Influence of hydrolysis on plasma homocysteine determination in healthy subjects and patients with myocardial infarction

143

Arherosclerosis, 88 (1991) 143-151 0 1991 Elsevier Scientific Publishers ADONZS 002191509100118T

ATHERO

Ireland.

Ltd. 0021-9150/91/$03.50

04635

Influence of hydrolysis on plasma homocysteine determination in healthy subjects and patients with myocardial infarction A. Andersson Departments

I, A. Isaksson ‘, L. Brattstriim 2, B. Israelsson 3 and B. Hultberg

’

of ’Clinical Chemistry and ’ Neurology, Unicersity Hospital, S - 22185 Lund (Sweden) and .’Department of Medicine, Malm6 General Hospital, S - 21401 Malmii (Sweden) (Received 5 October, 1990) (Revised, received 22 January, 1991) (Accepted 25 January, 1991)

Summary

After acid hydrolysis, mean plasma homocysteine concentrations, measured as homocysteine disulphides, of about 1000 and 40 pmol/l have recently been reported in 26 survivors of myocardial infarction and 26 matched control subjects, respectively. This finding contrasts sharply with those more than 50 times lower total homocysteine concentrations found by other research groups in non-hydrolysed plasma from survivors of myocardial infarction. Using the same hydrolysis conditions, we could not detect any homocysteine disulphides in plasma hydrolysates from 9 survivors of myocardial infarction and 10 healthy subjects, who had mean total homocysteine concentrations in non-hydrolysed plasma of 16.9 i 6.5 and 15.8 + 10.3 pmol/l, respectively. The chromatograms contained several peaks, probably representing peptides, which disappeared with more complete hydrolysis and which might have been misinterpreted as homocysteine disulphides in the reported study. Only after reduction of disulphides and by using a sulphydryl-selective extraction procedure were we able to determine mean homocysteine concentrations in hydrolysed plasma to be 26.2 k 7.9 pmol/l in the survivors of myocardial infarction and 24.5 k 12.2 pmol/l in the healthy reference subjects. Thus, we could not confirm that survivors of myocardial infarction have homocysteine concentrations that are many times higher than found in healthy subjects.

Key words: Homocysteine;

Hydrolysis; Plasma; Myocardial infarction

Introduction Correspondence to: B. H&berg, Department of Clinical Chemistry, University Hospital. S - 221835 Lund. Sweden. Tel: 046-l 73447; Fax: 046-1891;4.

Homocysteine, a sulphydryl amino acid, is the demethylated derivative of methionine. Certain inborn errors in homocysteine metabolism result

144 in severe hyperhomocysteinemia ( > 200 pmol/l) which is associated with the development of arteriosclerosis and frequent thromboembolism in children and young people [l]. Several studies have shown that moderate hyperhomocysteinemia is frequent in adult patients with vascular disease. Therefore even moderate elevations of plasma homocysteine have been suggested as a possible risk factor for vascular disease [2-91. In normal human plasma, 70-80% of homocysteine is disulphide-bound to proteins and the rest is mainly found in the form of the free homocysteine-cysteine mixed disulphide (MDS). In pathological conditions and after methionine loading the free disulphide homocysteine-homocysteine (homocystine) is also detectable. The sum of these fractions has been determined by different methods [8-141 after reduction of disulphide bonds prior to analysis and assumed to be the total content of homocysteine in plasma or serum. A value of 5-15 pmol/l [7-121 in normal fasting plasma has been considered normal. However, two studies have recently been published suggesting that homocysteine in plasma or serum may exist in other forms and in much higher concentrations than previously recognized. In the first of these studies McCully and Vezeridis [ 151 reported that homocysteine thiolactone was present in the range of 32-6700 pmol/l in serum from normal subjects and up to an incredibly high 47000 pmol/l in patients with coronary artery disease. It was, however, later shown by Mudd et al. [16] that no homocysteine thiolactone was detectable in human serum, and McCully [17] was unable to reproduce the earlier findings. In the second study, Olszewski and Szostak [18] reported that hydrolysis of plasma released hitherto unknown homocysteine fractions determined as MDS and homocystine. They found the extremely high total plasma homocysteine concentration of 958 k 84 Fmol/l (mean + SD> (n = 26) in survivors of myocardial infarction but only 38 + 11 Fmol/l in control subjects (n = 26). They proposed that homocysteine may also be bound to proteins by other linkages than disulphidebonds and that it is released by acid hydrolysis. A support for their hypothesis is that homocysteine thiolactone by a thiolation process can form a peptide linkage with the amino group of amino

acid residues in proteins. This principle has been used to label proteins with high density atoms in electron microscopy applications [19]. It has also been shown that thiolation of albumin can occur in vitro at physiological pH [20]. In this investigation we explored whether homocyst(e)ine liberated by hydrolysis, but not by reduction, is present in plasma. We first tried to determine homocysteine according to Olszewski and Szostak’s method [18]. With this method, however, we were not able to measure homocysteine levels below 100 Fmol/l. This was mainly because we could only use a highly diluted hydrolysate (80-fold dilution of the original plasma sample) in order not to overload the ion-exchange column. To keep the sample dilution low (about lo-fold), we developed a selective extraction procedure based on a covalent reaction between sulphydryls and Thiopropyl-Sepharosea 6B, a principle which has previously been used to isolate cysteine peptides from proteins [21]. By using this method on hydrolysed plasma, we found about 9 pmol/l higher homocysteine levels than in non-hydrolysed plasma, from both survivors of myocardial infarction and selected healthy reference subjects. Materials

and methods

Materials

Dr_-Homocysteine and L-homocysteine thiolactone hydrochloride were purchased from Sigma, St. Louis, U.S.A. L-Homocystine, DL-cystathionine and L-methionine were obtained from Fluka Chemie AG, Buchs, Switzerland, and dithiothreithol (DTT) from Calbiochem, San Diego, CA, U.S.A.. Sodium hydroxide (pellets), disodium EDTA, sodium phosphate (monobasic monohydrate, dibasic heptahydrate, tribasic dodecahydrate) were all purchased from Janssen Chimica, Beerse, Belgium. Hydrochloric acid (HCI) 30% (suprapur), disodium tetraborate decahydrate, sodium hydroxide (Titrisole 1 M) and citric acid monohydrate were from E. Merck, Darmstadt, Germany, and methanol was from Labscan Limited, Dublin, Ireland. Silica-based propylbenzenesulphonyl (Bound Elute, 500 mg, SCX, hydrogen form) disposable cation-exchange columns were supplied by Analytichem Int., Inc., Harbor City,

145 CA, U.S.A.. Thiopropyl-Sepharosea 6B was purchased from Pharmacia LKB Biotechnology AB, Uppsala, Sweden, and hydrolysate amino acid standard no. 20089 (not containing homocyst(e)ine) from Pierce, St. Louis, MO, U.S.A. The ninhydrin reagent, buffers for the amino acid analyser and the sample dilution buffer (0.15 M lithium, 0.07 M citrate, pH 2.2) were prepared according to Biotronic GmbH manuals (Munich, Germany). All reagents were of pro analys quality. The reagent water was produced by a Millipore Aqua Super Q-system. Subjects

Nine male patients, aged 50-69 years, who had survived myocardial infarction 3-4 months previously, 10 apparently healthy. men, aged 20-69 years, and one 46-year-old untreated female homozygote with hyperhomocysteinemia due to cystathionine P-synthase deficiency were studied. The diagnosis of myocardial infarction was based on criteria previously described [22]. The 10 healthy men were selected from a larger reference population in order to cover a wide range (3.5-38.0 /Lmol/ll in plasma homocysteine as determined by our previously described method for measurement of total plasma homocysteine (free plus protein-bound) after reduction of disulphide bonds [14]. Values for plasma total homocysteine after reduction in myocardial infarction patients were also compared with those in a control group of 33 healthy men, aged 50-69 years. Instruments

Hydrolysis tubes, with a valve for sealing under vacuum, were supplied by Werner-Glas Stockholm, Sweden. Samples were evaporated in a Savant SpeedVac Concentrator from Savant Inst., Inc., Farmingdale, NY, and hydrolysed in a funnel Memmert model UM 200 from Memmert GmbH, Schwabach, Germany. The analysis was performed on a one-column automatic amino acid analyser, Biotronic LC 5000, Biotronic GmbH, Munich, Germany, using an extended program for physiological amino acids according to the manufacturer’s manual. The column, 180 X 3.2 mm, was packed with a strong cation exchange resin, BIC 2710 Biotronic GmbH. The system gave the following retention times: cysteine 42

min, cystine 56 min, homocysteine 58 min, methionine 60 min, MDS 75 min, homocystine 95 min and homocysteine thiolactone 137 min. The chromatogram was evaluated with a Shimadzu CR2AX integrator. Sample preparation and analysis Sample collection. Plasma from the studied

individuals had been collected 2-8 months previously as follows. After an overnight fast: venous blood was collected in evacuated tubes containing EDTA. The blood was immediately centrifuged and the plasma was divided into several aliquots which were stored at -20°C. Hydrolysis. 50 ~1 plasma and 1000 ~1 of 4 M HCl were mixed in a hydrolysis tube. The tubes were frozen in solid carbon dioxide/acetone, evacuated to 0.2 bar while thawed, sealed and placed in a hydrolysis funnel at 110 ‘C. After 5 h the tubes were chilled on ice and the hydrolysate sample was transferred to another glass tube which was evaporated under vacuum in a centrifuge-evaporator. Analysis. Hydrolysed samples not exposed to thiopropyl-Sepharose extraction were either redissolved in the dilution buffer pH 2.2 and analysed directly on the automatic amino acid analyser or reduced for 1 h at 22 ’C in 1 ml 0.1 M sodium tetraborate buffer pH 8.5 with 0.5 mM EDTA and 30 ~1 1 M DTT in water. The reduced samples were then diluted 4 times with the dilution buffer and adjusted to pH 2.2 with 1 M HCl before analysis (resulting in a final dilution factor of 80 for both reduced and non-reduced samples). To obtain total plasma homocysteine after reduction of disulphides, plasma was analysed without hydrolysis as described earlier [14]. ThiopropyLSepharose extraction. The extraction procedure principally consists of the following steps. (11 Reduction of the disulphides in the hydrolysed plasma with DTT. (2) Separation of homocysteine from DTT on a disposable cation exchange column. (31 Homocysteine and other sulphydryls are bound to thiopropyl-Sepharose with liberation of 2-thiopyridone (Fig. 1. reaction A). The binding capacity of the gel is in excess (about 4 times) of total sulphydryls in the sample. (4) The gel is washed to remove non-bound amino

146 (1) A

(2)

&So /

+ FISH

-)

IS-S,

+ S5

N Y H (4)

B

;_ S-SR

(5) + HS-CH,(CHOH),CH,-SH

ZIZ

SH + RSH T

+ ?-

OH

2 OH

Fig. 1. Principal reaction scheme of thiopropyl-Sepharose

extraction. A detailed description is given in Materials and methods. (A) Sulphydryls (2) are bound to thiopropyl-Sepharose (1) with liberation of 2-thiopyridone (3). (B) Bound sulphydryls (4) are released after reduction with excess DTT (5).

acids (non-sulphydryls). (5) Release of bound sulphydryls after reduction with DTT (Fig. 1, reaction B). (6) Determination of homocysteine by ion-exchange chromatography on an amino acid analyzer. The detailed procedure is as follows: after evaporation, hydrolysed samples were reduced in 1 ml 0.1 M sodium tetraborate buffer pH 8.5 and 30 ~1 1 M DTT. The tubes were placed in an ultrasonic bath for 10 min and then left for 20 min at 22’ C. The pH was adjusted to 1.8-2.0 with 1 M HCl before 2 ml 0.01 M HCl was added to each tube. Thereafter the samples were applied to Bound Elute disposable cation exchange columns (500 mg) previously conditioned with 3 ml deaerated methanol, 2 ml 0.05 M and 2 ml 0.01 M HCl. The columns were washed with 3 ml 0.01 M HCl before homocysteine was eluted with 5 ml 0.5 M HCl in 10% methanol. To the eluate, 0.5 ml 0.5 M trisodium phosphate with 3 mM EDTA was added and the pH adjusted to 7.0-7.4 with 10 M sodium hydroxide. Thiopropyl-Sepharose, 67 mg per sample, washed in 10 mM phospate buffer pH 7.0 with 0.25 mM EDTA, was transferred to preweighed tubes and centrifuged 3 min at 3000 x g. The supernatant was discarded. Each tube then contained about 0.2 ml gel corresponding to approximately 4 pmol 2pyridyl disulphide groups [231. The pH-adjusted eluates were added to these tubes. The tubes were shaken for 60 min, centrifuged and the supernatant discarded. Each tube was washed 3 times with 6 ml 10 mM phosphate buffer pH 7.0 and once with 6 ml 0.025 M sodium tetraborate buffer pH 8.5 containing 0.5 mM EDTA. After aspiration of supernatants the tubes were weighed before 150 ~1 0.4 M DTT in 0.025 M sodium

tetraborate buffer pH 8.5 was added to each tube. The tubes were then shaken for 45 min. Thereafter 100 ~1 0.25 M citric acid pH 3.5 was added to each tube and after mixing the tubes were centrifuged. To 180 ~1 of supernatant 8 ~1 1 M HCl was added, which resulted in a pH of 2.0-2.2 and a final dilution factor of about 10. These samples were then analysed on the automatic amino acid analyser. Calculations

The concentrations of homocysteine in the hydrolysed and thiopropyl-Sepharose extracted plasma samples were calculated relative to an external standard of 50 ~1 of 50 PM r_-homocystine treated in the same way as the plasma samples. As different volumes of fluid can remain in the tubes after aspiration in the last step of gel-washing, an individual dilution factor is calculated for each sample based on the weight of the tube before the last DTT-treatment. The amino acid concentration in runs without thiopropylSepharose extraction were calculated relative to an external standard with each amino acid with the exception of homocysteine, where DTTtreated homocystine was used. Statistical significance (P < 0.05) was tested with the Mann-Whitney rank sum test. Linear regression analysis was used to compare the two methods. Errors in the slope and intercept were obtained by Student’s t-test of their standard deviations [241. Ecaluation of the hydrolysis and thiopropyl-Sepharose method

The interassay precision was determined by analysing a plasma sample 3 times (33.0 k 1.6 PM

147 homocysteine, mean k SD). The accuracy of the method was estimated by recovery studies. On two different occasions, 50 pmol homocystine per litre plasma was added before hydrolysis. The recovery of the additional homocysteine was 102 and 109910,respectively. The time necessary for binding of homocysteine and homocysteine thiolactone to the thiopropyl-Sepharose was estimated by mixing samples with known concentrations of homocysteine and homocysteine thiolactone with a 30-fold concentration of thiopropyl-Sepharose at pH 7.0. The binding was followed by monitoring the appearance of 2-thiopyridone at 343 nm (absorption coefficient: 8080 M-’ . cm-‘> [23]. The binding of homocysteine was complete in 24 min. At this time only 0.5% of homocysteine thiolactone was bound. After 200 min 3.6% of homocysteine thiolactone had been bound.

Results Hydrolysis of plasma and direct analysis on the amino acid analyser

When plasma was hydrolysed according to the conditions described by Olszewski and Szostak [18] (110°C 5 h, 4 M HCI) and analysed on an automatic amino acid analyser no peak corresponding to the elution time of MDS was seen, whereas a peak in homocystine position was found. To test whether this peak was homocystine, hydrolysed plasma was reduced with DTT and then analysed. The peak was not affected by the DTT reduction but a new peak in homocysteine position appeared in most but not in all chromatograms. No difference in the appearance of this peak was seen between the two groups of healthy subjects and the patients. The concentration of homocysteine in all plasma samples was below the detection limit, 100 pmol/l, when compared to a standard addition of homocystine to the plasma sample prior to hydrolysis. Not even when plasma from the hyperhomocysteinemit patient was hydrolysed was the peak in homocystine position larger than in the other subjects. This peak did not disappear after reduction, and no peak in MDS position was seen. After reduction the peak in homocysteine posi-

TABLE

1

THE DEGREE OF HYDROLYSIS OUS HYDROLYSIS CONDITIONS

OF PLASMA

IN VARI-

The degree of hydrolysis is expressed as the total area of all peaks in every chromatogram as a percentage of the chromatogram with the highest peak-sum area (20 h. 110°C and 4 M HCI). Conditions Time (h)

Temp. K?)

HCI (M)

20

110

5 5 5 2 5 5 5

110 110 110 110 I10 80 45

3 4 3 2 4 1 4 4

Degree of hydrolysis (9:)

too 94 90 83 73 70 39 4

tion was estimated to correspond to about 300 pmol/l in this patient. To test different hydrolysis conditions, a plasma sample was divided into several aliquots and hydrolysed at different times, temperatures, and acidities. Hydrolysis at 110°C in 4 M HCI for 20 h was found to be most complete and the chromatogram of this plasma did not show the peak in homocystine position (Table 1). The chromatogram obtained after the shorter hydrolysis used by Olszewski and Szostak [18] showed a 94% hydrolysis efficiency (Table 1) compared with that obtained after hydrolysis for 20 h and showed several small peaks including the peak in homocystine position. Hydrolysis of plasma and thiopropyl-Sepharose extraction

In order to obtain a higher sensitivity we extracted hydrolysed plasma samples with thiopropyl-Sepharose as described in the methods. A chromatogram is shown in Fig. 2. The homocysteine levels obtained in plasma from 10 healthy subjects, 9 patients with myocardial infarction and the hyperhomocysteinemic patient are presented in Table 2. This table also shows the value of total plasma homocysteine after reduction [14]. In each individual, except the patient with hyperhomocysteinemia, it is found that hydrolysis results in a somewhat higher level of homocysteine

c

b

I

0.05

AU

a

04 0

f I

I

I

1

0

I

20

40

60

80

Retention

time

(min)

Fig. 2. Chromatogram of a plasma sample after hydrolysis and thiopropyl-Sepharose extraction from a patient who has survived a myocardial infarction. Peak (a + b) oxidized and unoxidized DTT, peak (c) cysteine and peak (d) homocysteine 29.7 pmol/l. The other peaks are unidentified.

compared to the results obtained after reduction of non-hydrolysed plasma. The mean homocysteine concentration in plasma from healthy subjects and from patients with myocardial infarction was 15.8 t_ 10.3 (mean _t SD> and 16.9 rt 6.5 pmol/l, respectively, when assayed in non-hydrolysed plasma and 24.5 + 12.2 and 26.2 f 7.9, re-

TABLE 2 TOTAL HOMOCYSTEINE [HCY(ei)] IN NON-HYDROLYSED AND HYDROLYSED PLASMA Category

Total plasma HCY(ei) PM (mean k SD)

n

Non-hydrolysed Myocardial Infarction Healthy subjects Hyperhomocysteinemia Controls

Hydrolysed

9

16.9+ 6.5

26.2& 7.9

10

15.8 + 10.3

24.5 + 12.2

1 33

363 11.7+ 2.2

N.D. = not determined.

337 N.D.

.

,

,

10

20

Homocysteine

I

30 after

,

40 reduction

.

,

.

50 (PM)

Fig. 3. Comparison of plasma homocysteine content after reduction and after hydrolysis. 0, healthy subjects; 0, patients 3-4 months after a myocardial infarct.

spectively, when analysed after hydrolysis. A comparison between the two methods by linear regression analysis (n = 19, the hyperhomocysteinemic patient was excluded) showed the equation y = 1.11x + 7.14, r = 0.93 (x is the value obtained in non-hydrolysed plasma). The intercept deviated significantly from 0 (P < 0.OO.Q but the deviation of the slope from 1.0 was not significant (P = 0.32) (Fig. 3). When assayed in non-hydrolysed plasma (Table 21, the homocysteine level in the patients with myocardial infarction was significantly higher (P < 0.001) than that in the control group (n = 33). Hydrolysis of homocysteine and homocysteine-related amino acids

In order to study the effect of hydrolysis, 125 nmol homocysteine, homocystine, homocysteine thiolactone, cystathionine, methionine, or 50 ~1 of Pierce amino acid standard were hydrolysed for 5 h at 110 o C in 4 M HCl. The samples were evaporated and redissolved in the dilution buffer pH 2.2 before analysis. The results for each hydrolysed sample were compared to the same non-hydrolysed sample. When homocysteine or homocysteine thiolactone were hydrolysed, these

149 TABLE

3

CONVERSION OF HOMOCYSTINE [HCY(i)], (HCYTL) AmER 5 h AND 110°C HYDROLYSIS Hydrolysed

HCY(ei) HCY(i) HCYTL

substance

Amount

HOMOCYSTEINE IN 4 M HCI

after hydrolysis

expressed

[HCY(ei)lAND HOMOCYSTEINETHIOLACTONE

as homocysteine

in %

HCY(ei)

HCY(i)

HCYTL

HCY(ei + i) + HCYTL

6.4 0.0 8.7

7.5 99.5 13.1

68.5 0.6 80.3

82.4 100.1 102.1

species and also homocystine were detected with the thiolactone as the dominating component (Table 3). Homocystine showed no such conversion and only traces of the hydrolysed substance were found as homocysteine thiolactone. Recovery of hydrolysed homocysteine thiolactone was 100.2%, calculated as the sum of all 3 species. Only homocystine was quantitatively recovered (99.5%) in its original form. Homocysteine was most sensitive to hydrolysis, since only 82% was recovered as homocysteine, homocystine and homocysteine thiolactone (Table 3). Recovery of methionine and cystathionine after hydrolysis was 97 and 104%, respectively, and no formation of homocysteine, homocystine or homocysteine thiolactone could be seen. When Pierce amino acid standard was hydrolysed, no formation of those species could be noted. Discussion

We could not reproduce the results reported by Olszewski and Szostak [18]. The chromatograms obtained after hydrolysis of plasma using the conditions they described contained several small unidentified peaks, of which one in the position of homocystine was not seen when plasma was hydrolysed more extensively. This “homocystine” peak was not larger in plasma from a patient with severe hyperhomocysteinemia and it was resistant to reduction. In no plasma was a peak sensitive to reduction, corresponding to MDS, identified. Therefore, we believe that the peaks identified as MDS and homocystine by Olszewski and Szostak [18] represent undegraded peptides. Nonetheless, after reduction of disulphides in hydrolysed plasma, we could in most, but not in

all plasma samples identify a small peak in the homocysteine position, but this peak could only be quantified in the plasma from the hyperhomocysteinemic patient. Only after thiopropyl-Sepharose extraction were we able to quantify homocysteine satisfactorily in all hydrolysed plasma samples. The homocysteine concentrations were on average 9 pmol/l higher than those obtained in non-hydrolysed plasma and there was an excellent correlation between the two methods. Whether the higher mean homocysteine concentration found in hydrolysed plasma represents an extra homocysteine fraction is, however, uncertain. The difference between the methods might be because there is a small fraction of disulphide bound homocysteine which is not prone to reduction until after hydrolysis. Another explanation for the discrepancy is that homocysteine bound to plasma proteins by peptide linkage could exist as the disulphide of cysteine or homocysteine. A third possibility might be that some compound co-elutes with homocysteine even after thiopropyl-Sepharose extraction. We found that homocysteine in sulphydryl form to a great extent was converted to homocysteinc thiolactone by hydrolysis. However, the disulphide homocystine was very stable, which implies that MDS, homocysteine-protein disulphide and homocystine (or MDS) bound to proteins by linkages involving the a-carbon end of the homocysteine moiety, also should be stable to hydrolysis. As the samples had been stored for several months at -20°C it could be assumed that all homocysteine was present in the stable disulphide forms [25,261. Theoretically, if homocysteine is bound to proteins by the a-carbon end, in sulphydryl form, it may undergo ring closure during hydrolysis to homocysteine thiolactone which

150 is not measurable with the thiopropyl-Sepharose method. The same restriction is also applicable to the method used by Olszewski and Szostak [18] since they did not include homocysteine thiolactone in their calculations. However, we found no detectable homocysteine thiolactone in hydrolysed plasma not extracted with thiopropyl-Sepharose, not even in plasma from the patient with severe hyperhomocysteinemia. Therefore, we believe that no homocysteine fraction has been lost in the hydrolysis or the extraction procedure. When patients and the ten healthy reference subjects were compared, the difference in mean homocysteine values between the groups was of the same order whether homocysteine was measured in hydrolysed or non-hydrolysed plasma. Even if the significant difference (44%) in mean total homocysteine concentration in non-hydrolysed plasma between the present patients and the larger group of control subjects were similar in hydrolysed plasma our results are still in sharp contrast to the finding by Olszewski and Szostak [18] of about 25 times higher mean homocysteine concentration in survivors of myocardial infarctions than in matched control subjects and with no overlap between groups. Acknowledgements This work was supported by grants from the Medical Faculty, Lund University, the Swedish National Association Against Heart and Chest Disease, the Swedish Medical Research Council, the Ernhold Lundstrom Foundation and the Albert P%hlsson Foundation. References 1 Mudd, S.H. and Levy, H.L., Disorders of transsulfuration. In: Stanbury, Wyngaarden, Fredrickson, Goldstein and Brown (Eds.), The Metabolic Basis of Inherited Disease 5th edn., McGraw Hill, New York, 1983, p. 522. 2 Wilcken, D.E.L. and Wilcken, B., The pathogenesis of coronary artery disease: A possible role for methionine metabolism, J. Clin. Invest., 58 (1976) 1079. 3 Brattstrilm, LE., Hardebo, J.E. and Hultberg, B., Moderate homocysteinemia - a possible risk factor for arteriosclerotic cerebrovascular disease, Stroke, 1.5(1984) 1012. 4 Boers, G.H.J., Smals, A.G.H., Trijbels, F.J.M., Fowler B., Bakkeren, J.A.J.M., Schoonderwaldt, H.C., Kleijer, W.J. and Kloppenborg, P.W.C., Heterozygosity for homocystin-

uria in premature peripheral and cerebral occlusive arterial disease, N. Engl. J. Med., 12 (1985) 709. 5 Kang, S.S., Wang, P.W.K., Cook, H.Y., Norusis, M. and Messer J.V., Protein-bound homocysteine a possible risk factor for coronary artery disease, J. Clin. Invest., 77 (1986) 1482. 6 Murphy-Chutorian, D., Wexman, M.P., Grieco, A.J., Heininger, J.A., Classman, E., Gaull, G.E., Ng, S.K.C., Feit, F., Wexman, K. and Fox, A.C., Methionine intolerance: a possible risk factor for coronary artery disease, J. Am. COB. Cardiol., 6 (1985) 725. 7 Brattstrom, L., Israelsson, B., Norrving, B., Bergqvist, D., Thdrne, J., H&berg, B. and Hamfelt, A., Impaired homocysteine metabolism in early-onset cerebral and peripheral occlusive arterial disease. Effects of pyridoxine and foIic acid treatment, Atherosclerosis, 81 (1990) 51. 8 Malinow, M.R., Kang, S.S., Taylor, L.M., Wong, P.W.K., Coull, B., Inahara, T., Mukerjee, D., Sexton, G.J. and Upton, B.S., Prevalence of hyperhomocysteinemia in patients with peripheral arterial disease, Circulation, 79 (1989) 1180. 9 Israelson, B., Brattstrom, L. and Hultberg, B., Homocysteine and myocardial infarction, Atherosclerosis, 71 (1988) 227. 10 Stabler, S.P., Marcell, P.D., Podell, E.R. and Allen, R.H., Quantitation ot total homocysteine, total cysteine and methionine in normal serum and urine using capillary gas chromatography mass-spectrometer, Anal. Biochem., 162 (1987) 185. 11 Refsum, H., Helland, S. and Uealand, P.M., Radioenzymic determination of homocysteine in plasma and urine, Clin. Chem., 31 (19851624. 12 Araki, A. and Sako, Y., Determination of free and total homocysteine in human plasma by high-performance liquid chromatography with fluorescence detection, J. Chromatog., 422 (1987) 43. 13 Refsum, H., Ueland, P.M. and Svardal, A.M., Fully automated fluorescence assay for determining total homocysteine in plasma, Clin. Chem., 35 (1989) 1921. 14 Andersson, A., Brattstrom, L., Isaksson, A., Israelsson, B. and Hultberg, B., Determination of homocysteine in plasma by ion-exchange chromatography, Stand. J. Clin. Lab. Invest., 49 (1989) 445. 15 McCully, K.S. and Vezeridis, M.P., Homocysteine thiolactone in arterosclerosis and cancer, Res. Commun. Chem. Pathol. Pharmacol., 59 (1988) 107. 16 Mudd, S.H., Matorin, AI. and Levy, H.L., Homocysteine thiolactone: failure to detect. Res. Commun. Chem. Pathol. Pharmacol., 63 (1989) 297. 17 McCully, K.S., Homocysteinemia and arterosclerosis: Failure to isolate homocysteine thiolactone from plasma and lipoproteins, Res. Commun. Chem. Pathol. Pharmacol., 63 (1989) 301. 18 Olsewski, A.J. and Szostak, V.B., Homocysteine content of plasma proteins in ischemic heart disease, Atherosclerosis, 69 (1988) 109. 19 Kendall, P.A., Thiolation of proteins with homocysteine thiolactone: Preparation of Immunoglobulin G heavily la-

151 belled

with methylmercury.

Biochim.

Biophys.

Acta,

257

(1972) 83. 20 Dudman, N.P.B. and Wilcken, D.E.L., Homocysteine thiolactone and experimental homocysteinemia, Biochem. Med., 27 (1982) 244. 21 Carlsson, J., Svensson, A. and Ryden, L., Isolation of cysteine peptides from proteins by solid-phase techniques based on thiol-disulfide exange. In: Previero, A., ColletiPreviero, M.-A. (Eds.1, Solid phase methods in protein sequence analysis, INSERM Symp. No. 5. Elsevier. 1977, p. 29. 22 Henning, R. and Lundman, T., Swedish co-operative CCU-study, Acta Med. Stand. Suppl., 586 (1975) 3.

23 Pharmacia

Fine Chemicals

AB, Affinity

chromatography:

principles and methods, Pharmacia Fine Chemicals A.B., June. 1979. I. 24 Miller, J.C. and Miller, J.N., Errors in the slope and intercept of the regression line. In: Statistics for analytical chemistry. Ellis Horwood Limited, West Sussex, England, 1984. 25 Perry, T.L. and Hansen, S., Technical pitfalls leading to errors in the quantitation of plasma amino acids, Clin. Chim. Acta, 25 (1969) 53. 26 Kang, S., Wong, P.W.K. and Becker, N., Protein-bound homocysteine in normal subjects and in patients with homocystinuria. Pediatr. Res., 13 (1979l1141.