In recent years, a large body of evidence has indicated that inflammation, OS, and apoptosis are the primary causes of diabetic kidney injury. The lncRNA/circRNA–miRNA–mRNA network plays a central role in promoting the development of diabetic nephropathy by regulating inflammation, OS, and apoptosis in podocytes ( Figure 2 , Table 1 ).

Clinical trials have documented a marked elevation of lncRNA-DLX6-AS1 in renal biopsies from patients with DN, demonstrating a quantitative correlation between transcript abundance and urinary albumin–creatinine ratio (uACR) progression, thereby implicating DLX6-AS1 in DN progression (83). Further animal studies demonstrated that markedly elevated DLX6-AS1 expression in db/db murine models was associated with reduced podocyte integrity markers (podocin, WT-1, and SYNPO) and concomitant increases in the pro-inflammatory cytokine IL-17. These data mechanistically link DLX6-AS1 overexpression to podocyte injury, inflammatory exacerbation, and proteinuria, whereas DLX6-AS1 knockout (KO) substantially attenuates these pathological manifestations (83). Mechanistically, miR-346-5p directly targets the 3′ untranslated region (UTR) of GSK-3β, whereas DLX6-AS1 competitively sequesters miR-346-5p, thereby derepressing GSK-3β expression. This regulatory axis drives podocyte dysfunction and inflammatory cascades, thereby accelerating the pathogenesis of DN. Consequently, therapeutic strategies targeting DLX6-AS1 suppression or modulation of its downstream signaling pathways may be promising for alleviating DN pathology (83).

Additional studies revealed significantly elevated expression of lncRNA-SNHG5 and TRPC6 in the kidney tissues of db/db mice and HG-treated podocytes, concomitant with marked reductions in miR-26a-5p levels and the podocyte-specific markers, nephrin and podocin. Functional studies demonstrated that miR-26a-5p overexpression attenuated TRPC6 expression and mitigated podocyte apoptosis and inflammation, whereas KO-SNHG5 upregulated miR-26a-5p, suppressed TRPC6, and significantly ameliorated proteinuria, restored podocyte ultrastructure, and alleviated HG-induced podocyte injury in db/db mice (84). These findings indicate that SNHG5 acts as a molecular sponge for miR-26a-5p, thereby derepressing TRPC6 expression and driving podocyte dysfunction and DN progression. In addition, lncRNA-MIAT expression was markedly upregulated in HG-stimulated podocytes, whereas miR-130a-3p expression was suppressed, MIAT silencing substantially reduces HG-induced TNF-α, IL-6, and IL-1β secretion; downregulates pro-apoptotic markers, Bax and caspase-3; and upregulates anti-apoptotic marker, Bcl-2, thereby promoting podocyte survival under HG conditions (85). Mechanistically, KO-MIAT elevates miR-130a-3p expression, which subsequently inhibits TLR4 signaling and attenuates HG-mediated inflammation and apoptosis in podocytes (85). Therefore, MIAT functions as a sponge for miR-130a-3p and regulates TLR4 expression by competitively binding to miR-130a-3p. Inhibition of MIAT or modulation of its downstream pathway could help attenuate the pathological changes in DN.

Clinical investigations revealed significant downregulation of miR-16-5p and a concurrent rise of lncRNA-TCF7 and SEMA3A in serum samples from patients with DN, with these aberrant expression profiles demonstrating positive correlations with hyperglycemic indices (blood glucose and glycated hemoglobin) and uACR (86). Experimental studies have further shown that HG conditions elevate TCF7 and SEMA3A expression, while suppressing miR-16-5p, but also increase Bax, ROS, MDA, IL-1β, TNF-α, and IL-6. Concurrently, HG exposure reduced anti-apoptotic Bcl-2 expression and antioxidant enzyme activity (SOD and CAT), culminating in podocyte apoptosis, OS exacerbation, inflammatory activation, and diminished cellular viability. The genetic silencing of TCF7 or SEMA3A reversed these pathological alterations (86). Notably, TCF7 acts as a negative regulator of miR-16-5p and a positive modulator of SEMA3A, the latter being a direct miR-16-5p target. miR-16-5p directly binds to the 3′ UTR of SEMA3A to suppress its expression, establishing a regulatory triad. Collectively, these findings indicated that lncRNA-TCF7 functions as a ceRNA, sequestering miR-16-5p and derepressing SEMA3A expression (86). This axis drives HG-induced podocyte injury by amplifying OS, apoptosis, and inflammatory cascades, thereby positioning the TCF7/miR-16-5p/SEMA3A axis as a potential therapeutic target in DN.

Xia et al. showed that lncRNA-1500026H17Rik and EGR1 were significantly elevated, miR-205-5p was significantly reduced in the kidney tissues and HG-induced podocytes of DN mice, and 1500026H17-Rik was positively correlated with EGR1 and negatively correlated with miR-205-5p (87). KO-1500026H17Rik significantly downregulated Bax; cleaved caspase-3, α-SMA, FN, Col I, MDA, ROS, IL-6, IL-1β, and TNF-α expression; and upregulated Bcl-2, Podocin, and SOD. It also attenuates apoptosis, fibrosis of HG-induced podocytes, OS, and inflammatory responses, whereas inhibition of miR-205-5p or overexpression of EGR1 reversed the protective effects of KO-1500026H17Rik. In addition, 1500026H17Rik acts as a ‘sponge’ for miR-205-5p and inhibits miR-205-5p activity, whereas EGR1 is a downstream target of miR-205-5p and can be directly targeted by it to inhibit EGR1 expression (87). It has been shown that lncRNA-1500026H17Rik, by adsorbing miR-205-5p, deregulates the inhibition of EGR1, which in turn exacerbates podocyte injury.

Emerging evidence has indicated that although certain lncRNA–miRNA networks drive DN progression, others may exert protective effects. For instance, studies have demonstrated that HG conditions significantly impair podocyte survival and enhance apoptotic rates, concomitant with the downregulation of lncRNA-XIST expression. Notably, XIST overexpression not only attenuated HG-induced podocyte apoptosis and improved cellular viability but also suppressed the pro-apoptotic mediators such as cytochrome c, Bax, and caspase-3, while upregulating anti-apoptotic Bcl-2 expression (88). Mechanistically, miR-30 overexpression suppressed XIST, thereby abolishing its cytoprotective effects and promoting apoptosis, whereas miR-30 inhibition restored XIST-mediated podocyte protection. Furthermore, XIST overexpression rescues AVEN expression, a direct miR-30 target, by establishing a regulatory axis central to this pathway (88). These findings collectively suggest that lncRNA-XIST is a molecular sponge for miR-30 that modulates AVEN expression to govern podocyte survival and apoptosis. This axis highlights the potential therapeutic relevance of XIST in DN and offers novel targets for intervention.

In addition to the lncRNA–miRNA–mRNA network, the circRNA–miRNA–mRNA network had the same function ( Figure 2 , Table 1 ). It has been reported that that hsa_circ_00162 was upregulated in the peripheral blood and HG-induced podocytes of patients with DN. KO-hsa_circ_00162 suppressed MMP9 expression; downregulated IL-6, TNF-α, and IL-1β expression; restored nephrin and synaptopodin expression; inhibited HG-induced podocyte apoptosis; attenuated the inflammatory response; and alleviated podocyte injury (89). In addition, miR-149-5p inhibition reversed the inhibitory effect of KO-hsa_circ_0001162 on MMP9, which in turn upregulated MMP9 expression by adsorbing miR-149-5p to deregulate its inhibition of MMP9 (89). Thus, hsa_circ_0001162 has been reported to promote DN podocyte injury by regulating MMP9 expression through a ceRNA mechanism, which lays the groundwork for the molecular diagnosis and targeted therapy of DN.

Circ_Supr3 and G3BP2 were significantly upregulated and miR-185-5p was significantly downregulated in the kidney tissues of DM mice and HG-treated MPC5 cells in a previous study. KO-circ_Supr3 upregulated miR-185-5p expression and thus inhibited G3BP2; reversed HG-induced cell proliferation; and downregulated EdU, PCNA, Bax, Cleaved-PARP, TNF-α, IL-6, and IL-1β expression; increased Bcl-2 expression; and inhibited apoptosis and inflammatory responses, whereas G3BP2 overexpression counteracted the protective effect of miR-185-5p on HG-induced injury, and miR-185-5p inhibitor in turn reversed the protective effect of KO-circ_Supr3 on cell injury (90). Circ_Supr3 exacerbates HG-induced podocyte injury by adsorbing miR-185-5p to upregulate G3BP2 expression. Similarly, in the kidney tissues of the STZ-induced DN mouse model and HG-treated podocytes, circ_0000285 expression was significantly elevated and miR-654-3p expression was significantly reduced, overexpression of circ_0000285 inhibited podocyte proliferation, promoted podocyte apoptosis, and blocked the cell cycle. KO-circ_0000285 inhibited the release of IL -6, IL-1β, and TNF-α, whereas inhibition of miR-654-3p reversed the protective effect of KO-circ_0000285 on podocyte injury (91). In addition, MAPK6 is a direct target of miR-654-3p, and circ_0000285 activates the MAPK6 signaling pathway through the adsorption of miR-654-3p, leading to podocyte injury, an inflammatory response, and promotion of DN progression (91).

Research has identified significantly elevated expression levels of circ_GAB1 and MAPK6 in the serum of both patients with DN and HG-treated podocytes, concomitant with a marked reduction in miR-346 expression (92). Silencing circ_GAB1 suppressed MAPK6 expression, resulting in decreased levels of IL-6, TNF-α, ROS, MDA, cleaved caspase-3, and Bax within podocytes, conversely, this intervention elevated SOD, PCNA, and Bcl-2 expression, thereby mitigating HG-induced apoptosis, inflammatory responses, and OS, while promoting cellular proliferation and attenuating podocyte injury, notably, the administration of miR-346 inhibitors reversed the cytoprotective effects of KO-circGAB1. Furthermore, both miR-346 overexpression and MAPK6 suppression alleviated podocyte damage (92). These findings collectively indicated that circGAB1 acts as a molecular sponge for miR-346, thereby relieving its inhibitory effect on MAPK6 and subsequently activating the MAPK6-mediated pathways of podocyte apoptosis, inflammation, and OS mechanisms that ultimately drive DN progression.

In addition, Fang et al. demonstrated that sera from patients with DN and HG-induced podocytes exhibited significant upregulation of KLF9 and hsa_circ_0037128, along with marked downregulation of miR-31-5p. A negative correlation was observed between hsa_circ_0037128 and miR-31-5p, whereas a positive correlation existed with KLF9, KO-hsa_circ_0037128 reduces Bax and caspase-3 levels and increases Bcl-2 expression in HG-stimulated podocytes, this intervention suppressed HG-induced podocyte apoptosis as well as decreased inflammatory mediators such as TNF-α, IL-1β, and IL-6, reduced OS markers (ROS and MDA), and enhanced SOD activity, thereby collectively ameliorating podocyte injury (93). This study further revealed that miR-31-5p is a direct target of hsa_circ_0037128 and that KLF9 functions as a downstream target of miR-31-5p, notably, miR-31-5p inhibition or KLF9 overexpression counteracted the protective effects of KO-hsa_circ_0037128 in podocytes (93). These findings collectively indicate that hsa_circ_0037128 acts as a competitive endogenous RNA by sequestering miR-31-5p, thereby relieving the miR-31-5p-mediated suppression of KLF9. Subsequent KLF9 upregulation exacerbated HG-induced inflammatory responses, OS, and podocyte apoptosis.

Although quantitative studies have questioned the physiological effects of ceRNA interactions under typical conditions, the above-mentioned ceRNA network studies were conducted using in vivo and in vitro models that were functionally validated and supported by loss-of-function and gain-of-function experiments.

In addition to its regulatory effects on podocytes in DN progression through cell death pathways, the lncRNA–miRNA–mRNA network also modulated TECs to influence disease progression ( Figure 3 , Table 1 ). Investigations revealed a dose-dependent upregulation of lncRNA-XIST and HMGA2 in the renal tissues of patients with DN and HG-treated renal tubular HK-2 cells. Silencing of XIST or HMGA2 inhibited HG-induced cellular proliferation and enhanced apoptosis. Dual-luciferase reporter assays further demonstrated direct binding between XIST and miR-423-5p, with miR-423-5p targeting the 3′ UTR of HMGA2, KO-XIST upregulated miR-423-5p expression, consequently reducing HMGA2 levels—an effect partially reversed by miR-423-5p inhibition (94). Furthermore, miR-423-5p overexpression suppresses HMGA2 expression, attenuates cellular proliferation, and promotes apoptosis. These effects were counteracted by HMGA2 overexpression (94). Collectively, these findings indicate that lncRNA-XIST functions as a molecular sponge to sequester miR-423-5p, thereby derepressing HMGA2 expression. This regulatory axis promotes HG-driven renal cell proliferation, while suppressing apoptotic processes during DN pathogenesis.

The lncRNA–miRNA–mRNA network exhibits dual regulatory capacities in DN, demonstrating both promotive and suppressive effects on disease progression. Studies revealed that HG treatment significantly increased apoptotic rates and upregulated eERS-associated proteins GRP78, caspase-12, CHOP, p-PERK, p-eIF2α in HK-2 cells, concurrent with lncRNA-TUG1 downregulation, TUG1 overexpression suppressed miR-29c-3p expression, markedly attenuated HG-induced apoptosis, and reduced ERS-related protein levels (95). Notably, miR-29c-3p overexpression exacerbated HG-driven apoptosis and ERS protein expression, counteracting TUG1-mediated protection; this effect was reversed by miR-29c-3p inhibition. Mechanistic investigations have identified SIRT1 as a direct target of miR-29c-3p, whose expression is maintained through TUG1-dependent miR-29c-3p suppression (95). Collectively, these findings demonstrate that lncRNA-TUG1 functions as a molecular sponge to inhibit miR-29c-3p activity, thereby preserving SIRT1 expression and mitigating ERS-mediated cellular injury.

Tubular epithelial cell dysfunction is a pivotal mechanism in DN pathogenesis, in which inflammatory–apoptotic/pyroptotic signaling cascades contribute to dysregulated tubular reabsorptive processes and fibrotic changes in the kidney, leading to pathological advancement ( Figure 3 , Table 1 ). Emerging evidence highlights the critical regulatory role of ncRNAs in the modulation of these pathological processes. Studies have identified the significant upregulation of lncRNA-SNHG5 and concomitant downregulation of miR-26a-5p in the sera from patients with DN and HG-treated HK-2 cells, SNHG5 exhibited negative correlations with miR-26a-5p, eGFR, and albumin levels, whereas positive associations with fasting blood glucose, IL-6, TNF-α, and proteinuria. SNHG5 inhibition reversed HG-induced reductions in cell viability, attenuated apoptotic rates, and normalized IL-6, TNF-α, and ROS levels (84). Mechanistic investigations confirmed that miR-26a-5p is a direct target of SNHG5. SNHG5 silencing upregulates miR-26a-5p expression and ameliorates HG-induced tubular injury. Collectively, these findings demonstrate that lncRNA-SNHG5 exacerbates HG-mediated tubular damage through miR-26a-5p sequestration (84). Targeted SNHG5 suppression mitigates inflammatory responses, OS, and apoptotic signaling, thereby identifying novel therapeutic targets for DN.

Furthermore, Liu et al. demonstrated that HG treatment significantly upregulated lncRNA-MALAT1 alongside NLRP3, caspase-1, IL-1β, and IL-18 expression, while suppressing miR-30c levels in HK-2 cells. This dysregulation elevates pyroptotic rates and lactate dehydrogenase (LDH) release, MALAT1 silencing markedly attenuated HG-induced pyroptosis and reduced NLRP3 inflammasome components such as caspase-1, IL-1β, and IL-18, whilst restoring miR-30c expression—suggesting that KO-MALAT1 suppresses hyperglycemia-induced pyroptosis through miR-30c upregulation and NLRP3 downregulation (96). Dual-luciferase reporter assays confirmed direct binding between MALAT1 and miR-30c, with miR-30c targeting the 3′ UTR of NLRP3. miR-30c overexpression suppresses NLRP3 expression and ameliorates pyroptosis and inflammatory responses. Critically, co-transfection of MALAT1 with miR-30c inhibitors reversed the anti-pyroptotic effects of MALAT1 silencing (96). These findings establish that lncRNA-MALAT1 exacerbates inflammatory pyroptosis by sequestering miR-30c, thereby derepressing NLRP3 expression. Targeted disruption of the MALAT1/miR-30c/NLRP3 axis may provide novel therapeutic opportunities for the management of DN.

Further investigation revealed significant downregulation of lncRNA-UCA1 and concomitant upregulation of miR-206 in the renal tubular epithelial tissues of DN rats and HG-treated HK-2 cells, with an inverse correlation between miR-206 and UCA1 expression. UCA1 overexpression markedly reduced protein levels of pyroptosis-associated markers caspase-1, IL-1β, and NLRP3 under HG conditions, thereby alleviating apoptotic cell death and inflammatory responses. Conversely, KO-UCA1 exacerbated these pathological manifestations (97). Mechanistic validation using dual-luciferase reporter assays identified binding complementarity between UCA1 and miR-206. miR-206 mimics counteract UCA1 overexpression-mediated suppression of apoptosis, whereas miR-206 inhibition partially rescues the pro-apoptotic effects induced by UCA1 silencing (97). These findings demonstrated that lncRNA-UCA1 exerts cytoprotective effects by directly binding to and suppressing miR-206 activity.

The circRNA–miRNA–mRNA network regulates DN progression through the apoptosis–OS pathway, demonstrating dual therapeutic potential in both podocytes and TEC ( Figure 3 , Table 1 ). Experimental evidence revealed significant upregulation of circ_0003928 and HDAC4, along with marked downregulation of miR-506-3p in patients with DN and in HG-treated models. In HG-stimulated HK-2 cells, the characteristic pathological manifestations include reduced cell viability, elevated ROS and MDA levels, diminished SOD activity, upregulation of Bax and caspase-3 expression, and downregulation of Bcl-2 expression (98). Notably, KO-circ_0003928 ameliorated HG-induced reduction in cell viability and proliferation, attenuated ROS and MDA generation, restored SOD activity, and reduced apoptosis. Mechanistic studies further established that circ_0003928 silencing upregulates miR-506-3p expression, whereas miR-506-5p overexpression significantly suppresses HDAC4 expression at both the transcriptional and translational levels (98). Collectively, these findings demonstrated that circ_0003928 exacerbates OS and apoptosis by competitively sequestering miR-506-3p, thereby relieving its inhibitory effect on HDAC4. This novel mechanistic insight delineates the role of the circ_0003928/miR-506-3p/HDAC4 regulatory axis in pathogenesis of DN.

Wang et al. demonstrated that circ_0000064 is significantly upregulated in both serum samples from patients with DN and HG-stimulated HK-2 cells, concomitant with the downregulation of miR-532-3p and that circ_0000064 expression is positively correlated with disease severity (albuminuria levels) (99). Experimental interventions showed that either circ_0000064 silencing or miR-532-3p overexpression attenuated HG-induced pathological manifestations by reducing MDA activity; downregulating pro-apoptotic Bax, caspase-3, pro-fibrotic α-SMA, and Col-I markers; elevating SOD and anti-apoptotic Bcl-2 levels; mitigating OS, apoptosis, and fibrosis, while enhancing cellular proliferation. Notably, these protective effects were reversed by miR-532-3p inhibitors or ROCK1 overexpression, confirming the functional interplay within this regulatory axis (99). Mechanistic analyses revealed that under HG conditions, circ_0000064 acts as a competitive endogenous RNA by sponging miR-532-3p, thereby relieving its post-transcriptional repression of ROCK1 and leading to ROCK1 upregulation, which subsequently activates OS-related pathways, apoptotic cascades, and fibrotic signaling, collectively exacerbating tubular epithelial injury and DN progression. This study delineated the pathomechanistic landscape of circRNA-driven disease progression, establishing the circ_0000064/miR-532-3p/ROCK1 molecular interplay as a druggable pathway for DN intervention through multimodal regulatory network disruption (99).

In conclusion, the lncRNA/circRNA–miRNA–mRNA network exerts significant regulatory effects on podocytes and TECs through cell death, inflammation, OS, and apoptosis, with the cell death pathway serving as the common terminal pathway for ceRNA action. Regardless of their specific molecular mechanisms, all ceRNA networks studied ultimately converged directly to exacerbate or alleviate inflammatory responses, OS, and apoptosis in podocytes and TECs. These three processes constitute an inseparable ‘trinity’ of terminal effect axes through which ceRNA networks cause podocyte and TECs damage and DN progression, providing new insights into the molecular mechanisms and therapeutic strategies for DN. However, although the aforementioned ceRNA networks have been functionally validated and their interactions have been supported by loss-of-function/gain-of-function experiments, the roles of the lncRNA-TUG1/miR-29c-3p/SIRT1 and lncRNA-MALAT1/miR-30c/NLRP3 axes have only been validated in vitro and lack in vivo experiments. Therefore, their robustness and roles need to be further verified.

As a pivotal pathological process in DN, fibrosis is a critical therapeutic target for preventing disease progression (110). In the sera of both patients with DN and HG-treated human glomerular MCs (HGMCs), elevated expression of lncRNA-HCP5 and HMGA2 was observed, along with reduced miR-93-5p levels. HCP5 silencing inhibited HG-induced HGMC proliferation as well as downregulated Col-I, Col-IV, FN, TNF-α, IL-6, and IL-1β, while simultaneously upregulating apoptosis indicators cleaved-PARP and cleaved-caspase3. This dual modulation attenuates fibrotic and inflammatory responses while promoting apoptosis (111). Furthermore, HG stimulation markedly enhanced p-AKT and p-mTOR levels, which were reversed by KO-HCP5 treatment. Notably, miR-93-5p inhibition restored p-AKT/p-mTOR expression in HG-challenged HGMCs (111). Mechanistic investigations revealed that miR-93-5p was a direct target of HCP5, with HMGA2 functioning downstream of miR-93-5p. miR-93-5p overexpression significantly suppressed HMGA2 expression, mirroring the anti-fibrotic effects of HCP5 silencing. Conversely, miR-93-5p inhibition or HMGA2 overexpression reverses these protective effects (111).

The activation of lncRNA–miRNA networks exhibits dual regulatory effects on GMCs proliferation, demonstrating both promotive and inhibitory capacities. For instance, studies have revealed significantly lower lncRNA-GAS5 expression in the renal tissues of type 2 diabetic patients with DN than in those without DN, showing negative correlations with DN severity markers (proteinuria and eGFR) (112). In DN rat models and HG-treated GMCs, GAS5 downregulation coincided with pathological manifestations including mesangial hyperplasia, enhanced collagen deposition, and elevated expression of fibrotic/inflammatory mediators (FN, Col-IV, and TGF-β1), GAS5 overexpression upregulated SIRT1 expression as well as normalized cell cycle progression in GMCs, concurrently suppressing apoptotic proteins p53 and p21. Conversely, KO-GAS5 exacerbated these pathological parameters (112). Mechanistic investigations have demonstrated that GAS5 directly sequesters miR-221, which normally suppresses its target gene, SIRT1. miR-221 inhibition reduced MC proliferation and expression of fibrotic markers, which was reversed by SIRT1 silencing (112).

To sum up, the lncRNA-HCP5/miR-93-5p/HMGA2 axis promotes DN progression through the activation of the AKT/mTOR pathway, driving inflammatory, apoptotic, and fibrotic cascades. By contrast, the lncRNA-GAS5/miR-221/SIRT1 axis counteracted these pathological processes by suppressing mesangial cell proliferation. This dual regulatory paradigm provides novel strategic insights for precision therapeutics in DN management, highlighting the therapeutic potential of targeting specific lncRNA–miRNA networks based on the disease stage and molecular profiles ( Figure 5 , Table 3 ).

Emerging evidence has established apoptosis and inflammation as central pathogenic mechanisms in DN that play pivotal roles in disease acceleration and clinical deterioration. Recent advances have highlighted the ubiquitous involvement of ncRNAs in virtually all pathophysiological processes, including apoptotic regulation and cellular proliferation, that are directly relevant to DN progression (119, 120). Studies have revealed downregulation of circ_AKT3 and E-cadherin in the renal tissues of db/db mice and HG-treated GMCs, concomitant with miR-296-3p upregulation. A positive correlation was observed between circ_AKT3 and E-cadherin, whereas a negative correlation was observed between circ_AKT3 and miR-296-3p. Circ_AKT3 overexpression reduces the protein expression of caspase-3 and the Bax/Bcl-2 ratio, attenuates HG-induced apoptosis, and suppresses fibrotic protein accumulation in FN, Col-I, and Col-IV (17). Mechanistic studies showed that circ_AKT3 directly binds to miR-296-3p through complementary sequences, thereby effectively sequestering its activity. miR-296-3p in turn suppresses E-cadherin expression by targeting its 3′ UTR; therefore, E-cadherin overexpression reversed miR-296-3p-mediated enhancement of ECM deposition (17).

Distinct from apoptosis, pyroptosis is a novel pro-inflammatory apoptotic mechanism mediated by caspase-1 activation following NLRP3 inflammasome assembly, driving pyroptotic execution and inflammatory cytokine release (121, 122). Emerging evidence has implicated aberrant pyroptosis in DN pathogenesis, with ncRNAs critically regulating NLRP3 activation and pyroptosis pathways (4). For instance, studies identified significant upregulation of circ_LARP1B and TLR4 alongside downregulation of miR-578 in sera from patients with DN and in HG-stimulated renal GMCs. Overexpression of circ_LARP1B suppressed cellular proliferation, induced G0–G1 phase cell cycle arrest, and upregulated the expression of IL-6, TNF-α, IL-1β, and IL-18, as well as the pyroptosis-associated markers NLRP3 and caspase-1, thereby exacerbating pyroptosis and inflammatory pathology. Conversely, circ_LARP1B silencing promotes S-phase progression, attenuates pyroptotic cell death, and reduces inflammatory cytokine secretion (113). Mechanistically, TLR4 activation stimulated NLRP3 inflammasome formation and caspase-1 expression, thereby amplifying pyroptotic signaling. miR-578 directly suppresses TLR4 expression, with TLR4 overexpression negating the anti-inflammatory and anti-pyroptotic effects of miR-578; therefore, circ_LARP1B functions as a molecular sponge sequestering miR-578, thereby relieving miR-578-mediated TLR4 suppression. This cascade activates the TLR4/NLRP3/caspase-1 axis, ultimately accelerating the progression (113).

Collectively, circ_AKT3 deregulated the inhibition of E-cadherin by miR-296-3p through the adsorption of miR-296-3p, reducing the rate of HG-induced apoptosis and fibrosis-associated protein expression, whereas circ_LARP1B deregulated the inhibitory effect of miR-578 on TLR4 through the adsorption of miR-578, which in turn activated the TLR4/NLRP3/caspase-1 axis, promoting DN progression. The circRNA–miRNA–mRNA network plays a critical role in HG-induced apoptosis, pyroptosis, and fibrosis in renal GMCs ( Figure 5 , Table 3 ).

Emerging evidence has identified significant upregulation of circ_0060077, circ_0068087, circ_TAOK1, VASN, ROCK2, and SOX6 in peripheral blood samples from patients with DN and HG-treated HK-2 cells. KO-circ_0060077, circ_0068087, or circ_TAOK1 consistently showed therapeutic efficacy by enhancing cellular proliferation; downregulating pro-apoptotic markers (Bax and cleaved caspase-3); reducing ROS, MDA, IL-6, TNF-α, IL-1β, FN, Col-I, and Col-IV; elevating anti-apoptotic Bcl-2 expression and SOD activity; and concurrently attenuating apoptosis, OS, inflammation, and fibrosis, thereby collectively ameliorating HG-induced HK-2 cell injury (114–116). Furthermore, miR-145-5p, miR-106a-5p, and miR-142-3p inhibited VASN, ROCK2, and SOX6 expression by binding to their respective 3′ UTRs. Inhibition of these miRNAs or overexpression of VASN, ROCK2, and SOX6 reversed the protective effects of circ_0060077, circ_0068087, and KO-circ_TAOK1 (114–116).

Emerging studies have identified concomitant upregulation of circ_0003928, hsa_circ_0003928, MAPK6, and Anxa2 with downregulation of miR-31-5p and miR-151-3p in both patients with DN and HG-treated HK-2 cells, whereas KO-circ_0003928 and hsa_circ_0003928 upregulated miR-31-5p and miR-151-3p expression, respectively, both increased Bcl-2 expression and decreased Bax, TNF-α, IL-6, IL-1β, and ROS expression (117, 118). In addition, KO-circ_0003928 also downregulated FN, TGF-β1, and MDA expression and increased SOD activity (117), whereas KO-hsa_circ_0003928 decreased HG-induced cell viability and downregulated cleaved-caspase-3 expression (118), but both KOs reduced apoptosis, inhibited OS, and inflammatory responses (117, 118). Furthermore, circ_0003928 directly targeted miR-31-5p, and miR-31-5p overexpression inhibited the AKT signaling pathway by inhibiting MAPK6 and thus the AKT signaling pathway; overexpression of MAPK6 in turn reversed the protective effect of miR-31-5p (117), whereas inhibition of miR-151-3p reversed the KO-hsa_circ_0003928 effect, and overexpression of miR-151-3p significantly inhibited mRNA and protein expression of Anxa2 (118).

As shown, circ_0060077, circ_0068087, and circ_TAOK1 promoted the expression of VASN, ROCK2, and SOX6 by sequestering miR-145-5p, miR-106a-5p, and miR-142-3p, respectively, thereby exacerbating HG-induced renal tubular cell injury. Circ_0003928 deregulated the inhibition of MAPK6 by the adsorption of miR-31-5p, which in turn activated the AKT signaling pathway, whereas hsa_circ_0003928 deregulated the inhibition of Anxa2 by the adsorption of miR-151-3p, which led to HG-induced apoptosis, fibrosis, inflammation, and OS ( Figure 5 , Table 3 ).

In summary, cell death (such as apoptosis and pyroptosis) is a key pathological factor leading to podocyte damage, whereas fibrosis caused by inflammation and OS is the core driving factor of mesangial cell damage. Furthermore, the pathological changes caused by inflammation, OS, apoptosis, and fibrosis are the key factors in tubular epithelial cell damage, indicating that the terminal pathology of renal cells (podocytes, GMCs, and TECs) are closely coupled, and the key to regulating DN progression is to block this pathological process (such as inflammation and apoptosis/fibrosis). The ceRNA network regulates DN progression by acting on its key downstream molecules to activate or inhibit inflammation, OS, the inflammation–OS axis, cell death, and fibrosis processes, suggesting that the key molecules downstream of the ceRNA network and the dynamic balance within the network are central to DN progression. Therefore, research on target molecules and dynamic balance within the network should be prioritized for disease prevention and treatment. In addition, different ceRNA networks can target and regulate downstream target molecules. Different lncRNAs can adsorb the same miRNAs in a sponge-like manner, whereas the same lncRNA can adsorb different miRNAs. Therefore, it is questionable whether different circRNAs can adsorb the same miRNA in a sponge-like manner, whether different miRNAs can be adsorbed by the same circRNA to exert their functions, or whether this regulatory network plays a role in diseases with similar pathological processes.

Clinical studies have identified significant up-regulation of circ_0125310 in both serum and renal tissues of DN patients, demonstrating a positive correlation with proteinuria severity (136). In vitro studies revealed that HG-stimulated GMCs secrete exosomes enriched with circ_0125310, which undergo efficient cellular internalization, Notably, exposure to HG-conditioned GMCs exosomes induced: increased EdU-positive cell proportion; elevated G2/M phase cell population; up-regulated fibrotic markers p-cadherin, ZO-1, crucially, KO-circ_0125310 reversed these pro-fibrotic effects, confirming exosomal circ_0125310 role in promoting MC proliferation and ECM remodeling (136). Furthermore, circ_0125310 directly binds to miR-422a and inhibits its expression, miR-422a targets the 3’ UTR of IGF1R and inhibits its expression, and overexpression of IGF1R activates the downstream p38 signaling pathway, in addition, circ_0125310 overexpression in turn upregulates the miR-422a-dependent IGF1R/p38 axis, promoting GMCs proliferation and fibrosis (136). Suggests that HG-induced exosomes from GMCs promote MCs proliferation and fibrosis by delivering circ_0125310, adsorbing miR-422a, deregulating its inhibition of IGF1R, and activating the p38 signaling pathway.

Li et al. also found that circ_TAOK1 was greatly elevated in exosomes secreted by HG-treated GMCs and serum exosomes from DN patients, with significant up-regulation of the levels of PCNA, cytokinin D1, α-SMA, FN, N-cadherin, and SMAD3 and significant down-regulation of the level of E-cadherin, which promoted the proliferation, fibrosis, accelerated fibrosis and EMT, while KO-circTAOK1 reversed the above effects, suggesting that exosomal circ_TAOK1 promotes GMC injury (137). Furthermore, it was found that circ_TAOK1 directly binds to miR-520h and represses its expression, which in turn targets the 3’ UTR of SMAD3 and represses its transcription, and furthermore, KO-miR-520h or overexpression of SMAD3 in turn reverses the protective effect of KO-circTAOK1 on GMC (137). From the above, it can be seen that exosomal circ_TAOK1 delivery to GMC, by adsorbing miR-520h, deregulates its inhibition of SMAD3, which in turn promotes GMC proliferation, fibrosis and EMT.

It was also found that miR-143 expression was greatly down-regulated in HG-treated GMCs, DN patient serum and DN rat kidneys, exosomal circ_DLGAP4 expression was significantly elevated and positively correlated with UAER and eGFR, overexpression of circ_DLGAP4 up-regulated EdU positivity, and significantly promoted the proliferation of GMCs, G2/M phase of the cell cycle block, and fibrosis markers p-cadherin, Mcp-1 expression, and promoted GMCs injury, while KO-circ_DLGAP4 reversed the above phenotypes (138). It was also found that miR-143 is a direct target of circ_DLGAP4 and ERBB3 is a downstream target gene of miR-143. miR-143 inhibits the expression of ERBB3 by binding to its 3’ UTR, whereas circ_DLGAP4 up-regulates ERBB3 through the inhibition of miR-143, which in turn activates the NF-κB pathway and induces MMP-2 expression, exacerbating fibrosis (138). It suggests that exosomal circ_DLGAP4 exacerbates DN progression by adsorbing miR-143 to deregulate ERBB3 and activating the NF-κB/MMP-2 pathway to promote thylakoid proliferation and fibrosis.

Similarly, exosomes circ_0125310, circ_TAOK1 and circ_DLGAP4 up-regulated IGF1R, SMAD3 and expressed ERBB3 through sponge adsorption of miR-422a, miR-520h and miR-143, respectively, where circ_0125310/miR-422a/IGF1R axis activation in turn up-regulated its downstream p38 signaling pathway, whereas activation of the circ_DLGAP4/miR-143/ERBB3 axis up-regulated the NF-κB/MMP-2 pathway, both of which contributed to fibrosis via the circRNA-miRNA-mRNA network, thereby exacerbating DN progression ( Figure 7 , Table 4 ).

In HG-stimulated HK-2 cells and urinary exosomes from DN patients, circ_0008529 and SMAD2 expression were significantly up-regulated, and miR-185-5p expression was down-regulated. KO-circ_0008529 increased cell viability, inhibited caspase-3 activation, and reduced cell cycle arrest, TNF-α, IL -6 release and FN, Col-I, and α-SMA accumulation (139). In addition, inhibition of miR-185-5p or overexpression of SMAD2 up-regulated circ_0008529 expression, reversed the protective effect of KO-circ_0008529 on HK-2 cells, and exacerbated cellular injury, whereas circ_0008529 directly bound miR-185-5p and sponged to deregulate the effect of miR-185-5p on the SMAD2 inhibition (139). It is shown that exosome circ_0008529 upregulates SMAD2 expression through adsorption of miR-185-5p, promotes renal tubular cell apoptosis, inflammation and fibrosis, and exacerbates DN progression.

It was also found that exosomal circ_NUP98 expression was significantly elevated and miR-151-3p expression was marked reduction in serum and HG-stimulated GMCs from DN patients, KO-circ_NUP98 inhibited HG-induced cell proliferation, down-regulated FN, Col-IV, Bcl-2, TNF-α, IL-6, IL-1β, ROS and MDA, up-regulated cleaved-caspase3, SOD, inhibited fibrosis, inflammatory response and OS, promoted apoptosis, and attenuated HG-induced injury in HMCs, whereas inhibition of miR-151-3p or overexpression of HMGA2 both reversed the protective effect of KO-circ_NUP98 on HMCs (140). Furthermore, it was found that miR-151-3p directly targets HMGA2, whereas circ_NUP98 can upregulate HMGA2 expression by inhibiting miR-151-3p, and HMGA2 overexpression reverses the protective effect of miR-151-3p, revealing that circ_NUP98 upregulates HMGA2 expression by adsorbing miR-151-3p that promotes HG-induced proliferation, fibrosis and inflammation and OS in HMCs (140).

Ultimately, targeted inhibition of the circ_0008529/miR-185-5p/SMAD2 and circ_NUP98/miR-151-3p/HMGA2 axis is one of the pathways to inhibit HG-induced cell proliferation, fibrosis, and inflammation, which helps to provide a new perspective for the mechanistic study and treatment of DN ( Figure 7 , Table 4 ).

Experimental investigations revealed significantly elevated TRAF5 levels and reduced lncRNA-TUG1 expression in both DN rat models and HG-induced podocytes. Astragaloside IV treatment effectively reversed these pathological alterations (158). Mechanistic studies further demonstrated physical interaction between TUG1 and TRAF5, with TUG1 overexpression specifically promoting TRAF5 degradation via the ubiquitin-proteasome pathway. Notably, TUG1 transfection counteracted Astragaloside IV’s suppressive effects on TRAF5, confirming that Astragaloside IV inhibits TRAF5 expression through TUG1 up-regulation (158). Additionally, HG conditions markedly increased podocyte apoptosis rates and caspase-3 protein levels, both of which were significantly attenuated by Astragaloside IV treatment, TUG1 transfection abolished this cytoprotective effect (158). These findings collectively demonstrate that Astragaloside IV ameliorates DN progression by enhancing lncRNA-TUG1-mediated ubiquitin-proteasomal degradation of TRAF5 protein, thereby reducing podocyte apoptosis.

Furthermore, in vivo studies revealed that db/db mice exhibited mesangial expansion, glomerular basement membrane thickening, increased fibrosis, and tubulointerstitial fibrosis in the kidney tubular region, these animals also demonstrated significantly elevated 24-hour urinary protein levels, fasting blood glucose concentrations, and area under the curve (AUC) values in oral glucose tolerance tests (OGTT), along with marked up-regulation of lncRNA-Gm41268 and PRLR expression (30). Both in vivo and in vitro experiments demonstrated that Astragalus polysaccharide treatment effectively reversed these pathological alterations, ameliorated renal impairment, and concurrently reduced p-mTOR, p62, TGF-β, Col-IV, and FN, the intervention notably enhanced the LC3-II/I ratio, indicating that Astragalus polysaccharides (APS) alleviated HG-induced autophagy inhibition and renal fibrosis (30). Crucially, subsequent investigations revealed that either pharmacological inhibition of Gm41268 or genetic KO-PRLR abolished the renal protective effects mediated by APS (30). Collectively, these findings demonstrate that APS mitigate DN progression by enhancing autophagy and suppressing renal fibrosis through targeted modulation of the lncRNA-Gm41268/PRLR signaling pathway.

Furthermore, investigations demonstrated significantly elevated expression of LncRNA-XIST and TGF-β1 in renal tissues of db/db mice, with notable co-localization observed between XIST and TGF-β1 (159). Intervention with Dendrobium polysaccharides or KO-XIST substantially attenuated renal pathological manifestations, manifesting attenuation of tubulointerstitial fibrotic lesions, amelioration of GMCs hypertrophy, and preservation of podocyte foot process architecture (159). These interventions significantly decreased fasting blood glucose levels, urinary protein excretion rate, BUN, Cys-C concentrations, and kidney-to-body weight ratio. Concurrently, marked down-regulation was observed in XIST and TGF-β1 expression, alongside reduced levels of fibrosis biomarkers COL-I, COL-IV, α-SMA, and phosphorylated Smad2/3, indicating suppression of the TGF-β1/Smad2/3 signaling pathway (159). These findings collectively suggest that Dendrobium polysaccharides mitigate DN progression by inhibiting LncRNA-XIST expression, thereby downregulating TGF-β1 and its downstream profibrotic effector Smad2/3, ultimately delaying renal interstitial fibrosis in DN.

As natural small-molecule compounds, herbal monomers not only modulate ncRNA expression to influence DN progression but also regulate ceRNA networks to intervene in DN pathogenesis. For instance, studies demonstrated that under HG conditions, elevated levels of MDA, ROS, IL-1β, IL-6, TNF-α, FN, and Col-I were observed alongside reduced SOD activity, these alterations intensified OS, inflammatory responses, fibrotic processes, and cellular damage, all of which were dose-dependently reversed by emodin treatment (160). Mechanistically, HG-induced up-regulation of circ_0000064 expression was suppressed by emodin. Crucially, overexpression of circ_0000064 abolished the protective effects of emodin, while circ_0000064 was shown to directly bind miR-30c-5p and inhibit its activity. miR-30c-5p itself targets the 3’ UTR of Lmp7 to suppress its expression (159). KO-circ_0000064 or overexpression of miR-30c-5p significantly reduced Lmp7 levels, ameliorating HG-induced cellular injury. Furthermore, emodin counteracted HG-induced activation of the NF-κB and PI3K/AKT pathways by inhibiting Lmp7, whereas Lmp7 overexpression neutralized emodin’s suppressive effects on these pathways (160). Collectively, these findings demonstrate that emodin alleviates HG-induced mesangial cell OS, inflammation, and fibrosis through modulation of the circ_0000064/miR-30c-5p/Lmp7 axis, thereby suppressing NF-κB and PI3K/AKT signaling.

Cai et al. demonstrated significantly down-regulated miR-204-5p and up-regulated SNAI1 and lncRNA-CLYBL-AS2 in the serum of DN patients and murine models, CLYBL-AS2 exhibited a negative correlation with miR-204-5p and a positive association with SNAI1 expression and DN staging (161). Under HG conditions, CLYBL-AS2 overexpression in HK-2 cells reduced miR-204-5p levels while elevating SNAI1 expression, thereby promoting cellular migration, invasion, and the expression of fibrotic biomarkersFN, α-SMA, and ASK1. Conversely, KO-CLYBL-AS2 suppressed these pathological phenotypes (161). Dual-luciferase reporter assays confirmed that CLYBL-AS2 shares binding sites with miR-204-5p, competitively inhibiting miR-204-5p-mediated targeting of SNAI1, Intervention with Coptis chinensis significantly reduced CLYBL-AS2 expression in DN mice in a dose-dependent manner (161). These findings indicate that Coptis chinensis ameliorates DN progression by suppressing the lncRNA-CLYBL-AS2/miR-204-5p/SNAI1 axis, thereby reversing EMT and fibrotic processes.

Overall, herbal monomers regulate DN progression by modulating ncRNA expression or interfering with ceRNA networks ( Figure 8 , Table 6 ). This discovery not only expands the functional understanding of ncRNAs and ceRNA networks in DN pathogenesis but also provides novel evidence for the nephroprotective effects of herbal monomers. Furthermore, it establishes a critical theoretical foundation for developing natural product-based therapeutic strategies against DN and offers experimental insights into mechanistic exploration and targeted therapies for this condition.

Traditional herbal formulae, grounded in the Jun-Chen-Zuo-Shi compatibility principle, exert multi-dimensional interventions in DN progression through synergistic multi-target effects. These formulations not only circumvent drug resistance inherent to single-target pharmacological agents but also enable coordinated targeting of critical ncRNA nodes via their multi-component synergy, thereby achieving comprehensive modulation of DN pathogenesis (170, 171).

Jin et al. demonstrated that HG stimulation remarkably suppressed podocyte viability and promoted apoptosis, while treatment with Jixuepaidu Tang-1 (JPDT-1) markedly reversed these HG-induced alterations (162). Further investigations revealed that JPDT-1 up-regulated epithelial markers nephrin, P-cadherin and down-regulated mesenchymal markers desmin, vimentin, effectively counteracting HG-triggered EMT in podocytes (162). Additionally, JPDT-1 reduced HG-elevated levels of TGF-β1/SGK1, notably, recombinant TGF-β1 or SGK1 administration abolished the protective effects of JPDT-1, whereas SGK1 silencing produced opposing outcomes (162). These findings collectively indicate that JPDT-1 mitigates HG-induced podocyte injury and EMT progression through suppression of the TGF-β1/SGK1 signaling axis. Further investigations revealed elevated expression of lncRNA-LOC498759 in renal tissues of DN mice and HG-stimulated podocytes, JPDT-1 significantly suppressed LOC498759 expression, whereas LOC498759 overexpression attenuated JPDD-1-mediated inhibition of EMT. Conversely, KO-LOC498759 potentiated its protective effects (162). Additionally, TGF-β1/SGK1 pathway activation up-regulated LOC498759 expression, though LOC498759 overexpression itself exerted no significant regulatory effect on TGF-β1/SGK1 levels, suggesting LOC498759 operates downstream of this signaling axis (162). JPDT-1 treatment reduced kidney-to-body weight ratio, 24-hour urinary protein excretion, and blood glucose levels in DN mice. Histologically, it ameliorated glomerular matrix deposition and renal apoptosis while downregulating renal TGF-β1, SGK1, and mesenchymal marker expression. Notably, LOC498759 overexpression diminished JPDD-1’s renoprotective effects and EMT suppression, yet had no significant impact on TGF-β1/SGK1 signaling activity, further corroborating LOC498759’s position downstream of this pathway (162).

Collectively, these findings demonstrate that JPDT-1 ameliorates DN progression through modulation of the TGF-β1/SGK1/lncRNA-LOC498759 axis ( Figure 8 , Table 6 ). This not only provides experimental validation for the modernization of traditional herbal formulae but also offers critical mechanistic insights and potential clinical relevance for their application in DN treatment.

As can be seen from the previous discussion, although biological molecule-targeted regulation of the ceRNA network has shown some effectiveness in regulating DN progression; however, because of the complementary nature of the ncRNA seed region with non-target genes, the uptake of the vector by non-target cells, and the saturation of the RNAi mechanism at high ncRNA doses, as well as interference with the endogenous miRNA pathway, the specificity of ncRNA target genes is relatively poor. Consequently, off-target effects frequently occur when ncRNA-targeted therapies are used (172). Second, during DN, reactions such as inflammation and OS occur, causing changes in the renal microenvironment and significantly reducing intracellular release efficiency. The glomerular filtration barrier and the repulsion of charges on the basement membrane result in the clearance of numerous carriers, with only a small fraction reaching the kidneys. Unmodified renal targeting ligands on the carrier surface, as well as serum nuclease degradation and opsonin effects, cause the intervention drugs to accumulate non-specifically in the liver/spleen as well as reduce the half-life of the drug, leading to low delivery efficiency (173). Finally, because of the differing sensitive targets of ncRNAs across different DN subtypes (inflammatory and fibrotic types), generic therapies may be ineffective or damage normal kidney tissue, advanced DN glomerulosclerosis reduces carrier permeability, necessitating the development of penetration-enhancing peptides to enhance carrier penetration capacity (174). Therefore, the treatment of DN presents a unique challenge.

The low bioavailability of TCM components can be attributed to several factors. First, the low transmembrane permeability of large polar molecules limits their absorption. In addition, the metabolism by liver CYP450 enzymes results in a strong first-pass effect, and the degradation of glycoside components by intestinal microbiota further reduces bioavailability. Second, most TCM components cannot accumulate in diseased tissues, and their strong binding to plasma proteins reduces free drug concentrations, thus leading to impaired distribution of TCM components within the body. The complex interactions between multiple components and active metabolites in TCM also result in unpredictable metabolism. Furthermore, owing to the high molecular weight of TCM components, which cannot be filtered by the glomeruli, a high risk of accumulation and prolonged half-life exists in the enterohepatic circulation, leading to impaired drug excretion (175). Therefore, although strategies based on bioactive molecules and targeted regulation of ncRNAs in TCM show great potential in the prevention and treatment of DN, their clinical translation still faces multiple cross-scale bottlenecks (delivery efficiency, off-target effects, DN heterogeneity of diabetic nephropathy, and safety), requiring further study.