Relationships among biomarkers of one-carbon metabolism

Will D. King • Vikki Ho • Linda Dodds •

Sherry L. Perkins • R. Ian Casson •

Thomas E. Massey

Received: 7 July 2011 / Accepted: 16 April 2012 / Published online: 24 April 2012

� Springer Science+Business Media B.V. 2012

Abstract One-carbon metabolism is a network of meta-

bolic pathways, disruption of which has been associated with

cancer and other pathological conditions. Biomarkers of

these pathways include homocysteine (HCY), S-adenosyl-

methionine (SAM), and S-adenosylhomocysteine (SAH). A

better understanding of the relationships between these

biomarkers is needed for their utilization in research. This

study investigated the relationships between fasting con-

centrations of plasma HCY, SAM, SAH and the ratio of

SAM:SAH, and serum folate, vitamin B12 and creatinine in

a healthy adult population. A cross-sectional study recrui-

ted 678 volunteers; only subjects with complete data

(n = 581) were included in this analysis. Correlations were

used to examine bivariate relationships among the bio-

markers and multivariate linear regression determined

independent relationships with HCY, SAM and SAH

treated as dependent variables in separate models. Multi-

variate logistic regression examined determinants of a low

SAM:SAH ratio (defined as having a SAM:SAH ratio in

the bottom quartile and SAH value in the top quartile).

HCY correlated inversely with folate and vitamin B12 and

weakly correlated with SAH and creatinine. Both SAM and

SAH correlated with creatinine but were independent of

serum folate and vitamin B12. In multivariate analyses,

folate, vitamin B12, creatinine, sex and age were associated

with HCY; age and creatinine were determinants of SAM,

and sex and creatinine determinants of SAH. Finally, male

sex and increasing creatinine levels were associated with

having a low SAM:SAH ratio. Findings suggest that HCY,

SAM and SAH are relatively independent parameters and

reflect distinct aspects of one-carbon metabolism.

Keywords Homocysteine � S-Adenosylmethionine �S-Adenosylhomocysteine � Creatinine � One-carbon

metabolism

Introduction

There is considerable evidence that aberrant DNA methylation

patterns play a critical role in gene regulation; DNA methyl-

ation is an important feature of epigenetics that is increasingly

being implicated in disease states including cancer, cardio-

vascular disease, renal disease, neurodegenerative disease and

W. D. King (&) � V. Ho

Department of Community Health and Epidemiology, Queen’s

University, Carruthers Hall, Kingston, ON K7L 3N6, Canada

e-mail: kingw@queensu.ca

V. Ho

e-mail: vikki.ho@queensu.ca

L. Dodds

Department of Obstetrics and Gynaecology, Perinatal

Epidemiology Research Unit, IWK Health Centre, Halifax,

NS B3K 6R8, Canada

e-mail: L.Dodds@dal.ca

S. L. Perkins

Department of Pathology and Laboratory Medicine, The Ottawa

Hospital and University of Ottawa, Ottawa, ON K1H 8L6,

Canada

e-mail: slperkins@ottawahospital.on.ca

R. I. Casson

Centre for Studies in Primary Care, Queen’s University,

Kingston, ON K7L 5E9, Canada

e-mail: cassoni@hdh.kari.net

T. E. Massey

Pharmacology and Toxicology Graduate Program, Department

of Biomedical and Molecular Sciences, Queen’s University,

Botterell Hall, Kingston, ON K7L 3N6, Canada

e-mail: masseyt@queensu.ca

123

Mol Biol Rep (2012) 39:7805–7812

DOI 10.1007/s11033-012-1623-y

birth defects [1]. DNA methylation patterns are supported by

one-carbon metabolism as methyltransferases utilizes S-

adenosylmethionine (SAM), a metabolite of one-carbon

metabolism, as a methyl group donor [1]. As interest in the role

epigenetics play on the pathogenesis of many disease states

grows, metabolites of one-carbon metabolism including

homocysteine (HCY), SAM and S-adenosylhomocysteine

(SAH) are increasingly being used as predictors, intermediate

effects and outcomes in epidemiologic studies [2–5].

The known effects of SAM and SAH on cellular

methylation suggest that concentrations of SAM and SAH

may be useful measures representing an individual’s DNA

methylation capacity. SAM is a methyl donor for numerous

cellular methyltransferase reactions and SAH is the

demethylated product of SAM with a potent ability to

inhibit DNA methyltransferase reactions; thus, the ratio of

SAM to SAH has been referred to as the ‘‘methylation

index’’ [1, 6]. However, interpretation of the ratio of

SAM:SAH is complex since a low SAM to SAH ratio

could be due either to decreased levels of SAM (limited

substrate availability), increased concentrations of SAH

irrespective of SAM concentrations or their combination

[6–8]. On the other hand, HCY as a biomarker has been

investigated as a risk factor for many pathological condi-

tions, most notably applied in cardiovascular, neurode-

generative and certain types of cancer research. The two

main hypotheses that have been postulated to explain the

role of elevated HCY levels in these conditions include the

generation of reactive oxygen species and aberrant DNA

methylation patterns through an influence of HCY on SAM

and SAH concentrations [8]. Specifically, Gellekink et al.

[6] have shown that DNA hypomethylation is associated

with moderate elevations of HCY and SAH concentrations,

and a decreased SAM to SAH ratio. Due to the close

interrelationships between these biomarkers, it is unclear

which is causally implicated when considering these bio-

markers in relation to the pathogenesis of disease.

Detailed diagrams and descriptions of the biochemical

processes associated with one-carbon metabolism are

available in several sources [2, 9, 10]. Figure 1 depicts a

simplified schematic illustrating the pathways involved in

one-carbon metabolism. It is important to note that one-

carbon metabolism is complex and intersects with other

pathways necessary for the synthesis and utilization of

other significant metabolites such as folic acid, thymidyl-

ate, and cysteine and its derivatives [11, 12].

In one-carbon metabolism, methionine is activated by

adenosine triphosphate (ATP) to produce SAM. Transmeth-

ylation of SAM in which the methyl group of SAM is trans-

ferred to an acceptor results in the formation of SAH. These

reactions include the methylation of small molecules and

macromolecules such as creatine (the precursor of creatinine),

proteins, RNA and DNA [13]. Under physiologic conditions,

SAH is hydrolyzed by S-adenosylhomocysteine hydrolase

(SAHH) to HCY and adenosine in a reversible reaction that

strongly favours SAH synthesis rather than hydrolysis [4, 14].

Diet

TS

Cysteine & Taurine

Vitamin B12

ATP

FOLATE METABOLISM

ONE-CARBON METABOLISM Methylated

acceptor

Acceptor

SAHH

Dihydrofolate

Folic Acid

dUMP

dTMP

Tetrahydrofolate

5,10-Methylene Tetrahydrofolate

5-Methyl Tetrahydrofolatea

(Serum Folate)

Methionine

SAM

SAH

HCY

Vitamin B6

MS

Fig. 1 Simplified schematic of

one-carbon metabolism.

Biomarkers included in this

study appear in bold

(S-adenosylmethionine,

S-adenosylhomocysteine,

homocysteine, serum folate, and

vitamin B12). dUMPdeoxyuridine monophosphate,

TS thymidylate synthase, dTMPdeoxythymidine

monophosphate, MS methionine

synthase, ATP adenosine

triphosphate, SAMS-adenosylmethionine, SAHS-adenosylhomocysteine, SAHHS-adenosylhomocysteine

hydrolase, HCY homocysteine.a 5-methyl-THF may also be

derived from diet

7806 Mol Biol Rep (2012) 39:7805–7812

123

HCY is then metabolized by two biochemical pathways,

transsulfuration and remethylation. In the transsulfuration

pathway, HCY is catabolized to form the amino acids cysteine

and taurine in a series of irreversible reactions [10]. Alterna-

tively, to initiate a new cycle of one-carbon group transfer,

HCY is remethylated back to methionine by methionine

synthase (MS), a ubiquitous enzyme in most tissues, with

vitamin B12 acting as a co-factor and 5-methyltetrahydrofolate

(5-methyl-THF) a substrate. Thymidine synthesis competes

with the one-carbon metabolism cycle for the common sub-

strate 5,10-methylenetetrahydrofolate (5,10-methylene-THF),

a precursor of 5-methyl-THF produced during folate metab-

olism [9]. The reaction converts deoxyuridine monophosphate

(dUMP) to deoxythymidine monophosphate (dTMP) and is

catalyzed by thymidylate synthase (TS). The complexity of

the involved pathways clearly illustrates the importance and

the goal of metabolic regulation of one-carbon metabolism—

to balance and appropriately distribute metabolites among

these pathways while preventing the accumulation of the

involved intermediates (e.g., HCY, SAH and SAM) [11, 12].

As illustrated, HCY, SAM and SAH are essential

intermediates in one-carbon metabolism. However, a better

understanding of how these biomarkers are interrelated is

necessary for effective utilization in research and to

interpret findings. This study investigated the relationships

between HCY, SAM, SAH, the ratio of SAM to SAH,

folate, vitamin B12 and creatinine in a healthy middle-aged

Canadian population.

Methods

Study population

The data for this analysis were collected in a study focusing

on environmental exposures and markers of DNA meth-

ylation capacity. From 2006 to 2009, a cross-sectional

study recruited 678 healthy volunteer subjects, age 20–50,

from three study centers situated in Kingston, Ontario,

Ottawa, Ontario and Halifax, Nova Scotia, Canada. These

study centers were selected specifically to provide variation

in the environmental factors of interest; disinfection by-

products in public water supplies. Those with health con-

ditions that might be related to the biomarkers of interest

(e.g., history of coronary artery disease or other vascular

disease, or diabetes), or who had been pregnant within the

past year were not eligible. Eligible subjects consenting to

participate in this study provided 12 h fasting blood sam-

ples, completed a short questionnaire, and returned a take

home water sampling kit. Subjects with non-fasting blood

samples (n = 80) and incomplete questionnaire and bio-

chemical data (n = 17) were not included in the analysis,

resulting in a sample size of 581.

Laboratory analyses

Blood sampling, processing and aliquoting procedures were

standardized between the three study centers; a total of

15 mL of venous blood were drawn from each participant.

Specifically, concentrations of plasma SAM, SAH and HCY

were determined from 10 mL of blood drawn into tubes with

ethylenediaminetetraacetic acid (EDTA) and 5 mL of blood

was obtained in a serum separator tube (SST) for measure-

ments of serum folate, vitamin B12 and creatinine concen-

trations. After withdrawal, the EDTA tube was kept on ice

and centrifuged at 4 �C within 1 h of collection; 6 mL of

plasma was then extracted and stored in two separate aliquot

tubes for separate analysis of HCY, and SAM and SAH. For

the SST tube, 2 mL of serum was extracted after the speci-

men was allowed to clot and centrifuged at 4 �C within 1 h

of collection. Serum and plasma samples were kept frozen at

-80 �C prior to analysis; under these storage conditions,

levels of SAM, SAH and HCY have been reported to remain

stable for up to 4 years [14, 15]. Plasma HCY and serum

folate, vitamin B12 and creatinine were measured at the

Ottawa Hospital, Department of Pathology and Laboratory

Medicine, Division of Biochemistry, Ottawa, Canada.

Plasma SAM and SAH were measured at the Saskatchewan

Disease Control Laboratory, Saskatchewan, Canada. All

serum and plasma samples were shipped on dry ice.

Plasma HCY concentrations were quantified by the

Abbott AxSYM immunoassay based on fluorescence polar-

ization immunoassay technology (Abbott AxSYM, Abbott

Laboratories, Abbott Park, Illinois, USA). This assay has an

analytical range of 1–50 lmol/L and coefficients of variation

(CV) of 5.5 % at 7.4 lmol/L, 6.2 % at 13.5 lmol/L, and

5.4 % at 25.9 lmol/L. Plasma SAM and SAH were mea-

sured using a tandem mass spectrometry method developed

by Struys et al. [16] and Gellekink et al. [6]. Modification to

the published methods included the use of deuterium labelled

SAM (2H3-SAM) as an internal standard for both SAM and

SAH analyses; otherwise testing conditions of the assay were

conducted as described. Briefly, SAM and SAH were

obtained from Sigma (St. Louis, MO, USA) and 2H3-SAM

from CDN Isotopes (Pointe-Claire, Quebec, Canada). All

analyses were performed on an API 4000 triple quadrupole

tandem mass spectrometer and binary HPLC pump (Agilent

1200 series). This assay has CV of 7 and 15 % for SAM and

SAH, respectively. Recoveries for both SAM and SAH were

comparable to percentages published by Gellekink et al. [6]

(mean recoveries for SAM and SAH[90 %). Determination

of serum folate and vitamin B12 status was accomplished

using a two-step competitive immuno-enzymatic assay with

chemiluminescence detection on a Beckman Coulter DMXi

immunoassay analyzer (Beckman Coulter, Inc., Fullerton,

California, USA). This assay has analytical linearity of

1–45 nmol/L and a CV of 9.4 % at 3.4 nmol/L, 5.0 % at

Mol Biol Rep (2012) 39:7805–7812 7807

123

9.3 nmol/L, and 5.6 % at 22.0 nmol/L. Extreme values of

serum folate (n = 6) were truncated at three standard devi-

ations (SD) above the mean (mean serum folate ? 3

SD = 84.5 nmol/L). Finally, serum creatinine was deter-

mined colourimetrically by the Jaffe rate method on the

Beckman Coulter LX20PRO (Beckman Coulter, Inc., Ful-

lerton, California, USA) [17]. This assay has CV of 6 % at

60 lmol/L and 2.0 % or less at 180 and 610 lmol/L.

Statistical analyses

Distributions are presented according to quartiles, means

and SD. Pearson correlation coefficients were calculated to

evaluate bivariate relationships. Sensitivity analyses were

conducted to test the effect of non-normality on the Pear-

son correlation coefficient by comparing values obtained

through Pearson correlation to Spearman correlation.

Multivariate linear regression was used to determine

independent relationships with HCY, SAM and SAH

treated as dependent variables in separate models. In each

model the covariates included age, sex, serum folate, serum

vitamin B12 and serum creatinine.

In the multivariate linear regression analysis, all of the

variables were standardized (i.e., the difference between

each individual measurement and the overall mean was

divided by the SD). For serum creatinine, standardization

was conducted according to sex since serum creatinine

concentrations vary between males and females (e.g., the

difference between each individual female measurement of

serum creatinine and the overall female mean was divided

by the female SD). As a result all standardized biomarkers

have a mean of zero and SD of 1 and coefficients in the

regression model (e.g., slope [b] of the regression line) are

interpretable as the proportion of a SD change in the

dependent variable for a 1 SD change in the independent

variable. A variable representing the three study centers

was included as a random effects parameter to account for

the cluster sampling design.

The ratio of SAM to SAH has been proposed as a

methylation index; however, reports have indicated that a

decrease in the SAM:SAH ratio was only predictive of

reduced methylation capacity in the presence of high SAH

levels [4, 14]. For this study, a dichotomous variable rep-

resenting a low SAM:SAH ratio was defined as having a

SAM:SAH ratio in the bottom quartile of values and a SAH

value in the top quartile of values. Multivariate logistic

regression was used to examine relationships between the

dichotomous outcome of a low SAM:SAH ratio and stan-

dardized HCY, folate, vitamin B12 and creatinine; age and

sex were included in the regression model as covariates.

The cluster sampling design was accounted for by includ-

ing study center as a random effects parameter in the

logistic model.

Results

This analysis included 581 subjects with complete data on

all of the variables of interest. Subject recruitment was

designed to result in approximately balanced distribution

by sex and age. The final study had a greater proportion of

females (55 %) and a relatively balanced age distribution.

The distribution (quartiles), mean and SD of the bio-

markers under study are presented in Table 1.

Pearson correlation coefficients for the six biomarkers

are presented in Table 2. Plasma HCY did not correlate

with SAM, but had a positive correlation with SAH and

serum creatinine, and negative correlations with serum

folate and vitamin B12. SAM and SAH were found to only

correlate with serum creatinine. Folate and vitamin B12 had

a positive correlation. All of the markers have an approx-

imately normal distribution with a slight right skew and

Shapiro–Wilk tests for non-normal distribution were sta-

tistically significant (p value \ 0.05). However, violation

of the normality assumption tends to have little effect on

Pearson correlation coefficients given a sufficiently large

sample size [18]. In a sensitivity analysis (not presented)

Spearman correlation coefficients were found to be similar

to Pearson correlation coefficients presented.

The results of multivariate linear regression models with

standardized HCY, SAM and SAH treated as dependent

variables are presented in Table 3. Each coefficient is

interpretable as the proportion of a SD change of the

dependent variables for a one unit change in the indepen-

dent variable. Higher serum folate and vitamin B12 were

associated with decreased HCY levels and HCY increased

with serum creatinine, male sex and age. HCY was not

related to SAM or SAH. For the model using SAM as the

dependent variable, age and serum creatinine were the only

statistically significant predictors. With SAH as the

Table 1 Distribution of concentrations of HCY, SAM, SAH in

plasma, and folate, vitamin B12 and creatinine in serum (n = 581)

25th

percentile

Median 75th

percentile

Mean SD

HCY (lmol/L) 7.0 8.4 10.0 8.8 2.4

SAM (nmol/L) 54.7 62.2 73.2 64.0 13.6

SAH (nmol/L) 20.0 24.4 29.5 25.5 10.6

Folate (nmol/L) 21.8 28.2 35.5 29.4 11.1

Vitamin B12

(qmol/L)

161.0 213.0 284.0 234.9 110.0

Creatinine—

females

(lmol/L)

66.0 72.0 78.0 72.3 9.0

Creatinine—males

(lmol/L)

84.0 91.0 98.0 91.1 12.0

7808 Mol Biol Rep (2012) 39:7805–7812

123

dependent variable, SAH levels increased with serum

creatinine and were higher in males than in females.

A low SAM:SAH ratio was identified in 16.8 % of

subjects (e.g., a low SAM:SAH ratio in the presence of

high SAH). Table 4 presents odds ratios (OR) and 95 %

confidence intervals (95 % CI) for relationships with this

outcome. Higher serum creatinine and male sex were the

only factors associated with having a low SAM:SAH ratio.

Discussion

This study investigated the relationships between key

intermediates in one-carbon metabolism measured in fast-

ing blood samples obtained from 581 healthy adults. Our

mean SAM and SAH concentrations were found to be

similar to one of three studies conducted in fasting samples

of healthy subjects within a comparable age range [7].

Specifically, Loehrer et al. [7] reported mean SAM and

SAH levels that were similar to those observed in this

study. Tao et al. [19] and Valli et al. [20] both reported

higher mean SAM and SAH concentrations. Two addi-

tional studies that measured SAM and SAH levels in non-

fasting participants within a comparable age range both

reported higher mean SAM and lower SAH concentrations

[6, 21]. Reference values for plasma HCY measured in

healthy individuals are readily available in the literature;

our mean HCY was found to be consistent with those

reported in the literature [15].

The relationships between key intermediates in one-

carbon metabolism were investigated through bivariate and

multivariate analyses which address two different dimen-

sions of these relationships. Bivariate analysis gives

information on the extent to which each biomarker may

represent different aspects of one-carbon metabolism,

while multivariate analysis provides further insight into

underlying relationships among this set of biomarkers.

This discussion will focus on HCY, SAM and SAH.

With respect to bivariate analysis the main findings of note

were that: (1) HCY had only a weak positive association

with SAH and creatinine, was not associated with SAM,

and had strong negative correlations with folate and vita-

min B12 and; (2) SAM and SAH correlated with creatinine

but were independent of folate and vitamin B12. Few

studies have investigated the inter-relationships between

HCY, SAM and SAH in a healthy population sample; only

four studies have been identified [8, 22–24]. Becker et al.

[22] reported bivariate analysis of HCY, SAM and SAH

conducted among an older population (n = 97; aged

60–65); Yi et al. [23] investigated relationships with HCY,

SAM and SAH in a healthy female population (n = 58;

aged 19–53); Loehrer et al. [24] examined the correlation

between HCY, SAM and SAH as a sub-analysis conducted

among a healthy control group (n = 50; aged 41–87); and

Table 2 Pearson correlation coefficients between concentrations of

HCY, SAM, SAH in plasma, and folate and vitamin B12 in serum

HCY SAM SAH Folate Vitamin B12

SAM 0.00

SAH 0.09* 0.07

Folate -0.28* -0.03 -0.06

Vitamin B12 -0.18* -0.06 -0.01 0.20*

Creatinine 0.11* 0.10* 0.12* 0.00 -0.03

* p value \ 0.05

Table 3 Mulitvariate linear

regression analysis with

standardized HCY, SAM or

SAH as the dependent variable

Variable HCY SAM SAH

Coefficient p value Coefficient p value Coefficient p value

HCY (per 1 SD) -0.01 0.88 0.03 0.54

SAM (per 1 SD) -0.01 0.88 0.06 0.19

SAH (per 1 SD) 0.02 0.54 0.05 0.19

Folate (per 1 SD) -0.25 \0.01 0.02 0.71 -0.03 0.48

Vitamin B12 (per 1 SD) -0.16 \0.01 -0.03 0.53 0.00 0.98

Creatinine (per 1 SD) 0.13 \0.01 0.09 0.02 0.10 0.01

Sex (male vs female) 0.57 \0.01 -0.11 0.19 0.27 \0.01

Age (per 10 years) 0.11 \0.01 0.13 \ 0.01 0.04 0.43

Table 4 Multivariate OR and 95 % CI for a low SAM:SAH ratio

Variable OR (95 % CI)

HCY (per 1 SD) 1.20 (0.96–1.49)

Folate (per 1 SD) 1.01 (0.80–1.27)

B12 (per 1 SD) 1.05 (0.84–1.30)

Creatinine (per 1 SD) 1.26 (1.08–1.47)

Sex (male vs female) 2.10 (1.26–3.51)

Age (per 10 years) 0.92 (0.72–1.19)

A low SAM:SAH ratio was defined as the bottom 25th percentile of

SAM:SAH and top 25th percentile of SAH (16.8 % of subjects)

Mol Biol Rep (2012) 39:7805–7812 7809

123

Stabler and Allen [8] conducted bivariate analysis among a

healthy group of volunteers (n = 48; mean age 38.5). The

relationships between HCY, SAM and SAH reported in

these studies are inconsistent. Specifically, Becker et al.

[22] reported no correlations between HCY and SAH, and

HCY and SAM but a strong positive correlation between

SAM and SAH (r = 0.68; p \ 0.001). Yi et al. [23]

reported a stronger positive correlation between HCY and

SAH (r = 0.73; p \ 0.001) than the correlation observed

in our study but supported our finding of no association

between HCY and SAM; the bivariate relationship between

SAM and SAH was not reported. In bivariate analysis,

Loehrer et al. [24] reported no relationship between HCY,

SAM and SAH. Finally, Stabler and Allen [8] found a

moderate correlation between HCY and SAH (r = 0.29;

p = 0.047), no correlation between HCY and SAM and a

strong correlation between SAM and SAH (r = 0.50;

p \ 0.001).

Our finding of a positive correlation between HCY and

SAH was not unexpected since the hydrolysis of SAH to

HCY is a reversible reaction that favours the synthesis of

SAH; thus, any accumulation of HCY would be expected

to drive the reversal of the SAHH reaction and increase

SAH concentration [4]. However, evidence supports a

more facile transport of HCY across the plasma membrane

as compared to SAH. Therefore, concentration of plasma

HCY is almost threefold greater than SAH and correlation

between plasma levels of SAH and HCY is likely affected

by the dynamic balance between tissue and plasma con-

centrations [4]. The strong negative correlations between

HCY and folate, and HCY and vitamin B12 are consistent

with a large body of research, as both folate and vitamin

B12 play critical roles in the metabolism of HCY [24, 25].

Likewise, among a healthy population, the correlation

between creatinine and HCY has been consistently repor-

ted in the literature [26–33]. The lack of correlation found

between SAM and SAH in our study was unexpected, but

consistent with the finding reported by Loehrer et al. [24].

Finally, the correlations between creatinine, and SAM and

SAH in a healthy population have been investigated by

Stabler and Allen [8], serum creatinine was found to pos-

itively correlate with serum SAH but not SAM levels.

In multivariate analysis, the main findings of interest were

that: (1) there were no associations between HCY and SAM

and SAH levels; (2) lower serum folate and vitamin B12

levels, higher serum creatinine level, male sex and increasing

age were associated with higher HCY levels; (3) higher levels

of SAM were associated with increasing serum creatinine and

age and; (4) higher SAH levels were observed with increasing

serum creatinine levels and in males. Two studies that have

investigated similar determinants of HCY, SAM and SAH

were identified. In a study conducted by van Driel et al. [34]

among 336 healthy women (median age 32.7), only serum

folate and vitamin B12 were found as determinants of HCY

while serum creatinine and its associations with HCY, SAM

and SAH were not considered. Becker et al. [22] reported

similar findings where serum folate, serum creatinine and sex

were found to be determinants of HCY, and serum creatinine

and sex were determinants of SAH; no significant determi-

nants of SAM were found.

As SAM is the universal methyl donor for methylation

reactions and SAH is the product and inhibitor of meth-

yltransferase reactions, the ratio of SAM:SAH has been

suggested as a indicator of methylation capacity. However,

it has been reported that a decrease in SAM:SAH ratio was

predictive of methylation capacity only in the presence of a

high SAH value [4]. We defined a low SAM:SAH ratio as

those in the lowest quartile of SAM:SAH ratio and the

highest quartile of SAH. In this study, increasing serum

creatinine and male sex were found to be associated with a

low SAM:SAH ratio. Becker et al. [22] examined SAM:-

SAH as a continuous variable and also observed a rela-

tionship with serum creatinine and sex.

Limitations

The study population was aged 20–50, primarily Caucasian

and exposed to folic acid food fortification since 1998;

therefore generalizability to other age groups, ethnic pop-

ulations and unfortified populations may be limited. The

study was cross-sectional and included a one-time blood

draw for measurement of the relevant biomarkers. The

stability of these measures is therefore critical for mean-

ingful interpretation of the study results. HCY, SAM and

SAH concentrations in an individual are relatively constant

over at least 1 month [3, 35]. The CV for HCY, SAM, SAH,

folate, vitamin B12 and creatinine laboratory analyses

indicate a potential for random misclassification. Caution

was exercised to ensure that the samples were prepared in a

uniform manner between and within the three study centers.

In particular, prevention of any room temperature storage or

incubation at any time from phlebotomy to analysis was

strictly enforced as SAM is unstable and partially degrades

to SAH after relatively short intervals of room temperature

or 4 �C incubation [6, 16]. This degradation would have a

particular implication on the low SAM:SAH ratio variable.

Prior to analysis, all samples were stored at -80 �C and

shipped on dry ice to minimize degradation of SAM [14].

However, the potential for sample thawing in transit is a

possibility. Alternatively, stabilizing agents could have

been added for preservation; for example, plasma samples

are recommended to be acidified with acetic acid prior to

storage for SAM and SAH analysis and EDTA and ascorbic

acid are stabilizing agents for folate in blood [6, 36–38].

The use of a deuterium labelled SAH rather than SAM as an

internal standard for SAH analyses would have been ideal;

7810 Mol Biol Rep (2012) 39:7805–7812

123

however, at the time the samples were analyzed, deuterium

labelled SAH was not available. It is noteworthy that the

impact of the use of SAM as an internal standard for SAH

quantification may have biased the estimate of mean SAH

levels in our population, but is less likely to have biased

relative measures of SAH and therefore the analyses of the

interrelationships between SAM, SAH, HCY and relevant

dietary factors. Furthermore, our measure of folate status is

serum folate rather than red blood cell concentration. This

represents a potential limitation since serum folate tends to

represent short term folate and red blood cell concentrations

may be a better reflection of folate status over a longer

period of time [39]. It is also important to emphasize that

these biomarkers were measured in blood and that the

relationship between blood levels and critical target organs

has not been established and is likely complex; for instance,

the correlation between plasma concentrations of HCY and

SAH is likely influenced by tissue concentrations [4].

Finally, this study did not investigate whether HCY, SAM,

SAH or the ratio of SAM:SAH are valid markers of DNA

methylation and in the literature, it is not fully clarified

which of the moieties is most relevant to methylation.

Conclusion

This is the largest study to investigate the interrelationships

between markers of one-carbon metabolism among healthy

adults aged 20–50. The lack of strong correlations between

plasma HCY, SAM and SAH concentrations suggest that

HCY, SAM and SAH are relatively independent parame-

ters and therefore reflect different aspects of one-carbon

metabolism. However, it is noteworthy that serum creati-

nine levels correlated with those of plasma HCY, SAM and

HCY concentrations. These results do not support the use

of any single marker and illustrate the necessity to capture

each of these markers in risk factor research.

Acknowledgements The authors would like to thank the study

participants and study coordinator Gwyneth Fairfield. This research

was supported by an operating grant from the Canadian Institutes of

Health Research.

Conflict of interest No potential conflicts of interest to disclose.

References

1. Castro R, Rivera I, Struys EA, Jansen EE, Ravasco P, Camilo ME

et al (2003) Increased homocysteine and S-adenosylhomocysteine

concentrations and DNA hypomethylation in vascular disease.

Clin Chem 49(8):1292–1296

2. Mason JB (2003) Biomarkers of nutrient exposure and status in

one-carbon (methyl) metabolism. J Nutr 133(Suppl 3):941S–

947S

3. Poirier LA, Wise CK, Delongchamp RR, Sinha R (2001) Blood

determinations of S-adenosylmethionine, S-adenosylhomocys-

teine, and homocysteine: correlations with diet. Cancer Epi-

demiol Biomarkers Prev 10(6):649–655

4. James SJ, Melnyk S, Pogribna M, Pogribny IP, Caudill MA

(2002) Elevation in S-adenosylhomocysteine and DNA hy-

pomethylation: potential epigenetic mechanism for homocyste-

ine-related pathology. J Nutr 132(8 Suppl):2361S–2366S

5. Hobbs CA, Cleves MA, Melnyk S, Zhao W, James SJ (2005)

Congenital heart defects and abnormal maternal biomarkers of

methionine and homocysteine metabolism. Am J Clin Nutr

81(1):147–153

6. Gellekink H, van Oppenraaij-Emmerzaal D, van Rooij A, Struys

EA, den Heijer M, Blom HJ (2005) Stable-isotope dilution liquid

chromatography-electrospray injection tandem mass spectrome-

try method for fast, selective measurement of S-adenosylmethi-

onine and S-adenosylhomocysteine in plasma. Clin Chem

51(8):1487–1492. doi:10.1373/clinchem.2004.046995

7. Loehrer FM, Angst CP, Brunner FP, Haefeli WE, Fowler B

(1998) Evidence for disturbed S-adenosylmethionine : S-adeno-

sylhomocysteine ratio in patients with end-stage renal failure: a

cause for disturbed methylation reactions? Nephrol Dial Transpl

13(3):656–661

8. Stabler SP, Allen RH (2004) Quantification of serum and urinary

S-adenosylmethionine and S-adenosylhomocysteine by stable-

isotope-dilution liquid chromatography–mass spectrometry. Clin

Chem 50(2):365–372. doi:10.1373/clinchem.2003.026252

9. Jacobsen DW (2000) Biochemistry and metabolism. In: Robinson

K (ed) Homocysteine and vascular disease. Kluwer, Dordrecht,

pp 59–84

10. Selhub J (1999) Homocysteine metabolism. Annu Rev Nutr

19:217–246. doi:10.1146/annurev.nutr.19.1.217

11. Finkelstein JD (2001) Regulation of homocysteine metabolism.

In: Carmel R, Jacobsen DW (eds) Homocysteine in health and

disease. Cambridge University Press, Cambridge, pp 92–99

12. Alonso-Aperte E, Gonzalez MP, Poo-Prieto R, Varela-Moreiras

G (2008) Folate status and S-adenosylmethionine/S-adenosylho-

mocysteine ratio in colorectal adenocarcinoma in humans. Eur J

Clin Nutr 62(2):295–298. doi:10.1038/sj.ejcn.1602722

13. Kerins DM, Koury MJ, Capdevila A, Rana S, Wagner C (2001)

Plasma S-adenosylhomocysteine is a more sensitive indicator of

cardiovascular disease than plasma homocysteine. Am J Clin

Nutr 74(6):723–729

14. Wagner C, Koury MJ (2007) S-Adenosylhomocysteine: a better

indicator of vascular disease than homocysteine? Am J Clin Nutr

86(6):1581–1585

15. Blau N, Duran M, Gibson KM (2008) Laboratory guide to the

methods in biochemical genetics. Springer, Heidelberg,

pp 91–114

16. Struys EA, Jansen EE, de Meer K, Jakobs C (2000) Determina-

tion of S-adenosylmethionine and S-adenosylhomocysteine in

plasma and cerebrospinal fluid by stable-isotope dilution tandem

mass spectrometry. Clin Chem 46(10):1650–1656

17. Whelton A, Watson AJ, Rock RC (1994) Nitrogen metabolites

and renal function. In: Burtis CA, Ashwood ER (eds) Tietz

textbook of clinical chemistry, 2nd edn. WB Saunders, Phila-

delphia, pp 1513–1575

18. Zeller RA, Levin ZH (1974) The effects of violating the nor-

mality assumption underlying r. Sociol Method Res 2(511):

511–518. doi:10.1177/004912417400200406

19. Tao J, Yang M, Chen Z, Huang Y, Zhao Q, Xu J et al (2008)

Decreased DNA methyltransferase 3A and 3B mRNA expression

in peripheral blood mononuclear cells and increased plasma SAH

concentration in adult patients with idiopathic thrombocytopenic

purpura. J Clin Immunol 28(5):432–439. doi:10.1007/s10875-

008-9223-2

Mol Biol Rep (2012) 39:7805–7812 7811

123

20. Valli A, Carrero JJ, Qureshi AR, Garibotto G, Barany P, Axels-

son J et al (2008) Elevated serum levels of S-adenosylhomocys-

teine, but not homocysteine, are associated with cardiovascular

disease in stage 5 chronic kidney disease patients. Clin Chim

Acta 395(1–2):106–110. doi:10.1016/j.cca.2008.05.018

21. Kirsch SH, Knapp JP, Geisel J, Herrmann W, Obeid R (2009)

Simultaneous quantification of S-adenosyl methionine and

S-adenosyl homocysteine in human plasma by stable-isotope

dilution ultra performance liquid chromatography tandem mass

spectrometry. J Chromatogr B 877(30):3865–3870. doi:10.1016/

j.jchromb.2009.09.039

22. Becker A, Smulders YM, Teerlink T, Struys EA, de Meer K,

Kostense PJ et al (2003) S-Adenosylhomocysteine and the ratio

of S-adenosylmethionine to S-adenosylhomocysteine are not

related to folate, cobalamin and vitamin B6 concentrations. Eur J

Clin Invest 33(1):17–25

23. Yi P, Melnyk S, Pogribna M, Pogribny IP, Hine RJ, James SJ

(2000) Increase in plasma homocysteine associated with parallel

increases in plasma S-adenosylhomocysteine and lymphocyte

DNA hypomethylation. J Biol Chem 275(38):29318–29323. doi:

10.1074/jbc.M002725200

24. Loehrer FM, Tschopl M, Angst CP, Litynski P, Jager K, Fowler B

et al (2001) Disturbed ratio of erythrocyte and plasma S-adeno-

sylmethionine/S-adenosylhomocysteine in peripheral arterial

occlusive disease. Atherosclerosis 154(1):147–154

25. Refsum H, Nurk E, Smith AD, Ueland PM, Gjesdal CG, Bjelland

I et al (2006) The Hordaland homocysteine study: a community-

based study of homocysteine, its determinants, and associations

with disease. J Nutr 136(6 Suppl):1731S–1740S

26. Ueland PM, Refsum H, Schneede J (2000) Determinants of

plasma homocysteine. In: Robinson K (ed) Homocysteine and

vascular disease. Kluwer, Dordrecht, pp 59–84

27. Vrentzos GE, Papadakis JA, Malliaraki N, Bampalis DE, Repa A,

Lemonomichelaki V, Petinellis EG, Ganotakis ES (2006) Serum

homocysteine concentration as a marker of nutritional status of

healthy subjects in Crete, Greece. J Hum Nutr Dietet 19:117–123

28. Krivosıkova Z, Krajcovicova-Kudlackova M, Spustova V,

Stefıkova K, Valachovicova M, Blazıcek P, Nemcova T (2010)

The association between high plasma homocysteine levels and

lower bone mineral density in Slovak women: the impact of

vegetarian diet. Eur J Nutr 49:147–153. doi:10.1007/s00394-

009-0059-1

29. Rauh M, Verwied S, Knerr I, Dorr HG, Sonnichsen A, Koletzko

B (2001) Homocysteine concentrations in a German cohort of

500 individuals: reference ranges and determinants of plasma

levels in healthy children and their parents. Amino Acids 20:

409–418. doi:10.1007/s007260170037

30. Bjorkegren K, Svardsudd K (2001) Serum cobalamin, folate,

methylmalonic acid and total homocysteine as vitamin B12 and

folate tissue deficiency markers amongst elderly Swedes—a

population-based study. J Intern Med 249:423–432. doi:10.1046/

j.1365-2796.2001.00819.x

31. Herrmann W, Quast S, Ullrich M, Schultze H, Bodis M, Geisel J

(1999) Hyperhomocysteinemia in high-aged subjects: relation of

B-vitamins, folic acid, renal function and the methylenetetrahy-

drofolate reductase mutation. Atherosclerosis 144(1):91–101.

doi:10.1016/S0021-9150(99)00036-2

32. Marouf R, Mohammad Z, Olusegun MA, Mohammad Q, Nabila

AA, Hanan WA, Salah HA (2006) Determinants of plasma

homocysteine in relation to hematological and biochemical

variables in patients with acute myocardial infarction. South Med

J 99(8):811–816. doi:10.1097/01.smj.0000231245.44439.47

33. Jacques PF, Rosenberg IH, Rogers G, Selhub J, Bowman BA,

Gunter EW, Wright JD, Johnson CL (1999) Serum total homo-

cysteine concentrations in adolescent and adult Americans:

results from the third National Health and Nutrition Examination

Survey. Am J Clin Nutr 69(3):482–489

34. van Driel LM, Eijkemans MJ, de Jonge R, de Vries JH, van

Meurs JB, Steegers EA et al (2009) Body mass index is an

important determinant of methylation biomarkers in women of

reproductive ages. J Nutr 139(12):2315–2321. doi:10.3945/jn.

109.109710

35. Garg UC, Zheng ZJ, Folsom AR, Moyer YS, Tsai MY,

McGovern P et al (1997) Short-term and long-term variability of

plasma homocysteine measurement. Clin Chem 43(1):141–145

36. Elloitt P, Peakman TC (2008) The UK Biobank sample handling

and storage protocol for the collection, processing and archiving

of human blood and urine. Int J Epidemiol 37(2):234–244. doi:

10.1093/ije/dym276

37. Holland NT, Smith MT, Eskenazi B, Bastaki M (2003) Biological

sample collection and processing for molecular epidemiological

studies. Mutat Res 543(3):217–234. doi:10.1016/S1383-5742(02)

00090-X

38. Kerkay J, Coburn CM, McEvoy D (1977) Effect of sodium

ascorbate concentration on the stability of samples for determi-

nation of serum folate levels. Am J Clin Pathol 68:481–484

39. Herbet V (1998) Folic acid. In: Shils M, Olson JA, Shike M, Ross

AC (eds) Modern nutrition in health and disease, 9th edn. Lip-

pincott Williams & Wilkins, Baltimore, pp 443–446

7812 Mol Biol Rep (2012) 39:7805–7812

123