Dual Osteoprotective Actions of the Kefir Peptide KFP-1: Enhancement of Bone Formation and Suppression of Bone Resorption in Cells and Murine Models

Abstract

KFP-1, a kefir-fermented peptide, promotes osteoblast differentiation and bone formation while inhibiting osteoclast differentiation and bone resorption. We also confirmed the bioaccessibility of KFP-1 in bone tissue and its overall benefits for bone health. Kefir is a dairy beverage rich in bioactive peptides with diverse health benefits. We identified a kefir-fermented peptide, KFP-1 (TEVPAINTIASAEPTVH), which enhances intestinal calcium absorption and may modulate bone remodeling. This study investigated the effects of KFP-1 on osteoblast and osteoclast gene expression and differentiation in BMMSCs, MC3T3-E1, BMMs, and Raw264.7 cells. Using iodine-125 labeling, we tracked KFP-1 distribution in mice following oral and intravenous administration, and evaluated its osteoprotective effects in AKR1A1 knockout (AKR1A1-KO) osteoporotic mice. KFP-1 upregulated osteogenic markers (ALP, Col1a1, OCN, OPG, RUNX2, OSX, BMP-2, β-catenin) and promoted osteoblast differentiation and mineralization, while downregulating osteoclastic markers (CTK, CTR, DC-STAMP, TRAP, c-Fos, c-Src, NFATc1) and inhibiting osteoclast differentiation and resorption via the inhibition of NFATc1, c-Fos, and c-Jun nuclear translocation and attenuation of RANKL-induced p38 MAPK, JNK, ERK, and NF-κB signaling. Biodistribution analysis showed KFP-1 reached femur and tibia at approximately 0.4% (oral) and 1% (intravenous) of the administered dose. In AKR1A1-KO mice, oral KFP-1 prevented bone loss, with additional calcium supplementation further improving cortical bone mechanical properties. These findings highlight KFP-1’s dual action in promoting bone formation and inhibiting bone resorption, supporting its potential as a therapeutic adjuvant or functional food ingredient for osteoporosis prevention and bone health.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00223-026-01487-w.

Introduction

Bone remodeling occurs through a regulated mechanism that includes osteoclasts and osteoblasts, ensuring the preservation of bone homeostasis. Malfunctioning of this system could trigger conditions like osteoporosis, distinguished by severe bone deterioration and a heightened risk of fractures. Several variables, including age (> 50 years), gender (female), nutritional deficiencies, lifestyle habits (smoking, inactivity, alcohol use), and specific medications (e.g., cancer treatments, antidiabetics, and diuretics), may influence osteoporosis prevalence. Current literature indicates that the worldwide prevalence of osteoporosis stands at 18.3%, with a prevalence rate of 23.1% among women; moreover, estimations suggest that over 200 million individuals globally are afflicted by this disorder [1]. Bisphosphonates (alendronate, ibandronate, etc.), denosumab, raloxifene, teriparatide, and strontium ranelate are frequently utilized in osteoporosis management; however, these pharmaceuticals often produce adverse side effects such as osteonecrosis of the jaw from prolonged bisphosphonate use, thrombosis and hot flushes from raloxifene, and detrimental cardiovascular impacts from strontium ranelate [2, 3]. Teriparatide (hPTH) is a potent agent for enhancing bone formation, yet its cost presents affordability challenges. Despite existing treatment options, osteoporosis continues to exhibit significant morbidity and mortality rates from comorbidities annually. In the last two decades, natural substances with potential benefits for osteoporosis management have attracted significant interest due to their cost-effectiveness and suitability for prolonged application compared to synthetic pharmaceuticals.

Several bioactive compounds extracted from natural plant sources, such as flavonoids, terpenoids, and polyphenols, have demonstrated inhibitory effects on osteoclastogenesis and potential therapeutic value in conditions characterized by excessive bone resorption [4]. In addition, various bioactive peptides obtained from milk proteins, recognized for their antioxidative, antihypertensive, anti-inflammatory and immunomodulatory activities, have attracted attention because of their prospective role in promoting anabolic responses in osteoblast differentiation and bone formation processes. An earlier study showed that dietary calcium (Ca)-bound casein phosphopeptides (CPPs) increase bone mineral density (BMD) in aged ovariectomized (OVX) rats [5]. Over the past decade, an assortment of milk protein-derived peptides with characterized sequences have been demonstrated to possess osteoprotective effects. Narva et al. reported that the angiotensin-converting enzyme (ACE)-inhibitory tripeptides IPP and VPP, derived from milk fermentation with Lactobacillus helveticus, significantly enhanced bone formation in osteoblastic cultures without affecting osteoclasts [6]. Given that an excess of reactive oxygen species (ROS) has been shown to increase bone resorption by stimulating osteoclastogenesis [7] or to reduce bone formation by inducing osteoblast apoptosis [8], it is posited that antioxidative peptides derived from milk proteins may yield promising effects in treating osteoporosis. Reddi et al. [9–11] elucidated that casein-derived antioxidative peptides, specifically EDVPSER, NAVPITPTL, VLPVPQK, and HPHPHLSF, exhibit stimulatory effects on the proliferation and differentiation of osteoblasts. Pandey et al. demonstrated the in vitro osteoanabolic activities of whey-derived antioxidative (MHIRL and YVEEL) and ACE-inhibitory (YLLF, ALPMHIR, IPA and WLAHK) peptides in osteoblasts [12]; thereafter, they reported that the antioxidative peptide YVEEL and the ACE-inhibitory peptide YLLF mitigated inflammation and improved bone formation markers in OVX rats [13]. Recently, Shi et al. indicated that LFP-C (FKSETKNLL), a lactoferrin-derived peptide, enhances osteoblast proliferation and differentiation, suggesting its potential osteogenic role [14]. Moreover, Upadhyay et al. enhanced the osteogenic effect of the antioxidative peptide VLPVPQK through structure–activity-based sequence modification [15], potentially facilitating the development of functional peptides with superior osteogenic activity over the original peptide.

Kefir, an acidic-alcoholic dairy beverage, is increasingly recognized for its myriad health benefits in antihypertensive, anti-inflammatory, antioxidative, cholesterol-lowering, and immunomodulatory effects [16]. Beyond previous established benefits in ameliorating fatty liver-related diseases [17, 18], atherosclerosis [19], obesity [20], air pollutant-induced pulmonary inflammation [21], and depressive-like behaviors [22], we identified considerable promise of kefir peptides (KPs) in the management of osteoporosis. The antiosteoporotic effects of KPs have been demonstrated in our previous studies in OVX [23, 24], coagulation factor VIII knockout (F8-KO) [25], and AKR1A1-KO murine models [26], as well as in osteoporotic patients [27]. Recently, we further demonstrated that KPs mitigate atherosclerotic vascular calcification while concurrently preventing osteoporosis in high-cholesterol diet-fed ApoE KO mice [28]. Earlier, we isolated a κ-casein-derived KFP-1 peptide (TEVPAINTIASAEPTVH) from KPs and demonstrated its properties in calcium (Ca)-binding and promoting Ca absorption through transient receptor potential vanilloid subfamily member 6 (TRPV6) channels [29]. Ca is a crucial nutrient influencing bone health and osteoporosis risk; thus, peptides enhancing Ca absorption, bone formation, or inhibiting resorption are expected to benefit bone health. To clarify the role of KFP-1 in bone remodeling, we explored its influence on osteoblast and osteoclast differentiation in vitro, assessed its distribution in mice post-ingestion, and evaluated its osteoprotective efficacy in AKR1A1-KO mice.

Materials and Methods

KFP-1 Synthesis

KFP-1 (TEVPAINTIASAEPTVH) was identified from a mixture of kefir peptides (KEFPEP®, Phermpep Co., Ltd., Taichung, Taiwan) as previously described [29]. For this study, KFP-1 was custom-synthesized at > 98% purity by Yao-Hong Biotechnology Inc. (Taipei, Taiwan).

Isolation of Bone Marrow Mesenchymal Stem Cells (BMMSCs) and Bone Marrow Macrophages (BMMs)

BMMSCs and BMMs were freshly isolated from the femurs and tibias of 5–9-week-old female ICR mice, as previously described with modifications [26]. Briefly, both ends of the bones were removed, and the bones were placed in 0.6-mL microtubes pierced with a 23-gauge needle at the bottom. These were nested in 1.5-mL collection tubes and centrifuged at 15,000 × g for 30 s to collect bone marrow. The pellets were resuspended in RBC lysis buffer, centrifuged at 300 × g for 5 min, and then resuspended in α-MEM growth medium (89% α-MEM, 10% FBS, 1% penicillin/streptomycin). Cell suspensions were filtered through 70-mesh strainers and cultured in 10-cm dishes at 37 °C. After 24 h, nonadherent cells were transferred to a new dish and cultured for 72 h with 25 ng/mL M-CSF; the attached cells were designated as BMMs for osteoclast differentiation. Meanwhile, adherent cells in the original dish were maintained until 80–90% confluency and then designated as BMMSCs for subsequent osteoblast differentiation.

Cell Lines

The MC3T3-E1 cell line (EP-CL-0378; Elabscience Biotechnology Inc., Houston, TX, USA) was cultured in complete α-MEM growth medium. Subculturing was performed by rinsing with D-PBS, treating with trypsin–EDTA, and replating the cell suspension into new vessels. The Raw264.7 cell line (Bioresource Collection and Research Center, Hsinchu, Taiwan) was maintained in complete DMEM (89% DMEM, 10% FBS, 1% penicillin/streptomycin). At high density (every 2–3 days), Raw264.7 cells were detached with a cell scraper and passaged. All cells were incubated at 37 °C in a humidified atmosphere with 5% CO₂.

Osteoblast Differentiation and Mineralization

Osteoblast differentiation was induced in BMMSC and MC3T3-E1 cultures using 10 mM β-glycerophosphate, 50 μg/mL vitamin C, 100 nM dexamethasone, and KFP-1 at 10, 20, or 40 μg/mL. The KFP-1 concentrations applied in this study aligned with the KP concentrations from our prior investigation [26]. Differentiation proceeded for 21 days, with medium changed every two days. Osteoblast differentiation was assessed by alkaline phosphatase (ALP) staining (Sigma-Aldrich #86C, St. Louis, MO, USA) on Day 14, as well as an ALP activity assay (BioVision #K412, Waltham, MA, USA) on both Day 7 and Day 14. Alizarin red staining was performed on Day 21 to evaluate osteoblast mineralization, as previously described [26].

Osteoclast Differentiation

BMMs (6 × 10³ cells/well) and Raw264.7 cells (2 × 10³ cells/well) were seeded in 96-well plates and induced to differentiate with M-CSF (25 ng/mL) and RANKL (50 ng/mL) in the presence of KFP-1 (0, 2.5, 5, 10, or 20 μg/mL). Differentiation was carried out for 5 days, with media refreshed on Day 3. On Day 5, cells were stained for tartrate-resistant acid phosphatase (TRAP) using the Leukocyte Acid Phosphatase kit (Sigma-Aldrich #387A). TRAP-positive multinucleated cell (MNC) numbers and areas were quantified using ImageJ software.

Resorption Pit Assay

BMMs or Raw264.7 cells were differentiated into osteoclasts on 96-well Corning™ Osteo Assay plates with various concentrations of KFP-1. Media were refreshed on Day 3, and on Day 6, cells were removed with 5% sodium hypochlorite. Resorption pits were imaged, and the pit area relative to the total well area was quantified using ImageJ software.

CCK-8 Assay

Cell proliferation and viability during osteoblast and osteoclast differentiation were assessed using the CCK-8 assay. Cells were seeded in 96-well plates, treated with 10 μL/well CCK-8 reagent (MedChemExpress, Monmouth Junction, NJ, USA), and incubated at 37 °C for 2 h. Absorbance was measured at 450 nm using a microplate reader.

Quantitative Reverse Transcription-Polymerase Chain Reaction PCR (qRT-PCR)

Total RNA was extracted using a Total RNA Mini Kit (Geneaid, RTD300, Taipei, Taiwan) and reverse transcribed into cDNA with an RT Premix Kit (iNtRON, 25,087, Korea). qPCR was performed in 20 μL reactions containing cDNA (10–100 ng), 0.2 μM gene-specific primers, and 2 × SYBR Green PCR mix in a 384-well plate using the QuantStudio 6 Pro Real-Time PCR System (Applied Biosystems, Waltham, MA, USA). Cycling conditions were 95 °C for 5 min, followed by 40 cycles of 95 °C for 10 s and 60 °C for 40 s, with a melting curve from 65 to 95 °C. Relative mRNA expression was calculated by the 2^–ΔΔCt method with GAPDH as the reference gene. Primer sequences are provided in Supplementary Table S1.

Western Blot Analysis

Western blotting was performed as previously described [26]. For sample preparation, cell pellets were lysed in RIPA buffer, centrifuged, and total protein concentrations were quantified using a BCA protein assay kit (Thermo Fisher Scientific Inc., Waltham, MA, USA). Additionally, cytosolic and nuclear protein separation was performed utilizing a nuclear extraction kit (Signosis, Santa Clara, CA, USA) as per the manufacturer’s instructions. Protein samples (10 μg/lane) were subjected to separation via 12% SDS-PAGE and subsequently transferred to PVDF membranes. The membranes underwent a blocking procedure, followed by incubation with primary and secondary antibodies (with TBST washes between steps), and detected by HRP chemiluminescence using a luminescence imaging system. Relative protein levels were normalized to β-actin and quantified with ImageJ software. The specificity of the cytosolic and nuclear protein fractions was validated by lacking significant histone H3 and β-actin cross-contamination within the respective cytosolic and nuclear extracts. Antibody sources and dilutions are listed in Supplementary Table S2.

Radiolabeling of KFP-1 with Iodine-125 (¹²⁵I)

KFP-1 was radiolabeled with ¹²⁵I at its C-terminal histidine using Pierce precoated iodination tubes (Thermo #28,601, Rockford, IL, USA) following the manufacturer’s instructions with minor modifications. Briefly, the tube was wetted with PBS (100 μL remaining), then 10 μL (1560 μCi) Na¹²⁵I (NEZ033A005MC, PerkinElmer, Turku, Finland) was added and incubated at room temperature for 6 min. Next, 0.1 mg KFP-1 was added and reacted for 15 min at room temperature. The reaction was stopped with 50 μL scavenging buffer (10 mg/mL tyrosine in PBS), and the sample was purified using a Z-25 desalting column. A total of 675 μCi of ¹²⁵I-labeled KFP-1 (¹²⁵I-KFP-1) was recovered, with a labeling efficiency of 43.2%.

Animal Experiments

All animal procedures were approved by the Institutional Animal Care and Use Committee of National Chung Hsing University (IACUC No. 106–131). Female ICR mice for BMMSC, BMM preparation, and biodistribution assays were obtained from BioLASCO Taiwan Co., Ltd. (Taipei, Taiwan). The AKR1A1^eGFP/eGFP knockout (AKR1A1-KO) mouse line was generated in-house via pronuclear microinjection [30], and genotypes were confirmed by PCR. Mice were housed under controlled conditions (22 ± 4 °C, 50 ± 10% humidity, 12-h light/dark cycle) with free access to food and water.

Biodistribution Assay

Eight-week-old female ICR mice received ¹²⁵I-KFP-1 either orally (20 mg/kg, n = 9) or intravenously (2 mg/kg, n = 9). Isotopic counts in the brain, heart, lung, liver, spleen, kidney, intestine, stomach, pancreas, blood (1 mL), urine, bone (femur and tibia), and carcass were measured using an automatic gamma counter (2480 WIZARD2, PerkinElmer) at 30 min, 2 h, and 4 h post-administration (n = 3 per time point). The percentage of counts in each organ relative to the initial injection dose (% ID) was calculated.

Evaluation of Bone-Protective Effects of KFP-1 in AKR1A1-KO Mice

Twenty-four 8-week-old male AKR1A1-KO mice were randomly assigned to four groups (n = 6 each): Vit C (40 μg vitamin C, positive control), mock (distilled water, negative control), KFP-1 (1.5 mg/kg), and KFP-1 + Ca (1.5 mg/kg KFP-1 plus 8 mg/kg CaCO₃). Treatments were administered orally in 100 μL solution daily for 12 weeks. After treatment, serum was collected and femurs were harvested for analysis.

Microcomputed Tomography (μ-CT)

Left femurs were scanned using a SkyScan 1076 scanner (Bruker, Kontich, Belgium) at 9-μm resolution, as previously described [23–26, 30]. Bone mineral density (BMD) and trabecular morphometric parameters—including bone volume/total volume ratio (Tb.BV/TV), number (Tb.N), thickness (Tb.Th), and separation (Tb.Sp)—were calculated from the volume of interest using Bruker CT-Analyzer (CTAn) software. A 3D image of the trabecular bone was reconstructed by merging 105 layers of 2D images, corresponding to a 0.9-mm-thick region of the distal femoral metaphysis.

Nanoindentation

Nanoindentation was performed on the diaphyseal cortical bone of the right femur to assess mechanical strength following oral KFP-1 administration. Sample preparation and nanoindenter settings (Hysitron, Inc., Eden Prairie, MN, USA) were as previously described [31, 32]. Elastic modulus and hardness were calculated from load-depth curves using the Oliver-Pharr method.

Statistical Analysis

Data are presented as mean ± SD (histograms) or as box plots (median, quartiles, minimum, and maximum) using GraphPad Prism 8. Statistical significance was assessed by unpaired t-test or one-way ANOVA with Tukey’s or Duncan’s post hoc test. Differences were considered significant at p < 0.05.

Results

KFP-1 Promotes Osteoblast Differentiation and Mineralization

KFP-1, a 17-residue Ca-binding peptide isolated from kefir peptides, facilitates intestinal calcium absorption via TRPV6 channels. Given the potential of dietary Ca-bound peptides to support bone health, we investigated whether KFP-1 enhances osteoblast differentiation and mineralization in vitro. BMMSCs and MC3T3-E1 cells were differentiated in the presence of various KFP-1 concentrations, and osteoblast differentiation was assessed by ALP activity and alizarin red staining. As shown by ALP staining, KFP-1 treatments increased ALP expression in differentiated BMMSCs, with a positive correlation between ALP staining density and KFP-1 concentrations (Fig. 1A). Quantitative analysis revealed a dose-dependent enhancement of ALP expression by KFP-1, with 20 μg/mL significantly elevating osteoblastic ALP levels; however, treatment with 40 μg/mL KFP-1 accelerated the mineralization of mature osteoblasts, resulting in decreased ALP levels compared to 20 μg/mL KFP-1 (Fig. 1B). To avoid this condition, KFP-1 at 20 μg/mL was chosen for subsequent ALP activity detection, as well as mRNA and protein expression analysis via qRT-qPCR and Western blot. The results indicated that treatments with 20 μg/mL KFP-1 significantly enhanced ALP activity on both Day 7 and Day 14 of BMMSC differentiation (Fig. 1C). The effects of KFP-1 on ALP expression were similarly noted in osteoblast differentiation from MC3T3-E1 cells, showing a dose-dependent relationship (Fig. 1D). Furthermore, on Day 21 of BMMSC differentiation, alizarin red staining showed that KFP-1 significantly enhanced mineralization (Fig. 1E), with a dose-dependent increase confirmed by dye quantification (Fig. 1F). CCK-8 assays indicated that KFP-1 did not affect cell proliferation during early osteoblast differentiation, suggesting its effects are specific to differentiation (Supplementary Fig. S1A).

Fig. 1.

KFP-1 promotes osteoblastic differentiation and mineralization in BMMSCs and MC3T3-E1 cells. BMMSCs and MC3T3-E1 cells were induced to differentiate into osteoblasts with varying concentrations of KFP-1. A Representative ALP staining images of BMMSCs on Day 14 (scale bar: 1 mm). B Quantitative analysis of ALP staining by DMSO solubilization and measuring absorbance at 562 nm. C ALP activity in BMMSCs measured on Days 7 and 14 with 20 μg/mL KFP-1. D ALP activity in MC3T3-E1 cells on Day 14 with different KFP-1 concentrations. E Representative alizarin red staining images of BMMSCs on Day 21 (scale bar: 0.5 mm). F Quantification of mineralization by solubilizing bound alizarin red with 10% cetylpyridinium chloride and measuring absorbance at 405 nm. Data are mean ± SD (n ≥ 3). *p < 0.05, ***p < 0.001 vs. control (without KFP-1)

KFP-1 Upregulates Osteogenic Marker Gene Expression

The effects of KFP-1 on osteogenic marker gene expression were assessed by qRT-PCR at multiple time points during differentiation. KFP-1 significantly increased ALP and Col1a1 expression on Day 7, and enhanced OCN expression throughout differentiation (Fig. 2A–C). Osterix (OSX) expression was upregulated on Day 3, then returned to control levels (Fig. 2D). RUNX2 and BMP-2 mRNA levels were elevated on Day 7, with BMP-2 showing a non-significant increase (Fig. 2E, F). KFP-1 significantly increased OPG expression after Day 5, while RANKL mRNA was mildly increased on Day 7, resulting in a lower RANKL/OPG ratio compared to control (Fig. 2G–I). Consistently, protein levels of ALP, Col1a1, BMP-2, RUNX2, and OSX were elevated following KFP-1 treatment, along with increased β-catenin, a key regulator of Wnt signaling in osteoblast differentiation (Fig. 3).

Fig. 2.

KFP-1 upregulates osteogenic marker gene expression during BMMSC differentiation. BMMSCs were differentiated in the absence or presence of 20 μg/mL KFP-1, and mRNA was collected on Days 3, 5, 7, and 9 (D3, D5, D7, D9). Expression levels of A ALP, B Col1a1, C OCN, D OSX, E RUNX2, F BMP-2, G OPG, and H RANKL were quantified by qRT-PCR using the 2^–ΔΔCt method. Data are mean ± SD (n ≥ 4). Differences greater than twofold or less than 0.5-fold (± 1 log₂ ΔΔCt) are indicated by a hashtag (#). I The RANKL/OPG mRNA ratio is shown, with a significantly lower ratio in KFP-1-treated cells on Days 5 and 9 (*p < 0.05, ***p < 0.001)

Fig. 3.

KFP-1 promotes osteogenic marker protein expression during BMMSC differentiation. BMMSCs were differentiated with or without 20 μg/mL KFP-1, and protein lysates were collected on Days 3, 5, 7, and 9 (D3, D5, D7, D9). A Representative western blot images. Relative band intensities of B ALP, C type I collagen, D β-catenin, E BMP-2, F RUNX2, and G OSX (osterix) were quantified using ImageJ. Statistical significance: * (control vs. D3), # (KFP-1 vs. D3), and ϕ (control vs. KFP-1); one symbol p < 0.05, two symbols p < 0.01, three symbols p < 0.001

KFP-1 Activates Key Osteogenic Transcription Factors and Signaling Pathways

KFP-1 upregulates RUNX2 and OSX, two essential transcription factors for osteoblast differentiation (Fig. 3A, F, G). RUNX2 drives mesenchymal stem cell commitment to osteoblasts, while OSX promotes osteoblast maturation and inhibits chondrocyte differentiation. Notably, KFP-1 also increases β-catenin and BMP-2 expression (Fig. 3A, D, E), suggesting involvement in Wnt and TGF-β signaling.

KFP-1 Inhibits Osteoclast Differentiation and Resorptive Activity

KFP-1 suppressed osteoclast differentiation in both BMMs and Raw264.7 cells in a dose-dependent manner, as shown by reduced numbers and areas of TRAP-positive multinucleated cells (Fig. 4A–C). The estimated IC₅₀ for KFP-1 was ~ 20 μg/mL in BMMs and 40 μg/mL in Raw264.7 cells (Supplementary Fig. S2A–C), with 20 μg/mL KFP-1 showing inhibitory effects comparable to 10 nM alendronate. CCK-8 assays indicated that KFP-1 did not affect the proliferation of precursor cells, suggesting its inhibitory effects were not due to cytotoxicity (Supplementary Fig. S1B, C).

Fig. 4.

KFP-1 inhibits osteoclastic differentiation and bone resorption in BMMs. BMMs were induced to differentiate into osteoclasts with varying concentrations of KFP-1 or 10 nM alendronate (ALN) for 5 days, followed by TRAP staining. A Representative TRAP staining images (scale bar: 1 mm). Osteoclast differentiation was assessed by B counting TRAP-positive multinucleated cells (MNCs, nuclei ≥ 3) and C measuring TRAP-positive areas using ImageJ. D Representative images from the bone resorption pit assay, performed on 96-well Corning osteo plates after 6 days of differentiation. E Bone resorption quantified as the percentage of total pit area relative to the well area. Data are mean ± SD (n ≥ 3). **p < 0.01, ***p < 0.001 vs. control (without KFP-1); ns, not significant

The inhibitory effects of KFP-1 on differentiation further affected the functional performance of mature osteoclasts on subsequent resorption pit assays in a dose-dependent manner, with 20 μg/mL KFP-1 decreasing total resorption area by nearly 90% (Fig. 4D, E). Similar inhibition was observed in Raw264.7 cells, where 20 μg/mL KFP-1 reduced resorption by about 87%, comparable to 10 nM alendronate (Supplementary Fig. S2D, E). These results suggest the functionality of KFP-1 in the suppression of osteoclast differentiation, ultimately leading to a decrease in bone resorption.

KFP-1 Downregulates Osteoclastogenic Genes and RANKL-Induced Signaling

KFP-1 significantly decreased the mRNA expression of key osteoclast markers—including cathepsin K, calcitonin receptor, DC-STAMP, TRAP, c-Fos, and NFATc1—particularly at 5 and 10 μg/mL (Supplementary Fig. S3A–F). It also inhibited PLCγ2 mRNA and, to a lesser extent, calcineurin activity (Supplementary Fig. S3G, H). Western blot analysis confirmed reduced protein levels of NFATc1, c-Fos, c-Jun, cathepsin K, and c-Src, with no effect on RANK (Fig. 5A, B). Additionally, KFP-1 reduced nuclear translocation of NFATc1 and c-Jun, but not c-Fos (Fig. 5C, D), indicating suppression of RANKL-induced osteoclastogenic signaling pathways.

Fig. 5.

KFP-1 suppresses osteoclast marker gene expression and RANKL-induced signaling pathways. A, B BMMs were differentiated into osteoclasts with varying concentrations of KFP-1, and cytosolic protein lysates were collected on Day 3 for western blot analysis. A Representative blots of osteoclastic marker proteins; B quantification of cytosolic marker protein bands normalized to β-actin. C Representative blots showing nuclear translocation of NFATc1, c-Fos, and c-Jun; D quantification of nuclear protein bands normalized to histone H3. E For signaling analysis, BMMs were pretreated with 5 μg/mL KFP-1 for 6 h, then stimulated with M-CSF and RANKL; lysates were collected at 15-min intervals post-RANKL addition and analyzed by Western blotting. F Phosphorylated protein levels were normalized to total protein. Data are mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001 vs. control (without KFP-1)

To assess KFP-1’s effects on RANKL-induced intracellular signaling, BMMs were pretreated with KFP-1 or control, then stimulated with RANKL. KFP-1 markedly inhibited RANKL-induced phosphorylation of p38 MAPK, JNK, ERK, and NF-κB at 60 min, while having minimal effect on AKT phosphorylation (Fig. 5E, F).

Biodistribution of KFP-1

To determine whether KFP-1 crosses the gastrointestinal barrier and reaches bone tissue, we radiolabeled KFP-1 with iodine-125 (¹²⁵I) and conducted biodistribution assays in mice following oral (20 mg/kg) or intravenous (2 mg/kg) administration. After 30 min, ¹²⁵I-KFP-1 localized mainly to the stomach (32.3%), kidney (8.3%), intestine (6.9%), liver (2.3%), urine (7.0%), and carcass (25.3%) following oral dosing (Fig. 6A). Intravenous injection resulted in higher kidney (17.9%), intestine (4.8%), liver (2.4%), and carcass (31.2%) accumulation (Fig. 6B). Trace amounts were detected in femur and tibia after oral administration (0.19% and 0.21%, respectively) (Fig. 6C), with significantly higher levels observed after intravenous injection (0.65% and 0.40%, respectively) (Fig. 6D). Radioactivity declined at 2 and 4 h post-administration, particularly in the femurs following intravenous injection.

Fig. 6.

Biodistribution of KFP-1 in mice following oral and intravenous administration. KFP-1 was radiolabeled and administered orally (20 mg/kg) or intravenously (2 mg/kg), and isotopic counts were measured at 30 min, 2 h, and 4 h post-administration (n = 3 per group). A Organ distribution of KFP-1 after oral administration. B Organ distribution after intravenous injection. C KFP-1 accessibility in femur and tibia following oral administration. D KFP-1 accessibility in femur and tibia following intravenous injection. Data are mean ± SD (n = 3). **p < 0.01 vs. 30 min

Oral KFP-1 Administration Restores Bone Formation Markers and Reduces Inflammation in Vitamin C-Deficient AKR1A1-KO Mice

In AKR1A1-KO mice with vitamin C deficiency, oral KFP-1 (alone or with calcium) for 12 weeks prevented the decline in serum bone formation markers (P1NP, OCN, OPG) (Fig. 7A–C) and the rise in bone resorption markers (CTX-1, RANKL) seen in untreated controls, normalizing the RANKL/OPG ratio (Fig. 7D–F). KFP-1 treatments also reduced oxidative stress (lower DCF, higher SOD and CAT) (Fig. 7G–I) and decreased proinflammatory cytokines (IL-1β, TNF-α, IL-6) (Fig. 7J–L) compared to the mock group, with no apparent additive effect from calcium supplementation.

Fig. 7.

Oral KFP-1 administration induces serum changes favoring bone formation. After 12 weeks of oral administration, serum levels of A P1NP, B OCN, C OPG, D CTX-1, E RANKL, F RANKL/OPG ratio, G DCF, H SOD, I CAT, J IL-1β, K TNF-α, and L IL-6 were measured. Data are presented as box plots and analyzed by one-way ANOVA with Duncan’s post hoc test (n = 6); different letters indicate significant differences (p < 0.05)

KFP-1 Preserves Trabecular and Cortical Bone Microarchitecture in AKR1A1-KO Mice

μ-CT analysis revealed that the mock group exhibited significant trabecular bone loss in the distal femoral metaphysis compared to the Vit C group, while both KFP-1 and KFP-1 + Ca groups maintained trabecular structures similar to the Vit C group (Fig. 8A). KFP-1 treatment increased trabecular bone mineral density (Tb.BMD), bone volume fraction (Tb.BV/TV), and trabecular number (Tb.N), and reduced trabecular separation (Tb.Sp), with no effect on trabecular thickness (Tb.Th) (Fig. 8B–F). The mock group also showed decreased cortical bone mineral density (Cb.BMD) and cortical bone volume fraction (Cb.BV/TV) in the diaphysis, which were improved by KFP-1, although Cb.BMD did not reach statistical significance (Supplementary Fig. S4A–C). No significant changes were observed in the cortical bone of the distal metaphysis across groups (Supplementary Fig. S5). Calcium supplementation did not provide additional benefits over KFP-1 alone.

Fig. 8.

Micro-CT analysis of metaphyseal trabecular bone in femurs. A Reconstructed 3D images of trabecular bone. Box plots showing B trabecular bone mineral density (Tb.BMD), C bone volume fraction (Tb.BV/TV), D trabecular number (Tb.N), E trabecular separation (Tb.Sp), and F trabecular thickness (Tb.Th). Data were analyzed by one-way ANOVA with Duncan’s post hoc test (n = 6); different letters indicate significant differences (p < 0.05)

KFP-1 Improves Cortical Bone Mechanical Properties

Nanoindentation analysis showed that KFP-1 treatment enhanced cortical bone hardness and elastic modulus compared to the mock group (Supplementary Fig. S4D, E). Notably, combining KFP-1 with calcium supplementation produced an additive improvement in these mechanical properties.

Discussion

A novel KFP-1 peptide was isolated from kefir peptides in our current study. This 17-residue peptide has been characterized as a Ca-binding peptide (with a binding capacity of 29 μg Ca/μmole KFP-1) with an in vivo function to promote intestinal Ca absorption through TRPV6 channels [29]. Due to the potential of dietary Ca-bound CPPs to prevent bone loss, we hypothesized that KFP-1 could exert positive effects on bone health. To confirm this hypothesis, we first characterized whether KFP-1 affected in vitro osteoblastic differentiation and mineralization. As predicted, KFP-1 enhanced ALP activities during osteoblastic differentiation and accelerated extracellular matrix mineralization at the late stage of differentiation in a dose-dependent manner (Fig. 1). Several osteoblast marker genes were characterized for their expression at the mRNA and protein levels under the regulation of KFP-1. Regarding osteoblast differentiation, KFP-1 was found to concomitantly upregulate the expression of RUNX2 and OSX (Fig. 2), two central transcription factors in the osteogenic process. The main function of RUNX2 is to promote the differentiation of multipotent mesenchymal stem cells into immature osteoblasts, thereby serving as the most upstream transcription factor in the regulation of osteoblast differentiation, while OSX plays a role in the final commitment of osteoblasts by promoting the maturation of preosteoblasts and blocking the differentiation of mesenchymal stem cells into chondrocytes [30]. Several signaling systems upregulating RUNX2 and OSX expression (Fig. 3) have been demonstrated to be crucial for osteoblast differentiation, including Wnt, transforming growth factor-β (TGF-β), Hedgehog, and fibroblast growth factor (FGF) signaling [31]; conversely, the signaling mechanisms that downregulate RUNX2 and OSX expression have been shown to inhibit osteoblast differentiation, such as epidermal growth factor receptor (EGFR) and Notch signaling [32, 33]. Due to the finding that KFP-1 upregulates the expression of β-catenin and BMP-2, it is suggested that KFP-1 is at least involved in Wnt and TGF-β-related signal transduction. β-catenin is a key activator that modulates gene transcription in the context of canonical Wnt signaling. Cytoplasmic β-catenin normally remains at a lower level through interactions with the β-catenin destruction complex when cells are not exposed to Wnt ligands; however, the cytoplasmic β-catenin level will be increased once Wnt is bound to a Frizzled receptor and interacts with the LRP5/6 coreceptor [34]. BMP-2 is a powerful cytokine belonging to the TGF-β family that exhibits potent osteogenic effects by activating the Smad-dependent and Smad-independent signaling pathways [35]. The activation of these signaling pathways and transcription factors during osteoblast differentiation may enhance the transcription of downstream osteogenic effector genes such as ALP, Col1al, OCN, and OPG, which were observed in the present study.

In addition, KFP-1 exerted inhibitory effects on in vitro osteoclast differentiation and bone resorption (Fig. 4). In vitro osteoclast differentiation using BMMs requires the costimulation of M-CSF and RANKL, in which RANKL serves as a key cytokine to trigger the differentiation process. Mechanistically, the binding of RANKL to the membrane receptor RANK recruits TRAF6 to the cytoplasmic domain of RANK and subsequently activates downstream p38 MAPK, JNK, ERK, NF-κB, and NFATc1 signaling pathways [36]. Notably, KFP-1 attenuated RANKL-induced activation of p38 MAPK, JNK, ERK and NF-κB signaling as an early event but had little or no effect on AKT signaling (Fig. 5). The AKT signaling pathway (or PI3K/AKT signaling pathway) is known to promote cell survival and proliferation in response to extracellular stimuli, thus explaining why KFP-1 had no effect on the proliferation of BMMs, Raw264.7 cells, and even BMMSCs and MC3T3-E1 cells. The inhibition of MAPKs and NF-κB signaling by KFP-1 led to the downregulation of two master transcription factors, NFATc1 and AP-1 (formed by c-Jun and c-Fos), and their nuclear translocation. NFATc1 is likely the most crucial regulator of osteoclast differentiation. The expression of NFATc1 can be upregulated by the binding of c-Fos and NFATc1 itself (autoamplification) to the NFATc1 promoter and downregulated by several epigenetic modifications (ubiquitination, deacetylation, methylation, and miRNAs) to maintain normal bone homeostasis [37]. Several osteoclastic marker genes, such as c-Src [38], CTK [39], CTR [40], DC-STAMP [41], and TRAP [42], contain NFATc1-binding sites in their promoter regions; thus, the reduction of NFATc1 by KFP-1 also caused downregulation of these genes. Additionally, RANKL-dependent and RANKL-independent Ca signaling play a key role in osteoclast differentiation. In this regard, external signals from RANKL or other costimulatory factors synergistically activate PLCγ2 to increase the intracellular Ca²⁺ concentration, which in turn activates calcineurin to dephosphorylate NFATc1, and then, the dephosphorylated NFATc1 enters the nucleus to regulate downstream gene expression [43]. We found that KFP-1 downregulated PLCγ2 mRNA expression and reduced calcineurin activity, suggesting that KFP-1 also had an impact on intracellular calcium signaling. The calcium channels of the TRPV family have been reported to play important roles in the modulation of intracellular calcium signaling in both osteoclast and osteoblast differentiation. For example, TRPV4 acts as a mechanosensor in osteoblasts [44], whereas increasing intracellular Ca²⁺ concentration through TRPV4 was reported to promote osteoclast differentiation [45], and TRPV5 was shown to be a negative mediator of bone resorption [46]. Interestingly, our recent data showed that KFP-1 modulated TRPV4 and TRPV5 levels in differentiated BMMs and BMMSCs (unpublished data). Ongoing examinations concerning this topic are presently being conducted.

It is important to know whether KFP-1 crossed the gastrointestinal barrier and was transported to the bones after intake. To answer this question, we radiolabeled KFP-1 with the isotope ¹²⁵I and performed a biodistribution assay in mice (Fig. 6). Through the oral route, KFP-1 was largely taken up in the gastrointestinal tract and rapidly transported to the kidney, liver, and other tissues in 30 min, whereas KFP-1 was mostly delivered to the kidney in 30 min via the intravenous route. The amount of KFP-1 generally declined with time. Notably, nearly 0.4% and 1% of ¹²⁵I-labeled KFP-1 was delivered to the long bones (femurs and tibias) at 30 min after oral and intravenous administration, respectively. These data suggest that the in vivo bioaccessibility of KFP-1 is less than 0.5% in bone tissues through the oral route. Previously, using the Caco-2 Transwell model and LC/MS/MS analysis, Reddi et al. found that almost 1% of NAVPITPTL was transported in an intact form [11], suggesting that the in vivo bioaccessibility of this peptide may be less than 1%. There are some limitations to our data: (1) ¹²⁵I labeling may affect the original structure and function of KFP-1; (2) ¹²⁵I labeling may affect the absorption of KFP-1 by intestinal enterocytes; and (3) ¹²⁵I-KFP-1 may be further cleaved into shorter peptides. Non-labeling methods should be helpful to avoid the possible disturbance caused by labeling with an isotopic or fluorescent probe. For instance, Sánchez-Rivera et al. [47] conducted a kinetic study of the milk casein-derived antihypertensive peptide HLPLP by coupling ultra-high-performance liquid chromatography (UPLC) with a Q-TOF analyzer, and Gu et al. [48] clarified the single- and multiple-dose pharmacokinetics of the tufstin-derived T peptide, a potent antitumor agent, by using a reliable competitive ELISA method. However, high-affinity antibodies against KFP-1 and efficient methods for extracting KFP-1 from blood samples are currently unavailable. Based on our previous data, KFP-1 is moderately resistant to cleavage by pepsin and pancreatin, and approximately 50% of KFP-1 will remain intact after simulated gastrointestinal digestion [29]. Nonetheless, further digestion in the blood circulation may occur. In the future, we are attempting to clarify whether intact KFP-1 is necessary for its function in modulating osteoblast and osteoclast differentiation.

Our animal studies demonstrate that oral KFP-1 administration prevents osteoporosis in AKR1A1-KO mice with chronic vitamin C deficiency, as evidenced by increased serum bone formation markers, decreased bone resorption markers, reduced proinflammatory cytokines, preserved trabecular bone mass, and improved cortical bone mechanical properties (Figs. 7, 8, Supplementary Figs. S4, S5). While KFP-1 alone was effective, combining it with calcium supplementation further enhanced cortical bone strength. These results highlight KFP-1’s dual action on osteoblast and osteoclast differentiation and support its potential as a food additive or therapeutic adjuvant with antiosteoporotic effects.

To date, few milk protein-derived peptides have been identified with antiosteoporotic effects (see Introduction). While previous peptides have been characterized for their osteogenic properties, none have demonstrated simultaneous inhibition of in vitro RANKL-induced osteoclastogenesis and bone resorption. This study is the first to show that the kefir-derived KFP-1 peptide exhibits comprehensive antiosteoporotic effects both in vitro and in vivo.

Conclusions

This study demonstrates that KFP-1 promotes osteoblast differentiation and bone formation while inhibiting osteoclast differentiation and bone resorption. We also confirmed the bioaccessibility of KFP-1 in bone tissue and its overall benefits for bone health, supporting its potential as a functional food ingredient or therapeutic agent for osteoporosis prevention.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Min-Che Tung: Conceptualization, methodology, data curation, resources. Gary Ro-Lin Chang: Investigation, formal analysis, data curation, writing—original draft preparation. Min-Yu Tu: Conceptualization, methodology, data curation. Hueng-Chuen Fan: Methodology, data curation, resources. Chih-Ching Yen: Conceptualization, methodology, validation. Abdulkadir Cidem: Formal analysis, data curation. I-Chien Chen: Formal analysis, data curation. Chuan-Mu Chen: Conceptualization, methodology, funding acquisition, supervision, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

Open access funding provided by National Chung Hsing University. Ministry of Science and Technology, Taiwan, MOST-108-2313-B-005-039-MY3, Chuan-Mu Chen, Ministry of Education, MOE-114-S-0023-A, Chuan-Mu Chen.

Data Availability

All data generated or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of interest

Min-Che Tung, Gary Ro-Lin Chang, Min-Yu Tu, Hueng-Chuen Fan, Chih-Ching Yen, Abdulkadir Cidem, I-Chien Chen, and Chuan-Mu Chen declare that they have no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

Ethical Approval and Consent to Participate

All animal experiments were approved by the Institutional Animal Care and Use Committee of National Chung Hsing University (IACUC No. 106–131). Experimental protocols and animal care were provided according to the guideline for the care and use of animals established by National Chung Hsing University.

Consent for Publication

Not applicable.