Xenophagy

Selective macroautophagic capture and lysosomal degradation of intracellular pathogens — bacteria, viruses, and parasites that breach cytosolic or vacuolar compartments of the host cell. Xenophagy is the innate immune arm of the autophagy system: it diverts the bulk degradation machinery toward microbial targets, reducing pathogen burden before adaptive immunity is mobilized. The first demonstration of xenophagy as a functional defense mechanism was published by Gutierrez et al. in 2004 using Mycobacterium tuberculosis in macrophages ¹.

Xenophagy intersects with aging biology primarily through immunosenescence: aged macrophages show reduced xenophagy flux, impaired TBK1 signalling, and heightened susceptibility to intracellular pathogens. Mutations in the xenophagy cargo receptor optineurin (optn) and its activating kinase TBK1 also cause ALS/FTD, linking the same molecular machinery to both antimicrobial defense and neurodegenerative disease. needs-human-replication

Discovery

Gutierrez et al. (2004) demonstrated that M. tuberculosis (BCG and virulent H37Rv strains) survives phagosomal arrest in macrophages under normal conditions, but that pharmacological or cytokine-induced activation of autophagy (via rapamycin or IFN-γ) delivers mycobacteria to acidified autolysosomes and kills them ¹. Rapamycin (mTOR inhibitor) treatment of BCG-infected macrophages reduced bacterial colony-forming units significantly and increased co-localisation of LC3+ autophagosomes with mycobacteria-containing vacuoles. This established three foundational principles:

1. The autophagy machinery can be weaponised against pathogens occupying vacuolar compartments.
2. mTOR suppression (as by nutrient deprivation or rapamycin) potentiates xenophagy.
3. Pathogens actively resist this arm of innate immunity — wild-type virulent M. tuberculosis is substantially more autophagy-resistant than BCG.

Dimension	Status
Pathway conserved in humans?	yes — cargo receptors, LC3 family, and autophagy core are conserved
Phenotype conserved in humans?	yes — human macrophages kill intracellular bacteria via autophagy; HIV/TB/Salmonella xenophagy documented in human cell lines
Replicated in humans?	partial — human cell-line and ex vivo macrophage studies; controlled in vivo human xenophagy flux data absent needs-human-replication

Mechanism

Xenophagy proceeds through four conceptual stages, each involving dedicated molecular sensors or receptors that are distinct from (but share machinery with) other selective autophagy subtypes.

Stage 1 — Cytosolic pathogen detection

Two complementary detection strategies exist:

Galectin-based damage sensing (vacuolar pathogens): When bacteria such as Salmonella enterica Typhimurium escape their vacuole (the Salmonella-containing vacuole, SCV) — or when the vacuole is damaged by bacterial toxins — host galectins are rapidly recruited to the damaged membrane:

• Galectin-8 (LGALS8): senses SCV rupture by binding exposed intraluminal glycans (specifically β-galactosides on the luminal leaflet of the damaged vacuolar membrane) ². Galectin-8 then directly recruits NDP52 (ndp52), which engages the autophagy machinery. This galectin-8 → NDP52 axis was identified by Thurston et al. 2012 as an early damage signal that precedes and is independent of ubiquitin coating ².
• Galectin-3 and galectin-9: similarly bind damaged endolysosomal membranes; galectin-9 recruits TAX1BP1; galectin-3 recruits multiple downstream effectors including p62. These galectin arms help generate redundancy in the sensor layer. needs-replication — the relative contributions of each galectin are not cleanly resolved in human primary macrophages.

Direct ubiquitination (cytosolic pathogens): When bacteria escape into the cytoplasm (e.g., Listeria monocytogenes, cytosolic Salmonella), host E3 ubiquitin ligases ubiquitinate bacterial surface proteins and/or bacterial outer-membrane components, generating a poly-ubiquitin coat. The ubiquitin chains — predominantly K48- and K63-linked — serve as the docking platform for cargo receptors.

Stage 2 — Cargo receptor recruitment

Four primary selective autophagy receptors with distinct but overlapping specificities bind the ubiquitinated pathogen and bridge it to the autophagy membrane via their LC3-interacting region (LIR) motifs:

Receptor	Gene	Binding partner	Primary link to xenophagy	Notes
p62 / SQSTM1	SQSTM1	K63-Ub, K48-Ub (UBA domain)	Earliest described; broadly recruited to ubiquitinated bacteria	NF-κB scaffold; also involved in aggrephagy; not always required for xenophagy in cells with redundant receptors
ndp52 / CALCOCO2	CALCOCO2	Ubiquitin (ZnF domain); Galectin-8 (SKICH domain) 2	Ubiquitin-binding and TBK1 recruitment established by Thurston 2009 3; galectin-8-mediated recruitment to damaged SCVs established by Thurston 2012 2; primary receptor for Salmonella xenophagy in HeLa cells	TBK1 phosphorylation of NDP52 enhances LC3 affinity; NDP52 selectively binds LC3C isoform (not LC3B)
optn / Optineurin	OPTN	Ubiquitin (UBAN domain, K63 preference); myosin VI	Phosphorylation at Ser177 by TBK1 dramatically increases LC3 affinity 4; primary functional receptor for Salmonella growth restriction in HeLa	ALS/FTD disease gene; also mediates mitophagy
TAX1BP1 / CALCOCO3	TAX1BP1	K63-Ub (ZF2 domain, primary); K48-Ub and linear Ub also bound; myosin VI (ZF1+ZF2); galectin-9	Required for Salmonella clearance alongside NDP52; TAX1BP1 siRNA causes cytosolic accumulation of ubiquitin-positive bacteria (cargo-recognition defect); single knockdown of TAX1BP1 causes stronger xenophagy defect than NDP52 alone 5; TAX1BP1 is the dominant paralogue in vertebrates (NDP52 lost from Xenopus; truncated in mice)	Also called T6BP or CALCOCO3; binds LC3B and LC3C equally (unlike NDP52 which is LC3C-selective)

TBK1 (TANK-binding kinase 1) is the master kinase for xenophagy: it phosphorylates NDP52, OPTN (Ser177), and p62 (Ser403), enhancing their LC3 affinity and amplifying receptor clustering around the pathogen.

Stage 3 — Autophagosome encapsulation

Receptor clustering on the pathogen surface recruits the ULK1 complex and the Class III PI3K complex (Beclin-1/VPS34), which generates PI3P locally, nucleating a phagophore that expands around the pathogen. The ATG5–ATG12–ATG16L1 elongation complex and LC3 lipidation machinery (ATG7 → ATG3 → LC3-PE) coat the growing membrane ⁶.

For large targets (e.g., M. tuberculosis-containing vacuoles ~2–4 µm), complete engulfment requires coordinated phagophore extension; partial engulfment followed by membrane sealing has been documented. Myosin VI (a minus-end-directed actin-based motor) is recruited by NDP52 and TAX1BP1 (via their overlapping ubiquitin/myosin VI binding sites in the zinc-finger domains) to deliver endosomal membrane components to the nascent xenophagosome ⁵. Crucially, myosin VI loss causes accumulation of ubiquitin-positive Salmonella inside LC3-positive autophagosomes — a later maturation/degradation defect distinct from the cargo-recognition defect caused by receptor (TAX1BP1/NDP52/OPTN) depletion.

Stage 4 — Lysosomal fusion and bacterial degradation

The sealed xenophagosome fuses with the lysosome via SNARE machinery (STX17/SNAP29/VAMP8), forming the xenolysosome/autolysosome. Lysosomal hydrolases (cathepsins B, D, L) and acidification (pH 4.5–5.0) degrade bacterial cell walls and proteins. Reactive oxygen species generated by lysosomal NADPH oxidase (NOX2) contribute to bactericidal activity in some contexts.

Receptor specificity and the TBK1 axis

The four major xenophagy receptors are functionally non-redundant in vivo despite overlapping domain architecture. Key specificity findings:

• NDP52 is the primary restrictor of Salmonella growth in epithelial cells. Thurston et al. 2009 demonstrated that NDP52 siRNA causes hyper-proliferation of intracellular Salmonella in HeLa cells and that NDP52 is recruited to ubiquitin-coated cytosolic bacteria via its zinc-finger domain (ubiquitin binding) ³; an additional galectin-8 → NDP52 recruitment pathway (independent of ubiquitin) was established later by Thurston 2012 ².
• OPTN phosphorylation at Ser177 by TBK1 is functionally essential: Wild et al. 2011 showed that phospho-OPTN (pSer177) has dramatically higher LC3 affinity than unphosphorylated OPTN, and that a Ser177Ala mutant OPTN fails to restrict Salmonella growth, even though it still binds ubiquitin ⁴. This establishes TBK1-mediated OPTN phosphorylation as a rate-limiting step for xenophagy flux.
• TAX1BP1 acts cooperatively with NDP52 at the cargo-recognition step: TAX1BP1 siRNA causes accumulation of cytosolic ubiquitin-positive bacteria (a receptor/recognition defect), and simultaneous knockdown of TAX1BP1+NDP52 has a stronger effect than either alone, indicating partial functional redundancy ⁵. Myosin VI loss produces a distinct, later-stage defect: ubiquitin-positive bacteria accumulate inside LC3-positive autophagosomes, indicating a maturation/degradation failure rather than a recognition failure. Depletion of myosin VI alone causes a stronger Salmonella hyper-proliferation phenotype than simultaneous depletion of TAX1BP1 + NDP52 + OPTN (triple knockdown) ⁵. needs-replication — most mechanistic data is from HeLa and RPE cell lines; primary human macrophage validation is limited.

Pathogen evasion strategies

Several major intracellular pathogens have evolved counter-strategies that specifically subvert xenophagy:

Mycobacterium tuberculosis

Virulent M. tuberculosis resists xenophagy through multiple mechanisms:

• Phagosomal maturation arrest: The bacteria arrest their phagosome at an early endosomal stage (Rab5+/Rab7−), preventing maturation to the acidified phagolysosome that would activate lysosomal killing. Gutierrez et al. 2004 demonstrated that pharmacological or starvation-induced autophagy can overcome this block, forcing mycobacterial phagosomes to acquire late-endosomal/lysosomal markers (cathepsin D, LAMP-1, Vo-ATPase) and suppressing bacterial viability ¹.
• ESX-1 type VII secretion system (from later literature — not established in Gutierrez 2004 ¹): the ESX-1 effector EsxA (ESAT-6) has been proposed to disrupt SNARE-mediated autophagosome–lysosome fusion; this attribution is supported by subsequent studies not yet on this wiki. unsourced — a primary source for the ESX-1/SNARE disruption claim needs to be identified and added. no-mechanism
• IFN-γ, which transcriptionally upregulates autophagy and xenophagy factors, partially overcomes this block — providing a mechanistic rationale for why impaired IFN-γ signalling in immunosenescence increases TB susceptibility ⁶. needs-human-replication

Listeria monocytogenes

Listeria escapes from its vacuole into the cytoplasm via listeriolysin O (LLO) and recruits host actin (via ActA) to propel itself through the cytosol. ActA coating of the bacterial surface disguises the bacterium from autophagic recognition: ActA-dependent actin polymerization blocks bacterial association with ubiquitin and prevents p62 recruitment and autophagic targeting ⁶⁷. Only non-motile Listeria (ActA-deleted mutants) are efficiently cleared by xenophagy. The precise mechanism — whether ActA physically occludes receptor-binding sites or diverts actin polymerization to mask the bacterial surface as a host organelle — is not fully resolved. no-mechanism

Salmonella enterica Typhimurium

A subpopulation of Salmonella escapes the SCV and is targeted by xenophagy; however, the intravacuolar population manipulates the endocytic pathway to sustain its replicative niche. The bacterial effector SopF inhibits ATG16L1 recruitment, partially blocking autophagosome nucleation around the SCV.

Role in aging and immunosenescence

Xenophagy decline in aged macrophages

Aged macrophages show reduced autophagic flux that extends to xenophagy. Mechanistically proposed contributors include:

• Reduced expression of ATG proteins (ATG5, ATG7) in aged myeloid cells
• Impaired lysosomal acidification in aged macrophages (consistent with lipofuscin accumulation and reduced cathepsin activity)
• Chronic mTOR over-activation in aged macrophages, which suppresses ULK1-mediated autophagy initiation
• Reduced TBK1 signalling capacity, impairing phospho-OPTN(Ser177) and phospho-NDP52 amplification

The functional consequence is heightened susceptibility to intracellular bacterial pathogens in elderly individuals — a well-documented clinical epidemiological pattern for M. tuberculosis (TB is 2–3× more common in adults aged 65+ in high-income countries) and Listeria (>70% of invasive listeriosis in the EU occurs in individuals 60+). Whether impaired xenophagy specifically (vs. other immune deficits) drives this vulnerability is not yet established in controlled human studies. needs-human-replication no-mechanism

TBK1 — bridge between xenophagy and neurodegeneration

TBK1 loss-of-function mutations are a major Mendelian cause of ALS and frontotemporal dementia (FTD) ⁸. Because TBK1 is required for both xenophagy (via OPTN/NDP52 phosphorylation) and mitophagy (via p62 phosphorylation at Ser403 and OPTN at Ser177), ALS-associated TBK1 variants impair selective autophagy of both damaged mitochondria and (presumably) bacterial targets. This creates a mechanistic link between innate immune signalling, mitophagy, and neurodegeneration through a shared kinase.

Dimension	Status
Pathway conserved in humans?	yes — TBK1, OPTN, NDP52, and p62 are conserved
Phenotype conserved in humans?	yes — TBK1/OPTN loss-of-function causes human ALS/FTD (genetic epidemiology)
Replicated in humans?	yes (genetic) — xenophagy-specific flux in ALS patient macrophages not directly measured needs-replication

NF-κB and SASP amplification

p62 (p62) bridges xenophagy cargo recognition to NF-κB activation: p62 interacts with TRAF6 and atypical PKC isoforms to activate NF-κB, which transcriptionally upregulates inflammatory cytokines. In the context of aging, elevated chronic p62-NF-κB signalling (partly from impaired autophagy clearing p62 itself — a flux readout) contributes to chronic-inflammation / inflammaging. This is distinct from direct bacterial clearance but links failed xenophagy to the senescence-associated secretory phenotype (sasp)-like inflammatory milieu of aged tissues.

Pharmacology

Rapamycin (mTOR inhibition)

Rapamycin and rapalogs enhance xenophagy by relieving mTOR-mediated suppression of ULK1 and downstream autophagy initiation. This was the pharmacological proof-of-concept in the founding Gutierrez 2004 study ¹ and has since been replicated in multiple models of intracellular bacterial infection. Rapamycin reduces M. tuberculosis survival in macrophage culture and in murine infection models. needs-human-replication — no controlled human trial of rapamycin as an anti-TB adjunct has been completed as of 2026.

IFN-γ

IFN-γ is an endogenous xenophagy inducer: it transcriptionally upregulates LRG-47/IRGM (immunity-related GTPases) and other autophagy effectors, and enhances xenophagic flux in macrophages. The decline in IFN-γ responsiveness with age (a documented immunosenescence feature) is consistent with reduced xenophagy capacity. no-mechanism — the specific molecular steps linking IFN-γ receptor → TBK1/OPTN axis in aged macrophages are uncharacterized.

Caloric restriction

Caloric restriction reduces mTOR activity and increases AMPK signalling, which collectively upregulate autophagy including xenophagy. Whether CR specifically enhances xenophagy in aged animals at the organismal level (not just in young models) has not been tested directly. needs-human-replication

Measurement notes

• Xenophagy flux is typically measured by colony-forming unit (CFU) assays of intracellular bacteria ± autophagy inhibitors (3-methyladenine, bafilomycin A1, ATG5 siRNA). A decrease in intracellular CFU that is reversed by autophagy inhibition is considered evidence of xenophagy-mediated killing.
• LC3 co-localisation with bacteria-containing vacuoles (by immunofluorescence or live GFP-LC3) provides spatial evidence of autophagosomal targeting but does not confirm degradation.
• Receptor co-localisation with ubiquitin-coated bacteria (p62/NDP52/OPTN puncta) is a widely used early-step readout.
• Distinction from LC3-associated phagocytosis (LAP): LAP recruits LC3 to the outer face of phagosomes (single-membrane, Rubicon-dependent) rather than forming a canonical double-membrane autophagosome; LAP occurs in parallel with xenophagy and is mechanistically distinct. Distinguishing the two requires EM or Rubicon/ATG14L1 co-localisation experiments.

Limitations and gaps

• Most mechanistic xenophagy data is from epithelial cell lines (HeLa, U2OS) or murine macrophage lines; validation in primary human macrophages — especially aged primary macrophages — is limited. needs-human-replication
• Quantitative xenophagy flux in aged human macrophages has not been directly measured; the aging-immunosenescence connection is inferred from indirect evidence (reduced autophagy protein expression + clinical epidemiology). no-mechanism
• Receptor redundancy vs. non-redundancy differs by cell type and pathogen; the dominant receptor for each pathogen in the relevant primary human cell type is not always established. needs-replication
• Whether boosting xenophagy pharmacologically (e.g., via mTOR inhibition) has beneficial effects on infectious disease outcomes in elderly humans is untested. long-term-unknown
• The molecular mechanism by which M. tuberculosis ESX-1 blocks autophagosome–lysosome fusion is incompletely resolved at the molecular level. no-mechanism

Related pages

• autophagy — parent process; bulk macroautophagy machinery
• mitophagy — overlapping cargo receptors OPTN, NDP52, p62, TAX1BP1 with shared TBK1 axis
• p62 — cargo receptor; NF-κB link; Nrf2 activator
• ndp52 — primary receptor for galectin-8-mediated xenophagy; R6e parallel seed
• optn — TBK1-phosphorylated receptor; ALS disease gene; R6e parallel seed
• tbk1 — master xenophagy/mitophagy kinase; ALS/FTD Mendelian gene (implicit stub)
• tax1bp1 — fourth canonical receptor; myosin VI co-factor (implicit stub)
• galectin-8 — early damage sensor for ruptured vacuolar membranes (implicit stub)
• disabled-macroautophagy — hallmark linked to aging; xenophagy is an arm of this failure
• chronic-inflammation — downstream of impaired xenophagy via p62–NF-κB axis
• sasp — context for p62–NF-κB contribution to aging inflammation
• immunosenescence — phenotype of aged immune decline that encompasses xenophagy failure (implicit stub)
• inflammaging — chronic low-grade inflammation of aging; partly p62/NF-κB mediated

Footnotes

Footnotes

1. doi:10.1016/j.cell.2004.11.038 · in-vitro · model: RAW 264.7 murine macrophages + BCG/H37Rv M. tuberculosis; also validated in bone marrow-derived and human peripheral blood monocyte-derived macrophages · Gutierrez et al. 2004 Cell · founding xenophagy paper; starvation and rapamycin (50 µg/ml) + IFN-γ activate autophagic killing of mycobacteria; virulent H37Rv is suppressed by autophagy induction but paper does not identify ESX-1 as the evasion mechanism (ESX-1/EsxA attribution is from later literature) · archive: bronze OA (downloaded) ↩ ↩² ↩³ ↩⁴ ↩⁵

2. doi:10.1038/nature10744 · in-vitro · model: HeLa cells + Salmonella enterica Typhimurium · Thurston et al. 2012 Nature · galectin-8 binds damaged SCV membranes (exposed β-galactosides) and directly recruits NDP52 as an early pathogen-detection signal independent of ubiquitin coating; galectin-8 KD impairs xenophagy · archive: not_oa no-fulltext-access — claims not independently verified against full PDF ↩ ↩² ↩³ ↩⁴ ↩⁵

3. doi:10.1038/ni.1800 · in-vitro · model: HeLa cells + Salmonella enterica Typhimurium · Thurston et al. 2009 Nature Immunology · identifies NDP52 as a novel cytosolic autophagy receptor for ubiquitin-coated Salmonella; NDP52 binds ubiquitin via its zinc-finger domain and recruits TBK1 complexes (via Nap1/Sintbad adaptors); NDP52 siRNA causes hyper-proliferation of intracellular bacteria; NDP52 detected on ubiquitin-coated bacteria in LAMP1-negative (cytosolic) compartment · note: galectin-8 recruitment to NDP52 is from Thurston 2012, not this paper · local PDF available ↩ ↩²

4. doi:10.1126/science.1205405 · in-vitro · model: HeLa/U2OS cells + Salmonella enterica Typhimurium · Wild et al. 2011 Science · TBK1 phosphorylates OPTN at Ser177 → dramatically enhanced LC3 affinity; Ser177Ala mutant fails to restrict bacterial growth despite intact ubiquitin binding; OPTN is rate-limiting xenophagy receptor in this system · archive: not_oa no-fulltext-access — claims not independently verified against full PDF ↩ ↩²

5. doi:10.1371/journal.ppat.1005174 · in-vitro · model: HeLa/RPE cells + Salmonella Typhimurium; also MEFs from myosin VI KO (Snell’s waltzer) mice · Tumbarello et al. 2015 PLoS Pathogens · TAX1BP1 is a novel xenophagy receptor recruited to ubiquitin-positive cytosolic Salmonella via its ZF2 domain; TAX1BP1 siRNA causes accumulation of ubiquitin-positive bacteria (cargo-recognition defect); myosin VI loss causes distinct later-stage defect — ubiquitin-positive bacteria accumulate inside LC3-positive autophagosomes (maturation defect); myosin VI KD causes stronger Salmonella hyper-proliferation than simultaneous triple KD of TAX1BP1+NDP52+OPTN; ubiquitin-binding site in all three receptors overlaps with myosin VI binding site · gold OA (downloaded) ↩ ↩² ↩³ ↩⁴

6. doi:10.1038/nature09782 · review · model: multi-system (in-vitro + in-vivo + human epidemiology) · Levine B, Mizushima N & Virgin HW 2011 Nature Vol 469 pp 323–335 · comprehensive review of autophagy in immunity and inflammation; covers xenophagy mechanisms, IFN-γ, bacterial evasion (ActA/Listeria; IcsB/Shigella; mycobacterial evasion strategies), and host–pathogen co-evolution; Table 1 catalogs key autophagy proteins and their immune functions · local PDF available; note: task brief cited wrong DOI (10.1016/j.cell.2010.12.024 = Chromatin Remodeling SnapShot); corrected to 10.1038/nature09782 ↩ ↩² ↩³

7. doi:10.4161/auto.5.8.10177 · in-vitro · model: murine macrophages + L. monocytogenes · Bhatt 2009 Autophagy · ActA coating of cytosolic Listeria is required for autophagy evasion; ActA-deficient mutants are efficiently captured by autophagy · bronze OA (pending download) ↩

8. doi:10.1080/15548627.2021.1926656 · review · model: ALS patient cohorts + cell-culture · Chua, De Calbiac, Kabashi & Barmada 2022 Autophagy · TBK1 loss-of-function mutations are frequent in familial ALS; TBK1 regulates selective autophagy (xenophagy and mitophagy) via OPTN/NDP52/p62 phosphorylation; mechanistic links to neurodegeneration reviewed · pending download ↩