A urea-activated nanocarrier for site-specific SGLT2 inhibition and metabolic rescue against cardiovascular-kidney-metabolic syndrome

Abstract

Cardiovascular-kidney-metabolic (CKM) syndrome, driven by interlinked metabolic, renal and cardiovascular dysfunction, remains therapeutically challenging due to its multi-organ complexity. Although sodium-glucose cotransporter 2 (SGLT2) inhibitors such as empagliflozin (EMPA) confer cardiorenal benefits, their efficacy is limited by poor renal specificity and systemic off-target exposure. Here, we report a kidney-targeted and urea-responsive nanocarrier (T-PAAD NPs) that enables renal tubule-selective release of EMPA in response to pathological urea concentrations. This delivery strategy, rarely explored in nanomedicine, synergistically enhances therapeutic precision while integrating reactive oxygen species (ROS) scavenging to mitigate oxidative stress. In male mouse models of CKM, T-PAAD NPs/EMPA effectively reprogram cardiac and renal energy metabolism, restore filtration and contractile function, and achieve superior glycemic, renal, and cardiovascular outcomes compared to free EMPA. By coupling bioresponsive controlled release with metabolic modulation, this nanoplatform provides a promising approach to treat CKM and other systemic metabolic disorders.

Due to interacting metabolic, renal and cardiovascular dysfunction, cardiovascular-kidney-metabolic syndrome is difficult to treat. Here, the authors show a urea-responsive, kidney-targeted nanocarrier enables site-specific SGLT2 inhibition, restores cardiorenal metabolism and improves efficacy.

Introduction

Type 2 diabetes mellitus (T2DM), chronic kidney disease (CKD), and cardiovascular disease (CVD) have been defined as the most disruptive public health issues^1,2. Additionally, there exists a strong interplay among T2DM, CKD and CVD, and they have the propensity to initiate and perpetuate each other, leading to the increased morbidity and mortality^3–5. In 2023, the American Heart Association (AHA) coined a new term, cardiovascular-kidney-metabolic (CKM) syndrome, to describe a systemic disease arising from the interconnectedness of obesity, diabetes, CVD and CKD⁶. CKM health reflects the intricate interrelationships among metabolic risk factors, kidney function, and the cardiovascular system. Therefore, it is essential to develop a comprehensive strategy to manage the multidimensional risk factors from cardio‑renal‑metabolic connection for CKM therapy^7,8.

Currently, in addition to healthy lifestyle, the out comes of several clinical trials have demonstrated that sodium-glucose cotransporter-2 inhibitor (SGLT-2i) not only lowers blood glucose levels but also has renal- and cardio-protective effects^9–11. SGLT-2i acts on the SGLT-2 receptors expressed in the renal proximal convoluted tubules to inhibit reabsorption of glucose for T2DM therapy¹². Furthermore, SGLT2i can reduce cardiac load by promoting the osmotic excretion of sodium and urine through the kidneys, which in turn helps maintain cardiac ejection fraction and improve cardiac energy metabolism^13–16. Meanwhile, SGLT2i can also contribute to kidney protection by improving renal hemodynamics, decreasing the proteinuria excretion, and promoting the renal tubular energy metabolism^17–19. These multifaceted actions underscore the therapeutic potential of SGLT-2i in managing CKM syndrome. However, despite their promise, SGLT-2i also faces several limitations that may hinder their optimal clinical efficacy²⁰. On one hand, poor aqueous solubility of SGLT-2i, like empagliflozin (EMPA) makes them suffer from poor bioavailability²¹. Additionally, the heterogeneity, complexity, and unique physiological structure of the kidney present significant challenges for therapeutic interventions, especially with respect to safety, specificity, and efficacy of small-molecule drugs^22,23. Both of these factors necessitate the high doses of SGLT-2i for effective CKM management, causing the risk of adverse effects such as bone fracture²⁴, diabetic ketoacidosis^25,26, serious urinary tract infections²⁷, etc. On the other hand, oxidative stress plays a pivotal role in the pathogenesis of CKM. Although SGLT-2i can indirectly reduce oxidative stress through various mechanisms, they are unable to directly scavenge pre-existing ROS in the damaged kidney^28–30. Notably, several studies have demonstrated that the combination of antioxidants with SGLT-2i can further enhance both renal- and cardio-protective effects^31,32. Therefore, the strategies that improve the in vivo delivery efficiency of SGLT-2i and simultaneously incorporate antioxidant therapies hold promise for enhancing their overall therapeutic efficacy in CKM management.

Nano-drug delivery system is the results of merging nanotechnology with new drug development, possessing the improved pharmacokinetics, the targeted delivery and the controlled release of drugs to the specific tissues and cells^33,34. In addition, nanocarriers can be chemically modified to possess multifunctional properties, such as ROS-scavenging capability, which can suppress oxidative stress and provide additional therapeutic benefits for CKM management. Despite these advantages, the nanomedicines for treating kidney diseases remain significantly underdeveloped compared to nanomedicines in other fields, such as cancer nanomedicines, thereby hindering the development of SGLT-2i-based nanomedicine³⁵. Nevertheless, the kidney possesses several unique features that can be exploited for the rational design of renal-targeted nanomedicines³⁶. Notably, the glomerular filtration barrier becomes leaky resulting from impairment of the glomerular filtration barrier under diseased conditions, potentially facilitating the nanomedicine across the physiological barriers to proximal tubule³⁷. Additionally, the water reabsorption via aquaporins further concentrates urea in the lumen, resulting in the urea concentration in the renal tubule that is ~50% higher than in plasma^38–40. Essentially, urea levels within the S1 segment of the proximal tubule vary markedly across stages of kidney injury, increasing from ~5 mM in healthy kidneys to ~10 mM in early injury and up to ~50 mM in end-stage disease^41,42. This suggests that urea may serve as an endogenous stimulus to achieve the efficient controlled drug release from nanomedicines within the kidney; however, this approach has been rarely explored to date. Therefore, developing a urea-responsive nanocarrier with ROS-scavenging ability to enable targeted delivery and controlled release of SGLT-2i in the proximal tubule represents a promising strategy to improve therapeutic efficacy against CKM.

In our previous studies, a series of smart polymers exhibiting an upper critical solution temperature (UCST) have been designed and developed, with their phase inversion behavior primarily driven by polymer-polymer hydrogen bonds^43,44. Urea is known to disrupt intramolecular hydrogen-bonded network; therefore, nanomedicine formulated from UCST polymers are expected to disassemble in the presence of urea by breaking their intramolecular hydrogen bonds, enabling rapid drugs release from UCST nanomedicines. Herein, we engineered kidney-targeted and urea-responsive nanocarriers with intrinsic ROS-scavenging capacity by employing multifunctional UCST polymer to achieve controlled release of the SGLT-2i (empagliflozin, EMPA) specifically in the renal proximal tubules (Fig. 1a). This design synergistically integrates oxidative stress mitigation with cardio-renal energy metabolism reprogramming, offering enhanced therapeutic efficacy against CKM syndrome. Briefly, diallyl trisulfide (DATS), a sulfur-rich compound derived from garlic, was introduced into polyethylene glycol-b-poly(acrylonitrile-co-acrylamide) to synthesize the multifunctional UCST polymer (polyethylene glycol-b-poly(acrylamide-co-acrylonitrile-co-diallyl trisulfide), PEG-PAAD). This polymer self-assembles into nanoparticles at physiological temperature via intramolecular hydrogen bonding, while the trisulfide groups confer antioxidant and anti-inflammatory activities by scavenging existing ROS in the damaged kidney. Meanwhile, the renal tubule targeting peptide, (Lys-Lys-Glu-Glu-Glu)₃-Lys ((KKEEE)₃K), was conjugated to the surface of UCST nanoparticles to enhance proximal tubule epithelial cell targeting. Upon administration, these EMPA-loaded nanomedicines effectively accumulated in the kidney and targeted proximal tubules. Elevated urea levels in the renal tubule disrupted the intramolecular hydrogen bonds of the UCST nanomedicine, triggering nanoparticle disassembly and controlled EMPA release. Combined with the antioxidant properties of the UCST polymer, EMPA effectively lowered blood glucose and blood pressure, while improving cardiac and renal hemodynamics through reprogramming cardio-renal energy metabolism. This novel and intelligent drug delivery system triggered by endogenous urea levels, a mechanism rarely explored in nanomedicine, demonstrates promising potential for treating CKM. This approach represents a significant advance toward nanomedicine-based co-management of diabetes, cardiovascular disease, and kidney dysfunction, offering a compelling strategy to improve clinical outcomes in patients with CKM.

Fig. 1. Schematic illustration of the synthesis strategy and cardio-renal energy metabolism reprogramming for CKM syndrome therapy.

a Synthesis of the urea-responsive and kidney-targeted T-PAAD NPs and their ROS-scavenging mechanism. b Illustrative workflow of renal tubule-specific EMPA delivery and the subsequent reprogramming of cardiac and renal energy metabolism in CKM mice. Created in BioRender. Xue, R. (2026) https://BioRender.com/a8vy1qa.

Results

Synthesis and characterization of urea-responsive UCST copolymer with renal tubule targeting capability

The multifunctional UCST copolymer, poly(acrylamide-co-acrylonitrile-co-diallyl trisulfide) (PAAD), was synthesized via free radical polymerization. The (KKEEE)₃K peptide, a promising renal tubule-targeting ligand, was subsequently conjugated to the PAAD copolymer through a polyethylene glycol (PEG) linker (Supplementary Fig. 1). The chemical structures of the PAAD and PEG-PAAD copolymers were confirmed by ¹H-NMR spectroscopy (Supplementary Fig. 2a, b). The molecular weight of these two copolymers were 21,908 Da and 24,804 Da, respectively (Supplementary Fig. 2c). Furthermore, the bicinchoninic acid (BCA) assay confirmed successful conjugation of the (KKEEE)₃K peptide to the PEG-PAAD copolymer, with a conjugation efficiency of ~64.5% (Supplementary Fig. 2d).

As mentioned above, urea can disrupt the intramolecular hydrogen bonds among UCST copolymers that govern their phase transition behavior. Based on this property, we hypothesized that the UCST copolymer may exhibit urea-responsiveness. To demonstrate this hypothesis, the phase transition temperatures of the PEG-PAAD copolymer were measured under varying urea concentrations. As the urea concentration increased from 0 to 500 mM, the UCST of the copolymer decreased markedly from 41.3 °C to 30.6 °C. Corresponding optical images revealed concentration-dependent changes in turbidity when the copolymer was incubated with urea at different temperatures (Fig. 2a and Supplementary Fig. 3a). At 37 °C, the initially opaque PEG-PAAD emulsion became transparent under high urea concentrations, and the Tyndall effect was visibly diminished, indicating that the polymer aggregates disassembled in response to urea (Supplementary Fig. 3b). Meanwhile, the Fourier Transform Infrared Spectroscopy (FTIR) analysis of the PEG-PAAD copolymer was conducted in the presence of urea. As shown in Fig. 2b, c, the characteristic FTIR peaks of PEG-PAAD copolymer were observed at 3350 cm⁻¹ (N-H stretching vibration), 1616 cm⁻¹ (N-H bending vibration), and 1660 cm⁻¹ (C=O stretching vibration). However, due to the disruptive effect of urea on intramolecular hydrogen bonding, these peaks shifted to higher wavenumbers, appearing at 3446 cm⁻¹, 1624 cm⁻¹, and 1667 cm⁻¹, respectively. To further validate this observation, molecular simulation analysis revealed that the zipper-like intramolecular hydrogen bonds formed within PEG-PAAD copolymers were disrupted in the presence of urea. Both the number and energy of hydrogen bonds were significantly reduced (Fig. 2d and Supplementary Fig. 4), further confirming the urea-responsiveness of the PEG-PAAD copolymer. To evaluate whether the urea-responsiveness of UCST copolymer could facilitate urea-triggered drug release, the kidney-targeted UCST NPs (T-PAAD NPs) were fabricated using a molar ratio of (KKEEE)₃K-PEG-PAAD to PEG-PAAD of 1:10. The resultant T-PAAD NPs were employed to encapsulate EMPA, forming the spherical nanoparticles with an average diameter of 50.75 nm and a surface potential of −0.98 mV. Furthermore, T-PAAD NPs maintained excellent colloidal stability in serum-containing medium at 37 °C for 72 h without significant aggregation. These NPs exhibited favorable drug loading capacity (233.3 μg mg⁻¹ NPs) and encapsulation efficiency (47.6%) (Supplementary Fig. 5).

Fig. 2. Characterization of PEG-PAAD copolymer and T-PAAD NPs.

a Turbidity heating profiles of PEG-PAAD aqueous solutions (0.25 mg mL⁻¹) incubated with different concentrations of urea (100 mM, 300 mM and 500 mM). Insets: turbidity images of PEG-PAAD copolymer emulsion containing with different concentrations of urea at 37 °C. FTIR spectra of PEG-PAAD and PEG-PAAD+Urea from 4000 to 650 cm⁻¹ (b) and their enlarged spectra from 1900 to 900 cm⁻¹ (c). d Molecular simulations of PEG-PAAD polymer in response to urea. e Cumulative in vitro release of EMPA from T-PAAD NPs over 24 h under simulated physiological and pathological conditions. Three media were used to model normal plasma and the S1 segment of the renal proximal tubule across healthy-to-diseased states: Condition 1 (normal plasma/healthy S1, 5 mM urea), Condition 2 (early kidney injury S1, 10 mM urea), and Condition 3 (end-stage kidney injury S1, 50 mM urea) (n  =  3 independent experiments). H₂O₂ (f), O₂^·− (g), ·OH (h), and free radical (i) scavenging ability of T-PAAD NPs at different concentrations (n  =  3 independent experiments). j Representative CLSM images of ROS (green fluorescence) in PBS (Negative) and 150 μM H₂O₂-stimulated (Positive) HK2 cells after different treatments. Scale bars, 20 μm. (n  =  3 independent experiments). k Flow cytometry analysis of intracellular ROS levels in H₂O₂-pretreated HK2 cells after different treatment. Data were analyzed using one­way analysis of variance (ANOVA) with Tukey’s post hoc test (e–i) and shown as mean ± s.d. *P  <  0.05, **P  <  0.01, ***P  <  0.001 and ****P  <  0.0001. Source data are provided as a Source Data file.

Importantly, the in vitro release profiles show that T-PAAD NPs enable stimulus-responsive drug release across urea concentrations relevant to kidney injury. Under conditions mimicking normal plasma or healthy proximal tubules (5 mM urea), T-PAAD NPs/EMPA showed a sustained release pattern, reaching a cumulative release of 62.3% over 24 h. In contrast, as proximal tubular injury progresses and urea levels rise, EMPA release was correspondingly accelerated. Notably, exposure to a simulated severe-injury microenvironment (50 mM urea) triggered a rapid increase in EMPA release, reaching ~93% within 24 h (Fig. 2e). This pronounced difference in release kinetics supports the sensitivity of the T-PAAD NPs to the pathological microenvironment, driven by elevated urea concentrations. Mechanistically, UCST-based nanomedicines are expected to partially disassemble in the presence of urea, which disrupts intra- and interchain hydrogen bonding; this urea-induced weakening lowers the UCST of T-PAAD and thereby facilitates faster drug release. Taken together, because the urea concentration of the proximal tubule is markedly elevated in injured kidneys, these UCST nanoparticles offer a promising platform for on-demand drug release within the renal tubules.

Antioxidative capacity of T-PAAD NPs

In addition to their urea-responsiveness, T-PAAD NPs possess strong ROS-scavenging activity due to the incorporation of the antioxidant DATS in PAAD copolymer. This feature is expected to synergistically alleviate renal oxidative stress when combined with SGLT-2i therapy. To assess the antioxidant capacity of T-PAAD NPs, three representative ROS species, namely hydrogen peroxide (H₂O₂), superoxide anion (O₂^•−), and hydroxyl radical (·OH), were incubated with T-PAAD NPs. As shown in Fig. 2f–h, the ROS clearance rate increased with nanoparticle concentration, and effective scavenging activity was observed even at a low concentration of 0.125 mg mL⁻¹. Additionally, at a concentration of 1 mg mL⁻¹, T-PAAD NPs were able to decompose ~90% of H₂O₂, 70% of the O₂^•−, and 60% of the ·OH. Furthermore, the total antioxidant capacity of T-PAAD NPs was evaluated using 2, 2’-azinobis (3-ethylbenzothiazoline 6-sulfonate) (ABTS) radical scavenging assay. There were more than 60% of ABTS ^•+ radicals being scavenged in the group of T-PAAD NPs (1 mg mL⁻¹), further confirming their potent antioxidant activity (Fig. 2i). Given the favorable antioxidative properties of T-PAAD NPs, we further evaluated their intracellular ROS-scavenging capacity. As shown in Fig. 2j, treatment of proximal tubule epithelial cells (HK2 cells) with 150 μM H₂O₂ led to a marked increase in intracellular ROS levels, indicated by a strong green fluorescent signal. In contrast, HK2 cells treated with T-PAAD NPs exhibited a significant reduction in ROS levels. Quantitative analysis by flow cytometry confirmed that even at a low concentration (12.5 μg mL⁻¹), T-PAAD NPs could scavenge ~40.2% of intracellular H₂O₂ (Fig. 2k and Supplementary Fig. 6a). Consistently, T-PAAD NPs at this concentration improved HK2 cell viability from 78.9% to 93.5% under oxidative stress induced by 150 μM H₂O₂, further confirming their effective intracellular ROS-scavenging ability (Supplementary Fig. 6b). To further simulate the pathological environment of CKD, HK2 cells were exposed to high levels of glucose (33 mmol L⁻¹ D-glucose) and palmitate (0.3 mmol L⁻¹) to induce hyperglycemic and lipotoxic stress. This co-stimulation significantly elevated the levels of malondialdehyde (MDA), a marker of lipid peroxidation, to 0.84 μmol g⁻¹ protein, which were 1.67- and 1.80-fold higher than those observed in PBS and mannitol control groups, respectively (Supplementary Fig. 6c, d). Notably, treatment with T-PAAD NPs effectively normalized MDA levels, highlighting their potent antioxidative ability to counteract cellular damage under metabolic stress. Together, these findings suggest that T-PAAD NPs serve not only as a targeted nanocarrier for EMPA delivery but also as an effective antioxidant agent. Their dual functionality offers strong potential to synergize with EMPA and enhance its renal- and cardiovascular protective effects in CKM therapy.

Renal tubule targeting capability of T-PAAD NPs

In addition to controlled EMPA release and ROS-scavenging ability, the renal tubule-targeting ability of T-PAAD NPs/EMPA is a critical determinant of their therapeutic efficacy. The (KKEEE)₃K peptide targets megalin, a multiligand receptor highly expressed on renal tubular epithelial cells^45,46, thereby facilitating the endocytosis of T-PAAD NPs. This targeting ability was validated using coumarin 6 (C6)-loaded T-PAAD NPs, where flow cytometric analysis showed significantly enhanced uptake by HK2 cells (Supplementary Fig. 7). However, this uptake was remarkedly suppressed by pretreatment with free (KKEEE)₃K peptide, confirming the receptor-mediated and specific renal tubular cell-targeting ability of T-PAAD NPs (Fig. 3a, b).

Fig. 3. Evaluation of the renal targeting ability of T-PAAD NPs.

Flow cytometry analysis (a) and CLSM images (b) of HK2 cells incubated with C6-loaded T-PAAD NPs with or without free (KKEEE)₃K pre-incubation. Scale bars, 20 μm. (n  =  3 independent experiments). Ex vivo fluorescence images showing the biodistribution of Free IR780, PEG-PAAD NPs/IR780, and T-PAAD NPs/IR780 in CKM mice and RD mice, respectively (c), and quantitative analysis of IR780 fluorescence intensity in the harvested organs at 6 h post-injection (d) (n  =  3 biologically independent samples). e Fluorescence images of C6 (presenting T-PAAD NPs, green) and AQP1 (presenting proximal tubule marker, red) in the kidney after T-PAAD NPs/C6 i.p administration. Cell nucleus was stained with DAPI (blue). Scale bars, 50 μm. (n  =  3 biologically independent samples). Statistical analysis was performed with unpaired t-test two-tailed (a), and one-way ANOVA with Tukey’s post hoc test (d). Data were expressed as mean ± s.d. **P  <  0.01, ***P  <  0.001. Source data are provided as a Source Data file.

To further assess renal targeting in vivo, we evaluate the biodistribution of T-PAAD NPs in a CKM mouse model, established by feeding STZ-treated ApoE^−/− mice with Western diet (WD) for 12 weeks (Fig. 3c). Following intraperitoneal (i.p) injection, IR780-loaded T-PAAD NPs exhibited strong accumulation in the kidneys, with renal fluorescence intensities 1.60- and 1.80-fold higher than those observed with free IR780 and PEG-PAAD NPs/IR780, respectively (Fig. 3d). In contrast, this targeting effect was significantly attenuated in healthy mice fed a regular rodent chow (RD) diet, further highlighting the disease-dependent targeting capability of T-PAAD NPs. Moreover, time-dependent biodistribution analysis revealed rapid renal uptake and prolonged kidney retention of T-PAAD NPs, with significantly higher fluorescence than non-targeted PAAD NPs at 6, 12, and 24 h post-injection (Supplementary Fig. 8). To spatially evaluate the intrarenal distribution of T-PAAD NPs, the T-PAAD NPs/C6 were i.p injected into CKM mice. At 6 h post-injection, kidney tissues were collected, sectioned, and immune-stained for aquaporin-1 (AQP-1), a biomarker specific to renal tubules. Confocal imaging revealed clear colocalization of the green fluorescence from T-PAAD NPs/C6 with the red fluorescence from AQP-1, indicating that the NPs successfully accumulated within renal tubular epithelial cells (Fig. 3e).

This enhanced-renal tubule accumulation can be attributed to two major factors. Firstly, the CKM-induced kidney injury compromises the glomerular filtration barrier, facilitating the enhanced translocation of nanoparticles into renal tubules. Meanwhile, due to the steep pressure gradient along the nephron (50-10 mmHg) and the strong absorptive force of the peritubular capillaries, NPs have increased chances to interact with capillary endothelial cell membranes and can be effectively transported to renal tubule epithelial cells via the basolateral membrane^23,47. Secondly, the (KKEEE)₃K peptide serves as a renal tubule-targeting ligand that is specifically recognized and internalized by tubular epithelial cells, resulting in the high renal specificity and rapid accumulation of T-PAAD NPs in the kidney. Collectively, this enhanced renal residency is mediated by a composite mechanism of brush-border retention and cellular internalization, which overcomes the rapid tubular transit flow and allows the pathological urea microenvironment to trigger sustained and site-specific drug release.

Mitigating hyperglycemia and renal dysfunction by T-PAAD NPs/EMPA

CKM syndrome is a complex disorder comprising T2DM, kidney disease, cardiovascular disease, and obesity. These interconnected conditions share common risk factors and often exacerbate one another. To comprehensively investigate the therapeutic efficacy of T-PAAD NPs/EMPA against CKM syndrome in vivo, we established a CKM mouse model by feeding ApoE^−/− mice with WD for 16 weeks in combination with STZ injection. This model successfully mimicked the metabolic and pathophysiological features of CKM, including T2DM, renal dysfunction, cardiovascular disease, and obesity. Mice were randomly divided into three treatment groups: PBS (control), free EMPA, and T-PAAD NPs/EMPA (Fig. 4a). As a SGLT-2i, EMPA reduces renal glucose reabsorption, thereby lowering blood glucose levels. Following treatment, the blood glucose levels in CKM mice receiving T-PAAD NPs/EMPA dropped to 13.65 mmol L⁻¹, approaching levels observed in healthy controls, whereas free EMPA reduced blood glucose levels only to 20.60 mmol L⁻¹ (Fig. 4b). To further assess glucose homeostasis, intraperitoneal glucose tolerance test (IPGTT) and intraperitoneal insulin tolerance test (IPITT) were conducted after three weeks of treatment. The IPGTT results showed that the glycemic peak in CKM mice treated with T-PAAD NPs/EMPA was 19.75 mmol L⁻¹, which was 1.69- and 1.30- fold lower than that in the PBS and free EMPA groups, respectively. Furthermore, the area under the curves (AUCs) for both IPGTT and IPITT was significantly lower in the T-PAAD NPs/EMPA group, indicating improved glucose tolerance and insulin sensitivity (Fig. 4c, d and Supplementary Fig. 9a, b). However, the administration of blank T-PAAD NPs did not lead to any significant reduction in blood glucose levels or improvement in glucose tolerance (IPGTT) compared to the untreated CKM group (Supplementary Fig. 10). In addition, hemoglobin A1c (HbA1c), a key indicator of long-term glycemic control and prognostic marker for diabetes, decreased by 0.95% in the T-PAAD NPs/EMPA group, compared to only a 0.46% reduction in the free EMPA group (Supplementary Fig. 9c). Beyond enhancing urinary glucose excretion, T-PAAD NPs/EMPA also promoted natriuresis and osmotic diuresis without causing significant urinary tract infections (Supplementary Fig. 11).

Fig. 4. In vivo glycemic management and nephroprotective efficacy of T-PAAD NPs/EMPA.

a Schematic illustration of CKM model establishment and evaluation of T-PAAD NPs/EMPA treatment. Created in BioRender. Xue, R. (2026) https://BioRender.com/67vdpxr. b Dynamic changes of random blood glucose were continuously measured in CKM mice at indicated time points from 0 to 24 h (n  =  4 biologically independent samples). IPGTT (c) and IPITT (d) performed after 4 weeks of T-PAAD NPs/EMPA treatment (n  =  4 biologically independent samples). e Schematic illustration of the GFR evaluations. The transdermal mini-GFR monitor was employed to detect fluorescence signals following intravenous FITC-sinistrin administration in mice for glomerular filtration assessment, demonstrating rapid tracer clearance in normal renal function versus prolonged retention and reduced GFR in renal impairment. Created with BioRender.com. f Dynamic change in glomerular filtration rate at different times (n  =  3 biologically independent samples). g Histological PAS staining (scale bars, 50 µm), H&E staining (scale bars, 50 µm), and Masson’s trichrome staining (scale bars, 50 µm) of kidney slices from mice after different treatments. h TUNEL staining (Green fluorescence represents cellular apoptosis) and fluorescence images of KIM-1(red) and 4’,6-diamidino-2-phenylindole (DAPI, blue) of kidney tissues from different groups. Scale bars, 50 µm. i Color Doppler imaging of the kidney and pulsed wave Doppler imaging of blood flow in the right renal artery (Red indicates blood flow moving towards the probe, and blue indicates blood flow moving away from the probe). Data were shown as mean ±s.d., and analyzed using two-way ANOVA with Sidak multiple comparisons test (b–d) (*, CKM vs. T-PAAD NPs/EMPA; #, Free EMPA vs. T-PAAD NPs/EMPA.) and two-tailed Student’s t test (f), respectively. *^{, #}P  <  0.05, **^{, ##}P  <  0.01, ***^{, ###}P < 0.001, ****^{, #####}P < 0.0001. Source data are provided as a Source Data file.

Collectively, these findings confirm that targeted and controlled EMPA release from T-PAAD NPs to renal tubule cells plays a crucial role in glycemic regulation. This strategy not only achieves superior glycemic manage but also improves glucose tolerance and insulin sensitivity in CKM mice. These combined metabolic benefits contribute to reduced blood pressure, lower blood volume, decreased cardiac pre- and afterload, and enhanced renal hemodynamics, ultimately providing comprehensive cardio-renal protection.

Diabetic nephropathy (DN), a prevalent microvascular complication of diabetes, plays a pivotal role in the progression of CKM syndrome. Diabetes-induced kidney injury disrupts the cardiorenal axis, initiating bidirectional pathological crosstalk between the heart and kidneys⁴⁸. This mutual dysfunction further exacerbates systemic metabolic disturbances, ultimately contributing to the hallmark triad of CKM syndrome. Given this central role of DN, we evaluated the therapeutic efficacy of T-PAAD NPs/EMPA in mitigating diabetes-associated kidney damage. To assess renal function, glomerular filtration rate (GFR) was measured by transcutaneous monitoring of FITC-sinistrin clearance (Fig. 4e). The obvious reduction (24.6%) of GFR in untreated CKM mice indicated that there existed the severe kidney dysfunction in CKM mice. In contrast, EMPA treatments, especially T-PAAD NPs/EMPA, markedly preserved renal function. Notably, GFR in the T-PAAD NPs/EMPA group decreased by only 0.18%, suggesting significant renal protection (Fig. 4f and Supplementary Fig. 12). In addition to GFR, the urine albumin-creatinine ratio (UACR) was measured as a key indicator of kidney damage. As shown in Supplementary Fig. 13a, T-PAAD NPs/EMPA treatment effectively prevented albuminuria, reducing the UACR to a level 2.16-fold lower than that in untreated CKM mice. Furthermore, serum biomarkers of renal function, including creatinine (CRE), blood urea nitrogen (BUN), and uric acid (UA), were significantly improved in the T-PAAD NPs/EMPA group. Specifically, CRE, BUN, and UA levels were reduced to 39.2 μM, 11.9 mM, and 130.0 μM, respectively, closely approximating values observed in healthy controls (Supplementary Fig. 13b–d). In contrast, free EMPA treatment failed to achieve comparable improvements.

To further confirm the enhanced therapeutic effect of T-PAAD NPs/EMPA against DN, histopathological evaluations of kidneys were conducted following various treatments (Fig. 4g). Periodic acid-Schiff (PAS) staining revealed classic pathological features of DN in CKM mice, including glomerular enlargement, basement membrane thickening, and mesangial cell proliferation. Hematoxylin and eosin (H&E) staining further demonstrated serious tubular injury characterized by vacuolization, tubular dilation and atrophy, brush border loss, and thickened tubular basement membrane. However, these pathological alterations were significantly alleviated in both free EMPA and T-PAAD NPs/EMPA treatment groups. In particular, there appeared no distinguishable kidney pathological changes between the healthy mice and the T-PAAD NPs/EMPA-treated CKM mice. Moreover, Masson staining corroborated these findings, showing substantially reduced collagen deposition in the kidney after the T-PAAD NPs/EMPA treatment, indicative of effective amelioration of renal fibrosis. Quantitative analysis confirms a significant reduction in the fibrotic area, validating the enhanced anti-fibrotic efficacy of T-PAAD NPs/EMPA (Supplementary Fig. 14a). Furthermore, terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assays and immunostaining of kidney injury molecule-1 (KIM-1) were used to evaluate renal tubular apoptosis and injury. Both parameters were markedly elevated in untreated CKM mice but significantly reduced following EMPA-based therapies. Notably, T-PAAD NPs/EMPA treatment resulted in the greatest reduction in TUNEL-positive cells and KIM-1 expression (Fig. 4h and Supplementary Fig. 14b, c), highlighting its superior protective effect against renal tubular apoptosis and injury.

Taken together, these findings demonstrate that T-PAAD NPs/EMPA offer superior protection against kidney dysfunction in CKM mice compared to free EMPA. This enhanced therapeutic efficacy may be attributed to several key mechanisms. First and foremost, hyperglycemia is a major risk factor for kidney damage, and tight glycemic control is essential for renal protection. T-PAAD NPs/EMPA leverage a urea-triggered disassembly mechanism within the renal tubules to achieve spatiotemporally controlled release of EMPA, thereby enhancing drug accumulation at the site of SGLT-2 receptor expression in renal tubular epithelial cells. This precision targeted delivery strategy amplifies the blockade of SGLT-2, leading to more pronounced natriuretic and glycosuric effects and sustained urinary glucose excretion. Meanwhile, the improved glycemic control achieved by T-PAAD NPs/EMPA is expected to positively influence renal hemodynamics, which refers to the regulation of blood flow and pressure within the kidneys. As shown in Fig. 4i and Supplementary Fig. 15, color Doppler ultrasound was used to assess renal hemodynamic parameters, including the peak systolic velocity (PSV) and end-diastolic velocity (EDV) of the right renal artery in mice. The resistance index (RI) and pulsatility index (PI) were calculated to evaluate renal blood perfusion and the degree of kidney function impairment. Analysis of these parameters revealed significant abnormalities in PSV and EDV in CKM mice, along with a marked increase in PI and RI compared to the healthy control group, indicating elevated renal arterial blood flow resistance and reduced renal blood perfusion. However, following treatment with T-PAAD NPs/EMPA, renal hemodynamic parameters were largely restored to normal levels, suggesting that T-PAAD NPs/EMPA effectively preserved glomerular filtration function, modulated renal blood flow distribution, and alleviating renal dysfunction. Additionally, T-PAAD NPs could also scavenge the existed ROS in kidney. Thus, the synergistic effect of EMPA and T-PAAD NPs further normalized key oxidative stress markers (H₂O₂ and MDA), while enhancing renal antioxidant enzyme activity (catalase (CAT) and superoxide dismutase (SOD)), and improving total antioxidant capacity (T-AOC). These changes were accompanied by reduced renal inflammation and fibrosis, thereby offering enhanced protection against kidney damage (Supplementary Fig. 16).

Improving cardiovascular function by T-PAAD NPs/EMPA

The cardiovascular system and kidneys are vital for maintaining systemic homeostasis and are intricately linked through the bidirectional “cardio-renal axis”, which plays a central role in the regulation of blood pressure, hemodynamics, and metabolic balance^6,49. The functional interplay between these two organs is essential for sustaining energy homeostasis and overall physiological equilibrium⁵⁰. Based on this pathophysiological framework, we systematically evaluated the therapeutic potential of T-PAAD NPs/EMPA in improving cardiovascular function in CKM mice.

As shown in the Fig. 5a–c and Supplementary Fig. 17, CKM mice exhibited pronounced hemodynamic dysregulation compared with healthy controls, developing sustained hypertension characterized by elevated systolic blood pressure (SBP), diastolic blood pressure (DBP), and mean arterial pressure (MAP), reaching 142.29 mmHg, 95.26 mmHg, and 111.31 mmHg, respectively. Notably, after four weeks of treatment with T-PAAD NPs/EMPA, these values were significantly reduced to 117.85 mmHg (SBP), 79.30 mmHg (DBP), and 92.47 mmHg (MAP). This antihypertensive effect was superior to that of free EMPA, which reduced SBP, DBP, and MAP to 133.49, 84.69, and 99.84 mmHg, respectively. Similarly, multimodal cardiac imaging (M-mode echocardiography and pulsed Doppler flowmetry) quantitatively demonstrated the profound cardio-reparative effects of T-PAAD NPs/EMPA in CKM mice (Fig. 5d). The treatment of T-PAAD NPs/EMPA significantly restored systolic performance, elevating the ejection fraction (EF) from 41.5% to 53.4% (Fig. 5e) and improving fractional shortening (FS) by 8.1% (Fig. 5f) in comparison to the untreated CKM mice. In addition, the mitral ratio of peak early to late diastolic filling velocity serves as a critical hemodynamic biomarker for evaluating left ventricular compliance. Physiologically, the pressure gradient during early passive filling (E-wave) predominates over atrial contraction-driven flow (A-wave) in healthy myocardium, maintaining an E/A ratio >1.0. Pathological diastolic dysfunction fundamentally alters this dynamic, exhibiting characteristic E/A inversion (<1.0) due to impaired ventricular relaxation and elevated left atrial filling pressures. Thus, in CKM mice, echocardiographic quantification revealed severe diastolic impairment with E/A ratio depression to 0.91 (Fig. 5g). However, T-PAAD NPs/EMPA restored pseudo-normalized filling patterns, elevating E/A ratios to 1.49, demonstrating the capacity of T-PAAD NPs/EMPA to concurrently reverse the systolic and diastolic dysfunction of CKM mice. Collectively, these data demonstrate that T-PAAD NPs/EMPA effectively ameliorate cardiovascular abnormalities, attenuating pathological cardiac remodeling while helping to re-establish systemic circulatory homeostasis.

Fig. 5. Cardiovascular function of CKM mice after the treatment of T-PAAD NPs/EMPA.

SBP (a), DBP (b) and MAP (c) of CKM mice after four weeks of different treatments (n  =  5 biologically independent samples). Representative M-mode echocardiography images and pulsed-wave Doppler echocardiography images (d) and graphs showing EF % (e), FS % (f) and E/A ratio (g) measured by echocardiography (n  =  4 biologically independent samples). h Representative images of ORO stained en face aortas from mice after different treatments. i Representative images of aortic root sections stained with H&E, ORO and Masson’s trichrome, respectively. Scale bars, 500 μm. Quantitative data of plaque area of aortas (j) and collagen of plaques (k) in aortic roots sections using ORO and Masson’s staining, respectively (n  =  4 biologically independent samples). Data were analyzed using one­way ANOVA with Tukey’s post hoc tests (a–c, e–g, j, k) and shown as mean ± s.d. *P  <  0.05, **P  <  0.01 and ***P  <  0.001. Source data are provided as a Source Data file.

These benefits likely arise from two complementary mechanisms. Firstly, hemodynamic unloading: improved renal filtration and enhanced natriuresis lower systemic blood pressure and cardiac afterload, thereby reducing the hemodynamic stress that contributes to CKM-associated cardiac dysfunction. Secondly, systemic metabolic correction: T-PAAD NPs/EMPA markedly decreases circulating glucose and lipid levels, alleviating myocardial glucolipotoxicity and improving cardiac substrate utilization. In summary, by unloading the heart via improved renal function and blood pressure and correcting systemic metabolism via sustained glycosuria and lipid reduction, T-PAAD NPs/EMPA would improve the cardiovascular function.

Given the established role of hypertension in accelerating atherosclerotic plaque development through enhanced lipid deposition and endothelial injury, we further assessed the therapeutic efficacy of T-PAAD NPs/EMPA against atherosclerosis, which is a primary pathological basis of CVD and is closely related to the metabolic disorders underlying CKM syndrome. En face quantitative analysis of Oil Red O (ORO)-stained aortas demonstrated that T-PAAD NPs/EMPA significantly reduced atherosclerotic burden, decreasing the plaque area ratio from 38.29% in untreated CKM mice to 20.50% (Fig. 5h). Histopathological evaluation of aortic root sections supported these findings. H&E staining revealed a notable reduction in lesion area following T-PAAD NPs/EMPA treatment (Fig. 5i), while ORO staining of aortic roots showed a 1.39-fold greater reduction in plaque area compared to free EMPA, indicating the enhanced anti-atherosclerotic effect of the nanomedicine (Fig. 5j). Additionally, Masson’s trichrome staining indicated that collagen content within plaques increased to 49.21% in the T-PAAD NPs/EMPA group, which was 1.35 and 1.14 folds higher than those in the CKM and free EMPA groups, respectively (Fig. 5k). Since collagen deposition critically determined the integrity and stability of plaques, these data collectively suggest that T-PAAD NPs/EMPA not only attenuated plaque burden but also improved plaque stability, potentially reducing the risk of rupture and providing enhanced therapeutic benefits.

Alleviating hyperlipidemia of hepatic steatosis by T-PAAD NPs/EMPA

Hyperlipidemia increases the risk of cardiovascular and chronic kidney diseases through multiple mechanisms, including the promotion of atherosclerosis, endothelial injury, chronic inflammation, and oxidative stress. These factors collectively contribute to the progression of CKM syndrome^51,52. In addition to mitigating hyperglycemia, EMPA can also influence the lipid metabolism to manage hyperlipidemia, which is also crucial for treating CKM syndrome^53,54. Compared to free EMPA, T-PAAD NPs/EMPA exhibit superior blood lipid-lowering effects (Fig. 6a–d). In CKM mice, treatment with T-PAAD NPs/EMPA resulted in 1.68-, 1.60-, and 1.32-fold reductions in serum triglyceride (TG), total cholesterol (TC), and low-density lipoprotein cholesterol (LDL-C) levels, respectively. However, free EMPA failed to produce the comparable lipid-lowering effect. These results highlight the enhanced lipid-modulating capacity of T-PAAD NPs/EMPA and support their potential for effectively mitigating dyslipidemia in CKM mice.

Fig. 6. Alleviating hyperlipidemia of hepatic steatosis by T-PAAD NPs/EMPA in CKM mice.

Quantifications of blood triglyceride (a), total cholesterol (b) and LDL-C (c) (n  =  5 biologically independent samples). d Representative images of serum collected after different treatments. e Representative images of liver, H&E (scale bars, 200 µm) and ORO staining (scale bars, 100 µm) of fatty liver. f The proportion of liver/body weight (n  =  5 biologically independent samples). g Quantification of lipid area in ORO-stained fatty liver (n  =  5 biologically independent samples). h Gross appearance of epididymal white adipose tissues (eWAT) and interscapular brown adipose tissue (BAT) and representative H&E staining of eWAT and BAT. Scale bars, 100 µm. i The proportion of eWAT/body weight (n  =  4 biologically independent mice). Data were analyzed using one­way ANOVA with Tukey’s post hoc tests (a–c, f, g, i) and are shown as mean ± s.d. *P  <  0.05, **P  <  0.01, ***P  <  0.001 and ****P  <  0.0001. Source data are provided as a Source Data file.

Additionally, the liver plays a pivotal role as the metabolic command center for systemic lipid homeostasis, precisely regulating lipid synthesis, storage, and secretion. In CKM mice, this delicate balance is disrupted, leading to severe hepatic dysfunction characterized by ectopic lipid accumulation and dyslipidemia. To assess the therapeutic potential of T-PAAD NPs/EMPA in this context, its efficacy against hepatic steatosis was evaluated. As shown in Fig. 6e, f, the CKM mice developed obvious hepatomegaly, with liver/body weight ratio increasing to 8.96%, however, which was superiorly reversed by T-PAAD NPs/EMPA (6.26%) compared with free EMPA (6.83%). This was accompanied by the substantial decreases of serum ALT and AST levels, indicating the robust hepatoprotective effects of T-PAAD NPs/EMPA (Supplementary Fig. 18). Meanwhile, histological examination demonstrated that T-PAAD NPs/EMPA significantly resolved macrovesicular steatosis and achieved a 6.13% reduction in hepatic triglycerides (ORO-positive area), significantly outperforming the free EMPA treatment (2.03% reduction) (Fig. 6g).

Meanwhile, adipose tissue functions not only as an energy reservoir but also as a dynamic endocrine organ that plays a critical role in maintaining systemic metabolic homeostasis. In CKM mice, we observed substantial remodeling of adipose tissue architecture. Specifically, white adipose tissue (WAT) mass was reduced by 1.82% compared to the RD control group. Histological analysis of epididymal WAT (eWAT) revealed significantly smaller adipocytes, along with evidence of ectopic lipid deposition in visceral organs, reflecting severe metabolic dysregulation. Concurrently, brown adipose tissue (BAT) in CKM mice also exhibited increased lipid droplet accumulation compared to RD mice (Fig. 6h). Notably, T-PAAD NPs/EMPA exhibited the robust therapeutic efficacy, resulting in a 2.24-fold increase in eWAT mass recovery compared to the untreated CKM group and substantially outperforming the minimal 1.01-fold improvement observed with free EMPA treatment (Fig. 6i). Moreover, BAT morphology was restored following T-PAAD NPs/EMPA treatment, with lipid droplet size and distribution resembling those seen in healthy controls. Since BAT contributes to enhanced insulin sensitivity, glucose utilization, and overall metabolic regulation, these findings underscore the ability of T-PAAD NPs/EMPA to restore both the mass and morphology of WAT and BAT. These results demonstrate the therapeutic potential of T-PAAD NPs/EMPA for reversing systemic metabolic dysfunction associated with CKM syndrome.

Mechanistically, these effects likely stem from kidney-targeted delivery that increases local EMPA exposure in proximal tubules, thereby amplifying urinary glucose excretion. The resulting sustained glycosuric energy loss generates a stronger systemic negative energy balance than free EMPA, whose effective exposure can be attenuated by presystemic and hepatic metabolism. In response, metabolism may shift away from hepatic lipogenesis toward lipid mobilization and increased fatty acid oxidation. To rule out confounding by energy intake, we monitored food consumption and body weight throughout the treatment period (Supplementary Fig. 19). Daily food intake remained stable and comparable across groups, arguing against caloric restriction as a driver of the hepatic changes. Moreover, whereas untreated CKM mice showed pathological weight loss, mice treated with T-PAAD NPs/EMPA maintained physiologic weight stability. Together, these findings support that the benefits arise from targeted metabolic reprogramming rather than differences in energy intake or overall energy balance.

T-PAAD NPs/EMPA alleviating the CKM syndrome by reprogramming the cardio-renal energy metabolism

In CKM mice, extensive transcriptional dysregulation was observed in both kidney and heart (Fig. 7a–d). Volcano plots (Fig. 7a, c) illustrated differentially expressed genes when comparing CKM mice to healthy controls (RD group). In the kidneys of CKM mice, 1875 genes were significantly upregulated and 515 downregulated; while in the hearts, 511 genes were upregulated and 220 downregulated. Pathway enrichment analyses using Gene Ontology (GO) and Kyoto Encyclopedia of Genomes (KEGG) databases revealed strong activation of pro-diabetic and inflammatory pathways in CKM mice (Fig. 7b, d and Supplementary Fig. 20). Conversely, key metabolic pathways such as fatty acid degradation, pantothenate & CoA biosynthesis, and the PPAR signaling pathway, which were critical for lipid oxidation and peroxisomal fatty acid metabolism, were significantly downregulated in CKM mice. These pathways were closely related to energy metabolism, indicating that severe metabolic dysfunction and inflammation in the heart and kidney played an important role in CKM progression (Supplementary Fig. 21a, b). Consistent with these findings, Gene Set Enrichment Analysis (GSEA) showed downregulation of gene sets associated with oxidative phosphorylation and fatty acid metabolism, along with upregulation of gene sets associated with inflammatory response in CKM mice relative to the RD group (Fig. 7e). However, following the T-PAAD NPs/EMPA treatment, the RNA transcriptomic profiles shifted markedly toward a restorative pattern (Fig. 7f, h). T-PAAD NPs/EMPA treatment significantly downregulated 634 genes and upregulated 415 genes in the kidney, while 368 genes were downregulated and 121 upregulated in the heart. Notably, T-PAAD NPs/EMPA suppressed several pathological pathways that were elevated in the heart and kidney tissues of CKM mice, including those involved in pro-diabetic signaling, aberrant lipid metabolism, atherosclerosis, and IL-17-mediated inflammation (Fig. 7g, i). In parallel, beneficial pathways such as glutathione metabolism and drug metabolism were upregulated following treatment (Supplementary Figs. 21c, d, 22). Further GSEA results confirmed that T-PAAD NPs/EMPA treatment led to upregulated expression of genes involved in oxidative phosphorylation pathway and downregulated expression of genes related to inflammatory response in both heart and kidneys when compared with untreated CKM mice (Fig. 7j). Collectively, the observed transcriptomic reprogramming supported that T-PAAD NPs/EMPA effectively attenuates inflammation while reactivating energy metabolism pathways at the gene expression level.

Fig. 7. Cardiac and renal multi-omics analysis of CKM mice treated using T-PAAD NPs/EMPA.

a–j Transcriptomics analysis in the CKM mice. Volcano plot depicting differentially expressed genes (DEGs) in kidney and heart tissues between the CKM group and RD group (a, c), DEGs between the T-PAAD NPs/EMPA group and CKM group (f, h) and corresponding KEGG pathway enrichment analysis of upregulated differentially expressed genes between CKM group and RD group (b, d) and downregulated differentially expressed genes between T-PAAD NPs/EMPA group and CKM group (g, i). The horizontal axis represents the enrichment score, the vertical axis represents description and bubble sizes represent enriched gene counts. Cardd. cardiovascular disease, Cegad. cell growth and death, Diges. digestive system, Enamd. endocrine and metabolic disease, Immus. immune system, Signt. signal transduction, Simai. signaling molecules and interaction. GSEA was applied to compare the set of genes involved in oxidative phosphorylation, fatty acid metabolism, and inflammatory response in kidney and heart tissues between CKM and RD groups (e) and between T-PAAD NPs/EMPA and CKM groups (j). k–n Untargeted metabolomic analysis in the CKM mice with different treatments. Heatmap analysis of differential metabolites of kidney and heart in the T-PAAD NPs/EMPA group compared with the CKM group (k, m) and corresponding KEGG pathway enrichment analysis (l, n). ATP levels (o, q) and FAO activity (p, r) in kidney and heart tissues from CKM mice after different treatments were measured (n = 5 biologically independent samples). Data were analyzed using negative binomial Wald test with Benjamini-Hochberg False Discovery Rate (FDR) correction (q < 0.05) (a, c, f, h), hypergeometric test (b, d, g, i, l, n), Normalized Enrichment Score (NES) and FDR correction (e, j) and one­way ANOVA with Tukey’s post hoc test (o–r) and are shown as mean ± s.d. *P  <  0.05, **P  <  0.01, ***P  <  0.001. Source data are provided as a Source Data file.

To determine whether the observed transcriptional changes translated into downstream metabolites alterations, the untargeted metabolomic analysis were performed. Partial least squares Discriminant Analysis (PLS-DA) revealed that the untreated CKM mice formed a metabolically distinct cluster, clearly separated from the healthy control group. This separation reflected the profound metabolic disruptions caused by CKM syndrome. Importantly, the metabolic profiles of the free-drug group closely resembled those of the CKM group, whereas T-PAAD NPs/EMPA-treated group showed pronounced shifts, indicating that T-PAAD NPs/EMPA treatment substantially remodeled renal and cardiac metabolic signatures (Supplementary Fig. 23). Heatmaps and KEGG enrichment analysis further confirmed that T-PAAD NPs/EMPA significantly reprogrammed key metabolic pathways. In the kidneys, T-PAAD NPs/EMPA increased metabolites involved in glycerophospholipid metabolism and arginine metabolism, while decreasing those associated with glycolysis (Fig. 7k). KEGG pathway enrichment analysis further confirmed that arginine metabolism, glycolysis/gluconeogenesis, and lipid metabolism were among the most significantly altered pathways (Fig. 7l). These findings suggest that the energy metabolism of the kidney was substantially reprogrammed, consistent with transcriptomic evidence of restored metabolic gene expression. Similarly, in the heart, T‑PAAD NPs/EMPA treatment corrected abnormalities in lipid metabolism. The levels of free fatty acids and triglyceride derivatives shifted toward those observed in healthy controls (Fig. 7m and Supplementary Fig. 24). KEGG enrichment analysis (Fig. 7n) further identified fatty acid biosynthesis and glycerolipid metabolism as key pathways affected by the T‑PAAD NPs/EMPA treatment. Together, these coordinated metabolic improvements in both the kidney and heart provide strong evidences that T‑PAAD NPs/EMPA effectively reprograms the cardio-renal energy metabolism in CKM mice.

To further validate these findings, key metabolic readouts were quantitatively assessed to reinforce the therapeutic reprogramming effects mediated by T‑PAAD NPs/EMPA. As shown in Fig. 7o, q, CKM mice exhibited significantly reduced ATP levels in both heart and kidney, reflecting impaired energy production. In contrast, T‑PAAD NPs/EMPA treatment markedly increased ATP content in kidneys, nearly restoring it to the levels observed in healthy controls. A similar improvement in ATP levels was observed in heart, suggesting a systemic enhancement of energy metabolism. Fatty acid oxidation (FAO), a primary source of ATP in both renal and cardiac tissues under physiological conditions, was also evaluated. As shown in Fig. 7p, r, FAO activity was significantly suppressed in untreated CKM mice, consistent with the transcriptomic evidence of downregulated PPAR signaling and fatty acid catabolism (Supplementary Fig. 21a). Notably, T‑PAAD NPs/EMPA treatment restored the FAO capacity in both kidney and heart, indicating improved mitochondrial oxidative function. Given that FAO serves as the dominant energy source in normal cardiac and renal physiology, these results provide compelling functional evidence that T‑PAAD NPs/EMPA enhances therapeutic outcomes in CKM by reprogramming cardio-renal energy metabolism.

Overall, these findings demonstrate that T-PAAD NPs/EMPA therapy effectively reprogramed both the transcriptional and metabolic landscapes in the heart and kidneys of CKM mice. The treatment suppressed pro-inflammatory and diabetic signaling while restoring energy metabolism. This integrated multi-omics evidence strongly supports the conclusion that T-PAAD NPs/EMPA mitigates CKM syndrome by reprogramming the cardio-renal energy homeostasis.

Biocompatibility evaluation in vivo

Prior to in vivo toxicity assessment, the hemocompatibility of T-PAAD NPs was rigorously evaluated. Hemolysis assays performed at 2, 4, and 6 h revealed that the hemolysis rates remained consistently below the 5% safety threshold across all tested concentrations (up to 400 μg mL⁻¹), with no visible erythrocyte rupture observed in the supernatants (Supplementary Fig. 25). These results demonstrate the excellent hemocompatibility of T-PAAD NPs, supporting their suitability for systemic administration. Finally, to further evaluate the biocompatibility of T-PAAD NPs, a single high-dose injection (500 mg kg⁻¹, ten-fold the therapeutic dose) was administered to healthy C57BL/6J male mice (n = 6) to evaluate acute in vivo toxicity. Over a 14-day monitoring period, body weight fluctuations were negligible (Supplementary Fig. 25). At the end of the study, H&E staining of major organs revealed no signs of necrosis, hemorrhage, or tissue damage in the heart, liver, spleen, lungs, or kidneys. Additionally, liver and kidney function indicators, as well as routine blood parameters, remained within normal physiological ranges and showed no significant differences between treated and control groups. These results demonstrate the excellent biosafety and non-toxicity of T-PAAD NPs, supporting their potential for clinical translation (Supplementary Fig. 26).

Discussion

In this study, we developed a novel therapeutic strategy based on a urea-responsive, kidney-targeted nanoplatform designed for site-specific delivery and controlled release of EMPA in the renal tubules. This approach capitalizes on two key pathological features of CKM, namely elevated tubular urea levels and oxidative stress, to achieve localized drug activation and synergistic ROS scavenging. By simultaneously addressing disease specific microenvironmental cues and overcoming pharmacokinetic challenges, this system enhances the therapeutic performance of SGLT2 inhibitors through coordinated mitigation of oxidative stress, improved glycemic control, and restoration of renal and cardiovascular function in the mice with CKM. Treatment with T-PAAD NPs/EMPA significantly improved glucose tolerance, reduced hyperglycemia, and alleviated diabetic nephropathy, thereby offering a comprehensive approach to CKM management. Moreover, targeted delivery of EMPA also provide renal protection, superior blood pressure regulation, and favorable remodeling of adipose and hepatic tissues. Multi-omics analyses further revealed that T-PAAD NPs/EMPA reprogrammed energy metabolism in both cardiac and renal tissues, suppressed inflammation responses, and restored systemic metabolic homeostasis. Collectively, these findings highlight the promise of integrating nanomedicine with established pharmacological agents to achieve safer and more effective treatment of the complex, multi-organ diseases such as CKM. Despite these encouraging results, several limitations should be acknowledged. The kidney is an exceptionally complex and highly dynamic organ, exhibiting pronounced spatial and temporal gradients in solute composition, tubular flow, cellular phenotypes, and pathological remodeling along the nephron. While the present study demonstrates urea-responsive and kidney-preferential drug delivery under defined experimental conditions, the in vivo drug release process within the renal microenvironment is likely governed by the combined effects of tubular fluid dynamics, epithelial interaction, cellular uptake, and pathological heterogeneity, which cannot be fully recapitulated by current in vitro models. Future studies will therefore focus on developing more physiologically relevant experimental platforms to more precisely model renal microenvironments and to dissect the spatiotemporal behavior of nanocarrier disassembly and drug release within the kidney. Such efforts may include advanced microfluidic systems, refined ex vivo renal models, and high-resolution in vivo tracking approaches. By further integrating kidney-specific physiology into nanomedicine design and evaluation, this strategy may be optimized and extended to broader renal and cardio-metabolic disorders. In summary, this work presents a multifunctional nanotherapeutic platform that combines targeted drug delivery with metabolic reprogramming, offering a compelling strategy for the holistic management of CKM syndrome.

Methods

Materials

Acrylamide (AAm), Acrylonitrile (AN), DATS, 4,4’-Azobis(4-cyanovaleric acid) (ACVA), 1-ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDC), N–hydroxysuccinimide (NHS), Hoechst 33342, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT), 2′,7′- dichlorofluorescein diacetate (DCFH-DA), streptozotocin (STZ), ORO, empagliflozin (EMPA), coumarin 6 (C6), and IR780 were purchased from Sigma-Aldrich. NH₂-PEG2000-COOH were purchased from Titan. (KKEEE)₃K peptide were purchased from ChinaPeptides (QYAOBIO). Antibodies: AQP-1 (1:200, Catalog #PA5-78805), KIM-1 (1:600, Catalog #PA5-98302) were obtained from Thermo Fisher Scientific.

Cells and animals

HK2 cells (derived from adult male human kidney) were acquired from the Shanghai Cell Bank of the Chinese Academy of Sciences (China). Male apolipoprotein E-deficient (ApoE^−/−) mice (about 6 weeks old) and C57BL/6J mice (about 6 weeks old) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd (China). Mice were housed under a 12/12-h light/dark cycle at a constant temperature (22 °C) and humidity (50–60%), with free access to food and water ad libitum. All animal experimentation procedures were ethically approved by the Animal Experimentation Ethics Committee of School of Life Science and Technology in Xi’an Jiaotong University (approval no. XJTUSLST-2023-54).

Synthesis and characterization of PEG-PAAD copolymer

The PAAD copolymer was synthesized via free radical polymerization. Specifically, AAm (502.9 mg), AN (300 mg), DATS (63.1 mg), and ACVA (12.1 mg) were dissolved in N, N-dimethylformamide (DMF) (4 mL) at a monomer feed molar ratio of AAm: AN: DATS = 5:4:0.25. The solution was deoxygenated through three cycles of “freeze–vacuum–thaw”, with each cycle lasting 30 min and maintaining a constant vacuum pressure of 6×10⁻² Pa. The mixture was subsequently stirred at a constant temperature of 70 °C for 24 h. Upon completion of the reaction, the resulting copolymer (isolated with an approximate yield of 51.8%) was precipitated into an excess of cold diethyl ether to remove unreacted monomers and oligomers, followed by collection via centrifugation.

Subsequently, PAAD (100 mg) and NH₂-PEG-COOH (Mw:2000, 12 mg) were dissolved in dimethyl sulfoxide (DMSO) (3 mL), followed by the addition of NHS (0.7 mg), EDC (1.2 mg), and triethylamine (TEA) (0.6 mg). The reactant molar ratio was PAAD: NH₂-PEG-COOH = 1:1.2. The mixture was stirred at 50 °C for 24 h. The product was then purified by dialysis against ultrapure water for 48 h, utilizing a dialysis membrane with a Molecular Weight Cut-Off (MWCO) of 3500, which was sufficient to ensure the removal of organic solvents and unreacted reagents. Finally, the solution was lyophilized to obtain the PEG-PAAD copolymer (isolated with an approximate yield of 50%).

The synthesized polymers PAAD and PEG-PAAD were characterized using a Bruker Avance nuclear magnetic resonance spectrometer (400 MHz) for ¹H-NMR analysis. The average molecular weight (Mw) of the polymer was determined through gel permeation chromatography (Agilent PL-GPC 50) at a column temperature of 45 °C, employing DMSO as the mobile phase at a flow rate of 1 mL min⁻¹. Fourier transform infrared spectroscopy (FTIR) were recorded through the KBr pellet method using a FTIR spectrometer (Thermo Scientific Nicolet iS50). The UCST of polymer was measured using a variable-temperature ultraviolet spectrometer (Purkinje T6-1610F) with the wavelength set at 650 nm. The analysis began at 4 °C with a heating rate of 0.5 °C min⁻¹. Transmittance values were recorded at various temperatures to calculate the UCST, which was defined as the temperature when the transmittance reached 50%. The Tyndall effect of the polymer PEG-PAAD aqueous solution was observed via laser irradiation with the wavelength set at 650 nm.

Molecular dynamics simulation

To investigate the potential disruption of zipper-like intramolecular hydrogen bonds in PEG–PAAD copolymers by urea, density functional theory (DFT) calculations were performed using the Vienna Ab initio Simulation Package (VASP). To mimic the physiological aqueous environment, the implicit solvent model VASPsol was employed to account for dielectric screening and solvent polarization effects. The exchange-correlation interactions were described using the generalized gradient approximation of Perdew–Burke–Ernzerhof (GGA-PBE). The projector augmented-wave (PAW) method was used to treat the interactions between ion cores and valence electrons. A plane-wave cutoff energy of 500 eV was adopted. The structural models were fully relaxed until the Hellmann–Feynman forces on each atom were below 0.02 eV/Å and the total energy change between successive steps was less than 1 × 10⁻⁵ eV. During geometry optimization, the Brillouin zone was sampled using a Γ-centered 3 × 3 × 3 k-point grid, which was determined based on systematic convergence tests. Grimme’s DFT-D3 method was applied to account for dispersion interactions in the urea–polymer systems. Subsequently, urea-induced changes in hydrogen bond populations and binding energies within the polymeric system were systematically quantified.

ROS scavenging assays of PEG-PAAD copolymer

Measurement of H₂O₂ scavenging: H₂O₂ scavenging capacity of PEG-PAAD was tested by the hydrogen peroxide detection kit (Nanjing Jiancheng Bioengineering Institute). H₂O₂ reacts with ammonium molybdate to form a stable yellow complex, which displays an absorbance peak at 405 nm. Various concentrations of PEG-PAAD (0.125–1 mg mL⁻¹) were incubated with H₂O₂ at 37 °C, respectively. After reaction, the concentration of remaining H₂O₂ was determined according to the manufacturer’s instructions, and the H₂O₂-eliminating capacity was calculated.

Measurement of O₂^•− scavenging: The superoxide anion (O₂^•−) scavenging activity was assessed using a superoxide anion assay kit (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer’s instructions. Different concentrations of PEG-PAAD (0.125–1 mg mL⁻¹) were added to the working solution. The absorbance at 550 nm was measured using a microplate reader (Tecan M200) after standing for 10 min.

Measurement of ·OH scavenging: The scavenging activity of PEG-PAAD against ·OH was determined based on the effective removal of ·OH originating from the Fenton reaction. In detail, the working test solutions containing 500 μM 3,3′,5,5′-tetramethylbenzidine (TMB), 2 mM H₂O₂, 1 mM FeSO₄ and different concentrations of PEG-PAAD (0.125–1 mg mL⁻¹) in sodium acetate buffer (0.5 M, pH 4.5) were prepared in the dark and rest for 5 min. Afterwards, the absorbance at 630 nm of the solution was monitored using a microplate reader (Tecan M200).

Measurement of ABTS radical scavenging: The radical scavenging capacity of PEG-PAAD was evaluated using an ABTS radical scavenging capacity assay kit (Beyotime Biotechnology). Briefly, ABTS solution was incubated with potassium persulfate overnight to generate ABTS radicals. PEG-PAAD at various concentrations (0.125–1 mg mL⁻¹) were mixed with ABTS radical solutions and incubated for 10 min. The absorption of ABTS radicals at 734 nm was measured. The scavenging efficiency of ABTS radicals was calculated by the following equation:

$$\mathrm{ABTS}\mathrm{scavenging}\mathrm{efficiency}\left(\%\right)=\left(\left({\mathrm{A}}_{\mathrm{ABTS}}-{\mathrm{A}}_{\mathrm{Sample}}\right)/{\mathrm{A}}_{\mathrm{ABTS}}\right)\times 100\%$$ 1

where A_ABTS represented the absorbance of ABTS without additional processing. A_Sample represented the absorbance of ABTS with the addition of PEG-PAAD.

Synthesis and characterization of PEG-PAAD-(KKEEE)₃K copolymer

The PEG-PAAD (50 mg) was dissolved in DMSO (3 mL), and EDC (0.6 mg) and NHS (0.4 mg) were added for activation of the terminal carboxyl groups by stirring for 4 h. Subsequently, the (KKEEE)₃K peptide (5.7 mg) and TEA (0.3 mg) were added, and the reaction was carried out at 35 °C under 300 rpm stirring for 20 h. The final product was purified by dialysis against ultrapure water for 48 h, utilizing a dialysis membrane with a Molecular Weight Cut-Off (MWCO) of 3500, and lyophilized to obtain the PEG-PAAD-(KKEEE)₃K copolymer. The conjugation of (KKEEE)₃K peptides to PEG-PAAD copolymer was detected using the bicinchoninic acid (BCA) assay.

Preparation and characterization of T-PAAD NPs/EMPA

The amphiphilic polymers PEG-PAAD and PEG-PAAD-(KKEEE)₃K (10% w/w), along with empagliflozin (EMPA), were dissolved in DMSO (250 μL). The mass ratio of the drug to the polymer was set at 2:5 (w/w). The mixture was then added dropwise into an aqueous solution under probe sonication at 40% amplitude for 1 min with a pulse cycle of 3 s on and 3 s off to ensure uniform dispersion. The resulting emulsion was dialyzed using a dialysis membrane (MWCO 3500 Da) against a large volume (1 L) of PBS to thoroughly remove the organic solvent. The particle size, zeta potential, and polydispersity index (PDI) of the nanoparticles were determined using dynamic light scattering (Malvern Zetasizer Nano ZS90), while their morphology was characterized by transmission electron microscopy (Thermo Fisher Scientific FEI Talos F200C).

The encapsulation efficiency and drug loading of EMPA were measured by high-performance liquid chromatography (HPLC) using an Agilent 1100 system. A C18 column (250 mm × 4.6 mm, 5 μm) was maintained at 30 °C and eluted with a mobile phase consisting of acetonitrile (32%) and 0.01% (V/V) trifluoroacetic acid (68%) in an isocratic elution manner at a flow rate of 1 mL min⁻¹. The content of EMPA was quantified by monitoring the UV signals at 225 nm.

In vitro drug release profiles

To investigate the in vitro drug release profiles, T-PAAD NPs/EMPA were incubated at 37 °C in media simulating normal plasma/tissue (5 mM Urea) or pathological kidney injury conditions (10 and 50 mM Urea). To accurately separate the released drug and minimize non-specific adsorption, the samples were processed using Millipore Amicon Ultra centrifugal filters (100 kDa MWCO) with low-binding regenerated cellulose (RC) membranes. At predetermined time intervals, the samples were centrifuged to collect the filtrate, and the retentate was replenished with fresh solution of the same volume. The amount of EMPA released was determined using HPLC (Agilent 1100 system).

In vitro cellular uptake assessment

The targeting ability of T-PAAD NPs was investigated by incubating HK2 cells that were pretreated with free (KKEEE)₃K peptide or left untreated with C6-loaded T-PAAD NPs to assess their targeting capacity. After a 4 h incubation, the cells were collected and assessed using flow cytometry (BD Biosciences) at an excitation wavelength of 488 nm. Moreover, to visualize the intracellular distribution of T-PAAD NPs, Hoechst 33342 was used to stain the nuclei of the T-PAAD NPs/C6-treated cells. Subsequently, imaging was performed using a Confocal Laser Scanning Microscope (CLSM Leica TCS SP8 STED 3X).

In vitro ROS scavenging assays

HK2 cells were cultured under the standard condition. For the protection of cells from H₂O₂ induced oxidative stress, different concentrations of T-PAAD NPs (12.5, 25, 50 μg mL⁻¹) were cultured with HK2 cells in the absence or presence of H₂O₂ (150 μM) for 24 h, and the relative cell viabilities were tested using the MTT assay. To qualitatively and quantitatively describe the content of the intracellular ROS under H₂O₂ stimulation, DCFH-DA was used to stain intracellular ROS. The HK2 cells incubated with H₂O₂ (150 μM) and different concentrations of T-PAAD NPs (12.5, 25, 50 μg mL⁻¹) for 24 h were washed twice with serum-free medium and then DCFH-DA dye working solution was added for another 30 min. Finally, the unbound DCFH-DA probe was removed by washing twice with a serum-free medium. Then, the above treated cells were evaluated using CLSM and flow cytometry, respectively.

To establish a pathophysiologically relevant in vitro model of CKM syndrome, we developed a co-stimulation system using high glucose and palmitate to mimic systemic hyperglycemia and hyperlipidemia in renal tubular epithelial cells. Specifically, serum-starved HK2 cells were pretreated with glucose (33 mM) and palmitate (0.3 mM) in combination with varying concentrations of T-PAAD NPs for 24 h following 6-h serum deprivation. Lipid peroxidation was quantitatively assessed by measuring malondialdehyde (MDA) levels using a colorimetric assay kit, thereby evaluating the antioxidative capacity of T-PAAD NPs against oxidative stress in HK2 cells.

Establishment of the CKM model

To establish a CKM syndrome mouse model, 8-week-old APOE^−/− mice were fed a Western diet (WD, 21% fat and 0.15% cholesterol) for 8 weeks, followed by daily low-dose STZ injections (50 mg kg⁻¹) for 5 consecutive days, with a 12-h fasting period prior to each injection. Four weeks after STZ treatment, the mice exhibited significantly elevated fasting blood glucose levels (>11 mM), along with pronounced polydipsia, polyuria, and polyphagia. Additionally, early signs of kidney injury emerged, evidenced by the onset of microalbuminuria and a slight increase in the urinary albumin-to-creatinine ratio (UACR), indicating the development of T2DM with early renal impairment. By eight weeks post-STZ treatment, the glomerular filtration rate (GFR) was markedly reduced, and urinary albumin levels were significantly elevated, reflecting progressive renal dysfunction. Moreover, the mice displayed prominent atherosclerotic plaque formation, elevated blood pressure, dyslipidemia, cardiac remodeling, and metabolic abnormalities. Collectively, these findings demonstrate that this model integrates the pathophysiological interactions among T2DM, CKD, and CVD, providing a simple and effective system for investigating the interrelationship among metabolic disturbances, cardiac, vascular, and renal dysfunction. This model offers a valuable platform for the comprehensive study of the onset and progression of CKM syndrome.

Biodistribution of T-PAAD NPs in CKM mice

T-PAAD NPs/IR780 (0.35 mg kg⁻¹ of IR780) were intraperitoneally injected to healthy or CKM mice, while the control group received an equivalent dose of free IR780. At 6 h post-injection, major organs including the heart, liver, spleen, lungs, and kidneys were harvested and subjected to fluorescence imaging analysis using an in vivo imaging system (Lumazone, USA).

For fluorescence images of kidney tissue slices, the T-PAAD NPs/C6 were intraperitoneally injected into CKM mice. At 6 h post-injection, the kidneys were collected, and the proximal tubules and their cell nucleuses were stained with AQP1-PE (Thermo Fisher Scientific, PA5-78805) and DAPI, respectively.

In vivo therapeutic efficacy of T-PAAD NPs/EMPA against CKM syndrome

Male ApoE^−/− mice at 8 weeks of age were randomly assigned to 5 groups (n = 5): regular diet group (RD), Western diet group (WD), CKM model group (CKM), free EMPA treatment group (Free EMPA), and T-PAAD NPs/EMPA treatment group (T-PAAD NPs/EMPA). The specific treatment protocol is shown in Fig. 4a. The free EMPA and T-PAAD NPs/EMPA (the dose of EPMA:10 mg kg⁻¹ body weight) were intraperitoneally administered every other day for 4 weeks. During the treatment period, the mice’s body weight, blood glucose, and blood pressure were regularly monitored. At the end of the treatment, the GFR, cardiac ultrasound, and renal hemodynamics were assessed to evaluate heart and kidney function, respectively. Finally, the mice were euthanized by cervical dislocation after anaesthetization with isoflurane, and the major organ tissues and plasma were collected for subsequent biochemical assays, histological analysis, and transcriptomic and metabolomic analysis.

Measurement of blood pressure

During the experimental period, systolic blood pressure (SBP), and diastolic blood pressure (DBP) were measured weekly using a noninvasive blood pressure system for mice (Softron Biotechnology BP-2010A). The mean arterial pressure (MAP) was calculated according to the following formula:

$$\mathrm{MAP}=1/3\times \left(\mathrm{SBP}+2\times \mathrm{DBP}\right)$$ 2

Intraperitoneal glucose tolerance test and insulin tolerance test

In the final week of treatment, all mice underwent intraperitoneal glucose tolerance test (IPGTT) and intraperitoneal insulin tolerance test (IPITT) following a 12-h fasting period to evaluate the long-term efficacy of T-PAAD NPs/EMPA in regulating glucose tolerance and insulin sensitivity. For the IPGTT, mice were administered an intraperitoneal injection of glucose (2 mg g⁻¹ body weight). Blood glucose levels were measured at 0 (pre-injection), 15, 30, 60, 90 and 120 min post-injection. For the IPITT, insulin was intraperitoneally administered at a dose of 0.55 U kg⁻¹ body weight, and blood glucose levels were measured at the same time points as in the IPGTT. Blood glucose concentrations were determined using glucose test strips (Roche Mannheim), and the area under the curve (AUC) was calculated using GraphPad Prism 9.0 software.

Echocardiographic measurement

Following isoflurane anesthesia, transthoracic echocardiography was performed using the Vevo LAZR-X Imaging System (VisualSonics). M-mode echocardiography measurements in the left ventricular long-axis view quantified: left ventricular internal diameter at end-diastole and end-systole (LVIDd, LVIDs). Left ventricular ejection fraction (LVEF) was calculated as:

$$\mathrm{LVEF}\left(\%\right)=\left(\mathrm{LVEDV}-\mathrm{LVESV}\right)/\mathrm{LVEDV}\times 100\%$$ 3

where LVEDV represent left ventricular end-diastolic volume, LVESV represent left ventricular end-systolic volume. While fractional shortening (FS) was derived as:

$$\mathrm{FS}\left(\%\right)=\left(\mathrm{LVIDd}-\mathrm{LVIDs}\right)/\mathrm{LVIDd}\times 100\%$$ 4

the early (E) and late (A) diastolic mitral flow velocities were measured by pulsed Doppler in the four-chamber view and the ratio of E/A was calculated.

Renal doppler ultrasonography

Renal blood flow was visualized using color Doppler mode, with red indicating flow toward the transducer and blue representing flow away. Pulsed-wave Doppler measurements of renal artery hemodynamics were performed. And Resistance Index (RI) reflecting distal vascular resistance was calculated using following formulation:

$$\mathrm{RI}=\left(\mathrm{PSV}-\mathrm{EDV}\right)/\mathrm{PSV}$$ 5

where PSV represent peak systolic velocity, EDV represent end-diastolic velocity.

Pulsatility Index (PI) evaluating arterial compliance and impedance was measured as:

$$\mathrm{PI}=\left(\mathrm{PSV}-\mathrm{EDV}\right)/\mathrm{MV}$$ 6

where MV represent mean velocity.

Glomerular filtration rate (GFR) measurements

The percutaneous GFR measurement technique (MediBeacon Mannheim) was used to assess the GFR in awake mice. 24 h prior to the experiment, dorsal fur was removed to ensure optimal signal acquisition. Before formal testing, the mice were briefly anesthetized with isoflurane, and the MediBeacon transdermal Mini-GFR monitor was securely attached to the shaved area of the skin using medical tape. Subsequently, FITC-sinistrin (70 mg kg⁻¹ body weight) was injected via the tail vein. The skin fluorescence signal was continuously monitored for ~1.5 h while the mice were allowed to move freely in their cages. Afterward, the device was removed under isoflurane anesthesia, and data were analyzed using MB Studio v.22 software (MediBeacon).

Biochemical measurements

Serum creatinine (CRE), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglycerides (TG), malondialdehyde (MDA), glycated hemoglobin (HbA1c), urinary microalbumin, catalase (CAT), hydrogen peroxide (H₂O₂), uric acid (UA), alanine aminotransferase (ALT), and aspartate aminotransferase (AST) levels were measured using reagent kits (Nanjing Jiancheng Bioengineering Institute, catalogue number: c011-2-1, A111-1-1, A113-1-1, A110-1-1, A003-1-2, A056-1-1, A028-2-1, A007-2-1, A064-1-1, C012-2-1, C009-3-1, C010-3-1) according to the manufacturer’s instructions, respectively. Serum urea nitrogen (BUN) and renal tissue total antioxidant capacity (T-AOC) and fatty acid oxidation (FAO) levels were determined using reagent kits (Elabscience, catalogue number: E-BC-K329-S, E-BC-K136-M, E-BC-K784-M) according to the manufacturer’s instructions, respectively. Renal ATP content was quantified using reagent kits (Beyotime Biotechnology, catalogue number: S0026). All biochemical analyses were performed following standardized operating procedure.

Histological and immunofluorescence analysis

Upon treatment completion, mice were euthanized and major organs (heart, kidneys, liver, aortas, epididymal white adipose tissue (eWAT), and brown adipose tissue (BAT)) were harvested for organ index calculation (organ-to-body weight ratio). Fresh tissues were fixed in 4% paraformaldehyde and processed for either paraffin-embedded sections or frozen sections. Comprehensive histological analyses included: Periodic acid-Schiff (PAS) staining for glycogen detection, hematoxylin and eosin (H&E) for cellular morphology, Masson’s trichrome for fibrosis quantification, and ORO for lipid droplet visualization. Apoptosis was assessed by a one-step terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay (Servicebio), while kidney injury molecule-1 (KIM-1) expression was localized through immunofluorescence staining using anti-KIM-1 antibody (Thermo Fisher Scientific, PA5-98302). Histological image analysis was semi-quantitatively performed using Image-Pro Plus 6.0 software.

Transcriptomics analysis

Cardiac and renal tissues were collected from each treatment group in the CKM mice (RD group, n = 3; CKM group, n = 3; T-PAAD NPs/EMPA treatment group, n = 3) for total RNA extraction. mRNA enrichment, fragmentation, reverse transcription, library construction, transcriptome sequencing, and data analysis were performed by Genergy Biotechnology. Libraries were subjected to high-throughput sequencing using the Illumina NovaSeq 6000 platform. Differentially expressed genes (DEGs) between sample groups were identified using DESeq2 software, with the threshold set at q-value < 0.05 and fold-change >2 or <0.5. Functional annotation and pathway enrichment analysis of DEGs were conducted using the GO and KEGG databases.

Metabolomics analysis

Samples of heart, kidney, and serum were collected from the CKM mouse model across different treatment groups (RD group, n = 3; CKM group, n = 3; T-PAAD NPs/EMPA treatment group, n = 3). After grinding and extraction, these samples were analyzed using gas chromatography-mass spectrometry (GC-MS) (Agilent 7890 A and Agilent 5975 C). Subsequent data analysis was performed using the gas chromatography-specific SIMCA analysis software (version 14.1). The criteria for differential metabolites selection were listed as following: a variable importance projection (VIP) value > 1.0 obtained through orthogonal partial least squares discriminant analysis (OPLS-DA), and a two-tailed Student’s t test p-value < 0.05.

Bio-safety analysis of T-PAAD NPs

The hemolysis tendency of blank T-PAAD NPs was also evaluated. Briefly, the erythrocytes at the density of 1 × 10⁷ cells mL⁻¹ were incubated with blank T-PAAD NPs at concentrations ranging from 1.5625 μg mL⁻¹ to 400 μg mL⁻¹ at 37 °C for 2, 4, and 6 h. PBS (pH 7.4, 0.01 M) and Triton X-100 (1%, v/v) solution were also selected as negative and positive control, respectively. The absorbance of hemoglobin at the wavelength of 540 nm in the supernatant was measured using the microplate reader (Tecan M200). The collected blood samples were analyzed for routine blood tests (Sysmex KX21 Sysmex).

Acute toxicity assessment of T-PAAD NPs

Twelve healthy C57BL/6J male mice were randomly allocated into two groups (n = 6): the control group received intravenous saline injection, while the mice in another group were administered T-PAAD NPs at ten-fold higher than the therapeutic concentration (500 mg kg⁻¹). Body weight and general health status were recorded every other day with ad libitum access to food and water throughout the 14-day observation period. Following euthanasia under anesthesia, comprehensive analyses were performed including complete blood count, hepatic function markers (ALT, AST), renal function parameters (BUN, CRE), precise organ weight measurements (heart, liver, spleen, lungs, kidneys) for organ index calculation, and histopathological examination via H&E staining of major organs.

Statistical analysis

All the statistical data were presented as the mean ± standard deviation (s.d.) from a minimum of three independent experiments. One-way analysis of variance (ANOVA) with Tukey’s post hoc test or two-way ANOVA with Sidak multiple comparisons test was used for multiple comparisons. Unpaired Student’s t test was used to analyze the difference between two groups. *P  <  0.05, **P  <  0.01, ***P  <  0.001 and ****P < 0.0001.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

Zhe Yang and Zhongmin Tian conceived the concept and directed the research. Xuechun Ren and Di Gao carried out material synthesis. Xuechun Ren, Di Gao, Rong Yun and Xinyang Liu performed material characterization and electrochemical tests. Xuechun Ren, Rong Yun, Chenna Di, Zeyu Hu and Xinyuan Zhang conducted the cell and animal experiments. Zhe Yang and Zhongmin Tian wrote the paper. All authors discussed the results and commented on the paper.

Peer review

Peer review information

Nature Communications thanks Sabu Thomas, Kengo Kidokoro, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

All data generated or analyzed during this study are included in this published article and its Supplementary Information. Source data are provided with this paper. The transcriptome sequencing data generated in this study have been deposited in the NCBI Sequence Read Archive under accession code PRJNA1390588 and its corresponding hyperlink: https://www.ncbi.nlm.nih.gov/sra/PRJNA1390588.

The metabolomic data of kidney and heart have been uploaded into the MetaboLights database (https://www.ebi.ac.uk/metabolights/) (Accession Number: MTBLS13988 (kidney)[https://www.ebi.ac.uk/metabolights/MTBLS13988], MTBLS13991(heart) [https://www.ebi.ac.uk/metabolights/MTBLS13991]). Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-71424-w.