Calcif Tissue Int (2007) 81:450–454 DOI 10.1007/s00223-007-9089-y
The Association Between Cysteine, Bone Turnover, and Low Bone Mass M. Baines Æ M.-B. Kredan Æ A. Davison Æ G. Higgins Æ C. West Æ W. D. Fraser Æ L. R. Ranganath
Received: 22 May 2007 / Accepted: 26 October 2007 / Published online: 4 December 2007 Ó Springer Science+Business Media, LLC 2007
Abstract Background With the identification of hyperhomocysteinemia as a risk factor for developing osteoporosis, the contribution of thiols metabolically linked with homocysteine (tHcy) may be of importance. Cysteine (Cys) is formed from tHcy and is involved in bone metabolism via incorporation into collagen and cysteine protease enzymes. Methods We investigated the association of plasma Cys and related thiols, the bone turnover markers C-telopeptide (CTX) and procollagen type 1 N propeptide (P1NP) and folate and vitamin B6 with calcaneal bone mineral density (BMD) in 328 postmenopausal British women grouped according to their BMD measurement. Results Subjects with low BMD had a significantly lower plasma Cys concentration (146.3 vs. 177.7 lmol/l, p \ 0.0001), a significantly higher recent fracture rate (30.9% vs. 16.4%, p = 0.017), and a significantly higher percentage of current smokers (26.4% vs. 7.3%. p = 0.003) than those with normal BMD. Additionally, they had a significantly lower plasma Cys, and higher plasma tHcy and CTX, than those with osteopenia. In the whole population, Cys was significantly associated with BMD, weight, height, smoking habit, log creatinine, Cys-Gly, log tHcy, and log folate, but the significant positive association of
M. Baines (&) M.-B. Kredan A. Davison G. Higgins W. D. Fraser L. R. Ranganath Department of Clinical Biochemistry and Metabolic Medicine, Royal Liverpool University Hospital, Prescot Street, Liverpool L7 8XP, UK e-mail:
[email protected] C. West Department of Public Health, University of Liverpool, Liverpool L69 3GA, UK
123
Cys with BMD was maintained after correction for all other variables (r = 0.197, p = 0.003). After weight, Cys was the next most significant predictor of BMD in a stepwise multiple linear regression model. Conclusion Our study suggests a significant association between plasma Cys and BMD. A reduced Cys concentration, possibly modulated by smoking, or reduced flux from tHcy, may lead to reduced availability for collagen formation. Increased osteoclast activation, possibly as a result of relative hyperhomocysteinemia, may lead to increased Cys utilization in cysteine proteases. Keywords BMD CTX Cysteine Cysteine proteases Homocysteine Smoking
Recent research has implicated increased plasma total homocysteine (tHcy) as a risk factor for developing osteoporosis, though at present there is little data on the mechanistic effect of tHcy on bone metabolism [1]. Homocysteine is formed by demethylation of the essential amino acid methionine, and is metabolized either by remethylation to methionine (the methylation cycle), or by trans-sulfuration to cystathionine, then cysteine and eventually sulfate. Reduction in the activity of the enzyme cystathionine-b-synthase (CBS; EC 4.2.1.22), by either mutation or lack of the cofactor pyridoxal-phosphate (vitamin B6), can lead to a relative hyperhomocysteinemia with a consequent downstream reduction of cysteine (Cys). The Cys formed from tHcy has a number of possible metabolic fates. It may be incorporated into proteins and enzymes, keeping its functional –SH group which can be converted into the disulfide bridges necessary for the strength and stability of many structural proteins, including
M. Baines et al.: Cysteine, Bone Turnover, and Low Bone Mass
collagen. Alternatively, it may be converted to and stored intracellularly as the tripeptide glutathione (GSH), the major cellular antioxidant [2]. Up to 20% of the circulating concentration of tHcy is bound to Cys [3]; thus the metabolic fates of tHcy and Cys are inter-related. We and others have shown that increased plasma tHcy is associated with osteoporosis [4–6], though we are not aware of any studies of a possible involvement of Cys with bone metabolism. Studies have reported primarily on the relationship of tHcy and Cy in patients with cardiovascular disease [7–10]. Risk assessments of abnormal Cys concentrations in disease states other than vascular disease are few, and have produced variable findings. Zhang et al. [11] found that a higher plasma Cys concentration predicted a reduced risk of breast cancer, and similarly Goodman et al. [13] reported a possible inverse association between plasma Cys concentrations and the risk of cervical dysplasia. In contrast, though possibly due to different pathology, El-Khairy et al. [12] reported that higher Cys concentrations were associated with adverse pregnancy outcomes (pre-eclampsia, premature delivery, and low birth weight) in 5,883 Norwegian women. In the present study we investigated the association between Cys and the related thiols tHcy, cysteinyl-glycine (Cys-Gly) and GSH, vitamin B6, bone mineral density (BMD), and the bone turnover markers C-telopeptide (CTX) and procollagen type 1 N propeptide (P1NP) in a cohort of postmenopausal British women.
451
Subjects and Methods Three hundred and twenty-eight British women at least 12 months since their last menstrual period were invited to attend our Metabolic Bone Unit following a Peripheral Instantaneous X-ray Imager (PIXI) (GELunar, Madison, WI, USA) scan of the os calcis as part of a General Practitioner osteoporosis screening program. All subjects were ambulant and free-living and gave written consent to the study, which was approved by the local research ethics committee. The demographic details of the subjects are presented in Table 1. The subjects were assigned to one of three groups according to their WHO-equivalent ‘T’-score, established for the PIXI scanner following correlation with axial BMD measurement of T-scores: a normal group (N) with a T-score of –0.6 SD or higher, an osteopenic group (OPN) with a T-score between –0.7 and –1.6 SD, and an osteoporotic group (OP) with a T-score of lower than -1.6 SD [14]. Recruitment continued prospectively until each group had approximately equal numbers. Following an overnight fast of 12 hr, blood samples were taken for tHcy, Cys, Cys-Gly, GSH, vitamin B6, CTX, and P1NP. tHcy, Cys, Cys-Gly, and GSH were measured by a dedicated HPLC system (Drew DS 30, Drew Scientific, UK) [15] on an EDTA plasma sample that had been separated from the cells within 1 hr of venepuncture and stored at –20 °C. The intra-assay coefficients of variation (CV) were 2.7% at a concentration of 16.9 lmol/l (tHcy),
Table 1 Distribution of age, social habits, and measurements between the groups ranked by BMD Normal (N) BMD [–0.5 SD (n = 110) Age (years)
67.7 (45–84)
Alcohol (units/week)
Osteopenia (OPN) BMD -0.6 to -1.6 SD (n = 108) 66.3 (40–80)
Osteoporosis (OP) BMD \–1.7 SD (n = 110) 68.5 (41–86)
p (regression or chi-squared) 0.049c
2.2 (3.82)
2.4 (4.36)
3.2 (5.94)
72.3 (12.2) 1.61 (0.07)
64.2 (11.0) 1.59 (0.06)
60.1 (10.1) 1.56 (0.07)
\0.0001a,b 0.007c \0.0001a 0.003c
7.3
18.3
26.4
0.0003a 0.027b
Fracture in previous year (%)
16.4
23.1
30.9
0.017a
Age at menopause (years)
48.5 (5.41)
47.2 (5.56)
46.7 (4.49)
0.010a
Time since menopause (years)
19.3 (9.10)
19.1 (9.27)
21.8 (8.70)
0.038a 0.034c
177.4 (49.7)
164.7 (35.5)
146.2 (38.2)
33.6 (11.3)
29.7 (9.2)
31.5 (8.1)
4.5 (2.0)
4.4 (1.6)
4.6 (2.0)
12.0 (4.9)
11.2 (4.2)
13.6 (6.63)
0.003b
9.4 (6.43)
10.2 (4.63)
8.1 (8.69)
0.049c
50.0 (25.0)
51.6 (42.4)
43.8 (29.2)
Weight (kg) Height (m) Current smoker (%)
Cysteine (lmol/l) Cysteinyl-glycine (lmol/l) Glutathione (lmol/l) Homocysteine (lmol/l) Folate (lmol/l) Vitamin B6 (nmol/l) CTX (lg/l)
0.34 (0.15)
P1NP (lg/l)
28.4 (13.8)
0.32 (0.20) 28.7 (17.7)
0.40 (0.29) 32.7 (22.2)
NS
\0.0001a 0.050b 0.001c 0.011b NS
NS 0.024c NS
Data are given as mean (SD), except age which is given as mean (range) NS, no significant difference between any group a
N vs. OP;
b
N vs. OPN;
c
OPN vs. OP
123
452
3.3% at 292 lmol/l (Cys), 6.0% at 13.0 lmol/l (Cys-G), and 7.8% at 3.7 lmol/l (GSH). Vitamin B6 was measured as pyridoxal-5-phosphate by an HPLC method (Chromsystems, Munich, Germany) on a heparin plasma sample. The intra-assay CV was 2.6% at a concentration of 78 nmol/l. CTX and P1NP were measured on a plasma sample by electrochemiluminescence immunoassay (ECLIA) (Roche Elecsys 2010, Roche Diagnostics, Lewes, UK). For CTX, the intra-assay CV was 1.6% at a concentration of 0.84 lg/l and the inter-assay CV 1.9% at the same concentration. For P1NP, the intra-assay CV was 2.9% at a concentration of 29 lg/l and the inter-assay CV 3.8% at a concentration of 57 lg/l. Alcohol (ethanol) intake per week, current smoking status, and history of any fracture in the previous year were assessed by questionnaire. Statistical analysis was carried out using SPSS, version 14 (SPSS, Chicago, IL, USA). Variance across the groups was assessed by ANOVA (continuous data) or chi-squared test (descriptive data). The Pearson correlation coefficient was calculated as a measure of association, after log transformation of data where appropriate. Stepwise multiple linear regression was used to assess independent predictors of BMD. A p value of \0.05 was considered statistically significant for all analyses.
M. Baines et al.: Cysteine, Bone Turnover, and Low Bone Mass
weight (r = 0.320, p \ 0.0001), height (r = 0.0197, p = 0.001), smoking (r = –0.132, p = 0.026), log creatinine (r = 0.128, p = 0.035), Cys-Gly (r = 0.514, p \ 0.0001), log tHcy (r = 0.191, p = 0.001), and log folate (r = 0.130, p = 0.038). Pearson partial correlation analysis showed that the association of Cys with BMD was maintained (r = 0.220, p = 0.001) after correction for age, weight, height, log folate, log creatinine, log Hcy, and smoking status. Fracture incidence was not significantly associated with Cys or any other thiol, nor with any other biochemical or demographic measurement. Smokers as a whole had a significantly lower BMD (–1.50 vs. –0.80 SD, p \ 0.001) and lower Cys (149.7 vs. 165.4 lmol/l, p = 0.019) than nonsmokers, and a trend toward lower plasma vitamin B6 concentration (39.6 vs. 48.5 nmol/l, p = 0.080). Stepwise multiple linear regression analysis of the whole population showed that after weight, Cys was the next most significant predictor of BMD (t = 3.377, p = 0.001). Other significant contributors were log Hcy, smoking, and height (Table 2). When nonsmokers were analyzed separately, weight remained the main predictor, followed by age, height, and Cys (t = 1.378, p = 0.170).
Discussion Results Table 1 summarizes the variables distributed between the groups ranked according to BMD. The OP group were significantly lighter, shorter, and had significantly higher incidences of current smoking and recent fracture, but not alcohol intake, than the N group. The OP group were also significantly younger at menopause than the N group and were further from menopause than the other groups. Cys was significantly lower in the OP group when compared with both the other groups, though none of the other thiols (Cys-Gly, tHcy or GSH) were significantly different in the OP group compared with the N group. When adjustments for age, weight, height, folate, creatinine, tHcy, and smoking status were made, the mean Cys concentration in the OP group (149.7 lmol/l) remained significantly lower than that of the N group (171.6, p = 0.001) and the OPN group (162.5, p = 0.029). tHcy was significantly higher in the OP group than the OPN group, and Cys-Gly significantly lower in the OPN group compared with the N group. CTX was significantly higher in the OP group compared with the OPN group. There were no significant differences between any of the groups for P1NP, vitamin B6, or plasma creatinine. Pearson bivariate analysis showed that Cys was significantly associated with BMD (r = 0.309, p \ 0.0001),
123
Compared with a control population with normal BMD, postmenopausal osteoporotic women in this study had a significantly reduced plasma Cys concentration, and, in the population as a whole, BMD was significantly positively Table 2 Predictors of BMD: stepwise multiple linear regression analysis Independent variables
Unstandardized coefficients B
SE
Standardized beta coefficient
t
p value
All subjects Intercept
–5.516
2.684
Weight
0.040
0.007
0.353
5.677 \0.0001
Cys
0.006
0.002
0.210
3.377
0.001
Log Hcy
–1.548
0.607
–0.189
–2.552
0.011
Smoker
–0.418
0.210
–0.115
–1.987
0.048
Height
2.461
1.286
0.116
1.913
0.057
–1.276
5.293
Weight
0.042
0.008
0.374
5.342 \0.0001
Cys
0.010
0.007
0.340
1.378
0.170
Log Hcy
–1.979
2.635
–0.184
–0.751
0.454
Age
–0.023
0.013
–0.117
–1.791
0.075
2.391
1.482
0.113
1.614
0.108
Nonsmokers Intercept
Height
M. Baines et al.: Cysteine, Bone Turnover, and Low Bone Mass
correlated with Cys. An association between Cys and BMD has not previously been reported, and the cause of the association is intriguing. The main determinants of plasma Cys in this study are the demographic factors weight, height, and smoking habit; the related metabolites tHcy, folate, and Cys-Gly; and an indicator of kidney function, creatinine. The relationship between smoking and plasma Cys remains ill-defined. In the Hordaland Homocysteine study (16,176 subjects, age range 40–67 years), El-Khairy et al. [8] found a weak and inverse relationship between smoking and Cys in elderly men and women, and no relationship in young men and women. In a smaller, case-control study of 750 patients with vascular disease and a similar number of controls, all aged \60 years, the same authors found a significant negative relationship of Cys with smoking [16]. However, a recent study of 117 subjects, age range 19– 60 years, in which exposure to cigarette smoke was assessed by urine cotinine concentration, concluded that cigarette smoking was not a determinant of plasma Cys concentration, though the mean plasma Cys concentration was significantly lower in smokers when compared with nonsmokers [17]. Thus the contribution of smoking to the plasma Cys concentration requires further study. Our study shows that smokers have both significantly lower BMD and plasma Cys than nonsmokers. However, when we corrected for smoking in a partial correlation analysis, a significant positive correlation of BMD with Cys remained, suggesting that smoking may be only a partial cause of the changes in BMD seen in our patients. Smoking may be associated with a reduction in plasma vitamin B6, which may impair the conversion of tHcy to Cys via CBS. The OP group had, compared with the N group, a significantly higher percentage of subjects reporting fracture (of any kind) in the previous year, together with a significantly lower plasma Cys concentration. It could be that the reduction in Cys was evidence of incorporation of Cys into healing bone or was a reflection of higher bone resorption, demonstrated by significantly higher CTX, and therefore increased utilization in the OP group. However, we were unable to demonstrate a significant association between fracture incidence and Cys. The higher recent fracture incidence is likely to be due to a combination of lower BMD, higher bone resorption, and poorer bone quality. Less Cys available for incorporation into collagen may be contributing to a reduction in bone quality. The inverse relation between tHcy and Cys in the OP group might suggest that a trans-sulfuration defect could be impairing the irreversible conversion of tHcy to Cys. However, there is no difference among the groups in their vitamin B6 status, so the data could be explained by a high incidence of CBS deficiency in the OP group. Although we were unable to test for CBS mutation in our population, the
453
estimated prevalence of heterogeneity for CBS is only 1% in the population as a whole [18], and would therefore prove an unlikely cause. Our studies suggest that there is a significant association between plasma Cys and BMD, which remains after correction for demographic factors and determinants of Cys including plasma tHcy and smoking habit. After weight, Cys was the next significant predictor of BMD in our total population, and behind weight, age, and height when smokers were excluded. A reduced Cys concentration, possibly due to factors related to smoking, or reduced flux from tHcy, may lead to reduced availability for collagen formation. Alternatively, increased osteoclast activation, possibly as a result of relative hyperhomocysteinemia [19, 20], may lead to increased Cys utilization in cysteine proteases. If plasma Cys is related to increased bone turnover, it may alter as therapy modulates bone metabolism. Further studies are required to define the exact mechanisms involved in the associations we have reported and to follow any trends in plasma Cys in response to active treatment for osteoporosis. Acknowledgment We thank REMEDI for financial support for a large part of this work.
References 1. Hermann M, Widmann T, Herrmann W (2005) Homocysteine: A newly recognized risk factor for osteoporosis. Clin Chem Lab Med 43:1111–1117 2. Ingenbleek Y, Young VR (2004) The essentiality of sulfur is closely related to nitrogen metabolism; a clue to hyperhomocysteinaemia. Nutr Res Rev 17:135–151 3. Ueland PM (1995) Homocysteine species as components of plasma redox thiol status. Clin Chem 41:340–342 4. van Meurs JBJ, Dhonukshe-Rutten RAM, Pluijm SMF, van der Klift M, de Jonge R, Lindemans J, de Groot LCPGM, Hofman A, Witteman JCM, van Leeuwen JPTM, Breteler MMB, Lips P, Pols HAP, Uitterlinden AG (2004) Homocysteine levels and the risk of osteoporotic fracture. N Engl J Med 350:2033–2041 5. McLean RR, Jacques PF, Selhub J, Tucker KL, Samelson EJ, Broe KE, Hannan MT, Cupples LA, Kiel DP (2004) Homocysteine as a predictive factor for hip fracture in older persons. N Engl J Med 350:2042–2049 6. Baines M, Kredan M-B, Usher J, Davison A, Higgins G, Taylor W, West C, Fraser WD, Ranganath LR (2007) The association of homocysteine and its determinants MTHFR genotype, folate, vitamin B12 and vitamin B6 with bone mineral density in postmenopausal British women. Bone 40:730–736 7. Yardim-Akaydin S, Ozkan Y, Ozkan E, Torun M, Simsek B (2003) The role of plasma thiol compounds and antioxidant vitamins in patients with cardiovascular diseases. Clin Chim Acta 338:99–105 8. El-Khairy L, Ueland PM, Nygard O, Refsum H, Vollset SE (1999) Lifestyle and cardiovascular disease risk factors as determinants of total cysteine in plasma: The Hordaland Homocysteine Study. Am J Clin Nutr 70:1016–1024
123
454 9. Ozkan Y, Ozkan E, Simsek B (2002) Plasma total homocysteine and cysteine levels as cardiovascular risk factors in coronary heart disease. Int J Cardiol 82:269–277 10. Marcucci R, Brunelli T, Giusti B, Fedi S, Pepe G, Poli D, Prisco D, Abbate R, Gensini GF (2001) The role of cysteine and homocysteine in venous and arterial thrombotic disease. Am J Clin Pathol 116:56–60 11. Zhang SM, Willett WC, Selhub J, Manson JE, Colditz GA, Hankinson SE (2003) A prospective study of plasma total cysteine and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 12:1188–1193 12. El-Khairy L, Vollset SE, Refsum H, Ueland PM (2003) Plasma total cysteine, pregnancy complications, and adverse pregnancy outcomes: The Hordaland Homocysteine Study. Am J Clin Nutr 77:467–472 13. Goodman MT, McDuffie K, Hernandez B, Wilkens LR, Selhub J (2000) Case-control study of plasma folate, vitamin B(12), and cysteine as markers of cervical dysplasia. Cancer 89:376–382 14. Barr RJ, Adebajo A, Fraser WD, Halsey JP, Kelsey C, Stewart A, Reid DM (2005) Can peripheral DXA measurements be used to predict fractures in elderly women living in the community? Osteoporosis Int 16:1177–1183
123
M. Baines et al.: Cysteine, Bone Turnover, and Low Bone Mass 15. Zhang M, Gunter EW, Pfeiffer CM (2001) Evaluation of the Drew Scientific DS30 homocysteine assay in comparison with the Centers for Disease Control and Prevention Reference HPLC method. Clin Chem 47:966–967 16. El-Khairy L, Ueland PM, Refsum H, Graham IM, Vollset SE (2001) Plasma total cysteine as a risk factor for vascular disease: The European Concerted Action Project. Circulation 103:2544– 2559 17. Sobczak A, Wardas W, Zielinska-Danch W, Pawlicki K (2004) The influence of smoking on plasma homocysteine and cysteine in passive and active smokers. Clin Chem Lab Med 42:408–414 18. Boddie AM, Steen MT, Sullivan KM, Pasquali M, Dembure PP, Coates RJ, Elsas LJ 2nd (1998) Cystathionine-b-synthase deficiency: Detection of heterozygotes by the ratio of homocysteine to cysteine and folate. Metabolism 47:207–211 19. Herrmann M, Widman T, Colaianni G, Colucci S, Zallone A, Herrmann W (2005) Increased osteoclast activity in the presence of increased homocysteine concentrations. Clin Chem 51:2348–2353 20. Koh J-M, Lee Y-S, Kim Y, Kim D, Kim H-H, Park JY, Lee KV, Kim GS (2006) Homocysteine enhances bone resorption by stimulation of osteoclast formation and activity through increased intracellular ROS generation. J Bone Miner Res 21:1003–1011
