Irisin fails to rescue bone loss in a rat model of chronic kidney disease – mineral bone disorder (CKD-MBD) but interacts with parathyroid hormone (PTH) in vivo and in vitro in IDG-SW3 osteocytes

Abstract

Introduction

Chronic kidney disease progression leads to bone loss and increased fracture and mortality risk. The myokine irisin, produced by muscle during exercise, has been shown to exert protective effects on bone in healthy animals and exhibits a potential interaction with PTH in vitro. The aims of this study were to determine the effect of irisin on bone in CKD-MBD and determine if an interaction exists with PTH in vitro and in vivo.

Methods

Early and mature IDG-SW3 osteocytes were treated with recombinant irisin, PTH, or both. Mineralization and osteocyte cell markers were assessed. In a similar 2 × 2 design, male Cy/+_IU (CKD-MBD) rats were treated with or without irisin under high and low PTH conditions. Dynamic histomorphometry, bone structure, mechanical properties, and musculoskeletal gene markers were assessed.

Results

In vitro, irisin alone downregulated Dkk1 and Sost, inhibitors of bone formation, and PTH receptor mRNA expression, primarily in early osteocytes. However, in the presence of PTH, no effects of irisin were observed. In vivo, the administration of irisin to animals with progressive CKD also downregulated PTH receptor 1 expression but did not affect bone turnover or structure or mechanical properties.

Conclusions

Irisin did not meaningfully ameliorate bone loss in an animal model of CKD-MBD but did interact with PTH by downregulating its receptor in vitro in osteocytes and in vivo in bone.

Highlights

• •Irisin downregulated inhibitors of the Wnt/β-catenin bone formation pathway in osteocytes.
• •Irisin downregulated PTH receptor in early but not mature osteocytes and in CKD-MBD animals with high but not low PTH.
• •Bone loss was not attenuated by recombinant irisin injection in an animal model of CKD-MBD.

1. Introduction

Chronic kidney disease (CKD) affects 1 in 7 individuals in the U.S. (Chronic Kidney Disease in the United States, 2023) Disease progression is associated with alterations in bone and mineral metabolism, including abnormalities in calcium, phosphorus, parathyroid hormone (PTH), and fibroblast growth factor 23 (FGF23) levels; alterations in bone turnover, mineralization, volume, linear growth, and strength; and increased extraskeletal calcification. Collectively, these abnormalities are termed chronic kidney disease-mineral bone disorder (CKD-MBD) and are nearly ubiquitous in patients with advanced CKD (Chronic Kidney Disease in the United States, 2023; Mineral and Bone Disorder in Chronic Kidney Disease, 2021).

End stage kidney disease leads to a nearly 5-fold increased risk of hip fracture (Moe and Nickolas, 2016), which is associated with increased mortality risk compared to those without CKD (Brauer et al., 2009; Kim et al., 2016). Exercise can be an effective strategy to strengthen bones (Haque et al., 2024), but it has demonstrated variable success and is often not well tolerated in patients with advanced CKD due to severe fatigue (Kirkman et al., 2021), perhaps due in part to cardiovascular disease and abnormalities in muscle metabolism, findings we identified in rodent models and in some studies in humans with CKD (Avin et al., 2016; Avin et al., 2021). Historical medical treatment of bone disease in CKD-MBD is prevention of adverse outcomes associated with secondary hyperparathyroidism (HPT) (Tomasello, 2008). Despite the advent of effective PTH-lowering therapies, including calcimimetics and vitamin D analogs, fractures in patients with CKD have not substantially decreased (Nair et al., 2013), and a heterogeneity in bone formation rates for a given level of blood PTH persists (Barreto et al., 2008). This indicates that additional factors may contribute to CKD-MBD pathophysiology or modulate the skeletal effects of PTH. One such factor may be the myokine irisin.

Bone and muscle are known to affect one other's actions based on biomechanical function (Avin et al., 2015). However, physiologic changes in muscle can also induce bone changes locally through muscle-produced cytokines, known as myokines. Irisin, discovered in 2012 by Boström et al., is induced by exercise, additionally deeming it an exerkine. Exercise increases transcriptional coactivator PGC-1α expression in skeletal muscle, which upregulates transmembrane precursor fibronectin type III domain-containing 5 (FNDC5), the precursor to functional irisin (Bostrom et al., 2012). The cleavage mechanism of FNDC5 into irisin remains unknown, though some have proposed a role of A Disintegrin And Metalloproteinase Domain 10 (ADAM10) (Yu et al., 2019). Irisin can then act in an autocrine, paracrine, or endocrine fashion to affect several tissues throughout the body. Extensive research since its discovery has uncovered a vast number of functions, including browning of adipose tissue (Bostrom et al., 2012; Fox et al., 2018), improved insulin sensitivity (Song et al., 2021), neuroprotection via increased brain-derived neurotrophic factor (BDNF) (Wrann et al., 2013), reduced oxidative stress (Ren et al., 2020; Wen et al., 2025), and, importantly, improvements in bone homeostasis (Kornel et al., 2021). Irisin binds to the αV/β5 integrin receptor in bone cells to affect osteoblasts, osteoclasts, and osteocytes either directly or indirectly (Kim et al., 2018). In vitro treatment results in increased osteoblast differentiation and mineralization (Qiao et al., 2016), reduced osteocyte apoptosis (Storlino et al., 2020), and reduced osteoclast differentiation (Ma et al., 2018). In vivo studies in non-CKD animals have shown irisin injections improved cortical bone volume, thickness, and mass in C57BL/6 male mice and protected against osteoporosis in ovariectomized female rats (Colaianni et al., 2015; Morgan et al., 2021). Clinical studies in elderly, osteoporotic, and patients with advanced renal disease have found an association between reduced circulating irisin and increased risk of fracture and bone disease (Shen et al., 2024; Lu et al., 2021; Yan et al., 2018).

The effect of irisin in CKD is significantly less understood. Circulating irisin is reduced in CKD according to a 2022 systematic review (Gan et al., 2022), and levels decrease with progressive loss of kidney function (Sadeghi Shad et al., 2019), but the assays utilized to obtain these results are notoriously inconsistent (Albrecht et al., 2015; Barja-Fernandez et al., 2016; Maak et al., 2021). Studies examining the effects of recombinant irisin administration on bone in validated CKD animal models are lacking. Those published to date have administered irisin to C57BL/6 mice with CKD induced by partial nephrectomy (Wu et al., 2025; Kawao et al., 2022). This animal model presents significant limitations for studying CKD-MBD because C57BL/6 mice are resistant to important CKD-induced pathophysiological changes, including hypertension and renal fibrosis, by partial nephrectomy (Bao et al., 2022). Further, because secondary HPT is a key pathophysiological manifestation, understanding the involvement of PTH in irisin's musculoskeletal action is essential. Palermo et al. (2019) first investigated a potential interplay between irisin and PTH; their group found a reduction in circulating irisin in postmenopausal women with primary HPT and noted teriparatide treatment downregulated FNDC5 expression in myotube cultures, and irisin treatment downregulated PTH receptor expression in osteoblast cultures (Palermo et al., 2019). This interaction has not yet been confirmed in vivo in any animal model or in osteocytes, the master regulators of bone (Schaffler et al., 2014). The aim of this study was to determine the effect of irisin and its interaction with PTH in the setting of high and low PTH in vitro in osteocytes and in vivo on bone in a model of progressive CKD-MBD. We hypothesized that high PTH in CKD-MBD blunts the protective effects of irisin in osteocytes and on bone.

2. Materials and methods

2.1. Cell culture

For this study, the IDG-SW3 cell line was utilized to represent early and mature osteocytes. The cells were originally derived from the long bone of the Immortomouse, which expresses a thermolabile SV40 large T antigen regulated by interferon γ (IFN-γ) crossed with mice carrying the Dmp1 promoter driving GFP to visualize osteocytes (Woo et al., 2011). The cells expanded in permissive conditions (33 °C/5% CO2 in α-MEM with 10% fetal bovine serum, 100 units/mL of penicillin, 50 μg/mL of streptomycin, and 50 U/mL of IFN-γ, Gibco Life Technologies, Carlsbad, CA) on rat tail type 1 collagen–coated culture plates. To induce osteogenesis, cells were plated at 1 × 10⁴ cells/cm² in osteogenic conditions (37 °C/8% CO2 with 50 μg/mL of ascorbic acid and 4 mM β-glycerophosphate in the absence of interferon gamma) in which the cells differentiate from a late osteoblast to early osteocyte (minimal mineralization) at day 14 to a mature osteocyte at day 35 (Woo et al., 2011; Chen et al., 2024). This cell line was chosen because in addition to producing sclerostin, a marker of mature osteocytes, they secrete fibroblast growth factor (FGF23), an important hormone in CKD, unlike the MLO-Y4 osteocyte-like cell line.

2.2. In vitro experimental design

IDG-SW3 osteocytes were cultured on collagen coated dishes with osteogenic media for 14 (early) or 35 (mature) days along with no treatment, 8.33 nM (100 ng/mL) recombinant irisin (Storlino et al., 2020; He et al., 2020) (PeproTech), 100 nM (412 ng/mL), 1–34 PTH (Che et al., 2022; St John et al., 2015; Jilka et al., 1999) (Sigma-Aldrich) or both in a 2 × 2 design when cell culture media was changed thrice weekly. The PTH dose was chosen based on our published work where it increased the Tnfsf11/Tnfrsf11b ratio (Rankl/Opg) and cAMP secretion in these cells (Chen et al., 2024). The irisin dose was chosen based on results from Palermo et al. where it downregulated PTH receptor mRNA in osteoblasts (Palermo et al., 2019) and other papers showing irisin increased osteoblast proliferation and differentiation and reduced osteocyte Sost and Dkk1 mRNA (Qiao et al., 2016; Storlino et al., 2020; He et al., 2020).

2.3. Alkaline phosphatase activity and mineralization

Alkaline phosphatase (ALP) activity was measured using p-nitrophenyl substrate supplied in an ALP assay kit (Pointe Scientific, Canton, MI) and normalized by protein content. To assess mineralization, osteocytes were decalcified with 0.6 N HCl for 24 h and the calcium content of HCl supernatants determined colorimetrically by the o-cresolphthalein complexone method (Calcium kit; Pointe Scientific, Canton, MI) and normalized to protein content as previously described (Chen et al., 2002).

2.4. Western blotting for MAP kinase signaling

IDG-SW3 osteocytes were cultured for 14 or 35 days with osteogenic media and treated with 100 ng/mL recombinant irisin for 5, 10, 15 or 30 min prior to harvest. The activation of MAP kinase (p38 and p44/p42/ERK1/2) was assessed using Western blotting as previously published (Chen et al., 2021). Band intensity was analyzed by the ChemiDoc MP Imaging System (Imaging Lab 4.0; Bio-Rad), and p38 or p44/p42/ERK1/2 activation was quantified by normalizing phosphorylated p38 or p44/p42 to total protein.

2.5. In vivo experimental design

Male Cy/⁺_IU rats (CKD hereafter) demonstrate an autosomal dominant progressive cystic kidney disease that is not homologous to human polycystic kidney disease. CKD is spontaneous on a normal phosphate diet and slowly progressive. Animals develop the tripartite manifestation of CKD-MBD (biochemical abnormalities, extraskeletal calcification, and abnormal bone) (Moe et al., 2009). These animals also demonstrate impaired skeletal muscle strength (Organ et al., 2016). Female rats with or without an ovariectomy do not develop end-stage kidney disease (Cowley Jr. et al., 1997; Vorland et al., 2019), thus only male animals and their normal littermates were used. Cy/⁺_IU rats show that BUN and creatinine are elevated by 10 weeks; FGF23 and PTH are mildly elevated by 20 weeks (~50% normal kidney function); skeletal abnormalities are evident by 30 weeks (~20–25% normal kidney function) with rapid worsening of CKD-MBD between 30 and 33 weeks of age (~10% normal kidney function) (Moe et al., 2014; Biruete et al., 2023). These CKD-MBD changes include severe hyperparathyroidism, hyperphosphatemia, and increases in cortical porosity, paralleling changes seen in humans with CKD (Isakova et al., 2011).

CKD animals aged 22 weeks were divided into four groups (n = 12–14 per group): half the animals received 2% calcium gluconate in their water to continuously suppress PTH (Moe et al., 2014); half received normal drinking water. After 5 weeks, half from each group additionally received 30 μg/kg recombinant irisin (PeproTech) injections thrice weekly (dose based on review of published literature where irisin elicited bone changes (Metzger et al., 2019; Chen et al., 2022; Bulmus et al., 2022)) for 5 weeks (weeks 28–32). 21 and 14 days prior to sacrifice, rats were double labeled with subcutaneous calcein injections to permit dynamic histomorphometry. Four days prior to sacrifice, animals were placed in a cage with a voluntary activity wheel for 48 h as previously described, to determine if differences in activity could account for musculoskeletal changes (Avin et al., 2019). At 32–33 weeks, animals were anesthetized, underwent cardiac puncture for terminal blood draw, euthanized, and tissues were collected for analyses. Normal (non-CKD) littermates were untreated and euthanized at 33 weeks, serving as a reference range for results and not included in analysis as we have previously demonstrated dramatic differences between CKD and normal animals, and this difference was not of scientific interest in this study (Moe et al., 2009). Animals were assigned randomly from each litter. The Indiana Institutional Animal Care and Use Committee approved all procedures prior to initiating these studies. This study was run in parallel with another in which the same animals (Normal, CKD, CKD + 2% calcium) were also used for analysis (Allen et al., 2026).

2.6. Blood biochemistries

Serum creatinine and plasma blood urea nitrogen, calcium, and phosphorus were measured using colorimetric assays (Pointe Scientific or BioAssay Systems). Terminal blood collection, 48 to 72 h after the last irisin injection, was used to measure plasma intact PTH, intact FGF23, oxidative stress marker 8-hydroxy-20-deoxyguanosine (8-OHdG), and irisin were determined by enzyme-linked immunosorbent assay (ELISA) (Quidel, Enzo Life Sciences or Phoenix Pharmaceuticals). Serum bone markers of bone formation procollagen type 1 N-terminal propeptide (P1NP) and bone resorption Tartrate-resistant acid phosphatase (TRAcP 5b) were also determined by ELISA (Immuno Diagnostic Systems).

2.7. Bone mechanical testing

Following harvest, right femora were collected and wrapped in saline-soaked gauze before freezing. Immediately prior to testing, femora were left in saline-soaked gauze to maintain hydration and thawed at room temperature. Bones were measured for anterior-posterior diameter using calipers and tested to failure using a 3-point bending (lower support span 18 mm) mechanical test system (Test Resources). A 667 N load cell and a 2 mm/min displacement rate were used for all tests. Testing was conducted with the anterior surface of the femur in compression with a loading point at 50% bone length. All bones were pre-loaded with ~0.5 N. Mean polar moment of inertia was obtained from μCT at the midshaft (described below). Structural and estimated material properties were determined by a custom MATLAB script.

2.8. Bone architecture

Tibiae were harvested and fixed in 10% neutral buffered formalin for 48 h and stored in 70% ethanol. Left proximal tibia was scanned by μCT (Skyscan 1272; Bruker) with a 0.5 mm aluminum filter and a 12 μm voxel size. Trabecular bone volume fraction (BV/TV) of the secondary spongiosum was determined in a 1 mm region. Cortical measurements were obtained 4 mm below the distal end of the trabecular region. Cortical periosteal and endosteal surfaces were hand-traced to isolate the intracortical tissue from 5 contiguous slices. Cortical porosity was defined as the void space within the cortex. Cortical bone area was determined using a binary threshold of 110–255. Cortical thickness was obtained using a flooded binary threshold of 0–255 and defined as the width of the endosteal to periosteal surfaces without porosity. Following mechanical testing, right femora mid-diaphyses were also scanned using the same settings as above, near the site of fracture, at approximate mid-shaft. Cortical parameters were assessed to normalize mechanical properties. Nomenclature was reported as per American Society of Bone and Mineral Research (ASBMR) guidelines (Bouxsein et al., 2010).

2.9. Bone histomorphometry

Following μCT, proximal tibiae were serially dehydrated through graded steps of ethanol and embedded in methyl methacrylate (Sigma-Aldrich). Frontal sections were cut 4 μm thick. Unstained slides were coverslipped, and dynamic histomorphometry (BIOQUANT) was assessed using fluorochrome calcein labels. A trabecular region of interest was selected for each specimen to measure total bone surface (BS), single-labeled surface (sLS), double-labeled surface (dLS), and interlabel distances at 20× magnification. Primary spongiosa and endocortical surfaces were excluded. Mineralized surface to bone surface (MS/BS; [dLS+(sLS/2)]/BS*100), mineral apposition rate (MAR; average interlabel distance/7 days), and bone formation rate (BFR/BS; [MS/BS*MAR]*3.65) were calculated. Osteoclast-covered trabecular surfaces (Oc.S/BS) were assessed with an additional 4 μm slide with Von Kossa/MacNeal staining. A standard trabecular region of interest as listed above was analyzed at 40× magnification. Nomenclature follows ASBMR standard guidelines (Nickolas et al., 2020). All analyses within a data set were analyzed by the same individual.

2.10. RNA isolation and real-time qPCR

Total RNA from IDG-SW3 osteocytes, soleus muscle, extensor digitorum longus (EDL) muscle, and right tibia after flushing bone marrow with RNAase-free saline was isolated using miRNeasy Mini Kit (Qiagen) as previously published (Avin et al., 2019; Nickolas et al., 2020). Expression of specific gene markers were analyzed by real-time polymerase chain reaction (PCR) using Taqman gene expression assays (TaqMan MGP probes, FAM dye-labeled, Applied Biosystems) using QuantStudio 3 systems. The cycle number at which the amplification plot crosses the threshold was calculated (CT), and the ∆∆CT method was used to analyze the relative changes in messenger RNA (mRNA) expression and normalized by β-actin (Chen et al., 2018).

2.11. Statistics

The experimental design compared IDG-SW3 osteocytes or CKD rats with and without irisin in the setting of high or low PTH for a 2 × 2 design. For the osteocytes, high and low PTH were achieved through the addition or non-addition of 1–34 PTH. For the rat study, CKD animals have high PTH, and thus the low PTH was achieved by the addition of calcium in the drinking water. Data from age-matched normal (non-CKD) littermates are shown for comparison purposes only and were not included in the statistical analyses. Analysis was conducted by first excluding outliers via ROUT (Q = 1%) apart from cortical porosity and heart calcification, endpoints with expected wide inter-animal variability in which all data were included. This was followed by a normality test (p < 0.05 by Shapiro Wilk), and non-normal data was log transformed for analysis. A 2-way ANOVA followed by Tukey's multiple comparisons test was used to determine the effect of irisin, PTH, and their interaction on endpoints. For consistency and readability, we present the 2-way ANOVA main/overall and interaction effects in the top right corners and the Tukey's results as individual group comparisons. Western blotting was assessed by t-test; physical activity was assessed by 1-way ANOVA; and Fndc5/PTH associations were analyzed simple linear regression. Results are expressed as mean ± standard deviation (SD) with a p < 0.05 considered significant.

3. Results

3.1. In vitro osteocytes

IDG-SW3 osteocytes were cultured in osteogenic media with 100 ng/mL recombinant irisin, 100 nM (412 ng/mL) PTH, or both for 14 or 35 days, representative of early and mature osteocytes respectively. Irisin had a main effect of increasing mineralization assessed by calcification in early osteocytes (p = 0.003), particularly in those also treated with PTH (p = 0.02). However, irisin had no effect on mineralization in mature osteocytes or on alkaline phosphatase (ALP) activity at either timepoint (Fig. 1A–D). Importantly, irisin significantly downregulated PTH receptor 1 (Pth1r) mRNA expression by approximately 25% (p = 0.0001) in early osteocytes without PTH and demonstrated a significant interaction with PTH (p = 0.01) but had no effect in mature osteocytes (Fig. 1E, F). Tnfrsf11b (gene encoding osteoprotegerin [OPG]) was marginally increased in early cells and reduced in mature cells treated with PTH while Tnfsf11 (gene encoding receptor activator of nuclear factor kappa-Β ligand [RANKL]) was increased by PTH in both. PTH increased the ratio of Tnfsf11/Tnfrsf11b (Rankl/Opg) and reduced the expression of osteocyte marker fibroblast growth factor 23 (Fgf23) mRNA and osteoblast differentiation marker runt-related transcription factor 2 (Runx2). In contrast, irisin had no effect on the mRNA expression of these genes under either PTH condition (Supplementary Fig. 1). However, irisin had a main effect in reducing Wnt pathway inhibitor dickkopf-related protein 1 (Dkk1) mRNA expression in early (p = 0.003) but not mature osteocytes (Fig. 2A, B). The Wnt inhibitor sclerostin (Sost) expression was too low to quantify at D14 (Woo et al., 2011); however, at D35 (mature osteocytes), irisin reduced Sost expression without PTH (p = 0.04), but the reduction with PTH was much more pronounced, negating any evaluation of an additive effect (Fig. 2C).

Fig. 1.

Prolonged expsoure to irisin in early osteocytes in vitro alters mineralization and PTH receptor expression. IDG-SW3 osteocytes were cultured and treated with 100 ng/mL recombinant irisin, 100 nM (412 ng/mL) PTH, or both for 14 (early) or 35 days (mature). The effect of irisin (green bars) was compared with and without PTH treatment by two way ANOVA to examine the effects and interaction of irisin and PTH with p < 0.05 considered significant. The results demonstrate that irisin treatment in early osteocytes (day 14) had a main effect in increasing mineralization assessed by calcium content (A), had no effect on alkaline phosphatase activity (C) but reduced PTH receptor (E). However, there was no effect of irisin on mature osteoctyes for any of these measures (B, D, F). These effects were more pronounced in D14 cells, demonstrating a predominant effect of irisin in early (osteoblast-like) osteocytes, and were outweighed by the dominating effects of PTH. Data are shown as mean ± SD (n = 9–12). Abbreviations: alkaline phosphatase (ALP), parathyroid hormone receptor 1 (Pth1r). ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Fig. 2.

Irisin reduced Dkk1 and Sost gene expression in vitro in osteocytes. IDG-SW3 osteocytes were cultured and treated with 100 ng/mL recombinant irisin, 100 nM (412 ng/mL) PTH, or both for 14 (early) or 35 days (mature). The effect of irisin (green bars) was compared with and without PTH treatment by two way ANOVA to examine the effects and interaction of irisin and PTH with p < 0.05 considered significant. Irisin had a main effect reducing Dkk1 expression in early but not mature osteocytes (A, B). Sost was not expressed in early osteocytes but was reduced in mature osteoctyes (C). These results suggest that irisin may signal through Wnt/β-catenin pathway in osteocytes. Data are shown as mean ± SD (n = 9–12). Abbreviations: dickkopf-1 (Dkk1). ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Previous literature has suggested that irisin acts via the p38 and p44/p42/ERK1/2 pathway to promote differentiation in osteoblast and MLO-Y4 osteocyte cell lines (Qiao et al., 2016; He et al., 2020). Therefore, we treated IDG-SW3 cells with irisin for 5, 10, 15 or 30 min prior to harvest. At each of these timepoints, the expression of phosphorylated p38 and p44/p42/ERK1/2 protein normalized by total protein was not different between those treated with 100 ng/mL recombinant irisin and vehicle in both early and mature cells (30 min data shown in Supplementary Fig. 2).

3.2. In vivo animal study

3.2.1. Effect of recombinant irisin on blood biochemistries in a CKD-MBD rat model

We compared the effects of irisin under high and low PTH conditions, the latter induced by calcium supplementation in the drinking water (all data in Supplementary Table 1). There were no differences in body weights between treatment groups. There was no difference in circulating plasma irisin between normal and CKD control animals, but groups treated with irisin had significantly higher levels of circulating plasma irisin at euthanasia (CKD control: 14.63 ± 4.3 vs Irisin: 215.65 ± 146.69 ng/mL) (2% calcium only: 13.33 ± 1.68 ng/mL vs. 2% calcium + Irisin: 228.98 ± 211.56 ng/mL) (p < 0.0001), validating the delivery of recombinant irisin into the bloodstream. There were no effects of calcium/PTH level nor irisin on blood urea nitrogen (BUN), but an interaction effect (p = 0.03) and a main effect of irisin on creatinine (p = 0.01) but no differences in individual group comparisons (Fig. 3A, B). Treatment with 2% calcium water led to a modestly higher main effect of calcium/PTH level on plasma calcium (p = 0.003) but no effect in individual group comparisons and effectively suppressed PTH in those groups (p < 0.0001) (Fig. 3C, D). The calcium drinking water/reduced PTH also had a main effect in lowering plasma phosphorus compared to untreated CKD (p = 0.004) (Fig. 3E) while irisin had no effect. Irisin had a main effect in increasing serum FGF23 (p = 0.02), principally in the low PTH group, but there were no individual group differences in the post-hoc analyses (Fig. 3F). There was no main effect of irisin on serum tartrate resistant acid phosphatase (Supplementary Table 1). However, irisin reduced levels of procollagen type 1 intact N-terminal propeptide (P1NP) in the low PTH/calcium group (p = 0.002) (Fig. 3G). Treatment with 2% calcium water, irisin, and the combination resulted in lower 8-hydroxy-2′-deoxyguanosine (8-OHdG) (irisin: p < 0.0001), a serum marker of DNA oxidative stress, confirming a known effect of irisin action (Wang et al., 2020) (Fig. 3H).

Fig. 3.

Effect of irisin treatment on biochemical endpoints in vivo in rats with CKD-MBD. Cy/+_IU male rats were treated with irisin (30 μg/kg irisin thrice weekly for 5 weeks) (from wk 28 to 32, approximately CKD stage 4–5) with and without calcium in the drinking water to suppress PTH in a 2 × 2 design. Normal littermates were sacrificed at the same age and serve as normal controls shown in the dotted lines. Two-way ANOVA examined the effects and interaction of irisin (green bars) and PTH with p < 0.05 considered significant. Irisin had no effect on measures of kidney function (BUN (A), creatinine (B)) in individual group comparisons. As exprected, calcium treatment effectively suppressed PTH (D) and had a main effect in increasing plasma calcium (C) and reducing plasma phosphorus (E). However, there was no effect of irisin on any of these measures or on FGF23 (F) in individual comparisons. Irisin reduced a marker of bone formation (G) and a marker of DNA oxidative stress, serving as a positive control (H). Data are shown as mean ± SD (n = 12–14). Abbreviations: blood urea nitrogen (BUN), parathyroid hormone (PTH), fibroblast growth factor 23 (FGF23), procollagen type I N-propeptide (P1NP), 8-Hydroxy-2′-deoxyguanosine (8-OHdG). ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

3.2.2. Effect of irisin on irisin on bone architecture and remodeling

As expected, there were differences between the low and high PTH groups on bone, but despite its effects on P1NP, irisin did not affect bone formation rate, mineral apposition rate, mineralizing surface, or osteoclast surface and did not demonstrate a statistical interaction with PTH (Fig. 4A–D, Supplementary Table 2). Similarly, 3-point bending and μCT displayed a consistent effect of PTH level in affecting bone structure and mechanical properties, but no effects of irisin were observed under either PTH condition (Table 1).

Fig. 4.

Irisin did not change bone turnover but reduced PTH receptor expression in vivo in rats with CKD-MBD. Cy/+_IU male rats were treated with irisin (30 μg/kg irisin thrice weekly for 5 weeks) (from wk 28 to 32, approximately CKD stage 4–5) with and without calcium in the drinking water to suppress PTH in a 2 × 2 design. Normal littermates were sacrificed at the same age and serve as normal controls shown in the dotted lines. Two-way ANOVA examined the effects and interaction of irisin (green bars) and PTH with p < 0.05 considered significant. Dynamic histomorphometry of the proximal tibia and qPCR of total tibia bone were performed after euthanasia. Irisin did not affect bone formation or resorption (A–D) but downregulated PTH receptor under high PTH conditions (E). It had a main effect in reducing Sost (sclerostin) gene expression but not in individual comparisons (F) and did not affect the Tnfsf11/Tnfrsf11b (Rankl/Opg) ratio (G). Data are shown as mean ± SD (n = 12–14). Abbreviations: bone formation rate (BFR/BS), mineral apposition rate (MAR), mineralizing surface (MS/BS), osteoclast surface (OcS/BS), PTH receptor 1 (Pth1r), gene for RANKL/OPG (receptor activator of nuclear factor-κB ligand/osteoprotegerin) (Tnfsf11/Tnfrsf11b). ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

Table 1.

Bone outcomes. μCT was performed on the proximal tibiae and midshaft femora. 3-pt-bending was performed on femora. Two way ANOVA examined the effects and interaction of irisin and PTH with p < 0.05 considered significant (bolded). Data are shown as mean ± SD (n = 12–14).

Variable	Normal littermates	CKD control	Irisin	2% calcium	Irisin + 2% calcium	2-way ANOVA
Groups	For reference only	CKD (High PTH)	CKD (High PTH) AND irisin	CKD (Low PTH)	CKD (Low PTH) AND irisin	Main Effect: PTH-level (Calcium treatment vs non)	Main Effect: Irisin	Interaction Effect
	n = 14	n = 12	n = 12	n = 13	n = 14
μCT: Proximal Tibia
Cortical porosity (%)	0.02 ± 0.01	6.76 ± 10.4	7.31 ± 6.58	0.28 ± 0.2	0.18 ± 0.09	p < 0.0001	p = 0.36	p = 0.3
Cortical area (mm2)	5.95 ± 0.34	5.64 ± 0.47	5.41 ± 0.58	6.07 ± 0.25	6.22 ± 0.18	p < 0.0001	p = 0.72	p = 0.1
Cortical thickness (mm)	0.44 ± 0.03	0.44 ± 0.03	0.42 ± 0.03	0.45 ± 0.01	0.44 ± 0.02	p = 0.03	p = 0.05	p = 0.19
Bone volume/Total volume BV/TV (%)	6.47 ± 1.18	5.35 ± 2.71	5.22 ± 1.84	10.4 ± 1.99	10.3 ± 3.19	p < 0.0001	p = 0.83	p = 0.98
Trabecular thickness (mm)	0.075 ± 0.005	0.079 ± 0.006	0.078 ± 0.005	0.08 ± 0.005	0.08 ± 0.006	p = 0.21	p = 0.81	p = 0.77
Trabecular separation (mm)	0.48 ± 0.08	0.58 ± 0.11	0.6 ± 0.1	0.4 ± 0.07	0.38 ± 0.06	p < 0.0001	p = 0.97	p = 0.39
Trabecular number (#/mm)	0.87 ± 0.16	0.68 ± 0.35	0.68 ± 0.26	1.3 ± 0.26	1.3 ± 0.36	p < 0.0001	p = 0.84	p = 0.87
μCT: Midshaft Femur
Cortical porosity (%)	0.07 ± 0.05	7.4 ± 7.4	8.1 ± 6.6	0.16 ± 0.14	0.21 ± 0.13	p < 0.0001	p = 0.21	p = 0.83
Cortical area (mm2)	7.7 ± 0.4	6.9 ± 0.47	6.7 ± 0.45	8.0 ± 0.47	8.1 ± 0.37	p < 0.0001	p = 0.65	p = 0.14
Cortical thickness (mm)	0.7 ± 0.02	0.68 ± 0.06	0.66 ± 0.06	0.76 ± 0.02	0.75 ± 0.03	p < 0.0001	p = 0.37	p = 0.43
Cortical mean polar moment of inertia (mm4)	21.7 ± 2.19	20.9 ± 2.53	20.5 ± 2.38	21 ± 2.53	22.2 ± 2.15	p = 0.2	p = 0.57	p = 0.25
Mechanical Testing
Ultimate load (N)	145 ± 11	99 ± 19	99 ± 20	137 ± 13	140 ± 16	p < 0.0001	p = 0.76	p = 0.86
Total displacement (μm)	707 ± 82	581 ± 68	595 ± 56	677 ± 99	679 ± 57	p < 0.0001	p = 0.71	p = 0.8
Stiffness (N/mm)	342 ± 31	257 ± 45	231 ± 10	318 ± 37	344 ± 23	p < 0.0001	p = 0.93	p = 0.01
Total work (mJ)	63 ± 9.4	36 ± 9.5	37 ± 8.3	56 ± 13	59 ± 9.7	p < 0.0001	p = 0.54	p = 0.62
Ultimate stress (MPa)	108 ± 10	74 ± 18	76 ± 16	104 ± 14	99 ± 14	p < 0.0001	p = 0.86	p = 0.45
Total stress (MPa)	46,884 ± 5967	37,015 ± 5642	37,840 ± 4326	43,958 ± 6488	43,817 ± 5956	p = 0.0002	p = 0.76	p = 0.71
Modulus (GPa)	3.9 ± 0.51	3 ± 0.69	3 ± 0.52	3.7 ± 0.55	3.8 ± 0.35	p < 0.0001	p = 0.88	p = 0.69
Total toughness (MPa)	3.1 ± 0.55	1.7 ± 0.58	1.8 ± 0.39	2.8 ± 0.7	2.7 ± 0.61	p < 0.0001	p = 0.83	p = 0.7

3.2.3. Irisin downregulated bone PTH receptor

Consistent with our in vitro data, irisin had a main effect in downregulating Pth1r mRNA in the tibia (p = 0.008), particularly in the high PTH group (p = 0.05) (Fig. 4E). Further, it had a main effect in downregulating Sost (p = 0.04) mRNA but had no effect on Tnfsf11/Tnfrsf11b ratio (genes encoding Rankl/Opg) (Fig. 4F, G).

3.2.4. Irisin and PTH interact on muscle properties

We confirmed no difference in physical activity between those treated with and without irisin to rule out a potential effect on the results (Supplementary Fig. 3). Irisin did not downregulate its precursor Fndc5 mRNA expression in the extensor digitorum longus (EDL) or in the soleus. Low PTH/calcium treatment had a main effect in significantly upregulating Fndc5 expression in both muscles (EDL: p = 0.02, soleus: p = 0.01) (Fig. 5A, B). Further, there was a negative correlation observed between plasma PTH and muscle Fndc5 in the soleus (R = −0.64, p < 0.0001) (Fig. 5D) which was more dramatic than the EDL (R = −0.34, p = 0.01) (Fig. 5C). A positive correlation between plasma calcium and Fndc5 was also observed (soleus: R = 0.42, p = 0.002) (EDL: R = 0.34, p = 0.02). Irisin alone downregulated A Disintegrin and Metalloproteinase 10 (Adam10), candidate FNDC5 cleavage enzyme (Yu et al., 2019), expression in the soleus (p = 0.002) and not the EDL (Fig. 5E) under high PTH conditions, but this effect was reversed under low PTH conditions/with calcium treatment (Fig. 5F). Irisin had a main effect in downregulating Pth1r expression in the soleus (p = 0.047) but no effect in the EDL (Supplementary Fig. 4).

Fig. 5.

Irisin and PTH may interact in muscle in rats with CKD-MBD. Cy/+_IU male rats were treated with irisin (30 μg/kg irisin thrice weekly for 5 weeks) (from wk 28 to 32, approximately CKD stage 4–5) with and without calcium in the drinking water to suppress PTH in a 2 × 2 design. Normal littermates were sacrificed at the same age and serve as normal controls shown in the dotted lines. Two-way ANOVA examined the effects and interaction of irisin (green bars) and PTH with p < 0.05 considered significant. Reducing PTH had a main effect in increasing irisin precursor Fndc5 but not in individual comparisons (A, B), and PTH and Fndc5 were negatively correlated by simple linear regression (C, D). Irisin reduced candidate cleavage enzyme Adam10 only in the high PTH group in the soleus (F). Data are shown as mean ± SD (n = 12–14). Abbreviations: extensor digitorum longus (EDL), fibronectin type III domain-containing protein 5 (Fndc5), parathyroid hormone (PTH), a disintegrin and metalloproteinase 10 (Adam10). ns: p > 0.05, *: p ≤ 0.05, **: p ≤ 0.01, ***: p ≤ 0.001, ****: p ≤ 0.0001.

4. Discussion

Progression of CKD is associated with a high risk of fractures and associated mortality (Moe and Nickolas, 2016; Kim et al., 2016; Nair et al., 2013), and secondary hyperparathyroidism has long been presumed as the primary driver of increased fracture risk (Shen et al., 2024). However, heterogeneity in bone formation rates for a given level of PTH remains (Barreto et al., 2008), and the advent of effective PTH-lowering therapies has inconsistently affected fracture rates (Nair et al., 2013), supporting the notion that other factors are at play. In this study, we examined the effect of the myokine irisin on bone under low and high PTH conditions in vitro in IDG-SW3 osteocytes and in vivo in a CKD-MBD animal model. We found that irisin downregulated Pth1r and inhibitors of the bone formation promoting Wnt pathway, Dkk1 and Sost, in IDG-SW3 osteocytes. In CKD animals, we confirmed that irisin downregulated bone Pth1r expression but did not affect bone turnover, structure, or mechanical endpoints.

Studies examining the effects of irisin on osteocyte cell lines have found that irisin increases cell survival, increases the Rankl/Opg ratio, and acts via the p-Erk/p-38 and Wnt/β-catenin pathways (Storlino et al., 2020; He et al., 2020). However, no published work to our knowledge has examined its effects in IDG-SW3 cells, which produce FGF23 (Woo et al., 2011), important for CKD pathophysiology (Komaba and Fukagawa, 2010), unlike the commonly used MLO-Y4 line. Further, this study represents the first to examine the long-term effects of irisin and its interaction with PTH in osteocytes. Osteocytes, previously thought of as quiescent and relatively inactive, are now recognized as the primary orchestrator of bone, integrating biochemical and mechanical signals to regulate the activity of osteoblasts and osteoclasts, and making them a vital cell type to understand physiologically (Schaffler et al., 2014). Palermo et al. found that an eight-hour exposure of irisin, at the same concentration as this study, also downregulated PTH receptor in MC3T3-E1 osteoblasts (Palermo et al., 2019). In our study, treatment with recombinant irisin continuously for 14 days or 35 days downregulated Pth1r expression in early but not mature osteocytes, demonstrating an interplay between PTH and irisin and suggesting the fully differentiated osteocytes may be less responsive (Schaffler et al., 2014). Our finding that irisin increased mineralization and downregulated Dkk1 and Pth1r in early but not mature osteocytes further supports that irisin may have a greater effect in earlier, osteoblast-like cells. We hypothesize this may be due to reductions in irisin receptor expression or sensitivity in mature osteocytes compared to osteoblasts. Additionally, we found that irisin did not affect phosphorylated (active) p42/44/Erk or p38 at multiple time points, indicating it acts through an alternate pathway in this cell line. Given the changes observed with Dkk1 in early and Sost in mature osteocytes, our data suggest that irisin may alternatively act through the Wnt/β-catenin pathway, a well-known target of Sost and Dkk1 (Bafico et al., 2001; Semënov et al., 2005), rather than the MAPK pathway in IDG-SW3 osteocytes. Further, nearly all of our results demonstrate an important finding that continuously high PTH tends to outweigh the effects of irisin, consistent with our previous findings that PTH negates the osteocyte effects induced by the uremic toxin indoxyl sulfate (Chen et al., 2024). These data suggest that the potent signal of high PTH can mask physiological effects of other molecules, including myokines like irisin, and supports that despite limitations, clinical suppression of PTH in patients with CKD is critical for bone health.

To examine the effect of irisin in vivo, we examined the effect of irisin in low or high PTH in the Cy/⁺_IU model of CKD-MBD. We administered 30 μg recombinant irisin/kg body weight thrice weekly to CKD-MBD male rats treated with or without 2% calcium gluconate water to continuously suppress PTH (Moe et al., 2014) or allowed sustain high PTH levels. We hypothesized irisin would increase bone formation and rescue bone loss with a greater effect in the group treated with calcium where the dominant effects of continuously elevated PTH are blunted. However, despite significantly increasing circulating irisin levels, exogenous irisin treatment did not significantly affect bone architecture or turnover or structure under either condition. The bone formation/collagen synthesis marker, P1NP, was reduced, particularly under low PTH conditions. However, dynamic histomorphometry demonstrated no effect of irisin on bone remodeling. Although P1NP is a useful biomarker, monomeric P1NP is renally cleared, causing artificially elevated values in CKD, and it does not describe bone turnover at the tissue level as accurately as histomorphometry (Nickolas, 2018; Vasikaran et al., 2014; Koivula et al., 2010). Of additional concern, animals treated with irisin had marginally higher creatinine, demonstrating a potentially deleterious effect on kidney function—a common concern with effective bone therapies such as bisphosphonates (Perazella and Markowitz, 2008). This is in contrast to the studies published to date that administered exogenous irisin to a CKD animal model where either a neutral or protective effect on kidney function was observed (Wu et al., 2025; Kawao et al., 2022; Wang et al., 2022; Wang et al., 2024). Irisin's excretion mechanism is unclear, but it is hypothesized to be at least partially renally cleared (Lv et al., 2015). Advancements in our understanding of irisin metabolism are vital in order to understand its safety in CKD-MBD.

We propose that a primary rationale for the lack of effect by irisin on bone was the severity of CKD and secondary hyperparathyroidism. Animals treated with irisin had a 14- to 17-fold increase in plasma irisin, and a dramatic reduction in a marker of oxidative stress, a known effect of the irisin (Wang et al., 2020), validating efficacy of the treatment. There are limited prior studies that have examined irisin's effect on bone in CKD; those that have found protective effects. Kawao et al. (2022) found 5/6 nephrectomised female C57BL/6 mice had reduced expression of muscle irisin which correlated with cortical bone mineral density (BMD) loss (Kawao et al., 2022). Wu et al. (2025) also used a 5/6th nephrectomy model in C57BL/6 male mice and found recombinant 100 μg/kg irisin injections beginning in advanced CKD and given daily for four weeks significantly upregulated osteoblast activation markers and improved trabecular bone volume despite elevated PTH (Wu et al., 2025). These opposing findings are likely due to the mouse models of CKD utilized and the dose and frequency of irisin administration. The 5/6th nephrectomy model is an acute kidney injury model with a less severe CKD phenotype, particularly in C57BL/6 mice (Bao et al., 2022; Leelahavanichkul et al., 2010). This model, which demonstrates a phenotype closer to acute kidney injury, is not well validated to study severe CKD-MBD because animals are resistant to vital pathophysiological changes such as renal fibrosis, glomerular damage, and hypertension in contrast to our model which develops a slowly progressive strong CKD-MBD phenotype. In addition, the Wu study used a high dose of 100 μg/kg daily while we administered 30 μg/kg thrice weekly. The concentration and dosing of irisin used in our model paralleled other published reports (Storlino et al., 2020; Colaianni et al., 2017) but suggests that, in the setting of very high PTH, greater doses or more frequent dosing may be required. These results are also consistent with our in vitro findings that the effects of irisin are outweighed by elevated PTH.

Although this study focused on the interaction between irisin and PTH at the level of bone, we recognize that muscle could be another potential site of crosstalk or overlap. The finding that low PTH conditions overall led to slightly increased Fndc5 expression in muscle and a significant negative correlation between Fndc5 and PTH further alludes to this possibility. Palermo et al. found that the incubation of myotubes in vitro with teriparatide (1–34 PTH) downregulated FNDC5 expression. However, in vivo studies are lacking. PTH may also alter the metabolism of irisin, and our findings of a reduction in Adam10 expression in the group treated with irisin under high PTH conditions supports a role of PTH in FNDC5 cleavage to irisin. Another potential explanation is that irisin could have downregulated ADAM10 indirectly by reducing inflammation and/or oxidative stress. ADAM10 is linked with several inflammatory pathways and is upregulated in CKD (Maretzky et al., 2008; Pabois et al., 2014; Maas et al., 2023), and we demonstrated a remarkable reduction in 8-OHdG, a marker of DNA oxidative stress (Valavanidis et al., 2009), in both groups treated with irisin. This downregulation was reversed in the group also treated with calcium, supporting the former hypothesis. We propose that either high PTH is permissive for irisin's action on muscle or calcium opposes an intracellular cascade activated by irisin. No direct relationship has been previously established between ADAM10 and PTH, warranting further investigation into a potential physiological interaction of PTH on irisin metabolism.

The assay chosen to quantify circulating plasma irisin for this study has been validated by mass spectrometry (Hew-Butler et al., 2015; Lee et al., 2014). However, it is important to interpret these results with caution as the validity of irisin ELISA and western blots can be limited by non-specific binding of fragments and cross reactivity of antibodies in commercially available assays (Albrecht et al., 2015). Further, the Cy/⁺_IU animal model introduces an inherent sex-specific limitation, important to irisin physiology as our group (Wilson, et al. Manuscript under revision) and others have shown (Shimonty et al., 2024).

In conclusion, we found that irisin increased mineralization and downregulated inhibitors of the Wnt/β-catenin pathway and PTH receptor expression in IDG-SW3 osteocytes. These effects were more pronounced in early, osteoblast-like cells than mature osteocytes. In a rat model of CKD-MBD, irisin administration did not affect bone turnover, architecture, or mechanical properties, suggesting that the severity of bone loss in end stage CKD is too severe for amelioration by irisin. Although we did find pharmacologic levels of irisin at terminal end point, it remains possible that the frequency or dosing may have altered response. However, irisin did interact with PTH by downregulating PTH receptor in osteocytes and in bone in CKD rats. Our data also support a potential role of ADAM10 in the physiological intersection of irisin, FNDC5, and PTH. Additional studies are needed to determine the mechanism underlying this interaction and if differing dosing regimens at earlier CKD stage timepoints would affect bone structure and mechanics. Future studies should investigate muscle as a potential site of interplay, particularly in understanding the role of PTH on the metabolism of the FNDC5/irisin axis.

CRediT authorship contribution statement

Hannah E. Wilson: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Neal X. Chen: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Kalisha D. O'Neill: Writing – review & editing, Visualization, Validation, Supervision, Project administration, Methodology, Investigation, Data curation, Conceptualization. Corinne E. Metzger: Writing – review & editing, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Data curation, Conceptualization. Shruthi Srinivasan: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Investigation, Data curation. Daniel J. Stephen: Writing – review & editing, Investigation, Data curation. Karis Blacklock: Writing – review & editing, Investigation, Data curation. Emily K. Matter: Writing – review & editing, Investigation, Data curation. Lynda F. Bonewald: Writing – review & editing, Visualization, Validation, Resources, Methodology, Conceptualization. Matthew R. Allen: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization. Sharon M. Moe: Writing – review & editing, Writing – original draft, Visualization, Validation, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.

Ethics declaration

This study was conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines.

This study was approved by the Institutional Animal Care and Use Committee (IUCAC).

Funding

This work was supported by funding from the National Institutes of Health [R01DK138021 (S.M.M., M.R.A.), F30DK142313 (H.E.W)]. M.R.A is supported in part by a VA RCS award [BX006479].

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Hannah E Wilson reports financial support was provided by National Institute of Diabetes and Digestive and Kidney Diseases. Sharon M Moe reports financial support was provided by National Institute of Diabetes and Digestive and Kidney Diseases. Matthew R Allen reports financial support was provided by National Institute of Diabetes and Digestive and Kidney Diseases. Sharon M Moe reports financial support was provided by National Heart Lung and Blood Institute. Sharon M Moe reports financial support was provided by Indiana Clinical and Translational Sciences Institute. Matthew R Allen reports financial support was provided by MBX Biosciences, Inc. Matthew R Allen reports financial support was provided by Bluewater Biosciences. Matthew R Allen reports financial support was provided by US Department of Veterans Affairs. Sharon M Moe reports a relationship with Eli Lilly and Company that includes: equity or stocks. Sharon M Moe reports a relationship with Amgen Inc. that includes: consulting or advisory. Sharon M Moe reports a relationship with R1 Therapeutics that includes: consulting or advisory. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary Fig. 1.

Irisin did not affect expression of genes encoding Rankl/Opg, Fgf23 or Runx2 in vitro in osteocytes. IDG-SW3 osteocytes were cultured and treated with 100 ng/mL recombinant irisin, 100 nM (412 ng/mL) PTH, or both for 14 (early) or 35 days (mature). The effect of irisin (green bars) was compared with and without PTH treatment by two way ANOVA to examine the effects and interaction of irisin and PTH with p < 0.05 considered significant. PTH consistently upregulated the Tnfsf11/Tnfrsf11b ratio (Rankl/Opg), downregulated Fgf23 and had a main effect in downregulating Runx2 in early (A, C, E) and mature (B, D, F) cells, while irisin had no effect on either at any timepoint. Data are shown as mean ± SD (n = 9–12). Abbreviations: receptor activator of nuclear factor kappa-Β ligand (Rankl), osteoprotegerin (Opg), fibroblast growth factor 23 (Fgf23), runt-related transcription factor 2 (Runx2).

Supplementary Fig. 2.

Irisin did not affect p-p38 or p-p44/42/Erk1/2 in IDG-SW3 cells. IDG-SW3 osteocytes were cultured for 14 or 35 days with osteogenic media and treated with 100 ng/mL recombinant irisin for 30 min prior to harvest. The activation of MAP kinase (p38 and p44/p42/ERK1/2) was assessed using Western blotting and normalized by total protein. The effect of irisin was compared by unpaired t-test. In both early and mature cells, there was no difference between phosphorylated p38 and p44/p42/ERK1/2 (A–D). Data are shown as mean ± SD (n = 5–9).

Supplementary Fig. 3.

Irisin treatment did not alter physical activity. Cy/+_IU male rats were treated with irisin (30 μg/kg irisin thrice weekly for 5 weeks) (from wk 28 to 32, approximately CKD stage 4–5) with and without calcium in the drinking water to suppress PTH in a 2 × 2 design. One-way ANOVA examined the effects of irisin (green bars) with p < 0.05 considered significant. There was no difference between groups in voluntary physical activity prior to sacrifice, indicating that the effects of recombinant irisin injections were independent of activity. Data are shown as mean ± SD (n = 7–13).

Supplementary Fig. 4.

No effect of irisin on Pth1r expression in muscle in rats with CKD-MBD. Cy/+_IU male rats were treated with irisin (30 μg/kg irisin thrice weekly for 5 weeks) (from wk 28 to 32, approximately CKD stage 4–5) with and without calcium in the drinking water to suppress PTH in a 2 × 2 design. Normal littermates were sacrificed at the same age and serve as normal controls shown in the dotted lines. Two-way ANOVA examined the effects and interaction of irisin (green bars) and PTH with p < 0.05 considered significant. Irisin a main effect in downregulating Pth1r expression in the soleus (B) but not the EDL (A). Data are shown as mean ± SD (n = 12–14). Abbreviations: PTH receptor 1 (Pth1r), extensor digitorum longus (EDL).

Supplementary tables

Data availability

Data will be made available on request.