Higher oxidation and lower antioxidant levels in peripheral blood plasma and bone marrow plasma from advanced cancer patients

Authors


Abstract

BACKGROUND

Bone marrow (BM) is an important tissue in the generation of immunocompetent and peripheral blood cells. The precursors of hematopoietic cells in BM undergo continuous proliferation and differentiation and are highly vulnerable to acute and chronic oxidative stress. Little is known about the oxidant and antioxidant status in the BM of untreated patients with nonhematologic tumors. In this study, oxidative stress was evaluated in peripheral blood plasma (PBP) and BM plasma (BMP) from lung carcinoma (LC) and breast carcinoma (BC) patients.

METHODS

The sample included 13 consecutive untreated LC patients, 15 BC patients, and 11 healthy controls. Luminol-dependent chemiluminescence was used to evaluate oxygen radical generation by peripheral blood neutrophils. Lipid oxidation, assessed by 2-thiobarbituric acid-reactive substances (TBARS), and α-tocopherol, β-carotene, and total ubiquinol-10 levels were determined in PBP and BMP.

RESULTS

In LC and BC patients, neutrophil chemiluminescence was higher (128% and 264%, respectively) than in controls (P < 0.05). In cancer patients, TBARS levels were higher in both PBP (51% and 243% for LC and BC patients, respectively) and BMP (66% and 305% for LC and BC patients, respectively) than in plasma from controls (P < 0.01). α-Tocopherol and total ubiquinol-10 levels were significantly lower in BMP from BC patients compared with controls. In BC patients, α-tocopherol content in PBP was significantly lower than in controls.

CONCLUSIONS

Untreated cancer patients presented an imbalance between oxidant generation and lipid-soluble antioxidant levels in favor of the former. Cancer 2002;94:3247–51. © 2002 American Cancer Society.

DOI 10.1002/cncr.10611

Reactive oxygen species (ROS), including free radicals and nonradical species, are generated as byproducts of normal cell metabolism. They are can be produced and act inside the cell, or they can be generated within the cell and released to the extracellular space.1 ROS may also arise from exogenous sources, like electromagnetic and particulate radiation, air pollutants, and tobacco smoke, or through the metabolism of certain drugs, like anthracyclic antineoplastic agents (doxorubicin), pesticides, and solvents.2 ROS can damage proteins, lipids, carbohydrates, and nucleic acids. It is well known that lipid oxidation induces a decrease of membrane fluidity and changes in receptor distribution. Oxidative alterations to carbohydrates can modify cellular receptor functions, and damage to DNA can cause mutations that may be carcinogenic. Finally, some products of free radical degradation of polyunsaturated fatty acids, such as malondialdehyde, can induce cross-linking in lipids, proteins, and nucleic acids.2 As a consequence, ROS levels must be controlled to avoid damage to cellular structures. To counteract ROS-induced damage, aerobic organisms rely on defenses, including enzymes (catalase, superoxide dismutase, glutathione peroxidase), as well as on small molecular weight antioxidants both of endogenous synthesis (glutathione, ubiquinol) and of dietary origin (vitamin C, vitamin E, carotenes). As a result, a balance is established between the generation and destruction of ROS, minimizing the damage to physiologically relevant biomolecules.1

Johnke et al.3 reported that tumor-bearing mice (Lewis lung carcinoma [LC] and EMT6 breast carcinoma [BC] tumors) showed a significant suppression of antioxidant enzyme activity in the hematopoietic marrow. This suggests that tumor development may influence the capacity of the bone marrow (BM) to tolerate ROS production.

In recent years, in vivo and in vitro studies have focused on the effects of ROS on BM injury, especially in cancer patients receiving high doses of chemotherapy and total body irradiation.4–6 After treatment for BM transplantation, free radical-trapping capacity decreases and levels of 2-thiobabituric acid-reactive substances (TBARS) increase markedly in peripheral blood plasma (PBP).7, 8 Early posttransplant organ toxicity may be due to ROS damage.9 This increase in ROS activity might underlie both a progressive hematopoietic suppression and alterations in the composition and function of the BM hematopoietic microenvironment.10 There are no reports on the oxidant/antioxidant status in BM plasma (BMP) from untreated patients with advanced solid tumors.

In the current study, we measured ROS generation by peripheral blood neutrophils, as well as lipid oxidation and the levels of α-tocopherol, β-carotene, and total ubiquinol-10 in PBP and BMP from untreated LC and BC patients.

PATIENTS AND METHODS

Patients

Consecutive untreated patients with non-small cell (squamous cell) LC (n = 13) of Stages IIIA, IIIB, and IV disease or with ductal BC (n = 15) of Stages III and IV disease and healthy controls (n = 11) comprised the study population. Healthy controls were donors for allogeneic BM transplantation. The mean age of BC and LC was 47.7 years (range 41–55) and 51.6 years (range 40–62), respectively, and the mean age of healthy controls was 37.6 years (range 30–46). The International Union Against Cancer TNM classification system was used. Informed consent was obtained from all the individuals included in these studies. This work was performed in accordance with the principles of the Declaration of Helsinki. The present investigations were approved by the British and I. Iriarte Hospital Ethical Committees.

Isolation of Neutrophils from Peripheral Blood

Venous blood was drawn into syringes containing ethylene-diamine tetraacetic acid (EDTA; 1.4 mg/mL) and subjected to 3% (w/v) dextran T-200 (Pharmacia, Uppsala, Sweden) sedimentation (30 minutes, 37 °C). This was followed by Ficoll-hypaque (Sigma, St. Louis, MO) centrifugation (25 minutes at 340 × g) and hypotonic lysis of the erythrocytes with 0.206% (w/v) Tris-0.83% (w/v) ammonium chloride, pH 7.2. The neutrophils were washed once with phosphate-buffered saline, pH 7.4, containing EDTA 0.056% (w/v), and twice with phenol red-Ca2+/Mg2+-free Hank's balanced salt solution (HBSS), pH 7.3. The final preparations were adjusted to 107 cells/mL in HBSS (95% of viable cells as determined by trypan blue dye exclusion). All the surfaces in contact with the neutrophils were plastic or siliconized glass.

Isolation of PBP and BMP

Peripheral venous blood was drawn into syringes containing preservative-free heparin (25 U/mL, Gibco, Life Technologies, Gaithersburg, MD) and centrifuged (400 × g) for 30 minutes to isolate the plasma. BM samples, obtained from the posterior iliac crest under local anesthesia, were collected into preservative-free heparinized saline (25 U/mL). BM was aspirated within minutes from the time the peripheral blood was obtained. The BM samples were centrifuged (400 × g) for 30 minutes to isolate BMP. PBP and BMP were frozen at −20 °C until use.

Determination of Luminol-Dependent Chemiluminescence by Nonstimulated Neutrophils

Chemiluminescence was used to evaluate ROS generation by nonstimulated neutrophils. Chemiluminescence was measured in the dark, at room temperature, in a liquid scintillation LS100 counter (Packard, Meriden, CT) set in the out-of-coincidence mode, according to a previously reported method.11 For this assay, 106 neutrophils were placed in plastic liquid scintillation vials containing 2 mL HBSS and incubated for 50 minutes at 37 °C. Forty minutes into the incubation time, 10 μL of a luminol (5-amino-2,3-dihydro-1,4-phtalazinedione, Sigma) solution (1.25 μg/mL) was added to each vial. The results are expressed as counts per minute (cpm). Luminol was dissolved in dimethylsulfoxide (Sigma) at 10 mg/mL. This stock solution was diluted to 1.25 μg/mL with HBSS before use.

Determination of Plasma TBARS

Plasma lipid oxidation was assessed by determining the level of TBARS using a fluorescence method.12 Before storage plasma samples (200 μL) were added with 100 μL of butylated hydroxytoluene (4% w/vol in ethanol). Results are expressed as micromole TBARS (malondialdehyde equivalents) per liter of plasma. Malondialdehyde standards were prepared from 1,1,3,3-tetramethoxypropane.

Determination of Lipid-Soluble Antioxidants

The levels of α-tocopherol, β-carotene, and total ubiquinol-10 in plasma were evaluated using high-performance liquid chromatography with electrochemical detection.13 Total ubiquinol-10 measurement includes ubiquinone and its reduction product, ubiquinol. Plasma aliquots (200 μL) were added with 500 μL of methanol, vortexed, and added with 4 mL of n-hexane. The mixtures were vortexed for 1 minute and then centrifuged for 5 minutes at 1000 g. A 3-mL aliquot of the hexane layer was dried under N2. The residue was dissolved in 0.2 mL of ethanol:methanol (1:1, v/v) and filtered through a 0.22-μm pore nylon membrane. The samples were chromatographed on a C8 reversed-phase column, and the antioxidant levels measured in a BAS LC4C amperometric detector (Analytical Bio System, West Lafayette, IN) with a glassy-carbon working electrode at an applied potential of +0.6 V. Pure commercial compounds (Sigma) were used as standards.

Statistics

Values in the text and tables are expressed as mean ± standard error (SE). Nonparametric Kruskall–Wallis statistics (Statview, version 5.0; SAS Institute, Cary, NC) were used to establish the significance of between-group differences. A P value of less than 0.05 was considered significant. Spearman correlation coefficient (rho) was used to establish the relationships between variables.

RESULTS

Neutrophil Chemiluminescence

Chemiluminescence was significantly higher in neutrophils from LC and BC patients than in those from controls: control = 36.6 ± 3.8; LC = 83.6 ± 20.8 (P < 0.05 vs. control); and BC = 133.4 ± 44.4 cpm × 10−3 (P < 0.05 vs. control).

TBARS in PBP and BMP

In cancer patients, TBARS levels were significantly higher in both PBP (51% and 243% for LC and BC patients, respectively) and BMP (66% and 305% for LC and BC patients, respectively) than in controls (P < 0.01; Table 1). In LC patients, BC patients, and controls, PBP TBARS levels correlated positively with those in BMP (Table 2). Both in the control and in the BC and LC patient groups, no correlation was found between stage of the disease, age, sex, smoking/nonsmoking habits and TBARS content in either PBP or BMP (data not shown).

Table 1. α-Tocopherol, β-Carotene, Total Ubiquinol-10, and TBARS Levels in Peripheral Blood Plasma (PBP) and Bone Marrow Plasma (BMP) from LC and BC Patients
Variable (μM)ControlLC patientsBC patients
  • LC: lung carcinoma; BC: breast carcinoma; TBARS: 2-thiobarbituric acid-reactive substances.

  • a

    Values are expressed as mean ± SE.

  • b

    P < 0.01 vs. control.

  • c

    P < 0.05 vs. LC patients.

α-tocopherol   
 PBP20.8 ± 2.024.4 ± 1.64.7 ± 1.4bc
 BMP18.6 ± 2.219.0 ± 2.66.8 ± 2.6bc
β-carotene   
 PBP0.09 ± 0.030.13 ± 0.040.10 ± 0.05
 BMP0.08 ± 0.030.10 ± 0.030.07 ± 0.04
Total ubiquinol-10   
 PBP1.06 ± 0.181.23 ± 0.220.70 ± 0.16c
 BMP1.42 ± 0.221.06 ± 0.230.57 ± 0.14b
TBARS   
 PBP2.67 ± 0.264.05 ± 0.40b9.16 ± 2.02bc
 BMP2.35 ± 0.193.90 ± 0.42b9.53 ± 1.55bc
Table 2. Oxidative Stress Variables: Correlation Coefficients between Values in Peripheral Blood Plasma and Bone Marrow Plasma
VariableSpearman correlation coefficient (rho)P value
  1. TBARS: 2-thiobarbituric acid-reactive substances.

TBARS0.660.0019
α-tocopherol0.610.0034
β-carotene0.640.0022
Total ubiquinol-100.840.0001

Determination of Lipid-Soluble Antioxidants in PBP and BMP

Table 1 shows the α-tocopherol, β-carotene, and total ubiquinol-10 values in PBP and BMP from LC and BC patients. In PBP from LC patients, the levels of α-tocopherol, β-carotene, and total ubiquinol-10 were unchanged relative to those found in controls. In BMP from LC patients, α-tocopherol, β-carotene, and total ubiquinol-10 levels were not different from those found in controls. In BC patients, α-tocopherol and total ubiquinol-10 levels in PBP were lower than in both controls (77% and 34%, respectively) and LC patients (81% and 43%, respectively) The levels of α-tocopherol in BMP from BC patients were significantly lower relative to both control and LC patient values (63% and 64%, respectively). Total ubiquinol-10 level in BMP from BC patients was signficantly lower compared with controls (60%). However, β-carotene content in PBP and BMP from BC patients was similar to that in controls. Significant positive correlations were observed among the levels of α-tocopherol, β-carotene, and total ubiquinol-10 in PBP and BMP (Table 2). Combined data from LC and BC patients showed that in both PBP and BMP, levels of TBARS were related inversely to lipid-soluble antioxidant (α-tocopherol plus β-carotene plus total ubiquinol-10) levels (PBP: rho = −0.61, P = 0.0198; BMP: rho = −0.74, P = 0.004).

Both in the control and in the BC and LC patient groups, no correlation was found between stage of the disease, age, sex, smoking/nonsmoking habits and lipid-soluble antioxidant content in either PBP or BMP (data not shown).

DISCUSSION

This article shows that in neutrophils from both LC and BC patients oxygen radical production is higher than in those from healthy controls. Braun et al.14 demonstrated that nonstimulated peripheral blood mononuclear cells from patients with advanced solid tumors had a significantly higher chemiluminescence emission than those from normal controls. The higher level of neutrophil chemiluminescence may be attributed to different factors, including immunologic reactions to tumor antigens and/or the stimulatory influences of tumor-derived products. It is well known that insoluble and soluble circulating immune complexes enhance chemiluminescence emission in neutrophils.15 Patients with advanced LC and BC show higher levels of circulating immune complexes in sera relative to healthy controls, as we reported previously.16

The current results show elevated levels of TBARS in PBP and BMP from untreated LC and BC patients compared with controls, suggesting a significant increase in both lipid oxidation and ROS production in peripheral blood as well as in BM. The increase in TBARS levels might be associated with the observed higher production of ROS in neutrophils. TBARS levels in PBP strongly correlate with those in BMP, suggesting that the determination of PBP TBARS might serve as an indicator of lipid oxidation taking place in BMP.

α-Tocopherol levels in PBP and BMP from untreated BC patients were significantly lower compared with levels in controls. BC patients showed a lower level of total ubiquinol-10 than controls, but the difference was significant only in BMP. In contrast, α-tocopherol, β-carotene, and total ubiquinol-10 levels in PBP and BMP from LC patients were similar to those found in controls. The level of TBARS in PBP and BMP from BC patients was significantly higher than in LC patient samples, in agreement with lower levels of lipid-soluble antioxidants in PBP and BMP from BCP patients. Based on these findings, we suggest that the deficiencies of α-tocopherol and total ubiquinol-10 levels in PBP and BMP from BC patients may underlie the significant increase of lipid oxidation occurring in these patients.

However, to further define the significant increase in the TBARS levels in PBP and BMP from LC patients, it may be necessary to determine other antioxidants, such as vitamin C, which can regenerate the oxidized forms of lipid-soluble antioxidants. In this context, the similarity between the levels of α-tocopherol, β-carotene and total ubiquinol-10 in PBP and BMP from LC patients and those found in controls might be explained by our failure to determine a broader range of antioxidants. Moreover, considering that the generation of ROS by neutrophils is, at least in part, dependent on immunologic reactions to antigens, the higher chemiluminescence and TBARS levels observed in BC patients compared with those in LC patients might be related to the specific type of neoplastic antigen present in the circulating immune complexes associated with BC. It is well known that lipid oxidation and ROS are implicated in carcinogenesis and might be etiologically involved in the promotional phase. The oxidative stress level found in BC patients suggests that tumor progression is worse than in LC patients.

Lipid oxidation in both PBP and BMP correlates negatively with lipid-soluble antioxidant levels. In addition, the levels of α-tocopherol, β-carotene, and total ubiquinol-10 in PBP correlate positively with those in BMP. This suggests that the determination of both lipid oxidation and antioxidant levels in PBP might indicate the oxidative stress status in BMP.

Dietary antioxidant supplementation increases the antioxidant levels in PBP.17, 18 In addition, nutritional antioxidant strategies can prevent the genotoxic effects that some chemotherapeutic agents produce in BM myelogenous progenitor cells.6

The current results suggest that the supplementation of untreated LC and BC patients with antioxidants might also increase antioxidant levels in BMP, possibly ameliorating the oxidant/antioxidant status in the BM. Given both the role of ROS in tumor promotion and the danger of increased ROS production under chemotherapy/ irradiation treatments, the current results underscore the importance of antioxidant therapy to protect cells from further damage.

Ancillary