Linagliptin increases the expression of LRP2 independent of Nrf2 in streptozotocin‐induced diabetic mice

ABSTRACT

Aims/Introduction

Oxidative stress is thought to play a role in the onset and progression of diabetic nephropathy (DN). Nuclear factor erythroid‐related factor 2 (Nrf2) is a master regulator of antioxidant and detoxification pathways and provides cellular protection against oxidative stress. This study aimed to examine the antioxidant effects of linagliptin, a dipeptidyl peptidase‐4 inhibitor, on DN using streptozotocin (STZ)‐induced diabetic Nrf2 knockout (Nrf2KO) mice.

Materials and Methods

Six‐week‐old male Nrf2KO mice and their wild‐type (WT) littermates were intraperitoneally injected with STZ (55 mg/kg) for five consecutive days to induce diabetes (DM) and/or treated with linagliptin (10 mg/kg/day) for 8 weeks. We also investigated the effects of linagliptin and high glucose concentrations on Nrf2‐mediated signaling in human renal proximal tubular epithelial cells (HRPTECs) using a NRF2‐specific siRNA.

Results

Despite the lack of effect on the pathological changes and higher blood glucose levels, linagliptin reduced urinary albuminuria in diabetic mice. Interestingly, linagliptin increased the expression of NADP(H): quinone oxidoreductase‐1 (NQO1) and LRP2 (megalin) in renal proximal tubular cells of WT and Nrf2KO mice, and these effects were accompanied by the activation of the AKT pathway. Linagliptin increased the expression of LRP2 under high glucose conditions through a mechanism independent of NRF2. Finally, an AKT siRNA abolished the stimulatory effects of linagliptin on LRP2 expression in HRPTECs.

Conclusions

These results suggested that linagliptin reduced albuminuria in individuals with DN in an NRF2‐independent manner and might exert its effects by upregulating LRP2 expression via the AKT pathway.

INTRODUCTION

Diabetic nephropathy (DN) is recognized as a common complication for patients who suffer from diabetes, and it can progress to end‐stage renal disease (ESRD). The four pillars of guideline‐directed medical therapies for chronic kidney disease (CKD) include renin–angiotensin system (RAS) inhibitors, sodium–glucose cotransporter 2 (SGLT2) inhibitors, nonsteroidal mineralocorticoid receptor antagonists (MRAs), and glucagon‐like peptide‐1 receptor agonists (GLP‐1Rs) based on accumulating evidence obtained from clinical trials. ¹ Dipeptidyl peptidase‐4 (DPP4) inhibitors are frequently used to treat patients with type 2 diabetes because they are effective oral glucose‐lowering agents with a low risk of severe hypoglycemia. Among them, linagliptin (LINA) is the only DPP4 inhibitor that is well tolerated in patients with renal impairment, as it is primarily cleared by hepatobiliary mechanisms rather than renal filtration. Therefore, patients with DN tend to be treated with linagliptin. With respect to DN, linagliptin showed the potential to decrease albuminuria in patients with CKD in the CARMELINA trial on the safety concerns associated with cardiovascular and kidney disease without affecting the cardiovascular or kidney risk. ² To our knowledge, no study has investigated the mechanism by which linagliptin decreases albuminuria in patients with type 2 diabetes.

Linagliptin has been shown to exert pleiotropic effects on streptozotocin (STZ)‐induced renal impairment that are independent of its glucose‐lowering properties. ³ , ⁴ , ⁵ , ⁶ Moreover, a beneficial renoprotective effect on diabetic DBA/2 J mice was observed and was shown to be mediated by other mechanisms involving antioxidant enzyme system‐dependent increases in the levels of the antioxidants catalase and MnSOD rather than through a decrease in blood glucose levels. ⁷ Previous studies have also reported that linagliptin activates nuclear factor erythroid‐derived 2‐like 2 (Nrf2) signaling, resulting in renoprotective effects on LPS‐induced acute kidney injury, ⁸ tacrolimus‐induced kidney injury ⁹ and endotoxin‐induced acute kidney injury ¹⁰ in rats. Notably, linagliptin protects diabetic kidneys by enhancing insulin receptor substrate 1 (IRS1)/Akt insulin signaling in podocytes, which is partially mediated by the Keap1/Nrf2 pathway. ⁶

The overproduction of mitochondrial reactive oxygen species (ROS) causes oxidative stress because of an imbalance between an increase in free radical production and a reduction in antioxidant defenses. ¹¹ Nrf2 is a master regulator that mediates the activities of antioxidant and detoxication enzymes to protect cells from oxidative stress. Nrf2 subsequently interacts with antioxidant‐responsive elements (AREs), promoting the activity of detoxication and antioxidant enzymes, such as NAD(P)H dehydrogenase, quinone 1 (NQO1) and heme oxygenase 1 (HO‐1), to ultimately decrease ROS activity. ¹² , ¹³ Nrf2 activation protects against STZ‐induced kidney damage and preserves renal function in diabetic mice by reducing oxidative stress. ¹⁴ , ¹⁵ A derivative of the synthetic triterpenoid bardoxolone methyl (CDDO‐methyl ester) and its analogs, which are potent inducers of the Nrf2/Keap1 pathway, were used to treat CKD associated with type 2 diabetes in a phase II trial (BEAM study). ¹⁶ An increase in the estimated glomerular filtration (eGFR) rate was observed in patients treated with this pharmaceutical Nrf2 activator as well as in a phase III trial (BEACON study); however, the phase III trial was discontinued because of the high rates of cardiovascular side effects and mortality. ¹⁷ Pharmacological approaches directly targeting Nrf2 ¹⁸ could be interesting, novel therapies for DN; however, these approaches have shown limited clinical efficacy thus far.

A lack of understanding of the effects of linagliptin on antioxidant mechanisms in the kidney limits the current therapeutic strategies for DN. Therefore, the aim of the current study was to investigate the protective effect of linagliptin on diabetic kidneys through the antioxidant Nrf2 pathway in STZ‐induced diabetic Nrf2 knockout mice.

MATERIALS AND METHODS

Animals and experimental design

All animal experiments conformed to the National Health Guide for the Care and Use of Laboratory Animals and were approved by the Research Center for Animal Life Science at Asahikawa Medical University (Approval No. 24–03, Date of Approval; March 19, 2024). Male C57BL/6J mice were purchased from The Jackson Laboratory Japan, Inc. (Yokohama, Japan) at 4 weeks of age. The Nrf2 knockout (Nrf2^−/−) mouse strain (RBRC01390; C57BL/6J background) was provided by RIKEN BioResource Center (Tsukuba, Japan). ¹⁹ Wild‐type littermates were used as controls in all the experiments. All the mice were housed with free access to food and water on a 12/12 h light/dark cycle. At 6 weeks of age, the male wild‐type mice (WT) and Nrf2 knockout mice (KO) were divided into two groups: a diabetic (DM) group intraperitoneally injected with 55 mg/kg STZ (Sigma, St. Louise, MO, USA) for five consecutive days and a control group injected with saline alone (WT, n = 14; KO, n = 10). At 8 weeks of age, the STZ‐induced diabetic mice in each group (DM‐WT, n = 26; DM‐KO, n = 24) were randomly assigned to one of two groups: the first group did not receive a pharmacological treatment and was used as the diabetes mellitus group (DM), and the second group received linagliptin (0.0075% linagliptin in chow; 10 mg/kg body weight/day) in a regular rodent diet (MF) for 8 weeks to examine the effects of linagliptin on the diabetic kidney (DM‐WT + LINA, n = 14; DM‐KO + LINA, n = 12) (Figure S1). Body weight (BW) was monitored every week throughout this study. Glucose levels in whole blood collected from the tail vein were quantified using a One Touch glucose analyzer (LifeScan Inc., Milpitas, CA, USA). HbA1c levels were assessed using a DCA 2000 analyzer (Siemens Medical Solutions Diagnostics, Tokyo, Japan). Systolic blood pressure (SBP) and mean blood pressure (MBP) were measured in conscious mice using an automated tail‐cuff manometer system (MK‐2000ST; Muromachi Kikai, Tokyo, Japan). The mice in each group were placed in metabolic balance cages for 24 h of urine collection. Renal function was assessed by measuring urinary albumin excretion (UAE) (Exocell, Philadelphia, PA, USA) and plasma cystatin C levels (Quantikine® ELISA (R&D Systems®, Inc.; Minneapolis, MN, USA)). Plasma insulin levels were evaluated with an insulin ELISA kit in accordance with the manufacturer's instructions (Morinaga, Yokohama, Japan). After treatment with linagliptin for 8 weeks, all the mice were sacrificed under inhalation anesthesia. The right kidney was fixed with formalin for the histological examination, and the left kidney was snap‐frozen in liquid nitrogen and stored at −80°C until use.

Cell culture

Human renal proximal tubular epithelial cells (HRPTECs) were obtained from Lonza, Inc. (Walkersville, MD, USA) and were cultured with renal epithelial cell growth medium (REGM, Lonza) on collagen type 1‐coated dishes at 37°C in an incubator containing 5% CO₂ and 95% humidified air. Cells were exposed to 5.5 mM glucose (control) or 33.3 mM glucose (high glucose; HG) in the absence (control) or presence of linagliptin (100 nM) overnight and were then harvested for experiments as previously described. ²⁰

Quantitative RT–PCR

Total RNA was extracted from both kidney tissues and cultured cells using an RNeasy mini kit (Qiagen, Tokyo, Japan). Complementary DNA (cDNA) was subsequently synthesized by reverse transcription (Thermo Fisher Scientific, Waltham, MA, USA). Quantitative real‐time PCR was performed on a 7,300 Real‐Time PCR System (Applied Biosystems, Waltham, MA, USA) using SYBR™ Select Master Mix (Thermo Scientific, Waltham, MA, USA) and TB Green® Premix Ex Taq™ II (Tli RNaseH Plus) (Takara Bio Inc., Kyoto, Japan) for mouse kidney tissues ²¹ and a fluorescent TaqMan 57‐nuclease assay and a sequence detection system (Prism 7,300, Applied Biosystems, Carlsbad, CA, USA) for HRPTECs, ²² as previously described. The ribosomal protein lateral stalk subunit P0 (RPLP0) was used as the internal control to normalize the mRNA levels of the genes of interest. The primer sequences are listed in Tables S1 and S2.

Histological examination

Four‐micrometer‐thick 10% neutral‐buffered formalin‐fixed and paraffin‐embedded kidney tissues were stained with periodic acid–Schiff (PAS). Histological changes in the kidney were defined by quantitative grading as described previously. ²³ The mesangial area was determined by assessing the PAS‐positive and nucleus‐free areas in the mesangium using Fiji ImageJ (US National Institutes of Health, Bethesda, MD, USA; ImageJ.net/Fiji, v.1.52). Tubular damage was scored by evaluating the percentage of injured tubules showing dilation or atrophy, brush border loss, cellular lysis, and cast formation in the cortex of the kidney.

Western blotting

Before the dissected kidneys were homogenized and sonicated on ice, all the tissues were added to RIPA lysis buffer (Abcam, Massachusetts, USA) supplemented with a protease inhibitor cocktail and phosphatase inhibitors (Sigma–Aldrich, Missouri, USA). Afterward, the samples were centrifuged at 10,000 rpm for 10 min at 4°C, and the supernatant was discarded. The protein concentration of each lysate was determined using the BCA method according to the manufacturer's instructions. Protein samples from the kidney were loaded onto 5%–20% SuperSep™ Ace SDS–PAGE gels (Fujifilm, Osaka, Japan) and then transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were incubated with the primary antibodies of interest (Table S3) overnight and placed in a refrigerator. Afterward, the blots were blocked with Block‐ACE (WakenBtech Co., Ltd., Kyoto, Japan) for 1 h and subsequently incubated with anti‐mouse or anti‐rabbit horseradish peroxidase‐conjugated secondary antibodies for 1 h at 4°C. Then, the membranes were washed three times with TBS containing 0.1% Tween. In the next step, all the membranes were visualized with chemiluminescence reagents (Millipore, Bedford, Massachusetts, USA), and the signals for immunoreactive bands were detected with a Molecular Imager® ChemiDoc™ XRS+ (Bio‐Rad, Hercules, California, USA).

Immunohistochemistry

Immunohistochemical staining was performed as previously described. ²⁴ The tissue sections were incubated with the primary antibodies listed in Table S3. The stained sections were observed and visualized with a light microscope system (BZ‐X810; Keyence Co., Osaka, Japan). Images of immunohistochemical staining were quantified with Fiji ImageJ (US National Institutes of Health, Bethesda, MD, USA, ImageJ.net/Fiji, v.1.52) as described in a previous study. ²³

Statistical analysis

The sample sizes for the animal studies were determined according to previous publications. ¹⁴ , ²⁵ Descriptive statistics are shown as the means ± standard errors of the means (SEMs). The mouse urinary albumin levels were log(e) transformed. Multiple comparisons between groups were performed using one‐way anova followed by Tukey's post hoc tests. The Kruskal–Wallis test was used to analyze the expression levels of all proteins. All analyses were performed using the GraphPad Prism ver. 7.0 software (Boston, MA, USA, https://www.graphpad.com). Differences were considered significant when the p value was less than 0.05. Exclusion criteria were based on animal well‐being at the beginning of the study. No animals were excluded from this study. No power analysis was performed to determine the sample size.

RESULTS

Linagliptin ameliorated urinary albuminuria in diabetic Nrf2‐knockout mice without improving hyperglycemia

STZ‐induced diabetic animals were treated with linagliptin for 8 weeks to evaluate the effects of linagliptin on DN in vivo (Table S4). STZ‐induced diabetes in both wild‐type (WT) and Nrf2 knockout (KO) mice, as shown by the higher HbA1c levels in these mice than in the control (Ctrl) mice, and linagliptin failed to lower the HbA1c levels in diabetic WT and KO mice (Figure 1a). STZ‐induced diabetes significantly decreased the body weight of the KO mice but not the WT mice (Figure 1b), and linagliptin did not affect the body weight of either group of diabetic mice (Figure 1b, Figure S2). Compared with the control treatment, STZ‐induced diabetes decreased plasma insulin levels in DM‐WT and DM‐KO mice by approximately 39% (P = 0.3895) and 50% (P < 0.05), respectively. Linagliptin tended to restore plasma insulin levels to 64% in DM‐WT mice, but the difference was not statistically significant (P = 0.3644) because of the wide variation. In contrast, linagliptin significantly decreased plasma insulin levels in Nrf2‐KO mice (P < 0.001; Figure 1c). Neither diabetes nor linagliptin treatment changed the systolic blood pressure (SBP) (Figure 1d). Linagliptin increased relative kidney weight in DM‐KO mice compared with Ctrl‐KO mice (Figure 1e, P < 0.01). Diabetes significantly induced polydipsia and polyuria in DM‐WT and DM‐KO mice (Figure 1f, g). Linagliptin failed to correct polydipsia (Figure 1f) and polyuria (Figure 1g) in DM‐WT mice and DM‐KO mice. Moreover, compared with the control mice, linagliptin treatment significantly decreased plasma cystatin C levels in DM‐WT (P < 0.05) and DM‐KO (P < 0.01) mice by 32% and 30%, respectively (Figure 1h). Linagliptin tended to reduce urinary albuminuria in DM‐WT and DM‐KO mice, but this difference was not significant because of the wide variation (Figure 1i, Table S4).

FIGURE 1.

Effects of linagliptin (LINA) on plasma cystatin C levels and urinary albumin excretion in wild‐type (WT) and Nrf2 knockout (Nrf2‐KO) diabetic mice. (a) HbA1c levels (mmol/mol); (b) body weight (g); (c) plasma insulin concentration (ng/ml); (d) systolic blood pressure (SBP) (mmHg); (e) total kidney weight (% of body weight); (f) water intake (ml/day); (g) urinary volume (ml/day); (h) plasma cystatin C concentration (mg/l); (i) urinary albumin excretion (μg/day). The values are presented as the means ± SEMs (n = 10–14 animals per group). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Linagliptin failed to ameliorate diabetes‐induced tubular injury in DM‐WT and DM‐KO mice

STZ did not significantly induce mesangial expansion in diabetic WT or Nrf2‐KO mice (Figure 2b). Linagliptin had no renoprotective effects on glomerular or tubular injury in DM‐WT or DM‐KO mice (Figure 2). Compared with DM‐WT mice, DM‐KO mice presented more severe renal tubular injury despite being treated with linagliptin (Figure 2a), but the difference between these mice was not statistically significant because of the wide variation (Figure 2c).

FIGURE 2.

Linagliptin (LINA) fails to ameliorate diabetic renal injury in both wild‐type (WT) and Nrf2 knockout (Nrf2‐KO) mice. (a) PAS (periodic acid–Schiff) staining of renal tissues. The upper panel shows a low‐magnification overview; scale bar 100 μm. The lower panel shows a high‐magnification view of the boxed region in the upper panel highlighting finer structural detail; scale bar 20 μm; quantitative analysis of mesangial expansion (b) and tubular damage (c) scores. The values are presented as the means ± SEMs (10–14 animals per group). ***p < 0.001 and ****p < 0.0001.

Linagliptin increased NQO1 protein expression without affecting Nqo1 mRNA expression in diabetic mice through a mechanism independent of Nrf2

Hyperglycemia, a potent inducer of oxidative stress, ²⁶ causes the onset and progression of DN. ²⁷ Nrf2 is known to be the master regulator of antioxidant responses. ²⁸ Because linagliptin presented a tendency toward attenuating albuminuria in DM‐KO mice, we investigated whether the Nrf2 pathway might be involved in the ability of linagliptin to attenuate albuminuria in DN. We examined the expression of Nrf2 and its target genes in the kidneys of mice using RT–qPCR (Figure 3). The expression of Nrf2‐regulated genes such as Ho1 and catalase (Cat) was decreased in the kidneys of Nrf2‐KO mice (Figure 3b, c), which was consistent with the decrease in Nrf2 expression (Figure 3a). Linagliptin did not affect the expression of Ho1 or Cat in the kidneys of Nrf2‐KO mice (Figure 3b, c). In contrast, the expression of Nqo1, a gene targeted by Nrf2, did not decrease in Nrf2‐KO mice. Moreover, compared with that in control Nrf2‐KO mice, Nqo1 expression in diabetic Nrf2‐KO mice was significantly increased by 57% (P < 0.05; Figure 3d). Western blotting and immunohistochemical analyses revealed that in diabetic mice, the expression of the NQO1 protein was independent of that of Nrf2 (Figure 3e, f). In addition, linagliptin tended to increase NQO1 protein expression in diabetic mice independent of Nrf2 (Figure 3e, f).

FIGURE 3.

Expression of Nrf2 and its target genes in DN mice treated with LINA. Quantitative real‐time PCR analysis of the mRNA expression of Nrf2 (a) and its target genes, heme oxygenase‐1 (Ho1) (b), catalase (Cat) (c), and NAD(P)H, quinone oxidoreductase 1 (Nqo1) (d) in the renal cortex (normalized to 36b4). (e) Western blots showing the levels of NQO1 and β‐Actin in the kidneys of mice. Each lane represents a total kidney lysate from each group. The bar graphs show normalized data for NQO1/β‐Actin. (f) Immunohistochemical staining showing NQO1 protein expression in mouse kidney tissues (original magnification, ×20; scale bars, 50 μm), and the data in the bar graphs show the results of the quantitative assessment of NQO1 protein expression. The values are presented as the means ± SEMs (n = 10–14 animals per group). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. A.U.: arbitrary unit.

Linagliptin increased LRP2 expression in diabetic mice

An increase in urinary albumin levels is an indicator of the progression of DN. ²⁹ A decrease in albumin reabsorption by the proximal tubule resulted in increased urinary albumin levels. ³⁰ The major endocytic receptor mediating protein uptake in the kidney is low‐density lipoprotein receptor‐related protein 2 (LRP2, megalin), which is extensively localized in the apical brush border of proximal tubular epithelial cells. ³¹ We explored the effects of linagliptin on decreased urinary albumin excretion in diabetic subjects by determining its effects on the levels of LRP2 mRNA and protein expression in mouse kidney tissues. Compared with no treatment, linagliptin increased renal levels of the Lrp2 mRNA (P = 0.06 in WT mice and P < 0.05 in Nrf2‐KO mice) (Figure 4a), which was accompanied by increased expression of the LRP2 protein in both linagliptin‐treated groups (Figure 4b). We also examined albumin staining in the kidney and quantified albumin staining as described in previous studies ³² , ³³ (Figure S3). In WT‐DM mice, albumin exhibited diffuse, relatively homogeneous intracellular staining that was confined to discrete tubular segments in the juxtamedullary region (Figure S3). By contrast, in linagliptin‐treated diabetic mice (WT‐DM + LINA), albumin displayed a punctate, granular intracellular pattern with heterogeneous intensity (Figure S3), consistent with focal accumulation within vesicular compartments. Consequently, linagliptin might modulate lysosomal degradation or trafficking pathways into the systemic circulation. These results suggest that linagliptin reduces urinary albumin levels in diabetic mice via the upregulation of renal LRP2 protein expression.

FIGURE 4.

Linagliptin increased LRP2 protein expression in mouse kidney tissues via increased Akt phosphorylation independent of Nrf2. (a) Relative mRNA levels of Lrp2 in mouse kidney tissues (n = 10–14 mice/group). (b) Immunohistochemical staining showing LRP2 protein expression in renal tissues (original magnification, ×20; scale bars, 50 μm), and the bar graphs show the results of the quantitative assessment of LRP2 protein expression. (c) Western blots showing the levels of phosphorylated AKT (p‐AKT), total AKT (t‐Akt) and β‐Actin. The bar graphs show normalized data for p‐AKT/t‐AKT. The values are presented as the means ± SEMs. *p < 0.05, ***p < 0.001 and ****p < 0.0001. A.U.: arbitrary unit.

Linagliptin increased LRP2 expression by phosphorylating AKT

On the basis of prior evidence implicating Akt in the control of LRP2 expression, ³⁴ , ³⁵ , ³⁶ we examined the effect of linagliptin on phospho‐AKT (p‐AKT) levels using Western blotting to elucidate the mechanism by which linagliptin increases LRP2 expression in kidney tissues from diabetic mice. Linagliptin significantly increased p‐AKT levels in diabetic WT mice (P < 0.05) and tended to increase p‐AKT levels in diabetic Nrf2‐KO mice (Figure 4c).

We confirmed the roles of Nrf2 and AKT in LRP2 expression by studying the effects of NRF2 and AKT siRNAs on LRP2 expression in HRPTECs (Figure 5). The NRF2 siRNA efficiently decreased NRF2 expression, but linagliptin treatment did not result in any additional change (Figure 5a). Although the expression of the NRF2‐targeted gene NQO1 was decreased by the NRF2 siRNA, linagliptin tended to increase NQO1 expression independent of NRF2 (Figure 5b). In addition, linagliptin significantly increased LRP2 expression under high glucose conditions with NRF2 ablation (Figure 5c). Finally, the AKT1 siRNA abolished the effect of linagliptin on LRP2 expression (Figure 5d), suggesting that linagliptin increased LRP2 expression via AKT phosphorylation accompanied by hyperglycemia, resulting in decreased urinary albumin excretion in diabetic mice (Figure 5e).

FIGURE 5.

Linagliptin increased LRP2 expression under high glucose conditions via AKT signaling independent of NRF2 in human renal proximal epithelial cells (HRPTECs). (a) The Nrf2 siRNA efficiently decreased NRF2 expression in HRPTECs. (b) The expression of a Nrf2‐targeted gene, NQO1, decreased in cells transfected with the NRF2 siRNA, and linagliptin treatment tended to increase NQO1 expression in cells cultured under high glucose conditions. (c) The Nrf2 siRNA decreased LRP2 expression, but linagliptin treatment significantly increased LRP2 expression under high glucose conditions. (d) The AKT1 siRNA did not affect LRP2 expression and abolished the stimulatory effect of linagliptin on LRP2 expression. (e) Schematic of the mechanism by which linagliptin decreases urinary albumin excretion in individuals with diabetic nephropathy. Dashed lines indicate previously reported effects ²⁶ , ⁷² solid lines indicate findings from this study. The values are presented as the means ± SEMs. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

DISCUSSION

In this study, we demonstrate for the first time that linagliptin markedly upregulates renal expression of LRP2 (megalin) and NQO1 in an Nrf2‐independent manner in the STZ‐induced diabetic kidney, using Nrf2‐knockout (Nrf2‐KO) mice.

Extensive genetic and functional studies ³⁷ , ³⁸ , ³⁹ , ⁴⁰ indicate that LRP2‐dependent endocytosis underlies proximal tubular uptake of albumin and diverse plasma‐derived molecules, and that the loss of LRP2 function markedly impairs this reabsorptive capacity. Albumin staining in kidney sections may support the enhanced intracellular processing of albumin in linagliptin‐treated diabetic kidney, which return to the systemic circulation, thereby reducing urinary albumin excretion.

NQO1 is an antioxidant enzyme that protects against renal ischemia/reperfusion injury ⁴¹ and cisplatin‐induced nephrotoxicity ⁴² in mouse models. NQO1 overexpression results in suppressed proinflammatory cytokine secretion, extracellular matrix accumulation and epithelial‐mesenchymal transition (EMT) in the kidneys of type 2 diabetes model db/db mice. ⁴³ In contrast, NQO1 knockout mice exhibit severe STZ‐mediated DN. ⁴⁴ Moreover, a lack of NQO1 in vivo and in vitro dysregulates autophagy, resulting in exacerbated renal fibrosis. ⁴⁵ The linagliptin‐induced upregulation of NQO1 in the present study may contribute to, and at least partially underlie, its renoprotective effects on DN.

Although Nrf2‐KO mice presented decreased renal expression of Nrf2 and its target genes, Ho1 and Cat, compared with WT mice, no difference in Nqo1 expression in the kidney was detected between WT and Nrf2‐KO mice. Thus, Nqo1 displayed regulatory behavior distinct from other canonical Nrf2 target genes, suggesting that the expression of NQO1 not directly regulated by Nrf2 and may involve, alternative regulators such as Nrf1, ⁴⁶ , ⁴⁷ Nrf3 ⁴⁸ , ⁴⁹ and BATCH1, ⁵⁰ which appears to change with the progression of DN. ⁵¹ Additionally, diabetes and linagliptin treatment increased NQO1 protein expression in the kidneys of WT and Nrf2‐KO mice. These results are consistent with the discordance between transcript and protein levels reported for NQO1. ⁵²

Previous studies have reported decreased renal LRP2 expression in STZ‐induced diabetic models. ⁵³ , ⁵⁴ In contrast, a Zucker diabetic fatty (ZDF) rat model of type 2 diabetes has demonstrated significantly increased renal LRP2 expression at 12 weeks of age. ³⁴ The contrasting insulin milieus in these models have stimulated focused investigations into the interplay between insulin signaling cascades, notably those involving PI3K, ⁵⁵ , ⁵⁶ AKT, ³⁶ , ⁵⁷ and mTOR. ⁵⁸ Linagliptin also protects diabetic kidneys by enhancing IRS1/Akt insulin signaling, with additional partial protection via the Nrf2 pathway in podocytes. ⁶ Accordingly, linagliptin increased LRP2 expression and renal p‐AKT levels in diabetic mice independent of Nrf2. Furthermore, an AKT1 siRNA abolished linagliptin‐induced LRP2 expression in HRPTECs, suggesting that linagliptin increased LRP2 expression under high glucose conditions by increasing p‐AKT levels.

However, insulin was unlikely to be the upstream regulator of this effect. If insulin acted upstream of Akt, LRP2 expression would be expected to depend on insulin levels. Interestingly, a previous study has shown that linagliptin produces a significant reduction in urinary albumin excretion in db/db mice without any change in plasma insulin concentrations in comparison of untreated db/db control mice and linagliptin‐treated db/db mice. ²⁵ This finding was recapitulated in our own experiments. Akt is a convergence point for multiple upstream inputs ⁵⁹ beyond insulin, including receptor tyrosine kinases (RTKs), G‐protein coupled receptors (GPCRs), ⁶⁰ , ⁶¹ angiotensin II, ⁵⁹ , ⁶² integrin, ⁶³ starvation stress, ⁶⁴ and intracellular modulators such as phosphatidylinositol‐3‐kinase (PI3K), ⁶⁵ 3‐phosphoinositide dependent kinase 1 (PDK1), ⁶⁶ mTORC2, ⁶⁷ , ⁶⁸ and phosphatase and tensin homolog (PTEN). ⁶⁹ These alternative upstream pathways can explain the Akt activation in settings where systemic insulin is unchanged and support the plausibility of insulin‐independent mechanisms for the effects of linagliptin on tubular Akt signaling and renal outcomes.

In the current study, linagliptin significantly decreased plasma cystatin C levels compared with the controls in WT and Nrf2‐KO mice. LRP2 is an endocytic receptor of cystatin C in proximal tubular cells. ⁷⁰ , ⁷¹ Decreased plasma cystatin C levels suggest that linagliptin might augment cystatin C uptake as well as albumin in diabetic mice.

This study has several limitations. First, the precise mechanisms by which linagliptin increases LRP2 expression in proximal tubular cells via the activity of AKT1 but not Nrf2, including the effects on albumin absorption and the relationship with NQO1, need to be clarified. Because multiple factors described above increase p‐Akt expression, these alternative pathways will be addressed in future studies. Second, albumin immunostaining just showed the intracellular presence of albumin. In order to determine which intracellular process(es) in the renal proximal cells linagliptin targets, further detailed experiments are required. Third, clinical studies examining the changes in LRP2 expression in the kidney tissues of patients treated with linagliptin will be vital for the translation of our results to patient outcomes.

In conclusion, the results of the current study indicate, for the first time, that linagliptin increases the expression of NQO1 and LRP2 in STZ‐induced diabetic kidneys independent of Nrf2, and this phenomenon might be involved in the attenuation of albuminuria accompanied by the increase in AKT phosphorylation in diabetic kidneys.

Disclosure

This study was performed in collaboration with Boehringer Ingelheim, Japan.

Approval of the research protocol: N/A.

Informed Consent: N/A.

Approval date of Registry and the Registration No. of the study/trial: N/A.

Animal studies: approval no. 24–03, date of approval: March 19, 2024 by the Research Center for Animal Life Science at Asahikawa Medical University.

Author Contributions

SP and Yumi T designed and performed the experiments, analyzed the data, and wrote the manuscript. TT and Yuri T performed the experiments. HK and HN participated in discussions. All the authors approved the final version of the manuscript. Yumi T is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Funding

This work was partially supported by grants from the Japan Society for the Promotion of Science (22 K08301 [Yumi T], 25 K19648 [TT] and 23 K08001 [NH]).

Supporting information

Table S1. Primer sequences used for SYBR Green‐based qRT–PCR in mouse kidneys.

Table S2. qRT–PCR primers from Applied Biosystems using HRPTECs.

Table S3. Primary antibodies used for Western blotting and immunohistochemistry.

Table S4. Laboratory data from mice.

Figure S1. Experimental setup and time points of analysis.

Figure S2. Linagliptin has no effect on weight.

Figure S3. Linagliptin may promote the sequential process of albumin in proximal tubular cells.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.