Vitamin D and bone health

Abstract

Current data demonstrate that vitamin D deficiency contributes to the aetiology of at least two metabolic bone diseases, osteomalacia and osteoporosis. Osteomalacia, or rickets in children, results from a delay in mineralization and can be resolved by normalization of plasma calcium and phosphate homeostasis independently of vitamin D activity. The well characterized endocrine pathway of vitamin D metabolism and activities is solely responsible for vitamin D regulating plasma calcium and phosphate homeostasis and therefore for protecting against osteomalacia. In contrast a large body of clinical data indicate that an adequate vitamin D status as represented by the serum 25-hydroxyvitamin D concentration protects against osteoporosis by improving bone mineral density and reducing the risk of fracture. Interestingly adequate serum 1,25-dihydroxyvitamin D concentrations do not reduce the risk of fracture. In vitro human bone cell cultures and animal model studies indicate that 25-hydroxyvitamin D can be metabolised to 1,25-dihydroxyvitamin D by each of the major bone cells to activate VDR and modulate gene expression to reduce osteoblast proliferation and stimulate osteoblast and osteoclast maturation. These effects are associated with increased mineralization and decreased mineral resorption. Dietary calcium interacts with vitamin D metabolism at both the renal and bone tissue levels to direct either a catabolic action on bone through the endocrine system or an anabolic action through a bone tissue autocrine or paracrine system.

Key Words::

• Metabolic bone disease
• plasma calcium homeostasis
• osteomalacia/rickets
• osteoporosis
• vitamin D metabolism
• 25-hydroxyvitamin D critical serum concentrations

Introduction

Vitamin D was identified on the basis of its efficacy to treat the metabolic bone disease rickets in children [1]. The common form of rickets results from a delay in mineralization particularly at the growth plate while in adults the disease is known as osteomalacia [2]. Severe vitamin D depletion is one of a number of causes of rickets/osteomalacia and while there is a delay in mineralization, the synthesis of the protein matrix of bone (osteoid) by the bone forming cells (osteoblasts) is unimpaired [3]. A crucial observation on the role of vitamin D in the pathogenesis of rickets/osteomalacia was the rescue of this phenotype by feeding a diet containing high concentrations of calcium and phosphorus sufficient to normalize plasma concentrations [4]. These data indicate that the essential biological activity of vitamin D in the aetiology of rickets/osteomalacia is the regulation of intestinal calcium and phosphate absorption and the maintenance of plasma calcium and phosphate homeostasis.

It has been some 40 years since the issue of vitamin D deficiency was first raised in the context of increased risk of fracture [5]. Since that time the consistent finding from observational studies on elderly hip fracture patients has been that moderate but not necessarily severe vitamin D depletion increases fracture risk [6]. The increased risk of fracture relates to the serum 25-hydroxyvitamin D concentrations and not to serum 1,25-dihydroxyvitamin D concentrations [7]. Bone histomorphometry of hip fracture patients indicate that most patients demonstrate histology of osteoporosis with only a minority of patients in some studies but not all, demonstrating osteomalacia [5, 8]. Randomized clinical trial data of vitamin D supplementation, particularly when accompanied with calcium supplements to treat institutionalized, elderly subjects, demonstrate a reduction in the risk of fracture [9–11]. In contrast supplementation with 1,25-dihydroxyvitamin D does not demonstrate fracture risk reduction[11]. Finally population studies from the US indicate that bone mineral density increases with vitamin D status [12]. Thus there is a large body of clinical data to indicate that an adequate vitamin D status as represented by the serum concentration of the 25-hydroxyvitamin D metabolite improves bone mineral density and decreases the risk of osteoporotic fracture particularly in the elderly.

Actions of vitamin D on bone cells

Bone physiology is dependent on the activities of the 3 major cell types as well as the interaction with the haemopoietic system of the bone marrow. Osteoblasts are the bone forming cells derived from a mesenchymal cell lineage. Osteocytes are the most numerous cell type, embedded in bone mineral and are also derived from the mesenchymal cell lineage. Lastly osteoclasts are bone resorbing cells, derived from the haemopoietic lineage. Each of these cell types express the vitamin D receptor (VDR) and demonstrate vitamin D biological activities. As well each cell type is capable of metabolizing vitamin D, particularly 25-hydroxyvitamin D to form the active metabolite 1,25-dihydroxyvitamin D to exert biological activities [13].

The best characterized direct activity of vitamin D on bone cells is the action of 1,25-dihydroxyvitamin D on osteoblasts to increase osteoclastogenesis and bone resorption [14]. This action contributes to maintaining plasma calcium and phosphate homeostasis. 1,25-dihydroxyvitamin D activates VDR in osteoblast cells to increase expression of receptor activator of nuclear factor kappa (RANK) – ligand (RANKL). RANKL on the surface of osteoblasts binds to RANK expressed on the surface of osteoclast-progenitor cells to stimulate osteoclastogenesis and bone resorption. This pathway for activation of bone resorption is dependent on osteoblastic expression of VDR [15].

Vitamin D acts to increase expression of many genes in osteoblasts. Bone formation involves the coordinated synthesis of bone matrix, composed mainly of type I collagen and other proteins including osteocalcin, osteopontin, osteonectin, bone sialoprotein-1 and proteoglycans [16]. Mineralization occurs at discrete sites particularly along collagen fibrils incorporating calcium and phosphate to form a mature bone matrix. The life cycle of osteoblasts and their transformation to osteocytes are tightly linked to the mineralization process. The first activity requires cell proliferation followed by cell maturation, initially involving the synthesis and secretion of the components of osteoid including type I collagen. These activities are followed by mineralization accompanied by the maturation of some osteoblasts to pre-osteocytes which are embedded in unmineralized osteoid. With completion of mineralization most of the osteoblasts undergo apoptosis while the pre-osteocytes mature into osteocytes embedded in mineralized tissue. A small number of osteoblasts transform into lining cells localized to the surface of the bone. Mineralization is tightly controlled by the synthesis of stimulating and inhibitory factors by osteoblasts and osteocytes at various stages of maturation.

In vitro experimentation with primary bone cells isolated from rodents or humans demonstrate that inclusion of 1,25-dihydroxyvitamin D in the culture media inhibits osteoblast proliferation and enhances osteoblast maturation and mineral deposition [13,17,18]. The expression of many genes key to osteoblast maturation and mineral deposition are modulated by 1,25-dihydroxyvitamin D mostly with increased gene expression including type I collagen; alkaline phosphatase; matrix gla protein, an inhibitor of aberrant calcification of cartilage and arteries; osteopontin, another inhibitor of mineralization; bone sialoprotein and osteocalcin. Osteocytes in culture respond to 1,25-dihydroxyvitamin D with induction of fibroblast growth factor-23 (FGF23), which codes for the major phosphate-regulating hormone and dentin matrix protein 1 (DMP1), which plays a similar role as osteopontin as a source of inhibitors of mineralization [19].

Despite a report in the 1980's [20], it is only during recent times that the expression of critical genes coding for vitamin D metabolizing enzymes within bone cells and their physiological outcomes have been characterized [18,21]. These enzymes include 25-hydroxyvitamin D 1-alpha-hydroxylase (CYP27B1) responsible for the synthesis of 1,25-dihydroxyvitamin D and 25-hydroxyvitamin D 24-hydroxylase (CYP24) responsible for the catabolism of vitamin D metabolites. In vivo studies demonstrate that CYP27B1 is expressed extensively by many tissues including the skeleton where it is largely expressed in areas of trabecular bone and the growth plate [22–24]. Expression in bone tissue decreases with age [25], and is increased with high dietary calcium intake [26]. Expression is also dependent on the stage of osteoblast maturation and in vitro expression peaks immediately before mineralization reaches its peak. Regulation of CYP27B1 and CYP24 expression in bone cells is quite distinct from that in renal cells where there is an inverse relationship [25]. In bone cells and in other non-renal tissues, expression of these genes is positively related to each other suggesting that synthesis of 1,25-dihydroxyvitamin D in these non-renal tissues does not contribute to the plasma compartment but may perform an autocrine or paracrine activity. Consistent with this concept, expression of bone cell CYP24 is unrelated to serum 1,25-dihydroxyvitamin D concentrations, indicating that the endocrine action of plasma 1,25-dihydroxyvitamin D does not play a role in regulating bone CYP24 expression.

CYP27B1 is also expressed strongly in chrondrocytes [27]. Chrondrocyte-synthesised 1,25-dihydroxyvitamin D, activating chrondrocyte VDR, is responsible for the apoptosis of the hypertrophic chrondrocytes, osteoclastogenesis and resorption of mineral in the primary spongiosa region possibly as a result of angiogenesis [2,28]. In culture, expression of CYP27B1 by human osteoblasts is essential for the conversion of 25-hydroxyvitamin D at physiological concentrations to 1,25-dihydroxyvitamin D and stimulation of gene expression such as osteocalcin, osteopontin, and RANKL [29]. Under these conditions 25-hydroxyvitamin D inhibits human osteoblast proliferation and enhances mineral deposition at physiological concentrations similar to1,25-dihydroxyvitamin D which under these conditions requires supra-physiological concentrations [18]. Osteocytes respond to 25-hydroxyvitamin D and 1,25-dihydroxyvitamin D to up regulate FGF23 expression [30]. Recently osteoclasts have been reported to express CYP27B1 which is essential for 25-hydroxyvitamin D to stimulate osteoclast maturation in vitro from human peripheral blood monocytes in conjunction with RANKL and macrophage-colony stimulating factor (m-CSF) [31]. Most interesting is the observation that osteoclasts generated in the presence of 25-hydroxyvitamin D or 1,25-dihydroxyvitamin D, although more numerous, exhibit reduced resorptive activity on hydroxyapatite coated-slides when compared to osteoclasts matured simply in the presence of RANKL and m-CSF [32].

Thus vitamin D metabolites through the activation of osteoblast VDR by 1,25-dihydroxyvitamin D can exert catabolic actions on bone mineral to support plasma calcium homeostasis or alternatively exert actions to enhance mineralisation, at least with in vitro models, through stimulating osteoblast maturation and expression of genes associated with mineralization. One factor proposed to determine which of these two activities predominates is the stage of osteoblast differentiation. It has been reported that phenotypically immature osteoblast precursors predominately respond to 1,25-dihydroxyvitamin D through stimulation of the expression of the RANKL gene while mature osteoblasts predominately respond through stimulation of expression of osteocalcin [33]. The key questions for vitamin D physiology in the bone mineral field is whether vitamin D is capable of exerting these actions in vivo and if so, what are the regulating factors?

Vitamin D and Rickets/Osteomalacia:

The index disease of vitamin D deficiency is rickets in children or osteomalacia in adults when bone growth has ceased. It arises from a delay in mineralisation, which is measured as the average time between the synthesis of osteoid and its mineralization and is known as the mineralization lag time. Osteomalacia can only be definitively diagnosed from bone histology with the concurrence of prolonged mineralization lag time and increased osteoid width [34]. During growth, cartilage proliferation continues for a longer period with vitamin D deficiency with a marked delay in the mineralization of the growth plate [35]. In children, at skeletal sites which do not involve a growth plate such as the vertebral bodies, and in adults, vitamin D deficiency causes a delay in the mineralization of osteoid [34]. Therefore the ratio of mineral to osteoid is markedly reduced compared to healthy bone. With severe vitamin D deficiency or when vitamin D is inactivated, whether in humans or rodent models, rickets/ osteomalacia develops in association with hypocalcemia and/or hypophosphatemia, and often severe secondary hyperparathyroidism.

A highly significant finding was that the correction of the hypocalcaemia, hypophosphataemia and secondary hyperparathyroidism through dietary manipulation whether in the clinical condition or any of the animal models, also corrects the skeletal pathology [4,36]. Therefore the actions of vitamin D to regulate plasma calcium and phosphate homeostasis can completely account for its action to regulate the rate of mineralization as signified by the development of rickets/osteomalacia. These actions of vitamin D are exerted through the plasma 1,25-dihydroxyvitamin D metabolite activating VDR to stimulate intestinal calcium and phosphate absorption, renal tubular reabsorption of calcium and RANKL expression by osteoblasts to stimulate osteoclastogenesis in bone tissue. Each action enhances the flow of calcium and phosphate into the plasma compartment [37].

Vitamin D, Osteoporosis and Risk of Fracture:

A number of lines of evidence indicate that vitamin D status modulates the development of a second bone disease, osteoporosis. A recent meta-analysis of clinical observational studies clearly demonstrates the increased risk of hip fracture amongst the elderly with a depleted but not severely deficient vitamin D status [6]. While it was first considered that the skeletal pathology of a low vitamin D status with increased risk of fracture was osteomalacia [5], further bone histomorphometry studies on hip fracture patients revealed that only a minority of patients in some studies can be verified to exhibit this condition [8,38]. The vast majority of hip fracture patients demonstrate osteoporotic bone histology. Osteoporosis is the condition when the overall amount of mineralized tissue in the skeleton is reduced with a normal ratio of mineral to osteoid. Population studies indicate that bone mineral density increases with increasing serum 25-hydroxyvitamin D status until a plateau is reached at serum concentrations of approximately 75 nmol/L or greater [12]. Data from randomised clinical trials of vitamin D supplementation, particularly accompanied with calcium supplements and involving treatment of the institutionalized elderly, indicate fracture incidence is reduced clearly implicating a reduced vitamin D status as a cause of osteoporosis and increased risk of fracture [9–11].

Dietary studies in rodent models can mimic the data from clinical studies providing evidence for a mechanism by which a depleted vitamin D status can increase the risk of fracture [39]. Reducing serum concentrations of 25-hydroxyvitamin D from above 100 nmol/L to either 20 or 45 nmol/L in young, adult Sprague-Dawley rats for 3 months reduced trabecular bone volume in a dose dependent manner at a number of skeletal sites. There was no disruption to plasma calcium or phosphate homeostasis and mineralization lag time was not increased in these animals ruling out any contribution from osteomalacia. No relationship was found between serum PTH or 1,25-dihydroxyvitamin D concentrations and trabecular bone volume implicating the maintenance of an adequate serum concentration of 25-hydroxyvitamin D to be critical for maintaining bone mineral homeostasis. The loss of bone mineral volume with decreasing 25-hydroxyvitamin D was the result of increased expression of skeletal RANKL and increased osteoclast surface in bone tissue. Similar studies with 15-month old rats confirmed the loss of trabecular bone with decreasing vitamin D status and extended bone loss to cortical bone at similar serum 25-hydroxyvitamin D concentrations [40].

An alternative strategy to investigate whether vitamin D activity in bone cells can improve bone mineral homeostasis has involved the development of transgenic mouse models in which vitamin D activity is increased in specific bone cells. Increasing vitamin D activity in mature osteoblasts through over-expression of the gene for human VDR under the regulation of the human osteocalcin gene promoter increases trabecular and cortical bone volumes [41]. These mice have an increase of both cortical and trabecular bone volumes at various skeletal sites of some 15 % compared with wild type mice. An increase in bone formation and a decrease in bone resorption contribute to the increased bone volumes. A similar approach has been followed with the creation of a transgenic mouse line over-expressing the human CYP27B1 gene only in osteoblasts under the control of the human osteocalcin gene promoter. These mice also demonstrate a bone phenotype with increased trabecular and cortical bone volumes [42].

Critical concentrations for serum 25-hydroxyvitamin D and bone health: osteomalacia and osteoporosis

Effects of a depleted vitamin D status on osteoporosis and increased fracture risk have previously been attributed to effects on the calcium economy through a decrease in intestinal calcium absorption [43]. However this concept must be questioned since all the relevant data identify significant relationships between serum concentrations of the 25-hydroxyvitamin D metabolite and bone mineral density, bone volumes or risk of fracture whether from clinical or animal studies with no relationships found between these variables and serum 1,25-dihydroxyvitamin D concentrations. In contrast intestinal calcium absorption is strongly related to serum concentrations of 1,25-dihydroxyvitamin D [44]. Reports of a significant relationship between calcium absorption and serum 25-hydroxyvitamin D have been made, but this relationship is lost when serum 1,25-dihydroxyvitamin D concentrations are included in the analysis [44, 45]. Clinical studies on patients with a low vitamin D status (that is serum 25-hydroxyvitamin D concentrations of 40 nmol/L or lower) identified that ionised calcium, serum 1,25-dihydroxyvitamin D and intestinal calcium absorption were not clinically reduced until serum 25-hydroxyvitamin D concentrations fell below 20 nmol/L [46]. In the rodent model osteomalacia was not found until serum 25-hydroxyvitamin D fell below this concentration [39]. Thus robust clinical data indicate that with average dietary calcium intakes plasma calcium homeostasis is maintained at serum 25-hydroxyvitamin D concentrations of 20 nmol/L or greater. Data from rodent studies indicate that this concentration of serum 25-hydroxyvitamin D also prevents the development of osteomalacia.

As discussed above consistent observational studies have demonstrated that the risk of hip fracture amongst the elderly is significantly increased with a depleted but not necessarily deficient vitamin D status [6]. A mean serum concentration of 25-hydroxyvitamin D around 40 nmol/L has been reported for these patients [47,48]. Meta-regression analyses of data from randomised clinical trials of vitamin D supplementation indicate that whether hip fracture or other non-vertebral fracture is the outcome, a serum 25-hydroxyvitamin D concentration of approximately 75 nmol/L or greater is required to significantly reduce the risk of these fractures [9]. Furthermore depleted serum 25-hydroxyvitamin D concentrations have been shown to affect bone histology in clinical studies. Need et al. [49], found that serum 25-hydroxyvitamin D was significantly, inversely related to unmineralised osteoid thickness in ambulant patients presenting to an osteoporosis clinic. Although osteoid thickness is increased in osteomalacia this is not a specific finding and osteoid thickness can be increased when bone turnover is increased as for example with hyperparathyroidism, whether secondary or primary [34,50]. The patients in the Need et al. study were carefully evaluated by mineralization lag time, the diagnostic parameter for osteomalacia, which indicated that these patients did not exhibit osteomalacia. No bone parameters were related to serum 1,25-dihydroxyvitamin D concentrations and the concentration of serum 25-hydroxyvitamin D required to minimise osteoid width was between 80 and 100 nmol/L. A large cohort of Northern European road accident victims provided similar results when bone histology and serum vitamin D metabolites were analysed. Osteoid thickness was inversely related to serum 25-hydroxyvitamin D with an optimal serum concentration of approximately 75 nmol/L necessary to minimise this variable [51]. Once again no relationship was found between serum 1,25-dihydroxyvitamin D and any of the bone variables. Rodent studies involving dietary manipulation of vitamin D status indicate the critical serum concentration of 25-hydroxyvitamin D to abrogate any loss of trabecular bone volume in this model is 75–80 nmol/L or greater [39].

Discussion

The current data indicate that vitamin D exerts at least two actions to modulate bone health. Firstly it has an action to maintain plasma calcium and phosphate homeostasis preventing the development of osteomalacia. It is the plasma 1,25-dihydroxyvitamin D which contributes to maintaining plasma calcium homeostasis through an action on osteoblasts to stimulate osteoclastogenesis and bone resorption enhancing the flow of calcium and phosphate into the plasma compartment. The critical concentration of serum 25-hydroxyvitamin D to maintain adequate plasma concentrations of 1,25-dihydroxyvitamin D and intestinal calcium absorption is 20 nmol/L.

Vitamin D also acts to maintain mineralised bone tissue volume as assessed by bone mineral density and trabecular and cortical bone volumes and thus prevent the development of osteoporosis and reduce the risk of fracture. In this case increased risk of fracture, bone histology and bone mineral density all relate to serum 25-hydroxyvitamin D concentrations and not to serum 1,25-dihydroxyvitamin D. An interaction between vitamin D status and dietary calcium intake is probable as the most consistent data from randomised controlled trials for fracture risk reduction is demonstrated when these nutrients are combined [10,11].

The concept that vitamin D has at least two activities to maintain bone health is not new as studies in the 1950's demonstrate [52]. Swedish researchers experimenting with vitamin D deficient rats given graded doses of vitamin D found that serum calcium continued to rise even when intestinal calcium absorption was normalised. At low doses of vitamin D intestinal absorption of calcium was corrected. At higher doses of vitamin D, serum calcium continued to rise by a mechanism which has been assumed to arise from bone, described as a ‘calcemic effect’ of vitamin D [45].

The actions of vitamin D at any one time depend not only on vitamin D status but also on dietary calcium intake [53,54]. When dietary calcium intake is inadequate and vitamin D status is deficient we have demonstrated that osteomalacic bone develops with increased mineralization lag time [55]. Under these conditions plasma calcium and phosphate homeostatic mechanisms are activated which include the endocrine vitamin D system through plasma 1,25-dihydroxyvitamin D. Serum parathyroid hormone (PTH) concentrations may be increased elevating the activity of CYP27B1 enzyme in the kidney to allow maintenance of serum 1,25-dihydroxyvitamin D concentrations even though serum 25-hydroxyvitamin D concentrations may fall below 40 nmol/L [26]. Under these conditions intestinal calcium and phosphate absorption are maintained and osteoclastogenesis is stimulated through the interaction between PTH and 1,25-dihydroxyvitamin D acting on osteoblasts to increase bone resorption. Furthermore bone cell activity is augmented, increasing the number of proliferating osteoblasts and therefore enhancing the expression of RANKL to further stimulate osteoclastogenesis [33]. When serum 25-hydroxyvitamin D concentrations fall below 20 nmol/L there is insufficient substrate for the renal CYP27B1 enzyme and serum 1,25-dihydroxyvitamin D concentrations fall with a consequent fall in intestinal calcium absorption and the development of hypocalcemia and osteomalacia [46].

When dietary calcium intake is sufficient but vitamin D status is low, or vice versa, plasma calcium and phosphate homeostasis is maintained [55]. Depending on the concentrations of dietary calcium or serum 25-hydroxyvitamin D, the plasma calcium homeostatic mechanism can minimize serum PTH and 1,25-dihydroxyvitamin D concentrations. However the expression of RANKL in bone tissue remains elevated with increased osteoclast surface and bone resorption and bone volumes are decreased. Mineralization lag times are normalized and the bone histology demonstrates osteoporosis [39].

When dietary calcium is sufficient to meet the needs of calcium balance and serum 25-hydroxyvitamin D concentrations are 75 nmol/L or greater, serum PTH and 1,25-dihydroxyvitamin D concentrations are minimized. Serum 25-hydroxyvitamin D and bone tissue expression of CYP27B1 are maximized. Bone tissue expression of RANKL is minimised with decreased osteoclast surface and increased bone volumes. Another bone variable, mean wall thickness is increased. This is the thickness of mineral laid down by osteoblasts at each site of bone remodelling by the bone multicellular unit and indicates that the time period for bone formation is increased. It suggests that apoptosis of osteoblasts may be delayed [39].

These data further describe the relationship between vitamin D activities at various organs and dietary calcium intake to maintain plasma calcium homeostasis and bone mineral homeostasis. Hence bone architecture, and presumably strength, is maintained reducing the risk of fractures. Such interactions are complex and dependent on dietary calcium intake. They involve vitamin D exerting either endocrine activities to maintain plasma calcium homeostasis or the novel concept of autocrine /paracrine activities within bone tissue to maintain bone architecture.

Questions and Answers

JC Souberbielle, France

Recent publications suggest that bisphosphonates and other anti-resorptive drugs need to have sufficient vitamin D to be effective. Do you think that is related to a local action of 25(OH)D on bone cells, with local formation of 1,25(OH)₂D3 or is it only a matter of a control of secondary hyperparathyroidism because we have a decrease in bone resorption and an increase of efflux of calcium from bone?

H Morris

It is possible that the rate of bone turnover has an effect and that with a high bone turnover we have more immature osteoblasts and that these produce RANK ligand and therefore stimulate bone resorption, whereas with low bone turnover we will have more mature osteoblasts which will have an anabolic action on bone and preserve it. Goltzman has published interesting parallel data demonstrating that FGF23 is also affected by bone turnover, so phosphate homeostasis and calcium homeostasis are being influenced, if not regulated, by the rate of bone turnover. It is therefore an interesting proposition to look at something other than calcium, some other anti-resorptive agent which can suppress bone turnover, to observe these effects on FGF23.

G Jones, Canada

Your story is compelling. Work done in the early 1980's showed that bone cells could perform hydroxylations. However, this could only be demonstrated if the DBP concentration was dropped to less than 3 %. So when you show us the concentrations of 25(OH)D required to produce these effects, how much DBP or calf serum did you add and how is that influencing the dose?

H Morris

Concerning our in vivo study, the DBP is what the rat produces. In the other study, we used 10 % fetal calf serum and the only DBP the cell cultures got was what was in that serum. We have found that both megalin and cubulin are expressed by osteoblasts and both are receptors for DBP. Therefore, there is a very interesting paradigm for vitamin D as to whether it is the free hormone, i.e. that which is not bound to DBP, which is biologically active or, under conditions, is it the DBP binding the 25(OH) and, taking it into the cell through the megalin-cubulin receptor system.

G Jones

The other thing you have added to the work from the 1980's is studying the mRNA expression and showing that these messages are expressed, not just the enzymic activity data we relied upon before.

M Hewison, USA

Your findings are interesting in the context of patients with inflammatory bowel disease. Often young people with the condition have a bone mineral density of a 75 year old. It can be a chronic bone wasting disease. They tend to have higher concentrations of 1,25(OH)₂D3 than in the normal population. It is an independent marker of bone loss, which would fit in with your proposal.

H Morris

The clinical trials and the Cochrane meta-analysis showed that 1,25(OH)₂D3 supplementation had no benefit to reduce hip fracture. Some have used Rocaltrol, because of its bone forming properties. In one regime, it was given in low doses with a dietary calcium supplement in order to enhance the absorption of calcium. One study using a higher dose without calcium supplementation showed no benefit in bone mineral density. The issue is whether it is Rocaltrol or the circulating actions of 1,25(OH)₂D3 which move calcium from bone and to plasma.

I Young, UK

Coming back to the question of a relatively high calcium intake with vitamin D supplementation, are you concerned about the potential dangers with that?

H Morris

Yes. I have talked about an inadequate and an adequate calcium intake. Studies show that for men under 60, balance is achieved with a daily intake of 700–750 mg/day. All we have to achieve from dietary intake is the physiological concentrations to meet the obligatory losses of calcium through the kidney. To maintain plasma calcium homeostasis, we call on our reserves in bone, and it is only that we need to replace.

For post-menopausal women there are different considerations because oestrogens have major effects on the calcium economy, causing increased losses and decreased absorption. Their dietary requirements are therefore higher and although the studies have not been done, their requirement is estimated to be around 1,300 mg/day.

For vitamin D, we need to meet the requirement for the organ we are considering. I suggest that every organ system will have a different critical 25(OH)D concentration because it will have a different expression of the CYP27B1. Those with lower expression will require higher concentrations.

Acknowledgements

This manuscript was prepared and the work of the authors reported herein with the support of project grants funded by the National Health and Medical Research Council of Australia.

Disclosures: HAM has participated in Speakers Bureaux for Roche Diagnostics Australia Pty Ltd and Abbott Diagnostics International.

Declaration of interest: The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper.