Skeletal myopathy in CKD: a comparison of adenine-induced nephropathy and 5/6 nephrectomy models in mice

Abstract

Preclinical animal models of chronic kidney disease (CKD) are critical to investigate the underlying mechanisms of disease and to evaluate the efficacy of novel therapeutics aimed to treat CKD-associated pathologies. The objective of the present study was to compare the adenine diet and 5/6 nephrectomy (Nx) CKD models in mice. Male and female 10-wk-old C57BL/6J mice (n = 5–9 mice/sex/group) were randomly allocated to CKD groups (0.2−0.15% adenine-supplemented diet or 5/6 Nx surgery) or the corresponding control groups (casein diet or sham surgery). Following the induction of CKD, the glomerular filtration rate was reduced to a similar level in both adenine and 5/6 Nx mice (adenine diet-fed male mice: 81.1 ± 41.9 µL/min vs. 5/6 Nx male mice: 160 ± 80.9 µL/min, P = 0.5875; adenine diet-fed female mice: 112.9 ± 32.4 µL/min vs. 5/6 Nx female mice: 107.0 ± 45.7 µL/min, P = 0.9995). Serum metabolomics analysis indicated that established uremic toxins were robustly elevated in both CKD models, although some differences were observed between CKD models (i.e., p-cresol sulfate). Dysregulated phosphate homeostasis was observed in the adenine model only, whereas Ca²⁺ homeostasis was disturbed in male mice with both CKD models. Compared with control mice, muscle mass and myofiber cross-sectional areas of the extensor digitorum longus and soleus muscles were ∼18−24% smaller in male CKD mice regardless of the model but were not different in female CKD mice (P > 0.05). Skeletal muscle mitochondrial respiratory function was significantly decreased (19−24%) in CKD mice in both models and sexes. These findings demonstrate that adenine diet and 5/6 Nx models of CKD have similar levels of renal dysfunction and skeletal myopathy. However, the adenine diet model demonstrated superior performance with regard to mortality (∼20−50% mortality for 5/6 Nx vs. 0% mortality for the adenine diet, P < 0.05 for both sexes) compared with the 5/6 Nx surgical model.

NEW & NOTEWORTHY Numerous preclinical models of chronic kidney disease have been used to evaluate skeletal muscle pathology; however, direct comparisons of popular models are not available. In this study, we compared adenine-induced nephropathy and 5/6 nephrectomy models. Both models produced equivalent levels of muscle atrophy and mitochondrial impairment, but the adenine model exhibited lower mortality rates, higher consistency in uremic toxin levels, and dysregulated phosphate homeostasis compared with the 5/6 nephrectomy model.

INTRODUCTION

Chronic kidney disease (CKD) increases in prevalence with age, particularly in individuals with hypertension and diabetes (1). More than 35 million people in the United States are estimated to have CKD, which has no cure, and available treatment options remain limited to peritoneal dialysis, hemodialysis, and kidney transplantation (1, 2). CKD is a catabolic disease that results in loss of muscle mass and function that has been strongly associated with mortality (3). It has been established that muscle catabolism, but not malnutrition, is primarily responsible for this increased mortality risk in CKD (4). In addition, several uremic toxins are known to cause skeletal muscle atrophy through a variety of proposed mechanisms including activation of proteolytic and autophagy pathways (5–7), as well as mitochondrial dysfunction and oxidative stress (8–13). In this regard, recent studies have shown that skeletal muscle abnormalities including muscle wasting and mitochondrial dysfunction are prevalent in patients with renal failure and that uremic toxins may play a significant role in the developing skeletal myopathy (14–16).

Numerous preclinical models of CKD have been developed to investigate the molecular mechanisms underlying CKD pathobiology as well as to test new therapeutic approaches aimed to mitigate CKD-associated pathologies (14, 17, 18). The 5/6 nephrectomy (Nx) model, which involves surgical resection of kidney mass, is one of the most widely used techniques to successfully induce renal failure in animals (19, 20). However, complications such as the ureteral injury and bleeding contribute to a considerable mortality rate (up to 50%) (20). In addition, added experimental variance of this model due to slight differences in the size of the remnant kidney can significantly impact experimental outcomes.

Notably, the adenine diet model of CKD, a nonsurgical option, was originally developed to induce renal dysfunction in rats (21–24). Kidney pathology with adenine feeding stems from the formation of 2,8-dihydroxyadenine, an adenine metabolite that crystalizes within renal tubules and causes injury, inflammation, tubular atrophy, and fibrosis of the renal parenchyma (25). More recently, the adenine model of nephropathy has been successfully used in mice and provides better control over the disease severity with modification of adenine dosage (26, 27). For example, recent studies have reported that mice fed an adenine diet for 6 wk showed decreased kidney size and glomerular filtration rate (GFR), increased blood urea nitrogen (BUN), and renal fibrosis compared with mice fed a control diet (14, 28). Despite the growth in popularity of the adenine diet CKD model, no studies have directly compared the associated skeletal myopathy in the adenine model with the traditional surgical CKD models. Therefore, the primary goal of this study was to compare skeletal myopathy between the adenine diet and 5/6 Nx CKD models in both male and female mice. It was hypothesized that both CKD models would produce similar impairments in renal function and skeletal myopathy.

METHODS

Animals

Male and female 10-wk-old C57BL/6J mice (n = 5–9 mice/group/sex) were obtained from The Jackson Laboratory (Bar Harbor, ME). All rodents were housed in a light-controlled (12:12-h light-dark cycle), temperature-controlled (22°C), and humidity-controlled (50%) room, and maintained on standard chow (Teklad 2018, Envigo, 18.6% crude protein) before CKD induction with free access to food and water. All animal experiments adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals and any updates. All procedures were approved by the Institutional Animal Care and Use Committee of the University of Florida and Malcom Randall Veterans Affairs Medical Center.

Adenine Diet-Induced Nephropathy

We used an established adenine diet model (17, 27, 29–32) to induce kidney disease in mice. Mice were allocated to a casein-based chow diet (20% casein) for 7 days, followed by the induction of renal tubular injury by supplementing the casein-based chow with 0.2% adenine diet for 7 days, followed by a maintenance diet supplemented with 0.15% adenine diet for 7 wk. CKD mice were then placed back on the control casein diet for 2 wk before the terminal experiments. Control animals received casein-based chow for the entirety of the study (Fig. 1A).

Figure 1.

Comparison of adenine nephropathy and 5/6 nephrectomy (Nx) models. A: schematic overview of the study design for adenine-induced nephropathy and 5/6 Nx. Male and female C57BL/6J mice were allocated to control (casein diet and sham surgery) and chronic kidney disease (CKD; adenine diet and 5/6 Nx) groups. Glomerular filtration rate (GFR) and blood urea nitrogen (BUN) were assessed 4–7 days prior to euthanization. B: probability of survival throughout the study duration. C: body weights in control and CKD mice. D: GFR measured by FITC-inulin clearance. E: BUN measured using a commercial kit. GFR and BUN tests were conducted 8 wk after adenine diet in the group of diet-induced nephropathy and 4 wk after the second-step surgery in the group of 5/6 Nx. Values are represented as means ± SD; n = 4–9 mice/group/sex. Data were analyzed using two-way ANOVA, and Tukey’s post hoc analysis was performed when appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 vs. control; ###P < 0.001 vs. adenine. FITC

5/6 Nx Model

A modified 5/6 Nx model (20) was used to surgically induce renal dysfunction in mice. This consisted of two surgical steps: 1) initial upper and lower pole ligation of the left kidney during the first surgery and 2) right kidney resection 1 wk after the first surgery (Fig. 1A). All operating procedures were performed under aseptic techniques. Mice were anesthetized with 1.0−1.2% isoflurane mixture (2% during the induction phase) and placed on a warm pad (SpaceGel, Braintree Scientific) during the surgery. In the supine position, a midline abdominal incision was made, and the left kidney was exposed. The surgeon performed careful separation of perirenal fat, connective tissue, the adrenal gland, and the ureter with blunt forceps. A 3-0 silk suture was placed circumferentially around the upper and lower poles of the left kidney. Thereafter, application of the ligatures was performed, and the renal parenchyma was observed to ensure the presence of ischemia without bleeding. Subsequently, the contralateral right renal Nx was performed 7 days after the first operation. The Nx was performed by reopening the previous midline laparotomy incision, and the right kidney was isolated. Care was taken to preserve the adrenal gland, after which the renal pedicle with the artery, vein, and ureter was ligated. Finally, the right kidney was extirpated by transecting the vessels and ureter immediately distal to the ligature. The abdominal muscle layer and skin were closed with 5-0 sterile absorbable suture using a simple continuous technique. Following each surgery, mice were treated with buprenorphine (0.1 mg/kg body weight). The sham surgery group also underwent two-step surgeries but without pole ligation of the left kidney or right kidney resection.

Assessment of Kidney Function

Kidney function was assessed by measuring GFR via FITC-inulin clearance as previously described (14, 33, 34). Briefly, FITC-inulin was dissolved in 0.9% NaCl [5% (wt/vol)], and the solution was dialyzed with a 1,000-kDa dialysis membrane (Spectrum Laboratories) in 0.9% NaCl at room temperature in the dark for 24 h, followed by sterile filtering through a 0.22-µm filter (Thermo Fisher). Under brief isoflurane anesthesia, FITC-inulin (2 µL/g body weight) was injected into the retroorbital sinus. The mouse was then returned to a clean cage, and ∼20 µL of blood was collected in heparinized capillary tubes via a ∼1-mm tail snip in the conscious mouse at 3, 5, 7, 10, 15, 35, 56, and 75 min following the injection. Thereafter, blood was centrifuged for 10 min at 4,000 rpm (4°C), and the plasma supernatant was transferred to a new tube. To measure the FITC-inulin concentration, plasma was diluted (1:20) in 0.5 mol/L of HEPES buffer (pH 7.4) and loaded into a 96-well plate along with a FITC-inulin standard curve, and fluorescence was detected using a BioTek Synergy II plate reader. GFR was calculated using a two-phase exponential decay curve fit in GraphPad Prism as previously described (14, 33, 34), and was not normalized to body weight. BUN was also assessed using a commercial kit (K024, Arbor Assays) according to the manufacturer’s instructions.

Serum Biochemical Analyses

Serum levels of calcium and inorganic phosphate were determined by a spectrophotometer using commercial kits (Cat. No. P7516 and C7529, Pointe Scientific) according to the manufacturer’s instructions. Quantification of serum parathyroid hormone (PTH) and fibroblast growth factor-23 (FGF23) was performed using ELISA (Cat. No. 60–2305 and 60–6800, Immutopics/Quidel) according to the manufacturer’s instructions. Serum used in these analyses was collected at the time of euthanasia.

Serum Metabolomics

Analysis of serum metabolomics was carried out at the Southeast Center for Integrated Metabolomics at the University of Florida (http://secim.ufl.edu) as previously described (35). On the day of euthanization, whole blood was collected via cardiac puncture with a 21-gauge needle syringe, allowed to clot for ∼20 min at room temperature, and centrifuged at 4,000 rpm for 10 min. The supernatant was transferred to new Eppendorf tubes and stored at −80°C until analysis. Serum samples were processed using a metabolomics platform to quantify select tryptophan-derived metabolites that are established uremic toxins. Twenty-five microliters of serum from each mouse (n = 4 mice/sex/group) were mixed with 5 µL of internal standard solution containing [¹³C₁₁]tryptophan, serotonin D4, kynurenine D4, kynurenic Acid D5, xanthurenic acid D4, [¹³C₆]anthranilic acid, [¹³C]indoxyl sulfate, p-cresol sulfate D7, and 3-indole-acetate D7. After the addition of 200 µL of acetonitrile:methanol:acetone (8:1:1) with 0.1% formic acid, samples were placed in an ultrasonic bath for 10 min and then centrifuged at 20,000 g for 5 min at 4°C to pellet the protein. Then, 190 µL of supernatant were transferred from each sample into clean tubes and dried under a gentle stream of nitrogen at 30°C. The dried extracts were resuspended with 25 µL water with 0.1% formic acid and incubated at 4°C for 10−15 min, and the samples were then centrifuged at 20,000 g for 5 min at 4°C. Supernatants were transferred into clean LC vials for targeted LC-MS quantitation on a Thermo Q-Exactive Oribtrap mass spectrometer with Dionex UHPLC and autosampler. Tryptophan, serotonin, kynurenine, kynurenic acid, xanthurenic acid, and anthranilic acid were quantified in positive ionization, whereas indoxyl sulfate, p-cresol sulfate, and 3-indole-acetate were quantified in negative ionization. Separation was achieved on an ACE C18-PFP (C18 bonded HPLC column with a pentafluorophenyl phase) 100 × 2.1-mm, 2-µm column, using a gradient with mobile phase A as 0.1% formic acid in water and mobile phase B as acetonitrile. The flow rate was 350 µL/min with a column temperature of 25°C and injection volume of 2 µL. A nine-point calibration curve and quality control samples were prepared for targeted quantitation of tryptophan, serotonin, kynurenine, kynurenic acid, xanthurenic acid, anthranilic acid, indoxyl sulfate, p-cresol sulfate, and 3-indole-acetate. Twenty microliters of each calibrator and quality controls were supplemented with 5 µL of indoxyl sulfate. Peak areas of each analyte and the corresponding internal standards in the calibrator, quality controls, and samples were integrated using Xcalibur 4.0. A calibration curve was generated by plotting nominal concentration of the analyte in the calibrators versus the peak area ratio of the analyte and internal standard. Quality controls and samples were quantitated against the calibration curve.

Immunofluorescence Microscopy

Skeletal muscle fiber cross-sectional area (CSA) was assessed by immunofluorescence microscopy. Isolated extensor digitorum longus (EDL) and soleus (SOL) muscles were mounted in disposable base molds (Cat. No. 6235215, Electron Microscopy Science) using optimum cutting temperature mounting medium (Tissue-Tek, Sakura Finetek) and frozen in liquid nitrogen-cooled isopentane for cryosectioning. Transverse serial sections (10 µm) were cut from the midbelly of the skeletal muscles using a Leica 3050S cryostat at −20°C, mounted on frosted microscope slides, briefly air-dried at room temperature, and stored at −80°C until being stained. Prior to staining, frozen muscle sections were air-dried at room temperature for 10 min and fixed with 4% paraformaldehyde for 5 min. Following 2 × 3-min washes with 1× PBS, sections were permeabilized with 0.3% Triton X-100 in PBS for 10 min at room temperature. After 3 × 5-min washes with PBS, sections were incubated in blocking buffer (5% goat serum and 1% BSA in PBS) for 1 h at room temperature. To label the sarcolemma, sections were incubated with primary antibody against to laminin (L9393, Sigma, 1:100) overnight at 4°C. Following 4 × 5-min washes with 1× PBS, muscle sections were stained with secondary antibody (Alexa Fluor 488 goat anti-rabbit IgG, 1:250) diluted in blocking buffer. Following 4 × 5-min washes with 1× PBS, coverslips were mounted with Vectashield hardmount (H-1500, Vector Laboratories). Slides were imaged at ×20 magnification with an Evos FL2 Auto microscope (Thermo Scientific), and tiled images of the entire muscles were obtained for analysis. Quantification of myofibers CSA was performed using MuscleJ (36), an automated analysis software developed in Fiji.

Skeletal Muscle Mitochondrial Isolation

Skeletal muscle mitochondria were isolated as previously described (16, 37). Briefly, the following skeletal muscles were dissected: tibialis anterior (TA), quadriceps, hamstrings, gastrocnemius, plantaris, triceps, pectorals, gluteus maximus, and erector spinae muscles and placed in ice-cold mitochondrial isolation medium (300 mM sucrose, 10 mM HEPES, and 1 mM EGTA, pH 7.1). Muscles were then trimmed to remove excessive blood, connective tissue, and fat, minced in a petri dish on ice, transferred to ice-cold mitochondrial isolation medium supplemented with BSA (1 mg/mL), homogenized on ice using a glass-Teflon homogenizer (Wheaton), and subsequently centrifuged at 800 g for 10 min to pellet nonmitochondrial myofibrillar proteins, nuclei, and other cellular components. The resulting supernatant was centrifuged at 10,000 g for 10 min to pellet mitochondria. All centrifugation steps were performed at 4°C. The mitochondrial pellet was gently washed to remove damaged mitochondria and then resuspended in mitochondrial isolation medium (without BSA). The protein concentration was determined using a bicinchoninic acid protein assay (SL256970, Thermo Scientific).

Measurement of Mitochondrial Respiratory Function

Mitochondrial respiratory function was assessed using high-resolution respirometry (Oroboros Oxygraph-2K) at 37°C in mitochondrial assay buffer (105 mM K-MES, 30 mM KCl, 1 mM EGTA, 10 mM K₂HPO₄, 5 mM MgCl₂·6H₂O, and 2.5 mg/mL BSA, pH 7.2) supplemented with 20 mM creatine. Twenty micrograms of mitochondria were added to the Oxygraph-2K chamber filled with 2 mL of mitochondrial assay buffer. To energize the mitochondria, 5 mM pyruvate and 2.5 mM malate substrates were provided as fuel (state 2), followed by the sequential additions of 4 mM ADP (complex I state 3), 0.005 mM cytochrome c (to check the integrity of the mitochondrial outer membrane), and 10 mM succinate (complex I + II state 3).

Statistical Analysis

All statistical analyses were conducted using GraphPad Prism software (v. 8.3.1). Data were analyzed using two-way ANOVA with Tukey’s post hoc testing for multiple comparisons when appropriate. The two-stage step-up method of Benjamini, Krieger, and Yekutieli was used for multiple comparisons of mitochondrial respiration data. When sex differences or group × sex interactions were detected, data were presented accordingly. Pearson correlations were performed using two-tailed statistical testing. In all cases, P < 0.05 was considered statistically significant. All data were presented as means with error bars representing SDs.

RESULTS

Characterization of Renal Function and Mortality in the Adenine Diet and 5/6 Nx CKD Models

The widely used 5/6 Nx model of CKD involves two invasive surgeries that produce a rapid decrease in kidney mass that does not mimic the progressive nature of CKD in humans. The mortality rate in surgical models of CKD complicates preclinical study design. To this end, there was a notable difference in the survival rate between adenine and 5/6 Nx CKD models in both male and female mice. A total of 10 male mice and 11 female mice underwent the 5/6 Nx surgery, and five male mice and two female mice died throughout the experiment. In contrast, no mortality was observed for male or female CKD mice subjected to the adenine diet (Fig. 1B). Average body weights of mice fed with the casein diet were not different from mice with sham surgery in both sexes. Both adenine feeding and 5/6 Nx led to reduced body weight in male mice only (Fig. 1C). GFR was decreased in both adenine and 5/6 Nx mice compared with the corresponding control mice (Fig. 1D). Importantly, no difference in GFR was found between adenine and 5/6 Nx CKD models (Fig. 1D). Similar to GFR, BUN was significantly higher in animals with CKD (Fig. 1E). Interestingly, BUN levels were significantly greater (although highly variable) in 5/6 Nx mice compared with adenine diet-fed mice (Fig. 1E).

Dysregulated mineral and bone metabolism is common in patients with CKD. To characterize biochemical markers related to mineral and bone metabolism, we compared serum levels of phosphate, Ca²⁺, FGF23, and PTH in both CKD models (Fig. 2). Baseline concentrations of these serum biochemical markers between casein diet-fed and sham controls were similar in both male and female mice (Fig. 2). Interestingly, serum phosphate (Fig. 2A) and FGF23 (Fig. 2B) levels were significantly increased in adenine diet-fed mice but not in 5/6 Nx mice. Serum Ca²⁺ levels were significantly increased in both CKD models for male but not female mice (Fig. 2C), indicating a clear sex difference. PTH was significantly greater in male CKD mice compared with controls; however, male adenine diet-fed mice displayed higher PTH compared with 5/6 Nx male mice (Fig. 2D). Taken together, these data provide support for the clinical relevance of the adenine model and highlight sex differences in preclinical CKD models that are not examined often.

Figure 2.

Serum biochemical markers associated with dysregulated bone and mineral metabolism. Serum phosphate (A), fibroblast growth factor-23 (FGF23; B), calcium (C), and parathyroid hormone (PTH; D) were determined using a microplate spectrophotometer. Values are represented as means ± SD; n = 3 or 4 mice/group/sex. Data were analyzed using a two-way ANOVA, and Tukey’s post hoc analysis was performed when appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 vs. control; ##P < 0.01 and ###P < 0.001 vs. adenine. Nx, nephrectomy.

Adenine Diet and 5/6 Nx Models Increase Uremic Toxins

With impaired renal function, there is often increased accumulation of circulating uremic toxins, which have been strongly associated with CKD pathobiology (38–43). For this reason, we performed targeted serum metabolomics analyses to examine the levels of uremic toxins in adenine and 5/6 Nx CKD models (Fig. 3). Consistent with previous reports (28, 44, 45), mice with adenine-induced and 5/6 Nx-induced CKD displayed increases in serum levels of established uremic toxins, including kynurenine (Fig. 3A), kynurenic acid (Fig. 3B), xanthurenic acid (Fig. 3C), and indoxyl sulfate (Fig. 3E). However, there were some differences between sexes and CKD models that were noteworthy. Interestingly, p-cresol sulfate, a well-known uremic toxin linked to cardiovascular disease risk and insulin resistance (41, 46), was elevated only in adenine diet-fed mice (Fig. 3F). Similarly, indole-3-acetate was elevated in male adenine diet-fed mice (but not female mice), whereas 5/6 Nx male mice displayed no differences from sham controls (Fig. 3G). There were no changes in the concentrations of serum tryptophan (Fig. 3H) and serotonin (Fig. 3I) in male and female mice in either CKD model. Another striking observation was that the variability in uremic toxin levels was substantially greater in 5/6 Nx groups, suggesting that the adenine diet model produces more consistent levels of uremic toxins.

Figure 3.

Serum uremic toxins in adenine nephropathy and 5/6 nephrectomy (Nx) models. Serum metabolomics determined circulating uremic toxins derived from tryptophan metabolism measured via mass spectrometry (LC-MS/MS). A: kynurenine. B: kynurenic acid. C: xanthurenic acid. D: anthranilic acid. E: indoxyl sulfate. F: p-cresol sulfate. G: indole-3-acetate. H: tryptophan. I: serotonin. Values are represented as means ± SD; n = 4 mice/group/sex. Data were analyzed using two-way ANOVA, and Tukey’s post hoc analysis was performed when appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 vs. control; #P < 0.05, ##P < 0.01, ###P < 0.001, and ###P < 0.0001 vs. adenine.

Muscle Mass and Myofiber CSA Are Decreased in Male but Not Female CKD Mice

Muscle catabolism is a common feature of CKD contributing to the progressive myopathic symptoms including muscle atrophy, weakness, and impaired physical function. To compare the muscle pathology in these CKD models, we measured muscle mass and myofiber CSAs in fast-twitch EDL and slow-twitch SOL muscles. There were no differences in muscle mass for EDL and SOL muscles between the control groups (casein diet vs. sham surgery). Both adenine diet-fed and 5/6 Nx male CKD mice had significantly reduced muscle mass in both EDL and SOL muscles (Fig. 4, B and G). Interestingly, EDL and SOL muscle masses were not statistically different between CKD and control female mice (Fig. 4, B and G). Consistent with muscle masses, mean myofiber CSA was not different between casein diet-fed and sham-operated mice (Fig. 4, C and H). Adenine diet-fed and 5/6 Nx male mice displayed significantly smaller myofiber CSA in EDL and SOL muscles (Fig. 4, C and H). Similar to muscle mass, mean myofiber CSAs of EDL and SOL muscles were not different between CKD and control groups in female mice (Fig. 4, C and H).

Figure 4.

Chronic kidney disease causes muscle wasting and reduced fiber cross-sectional area (CSA) in male but not in female mice. A: representative immunofluorescence image of the extensor digitorum longus (EDL) muscle stained with laminin from a casein diet-fed control mouse. B: muscle mass for the EDL muscle. C: mean myofiber CSA of the EDL muscle. D and E: relative frequency distributions of EDL myofiber CSA for male mice (D) and female mice (E). F: representative immunofluorescence image of the soleus (SOL) muscle stained with laminin from a casein diet-fed control mouse. G: muscle mass for the SOL muscle. H: mean myofiber CSA of the SOL muscle. I and J: relative frequency distributions of SOL myofiber CSA for male mice (I) and female mice (J). Values are represented as means ± SD; n = 5 mice/group/sex. Data were analyzed using two-way ANOVA, and Tukey’s post hoc analysis were performed when appropriate. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001 vs. control; #P < 0.05 vs. casein. Nx, nephrectomy.

Skeletal Muscle Mitochondrial Respiratory Function Is Equally Impaired in Adenine Diet-Fed and 5/6 Nx Mice

Accumulating evidence from human and animal studies suggests that skeletal muscle mitochondrial abnormalities contribute to the myopathic symptoms in CKD (14, 28). Thus, we assessed mitochondrial respiratory function to determine how both CKD models impact muscle mitochondrial health. Figure 5A shows an overview of mitochondrial respiratory steps and conditions used in this experiment. First, we energized mitochondria in the absence of adenylates with pyruvate and malate to measure state 2 respiration, which displayed a significant group (CKD) effect (Fig. 5B). Next, maximal state 3 respiration was measured following the addition of 4 mM ADP, which assesses the maximal respiratory capacity supported by substrates that provide electrons to complex I (Fig. 5C). Next, exogenous cytochrome c was added to assess the integrity of the outer mitochondrial membrane (Fig. 5D). Finally, the subsequent addition of succinate allows additional electrons to enter the electron transport system through complex II (succinate dehydrogenase), facilitating analysis of maximal state 3 respiration supported by complex I + II substrates (Fig. 5E), which displayed a significant group (CKD) effect, but no significant effects of sex or interactions were detected. Notably, both adenine diet-fed and 5/6 Nx mice exhibited a similar magnitude (∼20%) of reduction in respiratory function, confirming that both murine CKD models impact muscle mitochondria to a similar degree.

Figure 5.

Mitochondrial respiratory function in skeletal muscle. Skeletal muscle mitochondria were isolated from control and chronic kidney disease mice. A: graphical depiction of the detailed respiratory protocol used. B: state 2 respiration supported by pyruvate and malate. C: ADP-stimulated mitochondrial respiration supported by pyruvate and malate. D: state 3 respiration following the addition of cytochrome c (Cyt C; used for checking mitochondrial integrity). E: state 3 respiration supported by pyruvate, malate, and succinate. Values are represented as means ± SD; n = 5–9 mice/group/sex. Data were analyzed using two-way ANOVA and the two-stage step-up method of Benjamini, Krieger, and Yekutieli for multiple comparisons. *P < 0.05, **P < 0.01, and ***P < 0.001 vs. control. Nx, nephrectomy.

Pearson Correlation Analysis across Outcome Measures

To examine potential relationships between outcome measures, we next performed Pearson correlation analyses within each sex and model (Fig. 6). The results of these analyses revealed strong associations among GFR, uremic toxin levels, muscle size, and mitochondrial respiratory capacity (P < 0.05; Fig. 6). Elevated uremic toxins, including kynurenic acid, had inverse relationships with muscle mass and myofiber CSA (P < 0.05). Notably, several uremic toxins derived from kynurenine and indole pathways exhibited inverse relationships with mitochondrial respiratory function consistent with previous work (16, 28).

Figure 6.

Pearson correlation analyses across outcome measures. Pearson correlation coefficients among glomerular filtration rate (GFR), serum metabolites, muscle mass, myofiber cross-sectional area (CSA), and mitochondrial respiratory capacity were performed according to model and sex. Heatmaps were created from the Pearson correlation matrixes for adenine-induced nephropathy in male (A) and female (C) mice and for 5/6 nephrectomized (Nx) male (B) and female (D) mice. n = 4–9 mice/group/sex. *Significant correlation (P < 0.05). AA, anthranilic acid; EDL, extensor digitorum longus; I-3-A, indole-3-acetic acid; IS, indoxyl sulfate; KA, kynurenic acid; p-CS, p-cresol sulfate; SOL, soleus; XA, xanthurenic acid.

DISCUSSION

In the present study, two murine models of CKD, adenine diet and 5/6 Nx, were compared in disease severity and in relation with their adverse impact on skeletal muscle physiology. Although both models produced equivalent levels of renal dysfunction and impaired mitochondrial respiratory capacity in both sexes and muscle atrophy in male mice, the 5/6 Nx model displayed higher mortality in both sexes compared with the adenine diet model. Both models resulted in increased levels of established uremic toxins, including kynurenine, kynurenic acid, and indoxyl sulfate. However, several uremic toxins (p-cresol sulfate and indole-3-acetate) were increased in the adenine diet model but not in the 5/6 Nx model. Moreover, 5/6 Nx mice displayed greater variability in uremic toxin levels compared with adenine diet-fed mice. Interestingly, the pathophysiological impact of CKD was greater in male mice than in female mice regardless of the CKD model. Finally, we observed strong correlations between kidney function and the level of uremic toxins with muscle atrophy and mitochondrial function.

To confirm renal dysfunction in both models, we measured GFR, a gold standard measure of kidney function (34), as well as the concentration of BUN. Several previous studies using adenine diet-induced nephropathy or 5/6 Nx have reported reduced GFR and high levels of circulating urea nitrogen compared with their respective control groups (14, 28, 47). As expected, in the present study, GFR was significantly decreased in both CKD models compared with control animals. Importantly, GFR was not different between adenine diet-fed and 5/6 Nx mice for either sex. Similarly, BUN was elevated after the induction of renal dysfunction in male adenine diet-fed and 5/6 Nx mice in both sexes. Despite similar GFRs, BUN was significantly higher in 5/6 Nx mice compared with adenine diet-fed mice. There are several possible explanations for this result. First, protein catabolism with kidney disease resulting in increased nitrogenous waste is the primary source of urea production. However, given the comparable changes in body weights and EDL and SOL muscle masses in the two CKD models, rates of muscle protein catabolism are likely similar between models. This presumption is supported by previous studies that have reported similar rates of muscle protein degradation in 5/6 Nx and adenine models (5, 28). Second, the amount of dietary protein intake is associated with elevated levels of BUN in the CKD condition (48). In this regard, the average daily food intake is less in adenine diet-fed mice due in part to their reluctance to consume adenine-supplemented food (26). In a current study, we estimated daily food consumption in male and female mice consuming the adenine diet to be 2.4 and 2.1 g/day, respectively (28). This food intake is lower than the standard rodent diets (3–4 g/day), which were provided to the 5/6 Nx mice. Consistent with this notion, previous reports in young male C57BL/6J mice subjected to subtotal Nx found food consumption of 3–4 g of standard chow with 17% crude protein (49, 50). Accordingly, it is likely that the greater BUN concentration in 5/6 Nx mice compared with adenine diet-fed mice related to greater dietary protein intake.

Various animal models have been developed to characterize the numerous clinical features of a patient with CKD. Although no model has yet to fully recapitulate the complex array of clinical pathophysiology of patients with CKD, animal models of CKD remain a crucial tool for discovery of disease mechanisms and therapeutic development. To better understand potential differences in the 5/6 Nx and adenine diet models, we assessed several factors related to phosphate and Ca²⁺ homeostasis that have been linked to vascular calcification and increased cardiovascular events (51), and dysregulated bone and mineral metabolism, which are desirable features of preclinical CKD studies (26, 52). In the present study, we observed that male and female mice exposed to adenine diet had increases in serum phosphate and FGF23 levels similar to the results of a previous study (26). In contrast, neither phosphate nor FGF23 levels were changed in the 5/6 Nx group in either male or female mice. These results provide clear evidence of bone and mineral dysregulation in adenine diet-fed male mice. Notably, serum Ca²⁺ and PTH levels were similar in control and CKD female mice, again highlighting the sex differences in preclinical CKD models. The underlying mechanisms for this protection in female mice is not known but suggests that young female mice with intact ovaries are resistant to Ca²⁺ and phosphate dysregulation in CKD models.

Although there was a similar degree of renal dysfunction observed in both models, it is important to note that there was no mortality in the adenine diet-fed group but that there was 20−50% mortality in the 5/6 Nx group, with a greater rate observed in male mice. This observation confirms previous reports (26–28, 53) that used adenine diet models with <0.2% adenine in mice. The lower survival rate in the 5/6 Nx groups may be due in part to kidney tissue damage stemming from forced ligation that may cause intrarenal bleeding, damage on the ureter, or possible infection during the operation and postoperative care (17, 20, 26, 27). With regard to the 5/6 Nx model, it is important to highlight that this study performed a pole ligation 5/6 Nx model, which has subtle differences, including lower mortality rates, compared with conventional electrocoagulation models that carry greater risk of bleeding and infection (54). Although these differences in surgical approach for the 5/6 Nx model should not be overlooked, both models have been shown to produce similar levels of renal dysfunction and histopathology (54). In addition to lower mortality rates, the adenine model circumvents the potential confounding impact of subtle surgical differences on renal and systemic pathophysiology. Considering the cost, time, technical ability, and effort to perform the surgery and postoperative care, as well as the increased variability in outcomes (i.e., GFR, BUN, FGF23, PTH, serum phosphate and Ca²⁺, and uremic toxins levels) reported herein, there is appreciable value in using the adenine diet model in preclinical research.

As a result of insufficient kidney function, patients with CKD display increased levels of uremic toxins, which have been linked to chronic inflammation, vascular calcification, and oxidative stress, contributing to the high risk of cardiovascular complications (55, 56). In addition, uremic toxins have been shown to negatively impact skeletal muscle health (15, 57, 58). For example, accumulation of serum uremic toxins in patients with kidney disease is associated with sarcopenia and mitochondriopathy (28, 39, 59). These pathological effects may be worsened in severe stages of renal disease (i.e., end-stage renal disease) despite treatment efforts such as hemodialysis because of incomplete filtration of uremic toxins by the conventional dialysis membrane, particularly for protein-bound toxins (60). In this study, we examined tryptophan-derived uremic toxins in both adenine and 5/6 Nx models of CKD. Both murine CKD models produced similar profiles of increased uremic toxins, including indoxyl sulfate and kynurenines, consistent with earlier studies (61, 62). Despite similar changes in kynurenines and indoxyl sulfate, there were noteworthy differences between models. For example, p-cresol sulfate, which plays a significant role in endothelial dysfunction and is associated with cardiovascular disease risk (41, 46), was only elevated in adenine diet-fed mice. The exact mechanisms underlying the difference of p-cresol sulfate levels between adenine and 5/6 Nx murine CKD models is not known at this time. As shown in Fig. 3, the adenine model produced more consistent increases in uremic toxins compared with 5/6 Nx mice. This is demonstrated by the standard deviations for key uremic toxins such as indoxyl sulfate, kynurenine, and kynurenic acid. Further analysis using Pearson correlations revealed that there were significant inverse correlations between GFR and uremic toxins in both models of CKD (Fig. 6).

Muscle wasting, defined as the loss of muscle mass and reduced fiber CSA, is one of the most notable characteristics observed in patients with CKD, and contributes to exercise intolerance, frailty, and poor quality of life (63, 64). In the CKD condition, negative net protein balance in which the rate of protein degradation is greater than the rate of protein synthesis has been known to contribute to the increased whole body protein turnover (7). Numerous cellular pathways have been shown to contribute to elevated protein catabolism in CKD, including impaired satellite cell function and muscle growth (65), suppressed anabolic cell signaling such as insulin/insulin-like growth factor-I-phospho-AKT-mammalian target of rapamycin-p70S6 kinase pathways (66), elevated protein degradation (i.e., caspase 3 and the ubiquitin-proteasome system) (67), increased myostatin (a negative regulator of muscle growth) (68), and altered microRNA expression (69). In addition to the factors such as inflammation and oxidative stress, uremic toxicity plays a significant role in the catabolic condition in CKD (39). The role of tryptophan-derived metabolites, such as kynurenine and indoxyl sulfate, in muscle wasting has been suggested (15, 57, 58). For example, Berru et al. (14) showed that C2C12 cells in DMEM growth medium supplemented with 5% serum acquired from adenine-induced CKD mice for 24 h resulted in myotube atrophy. McGee-Lawrence et al. (58) showed that treatment with kynurenine through intraperitoneal injection into young healthy mice (with normal kidney function) resulted in decreased muscle fiber size. Similarly, Sato et al. (15) reported that adenine diet-induced CKD mice accumulated indoxyl sulfate in muscle tissues and decreased myofiber CSA in the tibialis anterior muscle. In the present study, we observed that male CKD mice, regardless of model, displayed smaller muscle mass and myofiber CSA in the EDL and SOL muscles, which were strongly associated with elevated serum concentrations of uremic toxins (Fig. 6). The exact mechanisms by which uremic toxins may mediate muscle atrophy are not fully known but represents an exciting avenue for therapeutic discovery.

In contrast to male CKD mice, muscle mass and fiber CSA were modestly decreased (6–13%) in female CKD mice compared with their respective controls; however, this was not statistically significant. A plausible explanation of the sex-dependent impact of CKD on muscle wasting may be related to the protective effects of estrogen on skeletal muscle because the female mice used in this study were young (10 wk old) and had normal menstrual cycles: two characteristics that do not match female patients with CKD well. Clinical studies have reported that CKD is more prevalent in postmenopausal women (70), but also that menopause accelerates the disease progression, including skeletal muscle weakness (71). Accumulating evidence from animal studies indicates that estrogen has a protective role in the development of renal disease (72), bone metabolism (73), regulation of muscle mass (74, 75), and mitochondrial function (76, 77). Consistent with this hypothesis, estrogen deficiency induced by ovariectomy resulted in more severe glomeruloscelerosis and increased BUN compared with animals with intact ovaries (72). Kitajima et al. (78) reported that surgical removal of the ovaries from female mice results in muscle atrophy, which could be rescued by estradiol administration. Taken together, the results of the present study and others (28, 78) suggest the need to consider the age, menstrual status, and role of estrogens in preclinical CKD studies involving female mice.

In addition to muscle atrophy, skeletal muscle abnormalities prevalent in CKD include impaired mitochondrial biogenesis and/or mitochondrial function, which is believed to contribute to poor physical performance and exercise intolerance (9, 28, 79–81). Kestenbaum et al. (80) showed that compared with age-matched controls with normal renal function, patients with CKD had poor physical performance (6-min walk test), which was associated with impaired mitochondrial oxidative capacity as assessed by phosphorus magnetic resonance spectroscopy. Nishikawa et al. (11) showed that C57BL/6J mice that underwent partial Nx exhibited impaired mitochondrial biogenesis and reduced treadmill exercise capacity compared with sham-operated mice. Consistent with these clinical and preclinical studies, in the present study, we report that ADP-stimulated mitochondrial respiratory capacity was diminished (∼20%) in CKD mice regardless of the model, confirming a recent report in the adenine model (28). Taken together, these findings suggest that both adenine diet and 5/6 Nx models of CKD result in similar mitochondrial impairments in mice.

It has been established that the background genetic strain of mice can influence pathophysiology in numerous models of renal dysfunction (82–84). Indeed, C57BL6 mice, such as those used in this study, have been shown to be resistant to some aspects of CKD pathophysiology with the 5/6 Nx model, including cardiac fibrosis, proteinuria, glomerulosclerosis, and hypertension (82, 83, 85). Notably, the resistance displayed by the C57BL6 strain in the 5/6 Nx model does not appear in other models of renal dysfunction, including the adenine diet model (28), oxalate diet model (86), unilateral ureteral obstruction (84), and nephrotoxin treatment (87). Nonetheless, the results of the present study suggest that C57BL6 mice are not resistant to CKD-associated skeletal myopathy with the 5/6 Nx model, despite their relative resistance to some other CKD pathologies.

In conclusion, preclinical animal models of CKD are important tools required to understand the underlying mechanisms of renal disease as well as the pathophysiological consequences, including skeletal myopathy. Moreover, they serve as a critical research tool that allows testing of novel therapeutics that may translate to improve clinical outcomes in human patients with CKD. In the present study, we used two mouse models of CKD, adenine-induced nephropathy and 5/6 Nx, to compare their impact on the levels of uremic toxins, muscle mass, fiber CSA, and mitochondrial function. Both CKD models produced equivalent levels of renal dysfunction and uremia, which was accompanied near-identical reductions in skeletal muscle mass and mitochondrial respiratory function. Another significant finding in the present study was the striking sex differences in renal pathophysiology and muscle wasting between male and female mice. Notably, female mice displayed similar reductions in GFR compared with male mice but did not exhibit the same changes in BUN, serum Ca²⁺ or PTH levels, or muscle atrophy that male mice displayed. Considering that the prevalence of CKD is high in postmenopausal women (88) and that most preclinical CKD studies have been conducted in male animals (14, 15, 39), future studies should carefully consider the inclusion of female mice and possibilities of modeling the postmenopausal condition to better align clinically. Compared with the traditionally used 5/6 Nx model, the adenine model did not cause mortality and produced more consistent levels of GFR, BUN, and uremic toxin accumulation. Coupled with the ease of implementation, these findings support the use of the adenine diet model in mice for studies exploring adverse muscle impact.

SUPPLEMENTAL DATA

The data that support this study are available at https://doi.org/10.6084/m9.figshare.14317331.v2.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants R01HL149704 (to T.E.R.) and R01HL148597 (to S.T.S.), as well as a seed grant from the UF Office of Research (to T.E.R.). Salary support was also provided by American Heart Association Grant 18CDA34110044 (to T.E.R.).

DISCLAIMERS

The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

K.K., S.T.S., and T.E.R. conceived and designed research; K.K., E.M.A., T.T., G.L., Z.R.S., and T.A.C. performed experiments; K.K., E.M.A., T.T., T.A.C., and T.E.R. analyzed data; K.K., E.M.A., T.T., G.L., K.A.O., S.T.S., and T.E.R. interpreted results of experiments; K.K. and T.E.R. prepared figures; K.K. and T.E.R. drafted manuscript; K.K., E.M.A., T.T., G.L., Z.R.S., T.A.C., K.A.O., S.T.S., and T.E.R. edited and revised manuscript; K.K., E.M.A., T.T., G.L., Z.R.S., T.A.C., K.A.O., S.T.S., and T.E.R. approved final version of manuscript.