Am J Physiol Gastrointest Liver Physiol 308: G206–G216, 2015. First published December 4, 2014; doi:10.1152/ajpgi.00093.2014.

Segmental transport of Ca2⫹ and Mg2⫹ along the gastrointestinal tract Anke L. Lameris,1 Pasi I. Nevalainen,2,3 Daphne Reijnen,4 Ellen Simons,1 Jelle Eygensteyn,5 X Leo Monnens,1 René J. M. Bindels,1 and Joost G. J. Hoenderop1 1

Department of Physiology, Radboud University Medical Center, Nijmegen, The Netherlands; 2School of Medicine, University of Tampere, Tampere, Finland; 3Department of Internal Medicine, Tampere University Hospital, Tampere, Finland; 4Central Animal Facility, Radboud University, Nijmegen, The Netherlands; and 5Department of General Instrumentation, Faculty of Sciences, Radboud University, Nijmegen, The Netherlands Submitted 12 March 2014; accepted in final form 23 November 2014

Lameris AL, Nevalainen PI, Reijnen D, Simons E, Eygensteyn J, Monnens L, Bindels RJ, Hoenderop JG. Segmental transport of Ca2⫹ and Mg2⫹ along the gastrointestinal tract. Am J Physiol Gastrointest Liver Physiol 308: G206 –G216, 2015. First published December 4, 2014; doi:10.1152/ajpgi.00093.2014.—Calcium (Ca2⫹) and magnesium (Mg2⫹) ions are involved in many vital physiological functions. Since dietary intake is the only source of minerals for the body, intestinal absorption is essential for normal homeostatic levels. The aim of this study was to characterize the absorption of Ca2⫹ as well as Mg2⫹ along the gastrointestinal tract at a molecular and functional level. In both humans and mice the Ca2⫹ channel transient receptor potential vanilloid subtype 6 (TRPV6) is expressed in the proximal intestinal segments, whereas Mg2⫹ channel transient receptor potential melastatin subtype 6 (TRPM6) is expressed in the distal parts of the intestine. A method was established to measure the rate of Mg2⫹ absorption from the intestine in a time-dependent manner by use of 25Mg2⫹. In addition, local absorption of Ca2⫹ and Mg2⫹ in different segments of the intestine of mice was determined by using surgically implanted intestinal cannulas. By these methods, it was demonstrated that intestinal absorption of Mg2⫹ is regulated by dietary needs in a vitamin D-independent manner. Also, it was shown that at low luminal concentrations, favoring transcellular absorption, Ca2⫹ transport mainly takes place in the proximal segments of the intestine, whereas Mg2⫹ absorption predominantly occurs in the distal part of the gastrointestinal tract. Vitamin D treatment of mice increased serum Mg2⫹ levels and 24-h urinary Mg2⫹ excretion, but not intestinal absorption of 25Mg2⫹. Segmental cannulation of the intestine and time-dependent absorption studies using 25Mg2⫹ provide new ways to study intestinal Mg2⫹ absorption. stable magnesium isotopes; intestinal absorption; TRPM6; TRPV6; local absorption

(Ca2⫹) and magnesium (Mg2⫹) are essential for many physiological functions, such as muscle contraction, neuronal excitability, bone formation, and certain enzymatic activity. It is, therefore, of vital importance that the concentration throughout the body is tightly regulated (17). For Ca2⫹ as well as Mg2⫹, this regulation takes place by the concerted action of the intestine, where these divalents are absorbed from the diet; the bones, where Ca2⫹ and Mg2⫹ are stored; and the kidneys, where they are filtered from the blood and reabsorbed from the pro-urine (43). Since our dietary intake is the only source of minerals for the body, intestinal absorption is of key importance for the maintenance of normal homeostatic levels. Intestinal absorption of Ca2⫹ and Mg2⫹ can take place via a paracellular and/or a transcellular pathway (40). Paracellular transport takes place CALCIUM

Address for reprint requests and other correspondence: J. G. Hoenderop, Dept. of Physiology (286), Radboud Univ. Medical Center, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands (e-mail: [email protected]). G206

via tight junctions and is driven by the electrochemical gradient across the epithelium. The transepithelial electrical potential difference is lumen negative (24). The luminal concentrations of Ca2⫹ and Mg2⫹ depend on their dietary contents and are in general sufficient to provide a transepithelial concentration gradient that further enhances absorption (32). Transcellular absorption of Ca2⫹ and Mg2⫹ from the intestinal lumen is facilitated by members of the transient receptor potential (TRP) ion channel family, which represents the rate-limiting factor in the transcellular transport process. Intestinal absorption of Ca2⫹ is mediated via TRP vanilloid type 6 (TRPV6), whereas absorption of Mg2⫹ takes place via TRP melastatin type 6 (TRPM6) (20, 30, 43). Other key players in transcellular Ca2⫹ transport are calbindin-D9K and plasma membrane Ca2⫹ATPase 1b (PMCA1b), involved in intracellular Ca2⫹ buffering and basolateral Ca2⫹ transport, respectively (19). For transcellular Mg2⫹ transport in the intestine TRPM7 and cyclin M4 (CNNM4) seem to be important in addition to TRPM6, since they are though to play a role in apical and basolateral transport of Mg2⫹, respectively (38, 47). The relative contribution of the paracellular and transcellular pathway to the total amount of Ca2⫹ and Mg2⫹ that is absorbed from the intestine depends on dietary supply and bodily needs (40). Although occasionally debated, paracellular absorption is generally considered to be predominant under normal physiological circumstances (3, 4, 32, 40, 44). When paracellular absorption cannot meet the daily need of the body, transcellular absorption becomes essential (6, 40). Specific transport proteins, located in the luminal and basolateral membrane of the epithelial cells, govern this latter process (23, 26, 32). Previous studies in mice have shown that TRPV6 is primarily expressed in duodenum, cecum, and colon, whereas TRPM6 was only found in cecum and colon (14). This finding suggests that transcellular Ca2⫹ and Mg2⫹ absorption takes place only in distinct segments of the gastrointestinal tract. Functional data supporting this hypothesis are, however, lacking. Although the regulation of intestinal Ca2⫹ absorption by hormones is well documented, the role of hormones, such as vitamin D, in Mg2⫹ transport remains under debate (16). The aim of the present study was, therefore, to characterize the absorption of Ca2⫹ and Mg2⫹ along the gastrointestinal tract at the molecular and functional level. To this end, the expression patterns of TRPV6 and TRPM6 along the gastrointestinal tract of humans as well as mice were determined. In addition, methods were developed to measure the rate of Mg2⫹ absorption from various segments of intestine in a time-dependent manner. To this end, a technique was developed to measure local absorption at different segments of the intestine by using surgically implanted intestinal cannulas. The intestinal absorption of Ca2⫹ was studied with the use of the

0193-1857/15 Copyright © 2015 the American Physiological Society

http://www.ajpgi.org

G207

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

radioactive 45Ca2⫹ isotope. Although there are ample radioactive Mg2⫹ isotopes, none of them is particularly suitable for absorption studies because of their short half-lives, which range from nanoseconds to hours (22). In addition to the radioactive isotopes, three stable Mg2⫹ isotopes exist. These stable isotopes, i.e., 24Mg, 25Mg, and 26Mg, have a respective natural abundance of 79, 10, and 11% and can be distinguished by inductively coupled plasma mass spectrometry (ICP-MS) (39). Enriched (⫾ 99%) preparations of a 25Mg2⫹ or 26Mg2⫹ are commercially available. Absorption of the enriched Mg2⫹ isotope will change the isotopic ratio compared with the normal natural abundance (8). Both 25Mg2⫹ and 26Mg2⫹ have been used to study Mg2⫹ homeostasis in humans as well as in animal models (1, 9). The most commonly used and wellestablished method to determine Mg2⫹ absorption by stable isotopes techniques is via fecal monitoring (1, 7). However, this technique will underestimate intestinal absorption since part of the absorbed Mg2⫹ is secreted into the intestinal lumen via for example bile or pancreatic excretions. To correct for intestinal secretion, the double-labeling technique has been developed in which net absorption is measured by correcting for fecal excretion of an additional intravenously administered isotope (8). Both the single- and the double-labeling technique are laborious and, more importantly, neither of the techniques provides detailed information on time-dependent absorption kinetics. Thus in the present study stable 25Mg2⫹ isotopes were employed to measure intestinal Mg2⫹ absorption in a time- and concentration-dependent manner. The effect of dietary Mg2⫹ content on intestinal absorption was determined. In addition, with the use of intestinal cannulas the absorption of Ca2⫹ as well as Mg2⫹ was analyzed at in the proximal and distal regions of the gastrointestinal tract, at several concentrations, to determine the location of paracellular and transcellular absorption. Finally, the effect of the potential magnesiotropic hormone vitamin D on Mg2⫹ homeostasis was investigated. MATERIALS AND METHODS

Patient selection, sample collection, and ethical considerations. For the expression gradients, human mucosal gastrointestinal biopsies

were collected from the esophagus, fundus, antrum, duodenum (3–5 cm beyond the pylorus), terminal ileum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum. All biopsies were taken from healthy mucosa. Gastroscopies and colonoscopies were performed during routine patient care when needed for clinical diagnosis. The jejunum could not be reached for sample collection. None of the patients from whom biopsies was taken were pregnant, diabetic, or receiving estrogen substitution. One patient used 20 mg esomeprazole once daily. Patients were selected during routine patient care in Finland (Tampere University Hospital). Exclusion criteria included active hyperparathyroidism, kidney insufficiency (serum creatinine ⬎200 ␮M), and liver insufficiency (international normalized ratio, spontaneously ⬎1.5). The additional biopsy samples taken for the study did not increase the complication risk of patients. This study was approved by the ethical committee of Tampere University Hospital, Finland. Written, informed consent was obtained from each subject. Studies were carried out in accordance with the Declaration of Helsinki. Quantitative real-time PCR. Total RNA was extracted from tissues by using TRIzol reagent (Invitrogen, Paisley, UK) according to the manufacturer’s protocol. The obtained RNA was subjected to DNase treatment (Promega, Leiden, The Netherlands) to prevent genomic DNA contamination. Subsequently, RNA was reverse transcribed with murine leukemia virus reverse transcriptase. The obtained cDNA was used to determine mRNA expression levels of various calciotropic and magnesiotropic genes, as well as mRNA levels of glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as an endogenous control. The mRNA expression levels were quantified by real-time PCR on a CFX69 real-time detection system (Bio-Rad, Veenendaal, The Netherlands) by use of SYBR Green (Bio-Rad) and calculated by the ⌬⌬Ct method. Data was normalized for total expression along the gastrointestinal tract. Primers (Biolegio, Nijmegen, the Netherlands) were designed with Primer 3 software (Whitehead Institute for Biomedical Research, Cambridge, MA) and are listed in Table 1. Animal studies. Male C57BL/6J mice (8 wk old) were purchased from Harlan, the Netherlands. For the tissue panels, animals were euthanized by cervical dislocation. Subsequently, intestinal segments were isolated, cleaned, and snap frozen in liquid nitrogen. For all other studies mice were housed in a temperature- and light-controlled room with a standardized pelleted chow (containing 0.22% wt/wt Mg2⫹ and 1.00% wt/wt Ca2⫹) and drinking water available ad libitum. For the diet studies, mice were fed a low Mg2⫹-deficient diet (0.02% wt/wt Mg2⫹) a high-Mg2⫹ diet (0.48% wt/wt Mg2⫹), or a normal diet

Table 1. Primer sequences used for real-time PCR Gene

Species

Forward Primer

Reverse Primer

GAPDH TRPV6 TRPM6 TRPM7 CNNM4 Calb-D9K PMCA1b GAPDH TRPV6 TRPM6 TRPM7 CNNM4 Calb-D9K PMCA1b TRPV5 1␣OHase CLDN2 CLDN12

Homo sapiens Homo sapiens Homo sapiens Homo sapiens Homo sapiens Homo sapiens Homo sapiens Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus Mus musculus

5=-GGAGTCAACGGATTTGGTCGTA-3= 5=-GCTTTGCTTCAGCCTTCTATATCAT-3= 5=-GCGGACATGGCATAAAATCTTC-3= 5=-GAGTCCGCCCCGTGAGG-3= 5=-TCCTGGGCCAGTACATCTC-3= 5=-GACCAGTTGTCAAAGGATGAACTG-3= 5=-CCCGGAAAATTCATGGTGAA-3= 5=-TAACATCAAATGGGGTGAGG-3= 5=-CTTTGCTTCAGCCTTCTATATCAT-3= 5=-AAAGCCATGCGAGTTATCAGC-3= 5=-GGTTCCTCCTGTGGTGCCTT-3= 5=-TCTGGGCCAGTATGTCTCTG-3= 5=-CCTGCAGAAATGAAGAGCATTTT-3= 5=-CGCCATCTTCTGCACCATT-3= 5=-CCACAGTGATGCTGGAGAGG-3= 5=-CGGATTGCTAACGGCGGA-3= 5=-TGGTGGGCATGAGATGCA-3= 5=-TCACATTCAACAGAAACGAGAAGAA-3=

5=-GGCAACAATATCCACTTTACCAGAGT-3= 5=-TGGTAAGGAACAGCTCGAAGGT-3= 5=-TTGAGCAGCTCTTTGTTGTTGAA-3= 5=-TCCTGGAAGGCATCTGTG-3= 5=-GAAACTGAGGGCTGTTCTCC-3= 5=-TCTAGGGTGTTTGGACCTTTGAG-3= 5=-CTGTACCACAAAAGTGCCTAAAACA-3= 5=-GGTTCACACCCATCACAAAC-3= 5=-TGGTAAGGAACAGCTCGAAGGT-3= 5=-CTTCACAATGAAAACCTGCCC-3= 5=-CCCCATGTCGTCTCTGTCGT-3= 5=-CACAGCCATCGAAGGTAGG-3= 5=-CTCCATCGCCATTCTTATCCA-3= 5=-CAGCCATTGCTCTATTGAAAGTTC-3= 5=-GGATTCTGCTCCTGGTGGTG-3= 5=-GGCGCCTTAGTCGTCGCA-3= 5=-CTCCACCCACTACAGCCACTCT-3= 5=-CCATCATACCGGGCACACTT-3=

GAPDH, glyceraldehyde 3-phosphate dehydrogenase; TRPV6, transient receptor potential vanilloïd type 6; TRPM6, transient receptor potential melastatin type 6; TRPM7, transient receptor potential melastatin type 7; CNNM4, cyclin M4; Calb-D9K, calbindinD9K; PMCA1b, plasma membrane Ca2⫹-ATPase 1b; TRPV5, transient receptor potential vanilloïd subtype 5, 1␣OHase, 1-␣-hydroxylase; CLDN2, claudin 2, CLDN12, claudin 12. AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

G208

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

(0.25% wt/wt Mg2⫹) (14). All diets were purchased from SSNIFF Spezialdiäten (SSNIFF Spezialdiäten, Soest, Germany), and Mg2⫹ content of the diets was increased by the addition of MgO. For urine and feces collection, animals were individually housed in metabolic cages for 24 h. Blood was sampled from the submandibular facial vein at the end of the stay in the metabolic cages, and sera were collected for Mg2⫹ measurements. For studies on vitamin D, animals received daily intraperitoneal injections with 1,25-dihydroxyvitamin D3 (100 pg·g body wt⫺1·day⫺1) in peanut oil or vehicle for 14 days. All experiments were approved by the animal ethics board of the Radboud University Nijmegen (Ethical License number RU-DEC 2010-143, RU-DEC 2010-204, RU-DEC 2011-165). Placement of intestinal cannulas. Cannulas were made using PE-50 polyethylene tubing (Instech Laboratories, Plymouth Meeting, PA), with an inner diameter 0.58 mm and outer diameter of 0.97 mm. Suture bulbs with a diameter of ⬃2 mm were created on the end of pieces of polyethylene tubing by using Dow Corning 723 flowable sealant silicone elastant (Dow Corning, Midland, MI). Mice were anesthetized with 1 ␮g/g body wt medetomidine-hydrochloride (Sedastart, ASTfarma, Oudewater, The Netherlands) and 0.08 mg/g body wt ketamine (Nimatek, Eurovet, Bladel, The Netherlands), both administered via subcutaneous injection. The abdominal skin was opened via a lateral incision, after which the muscle layer was cut along the linea alba (Fig. 1A). Next, the cecum or stomach was localized, a small hole was created, and a pouch suture was placed around the hole, after which the tip of the cannula was inserted into the lumen (Fig. 1, B–F). Sutures were made with Vicryl Plus 5.0 threads (Johnson & Johnson Medical, Amersfoort, The Netherlands). The suture was closed around the cannula, in between the suture bulbs, fixing the cannula in place. The tubing was passed through the muscular layer of the intestinal wall on the lateral side, after which the abdominal cavity was closed with a running stitch (Fig. 1, G and H). A pocket was created between the muscular layers and the skin. A SIP22/4 miniature injection port (Instech Laboratories) was attached to the cannula and placed in the subcutaneous pocket on the ventrolateral side of the animal (Fig. 1, I and J). The wound was closed with a running stitch and covered with wound clips to prevent postsurgical opening of the wound. Medetomidine anesthesia was reversed with a subcutaneous injection of 1 ␮g/g body wt atipamezole hydrochloride (Sedastop, ASTfarma). During the first 3 days after surgery, animals received 0.05 ␮g/g body wt buprenorphine hydrochloride (Temgesic, Reckitt Benckiser, West Ryde, Australia) twice a day via subcutaneous injection to prevent postoperative pain. The cannula was accessible via the skin and was flushed every other day with a small volume (⬃50 ␮l) of saline to prevent clogging (Fig. 1K). Animals were allowed to recover for a minimum of 7 days, before absorption experiments were performed. In general, the mice fully recover during this period, as indicated by their return to the preoperative weight (Fig. 1M). Electrolyte determinations in serum, urine, and feces. Before analysis, fecal samples were homogenized and digested in 65% vol/vol nitric acid (Sigma-Aldrich) for 2 h at 70°C, followed by an overnight incubation at room temperature. Serum, urinary, and fecal total Ca2⫹ concentrations were measured with a colorimetric assay as described previously (18). Total Mg2⫹ concentrations were determined by using a colorimetric xylidyl-II blue assay kit according to the manufacturer’s protocol (Roche Diagnostics, Woerden, The Netherlands). With-

in-run precision and accuracy was controlled by means of an internal control Precinorm U (Roche Diagnostics, Almere, The Netherlands). Urinary Na⫹, K⫹, Cl⫺, and Pi were analyzed with a Hitachi auto analyzer (Hitachi, Laval, Quebec, Canada). In vivo 25Mg2⫹ absorption assay. Intestinal absorption of Mg2⫹ was measured by analyzing serum 25Mg2⫹ levels. In short, animals were fasted (food, not water) overnight on wire mesh raised floors to prevent coprophagia. At time point 0, mice were administered a solution containing 44, 10, or 1.8 mM 25Mg2⫹ (MgO, isotopic enrichment of ⬎98%, CortectNet, Voisins-Le-Bretonneux, France), 125 mM NaCl, 17 mM Tris·HCl pH 7.5, 1.8 g/l fructose. Animals were administered a volume of 15 ␮l/g body wt via oral gavage or via the subcutaneously placed injection ports. Subsequently, blood was withdrawn at serial time points via orbital puncture or via a small cut in the tail and collected in Microvette serum tubes (Sarstedt, EttenLeur, The Netherlands). After collection of serum from the coagulated blood samples, samples were digested in nitric acid (65% concentrated, Sigma, Zwijndrecht, The Netherlands) for 1 h at 70°C followed by an overnight incubation at room temperature. Subsequently, samples were diluted in milliQ and subjected to ICP-MS analysis (X1 series, Thermo Fisher Scientific, Breda, The Netherlands). Calculations. The three stable isotopes of Mg2⫹ have the following natural abundance: 24Mg 78.9%, 25Mg 10.0%, 26Mg 11.1%. The isotopic ratios at baseline are therefore 25Mg/24Mg ⫽ 0.1267 and 26 Mg/24Mg ⫽ 0.1407. The percentage of isotopic enrichment in the serum, urine, and fecal samples was calculated by using the following equation: (measured 25Mg/24Mg ratio ⫺ baseline 25Mg/24Mg ratio)/ (baseline 25Mg/24Mg ratio) * 100. In vivo 45Ca2⫹ absorption assay. Mice were fasted (food, not water) overnight on wire mesh raised floors to prevent coprophagia. At time point 0, mice were administered a solution containing 0.1, 10, or 44 mM CaCl2, 125 mM NaCl, 17 mM Tris·HCl pH 7.5, 1.8 g/l fructose, enriched with 20 ␮Ci 45CaCl2/ml (18 Ci/g, New England Nuclear, Newton, MA) in a volume of 15 ␮l/g body wt. The solution was administered via the subcutaneously placed injection port. Blood samples were obtained via orbital puncture at serial time points after isotope administration via orbital puncture. Blood was collected in Microvette tubes (Sarstedt, Nümbrecht, Germany), after clotting serum was removed and serum 45Ca2⫹ levels were analyzed by liquid scintillation counting. The change in the serum Ca2⫹ concentration was calculated from the 45Ca2⫹ content of the serum samples and the specific activity of the administered 45Ca2⫹. Statistical analysis. Values are expressed as means ⫾ SE. Statistical comparisons were analyzed by one-way ANOVA with a Bonferroni correction, t-tests corrected for multiple comparisons by Holm-Šídák method, or an unpaired Student’s t-test. Differences between groups were considered to be statistically significant at P ⬍ 0.05. Statistical analysis was performed with GraphPad Prism (Macintosh version, 4.51). RESULTS

TRPV6 and TRPM6 expression along the gastrointestinal tract in humans and mice. In the human gastrointestinal tract, TRPV6 is expressed predominantly in the stomach and duodenum (Fig. 2A). Along the murine gastrointestinal tract, the expression pattern of TRPV6 is slightly different. In addition to

Fig. 1. Surgical procedure to place intestinal cannulas connected to a subcutaneously positioned injection port. A: opening of the abdominal cavity. B: localization of the stomach or duodenum. C: creating of an opening for the insertion of the cannula. D: placement of a pouch suture around the opening. E: insertion of the cannula in the intestine. F: tightening the pouch suture around the cannula between the suture bulbs, fixing the cannula in place. G: cannula is passed through the muscular layer of the abdomen to the lateral side of the mouse. H: abdominal cavity is closed. I and J: injection port is attached to the cannula and placed in a subcutaneous pocket. K: abdominal skin is closed and the cannula is flushed via the injection port. L: cannula with suture bulbs and injection port. M: weight development of the mice throughout experiments. Data are presented as means ⫾ SE, n ⫽ 81. AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

A

B

C

D

E

F

G

H

I

J

K

L

M % of starting weight

105 100 95 90 85 0 0

1

2

3

4

5

6

7

8

9

Days after surgery

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

G209

G210

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

40 20 ND

ND

ND ND

0

ND

ND

ND ND

ND ND

op Es

D

Fu

op

ha

gu

s

nd An u s u o t ru m de n Je u m ju nu m Ile C ol C um o a C n ol a s e c on c u m e C o l tran n d e on s n d e ver s s c su m Si gm e n d oï e n s d co l R on ec tu m

ND

30 20 10 0

ND

ND

ND ND

ha g Fu u s nd A us D ntr u o um de n Je u m ju nu m Ile C ol u C ae m C on ol a s c on c u m e C o l tran n d e on s n d e ver s s s Si c e u m gm n d oï e n s d co l R on ec tu m

ND

0

TRPM7 mRNA (normalized expression)

50

F 20 15 10 5 ND

ND

ND ND

ha g Fu u s nd A us D nt r u o um de n Je u m ju nu m Ile C ol C um o a C n ol a s e c on c u m e C o l tran n d e on s n d e ver s s c su m Si gm e n d oï e n s d co l R on ec tu m

0

the expression in the stomach and the duodenum, TRPV6 is also strongly expressed in the cecum as well as in the distal regions of the colon (Fig. 2A). TRPM6 is expressed in the distal regions of the gastrointestinal tract of human subjects and mice (Fig. 2B). Interestingly, like TRPV6, TRPM6 in mice is expressed strongly in the cecum (Fig. 2B). Expression of the intracellular Ca2⫹ binding protein calbindin-D9k was mainly restricted to the duodenum (Fig. 2C),

40 30 20 10 0

ND

ND

ND ND

Es

op

ha g Fu u s nd A us D ntr u o um de n Je u m ju nu m Ile C ol u C ae m C on ol a s c on c u m e C o l tran n d e on s n d e ver s s s Si c e u m gm n d oï e n s d co l R on ec tu m

25

op

PMCA1b mRNA (normalized expression)

20

op

100

Es

150

E

Es

40

D

Es

Fig. 2. Localization of TRPV6 and TRPM6 along the gastrointestinal tract in humans and mice. Total gastrointestinal mRNA expression of TRPV6 (A) and TRPM6 (B), calbindin-D9k (C), TRPM7 (D), PMCA1b (E), and CNNM4 (F) corrected for GAPDH expression, in human (solid bars) and murine (shaded bars) samples (n ⫽ 5). Normalized data presented as means ⫾ SE. ND, not determined.

CalbindinD9k mRNA (normalized expression)

C

CNNM4 mRNA (normalized expression)

Es

op

ha g Fu u s nd A us D nt r u o um de n Je u m ju nu m Ile C ol C um o a C n ol a s e c on c u m e C o l tran n d e on s n d e ver s s c su m Si gm e n d oï e n s d co l R on ec tu m

0

60

ha g Fu u s nd A us D ntr u o um de n Je u m ju nu m Ile C ol C um o a n C ol a s e c on c u m e C o l tran n d e on s n d e ver s s s Si c e u m gm n d oï e n s d co l R on ec tu m

60

TRPM6 mRNA (normalized expression)

B

TRPV6 mRNA (normalized expression)

A

whereas the basolateral Ca2⫹-pump PMCA1b was ubiquitously expressed (Fig. 2E). TRPM7 was evenly expressed throughout the gastrointestinal tract (Fig. 2D). Expression of the basolateral Mg2⫹ transporter CNNM4 was most prominent in the distal gastrointestinal segments, resembling the expression profile of TRPM6 (Fig. 2F). However, in contrast to TRPM6, CNNM4 expression was not completely absent in the proximal gastrointestinal tract. Expression profiles of calbin-

Table 2. General characteristics and urinary electrolyte excretion of mice during metabolic cage studies Measurement

⬃0.02% (wt/wt) Mg

0.25% (wt/wt) Mg

0.48% (wt/wt) Mg

Animal weight, g Food intake, g/24 h Water intake, ml/24 h Feces, g/24 h Dry fecal weight Urinary volume, ml/24 h Urinary pH Na⫹ excretion, ␮mol/24 h K⫹ excretion, ␮mol/24 h Cl⫺ excretion, ␮mol/24 h Pi excretion, ␮mol/24 h

23.9 ⫾ 0.6 3.7 ⫾ 0.2 4.1 ⫾ 0.2 0.47 ⫾ 0.02 0.40 ⫾ 0.01 1.6 ⫾ 0.2 6.9 ⫾ 0.2 313 ⫾ 26 591 ⫾ 56 326 ⫾ 30 264 ⫾ 25

23.9 ⫾ 0.5 3.3 ⫾ 0.2 5.0 ⫾ 0.4 0.43 ⫾ 0.03 0.38 ⫾ 0.03 1.5 ⫾ 0.1 6.9 ⫾ 0.1 240 ⫾ 14 525 ⫾ 33 260 ⫾ 13 158 ⫾ 23†

24.4 ⫾ 0.8 3.3 ⫾ 0.2 5.2 ⫾ 0.2 0.70 ⫾ 0.09*† 0.46 ⫾ 0.03 2.0 ⫾ 0.2 6.7 ⫾ 0.2 236 ⫾ 42 542 ⫾ 90 272 ⫾ 49 97 ⫾ 17†

*Significantly different (P ⬍ 0.05) compared with 0.25% (wt/wt) Mg diet; †significantly different (P ⬍ 0.05) compared with 0.02% (wt/wt) Mg diet. AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

*

110

h H h

ig

N

or

m

Lo

h ig H

al N

or

m

w

h

1.5 1.2 0.9 0.6 0.3 0

ig

30

0

H

25

0.5

al

20

1.0

Fig. 3. Effect of dietary Mg2⫹ on intestinal Mg2⫹ absorption. A: serum Mg2⫹ values of mice 2 wk on diet with a low (0.02% wt/wt Mg2⫹), normal (0.25% wt/wt Mg2⫹), or high (0.48% wt/wt Mg2⫹) Mg2⫹ content (n ⫽ 5 for each group). B: urinary Mg2⫹ excretion. C: fecal Mg2⫹ excretion. D: serum Ca2⫹ levels. E: urinary Ca2⫹ excretion. F: fecal Ca2⫹ excretion. G: intestinal absorption of orally administered 25Mg2⫹ in mice on a diet with low (dotted line), normal (solid line), or high (dashed line) Mg2⫹ content. H: average difference in intestinal absorption of 25Mg2⫹ during 4 h after isotope administration (orally administered, 44 mM 25 Mg2⫹). I: TRPM6 mRNA expression levels in the colon, corrected for GAPDH expression and normalized to the normal-Mg2⫹-diet group. J: 1-␣hydroxylase (1␣OHase) mRNA expression levels in the kidney, corrected for GAPDH expression and normalized to the normal-Mg2⫹-diet group. Data are presented as means ⫾ SE. *P ⬍ 0.05 compared with normal-Mg2⫹ diet.

N

15

*

1.5

H

al

0

Lo

10

Time (minutes)

m

90

m

0

or

*

100

or

10

ig

w N

*

120

Lo

J

*

al

0

w

20

TRPM6 mRNA expression colon (relative to normal)

30

1 OHase mRNA expression kidney (relative to normal)

I

*

5

h

200

w

(relative to normal)

H

25Mg2+ absorption

*

40

H

m or

400

Lo

m

600

N

or

Lo

60

ig

w Lo

Fecal Ca2+ excretion (µmol/24hr)

0

w

h ig

m or

*

5

N

H

al

0

10

800

h

0.5

15

1000

ig

1.0

0

al

0

H

1.5

*

al

2.0

Lo

200

F 20

50

% enrichment 25Mg2+ in serum

400

N

or

m

Lo

Urinary Ca2+ excretion (µmol/24hrs)

E 2.5

w

Serum Ca2+ (mM)

D

G

*

N

w

h

0

ig

m or

*

N

Lo

20

H

al

0

40

h

0.5

60

600

ig

*

1.0

*

al

1.5

C 80

the high-Mg2⫹ diet were not significantly different from the group on the normal diet (Fig. 3A). Urinary and fecal Mg2⫹ excretion, however, were both significantly increased (Fig. 3, B and C). Interestingly, urinary excretion of Ca2⫹ was significantly decreased in the group receiving the low-Mg2⫹ diet and significantly increased in the group receiving the high-Mg2⫹ diet (Fig. 3E). Serum Ca2⫹ levels and fecal Ca2⫹ excretion remained unchanged (Fig. 3, D and F). Urinary excretion of Na⫹, K⫹, and Cl⫺ did not differ significantly between the groups, excretion of Pi was significantly higher in the lowMg2⫹-diet group compared with the groups receiving normal and high-Mg2⫹ diets (Table 2). Dietary Mg2⫹ restriction significantly stimulated the rate of intestinal absorption of Mg2⫹ (Fig. 3G). Conversely, intestinal absorption of Mg2⫹ in the high-Mg2⫹-diet group was diminished compared with the normal group. The absorption of

H

Urinary Mg2+ excretion (µmol/24hrs)

B 2.0

w

Serum Mg2+ (mM)

A

Fecal Mg2+ excretion (µmol/24hr)

din-D9k, PMCA1b, TRPM7, and CNNM4 were similar in humans and mice. Intestinal Mg2⫹ absorption is regulated by dietary content. To investigate whether Mg2⫹ absorption is regulated by dietary Mg2⫹ content, mice were fed a diet with a low (0.02% wt/ wt Mg2⫹), a normal (0.25% wt/wt Mg2⫹), or a high (0.48% wt/wt Mg2⫹) Mg2⫹ content for 2 wk. General characteristics such as weight, feeding behavior, and diuresis were similar in all groups (Table 2). Wet weight of the feces from the mice receiving the high-Mg2⫹ diet was significantly increased; however, stools were solid and mice did not display signs of dehydration (Table 2). The mice on the low-Mg2⫹ diet had significantly lower serum Mg2⫹ levels compared with the group on the normal diet (Fig. 3A). In addition, urinary Mg2⫹ excretion was diminished and fecal Mg2⫹ excretion was decreased (Fig. 3, B and C). Serum Mg2⫹ levels in mice receiving

G211

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

The proximal Mg2⫹ absorption was higher compared with the distal part of the intestine at a high luminal concentration of Mg2⫹ (44 mM) (Fig. 4E), but not at lower luminal concentrations (1.6 and 10 mM Mg2⫹) (Fig. 4, A and C). The relationship between the absorption of 25Mg2⫹ and the luminal Mg2⫹ concentration is displayed in Fig. 4G. In the proximal intestine, the absorption of 25Mg2⫹ resembles a linear relation to the buffer Mg2⫹ concentration, whereas in the distal part of the intestine the absorption of Mg2⫹ seems saturable at higher Mg2⫹ concentrations. The relation between 45Ca2⫹ absorption and the luminal Ca2⫹ concentration is comparable in the proximal and distal parts of the intestine and resembles a saturable and a nonsaturable linear component (Fig. 4H). Effect of vitamin D on Mg2⫹ homeostasis. To determine whether vitamin D has an effect on Mg2⫹ homeostasis, and more specifically on intestinal Mg2⫹ absorption, mice were treated with 1,25-dihydroxyvitamin D3 (100 pg·g body wt⫺1·day⫺1) for 7 days and subsequently the intestinal absorp-

Enrichment 25Mg2+ in serum (%)

A

B 3 2 1 0 -1 -2 0

10

20

(µmol)

Mg2⫹ in the low-Mg2⫹-diet group was significantly higher compared with the normal group (110.0 ⫾ 3.0% and 100.0 ⫾ 0.1%, respectively, P ⬍ 0.01) and in the group on the highMg2⫹ diet absorption was significantly lower compared with the normal group (95.0 ⫾ 1.0 vs. 100.0 ⫾ 0.1%, respectively, P ⬍ 0.05) (Fig. 3H). TRPM6 mRNA expression levels in the colon were slightly, but significantly, regulated by dietary Mg2⫹, showing an increase in the group on the low-Mg2⫹ diet and a decrease as a consequence of the high-Mg2⫹ diet (Fig. 3I). Importantly, the two diets did not affect the renal expression of 1-␣-hydroxylase (1␣OHase) (Fig. 3J). Absorption of Ca2⫹ and Mg2⫹ in intestinal segments. To investigate the mineral absorption specifically in the proximal and distal intestinal segments, cannulas were surgically implanted in either the stomach or the cecum of mice. The Ca2⫹ absorption in the proximal intestine was significantly higher compared with the more distal part at all luminal concentrations of Ca2⫹ tested (0.1, 10, and 44 mM) (Fig. 4, B, D, and F).

45Ca2+ absorption

G212

30

1.5

*

1.0

6

*

4 2 0 10

20

(µmol)

*

45Ca2+ absorption

Enrichment 25Mg2+ in serum (%)

D

0

0 0

10

30

*

20

**

10 0 10

20

30

Time (minutes)

G

*

50

60 40

*

20

*

**

40 30 20 10

*

*

0 0

15

30

*

*

0 0

10

20

30

*

150 100

* **

50

*

0 0

10

20

30

Time (minutes)

H 45Ca2+ absorption

0

Enrichment 25Mg2+ in serum (%)

45Ca2+ absorption

*

*

120

(µmol/30min.)

30

*

(µmol)

F 40

30

Time (minutes)

E 50

20

80

Time (minutes)

Enrichment 25Mg2+ in serum (%)

Fig. 4. Paracellular and transcellular absorption of Mg2⫹ in intestinal segments. Solid lines represent mice with a cannula placed in the stomach, dashed lines represent mice with a cannula placed in the cecum. 25Mg2⫹ absorption by using a buffer containing 1.6 (A), 10 (C), or 44 (E) mM Mg2⫹. 45Ca2⫹ absorption using a buffer containing 0.1 (B), 10 (D), or 44 (F) mM. Relationship between luminal Mg2⫹ concentration and 25Mg2⫹ absorption (G) and between luminal Ca2⫹ concentration and 45Ca2⫹ absorption (H) at t ⫽ 30 min. Data are presented as means ⫾ SE, n ⫽ 4 – 6. *P ⬍ 0.05 compared with the group with cannula placed in the stomach.

*

Time (minutes)

C 8

*

0.5

Time (minutes)

10

*

45

Concentration Mg2+ (mM) AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

90

*

60 30 0 0

15

30

Concentration Ca2+ (mM)

45

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

tion of 25Mg2⫹ was measured. Vitamin D treatment significantly increased serum Mg2⫹ and Ca2⫹ levels as well as 24-h urinary Mg2⫹ and Ca2⫹ excretion (Fig. 5, A–D). Importantly, the intestinal Mg2⫹ absorption was not significantly different between vehicle- and vitamin D-treated mice (Fig. 5E). Effect of vitamin D on intestinal Mg2⫹ absorption. The mRNA expression levels of TRPV6 in duodenum (Fig. 6C) as well as the colon (Fig. 6A) of vitamin D-treated mice were significantly increased compared with vehicle-treated controls. TRPM6 expression in the colon, however, remained unchanged (Fig. 6B). The duodenal expression levels of CLDN2 and CLDN12 were analyzed to determine the effect of vitamin D on paracellular mineral transport. No significant differences in the expression levels were observed between vitamin D-treated mice and controls (Fig. 6, D and E). In kidney, 1␣OHase expression was significantly decreased in the vitamin D-treated group, indicating appropriate suppression of vitamin D synthesis by vitamin D treatment (Fig. 6F). The renal TRPV5 expression of vitamin D-treated mice was significantly augmented compared with controls, whereas no changes were observed in renal TRPM6 expression between both groups.

3.0 2.0 1.0

80

*

60 40 20

D

tro

25

(O

C

H

)2

on

3

D )2 H (O 25

Urinary Mg2+ excretion (µmol/24hrs)

*

1.5 1.0 0.5

*

40 30 20 10 0

3

)2

Fig. 5. Effect of vitamin D on Mg2⫹ homeostasis. Mice were treated with vehicle (control) or 1,25-dihydroxyvitamin D3 [1,25(OH)2D3; 100 pg·g body wt⫺1·day⫺1 for 14 days]. A: serum Ca2⫹ levels in controls and vitamin D-treated mice. B: 24-h urinary Ca2⫹ excretion. C: serum Mg2⫹ levels. D: 24-h urinary Mg2⫹ excretion. E: intestinal absorption of orally administered 25Mg2⫹ (44 mM) in control (solid line) and vitamin D (dashed line)-treated mice. Data are presented as means ⫾ SE, n ⫽ 5–7. *P ⬍ 0.05 compared with vehicle-treated controls.

H

C

25

1,

1,

25

E

D

tro on

3

(O

C

H

)2

on

D

tro

l

l

0

1,

1,

D

2.0

3

l

0

l tro on C

C Serum Mg2+ (mM)

Compared to Ca2⫹, little is known about Mg2⫹ absorption in the intestine, including hormones regulating this process. The aim of this study was, therefore, to investigate the absorption of Mg2⫹ along the gastrointestinal tract on a molecular and functional level. Our findings indicate that 1) mRNA expression of TRPV6 is predominant in the proximal part of the human gastrointestinal tract, whereas TRPM6 is mainly expressed in the distal segments and expression patterns of TRPV6 and TRPM6 in the murine gastrointestinal tract are similar to those observed in humans; 2) intestinal absorption of Mg2⫹ is regulated by dietary supply; 3) Ca2⫹ is predominantly absorbed in the proximal parts of the intestine, irrespective of its luminal concentration; 4) at low luminal concentrations, Mg2⫹ is predominantly absorbed in the distal intestinal segments; and 5) vitamin D does not affect Mg2⫹ absorption in a physiological in vivo setting. The epithelial mineral channels, TRPV6 and TRPM6, form the gatekeepers of Ca2⫹ and Mg2⫹ absorption, respectively, along the gastrointestinal tract (10, 29). In humans, TRPV6 is expressed predominantly in the stomach and duodenum,

(O

Serum Ca2+ (mM)

*

0

% enrichment 25Mg2+ in serum

DISCUSSION

B 4.0

Urinary Ca2+ excretion (µmol/24hrs)

A

G213

100 80 60 40 20 0 0

10

20

30

40

50

60

Time (minutes) AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

B

0 )2 H

0.5

25 1, D H

)2

3

l

0 (O 25 1,

(O 25 1,

0.5

tro

D )2

3

l

0

1.0

on

1.0

1.5

C

2.0

TRPM6 mRNA expression (relative to control)

3.0

H

3

3

)2 H (O

C

1, (O

25 1,

whereas TRPM6 is mainly expressed in the ileum and colon. In mice, TRPV6 was also expressed in cecum and distal segments of the colon, whereas TRPM6 was in addition to ileum and colon also abundantly detected in cecum. The discrepancy between the expression patterns of these TRP channels in human and mice might be explained by the morphological and functional differences between their gastrointestinal tracts. Mice are nonruminant herbivores that have a large cecum compared with humans (21). Their cecum houses many types of microorganisms, which aid in the digestion of plant material by breaking down cellulose (11). Plant material is an important source of Ca2⫹ and Mg2⫹, which is only released if cellulose is digested. Since the digestion of cellulose takes place in the cecum of mice, these nutrients are released from the plant material in the more distal segments of the intestine. Intestinal absorption of Ca2⫹ is strictly regulated by bodily needs: increased when dietary Ca2⫹ supply is limited and decreased when Ca2⫹ is abundantly available (42). To determine whether the same holds true for of Mg2⫹, the effect of dietary Mg2⫹ supply on intestinal absorption of Mg2⫹ was investigated. A low-Mg2⫹ diet effectively induced a hypomagnesemia and urinary Mg2⫹ conservation. Conversely, enhanced renal Mg2⫹ excretion indicated that the high-Mg2⫹ diet

4.0

C

H

)2

on

D

tro

l

0

*

5.0

tro

*

0.5

H

on

1.0

TRPV5 mRNA expression (relative to control)

G 1.5

D

tro

l

0 on

3

D )2

1.0

25

25 1,

F

C

1.5

(O

C

(O

H

3

0

H

)2

on

D

tro

l

0

0.5

l

5

1.0

tro

10

1.5

on

15

CLDN2 mRNA expression (relative to control)

*

E

CLDN12 mRNA expression (relative to control)

1,

1,

D

20

C

TRPV6 mRNA expression (relative to control)

C

1 OHase mRNA expression (relative to control)

Fig. 6. Expression of calciotropic genes, but not TRPM6, is regulated by vitamin D treatment. mRNA expression levels of various calciotropic and magnesiotropic genes were determined in vehicle-treated controls and 1,25-dihydroxyvitamin D3-treated mice (100 pg·g body wt⫺1·day⫺1, for 14 days). All values were corrected for GAPDH expression and were normalized to controls. A and B: TRPV6 (A) and TRPM6 (B) mRNA expression in colon. C–E: TRPV6 (C), CLDN2 (D), and CLDN12 (E) mRNA expression in duodenum. F–H: 1␣OHase (F), TRPV5 (G), and TRPM6 (H) mRNA expression in kidney. Data are presented as means ⫾ SE n ⫽ 5–7. *P ⬍ 0.05 compared with vehicle-treated controls.

25

25

(O

C

(O

C

H

on

)2

3

D

tro

l

0

0.5

3

50

1.0

D

100

1.5

l

150

tro

*

200

on

TRPV6 mRNA expression (relative to control)

A

TRPM6 mRNA expression (relative to control)

G214

resulted in a surplus of Mg2⫹ in the body. Intestinal absorption of 25Mg2⫹ was significantly increased in mice receiving the low-Mg2⫹ diet, whereas Mg2⫹ absorption in the high-Mg2⫹diet group was reduced. This indicates that the rate of Mg2⫹ absorption, like for Ca2⫹, follows the dietary body needs. The alterations in Mg2⫹ absorption were accompanied by a slight, but significant, effect on the expression of TRPM6 in the mice on the low-Mg2⫹ diet but not the high-Mg2⫹ diet, in line with previous findings (14). These findings are in contrast with mice challenged with a Ca2⫹-deficient diet who showed a strong increase in intestinal Ca2⫹ transport, facilitated by the increased expression of proteins involved in Ca2⫹ transport, including TRPV6 (41, 42, 45). Renal expression of 1␣OHase, the enzyme responsible for the production of active vitamin D, of which the role in Mg2⫹ homeostasis has been under debate (16, 37), was unchanged during the various dietary Mg2⫹ regimes. This suggests that the regulation of intestinal Mg2⫹ absorption is vitamin D independent. New magnesiotropic hormones, including EGF and insulin, could potentially play a role in the regulation of intestinal Mg2⫹ absorption (15, 28, 41). Alternatively, the Ca2⫹-sensing receptor (CaSR), which senses the extracellular concentration of both Ca2⫹ and Mg2⫹ could be involved (13, 34). Interestingly, the dietary Mg2⫹

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

content also affected renal excretion of Ca2⫹. Our previous studies have shown that dietary Mg2⫹ content does not affect intestinal Ca2⫹ absorption (14). The coupling between the Ca2⫹ and Mg2⫹ balance has previously been observed (2, 14), but the underlying mechanism remains under debate and is of interest for future research (2, 5, 25, 33). The changes in renal Pi excretion during Mg2⫹ depletion or supplementation are most likely the result of intestinal chelation of Pi by Mg2⫹. In line with this observation, it was recently demonstrated that the reduced renal Pi excretion upon Mg2⫹ loading is independent of parathyroid hormone and the CaSR (33). Paracellular absorption of Ca2⫹ and Mg2⫹ mainly takes place in the duodenum and jejunum. On the basis of the expression patterns of TRPV6 and TRPM6 observed in the gastrointestinal tracts of humans and mice, we hypothesized that transcellular Ca2⫹ absorption mainly takes place in the proximal parts of the intestine whereas transcellular Mg2⫹ absorption is restricted to the distal parts of the gastrointestinal tract. To address this hypothesis in an in vivo setting with functional measurements, intestinal cannulas were placed in either the stomach or the cecum of mice, to allow local measurements of mineral absorption. Indeed, Ca2⫹ absorption in the proximal intestine was significantly higher compared with the more distal parts at all luminal concentrations of Ca2⫹ that were tested. Mg2⫹ absorption in the proximal intestine on the other hand was higher compared with the distal part of the intestine at a high luminal concentration of Mg2⫹, but not at lower luminal concentrations. At high luminal concentrations of Ca2⫹ or Mg2⫹ paracellular absorption is considered to be predominant, whereas low luminal concentrations favor transcellular absorption (4, 40). The findings in our study, therefore, suggest that transcellular Ca2⫹ absorption takes place in the proximal region of the intestine, whereas transcellular Mg2⫹ absorption mainly occurs in the distal gastrointestinal segments. Because of the similarities between Ca2⫹ and Mg2⫹ homeostasis, the calciotropic hormone vitamin D has often been suggested to play a role in Mg2⫹ handling (16). To examine the effect of vitamin D on Mg2⫹ homeostasis, mice were treated with 1,25-dihydroxyvitamin D3 for 14 days. Vitamin D treatment increased serum Ca2⫹ levels as well as urinary Ca2⫹ excretion, which is in line with previous findings (35). Interestingly, serum Mg2⫹ levels and 24-h urinary Mg2⫹ excretion were also significantly higher compared with controls, suggesting that intestinal absorption of Mg2⫹ was increased. However, changes in intestinal absorption were not detected by the 25 Mg2⫹ absorption assay. To gain more insight in the underlying molecular processes, mRNA expression levels or various genes were determined in duodenum, colon, and kidney of the vehicle-treated controls and vitamin D-treated mice. TRPM6 expression was unchanged both in colon and kidney, indicating that vitamin D does not affect transcellular absorption via TRPM6 at the transcriptional level. Vitamin D could also have an effect on paracellular transport of Ca2⫹ and Mg2⫹. However, expression of vitamin D-sensitive tight junction proteins CLDN2 and CLDN12 was unchanged by vitamin D treatment (12). Therefore, it is unlikely that they contributed to changes in intestinal Mg2⫹ absorption. So, despite the changes in serum Mg2⫹ levels and urinary Mg2⫹ excretion, no functional or molecular evidence was found that vitamin D affects intestinal Mg2⫹ absorption. High urinary Ca2⫹ levels may affect renal

G215

Mg2⫹ handling, explaining the hypermagnesiuria, but not the observed hypermagnesemia (31, 36). Alternatively, vitamin D-induced mobilization of Mg2⫹ from bone could provide an alternative explanation for the increased serum Mg2⫹ levels and urinary Mg2⫹ excretion in vitamin D-treated animals (27). Although Mg2⫹ homeostasis can be influenced by vitamin D in pharmacological doses, which is in line with previous findings, vitamin D seems to have no effect on Mg2⫹ homeostasis in the normal physiological range (16). Indeed, 1␣OHase and klotho knockout mice, which are vitamin D deficient or display hypervitaminosis D, respectively, both have normal serum Mg2⫹ levels, urinary Mg2⫹ excretion, and renal TRPM6 expression levels (14, 46). In conclusion, the stable isotope technique and intestinal cannulas were successfully applied to measure time- and location-dependent absorption of Mg2⫹ in the intestine in an in vivo setting. Using this technique demonstrated that transcellular Ca2⫹ and Mg2⫹ absorption in the intestine are spatially separated. In addition, it was shown that absorption of Mg2⫹ from the intestinal lumen is regulated by dietary supply, but not by vitamin D. Together, the intestinal cannulas and 25Mg2⫹ absorption assay provide a novel set of tools, which can be applied to investigate the intestinal absorption of Mg2⫹ and other ions in a physiological relevant in vivo setting. ACKNOWLEDGMENTS We thank B. Lemmers-van der Weem and M. School (Central Animal Facility, Nijmegen) for excellent technical assistance with the animal experiments. We thank Dr. Maxime Blanchard for scientific discussions and help in the preparation of the figures. GRANTS This work was supported by grants from the Netherlands Organization for Scientific Research (ZonMw 9120.8026, NWO ALW 818.02.001), a Vici Grant for J. Hoenderop (NWO-ZonMw 016.130.668), and EURenOmics funding from the European Union seventh Framework Programme (FP7/20072013, agreement n° 305608). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS A.L.L., L.M., and J.G.H. conception and design of research; A.L.L., D.R., and E.S. performed experiments; A.L.L., E.S., and J.E. analyzed data; A.L.L., P.I.N., L.M., and J.G.H. interpreted results of experiments; A.L.L. prepared figures; A.L.L. drafted manuscript; A.L.L., P.I.N., L.M., R.J.B., and J.G.H. edited and revised manuscript; R.J.B. and J.G.H. approved final version of manuscript. REFERENCES 1. Abrams SA. Assessing mineral metabolism in children using stable isotopes. Pediatr Blood Cancer 50: 438 –441; discussion 451, 2008. 2. Bonny O, Rubin A, Huang CL, Frawley WH, Pak CY, Moe OW. Mechanism of urinary calcium regulation by urinary magnesium and pH. J Am Soc Nephrol 19: 1530 –1537, 2008. 3. Bronner F. Recent developments in intestinal calcium absorption. Nutr Rev 67: 109 –113, 2009. 4. Bronner F, Pansu D. Nutritional aspects of calcium absorption. J Nutr 129: 9 –12, 1999. 5. Carney SL, Wong NL, Quamme GA, Dirks JH. Effect of magnesium deficiency on renal magnesium and calcium transport in the rat. J Clin Invest 65: 180 –188, 1980. 6. Christakos S, Dhawan P, Porta A, Mady LJ, Seth T. Vitamin D and intestinal calcium absorption. Mol Cell Endocrinol 347: 25–29, 2011. 7. Coudray C, Feillet-Coudray C, Rambeau M, Tressol JC, Gueux E, Mazur A, Rayssiguier Y. The effect of aging on intestinal absorption and

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org

G216

8.

9.

10. 11. 12.

13. 14.

15.

16. 17. 18.

19. 20. 21.

22. 23. 24.

25. 26. 27. 28.

ABSORPTION OF Ca AND Mg2⫹ ALONG THE GASTROINTESTINAL TRACT

status of calcium, magnesium, zinc, and copper in rats: a stable isotope study. J Trace Elem Med Biol 20: 73–81, 2006. Coudray C, Pepin D, Tressol JC, Bellanger J, Rayssiguier Y. Study of magnesium bioavailability using stable isotopes and the inductivelycoupled plasma mass spectrometry technique in the rat: single and double labelling approaches. Br J Nutr 77: 957–970, 1997. Coudray C, Rambeau M, Feillet-Coudray C, Gueux E, Tressol JC, Mazur A, Rayssiguier Y. Study of magnesium bioavailability from ten organic and inorganic Mg salts in Mg-depleted rats using a stable isotope approach. Magnes Res 18: 215–223, 2005. Dimke H, Hoenderop JG, Bindels RJ. Molecular basis of epithelial Ca2⫹ and Mg2⫹ transport: insights from the TRP channel family. J Physiol 589: 1535–1542, 2011. Fonty G, Gouet P. Fibre-degrading microorganisms in the monogastric digestive tract. Anim Feed Sci Technol 23: 91–107, 1989. Fujita H, Sugimoto K, Inatomi S, Maeda T, Osanai M, Uchiyama Y, Yamamoto Y, Wada T, Kojima T, Yokozaki H, Yamashita T, Kato S, Sawada N, Chiba H. Tight junction proteins claudin-2 and -12 are critical for vitamin D-dependent Ca2⫹ absorption between enterocytes. Mol Biol Cell 19: 1912–1921, 2008. Geibel JP, Hebert SC. The functions and roles of the extracellular Ca2⫹-sensing receptor along the gastrointestinal tract. Annu Rev Physiol 71: 205–217, 2009. Groenestege WM, Hoenderop JG, van den Heuvel L, Knoers N, Bindels RJ. The epithelial Mg2⫹ channel transient receptor potential melastatin 6 is regulated by dietary Mg2⫹ content and estrogens. J Am Soc Nephrol 17: 1035–1043, 2006. Groenestege WM, Thebault S, van der Wijst J, van den Berg D, Janssen R, Tejpar S, van den Heuvel LP, van Cutsem E, Hoenderop JG, Knoers NV, Bindels RJ. Impaired basolateral sorting of pro-EGF causes isolated recessive renal hypomagnesemia. J Clin Invest 117: 2260 –2267, 2007. Hardwick LL, Jones MR, Brautbar N, Lee DB. Magnesium absorption: mechanisms and the influence of vitamin D, calcium and phosphate. J Nutr 121: 13–23, 1991. Hoenderop JG, Bindels RJ. Calciotropic and magnesiotropic TRP channels. Physiology (Bethesda) 23: 32–40, 2008. Hoenderop JG, Muller D, Van Der Kemp AW, Hartog A, Suzuki M, Ishibashi K, Imai M, Sweep F, Willems PH, Van Os CH, Bindels RJ. Calcitriol controls the epithelial calcium channel in kidney. J Am Soc Nephrol 12: 1342–1349, 2001. Hoenderop JG, Nilius B, Bindels RJ. Calcium absorption across epithelia. Physiol Rev 85: 373–422, 2005. Hoenderop JG, Nilius B, Bindels RJ. ECaC: the gatekeeper of transepithelial Ca2⫹ transport. Biochim Biophys Acta 1600: 6 –11, 2002. Hume I. Fermentation in the hindgut of mammals. In: Gastrointestinal Microbiology. Gastrointestinal Ecosytems and Fermentations, edited by Mackie RI, White BA. New York: Chapman & Hall, 1997, vol. 1, p. 84 –115. Jahnen-Dechent W, Ketteler M. Magnesium basics. Clin Kidney J 5: i3–i14, 2012. Lambers TT, Bindels RJ, Hoenderop JG. Coordinated control of renal Ca2⫹ handling. Kidney Int 69: 650 –654, 2006. Larsson MH, Simren M, Thomas EA, Bornstein JC, Lindstrom E, Sjovall H. Elevated motility-related transmucosal potential difference in the upper small intestine in the irritable bowel syndrome. Neurogastroenterol Motil 19: 812–820, 2007. Le Grimellec C, Roinel N, Morel F. Simultaneous Mg, Ca, P, K, Na and Cl analysis in rat tubular fluid. II. During acute Mg plasma loading. Pflügers Arch 340: 197–210, 1973. Lieben L, Carmeliet G, Masuyama R. Calcemic actions of vitamin D: effects on the intestine, kidney and bone. Best Pract Res Clin Endocrinol Metab 25: 561–572. Lifshitz F, Harrison HC, Harrison HE. Effects of vitamin D on magnesium metabolism in rats. Endocrinology 81: 849 –853, 1967. Nair AV, Hocher B, Verkaart S, van Zeeland F, Pfab T, Slowinski T, Chen YP, Schlingmann KP, Schaller A, Gallati S, Bindels RJ, Konrad M, Hoenderop JG. Loss of insulin-induced activation of TRPM6 mag-

29.

30.

31. 32. 33.

34.

35.

36.

37.

38.

39.

40. 41.

42.

43.

44. 45.

46.

47.

nesium channels results in impaired glucose tolerance during pregnancy. Proc Natl Acad Sci USA 109: 11324 –11329. Nijenhuis T, Hoenderop JG, Bindels RJ. TRPV5 and TRPV6 in Ca2⫹ (re)absorption: regulating Ca2⫹ entry at the gate. Pflügers Arch 451: 181–192, 2005. Nilius B, Prenen J, Hoenderop JG, Vennekens R, Hoefs S, Weidema AF, Droogmans G, Bindels RJ. Fast and slow inactivation kinetics of the Ca2⫹ channels ECaC1 and ECaC2 (TRPV5 and TRPV6). Role of the intracellular loop located between transmembrane segments 2 and 3. J Biol Chem 277: 30852–30858, 2002. Quamme GA. Effect of hypercalcemia on renal tubular handling of calcium and magnesium. Can J Physiol Pharmacol 60: 1275–1280, 1982. Quamme GA. Recent developments in intestinal magnesium absorption. Curr Opin Gastroenterol 24: 230 –235, 2008. Quinn SJ, Thomsen AR, Egbuna O, Pang J, Baxi K, Goltzman D, Pollak M, Brown EM. CaSR-mediated interactions between calcium and magnesium homeostasis in mice. Am J Physiol Endocrinol Metab 304: E724 –E733, 2013. Renkema KY, Alexander RT, Bindels RJ, Hoenderop JG. Calcium and phosphate homeostasis: concerted interplay of new regulators. Ann Med 40: 82–91, 2008. Renkema KY, Nijenhuis T, van der Eerden BC, van der Kemp AW, Weinans H, van Leeuwen JP, Bindels RJ, Hoenderop JG. Hypervitaminosis D mediates compensatory Ca2⫹ hyperabsorption in TRPV5 knockout mice. J Am Soc Nephrol 16: 3188 –3195, 2005. Revusova V, Gratzlova J, Zvara V. Impaired renal tubular reabsorption of magnesium (TRMg) in Ca-containing kidney stone formers. Int Urol Nephrol 16: 237–242, 1984. Ritchie G, Kerstan D, Dai LJ, Kang HS, Canaff L, Hendy GN, Quamme GA. 1,25(OH)2D3 stimulates Mg2⫹ uptake into MDCT cells: modulation by extracellular Ca2⫹ and Mg2⫹. Am J Physiol Renal Physiol 280: F868 –F878, 2001. Ryazanova LV, Rondon LJ, Zierler S, Hu Z, Galli J, Yamaguchi TP, Mazur A, Fleig A, Ryazanov AG. TRPM7 is essential for Mg2⫹ homeostasis in mammals. Nat Commun 1: 109, 2010. Schuette S, Vereault D, Ting BT, Janghorbani M. Accurate measurement of stable isotopes of magnesium in biological materials with inductively coupled plasma mass spectrometry. Analyst 113: 1837–1842, 1988. Schweigel M, Martens H. Magnesium transport in the gastrointestinal tract. Front Biosci 5: D666 –D677, 2000. Thebault S, Alexander RT, Tiel Groenestege WM, Hoenderop JG, Bindels RJ. EGF increases TRPM6 activity and surface expression. J Am Soc Nephrol 20: 78 –85, 2009. Van Cromphaut SJ, Dewerchin M, Hoenderop JG, Stockmans I, Van Herck E, Kato S, Bindels RJ, Collen D, Carmeliet P, Bouillon R, Carmeliet G. Duodenal calcium absorption in vitamin D receptor-knockout mice: functional and molecular aspects. Proc Natl Acad Sci USA 98: 13324 –13329, 2001. van de Graaf SF, Bindels RJ, Hoenderop JG. Physiology of epithelial Ca2⫹ and Mg2⫹ transport. Rev Physiol Biochem Pharmacol 158: 77–160, 2007. Wasserman RH. Vitamin D and the dual processes of intestinal calcium absorption. J Nutr 134: 3137–3139, 2004. Woudenberg-Vrenken TE, Lameris AL, Weissgerber P, Olausson J, Flockerzi V, Bindels RJ, Freichel M, Hoenderop JG. Functional TRPV6 channels are crucial for transepithelial Ca2⫹ absorption. Am J Physiol Gastrointest Liver Physiol 303: G879 –G885, 2012. Woudenberg-Vrenken TE, van der Eerden BC, van der Kemp AW, van Leeuwen JP, Bindels RJ, Hoenderop JG. Characterization of vitamin D-deficient klotho⫺/⫺ mice: do increased levels of serum 1,25(OH)2D3 cause disturbed calcium and phosphate homeostasis in klotho⫺/⫺ mice? Nephrol Dial Transplant 27: 4061–4068, 2012. Yamazaki D, Funato Y, Miura J, Sato S, Toyosawa S, Furutani K, Kurachi Y, Omori Y, Furukawa T, Tsuda T, Kuwabata S, Mizukami S, Kikuchi K, Miki H. Basolateral Mg2⫹ extrusion via CNNM4 mediates transcellular Mg2⫹ transport across epithelia: a mouse model. PLoS Genet 9: e1003983, 2013.

AJP-Gastrointest Liver Physiol • doi:10.1152/ajpgi.00093.2014 • www.ajpgi.org