- Email: [email protected]
Relation of homocysteine, folate, and vitamin B12 to bone mineral density of postmenopausal women
Bone 33 (2003) 956 –959
www.elsevier.com/locate/bone
Relation of homocysteine, folate, and vitamin B12 to bone mineral density of postmenopausal women Angelo Cagnacci, M.D.,a,* Francesco Baldassari, M.D. Giovanni Rivolta, M.D.,b Serenella Arangino, M.D.,a and Annibale Volpe, M.D.a a
Department of Obstetrics Gynecology and Pediatrics, Obstetrics and Gynecology, Policlinico of Modena, Modena, Italy b Medical Scientific Service, Bracco S.p.A., Milano, Italy Received 6 June 2003; revised 14 July 2003; accepted 14 July 2003
Abstract Genetic hyperhomocysteinemia is associated with skeletal abnormalities and osteoporosis. We tested whether levels of homocysteine and critical co-enzymes of homocysteine metabolism, such as vitamin B12 and folate, are related to lumbar spine bone mineral density (BMD) measured by DEXA in 161 postmenopausal women. Folate but not homocysteine or vitamin B12, was lower in osteoporotic than normal women (7.2 ⫾ 0.9 ng/L vs 11.4 ⫾ 0.7 ng/L, P ⬍ 0.003). Folate, but not homocysteine or vitamin B12, was independently related to BMD (r ⫽ 0.254, P ⬍ 0.011). BMD progressively increased from the lowest to the highest folate quartile (1.025 ⫾ 0.03 g/cm2 vs 1.15 ⫾ 0.03 g/cm2, P ⬍ 0.01) even when covaried for weight, which was the only other variable related to BMD. The present data suggest a major association between folate and bone mineralization. © 2003 Elsevier Inc. All rights reserved. Keywords: Folic acid; Homocysteine; Viatamin B12; Osteoporosis; Bone; Menopause; Nutrition
Introduction Genetic hyperhomocysteinemia is associated with skeletal abnormalities and osteoporosis [1,2]. Similarly, in experimental animals administration of methionine, a homocysteine precursor, is associated with skeletal alteration and bone demineralization [3]. More recently, it has been reported that in postmenopausal Japanese women the VV mutation of the gene encoding methylene tetrahydrofolate reductase (MTHFR), a key enzyme involved in remethylation of homocysteine to methionine, is associated with a lower bone mineral density (BMD) at the spine and total body [4]. The association was not found in a case– control study investigating forearm BMD and MTHFR mutation [5], but it was reconfirmed in the large Danish Osteoporosis Prevention Trial, in which MTHFR mutation was associated * Corresponding author. Dipartimento Materno Infantile, Ginecologia e Ostetricia, Policlinico di Modena, via del Pozzo 71, 41100, Modena, Italy. Fax: ⫹39059-4224394. E-mail address: [email protected] (A. Cagnacci). 8756-3282/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2003.07.001
with lower BMD and increased fracture incidence at both the spine and hip [6]. Homocysteine can be deleterious even at moderate levels [7], such as those achieved during aging or diets poor in vitamin B12 and folic acid [8], the necessary coenzymes for its metabolism. The aim of this study was to test whether in postmenopausal women levels of homocysteine, folate, and vitamin B12 are related to BMD.
Materials and methods One-hundred-sixty-one volunteer postmenopausal women gave their informed consent to participate into the study, which was previously approved by the local ethics committee and institutional review board. Women were recruited from those of our menopause center between September 2001 and March 2002. Inclusion criteria were no previous osteoporotic fracture, 24 months of amenorrhea, and no previous hormone replacement therapy. Women with liver or renal disease or endocrine or metabolic abnormalities
A. Cagnacci et al. / Bone 33 (2003) 956 –959
957
Table 1 Characteristics of the 161 postmenopausal women, divided on the basis of BMD values
Age (years) Menarche (years) Fertile years (years) Years since menopause (years) Weight (kg) BMI (kg/m2) Sedentary women Drinking ⬎2 glasses milk or equivalents (%) Drinking spirits (%) Smoking women (%) Cigarettes/day (n) Hcy (mol/L) Folate (ng/L) Vitamin B12 (pg/L)
Osteporosis (n ⫽ 28)
Osteopenia (n ⫽ 61)
Normal (n ⫽ 72)
54.7 ⫾ 0.9 12.7 ⫾ 0.24 34.8 ⫾ 0.8 7.0 ⫾ 0.74 64.9 ⫾ 1.8 25.6 ⫾ 0.8 17.8%e 75.0% 10.7% 32.1% 7.0 ⫾ 1.8 11.09 ⫾ 1.0 7.2 ⫾ 0.9 578 ⫾ 4
53.5 ⫾ 0.6 12.6 ⫾ 0.17 34.9 ⫾ 0.6 5.9 ⫾ 0.74 63.8 ⫾ 1.1 25.9 ⫾ 0.5 44.2% 93.4% 6.5% 32.7% 7.3 ⫾ 1.4 10.5 ⫾ 0.9 8.3 ⫾ 0.7 625 ⫾ 75
52.5 ⫾ 0.6 12.5 ⫾ 0.19 35.6 ⫾ 0.7 4.3 ⫾ 0.60c 70.8 ⫾ 1.4d 27.5 ⫾ 0.6a 36.1% 86.1% 8.3% 31.9% 9.9 ⫾ 1.6 10.2 ⫾ 0.6 11.4 ⫾ 0.7b 563 ⫾ 38
P ⬍ 0.05, b P ⬍ 0.003 versus osteoporosis and osteopenia. P ⬍ 0.05, d P ⬍ 0.05 versus osteoporosis. e P ⬍ 0.05 versus normal and osteopenia. a c
and/or receiving medicine known to influence bone mineralization or homocysteine, folate, or vitamin B12 levels were excluded. The dietary habit of each woman was evaluated by an interview with a dietitian, and the following risk factors [9] were measured: present and previous use of caffeine, spirits, or cigarettes; daily use of milk or milk products; and recreational or working physical activity. BMD was evaluated at the lumbar spine (L2–L4) by DEXA (Lunar DPX, Madison, WI, USA). In each woman, a blood sample was collected from an antecubital vein between 8 AM and 9 AM, following a 24-h low-protein diet and overnight fasting. Blood was collected in tubes placed on ice and immediately centrifuged in the cold, and the serum was immediately stored in two different tubes at ⫺25°C until assayed. Samples of all women were assayed together. Total serum homocysteine was measured by highperformance liquid chromatography (Bio-Rad Laboratories GmbH, Munich, Germany) with a reversed-phase column and fluorometric detection with excitation wavelength of the detector at ⫽385 nm and emission wavelength at ⫽515 nm. Before measurement of total homocysteine in serum, the protein-bound fraction of homocysteine was released by a reduction process and derivatized by use of a fluorescent thiol-specific dye. After protein precipitation 20 l of supernatant was injected. The assay had a sensitivity of 0.5 mol/L and intra- and interassay coefficients of variation (CV) of 3.8 and 6%, respectively. Folate and vitamin B12 were measured in a single assay, simultaneously, with the SimulTRAC-SNB radioimmunological kit (ICN Pharmaceuticals, Orangeburg, NY, USA). The sensitivity was 0.6 ng/mL for folate and 75 pg/mL for vitamin B12. The intraassay CV was 4.8% for folate and 11.2% for vitamin B12. Stepwise regression analysis was used to evaluate variables independently related to BMD. Single regression anal-
ysis was used to express the relation between an independent variable and BMD. Single-factor analysis of variance or covariance was used to perform group comparisons. Statistical analysis was considered significant at P values of 0.05.
Results At the DEXA evaluation, 28 women had osteoporotic (T score below ⫺2.5), 61 had osteopenic (T score between ⫺2.5 and ⫺1), and 72 had normal BMD values. Osteoporotic women had lower values of serum folate, weight, and body mass index (BMI). Women with osteoporosis were in postmenopause for a longer period than were women with a normal BMD (Table 1). In all women considered together, stepwise regression analysis shows that only weight (F⫽12.7) and level of serum folate (F ⫽ 8.4) are independently related to BMD (r ⫽ 0.371, P ⬍ 0.0001). Instead, no significant influence is exerted by BMI, age, age at menarche, years of fertile life, years since menopause, previous or present smoking habit, daily physical activity, and daily use of milk, milk products, alcohol, or caffeine. The linear relationship between serum folate and BMD is reported in Fig. 1 (r ⫽ 0.254, P ⬍ 0.002). Stratification of BMD on serum folate quartiles shows a progressive increase in BMD from the lowest to the highest quartile (from 1.025 ⫾ 0.03 to 1.150 ⫾ 0.03 g/cm2, P ⬍ 0.01), which is significant even when corrected for weight (Fig 2). The highest serum folate quartile shows also the highest serum vitamin B12 (from 424.6 ⫾ 34.3 to 734.6 ⫾ 87.0 pg/L, P ⬍ 0.01) and the lowest serum homocysteine levels (from 12.6 ⫾ 1.2 to 7.7 ⫾ 0.3 mol/L, P ⬍ 0.01).
958
A. Cagnacci et al. / Bone 33 (2003) 956 –959
Fig. 1. Linear regression analysis between serum levels of folate and lumbar spine BMD observed in 161 postmenopausal women.
Discussion The present data show an association between serum folate and BMD that is stronger than the relationships with all other osteoporosis risk factors except body weight. Serum folate furnishes a short-term indication of folate status, more long-term information being furnished by erythrocyte folate [9]. However, in several studies low serum folate has been associated with a reduction in folate status, as documented by low erythrocyte folate and elevated homocysteine [9]. Also in our data low serum folate was associated with high homocysteine, indicating a true folate deficiency. The mechanism linking serum folate to BMD is unclear, and the relationship may only by chance. However, several hypotheses can be formulated. Low serum folate can be an index of an unbalanced diet leading to a deficiency of nutrients necessary for bone mineralization [10]. The dietary habit of each woman was evaluated by an interview with a dietitian, which is not sufficiently accurate to quantify folic acid intake, but rather appropriate to exclude major alimentary deficits. At this evaluation women with osteoporosis did not show any major difference in comparison to women with normal BMD. Alternatively, low serum folate may be due to an inappropriate absorption of folate [11], as well as of nutrients important for bone mineralization. A direct effect of folate at bone cells is also possible. Folate is essential for intracellular processes, prevention of DNA damage, reduction of oxidative stress, and prevention of apoptosis [12]. Some of these mechanisms, such as the reduction of oxidative stress, seem to be mediated by a reduction of homocysteine levels, but effects independent of serum homocysteine have already been reported for the cardiovascular system [13]. Reduced levels of folate are also associated with mutations of MTHFR gene, particularly when folic acid intake is low [14,15]. Accordingly, a greater prevalence of MTHFR gene mutations in individuals in the lower folate quartiles is possible. MTHFR gene mutations is associated with altered homocysteine metabolism, low BMD, and an increased incidence of fractures [4,6]. Individuals with raised homocysteine, as part of homocystinuria, exhibit skeletal abnormalities and low BMD [1,2], because homocysteine is believed to interfere with
cross links of newly formed collagen [16] and, consequently, with bone strength and bone mineralization [17– 19]. In our study the highest folate levels and the lowest BMD values were indeed associated with the highest homocysteine levels. However, at regression analysis, levels of homocysteine were not related to BMD. Lack of a relationship may be due to the variability of the homocysteine assay or, more than that, to the fact that actual measurement of fasting homocysteine is not sufficiently accurate to reliably predict the effect of this substance on bone mineralization. Indeed, measurement of fasting homocysteine does not evaluate the duration and extent of the homocysteine rise that follows the intake of animal protein rich in its precursor methionine [20]. It is estimated that almost 50% of individuals diagnosed as having hyperhomocysteinemia after a methionine test show normal homocysteine levels under fasting conditions [21]. Moreover, before its assay, a reduction process is necessary to cleave homocysteine from the proteins to which it is bound. This step does not cleave amide-linked homocysteine, which, consequently, is not measured by available methods [7]. The possibility that fasting homocysteine is inappropriate for predicting bone mineralization is further confirmed by studies in individuals with genetic hyperhomocystinuria or MTHFR gene mutations. In individuals with genetic hyperhomocystinuria, levels of homocysteine are not related to bone modifications [17]. Similarly, the Danish Osteoporosis Prevention Trial has shown that the MTHFR gene mutation is associated with decreased BMD at the spine and femoral neck, but that levels of homocysteine, measured in a subset of individuals, are not related to BMD values [6]. Although the mechanism linking folate to BMD can be debated, the strong relationship between the two sheds a new light on the possible effect that the decrease in folate associated with aging or unbalanced diets [8] can exert on bone mineralization. The relation furnishes a strong support for future studies aimed at fully exploring the pathophysiological and clinical implications of folate intake and BMD in postmenopausal women.
Fig. 2. Mean ⫾ SE lumbar spine bone mineral density (BMD) according to folate quartiles in 161 postmenopausal women. **P ⬍ 0.02, ***P ⬍ 0.01 versus quartiles I and II.
A. Cagnacci et al. / Bone 33 (2003) 956 –959
Acknowledgments This study was partially supported by Bracco SpA, Milano Italy. References [1] Brenton DP. Skeletal abnormalities in homocystinuria. Postgrad Med J 1977;53:488 –96. [2] Mudd SH, Skovby F, Levy HL, Pettigrew KD, Wilcken B, Pyeritz RE, et al. The natural history of homocystinuria due to cystathionine beta-synthase deficiency. Am J Hum Genet 1985;37:1–31. [3] Whiting SJ, Draper HH. Effect of achronic acid load as sulfate or sulfur amino acids on bone metabolism in adult rats. J Nutr 1981; 111:1721– 6. [4] Miyao M, Morita H, Hosoi T, Kurihara H, Inoue S, Hoshino S, et al. Association of methylene tetrahydrofolate-reductase (MTHFR) polymorphism with bone mineral density in postmenopausal Japanese women. Calcif Tissue Int 2000;66:190 – 4. [5] Jorgensen HL, Madsen JS, Madsen B, Saleh MMA, Abrahamsem B, Fenger M, Lauritzen JB. Association of a common allelic polymorphism (C677T) in the methylene tetrahydrofolate reductase gene with a reduced risk of osteoporotic fractures: a case control study in Danish postmenopausal women. Calcif Tissue Int 2002; 71: 386 –92. [6] Abrahamsen B, Madsen JS, Tofteng CL, Stilgren L, Bladbjerg EM, Kristensen SR, et al. A common methylenetetrahydrofolate reductase (C677T) polymorphism is associated with low bone mineral density and increased fracture incidence after menopause: longitudinal data from the Danish Osteoporosis Prevention Study. J Bone Miner Res 2003;18:723–9. [7] Krumdieck CL, Prince CW. Mechanisms of homocysteine toxicity on connective tissues: impications for the morbidity of ageing. J Nutr 2000;130:365s– 8s. [8] Selhub J, Jacques PF, Wilson PWF, Rush D, Rosenberg IH. Vitamin status and intake as primary determinants of homocysteinemia in an edely population. JAMA 1993;270:2693– 8. [9] Neuhouser ML, Beresford SAA. Folic acid: are current fortification levels adequate? Nutrition 2001;17:868 –72. [10] Johnell O, Gullberg B, Kanis JA, Allander E, Elffors L, Dequeker J, et al. Risk factors for hip fracture in European women: the MEDOS study. J Bone Miner Res 1995;10:1802–15.
959
[11] Neuhouser ML, Beresford SAA, Hickok DE, Monsen ER. Absorption of dietary and supplemental folate in women with prior pregnancies with neural tube defects and controls. J Am Coll Nutr 1998;17:625– 30. [12] Luckock M. Folic acid: nutritional biochemestry, molecular biology, and role in disease processes. Mol Genet Metab 2000;71:121–38. [13] Doshi SN, McDowell IFW, Moat SJ, Payne N, Durrant HJ, Lewis MJ, Goodfellow J. Folic acid improves endothelial function in coronary artery disease via mechanisms largely independent of homocysteine lowering. Circulation 2002;105:22– 6. [14] Silaste ML, Rantala M, Sampi M, Alfthan G, Aro A, Kesaniemi YA. Polymorphisms of key enzymes in homocysteine metabolism affect diet responsiveness of plasma homocysteine in healthy women. J Nutr 2001;131:2643–7. [15] Guinotte CL, Burns MG, Axume JA, Hata H, Urrutia TF, Alamilla A, et al. Methylenetetrahydrofolate reductase 677C-T variant modulates folate status response to controlled folate intakes in young women. J Nutr 2003;133:1272– 80. [16] Lubec B, Fang-Kircher S, Lubec T, Blom HJ, Boers GH. Evidence for McKusick’s hypothesis of deficient collagen cross-linking in patients with homocystinuria. Biochim Biophys Acta 1996;1315: 159 – 62. [17] Grant SF, Reid DM, Blake G, Herd R, Fogelman I, Ralston SH. Reduced bone density and osteoposis associated with a polymorphic Sp1 binding site in the collagen type I alpha 1 gene. Nat Genet 1996;14:203–5. [18] Uitterlinden AG, Burger H, Huang Q, Yue F, McGuigan FE, Grant SF, et al. Relation of alleles of the collagen type I alpha 1 gene to bone density and the risk of osteoporotic fractures in postmenopausal women. N Engl J Med 1998;338:1016 –21. [19] Langdahl BL, Ralston SH, Grant SFA, Eriksen EF. An Sp1 binding site polymorphism in the COL1A1 gene predicts osteoporotic fractures in both men and women. J Bone Miner Res 1998;13:1384 –9. [20] Guttormsen AB, Schneede J, Fiskestrand T, Ueland PM, Refsum HM. Plasma concentrations of homocysteine and other aminothiol compounds are related to food intake in healthy human subjects. J Nutr 1994;124:1934 – 41. [21] van der Griend R, Biesma DH, Banga JD. Postmethionine-load homocysteine determination for the diagnosis of hyperhomocysteinaemia and efficacy of homocysteine lowering treatment regimens. Vasc Med 2002;7:29 –33.
www.elsevier.com/locate/bone
Relation of homocysteine, folate, and vitamin B12 to bone mineral density of postmenopausal women Angelo Cagnacci, M.D.,a,* Francesco Baldassari, M.D. Giovanni Rivolta, M.D.,b Serenella Arangino, M.D.,a and Annibale Volpe, M.D.a a
Department of Obstetrics Gynecology and Pediatrics, Obstetrics and Gynecology, Policlinico of Modena, Modena, Italy b Medical Scientific Service, Bracco S.p.A., Milano, Italy Received 6 June 2003; revised 14 July 2003; accepted 14 July 2003
Abstract Genetic hyperhomocysteinemia is associated with skeletal abnormalities and osteoporosis. We tested whether levels of homocysteine and critical co-enzymes of homocysteine metabolism, such as vitamin B12 and folate, are related to lumbar spine bone mineral density (BMD) measured by DEXA in 161 postmenopausal women. Folate but not homocysteine or vitamin B12, was lower in osteoporotic than normal women (7.2 ⫾ 0.9 ng/L vs 11.4 ⫾ 0.7 ng/L, P ⬍ 0.003). Folate, but not homocysteine or vitamin B12, was independently related to BMD (r ⫽ 0.254, P ⬍ 0.011). BMD progressively increased from the lowest to the highest folate quartile (1.025 ⫾ 0.03 g/cm2 vs 1.15 ⫾ 0.03 g/cm2, P ⬍ 0.01) even when covaried for weight, which was the only other variable related to BMD. The present data suggest a major association between folate and bone mineralization. © 2003 Elsevier Inc. All rights reserved. Keywords: Folic acid; Homocysteine; Viatamin B12; Osteoporosis; Bone; Menopause; Nutrition
Introduction Genetic hyperhomocysteinemia is associated with skeletal abnormalities and osteoporosis [1,2]. Similarly, in experimental animals administration of methionine, a homocysteine precursor, is associated with skeletal alteration and bone demineralization [3]. More recently, it has been reported that in postmenopausal Japanese women the VV mutation of the gene encoding methylene tetrahydrofolate reductase (MTHFR), a key enzyme involved in remethylation of homocysteine to methionine, is associated with a lower bone mineral density (BMD) at the spine and total body [4]. The association was not found in a case– control study investigating forearm BMD and MTHFR mutation [5], but it was reconfirmed in the large Danish Osteoporosis Prevention Trial, in which MTHFR mutation was associated * Corresponding author. Dipartimento Materno Infantile, Ginecologia e Ostetricia, Policlinico di Modena, via del Pozzo 71, 41100, Modena, Italy. Fax: ⫹39059-4224394. E-mail address: [email protected] (A. Cagnacci). 8756-3282/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2003.07.001
with lower BMD and increased fracture incidence at both the spine and hip [6]. Homocysteine can be deleterious even at moderate levels [7], such as those achieved during aging or diets poor in vitamin B12 and folic acid [8], the necessary coenzymes for its metabolism. The aim of this study was to test whether in postmenopausal women levels of homocysteine, folate, and vitamin B12 are related to BMD.
Materials and methods One-hundred-sixty-one volunteer postmenopausal women gave their informed consent to participate into the study, which was previously approved by the local ethics committee and institutional review board. Women were recruited from those of our menopause center between September 2001 and March 2002. Inclusion criteria were no previous osteoporotic fracture, 24 months of amenorrhea, and no previous hormone replacement therapy. Women with liver or renal disease or endocrine or metabolic abnormalities
A. Cagnacci et al. / Bone 33 (2003) 956 –959
957
Table 1 Characteristics of the 161 postmenopausal women, divided on the basis of BMD values
Age (years) Menarche (years) Fertile years (years) Years since menopause (years) Weight (kg) BMI (kg/m2) Sedentary women Drinking ⬎2 glasses milk or equivalents (%) Drinking spirits (%) Smoking women (%) Cigarettes/day (n) Hcy (mol/L) Folate (ng/L) Vitamin B12 (pg/L)
Osteporosis (n ⫽ 28)
Osteopenia (n ⫽ 61)
Normal (n ⫽ 72)
54.7 ⫾ 0.9 12.7 ⫾ 0.24 34.8 ⫾ 0.8 7.0 ⫾ 0.74 64.9 ⫾ 1.8 25.6 ⫾ 0.8 17.8%e 75.0% 10.7% 32.1% 7.0 ⫾ 1.8 11.09 ⫾ 1.0 7.2 ⫾ 0.9 578 ⫾ 4
53.5 ⫾ 0.6 12.6 ⫾ 0.17 34.9 ⫾ 0.6 5.9 ⫾ 0.74 63.8 ⫾ 1.1 25.9 ⫾ 0.5 44.2% 93.4% 6.5% 32.7% 7.3 ⫾ 1.4 10.5 ⫾ 0.9 8.3 ⫾ 0.7 625 ⫾ 75
52.5 ⫾ 0.6 12.5 ⫾ 0.19 35.6 ⫾ 0.7 4.3 ⫾ 0.60c 70.8 ⫾ 1.4d 27.5 ⫾ 0.6a 36.1% 86.1% 8.3% 31.9% 9.9 ⫾ 1.6 10.2 ⫾ 0.6 11.4 ⫾ 0.7b 563 ⫾ 38
P ⬍ 0.05, b P ⬍ 0.003 versus osteoporosis and osteopenia. P ⬍ 0.05, d P ⬍ 0.05 versus osteoporosis. e P ⬍ 0.05 versus normal and osteopenia. a c
and/or receiving medicine known to influence bone mineralization or homocysteine, folate, or vitamin B12 levels were excluded. The dietary habit of each woman was evaluated by an interview with a dietitian, and the following risk factors [9] were measured: present and previous use of caffeine, spirits, or cigarettes; daily use of milk or milk products; and recreational or working physical activity. BMD was evaluated at the lumbar spine (L2–L4) by DEXA (Lunar DPX, Madison, WI, USA). In each woman, a blood sample was collected from an antecubital vein between 8 AM and 9 AM, following a 24-h low-protein diet and overnight fasting. Blood was collected in tubes placed on ice and immediately centrifuged in the cold, and the serum was immediately stored in two different tubes at ⫺25°C until assayed. Samples of all women were assayed together. Total serum homocysteine was measured by highperformance liquid chromatography (Bio-Rad Laboratories GmbH, Munich, Germany) with a reversed-phase column and fluorometric detection with excitation wavelength of the detector at ⫽385 nm and emission wavelength at ⫽515 nm. Before measurement of total homocysteine in serum, the protein-bound fraction of homocysteine was released by a reduction process and derivatized by use of a fluorescent thiol-specific dye. After protein precipitation 20 l of supernatant was injected. The assay had a sensitivity of 0.5 mol/L and intra- and interassay coefficients of variation (CV) of 3.8 and 6%, respectively. Folate and vitamin B12 were measured in a single assay, simultaneously, with the SimulTRAC-SNB radioimmunological kit (ICN Pharmaceuticals, Orangeburg, NY, USA). The sensitivity was 0.6 ng/mL for folate and 75 pg/mL for vitamin B12. The intraassay CV was 4.8% for folate and 11.2% for vitamin B12. Stepwise regression analysis was used to evaluate variables independently related to BMD. Single regression anal-
ysis was used to express the relation between an independent variable and BMD. Single-factor analysis of variance or covariance was used to perform group comparisons. Statistical analysis was considered significant at P values of 0.05.
Results At the DEXA evaluation, 28 women had osteoporotic (T score below ⫺2.5), 61 had osteopenic (T score between ⫺2.5 and ⫺1), and 72 had normal BMD values. Osteoporotic women had lower values of serum folate, weight, and body mass index (BMI). Women with osteoporosis were in postmenopause for a longer period than were women with a normal BMD (Table 1). In all women considered together, stepwise regression analysis shows that only weight (F⫽12.7) and level of serum folate (F ⫽ 8.4) are independently related to BMD (r ⫽ 0.371, P ⬍ 0.0001). Instead, no significant influence is exerted by BMI, age, age at menarche, years of fertile life, years since menopause, previous or present smoking habit, daily physical activity, and daily use of milk, milk products, alcohol, or caffeine. The linear relationship between serum folate and BMD is reported in Fig. 1 (r ⫽ 0.254, P ⬍ 0.002). Stratification of BMD on serum folate quartiles shows a progressive increase in BMD from the lowest to the highest quartile (from 1.025 ⫾ 0.03 to 1.150 ⫾ 0.03 g/cm2, P ⬍ 0.01), which is significant even when corrected for weight (Fig 2). The highest serum folate quartile shows also the highest serum vitamin B12 (from 424.6 ⫾ 34.3 to 734.6 ⫾ 87.0 pg/L, P ⬍ 0.01) and the lowest serum homocysteine levels (from 12.6 ⫾ 1.2 to 7.7 ⫾ 0.3 mol/L, P ⬍ 0.01).
958
A. Cagnacci et al. / Bone 33 (2003) 956 –959
Fig. 1. Linear regression analysis between serum levels of folate and lumbar spine BMD observed in 161 postmenopausal women.
Discussion The present data show an association between serum folate and BMD that is stronger than the relationships with all other osteoporosis risk factors except body weight. Serum folate furnishes a short-term indication of folate status, more long-term information being furnished by erythrocyte folate [9]. However, in several studies low serum folate has been associated with a reduction in folate status, as documented by low erythrocyte folate and elevated homocysteine [9]. Also in our data low serum folate was associated with high homocysteine, indicating a true folate deficiency. The mechanism linking serum folate to BMD is unclear, and the relationship may only by chance. However, several hypotheses can be formulated. Low serum folate can be an index of an unbalanced diet leading to a deficiency of nutrients necessary for bone mineralization [10]. The dietary habit of each woman was evaluated by an interview with a dietitian, which is not sufficiently accurate to quantify folic acid intake, but rather appropriate to exclude major alimentary deficits. At this evaluation women with osteoporosis did not show any major difference in comparison to women with normal BMD. Alternatively, low serum folate may be due to an inappropriate absorption of folate [11], as well as of nutrients important for bone mineralization. A direct effect of folate at bone cells is also possible. Folate is essential for intracellular processes, prevention of DNA damage, reduction of oxidative stress, and prevention of apoptosis [12]. Some of these mechanisms, such as the reduction of oxidative stress, seem to be mediated by a reduction of homocysteine levels, but effects independent of serum homocysteine have already been reported for the cardiovascular system [13]. Reduced levels of folate are also associated with mutations of MTHFR gene, particularly when folic acid intake is low [14,15]. Accordingly, a greater prevalence of MTHFR gene mutations in individuals in the lower folate quartiles is possible. MTHFR gene mutations is associated with altered homocysteine metabolism, low BMD, and an increased incidence of fractures [4,6]. Individuals with raised homocysteine, as part of homocystinuria, exhibit skeletal abnormalities and low BMD [1,2], because homocysteine is believed to interfere with
cross links of newly formed collagen [16] and, consequently, with bone strength and bone mineralization [17– 19]. In our study the highest folate levels and the lowest BMD values were indeed associated with the highest homocysteine levels. However, at regression analysis, levels of homocysteine were not related to BMD. Lack of a relationship may be due to the variability of the homocysteine assay or, more than that, to the fact that actual measurement of fasting homocysteine is not sufficiently accurate to reliably predict the effect of this substance on bone mineralization. Indeed, measurement of fasting homocysteine does not evaluate the duration and extent of the homocysteine rise that follows the intake of animal protein rich in its precursor methionine [20]. It is estimated that almost 50% of individuals diagnosed as having hyperhomocysteinemia after a methionine test show normal homocysteine levels under fasting conditions [21]. Moreover, before its assay, a reduction process is necessary to cleave homocysteine from the proteins to which it is bound. This step does not cleave amide-linked homocysteine, which, consequently, is not measured by available methods [7]. The possibility that fasting homocysteine is inappropriate for predicting bone mineralization is further confirmed by studies in individuals with genetic hyperhomocystinuria or MTHFR gene mutations. In individuals with genetic hyperhomocystinuria, levels of homocysteine are not related to bone modifications [17]. Similarly, the Danish Osteoporosis Prevention Trial has shown that the MTHFR gene mutation is associated with decreased BMD at the spine and femoral neck, but that levels of homocysteine, measured in a subset of individuals, are not related to BMD values [6]. Although the mechanism linking folate to BMD can be debated, the strong relationship between the two sheds a new light on the possible effect that the decrease in folate associated with aging or unbalanced diets [8] can exert on bone mineralization. The relation furnishes a strong support for future studies aimed at fully exploring the pathophysiological and clinical implications of folate intake and BMD in postmenopausal women.
Fig. 2. Mean ⫾ SE lumbar spine bone mineral density (BMD) according to folate quartiles in 161 postmenopausal women. **P ⬍ 0.02, ***P ⬍ 0.01 versus quartiles I and II.
A. Cagnacci et al. / Bone 33 (2003) 956 –959
Acknowledgments This study was partially supported by Bracco SpA, Milano Italy. References [1] Brenton DP. Skeletal abnormalities in homocystinuria. Postgrad Med J 1977;53:488 –96. [2] Mudd SH, Skovby F, Levy HL, Pettigrew KD, Wilcken B, Pyeritz RE, et al. The natural history of homocystinuria due to cystathionine beta-synthase deficiency. Am J Hum Genet 1985;37:1–31. [3] Whiting SJ, Draper HH. Effect of achronic acid load as sulfate or sulfur amino acids on bone metabolism in adult rats. J Nutr 1981; 111:1721– 6. [4] Miyao M, Morita H, Hosoi T, Kurihara H, Inoue S, Hoshino S, et al. Association of methylene tetrahydrofolate-reductase (MTHFR) polymorphism with bone mineral density in postmenopausal Japanese women. Calcif Tissue Int 2000;66:190 – 4. [5] Jorgensen HL, Madsen JS, Madsen B, Saleh MMA, Abrahamsem B, Fenger M, Lauritzen JB. Association of a common allelic polymorphism (C677T) in the methylene tetrahydrofolate reductase gene with a reduced risk of osteoporotic fractures: a case control study in Danish postmenopausal women. Calcif Tissue Int 2002; 71: 386 –92. [6] Abrahamsen B, Madsen JS, Tofteng CL, Stilgren L, Bladbjerg EM, Kristensen SR, et al. A common methylenetetrahydrofolate reductase (C677T) polymorphism is associated with low bone mineral density and increased fracture incidence after menopause: longitudinal data from the Danish Osteoporosis Prevention Study. J Bone Miner Res 2003;18:723–9. [7] Krumdieck CL, Prince CW. Mechanisms of homocysteine toxicity on connective tissues: impications for the morbidity of ageing. J Nutr 2000;130:365s– 8s. [8] Selhub J, Jacques PF, Wilson PWF, Rush D, Rosenberg IH. Vitamin status and intake as primary determinants of homocysteinemia in an edely population. JAMA 1993;270:2693– 8. [9] Neuhouser ML, Beresford SAA. Folic acid: are current fortification levels adequate? Nutrition 2001;17:868 –72. [10] Johnell O, Gullberg B, Kanis JA, Allander E, Elffors L, Dequeker J, et al. Risk factors for hip fracture in European women: the MEDOS study. J Bone Miner Res 1995;10:1802–15.
959
[11] Neuhouser ML, Beresford SAA, Hickok DE, Monsen ER. Absorption of dietary and supplemental folate in women with prior pregnancies with neural tube defects and controls. J Am Coll Nutr 1998;17:625– 30. [12] Luckock M. Folic acid: nutritional biochemestry, molecular biology, and role in disease processes. Mol Genet Metab 2000;71:121–38. [13] Doshi SN, McDowell IFW, Moat SJ, Payne N, Durrant HJ, Lewis MJ, Goodfellow J. Folic acid improves endothelial function in coronary artery disease via mechanisms largely independent of homocysteine lowering. Circulation 2002;105:22– 6. [14] Silaste ML, Rantala M, Sampi M, Alfthan G, Aro A, Kesaniemi YA. Polymorphisms of key enzymes in homocysteine metabolism affect diet responsiveness of plasma homocysteine in healthy women. J Nutr 2001;131:2643–7. [15] Guinotte CL, Burns MG, Axume JA, Hata H, Urrutia TF, Alamilla A, et al. Methylenetetrahydrofolate reductase 677C-T variant modulates folate status response to controlled folate intakes in young women. J Nutr 2003;133:1272– 80. [16] Lubec B, Fang-Kircher S, Lubec T, Blom HJ, Boers GH. Evidence for McKusick’s hypothesis of deficient collagen cross-linking in patients with homocystinuria. Biochim Biophys Acta 1996;1315: 159 – 62. [17] Grant SF, Reid DM, Blake G, Herd R, Fogelman I, Ralston SH. Reduced bone density and osteoposis associated with a polymorphic Sp1 binding site in the collagen type I alpha 1 gene. Nat Genet 1996;14:203–5. [18] Uitterlinden AG, Burger H, Huang Q, Yue F, McGuigan FE, Grant SF, et al. Relation of alleles of the collagen type I alpha 1 gene to bone density and the risk of osteoporotic fractures in postmenopausal women. N Engl J Med 1998;338:1016 –21. [19] Langdahl BL, Ralston SH, Grant SFA, Eriksen EF. An Sp1 binding site polymorphism in the COL1A1 gene predicts osteoporotic fractures in both men and women. J Bone Miner Res 1998;13:1384 –9. [20] Guttormsen AB, Schneede J, Fiskestrand T, Ueland PM, Refsum HM. Plasma concentrations of homocysteine and other aminothiol compounds are related to food intake in healthy human subjects. J Nutr 1994;124:1934 – 41. [21] van der Griend R, Biesma DH, Banga JD. Postmethionine-load homocysteine determination for the diagnosis of hyperhomocysteinaemia and efficacy of homocysteine lowering treatment regimens. Vasc Med 2002;7:29 –33.