still only accounted for approximately 50% of the publications in the past 3 years (Figure 3A). Of the 4 criteria in AMSS-A, changes in autophagosome number and autophagy-related biochemical events were more frequently analyzed (Figure 3B). However, records of autophagic substrate degradation and autophagic flux lacked in 1/2 and 1/3 of the studies, respectively (Figure 3B). Because the increased number of autophagosomes or LC3-II levels can be observed either under the induction of autophagosome biogenesis or the blockage of autophagosome maturation [1], it is problematic to reach a conclusion whether a chemical induces or inhibits autophagy without determining the autophagic flux and parameters related to autophagosome turnover and autophagic cargo degradation. We then selected the 142 studies (earning 4 points in AMSS-A) for further analysis. In most of these studies, the drug targets and autophagy-dependent pharmacological effects of chemical autophagy modulators were assessed (Figure 3C). However, approximately 1/3 of the studies (50/142) examined drug activity only in one cell type, and approximately 2/3 of the studies (102/142) did not confirm the drug activity in vivo (Figure 3C).

Figure 2.

Literature search and article selection process.

Figure 3.

Analysis of chemical autophagy modulators studies based on AMSS. (A) the number of studies scoring at 0 ~ 3 and 4 in each year through the past 20 years. (B and C) the number of studies completed in each category in AMSS-A and -B.

Ten studies that fulfilled all the AMSS criteria in a single study are listed in Table 2, and 2 studies involving corynoxine B (Cory B study) [12] and S130 (S130 study) [13] are summarized here as examples. In AMSS-A, first, the Cory B study used GFP-LC3, whereas the S130 study used 2 more methods, i.e., autophagic vesicles (transmission electron microscopy) and endogenous LC3 staining, to measure changes in the number of autophagosomes. Second, both studies performed an LC3 lipidation assay. Third, the Cory B study tested the mutant SNCA degradation, whereas the S130 study examined the SQSTM1/p62 protein level and used DQ-BSA to confirm autophagic degradation. Fourth, both studies used more than 2 methods to confirm autophagosome maturation (autophagosome flux) and autophagic cargo flux. In AMSS-B, first, both studies showed lysosome function-related assays by using LysoTracker or LAMP1 staining. Second, both studies explored drug targets. Cory B induced autophagy in an MTOR-independent and BECN1-dependent manner, whereas S130 directly bound to and inhibited ATG4B activity. Third, the Cory B study used 3-methyladenine or chloroquine to confirm the autophagy-dependent neuroprotective function in vitro, whereas the S130 study explored the anti-cancer effects using both in vitro (ATG4B KO) and in vivo models. Finally, both studies confirmed the autophagy modulation in vitro and in vivo; the Cory B study confirmed the autophagy induction in neuronal cell lines, primary neurons and Drosophila, whereas the S130 study examined the autophagy markers in colorectal cancer cells as well as in a tumor xenograft mice model.

Table 2.

Example of autophagy modulators that fulfilled all the AMSS criteria in a single study (listed by publication year).

	Methods for monitoring autophagy
Autophagy modulators	Autophagosome quantification	Biochemical changes	Autophagy substrate degradation	Autophagic flux assays	Lysosome assays	Drug targets	Pharmacological effects	Cell lines	In vivo models	Ref.
Corynoxine B/inducer	mRFP/GFP-LC3↑	LC3 lipidation↑	Aggregation-prone protein degradation↑	LC3-II turnover assay, autophagosome biogenesis inhibition assay, mRFP-GFP-LC3, autophagic cargo flux assay	LysoTracker	BECN1	Neuroprotection, mutant SNCA/α-synuclein clearance (use 3-MA and CQ)	N2a, PC12, SH-SY5Y, mice primary cortical neurons	Drosophila	[12]
Oblongifolin C/inhibitor	GFP-LC3↑	LC3 lipidation↑	SQSTM1/p62 protein level↑, DQ-BSA	LC3-II turnover assay	LysoTracker, LAMP1 staining, CTSB&D protein levels	Autophagosome-lysosome fusion inhibition, lysosome deacidification	Anti-cancer (ATG5&7 KD)	HeLa, MEFs, HepG2, CNE, HCT116, MCF-7, MDA-MB-231	Mice	[14]
Corynoxine/inducer	GFP-LC3↑	LC3 lipidation↑	Aggregation-prone protein degradation↑	LC3-II turnover assay, autophagosome biogenesis inhibition assay, autophagic cargo flux assay	CTSD protein level	AKT-MTOR inhibition	Neuroprotection, mutant SNCA/α-synuclein clearance (use 3-MA and CQ)	N2a, SH-SY5Y	Drosophila	[15]
Liensinine/inhibitor	TEM for autophagic vesicles↑, mRFP/GFP-LC3↑, LC3 staining↑	LC3 lipidation↑	SQSTM1/p62 protein level↑	LC3-II turnover assay, autophagosome biogenesis inhibition assay, mRFP-GFP-LC3	LysoTracker, LysoSensor, LAMP1 staining and protein level, CTSB, D&L protein levels	RAB7A recruitment inhibition/autophagosome-lysosome fusion inhibition	Anti-cancer (ATG5&7 KD)	MDA-MB-231, MCF-7, U937, LN229, A549	Mice	[16]
Elaiophylin/inhibitor	TEM for autophagic vesicles↑, GFP-LC3↑	LC3 lipidation↑	SQSTM1/p62 protein level↑, DQ-BSA	LC3-II turnover assay, autophagosome biogenesis inhibition assay	LysoTracker, LAMP1-RFP, CTSB&D protein levels	Lysosomal function impairment	Anti-cancer (BECN1 and ATG5 KD)	CaOV-3, SKOV3, OVCAR3, A2780, SW626	Mice	[17]
Salvianolic acid B/inducer	TEM for autophagic vesicles↑, GFP-LC3↑, LC3 staining↑	LC3 lipidation↑	SQSTM1/p62 protein level↓	LC3-II turnover assay	LysoTracker	MTOR inhibition	Anti-cancer (use 3-MA and ATG5 KD)	HCT116, HT29	Mice	[18]
Tioconazole/inhibitor	TEM for autophagic vesicles↑, GFP-LC3↑	LC3 lipidation↑	SQSTM1/p62 protein level↑	LC3-II turnover assay	RFP-LAMP1	ATG4 inhibition	Anti-cancer (ATG4 KD)	H4, HCT116, MDA-MB-231	Mice	[19]
Tubeimoside I/inhibitor	TEM for autophagic vesicles↑, mRFP/GFP-LC3↑, LC3 staining↑	LC3 lipidation↑, BECN1-BCL2 interaction	SQSTM1/p62 protein level↑, DQ-BSA	LC3-II turnover assay, autophagosome biogenesis inhibition assay, mRFP-GFP-LC3	LAMP1&2 and CTSD protein levels	Lysosomal function impairment	Anti-cancer (use CQ and BAF)	HeLa, SiHa	Mice	[20]
S130/inhibitor	TEM for autophagic vesicles↑, mRFP/GFP-LC3↑, LC3 staining↑	LC3 lipidation↑	SQSTM1/p62 protein level↑, DQ-BSA	LC3-II turnover assay, mRFP-GFP-LC3	LysoTracker, LAMP1 staining	ATG4B inhibition	Anti-cancer (ATG4B KO)	MEFs, HeLa	Mice	[13]
Regorafenib/inhibitor	TEM for autophagic vesicles↑, mRFP/GFP-LC3↑, LC3 staining↑	LC3 lipidation↑, BECN1-BCL2 interaction↓	SQSTM1/p62 protein level↑, DQ-BSA	Autophagosome biogenesis inhibition assay, mRFP-GFP-LC3	LAMP1 staining	PSAT1 (phosphoserine aminotransferase 1) stabilization, autophagosome-lysosome fusion inhibition	Anti-cancer (use 3-MA, CQ, BAF and ATG5&7 KD)	U251, U87, human primary glioblastoma cells	Mice	[21]

This table does not suggest that these compounds are the most established autophagy modulators in the field because the AMSS parameters are evaluated based on the single study in the context of this review.

Abbreviations in the Table: 3-MA, 3-methyladenine; ATG, autophagy-related; AMSS, Autophagy Modulator Scoring System; BAF, bafilomycin A₁; BECN1, beclin 1; CQ, chloroquine; CTSB, cathepsin B; CTSD, cathepsin D; CTSL, cathepsin L; GFP, green fluorescent protein; KD, knockdown; KO, knockout; LAMP, lysosomal associated membrane protein; LC3, microtubule associated protein 1 light chain 3; mRFP, monomeric red fluorescent protein; MTOR, mechanistic target of rapamycin kinase; TEM, transmission electron microscopy.

To validate the reliability of the AMSS scoring system, this research developed a third-party validation method, which was proposed to verify whether an article with the higher AMSS score also showed the higher value measured by the objective indicator. It is well known that a citation-based approach has been increasingly used for evaluation and comparison of the research performance. In this research, we retrieved all subsequent documents citing sample articles and analyzed the co-citation relationship between sample articles. Co-citation is defined as the frequency with which 2 sample articles are cited together by subsequent documents. The more co-citations 2 sample articles receive, the more possible they are semantically related. Thus, we established a co-citation network (Figure 4), where sample articles (nodes) can be linked by their co-citation relationship (edges). Moreover, we calculated the importance of each sample article in this network by authority score and further performed the co-efficiency analysis between authority score and AMSS score to test the reliability of the AMSS scoring system. Herein, the authority score is an estimation of the value of the content of an article itself based on network analysis and can reflect nodal importance within a network [22]. Because the co-citation network is a system based on semantic similarity, the authority score of a sample article in this network indicates the importance of this article in the field of chemical autophagy modulators. The results showed that there is a significant co-efficiency association between the authority score in the co-citation network and the AMSS score (p < 0.05). This finding means that a study with a higher AMSS score has a higher chance of gaining more frequent co-citations by subsequent studies. The analysis thus provides an objective evaluation of the reliability of the AMSS scoring system.

Figure 4.

The co-citation network of sample articles. In the network, nodes denote sample articles, and edges represent co-citation relationships between them. Node size is scaled to the authority score, which is an estimation of the value of the content of an article itself based on network analysis [22]. In a word, the node size indicates the importance of a sample article in the field of chemical autophagy modulators. The thickness of edges represents the co-citation frequency, with which 2 sample articles are cited together by subsequent documents. Nodes are colored according to the sub-network modules to which they belong. Nodes within each module are tightly interconnected, whereas there is sparse connectivity between modules. We used the Louvain algorithm to detect sub-network modules [54]. Moreover, nodes are distributed based on Fruchterman Reingold layout strategy, which results in a set of node positions where strongly interconnected sets of nodes are placed near one another [55]. Last and most importantly, the co-efficiency between authority score and AMSS score was analyzed by Spearman’s rank correlation test, showing a correlation coefficient of 0.230 and a p-value of 0.021.

Moreover, we also analyzed the “reproducibility” and “citation index” of the included studies. The AMSS is designed to evaluate the individual study. However, if a single study of a novel autophagy modulator did not cover all the AMSS criteria, while subsequent studies on this compound over time fulfilled all the criteria, the compound could still be considered as an established autophagy modulator, because its efficacy is robustly reproduced and implemented by various labs. Consequently, the original study pertaining to this compound would get high citation scores. To provide an important supplement to the AMSS, the autophagy modulator studies with the highest citation were analyzed. Among the 385 studies, 44 compounds had been reported at least twice, and the list of these compounds and their reported times are provided (Figure 5A). In addition, the top 23 autophagy modulators studies with the highest citation scores (data from Web of Science) are shown sequentially (Figure 5B) [6,23–44]. These 2 lists of compounds represent the highest “reproducibility” (or the “most established in the field”).

Figure 5.

Reproducibility of chemical autophagy modulators in the 385 studies. (A) the list of chemical autophagy modulators reported at least twice. (B) the list of chemical autophagy modulators studies most cited.

Discussion

Autophagy has been implicated in various physiological and pathological conditions, and its dysfunction is associated with many human diseases. In the past 20 years, increasing numbers of preclinical and clinical studies have been conducted to evaluate the therapeutic potential of chemical autophagy modulators on various disease models including aging, neurodegeneration, cancer, and infection [45–52]. However, the lack of highly specific and safe autophagy modulators is still the major obstacle for the development of autophagy modulators for clinical application [4]. As we can see, among hundreds of autophagy modulator candidates being reported, only a few can be further evaluated in clinical trials. There is thus an urgent need for developing novel autophagy modulators with high specificity and low toxicity.

Autophagy is a complex cellular process. In the studies related to autophagy, many available autophagy-monitoring methods have been updated along with the rapidly developing knowledge and techniques. However, a systematic summary of these autophagy-monitoring methods used in the studies of autophagy modulators is lacking. Here we propose a simple and efficient tool, termed the Autophagy Modulator Scoring System (AMSS), to analyze the methodological integrity of the studies on chemical autophagy modulators. The AMSS contains two parts, namely AMSS-A and AMSS-B. AMSS-A is designed to test the presence/absence of 4 categories of assays that help to characterize autophagy. We suggest that the studies on chemical autophagy modulators should include at least 1 assay from each of the 4 categories in order to make the conclusion robust and convincing. AMSS-B is designed to determine the presence/absence of 5 categories of assays that assess the pharmacological and functional characteristics of the putative autophagy modulators. The co-efficiency analysis indicates that studies with higher AMSS scores are more likely to be co-cited. We hope that researchers, especially those new to the autophagy field, will find the AMSS helpful and user-friendly in choosing appropriate assays that will ensure reliable outcomes of the studies on chemical autophagy modulators.

While we think the system has immense value, it is not perfect in covering all the scenarios pertaining to the studies of chemical autophagy modulators. AMSS aims to analyze the methodological integrity of studies on chemical autophagy modulators, but not the reliability of the studies and the chemical autophagy modulators. It is essential to consider that the autophagy assays have evolved over time, and certain assays were not available in the past. Therefore, some studies may have received lower or higher points, which do not necessarily mean that the quality of the respective studies is low or high. To determine whether a compound is an “established” chemical autophagy modulator, we suggest that researchers may use other parameters, e.g., “reproducibility,” to evaluate the reliability of the original study and whether the autophagy-modulating effects of the compound are confirmed in other studies. Furthermore, it is important to consider that many compounds have off-target effects. It is also important to characterize chemical autophagy modulators at different time-points, dosages, and environmental conditions because the molecular action of a compound can be context-dependent [53]. Finally, the AMSS mainly works for the mammalian system and may not be applicable for assessing the modulators of selective autophagy (e.g., mitophagy modulator) and for chemicals that only act on specific tissues. Additionally, the AMSS is certainly time-sensitive and needs to be updated along with the development of the field.

Materials and methods

Literature search strategy

A literature search (1999–2019) was performed in PubMed and Web of Science to identify all the publications involving chemical autophagy modulators; the search terms “autophagy” and different types of interventions (such as “inducer”, “enhancer”, “inhibitor”, “blocker”, and their variations) were used. We hand-searched and supplemented studies from the representative review articles related to autophagy modulators. Also, we excluded the compounds that have been widely used as positive controls (e.g., rapamycin, 3-MA, CQ, and BAF).

Inclusion criteria

Studies included were: 1) experimental studies; 2) mainly focused on the identification of the chemical autophagy modulators; 3) those that provided adequate information about autophagy-monitoring methods.

Data extraction and analysis

We extracted data for each eligible study and obtained the following information: experiment type, compound information, and methods for monitoring autophagy. The data were inserted into a Microsoft Excel spreadsheet, and each publication was scored according to the AMSS.

Funding Statement

This study was supported by the Science and Technology Development Fund, Macau SAR (File no. 092-2015-A3, 024-2017-AMJ) and the University of Macau grants (MYRG2017-00147-ICMS) awarded to Jia-Hong Lu, and the General Research Fund (HKBU 12101417,  HKBU 12100618) awarded to Min Li, and the Wellcome Trust Seed Award (109626/Z/15/Z), UKIERI (UK-India Education and Research Initiative) DST Thematic Partnership Award (2016-17-0087) awarded to Sovan Sarkar.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.