|
|
||||||||
General Clinical Chemistry |
Departments of1 Clinical Chemistry and Laboratory Medicine and 2 Neurology, Faculty of Medicine, University Hospital of Saarland, Homburg/Saar, Germany.
aAddress correspondence to this author at: Department of Clinical Chemistry and Laboratory Medicine, University Hospital of the Saarland, Kirrberger Straße, Gebäude 57, 66421 Homburg, Germany. Fax 49-6841-1630703; e-mail kchwher{at}uniklinikum-saarland.de.
| Abstract |
|---|
|
|
|---|
Methods: We studied blood and cerebrospinal fluid (CSF) samples from 72 patients who underwent lumbar puncture. We measured HCY, methylmalonic acid (MMA), and cystathionine by gas chromatography-mass spectrometry; S-adenosylmethionine (SAM) and S-adenosylhomocysteine (SAH) by liquid chromatography-tandem mass spectrometry; and the B-vitamins by HPLC or immunoassays.
Results: Concentrations were lower in CSF than serum or plasma for HCY (0.09 vs 9.4 µmol/L), SAH (13.2 vs 16.8 nmol/L), cystathionine (54 vs 329 nmol/L), and holotranscobalamin (16 vs 63 pmol/L), whereas concentrations in CSF were higher for MMA (359 vs 186 nmol/L) and SAM (270 vs 113 nmol/L; all P <0.05). CSF concentrations of HCY correlated significantly with CSF folate (r = 0.46), CSF SAH (r = 0.48), CSF-albumin (r = 0.31), and age (r = 0.32). Aging was also associated with lower concentrations of CSF-folate and higher CSF-SAH. The relationship between serum and CSF folate depended on serum folate: the correlation (r) of serum and CSF-folate was 0.69 at serum folate <15.7 nmol/L. CSF concentrations of MMA and holotranscobalamin were not significantly correlated.
Conclusions: CSF and serum/plasma concentrations of vitamin biomarkers are significantly correlated. Older age is associated with higher CSF-HCY and CSF-SAH and lower CSF-folate. These metabolic alterations may be important indicators of low folate status, hyperhomocysteinemia, and neurodegenerative diseases.
| Introduction |
|---|
|
|
|---|
Methionine, the methylation product of HCY, is a major source of S-adenosylmethionine (SAM) in the brain (1). Folate and vitamin B12 are important cofactors for SAM production in the brain (4). Folate and/or cobalamin deficiency can cause increased concentrations of HCY and disturbed methylation status. The importance of the transmethylation pathway in the central nervous system has been outlined (5)(6). SAM is the most important methyl donor in the brain. Disturbed methylation has been implicated in the etiology of psychiatric and neurologic illness (5)(6). Biological methylation by SAM is involved in the integrity and maintenance of myelin, synthesis and inactivation of neurotransmitters, and DNA and RNA synthesis and methylation.
Recent studies demonstrated that increased plasma HCY concentration is a risk factor for several disorders of the central nervous system (7)(8)(9). A causal role for HCY in neuronal damage has been shown by numerous in vitro and in vivo studies. Deficiencies of folate and cobalamin are common (10)(11), especially in patients with neurological and neuropsychiatric disorders (12), suggesting a causal role for B-vitamin deficiency in neuronal damage, either directly or via increasing HCY concentrations. The relationship between HCY and brain function seems stronger in observational (13) than in intervention studies (14)(15)(16), a finding that could be related to the inability of the central nervous system to regenerate.
Little is known about the influence of B-vitamin status on HCY metabolism and the methylation capacity of the brain. We investigated the role of folate and cobalamin as determinants of HCY concentrations in the cerebrospinal fluid (CSF) and the relationship between blood and CSF concentrations of HCY, cystathionine (Cys), holotranscobalamin (holoTC), and methylmalonic acid (MMA).
| Materials and Methods |
|---|
|
|
|---|
Nonfasting blood samples were collected from all patients. Serum and EDTA plasma were available. CSF samples were collected during clinically indicated lumbar punctures. Blood and CSF samples were collected within 24 h of each other. Blood and CSF samples were centrifuged within 30 min of collection, and several aliquots were prepared and stored at 80 °C until analysis. Aliquots of the EDTA-plasma and CSF were immediately deproteinized with perchloric acid (100 g/L). These samples were stored at 80 °C and were used for SAM and S-adenosylhomocysteine (SAH) assays. An aliquot of the CSF was used for determining cell count and protein and glucose concentrations. The study was approved by the Ethics Committee at the Saarland University Hospital, and written informed consent was obtained from all patients.
analytical methods
Concentrations of HCY, Cys, and MMA were measured in serum and CSF samples with gas chromatography mass spectrometry as described elsewhere (17). Day-to-day imprecision (CV) for HCY was <5% in serum (at 8.0 and 16.0 µmol/L) and <10% in CSF (at 0.30 µmol/L). The CV for MMA was <6% in serum and CSF (at 290 and 540 nmol/L, respectively), and for Cys, it was <8% in serum (at 300 nmol/L) and <10% in CSF (at 60 nmol/L). The recovery of the 3 metabolites in CSF was 99% to 107% (see Table 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol53/issue2 ). In-house prepared pool serum and pool CSF were run each time with the study sample and were used to calculate the assay CV. Concentrations of SAM and SAH were measured with a slightly modified liquid chromatography-tandem mass spectrometry method according to Gellekink et al. (18). CVs for SAM and SAH assays were 4.8% and 8%, respectively at 103 nmol/L for SAM and 15.6 nmol/L for SAH.
Concentrations of vitamin B12 and folate were measured with a chemiluminescence immunoassay (ADVIA Centaur System, Bayer), plasma vitamin B6 (pyridoxal-5-phosphate) with HPLC connected to a fluorescence detector (with reagents from Immundiagnostik), and serum and CSF holoTC with RIA (Axis-Shield). The CVs for the holoTC assay were 6% and 8% at 37 and 95 pmol/L, respectively. Serum concentrations of cholesterol and triglycerides were measured by enzymatic colorimetric tests (Roche Diagnostics), and HDL by an enzymatic homogeneous assay (Roche Diagnostics). LDL was calculated from total cholesterol, triglycerides, and HDL according to the Friedewald equation. Serum creatinine concentrations were measured in serum with a kinetic colorimetric assay (Roche Diagnostics), and quantitative glucose was measured by enzymatic ultraviolet test (hexokinase method). Albumin concentration was measured by nephelometry with specific antibodies (DADE Behring). Cell counts and routine variable were measured immediately, and concentrations of B vitamins and metabolites were measured within 6 months of sample collection.
statistical analyses
Data analyses were performed with SPSS (version 12). All variables were skewed and therefore were log-transformed to approach gaussian distribution before application of parametric tests. We used the paired t-test to compare means of the log-transformed variables in blood and CSF and the 1-way ANOVA test for multiple comparisons. The post hoc Tamhane-T test was performed to identify the significantly different group means when the ANOVA test was significant. Correlations between variables were examined by Spearman Rho test. All tests were 2-sided; P values <0.05 were considered statistically significant.
| Results |
|---|
|
|
|---|
119-fold higher than concentrations in the CSF. In contrast, concentrations of MMA were higher in the CSF than in serum (Table 1
|
Concentrations of folate were only slightly higher in the CSF than in serum (Table 1
), and concentrations of holoTC were lower in the CSF than in serum (Table 1
). Total cobalamin and vitamin B6 were not detectable in CSF samples.
CSF markers related to serum concentrations of folate, HCY, and MMA are shown in Table 2
. Serum concentrations of folate
25 nmol/L were associated with lower concentrations of CSF folate (P = 0.002) and higher concentrations of CSF-HCY (P = 0.010). Furthermore, serum concentrations of HCY >10.8 µmol/L were associated with lower concentrations of CSF folate (P = 0.004) and older age (P <0.001). CSF concentrations of HCY tended to increase with increasing serum HCY (Table 2
). Higher concentrations of CSF-HCY and CSF-MMA were found with serum MMA concentrations >217 nmol/L. Neither CSF-SAH nor CSF-SAM concentrations changed with increasing serum folate, serum HCY, or serum MMA (Table 2
).
|
The relationship between serum and CSF folate seemed to be dependent on serum concentrations of folate. Concentrations of CSF folate were higher than those of serum folate in the lowest tertile of serum folate (5.415.6 nmol/L), whereas in the highest tertile of serum folate, CSF folate was lower than serum folate (Fig. 1
). The correlation between serum and CSF folate was strong in the lowest tertile (5.415.6 nmol/L) of serum folate (r = 0.69; P = 0.002). In contrast, no significant correlation was found between serum and CSF folate in the higher range of serum folate.
|
Concentrations of HCY in the CSF correlated to CSF-folate (r = 0.46; P = 0.002; Fig. 2A
), CSF-Cys (r = 0.57; P <0.001; Fig. 2B
), and CSF-albumin (r = 0.31; P = 0.045; Fig. 2C
). CSF concentrations of HCY correlated to those of SAH (r = 0.48 and r = 0.68) after adjusting for age. Furthermore, serum and CSF concentrations of HCY correlated significantly (r = 0.34; P = 0.014). The last correlation was stronger after adjusting for age (r = 0.70; P <0.001), CSF folate (r = 0.62; P <0.001), or serum folate (r = 0.69; P <0.001). No significant association between CSF-HCY and vitamin B12 markers in the CSF (holoTC, MMA) was observed. CSF concentrations of SAH and HCY increased with increasing age (r = 0.48 and r = 0.32, respectively; P <0.001). Other correlations between SAH, SAM, and blood vitamin concentrations were not significant. The most important significant correlations between age and CSF markers and between serum and CSF markers are presented in Table 3
.
|
|
| Discussion |
|---|
|
|
|---|
We observed a weak association between CSF and serum concentrations of HCY (r = 0.34; P = 0.014), and this correlation remained significant after adjustment for age, CSF folate, or serum folate. These results suggest that increased concentrations of HCY in the circulation may lead to increased HCY concentration in the CSF, an effect that could be related to HCY exchange between the circulation and the CSF. In contrast to our current results, a previous study found no correlation between plasma and CSF concentrations of HCY in patients with multiple sclerosis (23). These different findings may be attributable to differences in the underlying diseases of the study groups.
Concentrations of methionine in CSF are low (
34 µmol/L, that is, 1/10 that in the plasma). Therefore, because HCY is produced from methionine, low HCY concentrations may be expected in CSF, but the detected amount of CSF-HCY is 0.96% (mean value) of that in the plasma. Possible explanations for the unexpectedly low CSF HCY values include lower flow of methionine into HCY in the brain compared with the liver or export of excess brain HCY to the plasma. The possibility of HCY exchange between the brain and the circulation is supported by findings from cystathionine ß-synthase-deficient patients (24). In case of excess HCY in the plasma (homocysteinuria), an increased amount of HCY has been detected in CSF (24). Moreover, decreasing plasma HCY by means of betaine caused a reduction in CSF-HCY (24), suggesting that in addition to HCY production within the brain, plasma and CSF HCY may be exchanged via a bidirectional receptor. In our study, in which only 10% of patients had plasma HCY >16.2 µmol/L, CSF-HCY was not related to the integrity of the bloodbrain barrier (i.e., the ratio of CSF albumin:serum albumin; data not shown).
We showed that lower concentrations of serum folate may predict higher concentrations of HCY in CSF (Table 2
). Folate is transported into the brain across the bloodbrain barrier via specific transporters. Concentrations of folate are generally higher in the CSF than in the blood (25). Moreover, because of the active transport of folate by the choroid plexus, CSF concentrations of folate remained stable when plasma folate was >45 nmol/L (26), data that correspond well with our results. Moreover, our data further suggest that CSF folate is relatively conserved and is not subject to strong fluctuations when serum concentrations change. As can be inferred from Fig. 1
, differences between serum folate of
19 nmol/L were associated with only a small difference in CSF folate (
4 nmol/L). These results also indicate that the transport of folate into the brain is subject to a rate-limiting step.
Low folate status is common in elderly people and in neurological and psychogeriatric patients. Folate deficiency has been related to several age-related neurological diseases (27). Low CSF folate can cause disturbances in the metabolism of pteridins and monoamins (28) or severe neurological symptoms(29). In accordance with previous studies, older age was associated with a higher concentration of CSF-HCY and a lower CSF-folate (30)(31), an association attributable to increased plasma concentrations of HCY or reduced concentrations of serum folate with age. Therefore, depletion of CSF folate and increment of CSF-HCY may explain the association between low folate status and hyperhomocysteinemia and neurodegenerative diseases.
Cobalamin deficiency is very common in patients with dementia and those with Alzheimer disease. These results stress the importance of vitamin B12 for brain function. Cobalamin is actively transported into the brain via a specific receptor. Cobalamin is mainly (60%99%) bound to the binding protein transcobalamin in the central nervous system (32). Cultured astrocytes from human brain secrete functionally active transcobalamin that can facilitate the uptake of cobalamin from the circulation (33). Blood cobalamin content seems to be the major determinant of brain and CSF cobalamin (34). We demonstrated that concentrations of holoTC in the CSF correlated to serum holoTC and cobalamin. Therefore, cobalamin deficiency may cause lower cobalamin content in the brain and thereby neuronal damage. In line with these results, CSF concentrations of cobalamin were lower in cobalamin-deficient than nondeficient persons (35).
In contrast to the situation in serum, our data do not support a major role for brain cobalamin as a determinant for HCY concentrations in the CSF, possibly because plasma HCY mostly reflects liver metabolism and can be modified by vitamin B12 status in addition to variations in renal function. We found higher concentrations of holoTC in serum than in CSF (Table 1
), but the holoTC:albumin ratio was higher in CSF than in plasma (65 x 102 vs 1.43 x 102; Table 1
). Compared with other proteins, holoTC seems to be present in higher amounts in the brain than in the circulation. Therefore, higher holoTC concentrations in the brain than in the blood may protect the brain from strong variations in circulating cobalamin.
We have shown that individuals with higher serum concentrations of MMA had higher concentrations of CSF HCY (Table 2
), suggesting that low vitamin B12 status (indicated by higher serum concentration of MMA) can cause neuronal damage at least partly, by increasing HCY in the brain. Because vitamin B12 has a limited role in determining concentrations of MMA in CSF (Table 3
), increased MMA in CSF could be related to increased substrate rather than to low B12 status or methyl malonyl-CoA mutase activity.
Branched chain amino acids, fatty acids, and methionine are important substrates for MMA production. Branched chain amino acids are important precursors for neurotransmitters synthesis in the brain, and branched chain amino acid aminotransferase concentrations are higher in the brain than the liver (36). The brain tends to decrease in size and weight at older age, and the synthesis of neurotransmitters may be reduced. Moreover, there is some selective loss in the number and size of neurons and a reduction in CSF flow (37). These factors might also explain the negative relationship between age and CSF-MMA in our study and a previous one (30).
SAM and SAH concentrations in the CSF may reflect the methylation status of the brain and thus be markers in some neurological diseases. The SAM:SAH ratio in the CSF indicates the methylation status in other brain regions (38). We found that CSF concentrations of SAH increased with increasing age, possibly because of slower removal of this conversion of product. Another possible cause is increased HCY to SAH, which is supported by the strong positive correlation between CSF-HCY and CSF-SAH (r = 0.68; P <0.001 after adjusting for age). Concentrations of CSF-SAM were not related to CSF-SAH or to any vitamin marker tested in this study, with the exception of plasma SAM and SAH. Brain SAM may also be influenced by turnover and exchange with plasma SAM and plasma methionine.
Concentrations of plasma and CSF SAM and SAH in our study were comparable with those from other studies (39)(40). SAM and SAH are unstable, and preanalytical conditions are important for the assessment of these 2 compounds. Moreover, our current results cannot be compared with those of older studies because of variations in the analytical methods and units used.
In summary, our current study demonstrated that age and serum concentrations of HCY and folate are significant determinants of CSF-HCY. A significant role for vitamin B12 status as a determinant of CSF-HCY was reflected by MMA in serum but not in CSF. Our results suggest that improved folate and vitamin B12 status may be associated with higher concentrations of CSF vitamin and lower concentrations of CSF-HCY. Future studies should investigate the clinical implications of these metabolic consequences.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Nonstandard abbreviations: HCY, homocysteine; SAM, S-adenosylmethionine; CSF, cerebrospinal fluid; Cys, cystathionine; holoTC, holotranscobalamin; MMA, methylmalonic acid; SAH, S-adenosylhomocysteine. ![]()
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
C. Wagner and M. J Koury S-Adenosylhomocysteine a better indicator of vascular disease than homocysteine? Am. J. Clinical Nutrition, December 1, 2007; 86(6): 1581 - 1585. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. Obeid, M. Kasoha, J.-P. Knapp, P. Kostopoulos, G. Becker, K. Fassbender, and W. Herrmann Folate and Methylation Status in Relation to Phosphorylated Tau Protein(181P) and {beta}-Amyloid(1-42) in Cerebrospinal Fluid Clin. Chem., June 1, 2007; 53(6): 1129 - 1136. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |