| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
1 Chair of Clinical Biochemistry, University of Sassari, Viale San Pietro 43/B, 07100 Sassari, Italy
aauthor for correspondence: fax 39-079228120, e-mail carru{at}uniss.it
Several studies have indicated that plasma proteins interact with homocysteine (Hcy) to form stable disulfide-linked products. Hcy in plasma is mainly bound to albumin, but interactions with ceruloplasmin, fibrin, annexin II, and transthyretin have been also reported (1)(2)(3)(4). In 1991, Olszewski and McCully (5) described the presence of Hcy in lipoproteins in patients with hypercholesterolemia. Because the analysis was performed after acidic hydrolysis of apoprotein, the Hcy measured by these authors was the sum of (a) Hcy incorporated in the primary structure of apolipoprotein B-100 (5), (b) Hcy thiolactone bound to lysine residues of protein by amide or peptide linkages and converted to Hcy by acidic conditions after release (6), and (c) Hcy linked to apolipoprotein B-100 (apoB-100) by a disulfide bond.
Recent studies have demonstrated that there are at least nine free sulfhydryl groups (–SH) in the apoB-100 primary structure that could potentially bind plasma free aminothiols by disulfide linkage (7). Because lysyl residues of apoB-100 could react in vivo with plasma Hcy thiolactone by an amide bond, the number of free apoprotein sulfhydryl groups could increase considerably, thus increasing the number of sites that may be bound with plasma aminothiols (6). These LDL modifications are accompanied by an increase in density and in electrophoretic mobility of lipoprotein and are associated with functional alterations that make Hcy-LDL more susceptible to aggregation and to spontaneous precipitation. Moreover, higher uptake of Hcy-LDL by membrane receptor and by phagocytosis and a higher accumulation of intracellular cholesterol have been observed in cultured macrophages, suggesting that homocysteinylation could increase the atherogenicity of LDL (8)(9). Thus, to study the association between Hcy and lipid metabolism, a highly sensitive method to measure Hcy and other thiols bound to apoB-100 is required.
Here we describe a simple capillary electrophoresis method with laser-induced fluorescence detection to measure physiologic thiols bound to apoprotein by a disulfide linkage. We assessed the performance of this analytical method by measuring apoB-100-bound thiols in 16 volunteers. Participants were not receiving dietary supplements of vitamin B6, B12, or folate or statin therapy as determined by medical interviews. Blood was collected into Vacutainer Tubes containing EDTA. Plasma was prepared by centrifugation at 2000g for 10 min at 4 °C. LDL was isolated by ultracentrifugation according to the methods of Himber et al. (10) and McDowell et al.(11). Recovered LDL was passed through Sephadex PD-10 column equilibrated with phosphate-buffered saline to remove salts, EDTA, and other interfering compounds. Lipoprotein (a) was isolated as described by Fless et al. (12).
Because the aim of our work was to measure thiols bound to apoB-100 only, it was important to verify that the LDL subfraction obtained by ultracentrifugation did not contain other lipoprotein types. We therefore checked the purity of the isolated LDL by capillary electrophoresis with diode array detection, using our previous method to monitor the oxidative state of the lipoproteins (13). The obtained electropherograms showed that the LDL fraction was free from the other lipoproteins. This result was confirmed by classic lipoprotein agarose electrophoresis of the isolated subfractions.
For quantification of apoB-100-bound thiols, we delipidated 200 µL of the LDL fraction (1 g/L protein) by adding 2 mL of methanol–chloroform (2:1 by volume) and centrifuging the sample at 2000g for 5 min. After protein precipitation, we discarded the supernatant; we then washed the protein pellet three times with 50 g/L 5-sulfosalicylic acid and successively dissolved it in 200 µL of 0.2 mmol/L NaOH at 60 °C for 15 min. We then reduced the disulfide bonds by incubation with 20 µL of 100 mL/L tributyl-n-phosphate (TBP) in dimethyl formamide for 10 min. We mixed 50 µL of the resulting sample with 100 µL of 150 mmol/L sodium phosphate buffer (pH 12.5) and 10 µL of 0.8 mmol/L 5-iodoacetamidofluorescein. After letting the mixture react for 15 min, we diluted the derivatized samples 100-fold in water and analyzed them by capillary electrophoresis. ApoB-100-bound thiols were analyzed by a P/ACE 5510 capillary electrophoresis system equipped with a laser-induced fluorescence detector (Beckman), as described previously with some modifications (14). Briefly, we used an uncoated fused-silica capillary 57 cm long (50 cm to the detection window) with a 75-µm i.d. Analysis was performed with 35 nL of sample under nitrogen pressure (0.5 psi) for 5 s with an electrolyte solution containing 18 mmol/L sodium phosphate, 14.5 mmol/L boric acid, and 75 mmol/L N-methyl-D-glucamine (pH 11.4). The separation conditions (22 kV, 150 µA at normal polarity) were reached in 20 s and held at a constant voltage for 8 min. Separations were carried out at 40 °C and monitored by fluorescence detection with excitation at 488 nm and emission at 520 nm. As seen in Fig. 1A
, use of the above electrolyte solution enabled baseline separation of cysteinylglycine (Cys-Gly), Hcy, and Cys calibrators in <8 min.
|
After LDL isolation by ultracentrifugation, we extracted the lipids from the lipoprotein by methanol–chloroform and then discarded that layer. The sample containing apoB-100 was successively washed three times with a solution of 5-sulfosalicylic acid to remove the water-soluble small residual components in the extracted samples. We reduced the disulfide bond between the thiols and apoB-100 by adding TBP to the apoprotein solution. Released thiols were successively derivatized by the selective thiol laser-induced fluorescence-labeling agent 5-iodoacetamidofluorescein and separated by capillary electrophoresis, as shown in Fig. 1B
.
Calibration curves for a water solution of Cys-Gly (y = 24.5x + 0.22 µmol/L), Hcy (y = 33.3x + 1.18 µmol/L), and Cys (y = 15.8x + 0.98 µmol/L) show good correlation (r2 >0.99 for all thiols), demonstrating the linear response over the concentrations tested (20–8000 nmol/L). We calculated the reproducibility of the injections by injecting the same calibration solution 10 times consecutively. We evaluated the within-run (intraassay) imprecision of the method by injecting the same extracted sample 10 times consecutively and the between-run (interassay) imprecision by injecting the same extracted sample on 10 consecutive days. Precision tests indicated good repeatability of our method for both migration times (CV <0.7%) and areas (CV<3.1%). Moreover, we obtained good intra- and interassay reproducibility (CV <7% and CV <10%, respectively).
We determined the recovery of the thiols by adding pure thiols to apoB-100 samples. In particular, we added oxidized thiols after isolation and purification of apoprotein from LDL, before the TBP addition. The analytical recoveries, evaluated at four different concentrations for each thiol, were 95.6–103.5% for Cys-Gly, 95.1–104.5% for Hcy, and 94.8–102.6% for Cys. The limit of detection, calculated based on 35-nL injections of a solution of calibrators (after dilution 100-fold in water), was 25 pmol/L, corresponding to an injected quantity of 
0.5 amol, with a signal-to-noise ratio of 3. The limit of quantification, also calculated by 35-nL injections of sample, was 15 nmol/L.
We tested the performance of our capillary electrophoresis method by measuring apoB-100 in 16 volunteers. Reduced and total thiols were also measured by capillary electrophoresis, as described previously, and the data are reported in Table 1
. Cys was the most abundant thiol bound to apoB-100 [mean (SD), 109.4 (39.3) µmol/L], but smaller amounts of Cys-Gly [49.12 (10.67) µmol/L] and Hcy [5.05 (1.83) µmol/L] were detected. To verify whether the Hcy linked to apoprotein is proportional to Hcy plasma concentrations, we performed the univariate Pearson correlation between these variables in the samples from our volunteers. ApoB-bound Hcy was correlated with both total (r = 0.69; P <0.02) and reduced plasma Hcy (r = 0.78; P <0.01).
|
In conclusion, by modifying our previous assay, we have developed a method that is able to simultaneously measure all thiols linked to apoB-100 by a disulfide bond. We obtained evidence for the method suitability by measuring the concentrations of different thiols in 16 volunteers. For the first time we demonstrate that LDL carries not only Hcy but also Cys-Gly and Cys and that the latter is present largely in the LDL fraction. Moreover, we show that there is a positive relationship between plasma Hcy concentrations (both total and reduced) and Hcy linked to apoprotein. Even if the mechanism by which Hcy exerts its deleterious effects is unknown, several findings have suggested that oxidation of LDL is a key step in atherogenesis. The etiology of vascular disease resulting from hyperhomocysteinemia has been attributed to thiol autooxidation, a process that generates reactive oxygen species such as superoxide and hydrogen peroxide. It has been suggested that these damaging species are generated under conditions of hyperhomocysteinemia and that they lead to oxidative damage of LDL (15). The integrated study of hyperhomocysteinemia and lipoprotein metabolism may help us to better understand the deleterious effects of increased Hcy concentrations as they relate to the atherogenic thiolated LDL.
Acknowledgments
This study was supported by the "Assessorato dell’Igiene e Sanità Regione Autonoma della Sardegna", by the "Ministero dell’Istruzione, dell’Università e della Ricerca", and by the "Ministero della Sanità (Attività di Ricerca Finalizzata–2002)" (Italy). We greatly appreciate the assistance of Maria Antonietta Meloni with the language in the manuscript.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  G. L. Hortin, N. Seam, and G. T. Hoehn Bound Homocysteine, Cysteine, and Cysteinylglycine Distribution between Albumin and Globulins Clin. Chem., December 1, 2006; 52(12): 2258 - 2264. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Zinellu, E. Zinellu, S. Sotgia, M. Formato, G. M. Cherchi, L. Deiana, and C. Carru Factors Affecting S-Homocysteinylation of LDL Apoprotein B Clin. Chem., November 1, 2006; 52(11): 2054 - 2059. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
