Abstract

Epidemiological data suggest a protective role of calcium and vitamin D against colorectal tumor pathogenesis. 1,25-dihydroxyvitamin D 3 (1,25-D 3 ) is a key determinant of calcium homeostasis, cell proliferation and differentiation. Calcium in the intestinal lumen functions as a growth regulator and may prevent cancer by direct reduction of colonocyte proliferation. While calcium or vitamin D can counteract proliferation by itself, they could also interact if nutritional calcium were to modulate colonic vitamin D synthesis. In this paper we demonstrate that colonic and renal vitamin D hydroxylases are regulated independently. When mice were fed a modified AIN-76 diet containing low dietary calcium (0.1 or 0.04%) fecal calcium content was as low as 5% of that found in mice on a 0.9% calcium containing diet. Low fecal calcium concentration enhanced proliferating cell nuclear antigen expression in the colon mucosa and reduced that of the cyclin dependent kinase inhibitor p21. While low dietary calcium did not affect colonic expression of VDR or 25-hydroxyvitamin D 3 1α-hydroxylase (CYP27B1) mRNA, it influenced their renal expression in the expected manner by elevating the CYP27B1 expression and reducing VDR and 25-hydroxyvitamin D 3 24-hydroxylase (CYP24) expression. In contrast, low calcium diets significantly augmented colonic CYP24 mRNA expression, but only in the ascending colon. This might result in reduced colonic accumulation of 1,25-D 3 during hyperproliferation caused by low dietary calcium and might support site-specific tumorigenesis. The important realization that low dietary calcium by itself is a risk factor for colorectal carcinogenesis and that colonic and renal vitamin D hydroxylases indeed are regulated differently from each other will provide novel approaches for colon cancer prevention.

Introduction

Epidemiological studies have demonstrated an inverse correlation between risk of several cancers and sun exposure, dietary fish consumption and serum levels of 25-hydroxyvitamin D 3 (25-D 3 ) ( 13 ). Cholecalciferol (vitamin D 3 ) is nutritionally provided only in oily fish or fortified foods, whereas a major part of this precursor is synthesized in the skin from 7-dehydrocholesterol, by solar UVB radiation. Cholecalciferol is 25-hydroxylated in the liver by CYP27A1 ( 4 ) and possibly by CYP2R1 ( 5 ). The resultant serum levels of 25-D 3 are highly dependent on season, age and skin pigmentation (reviewed in ref. 6 ). However, fluctuation of serum 25-D 3 is increasingly common even among younger people as a direct consequence of an indoor lifestyle. Recently it was shown that a large section of the Austrian population is vitamin D insufficient (<12 ng/ml serum) whereas outright vitamin D deficiency, as found in patients with rickets (<5 ng/ml) is much more rare ( 7 ).

While 25-D 3 is present in human serum at nanomolar concentrations, levels of the active vitamin D metabolite 1,25-dihydroxyvitamin-D 3 (1,25-D 3 ) synthesized in the kidney by the 1α-hydroxylation by 25-hydroxyvitamin D 3 1α-hydroxylase (CYP27B1) are in the picomolar range and are stringently regulated in order to maintain calcium homeostasis. 25-Hydroxyvitamin D 3 -24-hydroxylase (CYP24) is the enzyme responsible for the first step in 1,25-D 3 catabolism. In the proximal tubule of the kidney, expression and activity of both the synthesizing and catabolic vitamin D hydroxylases (CYP27B1 and CYP24) are tightly and reciprocally regulated by 1,25-D 3 and parathyroid hormone (PTH) ( 810 ). 1,25-D 3 synthesis is primarily upregulated by the low serum calcium concentration, and consequently by PTH, by increasing the transcription of CYP27B1 mRNA, whereas 1,25-D 3 by itself downregulates CYP27B1 expression by negative feedback control ( 11 ). Transcription of CYP24 mRNA is under the direct regulation of 1,25-D 3 : the ligand/vitamin D receptor (VDR) complex acts as a transactivation factor via two vitamin D response elements (VDRE) in the promoter of the CYP24 gene ( 12 , 13 ). PTH suppresses renal CYP24 transcription and activity probably by reducing CYP24 mRNA stability ( 14 , 15 ).

1,25-D 3 , besides maintaining mineral ion homeostasis, is a potent inhibitor of proliferation and promotes differentiation and apoptosis in a variety of cancer cells in vitro ( 1618 ), however, only at nanomolar concentrations. Therefore it seems unlikely that the picomolar levels of 1,25-D 3 present in the serum could potentially protect against malignancies. In epidemiological studies, normal to high levels of 1,25-D 3 indeed did not correlate inversely with tumor incidence ( 19 ). However, we have demonstrated that human colonic cell lines can synthesize and degrade 1,25-D 3 ( 20 ), and that cells isolated from human colon tumors also possess this ability ( 21 ). Expression of vitamin D hydroxylases in colonic cell lines and primary cultures is regulated by the epidermal growth factor (a common mitogen during colonic tumorigenesis) and by 1,25-D 3 ( 22 ). When we evaluated expression of CYP27B1 and of VDR mRNA in human colonic tissue, it became apparent that, in comparison with adjacent normal tissue, these two components of the vitamin D system were strongly expressed in high to medium differentiated tumors, whereas in undifferentiated tumors they were barely detectable ( 21 , 23 , 24 ). These observations suggested that extrarenal production of 1,25-D 3 in the human colon could potentially prevent tumorigenesis, and perhaps retard its progression as long as cells were still rather differentiated.

Another factor well known in colonic tumor prevention is dietary calcium. Garland et al . ( 25 ) have shown that the risk of colorectal cancer (CRC) is inversely correlated with dietary vitamin D and calcium. In a recent meta-analysis Cho et al . ( 26 ) pooled the primary data from 10 cohort studies in five countries and calculated multivariable relative risks for categories of milk intake and quintiles of calcium intake with 95% confidence intervals. The results were consistent across studies and sexes, and showed that higher consumption of milk and calcium is associated with lower risk of CRC. Kállay et al . ( 27 ) were able to demonstrate in vitro that high extracellular calcium concentration could reduce proliferation of the Caco-2 colon cancer cell line and that this was mediated, at least in part, by the calcium-sensing receptor. This indicates that nutritional calcium reaching the colon lumen is likely to affect colonocytes directly, and an interaction with the colonic 1,25-D 3 -synthesizing machinery suggests itself.

In this paper we present evidence that the regulation of colonic vitamin D hydroxylase expression differs from that observed in the kidney. In a mouse model we demonstrate that a diet deficient only in calcium triggers hyperproliferation of colonocytes and increases both CYP24 mRNA and protein levels in the proximal colon. It is not unreasonable to expect that increased levels of CYP24 protein lead to enhanced degradation of the secosteroid and thus to reduced colonic accumulation of 1,25-D 3 , which might then lead to increased tumor incidence.

Materials and methods

Animals

C57BL/6 mice were housed in the Center for Laboratory Animal Care of the Medical University of Vienna in a contained environment. Mice were weaned at 2–3 weeks of age and were then fed ad libitum a standard diet (basic AIN 76A plus 20% lactose instead of casein) containing 0.9, 0.1% or 0.04% calcium. To evaluate the effect of a lack of genomic vitamin D action on vitamin D hydroxylase expression, we used the vitamin D receptor knockout (VDR −/− ) mouse ( 28 ). VDR −/− mice were fed the standard diet with 0.9% calcium.

Blood was drawn from the tail vein before killing the animals at 5 months. Colon and kidney tissue samples were collected. A part of the proximal colon and of the distal colon was fixed in formalin. Colon lumina were cut open and rinsed, mucosal tissue was divided for the use in immunoblotting or real time PCR and frozen in liquid nitrogen. Treatment groups consisted of at least six animals. Study protocols were approved by the Committee of Animal Experimentation of the Medical University of Vienna and by the Austrian Ministry of Science and Education.

Serum parameters

25-D 3 and its bioactive metabolite 1,25-D 3 were measured with commercial radioimmunoassay kits from DiaSorin (Sillwater, MN) ( 29 ). Interassay coefficients of variation were 9 and 13.5% (14 and 138 nM, 25-D 3 ) and 20% (89 and 238 pM, 1,25-D 3 ). Total calcium and phosphate were measured by a Hitachi D2400-module clinical chemistry analyzer (Roche Diagnostics, Mannheim, Germany).

Calcium measurement in feces

Feces were collected for three consecutive days from different feeding groups, were dried and digested with 2 ml HNO 3 (65%, suprapur Merck, Germany), ashed and dissolved in 100 ml 1 M HNO 3 . Calcium measurements were performed in triplicate, simultaneously at two wavelengths (λ 1 = 315.887, λ 2 = 317.933) using inductively coupled plasma atomic emission spectrometry (ICP-AES; Perkin Elmer Optima 3000 XL). The calibration ranges were 1.0–50.0 mg/l. The calibration curve was based on five single element standard solutions in 1 M HNO 3 including a blank. The determined limit of detection was 50 µg/l digest.

Cell culture

The human colon adenocarcinoma-derived cell line Caco-2 was cultured until confluence in Dulbecco's modified Eagles' Medium (DMEM) supplemented with 10% fetal calf serum, glutamine (4 mM), 10 mM HEPES, penicillin (100 U/ml) and streptomycin (100 µg/ml) (all from Invitrogen, Paisley, UK). The final extracellular calcium concentration ([Ca 2+ ] o ) in this medium is 1.80 mM. After confluency, cells were cultured for 48 h in serum-free DMEM supplemented with 5 µg/ml insulin, 5 µg/ml transferrin and 5 ng/ml sodium selenite (Sigma, St Louis, MO) prior to treatment with calcium-free (0.0 mM) medium.

RNA isolation and real-time RT–PCR

Using random hexamer primer and SuperScript II (Invitrogen), 2 µg of total RNA extracted with TRIZOL reagent (Invitrogen Ltd, Paisley, UK) were reverse-transcribed into single-strand cDNA. The RT reaction was performed at 42°C for 60 min, 45°C for 15 min, followed by 70°C for 10 min to inactivate the reaction.

In order to assess mRNA levels accurately, we used real-time RT–PCR and quantified the expression levels by the comparative ΔΔC T method. The reliability of the RT–PCR experiment was improved by including an invariant endogenous control, 18S rRNA in the assay to correct for sample to sample variations in RT–PCR efficiency and errors in sample quantification. Relative abundance values were then calculated for 18S rRNA as well as for the experimental sequence (VDR, CYP27B1 and CYP24). For each experimental sample, the relative abundance value obtained was normalized to the value derived from the control sequence (18S rRNA) in the corresponding sample. The normalized values for different samples were directly compared. A pool of mouse colon cDNAs were designated as the ‘calibrator’, and the relative expression levels of all other samples were expressed relative to the calibrator. The real-time PCR was performed on an ABI Prism 7700 Sequence Detection System (Applied Biosystems, Foster City, CA). Triplicates were set up for each sample and transcript under investigation. PCR conditions were 50°C for 2 min (UDG (Uracil DNA Glycosylase) decontamination step), 94°C for 2 min, which was followed by 40 cycles of 94°C for 15 s and 60°C for 30 s. Primers and internal probes were designed using Primer Express (Applied Biosystems) and were located on different exons to prevent the amplification of contaminating genomic DNA.

Primer sequences

Mouse CYP24 (forward): 5′-CCA GAG CGT GCT GCC TG-3′; (reverse): 5′-TTA AAT AGG GCA TAT TCC TCA CAT CTT-3′; (TaqMan probe): 5′-CAA CCA GAC GCC ACG GGC G-3′.

Mouse VDR (forward): 5′-CGA TCT GTG GAG TGT GTG GAG ACC-3′; (reverse): 5′-CTT CAT CAT GCC AAT GTC CAC GCA G-3′; (TaqMan probe): 5′-CTG TTC ACC TGC CCC TTC AAT GGA GAT-3′.

Mouse CYP27B1 (forward): 5′-ATG TTT GCC TTT GCC CAG AG-3′; (reverse): 5′-GAC GGC ATA TCC TCC TCA GG-3′; (TaqMan probe): 5′-CCC TGG TTC CTC ATC GCA GCT TCA-3′.

Real-time PCR analysis of the RNA obtained from the human Caco-2 cell line was performed with FAM-labeled ‘Assays-on-Demand™Gene Expression’ products for human 18S rRNA, proliferating cell nuclear antigen (PCNA), VDR, CYP27B1 and CYP24 (Applied Biosystems).

Immunoblotting

Snap-frozen large bowel tissue was homogenized in 10 mM Tris buffer (pH 7.4) with protease inhibitors and 1% SDS, and was boiled for 5 min. Homogenates were centrifuged and the supernatant was frozen at –70°C. Proteins were separated by 12% SDS–PAGE, were transferred to nitrocellulose membranes and then exposed overnight to the following primary antibodies: rabbit polyclonal anti-vitamin D receptor (Santa Cruz Biotechnology, Santa Cruz, CA), sheep anti-CYP27B1 (The Binding Site, Heidelberg, Germany), rabbit polyclonal CYP24 antibody (courtesy Dr H.J.Ambrecht), rabbit polyclonal p21 antibody (Santa Cruz Biotechnology), mouse monoclonal anti-PCNA (DakoCytomation, Glostrup, Denmark) and mouse monoclonal anti-cytokeratin (CK) 8 antibody (Cymbus Biotechnology Ltd, Chandlers Ford, UK). CK8, an invariant marker of simple epithelia was used as internal control for the immunoblots. Horseradish peroxidase-conjugated secondary antibodies (Amersham Biosciences, Little Chalfont, UK) were used. The immunogenic peptide employed for deriving the CYP24 antibody, which was used as a negative control after pre-incubation with the antibody, was provided by Dr H.J.Armbrecht.

Detection was performed with the SuperSignal CL-HRP Substrate system (Pierce, Rockford, IL). Bands were evaluated by densitometry with a video camera imaging system (Herolab, Wiesloch, Germany).

Immunohistochemistry

Fresh tissues were formalin-fixed and paraffin-embedded. After deparaffinization and rehydration, slides were boiled in 0.01 M citrate buffer (pH 6.0) in a microwave oven at 600 W for antigen retrieval. Sections were stained according to the Histomouse-SP kit instructions (Zymed, San Francisco, CA) with a mouse monoclonal anti-PCNA antibody (DAKO, Copenhagen, DK). The negative control was mouse IgG (Jackson ImmunoResearch Laboratories, West Grove, PA). Sections were viewed in a Nikon Eclipse E400 microscope. Images were captured into a computer using a Coolpix 950 digital camera (Nikon, Tokyo, Japan).

Statistical analyses

Data are presented as mean ± SE. Student's t -test was used for statistical group analysis. Correlation coefficients were calculated for possible interrelations between variables. P -values <0.05 were considered statistically significant.

Results

Expression of vitamin D hydroxylases in the VDR −/− mouse model

VDR −/− mice grow up normally until weaning, but after weaning they develop a phenotype similar to vitamin D-dependent rickets type II, showing hypocalcemia, hypophosphatemia, elevated serum 1,25-D 3 , hyperparathyroidism and alopecia ( 30 ).

Supplementation of the basic AIN76A diet with 20% lactose, which enhances calcium diffusion in the small intestine, prevented development of the rickets phenotype ( 31 , 32 ) by the almost normalization of serum calcium levels in VDR −/− animals ( Table I ). Lack of VDR caused an enormous increase of 1,25-D 3 concentration in the serum of VDR −/− animals. Actually, synthesis of the steroid hormone was so high that 25-D 3 stores became depleted ( Table I ).

Table I.

Concentrations of calcium, 1,25-D 3 and 25-D 3 in serum of VDR +/+ and VDR −/− mice

Genotype
 
VDR +/+
 
VDR −/−
 
Calcium (nmol/l) 2.27 ± 0.04 1.95 ± 0.4 
1,25-D 3 (pmol/l)  86 ± 48  7841 ± 598 * 
25-D 3 (nmol/l)  189 ± 37  98.6 ± 65 ** 
Genotype
 
VDR +/+
 
VDR −/−
 
Calcium (nmol/l) 2.27 ± 0.04 1.95 ± 0.4 
1,25-D 3 (pmol/l)  86 ± 48  7841 ± 598 * 
25-D 3 (nmol/l)  189 ± 37  98.6 ± 65 ** 

Serum parameters were measured as described in Materials and methods. Data are presented as mean ± SD ( n = 10 animals).

*

P < 0.001.

**

P < 0.01 VDR −/− versus VDR +/+ .

The very high serum levels of 1,25-D 3 were reflected by CYP27B1 mRNA expression in the kidney of VDR −/− mice: it was at least 70-fold higher than in VDR +/+ animals ( Figure 1A ). This difference, however, was not at all apparent when comparing colonic CYP27B1 levels: evaluation by real time RT–PCR actually showed a reduction in CYP27B1 mRNA in the colon of VDR −/− mice when compared with VDR +/+ animals ( Figure 1B ).

Fig. 1.

Expression of CYP27B1 and CYP24 mRNA ( A ) in kidney and ( B ) in colon of VDR +/+ and VDR −/− mice. Data determined by real-time RT–PCR were quantified by the comparative ΔΔC T (C T : threshold cycle) method and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. Data are mean ± SE ( n = 6 mice). *P < 0.05; **P < 0.01 versus VDR +/+ mice.

Fig. 1.

Expression of CYP27B1 and CYP24 mRNA ( A ) in kidney and ( B ) in colon of VDR +/+ and VDR −/− mice. Data determined by real-time RT–PCR were quantified by the comparative ΔΔC T (C T : threshold cycle) method and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. Data are mean ± SE ( n = 6 mice). *P < 0.05; **P < 0.01 versus VDR +/+ mice.

Renal CYP24 mRNA expression, as expected, was also dependent upon the presence or absence of active VDR: in the kidneys of VDR +/+ mice we found significantly higher levels than in the kidneys of VDR −/− animals ( Figure 1A ). In the colon, however, we could not observe any differences in CYP24 mRNA expression between VDR +/+ and VDR −/− mice ( Figure 1B ).

Modulation of proliferation and vitamin D hydroxylase expression by low dietary calcium in normal C57BL/6 mice

In these studies we reduced the nutritional intake of calcium in order to achieve extremely low fecal calcium concentrations, whereas serum calcium levels were maintained normal. Control mice were fed 0.9% calcium in their AIN76A diet. We also fed AIN76A containing 0.1% calcium, which is comparable to the actual human dietary intake prevalent in most of the Western industrialized countries (400–500 mg/day) ( 33 ). This resulted in the reduction of fecal calcium content to 20% of that found in animals on 0.9% dietary calcium. Reducing dietary calcium further to 0.04% led to a lowering of fecal calcium content to 5%. Serum calcium levels in mice on low nutritional calcium were normal, whereas serum 1,25-D 3 increased significantly ( Table II ).

Table II.

Influence of dietary calcium concentrations on fecal calcium excretion and serum parameters

Dietary Ca conc.
 
0.90%
 
0.10%
 
0.04%
 
Fecal calcium (mg/g) 55.8 ± 4 9.7 ± 1 2.9 ± 0.4 
Serum calcium (nmol/l) 2.27 ± 0.04 2.18 ± 0.10 2.32 ± 0.13 
1,25-D 3 (pmol/l)  86 ± 48  198 ± 27 *  314 ± 100 * 
25-D 3 (nmol/l)  189 ± 37 134 ± 15 96 ± 43 
Dietary Ca conc.
 
0.90%
 
0.10%
 
0.04%
 
Fecal calcium (mg/g) 55.8 ± 4 9.7 ± 1 2.9 ± 0.4 
Serum calcium (nmol/l) 2.27 ± 0.04 2.18 ± 0.10 2.32 ± 0.13 
1,25-D 3 (pmol/l)  86 ± 48  198 ± 27 *  314 ± 100 * 
25-D 3 (nmol/l)  189 ± 37 134 ± 15 96 ± 43 

Parameters were measured as described in Materials and methods. Data are presented as mean  ± SD ( n = 10 animals).

*

P < 0.05 versus 0.90% calcium diet.

Influence of dietary calcium on PCNA and p21 protein expression

To evaluate the influence of low dietary calcium on colonocyte proliferation, we stained mouse colon sections for expression of PCNA ( Figure 2 ). Feeding low dietary calcium caused an increase in proliferation of colonocytes along the whole colon. Mice fed a diet with 0.9% calcium had low PCNA levels with positive cells located at the crypt base, whereas mice fed with 0.1 or 0.04% calcium responded with significantly enhanced positive staining for PCNA up to the upper half of the crypt ( Figure 2A , proximal colon shown). Quantitative evaluation of this proliferation marker by immunoblotting showed a more than 2-fold increase in PCNA expression even with 0.1% dietary calcium ( Figure 2B , proximal colon shown). We also evaluated expression of the cyclin dependent kinase inhibitor p21 by immunoblotting: this cell cycle inhibitor showed less sensitivity to lowered fecal calcium, since only the diet containing 0.04% calcium reduced p21 expression by 50%. This, however, was only apparent in the proximal colon ( Figure 2B ), and not at all in the distal colon (not shown).

Fig. 2.

Effect of dietary calcium concentration on proliferation and p21 expression in the colon of C57BL/6 mice. ( A ) Immunohistochemical detection of PCNA in formalin-fixed mouse colon sections. Animals received basic AIN 76A diet containing 0.9, 0.1 and 0.04% calcium. Bar = 50 µm. ( B ) Quantification of PCNA and p21 protein expression by densitometric evaluation of western blots compared with the epithelial cell marker cytokeratin (CK) 8. In the upper panel data are expressed as mean ± SE ( n = 6 animals). *P < 0.05; **P < 0.01 versus mice on 0.9% calcium diet. Lower panel shows representative immunoblots of 150 µg total proteins extracted from mouse colon and resolved by SDS–PAGE. (The size of the specific bands in kDa is shown on the left.)

Fig. 2.

Effect of dietary calcium concentration on proliferation and p21 expression in the colon of C57BL/6 mice. ( A ) Immunohistochemical detection of PCNA in formalin-fixed mouse colon sections. Animals received basic AIN 76A diet containing 0.9, 0.1 and 0.04% calcium. Bar = 50 µm. ( B ) Quantification of PCNA and p21 protein expression by densitometric evaluation of western blots compared with the epithelial cell marker cytokeratin (CK) 8. In the upper panel data are expressed as mean ± SE ( n = 6 animals). *P < 0.05; **P < 0.01 versus mice on 0.9% calcium diet. Lower panel shows representative immunoblots of 150 µg total proteins extracted from mouse colon and resolved by SDS–PAGE. (The size of the specific bands in kDa is shown on the left.)

Renal expressions of VDR, CYP27B1 and CYP24 mRNA

Renal VDR mRNA expression measured by real time RT–PCR was significantly down-regulated even by 0.1% dietary calcium, and there was no further decrement found in mice on 0.04% dietary calcium ( Figure 3A ). As expected, renal CYP27B1 was almost 30-fold elevated ( Figure 3B ) and renal CYP24 was almost undetectable when 0.1% calcium was fed ( Figure 3C ). Again there was no longer a significant difference when even less calcium (0.04%) was fed ( Figure 3C ). These modulations of renal vitamin D hydroxylase mRNA expressions typically reflect the enhanced production of serum 1,25-D 3 following a low calcium diet, as demonstrated in Table II . Decreased renal CYP24 levels in mice that were fed a low calcium diet are most likely due to downregulation by PTH, but could be attributed also to significantly lower VDR expression.

Fig. 3.

Effect of dietary calcium concentration on the expression of ( A ) VDR, ( B ) CYP27B1 and ( C ) CYP24 mRNA in the kidney. Animals received basic AIN 76A diet containing 0.9% (black bars), 0.1% (grey bars) and 0.04% calcium (white bars). Data expressed as mean 2 -ΔΔCT ± SE ( n = 6 mice) were assessed by real-time RT–PCR analysis and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. *P < 0.05; **P < 0.01 versus mice on 0.9% calcium diet.

Fig. 3.

Effect of dietary calcium concentration on the expression of ( A ) VDR, ( B ) CYP27B1 and ( C ) CYP24 mRNA in the kidney. Animals received basic AIN 76A diet containing 0.9% (black bars), 0.1% (grey bars) and 0.04% calcium (white bars). Data expressed as mean 2 -ΔΔCT ± SE ( n = 6 mice) were assessed by real-time RT–PCR analysis and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. *P < 0.05; **P < 0.01 versus mice on 0.9% calcium diet.

Colonic expression of VDR, CYP27B1 and CYP24 mRNA and protein

We quantified the expression of VDR and of vitamin D hydroxylases in the ascending and descending colon by real time RT–PCR and immunoblotting ( Figure 4 ). As already indicated in a previous study ( 29 ) VDR expression is higher in the ascending than in the descending colon. Lowered dietary calcium does not modulate significantly VDR mRNA ( Figure 4A ) or protein levels (not shown), either in the proximal or in the distal colon. As demonstrated in Figure 4B , CYP27B1 mRNA expression was similar in right and left colon at high nutritional calcium levels, and reducing nutritional calcium did not change this significantly. Similar results were obtained by immunoblotting (not shown). Expression of CYP24 mRNA was higher in ascending than in the descending colon of mice fed with a high calcium containing diet ( Figure 4C ), and this difference was even more pronounced when we evaluated CYP24 protein expression ( Figure 4D ). Decreasing the calcium concentration of the diet to 0.04% resulted in threefold elevation of CYP24 mRNA ( Figure 4C ). This, however, occurred solely in the ascending colon, whereas in the descending colon, low fecal calcium content had no effect on CYP24 mRNA expression ( Figure 4C ). CYP24 protein expression as shown by immunoblotting in Figure 4D , demonstrated its extreme site specificity also. In the ascending colon, presence of the 52 kDa band is pronounced (lane 1) and is upregulated by 0.1% as well as by 0.04% dietary calcium (lanes 2 and 3), whereas expression is almost nil in the descending colon (lanes 5, 6 and 7). Specificity of the antibody was demonstrated by the absence of staining at 52 kDa when the CYP24 antibody was pre-absorbed with the synthetic peptide against which it was raised (lane 4).

Fig. 4.

Effect of dietary calcium concentration on expression of ( A ) VDR, ( B ) CYP27B1 and ( C ) CYP24 mRNA in ascending (C. asc.) and descending (C. desc.) colon of mice. Animals received basic AIN 76A diet containing 0.9% (black bars), 0.1% (grey bars) and 0.04% (white bars) calcium. Data determined by real-time RT–PCR were quantified by the comparative ΔΔC T method and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. Data are mean ± SE ( n = 6 mice). ( D ) Western blot analysis of CYP24 protein expression in mouse colon. Data are expressed as mean ± SE ( n = 6 mice). Lower panel shows representative immunoblots: 150 µg total proteins were extracted from ascending (C. asc.) and descending (C. desc.) colon, resolved by SDS–PAGE, transferred to nitrocellulose and probed with a polyclonal antibody against CYP24 or CK 8. Specificity of the CYP24 antibody was demonstrated by the absence of staining when the antibody was pre-absorbed with the synthetic peptide against which it was raised (lane 4). *P < 0.05 versus mice on 0.9% calcium diet.

Fig. 4.

Effect of dietary calcium concentration on expression of ( A ) VDR, ( B ) CYP27B1 and ( C ) CYP24 mRNA in ascending (C. asc.) and descending (C. desc.) colon of mice. Animals received basic AIN 76A diet containing 0.9% (black bars), 0.1% (grey bars) and 0.04% (white bars) calcium. Data determined by real-time RT–PCR were quantified by the comparative ΔΔC T method and were calculated with respect to 18S rRNA as endogenous control and compared with the pooled cDNA calibrator. Data are mean ± SE ( n = 6 mice). ( D ) Western blot analysis of CYP24 protein expression in mouse colon. Data are expressed as mean ± SE ( n = 6 mice). Lower panel shows representative immunoblots: 150 µg total proteins were extracted from ascending (C. asc.) and descending (C. desc.) colon, resolved by SDS–PAGE, transferred to nitrocellulose and probed with a polyclonal antibody against CYP24 or CK 8. Specificity of the CYP24 antibody was demonstrated by the absence of staining when the antibody was pre-absorbed with the synthetic peptide against which it was raised (lane 4). *P < 0.05 versus mice on 0.9% calcium diet.

Regulation of CYP24, CYP27B1 and PCNA by low extracellular calcium in Caco-2 cells

We had to establish whether increased expression of CYP24 in the right-sided colon observed in our mouse model was related at all to the rise of serum 1,25-D 3 following the low calcium feeding or whether it was owing to a direct effect of low fecal calcium concentration in the colon lumen. For this purpose we evaluated in vitro , in the absence of FCS and of 1,25-D 3 , if lack of calcium in the culture medium has any effect on VDR, CYP24, CYP27B1 and PCNA mRNA expression in the human colon adenocarcinoma-derived Caco-2 cell line. We had shown that low extracellular calcium concentration ([Ca 2+ ] o ) increases proliferation of Caco-2 cells ( 34 ). Indeed, PCNA expression was at least doubled, and exposure to calcium-free conditions for 48 h tripled the CYP24 expression ( Figure 5A ). These results are similar to the results obtained in our in vivo experiments ( Figures 2 and 4 ). In a parallel experiment we treated confluent Caco-2 cells for 48 or 24 h with calcium-free medium (0.0 mM). In one group, after 24 h treatment with the calcium-free medium the calcium concentration was again raised to 1.80 mM for 24 h prior to the RNA extraction. CYP24 expression was calculated relative to the control (cells cultured in 1.8 mM calcium during the whole experiment). The results (presented in Figure 5B ) show that in the calcium free medium CYP24 mRNA is significantly increased (both after 48 and 24 h treatment). Raising calcium concentrations from 0.0 mM to 1.8 mM led to a drop in CYP24 mRNA concentration to levels similar to those measured in cells exposed only to 1.8 mM calcium.

Fig. 5.

( A ) Effect of low extracellular calcium concentration (1.8 mM: white bars, 0.0 mM: black bars) on VDR, CYP27B1, CYP24 and PCNA mRNA expression in the Caco-2 cell line. Data expressed as mean 2 -ΔΔCT ± SE ( n = 3 experiments) were assessed by real-time RT–PCR analysis and were calculated with respect to 18S rRNA as endogenous control and compared with the 1.8 mM calcium-treated control. The real-time PCR was performed with FAM-labeled ‘Assays-on-Demand™Gene Expression’ products for human 18S rRNA, PCNA, VDR, CYP27B1 and CYP24 (Applied Biosystems). *P < 0.05; **P < 0.01 versus 1.8 mM calcium-treatment. ( B) Time-dependent effect of calcium-free medium (0.0 mM) on CYP24 mRNA expression. Caco-2 cells were treated for the given time-periods with calcium-free medium. In the last group, after 24 h treatment with calcium-free medium the calcium concentration was raised again to 1.80 mM [Ca 2+ ] o for 24 h prior to RNA extraction. CYP24 expression was calculated relative to the control (treatment with 1.8 mM calcium for 48 h).

Fig. 5.

( A ) Effect of low extracellular calcium concentration (1.8 mM: white bars, 0.0 mM: black bars) on VDR, CYP27B1, CYP24 and PCNA mRNA expression in the Caco-2 cell line. Data expressed as mean 2 -ΔΔCT ± SE ( n = 3 experiments) were assessed by real-time RT–PCR analysis and were calculated with respect to 18S rRNA as endogenous control and compared with the 1.8 mM calcium-treated control. The real-time PCR was performed with FAM-labeled ‘Assays-on-Demand™Gene Expression’ products for human 18S rRNA, PCNA, VDR, CYP27B1 and CYP24 (Applied Biosystems). *P < 0.05; **P < 0.01 versus 1.8 mM calcium-treatment. ( B) Time-dependent effect of calcium-free medium (0.0 mM) on CYP24 mRNA expression. Caco-2 cells were treated for the given time-periods with calcium-free medium. In the last group, after 24 h treatment with calcium-free medium the calcium concentration was raised again to 1.80 mM [Ca 2+ ] o for 24 h prior to RNA extraction. CYP24 expression was calculated relative to the control (treatment with 1.8 mM calcium for 48 h).

Discussion

Through epidemiological studies we have realized ( 35 ) that high vitamin D levels may protect against CRC, whereas the concept that extrarenal colonic 1,25-D 3 synthesis may serve as a physiological defense against colon tumorigenesis has only recently been gaining acceptance ( 21 , 23 , 24 , 36 ). It has been assumed that vitamin D insufficiency or low dietary calcium intake contribute to the development of colon cancer by different pathogenic mechanisms. Therefore, a nutritional calcium deficit and a compromised vitamin D status were seen as independent risk factors ( 37 ). Interestingly, as shown by Grau et al . ( 38 ) in their study on the effect of vitamin D and calcium supplementation on recurrence of colorectal adenomas, calcium and vitamin D status appear to act largely together in the control of colon epithelial cell proliferation. Calcium supplementation was effective only in patients with normal serum 25-D 3 levels. Conversely, high 25-D 3 levels were associated with a reduced risk of adenoma recurrence only among subjects receiving calcium supplements. Thus our working hypothesis was that calcium supplementation, in addition to adequate 25-D 3 levels, may lower the risk for CRC because vitamin D metabolism in the colon mucosa might be affected by dietary calcium in favor of higher local 1,25-D 3 production. A prerequisite for this novel concept would be the possibility to regulate colonic vitamin D hydroxylases independently from their renal counterparts.

Data accrued in our studies in the VDR −/− mouse indeed show that vitamin D hydroxylases present in colon mucosal cells are regulated differently from renal hydroxylases. Renal CYP27B1 mRNA was increased 70-fold in VDR −/− mice ( Figure 1A ), which was reflected also by very high levels of serum 1,25-D 3 ( Table I ). In contrast, in the colon CYP27B1 was rather reduced in VDR −/− mice. The renal CYP24 mRNA was high in VDR +/+ mice and was lowered to almost zero in VDR −/− mice, reflecting the absence of VDR, even if the serum of these animals contained very high (ng/ml) amounts of 1,25-D 3 . In contrast, colonic CYP24 expression was low even in the presence of active VDR and did not decrease further in the absence of VDR in VDR −/− mice ( Figure 1B ). These data demonstrate for the first time that colonic hydroxylases indeed are regulated in a manner that is independent from that of renal hydroxylases.

In the past, several studies have been performed to examine the regulation of expression of vitamin D hydroxylases in the intestine ( 3942 ), though all were concerned with the small intestine, the major site of calcium and phosphate transport. However, the small intestine is associated with a very low rate of tumor occurrence, unlike the colon and rectum ( 43 ).

In order to study the regulation of colonic 1,25-D 3 synthesis, which could be an important factor for colon tumor prevention ( 36 ), we fed C57BL/6 mice with low dietary calcium, a nutritional condition suggested to be a risk factor for human CRC. The calcium deficient diets increased proliferation of mouse colonocytes in both the ascending and descending colon, whereas expression of the cell cycle inhibitor p21, a protein known to play an important role in regulating intestinal cell proliferation, maturation and tumorigenesis ( 44 ), was decreased in the ascending colon only ( Figure 2 ).

Calcium deficient diets (even the one containing 0.1% calcium, a level comparable to that consumed by the population of Western industrialized countries) significantly affected the expression of renal hydroxylases: decreased CYP24 and increased CYP27B1 mRNA levels. Regulation of renal vitamin D hydroxylases is consistent with the action of PTH in response to lowered serum calcium. In the colon, however, CYP24 mRNA expression was either significantly induced (in the ascending colon) or not changed at all (in the descending colon). While the increase of colonic CYP24 expression could potentially be an effect of increased serum 1,25-D 3 levels in response to low calcium ingestion ( Table I ), this is extremely unlikely since the elevation of CYP24 is found in ascending colon only where it is paralleled with a decrease in p21 expression. While VDR levels are somewhat higher in the ascending than the descending colon, this difference is too small to be a limiting factor for a potential serum 1,25-D 3 effect on colonic CYP24 expression. In addition, there is a trend towards upregulation of colonic CYP27B1 by lowered calcium ingestion, whereas this is not significant ( Figure 4B ) or remotely comparable with the huge increase demonstrated in the kidney ( Figure 3B ).

In order to substantiate even further our claim that low fecal calcium levels alone, even in absence of 1,25-D 3, are able to enhance colonic CYP24 mRNA expression, we availed ourselves of an in vitro model. Caco-2 cells were cultured serum-free and 1,25-D 3 -free. Under such growth conditions they respond to zero calcium by greatly enhanced proliferation and a shift in cell cycle distribution from the G0 to the S phase ( 27 ). This enhancement of proliferation by a 2-fold increased expression of PCNA mRNA is demonstrated in Figure 5 and, importantly, CYP24 mRNA was elevated 3-fold in these hyperproliferative cells. Furthermore, we were able to reverse this increase in CYP24 mRNA expression by raising again the calcium concentration of the medium ( Figure 5B ), demonstrating that extracellular calcium by itself is able to modulate CYP24 mRNA expression.

We have shown previously ( 29 ) that lack of genomic activity of 1,25-D 3 results in hyperproliferation, mainly in the distal colon. In this communication we now demonstrate that low nutritional calcium increased proliferation in both the proximal and distal colon segments. Parallel to the resultant hyperproliferation, both mRNA and protein levels of CYP24, the catabolic vitamin D hydroxylase, were upregulated, although in the proximal colon only. It is reasonable to expect that higher levels of CYP24 protein might lead to enhanced degradation and thus to reduced mucosal accumulation of the antimitogenic prodifferentiating 1,25-D 3 . The decrease of p21 protein expression (a target gene of 1,25-D 3 ( 45 )) in the same colon segment could be an indirect indication. This might lead to increased tumor incidence, since p21 was described to modulate the formation of tumors independent of the mechanisms of tumor initiation ( 46 , 47 ).

The present investigation characterizes for the first time regulation of CYP27B1 and CYP24 in the colon and demonstrates that colonic and renal vitamin D hydroxylases indeed are regulated differently from each other. We have shown recently in a mouse model that phytoestrogens present in soy reduced the expression of the colonic catabolic vitamin D hydroxylase CYP24, without affecting renal CYP24 expression at all ( 36 , 48 ). This insight could open new avenues for colon cancer prevention.

This work was supported financially by grant No. 9335 from the Austrian National Bank to E.K., by a grant from the American Institute of Cancer Research, Washington, to H.S.C. and by the Medical Research Service of the Department of Veterans Affairs to H.J.A.

Conflict of Interest Statement : None declared.

References

1.
Garland,C.F., Comstock,G.W., Garland,F.C., Helsing,K.J., Shaw,E.K. and Gorham,E.D. (
1989
) Serum 25-hydroxyvitamin D and colon cancer: eight-year prospective study.
Lancet
  ,
2
,
1176
–1178.
2.
Grant,W.B. (
2003
) Ecologic studies of solar UV-B radiation and cancer mortality rates.
Recent Results Cancer Res.
  ,
164
,
371
–377.
3.
Tuohimaa,P., Tenkanen,L., Ahonen,M. et al . (
2004
) Both high and low levels of blood vitamin D are associated with a higher prostate cancer risk: a longitudinal, nested case-control study in the Nordic countries.
Int. J. Cancer
  ,
108
,
104
–108.
4.
Sawada,N., Sakaki,T., Ohta,M. and Inouye,K. (
2000
) Metabolism of vitamin D(3) by human CYP27A1.
Biochem. Biophys. Res. Commun.
  ,
273
,
977
–984.
5.
Cheng,J.B., Motola,D.L., Mangelsdorf,D.J. and Russell,D.W. (
2003
) De-orphanization of cytochrome P450 2R1: a microsomal vitamin D 25-hydroxylase.
J. Biol. Chem.
  ,
278
,
38084
–38093.
6.
Holick,M.F. (
2003
) Vitamin D: a millenium perspective.
J. Cell Biochem.
  ,
88
,
296
–307.
7.
Kudlacek,S., Schneider,B., Peterlik,M., Leb,G., Klaushofer,K., Weber,K., Woloszczuk,W. and Willvonseder,R. (
2003
) Assessment of vitamin D and calcium status in healthy adult Austrians.
Eur. J. Clin. Invest.
  ,
33
,
323
–331.
8.
DeLuca,H.F. and Schnoes,H.K. (
1976
) Metabolism and mechanism of action of vitamin D.
Annu Rev. Biochem.
  ,
45
,
631
–666.
9.
Tanaka,Y. and DeLuca,H.F. (
1981
) Measurement of chick kidney 25-hydroxyvitamin D3-1 alpha-hydroxylase without use of radioactive substrate.
Anal. Biochem.
  ,
110
,
102
–107.
10.
Henry,H.L. and Norman,A.W. (
1984
) Vitamin D: metabolism and biological actions.
Annu. Rev. Nutr.
  ,
4
,
493
–520.
11.
Murayama,A., Takeyama,K., Kitanaka,S., Kodera,Y., Kawaguchi,Y., Hosoya,T. and Kato,S. (
1999
) Positive and negative regulations of the renal 25-hydroxyvitamin D3 1alpha-hydroxylase gene by parathyroid hormone, calcitonin, and 1alpha,25(OH)2D3 in intact animals.
Endocrinology
  ,
140
,
2224
–2231.
12.
Zierold,C., Darwish,H.M. and DeLuca,H.F. (
1994
) Identification of a vitamin D-response element in the rat calcidiol (25-hydroxyvitamin D3) 24-hydroxylase gene.
Proc. Natl Acad. Sci. USA
  ,
91
,
900
–902.
13.
Ohyama,Y., Ozono,K., Uchida,M., Shinki,T., Kato,S., Suda,T., Yamamoto,O., Noshiro,M. and Kato,Y. (
1994
) Identification of a vitamin D-responsive element in the 5′-flanking region of the rat 25-hydroxyvitamin D3 24-hydroxylase gene.
J. Biol. Chem.
  ,
269
,
10545
–10550.
14.
Zierold,C., Mings,J.A. and DeLuca,H.F. (
2001
) Parathyroid hormone regulates 25-hydroxyvitamin D(3)-24-hydroxylase mRNA by altering its stability.
Proc. Natl Acad. Sci. USA
  ,
98
,
13572
–13576.
15.
Zierold,C., Mings,J.A. and DeLuca,H.F. (
2003
) Regulation of 25-hydroxyvitamin D3-24-hydroxylase mRNA by 1,25-dihydroxyvitamin D3 and parathyroid hormone.
J. Cell Biochem.
  ,
88
,
234
–237.
16.
Palmer,H.G., Gonzalez-Sancho,J.M., Espada,J., Berciano,M.T., Puig,I., Baulida,J., Quintanilla,M., Cano,A., de Herreros,A.G., Lafarga,M. and Munoz,A. (
2001
) Vitamin D(3) promotes the differentiation of colon carcinoma cells by the induction of E-cadherin and the inhibition of beta-catenin signaling.
J. Cell Biol.
  ,
154
,
369
–387.
17.
Shabahang,M., Buras,R.R., Davoodi,F., Schumaker,L.M., Nauta,R.J., Uskokovic,M.R., Brenner,R.V. and Evans,S.R. (
1994
) Growth inhibition of HT-29 human colon cancer cells by analogues of 1,25-dihydroxyvitamin D3.
Cancer Res.
  ,
54
,
4057
–4064.
18.
Cross,H.S., Farsoudi,K.H. and Peterlik,M. (
1993
) Growth inhibition of human colon adenocarcinoma-derived Caco-2 cells by 1,25-dihydroxyvitamin D3 and two synthetic analogs: relation to in vitro hypercalcemic potential.
Naunyn Schmiedebergs Arch. Pharmacol.
  ,
347
,
105
–110.
19.
Tangrea,J., Helzlsouer,K., Pietinen,P., Taylor,P., Hollis,B., Virtamo,J. and Albanes,D. (
1997
) Serum levels of vitamin D metabolites and the subsequent risk of colon and rectal cancer in Finnish men.
Cancer Causes Control
  ,
8
,
615
–625.
20.
Cross,H.S., Peterlik,M., Reddy,G.S. and Schuster,I. (
1997
) Vitamin D metabolism in human colon adenocarcinoma-derived Caco-2 cells: expression of 25-hydroxyvitamin D3-1alpha-hydroxylase activity and regulation of side-chain metabolism.
J. Steroid Biochem. Mol. Biol.
  ,
62
,
21
–28.
21.
Bareis,P., Bises,G., Bischof,M.G., Cross,H.S. and Peterlik,M. (
2001
) 25-Hydroxyvitamin D metabolism in human colon cancer cells during tumor progression.
Biochem. Biophys. Res. Commun.
  ,
285
,
1012
–1017.
22.
Bareis,P., Kallay,E., Bischof,M.G., Bises,G., Hofer,H., Potzi,C., Manhardt,T., Bland,R. and Cross,H.S. (
2002
) Clonal differences in expression of 25-hydroxyvitamin D(3)-1alpha-hydroxylase, of 25-hydroxyvitamin D(3)-24-hydroxylase, and of the vitamin D receptor in human colon carcinoma cells: effects of epidermal growth factor and 1alpha,25-dihydroxyvitamin D(3).
Exp. Cell Res.
  ,
276
,
320
–327.
23.
Cross,H.S., Bareis,P., Hofer,H., Bischof,M.G., Bajna,E., Kriwanek,S., Bonner,E. and Peterlik,M. (
2001
) 25-Hydroxyvitamin D(3)-1alpha-hydroxylase and vitamin D receptor gene expression in human colonic mucosa is elevated during early cancerogenesis.
Steroids
  ,
66
,
287
–292.
24.
Bises,G., Kallay,E., Weiland,T., Wrba,F., Wenzl,E., Bonner,E., Kriwanek,S., Obrist,P. and Cross,H.S. (
2004
) 25-Hydroxyvitamin D3-1alpha-hydroxylase expression in normal and malignant human colon.
J. Histochem. Cytochem.
  ,
52
,
985
–989.
25.
Garland,C.F., Garland,F.C. and Gorham,E.D. (
1991
) Can colon cancer incidence and death rates be reduced with calcium and vitamin D?
Am. J. Clin. Nutr.
  ,
54
,
193S
–201S.
26.
Cho,E., Smith-Warner,S.A., Spiegelman,D. et al . (
2004
) Dairy foods, calcium, and colorectal cancer: a pooled analysis of 10 cohort studies.
J. Natl Cancer Inst.
  ,
96
,
1015
–22.
27.
Kállay,E., Bajna,E., Wrba,F., Kriwanek,S., Peterlik,M. and Cross,H.S. (
2000
) Dietary calcium and growth modulation of human colon cancer cells: role of the extracellular calcium-sensing receptor.
Cancer Detect. Prev.
  ,
24
,
127
–136.
28.
Yoshizawa,T., Handa,Y., Uematsu,Y. et al . (
1997
) Mice lacking the vitamin D receptor exhibit impaired bone formation, uterine hypoplasia and growth retardation after weaning.
Nat. Genet.
  ,
16
,
391
–396.
29.
Kallay,E., Pietschmann,P., Toyokuni,S., Bajna,E., Hahn,P., Mazzucco,K., Bieglmayer,C., Kato,S. and Cross,H.S. (
2001
) Characterization of a vitamin D receptor knockout mouse as a model of colorectal hyperproliferation and DNA damage.
Carcinogenesis
  ,
22
,
1429
–1435.
30.
Li,Y.C., Pirro,A.E., Amling,M., Delling,G., Baron,R., Bronson,R. and Demay,M.B. (
1997
) Targeted ablation of the vitamin D receptor: an animal model of vitamin D-dependent rickets type II with alopecia.
Proc. Natl Acad. Sci. USA
  ,
94
,
9831
–9835.
31.
Li,Y.C., Amling,M., Pirro,A.E., Priemel,M., Meuse,J., Baron,R., Delling,G. and Demay,M.B. (
1998
) Normalization of mineral ion homeostasis by dietary means prevents hyperparathyroidism, rickets, and osteomalacia, but not alopecia in vitamin D receptor-ablated mice.
Endocrinology
  ,
139
,
4391
–4396.
32.
Amling,M., Priemel,M., Holzmann,T., Chapin,K., Rueger,J.M., Baron,R. and Demay,M.B. (
1999
) Rescue of the skeletal phenotype of vitamin D receptor-ablated mice in the setting of normal mineral ion homeostasis: formal histomorphometric and biomechanical analyses.
Endocrinology
  ,
140
,
4982
–4987.
33.
Newmark,H.L., Lipkin,M. and Maheshwari,N. (
1991
) Colonic hyperproliferation induced in rats and mice by nutritional-stress diets containing four components of a human Western-style diet (series 2).
Am. J. Clin. Nutr.
  ,
54
,
209S
–214S.
34.
Kallay,E., Kifor,O., Chattopadhyay,N., Brown,E.M., Bischof,M.G., Peterlik,M. and Cross,H.S. (
1997
) Calcium-dependent c-myc proto-oncogene expression and proliferation of Caco-2 cells: a role for a luminal extracellular calcium-sensing receptor.
Biochem. Biophys. Res. Commun.
  ,
232
,
80
–83.
35.
Garland,C.F., Garland,F.C. and Gorham,E.D. (
1999
) Calcium and vitamin D. Their potential roles in colon and breast cancer prevention.
Ann. NY Acad. Sci.
  ,
889
,
107
–119.
36.
Kallay,E., Adlercreutz,H., Farhan,H., Lechner,D., Bajna,E., Gerdenitsch,W., Campbell,M. and Cross,H.S. (
2002
) Phytoestrogens regulate vitamin D metabolism in the mouse colon: relevance for colon tumor prevention and therapy.
J. Nutr.
  ,
132
,
3490S
–3493S.
37.
Feskanich,D., Ma,J., Fuchs,C.S., Kirkner,G.J., Hankinson,S.E., Hollis,B.W. and Giovannucci,E.L. (
2004
) Plasma vitamin D metabolites and risk of colorectal cancer in women.
Cancer Epidemiol Biomarkers Prev.
  ,
13
,
1502
–1508.
38.
Grau,M.V., Baron,J.A., Sandler,R.S., Haile,R.W., Beach,M.L., Church,T.R. and Heber,D. (
2003
) Vitamin D, calcium supplementation, and colorectal adenomas: results of a randomized trial.
J. Natl Cancer Inst.
  ,
95
,
1765
–1771.
39.
Lemay,J., Demers,C., Hendy,G.N., Delvin,E.E. and Gascon-Barre,M. (
1995
) Expression of the 1,25-dihydroxyvitamin D3-24-hydroxylase gene in rat intestine: response to calcium, vitamin D3 and calcitriol administration in vivo .
J. Bone Miner Res.
  ,
10
,
1148
–1157.
40.
Demers,C., Lemay,J., Hendy,G.N. and Gascon-Barre,M. (
1997
) Comparative in vivo expression of the calcitriol-24-hydroxylase gene in kidney and intestine.
J. Mol. Endocrinol.
  ,
18
,
37
–48.
41.
Armbrecht,H.J., Boltz,M.A., Hodam,T.L. and Kumar,V.B. (
2001
) Differential responsiveness of intestinal epithelial cells to 1,25-dihydroxyvitamin D3—role of protein kinase C.
J. Endocrinol.
  ,
169
,
145
–151.
42.
Shinki,T., Jin,C.H., Nishimura,A., Nagai,Y., Ohyama,Y., Noshiro,M., Okuda,K. and Suda,T. (
1992
) Parathyroid hormone inhibits 25-hydroxyvitamin D3-24-hydroxylase mRNA expression stimulated by 1alpha,25-dihydroxyvitamin D3 in rat kidney but not in intestine.
J. Biol. Chem.
  ,
267
,
13757
–13762.
43.
Neugut,A.I., Jacobson,J.S., Suh,S., Mukherjee,R. and Arber,N. (
1998
) The epidemiology of cancer of the small bowel.
Cancer Epidemiol. Biomarkers Prev.
  ,
7
,
243
–251.
44.
Gartel,A.L. and Tyner,A.L. (
1999
) Transcriptional regulation of the p21 ((WAF1/CIP1)) gene.
Exp. Cell Res.
  ,
246
,
280
–289.
45.
Jensen,S.S., Madsen,M.W., Lukas,J., Binderup,L. and Bartek,J. (
2001
) Inhibitory effects of 1alpha,25-dihydroxyvitamin D(3) on the G(1)-S phase-controlling machinery.
Mol. Endocrinol.
  ,
15
,
1370
–1380.
46.
Yang,W.C., Mathew,J., Velcich,A., Edelmann,W., Kucherlapati,R., Lipkin,M., Yang,K. and Augenlicht,L.H. (
2001
) Targeted inactivation of the p21 (WAF1/cip1) gene enhances Apc-initiated tumor formation and the tumor-promoting activity of a Western-style high-risk diet by altering cell maturation in the intestinal mucosa.
Cancer Res.
  ,
61
,
565
–569.
47.
Yang,W., Velcich,A., Lozonschi,I., Liang,J., Nicholas,C., Zhuang,M., Bancroft,L. and Augenlicht,L.H. (
2005
) Inactivation of p21WAF1/cip1 enhances intestinal tumor formation in Muc2 −/− mice.
Am. J. Pathol.
  ,
166
,
1239
–1246.
48.
Cross,H.S., Kallay,E., Lechner,D., Gerdenitsch,W., Adlercreutz,H. and Armbrecht,H.J. (
2004
) Phytoestrogens and vitamin D metabolism: a new concept for the prevention and therapy of colorectal, prostate, and mammary carcinomas.
J. Nutr.
  ,
134
,
1207S
–1212S.

Author notes

Department of Pathophysiology, 1Department of Medical and Chemical Laboratory Diagnostics and 2Center for Animal Care, Medical University of Vienna, A-1090 Vienna, Währinger Guertel 18-20, Austria, 3Institute of Analytical Chemistry, University of Vienna, Währinger Strasse 38, A-1090 Vienna, Austria, 4Department of Pathology and Biology of Diseases, Graduate School of Medicine, Kyoto University, Japan and 5Geriatric Research, Education, and Clinical Center, St Louis Veterans Administration Medical Center, St Louis, MO 63125, USA