Vitamin D in incidence and severity of frailty

Calciferol, or vitamin D, is a lipophilic vitamin that functions as a steroid hormone. It naturally exists in two main forms: ergocalciferol (vitamin D₂) and cholecalciferol (vitamin D₃). Vitamin D is acquired through the diet, but it is an inactive form that needs to be further metabolised in the liver and then in the kidney to be able to achieve its effects. Ergocalciferol is synthesised by irradiating ergosterol in plants and fungi with ultraviolet (UV)-B light, while cholecalciferol is derived from the photosynthesis of 7-dehydrocholesterol (7-DHC) within the skin in the presence of UV-B light^{(Reference Rebelos, Tentolouris and Jude88,Reference Nair and Maseeh89)} . Cholecalciferol is primarily abundant in fatty fish, such as tuna, salmon, trout, and fish liver oil. It is also present in smaller amounts in egg yolks, dairy products and beef liver^{(Reference Roseland, Phillips, Patterson, Pehrsson, Taylor and Feldman90)}. Many animal-derived foods, such as poultry, pork and beef, contain a notable quantity of 25-hydroxycholecalciferol (25(OH)D₃), also known as calcifediol, which is a hydroxylated form of cholecalciferol (inactive vitamin D)^{(Reference Taylor, Patterson, Roseland, Wise, Merkel and Pehrsson91)}. Furthermore, most dairy products, particularly milk, some brands of orange juice and ready-to-eat cereals, are fortified with cholecalciferol^{(Reference Calvo, Whiting and Barton92)}. The recommended dietary allowance of calciferol for individuals aged 19–70 years old is 15 µg (600 IU), a value that also applies to women during pregnancy and lactation. However, individuals older than 70 years should consume at least 20 µg (800 IU) of calciferol a day to maintain its status^{(Reference Ross, Taylor, Yaktine and Del Valle93)}.

In 2007, Holick reported in his review that 1 billion people worldwide had vitamin D insufficiency^{(Reference Holick94)}, which would represent 15% of the global population. Almost 10 years later, a study on a European population sample indicated that 13% had insufficient serum 25(OH)D levels^{(Reference Cashman, Dowling, Škrabáková, Gonzalez-Gross, Valtueña and De Henauw95)}, although figures worldwide are highly variable depending on the continent^{(Reference Amrein, Scherkl, Hoffmann, Neuwersch-Sommeregger, Köstenberger and Tmava Berisha96)}.

We have compiled in Supplementary Table 2 human population studies conducted between 2014 and 2022, from the PubMed database, that assessed the relationship between vitamin D and frailty. Most studies found that vitamin D insufficiency or deficiency was present in frail people, and some longitudinal studies^{(Reference Vogt, Decke, de Las Heras Gala, Linkohr, Koenig and Ladwig97–Reference van den Berg, Hegeman, van den Brink, Rhebergen, Oude Voshaar and Marijnissen100)} concluded that low serum vitamin D levels could be a pre-existing condition of frailty. Nonetheless, how direct and independent this relation could be was never ascertained, leaving the leading cause of frailty elusive. Few longitudinal studies that observed the association were limited to concluding a potential causality^{(Reference Buta, Choudhury, Xue, Chaves, Bandeen-Roche and Shardell98,Reference Buchebner, Bartosch, Malmgren, McGuigan, Gerdhem and Akesson99,Reference Dai, Yue, Zhang, Wang and Wu101)} . In studies that found a negative association between frailty and 25(OH)D^{(Reference Demircioglu102)}, the odds ratio of frailty was estimated to be between 2 and 3 in people with insufficient or deficient serum 25(OH)D when compared with normal levels^{(Reference Vogt, Decke, de Las Heras Gala, Linkohr, Koenig and Ladwig97,Reference Wang, Wang, Zhan, Tang, Huang and Tan103–Reference Zheng, Xu, Wang, Qiu and Xue110)} .

A longitudinal study found that progression of frailty was no more associated with 25(OH)D levels in people aged 80 years and older^{(Reference Buchebner, Bartosch, Malmgren, McGuigan, Gerdhem and Akesson99)}, which could explain why no association was detected in several cross-sectional analyses using data from populations with mean age above 80 years^{(Reference Krams, Cesari, Guyonnet, Abellan Van Kan, Cantet and Vellas111–Reference Xiong and Xue113)}. A different cross-sectional study found no correlation between frailty and serum 25(OH)D in adults with diabetes and chronic kidney disease (mean age of 70 years old)^{(Reference Adame Perez, Senior, Field, Jindal and Mager114)}. This suggests that any possible link between vitamin D status and frailty risk may be obscured or confused by underlying medical conditions. Ethnicity may also be a confounding factor in the vitamin D-frailty relationship. The cross-sectional analysis of a database of more than 5,000 North American people aged 60 years or older, found that higher odds of frailty were associated with serum 25(OH)D deficiency (<37·5 nmol/L) in ‘whites’ (3·7 odds ratio), and with deficiency and insufficiency (<75 nmol/L) in ‘non-whites’ (4·0 and 2·7 odds ratios)^{(Reference Wilhelm-Leen, Hall, Deboer and Chertow115)}. An important determining factor was sun exposure, with people living in states above the 40° parallel being at higher risk of frailty. Older African American women were also more likely to have lower 25(OH)D and thus be at a higher risk of frailty^{(Reference Buta, Choudhury, Xue, Chaves, Bandeen-Roche and Shardell98)}. However, the authors of the study observed no association when adding cardiometabolic diseases as a confounding factor. Schöttker et al. reported, from a large cohort study involving nearly 10,000 participants, that vitamin D levels were associated with self-rated health and frailty by cross-sectional analysis^{(Reference Schöttker, Saum, Perna, Ordóñez-Mena, Holleczek and Brenner116)}. However, this relationship was not observed in the longitudinal analysis, except for mortality, letting the authors hypothesise that serum 25(OH)D level could be used as a mortality marker.

To palliate improper nutritional status regarding vitamin D, supplements are usually recommended, but it is inconclusive to prevent frailty. The 4,000 IU/d dose had a lower risk of developing frailty compared with 200 IU/d^{(Reference Cai, Wanigatunga, Mitchell, Urbanek, Miller and Juraschek117)}. Bolzetta et al.^{(Reference Bolzetta, Stubbs, Noale, Vaona, Demurtas and Celotto118)} used data from an American dataset of which 3,083 participants (two-thirds of total participants), with a mean age of 62.1 years, had taken a daily dose of vitamin D supplement in the range of 57–600 IU. They were more likely to be women and older, to have a lower BMI (28·3 versus 29·4), a better dietary vitamin D₃ intake (145 versus 134 IU/d), a lower energy intake (1,391 versus 1,466 kcal/d) and to suffer less depression. However, after an 8-year follow-up, 362 participants had become frail, but vitamin D supplementation did not affect the incidence rate of frailty. This finding aligns with the following studies. Interventional studies^{(Reference Cai, Wanigatunga, Mitchell, Urbanek, Miller and Juraschek117,Reference Orkaby, Dushkes, Ward, Djousse, Buring and Lee119)} examined the effect of vitamin D supplementation at 2,000 IU/d and found no association with frailty. In addition, it was found in a Swedish cohort study that a serum concentration above 75 nmol/L did not have a particular beneficial effect on frailty^{(Reference Buchebner, Bartosch, Malmgren, McGuigan, Gerdhem and Akesson99)}.

Vitamin D in cognitive functions and depression

The role of vitamin D extends beyond bone health, with emerging research suggesting its potential influence on mental health and the nervous system. Numerous tissues, including the brain, have been demonstrated to synthesise active forms of vitamin D. Vitamin D binding proteins (DBP) and VDR have been found in the central nervous system (CNS), specifically in regions linked to depression and mood^{(Reference Kesby, Eyles, Burne and McGrath120,Reference Caldwell, Londe, Ochs, Hajdu, Rodewald and Gebhart121)} . Nonetheless, RCT have shown conflicting findings about how vitamin D supplementation affects the intensity of depression. In the 8-week double-blind RCT, 56 participants with mild-to-severe depression, aged 43 years on average, were given either 50,000 IU/2 weeks of vitamin D supplementation or a placebo. The intervention group experienced a substantial decrease in their depression score, whereas the control group did not^{(Reference Kaviani, Nikooyeh, Zand, Yaghmaei and Neyestani122)}. This finding is concomitant with a 12-week double-blind, placebo-controlled, randomised trial on 68 subjects with type 2 diabetes and mild-to-moderate depressive symptoms. After three months of vitamin D supplementation (4,000 IU/d), the intervention group demonstrated a significant decrease in depression score^{(Reference Omidian, Mahmoudi, Abshirini, Eshraghian, Javanbakht and Zarei123)}. Furthermore, a 1-year double-blind randomised clinical trial (24,000 IU of vitamin D₃/month, 60,000 IU of vitamin D₃/month or 24,000 IU of vitamin D₃ plus 300 μg calcifediol) on 200 participants, aged 77–78 years on average, showed that participants who achieved 25(OH)D levels in the highest quartile were associated with the best improvements in mental health at 12-month follow-up^{(Reference Gugger, Marzel, Orav, Willett, Dawson-Hughes and Theiler124)}. However, there was no benefit of vitamin D on depressive symptoms in four previous clinical trials at doses of 800 IU/d^{(Reference Dumville, Miles, Porthouse, Cockayne, Saxon and King125)}, 5,000 IU/d^{(Reference Dean, Bellgrove, Hall, Phan, Eyles and Kvaskoff126)}, 50,000 IU/week^{(Reference Arvold, Odean, Dornfeld, Regal, Arvold and Karwoski127)} and 500,000 IU once annually^{(Reference Sanders, Stuart, Williamson, Jacka, Dodd and Nicholson128)}.

Vitamin D in muscle functions

VDR has been presented in several tissues, including skeletal muscle^{(Reference Bischoff, Borchers, Gudat, Duermueller, Theiler and Stähelin129)}. Therefore, the important role vitamin D plays in preserving healthy muscular function is being increasingly recognised. A prospective cohort study in Australia aimed to screen for aortic aneurysm in men aged 70–88 years, a relationship between vitamin D levels, frailty, and mortality was revealed^{(Reference Wong, McCaul, Yeap, Hankey and Flicker130)}. The study determined that vitamin D levels below 52·9 nmol/L were associated with an odds ratio of 1·56 for being frail at 5 years. Low vitamin D levels were found to be predictive of fatigue (worn out or feeling tired most of the time), resistance (inability to climb a flight of stairs), and ambulation (inability to walk 100 m), as well as of all-cause mortality, without establishing a causal relationship^{(Reference Wong, McCaul, Yeap, Hankey and Flicker130)}. Gutiérrez-Robledo et al. demonstrated an independent association between 25(OH)D levels and weakness^{(Reference Gutiérrez-Robledo, Ávila-Funes, Amieva, Meillon, Acosta and Navarrete-Reyes131)}. In addition, other research studies have highlighted the link between low vitamin D levels and impaired physical performance^{(Reference Vitale, Caruso, Rapisarda, Cianci and Cianci132,Reference Carrazco-Peña, Farías-Moreno, Del Toro-Equihua, Aguilar-Mancilla, Trujillo-Magallón and Solórzano-Rodríguez133)} .

Moreover, the impact of vitamin D supplementation on muscular function has been examined in several RCT. In the 12-week RCT, forty elderly post-menopausal women, aged 70 years on average, were given either 0·5 mg/d of alfacalcidol, a synthetic vitamin D analogue plus 1,500 mg/d of calcium carbonate, or a placebo plus 1,500 mg/d of calcium carbonate. The intervention group effectively improved the quadriceps muscle strength measured by the isokinetic dynamometer device^{(Reference Songpatanasilp, Chailurkit, Nichachotsalid and Chantarasorn134)}. A year-long RCT in 261 elderly women (average age of 77 years) showed that daily intake of 1,000 IU of vitamin D2 plus 1,000 mg of calcium citrate positively impacted hip muscle strength and mobility^{(Reference Zhu, Austin, Devine, Bruce and Prince135)}. Ceglia et al.^{(Reference Ceglia, Niramitmahapanya, Da Silva Morais, Rivas, Harris and Bischoff-Ferrari136)} argued – through an RCT of vitamin D supplementation at 4,000 IU/d in mobility-limited older women with serum 25(OH)D₃ deficiency – that the nuclear VDR concentration in muscle was associated with vitamin D levels. However, Hangelbroek et al.^{(Reference Hangelbroek, Vaes, Boekschoten, Verdijk, Hooiveld and van Loon137)} found that vitamin D₃ supplementation at 400 IU/d (a double-blind RCT) did not affect gene expression in the muscle tissue of ten frail older people after 6 months. Moreover, in a 6-month RCT, 65 healthy men, aged 77 years on average, were given 1,000 IU/d of vitamin D₃ plus 500 mg/d of calcium or a placebo plus 500 mg/d of calcium. The intervention group did not increase muscle strength or improve physical performance^{(Reference Kenny, Biskup, Robbins, Marcella and Burleson138)}. Even when studying the effect of vitamin D supplementation on frailty status, an interventional study^{(Reference Cai, Wanigatunga, Mitchell, Urbanek, Miller and Juraschek117)} analysed the effect of vitamin D₃ in individuals aged 70 years and older who had vitamin insufficiency (<75 nmol/L). Over 2 years, they received a daily dose of 1,000 IU of vitamin D₃ compared with a control group at 200 IU. The optimal dose of 1,000 IU/d had been determined after comparing with subgroups at 2,000 IU/d, which was deemed to lead to a two-fold increased risk of frailty and slowness. Although the 4,000 IU/d dose of vitamin D₃ had a lower risk of developing frailty during follow-up (hazard ratio = 0·22), this might be the result of type 1 error. Therefore, the authors concluded the ineffectiveness of vitamin D₃ supplementation in the prevention of frailty.

Vitamin D in immunity

Vitamin D is known to control several important genes that modulate the immune system^{(Reference Aranow139,Reference Sîrbe, Rednic, Grama and Pop140)} . It primarily affects the metabolism, differentiation, maturation, and reaction of immune cells to cytokines and chemokines. Both the 1-α-hydroxylase and VDR genes are expressed more when toll-like receptor (TLR) binding occurs in monocytes and macrophages, leading to increased transcriptional activity of vitamin D^{(Reference Liu, Stenger, Li, Wenzel, Tan and Krutzik141,Reference Martens, Gysemans, Verstuyf and Mathieu142)} . Vitamin D deficiency can lead to immune system dysfunction, which increases the risk of infections such as tuberculosis^{(Reference Chocano-Bedoya and Ronnenberg143)}, sepsis^{(Reference Cutuli, Ferrando, Cammarota, Franchini, Caroli and Lombardi144)}, coronavirus disease 19^{(Reference Kaya, Pamukçu and Yakar145)}, cancers^{(Reference Seraphin, Rieger, Hewison, Capobianco and Lisse146)}, and autoimmune diseases^{(Reference Sîrbe, Rednic, Grama and Pop140)}. Moreover, vitamin D plays a vital role in the immune response to vaccines such as influenza, measles, rubella, and hepatitis B vaccines^{(Reference Sadarangani, Whitaker and Poland147)}.

In this review, we will focus on immunity in the elderly with/without frailty and vitamin D. High serum levels of inflammatory cytokines and acute phase proteins indicate a heightened inflammatory state in frail older persons, corroborating the idea that frailty is associated with a dysregulated immune system^{(Reference Yao, Li and Leng148)}. A Chinese study with 288 frailty patients with chronic coronary syndrome identified by cross-sectional analysis that frailty was associated with interleukin-6 (IL-6) and 25(OH)D levels in an independent manner^{(Reference Dai, Li, He, Huang and Li149)}. A population-based study conducted in the Netherlands found that there was a correlation between moderately elevated levels of C-reactive protein (CRP) (3–10 µg/mL) and severe vitamin D deficiency (<25 nmol/L) with the development of frailty, whereas no consistent relationships were found between IL-6 and insulin-like growth factor 1 with frailty^{(Reference Puts, Visser, Twisk, Deeg and Lips150)}. A retrospective analysis of 237 older adults (mean age of 86.5 years) admitted to an acute care geriatric unit revealed an inverse relationship between serum 25(OH)D and CRP, a marker of inflammation^{(Reference Boccardi, Lapenna, Gaggi, Garaffa, Croce and Baroni151)}. In addition, 25(OH)D levels were negatively correlated with the Cumulative Illness Rating Scale Severity Index (CIRS-SI) and Comorbidity Index (CIRS-CI), whereas CRP levels were positively correlated with CIRS-SI and CIRS-CI^{(Reference Miller, Paradis, Houck, Mazumdar, Stack and Rifai152)}. RCT are essential for assessing how vitamin D₃ affects immunity in older persons, and some studies have started to investigate this connection. In a 14-week intervention study including eighteen older persons given 6,400 IU/d of vitamin D₃, this supplementation dramatically improved the response to the cutaneous varicella zoster virus antigen challenge^{(Reference Chambers, Vukmanovic-Stejic, Turner, Shih, Trahair and Pollara153)}. In a 3-month RCT, 38 elderly participants, aged 70 years on average, were given 100,000 IU/15 d of vitamin D₃ or a placebo. The findings demonstrated that vitamin D supplementation raised the plasma level of transforming growth factor beta in response to influenza vaccination without enhancing the generation of antibodies^{(Reference Goncalves-Mendes, Talvas, Dualé, Guttmann, Corbin and Marceau154)}. Collectively, vitamin D may play a modulatory role in the inflammatory and immune pathways associated with frailty in older individuals, making it a potential target for interventions.

Meta-analyses of vitamin D research in frailty

Several observational studies have investigated the clinical relationship between vitamin D serum concentration and frailty. Data obtained can be conflicting, mainly owing to divergences in the types of studies, inclusion/exclusion criteria and analysis methods, but also in the definition of vitamin D deficiency itself. Indeed, although it is commonly admitted that serum 25(OH)D levels of 50–75 nmol/L are insufficient, 25–50 nmol/L are deficient and <25 nmol/L are severely deficient, not all studies use the same scale. Some researchers prefer to test the dose–response effect, whereas others choose to use an incremental scale without associating it with an evaluation of the degree of vitamin deficiency. The threshold of 75 nmol/L was initially chosen as it corresponds to the minimal concentration beyond which serum parathyroid hormone is kept at its minimum through intestinal calcium absorption^{(Reference Dawson-Hughes, Heaney, Holick, Lips, Meunier and Vieth155)}. Regarding these discrepancies, meta-analysis proves to be an essential approach to understand the associations between frailty and vitamin D.

The inverse association between vitamin D levels and the risk of frailty in older people is still controversial. Marcos-Pérez et al. concluded that the connection between low 25(OH)D levels and frailty severity, on the basis of their meta-analysis of 26 studies, was an inverse association, especially at the deficient level when comparing frails and pre-frails^{(Reference Marcos-Pérez, Sánchez-Flores, Proietti, Bonassi, Costa and Teixeira156)}. A daily intake of 1,000 IU of vitamin D₃ was also recommended by Ju et al., in their ten-study meta-analysis, which indicated the protective effect of a 25 nmol/L increase in serum^{(Reference Ju, Lee and Kim157)}. This daily dose seemed particularly effective on physical performance^{(Reference Marcos-Pérez, Sánchez-Flores, Proietti, Bonassi, Costa and Teixeira156)}. However, on the basis of a large meta-analysis, Bolland et al. showed that vitamin D supplementation was not able to reduce the risk of outcomes such as hip fracture, cerebrovascular disease and cancer beyond 15%^{(Reference Bolland, Grey, Gamble and Reid158)}.

In the face of the current state of knowledge, it is not yet possible to define the nature of the effect with certainty (although some research results argue for a threshold effect with a characteristic U-shaped curve^{(Reference Buchebner, Bartosch, Malmgren, McGuigan, Gerdhem and Akesson99,Reference Wang, Wang, Zhan, Tang, Huang and Tan103,Reference Ensrud, Ewing, Fredman, Hochberg, Cauley and Hillier159)} ) the risk of frailty increasing above a certain serum concentration. Nevertheless, cross-sectional studies constitute the majority of research on frailty, and their intrinsic methodology prevents drawing a stronger connection with vitamin D. The causality itself between vitamin D and the risk of frailty is still uncertain, as research has not yet been able to point out which is the stronger contributor. Through a disturbance in calcium metabolism, low vitamin D levels can lead to frailty. However, owing to decreased mobility and outdoor activities, frailty itself frequently results in less exposure to sunshine, which inhibits the skin’s ability to synthesise vitamin D^{(Reference Marcos-Pérez, Sánchez-Flores, Proietti, Bonassi, Costa and Teixeira156,Reference Ju, Lee and Kim157)} . Figure 2 shows the possible links existing between vitamin D deficiency and frailty.

Fig. 2 Mechanisms linking vitamin D to frailty. The ageing process can psychologically affect older adults, leading to behaviours that contribute to lower vitamin D levels. These behaviours include spending more time indoors, reducing exposure to sunlight (an essential factor for vitamin D metabolism) and consuming an insufficient and unbalanced diet. Vitamin D deficiency can result in muscle weakening and bone demineralisation, contributing to the physical component of frailty and increasing the risk of fractures. In addition, low vitamin D levels are associated with a high systemic inflammatory potential of pro-inflammatory factors and cytokines, which accelerate ageing and contribute to frailty and cognitive decline. Frail individuals, affected as such, are less likely to engage in physical activities, especially outdoors, perpetuating the cycle of frailty including low gait speed and weight loss.

Vitamins B₉ and B₁₂ in methylation

Vitamin B₉ and B₁₂ play critical roles in homocysteine metabolism and DNA methylation, both closely related to ageing and frailty in older adults. It is essential for the conversion of homocysteine, an amino acid, into methionine. Methionine is then converted into S-adenosylmethionine, serving as a methyl donor to various acceptors. After these methylation reactions, S-adenosylhomocysteine is produced as a by-product, which is then hydrolysed to regenerate homocysteine. Vitamin B₉ provides a methyl group to convert homocysteine to methionine, while vitamin B₁₂ acts as a cofactor to regenerate tetrahydrofolate, an active form of folate, allowing the reaction to proceed. Vitamin B₉ or B₁₂ deficiency impairs this conversion of homocysteine to methionine, leading to elevated homocysteine levels^{(Reference Miller, Nadeau, Smith, Smith and Selhub160,Reference Selhub161)} . Elevated homocysteine levels can lead to endothelial dysfunction, impairing blood vessel function and promoting inflammation. In addition, homocysteine can stimulate the production of reactive oxygen species (ROS), leading to oxidative stress^{(Reference Zou and Banerjee162–Reference Perna, Ingrosso, Lombardi, Acanfora, Satta and Cesare164)}. This dysregulation in homocysteine metabolism can contribute to the pathophysiology of ageing-related diseases and frailty. Guaita et al. conducted a retrospective longitudinal study to explore the influence of biomarkers on frailty incidence in a community-dwelling older population^{(Reference Guaita, Brunelli, Davin, Poloni, Vaccaro and Gagliardi56)}. The results demonstrated that a high homocysteine level is a risk factor for frailty onset in older adults. Furthermore, a cross-sectional study was performed to evaluate the association between inflammatory biomarkers, namely homocysteine and CRP, with frailty, revealing that homocysteine was independently associated with frailty^{(Reference Álvarez-Sánchez, Álvarez-Ríos, Guerrero, García-García, Rodríguez-Mañas and Cruz-Chamorro165)}. Deficiency of vitamin B₉ and elevated homocysteine levels in the blood have been associated with increased risk of cardiovascular disease^{(Reference Ma, Peng, Liu, Huang and Luo166,Reference De Koning, Werstuck, Zhou and Austin167)} , cognitive decline^{(Reference Baroni, Bonetto, Rizzo, Bertola, Caberlotto and Bazzerla48,Reference Kim, Choi, Nam, Kim, Oh and Yang168)} , osteoporosis^{(Reference Herrmann, Peter Schmidt, Umanskaya, Wagner, Taban-Shomal and Widmann169,Reference De Martinis, Sirufo, Nocelli, Fontanella and Ginaldi170)} , depression^{(Reference Kim, Stewart, Kim, Yang, Shin and Yoon171,Reference Permoda-Osip, Dorszewska, Skibinska, Chlopocka-Wozniak and Rybakowski172)} , and other age-related conditions^{(Reference Mattson, Kruman and Duan173,Reference Song, Song and Park174)} .

The dysregulation of DNA methylation can lead to increased expression of pro-inflammatory genes and decreased expression of anti-inflammatory genes, contributing to chronic inflammation, which is a hallmark of frailty^{(Reference Pansarasa, Pistono, Davin, Bordoni, Mimmi and Guaita175)}. Both vitamins B₉ and B₁₂ are cofactors in the one-carbon metabolism, i.e. folate metabolism, the homocysteine re-methylation cycle and the transsulfuration pathway. Bellizzi et al. suggested that global DNA methylation in the elderly is correlated with frailty^{(Reference Bellizzi, D’Aquila, Montesanto, Corsonello, Mari and Mazzei176)}. Collerton et al. also proposed that the acquisition of aberrant DNA methylation is associated with frailty in older people^{(Reference Collerton, Gautrey, van Otterdijk, Davies, Martin-Ruiz and von Zglinicki177)}. This study suggested a potential role for age-related changes in CpG island methylation in the development of frailty. It is currently assumed that the regular intake of micronutrients, particularly vitamin B₉ or vitamin B₁₂, can slow down the gradual hypomethylation observed during the ageing process^{(Reference Choi, Claycombe, Martinez, Friso and Schalinske178,Reference Kuo, Sorond, Chen, Hashmi, Milberg and Lipsitz179)} . However, the relationship between vitamin B₉ status and DNA methylation levels is complex and is likely influenced by vitamin B₉ availability^{(Reference Bae, Ulrich, Bailey, Malysheva, Brown and Maneval180)}. Recently, Michels et al. conducted a randomised intervention study to explore the impact of vitamin B₉ supplementation on the epigenetic profile (DNA methylation) in healthy participants on an unfortified diet^{(Reference Michels and Binder181)}. The results of this study showed that an increase in red blood cell vitamin B₉ had a modest impact on the epigenetic clock predicting chronologic age. Furthermore, both folate and vitamin B₁₂ are associated with genome stability, which contributes to frailty^{(Reference Sanchez-Flores, Marcos-Perez, Lorenzo-Lopez, Maseda, Millan-Calenti and Bonassi182)}. Folic acid is required for the synthesis of deoxythymidine monophosphate from deoxyuridine monophosphate (dUMP). In conditions of folic acid deficiency, dUMP accumulates, leading to the incorporation of uracil into DNA instead of thymine^{(Reference Fenech183)}. There is substantial evidence suggesting that excessive incorporation of uracil into DNA not only leads to point mutations but may also result in the generation of single- and double-stranded DNA breaks, chromosome breakage and micronucleus formation.

Vitamin B₁₂ and immunity

Vitamin B₁₂ plays a vital role in immunomodulation. Research has shown that transcobalamin, a vitamin B₁₂-specific non-glycosylated protein carrier, acts as an inflammatory suppressor by regulating anti-inflammatory cytokines, growth factors and other components within physiological levels^{(Reference Wheatley184,Reference Lee, Wang, Lin and Lin185)} . Vitamin B₁₂ itself serves as a down-regulator of the nuclear factor kappa-light-chain-enhancer of activated B cells, indirectly modulating the expression levels of pro-inflammatory cytokines, growth factors and cell adhesion molecules^{(Reference Veber, Mutti, Tacchini, Gammella, Tredici and Scalabrino186)}. Supplementation of vitamin B₁₂ has been shown to increase CD8⁺ T cell counts and natural killer T-cell activity in vitamin B₁₂-deficient subjects^{(Reference Tamura, Kubota, Murakami, Sawamura, Matsushima and Tamura187)}. Studies have demonstrated that vitamin B₁₂ possesses antioxidant activity by reducing cytosolic and mitochondrial superoxide levels in human aortic endothelial cells supplemented with cyanocobalamin^{(Reference Moreira, Brasch and Yun188)} and suppressing superoxide bursts in retinal ganglion cells of Long–Evans rats administered vitamin B₁₂ ^{(Reference Chan, Almasieh, Catrinescu and Levin189)}.

Regarding monocytes, which can further differentiate into tissue-specific phagocytes such as macrophages or antigen-presenting cells such as dendritic cells, hyperhomocysteinaemia-mediated inflammation triggered by vitamin B₁₂ deficiency activates macrophages^{(Reference Allen, Stabler, Savage and Lindenbaum190)}. This increases the production of pro-inflammatory cytokines, particularly tumour necrosis factor (TNF-α), exacerbating the pro-inflammatory environment. In addition to monocytes, vitamin B₁₂ regulates the production of T cells and the balance of the CD4⁺/CD8⁺ ratio. Therefore, a deficiency not only impairs the CD4⁺/CD8⁺ ratio but also reduces the overall T-cell count. Vitamin B₁₂ deficiency also impacts cytotoxic NK cells owing to impaired haematopoiesis, leading to a decrease in both cell counts and quality, which can be reversed by vitamin B₁₂ supplementation^{(Reference Tamura, Kubota, Murakami, Sawamura, Matsushima and Tamura187,Reference Erkurt, Aydogdu, Dikilitaş, Kuku, Kaya and Bayraktar191)} .

Vitamin B₁₂ in propionate metabolism regulation

Vitamin B₁₂ participates in the mitochondrial tricarboxylic acid (TCA) cycle as a coenzyme, in the form of adenosylcobalamin, for the vitamin B₁₂-dependent enzyme methylmalonyl-CoA mutase (MCM; EC 5.4.99.2). The short-chain fatty acid propionate is first converted to propionyl-CoA, which is then carboxylated to d-methylmalonyl-CoA in a reaction that requires vitamin B₇ (biotin) as a cofactor. Methylmalonyl-CoA is subsequently isomerised to l-methylmalonyl-CoA and then converted by MCM to succinyl-CoA, which enters the TCA cycle for energy production. A deficiency of vitamin B₁₂ impairs MCM activity, leading to the accumulation of methylmalonyl-CoA and its hydrolysed product, methylmalonic acid. This disruption affects the catabolism of cholesterols, fatty acids (specifically odd-chain fatty acids and those derived from propionate), and several amino acids (valine, isoleucine, threonine, and methionine), ultimately resulting in metabolic imbalance and potential mitochondrial dysfunction, characteristic of methylmalonic acidaemia^{(Reference Tejero, Lazure and Gomes192)}.

Vitamin B₁₂ deficiency impairs MCM activity, leading to increased accumulation of the substrate, methylmalonic acid (MMA)^{(Reference Tejero, Lazure and Gomes192)}. MMA is used as a surrogate biomarker for confirming vitamin B₁₂ deficiency, in addition to elevated serum homocysteine and macrocytic anaemia, with serum MMA values exceeding 270 nmol/L considered indicative of vitamin B₁₂ deficiency^{(Reference Rajan, Wallace, Beresford, Brodkin, Allen and Stabler193)}. Moreover, elevated circulatory MMA levels are common among the elderly and are associated with increased cardiovascular and all-cause mortality, independent of vitamin B₁₂ status^{(Reference Wang, Liu, Liu, Tian, Zhang and Cai194)}.

Pathologies of methylmalonic acidaemia are observed across a broad spectrum of biological systems. An in vitro study showed that MMA induces neuronal damage in embryonic rat striatal cell cultures through the inhibition of mitochondrial complex II and the TCA cycle^{(Reference Okun, Hörster, Farkas, Feyh, Hinz and Sauer195)}. In addition to complex II suppression, increased MMA levels also impaired mitochondrial respiration by inhibiting complex I in Wistar rat cerebral cortex homogenates^{(Reference Brusque, Borba Rosa, Schuck, Dalcin, Ribeiro and Silva196)}. Moreover, new-born mice injected with MMA demonstrated neurological deficits, caused by increased ROS, TNF-α, and IL-1β, leading to neuroinflammation and apoptosis^{(Reference Gabbi, Ribeiro, Jessié Martins, Cardoso, Haupental and Rodrigues197)}. Apart from neurological impacts, methylmalonic acidaemia facilitates non-alcoholic steatohepatitis (NASH) through epithelial–mesenchymal transition-mediated hepatic fibrogenesis^{(Reference Zhao, Zhu and Sun198,Reference Gomes, Ilter, Low, Endress, Fernández-García and Rosenzweig199)} . Moreover, independent of vitamin B₁₂ status, methylmalonic acidaemia is associated with an increased risk of hepatofibrosis in NASH subjects^{(Reference Li, Huang, Yang, Zhang, Gao and Han200)}. Skeletal muscles also suffer from increased MMA levels, owing to the high distribution of mitochondria in their tissues^{(Reference De Mario, Gherardi, Rizzuto and Mammucari201)}. Methylmalonic acidaemia-mediated inhibition of mitochondrial complex I, II, and the TCA cycle leads to impaired ATP production, causing reduced muscle power and muscular hypotonia^{(Reference Manoli, Sloan and Venditti202)}. In addition, mitochondrial dysfunction, oxidative stress, and inflammation caused by metabolic acidosis, including MMA acidaemia, can trigger proteolysis through autophagy or the ubiquitin–proteasome system, leading to muscle atrophy and sarcopenia^{(Reference Mitch, Medina, Grieber, May, England and Price203,Reference Yin, Li, Jia, Wang, Liang and Yang204)} .

Vitamin D transport

The main transport mode of vitamin D in blood, the vitamin D binding protein (DBP), is also a factor that could be linked to frailty. Wang et al. determined that DBP concentration was higher in frail participants^{(Reference Wang, Wang, Zhan, Tang, Huang and Tan103)}. More specifically, the first and second quartiles of serum 25(OH)D₃ (<31·5 and 31·5 to <41·8 nmol/L, respectively) were associated with an odds ratio greater than 2·5 compared with the highest quartile (≥56·6 nmol/L). In addition, having serum 25(OH)D₃ in the first quartile (<31·5 nmol/L) along with serum DBP in the fourth quartile (≥6,686 nmol/L) was associated with an odds ratio of 3·18. This suggests that the interaction with DBP could decrease 25(OH)D₃ bioavailability and, therefore, add to the frailty risk. Furthermore, frailty was associated with low cholesterol and triacylglycerols, as well as with high CRP and superoxide dismutase^{(Reference Xiong and Xue113,Reference Polat, Yalcin, Yazihan, Bahsi, Mut Surmeli and Akdas205)} . These findings could indicate a higher inflammatory level in frail older people independently of cholesterol.

Vitamin D cell absorption and signalling

Vitamin D can exert effects on the metabolism of cells at targeted tissues in a genomic or non-genomic way. In the genomic way, the formation of an activation complex between vitamin D–VDR and retinoid X receptor RXR in the cytoplasm leads to nuclear translocation. The complex binds to the vitamin D response element (VDRE), which promotes the expression of vitamin D-regulated genes. VDRE activation has been linked (to name only the most significant and documented ones) to calmodulin, implicated in muscle tone and contraction^{(Reference Li, Mihalcioiu, Li, Zakikhani, Camirand and Kremer206)}, and to insulin-like growth factor 1 (IGF-1), which contributes to muscle cell differentiation and proliferation, as well as type 2 muscle fibre development^{(Reference Barton-Davis, Shoturma, Musaro, Rosenthal and Sweeney207)}. In addition, IGF upregulates CYP27B1, leading to a rise in serum 1,25(OH)₂D levels. The expression of the calcium modulator calbindin-D_9k is also modulated via vitamin D^{(Reference Dardenne, Prud’homme, Arabian, Glorieux and St-Arnaud208)}.

Non-genomic effects originate from the transportation of free 1,25(OH)₂D₃ across the plasma membrane, which is facilitated through membrane VDR^{(Reference Capiati, Benassati and Boland209)} and from that of vitamin D–DBP via low-density lipoprotein receptor-related protein 2 (LRP-2)^{(Reference Abboud, Puglisi, Davies, Rybchyn, Whitehead and Brock210)}. VDR, once associated with 1,25(OH)₂D₃, triggers a cascade of mitogen-activated protein kinase (MAPK)-mediated reactions involved in protein synthesis and mitochondrial functions. In consequence, vitamin D deficiency has been associated with decreased mitochondrial calcium uptake and oxygen consumption^{(Reference Ryan, Craig, Folmes, Wang, Lanza and Schaible211)} and is also associated with ROS cytotoxicity^{(Reference Dzik, Skrobot, Flis, Karnia, Libionka and Kloc212)}. However, the discovery of protein disulphide isomerase family A member 3 (PDIA3), and subsequent research on it, indicate this membrane-bound receptor is the mediator of non-genomic calcium uptake regulation^{(Reference Nemere, Garbi, Hämmerling and Khanal213)}. However, a recent study advocates for both VDR and PDIA3 being implicated in both genomic and non-genomic responses, through different pathways^{(Reference Nowak, Olszewska, Wierzbicka, Gebert, Bartoszewski and Żmijewski214)}. Furthermore, MAPK-signalling inhibition and impairment of the proteasome catalytic activity were observed, along with anaerobic capacity, lean mass, and gait speed loss, which are associated with sarcopenia^{(Reference Sleeman, Aspray, Lawson, Coleman, Duncan and Khoo215)}. In comparison with the similarities observed in chronic kidney disease, it was proposed that with a long-lasting deficiency in vitamin D, LRP-2 expression itself could decrease, with the consequence of less vitamin uptake^{(Reference Dusso and Tokumoto216)}. Taken altogether, these effects produce dysregulation in cellular homeostasis and muscle atrophy.

Target sites

Although VDR has been found in a consequent number of tissues, it is predominant principally in the brain, type 2 fibre skeletal muscle cells and immune cells. VDR is partly responsible for the proliferation of skeletal muscle cells through gene upregulation of the cell cycle and protein production^{(Reference Bass, Nakhuda, Deane, Brook, Wilkinson and Phillips217)}. Allele polymorphism of the VDR gene was studied by Arosio et al.^{(Reference Arosio, Guerini, Costa, Dicitore, Ferri and Mari218)} with particular interest in intronic rs7975232 (C/A). Using blood analysis data and a frailty index based on Searle’s criteria^{(Reference Searle, Mitnitski, Gahbauer, Gill and Rockwood219)} from an Italian cohort study (participants aged ≥60 years), they found that women had higher frailty scores, parathyroid hormone (PTH) and inorganic phosphate levels than men, but lower 25(OH)D₃ amounts. The homozygous and heterozygous mutations were positively associated with frailty and phosphate levels in women, but no association was found between these mutations and vitamin D or PTH in men. The authors discussed that the difference between sexes could be caused by the imbalanced levels of PTH and 25(OH)D₃ responsible for low serum phosphate in women, adding to the observed risk of frailty in women.

With age, the regulation of serum PTH to low levels through 25(OH)D₃ mediation becomes less efficient and would require a greater intake of vitamin D than for younger people^{(Reference Vieth, Ladak and Walfish220)}. Ageing is associated with vitamin D resistance, characterised by decreased VDR abundance and impaired renal calcium reabsorption. This impacts calcium homeostasis and can lead to an imbalance in calcium levels, which is closely related to bone health in ageing. This resistance is linked to a reduction in VDR expression in various tissues, including bone, intestine, and muscle^{(Reference Oudshoorn, van der Cammen, McMurdo, van Leeuwen and Colin221)}. In addition, there is a decrease in the expression of transient receptor potential vanilloid (TRPV) calcium ion channels, specifically TRPV6 in the intestine and TRPV5 in the kidneys^{(Reference Van Abel, Huybers, Hoenderop, Van Der Kemp, Van Leeuwen and Bindels222)}. The baseline factors of individual participants highlighted by Lehmann et al., such as age, body weight, vitamin D levels, and triacylglycerol levels, can influence the effectiveness and outcomes of 12 weeks of daily oral doses of vitamin D₃ supplementation in relation to frailty, calcium and phosphate metabolism, bone health, and circulating vitamin D metabolites^{(Reference Lehmann, Riedel, Hirche, Brandsch, Girndt and Ulrich223)}.

Other factors conditioning vitamin activity

Although vitamin D has well-known direct effects, it is important to recognise that these effects can also be modulated upstream before reaching the target sites. This adds complexity for researchers in establishing causal relationships and for clinicians in developing tailored treatments. For instance, fibroblast growth factor 23 (FGF-23) was found as one of the core proteins identified by proteomic analysis in a Swedish cohort of 1604 women at the age of 75 years. Frailty was associated with increased serum FGF-23 levels over 5 and 10 years, alongside higher odds of frailty. FGF-23, among other core proteins, may contribute to frailty through its effects on multiple biological pathways, including skeletal homeostasis and muscle function, potentially leading to sarcopenia^{(Reference Mitchell, Malmgren, Bartosch, McGuigan and Akesson224)}. FGF-23 is produced mainly by osteocytes and regulates the expression of vitamin-D-metabolising enzymes CYP24A1 (up-regulation) and CYP27B1 (down-regulation)^{(Reference Shimada, Hasegawa, Yamazaki, Muto, Hino and Takeuchi225)}. It also targets proximal renal tubules where it reduces phosphate reabsorption^{(Reference Takeshita, Kawakami, Furushima, Miyajima and Sakaguchi226)}, under the control of PTH, via increasing levels of serum phosphate or 1,25(OH)₂D^{(Reference Bergwitz and Jüppner227)}. Beben et al. had previously associated FGF-23 as a potential biomarker of functional outcomes in a study involving nearly 3,000 older individuals^{(Reference Beben, Ix, Shlipak, Sarnak, Fried and Hoofnagle228)}. The research team demonstrated that higher FGF-23 levels were independently linked to frailty and pre-frailty in older community-dwelling individuals, regardless of demographics, cardiovascular disease, risk factors or kidney function. The activation pathway of FGF-23 requires α-klotho, another hormone expressed in the kidney. Based on a longitudinal study conducted in Italy^{(Reference Shardell, Semba, Kalyani, Bandinelli, Prather and Chia229)} on 649 people aged 65 years and older, the serum α-klotho concentration was inversely associated with frailty odds and exhaustion odds. However, no association was found in a Turkish cross-sectional study with 89 geriatric patients^{(Reference Polat, Yalcin, Yazihan, Bahsi, Mut Surmeli and Akdas205)}.