SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

The phenotypes of apolipoprotein E (Apo E) and their relationship with the bone mineral density (BMD) were examined in 284 unrelated postmenopausal Japanese women aged 47–82 years (64.0 ± 1.0 years, mean ± SE). The Apo E phenotype was analyzed by the isoelectric focusing method, followed by immunoblotting. The relationship between the Apo E phenotype and the vitamin D receptor (VDR) gene or estrogen receptor (ER) gene genotypes was also studied in the same population. The Apo E phenotypic frequencies in our population were 9.9% for E3/2, 66.5% for E3/3, 1.8% for E4/2, 19.7% for E4/3, and 2.1% for E4/4. We classified these phenotypes into three categories: Apo E4−/− (E3/2 and E3/3, n = 217), Apo E4+/− (E4/3 and E4/2, n = 61), and Apo E4+/+ (E4/4, n = 6). The age, body weight, body height, and years since menopause were not significantly different among these three categories. The lumbar BMD values in these three groups were significantly different in the order of E4−/− (0.91 ± 0.01 g/cm2), E4+/− (0.85 ± 0.02 g/cm2), and E4+/+ (0.83 ± 0.06 g/cm2) (p = 0.031). The same trend was also observed for the Z score of the total BMD (p = 0.022). The serum level of intact osteocalcin in E4+/+ (15.2 ± 5.7 ng/ml) was higher than in E4−/− (7.7 ± 0.3 ng/ml) or E4+/− (7.7 ± 0.7 ng/ml) (p = 0.004 by analysis of variance). However, there were no other significant differences in the serum or urinary levels of bone turnover markers. Serum cholesterol in the E4+/+ group tended to be higher than in the other two groups (p = 0.05). There were no significant associations of the VDR and ER genotypes with the Apo E4 phenotype. A multivariate linear regression analysis revealed Apo E4 to be a significant, independent predictor of the Z score of the lumbar BMD. The effect of the Apo E4 allele on the Z score of the lumbar BMD (−0.493 ± 0.152) was not significantly different from that in the AAB of VDR (−0.616 ± 0.225) or PPxx of ER (−0.785 ± 0.314). In conclusion, the Apo E4 allele is associated with a low bone mass in postmenopausal Japanese.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

OSTEOPOROSIS IS A MAJOR health problem in elderly women. Low bone mass is one of the major risk factors for osteoporotic fracture, and it may be partly hereditary.1,2 Therefore, genetic markers that correlate with bone mineral density (BMD) would be useful for predicting future bone loss and for clarifying the mechanisms of bone loss in osteoporosis. Recently, vitamin D receptor (VDR),3 and estrogen receptor (ER) gene4,5 polymorphisms were reported to be associated with the lumbar spine BMD. Since the pathogenesis of osteoporosis is complex, osteoporosis may have multiple genetic backgrounds. Therefore, we have been trying to look for other genetic markers in Japanese postmenopausal women.

Apolipoprotein E (Apo E) is a major component of high-density and low-density lipoproteins, and it is well known that the principal isoform, Apo E3, yields two isoforms,6 Apo E2 and E4. These phenotypes were reported to be related to some involutional diseases, such as late-onset Alzheimer's dementia7 and cardiovascular risk in diabetes mellitus.8 Saupe et al. reported that the level of plasma vitamin K (phylloquinone), which is known to activate osteocalcin molecules in bone, was related to the Apo E phenotype.9 Thus, in the present study, we have investigated the relationship between the Apo E phenotype and the BMD in postmenopausal Japanese women, and we have found a significant association of the BMD with the Apo E4 allele. We have also investigated the associations of the Apo E4 allele with the VDR and ER genotypes and the relative contributions of these genotypes to a low bone mass.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

Subjects

Serum samples were obtained from 284 unrelated postmenopausal Japanese women aged 47–82 years (64.0 ± 1.0 years; mean ± standard error) living in Nagano Prefecture, Japan. Subjects with diseases or conditions that are known causative factors of secondary bone loss, such as postgastrectomy, rheumatoid arthritis, surgical menopause, Grave's disease, or steroid use, were excluded. Patients with dementia or diabetes mellitus were also excluded. All the subjects were ambulatory, and the patients with cardiovascular diseases were also excluded.

Methods

Bone mineral measurement:

The total body (TBMD) and lumbar bone mineral densities (LBMD) were measured by dual-energy X-ray absorptiometry (DEXA) using a fast-scan mode (Lunar DPX-L; Lunar Rad Co., WI, U.S.A.). The coefficients of variation for the total and lumbar BMD (anteroposterior [AP] view) measurements in our institution were 1.5 and 0.5%, respectively. The Z score was calculated using the Lunar DPX-L software based on data for up to 20,000 Japanese women. The Z score was automatically adjusted for age (10-year intervals) and body weight.

Measurements of bone turnover markers and calcium-regulating hormones:

Serum and urinary samples were obtained in the morning from all the subjects. The serum levels of intact osteocalcin (IOC; intact osteocalcin ELISA kit; Teijin Ltd., Tokyo, Japan),10 N-terminal specific osteocalcin (NOC; N-Osteocalcin ELISA kit, Teijin Ltd.),11 intact PTH (intact PTH IRMA kit; Nichols Institute Diagnostics, San Juan Capistrano, CA, U.S.A.),12 25-hydroxyvitamin (25(OH)D) and 1,25-dihydroxyvitamin D (1,25(OH)2D) (competitive protein binding [CPB] method for 25(OH)D and radioreceptor assay for 1,25(OH)2D, after extraction and purification of samples)13 were measured. The urinary excretions of pyridinoline and deoxypyridinoline were measured by the high performance liquid chromatography (HPLC) method,14 and the values were standardized based on the creatinine concentration in the same urine sample.

Measurement of apolipoprotein E phenotype and serum levels of lipids:

For sample preparation, EDTA-treated plasma was obtained from the subjects and stored at −50°C until analysis. The Apo E phenotype determination procedure followed the method reported by Kataoka et al.15

Briefly, the plasma sample was pretreated with Tween 20 and dithiothreitol for 15 minutes. Then the prepared sample was subjected to isoelectric focusing electrophoresis. Fifteen minutes of prefocusing followed by 30 minutes of initial focusing was carried out, and subsequent focusing (final focusing) was conducted for an additional 90 minutes. Immunoblotting was carried out by the method described by Kamboh et al.16 Here, the first antibody for human Apo E was raised in goats, and anti-goat IgG antibody was prepared in rabbits and labeled with alkaline phosphatase. The serum levels of cholesterol, HDL-cholesterol, and triglycerides were also measured with an automated analyzer using enzyme methods.

Measurement of VDR gene and ER gene genotype:

For VDR genotype analysis, genomic DNA was extracted from peripheral white blood cells, and the extracted DNA was amplified by the polymerase chain reaction (PCR). The reaction mixture consisted of 10 mM Tris-HCl (pH 9.0), 50 mM KCl, 2.5 mM MgCl2, 1% Triton X-100, 200 μM of each deoxyribonucleotide triphosphate (dNTP), 1 U of Taq DNA polymerase, and 4 μM of each oligonucleotide primer. The reactions were carried out for 40 cycles by the following steps: denaturation at 94°C for 1 minute, annealing at 60°C for 1 minute, and extension at 72°C for 2 minutes. The amplified DNA fragments were purified by EtOH precipitation, and then the purified DNA fragments were digested with ApaI or BsmI. Each reaction mixture was electrophoresed in a 1.0% agarose gel, and the polymorphic site was detected.3 When the DNA fragment contained the ApaI site or BsmI site, we expressed it as “a” or “b,” and if not, we expressed it as “A” or “B.” For ER genotype analysis, we performed DNA extraction, PCR amplification, and digestion with PvuII and XbaI by exactly the same methods as previously reported.5

Statistical analyses:

Differences in the BMD and biochemical markers among Apo E phenotypes were tested using ANOVA. Fisher's protected least-significant-difference (PLSD) test was used to assess the relationship between each phenotype of Apo E and the Z scores for the BMD or the other markers. Differences in the Z scores for the lumbar BMD among the genotypes of VDR and ER, and Apo E4(+) or (−) were analyzed by Student's t-test (two-tailed). To determine whether the effect of the Apo E phenotype on the LBMD is independent or dependent on other genotypic effects, we have analyzed the frequency of VDR, and ER genotypes in the Apo E4 subjects were calculated by the χ2-test. Furthermore, a multivariate linear regression analysis was employed to estimate the independent contribution of each individual genotypic variable to the bone mass. A p value of less than 0.05 was considered to be statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

The phenotype of Apo E was analyzed in a total of 284 unrelated postmenopausal Japanese women, and the phenotypic distribution in our population was 9.9% for Apo E2/3 (n = 28), 66.5% for Apo E3/3 (n = 189), 1.8% for Apo E2/4 (n = 5), 19.7% for Apo E3/4 (n = 56), and 2.1% for Apo E4/4 (n = 6). These frequencies of Apo E phenotypes are in agreement with the data previously reported for the normal Japanese population17 (Table 1). The calculated allele frequencies in the present study were compared with the same database for a Japanese study17 and a Caucasian study,18 and significant racial differences in the allele frequencies of ε2 and ε3 were found, as reported previously17 (Table 2). However, there were no significant differences in the allele frequencies of ε3 and ε4 between the present results and the populational data for Japanese. Therefore, the present population was representative of the Japanese population.

Table Table 1. APO E PHENOTYPE FREQUENCIES
Thumbnail image of
Table Table 2. APO E ALLELE FREQUENCIES
Thumbnail image of

We classified these Apo E phenotypes into three categories: Apo E4−/− (E2/3 and E3/3, n = 217), Apo E4+/− (E3/4 and E2/4, n = 61), and Apo E4+/+ (E4/4, n = 6). Table 3 shows the basic clinical characteristics of these three groups. None of the parameters showed a significant difference among the three Apo E groups. The LBMDs in the Apo E4+/+, Apo E4+/−, and E4−/− groups were significantly different (p = 0.031 in ANOVA) (Fig. 1a). The Z scores for the LBMD in the Apo E4+/+, E4+/−, and E4−/− groups were also significantly different (p = 0.013 in ANOVA) (Fig. 1b). The same trend was also observed for the TBMD (Figs. 2a and 2b). The serum AI-P activity and the osteocalcin level, measured with N-terminal (NOC)-specific ELISA kits, were not significantly different among the groups (Figs. 3a and 3b). Only the serum level of intact osteocalcin in the Apo E4+/+ group (15.2 ± 5.7 ng/ml) was significantly higher than those in the Apo E4−/− (7.7 ± 0.3 ng/ml) and +/− (7.7 ± 0.7 ng/ml) groups (p = 0.004 by ANOVA). Urinary excretion of pyridinium cross-links (pyridinoline and deoxypyridinoline) in the three groups did not show any significant difference (Figs. 4a and 4b). There were no significant differences in serum levels of calcium, phosphate and calcium-regulating hormones among the groups (Table 4). The serum level of total cholesterol tended to be high in the order of Apo E4−/− < E4+/− < E4+/+. However, there were no significant differences in the serum HDL-Cho and triglyceride levels or the total fat mass as measured by DEXA (Table 5).

thumbnail image

Figure FIG. 1. Lumbar bone mineral density (LBMD) (a) and the Z score for the LBMD (b) in the subjects with and without the Apo E4 allele. The subjects were divided into three groups on the basis of their status for the Apo E4 allele. Group 1 is the negative homozygote for the Apo E4 allele (Apo E4−/−; Apo E2/3 and 3/3, n = 217), group 2 is the heterozygote (Apo E4+/−; Apo E3/4 and 2/4, n = 61), and group 3 is the homozygote (Apo E4+/+; Apo E4/4, n = 6). The mean and SEM of LBMD in each group are written in (a). The overall difference among the groups was statistically significant (p = 0.031) in ANOVA, and the difference between Apo E4−/− and +/− was significant (p = 0.013 in Fisher's PLSD). The Z scores for the LBMD in the three groups are illustrated in (b). The age- and body size–matched Z score was calculated automatically by the Lunar DPX-L software based on the normative database for Japanese women. The Z scores and the SEM for each group are indicated. The overall difference among the groups was statistically significant (p = 0.013) by ANOVA, and the difference between Apo E4−/− and +/− was significant (p = 0.008 by Fisher's PLSD). The boxes and the horizontal bars indicate the percentile lines in the order of the 90th, 75th, 50th (median), 25th, and 10th percentiles, from the top to bottom, as shown the below. The closed circles indicate values beyond the 90th and 10th percentile lines.

Thumbnail image of

Download figure to PowerPoint

thumbnail image

Figure FIG. 2. Total body bone mineral density (TBMD) (a) and the Z score (b) in the subjects with and without the Apo E4 allele. (a) The TBMD was measured by DEXA using a Lunar DPX-L with a fast scan mode. The means and the SEM for the TBMD are indicated. The overall difference among the groups was not significant (p = 0.072), but the difference between Apo E4−/− and +/− was statistically significant (p = 0.022 by Fisher's PLSD). (b) The Z score for the TBMD was calculated automatically by the Lunar DPX-L software based on the normative database for Japanese women. The mean and the SEM of the Z scores for each group are indicated. The overall difference among the groups was statistically significant (p = 0.009) by ANOVA, and the difference between Apo E4−/− and +/− was significant (p = 0.003 by Fisher's PLSD).

Download figure to PowerPoint

thumbnail image

Figure FIG. 3. Serum level of alkaline phosphatase (AI-P) activity (a) and osteocalcin (b) in the subjects with and without the Apo E4 allele. (a) The serum level of AI-P activity in the subjects with and without Apo E4 allele was shown. The mean and the SEM of AI-P for each Group are indicated in the figure. There were no significant differences among the groups. (b) The serum osteocalcin levels were measured with two different osteocalcin kits. NOC means the values obtained with the osteocalcin ELISA kit, which consists of two region-specific antibodies: one recognizes the N-terminal residue, while the other recognizes the midportion of the human osteocalcin molecule. IOC means the values obtained with intact human osteocalcin ELISA kit, which consists of N-terminal–specific and C-terminal–specific antibodies; this kit can only cross-react with the intact form of human osteocalcin. There were no significant differences in the serum levels of osteocalcin measured with the NOC kit (left: Apo E4−/−, 15.0 ± 0.5 ng/ml; +/−, 14.7 ± 0.9 ng/ml; +/+, 15.8 ± 3.0 ng/ml). For the serum level of intact osteocalcin in the three groups (Apo E4−/−, 7.7 ± 0.3 ng/ml; +/−, 7.7 ± 0.7 ng/ml; +/+, 15.2 ± 5.7 ng/ml), a significantly higher level was found in Apo E4+/+ compared with −/− and +/− (p = 0.0038 by ANOVA).

Download figure to PowerPoint

thumbnail image

Figure FIG. 4. Urinary excretion of pyridinoline (a) and deoxypyridinoline (b) in the subjects with and without the Apo E4 allele. Morning urine samples were subjected to HPLC measurement of urinary pyridinolium cross-links. There were no significant differences in these parameters among the groups.

Download figure to PowerPoint

Table Table 3. BACKGROUND CHARACTERISTICS OF THE APO E GROUPS
Thumbnail image of
Table Table 4. CALCIUM METABOLISM OF THE GROUPS
Thumbnail image of
Table Table 5. LIPID METABOLISM IN THE GROUPS
Thumbnail image of

The frequencies of VDR and ER genotypes in the Apo E4 subjects were calculated. The AABB or AABb (renamed AAB)3 of the VDR gene and PPxx of the ER gene5 were not significantly frequent in the subjects with Apo E4 allele (Table 6). To examine for a confounding influence on the BMD among the genotypes of VDR, ER, and Apo E, a multivariate linear regression analysis was performed. Here, the genotypes were coded as integers (1 or 2), and these values were used in the multiple regression analysis. The regression coefficients for the Z score of LBMD when the VDR, ER, and Apo E were taken as the predictors, were 0.353 (p = 0.233), 0.338 (p = 0.391), and 0.549 (p = 0.0035), respectively, and R2 was 0.037 with a y intercept of −1.148 (p = 0.0148). The Z scores for LBMD in the group having the genotype associated with a low bone density were compared with the group without. For VDR, the Z score of the group with AAB was −0.616 ± 0.225 (n = 22), while that of the other group was −0.026 ± 0.087 (n = 262) (p = 0.031). For ER, the Z score of the group with PPxx was −0.785 ± 0.314 (n = 11), while that of the others was 0.012 ± 0.086 (n = 273; p = 0.040). The Z score of the E4(+) group was −0.493 ± 0.152 (n = 67), whereas that of the other group was 0.050 ± 0.092 (n = 217; p = 0.004) (Fig. 5). The mean values of the Z score for AAB and PPxx were in good agreement with those determined in our previous studies.5,19

thumbnail image

Figure FIG. 5. Z scores for the lumbar bone mineral density (LBMD) in the subjects with and without genotypes associated with low bone density. The Z scores for the LBMD in the subjects with and without genotypes associated with a low bone mass (VDR, ER and Apo E) are shown. AAB for VDR means the genotype of AABB or AABb. E4(+) means the subjects with E2/4, 3/4, and 4/4. These genotypes were simultaneously analyzed in the same subjects. There were significant differences in the Z scores between the subjects with a known genotype related to low bone mass and the subjects without. However, there was no significant difference in the Z score among the subjects with AAB, PPxx or Apo E4(+).

Download figure to PowerPoint

Table Table 6. FREQUENCIES OF VDR AND ER GENOTYPES IN APO E4(+) AND (−) SUBJECTS
Thumbnail image of

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References

In the present study, we demonstrated a significant gene-dose effect of the Apo E4 allele on the BMD of the lumbar spine as well as the total body. The relationship between the Apo E phenotype and the BMD should be examined by the populational study. We analyzed the phenotypic distribution and the allelic frequency in the present study population for comparison with the previously reported Japanese database.17 The results showed no significant differences between them in either the phenotypic or allelic distributions. Therefore, our present population is surmised to be representative of the Japanese population.

Subjects with Apo E4(+) have been reported to show an association with some involutional diseases.7,8 These diseases may account for the secondary risk of bone loss due to disability. Thus, we excluded subjects with Alzheimer's disease, diabetes mellitus, and disabled subjects. None of the subjects had serious cardiovascular disease, and all of the subjects were ambulatory.

Subjects with homozygous for Apo E4 seemed to be associated with a relatively higher bone turnover, suggested by the higher serum level of intact osteocalcin. Morrison et al. also reported that the VDR genotype associated with a low BMD showed a higher serum level of osteocalcin.5 It is well accepted that a high bone turnover state usually leads to excessive bone loss.20,21 However, it is still necessary to clarify whether or not the Apo E4+/+ group has high bone turnover, because of the small number of cases with Apo E4+/+ in the present study and also the absence of a gene-dose effect on the serum level of intact osteocalcin. In addition, there was no significant difference in the urinary excretion of pyridinium cross-links. In the present study, the correlations between bone turnover markers and the LBMD were significant; namely, the correlation coefficients between the LBMD and Al-P, IOC, NOC, pyridinoline, and deoxypyridinoline were −0.253 (p < 0.0001), −0.191 (p = 0.0021), −0.219 (p = 0.0004), −0.249 (p < 0.0001), and −0.304 (p < 0.0001), respectively. These findings showed good agreement with earlier studies.20,21 We had reported that the ER genotype (PPxx) was associated with a low bone mass without any significant rise in Al-P, osteocalcin, or pyridinium cross-links. So it can be said that some genetic effects on the BMD may not be associated with a high bone turnover state.

A significant gene-dose effect was found for the serum cholesterol level. This finding is in good agreement with a previous study,22 and it may indicate that receptor-specific retardation of lipid metabolism exists in the Apo E4 group.23 However, there was no direct correlation between the serum levels of cholesterol and triglycerides and the BMD in the present study (data not shown).

Therefore, the exact reasons the subjects with the Apo E4 allele showed a lower BMD in the present study remain obscure. Saupe et al. reported that the serum level of phylloquinone depended on the Apo E phenotype, namely E2 > E3 > E4.9 Since vitamin K is a known modulator of bone metabolism in vitro24,25 as well as in vivo,26,27 Saupe et al.'s data may provide an explanation for our present findings.

Apo E has been reported to be produced by human monocytes28 and glial cells,29 and the Apo E produced by the latter cells may play an important role in the repair or growth of neural cells depending on their phenotype.29 Apo E produced by monocytes in the bone marrow may play an important role in bone repair or growth via the receptor-mediated pathway proposed by Mahley.30

There was no significant association of AAB of VDR or PPxx of ER with the Apo E4 allele. Furthermore, multivariate analysis revealed that Apo E4 was a significant and independent predictor of the Z score of the LBMD. The VDR and ER genotypes did not show any significant correlation with the Z score of LBMD in the multivariate analysis, and this may be due to the smaller number of cases with AAB or PPxx. Judging from the R2 value in the multivariate linear regression analysis, the total contribution of genomic effects on the Z score of LBMD was only 4%. This may be partly explained by the low frequencies of the significant genotypes; namely, AAB was 6.7% (19/284), PPxx was 4.2% (12/284), while Apo E4(+) was 23.6% (67/284) in our population. Therefore, the genomic effect of these genotypes on LBMD may be important in each individual, but not in the general population.

The Z scores of the LBMD in the groups with AAB, PPxx, or Apo E4 were not significantly different. Therefore, the effect of Apo E4 on the BMD is thought to be comparable to the effects of AAB and PPxx. Although the findings of the present study are the first evidence of a possible relationship between the Apo E isoform and the BMD, it will be necessary to show direct evidence of the effect of Apo E isoforms on bone cell biology as seen in the neurons.29

In conclusion, the Apo E4 allele was associated with a low bone mass, without any association with the VDR and ER genotypes. Therefore, Apo E4 is considered to be an independent risk factor for low bone mass in postmenopausal women.

References

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. References
  • 1
    Slemenda CW, Christian JC, Williams CJ, Norton JA, Johnston CC Jr 1991 Genetic determinations of bone mass in adult women: A reevaluation of twin model and the potential importance of gene interaction on heritability estimates J Bone Miner Res 6:561567.
  • 2
    Pocock NA, Eisman JA, Hopper JL, Yeates MG, Sambrook PN, Ebert S 1987 Genetic determinants of bone mass in adults: A twin study J Clin Invest 80:706710.
  • 3
    Morrison NA, Qi JC, Tokita A, Kelly PJ, Crofts L, Nguyen TV, Sambrook PN, Eisman JA 1994 Prediction of bone density from vitamin D receptor alleles Nature 367:284287.
  • 4
    Sano M, Inoue S, Hosoi T, Ouchi Y, Emi M, Shiraki M, Orimo H 1995 Association of estrogen receptor dinucleotide repeat polymorphism with osteoporosis Biochem Biophys Res Commun 217:378383.
  • 5
    Kobayashi S, Inoue S, Hosoi T, Ouchi Y, Shiraki M, Orimo H 1996 Association of bone mineral density with polymorphism of the estrogen receptor gene J Bone Miner Res 11:306311.
  • 6
    Weisgraber KH, Rall SC Jr, Mahley RW 1981 Human E apoprotein heterogeneity: Cysteine-arginine interchanges in the amino acid sequence of apo E isoforms J Biol Chem 256:90779083.
  • 7
    Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Gaskell PC, Small GW, Roses AD, Haines JL, Pericak-Vance MA 1993 Gene dose of apolipoprotein E type 4 allele and risk of Alzheimer's disease in late onset families Science 261:921923.
  • 8
    Laakso M, Kesaniemi A, Kervinen K, Jauhiainen M, Pyorala K 1991 Relation of coronary heart disease and apolipoprotein E phenotype in patients with non-insulin dependent diabetes Br Med J 303:11591162.
  • 9
    Saupe J, Shearer MJ, Kahlmeier M 1993 Phylloquinone transport and its influence on γ-χαλβoζψγλνταματε residues of osteocalcin in patients on maintenance hemodyalysis Am J Clin Nutr 58:204208.
  • 10
    Hosoda K, Eguchi H, Nakamoto T, Kubota T, Honda H, Jindai S, Hasegawa R, Kiyoki M, Yamaji T, Shiraki M 1992 Sandwich immunoassay for intact human osteocalcin Clin Chem 38:22332238.
  • 11
    Eguchi H, Hosoda K, Kurihara N, Nakamoto T, Kobayashi S, Arita M, Hasegawa R, Ohta T, Kiyoki M, Yamaji T, Shiraki M 1995 Sandwich immunoassay specific for the N-terminal sequence of osteocalcin J Immunol Meth 184:231240.
  • 12
    Brown RC, Aston JP, Weeks I, Woodhead JS 1987 Circulating intact parathyroid hormone measured by a two-site immunochemiluminometric assay J Clin Endocrinol Metab 65:407414.
  • 13
    Ishizuka S, Naruchi T, Hashimoto Y, Orimo H 1981 Radioreceptor assay for 1 alpha 24(R), 25 trihydroxyvitamin D in human serum J Nutr Sci Vitaminol 27:7179.
  • 14
    Ubelhart D, Gineyts E, Chapuy MC, Delmas PD 1990 Urinary excretion of pyridinium crosslinks: A new marker of bone resorption in metabolic bone disease Bone Miner 8:8796.
  • 15
    Kataoka S, Paidi M, Howard BV 1994 Simplified isoelectric focusing/immunoblotting determination of apoprotein E phenotype Clin Chem 40:1113.
  • 16
    Kamboh MI, Ferrell RE, Kottke B 1988 Genetic studies of human apolipoproteins. V. A novel rapid procedure to screen apolipoprotein E polymorphism J Lipid Res 29:15351543.
  • 17
    Eto M, Watanabe K, Ishii K 1986 A racial difference in apolipoprotein E allele frequencies between the Japanese and Caucasian populations Clin Genet 30:422427.
  • 18
    Sing CF, Davignon J 1985 Role of the apolipprotein E polymorphism in determining normal plasma lipid and lipoprotein variation Am J Hum Genet 37:268285.
  • 19
    Shiraki M, Shiraki Y, Aoki C, Eguchi H 1995 Genotypes of vitamin D receptor gene and response to vitamin D3 treatment in postmenopausal osteoporosis J Bone Miner Res 10(Suppl 1):S187.
  • 20
    Johansen J, Riis BJ, Delmas PD, Christiansen C 1988 Plasma BGP: An indicator of spontaneous bone loss and the effect of estrogen treatment in postmenopausal women Eur J Clin Invest 18:191195.
  • 21
    Ubelhart D, Schlemmer A, Johansen JS, Gineyts E, Christiansen C, Delmas PD 1991 Effect of menopause and hormone replacement therapy on the urinary excretion of pyridinium cross-links J Clin Endocrinol Metab 72:367373.
  • 22
    Davignon J, Gregg RE, Sing CF 1988 Apolipoprotein E polymorphism and atherosclerosis Arteriosclerosis 8:121.
  • 23
    Weisgraber KH, Innerarity TL, Mahley RW 1982 Abnormal lipoprotein receptor-binding activity of human E apolipoprotein due to cysteine-arginine interchange at a single site J Biol Chem 257:25182521.
  • 24
    Hara K, Akiyama Y, Tajima T, Shiraki M 1993 Menatetrenone inhibits bone resorption partly through inhibition of PGE2 synthesis in vitro J Bone Miner Res 8:535542.
  • 25
    Hara K, Akiyama Y, Nakamura T, Murota S, Morita I 1995 The inhibitory effect of vitamin K2 (Menatetrenone) on bone resorption may be related to its side chain Bone 16:179184.
  • 26
    Plantalech L, Guillaumont M, Vergnaud P, Leclercq M, Delmas PD 1991 Impairment of gamma carboxylation of circulating osteocalcin (bone gla protein) in elderly women J Bone Miner Res 6:12111216.
  • 27
    Hart JP, Shearer MJ, Klenerman L, Catterall A, Reeve J, Sambrook PN, Dodds RA, Bitensky L, Chayen J 1985 Electrochemical detection of depressed circulating levels of vitamin K1 in osteoporosis J Clin Endocrinol Metab 60:12681269.
  • 28
    Basu SK, Ho Y, Brown MS, Bilheimer DW, Anderson GW, Goldstein J 1982 Biochemical and genetic studies of the apoprotein E secreted by mouse macrophages and human monocytes J Biol Chem 257:97889795.
  • 29
    Nathan BP, Bellosta S, Sanan DA, Weisgraber KH, Mahley RW, Pitas RE 1994 Differential effects of apolipoproteins E3 and E4 on neuronal growth in vitro Science 264:850852.
  • 30
    Mahley RW 1988 Apolipoprotein E: Cholesterol transport protein with expanding role in cell biology Science 240:622630.