The gut–bone axis: microbial metabolism and nutritional interventions for bone health Authors: Alba Rodriguez-Bryant, Maria Papageorgiou, Marie-Noëlle Horcajada, Emmanuel Biver, Serge Ferrari, Nicolas Bonnet Published: 2026 ABSTRACT ABSTRACTOsteoporosis is a prevalent condition characterized by a rapid decline in bone mineral density and distorted microarchitecture, which leads to increased bone fragility. The gut microbiota and its metabolites play a crucial role in the development and progression of osteoporosis by influencing gut permeability, nutrient digestion and absorption, pH balance, and immune regulation. Nutritional interventions aimed at modulating gut microbiota, through postbiotics (butyrate), probiotics (Bifidobacterium animalis or Lacticaseibacillus rhamnosus LGG), prebiotics (fructo-oligosaccharides), synbiotics, and specific dietary patterns (e.g. Mediterranean and vegetarian diets) have emerged as promising strategies to mitigate bone loss associated with aging. This review investigates the communication between the gut and bone, summarizing the underlying mechanisms and providing an updated review on microbial metabolites. Additionally, it examines recent research on probiotics and their metabolic contributions, highlighting potential gaps in the current understanding of this field. Introduction Bone remodeling is a dynamic process in which old bone is continuously replaced by new bone. This balance is essential for maintaining bone homeostasis. With aging and under certain pathological conditions, bone remodeling becomes imbalanced favoring bone resorption over formation. Such changes mediate decreases in bone mineral density (BMD), impairments in bone microarchitecture, and eventually increased risk of fragility fractures. This condition is known as osteoporosis. Latest reports estimated that 25.5 million women and 6.5 million men in Europe are affected by this disease sustaining about 4.3 million fractures. 1 , 2 Osteoporosis-related fractures often lead to additional complications, such as prolonged disability, reduced quality of life, increased risk for subsequent fractures, loss of autonomy, and increased mortality up to 30% in the first year postfracture. 2 , 3 These complications entail a substantial economic burden to healthcare systems worldwide. 2 Therefore, it is a priority to establish strategies aiming at preventing bone loss. The gut is home to trillions of microorganisms that have co-evolved through complex mutualistic interactions with the host. While microbes benefit from a stable, nutrient-rich environment, the host gains in turn essential functions such as digestion of complex nutrients, synthesis of vitamins, or immune modulation—processes that are also critical for bone metabolism. Metabolites produced by the gut microbiota act as key mediators of these functions and therefore play a central role in gut–host communication. 4 , 5 Aging is associated with changes in microbiota composition, contributing to dysbiosis and potentially influencing disease risk. 6 Observational data indicate that osteopenic and osteoporotic patients exhibit decreased gut microbiota diversity. 7-11 Additionally, experimental studies exploring the effects of germ-free animal models have reported increased bone mass, 12 suggesting that certain bacterial species have a negative impact on bone during aging. 13 A well-balanced diet is key to maintaining microbial ecological equilibrium. The choice of dietary patterns—such as Western or Mediterranean diets—exerts a significant impact on the composition and functionality of the gut microbiota, thus influencing overall host health. 14 Certain dietary components are receiving increased interest for their potential benefits in bone remodeling via effects on the microbiota and its metabolites. Biotic interventions (prebiotics, probiotics, postbiotics, and synbiotics, see Figure 1) have gained considerable interest as adjuvant nutritional strategies. Their ability to modulate the gut microbiota has been shown to exert beneficial effects on bone metabolism in both preclinical and clinical studies. 5 , 15 , 16 Microbiome-based nutritional strategies: key definitions and examples. Created in https://BioRender.com. In this review, we provide an updated overview of the gut–bone axis communication. We highlight the main gut microbiota-related mechanisms involved in bone regulation, outline the latest evidence on the contribution of key microbial metabolites and discuss biotic interventions for the prevention and management of osteoporosis. Importantly, this review addresses a critical gap in the current literature by connecting these interventions with their potential metabolic consequences, thereby offering a mechanistic perspective on how microbial metabolism influences bone physiology. Additionally, we explore the potential role of fermented dairy products in supporting bone homeostasis through the gut–bone axis. Methods The present review aims to collect existing literature exploring the role of gut microbiota and its metabolism on adult bone homeostasis, as well as nutritional strategies to prevent bone loss. To this purpose, we conducted a targeted search using the PubMed database up to January 2026. We searched for in vitro, in vivo, observational, and randomized controlled trial studies. Relevant studies were identified by using a Boolean query combining keywords for bone health outcomes (e.g. bone, bone mineral density, osteoporosis, fracture, bone turnover, osteoblast, osteoclast), microbiota (e.g. microbiome, diversity, abundance, dysbiosis, SCFA, metabolites), and nutrition (e.g. diet, probiotics, prebiotics, synbiotics, diet, fermented food, fermented dairy, polyphenols). The search initially yielded 2033 records considering only English papers and full-text papers. We manually excluded studies on animal models other than humans or rodents, articles on no relevant bone conditions (e.g. cancer, periodontitis), and studies on children, adolescents, lactating, or pregnant women. Only studies in women and men adults (+45 y) were included. Under this method, 212 studies were identified. This literature research was complemented by a manual search and the use of the tool Connected Papers | Find and explore academic papers in order to identify relevant background literature and mechanistic studies. Gut–bone axis and its mechanisms In the following section, we summarize the most well-characterized mechanisms linking gut microbiota to bone pathophysiology. Gut integrity and permeability The intestinal barrier controls the entry of essential ions, nutrients, and water from the gut lumen into the submucosa. Under healthy conditions, intestinal cells are tightly connected to one another, enabling selective and regulated transport while preventing the passage of harmful substances. During aging, alterations in the gut epithelial barrier are commonly observed. 17 One notable change is the thinning of the mucus layer, resulting from reduced mucus secretion by goblet cells, which leaves the epithelium exposed to pathogenic agents. 18 Additionally, the expression of tight junction proteins such as zonula occludens (ZOs), claudins, and junctional adhesion molecules decreases, leading to epithelial disruption and increased gut permeability. Loss of epithelial integrity increases the translocation of endotoxins, such as lipopolysaccharides (LPS), into the systemic circulation, promoting systemic inflammation. This process is known as inflammaging, is a well-known contributing factor to bone deterioration. 19 , 20 Some biotics have demonstrated the ability to enhance epithelial barrier integrity by upregulating tight junction proteins such as claudin, ZO-1, and occludin, and by reducing LPS translocation into the systemic circulation. 21–23 Mineral absorption Calcium, phosphorus, and magnesium are minerals essential for bone remodeling. These minerals are primarily provided by the diet and subsequently absorbed by the gut. The intestine absorbs only about 10–40% of the total calcium from the diet. 24 , 25 The most efficient site for calcium absorption is the duodenum; however, due to its short length, it accounts for only 10% intake of the total ingested dietary calcium, whilst the jejunum contributes an additional 10%. 26 Although the colon has comparatively less Ca absorption capability than the small intestine, its contribution can reach up to 30% in fiber-rich diets. 25 , 27 The gut microbiota may influence this process directly by breaking down undigested substrates and releasing minerals bound to complex molecules. 28 Indirect mechanisms have also been described through the production and action of microbial metabolites. These include reducing luminal pH, which enhances mineral solubility, 27 , 29 increasing the intestinal absorptive surface via epithelial cell hypertrophy, 30 and altering the expression of transport proteins involved in calcium uptake, such as Transient Potential Receptor Vanilloid 6 (TRPV6), Plasma membrane Ca2+-ATPases (PMCA1), Na + /Ca2 + exchange (NCX1) or vitamin D receptor (VDR) 31 (Figure 2). Gut‒bone axis Mechanisms. Gut microbiota-derived metabolites can impact bone remodeling through both direct and indirect mechanisms. Directly, they can cross the intestinal barrier, enter the bloodstream, and influence bone remodeling by promoting or inhibiting osteoblast and osteoclast differentiation and activity. Indirectly, several chemical and physiological mechanisms have been described (from left to right). Microbial metabolites can modulate the expression of tight junctions (claudins, ZO-1, and occludins), thereby regulating intestinal epithelial cohesion and controlling gut permeability. These two factors have a key influence on nutrient absorption and inflammation. Some microbial metabolites have demonstrated immunomodulatory effects by either promoting proinflammatory responses, such as increased Th17 activity and IL-17 production, or anti-inflammatory pathways, including the expansion of regulatory B and T cells. Last, SCFA have been shown to reduce luminal pH, enhancing mineral solubility and promoting expression of specific transporters (e.g. TRPV6, PMCAb) that facilitate the uptake of micronutrients essential for the bone matrix formation. Collectively, these interconnected gut-driven mechanisms act systemically to significantly impact skeletal health, favoring bone formation or promoting resorption. Created in https://BioRender.com. Osteoimmunology This field of research explores how immune cells and inflammatory signals influence bone remodeling and health. Immune cells contribute to bone homeostasis through diverse molecular mechanisms, among which the receptor activator of nuclear factor κB (RANK)/receptor activator of nuclear factor κB ligand (RANKL)/osteoprotegerin (OPG) signaling axis is one of the most critical, as it regulates osteoclast differentiation and activity. RANKL is a key ligand released by osteoblasts and immune cells. It fosters osteoclast differentiation by binding to RANK on the surface of osteoclast precursors and initiating nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), mitogen-activated protein kinase (MAPK), and nuclear factor of activated T cells, cytoplasmic 1 (NFATc1) signaling cascades, thereby promoting osteoclastogenesis and bone resorption. OPG secreted by osteoblasts and immune cells, protects bone from excessive resorption by binding to RANKL and preventing its interaction with RANK. 32 Several immune cell subpopulations have been identified as significant regulators within the field of osteoimmunology. In the gut lamina propria, dendritic cells participate to mucosal protection by sampling antigens from the gut lumen and activating effector cells in mesenteric lymph nodes. 33 The most abundant T helper cell subset include Th17 cells (CD4⁺ RORγt⁺), which secrete proinflammatory cytokines such as IL-17, RANKL, and TNF-α, 34 and their counterparts, Tregs (CD4⁺/CD8⁺ Foxp3⁺), which release IL-10, IL-4, and TGF-β to promote anti-inflammatory responses. 35 , 36 These two cell subsets are modulated by the gut microbiota both locally in the gut and mesenteric lymph nodes, and systemically in the spleen and bone marrow. 37 The balance between Th17 and Treg cells in these compartments is critical for maintaining bone homeostasis. Recently, the role of regulatory B cells (Bregs) has been introduced. Regulatory B cells (CD19+CD1dhiCD5+ Bregs) function as a key regulatory switch in maintaining the balance between Th17 and Treg cells, establishing the Breg–Treg–Th17 axis 38 (Figure 2). Microbiota plays a critical role in modulating Breg differentiation. 39 Interestingly, some probiotics have been shown to effectively modulate B-cell populations and reduce ovariectomy (OVX)-induced bone loss. 15 , 40 , 41 Microbial metabolites associated with bone health Microbial metabolites are bioactive compounds produced by gut bacteria during the fermentation of dietary components, primarily in the large intestine but also in the small intestine (Figure 3). These metabolites can affect bone health directly by influencing the differentiation and proliferation of osteoblasts and osteoclasts, or indirectly by modulating physiological processes described above. In this section, we focus on microbial metabolites that may contribute to the regulation of bone health. Gut-microbial-derived metabolites and their dietary substrates. Microbial metabolites originate from two primary sources: undigested carbohydrates and amino acids derived from undigested proteins or peptides. The main short-chain fatty acids (SCFAs)—acetate, propionate, and butyrate—are produced through the fermentation of undigested carbohydrates. In contrast, amino acids give rise to a wide variety of microbial metabolites. For example, bacterial degradation of tryptophan generates several metabolites important for bone physiology, such as indole-3-propionic acid (via the indole pathway), kynurenic acid (via the kynurenine pathway), and 5-hydroxytryptophan (via the serotonin pathway). Similarly, the microbial degradation of branched-chain amino acids (BCAAs) yields branch-chain fatty acids (BCFAs) or degradation, while the degradation of ornithine (Orn), arginine (Arg), and lysine (Lys) produces polyamines. Sulfur- or sulfate-containing amino acids lead to the formation of hydrogen sulfide (H2S) and choline- and carnitine- amino acids lead to the production of trimethylamine (TMA). Another important group of microbial metabolites includes vitamins and secondary bile acids, which arise from diverse substrates. Created in https://BioRender.com. Short and branch-chain fatty acids Short-chain fatty acids (SCFAs) are a class of fatty acids characterized by having fewer than six carbon atoms, produced primarily through the microbial degradation of indigestible fibers such as inulin-type fructans or resistant starches. 42 The most abundant SCFAs, acetate, propionate, and butyrate, are usually found in the gut at a molar ratio of 60:20:20, with other SCFAs, such as valerate or formate, found in smaller amounts. Although formate is increasingly recognized as a gut microbial metabolite involved in shaping microbiota composition, modulating gut inflammation, 43 and potentially contributing to host one-carbon metabolism 1 1 The one-carbon metabolism is a biochemical pathway comprising two interconnected modules: the folate cycle and methionine cycle. This pathway contributes to the regulation of critical processes including, amino acid homeostasis (e.g. methionine and homocysteine), and epigenetic mechanisms. Thereby regulating core cellular processes such as cell survival, growth, division, and stress responses. , 44 its role in bone physiology remains unexplored. In humans, intestinal SCFA concentrations range from 10 to 130 mM, while blood concentrations range from 80 to 375 μM. In mice, intestinal concentrations vary between 0.1 and 40 mM per gram of intestinal content. 45 The one-carbon metabolism is a biochemical pathway comprising two interconnected modules: the folate cycle and methionine cycle. This pathway contributes to the regulation of critical processes including, amino acid homeostasis (e.g. methionine and homocysteine), and epigenetic mechanisms. Thereby regulating core cellular processes such as cell survival, growth, division, and stress responses. SCFAs have been shown to exert notable effects on bone metabolism in both in vitro and in vivo mice studies. 4 , 5 , 40 Lucas et al. demonstrated that individual SCFA at concentrations of 0.5 mM could suppress osteoclast formation by altering cellular metabolism in bone-marrow-derived mouse cells. This finding was further supported by another in vitro study, which demonstrated an inhibitory effect at 0.125 mM. 40 However, no effects on bone formation could be observed. In contrast, in vitro experiments have shown that valeric acid can modulate both osteoblast and osteoclast differentiation. These findings were consistent with preclinical in vivo evidence. Supplementation with valeric acid was shown to be protective against bone resorption and improved bone microstructure in ovariectomized (OVX) mice. 13 Similarly, treatment of mice with individual SCFA significantly increased bone mass and prevented postmenopausal-induced bone loss. 4 To date, several mechanisms have been elucidated through which SCFAs exert their beneficial effects on bone health. SCFAs facilitate calcium absorption by expanding the colonic epithelial surface and reducing luminal pH, thereby enhancing mineral solubility. 27 , 29 , 30 They can also influence insulin-like growth factor-1 (IGF-1), a critical regulator of bone growth and development. 46 Several studies have demonstrated that SCFAs might contribute to the regulation of hormones with anabolic effects on bone, such as glucagon-like peptide-1 (GLP-1), although current evidence for GLP-1's impact on bone is still limited. 47 Another key mechanism through which SCFAs act is the inhibition of histone deacetylase (HDAC) activity, particularly by butyrate, which fosters the expansion of regulatory Tregs and preserves immune system homeostasis. Finally, SCFAs engage G-protein-coupled receptors (GPCRs), 48 which are expressed on both immune and bone cells, allowing them to influence bone metabolism both directly and indirectly. Clinical efficacy of SCFA supplementation remains limited. 49 Nevertheless, a few observational studies have revealed potential associations. In an observational study of 499 community-based peri- and early postmenopausal women, Greenbaum et al. reported that predicted levels of fecal butyrate and valerate were associated with femoral neck BMD. 50 Supporting these findings, a recent study showed that serum valeric acid was positively associated with L1–L4 BMD and was causally downregulated by B. vulgatus in a Chinese cohort of peri- and postmenopausal women. 13 Similarly, in a study involving 684 middle-aged Chinese individuals, researchers found a positive association between lumbar spine BMD and butyrate biosynthesis. 51 While SCFAs are the most abundant microbial fatty acids, the gut also produces other fatty acids in lesser quantities, such as branched-chain fatty acids (BCFAs). These include isovalerate, 2-methylbutyrate, and isobutyrate, which arise from the microbial fermentation of the branched-chain amino acids leucine, isoleucine, and valine. Isovalerate has been reported to stimulate chemotactic migration of osteoclast precursor cells but to suppress osteoclast differentiation in vitro. Additionally, administration of isovaleric acid to ovariectomized mice decreased the expression of osteoclast-related genes and reduced osteoclast generation, thereby reducing the bone porous area. 52 While no randomized control studies have been conducted, a recent observational study found significantly reduced levels of isovalerate in the serum of osteoporotic women compared with normal-BMD controls. 11 Tryptophan-derivates Tryptophan is an essential amino acid that must be obtained through the diet. Upon reaching the colon, commensal bacteria degrade it, resulting in a variety of microbial-derived compounds via three major pathways: the indole pathway, the serotonin pathway, and the kynurenine pathway. The indole pathway represents the first route for tryptophan degradation, producing metabolites such as indole-3-propionic acid (IPA), indole-pyruvic acid, indole-3-acetic acid (IAA), indole-3-acetamide, and tryptamine. In vitro experiments with primary mouse osteoblasts have shown that IPA treatment upregulates the expression of key genes involved in osteogenic differentiation such as runt-related transcription factor 2 (Runx2), secreted phosphoprotein 1(spp1), bone gamma-carboxyglutamic acid–containing protein 2 (Bglap2), alkaline phosphatase (ALP), and collagen type I (Col1al), indicating its role in promoting osteoblast differentiation. IPA treatment has also been shown to prevent bone marrow-derived osteoclast differentiation in vitro, with the strongest inhibitory effect at early stages of differentiation. 53 , 54 In line with these findings, targeted levels of indoleacrylic acid (IA), IAA, and IPA in serum have been shown to be significantly lower in OVX mice compared with controls. Among these, IA and IPA have demonstrated strong positive correlations with BV/TV. 54 Similarly, IPA treatment has also been shown to be effective in reducing obesity-induced bone loss in mice. 53 , 54 In humans, indoles were found to be the most discriminant pathway between postmenopausal women with low and normal BMD. 11 Indole metabolites frequently act as ligands of the aryl hydrocarbon receptor, a cytosolic ligand-activated transcription factor involved in immune regulation. 55 Additionally, indoles can upregulate the expression of tight junction proteins, mucus proteins, and the anti-inflammatory factor IL-10 in intestinal epithelial cells, while downregulating the expression of the proinflammatory factor IL-8. 56 , 57 Through these mechanisms, indoles contribute to improved nutrient absorption and immune regulation and thus support greater bone remodeling. However, excessive accumulation of indoles should be tightly regulated, as they serve as precursors for indoxyl sulfate, a hepatic metabolite excreted by the kidneys. Acute low doses of indoxyl sulfate in circulation have been associated with bone damage via osteoclast differentiation. Conversely, long-term and/or high-dose indoxyl sulfate exposure is associated with reduced osteoclast formation. 58 The second pathway is the serotonin pathway, which produces serotonin, melatonin, and 5-hydroxytryptophan. 59 In recent years, serotonin [5-hydroxytryptamine (5-HT)], a well-studied neurotransmitter, has been recognized for its role in regulating bone metabolism. 12 5-HT exists in two forms based on its site of synthesis: brain-derived serotonin (BDS) and gut-derived serotonin (GDS). Notably, these two forms have opposing effects: GDS inhibits bone formation, whereas BDS promotes it. 60 As evidence, when GDS was inhibited, OVX rats developed significantly more bone mass. 61 Last, the kynurenine pathway results in the production of several intermediates, including kynurenic acid, niacin, xanthurenic acid, and ultimately NAD+ 59 . The gut microbiota plays a significant role in driving the enzyme indoleamine 2,3-dioxygenase, which is responsible for converting tryptophan into kynurenine within intestinal epithelial cells. The impact of kynurenine on bone health has shown contradictory findings. On one hand, an in vitro study demonstrated that kynurenine treatment enhanced RANKL-mediated osteoclast differentiation and the resorption ability of Raw 264.7 mouse cells. Mechanistically, kynurenine was shown to enhance osteoclast genesis by binding to the aryl hydrocarbon receptor and upregulating c-fos and nuclear factor of activated T-cells (NFATc). 62 On the other hand, a study demonstrated that kynurenine could inhibit osteoclast differentiation in human peripheral blood mononuclear cells (PBMCs) cultured under osteoclast differentiation conditions. 63 While these differences could be attributed to species-specific variations between humans and mice, several contrasting observations can also be also found in clinical studies. For example, physical activity has been shown to increase the downstream metabolites of kynurenine in older adults, 64 but elevated levels of kynurenine have been associated with aging 65 and certain cancers. 66 Vitamins B and K Alongside dietary sources, the gut microbiota contributes to the synthesis of water-soluble vitamins, including B vitamins (B1, B2, B3, B5, B6, B7, B9, and B12), as well as fat-soluble vitamins such as vitamin K2 (menaquinones, MK-n). 67 A genomic analysis of 256 common human gut bacteria revealed that 40–65% of these bacteria have the potential to synthesize some or all B-vitamins. 68 In a separate study, analyzing 254 genomes, approximately 10% of the genomes had the complete ability to produce menaquinones, including bacteria from the Firmicutes and Proteobacteria. 69 These microbial pathways may contribute up to 37% of the daily folate requirement and approximately 10–50% for vitamin K needs. 70 Of note, many vitamins present in dietary sources—particularly fermented foods—are themselves produced through microbial activity (e.g., natto, a traditional Japanese food produced by fermenting soybeans with Bacillus subtilis, is particularly rich in menaquinone-7). 71 Low folate availability, which commonly occurs with aging, can impair one-carbon metabolism and consequently leading to elevated levels of homocysteine in the blood. 72 In vitro experiments on human PBMCs have confirmed that homocysteine induces osteoclastogenesis. 73 , 74 In osteoblasts, the adverse effects of homocysteine are dose dependent and treatment with folic acid is protective against these effects. 75 In mice, folic acid supplementation reduced high-fat-diet-induced bone loss, with evidence pointing to microbiota-mediated pathways. 76 However, clinical proof of the efficacy of folate supplementation is inconsistent. While some population-wide observational studies have identified a positive correlation between dietary folate intake and BMD, 77 , 78 randomized controlled trials did not reveal beneficial effects of B vitamins on bone outcomes. 79-81 However, these studies did not account for variations in treatment regimens, such as vitamin type, dosage, and duration. Vitamin K is essential for bone homeostasis primarily through its role in osteocalcin γ-carboxylation. This post-translational modification enables osteocalcin to bind calcium ions and hydroxyapatite within the bone matrix. As a result, when vitamin K levels in serum are low, the carboxylation of osteocalcin is often impaired, thereby compromising bone integrity. 82 In vitro experiments have shown that vitamin K2 treatment exerts direct effects on osteoblasts by inducing autophagy 83 and inhibiting ferroptosis. 84 In osteoclasts, vitamin K2 treatment inhibited 1,25-dihydroxyvitamin D3-induced osteoclast formation by decreasing the expression of RANKL and OPG. 85 An in vivo mouse study has shown that Vitamin K3 (menadione bisulfite) treatment, a precursor of vitamin K2, significantly increased blood osteocalcin levels and improved bone ultrastructure by modulating osteoblast function in OVX animals. 84 , 86 Similarly, treatment with vitamin K2 improved femur BV/TV and CtBMD by promoting osteoblast function. Several lines of evidence suggest a potential role of vitamin K₂ in human bone health, although findings remain inconsistent. In a small observational study, postmenopausal women with osteoporosis had significantly lower serum levels of vitamin K₂ compared with women with normal bone mass. 84 In a 24-month randomized open-label trial, daily supplementation with vitamin K₂ was associated with a lower incidence of fractures and higher levels of carboxylated osteocalcin relative to the control group; however, no significant differences were observed in lumbar spine BMD. 87 Conversely, a separate 12-month trial in healthy postmenopausal Norwegian women found that vitamin K₂ supplementation did not influence rates of bone loss in the total hip, femoral neck, or lumbar spine. 88 Additionally, some studies have examined the relationship between dietary vitamin K intake—which includes both vitamin K₁ and vitamin K₂—and bone health outcomes. A meta-analysis reported an increase in BMD at the lumbar spine, whereas no consistent effect was observed at the femoral neck. Further subgroup analyzes revealed that these results were largely driven by certain sub-populations (Asian women and individuals who received vitamin K2 (MK-4) supplementation), suggesting a superior effect of vitamin K2 over vitamin K1. Of note, some concerns have been raised regarding the methodological rigor of several studies on this topic. 89 Therefore, future research is needed to clarify the effects of vitamin K across diverse populations and dosage ranges. Polyamines Polyamines (putrescine, spermine, cadaverine, and spermidine) are small organic molecules ubiquitously present in living organisms. In the diet, they are especially abundant in soybeans, mushrooms, wheat, animal livers, oranges, and green tea leaves. Experiments using germ-free mice have demonstrated that the main source of polyamines in the lower luminal part of the intestine is the gut microbiota. 90 Protein fermentation in the colon by the gut microbiota provides free amino acids, such as L-arginine or L-ornithine, that serve as precursors for polyamines. For example, L-arginine can be converted via citrulline into L-ornithine, leading to the production of putrescine. Alternatively, arginine can be converted into agmatine, which is then subsequently converted into putrescine. Putrescine can be further transformed into spermidine and spermine. 91 In vitro experiments have demonstrated that spermidine enhanced osteoblast activity by increasing osteopontin, OPG, and RANKL expression. 92 Other mechanisms, including the reduction of oxidative stress and the promotion of autophagy, have also been investigated in this context. 93 Additionally, spermidine and spermine have also been shown to inhibit osteoclast differentiation at high, but not low, doses by downregulating cathepsin K, TRAP, and matrix metalloprotease 9. 92 In mice, several studies have demonstrated the effects of polyamines on bone health. Spermidine or spermine supplementation in drinking water for 28 d was shown to significantly inhibit OVX-induced bone loss by decreasing osteoclast formation. 94 Another study demonstrated that warmth exposure prevented osteoporosis in mice through the production of microbial polyamines. 92 Indirect mechanisms i.e. modulation of inflammation and gut integrity have also been described. For example, spermidine appears to protect against collagen-induced arthritis by inhibiting the polarization of M1 macrophages, reducing the proinflammatory cytokines IL-6 and IL-1β and increasing levels of the anti-inflammatory cytokine IL-10. 95 Additionally, polyamines can support gut epithelial integrity and function by upregulating the expression of intracellular junction proteins and enhancing TLR4-mediated microbial signal transduction. 96 , 97 To date, clinical evidence on polyamines is largely limited to epidemiological studies examining associations between dietary polyamines and cognitive, cardiovascular, and morbidity outcomes. 98-100 With respect to bone health, a 9-y follow-up epidemiological study in individuals aged 40 to 69 y reported that high levels of spermidine and low levels of spermine and putrescine were associated with the risk of osteoporotic fractures. 101 The lack of RCTs on polyamines may be due to limited safety data and insufficient insights into their bioavailability in relation to human health outcomes. Only recently was a short-term safety trial testing high-purity spermidine supplementation (40 mg/d) in older adults published. While spermidine supplementation was well tolerated in healthy older adults, minimal changes in circulating polyamines were observed, suggesting tight homeostatic control. 102 Bile acids Bile acids are steroidal acids that act as emulsifying agents to facilitate the digestion, transport, and absorption of lipids and fat-soluble vitamins in the body. Primary bile acids, namely cholic acid and chenodeoxycholic acid, are synthesized from cholesterol by hepatocytes in the liver. Amino acid-conjugated primary bile acids are then released into the intestinal lumen, where they undergo transformation into secondary bile acids by the gut microbiota. This process begins with deconjugation of the amino acids, followed by dehydrogenation, dihydroxylation, and epimerization. As a result, cholic acid is converted into deoxycholic acid, and chenodeoxycholic acid into lithocholic acid and ursodeoxycholic acid. Primary and secondary bile acids are then recycled and reincorporated into the host circulation through a process known as enterohepatic circulation. The gut microbiota is involved in the synthesis of secondary bile acids and can also influence the quantity and composition of primary bile acids. 103 Bile acids act as signaling molecules and metabolic regulators in various transduction pathways. They exert their effects primarily through two receptors: the farnesoid X receptor (FXR) and the Takeda G-protein-coupled receptor 5 (TGR5), which are expressed across various tissues and notably present in both osteoblasts and osteoclasts. Although bile acids have also been shown to signal through other receptors such as the vitamin D receptor (VDR/NR1I1), the liver X receptors α and β (LXRα/β), the pregnane X receptor (PXR), the constitutive androstane receptor (CAR/NR1I3), and sphingosine-1-phosphate receptor 2 (S1PR2). 104 Through the activation or inhibition of these signaling pathways, bile acids may contribute to the regulation of bone health. Some studies have shown that lithocholic acid (LCA), for example, have deleterious effects on bone. LCA can bind to the VDR in osteoblasts; this binding results in reduced bone growth due to the disruption of mitochondria activity in osteoblasts. LCA can also decrease the effect of vitamin D on osteoblast through binding site competition, leading to inhibition of OCN and RANKL expression and thereby reducing bone formation. 105 Conversely, LCA may impair osteoclast differentiation by inhibiting the AMP-activated protein kinase signaling pathway. 106 Additionally, LCA has been shown to indirectly support bone health by modulating the host immune system. Two distinct derivatives of LCA, 3-oxoLCA and isoalloLCA, have been shown to modulate T-cell regulators in mice. 3-oxoLCA suppresses Th17 cell differentiation by directly binding to the key transcription factor RORγt, whereas isoalloLCA promotes Treg differentiation by inducing mitochondrial reactive oxygen species. 107 Ursodeoxycholic acid (UDCA) has, by contrast, shown consistently beneficial effects on bone health via different mechanisms. First, UCDA have been shown to influence osteoblasts by regulating ROS and cytokine production. The accumulation of ROS facilitates the activation of mitochondrial permeability transition pores, which play a crucial role in initiating cell-stress-induced death. UDCA and its amidated conjugates, tauroursodeoxycholic acid and glycoursodeoxycholic acid, have been shown to reduce ROS production, inhibit the formation of these pores in mitochondria, and downregulate the expression of pro-apoptotic proteins. 108 Additionally, UDCA has been shown to upregulate the expression of osteogenic factors such as Osterix and Runx2 while inhibiting adipogenic differentiation in myeloid-derived stem cells and downregulating inflammation factors. 109 , 110 Another study also reported that UDCA can inhibit pro-inflammatory responses in RAW 264.7 macrophages induced by LPS, suggesting that UDCA can act as an osteoclast inhibitor. 111 Lastly, it has been demonstrated that UDCA can influence VDR expression and promote calcium absorption. 112 Although bile acid supplementation studies on BMD are extremely limited, various human observational studies suggest a relationship between bile acid levels and bone health. 113-116 For example, an observational study examining serum bile acid levels in healthy, osteopenic, and osteoporotic individuals reported that bile acids were positively associated with BMD at the lumbar spine, femoral neck, and total hip. However, the correlation with serum markers was not consistent. 113 Overall, these findings underline the important role of the gut microbiota in modulating bone health via secondary bile acids. Nonetheless, it is important to note that bile acid composition in mice differs considerably from that in humans. 117 Therefore, some discoveries observed in mice should be interpreted with caution when applying them to human physiology. Hydrogen sulfide Hydrogen sulfide (H2S) is a gaseous microbial metabolite produced through the anaerobic respiration of sulfite-/sulfur-containing substrates, such as methionine, cysteine, or taurine. H2S levels in the body can originate from both the gut microbiota and the host. A study using germ-free mice reported that H2S blood levels are markedly altered in mice lacking microbiota when compared to conventional mice. Interestingly, the presence of gut microbiota can influence host endogenous H2S production in different tissues, 118 suggesting that the microbiota may play a role in regulating H2S production. H₂S involvement in bone regulation is supported by a growing body of in vitro and in vivo experiments. In osteoblasts, H2S has been shown to promote the self-renewal and osteogenic differentiation of BMSCs. 119 In preosteoclasts, sodium hydrogen sulfide has been shown to inhibit osteoclast differentiation through direct and indirect mechanisms by reducing intracellular reactive oxygen species (ROS) levels and triggering nuclear factor erythroid 2-related factor 2-dependent antioxidant responses. 120 As another potential mechanism, H2S has been shown to activate and upregulate numerous calcium channels in the colon, including TRPV6, TRPV4, and TRPV3, suggesting that H2S could increase Ca absorption in the intestine. 26 , 121 In mice, knocking down cystathionine β-synthase, the endogenous enzyme responsible for converting cysteine to H2S, results in an osteoporotic phenotype in bone BMSCs. 122 Restoration of H2S levels reverses bone-loss phenotype and b osteogenic deficiency of BMSCs. 123 In humans, observational data reveal a different aspect of sulfur metabolism. In a breath-metabolomics study conducted in postmenopausal women, levels of dimethyl sulfide and allyl-methyl sulfide (microbial-derived methylations of H2S) were markedly elevated in women with low BMD and high fracture risk compared with women with normal BMD. These findings were replicated in an independent cohort. The authors suggested that these metabolites could be serve as noninvasive biomarkers of altered gut microbial sulfur metabolism associated with osteoporosis. 124 , 125 Choline metabolism Choline and L-carnitine are essential nutrients for humans. Although the body can synthesize small amounts of them, they still need to be obtained through the diet. Once they reach the large intestine, they undergo microbial degradation, leading to the production of trimethylamine (TMA), which is subsequently converted into trimethylamine-N-oxide (TMAO) in the liver. An in vitro study reported that increased levels of TMAO impair endoplasmic reticulum homeostasis and the mitochondrial unfolded protein response, thereby aggravating osteoblast dysfunction and contributing to osteoporosis. 126 Another study investigating mechanisms of action of TMAO on bone health highlighted that TMAO may promote osteoclast maturation via activation of the ROS-dependent NF-kB signaling pathway. 127 In mice, ingestion of TMAO for 8 weeks has been shown to significantly reduce bone loss compared with the control. 127 More recently, the use of gold nanoparticles has been studied as a potential therapeutic approach for osteoporosis due to their antimicrobial and anti-inflammatory activity. Gold nanoparticles were able to reduce OVX-induced bone loss in a gut-microbiota-dependent manner by reducing TMAO metabolism and inflammatory cytokines. 128 Conclusions from clinical studies investigating TMAO serum levels and bone health are diverse. In a clinical study involving 286 postmenopausal women, serum TMAO levels were elevated in those with a higher risk of fractures. 129 Another study suggested that the TMAO association with fracture risk appeared to be especially relevant only when body mass index was taken into account. The association between TMAO and BMD was not significant. 130 Lastly, in a study involving 264 overweight participants, a greater reduction in TMAO levels in blood was associated with a greater loss of whole-body BMD after 6 months; this association was maintained after 2 y, independently of diet. 131 Nutritional strategies In the previous section, we reviewed different categories of bioactive compounds and their potential influence on bone metabolism. The use of microbial metabolites as a nutritional strategy is referred to as postbiotics. However, these compounds often present challenges such as unpleasant taste and instability, which can lead to degradation or transformation of the bioactive component before reaching the site of absorption. To date, clinical evidence supporting their efficacy remains limited (Table 1). In contrast, strategies aimed at modulating the microbiota and its metabolic activity through the use of probiotics, prebiotics, or synbiotics have become key approaches in the field of bone research. Additionally, in this section, we will discuss emerging findings on fermented dairy products and nutritional patterns. Table 1.Preclinical and clinical evidence on the microbial metabolites' benefits on bone health.MetaboliteLink to microbiotaMoA (indirect)Level of evidence related to bone health In vitro (concentration used) In Vivo (concentration used)Epidemiological evidence(population)Intervention evidence(dose and population)Butyrate [4,40,50,51] Clostridiaceae, Eubacteriaceae, Lachnospiraceae, and Ruminococcaceae via fermentation of dietary fiber↑Ca absorption (↓pH)Endocrine andImmune regulation↑Gut permeability ↓Osteoclast(125–500 uM) ↑BMD in OVX(45 d, 150 mM)↑BV/TV and TbSp WT & OVX(DW, 150 mM) ↑FN BMD and ↑L1 BMDPeri and Postmenopausal Χ Propionate [4,40] Akkermansia muciniphila, Eubacterium dolichu, Ruminococcus bromii, Bacteroides eggerthii, Bacteroides fragilis, and Veillonella parvula via fermentation of dietary fiber ↑Ca absorption (↓pH)Immune regulation ↓Osteoclast(250–500 uM) ↑BV/TV WT & OVX(DW, 8w 150 mM) Χ Χ Acetate [4,40] Lactobacillus, E. coli, K. pneumoniae, Proteus mirabilis, Enterococcus faecalis, and Staphylococcus saprophyticus via fermentation of dietary fiber↑Ca absorption (↓pH) Immune regulation ↓Osteoclast(500 uM) ↑BV/TV and ↓TbSp in WT(DW, 8w, 150 mM) Χ Χ Valerate [13,50] Bacteroides vulgatus via fermentation of Fiber/protein↑Ca absorption (↓pH) ↑ Osteoblast (1000 uM) ↓Osteoclast(1000 uM)↑BV/TV, ↑TbTh,(DW, 8w, 1.5 mM)↑L1-L4 BMDAnd ↑FN BMDPeri and Postmenopausal Χ Isovalerate [11,52] Bacillus spp Bacteroides spp. Propionobacterium via fermentation of protein (Leu-rich)anti-inflammatory ↓Osteoclast(100–500 uM) ↓ Osteoclast-related genes in OVX(150 mM by GV) Low in women with osteoporosis Χ Kynurenine[62,132,133] Lactobacillus spp. Bifidobacteria Spp. via fermentation of protein (Trp-rich)Immune regulation ↑Osteoclast(25 uM) ↓Osteoclast(100 uM) ↓CtBMD in ORX mice(10 mg/kg IP injection) No association was found with BMD in adults and elders women and men Χ Indoles[11,54,122] Echerichia coli, Clostridium spp, Bifidobacteria spp. Bacteroides spp,via fermentation of protein (Trp-rich)↓ROS↑ Immune regulation↑Gut integrity ↑Osteoblast(25–100 uM) ↓Osteoclast(100 uM) ↓ blood levels of indoles in OVX(10–20 mg/kg by GB) Low in women with osteoporosis Χ Folic acid (Vitamin B9) [75–77,79–81] Lacticaseibacillus spp.Effect on bone viahomocysteine levels on bone ↑Osteoblast(1–100 uM) ↓ blood levels of Folate in aged rats ↑BMC, PINP and ↓CTX in Obese mice(20 μg/ml in DW) ↑Folate intake correlated to ↑BMD in elders +80 Inconsistent findingsVitamin K (menaquinones)[83–86,89,134–136] Firmicutes and Proteobacteria B. subtilis Effect on bone viaOsteocalcinCa absorption↑Osteoblast(1–10 uM(effect on autophagy andferropsis) ↓Osteoclast(0.5–10 uM) ↑BV/TV, CtBMD and osteoblasts function in OVX(20 mg/kg/d by GV) ↑Bone metabolism and ultrastructure in OVX(10 mg/L/kg in DW)↓ blood levels of vitamin K, ↑ risk of fracture ↓ blood levels of vitamin K in postmenopausal women with osteoporosisBeneficial effect on BMD with vitamin K2 but findings are inconsistentSpermine[92–94,96,97] A. muciniphila via fermentation of protein (Arg or Orn-rich)Oxidative stressAutophagyImmune regulationGut permeability ↓Osteoclast(0.01–10 uM) ↑Osteoblast(10 uM) ↑BV/TV, ↓ CTX-1 in OVX by decreasing osteoclast number(0.3–3 mM in DW) Mix of polyamine improves bone strength(0.5 mM in DW) Χ Χ Spermidine[92–97,101] A. muciniphila via fermentation of protein (Arg or Orn-rich)Oxidative stressAutophagyImmune regulation (via macrophage polarization)Gut permeability ↓Osteoclast (0.1–10 uM) ↑BV/TV, ↓ CTX-1 in OVX by decreasing osteoclast number(0.3–3 mM in DW) Mix of polyamine improves bone strength(0.5 mM in DW) Spermidine concentrations > 1.57 μM had a 2.2-fold higher risk of fractures (69yo) Well tolerability but, no available studies on boneLCA[106,107,137] Lacticaseibacillus spp. (BSH + taxa)Metabolism regulation through (Cholesterol and lipid metabolism)Anti-oxidantImmune regulation (↑ Treg/↓TH17 cells) ↓Osteoblast(<10 uM)(↓ Mitochondria activity by binding to VDR) ↓Osteoclast(0.01-0.05uM) Bile acid receptor (Tgr5 -/- ) knock-out markedly decreased bone mass in aged and OVX mice Bile acids positively associated with LS, FN, Hip BMD Χ UDCA [108–111,137,138] Lactobacillus spp. via hydrolyzation of primary bile acids Metabolism regulation through (Cholesterol and lipid metabolism)Anti-oxidantAnti-inflammatory ↑ Osteoblast(200 uM) ↓Osteoclast(500–2000 uM) Injection of 100 ug UDCA improved BV/TV and TbTh in bone defects Χ H2S [119–124] Fusobacterium, Clostridium, Escherichia, Salmonella, Klebsiella, Streptococcus, Desulfovibrio, Enterobacter, Desulfovibrio, Desulfobacter, Desulfobulbus, and Desulfotomaculum via fermentation of protein (Met or Cys-rich) Ca absorption (increase of Ca channels TRPV3,4 and 6) ↓Osteoclast(NaSH, 100–300 uM) Knocking down the enzyme for H2S production leads to a osteoporotic phenotype in mice. Restoration of H2S reversed osteogenic deficiency↑ H2S in the breath of women with low BMD and high risk of fracture Χ TMAO [126–129,131] (from TMA) Enterococcus, Clostridium via fermentation of carnitine and Choline Metabolismgut microbiota ↓ Osteoblast(200 uM) ↑Osteoclast(200–400 uM) ↓BV/TV, ↓BMD, ↓TbTh, ↓TbN, ↑TbSp(1.5% in DW) Gold particles reduced OVX-induced bone loss via gut microbiota, along with, TMAO levels ↓BV/TV, ↓TbBMD, ↓TbTh, ↓TbN, ↑TbSp, ↓CtBMD, ↑SMI(0.2% in DW) ↑ TMAO serum levels in women with ↑ fracture risk No association between TMAO and BMD and risk of fracture ↓ TMAO serum levels associated to ↑ Bone loss in SPINE BMD Χ Abbreviations: WT: wild-type; OVX: ovariectomy; ORX: orchidectomy; DW: drinking water; GV: gavage; IP: intraperitoneal injection; FN BMD: femoral neck BMD; L(n) BMD: lumbar spine BMD; LCA: lithocholic acid; UDCA: ursodeoxycholic acid. TMAO: trimethylamine-N-oxide; TMA: trimethylamine; BV/TV: bone volume per tissue volume; TbTh: trabecular thickness; TbN: trabecular number; TbSp: trabecular Space; BMC: bone mineral content; SMI: structure model index; CtTh: cortical thickness; d: days; CtBMD: cortical bone mineral density; TbBMD: trabecular bone mineral density. Preclinical and clinical evidence on the microbial metabolites' benefits on bone health. Abbreviations: WT: wild-type; OVX: ovariectomy; ORX: orchidectomy; DW: drinking water; GV: gavage; IP: intraperitoneal injection; FN BMD: femoral neck BMD; L(n) BMD: lumbar spine BMD; LCA: lithocholic acid; UDCA: ursodeoxycholic acid. TMAO: trimethylamine-N-oxide; TMA: trimethylamine; BV/TV: bone volume per tissue volume; TbTh: trabecular thickness; TbN: trabecular number; TbSp: trabecular Space; BMC: bone mineral content; SMI: structure model index; CtTh: cortical thickness; d: days; CtBMD: cortical bone mineral density; TbBMD: trabecular bone mineral density. Probiotics Probiotics are live microorganisms that, when provided in adequate amounts, can confer a benefit to the host. 139 Their benefits for bone metabolism have been extensively studied, highlighting multiple physiological mechanisms underlying their positive effects. In this section, we will review the main probiotic groups studied in relation to BMD, bone quality and, fracture risk focusing on their metabolic activity. Lactobacilliales Lactobacilliales are the most comprehensively studied group of probiotics, with commonly isolated species including Lactobacillus acidophilus, L. salivarius, L. casei, L. plantarum, L. fermentum, L. reuteri, L. rhamnosus, L. gasseri, and L. brevis. These bacteria are ubiquitous in nature and typically present in the gastrointestinal tract. Their metabolism is characterized by the fermentation of simple carbohydrates which leads to the production of lactic acid. The production of lactic acid enables them to rapidly lower the pH of their surroundings, giving them a competitive advantage within microbial communities. A meta-analysis of 71 preclinical studies, of which 71% investigated Lactobacillus species, examined the effects of probiotics on the prevention of bone loss in both intact and ovariectomized rodent models. In intact rodents, probiotics resulted in greater vBMD and higher BV/TV, without changes in cortical bone structure. In ovariectomized models, probiotic supplementation resulted in greater vBMD, BV/TV in femur, tibia, and spine, and CtTh at the tibia. 140 Among all Lactobacillus, L. rhamnosus and L. reuteri have been the most extensively studied in the context of skeletal health. 5 , 21 , 22 , 141-144 Their beneficial effects on bone have been consistently associated with immune modulation and alterations in gut microbiota composition. L. rhamnosus, for example, has been associated with improved gut epithelial integrity and reduced proinflammatory gut environment, reflected by decreased levels of Th17 and increased Tregs in Peyer's patches, and their cytokines in the gut (Table 2). 22 , 41 Table 2.Probiotic from the genus Lactobacillus/Lacticaseibacillus have been investigated in the gut–bone axis.BacteriaEvidenceAnimal modelpopulation interventionEffect on boneGut and inflammationMicrobiota and metabolismRef L. reuteri PreclinicalAdult male and female C57Bl/6 mice3 × /week,4 weeks3 × 108 cfu/ml by GVFemur:↑BV/TV, ↑Tb.N., ↑Tb. Th., ↓Tb.Sp., ↑BMC and ↑BMD. Vertebrae: ↑BV/TV, ↑Tb. N., ↑Tb. Th., ↓Tb. Sp, ↑BMC and ↑BMD. No bone effect on female mouse↑Serum osteocalcin ↑Femur BFR↓TNFα in ileum and Jejunum [141]OVX female Balb/c mice3 × /week,4 weeks,4.5 × 107 CFU/ml by GV + 1.5 × 108 CFU/ml in DWFemur: ↑BV/TV, ↑BMD, ↑BMC, ↑Tb. N, ↑Tb.Th. and↓Tb.Sp. Vertebrae: ↑tb.Th.Tibia: ↓RANKL, ↓OsteoclastBM: ↓CD4 + T-lymphocytes↑ Richness (Chao1 and ACE), ↑Clostridiales, ↓ Bacteriodales[142]ClinicalRCT, n = 70 women with low BMD(75–80yo)12 months, 1 × 1010 CFU/d↑ total tibia vBMD [145] RCT, n = 70 women(75–80yo)12 months, 1 × 1010 CFU/d Fecal metabolites: BCAA and their derivatives, isovalerylcarnitine (C5), 2‐methylbutyrylcarnitine (C5), and isovalerylglycine, especially increased with the supplementation of L. reuteri. isovalerylcarnitine showed a positive correlation with total tibia vBMD at both baseline and 12 months.[146] L. rhamnosus PreclinicalSex steroid deficiency (pharmacologically)in female C57BL6/J mice2x/week4-weeks,1 × 109 CFU/dFemur and vertebrae: ↑BV/TV↑Tb.Th, ↑Tb.N, and ↑Tb.Sp, ↑Ct.V Serum: ↓CTX, ↑OsteocalcinIntestine: ↓FITC-Dextran ↓ Endotoxin, ↑claudin 2,3, 15, and Jam3, ↓Rankl,↓ TNF, and ↓ Il17 mRNAs.BM:↓ Rankl,↓ TNF, and ↓IL-17 [21]Intact female, C57BL/6 mice5x, 4-weeks, 1 × 109 CFU/dVertebrae↑BV/TV,↑ Tb.N. and ↑Tb.Th. Femur↑ BV/TV,↑ Tb.N., ↓Tb.Sp., ↑ P1NPBM: CD4 Foxp3, CD8 + T cellsSpleen CD4 Foxp3,↑ ClostridiaSmall intestine and serum metabolites:↑Butyrate[5] OVX female, BALB/c mice4 weeks,4 × 108 CFU/d in DWVertebrae: ↑BMD, ↑ BV/TV, ↑Tb.Th., ↓Tb.Sep. Femur Tb: ↑BMD, ↑ BV/TV, ↑Tb.Th., ↓Tb.Sep. Tibia tb: ↑BMD,↑ BV/TV, ↑Tb.Th., ↓Tb.Sep., Femur Ct: ↑T.Ar., ↑T.Pm., ↓Ct.Po., Tibia ct: ↑ T.Ar., ↓CtPo.-BM, LN, Spleen and PP: ↑CD4 + Foxp3 + , ↑CD8 + Foxp3 + , ↓CD4 + Roryt + , ↓CD4 + RANKL + Serum: ↓IL-17, ↓IL-6, ↓TNF-a, IL-10,↑ ↑IL-4,↑ IFN-Y [41]OVX femaleSprague-Dawley rats6-weeks,1 × 109 CFU/d by GVFemur: ↑BMD,↑ BV/TV, ↓Tb.Sp, ↑Tb.N, ↑Tb.Th, ↑Stiffness, ↑Ultimate load, ↑Deflection. Serum: ↓ Ca, ↑P1NP, ↓CTX-1, ↓CTX/PINPColon:↓inflammation infiltration, ↓crypt damage, ↓goblet cells damage, ↑ZO-1, ↑Occludin, ↑Claudin I, ↑GLP-2, ↓LPS,↓IL-17, ↓TNF-a, ↑TGF-B,↑ IL-10,↓TH17/Tregs, BM: ↓IL-17, ↓TNF-A, ↑IL-10, ↑ TGF-B,↓ RANKL↑ Abundance and community diversity. ↑Lactobacillaceae, ↑Ruminococcaceae, ↑Eubacterium, ↑ Peptostrepto-coccaceae, ↓Lachnospiraceae, ↓Oscillospiraceae, ↓Desulfovibrionaceae Fecal metabolites: 7-Ethoxy-4-methyl-2 H–1-benzopyran-2-one, Anofinic acid, Bunitrolol, Pindolol, 9-Octadecenamide, Oleamide, 1,3-Diacetoxy-4,6,12-tetradecatriene-8,10-diyne, Myrigalon B, Myrigalone A, Myrigalone E, Verimol C showed a significant positive correlation with the anti-inflammatory cytokines IL-10, TGF-β[22] L. paracasei PreclinicalOVX female ICR mice7x/week4 weeksHeat-killed1 × 1010 cells/mL by GVTibia ↑BV/TV,↑ BMD, ↓TRAP-5Bone: ↑TGF-β, ↑IL-10, ↓RANKL levelsSerum: ↓IL-17A and LPS↑Lachnospiraceae, ↑Rhizobiales, ↑Aeromonadaceae, ↑Bifidobacteriales,↑ Bifidobacterium, ↑Bifidobacteriaceae and↑ Rikenellaceae.[147] OVX mice C57BL/6N female7x/weekStarting 2-weeks before OVX6 weeks (Total treatment)1 × 1010CFU/ml in DWFemur: ↑ BV/TV, ↑BMD, ↓Tb.Th., Cortical area, Serum: ↓OsteocalcinUrine: ↓CaBM: ↓TNFa, ↓IL-1b, ↑OPG, TGF-B, ↑ Treg (with a mix of two strains of L. paracasei) [148] L. plantarum PreclinicalOVX female C57BL/6 mice7x/week1 × 109 CFU/d by GV54 dFemur: ↑BV/TV, ↑BMD ↑Tb.Th., ↑Tb.N., ↑Tb.Th., ↑ Max. breaking load, ↑Max. Bending stress.Intestine: ↓FITC-dextran, ↑claudin-family (1,2,3,15) proteins, ↑ZO-1, ↑JAM-3↓IL-1, ↓IL-6,↓ IL-17, ↓RORy,↓ TNF-α, ↓TLR4, ↓NF-kβ, ↓RANKL/OPGBone: ↓RANKL/OPG, ↓IL-1, ↓IL-6, ↓IL-17, ↓RORα. ↓RORy, ↓TNFαThe intestinal microbial communities of mice in the AR495 group exhibited greater differences from those of mice in the OVX group and smaller differences from those of mice in the SHAM group (↑Clostridia, ↑Lachnospiraceae,↑ E. xylanophilum)Fecal metabolites: ↑acetate, ↑propionate, and ↑butyrate[23] L. helveticus PreclinicalOVX femaleSprague-Dawley rats1 × 108–109CFU/ml16 weeksFemur: ↑BMD, ↑ ObS/BS, ↓OcS/BS, ↑OS/BS, ↑OV/BV, ↓ES/BS Serum: ↑OC, ↓ CTX-1 [149] L. gasseri PreclinicalOVX femaleSprague-Dawley rats1 × 1010 CFU/rat14 weeksCalcitonin, Bone volume [150] L. fermentum ClinicalRCT, n = 53 women(55–62yo)2×/d1 × 1010 CFU/12 months↑ Neck BMD and osteocalcin [151]OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: gavage, BM: bone marrow, PP: Peyer's patch, LN: lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular Space. BMC: bone mineral content, SMI: structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: Procollagen Type 1 N‑terminal Propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. Probiotic from the genus Lactobacillus/Lacticaseibacillus have been investigated in the gut–bone axis. OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: gavage, BM: bone marrow, PP: Peyer's patch, LN: lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular Space. BMC: bone mineral content, SMI: structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: Procollagen Type 1 N‑terminal Propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. The primary metabolic link between Lactobacillus and bone is the production of SCFA. Tyagi et al. demonstrated that administration of Lactobacillus rhamnosus GG effectively reduced bone loss in mice by increasing BV/TV and trabecular microstructural parameters. 5 Notably, increased LGG levels were associated with elevated butyrate concentrations in both the small intestine and serum. LGG lacks the genetic machinery to synthesize butyrate directly. 152 Instead, LGG appears to modulate gut microbiota composition, leading to an increased abundance of SCFA-producing Clostridium. 5 Similarly, treatment of OVX mice with Lactobacillus plantarum resulted in elevated levels of acetate, propionate, and butyrate in the intestinal contents. Probiotic treatment also led to a higher abundance of Lachnospiraceae, particularly the Eubacterium xylanophilum group, a known butyrate-producing genus. 23 Probiotics effects on gut microbiota and bone can vary depending on the dietary context. For example, OVX animals receiving a low-fat diet together with a probiotic mixture (Lacticaseibacillus paracasei DSM13434, Lactiplantibacillus plantarum DSM15312 and DSM15313) exhibited a significant increase in total body BMD. This effect was absent in animals consuming a high-fat diet. OVX animals treated with the probiotic mix showed elevated lactic acid production in comparison to SHAM animals, suggesting lactic acid as their potential cross-feeding mechanism. 153 Beyond lactic acid and SCFA, LAB are known to be involved in a range of biochemical processes such as bile acid hydrolysis or synthesis B vitamins such as folate (B9), thiamine (B1) and cobalamin (B2). 68 , 154 However, these mechanisms have been comparatively less investigated as potential therapeutic pathways for bone health than SCFA. Several RCTs have explored the effect of Lactobacillus on BMD, yielding both positive 145 , 155 , 156 and inconclusive results. 157 , 158 These findings have been pooled in a meta-analysis focusing on postmenopausal women. Overall, the results suggest that women consuming probiotics may benefit from increased lumbar spine and hip BMD. A subanalysis indicated that this evidence was specific to women with osteopenia rather than osteoporosis. 159 A previous meta-analysis had reported rather small effect sizes and exclusively at lumbar spine BMD. 160 The inconsistencies in observed effects may be attributed to variations in sample size, probiotic dose, and probiotic strains across studies, as the meta-analysis pooled data from different probiotic formulations. To date, the clinical efficacy of Lactobacillus formulations remains unclear. Bifidobacteria Bifidobacteria are natural inhabitants of the human gut but can also be found in breast milk and various dairy products. The most commonly isolated species are B. longum subsp. infantis, B. breve, B. bifidum, B. longum, B. lactis subsp. animalis and B. adolescentis. They are the dominant intestinal microorganisms in healthy breastfed infants, and their presence has been linked to healthy growth in infants and rodents. 161 An observational study reported that as individuals transition to adulthood, the abundance of Bifidobacteria gradually declines. 162 Interestingly, B. longum supplementation has been shown to accelerate bony callus formation, thereby enhancing limb fracture repair in aged mice. 163 Another study reported that six weeks of B. longum supplementation significantly ameliorated bone loss induced by surgical menopause in rats. Several mechanisms have been highlighted such as improved gut integrity, decreased inflammation and changes in gut microbiota composition. 15 Bifidobacteria are well-studied for their ability to metabolize complex sugars by releasing glycosyl hydrolases, which cleave dietary glycans and facilitate their intake into the cells for fermentation. 164 In infants, Bifidobacteria metabolize human milk oligosaccharides (HMOs) and mucins while in adults bifidobacterial species cleaves dietary plant glycans, such as inulin-type fructan, resistant starch, galactan, xylan, and arabinan. As a result of glycan fermentation, Bifidobacteria produce various metabolites, including acetate, ethanol, and formate. 165 However, Bifidobacteria supplementation benefits have not been attributed to these metabolites specifically, but mainly to increased butyrate levels. A study on Bifidobacteria's effects in DSS-induced bone and muscle loss found increased levels of butyrate-producing bacteria like Clostridium species, along with elevated butyrate in serum and bone, suggesting that butyrate may play a key role in the protective effects. In vitro co-culture of Bifidobacterium lactis A6 and Clostridium butyricum significantly boosted butyrate production, likely due to acetate and lactate from bifidobacteria serving as substrates for C. butyricum. 166 These findings suggest that bifidobacteria could influence bone metabolism by fostering the growth of butyrate-producing bacteria. Another mechanism by which Bifidobacteria may impact bone health is through contributing to the production or activation of vitamin D. Fecal fermentation studies using Bifidobacterium longum FSHHK13M1 have shown increased levels of vitamin D metabolites. Furthermore, treatment with FSHHK13M1 increased 1,25-dihydroxyvitamin D levels and reduced retinol-induced bone loss in male mice by activating the expression of the VDR, OPG, Wnt10b/β-catenin, and Runx2/Osterix pathways and inhibiting the expression of RANKL/RANK pathway. 167 However, it remains unclear whether the vitamin D was directly produced by FSHHK13M1 or resulted from cross-feeding interactions. Clinically, only one RCT to date has investigated the effect of Bifidobacteria. In this study oral administration of Probio‑M8 did not lead to changes BMD after three months of intervention. However, it is uncertain reflects true absence of treatment effects or whether the interventions was too short to capture small changes in bone mass. On the other hand, Probio-M8 improved the bone metabolism, reflected by increased vitamin D3 levels and decreased PTH and procalcitonin levels in serum (Table 3). 16 Table 3.Probiotic from the genus Bifidobacterium that have been investigated in the gut–bone axis.BacteriaEvidenceAnimal modelpopulation interventionEffect on boneGut and inflammationMicrobiota and metabolism Ref B. longum PreclinicalOVX Sprague-Dawley rats1 × 108–9CFU/d by GV16 weeksFemur: ↑BV/TV, ↑TbN, ↓ Porosity %, ↑TbTh, ↑BMD,↑ObS/BS, ↑OS/BS, ↑OV/BV,↓OcS/BS, ↓ES/BS, Serum: ↑OC, ↓ CTX-1 [168]Aged (18th months) femaleC57BL/6JN mice1 × 108–9CFU/d by GV3,7,14,21 or 35 d postfractureVertebrae: ↑BMD, ↑BMC↑BV/TV (trend), ↑ TbTh, ↓BS/BV, ↓SMI Callus: ↑BV/TV at 14 and 21 d, ↓ Callus area at 14D, ↑Max torque and ↑Strength, ↑ Energy and Failure.Intestine: ↓FITC-Dextran from day 3 ↓ Endotoxin at 21 d, ↑ LBP,↓Occludin,↓ Tight junction protein, ↓Jam3, ↓Clnd2Serum:↓ IL‐10Stabilizes α and β diversity (↑abundance of Ralstonia mannitolilytica, Halobacterium hubeiense, and Butyricimonas)[163]OVX female C57BL/6 J mice1 × 108-9CFU/d by GV6 weeksVertebrae: ↑BTVTV, ↑Tb.Th., ↓Tb.S., BMD Femur:↑ BV/TV,↑ Tb.Th., ↓Tb.Sp., ↑ Total pm, ↑ Ct.Th., ↑ Metaphysis BMD and ↑Diaphysis BMD. Tibia: ↑BV/TV, ↑Tb.Th., ↓Tb.Sp., ↑ T. area, ↑ Ct.Th., ↑Metaphysis BMD and↑ Diaphysis BMD↑Max. power, ↑energy, ↑StiffnessBM: ↑Breg, ↑Treg, ↓TH17Spleen: ↑Breg, ↑Treg, ↓TH17.Serum: ↓IL-17, ↓IL-6, ↓TNFα,↑IFNy, ↑ IL-10 [15]Male C57BL/6J mice,Retinol induced osteoporosis1 × 109 CFU/d by GV,3 weeksFemur:↑ Bone BMD, ↑ Tb.BMD, ↑Conn.D., ↓Tb.Sp., ↑Tb.Th.,↑Tb.N., and ↑BV/TVSerum: ↓ALP,↓ Ca, ↑ OCN Treatment did not affect Pielou evenness, Shannon, or Simpson diversitySerum metabolites: ↑ 1,25(OH)2D level[167]OVX female Balb/c mice5×/week1 × 109 CFU/dStarting 2-weeks before OVX4 weeks treatment + 4 weeks washoutFemur: Probiotic did not significantly improve BV/TV or BMD compared to OVX alone at 4 weeks. Only an effect on BMD at 8 weeks Purified exopolysaccharide produced by the probiotic prevented osteoclast formation under both normal and a TNFα-mediated proinflammatory environments in vitro.[169] B. animalissubsp. lactis PreclinicalOVX female C57BL/6J mice5×/week1 × 109 CFU/d by GV10 weeks Femur: ↑BMD, ↑BV/TV, ↑Tb.N., ↓Tb.Sp., ↓SMI,Serum: ↓CTX Colon: ↑Occludin, ↑ZO-1, ↑Claudin 2, ↑Muc2, ↓ IL-17A, ↓TNF-α, ↓IL-6, ↓IL-1β and ↓RANKL Serum: ↓ endotoxin↑ Firmicutes/Bacteroidetes, Prevotellaceae_UCG-001, Bifidobacterium, Muribaculum, Ruminococcus, Parabacteroides, Parasutterella, Colidextribacter, and norank_f__UCG-010 norank_f__norank_o__Clostridia_vadinBB60_group, Colidextribacter and Parabacteroides[170] PreclinicalMale adult C57BL/6J miceDSS-induced gut dysfunction model and wide-spectrum antibiotics -treated mice model7×/week4 × 109CFU/d by GV2 weeksFemur:↑ BMD,↑ BV/TV, ↑Tb.Th., ↑Tb.N.Colon: ↑ length, ↓FITC-dextran, ↓Histological score,↑ Muscus,↑ ZO-1, ↑Occudin and ↑Claudin, ↓ TNF-α, ↓IL-6 and ↓ IL-1β↑Diversity. At Phylum level: ↑Firmicutes, ↑Actinobacteria, and ↑Proteobacteria, ↓ Bacteroidetes. At genus level: ↑ Bifidobacterium, ↑Faecalibaculum, and ↑Clostridium, ↓ Prevotella Metabolites:↑ Butanoate metabolism, ↑ butyrate in serum, muscle, and bone[166]ClinicalRCT, n = 40 women(55–56yo)1.5 × 1010 CFU/d + Ca and calcitriol7×/week12 weeksNo significant changes in BMD after 3 months.Serum: ↓PTH and ↓procalcitonin, ↓ALP No significant intra-group or inter-group differences were observed in the α-diversity. However, treatment affected the gut microbial interactive correlation network, particularly the SCFA-producing bacteria.Serum: ↑ Vitamin D3.Feces: ↑ genes encoding some carbohydrate metabolism pathways and choline‒phosphate cytidylyltransferase.[16]OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: Gavage, BM: bone marrow, PP: peyer patch, LN: Lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular space. BMC: bone mineral content, SMI: structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: procollagen type 1 N‑terminal propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. Probiotic from the genus Bifidobacterium that have been investigated in the gut–bone axis. OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: Gavage, BM: bone marrow, PP: peyer patch, LN: Lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular space. BMC: bone mineral content, SMI: structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: procollagen type 1 N‑terminal propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. Bacilliales Bacillus belong to the Bacillota phylum, and they are generally characterized as rod-shaped, Gram-positive, spore-forming organisms that are either obligate aerobes or facultative anaerobes. Some types of Bacillus species are harmful to humans and plants; however, B. subtilis, B. coagulans, and B. licheniformis have been generally recognized as safe by the U.S. Food and Drug Administration. 171 In a mouse study, animals treated with B. coagulans were protected from OVX-induced bone loss, showing improvements in femur BV/TV, BMD, Tb.Th, and Tb.Sp compared with the vehicle group. Furthermore, treatment also showed increased colon length, a sign of reduced colonic inflammation. Immune cell phenotyping in the bone marrow revealed that treated animals exhibited a reduced inflammatory environment, characterized by decreased Th17 cells and increased Treg cells. Notably, in this study, fecal levels of SCFAs significantly increased following supplementation, exceeding even the levels observed in the SHAM group, 40 demonstrating that this intervention may shift microbiota composition and promote certain genera, ultimately fostering SCFA production. The genus Bacillus is considered one of the most important bacterial sources of proteases, known for their exceptional properties, including strong stability under extreme temperature and pH conditions. 172 Several of these proteases also exhibit antimicrobial activity by targeting bacterial cell walls and membranes. Notable examples of proteases with antimicrobial activity include bacitracin A, bacilysin, chlorotetaine, kanosamine, and bubtilin, which primarily target Gram-positive bacteria. For example, metabolites released by B. subtilis, such as surfactin, fengycin, and iturin can inhibit the growth of Staphylococcus aureus with up to 97% efficacy and prevent chronic osteomyelitis following orthopedic surgery. 173 This indicates that metabolites released by B. subtilis may exert antimicrobial activities that confer them competitive advantage (Table 4). Table 4.Probiotic from the genus Bacillus that have been investigated in the gut–bone axis.BacteriaEvidenceAnimal modelpopulation interventionEffect on boneGut and inflammationMicrobiota and metabolism Ref B. coagulans PreclinicalOVX female C57BL/6 mice1 × 109 CFU/d by GV8.5 weeksFemur: ↑BMD, ↑BV/TV, ↑Tb.Th., ↓Tb.Sp.,Intestine: ↑colon length, ↓ FITC DextranSpeen, Peyer Patch and BM: ↑Breg, ↑Treg,↓ TH17Serum:↓ IL-6,↓TNFa, ↓IL17, ↑IL10,↑ IFNYFeces: ↑acetate, ↑propionate, and ↑Butyrate[40] B. clausii PreclinicalOVX female BALB/c mice2 × 108 CFU/d by GV7×/week6 weeksVertebrae: ↑BMD, ↑BV/TV,↑Tb.Th., ↑Tb.N., ↑Conn.D, ↓Tb.Sp., ↓Tb.Pf. Femur Tb: ↑BMD, ↑BV/TV,↑Tb.Th., ↑Tb.N.,↑ Conn.D, ↓Tb.Sp., ↓Tb.Pf. Tibia Tb:↑BMD,↑ BV/TV,↑Tb.Th., ↑Tb.N., ↑Conn.D, ↓Tb.Sp, ↓Tb.Pf. Femur Ct: ↑BMD,↑Tt.Ar, ↑T.pm, ↑Ct.Ar,↑B.Pm,↑ Ct.Th, ↑J.Tibia Ct: ↑BMD,↑Tt.Ar,↑ T.pm,↑ Ct.Ar, ↑b.Pm,↑ Ct.Th,↑ J.BM:↑CD4 Treg,↑ CD8 Treg, ↓CD4-TH17Spleen:↑CD4 Treg,↑ CD8 Treg,↓CD4-TH17Serum: ↓IL-6,↑IL10, ↓IL17,↓TNFα,↓ IFNy [174] B. subtilis ClinicalRCT N = 76 healthy women(52–61yo)Spore-containing tablets3×/d 7×/week3.4 × 109 CFU/tablet24 weeks↑ Total hip BMD.Urine: ↓ CTX at 12 weeks, ↓TRACP-5b (trend) ↑ Bifidobacterium, ↓ Fusobacterium [175]OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: Gavage, BM: bone marrow, PP: Peyer's patches; LN: Lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular Space. BMC: bone mineral content, SMI: Structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: procollagen type 1 N‑terminal propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. Probiotic from the genus Bacillus that have been investigated in the gut–bone axis. OVX: ovariectomy, RCT: randomized control trial, DW: drinking water, GV: Gavage, BM: bone marrow, PP: Peyer's patches; LN: Lymph node. BV/TV: Bone volume per tissue volume, Tb.Th.: trabecular thickness, Tb.N.: trabecular number. Tb.Sp.: trabecular Space. BMC: bone mineral content, SMI: Structure model index. Ct.Th.: Cortical thickness, Tb.Pf: trabecular pattern factor, Total pm: total perimeter, J: polar moment of inertia, BFR: bone formation rate, Ct.V: cortical volume, ObS/BS: osteoblast surface/bone surface, OcS/BS: osteoclast surface/bone surface, ES/BS: eroded surface/bone surface/bone surface, OS/BS: osteoid surface/bone surface, and OV/BV: osteoid volume/bone volume, CTX: C‑terminal telopeptide of type I collagen, P1NP: procollagen type 1 N‑terminal propeptide, TRAP5: Tartrate‑Resistant Acid Phosphatase 5, RANKL: receptor activator of nuclear factor‑κB ligand, ICR: Institute for Cancer Research, OPG: osteoprotegerin, LBP: lipopolysaccharide‑binding protein, FICT dextran: fluorescein isothiocyanate–labeled 4.4-kDa dextran. Research on Bacillus species as modulators of bone health is still in its early stages, with only one randomized controlled trial conducted to date. This trial, conducted in 76 healthy postmenopausal Japanese women, examined the effects of Bacillus subtilis spore supplementation on BMD over 24 weeks. Those participants receiving B. subtilis exhibited a significant increase in total hip BMD compared with the placebo group. In this study, modulation of the gut microbiota was proposed as a potential mechanism influencing bone health, as treatment increased the relative abundance of Bifidobacterium whereas Fusobacterium was decreased. 175 Although this study did not examine metabolites, other research suggests that Bacillus-derived metabolites may play a key role in bone health. For example, fermentation of soybean by Bacillus subtilis results in high amounts of vitamin K2. 176 Prebiotics According to the International Scientific Association for Probiotics and Prebiotics (ISAPP), prebiotics are substrates that are selectively utilized by the host microbiota, providing a health benefit. Currently established prebiotics are predominantly carbohydrate-based; however, other substances such as polyphenols, some peptides, and polyunsaturated fatty acids (PUFA) may also meet the ISAPP definition. 177 Few observational studies have investigated the association between total dietary fiber intake and BMD. 178-181 However, it is important to note that not all dietary fibers have prebiotic properties. Furthermore, a limitation of the current literature is that studies integrating fiber intake, microbial community composition, and bone health outcomes are scarce. In an observational study including 224,630 participants from the UK-Biobank, a higher dietary fiber intake was significantly associated with higher heel BMD. Strikingly, this association was more pronounced among participants with lower genetically predicted propionate production, suggesting a potential role for gut microbiota-related mechanisms, including dysbiosis, in mediating this relationship. Additional studies have investigated whether specific foods rich in fibers can influence bone metabolism and bone health. 182-184 For example, a recent 12-month RCT showed that daily consumption of dried prunes (fiber content 7g/100g) preserved cortical volumetric BMD and estimated bone strength at the tibia. 185 These results are in line with previous investigations reporting a preventive effect on total hip BMD loss. 186 Several mechanisms have been proposed to explain the beneficial bone effects of prune consumption, including the prebiotic effects of fibers through selective fermentation and SCFA production, the antioxidant and anti-inflammatory properties of polyphenols, and the provision of micronutrients relevant to bone health such as potassium and vitamin K. 187 Given that diets and foods rich in fibers provide a heterogeneous mix of fermentable substrates, of which only specific compounds exhibit prebiotic activity, there has been growing interest in prebiotic supplementation as a more standardized and targeted approach to influence the gut‒bone axis. Several studies have investigated the effects of specific prebiotic supplementation on mineral content, BMD, and bone turnover markers. Evidence from multiple adult studies, including trials in older adults, indicates that prebiotic supplementation (e.g. inulin and oligofructose, soluble corn fibers, lactulose, galacto-oligosaccharides and fructo-oligosaccharides) effectively enhances calcium absorption. 25 , 188-192 The magnitude of treatment effects appears to be influenced by both dosage and the specific type of prebiotic administered. 193 Two RCTs reported no effects on BMD, likely due to the short duration of treatment. 194 In another RCT, 24 months of supplementation with calcium–fructo-oligosaccharide (CaFOS) reduced BMD loss at the spine compared with Ca-only or placebo control among osteopenic women. Notably, CaFOS group showed decreased serum CTX-1 at 12 months and reduced serum osteocalcin at 24 months compared with the control group, indicating early signs of lower bone turnover. 195 Overall, the effects of dietary fibers on bone are largely mediated by the production of SCFAs. SCFAs not only enhance epithelial barrier integrity by increasing cell density and crypt depth but also lower luminal pH, thereby limiting the growth of pathogenic bacteria, allowing mineral release from fibers and improving mineral solubility and absorption. 196 Other dietary components may also exert prebiotic effects. Resveratrol is a polyphenol present in grapes, berries, and red wine. Several studies have demonstrated its low bioavailability, likely due to its rapid biotransformation into glucuronide and sulfate conjugates. 197 , 198 Once in the colon, these conjugated forms are further metabolized by the gut microbiota, while simultaneously modulating the composition of the microbial community. 199 , 200 A substantial amount of in vitro and in vivo evidence supports the beneficial effects of resveratrol on bone. 201 However, to date, resveratrol efficacy on BMD and serum bone biomarkers has not been demonstrated in human RCTs. 202 Synbiotics The term synbiotic refers to the synergistic mixture comprising live microorganisms and substrate(s) selectively utilized by the host microorganisms that confer a one or more health benefits to the host. 203 Synbiotic approaches still remain a developing area of research in the bone field, yet recent studies have started to shed light on this topic. In a preclinical study, a blend of oligofructose and acacia gum combined with Lactobacillus acidophilus NCC90, reduced calcium loss in lumbar vertebrae along with increases in pH and in bifidobacteria and Bacteroides levels. Interestingly, only the synbiotic, not the probiotic alone affected mineralization and gut ecology. The prebiotic group showed a trend toward calcium retention. 204 In a 12-month RCT, postmenopausal women were supplemented with SBD111, a synbiotic medical food containing prebiotic dietary fibers and microbial strains isolated from fresh fruits and vegetables. 205 In the full cohort, supplementation did not significantly reduce bone loss. However, a subset analysis revealed that SBD111 was associated with reduced bone loss in postmenopausal women with osteopenia and a BMI ≥ 30 kg/m2. Although this study did not evaluate prebiotics and probiotics separately, prior in vitro work showed that acetate levels produced by the individual strains were significantly lower than those achieved with the synbiotic combination. The potential contribution of the prebiotic component was not addressed. 206 Although synbiotic formulations are often examined in supplement form, certain everyday foods can also function as synbiotics when they naturally combine probiotics with prebiotics, as is the case with some fermented dairy products. 191 Fermented dairy products Fermented dairy products (FDPs) or cultured dairy products are produced by fermenting milk by lactic acid bacteria species. Popular options in this category include yogurt, kefir, cream, and cheese, although many variations of these products exist depending on regional traditions, bacteria strains, and type of milk. FDPs have shown promising associations with skeletal outcomes in preclinical studies. Lee et al. demonstrated that milk fermented with Lactobacillus plantarum A41 and Lactobacillus fermentum SRK414 enhanced bone volume, BMD, and serum bone biomarkers in a postmenopausal osteoporosis rat model, compared to non-fermented milk. The effects observed in rats treated with milk fermented by A41 were comparable to those achieved with alendronate. 207 Narva et al. reported a significant increase in BMD and bone mineral content in OVX mice after L. helveticus-fermented milk in comparison to sour milk, skim milk, or water. 208 Together, these preclinical findings support the hypothesis that FDPs may exert more favorable effects on bone parameters than nonfermented milk. In humans, the health effects of dairy product consumption have been extensively investigated. A systematic review and meta-analysis indicated that consumption of yogurt and cheese was associated with lower risk of hip fracture in cohort studies. 209 In another study, higher yogurt consumption was associated with reduced hip fracture risk in postmenopausal women, while no associations were found with cheese. 210 Lastly, the most recent analysis, which included only observational studies, found no significant association between yogurt consumption and hip fracture risk. Although a positive effect on femoral BMD was observed, it was considered as minimal and of limited clinical relevance. 211 These inconsistencies could be explained by differences in study populations, methods, and the types of FDPs consumed, which vary in both nutrients and microbial content. To date, according to current dietary guidelines for rheumatology, adults are advised to consume two servings of dairy products per day, with FDPs being preferentially recommended. 212 , 213 Despite the fact that FDP are an excellent source of bone-supporting nutrients, their positive impact on bone health cannot be fully explained by their macro- and micronutrient content alone. Fermentation induces structural and biochemical modifications within the dairy matrix. The proteolytic activity of fermenting bacteria leads to the hydrolysis of caseins, resulting in the release of bioactive peptides and the release of calcium previously entrapped within casein micelles. Bacterial metabolism of lactose generates organic acids, including SCFA, which may lower the pH of the matrix and enhance mineral solubility, thereby facilitating calcium absorption. Furthermore, FDPs like yogurt contain live bacteria (about 1.5 billion CFUs per serving) that may contribute to microbial composition and metabolomic changes. FDPs consumption has been associated with changes in the gut microbiota and metabolome, suggesting that FDPs may contribute to bone health via the gut microbiota. 214 , 215 Using an ex vivo human colonic fermenter, the addition of FDPs significantly increased levels of butyrate, propionate, and total SCFAs, with butyrate showing the most substantial increase. 216 A recent study demonstrated a trend towards increased fecal butyrate concentrations in patients with atopic dermatitis following the consumption of yogurt containing Bifidobacterium animalis subsp. lactis LKM512 for four weeks. 3 In an RCT, milk kefir was shown to modulate gut microbiota composition, with reported increases in lactate-producing and SCFAs–associated bacterial taxa. 217 , 218 Lastly, an intervention study using semi-hard cow's cheese reported alterations in urinary choline and TMAO levels, along with changes in stool concentrations of propionate and butyrate in comparison to controls. This raises an important question: Are the health benefits of FDPs mainly due to bioactive compounds formed during fermentation, or to the live probiotics themselves? A 16-week intervention study compared the effects of milk, yogurt, heat-treated yogurt, and acidified milk on the fecal metabolome. Greater metabolic changes were observed with heat-treated yogurt and acidified milk, suggesting that the metabolic effects may be mainly driven by fermentation-derived compounds rather than by probiotics. 219 This hypothesis remains debated as some suggest that fermentation-derived metabolites reach the colon where they exert their benefits, while others argue that metabolites may degrade before absorption. Another view, not mutually exclusive, is that the dairy matrix could act as a carrier for probiotics, buffering stomach acid and providing nutrients that enhance their survival. Beyond FDPs, other fermented foods have also been investigated for their potential benefits for bone health. Natto is a fermented soybean, rich in Vitamin K2 (MK7), which is popularly consumed in Japan. A systematic analysis of prospective cohorts and RCT demonstrated that natto is associated with higher levels of serum MK-7, osteocalcin, and modestly greater BMD. However, most studies were conducted in Japan, and the certainty of evidence was low, making it difficult to draw conclusions. Therefore, larger and well-designed studies in more diverse populations are needed to clarify the effect of natto on bone health. 220 For other fermented foods (fermented vegetables, cereals/legumes, and beverages), evidence remains limited to mechanistic pathways, with little to no direct data on BMD or calcium balance. Dietary patterns Emerging evidence suggests that holistic approaches, such as the focus on dietary patterns to modulate the gut microbiome, are promising and pragmatic for promoting health, including skeletal health. 193 Such approaches offer several advantages: they are generally cost-effective, can promote dietary adequacy and diversity, and facilitate implementation. Notably, the study of dietary patterns, rather than isolated nutrients or food components, aligns with the growing recognition of food matrices, whereby whole foods exert biological effects beyond the sum of their individual components, influencing digestion, microbial metabolism, and nutrient bioavailability. 221 This is particularly pertinent for the gut-microbiota-derived metabolites, as the structural complexity of foods and nutrient interactions can influence microbial fermentation and metabolic outputs. The Mediterranean and other plant-based dietary patterns (e.g. dietary approaches to stop hypertension [DASH] diet and adherence to the American dietary guidelines), characterized by high intakes of whole grains, legumes, fruits, vegetables, and nuts, have been consistently associated with favorable skeletal health outcomes. 222-225 These dietary patterns are rich in fiber, monounsaturated fatty acids, and bioactive compounds such as polyphenols and antioxidants, all of which exhibit potential prebiotic and immunomodulatory properties, while they may also include live microorganisms through the consumption of fermented foods. In line with these compositional characteristics, adherence to the Mediterranean diet and other plant-based diets has been associated with increased abundance of beneficial microbial taxa, enhanced production of SCFAs, reduced gut inflammation, and improved gut barrier integrity. 225-228 However, these effects have rarely been demonstrated in the same studies reporting skeletal health benefits, 225 highlighting a gap in integrative approaches. In contrast, unhealthy or Western-type dietary patterns, typically rich in ultra-processed food, saturated fat, and refined carbohydrates have been negatively associated with BMD 229 and adverse alterations in the gut microbiome and metabolome, including reduced diversity, lower SCFA production, and increased pro-inflammatory and endotoxin-related metabolites. 230 Collectively, these findings support holistic dietary strategies that target microbiome-derived metabolites as potential mediators of bone health and underpin the need for future well-designed prospective and interventional studies that simultaneously assess dietary patterns, gut-microbiome-associated biomarkers and bone health outcomes. Conclusions Accumulating evidence supports the existence of communication between the gut and the bone, with microbial metabolites acting as messengers in this cross-talk. Several of these microbially derived metabolites, mainly butyrate, have been shown to directly and indirectly affect key determinants of bone strength in preclinical studies. However, the relevance of these microbial metabolite effects in humans remain uncertain because of the supraphysiological concentrations used in preclinical experiments and the lack of randomized clinical studies. Whether such levels can be achieved in circulation after targeted supplementation with microbiota-based interventions such as postbiotics, probiotics, prebiotics, or synbiotics is still debatable. Probiotics have shown promising positive effects on BMD and trabecular bone structure in preclinical studies. Mechanistically, probiotics have been shown to modulate microbial community composition, although the exact metabolic mechanisms by which probiotics interact with the broader microbiota remain poorly understood. Current evidence suggests that their major metabolic contribution may involve the production of microbial metabolites that promote cross-feeding interactions, particularly by amplifying butyrate-producing species. However, other metabolite-producing taxa may be involved. Clinical proof of the efficacy of probiotics on bone shows only a modest reduction in BMD loss in postmenopausal women, with disparate results among skeletal sites. Nevertheless, no studies have explored their effect on bone microstructure and microbial metabolism. In addition to probiotics, prebiotics represent another avenue for modulating the gut–bone axis through their ability to stimulate SCFA-producing microbes and enhance mineral absorption. However, the extent to which these nutritional approaches exert clinically meaningful skeletal effects remains to be fully elucidated. Future microbiome-based nutritional strategies for bone health, including synbiotics, functional foods such as FDPs may offer a more targeted and physiologically relevant means of modulating the gut–bone axis. Likewise, dietary patterns such as the Mediterranean or other plant-forward diets may offer a holistic, food-based strategy to promote bone health by fostering beneficial microbial communities and metabolite profiles. The successful translation of these strategies will require a deeper understanding of microbial contributions, host–microbiome interactions, strain–substrate specificity, and interindividual variability (including ethnicity, gender).