New insights into protein balance and treatments for human diseases

Cardiomyocytes exhibit intrinsically high expression levels of HSPs, reflecting the heart’s continuous exposure to mechanical load, metabolic demand, and oxidative stress. Among the HSP families, small heat shock proteins (sHSPs), also referred to as HSPB proteins, are particularly abundant in cardiac tissue (Fig. 2a). These proteins play essential roles in maintaining cytoskeletal integrity and promoting cardiomyocyte survival under both basal and stress conditions. During myocardial ischemia, arising from pathological events such as unstable atherosclerotic plaque rupture, coronary occlusion, or congestive heart failure, the expression of several sHSPs is markedly upregulated. These include HSPB1 (HSP27), HSPB5 (αB-crystallin), HSPB6 (HSP20), HSPB7 (cardiovascular HSP), and HSPB8 (HSP22). Induction of these chaperones represents an adaptive stress response that protects cardiomyocytes by stabilizing the contractile apparatus and preserving sarcomeric organization during ischemic injury.¹⁸ The coordinated upregulation of sHSPs underscores their central role in sustaining myocardial structural integrity under acute and chronic stress.

In addition to their structural roles, HSPs are integral components of the cardiac defense system against oxidative stress. Myocardial ischemia and reperfusion are accompanied by excessive production of reactive oxygen species (ROS), which contribute to ventricular dysfunction, arrhythmogenesis, and cardiomyocyte death. HSPs mitigate these deleterious effects by functioning as part of a broader free radical scavenger network, thereby reducing intracellular ROS accumulation and limiting oxidative damage (Fig. 1a). HSPs act synergistically with classical antioxidant enzymes, including superoxide dismutase and glutathione peroxidase, to enhance cardioprotection. This cooperative interaction amplifies the heart’s capacity to neutralize oxidative insults and preserve cellular homeostasis. Notably, HSP70 has been shown to interact with catalase-associated pathways, further augmenting antioxidant defenses and improving cardiac function following ischemic injury.¹⁹ Through these mechanisms, HSPs not only prevent protein misfolding and aggregation but also actively modulate redox balance, reinforcing their multifaceted role in cardiomyocyte survival.

Among cardiac HSPs, HSPB5 (αB-crystallin) has emerged as a particularly critical and disease-relevant chaperone. This ~20 kDa protein exhibits highly specific subcellular localization within cardiomyocytes, concentrating at the Z-band where it associates with desmin intermediate filaments. Under stress conditions, such as heat shock or ischemia, αB-crystallin binds to actin filaments, preventing their aggregation and thereby preserving cytoskeletal integrity and contractile function (Fig. 2a).²⁰ The pathological significance of αB-crystallin is underscored by genetic evidence. A missense mutation (R120G) in αB-crystallin leads to abnormal aggregation of both desmin and αB-crystallin, resulting in severe cardiomyopathy known as desminopathy. This condition is characterized by disrupted cytoskeletal architecture and progressive cardiac dysfunction.²¹ Beyond its role in inherited disease, αB-crystallin also contributes to myocardial mechanics in acquired heart failure. In failing human myocardium, αB-crystallin has been shown to reduce elevated cardiomyocyte stiffness, potentially by limiting titin aggregation and improving sarcomeric compliance.²²

Collectively, these findings establish HSPs, particularly members of the sHSP family, as central determinants of cardiac proteostasis, structural stability, and stress resilience. By stabilizing cytoskeletal elements, preventing protein aggregation, mitigating oxidative stress, and preserving the mechanical properties of cardiomyocytes, HSPs provide robust protection against ischemic and oxidative injury. The stress-induced upregulation of HSPs during myocardial ischemia highlights their adaptive significance, while disease-associated mutations underscore their pathological relevance. A deeper understanding of HSP regulation and function in the heart offers promising opportunities for the development of targeted therapeutic strategies in cardiovascular diseases.

In the absence of effective autophagy, cellular waste and misfolded proteins accumulate, generating proteotoxic stress, structural disorganization, and organelle dysfunction. Cardiomyocytes, which lack meaningful regenerative potential, are particularly vulnerable to the cumulative effects of aging and chronic stress. Studies using cardiac-specific Atg5 deletion models have been pivotal in establishing this concept (Fig. 2b). Although these mice initially exhibit normal cardiac function, they progressively develop left ventricular dilation, reduced fractional shortening, disorganized sarcomeres, and damaged mitochondria with aging, highlighting the essential role of sustained autophagy in maintaining structural integrity and contractile function over time.²³ The importance of autophagy extends to cardiac development, as evidenced by embryonic lethality and myocardial developmental defects in Atg13-knockout mice, underscoring the indispensable roles of Atg13 and FIP200 in both autophagy and heart formation (Fig. 2b).²⁴

Autophagy is dynamically activated in response to hemodynamic stress. In transverse aortic constriction models, autophagic activity rises rapidly and remains elevated during hypertrophic remodeling. Modulating autophagic regulators profoundly influences these outcomes. Reduction of Beclin1 impairs adaptive remodeling, whereas excessive Beclin1 promotes autophagy beyond beneficial levels and exacerbates adverse remodeling.²⁵ Thus, while autophagy is clearly protective under physiological and early stress conditions, its chronic hyperactivation may become maladaptive. Overall, constitutive autophagy maintains cardiomyocyte size, architecture, and basal performance in the healthy heart, whereas in failing myocardium, enhanced autophagy functions predominantly as a compensatory survival mechanism.²⁶ This dualistic behavior reflects the complexity and context-dependence of autophagy in cardiac physiology.

Several molecular regulators refine this autophagic network. FYCO1 is an emerging mediator that interacts with LC3, Rab7, and phosphatidylinositol-3-phosphate to regulate autophagosome trafficking and maturation.²⁷ Its overexpression increases autophagic flux, promotes the formation of autophagosomes even under nutrient-rich conditions, and improves contractile adaptation under stress, partly by modulating Myh7 expression.²⁸ Epigenetic regulators also influence autophagy. Histone deacetylases (HDACs) integrate stress-responsive signaling with autophagic activation; HDAC inhibition modulates autophagy, limits phenylephrine-induced hypertrophy, and preserves near-normal cardiac function in pre-hypertrophic models (Fig. 2b).²⁹ Collectively, these insights reinforce that autophagy is neither merely a degradation pathway nor a binary process but a carefully tuned homeostatic program that determines whether the myocardium adapts or deteriorates under stress.

Parallel to macroautophagy, mitophagy plays a particularly critical role in safeguarding mitochondrial quality control (MQC), especially given the extreme energetic demands of the heart. When mitophagy is dysregulated, damaged mitochondria accumulate, resulting in mitochondrial DNA damage, ROS amplification, energetic collapse, and cardiomyocyte death. In cardiac stress states such as myocardial infarction, mitophagy is frequently upregulated as a compensatory survival response to remove dysfunctional mitochondria and protect myocardial tissue.³⁰ Impaired mitophagy contributes to cardiac dysfunction and fibrosis in obesity, primary cardiomyopathies, and HIV-associated cardiomyopathy (HIVAC).^31,32 In HIVAC, disruptions in autophagy machinery lead to accumulation of abnormal mitochondria independent of viral entry pathways, whereas Highly Active Antiretroviral Therapy (HAART) agents such as azidothymidine and abacavir further exacerbate autophagy defects. Conversely, enhancing autophagy alleviates mitochondrial dysfunction and restores cardiomyocyte performance.

Among mitophagy pathways, the PINK1-Parkin axis is one of the best characterized (Fig. 1a). Altered expression of PINK1 and Parkin, including reduced Parkin levels, correlates strongly with mitochondrial dysfunction and contractile impairment in heart failure.^33,34 During ischemia‒reperfusion injury, ROS generation and mitochondrial fragmentation intensify; activation of PINK1-Parkin mitophagy removes damaged mitochondria, reduces oxidative stress, and limits apoptosis. Several therapeutic strategies exploit this pathway. Mitochondrion-targeted antioxidants such as MitoQ enhance PINK1-Parkin activity and reduce cardiomyocyte death.³⁵ Gene therapy restoring PINK1 or Parkin expression yields cardioprotective effects.³⁶ Interestingly, commonly used cardioprotective agents, including statins and ACE inhibitors, appear to indirectly enhance mitophagy signaling, highlighting their broader mechanistic relevance.³⁷ Rapamycin and other mTOR inhibitors similarly promote combined autophagy-mitophagy responses, improving mitochondrial integrity and cardiac outcomes in multiple disease models.³⁸

Beyond PINK1-Parkin, FUNDC1 has emerged as a key mitophagy regulator, particularly under hypoxia. Its activation during myocardial ischemia limits injury, whereas its deletion exacerbates ischemic damage.^39,40 FUNDC1 also links mitophagy to mitochondrial fission through DRP1 and is modulated by upstream regulators such as ULK1 and PGAM5, offering multiple therapeutic entry points.^39,40 Additional mitophagy receptors, including BNIP3 and NIX, contribute to stress-responsive mitochondrial clearance, particularly in hypoxic conditions.⁴¹ A complementary mechanism involves cardiolipin-mediated mitophagy, in which cardiolipin externalizes to the outer mitochondrial membrane and directly interacts with LC3 to trigger mitochondrial turnover.⁴² Stabilizing cardiolipin or enhancing its translocation improves mitochondrial clearance and reduces oxidative stress in heart failure models,⁴³ while enzymes involved in cardiolipin metabolism, such as tafazzin, represent additional therapeutic opportunities.⁴⁴

Together, autophagy and mitophagy form an integrated multilayer quality-control network essential for sustaining cardiac proteostasis, mitochondrial integrity, and mechanical performance. Their coordinated activation determines whether the heart maintains resilience under physiological demand and pathological stress or undergoes progressive failure. Understanding how to precisely modulate these pathways remains fundamental to advancing targeted therapies in cardiovascular disease.

Alongside autophagy and mitophagy, the ubiquitin‒proteasome system (UPS) represents a central pillar of cardiac proteostasis, responsible for the highly selective degradation of damaged, misfolded, or regulatory proteins and thereby safeguarding cellular homeostasis and protein turnover (Fig. 1a). Whereas autophagy can degrade large macromolecular complexes and even entire organelles, the UPS is characterized by exquisite substrate specificity, targeting only ubiquitinated proteins while preserving intact cellular structures. This selectivity is indispensable for normal cellular physiology, governing key processes such as cell cycle control, transcriptional and signal transduction regulation, receptor downregulation, endocytosis, immune responses, development, and apoptosis.

UPS-mediated degradation proceeds through a tightly ordered enzymatic cascade. Ubiquitin, a conserved 76-amino acid protein, is first activated in an ATP-dependent manner by E1 ubiquitin-activating enzymes, then transferred to E2 conjugating enzymes, and finally attached to target proteins by E3 ubiquitin ligases, which provide the critical element of substrate specificity (Fig. 2c). Ubiquitin is typically conjugated to lysine residues on substrate proteins, most classically via K48 linkages, forming polyubiquitin chains that are recognized by the 26S proteasome. Proteasomal processing results in complete proteolysis of the substrate, while deubiquitinating enzymes (DUBs) recycle ubiquitin molecules for subsequent rounds of degradation, ensuring the efficiency and sustainability of the pathway.

Although K48-linked ubiquitination is the canonical signal for proteasomal degradation, the UPS exhibits remarkable versatility through alternative ubiquitin linkages and chain architectures (Fig. 2c). K11-linked branched chains, for example, facilitate rapid turnover of key cell cycle regulators, whereas linkages such as K27, K29, and K33 can attenuate degradation, illustrating how chain topology fine-tunes proteolysis. Beyond proteasomal targeting, non-canonical ubiquitin modifications, including K63-linked polyubiquitination, regulate signaling, endocytosis, kinase activity, and DNA damage responses, while mono- and multi-ubiquitination influence chromatin dynamics, DNA repair, and subcellular localization (Fig. 2c). This combinatorial coding capacity underscores the adaptability of the UPS and its capacity to orchestrate diverse cellular outcomes beyond protein destruction.

In the heart, an organ composed largely of post-mitotic cardiomyocytes with limited regenerative capacity, UPS integrity is particularly critical. Failure of UPS surveillance promotes the accumulation of misfolded proteins, proteotoxic aggregates, fibrosis, and ultimately contractile dysfunction and heart failure. Thus, dissecting the contributions of E1, E2, and E3 components in cardiomyocytes is fundamental to understanding how UPS perturbations drive cardiovascular diseases.

Among UPS components, E3 ubiquitin ligases play a particularly decisive role owing to their ability to recognize specific substrates. Their diversity reflects the complexity of the proteome: mammals encode over 600 E3 ligases, far exceeding the number of E1 and E2 enzymes. E3 ligases are broadly categorized according to their catalytic domains. HECT and RBR ligases form covalent thioester intermediates with ubiquitin prior to substrate transfer, whereas RING-type ligases, the most abundant class, mediate the direct transfer of ubiquitin from E2 to the substrate.⁴⁸ Together with the diversity of ubiquitin linkage types, this architectural variability gives the UPS its extraordinary precision, enabling tight regulation of proteostasis and preventing the accumulation of toxic proteins under physiological and pathological stress.

E3 ligases are the most diverse and functionally specialized components of the UPS and are responsible for substrate recognition and the final transfer of ubiquitin to target proteins. They regulate cardiomyocyte growth, stress adaptation, signal transduction, ion channel turnover, autophagy-proteasome crosstalk, congenital heart disease mechanisms, and cardiomyopathies. Most cardiac E3 ligases belong to the RING-type family, although both single- and multi-subunit variants exist, each executing distinct regulatory roles. For example, the cardioprotective E3 ligase HectD3 suppresses Calcineurin-NFAT signaling and inflammatory stress by targeting SUMO2 and Stat1, thereby mitigating hypertrophic remodeling.⁴⁹ Conversely, individual TRIM family ligases exert opposing influences: TRIM32 promotes hypertrophy via dysbindin regulation, whereas TRIM24 restrains hypertrophic signaling, illustrating the nuanced, context-dependent contribution of E3 ligases to cardiac physiology and pathology.⁵⁰

Molecular chaperones, particularly HSPs, represent attractive therapeutic targets due to their central role in stabilizing misfolded proteins and suppressing aggregation under cardiac stress. HSP90 inhibitors such as ganetespib (STA-9090) and AUY922 have been extensively evaluated in oncology,^54,55 where they disrupt the stabilization of oncogenic client proteins. Although their clinical development has focused primarily on cancer, emerging studies are beginning to explore whether controlled modulation of HSP90 activity may also influence maladaptive proteostasis in cardiovascular disease.

In contrast, therapeutic strategies targeting other HSPs, including HSP70, HSP27, and αB-crystallin, remain at an earlier stage but have demonstrated encouraging results in preclinical cardiovascular models. Overexpression of HSP70 enhances the resistance of cardiomyocytes to ischemia-like stress and improves functional recovery following reperfusion. Similarly, HSP27 overexpression reduces infarct size and preserves cardiac function in animal models of myocardial infarction.^56–58 Collectively, these findings suggest that selective enhancement of cardioprotective chaperone functions, rather than broad inhibition, may offer a more viable and safer strategy for HSP-based interventions in heart disease.

Autophagy has emerged as one of the most actively investigated proteostasis pathways for therapeutic intervention in cardiovascular disease. Among currently available agents, hydroxychloroquine (HCQ) and chloroquine (CQ) are the most widely studied autophagy inhibitors in clinical use. Originally developed as antimalarial drugs, these compounds inhibit autophagy by blocking autophagosome-lysosome fusion, thereby preventing the degradation of autophagic cargo. HCQ and CQ are now being evaluated in combination regimens for multiple cancers, including pancreatic, breast, and lung malignancies.⁵⁹

However, their application in cardiovascular disease has raised important safety concerns. In patients with pre-existing heart failure, autophagy inhibition may exacerbate proteotoxic stress rather than confer benefit, prompting cautious evaluation in cardiac clinical trials.⁶⁰ These observations highlight the necessity of disease-stage-specific and quantitatively controlled autophagy modulation.

To overcome the limitations of first-generation inhibitors, next-generation autophagy inhibitors such as Lys05 and DQ661 have been developed. These compounds demonstrate greater potency and improved specificity in preclinical models. Lys05 exhibits enhanced autophagy inhibition and antitumor activity, while DQ661 targets both autophagy and lysosomal function, offering a dual mechanism of action.⁶¹ Although currently pursued primarily in oncology, these agents may ultimately enable more refined autophagy modulation in cardiovascular settings.

In parallel, several established drugs are being repurposed for their autophagy-modulating properties. Metformin, an AMP-activated protein kinase (AMPK) activator widely used in diabetes, is currently in phase I clinical trials for heart failure with preserved ejection fraction (HFpEF) (Tables 1 and 2). Metformin enhances autophagy, reduces cardiac hypertrophy, and shows promise for improving functional outcomes in heart failure. Rapamycin, an mTOR inhibitor, similarly promotes autophagy, attenuates pathological remodeling and is now in phase II clinical trials for HFpEF (Tables 1 and 2).

Sodium-glucose cotransporter-2 (SGLT2) inhibitors, including empagliflozin, canagliflozin, and dapagliflozin, represent another important class of repurposed agents with robust cardiovascular benefits. Originally developed for glycemic control, these drugs are increasingly recognized for their capacity to induce autophagy, which may contribute to their cardioprotective effects. Notably, the EMPEROR-Preserved trial demonstrated that empagliflozin significantly reduced cardiovascular mortality and hospitalization in HFpEF patients without increasing adverse events.⁶²

Recent advances in targeted protein degradation have transformed the ubiquitin‒proteasome system into a highly versatile therapeutic platform. Central to this approach are degrons, short amino acid motifs, typically 6–10 residues in length, that mark proteins for recognition by E3 ubiquitin ligases. Large-scale CRISPR-based screens have identified thousands of degron motifs, particularly at protein C-termini, regulated by Cullin-RING E3 ligase complexes such as CRL2 and CRL4.⁶³ Proteomic analyses further suggest selective depletion of C-terminal degrons in eukaryotic proteins, supporting the existence of “C-end rules” that shape proteome composition.⁶³ These insights have important implications for therapeutic manipulation of cardiac proteostasis.

Building upon degron biology, molecular glues have been developed to induce selective protein degradation. Compounds such as thalidomide and its analogs lenalidomide and pomalidomide promote interactions between E3 ligases (e.g., cereblon) and specific target proteins, triggering their ubiquitination and proteasomal degradation. Molecular glues are already approved for multiple myeloma and myelodysplastic syndromes and are under investigation for additional malignancies.⁶⁴ Their success has opened new opportunities for targeting disease-associated proteins previously considered undruggable, with emerging relevance for cardiovascular disorders.

Another major innovation is the development of proteolysis-targeting chimeras (PROTACs). These bifunctional molecules link a ligand for the target protein to a ligand recruiting an E3 ubiquitin ligase, such as VHL, MDM2, CRBN, or IAPs.^65,66 By forcing proximity between the target and ligase, PROTACs induce polyubiquitination and proteasomal degradation. Although initially developed for oncology, PROTAC-based strategies are now being explored in cardiovascular disease, where selective degradation of pathogenic proteins could mitigate hypertrophy, fibrosis and heart failure progression.

Deubiquitinating enzymes (DUBs) represent an additional regulatory layer with therapeutic potential. Ubiquitin-specific peptidase 14 (USP14) removes ubiquitin chains from proteasome-bound substrates, thereby modulating protein stability. Inhibitors of USP14 are being developed to enhance the degradation of oncogenic proteins.⁶⁷ While current efforts focus on cancer, similar strategies may be adaptable to cardiovascular disease, where enhanced clearance of misfolded or damaged proteins could improve outcomes in heart failure and ischemic injury.

ER-associated degradation and unfolded protein response signaling represent critical targets for alleviating proteotoxic stress in cardiomyocytes. One promising strategy involves modulation of activating transcription factor 6 (ATF6), a key mediator of adaptive UPR signaling. Small-molecule activators of ATF6 are being developed to enhance ER folding capacity, stimulate ERAD, and reduce ER stress, with potential benefits in heart failure and ischemic cardiomyopathy.⁶⁸

Enhancement of specific ERAD components offers another therapeutic avenue. E3 ubiquitin ligases such as Hrd1 play a central role in targeting misfolded ER proteins for degradation. Pharmacological upregulation of Hrd1 expression or activity may accelerate the clearance of misfolded proteins, thereby reducing ER stress and preventing cardiomyocyte apoptosis, an approach particularly relevant in diabetic cardiomyopathy.⁶⁹ Proteasome activators provide an additional strategy to support ERAD function. By increasing the overall degradation capacity, these agents facilitate more efficient clearance of ubiquitinated proteins and reduce the proteotoxic burden. Preclinical studies have demonstrated beneficial effects of proteasome activation in models of heart failure and ischemia‒reperfusion injury,⁶⁹ further supporting ER proteostasis as a therapeutically actionable axis.

Together, these therapeutic advances underscore the emergence of cardiac proteostasis as a druggable, systems-level target. Rather than manipulating individual pathways in isolation, effective intervention will likely require coordinated, context-dependent modulation of chaperones, autophagy, UPS activity, and ER stress responses. As precision medicine approaches advance, targeting proteostasis networks holds substantial promise for preserving myocardial function, slowing disease progression and reshaping the future of cardiovascular therapeutics.

Metabolic diseases, including obesity, T2DM, and NAFLD, now classified as MASLD, are characterized by chronic cellular stress that disrupts protein homeostasis.⁷⁰ This sustained imbalance manifests as ER stress, oxidative damage, mitochondrial dysfunction, and chronic low-grade inflammation, all of which accelerate disease progression. Central to the cellular defense machinery against such stress are heat shock proteins. By facilitating proper protein folding, stabilizing protein conformations, and limiting proteotoxic stress, HSPs play essential protective roles in metabolically active tissues. However, when dysregulated, these same pathways can exacerbate inflammation and metabolic dysfunction. Accordingly, altered HSP expression has been implicated in the pathogenesis of multiple metabolic disorders, positioning HSPs as both biomarkers of disease severity and promising, albeit complex, therapeutic targets (Fig. 3).

Among the HSP family, HSP70, particularly its inducible isoform HSP72 (also referred to as intracellular HSP72 or iHSP72), has been most extensively studied in metabolic disease contexts. Reduced iHSP72 levels in skeletal muscle and other insulin-responsive tissues are strongly associated with insulin resistance, a defining feature of T2DM. Experimental evidence indicates that iHSP72 mitigates metabolic dysfunction by suppressing inflammatory signaling, enhancing mitochondrial function, and stabilizing proteostasis. Mechanistically, iHSP72 inhibits the activation of stress kinases such as c-Jun N-terminal kinase (JNK), which plays a central role in promoting insulin resistance and metabolic inflammation. Consistent with these observations, animal models with iHSP72 overexpression exhibit improved insulin sensitivity and reduced systemic inflammation, reinforcing its therapeutic relevance.⁷¹

In the context of MASLD, iHSP72 exerts hepatoprotective effects by attenuating oxidative stress, reducing lipid accumulation, and limiting inflammatory responses. Increased iHSP72 expression in hepatocytes prevents steatosis and liver injury, underscoring its role in preserving hepatic metabolic homeostasis.^71,72 In contrast, extracellular HSP72 (eHSP72), which is released into circulation under conditions of cellular stress, is often associated with adverse metabolic outcomes. Elevated circulating eHSP72 levels correlate with chronic inflammation, insulin resistance, and pancreatic β-cell dysfunction in T2DM patients. Moreover, high eHSP72 levels have been linked to diabetic complications such as ketoacidosis, highlighting the importance of maintaining a finely tuned balance between intracellular and extracellular HSP72 for therapeutic intervention.⁷³

Beyond HSP72, several other HSP family members contribute to metabolic disease development in tissue-specific and context-dependent manners.

Adipose tissue represents a primary site where autophagy contributes to metabolic health. In obesity, autophagy markers such as LC3-II and autophagosome formation are elevated, accompanied by reduced mTORC1 activity, particularly in visceral depots.⁷⁹ This upregulation correlates with obesity severity, adipocyte hypertrophy, and adverse glucometabolic status, reflecting a tightly regulated interplay between autophagy, tissue remodeling, and metabolic stress.⁸⁰ Early obesity appears to engage autophagy as a compensatory defense against inflammation and lipid burden⁸¹; however, chronic exposure skews this response toward dysfunction. Maternal obesity similarly disrupts autophagy/mitophagy pathways, contributing to metabolic disturbances that adversely impact fetal development and postnatal metabolic health.⁸²

Genetic models reinforce the essential role of autophagy in adipose biology. Deletion of Becn2, Atg7, Tfeb, or Bif predisposes individuals to obesity and insulin resistance, highlighting autophagy’s role in systemic metabolic regulation.^83–85 Mechanistically, autophagy stabilizes PPARγ2, a master transcriptional regulator of adipogenesis; inhibition accelerates its proteasomal degradation and suppresses adipocyte differentiation.⁸⁶ Moreover, Atg5- and Atg7-dependent autophagy supports lipid droplet biogenesis and hormone-driven lipolysis, as their inhibition reduces lipid accumulation and adipogenic marker expression.⁸⁷ Together, these observations underscore autophagy as a determinant of adipocyte function, lipid storage dynamics, and inflammatory tone.

Beyond peripheral tissues, autophagy in the hypothalamus governs systemic energy homeostasis by integrating nutrient signals and adiposity cues. Impaired hypothalamic autophagy activates IKKβ signaling, promoting inflammation and disrupting metabolic regulation.⁸⁸ In POMC neurons, autophagy is required for intact leptin signaling and appetite suppression; its disruption aggravates leptin resistance, hyperphagia, and weight gain.⁸⁹ Importantly, restoration of hypothalamic autophagy via nutritional interventions such as docosahexaenoic acid-acylated astaxanthin monoester (AST-DHA) supplementation or genetic activation of Atg5 ameliorates obesity and improves glucose metabolism in experimental models,⁹⁰ demonstrating that hypothalamic autophagy is not merely correlative but mechanistically regulatory.

In the liver, autophagy is indispensable for maintaining cellular quality control by eliminating lipid droplets, damaged organelles, and toxic aggregates. Its impairment is a hallmark of MASLD and a key driver of disease progression toward MASH, fibrosis, and hepatocellular carcinoma.⁹¹ Although autophagosomes are often increased in MASLD, reductions in BECN1 and accumulation of p62/SQSTM1 indicate defective lysosomal degradation and impaired autophagic flux.1, 92 This dysfunction affects lipophagy, insulin signaling, inflammatory responses, and cellular stress pathways.

Lipophagy, a selective form of autophagy targeting lipid droplets, plays a central role in regulating hepatic triglyceride turnover. In early MASLD, lipophagy is induced to mitigate lipid overload; however, persistent metabolic stress progressively impairs this process, promoting triglyceride accumulation, inflammation, and hepatocellular injury.^93–96 Dysfunctional lysosomes and genetic variants such as IRGM further exacerbate disease progression.⁹⁷ mTORC1 functions as a major suppressor of lipophagy, and its inhibition restores autophagic flux and improves hepatic lipid handling.⁹⁸ TFEB-mediated lysosomal biogenesis is similarly essential for efficient lipophagy, while natural compounds such as naringin enhance autophagic flux and reduce steatosis in experimental MASLD.⁹⁹ These findings place autophagy-lipophagy failure at the mechanistic core of MASLD pathogenesis.

Complementing macroautophagy, mitophagy preserves mitochondrial quality control and thereby protects against cellular stress, inflammation, and metabolic deterioration. In obesity and insulin resistance, mitophagy markers, including LC3BII, Parkin, FUNDC1, and BNIP3, are generally reduced in both human and animal studies, reflecting compromised mitochondrial turnover. Although initially activated to buffer metabolic stress, the PINK1-Parkin pathway becomes suppressed under chronic overload, such as prolonged palmitate exposure, resulting in dysfunctional mitochondria, heightened oxidative stress, and activation of the NLRP3 inflammasome.^100–102

A similar biphasic pattern occurs in MASLD: mitophagy is enhanced during early disease to counteract lipid excess and maintain mitochondrial homeostasis, but prolonged metabolic stress suppresses this adaptive response.^103,104 Consequences include excessive ROS production, progressive lipid accumulation, mitochondrial swelling, cristae disruption, impaired β-oxidation, fibrosis, inflammasome activation, and transition from steatosis to MASH.^103,105,106 Thus, mitophagy impairment functions as a critical molecular turning point converting compensatory metabolic adaptation into persistent injury and organ failure.

Metabolic diseases are often accompanied by heightened oxidative stress, driven by ROS produced by malfunctioning mitochondria and persistent inflammation. When oxidative stress overwhelms cellular defenses, it can impair the 26S proteasome, leading to the accumulation of ubiquitinated proteins and disrupting protein homeostasis.¹⁰⁷ Similarly, ER stress, a hallmark of metabolic diseases, places additional demand on the UPS through the activation of the UPR. While the UPR initially enhances proteasome activity to manage misfolded proteins, chronic ER stress can overwhelm the UPS, compromise proteasome function and exacerbate cellular dysfunction. Proteasome activity is notably reduced in several tissues in obesity and T2DM. This has been demonstrated in animal studies and confirmed in humans, where impaired proteasome function in adipocytes is strongly linked to insulin resistance.¹⁰⁸ Interestingly, plasma levels of ubiquitin and 26S proteasomes are lower in obese individuals, showing an inverse correlation with body mass index (BMI).¹⁰⁹ Preclinical studies further indicate that alterations in the UPS contribute to the progression and perpetuation of obesity and its complications.¹¹⁰ However, the role of the UPS is tissue specific, e.g., in muscle cells from insulin-resistant individuals, proteasome activity is increased, contributing to muscle protein degradation and atrophy.¹¹¹ This highlights the complex and context-dependent role of the UPS in metabolic diseases.

Insulin receptors and insulin receptor substrates (IRS-1 and IRS-2) are crucial for insulin signaling, and E3 ligases such as MG53, c-Cbl, and Nedd4 can directly degrade these substrates, disrupting signaling and promoting insulin resistance.¹¹² Animal studies have shown that suppressing the activity of MG53 and SOCS1 improves insulin sensitivity and protects against metabolic dysfunction.^113,114 In MASLD and MASH, E3 ligases play dual roles, with both protective and pathological implications. Some proteins, such as TRIM31 and Gp78, mitigate liver damage by promoting lipid turnover and reducing inflammation.¹¹⁵ Others, such as Siah1 and TRIM8, exacerbate the condition by driving lipid accumulation and inflammation.^116,117 Additionally, Smurf1 stabilizes sterol regulatory element-binding protein-1c (SREBP-1c), a key transcription factor in lipid metabolism, by preventing its ubiquitination and degradation, while BTRC promotes the proteasomal degradation of adipose triglyceride lipase.^118,119

Chronic inflammation, a hallmark of metabolic disorders, is also influenced by E3 ligases. For instance, deleting Pellino3 exacerbates insulin resistance by increasing pro-inflammatory macrophages, while inhibiting ITCH protects against steatohepatitis by promoting anti-inflammatory macrophages.^120,121 In atherosclerosis, a metabolic disease driven by chronic inflammation and faulty lipid metabolism, E3 ligases such as FBXW2 and FBXO3 have been identified as key modulators of vascular inflammation.^122,123 Other E3 ligases, including Peli1, play vital roles in atherosclerotic plaque stability and foam cell formation, suggesting that targeting these ligases could impact disease progression.¹²⁴

DUBs, which reverse ubiquitination and regulate protein stability, localization, and activity, play critical roles in metabolic pathways and demonstrate both detrimental and protective effects in metabolic diseases. For example, USP19 promotes adipogenesis and exacerbates high-fat-diet-induced obesity and glucose intolerance in mice.¹²⁵ USP21, upregulated in the skeletal muscles of obese individuals, impairs mitochondrial activation and carbohydrate metabolism. Ablation of USP21 in mouse skeletal muscle ameliorates obesity and insulin resistance.¹²⁶ USP20 stabilizes HMG-CoA reductase (HMGCR), the rate-limiting enzyme in cholesterol biosynthesis, and its inhibition reduces lipid levels and improves insulin sensitivity in preclinical models.¹²⁷ USP14 and USP2 deubiquitinate fatty acid synthase (FASN), a key enzyme in lipogenesis, aggravating hepatic steatosis and insulin resistance.^128,129 Conversely, USP4, USP10, and USP18 mitigate metabolic dysfunction by reducing hepatic steatosis, hyperglycemia, and inflammation, suggesting therapeutic potential against MASLD and T2DM.^130–132 USP7 improves insulin sensitivity and suppresses gluconeogenesis, offering another example of a DUB with beneficial effects on glucose homeostasis.¹³³

Among ERAD components, the Sel1L-HRD1 complex is highly conserved across species and has emerged as a central regulator of systemic metabolism, influencing diverse processes ranging from appetite control to lipid handling and insulin secretion.¹³⁴ The Sel1L-HRD1 ERAD complex functions as a metabolic checkpoint that integrates nutrient availability, hormonal signaling, and cellular stress responses across multiple tissues. By governing the degradation of key metabolic regulators, this complex ensures adaptive metabolic flexibility under fluctuating physiological demands. Disruption of Sel1L-HRD1 activity leads to tissue-specific ER stress and systemic metabolic imbalance, underscoring its pivotal role in maintaining whole-body homeostasis.¹³⁴

At the level of the central nervous system, ERAD plays a decisive role in the neuroendocrine regulation of energy balance. In the hypothalamus, the Sel1L-HRD1 complex is indispensable for proper leptin signaling and the folding of POMC, a precursor of neuropeptides that suppress appetite and regulate energy expenditure. Disruption of ERAD in this context impairs leptin responsiveness, resulting in hyperphagia, excessive weight gain, and heightened susceptibility to diet-induced obesity in mice.^135,136 ERAD additionally contributes to metabolic regulation in the gastrointestinal tract. In the small intestine, ERAD controls the stability of enzymes involved in lipid absorption. Inhibition of ERAD leads to the accumulation of these enzymes, enhancing dietary fat uptake and promoting obesity.¹³⁷ Together, these findings position ERAD as a critical regulator of energy intake by coordinating central appetite control with peripheral nutrient absorption.

Beyond energy intake, ERAD exerts profound effects on energy expenditure and lipid distribution through its actions in adipose tissue. In brown adipose tissue, the Sel1L-HRD1 complex is essential for maintaining mitochondrial integrity and thermogenic capacity. BAT-specific disruption of ERAD results in abnormal mitochondrial morphology, impaired heat production, and increased cold sensitivity, highlighting the importance of ERAD in adaptive thermogenesis.^138,139 Interestingly, adipocyte-specific Sel1L deficiency produces a distinct systemic phenotype. Although these mice are resistant to diet-induced obesity, they exhibit elevated postprandial triglyceride levels, reflecting dysregulated lipid handling rather than reduced caloric intake. Mechanistically, Sel1L stabilizes lipoprotein lipase (LPL) in adipocytes, thereby controlling its secretion and regulating lipid clearance from circulation.¹³⁴ These observations underscore the complex and context-dependent effects of ERAD in adipose tissue, linking local proteostasis to systemic lipid homeostasis.

The liver represents a major metabolic hub where ERAD intersects with circadian and transcriptional control of metabolism. Hepatic Sel1L-HRD1 activity exhibits circadian rhythmicity mediated by the transcription factor CREBH and the nuclear receptor PPARα. Through regulated turnover of CREBH, ERAD orchestrates the rhythmic expression of genes involved in fatty acid oxidation, triglyceride lipolysis, and gluconeogenesis. Disruption of this regulatory axis perturbs energy balance and compromises metabolic homeostasis.¹⁴⁰ ERAD also modulates hepatic endocrine signaling by controlling the transcription of FGF21, a hormone that enhances energy expenditure and insulin sensitivity. HRD1-mediated degradation of CREBH regulates FGF21 expression, linking ERAD directly to systemic metabolic adaptation.¹⁴¹ Notably, mice with hepatocyte-specific HRD1 deficiency display increased energy expenditure and resistance to high-fat diet-induced obesity and insulin resistance.¹⁴² In addition, ERAD directly restrains hepatic lipogenesis by targeting ATP-citrate lyase (Acly), a key enzyme that channels citrate toward fatty acid synthesis. HRD1-mediated degradation of Acly reduces hepatic lipid accumulation, thereby mitigating MASLD and reinforcing ERAD’s protective role against metabolic liver disease.¹⁴³

ERAD further shapes metabolic health by regulating cholesterol and lipoprotein metabolism. A central function of ERAD is the degradation of 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), the rate-limiting enzyme in cholesterol biosynthesis. Dysregulation of this pathway leads to excessive cholesterol accumulation and contributes to hypercholesterolemia, a common feature of insulin resistance and metabolic syndrome.¹⁴⁴ ERAD also governs the quality control of apolipoprotein B (ApoB), a critical structural component of very-low-density lipoproteins (VLDL). When ApoB fails to fold correctly, it is targeted for ERAD-mediated degradation, preventing aberrant VLDL assembly and secretion.¹⁴⁵ Perturbations in this quality-control mechanism exacerbate hyperlipidemia and hepatic steatosis, further illustrating how ERAD links intracellular proteostasis to circulating lipid levels and cardiometabolic risk.

Pancreatic β-cells are uniquely vulnerable to ER stress due to their high insulin biosynthetic demand. ERAD plays a protective role by clearing misfolded insulin and maintaining ER function, thereby supporting glucose-stimulated insulin secretion (GSIS). The Sel1L-HRD1 complex is essential for preserving β-cell identity and secretory competence.¹⁴⁶ Mice with β-cell-specific disruption of Sel1L-HRD1 develop early-onset diabetes and glucose intolerance, reflecting impaired insulin secretion and β-cell dysfunction. Importantly, Sel1L expression is reduced in human islets from T2DM patients, suggesting that ERAD impairment contributes directly to β-cell failure and disease progression.¹⁴⁶ These findings position ERAD as a critical safeguard of endocrine pancreatic function and a potential therapeutic target for diabetes.

Given their central roles in metabolic homeostasis, heat shock proteins (HSPs) have emerged as attractive therapeutic targets in metabolic disease. Multiple strategies have been explored to modulate HSP activity, including pharmacological agents, lifestyle interventions, and emerging gene therapy approaches.

HSP90 is frequently overexpressed in cancers such as cholangiocarcinoma, lung cancer, gastric cancer, breast cancer, and glioblastoma, and its elevated levels are often associated with poor prognosis.¹⁸⁹ HSP90 also promotes the activation of oncogenic protein kinases such as JAK2/STAT3, PI3K/AKT, and MAPK, facilitating cancer progression.¹⁹⁰ Additionally, HSP90 activates HIF-1α and NF-kB, driving critical oncogenic processes such as EMT, invasion, motility, and metastasis.¹⁹¹ HSP90 interacts with and stabilizes HMGCR, the rate-limiting enzyme in the mevalonate pathway, which is crucial for cancer progression. HSP90 also promotes tumor angiogenesis by enhancing the transcription and expression of vascular endothelial growth factor receptors (VEGFRs), which are essential for endothelial cell-dependent angiogenesis.¹⁹² In prostate cancer, HSP90 stabilizes the androgen receptor, enhancing its affinity for androgen and supporting tumor formation.^188,193

In early tumorigenesis, autophagy functions predominantly as a tumor-suppressive mechanism by removing damaged organelles, misfolded proteins, and oncogenic substrates such as p62, thereby preventing genomic instability and malignant transformation.²⁰⁵ Loss of key autophagy genes, such as BECN1 and ATG5, predisposes mice to diverse cancers, including lung adenocarcinoma, B-cell lymphoma, and hepatocellular carcinoma,^206,207 highlighting the essential role of autophagy in maintaining cellular integrity. Failure of autophagy in this context promotes the accumulation of oxidative stress, aberrant signaling, mitochondrial damage, and chronic inflammation, collectively establishing a pro-tumorigenic cellular environment.

Mitophagy contributes to this tumor-suppressive function by maintaining mitochondrial quality control (MQC). Dysfunctional mitochondria generate excessive ROS, mtDNA instability, and oncogenic signaling. Mutations in mitochondrial DNA (mtDNA), commonly detected across cancers such as renal oncocytoma, gastric adenocarcinoma, renal cell carcinoma, bladder cancer, ovarian cancer, thyroid cancer, and pancreatic cancer,^208–215 can disrupt OXPHOS, enhance ROS production, and promote tumorigenesis. Some mutations accelerate intestinal tumor formation,²¹⁰ while MT-CYTB and MT-CO1 variants enhance tumor growth through ROS-induced NF-κB2 activation.^211–213 Interestingly, selective pressure against truncating mutations in complex V suggests that mitochondrial function remains necessary for survival.²¹⁶ Thus, both autophagy and mitophagy restrain early tumorigenic events by eliminating damaged mitochondria, suppressing oxidative stress, and stabilizing metabolic equilibrium.

Once tumors are established, the role of autophagy frequently shifts toward tumor promotion. Autophagy provides metabolic substrates through recycling pathways, helping cancer cells survive under nutrient deprivation, hypoxia, immune stress, and therapeutic challenge.^217,218 It supports high biosynthetic and energetic demands of rapidly proliferating cells and may facilitate dormancy of residual tumor cells following chemotherapy or radiation, contributing to recurrence and metastasis.²¹⁹ Hypoxia, a defining feature of solid tumors, potently induces autophagy through both HIF-dependent and HIF-independent mechanisms.²²⁰ HIF-1 interacts with BCL2, BNIP3, BNIP3L/NIX, Beclin-1, ATG proteins, and PIK3C3 to stimulate autophagy,^221,222 while AMPK and PTEN suppress mTORC1, activating ULK1 and enhancing autophagosome-lysosome fusion.²²³ Under hypoxic pressure, autophagy orchestrates selective degradation of cellular components, including mitochondria (mitophagy), ER (ER-phagy), lipid droplets (lipophagy), ribosomes (ribophagy), nuclei (nucleophagy), and peroxisomes (pexophagy),^224–227 thereby sustaining tumor metabolism and viability.

Mitophagy is a crucial contributor to this adaptive response. The PINK1-Parkin pathway, frequently altered in cancers such as lung, breast, and glioma,²²⁸ governs mitochondrial turnover, iron homeostasis, and metabolic flexibility. Parkin deficiency promotes spontaneous hepatocellular carcinoma by enhancing hepatocyte proliferation and apoptosis resistance.²²⁹ The pathway controls the mitochondrial iron importers SLC25A37 and SLC25A28²³⁰; pathway failure drives iron accumulation, HIF1α stabilization, inflammasome activation, and tumor progression.²³¹ Loss of PINK1-Parkin exacerbates KRAS-driven pancreatic tumorigenesis,²³⁰ whereas PINK1 acts as a tumor suppressor in glioblastoma by reducing ROS and HIF activation.²³² Conversely, in lung and esophageal cancers, PINK1-driven mitophagy promotes proliferation and chemoresistance. These observations illustrate cancer type- and context-dependent consequences of mitophagy modulation.

Mitochondrial dynamics further intersect with mitophagy regulation. Enhanced mitochondrial fission, frequently observed in tumors, supports glycolysis, invasiveness, and resistance to apoptosis,²³³ while impaired fusion accentuates mitochondrial and metabolic instability. Mitochondrial chaperonins such as TRAP-1 also shape mitochondrial proteostasis and tumor metabolism. TRAP-1 supports tumor growth by promoting glycolysis, inhibiting succinate dehydrogenase, and activating cytochrome-c oxidase,^234–236 although reduced TRAP-1 expression at later stages has been linked to enhanced motility and invasion, suggesting context-dependent roles.^237,238

Autophagy and mitophagy critically influence therapeutic resistance. In chronic myeloid leukemia, imatinib resistance is associated with autophagy activation in leukemic stem cells; combined autophagy inhibition enhances cell death and eradicates resistant populations.²³⁹ Elevated expression of ATG genes, including ATG4B, ATG4D, ATG5, and BECN1, underscores its mechanistic relevance,²⁴⁰ and ATG4B knockdown sensitizes cells to therapy. Similarly, in glioblastoma stem-like cells, autophagy inversely correlates with the response to temozolomide, and its inhibition improves treatment efficacy.²⁴¹ Parallel resistance mechanisms are increasingly attributed to mitophagy-mediated mitochondrial preservation and metabolic resilience, reinforcing the therapeutic relevance of targeting these pathways.

The UPS plays a significant role in the development of skin cancer, particularly through its regulation of apoptotic pathways and the TGF-β signaling cascade.²⁴³ TGF-β signaling has a dual role in cancer, acting as a tumor suppressor in early stages and promoting invasion and metastasis in advanced malignancies.²⁴⁴ The majority of TGF-β pathway components, including TβRI, TβRII, and SMAD proteins, are regulated by proteasomal degradation.²⁴⁵ E3 ligases from the NEDD4 family, such as NEDD4-2, SMURF1, SMURF2, and WWP1, are key regulators of this pathway.²⁴⁶ For example, SMAD2 ubiquitination by Itch/AIP4 promotes its phosphorylation by TβRI, facilitating signal transduction.²⁴⁷ Mutations in SMAD2/4 or their receptors, which are prone to ubiquitination, disrupt TGF-β signaling and contribute to tumorigenesis.²⁴⁸ These mutations are more prone to ubiquitination, leading to an imbalanced TGF-β pathway and thus to tumorigenesis. In melanoma, receptor-interacting protein 1 (RIP1) is often upregulated and promotes cancer cell proliferation by activating NF-κB.²⁴⁹ Ubiquitination of RIP1 reduces NF-κB activity and inhibits melanoma growth, underscoring the role of the UPS in regulating oncogenic signaling pathways.²⁵⁰ Environmental factors, such as chemicals or radiation, can also disrupt the UPS balance. Low-LET ionizing radiation, for instance, inhibits proteasomal activity and induces free-radical damage to UPS-associated molecules, increasing the risk of tumorigenesis.^251,252

Elevated proteasomal activity is frequently observed in colorectal cancer (CRC), where deregulation of specific 20S proteasome subunits has been linked to carcinogenesis.²⁵³ A pivotal event in CRC development is the accumulation of oncogenic β-catenin due to disruptions in its regulatory pathways.²⁵⁴ β-catenin degradation via the UPS occurs through either the APC-dependent p53/Siah1 pathway or the APC-independent retinoid X receptor α (RXRα) pathway.²⁵⁵ Siah1, a critical link between p53 and Wnt/β-catenin signaling, regulates β-catenin turnover independently of GSK3β, and its dysfunction contributes to CRC progression.²⁵⁵

The UPS tightly controls cell cycle regulators, such as cyclins and cyclin-dependent kinase (CDK) inhibitors, whose dysregulation is common in cancer. Cyclin E, for example, is degraded by the SCF ubiquitin ligase complex, which uses the F-box protein Fbw7 to recognize and ubiquitinate cyclin E.²⁵⁶ Mutations in Fbw7, observed in breast and ovarian cancers, lead to cyclin E accumulation, premature S-phase entry, chromosomal instability, and tumor formation.^257,258 Similarly, p27Kip-1, a CDK inhibitor, is regulated by ubiquitin-mediated degradation. Reduced p27Kip-1 levels, often due to overexpression of the F-box protein Skp2, are associated with tumor progression and poor prognosis in various cancers.²⁵⁹

Cancer cells exhibit a metabolic shift from OXPHOS to glycolysis, known as the Warburg effect. This reprogramming is driven by alterations in glycolytic enzymes regulated by the UPS. Hexokinase 2 (HK2), which catalyzes the first step of glycolysis, is stabilized by the deubiquitinase CSN5 in HCC, promoting metastasis.²⁶⁰ Conversely, TRAF6-mediated ubiquitination of HK2 facilitates its degradation via autophagy, reducing glycolytic activity.²⁶¹ Pyruvate kinase M2 (PKM2), another key glycolytic enzyme, is stabilized by the deubiquitinase USP20, making USP20 a potential therapeutic target.²⁶²

The UPS also plays a critical role in cancer metastasis and treatment resistance. The ubiquitin-conjugating enzyme E2N (UBE2N) is overexpressed in various cancers and promotes breast cancer metastasis to the lung by activating TGFβ-mediated signaling pathways.²⁶³ In castration-resistant prostate cancer, the UPS supports androgen receptor (AR) function by facilitating its nuclear localization and stabilizing mutated AR proteins, enabling tumor cells to survive and proliferate under low androgen levels. Additionally, the UPS contributes to treatment resistance by degrading apoptosis-promoting factors and enhancing pro-survival signaling pathways, such as NF-κB and PI3K/AKT.²⁶⁴

The UPR is frequently upregulated in cancer, reflecting its central role in enabling tumor cells to cope with sustained proteotoxic stress. Rapid proliferation, high protein synthesis rates, hypoxia, nutrient deprivation, acidosis, and oxidative stress collectively impose a severe folding burden on the ER. By enhancing ER protein-folding capacity, reducing the accumulation of misfolded proteins, and activating adaptive signaling programs, the UPR allows cancer cells to survive and proliferate in hostile microenvironments that would otherwise compromise cellular viability. This adaptive capacity also contributes to resistance against a wide range of anticancer therapies, many of which further exacerbate ER stress. Importantly, the UPR does not function as a unidirectional survival pathway. When ER stress is excessive or prolonged beyond the adaptive threshold, UPR signaling can shift from cytoprotective to pro-apoptotic outcomes, frequently through crosstalk with mitochondrial death pathways. This critical balance between survival and cell death is orchestrated by the coordinated activity of the three canonical UPR branches, IRE1, PERK, and ATF6, which together determine cell fate decisions under stress. Despite extensive investigation, the precise molecular determinants governing the transition from adaptive UPR signaling to apoptosis remain incompletely understood. This knowledge gap represents a major challenge for therapeutic strategies aimed at selectively targeting UPR signaling in cancer cells without inducing unacceptable toxicity in normal tissues.^265,266

Persistent activation of the cytoprotective arm of the UPR can actively drive tumorigenesis and malignant progression. Among the UPR branches, the PERK-eIF2α-ATF4 signaling axis has emerged as a particularly influential regulator of cancer cell adaptation to stress. PERK activation enables tumor cells to transiently attenuate global protein synthesis while selectively enhancing the translation of stress-responsive genes that promote survival, invasion, and metabolic rewiring. One of the most prominent tumor-promoting consequences of PERK activation is its ability to stimulate angiogenesis. The PERK-ATF4 pathway induces the expression of multiple vascular endothelial growth factors, including VEGFA, VEGFB, VEGFC, VEGFD, VEGFF, and placental growth factor, which collectively support endothelial cell survival, neovascularization, and sustained tumor growth.²⁶⁷ In parallel, PERK-ATF4 signaling is tightly linked to EMT, a transcriptional reprogramming process that enables epithelial cancer cells to acquire migratory, invasive, and stem-like properties. Through EMT induction, PERK signaling directly contributes to metastatic dissemination and colonization of distant organs. Thus, sustained PERK pathway activation integrates proteostasis control with angiogenic and invasive programs, positioning this UPR branch as a central driver of tumor progression rather than a mere stress-response mechanism.

Beyond its role in active tumor growth, the UPR plays a pivotal role in cancer cell dormancy, long-term survival, and resistance to anticancer therapies. Under therapeutic pressure, cancer cells frequently exploit UPR signaling to enter a quiescent or dormant state characterized by reduced proliferation and heightened stress tolerance. This adaptive dormancy allows tumor cells to evade cytotoxic treatments and persist as minimal residual disease until conditions favor reactivation and relapse. Within this context, PERK signaling again emerges as a key determinant of therapy resistance. Sustained activation of the PERK-eIF2α-ATF4 pathway has been strongly associated with EMT induction, survival under chemotherapeutic stress, and metastatic competence.²⁶⁸ Genetic studies further highlight the complexity of PERK signaling in cancer: deletion of PERK reduces Neu-dependent mammary tumor development and lung metastasis, underscoring its role in tumor progression. However, prolonged or complete PERK inactivation can paradoxically increase genomic instability, thereby predisposing cells to tumorigenesis under certain conditions.²⁶⁹ These findings illustrate that PERK signaling exerts dose-, duration-, and context-dependent effects, complicating efforts to therapeutically manipulate this pathway.

In parallel to PERK, the IRE1/XBP1 branch of the UPR plays a critical role in sustaining tumor growth and mediating resistance to therapy. Upon activation, IRE1 catalyzes the unconventional splicing of XBP1 mRNA, generating the transcriptionally active XBP1 spliced isoform (XBP1s). XBP1s drives the expression of genes involved in ER protein folding, ER-associated degradation (ERAD), lipid biosynthesis, and stress adaptation, thereby enhancing tumor cell fitness under proteotoxic conditions. In estrogen receptor-positive breast cancer, XBP1s directly regulates the transcription of nuclear receptor coactivator 3 (NCOA3), linking UPR signaling to hormone receptor activity and therapy resistance.²⁷⁰ In pancreatic adenocarcinoma, XBP1s promotes angiogenesis through VEGF-independent mechanisms, further underscoring its capacity to sustain tumor vascularization and growth.²⁷¹ Moreover, overexpression of XBP1s pushes plasma cells toward a premalignant state, while genetic or functional XBP1 deficiency suppresses tumor growth and angiogenesis.²⁷² Collectively, these observations establish the IRE1/XBP1 pathway as a central regulator of tumor progression, metabolic adaptation, and chemoresistance.

Taken together, these elaborate mechanisms highlight the UPR not as a uniform survival pathway but as a dynamic, branch-specific signaling network whose outputs depend on tumor type, microenvironment, and stress duration. The intertwined roles of PERK and IRE1 in angiogenesis, EMT, dormancy, and therapy resistance emphasize why indiscriminate UPR inhibition is unlikely to succeed clinically and why pathway-selective, context-aware strategies are needed.

HSP70 and HSP90 interact with a wide spectrum of disease-associated proteins implicated in NDDs, including amyloid-β (Aβ), tau, α-synuclein, and TAR DNA-binding protein 43 (TDP-43). Their ability to dynamically regulate the folding, stabilization, and degradation of these proteins places them at the center of neuronal protein homeostasis. Depending on client protein conformation and cellular context, HSP70 and HSP90 can either promote refolding into nontoxic states or facilitate clearance through proteasomal or autophagic pathways. In AD, HSP70 and HSP90 directly interact with amyloid-β, preventing its aggregation and promoting its clearance. By stabilizing Aβ in a soluble, non-aggregated form, these chaperones limit the formation of toxic oligomers and extracellular plaques, which are hallmark pathological features of AD.³¹⁸ This chaperone-mediated buffering of early aggregation events is particularly important, as soluble Aβ oligomers are now recognized as major drivers of synaptic dysfunction and neurotoxicity.

Tau pathology represents another central target of HSP70 and HSP90 activity. In AD and other tauopathies, these chaperones modulate tau homeostasis by preventing aggregation of hyperphosphorylated tau species and promoting their degradation. Through these mechanisms, HSP70 and HSP90 reduce the formation of neurofibrillary tangles, thereby mitigating one of the core pathological processes underlying tau-mediated neurodegeneration.³¹⁹

In PD, HSP70 and HSP90 interact with α-synuclein, inhibiting its misfolding and aggregation into Lewy bodies. By promoting α-synuclein clearance and preventing the accumulation of toxic oligomeric species, these chaperones reduce α-synuclein-induced neuronal toxicity and cell death.³²⁰ Similar protective roles are observed in amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD), where HSP70 and HSP90 suppress the aggregation of TDP-43, a protein that forms pathological cytoplasmic inclusions in affected neurons. By stabilizing TDP-43 and facilitating its degradation, these chaperones attenuate its toxic gain-of-function effects.^319,321

Collectively, the ability of HSP70 and HSP90 to engage multiple aggregation-prone client proteins across distinct NDDs underscores their central role in neuronal proteostasis and highlights their broad therapeutic potential.

In contrast to HSP70 and HSP90, small heat shock proteins do not actively refold misfolded proteins. Instead, they act as ATP-independent “holdases” that bind misfolded or partially unfolded proteins, preventing their aberrant aggregation and facilitating subsequent degradation. This function is particularly relevant in neurons, where sHSPs frequently colocalize with disease-associated protein aggregates and serve as an early line of defense against proteotoxic stress. αB-Crystallin (CryAB) is among the most extensively studied sHSPs in NDDs. In AD, CryAB is upregulated in the brain and colocalizes with amyloid-β within extracellular senile plaques. Functionally, CryAB counteracts the initial seeding of Aβ oligomers and reduces their neurotoxicity.^322,323 CryAB also associates with tau fibrils and shows an inverse correlation with granular tau species, suggesting a protective role in limiting tau aggregation and toxicity in tauopathies.³²⁴

HSP27 is another prominent sHSP implicated in multiple NDDs. In PD, HSP27 protects against α-synuclein-induced toxicity by inhibiting its aggregation and reducing oxidative stress, a major contributor to dopaminergic neuronal loss.^325,326 In HD, HSP27 exerts neuroprotective effects by suppressing polyglutamine toxicity and lowering ROS levels, thereby slowing neurodegenerative progression.³²⁷

Additional sHSP family members, including HSBP6, HSBP7, HSBP8, and HSBP9, play important roles in mitigating polyQ-mediated toxicity in HD. Notably, HSBP7 is particularly effective against long polyQ repeats (119Q), whereas HSBP6, HSBP8, and HSBP9 preferentially counteract shorter polyQ expansions (43Q), indicating substrate-length specificity within the sHSP family.³²⁸ In ALS, HSBP8 protects against aggregation of TDP-43 and its cleavage products (TDP-25 and TDP-35) by promoting their removal via autophagy, further illustrating the tight functional coupling between sHSPs and degradative pathways.³²⁹

The ability of sHSPs to engage a wide range of aggregation-prone proteins, coupled with their rapid upregulation in response to cellular stress, positions them as critical first responders in neurodegenerative proteostasis. Together with HSP70 and HSP90, sHSPs form an integrated chaperone network that buffers neurons against proteotoxic stress, delays aggregate formation, and promotes clearance of damaged proteins. Understanding how these chaperone systems are coordinated and how their protective capacity declines with age and disease will be essential for developing effective proteostasis-based therapies for NDDs.

Autophagy genes play critical roles in maintaining neuronal proteostasis by degrading ubiquitinated and aggregated proteins. Disruption of autophagic clearance leads to the accumulation of toxic protein aggregates and neuronal degeneration. The autophagy adaptor p62/SQSTM1, essential for targeting ubiquitinated substrates to autophagosomes, is mutated in ALS and FTD, where impaired p62 function results in aggregation-prone protein accumulation and neuronal toxicity.^331,332 Multiple PD-associated genes, such as LRRK2, SYNJ1, DNAJC6, and DNAJC13, regulate synaptic endocytosis, autophagy, lysosomal function, and endolysosomal trafficking, underscoring the intertwined regulation of vesicular dynamics and proteostasis in dopaminergic neurons.

PINK1 and Parkin, best known for their role in mitophagy, are also central to maintaining neuronal quality control (Fig. 5). Mutations that impair their function promote mitochondrial damage, oxidative stress, and neuronal death, advancing PD pathology. Excessive α-synuclein aggregation in PD further exacerbates autophagy dysfunction while contributing to neuroinflammation. Misfolded α-synuclein can be released extracellularly, activating microglia and spreading pathogenic seeds or DAMPs to neighboring neurons, amplifying degeneration.³³³ Similarly, aggregation-prone proteins, including mutant huntingtin (Htt) and Ataxin-3, disrupt the autophagy machinery in HD and spinocerebellar ataxias, linking impaired autophagy directly to disease progression.

Lysosomal dysfunction is a prominent pathological hallmark of NDDs. Loss-of-function mutations in lysosomal proteins, as seen in Niemann-Pick disease type C1 and Gaucher disease, disrupt lipid and cholesterol metabolism and result in profound neurodegeneration. Importantly, several NDD-associated genes, including GBA, GRN, and ATP13A2, are shared with lysosomal storage disorders, reinforcing lysosomal integrity as a critical determinant of neuronal health.³³⁴ Beyond degradation failure, autophagy-lysosomal dysfunction contributes to the pathological propagation of aggregation-prone proteins. Tau seeds taken up through endolysosomal pathways can trigger lysosomal rupture, releasing aggregates into the cytosol and facilitating their spread.³³⁵ Similar mechanisms underlie α-synuclein propagation in PD, whereby lysosomal impairment enhances seeding, transmission, and progressive network degeneration.³³⁶ Thus, autophagy and lysosomal dysfunction not only initiate neuronal stress but also actively drive disease progression and neuroanatomical spread.

Mitochondrial dysfunction is a defining molecular hallmark across NDDs. The mitochondrial genome is especially susceptible to oxidative damage due to proximity to ROS generated during respiration. The accumulation of mtDNA mutations with aging and disease jeopardizes respiratory capacity, leading to bioenergetic failure, oxidative stress, and neuronal vulnerability.^337,338 Structural mitochondrial abnormalities such as disrupted cristae are widely documented in FTLD-TDP, AD, mixed AD/PD, and HD, linking altered mitochondrial architecture to impaired ATP production, calcium dysregulation, synaptic dysfunction, and eventual neuronal death.^339–343 Age-related mitochondrial decline further amplifies these defects, positioning mitochondrial deterioration as both a cause and consequence of neurodegeneration.

Mitophagy serves as the critical defensive mechanism preventing these mitochondrial insults from overwhelming neurons. Impaired mitophagy in AD contributes to the accumulation of dysfunctional mitochondria and exacerbates Aβ and hyperphosphorylated tau pathology. Activation of the PINK1/Parkin pathway restores mitophagy, reduces pathological protein burden, reverses cognitive decline, and improves synaptic function in experimental AD models.³⁴⁴ Key regulatory proteins such as VDAC1 and DRP1 influence this process; targeted reduction of these proteins enhances mitophagy and improves neuronal outcomes.³⁴⁵ In PD, defective PINK1/Parkin-dependent mitophagy leads to the accumulation of damaged mitochondria, neuronal stress, and dopaminergic neurodegeneration. Mutations in GBA and LRRK2 further impair mitophagy signaling, compounding mitochondrial dysfunction and accelerating disease progression.^346,347 In HD, mutant huntingtin interferes with lysosomal degradation of damaged mitochondria, disrupting GAPDH-mediated mitophagy and resulting in mitochondrial accumulation and neuronal death; enhancing mitophagy restores mitochondrial function and attenuates neurodegeneration.³⁴⁸

The indispensable role of the UPS in neuronal health is most clearly illustrated by familial forms of NDDs caused directly by mutations in UPS constituents. These genetic disorders demonstrate that intact ubiquitin-dependent protein degradation is not merely supportive but fundamentally required for neuronal integrity.

Mutations in UBQLN2, which encodes a ubiquitin-like protein involved in shuttling ubiquitinated substrates to the proteasome, are linked to ALS and FTD. UBQLN2 dysfunction compromises substrate delivery to the proteasome, resulting in the accumulation of toxic protein aggregates and progressive neurodegeneration.³⁴⁹ Similarly, mutations in valosin-containing protein (VCP), a multifunctional AAA-ATPase essential for extracting ubiquitinated proteins from cellular structures and directing them toward degradation, cause multisystem proteinopathies that include ALS, FTD, and inclusion body myopathy. VCP dysfunction disrupts ubiquitin-dependent protein turnover, leading to widespread proteostasis failure and aggregate accumulation.^350,351

Loss-of-function mutations in Parkin (PARK2), an E3 ubiquitin ligase, underlie autosomal recessive Parkinson’s disease. Parkin plays a dual role in ubiquitinating misfolded proteins and targeting damaged mitochondria for degradation. Its inactivation impairs MQC, promotes the accumulation of dysfunctional mitochondria, and accelerates dopaminergic neuronal loss.³⁵²

Mutations in UCHL1, which encodes a deubiquitinating enzyme critical for ubiquitin recycling and proteasome efficiency, are associated with rare progressive neurodegenerative syndromes. UCHL1 dysfunction reduces the availability of free ubiquitin and compromises proteasomal activity, further destabilizing neuronal proteostasis.³⁵³

Collectively, these genetic disorders underscore the non-redundant role of the UPS in preventing protein aggregation and preserving neuronal viability. Importantly, they also illustrate how defects at distinct levels of the UPS, substrate recognition, ubiquitin recycling or proteasome targeting, can converge on a common outcome of proteotoxic stress and neurodegeneration.

In addition to genetic mutations, UPS dysfunction is also a defining feature of sporadic NDDs, where aggregation-prone proteins directly impair proteasomal function. Proteins such as tau, TDP-43, α-synuclein, and polyglutamine-expanded proteins misfold and aggregate into toxic species that overwhelm the UPS and interfere with normal degradation processes.

In AD and related tauopathies, hyperphosphorylated tau assembles into neurofibrillary tangles that impair proteasome activity and disrupt cellular homeostasis.³⁵⁴ These aggregates not only evade degradation but also obstruct the processing of other UPS substrates. In ALS and FTD, cytoplasmic aggregates of TDP-43 sequester UPS components, thereby preventing the clearance of additional misfolded proteins.³⁵⁵ Moreover, TDP-43 aggregates can translocate into mitochondria, triggering mitochondrial DNA release into the cytoplasm and activating inflammatory signaling through the cGAS-STING pathway.^333,356 This links UPS dysfunction to innate immune activation and neuroinflammation.

In PD and other synucleinopathies, aggregated α-synuclein impairs proteasomal activity and disrupts mitochondrial function, further exacerbating neuronal stress.³⁵⁷ Similarly, in HD and related polyQ disorders, expanded polyQ-containing proteins form insoluble aggregates that physically inhibit proteasome function, amplifying the proteotoxic burden.³⁵⁸

Together, these findings reveal a feedforward loop in which misfolded proteins evade degradation while simultaneously crippling the UPS, accelerating neuronal dysfunction and degeneration. This mechanism provides a unifying explanation for UPS impairment in both familial and sporadic NDDs.

In addition to canonical K48-linked ubiquitination that targets proteins for proteasomal degradation, alternative ubiquitin chain architectures and ubiquitin-like modifications play critical roles in neuronal signaling and survival (Fig. 2c). Dysregulation of these non-canonical pathways contributes substantially to NDD pathogenesis.

K63-linked ubiquitination, which typically regulates signaling, DNA repair, and autophagy rather than degradation, is frequently altered in PD, AD, and ALS. Aberrant K63-linked ubiquitination disrupts intracellular signaling networks and autophagic flux, contributing to the accumulation of toxic protein species and impaired stress responses.³⁵⁹

Linear ubiquitination, mediated by the linear ubiquitin chain assembly complex (LUBAC), regulates NF-κB signaling and inflammatory pathways. Dysregulation of LUBAC activity has been linked to NDDs, where excessive or sustained inflammatory signaling exacerbates neuronal injury and cell death.³⁶⁰

SUMOylation, a ubiquitin-like modification, further expands the regulatory complexity of neuronal proteostasis. SUMOylation modulates protein stability, subcellular localization, transcriptional activity, synaptic plasticity, and neuronal excitability. Disruption of SUMOylation has been implicated in AD, HD and PD, where it contributes to protein aggregation, synaptic dysfunction, and neuronal vulnerability.³⁶¹

These non-canonical ubiquitin-dependent modifications emphasize that UPS dysfunction in NDDs extends beyond proteasomal degradation alone. Instead, it reflects a broader collapse of ubiquitin-centered regulatory networks that integrate proteostasis, signaling, inflammation, and neuronal function.

Patients with SLE exhibit elevated levels of IgM anti-65 kDa HSP antibodies compared to healthy controls,⁴²¹ and a significant subset overexpress HSP90.⁴²² HSP60, which is present on the surface of endothelial cells (ECs), serves as a target for anti-HSP60 antibodies in SLE sera. These antibodies bind to ECs and induce apoptosis, contributing to tissue damage (Fig. 6).⁴²³ In RA, HSPs are implicated in driving inflammation and shaping the autoimmune environment through their immunogenic properties. Patients with RA exhibit increased levels of IgM anti-65 kDa and IgG anti-70 kDa HSP antibodies compared to healthy controls.⁴²¹ Synovial tissues from RA patients show significant upregulation of HSP27 mRNA,⁴²⁴ and elevated levels of pro-inflammatory IL-6 are associated with increased IgG autoantibodies against HSP40, making HSP40 a potential marker of disease activity and progression.^425,426 Additionally, IgG, IgM, and IgA autoantibodies against HSP60, HSP70, and HSP90 are significantly elevated in RA patients.⁴²⁷

In MS, HSPs contribute to the regulation of neuroinflammation and are linked to autoantigenicity and immune dysregulation in the central nervous system (CNS).^428,429 HSPs are induced in response to various CNS pathologies, including neurodegenerative diseases, ischemia, and epilepsy, where they play a neuroprotective role by preventing protein aggregation and misfolding.^430,431 However, in MS, HSPs can exacerbate the immune response by acting as pro-inflammatory cytokines and adjuvants for myelin peptides.⁴³² Patients with MS exhibit elevated levels of IgG antibodies targeting HSP70 family members, although autoantibodies against HSP27, HSP60, and HSP90 are not typically detected.^429,433 HSP27 is strongly implicated in MS pathogenesis as the most immunodominant myelin T-cell antigen, with up to 20-fold increased expression in MS lesions (Fig. 6).⁴³⁴ While HSP27 is induced in glial cells by oxidative stress and acts as an intracellular signaling factor, its upregulation is generally protective. HSP27-deficient mice develop more severe symptoms, including increased inflammation, demyelination, and immune cell infiltration, which are ameliorated by exogenous HSP27 treatment.⁴³⁵

Inflammasomes are multiprotein complexes that play a central role in innate immunity by regulating the production of pro-inflammatory cytokines, such as interleukin-1β (IL-1β) and IL-18. HSPs are essential for the proper folding, stabilization, and assembly of key inflammasome components, including NLRP3, ASC, and caspase-1. By maintaining protein homeostasis under stress conditions, HSPs directly influence the activation and function of inflammasomes. Dysregulation of HSP activity can lead to aberrant inflammasome activation, resulting in the excessive production of pro-inflammatory cytokines and contributing to the pathogenesis of autoimmune diseases such as SLE, RA, and MS.^436–438

A key feature of autoimmune diseases is the presence of autoantibodies that target self-proteins, including HSPs. Anti-HSP autoantibodies are elevated in patients with a wide range of inflammatory and autoimmune conditions, including RA,⁴²⁷ juvenile idiopathic arthritis,⁴³⁹ psoriasis,⁴⁴⁰ SLE,⁴⁴¹ atopic dermatitis,⁴⁴² and celiac disease.⁴⁴³ These autoantibodies can amplify inflammatory processes, disrupt normal protein homeostasis, and exacerbate tissue damage, playing a critical role in disease pathogenesis. Interestingly, anti-HSP autoantibodies are also found in healthy individuals, likely as a result of prior bacterial infections or tissue degradation.⁴⁴⁴ Mutations in HSP genes can further contribute to immune dysregulation. For example, a variant of the HSPA1L gene has been linked to susceptibility to vitiligo, an autoimmune condition characterized by the destruction of melanocytes.⁴⁴⁵ Additionally, intracellular HSP90, which is involved in the activation of both innate and adaptive immune responses, has been implicated in autoimmune processes. HSP90 inhibitors have shown promise in preclinical models of autoimmune diseases such as arthritis and SLE by suppressing inflammatory signaling pathways.⁴⁴⁶ These inhibitors, such as geldanamycin, bind to the N-terminal nucleotide-binding pocket of HSP90 with higher affinity than ATP, leading to the proteasomal degradation of HSP90-dependent client proteins. This results in the inactivation of NF-κB-dependent inflammatory factors such as TNF-α and IL-6, as well as the blockade of cell signaling molecules such as MAP kinases.⁴⁴⁶ In contrast, upregulation of HSP70 inhibits NF-κB activity, reducing inflammation and highlighting the complex roles of HSPs in immune regulation.

Autophagy plays a dual role in autoimmunity by regulating immune cell function, antigen processing, and inflammatory responses. Defective autophagic degradation promotes accumulation of intracellular autoantigens and enhances presentation of self-derived peptides, contributing to loss of self-tolerance. Autophagy also normally constrains inflammasome activation; failure of this regulation allows excessive IL-1β and IL-18 production, amplifying systemic inflammation, which is particularly relevant in SLE and RA.

In SLE, dysregulated autophagy contributes to immune hyperactivation, defective apoptotic cell clearance, and increased autoantigen availability. Genetic studies strongly support the involvement of autophagy machinery in SLE susceptibility, with GWAS identifying risk polymorphisms in ATG5, ATG7, and MTMR3.^447–450 Upregulated autophagy in macrophages enhances the production of pro-inflammatory cytokines such as IL-6 and TNF-α, while autophagy deficiency promotes exaggerated inflammasome activation, as demonstrated by ATG16L1 ablation increasing IL-1β and IL-18 secretion.^451,452 Together, these abnormalities destabilize immune tolerance and promote lupus nephritis, immune complex deposition, and tissue injury.

In RA, autophagy supports many of the pathological hallmarks of disease progression. Upregulation of autophagy-related proteins (including BCL-1, ATG7, ATG5, and LC3-II) in synovial tissues contributes to synovial fibroblast proliferation, resistance to apoptosis, and chronic inflammatory activation.^453–458 Autophagy also enhances osteoclastogenesis through an overactive autophagy-lysosomal system, facilitating cartilage erosion and bone destruction.⁴⁵⁵ Experimental inhibition of autophagy alleviates synovial inflammation and promotes apoptosis through PI3K/AKT modulation, underscoring autophagy’s pathogenic role in RA.

In MS, autophagy plays a uniquely context-dependent role. It can promote neuroprotection by limiting inflammation and oxidative stress, yet excessive or dysregulated autophagy also enhances autoreactive immune responses and demyelination. Differential expression of autophagy genes in MS patients, downregulation of ATG9A and BECN1 and upregulation of ULK1, ULK2, and ATG5 reflect complex autophagy modulation.^459,460 Reduced LC3 and BCL-1 in spinal cord tissues, along with evidence that mTORC1 inhibition ameliorates disease, supports a pathogenic dimension of autophagy in MS.⁴⁶¹ Conversely, interventions such as caloric restriction enhance autophagy, improve remyelination, and reduce disease severity in relapsing-remitting MS, emphasizing its therapeutic potential when properly modulated.⁴⁶²

Mitochondria are essential regulators of immune signaling, apoptosis, metabolism, and stress responses. In autoimmune diseases, mitochondrial dysfunction and impaired mitophagy fuel chronic inflammation by releasing ROS and mtDNA, which activate innate immune pathways and promote autoreactive immune responses. Damaged mitochondria that escape degradation act as persistent inflammatory stimuli, linking mitochondrial instability directly to autoimmune disease mechanisms. In SLE, mitochondrial dysfunction contributes to oxidative stress, altered apoptosis, mtDNA mutations, and impaired immune complex clearance, aggravating the loss of self-tolerance.³⁰ SLE T cells display increased mitochondrial mass, reflecting enhanced biogenesis but insufficient mitophagy, alongside elevated IFN-α signaling that further suppresses mitophagic clearance and primes dendritic cell activation.^463,464 mtDNA damage and immunogenic modifications promote autoantibody production and lupus nephritis, while genetic associations such as mtDNA haplotype N and D-loop heterogeneity support a mitochondrial genetic risk component.^465–468

In RA, mitochondrial dysfunction drives chronic synovial inflammation despite preserved mitochondrial mass. RA T cells exhibit reduced respiratory capacity and abnormal mitochondrial dynamics influenced by microRNAs such as miR-125b, which regulates the mitochondrial fission protein MTP18.^469–471 Hypoxic synovial environments enhance mitochondrial activity in osteoclasts, exacerbating bone erosion.⁴⁷² Elevated extracellular mtDNA in synovial fluid and plasma correlates with autoantibody presence and inflammation, while mtDNA mutations in synovial tissues are associated with increased TNF-α and IFNγ signaling, establishing a self-reinforcing inflammatory cycle.^473–475

In MS, mitochondrial injury contributes to both neuroinflammation and neurodegeneration. The CNS’s high metabolic demand renders it particularly vulnerable to oxidative injury.⁴⁷⁶ Oligodendrocyte differentiation and myelination are impaired by mtDNA damage and oxidative stress, leading to altered transcription, double-strand breaks, and mitochondrial deletion events, especially involving complex IV.^477,478 Reduced N-acetyl aspartate levels reflect compromised mitochondrial integrity and correlate with poor myelin maintenance.⁴⁷⁹ Mitochondrial dysfunction disturbs T-cell homeostasis by activating cell death pathways involving Bcl2, OPA1, PHB2, ATG5, OMA1, and SIRT3, while reduced OXPHOS, membrane potential instability, and altered glycolysis further destabilize immune function.^480–483

In SLE, aberrant UPS activity contributes to inflammatory pathways, defective antigen presentation, and impaired apoptotic processes. SLE patients exhibit autoantibodies against the α and β subunits of the 20S proteasome and other UPS components, which can block proteasome activation by PA28 in vitro, suggesting a regulatory role for these autoantibodies in proteasome function (Fig. 6).⁴⁸⁶ Elevated levels of circulating proteasomes correlate with SLE disease activity, making them potential biomarkers for lupus.⁴⁸⁷ Neutrophil extracellular traps (NETs), which are implicated in SLE pathogenesis, contain lower levels of polyubiquitinated proteins in SLE patients. Anti-ubiquitinated myeloperoxidase antibodies in SLE sera enhance cytokine production by macrophages upon NET stimulation, suggesting a role for extracellular ubiquitin in lupus-related inflammation.⁴⁸⁸

Polymorphisms in TNFAIP3 (A20), a DUB that removes K63-linked ubiquitin chains and adds K48-linked chains to signaling intermediates, are associated with SLE. Targeting TNFAIP3 and UBE2L3 represents a potential therapeutic strategy for SLE.⁴⁸⁹ E3 ubiquitin ligases, such as STUB1 and FBXW7, are also implicated in SLE. STUB1 is upregulated in CD4+ T cells of SLE patients and promotes the expression of CD70 and CD11a, making it a potential therapeutic target.⁴⁹⁰ FBXW7, which catalyzes the degradation of the anti-apoptotic protein MCL1, plays a key role in myeloid cells and SLE progression. FBXW7 deficiency in mice reduces immune complex deposition and glomerulonephritis, highlighting its importance in SLE pathogenesis.⁴⁹¹

In RA, the UPS is involved in multiple interconnected pathways that drive chronic inflammation and joint damage. Polyubiquitination of signaling molecules such as TRAF2, TRAF6, RIP-1, and NEMO activates the TAK1 and IκB kinase (IKK) complexes, leading to NF-κB activation and pro-inflammatory cytokine production.⁴⁹² The E3 ligases TRAF2 and TRAF6 play a central role in this process, linking the UPS to RA pathogenesis. Oxidative stress in RA overloads the UPS, leading to the accumulation of ubiquitin conjugates and exacerbating cellular damage and inflammation. Proteasome inhibitors have shown promise in RA by reducing NF-κB activation, IL-6 expression, and angiogenesis, thereby alleviating arthritis symptoms.⁴⁹³

The E3 ligase cIAP2 promotes FLS survival and inflammatory responses in RA (Fig. 6), whose inhibition reduces cytokine secretion and FLS proliferation, making it a potential therapeutic target.⁴⁹⁴ Similarly, inhibition of the E3 ligase synoviolin suppresses synovial cell proliferation and reduces disease severity in RA models. Polymorphisms in TNFAIP3 and elevated levels of circulating proteasomes are also associated with RA, further implicating the UPS in disease pathogenesis.^487,495 Inhibitors targeting TRAF6-Ubc13 interactions, such as C25-140, ameliorate inflammation and improve symptoms in RA models by blocking NF-κB activation (Fig. 6).⁴⁹⁶ Meta-analyses have identified TRAF6 and RAG1 as true RA risk alleles, underscoring the importance of UPS components in RA.⁴⁹⁷

In MS, the UPS regulates T-cell responses and CNS inflammation. T-cell receptor (TCR) signaling and Th17 cell differentiation, which are critical in MS, are influenced by ubiquitination and deubiquitination processes.⁴⁹⁸ DUBs such as USP9X, USP15, and CYLD regulate TCR signaling, while USP4 and USP17 stabilize the transcription factor RORγt, a key driver of Th17 differentiation (Fig. 6).^499–505 The DUB OTUD5 (DUBA) promotes RORγt degradation by stabilizing the E3 ligase UBR5, highlighting the intricate regulation of Th17 cells by the UPS.⁵⁰⁶ Polymorphisms in CBL and CBL-B are associated with MS, with reduced CBL-B expression enhancing CD4+ T-cell proliferation in response to type I interferons.⁵⁰⁷ Loss-of-function mutations in CBL are also linked to autoreactive inflammatory disorders, further implicating the UPS in MS pathogenesis.⁵⁰⁸ The UPS regulates myeloid leukemia 1 (MCL-1), an anti-apoptotic protein upregulated in MS. Dysregulation of the UPS and interferon-beta (IFN-β) signaling contributes to MS development, making UPS components potential therapeutic targets.

Autoantibodies associated with SLE can exacerbate ER stress and activate the ER stress response. For instance, antibodies targeting the ER stress-related protein GRP78 have been detected in the blood of SLE patients, with increased GRP78 and XBP1 expression observed in antibody-secreting plasma cells. This suggests that ER stress drives B-cell differentiation into plasma cells, enhancing antibody production. Activation of the IRE1-XBP1 pathway is essential for ER expansion and antibody secretion in plasma cells, further implicating ER stress in SLE pathogenesis.⁵⁰⁹ Clinical evidence indicates that the expression of ER stress-related proteins such as IRE1, PERK, and CHOP is reduced in SLE patients, while the levels of XBP1 and midbrain astrocyte-derived neurotrophic factor (MANF) are elevated, suggesting their potential role in disease development.⁵¹⁰

In RA, hyperactivation of the ERAD system, known as hyper-ERAD, contributes to synoviocyte proliferation and mitigates severe ER stress by promoting an anti-apoptotic response. Synoviolin, an ER-resident E3 ubiquitin ligase involved in the ERAD pathway, is highly expressed in the synovium of RA patients and plays a key role in disease progression.⁵¹¹ It facilitates the ubiquitination and degradation of IRE1α in FLS from mice with collagen-induced arthritis (CIA), protecting synovial cells from ER stress-induced apoptosis. Elevated levels of ER stress markers, including GRP78, IRE1, XBP1s, ATF6, and eIF2α-P, have been observed in macrophages and synovial tissues of RA patients.⁴⁵⁸

Inflammatory cytokines and autoantibodies implicated in RA pathogenesis can activate ER stress pathways, which in turn drive inflammatory responses that contribute to disease progression.⁵¹² Studies on FLS and macrophages from RA synovial fluid have revealed heightened activation of the IRE1/XBP1 axis.⁴⁵⁸ In mice, deletion of IRE1α specifically in myeloid cells reduces cytokine production and protects against inflammatory arthritis, highlighting its role in disease progression.⁵¹³ IRE1α activation enhances TLR-mediated pro-inflammatory cytokine production in macrophages and neutrophils, with its regulated IRE1-dependent decay (RIDD) activity stabilizing cytokine mRNA and sustaining a robust inflammatory response in the RA synovium.⁵¹⁴

ER stress is widely observed in demyelinating lesions in MS, with elevated levels of markers such as CHOP, BiP, phosphorylated PERK, phosphorylated eIF2α, and XBP-1 in microglia, oligodendrocytes (OLs), T cells, and astrocytes.⁵¹⁵ The pro-apoptotic molecule CHOP underscores the connection between ER stress and cell fate in MS, with its expression in various cell types suggesting immune-mediated processes contributing to ER stress-associated damage. UPR activation is particularly prominent in active MS lesions, driven by increased levels of inflammatory cytokines, reactive nitrogen species (RNS), and ROS.

Immunohistochemical analysis of biopsy and postmortem samples revealed increased expression of UPR markers, including CHOP and BiP, in oligodendrocytes, astrocytes, T cells, and macrophages/microglia within MS lesions.⁵¹⁶ In experimental autoimmune encephalomyelitis (EAE) mice, CNS expression of IFNγ before disease onset prevents oligodendrocyte death, demyelination and axonal degeneration. Modest activation of the PERK-eIF2α pathway in oligodendrocytes of control EAE mice is significantly enhanced by IFNγ, highlighting its protective role in the disease process.⁵¹⁷ However, PERK heterozygous deficiency exacerbates IFNγ-induced apoptosis of remyelinating oligodendrocytes and impairs remyelination in cuprizone-induced demyelinated lesions.⁵¹⁸ These findings suggest that activation of the PERK-eIF2α pathway protects (re)myelinating oligodendrocytes from the harmful effects of IFNγ in immune-mediated demyelinating diseases.⁵¹⁹

High-resolution proteomic technologies, including mass spectrometry and affinity purification coupled with mass spectrometry (AP-MS), enable systematic identification of protein‒protein interaction networks and post-translational regulatory nodes. These studies have already revealed expanded roles for classical chaperone systems. For example, mapping the interactomes of HSP70 and HSP90 has clarified that these proteins not only ensure correct folding but also influence autophagic decision-making and UPS performance,^548,549 highlighting their integrative role in proteostasis governance. Likewise, p62/SQSTM1 interactome analyses have elucidated how it couples ubiquitinated substrates to the autophagic machinery, placing it at a crucial signaling intersection.⁵⁵³

Beyond interaction networks, spatial proteomics adds an essential layer of biological meaning. Cellular proteostasis is governed not only by molecular availability but also by spatial proximity and compartmentalization. Technologies such as BioID and APEX labeling are uncovering how subcellular localization influences proteostasis pathway coordination. The localization of FUNDC1 at ER-mitochondria contact sites (MAMs) and its role in integrating mitophagy, mitochondrial adaptation, and ER stress responses exemplify how spatial organization determines functional proteostasis outcomes.⁵⁵¹ Understanding these spatial dynamics will be indispensable for deciphering how cells coordinate proteostasis surveillance across organelles and how diseases disrupt these positional networks.

For instance, CRISPR screens in the PINK1-Parkin mitophagy pathway have identified novel cofactors and signaling partners, expanding our understanding of mitochondrial quality control and offering promising therapeutic entry points in disorders such as Parkinson’s disease.³³ Similarly, systematic screening under ER stress conditions continues to uncover new UPR modulators that could be pharmacologically leveraged in metabolic and inflammatory pathologies.⁵⁵⁶ These efforts illustrate how functional genomics is redefining the proteostasis landscape.

Importantly, the integration of CRISPR screening with high-content phenotyping modalities, including live-cell imaging, multiplex transcriptomics, and single-cell RNA sequencing, enables real-time visualization of how cells dynamically reorganize proteostasis resources. Such platforms have demonstrated how proteasome inhibition induces a functional handover toward autophagy-dependent degradation,⁹ reflecting intrinsic compensatory resilience. These tools therefore not only identify genetic regulators but also illuminate systems-level adaptation principles that can be therapeutically harnessed.

Graph neural networks and other advanced computational tools now enable simulations wherein perturbation of a single node, for example, HSP90 inhibition, can be computationally modeled to predict downstream shifts in UPS engagement or autophagy induction.⁵⁵⁹ These predictions inform rational design of combination therapies that strategically exploit compensatory relationships rather than acting blindly upon isolated mechanisms.

AI-based drug discovery is also reshaping therapeutic development. Machine learning-driven compound prediction platforms have successfully identified novel HSP90 inhibitors with clinical potential.¹² Meanwhile, computational design approaches are accelerating the development of PROTACs and AUTACs, enabling targeted degradation of disease-associated proteins through UPS or autophagy, respectively.^66,168 Such tools represent a paradigm shift toward precision proteostasis engineering.