AJCN EB Program 2010
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M
Right arrow Search for Related Content
PubMed
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M
Agricola
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M
American Journal of Clinical Nutrition, Vol. 87, No. 5, 1513-1520, May 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Vitamin K1 intake is associated with higher bone mineral density and reduced bone resorption in early postmenopausal Scottish women: no evidence of gene-nutrient interaction with apolipoprotein E polymorphisms1,2,3,4

Helen M Macdonald, Fiona E McGuigan, Susan A Lanham-New, William D Fraser, Stuart H Ralston and David M Reid

1 From the Department of Medicine and Therapeutics, University of Aberdeen, Foresterhill, Aberdeen, United Kingdom (HMM, FEM, and DMR); the Centre for Nutrition and Food Safety, School of Biomedical and Molecular Sciences, University of Surrey, Guildford, Surrey, United Kingdom (SAL-N); the Department of Clinical Biochemistry, Royal Liverpool University Hospital, Liverpool, United Kingdom (WDF); and the Rheumatic Diseases Unit, Molecular Medicine Centre, University of Edinburgh, Western General Hospital, Edinburgh, United Kingdom (SHR)

2 Any views expressed are those of the authors.

3 Supported by the UK Food Standards Agency (N 05043) and the Arthritis Research Campaign for continuing infrastructure support (DMR).

4 Reprints not available. Address correspondence to HM Macdonald, Osteoporosis Research Unit, Health Sciences Building, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, United Kingdom. E-mail: h.macdonald{at}abdn.ac.uk.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: Polymorphisms in the apolipoprotein E (APOE) gene are associated with fracture risk, and a potential mechanism is through vitamin K transport.

Objective: We investigated the relation between dietary vitamin K1 intake, APOE polymorphisms, and markers of bone health.

Design: We measured bone mineral density (BMD) at the lumbar spine (LS) and femoral neck (FN) in a cohort of Scottish women aged 49–54 y in 1990–1994 (baseline) and in 1997–2000 (visit 2). At visit 2, bone markers (urinary pyridinoline crosslinks and serum N-terminal propeptide of type 1 collagen) were measured, 3199 women completed a food-frequency questionnaire, and 2721 women were genotyped for APOE.

Results: Compared with quartile 3 (Q3) of energy-adjusted vitamin K1 intake (x: 116 µg/d), women in the lowest quartile (x: 59 µg/d) had lower BMD (analysis of variance; FN, Q1: 0.831 ± 0.122 g/cm2; Q3: 0.850 ± 0.126 g/cm2; P < 0.001; LS, Q1: 1.000 ± 0.170 g/cm2; Q3: 1.020 ± 0.172 g/cm2; P = 0.009), remaining significant at the FN after adjustment for age, weight, height, menopausal status or use of hormone replacement therapy, socioeconomic status, and physical activity (P = 0.04). Vitamin K1 intake was associated with reduced concentrations of pyridinoline crosslinks (Q1: 5.4 ± 2.0 nmol/mmol; Q4: 5.1 ± 1.9 nmol/mmol; P = 0.003). Carriers of the E2 allele had greater LS BMD at visit 2 and lost less BMD than did carriers of the E4 allele (E2: –0.50 ± 1.22%/y; E4: –0.71 ± 1.17%/y; P = 0.05). After adjustment for confounders, the P value for BMD loss (0.03 for LS and 0.04 for FN) did not reach the level of significance required for multiple testing (P = 0.012). No interaction was observed between dietary vitamin K and APOE on BMD.

Conclusions: Vitamin K1 intake was associated with markers of bone health, but no interaction was observed with APOE alleles on BMD or markers of bone turnover.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The apolipoprotein E (APOE) gene has 3 alleles: E2, E3, and E4, defined by 2 common polymorphisms that affect the protein coding sequence. The E4 allele was associated with increased risk of heart disease, strokes, and diabetes (1), with Alzheimer disease, and with greater fracture risk in a number of populations (2), although the association was not consistent across studies (3, 4). One explanation for the increased fracture risk is that APOE is involved in vitamin K transport in the blood and that less vitamin K is available in carriers of the E4 allele than with carriers of the other alleles because of lower plasma ApoE concentrations. Vitamin K is required for carboxylating osteocalcin. Concentrations of undercarboxylated osteocalcin were associated with increased risk of hip fracture (5, 6), and lower concentrations of circulating vitamin K were found in women with osteoporotic fractures (7, 8). Short-term supplementation with vitamin K corrected the undercarboxylation of osteocalcin in women with Colles fractures (9). The anticoagulant drug warfarin competitively inhibits vitamin K, but studies of warfarin on bone mineral density (BMD) have produced conflicting results, and there is no evidence that warfarin increases fracture rates in women (10). Currently, there are no reference nutrient intakes for vitamin K in the United Kingdom, but a daily intake of 1 µg/kg body weight is both safe and adequate (11). The term vitamin K encompasses phylloquinone (vitamin K1) and menaquinone (vitamin K2). Menaquinone is produced by bacteria in the bowel and can originate from fermented foods (12). Its contribution to a Western diet is low. A recent meta-analysis of vitamin K and fracture included only vitamin K2 and at pharmacologic values (generally 45 000 µg/d) (13). Vitamin K1 is the major circulating form and the dominant form found in bone (14). A major source of vitamin K1 is green leafy vegetables (a 100-g portion provides 200 µg vitamin K1) and some vegetable oils. Fruit and vegetable intake was consistently associated with improved bone health status in populations from the United States and the United Kingdom (15-17). Although a plausible hypothesis receiving much renewed attention is that fruit and vegetables protect against excessive bone resorption by providing alkaline metabolites to balance the acidity caused by eating a diet rich in protein (18), an alternative explanation is that fruit and vegetables are beneficial because of the nutrients they provide such as vitamin K1.

Nutrient interactions were reported between alcohol intake and APOE genotype on LDL concentrations (19) and between polyunsaturated fatty acid intake and APOA1 genotype on HDL cholesterol (20). Yan et al (21) showed that higher concentrations of vitamin K1 were associated with E4 genotype in older persons in China and the United Kingdom. However, to the best of our knowledge, no one has specifically tested for an interaction between APOE genotype and dietary vitamin K in relation to bone health. The aim of this study was to determine whether APOE polymorphisms and dietary intake of vitamin K1 are associated with markers of bone health in Scottish early postmenopausal women and whether there is an interaction between the vitamin K1 intake and APOE genotype on early postmenopausal bone loss.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
The subjects for this study were taken from a population-based screening program for osteoporotic fracture risk, involving 5119 women aged 45–54 y, that took place between 1990 and 1994 with further assessment for 3883 of the women between 1997 and 2000 (22). Blood and urine samples were collected at the second visit only. A subset of 898 women completed a food-frequency questionnaire (FFQ) at both visits, and an additional 2340 women completed the FFQ shortly after the follow-up visit so that there were 3238 completed FFQs in total at visit 2. In 16 subjects, data on weight and hip or spine BMD were missing, and 23 subjects were excluded from the analysis because of extreme daily energy intakes < 3.2 MJ and > 18 MJ. In total 2731 women were genotyped for APOE polymorphism (of whom 10 had missing weight, hip, or spine data), and 2471 women had information on both diet and APOE genotype for study of APOE-nutrient interaction. At the time of the second visit when women were asked about their current medication, 6 women reported being on anticoagulant therapy.

The women were weighed on both occasions wearing light clothing and no shoes on a set of balance scales (Seca, Hamburg, Germany) calibrated to 0.05 kg. Height and sitting height were measured with a stadiometer (Holtain Ltd, Crymych, United Kingdom). Information was collected on general health (whether the subject had a disease), smoking (past, current, and never and how many cigarettes were smoked over how many years), menopausal status (regularity of menstrual periods or date of last menstrual period), and use of hormone replacement therapy (HRT; times started and stopped therapy) and other medication (what medication was currently being taken). Written informed consent was obtained for all the women, and the study was approved by Grampian Research Ethics Committee.

Genotyping
Genomic DNA was amplified by the polymerase chain reaction (PCR) with the use of the Qiagen Taq polymerase system and primers APOE forward (AGA CGC GGG CAC GGC TGT) and APOE reverse (CTC GCG GAT GGC GCT GAG). The reaction mixture consisted of 2.5 µL 10x Q buffer, 2 µL deoxyribonucleotide triphosphates at 2.5 mmol/L each, 5 µL of Q solution, 0.5 µL of each forward and reverse primer (10 µmol/L), 0.125 µL Q Taq (5 uµl–1) and 2.5 µL genomic DNA 20 ng µL–1 made up to 25 µL with 11.875 µL sterile water. The samples were thermocycled on a DYAD DNA Engine (MJ Research Inc, Waltham, MA) as follows: denaturation at 95 °C for 5 min followed by 50 cycles of denaturation at 95 °C for 30 s, annealing at 62 °C for 30 s, and extension at 72 °C for 30 s with a final extension cycle of 72 °C for 5 min. The PCR product (6 µL) was cleaned to remove any unused deoxyribonucleotide triphosphates and primers that would interfere in the sequencing reaction with the use of ExoSAP-IT (USB Corp, Cleveland, OH), which contains exonuclease 1 and shrimp alkaline phosphatase. The samples were labeled for sequencing by adding 2 µL of DYEnamicET terminator reagent premix (Amersham Biosciences, Amersham, United Kingdom) and 1 µL of the forward primer (2.5 mmol/L) to each aliquot of 7 µL of purified PCR product and then cycled for 31 cycles by the following steps: 95 °C for 20 s, 60 °C for 15 s, and 65 °C for 1 min. The labeled PCR products were ethanol-precipitated. The sequencing reactions were analyzed on the Amersham Biosciences MegaBACE 1000 DNA Analysis System with the use of downstream base caller (Cimarron 3.12), and the base-called sequences were viewed with the use of the SEQ MAN II expert sequence analysis software (version 5.07; DNASTAR Inc, Madison, WI). The single nucleotide polymorphisms (SNPs) were called individually from the sequence and analyzed according to the genotype of the SNPs situated in the codons at amino acid positions 112 and 158. Both SNPs had C/T base changes with TGC coding for cysteine and CGC coding for arginine. For E3 there is a cysteine in position 112 and arginine in position 158; E2 has a cysteine in both positions; and E4 has an arginine in both positions.

Markers of bone health
BMD of the left femoral neck (FN) and lumbar spine (LS; L2–L4) was measured by dual-energy X-ray absorptiometry. At baseline all women were measured on the same Norland scanner (Cooper Surgical Inc, Trumbull, CT). At follow-up, most of the women were scanned again with the use of a Norland XR26; however, 12% of women had measurements on the XR36 model. Because the XR36 machine gave slightly higher readings (1.258%), a correction factor was applied to bring these measurements in line with the XR26 measurements. A second, early morning fasting urine sample, collected 2 h after first void, was provided at the second visit only for analysis of free pyridinoline (fPYD) crosslinks and free deoxypyridinoline (fDPD) crosslinks by HPLC (23, 24). Creatinine was measured in urine by standard automated techniques (Roche, Lewes, United Kingdom), and results were expressed as fPYD/creatinine and fDPD/creatinine (in nmol/mmol). Serum N-terminal propeptide of type 1 collagen was measured with the use of an enzyme chemiluminescent immunoassay supplied by Roche Products Ltd (Penzberg, Germany) (25).

Diet and physical activity
Dietary assessment was made at the follow-up visit with the use of an FFQ that was validated with the use of 7-d weighed intakes [correlation coefficients (r) ranging from 0.31 for vitamin D, 0.40 for iron, 0.66 for phosphorus, 0.52 for calcium, 0.78 for potassium, 0.61 for protein, 0.81 for vitamin C, and 0.82 for fiber, with 45–90% correctly classified to within 1 quintile] (26) and serum concentrations of antioxidants (r = 0.14 for vitamin A and 0.42 for vitamin C, with 68–89% of subjects correctly classified within 1 tertile) (27). A subgroup of women (n = 898) also completed the same FFQ at baseline. For most women there was little change in nutrient intake, although mean (±SD) energy intake had decreased from 8.1 ± 1.2 MJ/d to 7.7 ± 1.1 MJ/d (28). Vitamin K1 concentrations were obtained by analyzing the FFQ with the use of a UK database, compiled by Bolton-Smith et al (29), which specifically detailed the vitamin K1 content of foods listed in our FFQ.

Physical activity levels (PALs) were obtained with the use of the same questions as used for the Scottish Heart Health Study (30). The PAL was calculated from the numbers of hours in a 24-h period doing heavy, moderate, or light activities and how many hours were spent sleeping or resting in bed. These questions were asked separately for working and nonworking days. PAL is normally defined as the ratio of energy expenditure divided by the basal metabolic rate, which is calculated from Schofield equations (31).

Statistical analysis
Analyses were performed with the use of SPSS version 14.0 (SPSS Inc, Chicago, IL). Dietary vitamin K1 intakes were adjusted for dietary energy intake by the residual method (32). Categorization of nutrient intake was carried out because this is a particularly appropriate method for analysis of FFQ data. For comparison of the 3 APOE alleles, the approach used by Corella et al (19) was adopted, which categorized the 6 possible genotypes for APOE into 3 groups as follows: carriers of E2 (E2/E2 and E2/E3), carriers of E4 (E4/E4 and E3/E4), and the E3/E3 genotype, omitting the E2/E4 genotype from the analysis (because this genotype could belong to either the E2 or the E4 carrier group). Because a number of independent tests were performed, this raises the possibility of false-positive associations. We took into account the correlation of the BMD sites (r = 0.66), and 6 tests resolve into 4.15 independent tests, so that a significance level of 0.012 would be equivalent to P = 0.05.

One-factor analysis of variance (ANOVA; with Scheffe post hoc test) and analysis of covariance (adjusting for confounding variables) were used to examine differences in characteristics and bone health indexes between different genotypes and quartiles of energy-adjusted dietary intake of vitamin K1. Chi-squared tests were used for categorical variables. Stepwise linear multiple regression analysis was used to determine independent predictors, including energy-adjusted dietary vitamin K1 as a continuous variable, of bone resorption markers and BMD. Dummy variables for menopausal status (premenopausal, perimenopausal, postmenopausal not using HRT, and postmenopausal with past use of HRT) were used to account for menopausal status or HRT use (with current HRT use as the reference). Interaction between genotype and vitamin K1 was tested by including the following variables in the general linear model: genotype (as carriers of the E2 allele compared with noncarriers of the E2 allele), energy-adjusted intake of vitamin K1, and an additional multiplicative term of genotype with energy-adjusted intake of vitamin K1. Estimated power was calculated with the use of the standardized β values obtained from the linear regression between the nutrient and BMD according to Luan et al (33).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
APOE genotype
Genotypes were in Hardy Weinberg equilibrium (SNP1 at position 112 X2 = 0.311; P = 0.578; SNP2 at position 158 X2 = 0.90; P = 0.343) with frequencies in descending order as follows: E3/E3 (n = 1664), E3/E4 (n = 619), E2/E3 (n = 310), E2/E4 (n = 62), E4/E4 (n = 50), and E2/E2 (n = 16). The allele frequencies were similar to those observed in other North European populations (3): E3 was 78.2%, E4 was 14%, and E2 was 7.8%.

Subject characteristics
The women who were genotyped for ApoE (n = 2731) were not significantly different from women who were not genotyped (n = 1150) in terms of BMD measurements and dietary energy intake, but they were marginally younger (x ± SD: 54.7 ± 2.2 y and 55.1 ± 2.3 y, respectively) and lighter (68.4 ± 12.7 kg and 69.9 ± 13.4 kg, respectively). No differences were observed in mean age, weight, height, and type of scanner used between carriers of the E2 allele, the E3/E3 genotype, and carriers of the E4 allele (Table 1Go). The characteristics of subjects who had one copy each of the E2 and E4 alleles (n = 62) and who were excluded from further analysis were not significantly different from the other genotypes (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Characteristics and measures of bone health according to apolipoprotein E (APOE) genotype1

 
APOE associations with markers of bone health
BMD at the second visit was greater for carriers of the E2 allele (LS: 1.03 g/cm2; FN: 0.85 g/cm2) than for carriers of the E4 allele (LS: 1.00 g/cm2, FN: 0.83 g/cm2), and there was less bone loss for carriers of the E2 allele (LS: –0.50%/y compared with –0.71%/y; FN: –0.63%/y compared with –0.79%/y) (Table 1Go). This was significant at the LS (ANOVA; P = 0.04 for visit 2; P = 0.05 for change in BMD) but not the FN (ANOVA; P = 0.11 for visit 2; P = 0.10 change in BMD). After adjustment for weight, height, smoking, socioeconomic status, menopausal status or HRT use, physical activity, and baseline BMD, the relation between APOE genotype and BMD was significant both at the spine (P = 0.03) and the hip (P = 0.04). No relation was observed between baseline BMD and APOE genotype. For the markers of bone turnover, the E3/E3 genotype was found to have significantly higher fDPD/creatinine (5.4 nmol/mmol) and fPYD/creatinine (19.5 nmol/mL) than carriers of the E2 allele (fDPD/creatinine: 5.0 nmol/mmol; fPYD/creatinine: 18.7 nmol/mmol) and carriers of the E4 allele (fDPD/creatinine: 5.2 nmol/mmol; fPYD/creatinine: 18.7 nmol/mmol), but this relation was not significant after adjustment for confounders for fPYD/creatinine (Table 1Go). A trend was observed for lower concentrations of propeptide of type 1 collagen in the carriers of the E2 allele (33.6 µg/L) than in carriers of the E4 allele (35.7 µg/L) and the E3/E3 genotype (36.4 µg/L) (P = 0.07), but the relation was not significant after adjustment for confounders (P = 0.17).

Vitamin K
For the subset of women that completed the FFQ on both occasions (n = 898), no change was observed in mean (±SD) vitamin K1 intakes between baseline (100 ± 39µg/d) and visit 2 (103 ± 44 µg/d; P = 0.31). The vitamin K1 intake of the women who completed the FFQ at the second visit only was significantly greater at 109 ± 55 µg/d (P = 0.01). This may be explained by the former group being younger (mean age at baseline: 47.6 ± 1.5 y compared with 48.8 ± 2.51 y; P < 0.01) because the women were selected because they were premenopausal and also had no medical condition and were not taking any medication known to affect bone metabolism. When the groups were combined, mean daily vitamin K1 intake (at visit 2) was 107 ± 50 µg, ranging from 8 µg to 494 µg.

Vitamin K1 and markers of health
A significant association was observed between energy-adjusted intake of vitamin K1 and both BMD and markers of bone resportion (Table 2Go). BMD was found to be lower in the lowest quartile of energy-adjusted intake of vitamin K1 (daily vitamin K1 intake: 64 ± 14 µg, energy-adjusted intake; 59 ± 17 µg, actual) than in the upper quartiles of vitamin K1 intake [daily vitamin K1 intake, quartile 3 (Q3): 124 ± 9 µg energy-adjusted, 116 ± 19 µg actual; Q4: 181 ± 55 µg energy-adjusted, 162 ± 57 µg actual] (ANOVA; second visit: P < 0.01 for FN BMD; P = 0.01 for LS BMD). Scheffe post hoc analysis showed significant differences between Q1 (0.83 ± 0.12 g/cm2) and Q3 (0.85 ± 0.13 g/cm2) (P = 0.01) and between Q2 (0.83 ± 0.12 g/cm2) and Q3 (P < 0.01) for FN and between Q2 (1.00 ± 0.16 g/cm2) and Q3 (1.02 ± 0.17 g/cm2) (P = 0.04) for LS. The same findings were observed if women who were scanned on the XR36 scanner were excluded or if women on warfarin treatment were excluded. The results were significant after adjustment for the following confounders: age, weight, height, smoking, socioeconomic status, menopausal status and use of HRT, and PAL at the hip but not at the spine (analysis of covariance; P = 0.04 for FN BMD; P = 0.38 for LS BMD) (Table 2Go). With linear regression analysis, adjusting for confounders, energy-adjusted intake of vitamin K1 as a continuous variable was not found to be a predictor of either LS or FN BMD. Weight accounted for most of the variation of BMD (13.5%) followed by menopausal status (1.4%). For LS BMD, again, weight was the major predictor (11.3%) with menopausal status accounting for a further 3.7% variation. In total, the variables we tested accounted for 17–18% of the total variation of BMD.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Differences in bone markers according to quartile (Q) of energy-adjusted intake of vitamin K1 at visit 21

 
Bone resorption markers were significantly higher in the lowest quartile of energy-adjusted intake of vitamin K1 (fDPD/creatinine: 5.4 ± 2.0 nmol/mmol; fPYD/creatinine: 19.8 ± 6.8 nmol/mmol) than in the highest quartile (fDPD/creatinine: 5.1 ± 1.9 nmol/mmol; fPYD/creatinine: 18.7 ± 6.7 nmol/mmol) (Scheffe post hoc analysis; P = 0.002 for fDPD/creatinine and P = 0.001 for fPYD/creatinine) (Table 2Go), and the difference remained significant after adjustment for confounders. Linear regression showed that energy-adjusted intake of vitamin K1 predicted 2.7% of the variation of fDPD/creatinine and 2.1% of the variation in fPYD/creatinine. The other predictors for fDPD/creatinine were menopausal status or HRT use (9.9%), weight (2.7%), height (1.4%), and smoking (0.3%), and predictors for fPYD/creatinine were menopausal status or HRT use (12.0%), weight (2.1%), height (1.1%), and smoking (0.3%).

Interaction between APOE and vitamin K1
No interaction was found between APOE genotype and dietary intake of vitamin K1 in regulating BMD or bone loss (BMD change: P = 0.76 for LS; P = 0.16 for FN). Postanalysis power was determined after excluding E2/E4 genotypes by comparing standardized β values between FN BMD change and energy-adjusted intake of vitamin K1 for carriers of the E2 allele and noncarriers of the E2 allele. For carriers of the E2 allele, the relation between FN BMD change and energy-adjusted intake of vitamin K1 was weakly positive (standardized β: 0.065), whereas for noncarriers of the E2 allele the relation was weakly negative (standardized β: –0.023). Power was calculated as 31.9% at the 5% significance level.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In previous studies the apolipoprotein E4 allele was associated with increased fracture risk in older men and women (34) and patients receiving hemodialysis (35) and was associated with lower BMD in men only but not women (36). In this population of early postmenopausal women, carriers of the E2 allele of the APOE polymorphism had the highest BMD at the second visit and carriers of the E4 allele had the lowest BMD, which was a result of less bone loss occurring in the E2 group than in the E4 carriers. No association was observed between APOE genotype and BMD at the baseline visit, indicating that any influence of the APOE genotype was not on peak bone mass. After adjustment for confounders, the level of significance of the tests with BMD did not reach the lower threshold required by multiple independent testing. It is possible that nutrient interaction may mask a clear-cut association.

However, our data suggest that vitamin K1 associations on bone health operate by a different mechanism from the trends seen with APOE. The association was between vitamin K1 and BMD (visit 2 and baseline) and bone markers and was not with a change in BMD. A 3-y vitamin K intervention study in healthy postmenopausal women saw a benefit at the FN only and no influence at the LS, which is consistent with our data (37). In a depletion-repletion study phylloquinone (vitamin K1) did not produce a marked effect on bone metabolism, although 450 µg (5–7 times adequate intake) resulted in a decrease in the bone resorption marker urine N-terminal telopeptide (38). We observed a significant difference in deoxypyridinoline markers between women with a low mean daily intake of vitamin K1 (59 µg) and women with a high mean intake (162 µg). For BMD, women with a low mean intake of vitamin K1 (59 µg/d) had a lower BMD than did women with a higher mean intake (116 µg/d), but BMD did not increase further for higher concentrations of vitamin K1 (mean intake: 162 µg/d). This, together with the observation that vitamin K1 as a continuous variable was not found to be a significant predictor of BMD, may suggest a threshold effect for this nutrient. A recent Danish study found no association between vitamin K1 intake and BMD or fracture risk in 1139 perimenopausal and postmenopausal women in whom mean intakes were estimated to be 60 µg/d (39). Although the scale of our findings was small, as is often the case for nutritional factors (2–3% BMD difference between Q1 and Q3 of vitamin K), it was 15% of the SD in BMD for this population. It is known that a 0.5-SD difference in BMD, if maintained into old age, could reflect a 30% reduction in fractures. Similarly, vitamin K intake accounted for 2–3% of the total variation in bone resorption markers, and there was a 5% difference in BMD between Q1 and Q4. We consider that the size of the effect, although modest, was biologically meaningful.

A study examining both British and Chinese populations found that the E4 allele was associated with higher plasma concentrations of vitamin K1 (and also lower concentrations of undercarboxylated osteocalcin in the Chinese population) compared with E2/3 or E3/3 genotypes (21). The mean daily intake of vitamin K1 in their UK population (103 µg) was similar to the mean intake in our study (107 µg), whereas vitamin K1 intake in their Chinese population was more than twice as much (247 µg/d). If, as has been hypothesized, the ApoE gene influences transport of vitamin K1, their study would indicate that E4 carriers had the advantage for bone health. This is the opposite of our own findings, whereby we found the E2 allele to be associated with reduced bone loss and again suggests that the influences of APOE and vitamin K on bone health are likely to be independent. The sample size in the study by Yan et al (21) was small (n = 132 in the United Kingdom; n = 177 in China) and included men and women, and it is possible that there may be sex differences in how APOE affects circulating vitamin K1.

We found no significant interaction between the ApoE genotype and vitamin K1 on BMD, although it is recognized that this analysis may be limited in terms of power. The observation that vitamin K1 and the APOE genotype affect different phenotypes (predominantly affecting different bone sites) and the trend observed with the APOE genotype was with bone loss, whereas for vitamin K1 it was with BMD and bone markers, suggest that the mechanisms behind the associations with markers of bone health are different for the APOE genotype and vitamin K1 intake. If nutrient interaction has masked the association of APOE with bone health, our data would not support that it is through vitamin K transport. Alternative hypotheses that could be explored include ApoE effects mediated through transport of fats or fat-soluble vitamins other than vitamin K such as vitamin A or vitamin D.

As with all observational studies that investigate diet, there will be some correlation between nutrients, and it is possible that vitamin K1 is simply a marker of green vegetable or overall nutritional intake. We previously reported an association with net endogenous acid production and BMD in premenopausal but not in early postmenopausal women (22). Prynne et al (40) found that fruit and vegetable intake was associated with the size-adjusted bone mineral content in postmenopausal women and concluded that it was not due to the acidity generated by the diet. We would concur that further research is required to determine which component in fruit and vegetables is responsible for better bone health.

A limitation of our study is that, because of its large size (essential for determining genetic effects on BMD), volunteers were seen throughout the day, and it was not possible to collect blood samples at a fixed time of day. The timing of sampling is crucial for measurements that are dependent on circadian rhythm, and, consequently, we have no measurements of circulating vitamin K or estimates of the ratio of carboxylated to undercarboxylated osteocalcin. A recent study of postmenopausal women in Ireland found that vitamin K1 was inversely related to undercarboxylated osteocalcin (41). That study found that the mean daily intake from food was 107 µg (with 44% coming from broccoli, cabbage, and lettuce), which is comparable to the mean intake of vitamin K1 of the women in our study.

A further limitation of our study is that we have only estimated vitamin K1. Evidence suggests that hydrogenation of vegetable oils produces a form of vitamin K that is not as well absorbed and is less active, which could influence the estimate of vitamin K obtained from spreadable fats (42). We do not have estimates of vitamin K2, although intakes of vitamin K2 are likely to be low in our population. In populations in which intakes of vitamin K2 are higher, through greater consumption of fermented foods (eg, in Japanese diets), these intakes still fall short of the amounts used in the vitamin K intervention studies that are published (up to several thousand-fold of what would be found in a normal diet) (13).

In conclusion, in this large cohort of well-characterized early postmenopausal women, dietary intake of vitamin K1 was associated with several markers of bone health. No interaction was observed between dietary vitamin K and APOE alleles on BMD or biochemical markers of bone turnover. Although advice about dietary guidelines is premature, these data add to the evidence that a healthy diet, which includes green vegetables, would appear to benefit bone.


   ACKNOWLEDGMENTS
 
We thank Martin Shearer and Caroline Bolton Smith with regard to the provision of the UK vitamin K database, Antonia Hardcastle for assistance with estimating dietary vitamin K, and N Hoyle (Roche) for the reagents for propeptide of type 1 collagen measurements. We thank Dr Luan for the SPSS version of power calculation for gene-nutrient interaction. We are also extremely grateful for the hard work of the radiographers and research nurses at the Osteoporosis Research Unit and to all the women who kindly participated in the study.

The author's responsibilities were as follows—HMM: carried out the study, analyzed the data, and wrote the manuscript; FEM: was responsible for the genotyping and wrote the manuscript; SAL-N: was involved in the design of the baseline dietary study and critically reviewed the manuscript; WDF: provided the bone marker results and critically reviewed the manuscript; SHR: gave advice on genetic aspects of the study, interpreted the data, and critically reviewed the manuscript; DMR: was responsible for the study design of APOSS and critically reviewed the manuscript. None of the authors had a financial or commercial interest in any company or organization sponsoring the research for this study.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Eichner JE, Dunn ST, Perveen G, Thompson DM, Stewart KE, Stroehla BC. Apolipoprotein E polymorphism and cardiovascular disease: a HuGE review. Am J Epidemiol 2002;155:487–95.[Abstract/Free Full Text]
  2. Feskanich D, Weber P, Willett WC, Rockett H, Booth SL, Colditz GA. Vitamin K intake and hip fractures in women: a prospective study. Am J Clin Nutr 1999;69:74–9.[Abstract/Free Full Text]
  3. Schoofs MW, van der Klift M, Hofman A, et al. ApoE gene polymorphisms, BMD, and fracture risk in elderly men and women: the Rotterdam study. J Bone Miner Res 2004;19:1490–6.[Medline]
  4. Booth SL, Tucker KL, Chen H, et al. Dietary vitamin K intakes are associated with hip fracture but not with bone mineral density in elderly men and women. Am J Clin Nutr 2000;71:1201–8.[Abstract/Free Full Text]
  5. Szulc P, Chapuy MC, Meunier PJ, Delmas PD. Serum undercarboxylated osteocalcin is a marker of the risk of hip fracture: a three year follow-up study. Bone 1996;18:487–8.[Medline]
  6. Szulc P, Chapuy MC, Meunier PJ, Delmas PD. Serum undercarboxylated osteocalcin is a marker of the risk of hip fracture in elderly women. J Clin Invest 1993;91:1769–74.[Medline]
  7. Hodges SJ, Pilkinton MJ, Stamp TC, et al. Depressed levels of circulating menaquinones in patients with osteoporotic fractures of the spine and femoral neck. Bone 1991;12:387–9.[Medline]
  8. Gage BF, Birman-Deych E, Radford MJ, Nilasena DS, Binder EF. Risk of osteoporotic fracture in elderly patients taking warfarin: results from the National Registry of Atrial Fibrillation 2. Arch Intern Med 2006;166:241–6.[Abstract/Free Full Text]
  9. Douglas AS, Robins SP, Hutchison JD, Porter RW, Stewart A, Reid DM. Carboxylation of osteocalcin in post-menopausal osteoporotic women following vitamin K and D supplementation. Bone 1995;17:15–20.[Medline]
  10. Jamal SA, Browner WS, Bauer DC, Cummings SR. Study of Osteoporotic Fractures Group. Warfarin use and risk for osteoporosis in elderly women. Ann Intern Med 1998;128:829–32.[Abstract/Free Full Text]
  11. Department of Health. Dietary reference values for food energy and nutrients for the United Kingdom. London, United Kingdom: HMSO, 1991.
  12. Shearer MJ, Bach A, Kohlmeier M. Chemistry, nutritional sources, tissue distribution and metabolism of vitamin K with special reference to bone health. J Nutr 1996;126(suppl):1181S–6S.[Abstract/Free Full Text]
  13. Cockayne S, Adamson J, Lanham-New S, Shearer MJ, Gilbody S, Torgerson DJ. Vitamin K and the prevention of fractures: systematic review and meta-analysis of randomized controlled trials. Arch Intern Med 2006;166:1256–61.[Abstract/Free Full Text]
  14. Kohlmeier M, Salomon A, Saupe J, Shearer MJ. Transport of vitamin K to bone in humans. J Nutr 1996;126(suppl):1192S–6S.[Abstract/Free Full Text]
  15. Macdonald HM, New SA, Golden MH, Campbell MK, Reid DM. Nutritional associations with bone loss during the menopausal transition: evidence of a beneficial effect of calcium, alcohol, and fruit and vegetable nutrients and of a detrimental effect of fatty acids. Am J Clin Nutr 2004;79:155–65.[Abstract/Free Full Text]
  16. Tucker KL, Hannan MT, Chen H, Cupples LA, Wilson PW, Kiel DP. Potassium, magnesium, and fruit and vegetable intakes are associated with greater bone mineral density in elderly men and women. Am J Clin Nutr 1999;69:727–36.[Abstract/Free Full Text]
  17. New SA, Bolton-Smith C, Grubb DA, Reid DM. Nutritional influences on bone mineral density: a cross-sectional study in premenopausal women. Am J Clin Nutr 1997;65:1831–9.[Abstract/Free Full Text]
  18. Wachman A, Bernstein DS. Diet and osteoporosis. Lancet 1968;i:958–9.
  19. Corella D, Tucker K, Lahoz C, et al. Alcohol drinking determines the effect of the APOE locus on LDL-cholesterol concentrations in men: the Framingham Offspring Study. Am J Clin Nutr 2001;73:736–45.[Abstract/Free Full Text]
  20. Ordovas JM, Corella D, Cupples LA, et al. Polyunsaturated fatty acids modulate the effects of the APOA1 G-A polymorphism on HDL-cholesterol concentrations in a sex-specific manner: the Framingham Study. Am J Clin Nutr 2002;75:38–46.[Abstract/Free Full Text]
  21. Yan L, Zhou B, Greenberg D, et al. Vitamin K status of older individuals in northern China is superior to that of older individuals in the UK. Br J Nutr 2004;92:939–45.[Medline]
  22. Macdonald HM, New SA, Fraser WD, Campbell MK, Reid DM. Low dietary potassium intakes and high dietary estimates of net endogenous acid production are associated with low bone mineral density in premenopausal women and increased markers of bone resorption in postmenopausal women. Am J Clin Nutr 2005;81:923–33.[Abstract/Free Full Text]
  23. Black D, Duncan A, Robins SP. Quantitative analysis of the pyridinium crosslinks of collagen in urine using ion-paired reversed-phase high-performance liquid chromatography. Anal Biochem 1988;169:197–203.[Medline]
  24. MacDonald AG, Birkinshaw G, Durham B, Bucknall RC, Fraser WD. Biochemical markers of bone turnover in seronegative spondylarthropathy: relationship to disease activity. Br J Rheumatol 1997;36:50–3.[Abstract/Free Full Text]
  25. Macdonald HM, McGuigan FE, Stewart A, et al. Large-scale population-based study shows no evidence of association between common polymorphism of the VDR gene and BMD in British women. J Bone Miner Res 2006;21:151–62.[Medline]
  26. New SA. An epidemiological investigation into the influence of nutritional factors on bone mineral density and bone metabolism. PhD thesis. University of Aberdeen, Scotland, United Kingdom, 1995.
  27. Bodner CH, Soutar A, New SA, et al. Validation of a food frequency questionnaire for use in a Scottish population: correlation of antioxidant vitamin intakes with biochemical measures. J Hum Nutr Diet 1998;11:373–80.
  28. Macdonald HM, New SA, Reid DM. Longitudinal changes in dietary intake in Scottish women around the menopause: changes in dietary pattern result in minor changes in nutrient intake. Public Health Nutr 2005;8:409–16.[Medline]
  29. Bolton-Smith C, Price RAG, Fenton ST, Harrington DJ, Shearer MJ. Compilation of a provisional UK database for the phylloquinone (vitamin K1) content of foods. Br J Nutr 2000;83:389–99.[Medline]
  30. Tunstall-Pedoe H, Smith WC, Crombie IK, Tavendale R. Coronary risk factor and lifestyle variation across Scotland: results from the Scottish Heart Health Study. Scot Med J 1989;34:556–60.[Medline]
  31. Department of Health. Dietary reference values for food energy and nutrients for the United Kingdom. London, United Kingdom: HMSO, 1991.
  32. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr 1997;65(suppl):1220S–8S.[Medline]
  33. Luan JA, Wong MY, Day NE, Wareham NJ. Sample size determination for studies of gene-environment interaction. Int J Epidemiol 2001;30:1035–40.[Abstract/Free Full Text]
  34. Johnston JM, Cauley JA, Ganguli M. APOE 4 and hip fracture risk in a community-based study of older adults. J Am Geriatr Soc 1999;47:1342–5.[Medline]
  35. Kohlmeier M, Saupe J, Schaefer K, Asmus G. Bone fracture history and prospective bone fracture risk of hemodialysis patients are related to apolipoprotein E genotype. Calcif Tissue Int 1998;62:278–81.[Medline]
  36. Pluijm SM, Dik MG, Jonker C, et al. Effects of gender and age on the association of apolipoprotein E epsilon4 with bone mineral density, bone turnover and the risk of fractures in older people. Osteoporos Int 2002;13:701–9.[Medline]
  37. Braam LA, Knapen MH, Geusens P, et al. Vitamin K1 supplementation retards bone loss in postmenopausal women between 50 and 60 years of age. Calcif Tissue Int 2003;73:21–6.[Medline]
  38. Martini LA, Booth SL, Saltzman E, do Rosario Dias de Oliveira Latorre M, Wood RJ. Dietary phylloquinone depletion and repletion in postmenopausal women: effects on bone and mineral metabolism. Osteoporos Int 2006;17:929–35.[Medline]
  39. Rejnmark L, Vestergaard P, Charles P, et al. No effect of vitamin K1 intake on bone mineral density and fracture risk in perimenopausal women. Osteoporos Int 2006;17:1122–32.[Medline]
  40. Prynne CJ, Mishra GD, O'Connell MA, et al. Fruit and vegetable intakes and bone mineral status: a cross-sectional study in 5 age and sex cohorts. Am J Clin Nutr 2006;83:1420–8.[Abstract/Free Full Text]
  41. Collins A, Cashman KD, Kiely M. Phylloquinone (vitamin K1) intakes and serum undercarboxylated osteocalcin levels in Irish postmenopausal women. Br J Nutr 2006;95:982–8.[Medline]
  42. Booth SL, Lichtenstein AH, O'Brien-Morse M, et al. Effects of a hydrogenated form of vitamin K on bone formation and resorption. Am J Clin Nutr 2001;74:783–90.[Abstract/Free Full Text]
Received for publication July 17, 2007. Accepted for publication January 16, 2008.





This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M
Right arrow Search for Related Content
PubMed
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M
Agricola
Right arrow Articles by Macdonald, H. M
Right arrow Articles by Reid, D. M

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS