Tau beyond the Brain: Endoplasmic Reticulum Stress and Tau Phosphorylation as Emerging Drivers of Diabetic Nephropathy Authors: Na-Yeong Kim, Chang-Myung Oh Published: 2026 Diabetic nephropathy (DN) is the leading cause of end-stage kidney disease worldwide, affecting approximately 40% of all individuals with diabetes mellitus and contributing to a nearly 38% increase in age-standardized mortality rate over the past three decades [1]. Despite notable advances in treatment methods based on renin-angiotensin system blockade and sodiumglucose cotransporter 2 inhibition, a substantial proportion of patients progress to kidney failure, underscoring the need for a deeper understanding of the pathogenic mechanisms that drive glomerular injury and tubular damage. Investigation into potential mediators of renal injury has recently focused on tau, a microtubule-associated protein canonically associated with neurodegenerative tauopathies such as Alzheimer’s disease (AD). In this issue of the Diabetes & Metabolism Journal, Lee et al. [2] provide mechanistic evidence that site-specific tau phosphorylation and endoplasmic reticulum (ER) stress contribute to DN pathogenesis, expanding the scope of therapeutic opportunities. Tau in non-neuronal tissues: a paradigm shift The pathological hyperphosphorylation and aggregation of tau into neurofibrillary tangles are hallmarks of tauopathies, including AD and frontotemporal dementia. While tau is classically considered a predominant neuronal protein, accumulating evidence has challenged this view. Tau is expressed in several peripheral tissues, including the kidney and pancreas, albeit at substantially lower levels than in the central nervous system [3,4]. Tau may also be actively transported from the central nervous system to peripheral tissues via cerebrospinal fluid drainage pathways, with the kidney being a major site of tau uptake and excretion [5]. Tau localizes to glomerular podocytes and participates in maintaining cytoskeletal architecture. Consequently, tau knockout studies have revealed a disruption of glomerular morphology [6]. Moreover, the detection of both tau and amyloid-β in the pancreas of individuals with type 2 diabetes mellitus suggests that neurodegenerative pathways can influence metabolic pathways [7]. Lee et al. [2] now extend this framework to the diabetic kidney, providing one of the first detailed assessments of site-specific tau phosphorylation in DN. Key findings and mechanistic insights Lee et al. [2] used NADPH oxidase 5 (NOX5) pod+ transgenic mice fed a high-fat diet as a model of DN. As early as 6 weeks post-treatment, tau phosphorylation at Ser202 and Thr205 was significantly elevated and persisted at a high level through 12 weeks. This elevation coincided with progressive albuminuria, glomerular basement membrane thickening, and podocyte injury. The Ser202/Thr205 upregulation was the most consistent and sustained of all tau phosphorylation sites examined under diabetic conditions. This outcome aligns with the stage-dependent phosphorylation hierarchy in tauopathies, where Ser202/Thr205 accumulation predominates at advanced disease stages [8]. Lee et al. [2] then validated the animal experiment’s translational relevance via immunohistochemical analysis of human renal biopsy specimens. Glomerular phosphorylated tau (pTau) Ser202/Thr205 accumulation was absent in controls but elevated in DN patients and positively correlated with the urinary protein-to-creatinine ratio. Hence, both the murine model and human pathological data support tau phosphorylation as a biomarker and pathogenic contributor in DN. At the mechanistic level, Lee et al. [2] identified a bidirectional relationship between ER stress and tau phosphorylation. Both transforming growth factor-β (TGF-β)-induced fibrosis and thapsigargin-induced ER stress augmented tau phosphorylation at Ser202 and Thr205 in mesangial cells. Additionally, upon blocking tau phosphorylation with thousand and one amino acid kinase (TAOK) inhibitor, canonical ER stress markers (binding immunoglobulin protein [BiP], phosphorylated eukaryotic initiation factor 2α [p-eIF2α], C/EBP homologous protein [CHOP], and cleaved caspase-3) were downregulated. This crosstalk mirrors the vicious cycle described in AD: first, ER stress promotes tau hyperphosphorylation; in turn, hyperpTau amplifies unfolded protein response signaling and ER dysfunction [9]. The demonstration that this mechanism operates in the diabetic kidney provides a plausible framework for understanding how metabolic stress, fibrotic remodeling, and cytoskeletal dysfunction converge to drive glomerular injury. Furthermore, the DN-ameliorating compound dehydrozingerone significantly limited tau phosphorylation at Ser202/Thr205 and attenuated TGF-β–induced fibrotic markers, offering pharmacological proof of concept for targeting this pathway. Limitations First, the human validation cohort comprised only four DN biopsy samples and five controls. While correlations were significant, they need to be replicated in larger, more ethnically diverse cohorts across defined chronic kidney disease stages. Second, enzyme-linked immunosorbent assay-based measurements of circulating pTau in serum and urine were inconsistent with results from tissue-based Western blots. This discrepancy highlights a critical question: can site-specific pTau Ser 202/Thr205 function as a reliable non-invasive biomarker for DN? Future studies employing site-specific immunoassays in well-characterized longitudinal cohorts are needed to resolve this issue. Third, while the NOX5 pod+ model provides a valuable platform for studying oxidative stress-driven podocyte injury, it does not fully capture the complex metabolic milieu of human DN. Indeed, tau phosphorylation is unlikely to act in isolation but instead as part of a broader stress-response network encompassing oxidative stress, mitochondrial dysfunction, and impaired autophagy. Before considering clinical applications, results from Lee et al. [2] should be validated in streptozotocin-induced or db/db diabetic mouse models to assess interactions with other molecular mechanisms of diabetes and DN. Cell-type-specific gain- and loss-of-function experiments will also help clarify tau’s independent contributions to DN pathogenesis. Therapeutic implications and future directions The identification of the tau–ER stress axis as a pathogenic contributor to DN has important clinical implications. TAOK inhibitors and ER stress modulators such as 4-phenylbutyric acid (4PBA) represent potential intervention points that could complement existing nephroprotective strategies. However, the therapeutic window, long-term safety, and off-target effects of these agents require careful evaluation in preclinical studies before clinical translation can be considered. Particularly noteworthy is the observation that low-dose TGF-β stimulation upregulated tau phosphorylation, but higher doses had a downregulatory effect. This biphasic response underscores the context-dependent complexity of tau regulation and cautions against simplistic therapeutic assumptions. Tau as a pathological mediator in DN emphasizes potential links between metabolic disease and neurodegeneration. Individuals with type 2 diabetes mellitus face an elevated risk of cognitive decline and AD, a connection increasingly attributed to shared mechanisms including insulin resistance, ER stress, glycogen synthase kinase-3β activation, and tau dysregulation [10]. Whether renal tau pathology is mechanistically linked to neurodegenerative processes remains unclear; both phenomena may simply reflect shared upstream stress pathways. Their relationship could be clarified with prospective studies incorporating renal and neurocognitive outcomes in diabetic cohorts, complemented by biomarker profiling of site-specific phospho-tau species. In conclusion, the study by Lee et al. [2] advances our understanding of DN pathogenesis by demonstrating that site-specific tau phosphorylation—both driven by and reciprocally amplifying ER stress—contributes to glomerular injury in both mouse models and human tissues. These data position tau as an active participant in renal stress response under diabetic conditions. Further research is needed to validate tau and its upstream kinase pathways as clinically actionable targets, as well as to develop site-specific biomarker assays suitable for longitudinal monitoring. Ultimately, determining whether the tau–ER stress axis represents a shared molecular vulnerability will be crucial to understanding multi-organ complications of diabetes mellitus. Footnotes CONFLICTS OF INTEREST Chang-Myung Oh has been an associate editor of the Diabetes & Metabolism Journal since 2024. He was not involved in the review process of this article. Otherwise, there are no conflicts of interest. REFERENCES