Mechanisms and implications of podocyte autophagy in chronic kidney disease

Abstract

Autophagy is a protective mechanism through which cells degrade and recycle proteins and organelles to maintain cellular homeostasis and integrity. An accumulating body of evidence underscores the significant impact of dysregulated autophagy on podocyte injury in chronic kidney disease (CKD). In this review, we provide a comprehensive overview of the diverse types of autophagy and their regulation in cellular homeostasis, with a specific emphasis on podocytes. Furthermore, we discuss recent findings that focus on the functional role of different types of autophagy during podocyte injury in chronic kidney disease. The intricate interplay between different types of autophagy and podocyte health requires further research, which is critical for understanding the pathogenesis of CKD and developing targeted therapeutic interventions.

INTRODUCTION

Chronic kidney disease (CKD) is one of the leading causes of mortality worldwide, affecting more than 10% of the global population (1). CKD is characterized by distinct histological and morphological features, including glomerulosclerosis, tubulointerstitial fibrosis, and podocyte injury (2). Podocytes are terminally differentiated epithelial cells that play a crucial role in maintaining the glomerular filtration barrier. A decrease in podocyte count is sufficient to drive CKD progression, although the underlying mechanisms remain largely unknown. Although current therapeutic approaches effectively slow down the progression of CKD, they are insufficient to completely halt or reverse the disease (3, 4). Consequently, further studies are necessary to understand the contribution of podocyte injury to CKD development and progression and to identify novel therapeutic targets.

Autophagy, a term derived from the Greek words “auto,” meaning self, and “phagy,” meaning eating, is a fundamental catabolic process that plays a key role in regulating and maintaining cellular homeostasis (5). Recent research has shed light on the critical role of autophagy in maintaining podocyte integrity (6, 7). Autophagy is commonly associated with survival or cellular protection and is activated in response to various cellular stressors, such as starvation, hypoxia, oxidative stress, cellular damage, and pathogens (8). Dysregulated autophagy has been increasingly implicated in the development and progression of several diseases, including kidney diseases (9–11). Dysregulated or maladaptive autophagy can lead to the accumulation of damaged cellular components and toxic substances within kidney cells. This accumulation can impair normal cellular functions and contribute to the pathogenesis of kidney diseases (7, 12–14).

This review provides a comprehensive understanding of different types of autophagy and their molecular regulation, as well as their intricate role in podocytes under pathological conditions. Furthermore, this review offers new avenues for autophagy-relevant therapeutic strategies to prevent the progression of CKD.

REGULATION OF AUTOPHAGY

Autophagy is a cellular degradation process that is highly conserved in all eukaryotic cells. In mammalian cells, autophagy can be classified into macroautophagy, microautophagy, and chaperone-mediated autophagy (CMA). Macroautophagy, commonly known as “autophagy,” can either be nonselective or selective—the latter referring to the degradation of specific subcellular structures, such as of mitochondria by mitophagy, of lipid droplets (LDs) by lipophagy, of protein aggregates by aggrephagy, of endoplasmic reticulum (ER) by ER-phagy, and of invading microbes by xenophagy. Macroautophagy involves the sequestration of substantial portions of cytoplasm and various cellular components, including damaged organelles (i.e., mitochondria, peroxisomes), aggregate proteins, carbohydrates, lipids, nucleic acids, and intracellular pathogens, into a double-membraned vacuole termed autophagosome. This autophagosome subsequently fuses with lysosomes to form an autolysosome, where autolysosomal contents are degraded (10, 15, 16). Microautophagy is a nonselective lysosomal degradation process in which small portions of the cytoplasm are directly engulfed by the lysosomes through invagination of the lysosomal membrane (17, 18). Although the molecular regulation of microautophagy is comparatively less understood than that of macroautophagy, recent research efforts are increasingly aimed at unraveling its intricate mechanisms and signaling pathways (19). However, its specific role in the kidney remains largely unknown, necessitating further research to understand its functions within the context of renal pathology. CMA involves a process in which proteins targeted for degradation form complexes with the heat shock cognate protein 70 (HSC70) via KFERQ-pentapeptide motifs and are subsequently delivered across lysosomal membranes into the lysosomal lumen by interacting with lysosome-associated membrane protein 2A (LAMP2A) (20). In the following section, we will explore the molecular regulation of various types of autophagy.

Nonselective Macroautophagy

The most distinguishing characteristic of macroautophagy is the formation of the double membrane bound phagophore and autophagosome. Formation of the autophagosome requires the complex interaction of autophagy-related proteins (ATG) and proteins of other accessory pathways (21, 22). The initial complex responsible for regulating the induction of autophagosome formation is the ATG1/Unc-51-like kinase (ULK) complex, consisting of ATG1, ATG11, ATG13, ATG17, ATG29, and ATG31. After the assembly of the ATG1 complex at the phagophore assembly site (PAS), ATG9 and its cycling system (ATG2, ATG9, and ATG18) mediate membrane delivery to the expanding phagophore (23). The class III phosphatidylinositol 3-kinase (PI3K) complex (VPS34, VPS15, VPS30/ATG6, and ATG14) is involved in vesicle nucleation and in recruiting phosphatidylinositol 3-phosphate (PtdIns3P)-binding proteins to the PAS (23). The ubiquitin-like (Ubl) conjugation systems, ATG12 system (involving ATG5, ATG7, ATG10, ATG12, and ATG16) and the ATG8 system (involving ATG3, ATG4, ATG7, and ATG8) play a role in vesicle expansion (23). After the autophagosome forms, it fuses with lysosomes, resulting in the proteolytic degradation of the engulfed cargo (21). Several signaling pathways are involved in the regulation of nonselective macroautophagy, including 5′-adenosine monophosphate-activated protein kinase (AMPK), mechanistic target of rapamycin (mTOR), and sirtuin signaling pathways (Fig. 1).

Figure 1.

Regulation of macroautophagy signaling pathway. The 5′-AMP-activated protein kinase (AMPK), mechanistic target of rapamycin (mTOR), and sirtuin 1 (SIRT1) signaling pathways are considered master regulators of the autophagic process. AMPK promotes autophagy by directly phosphorylating and activating Unc-51 like autophagy activating kinase 1 (ULK1) or by inhibiting mechanistic target of rapamycin complex 1 (mTORC1). AMPK inhibits mTORC1 both directly by phosphorylation of rapamycin-sensitive adaptor protein of mTOR (Raptor) and indirectly by phosphorylation of the tuberin (TSC2). AMPK can also phosphorylate beclin-1, promoting its interaction with VPS34 and autophagy related proteins (ATG)14, and facilitating the formation of autophagosomes. mTORC1 inhibits autophagy by phosphorylating ULK1 and ATG13, thereby disrupting their interaction and inhibiting the formation of the autophagy-initiating complex. mTORC1 also inhibits autophagy by phosphorylating and sequestering transcription factor EB (TFEB) in the cytoplasm, preventing its nuclear translocation, which is necessary to promote the expression of autophagy and lysosomal biogenesis-related genes. On the other hand, mechanistic target of rapamycin complex 2 (mTORC2) inhibits autophagy indirectly by phosphorylating its downstream effector Akt. Akt can phosphorylate and inhibit components of the hamartin (TSC1)/TSC2 complex, leading to the activation of mTORC1, which in turn suppresses autophagy. Akt can directly phosphorylate beclin-1, enhancing its binding to intermediate filaments and inhibiting autophagy. SIRT1 deacetylates key autophagy-related proteins, including ATG5, ATG7, and ATG8/ microtubule-associated protein 1 light chain 3 (LC3), facilitating their involvement in autophagosome formation. In addition, SIRT1 activates forkhead box protein O1 (FoxO1) transcription factors, enhancing the expression of autophagy-related genes. CAMKK2, calcium/calmodulin-dependent protein kinase 2; Deptor, DEP domain TOR-binding protein; LKB1, liver kinase B1; mLST8, mammalian lethal with Sec-13 protein 8; mSin1, mammalian stress-activated protein kinase-interacting protein 1; PI3K, phosphatidylinositol 3-kinase; PIP2, phosphatidylinositol 4,5-bisphosphate; PIP3, phosphatidylinositol (3,4,5)-trisphosphate; PRAS 40, proline-rich Akt substrate 40 kDa; RHEB, Ras homolog enriched in brain; TAK1, transforming growth factor β-activated kinase 1.

5′-Adenosine monophosphate-activated protein kinase signaling pathway.

Macroautophagy is tightly regulated by complex signaling pathways, particularly in response to conditions of starvation or low energy levels (24). One of the key players in the metabolic regulation of macroautophagy is the AMPK signaling pathway. AMPK is activated in response to an increase in the AMP to ATP ratio, and therefore acts as a sensor for nutrient deprivation and low energy levels. Binding of AMP activates AMPK allosterically and induces phosphorylation of Thr172 within the activation domain of the α subunit by upstream kinases, such as serine/threonine protein kinase (STK11)/liver kinase B1 (LKB1), calcium/calmodulin dependent protein kinase 2 (CAMKK2), and mitogen-activated protein kinase 7 (MAP3 K7)/transforming growth factor β-activated kinase 1 (TAK1) (25–29). In response to nutrient deprivation, AMPK activates autophagy via the inhibition of the mTOR pathway or direct phosphorylation of ULK1. AMPK inhibits mTOR activation by phosphorylating tuberous sclerosis complex 2 (TSC2), an upstream regulator of mTOR, on Thr1227 and Ser1345, thus, promoting autophagy (30). The impact of the mTOR signaling pathway on autophagy is discussed below in mTOR signaling pathway. AMPK also directly phosphorylates ULK1 at Ser317, Ser467, Ser555, Ser637, Ser777, and Thr574, recruiting FIP200, ATG13, and ATG101 to form the ULK1 complex (31). The ULK1 complex in turn activates the PI3K complex 1, including VPS14, VPS34, BECN1, and ATG14 L, generating PtdIns3P, which helps shape the phagophore membrane (32). AMPK can also phosphorylate beclin-1 at Thr388, enhancing beclin-1 association with VPS34 and ATG14 and autophagosome formation rate under nutrient deprivation (33).

mTOR signaling pathway.

mTOR, a serine threonine kinase, is a member of the PI3K-related kinase (PIKK) family (34). As part of the PI3K/Akt/mTOR signaling axis, mTOR plays a central role in the regulation of various physiological functions such as cell growth, metabolism, cell cycle progression, transcription, translation, migration, motility, as well as apoptosis and autophagy (35, 36). mTOR is the common catalytic subunit of two functionally different protein complexes, mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2) (35). The three main components of mTORC1 are the mTOR subunit, rapamycin-sensitive adaptor protein of mTOR (Raptor), and mammalian lethal with Sec-13 protein 8 (mLST8), which is also known as GβL (35). Proline-rich Akt substrate 40 kDa (PRAS40) and DEP domain TOR-binding protein (Deptor) are two inhibitory subunits that are also part of the mTORC1 complex (37–39). mTORC1 is regulated by growth factors, amino acids, energy, and oxygen among others, thus making it a potent orchestrator of cell growth and metabolism (40). The heterodimeric TSC complex composed of hamartin (TSC1) and tuberin (TSC2) negatively regulates mTORC1 (41). TSC2 is phosphorylated by the PI3K/Akt signaling, disturbing the TSC complex subcellular localization and rendering it inactive (42). This leads to the activation of RHEB, which strongly enhances mTORC1 activity (42). mTORC1 was described to regulate autophagy through ATG1 homologue, ULK1. mTORC1 phosphorylates ULK1 at Ser757, thereby preventing ULK1 activation and the formation of ULK1 and PI3K complexes, and consequently suppressing the process of autophagy (43). mTORC1 also phosphorylates transcription factor EB (TFEB) at Ser142 and Ser211, preventing TFEB translocation to the nucleus and decreasing autophagy gene transcription (44). Therefore, in conditions of nutrient or growth factor deprivation, mTORC1 signaling pathway is inhibited, thereby activating autophagy via ULK1 and PI3K complexes.

TOR-autophagy spatial coupling compartment (TASCC) is a distinct cellular compartment at the trans side of the Golgi apparatus, where autolysosomes and mTORC1 complexes colocalize. Amino acids derived from autolysosomal breakdown products are essential for the recruitment of mTOR to TASCC. This then mediates protein synthesis, reinforcing lysosome and autophagy biogenesis. Since mTORC1 is known to inhibit autophagy during the initial stages of autophagosome formation, TASCC facilitates autophagy by sequestering mTORC1, thereby, creating a spatiotemporal arrangement of mTORC1 (45). Importantly, TASCC were observed in high autophagy cells including glomerular podocytes, sequestering mTOR and regulating autophagy (45, 46).

Similar to mTORC1, mTORC2 has been shown to be a negative regulator of autophagy (47). The mTORC2 complex, also known as the rapamycin-insensitive complex, includes the rapamycin-insensitive companion of mTOR (Rictor), and the mammalian stress-activated protein kinase (SAPK)-interacting protein 1 (mSin1), mTOR, and mLST8 (48). mTORC2 is an effector of insulin/PI3K signaling. In the absence of insulin, mTORC2 is inhibited by the pleckstrin homology domain of mSin1. The binding of phosphatidylinositol (3,4,5)-trisphosphate (PIP3) to mSin1 in the plasma membrane reverses the inhibition by mSin1 (49). By phosphorylating mSin1 at Thr86, Akt activates mTORC2, which in turn further stimulates Akt by direct phosphorylation at Ser473, forming a positive feedback loop (50). Akt can directly phosphorylate beclin-1 at Ser234 and Ser295, promoting binding of beclin-1 to intermediate filaments and autophagy suppression (51).

Sirtuin signaling pathway.

Sirtuins are a family of NAD+-dependent class III histone deacetylases activated under conditions of low nutrient availability or caloric restriction. Sirtuin 1 (SIRT1) has been found to deacetylate essential autophagic proteins, such as ATG5, ATG7, and ATG8, enhancing their activity and promoting autophagic flux. SIRT1 also deacetylates forkhead box protein O1 (FoxO1), FoxO3, and FoxO4, and regulates FoxO-dependent gene transcription. By deacetylating these proteins, SIRT1 facilitates the efficient formation of autophagosomes, contributing to the initiation and progression of macroautophagy (52).

Selective Macroautophagy

Unlike nonselective macroautophagy, selective macroautophagy targets and engulfs specific cellular components such as protein aggregates (aggrephagy), damaged mitochondria (mitophagy), LDs (lipophagy), endoplasmic reticulum (ER-phagy), and invading microbes (xenophagy). The process of selective macroautophagy begins with the recognition of specific cargo by autophagy receptors or adaptors (53). In this review, we focus specifically on mitophagy and lipophagy due to their relevance in podocytes and role in both physiological and pathological conditions within the kidney.

Mitophagy.

Mitophagy is the process of selective elimination of damaged and dysfunctional mitochondria via macroautophagy (54). Numerous research studies have identified various mechanisms of mitophagy that are triggered in response to different types of stressors, such as oxidative stress, hypoxia, mitochondrial depolarization, and mitochondrial DNA damage (55–60). Based on specific signals on impaired mitochondria that initiate mitophagy, this process can be categorized into ubiquitin-dependent mitophagy, ubiquitin-independent or receptor-based mitophagy, lipid-based mitophagy, and micromitophagy (61) (Fig. 2).

Figure 2.

Molecular mechanisms of mitophagy. A: ubiquitin-dependent mitophagy: in response to mitochondrial damage, phosphatase and tensin homolog deleted in chromosome 10 (PTEN)-induced putative kinase 1 (PINK1) accumulates on the outer mitochondrial membrane, phosphorylating Parkin. Parkin then ubiquitinates outer mitochondrial proteins such as mitofusins 1/2 (Mfn1/2), mitochondrial GTPase Miro1/2 (Miro1/2), voltage-dependent anion channel (VDAC), and translocase of the outer mitochondrial membranes (TOMs) leading to mitophagy. Ubiquitin-dependent mitophagy can also occur independent of the Parkin. Glycoprotein 78 (GP78) is an E3 ubiquitin ligase anchored to the endoplasmic reticulum membrane that can also ubiquitinate mitochondrial proteins such as Mfn1/2. Mitochondrial E3 ubiquitin ligase 1 (MUL1) works in tandem with the PINK1-Parkin pathway and has the capacity to compensate for PINK1 or Parkin loss. In addition, PINK1 can recruit SYNPHILIN1 to the mitochondria which will recruit the E3 ubiquitin ligase seven in absentia homolog 1 (SIAH1). SIAH1 subsequently ubiquitinates mitochondrial proteins, initiating mitophagy. B: receptor-based mitophagy: mitophagy is initiated by receptors on the outer mitochondrial membranes that include microtubule-associated protein 1 light chain 3 (LC3)-interacting region (LIR) motifs, such as FUN14 domain containing 1 (FUNDC1), NIP3-like protein X (NIX)/ BCL2 interacting protein 3 (BNIP3)L, BNIP3, BCL2 like 13 (BCL2L13), and FKBP prolyl isomerase 8 (FKBP8), allowing them to engage directly with LC3 on the phagophore. C: lipid-based mitophagy: cardiolipin translocates to the outer mitochondrial membranes upon mitochondrial damage where it acts as an eat-me signal for mitophagy by directly binding to LC3 on the phagophore. Phospholipid scramblase-3 (PLSCR3) flips cardiolipin from the inner to the outer mitochondrial membrane, while nucleoside diphosphate kinase (NDPK)-D transports CL across the intermembrane space. Ceramide buildup on the outer mitochondrial membranes can also stimulate mitophagy. D: micromitophagy: in micromitophagy mitochondrial cargo is directly encapsulated by mitochondria-derived vesicles (MDVs). There are two types of MDVs: TOM20-positive MDVs and pyruvate dehydrogenase (PDH)-positive MDVs. MDVs are then transported to lysosomes for degradation. CerS1, ceramide synthase 1; DNM1L, dynamin 1-like protein; HOPS complex, homotypic fusion and protein-sorting complex; IMM, inner mitochondrial membrane; MVB, multivesicular bodies; NDP52, nuclear domain 10 protein 52; OMM, outer mitochondrial membrane; OPTN, optineurin; PINK1, transmembrane Ser/Thr kinase; SNARE, complex consisting of syntaxin 17 (STX17), synaptosome-associated protein 29 (SNAP29), and vesicle-associated membrane protein 7 (VAMP7); STX17, syntaxin 17; TOLLIP, Toll-interacting protein.

Ubiquitin-dependent mitophagy.

Ubiquitin-dependent mitophagy is tightly regulated by the transmembrane Ser/Thr kinase PINK1 [phosphatase and tensin homolog deleted in chromosome 10 (PTEN)-induced putative kinase 1]. When mitochondrial damage occurs, PINK1 stabilization on the mitochondrial outer membrane recruits the E3 ubiquitin ligase Parkin (PRKN, also known as PARK2) (54, 62), which ubiquitinates outer mitochondrial proteins, including mitofusins 1/2 (Mfn1/2) (63, 64), mitochondrial GTPase Miro1/2 (65, 66), voltage-dependent anion channel (VDAC) (67), translocase of the outer mitochondrial membranes (TOMs) (68), and mitochondrial hexokinase (66). Ubiquitin-modified mitochondrial proteins are identified and targeted toward autophagosomes by specific adaptor proteins. Among these adaptor proteins, nuclear domain 10 protein 52 (NDP52) and optineurin are thought to facilitate the recruitment of mitochondria to autophagosomes either by their inherent microtubule-associated protein 1 light chain 3 (LC3)-interacting region (LIR) (69) or by directly assisting in the assembly of autophagosomes through interactions with ATG8 homologs (70).

Several studies have identified mitophagy pathways that are ubiquitin-dependent, but independent of Parkin. These pathways act either in parallel or in addition to the PINK1-Parkin pathway. For instance, glycoprotein 78 (GP78) mitophagy pathway is ubiquitin dependent, but independent of Parkin and PINK1. GP78 is an ER membrane-anchored E3 ubiquitin ligase that plays a crucial role in ER-associated degradation (ERAD) and is localized to the mitochondria-associated ER domain (71). Overexpression of GP78 has been found to ubiquitinate Mfn1 and Mfn2, mediating their proteasomal degradation, and leading to mitochondrial fragmentation (72). Another reported ubiquitin-dependent but Parkin-independent mitophagy pathway involves PINK1, SYNPHILIN1, and seven in absentia homolog 1 (SIAH1). PINK1 recruits SYNPHILIN1 to mitochondria, causing mitochondrial depolarization and stabilizing uncleaved PINK1 at the organelle. SYNPHILIN1 subsequently recruits the E3 ubiquitin ligase SIAH1, which ubiquitinates mitochondrial proteins. The additional recruitment of autophagosome marker LC3 and lysosome marker Lamp1 to mitochondria then initiates mitophagy (73). The mitochondrial E3 ubiquitin ligase 1 (MUL1) acts in parallel to the PINK1-Parkin pathway and has the potential to compensate for the loss of PINK1 or Parkin (74). Under stress conditions or high reactive oxygen species (ROS) levels, MUL1 interacts with ULK1, leading to its ubiquitination and proteasomal degradation, promoting mitophagy (75). P62/SQSTM1 is another ubiquitin-based mitophagy pathway. In a recent study, it was found that the ubiquitination of mitochondrial proteins during mitophagy necessitated p62 but not Parkin. These findings were observed in Dnm1l/Drp1 knockout mice, in which reduced mitophagy occurred due to the loss of dynamin 1-like protein (DNM1L), a dynamin-related GTPase that mediates mitochondrial division, leading to enlarged mitochondria and decreased mitophagy (76).

Ubiquitin-independent or receptor-based mitophagy pathways.

Apart from ubiquitin-dependent mitophagy, various other mitophagy mechanisms have been identified, which do not rely on mitochondrial ubiquitination but where autophagy receptors directly interact with LC3 and/or GABA receptor-associated protein (GABARAP) through LIR motifs, such as BCL2/adenovirus E1B 19 kDa protein interacting protein 3 (BNIP3) (77), FUN14 domain containing 1 (FUNDC1) (78), NIP3-like protein X (NIX)/BNIP3L (79), BCL2 like 13 (BCL2L13) (80), and FKBP prolyl isomerase 8 (FKBP8) (81).

Lipid-based mitophagy.

Specific mitochondrial lipids can also function as mitophagy receptors, including cardiolipin (CL) and ceramide (61). CL is a phospholipid primarily found in the inner mitochondrial membrane. Upon mitochondrial stress or damage, CL is translocated to the outer mitochondrial membrane. This process is mediated by mitochondrial phospholipid scramblase-3 (PLSCR3) and by the intermembrane space protein complex NDPK-D. PLSCR3 flips CL from one membrane side to other, whereas NDPK-D transports CL across the intermembrane space. Once at the outer mitochondrial membrane, CL directly recruits LC3 to mitochondria by binding to LC3’s N-terminal helix, initiating mitophagy without requiring mitochondrial depolarization (82–84). Mitophagy can also be induced by ceramide accumulation at mitochondria. Ceramide can directly interact with LC3 on autophagosomes. Recent evidence highlights that mitochondrial stress triggers the trafficking of ceramide synthase 1 (CerS1) from the ER to the outer mitochondrial membrane through mitochondria-associated membranes (MAMs). This translocation, facilitated by the ER-mitochondria trafficking protein p17/PERMIT interacting with translocase of the outer mitochondrial membrane 20 (TOM20), leads to C18-ceramide generation in mitochondria, which in turn triggers LC3B-II-mediated autophagosome targeting to mitochondria, resulting in lethal mitophagy (85).

Micromitophagy.

In addition to macromitophagy, which involves the selective macroautophagic removal of damaged mitochondria, another process known as micromitophagy has been identified. Micromitophagy is a mechanism where parts of mitochondria are directly engulfed by lysosomes for degradation without the formation of autophagosomes (58, 60). This selective process targets only the damaged portion of mitochondria, rather than the entire organelle. It occurs through the formation of mitochondria-derived vesicles (MDV) that bud off from mitochondria and then translocate to lysosomes for degradation (86). Two types of MDVs have been well described in the literature: TOM20-positive MDVs and pyruvate dehydrogenase (PDH)-positive MDVs (19). In TOM20 MDVs, MDV formation involves Miro1 and Miro2 driving thin membrane protrusions, followed by the recruitment of DNM1L to form foci and catalyze the scission of the mitochondrial protrusion to complete the formation of MDV (87). Following budding, trafficking of MDVs to multivesicular bodies (MVBs) and lysosomes is mediated by the cooperation of Toll-interacting protein (TOLLIP) and Parkin (88). Under oxidative stress, PDH-positive MDVs form via PINK1 accumulation at the outer mitochondrial membrane (OMM), recruiting PRKN, which induces ubiquitylation of OMM proteins to promote the formation and budding of MDVs (89). The fusion of MDVs with endolysosomes is mediated by a SNARE complex consisting of syntaxin 17 (STX17), synaptosome-associated protein 29 (SNAP29), and vesicle-associated membrane protein 7 (VAMP7), in a homotypic fusion and protein-sorting complex (HOPS complex)-dependent manner (90). Another process of micromitophagy involves the formation of a spermatogenesis-associated protein 18 (SPATA18; also known as MIEAP)-induced vacuole (MIV) that engulfs entire dysfunctional mitochondria [reviewed in detail elsewhere (19)].

Lipophagy.

Lipophagy, originally described in hepatocytes (91), is a form of selective macroautophagy that targets LDs for degradation within lysosomes, breaking down stored lipids into free fatty acids (FFAs) (92). There are two forms of lipophagy: macrolipophagy and chaperone-mediated lipophagy. Recent research has highlighted the pivotal role of mTORC1-perilipin-3 pathway and RabGTP enzymes, specifically Rab7 and Rab10, in LD homeostasis. Rab10 facilitates the engulfment of LDs into autophagosomes by recruiting LC3-positive autophagic membranes through interactions with adaptor proteins EH domain binding protein 1 (EHBP1) and membrane deformed ATP EH domain-containing 2 (EHD2) (93). Rab7 regulates the recruitment of lysosomes toward LDs. LDs are then degraded by lysosomal acid lipase (LAL) (94). The recruitment of autophagic components is predominantly mediated by LIR-containing adapter proteins, such as p62 (95). p62 is recruited to LDs via the interaction with perilipin 1 (PLIN1), an LD-associated protein (96). Notably, p62 coimmunoprecipitates with PLIN1 (97) along with two additional LD proteins, PLIN2 (98) and phospholipase A2 group IVA (PLA2G4A)/cPLA2-α (99). The interaction of PLIN1 and p62 is essential for colocalization of LDs with LC3 (96). Activation of lipophagy reduces levels of p62 and LDs. In contrast, the presence of autophagosome-lysosome fusion inhibitors, such as bafilomycin A and chloroquine, leads to p62 and LDs accumulation (96, 98, 99). Furthermore, knockdown of SQSTM1 was found to increase LDs accumulation (96). Collectively, these findings suggest that p62 is a lipophagy specific autophagy receptor (SAR).

Chaperone-Mediated Autophagy

During CMA, substrates containing KFERQ-like motifs are initially recognized by cytosolic HSC70. Subsequently, the substrate-chaperone complexes are directed to the lysosomal surface where they bind to LAMP2A (100). The substrate then undergoes unfolding, a crucial step for translocation. The interaction between the substrate and LAMP2A leads to the multimerization of LAMP2A, facilitating the translocation of the substrate into the lysosomal lumen (101). Notably, only LAMP2A outside of the lipid-rich domain can multimerize (102). HSP90 interacts with LAMP2A to maintain the stability of the multimeric assembly (103). Once inside the lysosomal lumen, substrates are degraded by lysosomal proteases. Cytosolic HSC70 is then released from the translocation complex to bind new substrates, ensuring continuous CMA activity. Furthermore, HSC70 in the lysosomal lumen promotes the disassembly of the LAMP2A multimerization complex into monomers, enabling them to bind new substrates. The degradation of LAMP2A occurs within lipid microdomains of lysosomal membranes and is facilitated by cathepsin A and metalloproteinase enzymes (104).

The rate-limiting step of CMA is the binding of substrates to LAMP2A. Hence, the levels of LAMP2A at the lysosomal membranes directly correlate with CMA activity (105). The mTORC2/serine-threonine kinase 1 (Akt1)/pleckstrin homology (PH) domain and leucine-rich repeat protein phosphatase 1 (PHLPP1) axis has also been shown to regulate CMA. Phosphorylation of lysosomal Akt by mTORC2 increases the phosphorylation of Glial fibrillary acidic protein (GFAP), leading to the disassembly of the CMA translocation complex at the lysosomal membrane (106). In contrast, PH Domain And Leucine Rich Repeat Protein Phosphatase 1 (PHLPP1) is recruited to the lysosomal membrane under stress in a Rac1-dependent manner and dephosphorylates lysosomal Akt, stabilizing the translocation complex (106). Notably, Akt is also regulated by the insulin PI3K–3-phosphoinositide-dependent protein kinase 1 (PDPK1) pathway, where the inhibition of class-I PI3K or PDPK1 decreases Akt phosphorylation, thereby activating CMA (107).

Recent studies have highlighted a connection between CMA and LD breakdown, referred to as chaperone-mediated lipophagy. During chaperone-mediated lipophagy, PLINs, notably PLIN2 and PLIN3, are phosphorylated by the heat-shock cognate protein HSPA8/HSC70 and bind to LAMP2A. Subsequently, they are targeted for CMA-mediated degradation in lysosomes (108–111). This degradation activates esterases and lipases, which interact with Patatin-like phospholipase domain-containing 2 (PNPLA2) to initiate lipolysis. PNPLA2, a calcium-independent phospholipase, can interact with LC3 to bridge LDs to autophagosome membrane (112, 113) or engage with cargo receptor isolator-1/p62 to recruit LDs (114). Of note, LAMP2A deficiency was found to inhibit LD breakdown (111).

AUTOPHAGY IN PODOCYTE INJURY

Autophagy plays a crucial role in maintaining podocyte homeostasis (115). The fundamental mechanisms of autophagy are conserved across different cell types, including podocytes. Nevertheless, the regulation and function of autophagy in podocytes may be specifically adapted to preserve glomerular function and integrity (24, 116).

In transgenic mice with GFP-tagged LC3, a distinct punctate distribution of LC3-GFP was primarily observed in podocytes under physiological conditions. In contrast, kidney tubules and other glomerular cells exhibited minimal staining. This observation implies that podocytes maintain high levels of basal autophagy. Moreover, inhibiting autophagosomal degradation with the lysosomal inhibitor chloroquine resulted in the rapid accumulation of LC3-GFP puncta in podocytes, indicating the presence of a normal autophagic flux in these cells (7). In 12-wk-old podocyte-specific Atg5 knockout mice, a rapid onset of albuminuria was reported, emphasizing the essential role of autophagy in preserving podocyte function. Nevertheless, no significant differences were observed in kidney histology, glomerular ultrastructure, or albuminuria between 2- and 4-mo-old mice with a constitutive podocyte-specific ATG5 deletion and their wild-type littermates for up to 4 mo, suggesting that increased proteasome activity may compensate for the absence of autophagy (7). In support of an important role of autophagy in podocyte function, mice with podocyte-specific deletion of genes encoding mTOR, prorenin, or PI3K catalytic subunit type 3, developed severe glomerulosclerosis and proteinuria (14, 117, 118). These observations underscore the critical role of a high basal autophagic flux in maintaining podocyte homeostasis and suggests that dysregulated autophagy may contribute to the development of kidney diseases. This review will focus on recent advances in understanding the role of diverse forms of autophagy in the pathogenesis of podocyte injury in CKD.

Nonselective Macroautophagy

Impaired macroautophagy has been implicated in the pathogenesis of various kidney diseases, such as diabetic kidney disease (DKD), focal segmental glomerulosclerosis (FSGS), lupus nephritis (LN) (24), membranous nephropathy (7), IgA nephropathy (119), and human immunodeficiency virus (HIV)-associated nephropathy (HIVAN) (119) (Table 1).

Table 1.

Kidney disease phenotype following podocyte-specific genetic modulation of autophagy-related genes in CKD models

Kidney Disease	Model and Target Gene	Kidney Phenotypical Changes	References
Macroautophagy
DKD	STZ-induced diabetic podocyte-specific Atg5 knockout mice (podo-Atg5−/−)	Accelerated diabetes-induced podocytopathy with a leaky GFB and glomerulosclerosis	(115)
	High-fat diet (HFD)-fed podocyte-specific Atg5 knockout mice (podo-Atg5−/−)	Proteinuria accompanied by podocyte loss and enlarged damaged lysosomes in podocytes	(120)
	Podocyte-specific Tsc1 knockout mice (PcKOTsc1 mice)	DKD-like features including podocyte loss, glomerular basement membrane thickening, mesangial expansion, proteinuria, mislocalization of slit diaphragm proteins, and enhanced ER stress in podocytes, which can be prevented by rapamycin treatment	(121)
	Podocyte-specific Raptor knockout diabetic mice (Raptorfl/fl, db/db mice)	Protected from the development of DKD	(122)
	Human podocytes transfected with shSPAG5	Attenuated apoptosis, induced autophagy, and suppressed AKT/mTOR pathway in high glucose-treated human podocytes	(123)
	Podocyte-specific Sirt6 deletion mice (Podocin-Cre Sirt6fl/fl mice)	Accelerated podocyte injury and proteinuria in STZ-induced and non-STZ-induced models	(124)
	GFP-LC3 calpastatin transgenic (CSTTg) mice (unilateral left nephrectomy followed by STZ to induce type 1 diabetes)	Normalized podocyte autophagic flux, reduced nephrin loss, and prevented the development of albuminuria in diabetic mice	(125)
FSGS	Podocyte-specific Atg5 knockout mice (Adriamycin, PAN)	Albuminuria, glomerulosclerosis, foot process fusion, and podocyte loss	(7)
	Podocyte-specific Atg7 knockout mice (Adriamycin)	Podocyte injury, glomerulopathy, and proteinuria	(126)
LN	Immortalized human podocyte cell line infected with lentivirus vector ATG5 shRNA	Exacerbated IFN-α-induced derangement of podocin and impairment of GFB	(127)
HIVAN	Podocyte-specific SIRT1-overexpressing mice (Tg26;Pod-SIRT1OV mice)	Attenuated albuminuria, kidney lesions, and expression of inflammatory markers	(121)
Mitophagy
DKD	Conditionally immortalized mouse podocytes transfected with LV-CA-FoxO1	Activated the PINK1/Parkin pathway and protected against mitochondria-induced apoptosis in high glucose-treated podocytes	(128)
	Conditionally immortalized mouse podocytes transfected with LV-CA-FoxO1	Prevented high glucose-dependent decrease in the expression of the mitophagy-related proteins PINK1, Parkin, Drp-1, and Mfn1 and protected mitochondrial function	(129)
	STZ-induced diabetic mice transfected with LV-CA-FoxO1	Protected against apoptosis of glomeruli and ameliorated the progression of DKD	(128, 129)
	Conditionally immortalized mouse podocytes transfected with siRNA-SNHG17	Promoted Parkin-dependent mitophagy and reduced apoptosis of high glucose-treated podocytes through regulating the degradation of Mst1	(130)
	STZ-induced diabetic mice injected with adeno-associated virus serotype 9 (AAV9) system carrying siRNA-SNHG17	Promoted mitophagy and attenuated the progression of DKD	(130)
	Conditionally immortalized mouse podocytes transfected with Parkin siRNA	Aggravated palmitic acid-induced mitochondrial dysfunction, mitoROS production, and podocyte apoptosis	(131)
LN	Conditionally immortalized mouse podocytes transfected with nestin shRNA	Decreased nephrin, p-nephrin (Y1217), and mitophagy-associated proteins in podocytes stimulated with LN plasma	(132)
	MRL/lpr mice renally injected with nestin-shRNA-Ad	Aggravated proteinuria and renal pathological changes in MRL/lpr mice	(132)

Ad, adenovirus; ATG, autophagy related gene; CA, constitutively active; CKD, chronic kidney disease; DKD, diabetic kidney disease; DRP1, dynamin-related protein; ER, endoplasmic reticulum; FoxO1, forkhead box protein O1; FSGS, focal segmental glomerulosclerosis; GFB, glomerular filtration barrier; GFP, green fluorescent protein; HIVAN, HIV-associated nephropathy; IFN-α, interferon-α; LC3, microtubule-associated protein 1 light chain 3; LN, lupus nephritis; LV, lentivirus; Mfn1, mitofusin 1; mitoROS, mitochondrial reactive oxygen species; Mst1, mammalian sterile 20-like kinase 1; mTOR, mechanistic target of rapamycin; PAN, puromycin aminonucleoside; Raptor, rapamycin-sensitive adaptor protein of mTOR; Sirt6, sirtuin 6; SPAG5, sperm-associated antigen 5; STZ, streptozotocin; Tsc1, tuberous sclerosis complex 1.

Nonselective macroautophagy in DKD.

In cultured mouse podocytes, exposure to elevated glucose levels for 48 h was shown to trigger autophagy, whereas prolonged exposure for 15 days resulted in the suppression of autophagy (133). Consistent with the in vitro observations, LC3 staining was more pronounced in transgenic mice systemically expressing GFP-LC3, 4 wk post-streptozotocin (STZ) injection, at which time the mice are in a hyperglycemic state without evident glomerular lesions, compared with GFP-LC3 control mice. However, LC3 staining decreased at 8 wk post-STZ injection when mice exhibited glomerular lesions. Furthermore, LC3B-II accumulation was observed in the glomeruli of mice at 4 wk post-STZ injection, treated with chloroquine, suggesting an active autophagic flux (133). In addition, podocyte-specific Atg5 knockout mice showed exacerbated STZ-induced glomerulopathy compared with wild-type mice (133). These findings suggest that autophagy is activated at early stages of diabetes for renoprotection, but suppressed at later stages, potentially contributing to the progression of DKD. In support, kidney biopsy samples from patients with diabetes revealed a decrease in autophagy within podocytes when compared with kidney specimens from individuals with minimal changes in histology, low urinary protein excretion, or only hematuria (134). Another study reported insufficient podocyte autophagy in patients with type 2 diabetes and rats with obesity-induced diabetes (OLETF rats) and massive proteinuria, but not in those with no or minimal proteinuria (120). Furthermore, high-fat diet (HFD)-fed podocyte-specific Atg5 knockout mice exhibited podocyte loss and substantial proteinuria, accompanied by enlarged damaged lysosomes in their podocytes. Interestingly, the stimulation of cultured podocytes with serum from individuals with type 2 diabetes or OLETF rats with pronounced proteinuria inhibited autophagy and induced lysosomal dysfunction and apoptosis (120). This implies that factors derived from the serum may exert a negative regulatory influence on podocyte autophagy during the progression of DKD.

Reduced autophagy can be observed in the kidneys of patients with DKD and correlates with alterations in metabolic regulators, including mTOR, AMPK, and SIRT1. Enhanced mTORC1 activity is observed in both human and experimental type 1 and type 2 DKD (135, 136). In addition, podocyte-specific mTORC1 activation induced via Tsc1 deletion recapitulated many DKD features, including podocyte damage, effacement of foot processes, and albuminuria in nondiabetic mice (122). Conversely, genetic reduction of podocyte-specific mTORC1 in diabetic mice suppressed the progression of DKD (122). Moreover, a recent study demonstrated that sperm-associated antigen 5 (SPAG5), a mitotic spindle-associated protein, inhibits podocyte autophagy through the SPAG5/Akt/mTOR pathway, leading to podocyte injury and apoptosis, whereas downregulation of the lncRNA SPAG5-AS1 was found to enhance autophagy and reduce podocyte apoptosis (123). Increasing studies have demonstrated that the AMPK signaling pathway is inactivated in the diabetic kidney, leading to reduced autophagy, proteinuria, and renal pathological changes (137–139). AMPK activation has been shown to promote autophagy and alleviate DKD by activating SIRT1 (139, 140). Similar to AMPK, SIRT1 expression is decreased in the kidneys of experimental type 1 and type 2 diabetic animals (141), and the activation of the SIRT1 signaling pathway was found to activate podocyte autophagy (142). Similarly, knockdown of METTL14, a protein that facilitates the degradation of Sirt1 mRNA m6A in injured podocytes, promotes autophagy while alleviating apoptosis and inflammation (143) and silencing of miR-150-5p, which mediates the interaction between SIRT1 and p53, results in increased AMPK-dependent autophagy, thereby attenuating renal injury in DKD (144). Activation of liver X receptors (LXRs) inhibits autophagy by modulating the mTOR, AMPK, and SIRT1 signaling pathways, thereby leading to podocyte injury (145). Recently, several other pathways that contribute to podocyte injury in glomerular diseases were also found to regulate autophagy. For example, SIRT6 plays an important role in podocyte health by maintaining autophagy levels via the inhibition of NOTCH signaling. In support, reduced expression of SIRT6 was found in kidney biopsy samples from patients with DKD and podocyte-specific Sirt6 deletion in mice accelerates podocyte injury and proteinuria in both STZ-induced and non-STZ-induced models (124). Furthermore, a new mechanism that connects transient receptor potential cation channel subfamily C member 6 (TRPC6) and calpain activity to impaired podocyte autophagy has recently been reported (125). Kidney biopsies from patients with type 2 diabetes revealed increased TRPC6 expression in association with decreased calpastatin (an endogenous calpain inhibitor) expression, reduced autophagy, and podocyte injury. Transgenic overexpression of calpastatin, or pharmacologic inhibition of calpain activity were found to normalize podocyte autophagic flux, reduced nephrin loss, and prevent the development of albuminuria in diabetic mice (125). Other intracellular stress signaling pathways such as oxidative stress and endoplasmic reticulum stress can also negatively regulate autophagy in DKD, thereby contributing to disease progression (146). Taken together, recent studies have established an important role for deficient podocyte nonselective macroautophagy in DKD progression and identified several novel druggable targets that are amenable to increase podocyte autophagy as a mean to prevent the progression of DKD.

Nonselective macroautophagy in FSGS.

Several studies indicate that podocyte autophagy also plays a crucial role in the development of FSGS. Variants of the apolipoprotein L1 (APOL1) gene known to be associated with FSGS have been shown to inhibit autophagic flux, ultimately leading to podocyte injury (147, 148). In experimental models of FSGS, a positive correlation was observed between LC3-II levels and the number of autophagic vacuoles in podocytes. This correlation was also linked to the recovery from the damage induced by puromycin aminonucleoside (PAN) nephrosis (149). In addition, podocyte-specific Atg5 knockout mice were more susceptible to PAN-induced glomerulosclerosis than wild-type controls (7). These findings suggest that podocyte autophagy might counter the development of FSGS, similarly to what is observed in experimental models of DKD. Moreover, it has been previously shown that treatment with rapamycin mitigated adriamycin (ADR)-induced podocyte injury in vitro and in vivo, while chloroquine, an autophagy inhibitor, exacerbated ADR-induced podocyte apoptosis. Analogously, podocyte loss, glomerulosclerosis, and proteinuria were increased in podocyte-specific Atg5 or Atg7 knockout mice upon the ADR treatment (7, 126). Autophagy in podocytes of patients with FSGS has also been investigated. Specifically, patients with FSGS exhibit a markedly lower percentage of autophagosome-positive podocytes and significantly reduced levels of kidney beclin-1 compared with those with minimal change disease (MCD). Repeated kidney biopsies from patients with MCD indicated that individuals with diminished podocyte autophagy are more prone to progress to FSGS (150).

Nonselective macroautophagy in other glomerular disorders.

In addition to its role in DKD and FSGS, autophagy is also implicated in various other glomerular diseases, including lupus nephritis (LN), membranous nephropathy (MN), IgA nephropathy (IgAN), and HIV-associated nephropathy (HIVAN). Autophagy was found to be increased in MRL^lpr/lpr mice, patients with LN, and podocytes treated with IFN-α and sera from patients with LN. The inhibition of autophagy by 3-methyladenine and ATG5 siRNA exacerbated podocyte injury (127). Similarly, renal biopsies from patients with MN revealed a 10-fold higher expression of ATG3 mRNA and increased autophagosome formation compared with control biopsies (7). This suggests a potential role for enhanced autophagic activity in maintaining podocyte integrity during the progression of glomerular diseases. In the context of IgA nephropathy, mouse podocytes cultured with aggregated IgA1 from patients with IgAN showed impaired autophagy, characterized by decreased LC3II and increased p62 levels, which was reversed by rapamycin treatment (119), indicating that modulating autophagy could be a potential approach to alleviate podocyte injury and the progression of IgAN. Moreover, a recent study highlighted SIRT1 as a potential drug target in HIVAN. SIRT1 expression was found to be reduced in the glomeruli of human and mouse HIVAN kidneys. In immortalized human podocytes transduced with HIV-1 pseudovirus, SIRT1 expression was decreased, whereas NF-κB p65 and STAT3 were increased. In vivo, the administration of the small-molecule SIRT1 agonist BF175 or inducible overexpression of SIRT1 specifically in podocytes significantly attenuated albuminuria, renal lesions, and the expression of inflammatory markers in HIV-1 transgenic mice, supporting a critical role of SIRT1 in HIVAN (121).

Selective Macroautophagy

Mitophagy.

Under pathophysiological conditions, mitophagy is considered a defense mechanism to eliminate damaged and dysfunctional mitochondria. Hence, it has been suggested that in kidney diseases, mitophagy is initially activated to ensure mitochondrial quality. However, as the disease progresses, mitophagy may become impaired, resulting in the activation of apoptotic pathways to mitigate organelle and tissue damage (151) (Table 1). High glucose-induced mitochondrial dysfunction is strongly linked to podocyte injury in DKD. Previous findings indicate that culturing mouse podocytes in high-glucose medium decreases PINK1 expression, thereby inhibiting mitophagy and preventing proper mitochondrial turnover. This inhibition is associated with a reduction in the transcriptional activity of FoxO1 and an increase in podocyte apoptosis. Under high-glucose conditions, FoxO1 is inactivated by the PI3K/Akt pathway. Overexpressing FoxO1 has been shown to mitigate podocyte apoptosis and reduce podocyte cell damage by upregulating PINK1 expression in high glucose-treated podocytes and in the glomeruli of diabetic mice (128). The same research group further confirmed that PINK1 is a direct downstream target of FoxO1, primarily through the PINK1-binding site. Upregulation of FoxO1 expression has been shown to have protective effects against high glucose-induced mitochondrial dysfunction and podocyte injury in both in vitro and in vivo diabetic models, mediated through the activation of PINK1/Parkin-dependent mitophagy (129). In line with these findings, Zhou et al. (152) recently reported significant decreases in the expression of PINK1, PARK2, peroxisome proliferator-activated receptor-γ coactivator (PGC-1α), and Sirt1 in human podocytes treated with high glucose, suggesting impaired mitophagy and mitochondrial biogenesis. In addition, progranulin (PGRN), an autocrine growth factor with anti-inflammatory properties, was found to be significantly reduced in the kidneys of STZ-induced diabetic mice and in biopsies from patients with DKD. Treatment with recombinant human PGRN prevented the progression of DKD by activating mitophagy and mitochondrial biogenesis, thereby alleviating mitochondrial dysfunction and podocyte injury. A plausible mechanism underlying PGRN-mediated mitophagy and mitochondrial biogenesis in podocytes of diabetic mice involves PGRN-SIRT1-peroxisome proliferator-activated receptor-γ coactivator (PGC1α) mediated regulation of FoxO1 (152). Consistently, inhibition of AMPK or SIRT1 or downregulation of PGC-1α through siRNA significantly decreased ATP production and reduced the expression of mitochondrial biogenesis markers such as Nrf-1 and mitochondrial factor A (TFAM) in a DKD mouse model (153). TFAM also plays a crucial role in maintaining mitochondrial DNA integrity in DKD. In addition, research has shown that individuals with DKD exhibit significantly reduced PGC-1α expression levels in their kidney biopsies compared with individuals without diabetes (154). Similarly, diminished SIRT1 expression levels were associated with increased albuminuria in patients with DKD (155). In a recent study, podocytes exposed to high-glucose conditions exhibited exacerbated podocyte injury in association with reduced PINK1/Parkin-mediated mitophagy, as indicated by a decrease in the expression of beclin-1, LC3II/LC3I ratio, Parkin, and PINK1 and increased p62 expression. Importantly, these changes were reversed by treatment with placental mesenchymal stem cells (P-MSCs). Mechanistically, P-MSCs alleviated podocyte injury and mitigated PINK1/Parkin-mediated mitophagy inhibition in DKD by activating the SIRT1-PGC-1α-TFAM pathway (156).

Increasing evidence has emerged linking Src, the prototypical member of the Src family kinases, to the development of DKD (157). Indeed, Src activation in kidneys of db/db mice positively correlates with mitochondrial damage, podocyte apoptosis, and renal dysfunction, whereas inhibition of Src with PP2 demonstrated protective effects against mitochondrial damage and podocyte apoptosis. In podocytes subjected to high glucose in vitro, increased Src activation leads to the phosphorylation of FUNDC1 and the inhibition of mitophagy. In line with the findings, inhibition of Src activity protected podocytes from mitochondrial damage and silencing FUNDC1 counteracted the effects of PP2, suggesting that FUNDC1-mediated mitophagy is a downstream pathway regulated by Src (157). These findings indicate that Src plays a pivotal role in diabetic renal damage by suppressing FUNDC1-mediated mitophagy, thereby contributing to the development of DKD.

Numerous studies have indicated the significant involvement of long noncoding RNAs (lncRNAs) in the control of podocyte apoptosis in DKD (158, 159). In glomeruli and podocytes of diabetic mice and podocytes treated with high glucose, the expression of lncRNA SNHG17 and mammalian sterile 20-like kinase 1 (Mst1) was elevated, whereas Parkin expression was diminished. Overexpression of lncRNA SNHG17 hindered mitophagy and triggered podocyte apoptosis, whereas suppressing lncRNA SNHG17 promoted mitophagy and decreased podocyte apoptosis. In addition, lncRNA SNHG17 interacts with Mst1, influencing the degradation of Mst1. Furthermore, lncRNA SNHG17 was identified as a regulator of Parkin expression through Mst1. Mechanistically, lncRNA SNHG17 regulates Parkin-dependent mitophagy and podocyte apoptosis by regulating Mst1. Ultimately, the in vivo silencing of lncRNA SNHG17 enhanced mitophagy and alleviated DKD progression (130).

Work from both our group and others suggests that altered lipid metabolism in diabetes may be one of the major risk factors for the development and progression of DKD (160–162). Notably, exposure to palmitic acid (PA) results in an upregulation of PINK1 and Parkin expression in podocytes, indicating that mitophagy can be activated in response to lipotoxicity. This observation was further confirmed in a rat model of HFD-induced obesity. Furthermore, inhibition of mitophagy by silencing Parkin in podocytes exacerbated PA-induced mitochondrial impairment, mitochondrial reactive oxygen species (mitoROS) production, and podocyte apoptosis (131). These findings suggest that PINK1/Parkin-mediated mitophagy plays a protective role against lipotoxicity-induced podocyte injury in the context of DKD.

Mitophagy has also been shown to play a protective role in LN (132). A recent study reported a negative correlation between nestin, a cytoskeletal protein expressed in podocytes, and proteinuria in patients with lupus nephritis and in MRL/lpr lupus-prone mice and a positive correlation between nestin and nephrin. Interestingly, the knockdown of nestin decreased the expression of mitophagy-associated proteins in podocytes of MRL/lpr mice and induced mitochondrial dysfunction in podocytes stimulated with LN plasma. Notably, reducing mitophagy or the generation of reactive oxygen species led to a significant decrease in both the expression and phosphorylation of nephrin (132). These findings suggest that nestin regulates the expression of nephrin through mitophagy, protecting podocytes from injury in LN. Taken together, understanding and targeting mitophagy pathways in podocytes could provide practical therapeutic strategies to slow down the progression of kidney diseases, emphasizing the need for further investigation in this specific area.

Lipophagy.

Increasing evidence suggests a role for lipid accumulation in podocytes in the pathogenesis of proteinuric kidney diseases of metabolic and nonmetabolic origin (161, 163–167). A recent study revealed a decrease in lipophagy and an increase in ectopic lipid deposition (ELD) and lipotoxicity in the tubular cells of individuals with DKD. These changes were accompanied by a reduction in the expression of adiponectin receptor 1 (AdipoR1) and phosphorylated AMPK. Similar outcomes were replicated in db/db mice, and the observed alterations were effectively reversed by AdipoRon, an activator of adiponectin receptors known to promote autophagy. Moreover, the study uncovered a significant reduction in lipophagy levels in a human proximal tubular cell line (HK-2 cells) subjected to high-glucose treatment. This reduction correlated with an increase in lipid deposition, apoptosis, and fibrosis, which was partially alleviated by the administration of AdipoRon. However, the renoprotective effects of AdipoRon were abolished when cells were pretreated with the ULK1 inhibitor SBI-0206965 and the autophagy inhibitor chloroquine and amplified by the AMPK activator AICAR (168). This study highlights the significance of lipophagy in CKD progression. However, studies investigating the role of lipophagy in podocytes are sparse and further research into lipophagy mechanisms in podocytes is essential to identify new targets and develop potential therapeutic approaches for proteinuric kidney diseases.

Chaperone-Mediated Autophagy

Approximately 30% of kidney proteins contain a KFERQ motif, including glycolytic enzymes, pyruvate kinase, α-microglobulin, transcription factors such as Pax2, and other substrates, all of which have demonstrated relevance to kidney diseases (169). The involvement of CMA in kidney diseases is context-dependent and exhibits a reciprocal relationship with macroautophagy. In cases where CMA is compromised, there is an upregulation of macroautophagy, though complete compensation is not achieved. Consequently, cells with defective CMA are more susceptible to various stressors. Furthermore, this compensatory response tends to diminish under increased stress or advancing age (170). Of note, podocytes exhibit a high basal level of macroautophagy, while CMA predominates in the tubular system (7). Hence, in glomerular diseases, where podocyte loss is a predominant factor, the investigation of CMA has been relatively underexplored. CMA is activated during starvation, whereas in conditions related to renal growth, CMA is downregulated. In vitro, various growth factors have been demonstrated to inhibit CMA activity. For instance, in renal tubular cells, epidermal growth factor (EGF) extended the half-life of proteins targeted for CMA by 30% (171). The inhibition of proteolysis by EGF is associated with EGF-receptor tyrosine autophosphorylation, activating Ras, and subsequently activating class-I PI3K (172). These findings suggest that CMA could be downregulated in kidney diseases associated with renal hypertrophy. Indeed, levels of intracellular proteins containing the CMA-targeting motif have been shown to be significantly increased in kidneys of STZ-induced diabetic mouse model, whereas the levels of LAMP2A and HSC70 were markedly decreased, indicating that reduced CMA, rather than increased protein synthesis, may contribute to renal hypertrophy in diabetes mellitus (173). In the context of kidney aging, CMA disruptions are particularly significant, contributing to the onset of CKD (174). Studies indicate a decline in the levels of LAMP2A with aging, and experiments involving the overexpression of LAMP2A have demonstrated the potential to mitigate the age-related reduction in CMA, thereby alleviating the accumulation of damaged proteins and enhancing organ function (175). Furthermore, HFD or obesity has been shown to negatively regulate CMA, contributing to the progression of obesity-related CKD (176). Although macroautophagy is the predominant form of autophagy in glomerular podocytes under basal conditions, understanding the role of CMA in podocyte function and its dysregulation in kidney diseases could provide valuable insights for developing targeted therapies to preserve or restore kidney homeostasis.

CONCLUSIONS AND PERSPECTIVES

In conclusion, substantial evidence supports the crucial role of autophagy in both podocyte physiology and pathology. The beneficial effect of pharmacological activation of autophagy has been consistently observed in various experimental models of kidney diseases of metabolic and nonmetabolic origin. The expanding array of compounds known to activate autophagy comprises both mTOR inhibitors and mTOR-independent agents (177–180). mTOR inhibitors, such as rapamycin and rapamycin analogs, are commonly used as immunosuppressive agents in clinical settings (180). mTOR-independent autophagy inducers, including AMPK and SIRT1 activators, have been proven to activate autophagy in vivo and attenuate renal injury in animal models of CKD (181). However, despite significant progress in understanding autophagy in podocyte injury, the clinical translation of autophagy modulators remains challenging. Autophagy modulators require prospective and randomized trials to assess their therapeutic potentials and adverse events when administered in humans. In summary, more comprehensive research is needed to bridge the gap between experimental findings and clinical application in the field of autophagy modulation for kidney diseases.

GRANTS

R.N. is supported by National Institutes of Health Grant R56DK10475. A.F. and S.M. are supported by National Institutes of Health Grants R01DK136679 and R56DK104753 and by Aurinia Pharmaceuticals and Pfizer Inc. A.F. is supported by National Institutes of Health Grants UE5DK137308, U54DK083912, U01DK100846, U01DK116101, and UM1TR004556 (Miami Clinical Translational Science Institute) and K12TR004555.

DISCLOSURES

A.F. and S.M. are inventors on pending (PCT/US2019/032215; US 17/057,247; PCT/US2019/041730; PCT/US2013/036484; US 17/259,883; US17/259,883; JP501309/2021, EU19834217.2; CN-201980060078.3; CA2,930,119; CA3,012,773; CA2,852,904) or issued patents (US10.183.038, US10.052.345) aimed at diagnosing or treating proteinuric kidney diseases and therefore stand to gain royalties from their future commercialization. A.F. is Chief Scientific Officer of L&F Health LLC, holds equity interests in L&F Research, and is the inventor of assets developed by ZyVersa Therapeutics. ZyVersa has licensed worldwide rights to develop and commercialize hydroxypropyl-β-cyclodextrin for the treatment of kidney disease from L&F Research. A.F. also holds equity in River 3 Renal Corporation. S.M. holds equity interest in L&F Research. R.N. declares no competing interests.

AUTHOR CONTRIBUTIONS

R.N. prepared figures; R.N. drafted manuscript; R.N., S.M., and A.F. edited and revised manuscript; R.N., S.M., and A.F. approved final version of manuscript.