Effects of homocysteine on apoptosis-related proteins and anti-oxidant systems in isolated human lymphocytes


F. Blandini, Laboratory of Functional Neurochemistry, Neurological Institute ‘C. Mondino’, Via Palestro 3, 27100 Pavia, Italy.
Fax: + 39 0382 380396, Tel.: + 39 0382 380333,
E-mail: fabio.blandini@mondino.it


Homocysteine (Hcy) is a nonprotein-forming sulphur amino acid that plays an important role in remethylation and trans-sulphuration processes. In recent years, it has been suggested that increased levels of plasma Hcy may play a role in the pathogenesis of various diseases, particularly at the cardiovascular level. The pathogenic mechanism of hyperhomocysteinemia, however, has not been clarified. Because oxygen radicals can be generated by the auto-oxidation of this amino acid, it has been suggested that Hcy may cause cellular damage through oxidative mechanisms, ultimately leading to apoptotic cell death. In this study, we sought to investigate the effects of Hcy on oxidative damage and antioxidant agent levels, as well as on apoptosis-related proteins and apoptosis occurrence in human cells. For this purpose, we measured levels of Bcl-2, caspase-3 and caspase-9 activity, Cu/Zn superoxide dismutase, reduced glutathione, lipid peroxidation [malondialdehyde and 4-hydroxy-2 (E)-nonenal concentrations], apoptotic single-stranded DNA and nuclear changes in human isolated lymphocytes exposed to increasing concentrations of Hcy. Incubation with Hcy did not induce significant changes in any of these biomarkers. Therefore, our results do not support the existence of a direct link between increased levels of Hcy and the occurrence of a pro-apoptotic state mediated by enhanced oxidative stress.


analysis of variance


Cu/Zn superoxide dismutase


reduced glutathione


single-stranded DNA




5,5′-dithiobis(2-nitrobenzoic acid)

Increased levels of homocysteine (Hcy), a thiol-containing amino acid involved in the methionine conservation cycle and in trans-sulfuration pathways, have been associated with increased risk for atherosclerotic and thrombotic vascular diseases [1–3]. Increased Hcy levels have also been reported in patients with neurodegenerative disorders, such as Alzheimer's disease [4,5] or Parkinson's disease [6–9], and the importance of a balanced Hcy metabolism for the maintenance of neuronal homeostasis has been recently pointed out [10].

Hcy concentrations may increase as a result of deficiency in folate [11], vitamin B6 or vitamin B12 [12,13], or because of genetic mutations, particularly within the genes that encode the enzymes methylenetetrahydrofolate reductase [14] and cystathionine β-synthase [15]. The cellular and molecular mechanisms underlying the adverse effects of hyperhomocysteinemia have not been fully elucidated. Various authors have suggested that the amino acid may act as a pro-oxidant agent; indeed, the metabolism and metal-catalysed auto-oxidation of Hcy are paralleled by the formation of oxygen-free species, which may play a role in the endothelial damage associated with hyperhomocysteinemia [16–19]. In addition, in vitro studies have reported that Hcy, particularly at high concentrations, induces apoptosis (programmed cell death) in various cell lines [20,21]. This may be related, again, to the putative pro-oxidant properties of the amino acid, as oxidative stress is a major trigger of apoptosis [22].

Peripheral blood lymphocytes have been extensively used to study the involvement of oxidative stress and/or apoptosis in the pathogenesis of numerous pathological conditions, including cardiovascular and neurodegenerative disorders [23–26], as they provide a convenient and accessible model for in vitro studies. In addition, there is evidence that the humoral and cellular immune systems play a role in atherogenesis and that hyperhomocysteinemia may intervene in the phenomenon by inducing B lymphocyte proliferation [27]. This effect would be mediated by the production of intracellular reactive oxygen species, which act as second messengers to regulate signal transduction pathways controlling gene expression and post-translational modification of proteins [27].

The aim of our study was to investigate, in vitro, whether Hcy induces oxidative stress and/or apoptotic cell death in human lymphocytes, by interfering with the regulatory mechanisms of these pathological conditions. For this purpose, we measured a number of oxidative stress-related markers, including lipid peroxidation, antioxidant enzyme Cu/Zn superoxide dismutase (Cu/Zn SOD) and reduced glutathione (GSH), in isolated human lymphocytes exposed to increasing concentrations of Hcy. In the same samples, we determined the effect of Hcy on a selected set of proteins that participate in the apoptotic process with the opposite role: Bcl-2, which regulates cell survival by counteracting apoptosis [28]; caspase-9, which initiates cell death signalling events [29]; and caspase-3, the major effector of the apoptotic cascade [30,31]. We also investigated the actual occurrence of apoptosis through the analysis of the typical changes in DNA and membrane permeability associated with the phenomenon.

Materials and methods


d,l-homocysteine, RPMI-1640 medium (modified with 20 mmol·L−1 Hepes and l-glutamine, without NaHCO3), stabilized penicillin–streptomycin solution (10 000 U·mL−1 penicillin and 10 mg·mL−1 streptomycin), NaCl/Pi tablets, Ellman's reagent [5,5′-dithiobis(2-nitrobenzoic acid); (DTNB)], campthothecin, methanol and Trypan blue were purchased from Sigma. Foetal bovine serum was from BioWhittaker Inc., Yopro-1 and propidium iodide dyes were provided by Molecular Probes Inc.

Sample preparation

Venous blood samples (30 mL) were obtained from 10 healthy volunteers ranging in age between 26 and 33 years (mean ± SD, 29 ± 2.5 years). Experiments were carried out with the understanding and written consent of each subject. Blood was collected from the antecubital vein in vacuum tubes containing EDTA (Terumo Europe, Belgium) and immediately centrifuged at low speed (250 g, 15 min, room temperature) to remove platelets. After removal of the platelet-rich plasma layer, the original volume was reconstituted with NaCl/Pi pH 7.4, and the mixture layered over half of its volume of Lympholite®-H. The discontinuous gradient thus formed was centrifuged at 450 g for 20 min, at room temperature. After removal of the upper layer, the lymphocyte band at the interface was removed and washed twice with NaCl/Pi. Isolated lymphocytes were counted using an automated cell counter (MAX-M, Beckman Coulter Inc.), and resuspended in modified RPMI-1640 supplemented with 10% heat-inactivated foetal bovine serum and 0.5% penicillin–streptomycin solution. Different aliquots (each containing 2 × 106 lymphocytes·mL−1) were prepared from each sample in 15 mL sterile tubes. Increasing amounts of d,l-Hcy dissolved in water were added, in duplicate, to the samples to obtain final concentrations of 10, 100 and 1000 µm. Isolated lymphocytes were then incubated for 24 h at 37 °C in the dark. In order to have positive controls for the apoptosis study, subsets of lymphocytes were incubated with 10 µm campthothecin, an apoptosis inducer, for 4 h at 37 °C. Further evaluation of the apoptotic process was carried out on aliquots of lymphocytes incubated with d,L-Hcy at the highest concentration (1000 µm).

After incubation, cell viability was verified using the Trypan blue dye exclusion method and confirmed by propidium iodide fluorescence microscopy (Axioskope-2, Zeiss, Göttingen, Germany) [32].

Aliquots to be used for the assays of oxidative stress markers and apoptosis regulatory proteins were pelleted with a final centrifugation at 1000 g for 10 min and frozen, after discarding the supernatant, at −80 °C, until the assays were carried out; the aliquots for the DNA analysis and membrane permeability study underwent different treatments, according to the specific procedures (see below).


It has been reported that when thiol compounds, including Hcy, are added to commonly used cell culture media, a rapid, time-dependent loss of detectable –SH groups occurs, thus reducing the biological effects of the compounds being tested [33]. To assess the potential influence of such a phenomenon, in a preliminary experiment we added 100 or 1000 µm Hcy to aliquots of the medium used in this study (modified RPMI-1640, 10% heat-inactivated foetal bovine serum, 0.5% penicillin–streptomycin solution). Aliquots were incubated for increasing periods of time (up to 24 h) and free reduced Hcy levels were determined spectrophotometrically using the traditional Ellman's reagent (DTNB) and measuring the absorbance at 412 nm.

For the determination of the oxidative stress markers and apoptosis regulatory proteins, frozen lymphocyte pellets were resuspended in 0.6 mL ice-cold NaCl/Pi and homogenized by ultrasound (Vibra Cell, Sonics & Materials, Inc., Danbury, CT, USA). Homogenates were then centrifuged at 15 000 g, for 10 min at room temperature, and the resulting supernatants were used for the biochemical assays. Cu/Zn SOD and Bcl-2 were assayed using two commercially available ELISA kits (Bender MedSystems Diagnostics GmbH, Vienna, Austria) on a 550 Model Bio-Rad Microplate Reader (440 nm) (Bio-Rad Laboratories). Fluorometric microtitration was performed for the determination of GSH levels (Chemicon International Inc.), caspase-3 (Molecular Probes Inc.) and caspase-9 (Oncogene) protease activities, using a fluorimetric microplate reader SpectraMax Gemini XS (Molecular Devices Co.). Lipid peroxidation was evaluated using a colorimetric assay kit for the detection of malondialdehyde and 4-hydroxyalkenals, which are generated by the oxidation of polyunsaturated fatty acids in cell-membrane phospholipids [34] (Calbiochem).

Apoptosis detection was carried out using the fluorescent probe Yopro-1 iodide, which selectively binds to apoptotic nuclei [32], and by measuring the formation of single-stranded (ss) DNA. For Yopro-1 labelling, the cell suspension was added directly with 5 µm of the fluorescent dye and scanned by microtitration at 485/530 nm excitation/emission wavelengths [32]. For the ssDNA assay, lymphocytes were fixed with methanol (80% methanol in NaCl/Pi) at room temperature for 30 min and then transferred directly to a microtiter plate, according to the ssDNA Apoptosis ELISA Kit (Chemicon International Inc.); this procedure is based on the selective, formamide-induced denaturation of DNA, which identifies apoptotic cells [35,36] and subsequent staining of ssDNA, using a mixture of anti-ssDNA mAb and peroxidase-conjugated antimouse IgM.

Plasma levels of total Hcy in the volunteers were measured by HPLC with fluorometric detection, using a commercially available kit (Bio-Rad). The HPLC system consisted of an isocratic pump (Agilent 1100 series, Agilent Technologies, Waldbronn, Germany) equipped with an analytical column (70 × 3.2 mm i.d.) and a precolumn (microguard™; Bio-Rad). Excitation and emission wavelengths on the fluorometric detector (Jasco Corp.) were 385 and 515 nm, respectively. Data obtained from the detector were collected and integrated by a dedicated personal computer, equipped with a chromatography software package and an instrument interface (clinical data managementSystem, Bio-Rad).

Statistical analysis

One-way analysis of variance (ANOVA) and Fisher's post hoc test were used to evaluate the effect of increasing concentrations of Hcy on the lymphocyte levels of the variables considered in this study. The minimum level of statistical significance was set at P < 0.05.

Results and Discussion

In this study, we assessed the direct influence of increasing concentrations of Hcy on various mechanisms that subserve cellular and apoptotic processes in isolated human lymphocytes. The mean (± SD) basal level of total plasma Hcy in the subjects participating in the study was 6.8 ± 1.3 µm (range, 5.6–8.9 µm). A preliminary experiment, conducted to verify the time-dependent changes of free reduced Hcy levels in the medium (RPMI), showed a progressive decrease, peaking at 24 h (−90%), when the medium was spiked with 100 µm Hcy; however, when 1000 µm Hcy was added to the medium, only a moderate (−40%), late reduction was found after incubation (Table 1). This partially confirmed previous observations [33], but also excluded an early, significant loss of Hcy from the medium, particularly when using the highest concentration (1000 µm).

Table 1. Determination of free reduced Hcy levels (µm) in RPMI aliquots spiked with 100 or 1000 µm Hcy and incubated for increasing periods of time (up to 24 h).
Time100 µm Hcy1000 µm Hcy
1 h 53 906
2 h 49 884
4 h 30 840
8 h 24 786
12 h 15 734
24 h 10 603

As shown in Table 2, the 24-h incubation of isolated lymphocytes with increasing concentrations of Hcy (10, 100 and 1000 µm) did not induce statistically significant changes in any of the variables considered. In fact, we found that Hcy does not modify the lymphocyte levels of GSH, while inducing modest, not significant increases in the levels of Cu/Zn SOD or lipid peroxidation when lymphocytes were incubated with 100 µm Hcy. Similarly, a slight increase was observed in the activity of caspase-3, a major effector of apoptosis, for Hcy concentrations of 10 µm, while intracellular levels of antiapoptotic protein Bcl-2 were unchanged. We also failed to demonstrate any significant Hcy-induced increase in the lymphocyte levels of ssDNA, a specific biochemical marker of ongoing apoptosis [36–38].

Table 2. Mean (± SD) levels of Bcl-2, caspase-3 activity, Cu/Zn SOD, GSH, MDA + 4-HNE (lipid peroxidation) and ssDNA in human isolated lymphocytes, incubated for 24 h in the presence of increasing concentrations of homocysteine. The last row shows changes in apoptosis-related markers (mean ± SD) following incubation with 10 µm campthothecin (apoptosis inducer). ly, lymphocytes; AMC, 7-amino-4-methylcoumarin. ANOVA: Caspase-3: F = 23.9, P < 0.001; ssDNA: F = 3.0, P < 0.05. Fisher's post hoc test vs. untreated cells (mean ± SD): * P < 0.05 ** P < 0.001.
(U·106 ly−1)
(pmol AMC·106 ly−1)
(ng·106 ly−1)
(nmol·106 ly−1)
(pmol·106 ly−1)
(ng·106 ly−1)
014.6 ± 2.0879.2 ± 97.567.7 ± 11.43.6 ± 0.411.3 ± 7.0872.4 ± 94.1
1015.3 ± 1.1974.4 ± 170.963.4 ± 8.03.8 ± 0.510.1 ± 1.3864.3 ± 37.0
10015.3 ± 0.6893.3 ± 186.881.8 ± 32.53.6 ± 0.612.9 ± 4.8942.4 ± 51.1
100015.7 ± 2.0814.4 ± 179.170.3 ± 12.93.7 ± 0.65.7 ± 1.5878.5 ± 58.7
Campthothecin (10 µm) 1744.1 ± 224.6**   961.9 ± 18.7*

Analogously, incubation of lymphocytes with 1000 µm Hcy did not cause significant modifications of Yopro-1 fluorescence or increases in caspase-9 activity (Table 3).

Table 3. Caspase-9 activity and Yopro-1 fluorescence, expressed as arbitrary fluorescence units (AFU), in human isolated lymphocytes incubated for 24 h in the presence of 1000 µm Hcy or 10 µm campthothecin (apoptosis inducer). ANOVA: Caspase-9 F = 608.94, P < 0.001; Yopro-1 F = 11.62, P < 0.005. Fisher's post hoc test vs. untreated cells (mean ± SD): * P < 0.05 ** P < 0.001.
d,l-Hcy (µm)Caspase-9 (AFU)Yopro-1 (AFU)
022462 ± 256.2114.4 ± 5.3
100018972 ± 556.8109.8 ± 17.6
Campthothecin (10 µm)38670 ± 907.2**141.4 ± 7.2*

Our findings seem, therefore, to contradict the assumption that hyperhomocysteinemia exerts its cytotoxic effects by interacting with the apoptotic regulatory mechanisms and/or increasing cellular levels of oxidative stress, as previously suggested by in vitro and in vivo studies [16–21,39–42]. The reasons for this discrepancy are unclear. One possible explanation is that most of the diseases associated with hyperhomocysteinemia involve the cardiovascular system. In fact, it has recently been reported that the endothelial and smooth muscle cells of the cardiovascuar system may have limited ability to process Hcy, due to the low expression of cystathionine β-synthase [42], the key enzyme in the catabolic transformation of Hcy into cysteine [43]. The impairment of such an important Hcy removal system may account for an increased susceptibility of endothelial and smooth muscle cells, compared to lymphocytes. Zhang et al. have recently reported that, in isolated murine splenic T lymphocytes, exposure to increasing Hcy concentrations enhances the production of reactive oxygen species induced by Concanavalin A, but, on the other hand, reduces Concanavalin A-induced apoptosis [44]. In addition, Fenech [45] has reported that the micronucleus frequency, an index of genetic damage, in human lymphocytes is positively correlated with plasma Hcy and inversely correlated with plasma vitamin B12. Therefore, it may be hypothesized that combination of increased Hcy levels with other concurrent agents (reduced vitamin B12 and/or folate levels, for example) is required for the intracellular damage to occur, at least in lymphocytes.

It must also be noted that about 99% of circulating Hcy exists in various oxidized forms, indistinguishable by most Hcy assays. Therefore, the cytotoxic potential of Hcy may be due to alternative forms of Hcy. However, although free reduced Hcy represents the minor circulating component, it may reliably be considered the most active form for the vascular endothelial function in vivo[46]. In addition, aminothiol species, including Hcy, comprise a dynamic system referred to as ‘redox thiol status’, which is involved in the extracellular antioxidant defence system. Hyperhomocysteinemia may cause changes in the redox thiol status and imbalance between pro-oxidant and antioxidant process [47,48]. On the other hand, the pro-oxidant potential of homocysteine itself has been recently questioned. Zappacosta et al. [49], for example, have shown that the oxidative catabolism of Hcy does not produce significant amounts of H2O2, as previously reported by other authors [50–52] and Sengupta et al. have further questioned the role of circulating Hcy as a substantial source of reactive oxygen species [53,54]; moreover, Hcy proved able to counteract the effects of powerful oxidizing species, such as hypochlorite, peroxynitrite and ferrylmyoglobin, in addition to counteracting, as mentioned above, Concanavalin A-induced apoptosis [44]. This is not entirely surprising if one considers that thiolic compounds, such as Hcy itself, are generally considered to be powerful antioxidants [55]. Another recent study, which confirmed the association between hyperhomocysteinemia and coronary artery disease, showed also enhanced plasma levels of malondialdehyde, an index of lipid peroxidation, in these patients. The authors, however, failed to demonstrate any direct correlation between Hcy and malondialdehyde levels [56].

All of these considerations have led to the hypothesis that hyperhomocysteinemia, rather than playing a causative role, may simply be a marker for tissue damage or repair, particularly at the cardiovascular level [57]. The Hcy increase associated, for example, with coronary artery disease may be therefore an epiphenomenon of the disease itself [58,59].

In conclusion, the incubation of human, isolated lymphocytes with increasing concentrations of Hcy did not induce significant changes in the intracellular levels of caspase-3 and 9, Bcl-2, Cu/Zn SOD, GSH or lipid peroxides, or in the occurrence of apoptosis. These data seem therefore to question the hypothesized influence of Hcy, at least as an isolated agent, on the regulatory mechanisms of oxidative stress and apoptotic cell death, which have been suggested to underlie hyperhomocysteinemia-related citotoxicity.


This study was supported by grant ICS030.9 (Italian Ministry of Health). The authors thank R. Fancellu for his assistance in data analysis.