A Simple Rapid Immunoassay for S-adenosylhomocysteine In Plasma

Abstract

The measurement of plasma S-adenosylhomocysteine is a more sensitive indicator of the risk for vascular disease than is plasma homocysteine. Because the level of S-adenosylhomocysteine is normally in the nanomolar range it has been difficult to measure and necessitated the development of complex fluorometric and mass-spectrophotometric methods. We have now adapted an existing immunoassay used for the measurement of homocysteine to the measurement of S-adenosylhomocysteine in plasma. This assay is sensitive down to the level of less than 0.1 picomole and there is no interference by S-adenosylmethionine. The assay is carried out in microplates, allows the measurement of 12 samples per plate and can easily be carried out in a 4 hour period. The method is applicable to plasma samples having S-adenosylhomocysteine concentrations ranging from 10 to 150 nM without dilution. . The mean value for 16 normal subjects by this method was 18.9 +/− 1.4 nM (S.E.M.) compared with 17.8 +/− 1.4 nM obtained by a previously described method using two HPLC columns with fluorescence derivatization. Mean values for 7 cirrhotic patients were 46.5 +/− 3.3 nM by this new method compared with 44.6 +/− 5.3 by the former method. The ease and speed of this method should allow the widespread measurement of this important metabolite in laboratories without access to sophisticated equipment.

Introduction

S-adenosylhomocysteine (SAH) is the product of most methyltransferase reactions that are carried out in living organisms with S-adenosylmethionine (SAM) as the methyl donor. It is the metabolic precursor of homocysteine (Hcy) which has been identified as a risk factor for vascular disease in many studies [1] and has also been suggested to be a risk factor in a number of neuropsychiatric disorders [2] [3] [4]. The measurement of plasma Hcy is relatively straightforward since the normal level in human plasma is in the μM range and a number of methods have been published and several commercial kits are available for its measurement. On the other hand, plasma SAH is in the nM range and the published methods for its measurement are complicated usually requiring specialized equipment. A method we had developed several years ago for the measurement of SAH in plasma employs two HPLC steps and derivatization with naphthalene dicarboxaldehyde to produce a fluorescent isoindole [5]. Although time consuming, we have used this method successfully to show that plasma SAH is a more sensitive indicator of the risk for cardiovascular disease [6] and for renal disease than is plasma Hcy [7]. More recently, we have shown that in children with renal disease, uncomplicated by other risk factors, plasma SAH is highly correlated with the glomerular filtration rate while plasma Hcy is not [8]. For this reason, a simple, rapid method for the measurement of plasma SAH would permit its use to determine whether elevated SAH is a risk factor in other conditions that may be involved in vascular disease.

Because plasma SAH is in the nM range a method of high sensitivity as well as specificity is needed. A number of methods for the measurement of SAH in plasma or serum have been developed. For the most part, they are cumbersome or require highly specialized equipment. Loehrer et al. adapted a method utilizing conversion of the adenosine ring to a fluorescent derivative that was then subjected to HPLC chromatography. This required an 8 hour period for the derivitization [9]. Castro et al. [10]were able to shorten the reaction time to 4 hours. A number of other methods used tandem mass-spectrometry [11] [12] [13]. A competitive immunoassay for total plasma Hcy was developed by Frantzen et al. [14]. Plasma Hcy was first converted to SAH using the enzyme, S-adenosylhomocysteine hydrolase, and was then quantitated with anti-SAH antibody and adapted for microtitre plates. We eliminated the first part of this procedure involving the enzymatic conversion of Hcy to SAH and then modified the conditions of the procedure so that the range of the method runs from 0.05 to 0.5 pmoles of SAH in the assay. Although this is a narrow range, different amounts of sample can be used and those with higher concentrations can be diluted. At this level of sensitivity SAM does not interfere. Using this method we obtained a mean value of plasma SAH of 18.9 +/− 1.4 nM for 16 normal individuals. This is compared with a value of 17.8 +/− 1.4 nM for the same samples analyzed using our fluorescence method. When plasma samples from 7 patients with liver disease were analyzed using this method a mean value of 44.6 +/− 5.3 nM was obtained.

Methods and Materials

Subjects

Blood was drawn from normal human volunteers or from patients having Child-Pugh Class C cirrhosis using EDTA as anticoagulant. In two cases, samples were drawn from a normal individual at different times after different oral doses of SAM were administered, 1600 mg and 800 mg. The plasma was obtained by centrifugation within one hour of collection. Studies were approved by the Institutional Review Boards as appropriate.

Materials

SAH, casein, and bovine serum albumin (BSA) were obtained from Sigma. The microtiter plates were Maxisorp and obtained from Nunc. The ultrafiltration units were ULTRAFREE-MC from Amicon. The SAH-BSA conjugate, anti-SAH antibody, horseradish peroxidase (HRP)-conjugated rabbit anti-mouse antibody and the HRP substrate were obtained from Axis-Shield, Dundee, Scotland. Pooled plasma from 13 normal individuals was from Innovative Research, Michigan.

Reagents

The Assay Buffer contains 100 mM Na₂HPO₄, 150 mM NaCl, 14 mM NaN₃ adjusted to pH 8.5. The wash solution was 100 mM Na₂HPO₄, 150 mM NaCl, 0.1 g/L merthiolate, adjusted to pH 7.4 and then 2 g/L BSA and 0.5 ml/L Tween 20 were added. The dilution buffer was 5.0 mM K₂HPO₄, pH 7.0.

Sample preparation

100 μl of plasma were added to an Amicon ULTRAFREE-MC filter unit (#UFC3LGCOO) and centrifuged at 18,000 × g for 45 min at 5° C. The filtrate (∼ 70−80 μl ) was used for assay.

Coating of microtiter plates

To each well was added 250 μl of a solution (0.1 mg/ml) of SAH-bovine serum albumin (BSA) conjugate in PBS containing 2 μg/ml BSA and incubated overnight at 4° C. The plates were emptied by inversion and the excess liquid removed by tapping the plates on paper towels. Then, to each well was added 250 μl of blocking solution containing 25 g sodium caseinate/ L in PBS and incubated overnight at 4° C. The plates were emptied by inversion, blotting as before and washed 4 times with 250 μl of 10 fold-diluted Assay Buffer. The plates were then dried by inverting for 30 min on paper towels and stored inverted and wrapped in aluminum foil in a sealed plastic bag at −20° C. They are stable for 5−6 months in this condition.

Preparation of Standards

A 10 μM stock solution of SAH in 5 mM potassium phosphate, pH7.0 was diluted 100 fold in Assay Buffer to give a working solution of 0.1 μM. Further dilutions are made to give working solutions of 0.05 μM, 0.02 μM and 0.01 μM. These solutions are stable for at least one month at −20°.

Samples

Routinely, aliquots of 5, 10 and 15 μl of an undiluted, ultrafiltered plasma sample are assayed in duplicate. This usually ensures that at least two of the aliquots fall on the useful portion of the standard curve, giving four measurements that are used to obtain a mean value. In this way a range of concentrations from 3 to 100 nM can be measured without dilution. Samples with higher concentrations of SAH can be measured by dilution of the ultrafiltered plasma samples in 5 mM potassium phosphate buffer, pH 7.0.

Assay procedure

The standards or the samples were diluted to 50 μl in Assay Buffer ins the coated microtiter wells followed by 100 μl of anti-SAH antibody solution. The plates were incubated in the dark at room temperature on a shaking platform for one hr. Each well was then washed 4 times with 250 μl of the wash solution. Each well then received 50 μl of horseradish peroxidase-conjugated rabbit anti-mouse antibody and the plates incubated again for 20 min in the dark at room temperature with shaking. The wells were washed again 4 times with 250 μl of wash solution and 50 μl of HRP substrate solution was added. The plates were incubated for 10 min at room temperature and the reaction stopped by adding 50 μl of H₂SO₄ (0.8 M ). The yellow color is read at 450 nm in a Bio-RAD microplate reader within 5 min. The values were fitted to a quadratic equation as provided by the instrument.

Statistical analysis

This was carried out using GraphPad PRISM. All values presented are means +/− standard error unless otherwise indicated. Figures were drawn with Kaliedagraph (Synergy Software).

Results

Standards providing 0.0, 0.1. 0.2, 0.3, 0.4, 0.5, 0.75, and 1.00 pmoles of SAH per well were run in duplicate on each microtitre plate. The results of a typical standard curve is shown in Fig. 1. The useful portion of the curve is from 0.05 to 0.5 pmoles.

Figure 1. Standard curve for measurement of SAH.

The standard curve was measured as described in the text and absorbance was measured in a microplate reader.

The recovery was measured after the addition of 10, 20 and 50 nM SAH to five individual plasma samples. The recovery of 10 nM (1 pmole) added to the samples was 90.2 +/− 3.3 % (% CV = 8.1). The recovery of 20 nM (2 pmole) was 81.3 +/−6.6 % (% CV = 18.2) and the recovery of 50 nM (5 pmole) was 107.8 +/− 7.9 % (% CV = 16.5).

The between-assay precision was determined using plasma pooled from 13 normal individuals. Seven individual measurements were carried out in quadruplicate over a period of two weeks.. A mean value of 31.7 +/−1.5 was obtained with a % CV = 12.5. Within assay precision was carried out also using pooled plasma from 13 individuals. Eleven separate assays were carried out on the same microtiter plate. A mean value of 29.6 +/− 1.2 was obtained with a %CV = 13.6.

The original paper by Frantzen et al. [14] indicated that SAM was recognized to a slight extent by the anti-SAH antibody and could result in falsely high values for Hcy when SAM was present at concentrations greater than 10 μM. Surprisingly, addition of as much as 2000 nM SAM to a plasma sample containing 35 nM SAH did not change the measured value of SAH (Table 1). It is possible that this is because the amount of anti-SAH antibody has been greatly reduced in this method.

Table 1.

Effect of added SAM on plasma SAH values

*SAM added (nM +/− SEM)
0	50	100	500	1000	2000
35.0 +/− 0.7	35.2 +/− 4.4	35.4 +/− 4.7	35.2 +/− 4.1	36.1 +/− 0.5	37.5 +/− 4.6

^*The indicated amount of SAM was added to different aliquots of the same plasma and processed as indicated in Methods and Materials. The values are the means of duplicate determinations.

The values obtained by the immunoassay were compared with the values obtained using our original fluorescence assay [5]. We compared values for normal individuals, an individual at several times after receiving a single oral dose of SAM as well as individuals with cirrhotic liver disease. The values for 16 control subjects were 18.9 +/− 1.4 and 17.8 +/− 1.4 nM, respectively, for this new method and for the older HPLC/isoindole fluorescence assay. The values for 7 subjects with cirrhotic liver disease were 44.6 +/−5.3 and 46.5 +/− 3.3 for the former and current methods, respectively. The results are shown in Fig. 2 for 40 individual samples covering a long range of values. There is good agreement between the two methods and there is no significant difference between the values obtained with either of the two methods. Table 2 shows a comparison of the published values for plasma or serum SAH of normal individuals. The values obtained using this method are comparable to the others and were obtained in less than 4 hours time.

Figure 2. Comparison of values for plasma SAH obtained by the HPLC/ Fluorescence method and the immunoassay described here.

R = 0.91; slope = 0.97.

Table 2.

Comparison of Plasma Normal SAH Values

Method	nM	Reference
Fluorescence 1,N6-etheno conversion plus HPLC	24 +/− 1 (SEM)	[23]
Modified etheno conversion plus HPLC	28 +/− 3 (SD)	[10]
Fluorescence 1,N5-etheno conversion plus capillary electrophoresis	29 +/− 2 (SD)	[24]
Fluorescence isoindole formation plus HPLC	23 +/− 3 (SEM)	[5]
Coulometric electrochemical detection with HPLC	20 +/− 6 (SD)	[17]
Stable isotope dilution Tandem mass-spectrometry	12 +/− 4 (SD)	[13]
Stable isotope dilution LC/MS	15 (8−26, 95% CI)	[12]
Tandem mass-spectrometry	26 +/− 6 (SD)	[11]
Immunoassay	19 +/− 4 (SEM)	This study

Discussion

Elevation of plasma or serum Hcy is generally accepted as an independent risk factor for vascular disease [15] [16] [1]. Several studies have indicated that elevation of plasma SAH may be a more sensitive indicator of the risk of vascular disease [6] [7] [8]. Measurement of plasma or serum total SAH, however, is challenging because of the low levels present in normal subjects (approximately 20 nM). A number of methods have been developed. Most methods provided results for both SAM and SAH. Loehrer et al. developed a method based upon the formation of the fluorescent 1,N⁶-etheno derivatives of SAH and SAM followed by HPLC separation [9]. This method required a long period of time for the reaction to proceed although the results for measurement of SAH compared favorably with other methods. More recently Castro et al. [10] was able to shorten the derivatization time. Other methods involved the use of highly specialized equipment such as tandem mass-spectrometry [11] [12] [13] and coulometric electrochemical detection [17]. We had developed a reliable assay for the measurement of both SAM and SAH based on the conversion of these compound to their fluorescent isoindoles. Although reliable, the method required two HPLC separations with derivatization before the second column [5].

Using this method we showed that measurement of plasma SAH was a more sensitive indicator of the risk for cardiovascular disease than measurement of plasma total Hcy [6]. This was also the case for renal disease [7]. Because adults with cardiovascular disease or renal disease are likely to have other confounding factors such as diabetes and hypertension that can result in elevated Hcy, we compared plasma Hcy and plasma SAH in a group of children with varying degrees of renal insufficiency but no other risk factors [8]. It was very clear that there was a strong and significant inverse association of SAH with glomerular filtration rate but no such association with plasma Hcy. Yi et al. [18] found a positive correlation of plasma SAH levels and plasma Hcy levels in healthy young women but there was no correlation of Hcy with plasma SAM. This was also associated with lymphocyte DNA hypomethylation. SAH is a potent product inhibitor of most methyltransferases [19]. This group speculated on the toxic effects of elevated SAH on DNA methyltransferase and the effects this might have on DNA methylation and gene expression [20].

Because SAH is the direct metabolic precursor of Hcy and the enzyme catalyzing this conversion, SAH hydrolase, is reversible it might be expected that changes in the plasma concentrations of these two metabolites might follow each other. A discrepancy between changes in plasma SAH and Hcy has been shown by the study by Becker et al. [21]. They showed that changes in folate, cobalamin and vitamin B₆ concentrations did not affect plasma SAH concentrations although plasma Hcy was affected. The fact that SAH and Hcy are not equally affected in the case of the children with decreased renal function [22] and that SAH is more sensitive to changes in vascular function suggests that it may be more appropriate to monitor plasma SAH values. It should be noted, also, that there is no generally accepted mechanism for the pathophysiology of elevated Hcy while the potent end product inhibition of SAM-mediated methyltransferases by SAH is well known [19]. For these reasons this rapid and simple immunoassay for the measurement of SAH in plasma may be a useful tool in the study of methionine metabolites in vascular disease.