Endosomal-Lysosomal and Autophagy Pathway in Alzheimer’s Disease: A Systematic Review and Meta-Analysis

Abstract

Background:

The endosomal-lysosomal and autophagy (ELA) pathway may be implicated in the progression of Alzheimer’s disease (AD); however, findings thus far have been inconsistent.

Objective:

To systematically summarize differences in endosomal-lysosomal and autophagy proteins in the cerebrospinal fluid (CSF) of people with AD and healthy controls (HC).

Methods:

Studies measuring CSF concentrations of relevant proteins in the ELA pathway in AD and healthy controls were included. Standardized mean differences (SMD) with 95% confidence intervals (CI) between AD and healthy controls in CSF concentrations of relevant proteins were meta-analyzed using random-effects models.

Results:

Of 2,471 unique studies, 43 studies were included in the systematic review and meta-analysis. Differences in ELA protein levels in the CSF between AD and healthy controls were observed, particularly in lysosomal membrane (LAMP-1: N_AD/N_HC = 348/381, SMD [95% CI] = 0.599 [0.268, 0.930], I² = 72.8%; LAMP-2: N_AD/N_HC = 401/510, SMD [95% CI] = 0.480 [0.134, 0.826], I² = 78.7%) and intra-lysosomal proteins (GM2A: N_AD/N_HC = 390/420, SMD [95% CI] = 0.496 [0.039, 0.954], I² = 87.7%; CTSB: N_AD/N_HC = 485/443, SMD [95% CI] = 0.201 [0.029, 0.374], I² = 28.5%; CTSZ: N_AD/N_HC = 535/820, SMD [95% CI] = –0.160 [–0.305, –0.015], I² = 24.0%) and in proteins involved in endocytosis (AP2B1:N_AD/N_HC = 171/205, SMD [95% CI] = 0.513 [0.259, 0.768], I² = 27.4%; FLOT1: N_AD/N_HC = 41/45, SMD [95% CI] = –0.489 [–0.919, –0.058], I² <0.01). LC3B, an autophagy marker, also showed a difference (N_AD/N_HC = 70/59, SMD [95% CI] = 0.648 [0.180, 1.116], I² = 38.3%)), but overall there was limited evidence suggesting differences in proteins involved in endosomal function and autophagy.

Conclusion:

Dysregulation of proteins in the ELA pathway may play an important role in AD pathogenesis. Some proteins within this pathway may be potential biomarkers for AD.

Keywords

Alzheimer’s disease autophagy biomarkers dementia endosomes lysosomes meta-analysis proteins systematic review

INTRODUCTION

Alzheimer’s disease (AD) is a neurodegenerative disease that leads to gradual cognitive and functional deficits which can affect daily life. Although AD is characterized pathologically by the accumulation of amyloid-β (Aβ) and phosphorylated tau proteins in the brain [1], the pathogenesis and mechanisms of AD progression remain unclear; however, mounting evidence implicates the involvement of the endosomal-lysosomal and autophagy (ELA) systems which warrant further investigation [2].

The endosomal-lysosomal pathway is responsible for the internalization, recycling, and degradation of cargo from the plasma membrane [1]. Intersecting with the endosomal-lysosomal system, the autophagy pathway transports organelles within the cell to the lysosome for degradation [1]. Both pathways have been shown to be involved in the processing of the Aβ protein precursor (AβPP) to Aβ [3, 4]. Aβ has been detected in multiple vesicles in the endosomal-lysosomal and autophagy pathways including early endosomes, multivesicular bodies, autophagic vacuoles, and lysosomes [4 –7]. Inhibition of AβPP internalization by clathrin-mediated endocytosis was shown to reduce Aβ release in animal cells [3]. Likewise, inhibition of macroautophagy reduced the secretion of Aβ₄₀ [4].

A previous meta-analysis has demonstrated that several protein markers differ in cerebrospinal fluid (CSF) concentrations between AD and controls [8], but endosomal-lysosomal and autophagy related proteins were not meta-analyzed. Here we examine those markers to determine their involvement in AD. Numerous studies have compared concentrations of endosomal-lysosomal and autophagy proteins in the CSF of AD patients; however, consistent findings have not been observed. This systematic review aims to summarize and meta-analyze whether there is a difference in the CSF concentrations of selected ELA proteins between AD and healthy controls. The primary objective of the current systematic review is to identify involvement of the ELA system in AD pathogenesis. The study may also suggest whether these proteins in CSF serve as potential biomarkers for this condition.

MATERIALS AND METHODS

Data sources and search strategy

This systematic review was conducted according to our pre-defined protocol (not registered) and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [9]. A literature search was conducted on January 8, 2021, in EMBASE, Medline, PubMed, Web of Science, Cochrane Trials, and Cochrane Reviews. Search strategies for all databases used key and MeSH terms pertaining to AD, CSF, and the selected key ELA proteins and regulators of interest. ELA proteins of interest were selected from the hundreds of proteins that function in these pathways if they were 1) highly critical to key functions of the pathway; 2) highly specific to the pathway (i.e., only function within the endo-lysosomal and/or autophagy pathways, and nowhere else according current knowledge in the literature); or 3) known to have interactions or associations with AD pathology. An example search strategy can be found in Supplementary Table 1. Conference abstracts were filtered out of the EMBASE search results via a setting specific to that database. No other filters were applied to any database.

Study selection

Inclusion criteria were 1) original research articles measuring concentrations of any selected proteins (Table 1) in CSF and 2) articles which included people with late-onset AD versus healthy controls. Only articles in English and Chinese were included. When studies reported concentrations from overlapping groups of subjects, the study with a larger sample size was included. Inquiry emails were sent to the corresponding authors when sample overlapping between studies was suspected. A minimum of three studies were required for meta-analysis. The findings from the included studies were summarized qualitatively when there were insufficient degrees of freedom for a meta-analysis, when the reported data could not be meta-analyzed (e.g., fold change, inadequate statistical reporting), and when precursor proteins were measured. Comparisons in the qualitative synthesis were summarized as numerically increased, decreased, or equivocal. Inclusion eligibility was assessed by two independent reviewers, and disagreement was settled by consensus.

Table 1

Abbreviations for proteins included in the present study

Protein name	Abbreviation	Protein name	Abbreviation
Adaptor-related protein complex 2, beta-1 subunit	AP2B1	Microtubule-associated proteins 1A/1B light chain 3A	LC3A
Autophagy related 4b cysteine peptidase	ATG4B	Microtubule-associated proteins 1A/1B light chain 3B	LC3B
Autophagy related 4c cysteine peptidase	ATG4C	Lysosomal integral membrane protein-2	LIMP2
Autophagy related 5	ATG5	Progranulin	PGRN
Autophagy related 6	ATG6	Phosphatidylinositol-binding clathrin assembly protein	PICALM
Autophagy related 7	ATG7	Prosaposin	PSAP
Autophagy related 9A	ATG9A	Ras-related protein 4a	RAB4A
Autophagy related 12	ATG12	Ras-related protein 4b	RAB4B
Cystatin C	CSTC	Ras-related protein 5	RAB5
Cathepsin A	CTSA	Ras-related protein 5a	RAB5A
Cathepsin B	CTSB	Ras-related protein 5b	RAB5B
Cathepsin D	CTSD	Ras-related protein 5c	RAB5C
Cathepsin L	CTSL	Ras-related protein 7	RAB7
Cathepsin Z	CTSZ	Ras-related protein 7a	RAB7A
Dipeptidyl peptidase 2	DPPII	Ras-related protein 9a	RAB9A
Early endosome antigen 1	EEA1	Ras-related protein 11a	RAB11A
Flotillin 1	FLOT1	Ras-related protein 11b	RAB11B
Alpha-L-fucosidase 1	FUCA1	Sphingomyelin Phosphodiesterase 1	SMPD1
Ganglioside activator	GM2A	Sorting nexin 3	SNX3
Hexosaminidase subunit beta	HEXB	Sorting nexin 6	SNX6
Lysosomal associated membrane protein 1	LAMP1	Sorting nexin 17	SNX17
Lysosomal associated membrane protein 2	LAMP2	Tripeptidyl peptidase 1	TPP1
Microtubule-associated protein 1A/1B-light chain 3	LC3

Data extraction

Means and standard deviations (SD) of concentrations in AD and healthy control groups were extracted from included studies. Mean and SD were estimated when descriptive

statistics were reported in quartiles [10] or if data were log-transformed [11]. When only the mean difference and p-value were reported, the SD was estimated under the assumption that both groups had equal variance and the reported p-value was obtained from a pooled t-test. When a study had one control but had more than one AD group (e.g., prodromal and dementia AD groups), means and SDs were pooled. Other characteristics including mean age, male proportion, mean Mini-Mental State Examination (MMSE) score, mean CSF Aβ₄₂, mean CSF total tau, apolipoprotein E (APOE) ɛ4 allele carrier proportion, AD diagnostic criteria, and protein quantification assay were also extracted.

Methodological quality assessment

Risk of bias was evaluated based on criteria adapted from the Newcastle Ottawa Scale and the Cochrane Collaboration’s Risk of Bias Assessment Tool [12, 13]. Overall risk of bias of a study was judged based on whether the study methodology may greatly contribute to bias in the present review question. Each paper included was assessed by two independent reviewers, and disagreement was resolved by consensus.

Statistical analyses

Random effects meta-analyses were used because between-study variation was anticipated. Standardized mean differences (SMD) with 95% confidence intervals (CI) were calculated to summarize the pooled effect sizes between AD and healthy controls. Heterogeneity was quantified by the I² statistic, and Cochran’s Q test was used to test significant heterogeneity. If heterogeneity was detected (I² >50%), a pre-defined systematic exploration was conducted using meta-regressions and subgroup analyses. Meta-regressions with 1) mean age, 2) percent male, 3) mean MMSE score, 4) percent APOE ɛ4 carriers, 5) mean CSF Aβ₄₂, and 6) mean CSF total tau in the AD group were conducted when 10 observations were available for analysis. A subgroup analysis was performed excluding articles with a high potential risk of bias. Another subgroup analysis was performed stratifying by protein quantification assay. Publication bias was assessed using trim and fill procedures and funnel plots. Data were meta-analyzed using the “metafor” package [14], and bubble plots were created using the “ggplot2” package in R 4.0.5 [15].

RESULTS

Literature search findings

A total of 2,471 studies were identified through literature search, and two additional studies were identified from their references. To summarize findings from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), data were obtained directly from the ADNI website (https://adni.loni.usc.edu). Of the 43 studies (including ADNI) that met the inclusion criteria [16 –57], 36 studies were included in the meta-analyses and 13 studies were included in the qualitative synthesis (Supplementary Figure 1). Information for each included study is presented in Supplementary Table 2.

A total of 20 markers were included in the meta-analyses, and 64 markers were included in the qualitative synthesis (see Table 1 for protein names and their abbreviations). A summary of the ELA pathway and findings from the meta-analyses is illustrated in Fig. 1.

Fig. 1

ELA pathway and an illustrative summary of results. The figure summarizes proteins included in the search for this systematic review, their roles in the ELA pathway, and the results of the current meta-analysis (color printed in the online version).

Endosomal

In the meta-analyses, CSF concentrations of AP2B1 were significantly higher in AD compared to controls, while CSF concentrations of FLOT1 were significantly lower in AD compared controls (Table 2, Fig. 2). No differences in CSF concentrations were found between groups for RAB4A (Supplementary Figure 2).

Table 2

Summary of findings from the meta-analyses

Protein	Number of studies	N_AD/N_HC	Effect Estimates			Heterogeneity Estimates			Potential Risk of
			SMD [95% CI]	Z	p	χ²	p	I² (%)	Publication Bias
AP2B1	6	171/205	0.513 [0.259, 0.768]	3.95	<0.001	6.88	0.229	27.4	Not detected
CSTC	27	1222/1344	–0.187 [–0.421, 0.046]	–1.57	0.116	176.96	<0.001	85.3	Detected
CTSA	4	235/257	–0.035 [–0.330, 0.259]	–0.23	0.815	5.65	0.130	46.9	Not detected
CTSB	11	485/443	0.201 [0.029, 0.374]	2.29	0.022	13.99	0.174	28.5	Detected
CTSD	13	615/921	–0.136 [–0.321, 0.049]	–1.44	0.149	26.35	0.010	54.5	Detected
CTSL	6	178/193	0.149 [–0.145, 0.443]	0.99	0.320	8.46	0.133	40.9	Not detected
CTSZ	10	535/820	–0.160 [–0.305, –0.015]	–2.16	0.031	11.84	0.223	24.0	Not detected
DPPII	8	325/358	–0.008 [–0.160, 0.144]	–0.10	0.920	6.14	0.523	0	Not detected
FLOT1	3	41/45	–0.489 [–0.919, –0.058]	–2.23	0.026	0.04	0.978	0	Not detected
FUCA1	8	224/256	0.031 [–0.152, 0.214]	0.33	0.741	4.21	0.756	0	Not detected
GM2A	10	390/420	0.496 [0.039, 0.954]	2.13	0.034	73.39	<0.001	87.7	Not detected
HEXB	6	280/278	–0.070 [–0.237, 0.098]	–0.82	0.414	3.76	0.585	0	Detected
LAMP1	9	348/381	0.599 [0.268, 0.930]	3.54	<0.001	29.42	<0.001	72.8	Not detected
LAMP2	12	401/510	0.480 [0.134, 0.826]	2.72	0.007	51.62	<0.001	78.7	Not detected
LC3B	3	70/59	0.648 [0.180, 1.116]	2.71	0.007	3.24	0.198	38.3	Not detected
PGRN	11	524/704	0.026 [–0.168, 0.220]	0.26	0.793	17.86	0.057	44.0	Not detected
PSAP	9	371/364	0.062 [–0.358, 0.482]	0.29	0.772	49.93	<0.001	84.0	Not detected
RAB4A	3	61/44	0.183 [–0.414, 0.779]	0.60	0.549	4.30	0.116	53.5	Not detected
SMPD1	4	218/458	0.090 [–0.337, 0.517]	0.41	0.680	7.40	0.060	59.5	Detected
TPP1	9	336/445	–0.120 [–0.270, 0.029]	–1.58	0.115	8.03	0.431	0.36	Detected

Fig. 2

Forest plots for endocytosis markers. AP2B1 and FLOT1 showed significant differences between AD and control groups.

In the qualitative synthesis, AP2B1, FLOT1, PICALM, RAB5, RAB7, RAB7A, RAB9A, SNX3, SNX6, and SNX17 showed a higher trend in AD compared to controls, while RAB4B, RAB5B, RAB5C, RAB11A, and RAB11B showed a lower trend in AD. One study reported equivocal CSF RAB5A concentrations between groups. Studies showed inconsistent findings for EEA1, RAB4A, and RAB5C (Supplementary Table 3).

Lysosomal

In the meta-analyses, CSF concentrations of LAMP1, LAMP2, CTSB, and GM2A were significantly higher in AD (Table 2, Figs. 3 and 4), whereas CTSZ concentrations were significantly lower in AD compared to controls (Table 2, Fig. 3). No differences in CSF concentrations were observed between both groups for CTSA, CTSD, CTSL, SMPD1, TPP1, FUCA1, DPPII, HEXB, CSTC, PSAP, and PGRN (Table 2, Supplementary Figures 3–13).

Fig. 3

Forest plots for intra-lysosomal markers. GM2A, CTSB, and CTSZ showed significant differences between AD and control groups.

Fig. 4

Forest plots for lysosomal membrane proteins. LAMP1 and LAMP2 showed significant differences between AD and control groups.

In the qualitative synthesis, most studies reported trends towards elevation in AD for CSF concentrations of CTSA, CTSB or its precursors, CTSL or Prepro CTSL, GM2A, Prepro HEXB, LAMP1 or its precursor, LAMP2 or its precursor, SMPD1 or its precursor, and Prepro TPP1. CSF concentrations of CSTC, CTSA, Pro CTSL, and GM2A precursors showed a trend towards lower concentrations in AD. Most studies reported a lower trend in AD for CSF concentrations of TPP1. Inconsistent results were observed for CSTC, DPP-II, FUCA1, HEXB, LIMP2, PGRN, and PSAP (Supplementary Table 3).

Autophagosomal

CSF concentrations of LC3B was significantly higher in AD compared to controls by meta-analyses (Table 2, Fig. 5).

Fig. 5

Forest plot for LC3B, an autophagy marker. LC3B showed a significant difference between AD and control groups.

In the qualitative synthesis, CSF concentrations of ATG5, ATG4B, ATG6, ATG12, LC3, and LC3A showed a trend towards higher concentrations in AD compared to controls, while Atg7 showed a trend towards lower concentrations in AD. The results for ATG4C, ATG9A, and LC3B were inconsistent (Supplementary Table 3).

Assessment of potential risk of bias and publication bias

Thirteen studies were found to have high potential risk of bias due to 1) use of post-mortem samples, 2) controls had subjective cognitive complaints, 3) participants were recruited from different centres, 4) large age differences between groups, 5) late-onset AD group consists of several early-onset AD cases, or 6) controls were too young (Supplementary Table 4).

Risk of publication bias was detected in CSTC, CTSB, CTSD, HEXB, SMPD1, and TPP1 (Supplementary Figures 14–33).

Exploration of heterogeneity

Subgroup analyses was performed to determine whether studies with a high risk of bias contributed to heterogeneity (Supplementary Figures 34–39). When studies with a high potential risk of bias were excluded, differences in CSF concentrations of LAMP1 remained significant with low heterogeneity. Likewise, LAMP2 concentrations remained significantly higher in AD, but had high heterogeneity. Contrary results were observed for GM2A, where no differences in CSF GM2A concentrations were observed between AD and controls among studies without a high potential risk of bias. There were no differences in CSTC, CTSD, and PSAP concentrations between both groups after the exclusion of studies with high risk of bias. Studies with high risk of bias contributed to heterogeneity between studies for CTSD, while this factor did not explain the heterogeneity for CSTC and PSAP.

Subgroup analyses was also performed to determine whether the type of protein quantification assays used contributed to heterogeneity (Supplementary Figures 40–44). Studies were separated based on the use of mass spectrometry or other assays. The differences in CSF LAMP1 concentrations remained significant among studies using mass spectrometry, and this protein quantification assay contributed to the heterogeneity between studies. CSF LAMP2 concentrations remained significantly different between AD and controls for studies that measured proteins using mass spectrometry; however, a significant amount of heterogeneity was still present. For SMPD1, use of mass spectrometry contributed to the heterogeneity between studies, whereas this protein quantification assay did not explain the heterogeneity between studies for CSTC and CTSD.

Results for all meta-regression analyses are presented in Supplementary Figures 45–61. Lower mean MMSE score in the AD group was associated with greater difference in CSF CSTC concentrations favoring control group (Supplementary Figure 48; B [95% CI] = 0.100 [0.007, 0.193] SMD per MMSE, p = 0.038).

DISCUSSION

The present meta-analyses showed substantial differences between AD and healthy controls in CSF concentrations of proteins involved in lysosomal function and endocytosis, while there was limited evidence suggesting differences in proteins involved in endosomal function and autophagy.

Lysosomal markers

Meta-analyses of LAMP-1 and LAMP-2 showed higher CSF concentrations in AD with substantial heterogeneity. Heterogeneity could be partly explained by the different protein quantification assays used between studies and by the potential risk of bias detected in a few studies; however, the trends remained in subgroup analyses removing studies with high risk of bias, suggesting that the results are robust and provide clinical evidence supporting dysregulation of lysosomal pathways in AD. LIMP-2, another known lysosomal membrane protein, could not be meta-analyzed in this study. In the available qualitative data, LIMP-2 showed no directional trend, although more studies are needed to replicate these findings. The current data suggest that LAMP-1 and LAMP-2 may play a unique role in AD.

LAMP-1 and LAMP-2 have overlapping roles in maintaining the integrity and function of lysosomes and are widely considered to be biomarkers of lysosomal function [58]. Lysosomal function is integral to the processing and degradation of AβPP, Aβ, and other AβPP products [59, 60]; enzymatic processing within the lysosome is responsible for the hydrolysis of these proteins [1]. Preclinical studies suggested that lysosomal functioning is important for amyloid clearance in neurons and glial cells [61 –65]. Recent findings showed that amyloid secretion and tau hyperphosphorylation induced via infectious agents and oxidative stress were attenuated in a LAMP-2-deficient animal model [66]. Amyloid and tau dysregulations, characterized by a redistribution or overabundance of the proteins, have been seen in various murine and human models of genetically predisposed lysosomal storage disorders [67]. In this current meta-analysis, we demonstrated that lysosomal dysregulation in the brain observed in previous cell, animal, and histological studies is likely to generalize to a broader human population with AD.

This meta-analysis also demonstrated that CSF concentrations of some intra-lysosomal proteins might be affected in AD. In people with AD compared to healthy controls, concentrations of CTSB were higher and concentrations of CTSZ were lower; these effect sizes were small, but the differences were consistent between studies and not likely affected by publication bias. Concentrations of CTSD were lower in AD, but this difference was small and did not reach statistical significance. Preclinical studies suggested that deficiencies in CTSD promote amyloidosis and tau-mediated neurotoxicity [68, 69]. Similarly, CTSD concentrations were lower in the brain tissues of people with AD compared to mild cognitive impairment [70]. Interestingly, this study also found that concentrations of CTSD were lower in people with normal cognition compared to mild cognitive impairment; however, this difference did not reach statistical significance [70]. One in vitro study demonstrated that inhibition of CTSD did not result in lysosomal dysfunction [71], which is consistent with our finding that CTSD may be less important in this context. Furthermore, inhibition or deficiency of CTSB was shown to enhance amyloid accumulation [71 –74]. The immunoreactivity pattern of CTSZ was also altered in the human tissue of AD patients compared to healthy controls [75], suggesting that important changes to CTSZ protein expression in the brain may occur in AD. The existing evidence along with the results from this meta-analysis strengthen the confidence that intra-lysosomal dysregulation may be partly involved in AD pathogenesis.

Other intra-lysosomal proteins in this meta-analysis showed either no difference between AD and control groups, or in the case of GM2A, a significant difference with high heterogeneity. The mechanistic role of GM2A in the brain is not well-studied; therefore, future studies analyzing extracellular vesicle protein contents may be informative for studying GM2A and other uncommonly secreted intra-lysosomal proteins [76]. CSTC showed no significant difference with high heterogeneity, which implies that it may not be a useful marker for differentiating AD and cognitively healthy individuals; however, MMSE score likely explained the heterogeneity, suggesting that there may be cross-sectional correlations between CSTC and cognition.

Endosomal and endocytic markers

Many of the representative endosomal proteins reviewed in the current study had insufficient data for meta-analysis. Although it is well-known that endosomal and autophagosomal function is important in Aβ processing, there remains a lack of clinical evidence regarding changes in individual proteins within the system. The meta-analysis of RAB4A showed non-significant findings and no strong trends were observed in the qualitative synthesis; thus, based on current evidence, there were no significant differences in CSF concentrations of early endosome, late endosome, or endosome recycling biomarkers in AD. Further evidence is required to support the role of this pathway in AD pathogenesis; however, if endosomal biomarkers continue to show no difference with increased data availability, this could imply that 1) the endosomal functions within AD remain grossly intact, while other aspects of the ELA pathway are more prominently affected and/or are causative in the neurodegenerative processes; or 2) despite dysregulation in AD, endosome-specific markers are not reliable indicators of such dysregulation. The latter seems more likely, given post-mortem and in vitro evidence suggesting Rab GTPase upregulation in AD and endosome enlargement in AD or AD models [77, 78].

As for proteins involved in endocytosis, FLOT1 was higher and AP2B1 was lower in the CSF of individuals with AD, with medium effect sizes and low heterogeneity. This finding implies that FLOT1 and AP2B1 may be potential biomarkers for AD and highlights that endocytosis dysregulation may be important in AD pathogenesis. Our result supports a previous study which found lower transcript expression levels of FLOT1 in the hippocampi of people with AD [79]. AP2B1 is a subunit of the adaptor protein 2 complex which participates in clathrin-mediated endocytosis. AP2 was shown to degrade Aβ via autophagy, a process that also involved LC3 and PICALM [80]. A genome-wide association study found that a PICALM polymorphism was a potential genetic risk factor for AD [81]. PICALM could not be meta-analyzed due to an insufficient number of existing studies; therefore, further investigation into PICALM in biofluids is needed.

Autophagosomal markers

Most of the autophagy biomarkers assessed had insufficient data for meta-analysis and too few studies per biomarker to assess possible trends in the qualitative review. However, it was found that LC3B was higher in individuals with AD compared to healthy controls. LC3B is an essential regulator of autophagosome formation and maturation [82] and is a specific marker for autophagosomes and autolysosomes [83]. Although only three studies contributed to the meta-analysis of LC3B, there was relatively minimal heterogeneity. Therefore, LC3B could be a potential biomarker of interest for AD. It would be particularly interesting to compare temporal relationships with the lysosomal biomarkers to identify specific instigators of dysregulation in the pathway. Proteins involved in autophagosome formation that were outside of the scope of this review, such as vacuolar sorting proteins, might also be relevant to explore further.

Strengths and limitations

As a potential limitation, many proteins were unable to be meta-analyzed within the relevant pathways due to insufficient existing evidence; however, we searched for more than 60 relevant proteins in this systematic review, offered a comprehensive review of representative markers in the CSF, and reported gaps of knowledge in current literature. In addition, it has been suggested that dysregulation in the ELA pathway is an early event in AD progression [70]; therefore, it would be of interest to compare biomarker concentrations at early or precursor stages of AD to controls. Furthermore, the concentrations of these markers in the CSF may not represent the levels of dysregulation within the cells; however, we still identified differences in CSF levels between AD and controls, implying the potential involvement of these pathways in AD. Further studies might examine the extent to which these CSF markers reflect brain tissue proteomic profiles. Lastly, limited studies in this meta-analysis measured post-mortem ventricular CSF samples [20 , 49]; therefore, subgroup analyses could not be performed, and additional proteomic studies with post-mortem CSF samples would be informative.

Conclusion

This systematic review and meta-analysis of endosomal, lysosomal, and autophagosomal CSF biomarkers in AD revealed consistent differences in these biomarkers between AD and controls, particularly in lysosomal membrane proteins (i.e., LAMP-1 and LAMP-2) and intra-lysosomal proteins (i.e., GM2A, CTSB, and CTSZ). Meta-regression analyses showed that CSF Aβ₄₂ and t-tau did not explain heterogeneity in CSTC, CTSD, LAMP2, or GM2A; however, heterogeneity in CSTC concentrations was related to mean MMSE score, suggesting that this marker could be related to cognitive progression. Endocytosis markers AP2B1 and FLOT1, and the autophagy marker LC3B, also showed differences between AD and controls; however, most autophagy markers selected in this study had insufficient data to meta-analyze, suggesting a need for further investigation. The study of these and additional key proteins in the ELA pathway will provide focus, and allow the identification of the key parts of the pathway involved in AD.

Footnotes

ACKNOWLEDGMENTS

Walter Swardfager gratefully acknowledges financial support from the following: Canadian Institute of Health Research (CIHR), Grant/Award Number: PJT-159711; The Natural Sciences and Engineering Research Council of Canada (NSERC), Grant/Award Number: RGPIN-2017-06962; Alzheimer’s Association & Brain Canada, Grant/Award Number: AARG501466; Weston Brain Institute & Alzheimer’s Research UK, Alzheimer’s Association and Michael J. Fox Foundation, Grant/Award Number: BAND3. The work was supported in part through funding from the Canada Research Chairs Program.

BXS gratefully acknowledges the financial support from MITO2i and the Thomas Zachos Scholars. We gratefully acknowledge correspondence from Drs. Estrella Morenas-Rodríguez, Alberto Lleó, and David Holtzman [30 , ].

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Whyte

, Lau

, Hemsley

, Hopwood

, Sargeant

(2017) Endo-lysosomal and autophagic dysfunction: A driving factor in Alzheimer’s disease? J Neurochem 140, 703–717.

Cao

, Zhong

, Toro

, Zhang

, Cai

(2019) Endo-lysosomal pathway and ubiquitin-proteasome system dysfunction in Alzheimer’s disease pathogenesis. Neurosci Lett 703, 68–78.

Koo

, Squazzo

(1994) Evidence that production and release of amyloid beta-protein involves the endocytic pathway. J Biol Chem 269, 17386–17389.

, Cuervo

, Kumar

, Peterhoff

, Schmidt

, Lee

, Mohan

, Mercken

, Farmery

, Tjernberg

, Jiang

, Duff

, Uchiyama

, Näslund

, Mathews

, Cataldo

, Nixon

(2005) Macroautophagy— a novel β-amyloid peptide-generating pathway activated in Alzheimer’s disease. J Cell Biol 171, 87–98.

Langui

, Girardot

, El Hachimi

, Allinquant

, Blanchard

, Pradier

, Duyckaerts

(2004) Subcellular topography of neuronal Aβ peptide in APPxPS1 transgenic mice. Am J Pathol 165, 1465–1477.

Cataldo

, Petanceska

, Terio

, Peterhoff

, Durham

, Mercken

, Mehta

, Buxbaum

, Haroutunian

, Nixon

(2004) Aβ localization in abnormal endosomes: Association with earliest Aβ elevations in AD and Down syndrome. Neurobiol Aging 25, 1263–1272.

Takahashi

, Milner

, Li

, Nam

, Edgar

, Yamaguchi

, Beal

, Xu

, Greengard

, Gouras

(2002) Intraneuronal Alzheimer Aβ42 accumulates in multivesicular bodies and is associated with synaptic pathology. Am J Pathol 161, 1869–1879.

Olsson

, Lautner

, Andreasson

, Öhrfelt

, Portelius

, Bjerke

, Hölttä

, Rosén

, Olsson

, Strobel

, Wu

, Dakin

, Petzold

, Blennow

, Zetterberg

(2016) CSF and blood biomarkers for the diagnosis of Alzheimer’s disease: A systematic review and meta-analysis. Lancet Neurol 15, 673–684.

Page

, McKenzie

, Bossuyt

, Boutron

, Hoffmann

, Mulrow

, Shamseer

, Tetzlaff

, Akl

, Brennan

, Chou

, Glanville

, Grimshaw

, Hróbjartsson

, Lalu

, Li

, Loder

, Mayo-Wilson

, McDonald

, McGuinness

, Stewart

, Thomas

, Tricco

, Welch

, Whiting

, Moher

(2021) The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 372, n71.

10.

Wan

, Wang

, Liu

, Tong

(2014) Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol 14, 135.

11.

Higgins

JPT

, White

, Anzures-Cabrera

(2008) Meta-analysis of skewed data: Combining results reported on log-transformed or raw scales: Meta-analysis of skewed data. Statist Med 27, 6072–6092.

12.

Higgins

JPT

, Altman

, Gøtzsche

, Jüni

, Moher

, Oxman

, Savovic

, Schulz

, Weeks

, Sterne

; Cochrane Bias Methods Group; Cochrane Statistical Methods Group (2011) The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343, d5928.

13.

Stang

(2010) Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol 25, 603–605.

14.

Viechtbauer

(2010) Conducting meta-analyses in with the metafor package. J Stat Softw 36, 1–48.

15.

Wickham

(2016) ggplot2: Elegant Graphics for Data Analysis, Springer-Verlag, New York.

16.

Wilke

, Gillardon

, Deuschle

, Dubois

, Hobert

, Müller vom Hagen

, Krüger

, Biskup

, Blauwendraat

, Hruscha

, Kaeser

, Heutink

, Maetzler

, Synofzik

(2016) Serum levels of progranulin do not reflect cerebrospinal fluid levels in neurodegenerative disease. Curr Alzheimer Res 13, 654–662.

17.

Higginbotham

, Ping

, Dammer

, Duong

, Zhou

, Gearing

, Hurst

, Glass

, Factor

, Johnson

ECB

, Hajjar

, Lah

, Levey

, Seyfried

(2020) Integrated proteomics reveals brain-based cerebrospinal fluid biomarkers in asymptomatic and symptomatic Alzheimer’s disease. Sci Adv 6, eaaz9360.

18.

Khoonsari

, Shevchenko

, Herman

, Remnestål

, Giedraitis

, Brundin

, Degerman Gunnarsson

, Kilander

, Zetterberg

, Nilsson

, Lannfelt

, Ingelsson

, Kultima

(2019) Improved differential diagnosis of Alzheimer’s disease by integrating ELISA and mass spectrometry-based cerebrospinal fluid biomarkers. J Alzheimers Dis 67, 639–651.

19.

Carrette

, Demalte

, Scherl

, Yalkinoglu

, Corthals

, Burkhard

, Hochstrasser

, Sanchez

(2003) A panel of cerebrospinal fluid potential biomarkers for the diagnosis of Alzheimer’s disease. Proteomics 3, 1486–1494.

20.

Maarouf

, Andacht

, Kokjohn

, Castano

, Sue

, Beach

, Roher

(2009) Proteomic analysis of Alzheimers disease cerebrospinal fluid from neuropathologically diagnosed subjects. Curr Alzheimers Res 6, 399–406.

21.

Philips

, De Muynck

, Nguyen Thi Thu

, Weynants

, Vanacker

, Dhondt

, Sleegers

, Schelhaas

, Verbeek

, Vandenberghe

, Sciot

, Van Broeckhoven

, Lambrechts

, Van Leuven

, Van Den Bosch

, Robberecht

, Van Damme

(2010) Microglial upregulation of progranulin as a marker of motor neuron degeneration. J Neuropathol Exp Neurol 69, 1191–1200.

22.

Collins

, An

, Hood

, Conrads

, Bowser

(2015) Label-free LC–MS/MS proteomic analysis of cerebrospinal fluid identifies protein/pathway alterations and candidate biomarkers for amyotrophic lateral sclerosis. J Proteome Res 14, 4486–4501.

23.

Brinkmalm

, Sjödin

, Simonsen

, Hasselbalch

, Zetterberg

, Brinkmalm

, Blennow

(2018) A parallel reaction monitoring mass spectrometric method for analysis of potential CSF biomarkers for Alzheimer’s disease. Proteomics Clin Appl 12, 1700131.

24.

Simonsen

, McGuire

, Podust

, Hagnelius

N-O

, Nilsson

, Kapaki

, Vassilopoulos

, Waldemar

(2007) A novel panel of cerebrospinal fluid biomarkers for the differential diagnosis of Alzheimer’s disease versus normal aging and frontotemporal dementia. Dement Geriatr Cogn Disord 24, 434–440.

25.

Khoonsari

, Häggmark

, Lönnberg

, Mikus

, Kilander

, Lannfelt

, Bergquist

, Ingelsson

, Nilsson

, Kultima

, Shevchenko

(2016) Analysis of the cerebrospinal fluid proteome in Alzheimer’s disease. PLoS One 11, e0150672.

26.

Lauridsen

, Sando

, Møller

, Berge

, Pomary

, Grøntvedt

, Salvesen

, Bråthen

, White

(2017) Cerebrospinal fluid Aβ43 is reduced in early-onset compared to late-onset Alzheimer’s disease, but has similar diagnostic accuracy to Aβ42. Front Aging Neurosci 9, 210.

27.

Hansson

, Andréasson

, Wall

, Skoog

, Andreasen

, Wallin

, Zetterberg

, Blennow

(2009) Reduced levels of amyloid-β-binding proteins in cerebrospinal fluid from Alzheimer’s disease patients. J Alzheimers Dis 16, 389–397.

28.

Brettschneider

(2004) Meningeal derived cerebrospinal fluid proteins in different forms of dementia: Is a meningopathy involved in normal pressure hydrocephalus? . J Neurol Neurosurg Psychiatry 75, 1614–1616.

29.

Jahn

, Wittke

, Zürbig

, Raedler

, Arlt

, Kellmann

, Mullen

, Eichenlaub

, Mischak

, Wiedemann

(2011) Peptide fingerprinting of Alzheimer’s disease in cerebrospinal fluid: Identification and prospective evaluation of new synaptic biomarkers. PLoS One 6, e26540.

30.

Craig-Schapiro

, Kuhn

, Xiong

, Pickering

, Liu

, Misko

, Perrin

, Bales

, Soares

, Fagan

, Holtzman

(2011) Multiplexed immunoassay panel identifies novel CSF biomarkers for Alzheimer’s disease diagnosis and prognosis. PLoS One 6, e18850.

31.

Morenas-Rodríguez

, Alcolea

, Suárez-Calvet

, Muñoz-Llahuna

, Vilaplana

, Sala

, Subirana

, Querol-Vilaseca

, Carmona-Iragui

, Illán-Gala

, Ribosa-Nogué

, Blesa

, Haass

, Fortea

, Lleó

(2019) Different pattern of CSF glial markers between dementia with Lewy bodies and Alzheimer’s disease. Sci Rep 9, 7803.

32.

Wildsmith

, Schauer

, Smith

, Arnott

, Zhu

, Haznedar

, Kaur

, Mathews

, Honigberg

(2014) Identification of longitudinally dynamic biomarkers in Alzheimer’s disease cerebrospinal fluid by targeted proteomics. Mol Neurodegeneration 9, 22.

33.

Kalman

, Marki-Zay

, Juhasz

, Santha

, Dux

, Janka

(2000) Serum and cerebrospinal fluid cystatin C levels in vascular and Alzheimer’s dementia. Acta Neurol Scand 101, 279–282.

34.

Mareš

, Kanovsky

, Herzig

, Stejskal

, Vavrouskova

, Hlustik

, Vranova

, Burval

, Zapletalova

, Pidrman

, Obereigneru

, Suchy

, Vesely

, Podivinsky

, Urbanek

(2009) New laboratory markers in diagnosis of Alzheimer dementia. Neurol Res 31, 1056–1059.

35.

, Wang

, Li

, Zhu

, Zhao

, Deng

, Rogalski

, Bigio

, Rademaker

, Xia

, Mao

(2019) A highly sensitive sandwich ELISA to detect CSF progranulin: A potential biomarker for CNS disorders. J Neuropathol Exp Neurol 78, 406–415.

36.

Abdullah

, Kimura

, Akatsu

, Hashizume

, Ferdous

, Tachita

, Iida

, Zou

, Matsubara

, Michikawa

(2019) Flotillin is a novel diagnostic blood marker of Alzheimer’s disease. J Alzheimers Dis 72, 1165–1176.

37.

Park

, Jung

, Park

, Lee

, Park

, Kim

, Park

, Chae

, Jeong

, Choi

, Baek

(2020) SWATH-MS analysis of cerebrospinal fluid to generate a robust battery of biomarkers for Alzheimer’s disease. Sci Rep 10, 7423.

38.

Goossens

, Bjerke

, Van Mossevelde

, Van den Bossche

, Goeman

, De Vil

, Sieben

, Martin

, Cras

, De Deyn

, Van Broeckhoven

, van der Zee

, Engelborghs

(2018) Diagnostic value of cerebrospinal fluid tau, neurofilament, and progranulin in definite frontotemporal lobar degeneration. Alzheimers Res Ther 10, 31.

39.

Abdi

, Quinn

, Jankovic

, McIntosh

, Leverenz

, Peskind

, Nixon

, Nutt

, Chung

, Zabetian

, Samii

, Lin

, Hattan

, Pan

, Wang

, Jin

, Zhu

, Li

, Liu

, Waichunas

, Montine

, Zhang

(2006) Detection of biomarkers with a multiplex quantitative proteomic platform in cerebrospinal fluid of patients with neurodegenerative disorders. J Alzheimers Dis 9, 293–348.

40.

Perrin

, Craig-Schapiro

, Malone

, Shah

, Gilmore

, Davis

, Roe

, Peskind

, Li

, Galasko

, Clark

, Quinn

, Kaye

, Morris

, Holtzman

, Townsend

, Fagan

(2011) Identification and validation of novel cerebrospinal fluid biomarkers for staging early Alzheimer’s disease. PLoS One 6, e16032.

41.

Wang

, Dey

, Chen

, Li

, Niu

, Cho

, Wang

, Bai

, Jiao

, Chepyala

, Haroutunian

, Zhang

, Beach

, Peng

(2020) Integrated analysis of ultra-deep proteomes in cortex, cerebrospinal fluid and serum reveals a mitochondrial signature in Alzheimer’s disease. Mol Neurodegeneration 15, 43.

42.

Sundelöf

, Sundström

, Hansson

, Eriksdotter-Jönhagen

, Giedraitis

, Larsson

, Degerman-Gunnarsson

, Ingelsson

, Minthon

, Blennow

, Kilander

, Basun

, Lannfelt

(2011) Higher cathepsin B levels in plasma in Alzheimer’s disease compared to healthy controls. J Alzheimers Dis 22, 1223–1230.

43.

Zhong

, Hou

, Luo

, Shi

, Hu

, He

, Chen

, Zheng

, Zhang

, Tan

, Liu

, Mu

, Chen

, Ning

(2013) Alterations of CSF cystatin C levels and their correlations with CSF Aβ40 and Aβ42 levels in patients with Alzheimer’s disease, dementia with lewy bodies and the atrophic form of general paresis. PLoS One 8, e55328.

44.

Yamamoto-Watanabe

, Watanabe

, Jackson

, Akimoto

, Sugimoto

, Yasujima

, Wakasaya

, Matsubara

, Kawarabayashi

, Harigaya

, Lyndon

, Shoji

(2010) Quantification of cystatin C in cerebrospinal fluid from various neurological disorders and correlation with G73A polymorphism in CST3. Brain Res 1361, 140–145.

45.

Lleó

, Núñez-Llaves

, Alcolea

, Chiva

, Balateu-Paños

, Colom-Cadena

, Gomez-Giro

, Muñoz

, Querol-Vilaseca

, Pegueroles

, Rami

, Lladó

, Molinuevo

, Tainta

, Clarimón

, Spires-Jones

, Blesa

, Fortea

, Martínez-Lage

, Sánchez-Valle

, Sabidó

, Bayés

, Belbin

(2019) Changes in synaptic proteins precede neurodegeneration markers in preclinical Alzheimer’s disease cerebrospinal fluid. Mol Cell Proteomics 18, 546–560.

46.

Zhang

, Goodlett

, Quinn

, Peskind

, Kaye

, Zhou

, Pan

, Yi

, Eng

, Wang

, Aebersold

, Montine

(2005) Quantitative proteomics of cerebrospinal fluid from patients with Alzheimer disease. J Alzheimers Dis 7, 125–133.

47.

Andersson

, Remnestål

, Nellgård

, Vunk

, Kotol

, Edfors

, Uhlén

, Schwenk

, Ilag

, Zetterberg

, Blennow

, Månberg

, Nilsson

, Fredolini

(2019) Development of parallel reaction monitoring assays for cerebrospinal fluid proteins associated with Alzheimer’s disease. Clin Chim Acta 494, 79–93.

48.

Bader

, Geyer

, Müller

, Strauss

, Koch

, Leypoldt

, Koertvelyessy

, Bittner

, Schipke

, Incesoy

, Peters

, Deigendesch

, Simons

, Jensen

, Zetterberg

, Mann

(2020) Proteome profiling in cerebrospinal fluid reveals novel biomarkers of Alzheimer’s disease. Mol Syst Biol 16, e9356.

49.

Castaño

, Roher

, Esh

, Kokjohn

, Beach

(2006) Comparative proteomics of cerebrospinal fluid in neuropathologically-confirmed Alzheimer’s disease and non-demented elderly subjects. Neurol Res 28, 155–163.

50.

Heywood

, Galimberti

, Bliss

, Sirka

, Paterson

, Magdalinou

, Carecchio

, Reid

, Heslegrave

, Fenoglio

, Scarpini

, Schott

, Fox

, Hardy

, Bahtia

, Heales

, Sebire

, Zetterburg

, Mills

(2015) Identification of novel CSF biomarkersfor neurodegeneration and their validation by ahigh-throughput multiplexed targeted proteomic assay. MolNeurodegener 10, 64.

51.

Heywood

, Hallqvist

, Heslegrave

, Zetterberg

, Fenoglio

, Scarpini

, Rohrer

, Galimberti

, Mills

(2018) CSF pro-orexin and amyloid-β38 expression in Alzheimer’s disease and frontotemporal dementia. Neurobiol Aging 72, 171–176.

52.

Sathe

, Na

, Renuse

, Madugundu

, Albert

, Moghekar

, Pandey

(2019) Quantitative proteomic profiling of cerebrospinal fluid to identify candidate biomarkers for Alzheimer’s disease. Proteomics Clin Appl 13, 1800105.

53.

Sjödin

, Brinkmalm

, Öhrfelt

, Parnetti

, Paciotti

, Hansson

, Hardy

, Blennow

, Zetterberg

, Brinkmalm

(2019) Endo-lysosomal proteins and ubiquitin CSF concentrations in Alzheimer’s and Parkinson’s disease. Alzheimer Res Ther 11, 82.

54.

Whelan

, Mattsson

, Nagle

, Vijayaraghavan

, Hyde

, Janelidze

, Stomrud

, Lee

, Fitz

, Samad

, Ramaswamy

, Margolin

, Malarstig

, Hansson

(2019) Multiplex proteomics identifies novel CSF and plasma biomarkers of early Alzheimer’s disease. Acta Neuropathol Commun 7, 169.

55.

Armstrong

, Mattsson

, Appelqvist

, Janefjord

, Sandin

, Agholme

, Olsson

, Svensson

, Blennow

, Zetterberg

, Kågedal

(2014) lysosomal network proteins as potential novel CSF biomarkers for Alzheimer’s disease. Neuromol Med 16, 150–160.

56.

Duits

, Brinkmalm

, Teunissen

, Brinkmalm

, Scheltens

, Van der Flier

, Zetterberg

, Blennow

(2018) Synaptic proteins in CSF as potential novel biomarkers for prognosis in prodromal Alzheimer’s disease. Alzheimers Res Ther 10, 5.

57.

Simonsen

, McGuire

, Podust

, Davies

, Minthon

, Skoog

, Andreasen

, Wallin

, Waldemar

, Blennow

(2008) Identification of a novel panel of cerebrospinal fluid biomarkers for Alzheimer’s disease. Neurobiol Aging 29, 961–968.

58.

Eskelinen

(2006) Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy. Mol Aspects Med 27, 495–502.

59.

Evrard

, Kienlen-Campard

, Coevoet

, Opsomer

, Tasiaux

, Melnyk

, Octave

, Buée

, Sergeant

, Vingtdeux

(2018) Contribution of the Endosomal-lysosomal and proteasomal systems in Amyloid-β precursor protein derived fragments processing. Front Cell Neurosci 12, 435.

60.

Zhang

, Thompson

, Zhang

, Xu

(2011) APP processing in Alzheimer’s disease. Mol Brain 4, 3.

61.

Lee

, Sato

, Nixon

(2011) Lysosomal proteolysis inhibition selectively disrupts al transport of degradative organelles and causes an Alzheimer’s-like axonal dystrophy. J Neurosci 31, 7817–7830.

62.

Tammineni

, Jeong

, Feng

, Aikal

, Cai

(2017) Impaired axonal retrograde trafficking of the retromer complex augments lysosomal deficits in Alzheimer’s disease neurons. Hum Mol Genet 26, 4352–4366.

63.

Hassiotis

, Manavis

, Blumbergs

, Hattersley

, Carosi

, Kamei

, Sargeant

(2018) Lysosomal LAMP1 immunoreactivity exists in both diffuse and neuritic amyloid plaques in the human hippocampus. Eur J Neurosci 47, 1043–1053.

64.

Söllvander

, Nikitidou

, Brolin

, Söderberg

, Sehlin

, Lannfelt

, Erlandsson

(2016) Accumulation of amyloid-β by astrocytes result in enlarged endosomes and microvesicle-induced apoptosis of neurons. Mol Neurodegener 11, 38.

65.

, Lv

, Huo

, Liu

, Shang

, Fei

, Li

, Zhao

, Wei

, Deng

, Qu

(2017) Autophagy-lysosome dysfunction is involved in Aβ deposition in STZ-induced diabetic rats. Behav Brain Res 320, 484–493.

66.

Kristen

, Sastre

, Aljama

, Fuentes

, Recuero

, Frank-García

, Martin

, Sanchez-Juan

, Lage

, Bullido

, Aldudo

(2021) LAMP2 deficiency attenuates the neurodegeneration markers induced by HSV-1 infection. Neurochem Int 146, 105032.

67.

Lloyd-Evans

, Haslett

(2016) The lysosomal storage disease continuum with ageing-related neurodegenerative disease. Ageing Res Rev 32, 104–121.

68.

Suire

, Abdul-Hay

, Sahara

, Kang

, Brizuela

, Saftig

, Dickson

, Rosenberry

, Leissring

(2020) Cathepsin D regulates cerebral Aβ42/40 ratios via differential degradation of Aβ42 and Aβ40. Alzheimers Res Ther 12, 80.

69.

Khurana

, Elson-Schwab

, Fulga

, Sharp

, Loewen

, Mulkearns

, Tyynelä

, Scherzer

, Feany

(2010) Lysosomal dysfunction promotes cleavage and neurotoxicity of Tau. PLoS Genet 6, e1001026.

70.

Perez

, He

, Nadeem

, Wuu

, Ginsberg

, Ikonomovic

, Mufson

(2015) Hippocampal endosomal, lysosomal, and autophagic dysregulation in mild cognitive impairment: Correlation with AA and Tau pathology. J Neuropathol Exp Neurol 74, 14.

71.

Cermak

, Kosicek

, Mladenovic-Djordjevic

, Smiljanic

, Kanazir

, Hecimovic

(2016) Loss of cathepsin B and L Leads to lysosomal dysfunction, NPC-Like cholesterol sequestration and accumulation of the key Alzheimer’s proteins. PLoS One 11, e0167428.

72.

Oberstein

, Utz

, Spitzer

, Klafki

, Wiltfang

, Lewczuk

, Kornhuber

, Maler

(2021) The role of cathepsin B in the degradation of Aβ and in the production of Aβ peptides starting with Ala2 in cultured astrocytes. Front Mol Neurosci 13, 615740.

73.

Hook

VYH

, Kindy

, Hook

(2008) Inhibitors of cathepsin B improve memory and reduce β-amyloid in transgenic Alzheimer disease mice expressing the wild-type, but not the Swedish mutant, β-secretase site of the amyloid precursor protein. J Biol Chem 283, 7745–7753.

74.

, Ni

, Liu

, Teeling

, Takayama

, Collcutt

, Ibbett

, Nakanishi

(2017) Cathepsin B plays a critical role in inducing Alzheimer’s disease-like phenotypes following chronic systemic exposure to lipopolysaccharide from Porphyromonas gingivalis in mice. Brain Behav Immun 65, 350–361.

75.

Thygesen

, Ilkjær

, Kempf

, Hemdrup

, von Linstow

, Babcock

, Darvesh

, Larsen

, Finsen

(2018) Diverse protein profiles in CNS myeloid cells and CNS tissue from lipopolysaccharide- and vehicle-injected APPSWE/PS1ΔE9 transgenic mice implicate cathepsin Z in Alzheimer’s disease. Front Cell Neurosci 12, 397.

76.

Thompson

, Gray

, Mager

, Fischer

, Thézénas

, Charles

, Talbot

, El Andaloussi

, Kessler

, Wood

, Turner

(2018) UFLC-derived CSF extracellular vesicle origin and proteome. Proteomics 18, 1800257.

77.

Ginsberg

, Mufson

, Alldred

, Counts

, Wuu

, Nixon

, Che

(2011) Upregulation of select rab GTPases in cholinergic basal forebrain neurons in mild cognitive impairment and Alzheimer’s disease. J Chem Neuroanat 42, 102–110.

78.

Zhang

, Huang

, Yancey

, Luo

, Zhang

(2019) Role of Rab GTPases in Alzheimer’s disease. ACS Chem Neurosci 10, 828–838.

79.

Dharshini

SAP

, Taguchi

, Gromiha

(2019) Investigating the energy crisis in Alzheimer disease using transcriptome study. Sci Rep 9, 18509.

80.

Tian

, Chang

, Fan

, Flajolet

, Greengard

(2013) Adaptor complex AP2/PICALM, through interaction with LC3, targets Alzheimer’s APP-CTF for terminal degradation via autophagy. Proc Natl Acad Sci U S A 110, 17071–17076.

81.

Lambert

, Ibrahim-Verbaas

, Harold

, Naj

, Sims

, Bellenguez

, Jun

, DeStefano

, Bis

, Beecham

, et al., (2013) Meta-analysis of 74,046 individuals identifies 11 new susceptibility loci for Alzheimer’s disease. Nat Genet 45, 1452–1458.

82.

Satyavarapu

, Das

, Mandal

, Mukhopadhyay

, Mandal

(2018) Autophagy-independent induction of LC3B through oxidative stress reveals its non-canonical role in anoikis of ovarian cancer cells. Cell Death Dis 9, 934.

83.

Schaaf

MBE

, Keulers

, Vooijs

, Rouschop

KMA

(2016) LC3/GABARAP family proteins: Autophagy-(un)related functions. FASEB J 30, 3961–3978.

84.

Morenas-Rodríguez

, Cervera-Carles

, Vilaplana

, Alcolea

, Carmona-Iragui

, Dols-Icardo

, Ribosa-Nogué

, Muñoz-Llahuna

, Sala

, Belén Sánchez-Saudinós

, Blesa

, Clarimón

, Fortea

, Lleó

(2015) Progranulin protein levels in cerebrospinal fluid in primary neurodegenerative dementias. J Alzheimers Dis 50, 539–546.

85.

Perrin

, Payton

, Malone

, Gilmore

, Davis

, Xiong

, Fagan

, Townsend

, Holtzman

(2013) Quantitative label-free proteomics for discovery of biomarkers in cerebrospinal fluid: Assessment of technical and inter-individual variation. PLoS One 8, e64314.