Calcium and Phosphate: A Duet of Ions Playing for Bone Health

Abstract

The acquisition and maintenance of bone mass and strength are influenced by environmental factors, including physical activity and nutrition. Among micronutrients, calcium (Ca) and inorganic (i) phosphate (P) are the two main constituents of hydroxyapatite, the bone mineral that strengthens the mechanical resistance of the organic matrix. Bone contains about 99% and 80% of the body's entire supply of Ca and P, respectively. The Ca/P mass ratio in bone is 2.2, which is similar to that measured in human milk. The initial step of Ca-Pi crystal nucleation takes place within matrix vesicles that bud from the plasma membrane of osteogenic cells and migrate into the extracellular skeletal compartment. They are endowed with a transport system that accumulates Pi inside the matrix vesicles, followed by the influx of Ca ions. This process leads to the formation of hydroxyapatite crystal and its subsequent association with the organic matrix collagen fibrils. In addition to this structural role, both Ca and Pi positively influence the activity of bone-forming and bone-resorbing cells. Pi plays a role in the maturation of osteocytes, the most abundant cells in bone. Osteocytes are implicated in bone mineralization and systemic Pi homeostasis. They produce fibroblast growth factor-23, a hormonal regulator of renal Pi reabsorption and 1,25-dihydroxy vitamin D production. This relationship is in keeping with the concept proposed several decades ago of a bone-kidney link in Pi homeostasis. In contrast to their tight association in bone formation and resorption, Ca and Pi renal reabsorption processes are independent from each other, driven by distinct molecular machineries. The distinct renal control is related to the different extraskeletal functions that Ca and Pi play in cellular metabolism. At both the renal and the intestinal levels, interactions of Ca and Pi have been documented that have important implications in the acquisition and maintenance of bone health, as well as in osteoporosis management. In the kidney, increased Pi intake enhances Ca reabsorption and Ca balance. During growth and adulthood, administration of Ca-Pi in a ratio close to that of dairy products leads to positive effects on bone health. In contrast, when separately ingested as pharmaceutical salt supplements, thus inducing large differences between Ca and Pi concentrations in the intestinal lumen, they might have adverse effects on bone health. In osteoporotic patients treated with anabolic agents, a Ca-Pi supplement appears to be preferable to carbonate or citrate Ca salt. In conclusion, Ca and Pi constitute a key duo for appropriate bone mineral acquisition and maintenance throughout life. Outside the skeleton, their essential but distinct physiological functions are controlled by specific transporters and hormonal systems that also serve to secure the appropriate supply of Ca and Pi for bone health.

Key teaching points

•	Bone contains about 99% and 80% of the body's supply of Ca and P, respectively, as hydroxyapatite and has a Ca/P mass ratio of about 2.2, close to that measured in human milk.
•	The first step of Ca-Pi crystal nucleation takes place within matrix vesicles that bud from the plasma membrane of osteogenic cells.
•	In addition to their structural role, both Ca and Pi influence bone-forming and bone-resorbing cells.
•	There is a bone-kidney link in Pi homeostasis in which fibroblast growth factor-23, a molecule produced by osteocytes, appears to play a pivotal role.
•	In contrast to their tight association during bone formation and resorption, both intestinal and renal Ca and Pi processes are independent of each other.
•	Observational and interventional studies suggest that Ca-Pi salt or dairy products can exert positive effects on bone acquisition and maintenance.

Key words::

• bone mineralization
• matrix vesicles
• Pi-osteocytes interaction
• FGF-23
• osteoblast
• osteoclast
• Ca sensing receptors
• extraskeletal Ca and Pi handling

ACKNOWLEDGMENTS

The author is grateful to both Dr Marie-Claude Bertière, CERIN, Paris, France, and Dr Jean-Michel Lecerf, Institut Pasteur, Lille, France, for critically reading the manuscript.

Fig. 1. Bone mineralization process. Depicted are the relationships between bone-forming cells, MVs, and Pi transport. (A) Bone-forming cells, either osteoblasts or epiphyseal chondrocytes, are involved in endochondral ossification consisting of the replacement of cartilage by bone in the growth plate. These cells form vesicles, which bud from their plasma membrane and migrate into the unmineralized organic matrix. The MVs have the capacity to accumulate, from the surrounding unmineralized organic matrix, Pi and Ca, the two ions which constitute, once their solubility point is exceeded, the bone hydroxyapatite crystal. This crystalline structure penetrates the MV membrane, then propagates as extravesicular clusters and fills the space between the collagen fibrils, forming the new skeletal mineralized matrix. (B) This scheme illustrates the connection between bone-forming cells, particularly osteoblasts, and the Pi transporter. The osteogenic cells are endowed with a genetic program that codes for a specific Pi transporter protein (type III Pi transporter), which is inserted into the plasma membrane. The translocation of the Pi from the extracellular to the intracellular compartment is dependent upon the sodium (Na) gradient. The MVs are riched in this Na-coupled Pi transport system. The Pi transporter present in the plasma membrane of bone forming cells is regulated by PTH and growth factors such as IGF-I and BMP-2. With their respective plasma membrane receptors (R), these regulators are connected to the Na-Pi transporter genetic (DNA) and protein expression (mRNA) program. The stimulation in the plasma membrane of the type III Pi transporter Pit-1 by PTH, IGF-I, and BMP-2 is also expressed in the enhanced Pi translocation activity into the MVs. (Figure adapted from Caversazio and Bonjour [10], Palmer et al [11], and Suzuki et al [12]. See text for further details.)

Fig. 2. Respective role in MV mineralization of active Pi transporter, Ca channels, and membrane-bound phosphatases. The initial hydroxyapatite crystal formation requires as a driving force the active Na-dependent Pi translocation into the MV space and the influx of Ca through annexin channels. The increase in extravesicular Pi concentration is secured by MV membrane–bound phosphatases or pyrophosphatases. Tissue nonspecific alkaline phosphatase (TNAP) hydrolyzes PPi, which is a an inhibitor of hydroxyapatite formation. PHOSPHO1 hydrolyzes phosphocholine (PCho) and phosphoetholamine (PEA). The extravesicular accumulation of PPi comes from the hydrolyse of adenosine triphosphate (ATP) by nucleotide pyrophosphatase phosphodiesterase 1(NPP1) and also directly from the bone-forming cell metabolism. The accumulation of Ca within the MVs is promoted by several Ca-binding phospholipids and proteins not depicted in this scheme. (Adapted from Caversazio and Bonjour [10] and Orimo [15]. See text for further details.)

Fig. 3. Direct effects of ionized Ca on bone cell activity. Ca-operated sensing receptors (Rs) have been detected in precursors and mature osteoclastic and osteoblastic cells. In vitro, these Ca receptors are coupled with various intracellular signaling molecular mechanisms that may be involved in vivo to decrease osteoclastic resorption and increase osteoblastic formation. See Marie [16] for review.

Fig. 4. Transitional stages from preosteoblasts to osteocytes: model for positive effects of both dentin matrix protein-1 (DMP-1) and Pi for mineralization and osteocyte maturation. 1 = proliferating preosteoblast, 2 = preosteoblastic osteoblast, 3 = osteoblast, 4 = osteoblastic osteocyte (type I osteocyte), 5 = osteoid osteocyte (type II osteocyte), 6 = type III osteocyte, 7 = young osteocyte, 8 = old osteocyte, osteoid = unmineralized bone. The arrow indicates the transition from osteoblast to osteocyte differentiation and maturation. Both DMP-1 and Pi appear to be required for mineralization and maturation of osteoblasts into osteocytes. As described in the text, osteocytes are the main cells producing fibroblast growth factor-23 (FGF-23). (Adapted from diagrams presented in Franz-Odendaal et al. [33], Dallas and Bonewald [36], and Zhang et al. [37].)

Fig. 5. Experimental setting at the origin of the bone-kidney link concept in Pi homeostasis. Treatment of rats with the bisphosphonate EHDP (10 mg P equivalent per kg.day⁻¹ subcutaneously for 7 days) resulted in a blockage of bone mineralization. This restraint in the capacity of the bone organic matrix to incorporate Pi and Ca was followed at the kidney level by a selective inhibition of both the tubular Pi reabsorption and 1,25D production. As consequences of this transport and endocrine renal effects, the circulating level of Pi and 1,25D decreased. This latter effect led to a reduction in the intestinal absorption of both Ca and Pi. As discussed further in the text, FGF-23 may well be the factor involved in the kidney response observed after EHDP-induced inhibition of bone mineralization. Microradiographs reprinted from Schenk et al. [44]. EHDP = ethane-1-hydroxy-1, 1-bisphosphonate.

Fig. 6. Renal response to Pi intake in relation with changes in FGF-23 in healthy young men. (A) In 29 healthy volunteers (age range: 21–39 years), daily Pi intake was decreased from about 1500 to 500 mg and then elevated to nearly 3000 mg. To minimize directional changes in serum PTH by Pi, dietary Ca was also decreased and increased during low and high Pi intake, respectively. The maximal renal Pi reabsorption per milliliter of glomerular filtrate (TmPi/GFR) and the plasma levels of 1,25D and FGF-23 were measured before and after each variation in Pi intake. (B,C) The relationships between the changes (Δ) in TmPi/GFR (B) or 1,25D (C) were inversely related to the variations in the circulating levels of FGF-23. The arrows on the right side of diagram A symbolize the bidirectional changes of FGF-23 in response to dietary Pi variations. This study supports the concept that, in humans, FGF-23 is involved in systemic Pi homeostasis by regulating both renal Pi reabsorption and 1,25D production. (Adapted from Ferrari et al. [57]. See text for further details.)