As stated earlier, osteocytes came from embedded osteoblasts within the lacuna-canalicular network. Osteocytes generate signals that are essential for controlling osteoclast formation and regulating phosphate metabolism, including proteins such as DMP1, SOST, and FGF23. Additionally, they serve as a significant source of RANKL (Wu et al., 2025). All these factors help osteocytes function as central regulators of bone remodelling (Figure 1).

The processes of mechanotransduction play a critical role in bone remodelling, which is regulated by osteocytes. Notably, the activation of Wnt/β-catenin signalling is partly mediated through Piezo one and integrin-linked pathways. This activation results in the suppression of SOST, thereby establishing a critical nexus between mechanical loading and the processes of bone formation and resorption (Wu et al., 2025). Furthermore, Wnt signalling in osteocytes increases the formation of osteoclasts by increasing RANKL expression through interactions with transforming growth factor-β (TGF-β) and Smad4 at the promoter region of RANKL (Ye et al., 2025).

Disruption of osteocyte function contributes to pathological remodelling. Processes such as apoptosis, ferroptosis or pyroptosis can disrupt the lacuna-canalicular network. This will lead to impairment of remodelling and repair capacity (Wu et al., 2025). Various osteocyte-targeted strategies show significant potential in restoring osteocyte function. These include anticytokine therapy, gene editing, stem cell interventions, and multi-omics approaches (Wu et al., 2025; Liu et al., 2025).

Oxidative stress is highly recognised as the central drive of skeletal deterioration. Elevated reactive oxygen species (ROS) will impair osteocyte homeostasis. Oxidative stress serves as a central mechanism in response to various insults, including glucocorticoids and homocysteine. These sources impaired osteocyte biology and activated maladaptive bone remodelling. These exposures upregulate the NOX family of NADPH oxidases, driving the cytokine expression and apoptosis in osteocyte-like cells (Notsu et al., 2019; Zhang et al., 2022). Mitochondrial dysfunction due to ageing or injury-related causes will also contribute to the upregulation of ROS, resulting from impaired mitophagy (Ardura et al., 2020).

The resulting reactive oxygen species activate a network of stress-responsive signalling pathways that reshape osteocyte survival, communication, and remodelling output. ROS stimulate the NLRP3 inflammasome, promoting caspase-1 activation, IL-1β maturation, and pyroptotic cell death (Kar et al., 2019). Activation of NF-κB and MAPK pathways leads to elevations in RANKL expression and the production of pro-resorptive cytokines (Marcucci et al., 2023). Oxidative stress also inhibits Wnt/β-catenin signalling by diverting β-catenin towards FOXO transcription factors, suppressing anabolic bone formation (Marcucci et al., 2023).

Similar to the classic vitamin E, tocopherols (TPs), TTs can be categorised into four subtypes, such as alpha-, beta-, gamma-, and delta-homologues, differing by arrangement of the methyl group on the chromanol ring (Chin et al., 2013). TT and TF homologues exist in a mixture in natural sources (Aggarwal et al., 2010).

TTs exert both oxidative and non-oxidative actions on the skeletal system to achieve their anti-osteoporotic effects. TTs stimulate bone growth by facilitating the differentiation of osteoblasts and protecting these cells from oxidative stress (Abd Manan et al., 2012; Casati et al., 2020). It upregulates osteoblastogenic transcription factors, such as Runt-related transcription factor 2 (Runx2) and Osterix, thereby promoting the differentiation of osteoblasts (Wan Hasan et al., 2018). This action is potentially modulated by 3-hydroxy-3-methylglutaryl-coenzyme A reductase and GTPase activities (Wan Hasan et al., 2020). TTs also stimulate the Wnt signalling pathway to achieve pro-osteoblastogenesis effects (Xu et al., 2022). The anti-osteoclastogenic effects of TTs are mediated by the RANKL/OPG pathway and its anti-inflammatory effects. TT supplementation has been shown to modify the RANKL/OPG ratio and lower inflammation in the bone, favouring an anti-osteoclastogenesis milieu (Mohamad et al., 2021; Wong et al., 2018b).

There are some preliminary studies on the effects of TT on osteocytes. Delta-TT protects osteocytes (MLO-Y4) from oxidative stress induced by tert-butyl hydroperoxide, a commonly used in vitro inducer of oxidative damage (Casati et al., 2020). Its cytoprotective properties combat oxidative stress by activating signalling pathways, such as PI3K/Akt and Nrf2. Through PI3K/Akt activation, delta-TT enhances cell survival signalling and promotes the phosphorylation of downstream targets that inhibit apoptosis (Casati et al., 2020). Simultaneously, through Nrf2 activation, it induces the expression of antioxidant defence genes, leading to increased production of protective enzymes. These coordinated actions result in a reduction of reactive oxygen species (ROS) levels by improving the glutathione (GSH)/glutathione disulfide (GSSG) ratio. GSH is a primary cellular antioxidant molecule, and maintaining a high GSH/GSSG ratio indicates a healthy cellular redox state (Vázquez-Meza et al., 2023).

However, in this study, the redox signalling was established in the osteoblast cell line rather than the osteocyte cell line. Although osteoblasts and osteocytes originate from the same lineage, they have distinct physiological roles and metabolic demands (Karthik and Guntur, 2021; Shen et al., 2022). Therefore, direct confirmation of these signalling mechanisms specifically in osteocytes would strengthen the conclusions about delta-TT’s protective effects in these cells.

The osteoporosis-reversing effects of emulsified and non-emulsified palm TT mixtures were investigated in a unique experimental design using ovariectomised rats (Mohamad et al., 2021). This study differed significantly from previous studies using TTs by employing a therapeutic rather than preventive approach, in which supplementation was initiated 2 months post-ovariectomy in a group of older rats (12 months old by the end of treatment). This experimental design is clinically relevant because it mimics the real-world scenario where osteoporosis treatment typically begins after significant bone loss has already occurred, rather than as a preventive measure.

The results showed no significant changes in the number of osteocytes and the number of empty lacunae of trabecular bone after a 2-month treatment (Mohamad et al., 2021). The lack of change in osteocyte numbers and empty lacunae suggests that the relatively short treatment duration (2 months) in rats with established, advanced bone loss may not have been sufficient to demonstrate recovery of the osteocyte population. Osteocyte turnover is a slow process; thus, the restoration of osteocyte networks takes considerable time (Lerebours and Buenzli, 2016). Empty lacunae represent sites of osteocyte death, and their persistence indicates ongoing or incompletely reversed bone deterioration.

Despite the lack of changes in osteocyte parameters, both emulsified and non-emulsified palm TT formulations significantly reduced skeletal SOST protein expression levels (Mohamad et al., 2021). This reduction in SOST expression is mechanistically important because a lower level would allow activation of Wnt/beta-catenin signalling in osteoblasts, potentially promoting bone formation (Delgado-Calle et al., 2017). This molecular change could represent an early adaptive response that precedes observable changes in bone structure and cell populations. This observation suggests that TTs can modulate osteocyte gene expression even in established osteoporosis, which may translate into improved bone formation with longer treatment duration.

The observation that both emulsified and non-emulsified palm TT produced similar effects on SOST expression suggests that the bioactive TT compounds themselves, rather than the delivery vehicle, are responsible for this molecular effect. The reduction in SOST represents a favourable shift in the bone remodelling environment that could potentially lead to increased bone formation if treatment were continued for a longer period.

In a subsequent study using a preventive model, ovariectomised rats were supplemented with emulsified and non-emulsified palm TT mixtures for 10 weeks, with treatment initiated 1 week after ovariectomy (Ekeuku et al., 2025). This preventive approach differs from the therapeutic approach described in the previous study, as treatment begins before substantial bone loss occurs, allowing assessment of TTs’ ability to prevent rather than reverse osteoporosis.

The results showed a significantly higher number of trabecular and cortical osteocytes than in both the palm TT-treated groups (Ekeuku et al., 2025). The preservation of osteocyte numbers is significant because osteocyte loss is a hallmark of postmenopausal osteoporosis, where oestrogen deficiency increases osteocyte apoptosis (McNamara, 2021).

A significantly lower Dmp1 gene expression was noted in the TT-treated group, but no significant changes in Sost and Dkk1 were observed (Ekeuku et al., 2025). The interpretation of reduced DMP1 expression is complex. The lower expression might indicate a more mature, stable osteocyte population, as DMP1 is highly expressed during osteocyte differentiation but may decrease in fully mature osteocytes (Kamiya and Takagi, 2004). The lack of changes in SOST and DKK1 expression in this study contrasts with the previous therapeutic study that showed reduced SOST expression. These observations, combined with preserved osteocyte numbers, suggest that in the preventive model, TTs primarily work by protecting osteocytes from oestrogen deficiency-induced apoptosis rather than by modulating Wnt inhibitor expression.

Metabolic syndrome has been shown to exert degenerative effects on bone health, particularly on the trabecular structures, by elevating inflammation (Wong et al., 2018a). In a study using rats with metabolic syndrome induced by high-fat high-carbohydrate diet, palm and annatto TT supplementation reduced the expression of several key osteocyte-derived factors, namely, SOST, DKK1, FGF23, and RANKL (Wong et al., 2019a).

The concurrent reduction in both SOST and DKK1 represents a coordinated decrease in Wnt pathway inhibition, which would synergistically enhance osteoblast differentiation and activity (Hu et al., 2024). FGF23 is often elevated in metabolic syndrome and obesity (Kawai, 2016). Its reduction suggests improved osteocyte metabolic health and reduced inflammatory stress. RANKL is the primary regulator of osteoclastogenesis (Weitzmann, 2013). Its reduction indicates a shift toward an anti-resorptive profile, with decreased osteoclast formation and activity. This would be expected to reduce bone resorption and help preserve bone mass.

These changes collectively enhance osteocyte-mediated signalling and shift towards an anti-resorptive profile, with reduced bone resorption (via decreased RANKL) and potentially increased bone formation (via decreased SOST and DKK1).

A subsequent study using the same metabolic syndrome model found decreased cortical osteocyte numbers, skeletal DMP1 and PHEX levels and increased empty lacunae numbers in rats with metabolic syndrome (Wong et al., 2022). These findings suggest that metabolic syndrome leads to osteocyte death and dysfunction, characterised by impaired expression of proteins crucial for bone mineralisation and phosphate metabolism.

Only annatto TT reversed the decreased empty lacunae, but palm TT did not (Wong et al., 2022). Despite these changes, DMP1 and PHEX remained unchanged with annatto or palm TT treatments (Wong et al., 2022). The differential effects between annatto and palm TTs on empty lacunae but not on DMP1/PHEX expression suggest that annatto TT may have superior antioxidant or anti-apoptotic properties that better protect osteocytes from metabolic syndrome-induced death. The composition of TT isomers differs between palm (rich in γ and α-TT) and annatto (predominantly δ-TT), which may account for functional differences. Additionally, preventing osteocyte death (by reducing empty lacunae) may not immediately restore DMP1 and PHEX expression, which might require a longer treatment duration or additional interventions.

Collectively, these studies suggest dual protective actions of TTs on bone health through osteocytes. Firstly, TTs function as potent cytoprotective agents by activating the PI3K/Akt and Nrf2 pathways, thereby preserving the integrity of the osteocyte lacunocanalicular network (Casati et al., 2020). This is particularly evident in preventive models, where early supplementation forestalls the apoptosis typically induced by oestrogen deficiency (Ekeuku et al., 2025). Secondly, TTs appear to modulate the osteocyte secretome by downregulating key Wnt inhibitors, specifically SOST and DKK1, across both established osteoporosis and metabolic syndrome models (Mohamad et al., 2021; Wong et al., 2019a). By downregulating these inhibitors, TTs facilitate the activation of Wnt/β-catenin signalling, thereby promoting an environment conducive to bone formation. Ultimately, these findings indicate that the primary bone-sparing effect of TTs may not rely solely on maintaining osteocyte numbers, but also on driving the remaining osteocytes toward a pro-anabolic and anti-resorptive profile by reducing RANKL and Wnt antagonists. The findings of these studies are summarised in Table 1.

These limited studies have their shortcomings. Firstly, most in vivo studies also use only one or two doses (60 mg/kg or 100 mg/kg), previously established to prevent bone loss in animal models (Chin, 2024b). A dose-dependent effect on osteocytes could not be established by these studies. Secondly, most studies used young and growing rats as samples. Despite being sexually mature, the skeleton of rats continues to grow until an advanced age (Fukuda and Iida, 2004). Any assault at this age (ovariectomy or diet modification) may produce stunted growth rather than degenerative changes as in osteoporosis. Thus, researchers continue to debate the relevance and representativeness of these models, despite their widespread use. Lastly, some outcomes were only measurable at the end of the experiment (e.g., osteocyte number and skeletal expression of critical markers), preventing the tracking of changes over time. Therefore, missing the window may explain the lack of changes between treatments and the osteoporosis control.