Serum Ca (1 mmol/L =4 mg/dL) occurs in three forms: approximately 40% is bound, predominantly to albumin; approximately 10% is chelated and complexed to small molecules such as bicarbonate, phosphate, or citrate; and approximately 50% is ionized. Complexed and ionized Ca are ultrafiltrable.

Total Ca concentrations (tCa) in cord sera increase with increasing gestational age. Serum tCa may be as high as 3 mmol/L in cord blood of infants born at term, and the concentrations are significantly higher than paired maternal values at delivery (3,4,5,6). Serum tCa reaches a nadir during the first 2 days after birth (7,8,9,10,11,12,13); thereafter, concentrations increase and stabilize at a level generally above 2 mmol/L (14). In infants exclusively fed human milk, the mean serum tCa increases from 2.3 to 2.7 mmol/L over the first 6 months postnatally. Normally serum tCa in children and adults remains stable, with a diurnal range of less than 0.13 mmol/L. During the third trimester of pregnancy, a modest reduction in maternal serum tCa concentration (average 0.1 mmol/L) is associated with a decrease in serum albumin concentration.

Serum ionized Ca (iCa) concentration is the best indicator of physiologic blood Ca activity. Measurement of serum iCa is firmly established in clinical medicine, and simple, rapid, and direct determination of iCa from whole blood, plasma, and serum by ion-selective electrodes is freely available. Whole-blood iCa analyzers are gaining popularity because they are adaptable for “point of care” testing. However, some differences exist in values from different iCa analyzers, particularly for whole-blood iCa values (15), as a result of differences in the design of the reference electrode, formulation of calibrating solutions, and lack of a reference system for iCa. Thus, normative data should be generated according to the subject’s age, the instrument, and the type of sample used for iCa measurement.

Serum iCa decreases in the presence of high serum albumin, P, bicarbonate, and heparin. Serum iCa increases with increased Mg and is inversely related to blood pH. The effect of the latter may be minimized by the immediate analysis of serum samples for iCa. Freezing serum samples in 5% carbon dioxide-containing tubes may minimize the impact of pH variations if measurement of iCa is delayed for 1 week.

One report showed a wide range of cord whole-blood iCa of 0.4 to 1.85 mmol/L from apparently normal pregnancies (16). This is a much wider range compared with multiple reports based on cord sera, although the range for whole-blood iCa becomes much narrower within hours after birth and similar to serum iCa values. Cord-serum iCa increases with increasing gestational age and is higher than values in paired maternal sera. In healthy term neonates, serum iCa averages 1.25 mmol/L with 95% confidence limits of 1.1 to 1.4 mmol/L (4.4 to 5.6 mg/dL), and there is a decline in serum iCa in the first 48 hours of life with a nadir at 24 hours (17). Serum iCa generally changes in parallel with tCa in healthy humans. However, serum iCa is stable and normal during pregnancy in contrast to a slight reduction in tCa. Serum tCa and iCa are correlated in infants and adults, but the correlation is inadequate to predict one from the other with sufficient accuracy.

The concentration of iCa is critical to many important biologic functions with the extracellular Ca concentration normally maintained within a narrow range. The Ca ion is well established as an intracellular second messenger, but identification of the extracellular CaR has established that iCa also functions as a messenger outside cells. Ca homeostasis also involves the maintenance of an extremely large Ca concentration gradient across the cellular plasma membrane. In the cell, distribution of Ca is not uniform. The cytosolic compartment contains 50 to 150 nmol of Ca per liter of water; a larger intramitochondrial Ca pool contains 500 to 10,000 nmol of Ca per liter of cell water. In contrast, the concentration of iCa in extracellular fluid is 1 million nmol/L (1 mmol/L). There are at least two adenosine triphosphate-dependent mechanisms involved in the maintenance of the Ca concentration gradient across the plasma membrane. The measurement of intracellular Ca continues to improve with better instrumentation and probes, but it is not yet freely available.

Approximately 30% of serum Mg (1 mmol/L =2.4 mg/dL) is in the protein-bound form, with the remainder in the ultrafiltrable portion. Seventy percent to 80% of ultrafiltrable Mg is in ionic form, and the remainder is complexed to anions, particularly phosphate, citrate, and oxalate. Cord-serum total Mg (tMg) is higher than paired maternal values. Serum tMg of 0.92 ±0.13 mmol/L (mean ±2 SD) in children is slightly higher than the adult values of 0.88 ±0.13 mmol/L (18). Ion-selective electrodes are used in the measurement of ionized Mg (iMg) in whole blood and sera. iMg concentrations average 62% to 70% of the tMg concentration in cord and postnatal sera. Cord-serum iMg is also higher than that in maternal serum (19,20,21). The clinical role of iMg (versus tMg) in a number of disease states appears limited (22).

Cellular Mg content of most tissues is 6 to 9 mmol/kg wet weight, and most of this Mg is localized in membrane structures (e.g., microsome, mitochondria, plasma membrane). The much smaller pool of free Mg in the cell is maintained at about 1 mmol/L and is in an exchanging equilibrium with membrane-bound Mg. Unbound intracellular Mg has a critical role in cellular physiology and catalyzes enzymatic processes concerned with the transfer, storage, and use of energy. Intracellular Mg usually remains stable despite wide fluctuations in serum Mg. In Mg-deficient states, however, the intracellular content of Mg can be low despite normal serum concentrations. The measurement of intracellular Mg continues to improve with better instrumentation and probes but is not freely available.

In humans, parathyroid glands are derived from the third and fourth pharyngeal pouch. The PTH gene, along with the genes for insulin,β-globulin, and CT, is located on chromosome 11p15 (23), and restriction site polymorphisms near the PTH gene have been detected. The initial translational product of the mRNA is a 115-amino-acid prepro-PTH. Prepro-PTH then undergoes proteolytic cleavage in the endoplasmic reticulum to remove a 25-residue amino-terminal signal sequence to form pro-PTH. The prohormone-specific region is cleaved further during subsequent intracellular processing to generate the 84-amino-acid secreted form of the intact hormone with a relative molecular mass (M_r) of 9,500. PTH is synthesized by the chief cells and stored in secretory granules. It is colocated and secreted with chromogranin A, a protein that may act in autocrine- or paracrine-regulated release of PTH.

About 50% of the newly generated PTH is proteolytically degraded intracellularly, and some of the inactive fragments are also secreted. After release into the circulation, the intact PTH molecule has a serum half-life of 5 to 8 minutes and undergoes a series of cleavages by endopeptidases in the liver and kidney. The amino-terminal fragments contain the biologically active fractions, with the 1-34 fragment having the most calcemic activity; modifications at the amino terminal, particularly at the first two residues, can abolish its biologic activity. The midregion and carboxyl-terminal fragments are biologically inert, although the latter may have some in vitro biological activity.

Circulating immunoreactive PTH is a complex mixture of intact 1-84 PTH, multiple peptide fragments from the amino- and carboxyl-terminals, and mid-molecular regions. Normally there are greater amounts of middle and carboxyl fragments than intact hormone in the circulation because of metabolic breakdown of the short-lived, intact hormone, coupled with glandular secretion of inactive fragments. The fragments are cleared from the blood virtually exclusively by glomerular filtration. Intact PTH and amino-terminal fragments constitute less than 20% of PTH immunoreactivity in the peripheral circulation. PTH molecules reactive in the widely used commercial immunoradiometric assays (IRMA), designed to detect both amino- and carboxyl-terminal epitopes of the peptide, have been considered as “intact” PTH (IPTH). However, there have been reports that the large 7-84 fragment of PTH is also detected by these assays. This large fragment is biologically inactive and present in greater concentrations in uremic states or hyperparathyroidism. The conventional IPTH technique increases the PTH concentration by 30% to 50%, compared to the latest chemiluminescence or IRMA techniques that measure the “whole” or “bio-intact” PTH. Therefore, the treatment of secondary hyperparathyroidism based on data from conventional IPTH assays theoretically may lead to overtreatment and oversuppression of biologically active PTH, although the clinical significance of this possibility remains to be defined. In any case, consistency of the PTH assay methodology and serial measurements are critical to the interpretation and management of pathologic states.

PTH concentrations in cord blood frequently are low and do not correlate with PTH concentrations in maternal sera (4,24). Earlier studies of higher levels of bioactive PTH in cord sera from cytochemical assays may be related to elevated concentrations of parathyroid hormone-related protein (PTHrP), since PTH-like bioactivity was tightly correlated with levels of PTHrP in sheep and pig. Small amounts (approximately 5%) of perfused fragments (35-84, 44-68, and 65-84 amino acids) but probably not the whole PTH molecule are reported to cross the human placenta. Serum PTH concentrations increase postnatally coincidentally with the fall in serum Ca in both term and preterm infants (4,24-27). The rise in serum IPTH is greater for preterm infants with hypocalcemia compared to term infants reflecting appropriate PTH response. Serum PTH concentrations are similar for children and adults but are increased in the elderly. Serum concentrations of intact PTH as measured by IRMA showed no change during normal pregnancy. In adults, serum intact PTH is present in picomolar concentrations. It has a significant circadian periodicity, spontaneous episodic pulsatility with distinct peak property and a significant temporal coupling with serum iCa and P concentrations and prolactin secretion.

PTH effects on end-organ systems appear to be mediated through its binding to specific receptors. The type 1 PTH/PTHrP receptor has been identified in bone, cartilage, kidney, intestine, aorta, urinary bladder, adrenal gland, brain, and skeletal muscle. It binds equally to PTH and PTHrP and belongs to a superfamily of guanine-nucleotide-binding (G) protein-coupled cell membrane receptors (GPCR) including those for CT, secretin, growth hormone-releasing hormone, corticotrophin-releasing hormone, glucagon, vasoactive intestinal polypeptide, and others. Another PTH receptor (type 2) responds only to PTH, although its main endogenous ligand appears to be a 39-amino-acid peptide, hypothalamic tubular infundibular peptide. It has been found in the brain, pancreas, and intestines, but the physiologic significance of this receptor remains ill defined.

The gene for the PTH/PTHrP receptor is located on chromosome 3p21.1-p24.2. It contains 17 exons and encodes a mature glycoprotein of 593 amino acids (28). The type 1 PTH receptor consists of extended extracellular, ligand-binding amino-terminal and intracellular G-protein-associated carboxyl-terminal domains and seven transmembrane domains. Signal transduction mediated by G proteins results in multiple second messenger pathways to affect both stimulatory and inhibitory end-organ responses. The strongest and best-characterized second messenger signaling pathway is the PTH-stimulated coupling of the type 1 PTH receptor to the stimulating G protein (G_s; composed of 3 subunits: α, β, and γ, and encoded by the GNAS1 gene localized to 20q13.3), which activates adenyl cyclase, an enzyme that generates cyclic AMP (cAMP). However, coupling of the type 1 PTH receptor to the G_q class protein activates phospholipase C, which generates inositol phosphate (IP₃) and diacylglycerol (DAG). These second messengers in turn lead to stimulation of protein kinases A and C and Ca transport channels and result in a variety of hormone-specific tissue responses.

In physiologic terms, PTH acts synergistically with 1,25(OH)₂D and is the most important regulator of extracellular Ca concentration. PTH acts directly on bone and kidney, and indirectly on intestine. Immediate control of blood Ca is probably due to PTH-induced mobilization of Ca from bone and increased renal distal tubular reabsorption of Ca. PTH also decreases proximal tubular and thick ascending limb reabsorption of sodium, Ca, phosphate, and bicarbonate. PTH effects on kidney are mediated primarily through stimulation of sodium/calcium exchange, calcium transport proteins and renal 25 OHD-1α-hydroxylase, but a decrease in sodium-dependent phosphate cotransporter, NPT-2. Maintenance of steady-state Ca balance is probably from increased intestinal Ca absorption secondary to increased 1,25(OH)₂D production. PTH increases acutely within minutes the rate of Ca release from bone into blood. Chronic effects of PTH are to increase the numbers of osteoblasts and osteoclasts and to increase the remodeling of bone. Continuous exposure to elevated concentrations of PTH leads to increased osteoclastic resorption of bone. In contrast to its classic action on Ca mobilization from bone, the amino-terminal fragments of PTH and PTHrP, and small pulses of PTH have an anabolic effect on bone, independent of its resorptive action. Other PTH effects on bone include enhanced collagen synthesis, activities of alkaline phosphatase, ornithine and citrate decarboxylases, and glucose-6-phosphate dehydrogenase; DNA, protein and phospholipid synthesis.

Extracellular Ca is the most potent regulator of PTH secretion and is mediated by the cell-surface CaR, which detects minute perturbations in the extracellular iCa concentration and responds with alterations in cellular function that normalize iCa. Thus, iCa functions as extracellular as well as intracellular messenger. The human CaR gene is located on chromosome 3q13.3-q21 and encodes a cell-surface protein of 1,078 amino acids. The CaR gene is developmentally upregulated, and CaR transcripts are present in numerous tissues including chief cells of the parathyroid glands, kidneys (in particular the thick ascending limb), brain and nerve terminals, breast, lung, intestine, adrenal and skin, and also the precursor and mature osteoblasts and osteoclasts. CaR is a member of the GPCR superfamily with a seven-member membrane-spanning domain. It contains at least seven exons, of which six encode the large (approximately 600-amino-acid) amino-terminal extracellular domain and/or its upstream untranslated regions, while a single exon codes for the remainder of the receptor including a cytoplasmic carboxyl-terminal intracellular domain. Signal transduction mediated by G proteins results in activation of phospholipase C that generates IP₃ and DAG, and subsequent stimulation of protein kinase C (PKC) and Ca transport channels.

Low or falling serum Ca concentrations result in active secretion of preformed PTH within seconds. There is a sigmoidal type of PTH secretion in response to decreased serum Ca, which is most pronounced when serum Ca is in the mildly hypocalcemic range. PTH secretion is 50% of maximal at a serum iCa of about 1 mmol/L (4 mg/dL); this is considered the calcium set point for PTH secretion (29). Sustained hypocalcemia increases PTH mRNA within hours. Protracted hypocalcemia leads within days to cellular replication and increased gland mass. High serum Ca suppresses PTH secretion via activation of CaR. It in turn activates phospholipase C and generation of IP₃ and DAG and probably increases the proteolytic destruction of preformed PTH. Hyperphosphatemia stimulates PTH secretion, probably by lowering the serum Ca concentration.

In the kidney, CaR decreases the basal and PTH-stimulated paracellular reabsorption of Ca, Mg, and sodium via multiple mechanisms including inhibition of cAMP accumulation; stimulation of phospholipase A2 activity. The release of free arachidonic acid is promoted which is metabolized via the lipooxygenase pathway to P450 metabolites that inhibit the activities of the sodium-potassium-chloride cotransporter and potassium channel; and possibly affect renal water regulation by inhibition of vasopressin-abated water flow. In chronic renal failure, downregulation in the
expression of renal CaR may account for the development of secondary hyperparathyroidism (30), and downregulation of PTH receptors may account for the skeletal resistance to the calcemic effect of PTH (31). Extracellular Ca exerts numerous other actions on parathyroid function, including modulation of the intracellular degradation of PTH, cellular respiration, membrane voltage, and the hexose monophosphate shunt.

Maintenance of Ca homeostasis through other organs also may be possible, for example, through the presence of CaR in intestinal cells (32), and probable modulation of CT secretion from changes in intracellular Ca (33). Furthermore, expression of CaR in gastrin-secreting G cells and acid-secreting parietal cells, together with data indicating that CaR exhibits selectivity for L-aromatic amino acids, appears to provide a molecular explanation for amino acid sensing in the gastrointestinal tract, regulation of PTH secretion and urinary Ca excretion, and the physiologic interaction between Ca and protein metabolism.

Decrease in serum Mg concentration stimulates PTH secretion (34,35), although chronic hypomagnesemia inhibits secretion of PTH (34,36). Hypomagnesemia is also associated with an increased target tissue resistance to PTH probably from inactivity of adenylate cyclase, a Mg-requiring enzyme. Hypermagnesemia rapidly decreases the secretion of PTH in vivo in human subjects, and PTH concentration remains depressed despite concomitant hypocal- cemia, presumably in part due to stimulation of CaR by other divalent cations such as Mg. Vitamin D and its metabolites, 25-hydroxyvitamin D (25-OHD) and 1,25(OH)₂D, acting through vitamin D receptors, decrease the level of PTH mRNA. Additional systemic factors (growth hormone, insulin-like growth factor 1, estrogen, progesterone, CT, cortisol, catecholamines, prostaglandins, and somatostatin) and local factors (interleukin-l [IL-1]) modulate PTH secretion and function, although their role in the regulation of Ca and Mg metabolism under physiologic conditions is not clear.

CT is secreted primarily from thyroid C cells and also from many extrathyroidal tissues including placenta, brain, pituitary, mammary gland, and other tissues. Developmentally, CT-containing cells and parathyroid gland cells are thought to derive from the same tissue source as the neural crest. In the rat, the number of thyroid C cells and secretion of CT increase from fetal life to suckling, a period of rapid growth (37). There is probably no placental crossover of CT; the human placental tissue is able to produce CT in response to the presence of Ca in the culture medium. In human neonates, the CT content in crude tissue preparations of thyroid is larger than that of the adult thyroid (38).

There are two calcitonin genes, α and β, located on chromosome 11p15.2 near the genes for β-globulin and PTH. Two different RNA molecules are transcribed from the α gene. It is comprised of six exons with the fourth exon translated into the precursor for CT and the fifth translated into the precursor for CT gene-related peptide-I (CGRP-I). The initial translational product of the mRNA is prepro-CT, a 141-amino-acid peptide. It is cleaved by endopeptidase at the endoplasmic reticulum to form pro-CT, a 13-kDa 116-amino-acid peptide. CT (between the sixtieth and ninety-first positions of the pro-CT peptide) and equimolar amounts of non-CT secretory peptides, corresponding to the flanking peptides linked to the amino and carboxyl terminals of the prohormone, are generated during precursor processing. Further structural modifications to the CT molecule occur intracellularly prior to release into the circulation. These modifications include formation of a disulfide bridge between two cysteine remnants in position 1 and 7, and hydroxylation of the C-terminal proline residue; both are essential for binding of CT to its receptor. The CT monomer is a 32-amino-acid peptide (M_r 3,400). CGRP-I is synthesized wherever the CT mRNA is expressed, e.g., in medullary carcinoma of thyroid, although there is no translational product from the CGRP-I sequence.

The β or CGRP-II gene is transcribed into the mRNA for CGRP predominantly in nerve fibers in the central and peripheral nervous systems, blood vessels, thyroid and parathyroid glands, liver, spleen, heart, lung, and possibly bone marrow. CGRP, a 37-amino-acid peptide (M_r 4,000), is also generated from the larger precursor molecule pro-CGRP, a 103-amino-acid peptide. Seventy-five amino-terminal residues of the preprohormones for CT and CGRP are predicted to be identical.

Classic bioactivity of human CT (hCT) is present in the full 32-amino-acid structure or its smaller fragments, such as hCT 8-32 and hCT 9-32; the ring structure of CT enhances, but is not essential for, hormone action. Basic amino acid substitutions confer a helical structure in this region as found in salmon and other nonmammalian CT, resulting in greater potency in lowering serum Ca and probably a longer circulating half life. The kidney appears to be the dominant organ in the metabolism of human CT. A small percentage of the metabolic clearance rate of CT in humans may be accounted for by enzymatic degradation in blood. Injected hCT monomer disappears from the blood in vivo with a half-life of approximately 10 minutes; in contrast, the half-life of hCT in plasma incubated in vitro at 37°C may be longer than 20 hours (39). Depending on the animal species, other sites such as the liver, intestine, and bone may be involved in the metabolism of CT.

Circulating immunoreactive CT and CGRP are a heterogeneous mixture of different molecular forms and are recognized as long as the antigenic epitopes recognized by the antiserum are expressed. Immunoreactive CT or CGRP concentration is expressed in gravimetric or molar equivalents of synthetic CT or CGRP. Sample preparation with initial extraction, gel chromatography, and high-performance liquid chromatography separation, and the use of two-site immunoassay can improve the sensitivity and specificity of CT measurements. Serum CT concentrations during pregnancy are variable. They are high at birth compared to paired maternal CT concentrations (40).
Serum CT further increases during the first few days after birth and may reach levels five- to ten-fold higher than adult CT concentrations. Serum CT concentrations decrease progressively during infancy; however, in preterm infants up to 3 months after birth, the mean serum CT concentrations may remain twice the adult value. There is also a small peak of serum CT concentration during late childhood. In human adults, the basal serum CT concentration may be lower in women than in men, but the concentration is not affected by increasing age. The CT secretory response to Ca infusion is lower in women and with increasing age. In human adults, serum CT and CGRP concentrations are found in the picomolar range. Diurnal variability has been reported for serum CGRP but not for serum CT. In normal individuals, larger precursor molecules of CT such as procalcitonin are not detected.

CT function is mediated by binding to receptors linked to G proteins, members of the GPCR superfamily, and by activation of adenylate cyclase and phospholipase C (41). CT receptors (CTR) have been identified in the central nervous system, testes, skeletal muscle, lymphocytes, and placenta. The function of CTR can be influenced by accessory proteins, receptor isoforms, genetic polymorphisms, developmental and/or transcriptional regulation, feedback inhibition, and the specific cellular or tissue background. The CTR gene is located on chromosome 7q2l.2-q21.3 and encodes a 490-amino-acid G-protein-linked receptor with seven transmembrane domains. Two isoforms of human CTR arise by alternative splicing of an exon of 48 nucleotides that encodes a l6-amino-acid insertion within the first intracellular loop. The isoform with the insertion (hCTR-l) activates only adenylate cyclase, whereas the other isoform (hCTR-2) activates both adenylate cyclase and phospholipase C. CGRP functions are also mediated by receptors (42).

The presence of receptor-activity-modifying proteins (RAMPs) can posttranslationally modify the initially orphan calcitonin receptor-like (CL) receptor and CTR to exhibit different receptor functions, i.e., functional modification of G-protein-coupled receptors is possible. RAMP1, 2, and 3 thus far identified are single transmembrane domain proteins with intracellular C-terminals of up to 10 amino acids and extracellular N-terminals of about 120 amino acids. Noncovalent association of a RAMP with the CL receptor or CTR results in heterodimeric RAMP/receptor complexes at the cell surface. The CL receptor functions as a CGRP receptor when coexpressed with RAMP1, whereas CL receptor/RAMP2 and CL receptor/RAMP3 are adrenomedullin 1 and 2 receptor subtypes, respectively. CTR, with 60% homology to the CL receptor, predominantly recognizes CT in the absence of RAMP. When CTR is coexpressed with RAMP1, it transforms into an amylin/ CGRP receptor. When CTR is coexpressed with RAMP3, it interacts only with amylin. Thus, two class II G-protein- coupled receptors, the CL receptor and CTR, are associated with three RAMPs to form high-affinity receptors for CGRP, adrenomedullin, or amylin.

Secretion of CT is stimulated by an increase in serum Ca and Mg concentrations and by gastrin, glucagon, and cholecystokinin (along with several other structural analogues of these hormones, e.g., pentagastrin, prostaglandin E2), glucocorticoid, norepinephrine, and CGRP; secretion is suppressed by hypocalcemia, propranolol and other adrenergic antagonists, somatostatin, chromogranin A and vitamin D. CT gene transcription is positively regulated by glucocorticoids and negatively regulated by PKC, Ca, and vitamin D. Calcitonin may activate the l-hydroxylase system independent of PTH (43), whereas 1,25(OH)₂D decreases CT gene expression in adult rats but is ineffective in 13-day-old suckling rats (44). The latter observation may be related to fewer 1,25(OH)₂D receptors in C cells of immature rats. Calcitonin induces refractoriness to its own actions by downregulation, and functional reduction of receptor mRNA is a well-known phenomenon. Clinically, it is manifested as the “escape” phenomenon during therapy with calcitonin.

In humans, changes in Ca (and P) metabolism are not seen despite extreme variations in CT production. In the neonate, there is neither an identifiable hypocalcemic response to the postnatal surge in serum CT nor a blunting of CT secretion in the presence of hypocalcemia. In adults, there are no definite effects attributable to CT deficiency (e.g., totally thyroidectomized patients receiving only replacement thyroxin) or CT excess (e.g., patients with medullary carcinoma of thyroid), except for the chronic suppression of bone remodeling. The clinical significance of CT is related to its use as a tumor marker in the management of medullary carcinoma of the thyroid and its pharmacologic effect to inhibit osteoclast-mediated bone resorption and increase renal Ca clearance. The pharmacologic activities of CT are useful for the suppression of bone resorption in Paget disease, for limited use in the treatment of osteoporosis, and for early phase treatment of severe hypercalcemia. In addition, CT also increases renal clearance of Mg, P, and sodium and free water clearance. The net effect of CT is a lowering of serum Ca and P concentrations. Thus, the bioactivity of CT on calcium metabolism frequently is opposite that of PTH; CT probably modulates the effect of PTH on target organs.

Noncalcium-related actions of CT and associated molecules are increasingly being discovered. For example, CT and CTR may play important roles in a variety of processes as wide ranging as embryonic development and sperm function/physiology. In addition, pro-CT detectable in the plasma is not produced in the C cells of the thyroid and is being explored as a marker of bacterial-induced inflammation/sepsis. Production of pro-CT after exposure to bacterial endotoxin and inflammatory cytokines tumor necrosis factor (TNF) and IL-6 appears to be primarily from neuroendocrine cells in the lung and intestine. Cells of neuroendocrine origin express all proteins related to CT (CGRP-I and -II and amylin) derived from the same family of genes, and it is speculated that “inflammatory” pro-CT may be coded by the same gene family. There are no enzymes in the plasma that can break down pro-CT, and when it is secreted into the circulation, it has a half-life of 25 to 30 hours, thus increasing serum pro-CT. After
administration of endotoxin, the peak circulating concentrations of TNF, IL-6, pro-CT, and C-reactive protein occur at approximately 90 minutes, 180 minutes, 6 to 8 hours, and 24 hours, respectively.

CGRP primarily affects catecholamine release, vascular tone and blood pressure, and cardiac contractility. Its clinical role probably also lies in its potential pharmacologic effect. The influence of CGRP on Ca and P homeostasis is minor compared to that of CT. However, amylin, a pancreatic islet-derived or synthetic 37-amino-acid peptide, is a member of the CGRP family with a potent hypocalcemic effect despite sharing only 15% of its amino acid sequence with human CT. The hypocalcemic effect of amylin is probably mediated by the CTR on osteoclasts, and it is 100-fold more potent than CGRP (45). Both CT and CGRP inhibit gastric acid secretion and food intake.

Vitamin D (M_r 384) can be obtained from diet or synthesized endogenously. It must undergo several metabolic transformations primarily in the liver and kidney to form the physiologically most important metabolite, 1,25(OH)₂D, which functions as a hormone in the maintenance of mineral homeostasis. Under in vivo conditions, there are more than 30 other vitamin D metabolites, with and without putative functions.

Dietary vitamin D (1 μg =40 IU) is derived from plants as ergocalciferol (vitamin D₂) and from animals as cholecalciferol (vitamin D₃). Dietary vitamin D is absorbed from the duodenum and jejunum into lymphatics, and about 50% of the vitamin D in chylomicron is transferred to vitamin D-binding protein (DBP) in blood before uptake by the liver.

In animals, vitamin D₃ can be synthesized endogenously in the skin (46). During exposure to sunlight, the high-energy UV photons (290 to 315 nm) penetrate the epidermis and photochemically cleave the bond between carbons 9 and 10 of the sterol B-ring of 7-dehydrocholesterol (7-DHC or provitamin D₃) to produce previtamin D₃. It then undergoes a thermally induced isomerization to vitamin D₃ that takes 2 to 3 days to reach completion. Thus, cutaneous synthesis of vitamin D₃ continues for many hours after a single sun exposure. Previtamin D₃ is photolabile; continued exposure to sunlight causes the isomerization of previtamin D₃ to biologically inert products, principally lumisterol. No more than 10% to 20% of the initial provitamin D₃ concentrations ultimately end up as previtamin D₃, thus preventing excessive production of previtamin D₃ and vitamin D₃.

Vitamin D₃ synthesis in the skin is directly dependent on the amount of sunlight exposure and is affected by time of day, season, and latitude. Peak sunlight at midday, in summer, and at lower latitudes are optimal conditions; the amount of skin area exposed and duration of sunlight exposure directly affect vitamin D₃ synthesis. Melanin in the skin competes with 7-DHC for ultraviolet photons, but the production of vitamin D₃ can be adjusted by increasing the duration of sunlight exposure. Use of topical sunscreen blocks ultraviolet photons, and aging decreases the capacity for cutaneous synthesis of vitamin D₃.

The term “vitamin D” is frequently used generically to describe vitamins D₂ and D₃ and their metabolites. In mammals, vitamins D₂ and D₃ appear to metabolize along the same pathway, and there is little functional difference between their metabolites. However, differences in affinity to DBP and receptors between D₂ and D₃ and their metabolites support the contention that vitamin D₃ is more bioavailable than D₂.

In the circulation, vitamin D and its metabolites are bound to proteins, mainly DBP (approximately 85%) and albumin (approximately 15%). The DBP gene is located on chromosome 4q11-13. It is a member of the albumin multigene family of proteins that includes albumin and α-fetoprotein. DBP is an approximately 53-kDa globulin in humans, and its x-ray crystallographic structure has been determined (47). Plasma DBP concentration (4 to 8 μM) is approximately 20-fold higher than that of the total circulating vitamin D metabolites (approximately 100 nM), i.e., it is normally less than 5% saturated with vitamin D metabolite. The amount of unbound or free 25-OHD and 1,25(OH)₂D, important in determining bioactivity, is less than 1% of the total concentration.

In the liver, vitamin D is hydroxylated at carbon 25 to 25-OHD. Quantitatively, 25-OHD (1 nmol/L =0.4 ng/mL) is the most abundant vitamin D metabolite in the circulation and is a useful index of vitamin D reserve. Regulation of 25-hydroxylase activity is limited, and there are few limitations to the production of 25-OHD. However, in vivo administration of 1,25(OH)₂D (48) inhibits hepatic production of 25-OHD, and Ca deficiency (49) increase the metabolic clearance of 25-OHD with subsequently decrease circulating 25-OHD.

In the kidney, 25-OHD is hydroxylated further to 1,25(OH)₂D by 25-OHD-1α-hydroxylase (CYP1α), and to 24,25-dihydroxyvitamin D (24,25(OH)₂D) by 25-OHD-24-hydroxylase (CYP24). The hydroxylation occurs primarily in the mitochondria of renal proximal tubules. The genes for these enzymes have been localized to chromosome 12q13-14 and 20q13.3, respectively. The human gene encoding CYP1α is 5 kb in length and is comprised of nine exons and eight introns; its exon/intron organization is similar to other cloned mitochondrial P450 enzymes.

The activity of CYP1α and therefore production of 1,25(OH)₂D are tightly regulated. It is the rate-limiting hormonally regulated step in the bioactivation of vitamin D. PTH increases transcriptional activity of the CYP1α gene promoter and therefore increases mRNA for 1,25(OH)₂D. Decreases in serum or dietary Ca or P increase mRNA for and serum concentration of 1,25(OH)₂D independent of PTH (49,50,51). However, hypophosphatemia in renal wasting disorders does not elicit appropriate phosphate conservation or an increase in 1,25(OH)₂D production. These disorders include X-linked hypophosphatemic rickets (XLH), autosomal-dominant hypophosphatemic rickets (ADHR), and tumor-induced osteomalacia. They have
similar phenotypic manifestations characterized by hypophosphatemia, decreased renal phosphate reabsorption, normal (and thus inappropriately low) or low serum calcitriol concentrations, normal serum Ca and PTH, and defective skeletal mineralization.

XLH results from mutations in the PHEX (phosphate-regulating gene with homologies to endopeptidases on the X chromosome, Xp22.1) gene, which encodes a membrane-bound endopeptidase (52), whereas ADHR is associated with mutations of the gene encoding fibroblast growth factor 23 (FGF23) and is linked to chromosome 12p13.3 (53). The latter is a small heat-sensitive molecule less than 25 kDa that inhibits sodium-dependent phosphate wasting and probably inhibits CYP1α. The endopeptidase, PHEX, degrades native FGF23 which provides the biochemical link among these clinical syndromes. XLH rickets also has been associated with mutations in CLCN5, a voltage-gated chloride channel gene located on Xp11.22.

Other factors that enhance 1,25(OH)₂D production include estrogen, prolactin, growth hormone, insulin-like growth factor I, and PTHrP. 1,25(OH)₂D production is feedback regulated and is inhibited by chronic deficiency of or low circulating Mg (36). Mg deficiency also lowers serum 1,25(OH)₂D response to a low-Ca diet but does not appear to limit 1,25(OH)₂D production in animals (54). The effect of Mg on 1,25(OH)₂D metabolism is presumably related in part to its role as a cofactor of the CYP1α enzyme. In contrast to the rapid increase within minutes in the secretion and serum concentration of PTH, measurable alterations in serum 1,25(OH)₂D concentrations usually occur hours after exposure to an appropriate stimulus. Extrarenal production of 1,25(OH)₂D in macrophages, particularly in granulomatous disease states, may not be tightly regulated; is stimulated by γ-interferon (55), but is not responsive to changes in dietary Ca intake.

The degradation of 1,25(OH)₂D is also tightly regulated. 1,25(OH)₂D strongly induces the enzyme 25-hydroxyvitamin D-24 hydroxylase (CYP24) in all target cells for vitamin D. CYP24 catalyzes several steps of 1,25(OH)₂D degradation, collectively known as the C24 oxidation pathway, which starts with 24-hydroxylation and culminates in the formation of the biliary excretory form, calcitroic acid. CYP24 expression is inhibited by PTH and by dietary phosphate restriction. In the kidney and intestine in particular, upregulation of the CYP24 enzyme in response to 1,25(OH)₂D treatment is rapid and occurs within 4 hours (56). Physiologic production of 24,25(OH)₂D is therefore an important means to regulate the circulating concentration of 1,25(OH)₂D and catabolism of vitamin D, although it may also have a role in bone integrity and fracture healing in the chick model. Most of the other vitamin D metabolites are derived primarily from further metabolic alterations to 25-OHD and 1,25(OH)₂D through oxidation or side chain cleavage and have poorly defined physiologic functions. However, many analogues of vitamin D metabolites are being studied for the numerous potential pharmacologic actions that involve less calcemic-inducing and more maturation and differentiation effects.

Like other steroid hormones, 1,25(OH)₂D function is mediated primarily through modulation of the cellular genome by binding to a specific nuclear receptors, vitamin D receptor, VDR, a 424-amino-acid phosphoprotein for which the x-ray crystallographic structure has been determined (47). The VDR gene contains nine exons and is located on chromosome 12q13-14 near the site of the gene for CYPlα (57). VDR is a member of the subfamily of nuclear receptors with ligand-binding domains that bind classic hormones including thyroid hormone, androgens, estrogens, progesterone, glucocorticoids, aldosterone, hormonal forms of vitamin A, and 1,25(OH)₂D. It has several functional domains including a 110-residue N-terminal DNA-binding domain with two zinc fingers, C-terminal hormone-binding domain, and hinge region important for nuclear localization. The VDR interacts with the 9-cis retinoic acid nuclear receptor retinoid X receptor (RXR) to form a heterodimeric RXR-VDR complex that binds to specific DNA sequences, termed vitamin D-responsive elements (VDREs). After 1,25(OH)₂D binds to the receptor, it induces conformational changes (58) that result in the recruitment of a multitude of transcriptional coactivators that stimulate the transcription of target genes. VDR also can adopt a dual role as a repressor in the absence of ligand and then subsequently as a coactivator when a ligand binds. VDR is upregulated by 1,25(OH)₂D at both the mRNA and protein levels and is increased during growth, gestation, and lactation; however, it shows an age-dependent decrease in mature animals and humans, supporting the notion that VDR may be up- or downregulated, depending on Ca needs.

Although 1,25(OH)₂D regulates more than 60 genes whose actions include those associated with Ca homeostasis and immune responses, as well as cellular growth, differentiation, and apoptosis, two basic clinical functions define the major classic actions of vitamin D. The first is that vitamin D is required to prevent rickets in children and osteomalacia in adults. The second is the prevention of hypocalcemic tetany. These functions are maintained by 1,25(OH)₂D through its effect on a number of target tissues, primarily intestine, kidney, and bone, with modulating effects from other hormones including PTH and CT.

The genomic action of 1,25(OH)₂D can be preceded by more rapid nongenomic actions that occur in minutes and involve membrane-associated events such as increased Ca transport, and PKC and mitogen-activated protein kinase activation. This nongenomic rapid increase in cytosolic Ca within seconds to minutes is reported to occur in various cell types from the intestine and parathyroid, osteoblasts, myocytes, and leukemic cells (59).

Quantification of vitamin D and its metabolites has been achieved by several different methods, including high-performance liquid chromatography, with detection by ultraviolet absorbance or binding assays, and immunoas- says based on antibodies raised to vitamin D metabolite conjugates. Values from different laboratories cannot be compared without making direct comparison of their assay procedures. Interlaboratory coefficients of variation for the
measurement of 25-OHD, 24,25(OH)₂D, and l,25(OH)₂D may range between 35% and 52%. Furthermore, differences between the affinities of vitamins D₂ and D₃ to DBP and VDR, and different chromatographic behavior on various preparative chromatographic systems demand that great care be taken with assay techniques when dealing with patients who have significant vitamin D₂ intake. To ensure reliable results, appropriate vitamin D standards must be used for standard curve generation in performing competitive protein binding assays of these compounds.

Maternofetal transfer of vitamin D and its metabolites varies, depending on the species. In humans, the cord-serum vitamin D concentration is very low and may be undetectable; the 25-OHD concentration is directly correlated with, but is lower than, maternal values, consistent with placental crossover of this metabolite; and 1,25(OH)₂D concentrations also are lower than maternal values, but there is no agreement on the maternofetal relationship of this and other dihydroxylated vitamin D metabolites (3,60-62). However, the placenta, like the kidney, produces 1,25(OH)₂D, making it difficult to ascertain just how much fetal 1,25(OH)₂D results from placental crossover versus placental synthesis. 24,25(OH)₂D also crosses the placenta, and its concentration in the sera of mothers and newborns varies with the seasons, being highest in autumn. It appears that the human fetus receives the bulk of its vitamin D already metabolized to 25-OHD.

Seasonal and racial variations in serum 25-OHD concentrations occur, presumably from variations in endogenous production. The concentration of serum 25-OHD, as with 24,25(OH)₂D, is lower in winter. These changes may be reflected in cord-serum values. In normal adults, serum 1,25(OH)₂D concentrations are relatively constant and maintained within approximately 20% of the overall 24-hour mean, and show no seasonal variation, which is consistent with the tightly regulated CYPlα activity. African American mothers, infants, and young children tend to have lower 25-OHD and higher 1,25(OH)₂D concentrations than their white counterparts. Serum 1,25(OH)₂D in the newborn becomes elevated within 24 hours after delivery and appears to vary with to Ca and P intake.

The circulating half-life of vitamin D is about 24 hours and for 25-OHD is 2 to 3 weeks, although the latter half-life is decreased in vitamin D-deficient individuals. 1,25(OH)₂D has a much shorter half-life of 3 to 6 hours. Metabolites of 25-OHD and 1,25(OH)₂D may undergo enterohepatic circulation after exposure to intestinal β-glucuronidase. The physiologic role of enterohepatic circulation of vitamin D metabolites has not been precisely quantitated.

Neonatal hypocalcemia may be defined as a serum tCa concentration of less than 2 mmol/L (8 mg/dL) in term infants and 1.75 mmol/L (7 mg/dL) in preterm infants with iCa below 1.0 to 1.1 mmol/L (4.0 to 4.4 mg/dL), depending on the particular ion-selective electrode used. Whole-blood iCa shows similar values to serum iCa and is often used to determine hypocalcemia. However, the appropriate range used is also subject to the type of instrument used (15).

The definition of hypocalcemia is based on the clinical perspective because serum Ca concentrations are maintained within narrow ranges under normal circumstances, and the potential risk for disturbances of physiologic function increases as the serum Ca concentration decreases below the normal range. Furthermore, improvements in physiologic function, e.g., changes in cardiac contractility, blood pressure, and heart rate, are reported in hypocalcemic infants undergoing Ca therapy (70,71,72), and a higher mortality rate has been reported for children with hypocalcemia in pediatric intensive care settings (73).

Clinically, there are two peaks in the occurrence of neonatal hypocalcemia. An early form typically occurs during the first few days after birth, with the lowest concentrations of serum Ca being reached at 24 to 48 hours of age; late neonatal hypocalcemia occurs toward the end of the first week. These findings reflect in part the traditional clinical practice of screening for biochemical abnormalities in small or sick hospitalized infants during the first few days, and in symptomatic infants during hospitalization and after hospital discharge. However, the nadir of serum Ca concentration may occur at less than 12 hours (9 to 12) or not until some weeks after birth (74,75), and many neonates, particularly those with genetic defects in Ca metabolism, may be hypocalcemic but remain asymptomatic and undetected during the early neonatal period. This also may contribute to the less frequent diagnosis of late neonatal hypocalcemia compared to early neonatal hypocalcemia. Additionally, increased understanding of the mechanisms of disturbed Ca metabolism would support the approach to neonatal hypocalcemia based on risk factors and pathophysiologic basis rather than the traditional “early” or “late” onset.