J Nutr Sci Vitaminol, 58, 442–445, 2012

Note A Study of the Association between Serum -Specific Alkaline Phosphatase and Serum Phosphorus Concentration or Dietary Phosphorus Intake

Mayu Haraikawa1, Rieko Tanabe1, Natsuko Sogabe2, Aoi Sugimoto1, Yuka Kawamura1, Toshimi Michigami3, Takayuki Hosoi4 and Masae Goseki-Sone1,*

1 Department of Food and Nutrition, Faculty of Human Sciences and Design, Japan Women’s University, Bunkyo-ku, Tokyo 112–8681, Japan 2 Department of Health and Nutrition Sciences, Faculty of Human Health, Komazawa Women’s University, Tokyo 206–8511, Japan 3 Research Institute, Osaka Medical Center for Maternal and Child Health, Osaka 594–1101, Japan 4 Department of Clinical Research and Development, National Center for Geriatrics and Gerontology, Aichi 474–8511, Japan (Received March 27, 2012)

Summary Alkaline phosphatase (ALP) hydrolyzes a variety of monophosphate esters into phosphoric acid and alcohol at a high optimum pH (pH 8–10). Human ALPs are clas- sified into four types: tissue-non specific (TNSALP, liver/bone/), intestinal, placen- tal, and germ cell types. Based on studies of (HPP), which is a systemic caused by the presence of either one or two pathologic mutations in ALPL that encodes TNSALP, TNSALP was suggested to be indispensable for skeletal mineralization. In this study, we explored the possibility that dietary nutrients contribute to regulate serum bone-specific ALP (BAP) activity. Serum biochemical parameters, such as serum ALP, BAP, osteocalcin, and fibroblast growth factor 23 (FGF23), were measured in healthy young sub- jects (n5193). Dietary nutrient intakes were measured based on 3-d food records before the day of blood examinations. The presence of a carrier of the deletion of T at nucleotide 1559 (c.1559delT), which has been reported to be the most frequent in Japanese HPP, was not detected in any subject. By the analysis of BAP activity and other biochemical parameters or dietary nutrient intakes, we obtained significant correlations between BAP activity and serum phosphorus (r520.165, p50.022), intake (mg/1,000 kcal/d) (r520.186, p50.010), or phosphorus intake (mg/1,000 kcal/d) (r520.226, p50.002). Further study on the regulation of BAP activity and calcium and/or phosphorus will provide useful data for improving skeletal health. Key Words bone-specific alkaline phosphatase, hypophosphatasia, dietary nutrient intake, serum phosphorus, fibroblast growth factor 23

Alkaline phosphatase (ALP, EC 3.1.3.1) is an enzyme inborn error in bone metabolism involving the presence containing zinc which hydrolyzes monophosphate esters of either one or two pathologic mutations in ALPL and into phosphoric acid and alcohol at a high optimal pH bone and/or teeth in the presence of a low activity of (pH 8–10). In humans, there are at least four distinct serum and bone ALP (4). but related ALP types: tissue non-specific (TNSALP, More than 200 mutations of ALPL in HPP patients liver/bone/kidney), intestinal, placental, and germ cell have been identified to date (http://www.sesep.uvsq. types (1–3). fr/03_hypo_mutations.php). The deletion of T at TNSALP has been shown to be indispensable for bone nucleotide 1559 (1559delT) has been detected only in mineralization by controlling the concentration of inor- Japanese, and is the most frequent mutation of ALPL in ganic pyrophosphate (PPi), a potent calcification inhibi- Japanese HPP (8–13). Previously, we demonstrated that tor (4, 5). Recently, we identified a single nucleotide 1559delT causes the loss of ALP activity and results in polymorphism (SNP) in ALPL (787T.C) (rs3200254) the synthesis of an abnormal TNSALP molecule which that encodes TNSALP associated with the bone mineral is not expressed on the plasma membrane (10, 12). density (BMD) among 501 postmenopausal women (6, Nutrition management is useful for the control and 7). treatment of this disease. In the severe form of HPP, Hypophosphatasia (HPP) is a disease caused by a rare patients often suffer from intractable convulsions and tend to have hypercalcemia because of low-level bone * To whom correspondence should be addressed. formation. These complications were also observed in E-mail: [email protected] TNSALP knockout [Akp2(2/2)] mice (5). Treatment

442 Association between BAP and Phosphorus 443 with the administration of vitamin B6 is effective for Table 1. Correlations between BAP and serum param- convulsions, and low calcium-containing milk is recom- eters or dietary nutrient intakes. mended for hypercalcemia in HPP (14). In the present study, we aimed to clarify the associa- r values p-values tion between bone-specific alkaline phosphatase (BAP) activity and serum biochemical parameters or dietary Serum parameters nutrients to obtain basic information for the planning Calcium (mg/dL) 20.046 0.524 Phosphorus (mg/dL) 2 0.022* of desirable nutritional management for bone health. 0.165 ALP (U/L) 0.797 ,0.001*** Furthermore, we examined the carrier frequency of Osteocalcin (ng/dL) 0.339 ,0.001*** the common mutation c.1559delT in ALPL, which is FGF23 (pg/mL) 0.191 0.008** one of the most frequent ALPL mutations in the lethal Dietary nutrient intakes form of HPP in Japanese (15). Calcium (mg/1,000 kcal/d) 20.186 0.010* Phosphorus (mg/1,000 kcal/d) 20.226 0.002** Materials and Methods Vitamin D (mg/1,000 kcal/d) 0.032 0.655 Subjects. Young subjects living in Tokyo, Japan, Vitamin B6 (mg/1,000 kcal/d) 20.085 0.239 were recruited. Participants were excluded if they had metabolic disease. The study population consisted of BAP: bone-specific alkaline phosphatase, ALP: alkaline phosphatase, FGF23: fibroblast growth factor 23. 97 healthy Japanese males and 96 females. All subjects * p,0.05, ** p,0.01, *** p,0.001. were unrelated volunteers and were aged 22.161.8 (mean6standard deviation (SD)), with a height of 164.968.9 cm, weight of 57.269.2 kg, and body mass index (BMI) of 21.062.3 kg/m2. SD. Spearman rank correlation coefficients were calcu- In accordance with the Helsinki Declaration on lated to analyze the relation between two parameters. human studies, the study protocol was approved by the Significance was considered at p,0.05. Statistical anal- ethical committee of Japan Women’s University, and ysis was conducted using PASW (Version 18; SPSS, Inc., written informed consent was obtained from all study Chicago, IL, USA). subjects. Measurements. Fasting blood samples were obtained Results and Discussion and sera were kept frozen at 280˚C until measurement. We examined the presence of carriers of the deletion A bone formation marker, serum BAP, was determined of T at nucleotide 1559 (c.1559delT) in ALPL, which by enzyme immunoassay (Mitsubishi Kagaku Bio Clini- has been reported to be the most frequent in Japanese cal Laboratories Inc., Tokyo, Japan), which can detect HPP. The mutation c.1559delT was not detected in any only serum BAP activity in total ALP. ALP activity was subjects of this study, and no subjects were heterozy- determined employing the method of Bessey et al. (16). gous carriers of c.1559delT. Recently, it was reported Calcium was measured employing the o-cresol-phtha- that the 1559delT carrier frequency is 1/480 (95% con- lein complexone color development method (17), and fidence interval, 1/1,562–1/284), and ~1 in 900,000 inorganic phosphorus was determined using the method individuals have the perinatal lethal form of HPP caused of p-methylaminophenol reduction (18). Serum-intact by a homozygous 1559delT mutation in Japanese (15). osteocalcin [bone gamma-carboxyglutamic acid (Gla) Therefore, we considered that more subjects must be protein: BGP], which is one of the secreted - accumulated in order to reveal the frequency. specific proteins and is associated with the mineralized In all subjects (n5193), the mean (6SD) levels matrix of bone tissue (19), was measured by immuno- of serum BAP and ALP activity were 26.967.8 and radiometric assay (Mitsubishi Kagaku Bio Clinical 193648 U/L, respectively. The levels of serum calcium, Laboratories Inc.). Fibroblast growth factor (FGF23) phosphorus, osteocalcin, and FGF23 were 9.760.4 mg/ was measured employing the FGF23-enzyme-linked dL, 3.660.5 mg/dL, 7.963.1 ng/mL, and 43613 pg/ immunosorbent assay (Kainos Laboratories, Inc., Tokyo, mL, respectively. As the results of 3-d food records, Japan). the mean (6SD) dietary energy, calcium, phosphorus, Dietary nutrient intakes were measured based on 3-d vitamin D, and vitamin B6 intakes of the subjects were food records taken up to the day before blood collection. calculated as 2,0786555 (kcal/d), 5566223 (mg/d), Trained personnel reviewed the food records, and the 1,0596302 (mg/d), 5.864.8 (mg/d), and 3.10611.08 nutrient content was determined with the use of Eiyo- (mg/d), respectively. Kun software (Kenpaku-sha, Japan). We analyzed the relations between BAP activity and All subjects were analyzed for ALPL mutation (a dele- serum parameters or nutrient intakes. As shown in Table tion of T at nucleotide position 1559: 1559delT). DNA 1, serum BAP activity was significantly correlated with was extracted from whole blood (QIAamp DNA Blood the concentration of serum phosphorus (r520.165, Kit, Qiagen). The segment of ALPL including mutation p50.022), osteocalcin (r50.339, p,0.001) and FGF23 sites was amplified by polymerase chain reaction (PCR) (r50.191, p50.008), but not with serum calcium. (13). 1559delT was determined by the digestion of the Serum BAP activity was also correlated with the cal- PCR amplified fragment with DdeI. cium (mg/1,000 kcal/d) (r520.186, p50.010), and Statistical analysis. Values are shown as the mean6 phosphorus (mg/1,000 kcal/d) (r520.226, p50.002) 444 Haraikawa M et al.

Table 2. Correlations between serum phosphorus and References dietary nutrient intakes. 1) Weiss MJ, Ray K, Henthorn PS, Lamb B, Kadesch T, r values p-values Harris H. 1988. Structure of the human liver/bone/ kidney alkaline phosphatase gene. J Biol Chem 263: Calcium (mg/1,000 kcal/d) 0.063 0.380 12002–12010. Phosphorus (mg/1,000 kcal/d) 0.183 0.011* 2) Henthorn PS, Raducha M, Kadesch T, Weiss MJ, Harris Vitamin D (mg/1,000 kcal/d) 20.038 0.598 H. 1988. Sequence and characterization of the human Vitamin B6 (mg/1,000 kcal/d) 0.012 0.868 intestinal alkaline phosphatase gene. J Biol Chem 263: 12011–12019. * p,0.05. 3) Knoll BJ, Rothblum KN, Longley M. 1988. Nucleotide sequence of the human placental alkaline phosphatase gene. Evolution of the 5′ flanking region by deletion/ intakes, but not with the vitamin D (mg/1,000 kcal/d) substitution. J Biol Chem 263: 12020–12027. nor vitamin B6 (mg/1,000 kcal/d) intake. 4) Whyte MP. 1989. Alkaline phosphatase: physiological We suggest that the difference in dietary phospho- role explored in hypophosphatasia. In: Bone and Mineral rus intake influenced the level of BAP activity or serum Research (Peck WA, ed), Vol. 6, p 175–218. Elesevier phosphorus. Actually, the concentration of serum Science Publishers, Amsterdam. 5) Fedde KN, Blair L, Silverstein J, Coburn SP, Ryan LM, phosphorus was significantly correlated with the Weinstein RS, Waymire K, Narisawa S, Millán JL, Mac- dietary phosphorus intake (mg/1,000 kcal/d) (Table 2) Gregor GR, Whyte MP. 1999. Alkaline phosphatase (r50.183, p50.011), but there were no significant cor- knock-out mice recapitulate the metabolic and skeletal relations between serum phosphorus and the calcium, defects of infantile hypophosphatasia. J Bone Miner Res vitamin D, or vitamin B6 intake. 14: 2015–2026. Previously, we reported that phosphate starvation 6) Goseki-Sone M, Sogabe N, Fukushi-Irie M, Mizoi L, increased BAP activity and regulated its expression in Orimo H, Suzuki T, Nakamura H, Orimo H, Hosoi T. mouse stromal ST2 cells (20, 21). In mineralizing tis- 2005. Functional analysis of the single nucleotide poly- sues, BAP may regulate extracellular PPi concentrations morphism (787T.C) in the tissue-nonspecific alkaline by hydrolyzing PPi, which is an inhibitor of the forma- phosphatase gene associated with BMD. J Bone Miner Res tion of crystals (22). These findings sug- 20: 773–782. 7) Sogabe N, Oda K, Nakamura H, Orimo H, Watanabe H, gest that the extracellular concentration of phosphate is Hosoi T, Goseki-Sone M. 2008. Molecular effects of the one of the factors which modulate BAP activity. tissue-nonspecific alkaline phosphatase gene polymor- Furthermore, there were significant negative cor- phism (787T.C) associated with bone mineral density. relations between the serum level of BAP and serum Biomed Res 29: 213–219. phosphorus or phosphorus intake. Kemi et al. reported 8) Orimo H, Hayashi Z, Watanabe A, Hirayama T, Shimada that high phosphorus intake adversely affects bone T. 1994. Novel missense and frameshift mutations in the metabolism by decreasing the marker of bone forma- tissue-nonspecific alkaline phosphatase gene in a Japa- tion, BAP activity, in healthy young females (23). In the nese patient with hypophosphatasia. Hum Mol Genet 3: present study, a significant positive correlation was also 1683–1684. observed between BAP and the concentration of osteo- 9) Goseki-Sone M, Orimo H, Iimura T, Takagi Y, Watanabe calcin or FGF23 (Table 1). FGF23 is a 32-kDa peptide H, Taketa K, Sato S, Mayanagi H, Shimada T, Oida S. 1998. Hypophosphatasia: identification of five novel secreted by osteocytes involved in the control of phos- missense mutations (G507A, G705A, A748G, T1155C, phate homeostasis (24). Although there are limitations G1320A) in the tissue-nonspecific alkaline phosphatase due to the small sample size and results of single cor- gene among Japanese patients. Hum Mutat 1: 263–267. relation analysis, our results indicated that the dietary 10) Goseki-Sone M, Orimo H, Iimura T, Miyazaki H, Oda assessment of phosphorus intake is needed to evaluate K, Shibata H, Yanagishita M, Takagi Y, Watanabe H, bone metabolism and phosphorus homeostasis. Shimada T, Oida S. 1998. Expression of the mutant Concerning the relationship between serum BAP (1735T-DEL) tissue-nonspecific alkaline phosphatase activity and calcium intake, Fardellone et al. reported gene from hypophosphatasia patients. J Bone Miner Res that calcium supplementation resulted in a significant 13: 1827–1834. reduction of BAP activity and bone-resorption markers 11) Orimo H, Goseki-Sone M, Inoue M, Tsubakio Y, Saki- in postmenopausal women (25). In the present study, we yama T, Shimada T. 2002. Importance of deletion of T at nucleotide 1559 in the tissue-nonspecific alkaline demonstrated that serum BAP activity was significantly phosphatase gene in Japanese patients with hypophos- and negatively correlated with calcium intake (Table 1). phatasia. J Bone Miner Metab 20: 28–33. In conclusion, our results demonstrated that there 12) Komaru K, Ishida Y, Amaya Y, Goseki-Sone M, Orimo H, were significant correlations between BAP activity and Oda K. 2005. Novel aggregate formation of a frame-shift serum phosphorus, calcium intake, or phosphorus mutant protein of tissue-nonspecific alkaline phospha- intake. Further investigations of the effect of adequate tase is ascribed to three residues in the C-ter- nutritional management on the regulation of BAP minal extension. Retarded secretion and proteasomal activity and calcium and/or phosphorus homeostasis degradation. FEBS J 272: 1704–1717. will provide useful data for improving skeletal health. 13) Michigami T, Uchihashi T, Suzuki A, Tachikawa K, Nakajima S, Ozono K. 2005. Common mutations F310L Association between BAP and Phosphorus 445

and T1559del in the tissue-nonspecific alkaline phos- 19) Price PA. 1988. Role of vitamin-K-dependent proteins in phatase gene are related to distinct phenotypes in Japa- bone metabolism. Annu Rev Nutr 8: 565–583. nese patients with hypophosphatasia. Eur J Pediatr 164: 20) Goseki-Sone M, Yamada A, Asahi K, Hirota A, Ezawa I, 277–282. Iimura T. 1999. Phosphate depletion enhances tissue- 14) Ozono K, Michigami T. 2011. Hypophosphatasia now nonspecific alkaline phosphatase gene expression in a draws more attention of both clinicians and researchers: cultured mouse marrow stromal cell line ST2. Biochem a commentary on prevalence of c.1559delT in ALPL, Biophys Res Commun 265: 24–28. a common mutation resulting in the perinatal (lethal) 21) Goseki-Sone M, Yamada A, Hamatani R, Mizoi L, Iimura form of hypophosphatasias in Japanese and effects of T, Ezawa I. 2002. Phosphate depletion enhances bone the mutation on heterozygous carriers. J Hum Genet 56: morphogenetic protein-4 gene expression in a cultured 174–176. mouse marrow stromal cell line ST2. Biochem Biophys 15) Watanabe A, Karasugi T, Sawai H, Naing BT, Ikegawa S, Res Commun 299: 395–399. Orimo H, Shimada T. 2011. Prevalence of c.1559delT 22) Millán JL. 2006. Mammalian Alkaline Phosphatases. in ALPL, a common mutation resulting in the perina- Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim. tal (lethal) form of hypophosphatasia in Japanese and 23) Kemi VE, Kärkkäinen MUM, Lamberg-Allardt CJE. 2006. effects of the mutation on heterozygous carriers. J Hum High phosphorus intakes acutely and negatively affect Genet 56: 166–168. Ca and bone metabolism in a dose-dependent manner in 16) Bessey OA, Lowry OH, Brock MJ. 1946. A method for healthy young females. Br J Nutr 96: 545–552. the rapid determination of alkaline phosphates with five 24) Martin A, David V, Quarles LD. 2012. Regulation and cubic millimeter of serum. J Biol Chem 164: 321–329. function of the FGF23/klotho endocrine pathways. 17) Gitelman HJ. 1967. An improved automated procedure Physiol Rev 92: 131–155. for the determination of calcium in biological speci- 25) Fardellone P, Brazier M, Kamel S, Guéris J, Graulet AM, mens. Anal Biochem 18: 521–531. Liénard J, Sebert JL. 1998. Biochemical effects of cal- 18) Drewes PA. 1972. Direct colorimetric determination cium supplementation in postmenopausal women: of phosphorus in serum and . Clin Chim Acta 39: influence of dietary calcium intake. Am J Clin Nutr 67: 81–88. 1273–1278.