Logout

Masahiro Irie^a^, ^,, Masakazu Miyata^b^, ¹ and Hiroshi Kasai^c^, ²

^aSecond Department of Health Sciences (Clinical Science and Psychology), Institute of Health Science, Kyushu University, 6-1 Kasuga Park, Kasuga, Fukuoka 816-8580, Japan
^bHealth Center, Fukuoka University of Education, 729-1 Akama, Munakata, Fukuoka 811-4192, Japan
^cDepartment of Environmental Oncology, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan

Received 20 August 2004; revised 16 December 2004; accepted 2 January 2005. Available online 26 April 2005.

1. Introduction

Depression is the most common psychiatric disorder among patients with cancer. It is well known and documented that cancer patients are likely to have depressive symptoms after a diagnosis of cancer or during the clinical course of cancer (McDaniel et al., 1995 and Cella et al., 1989). For example, the prevalence of depression in cancer patients has been reported to range from 1.5% up to 50% at various times in its clinical course (McDaniel et al., 1995). Because of such an association, depression has been considered to have an impact not only on the progression, but also on the initiation, of cancer (Dalton et al., 2002). There have been several prospective studies in which the relationship between depression and cancer initiation was observed; however, results have remained inconclusive (Persky et al., 1987, Hahn and Petitti, 1988, Kaplan and Reynolds, 1988, Bleiker et al., 1996, Penninx et al., 1998, Gallo et al., 2000 and Jacobs and Bovasso, 2000).

Certain methodological limitations may have contributed to the conflicting results. First, the time-lag between the time of psychological investigation and the onset of cancer could have affected the results. It is often difficult to confirm whether a stress assessment actually preceded the initiation of cancer, because the transformation of normal cells into clinically apparent malignant cells can take years. Second, whether a single psychological assessment, as is used in most studies (Persky et al., 1987, Hahn and Petitti, 1988, Kaplan and Reynolds, 1988, Bleiker et al., 1996 and Gallo et al., 2000), can reflect long-term psychological state is unknown. Penninx and co-workers (1998) highlighted this problem and found that older persons, who suffered from depressive moods over a 6-year period, had a generally increased risk of cancer. Third, many epidemiological studies have focused their attention on the direct relationships between depression and cancer (Persky et al., 1987, Hahn and Petitti, 1988, Kaplan and Reynolds, 1988, Bleiker et al., 1996, Penninx et al., 1998, Gallo et al., 2000 and Jacobs and Bovasso, 2000). On the other hand, genetic examination relevant to cancer initiation at the time of psychological assessment is very limited. Glaser, Kiecolt-Glaser, and their associates have investigated stress-related genetic changes and have found that stress was associated with poor DNA repair in X-irradiated lymphocytes of high-distress psychiatric inpatients (Kiecolt-Glaser et al., 1985), the suppression of synthesis of a DNA repair enzyme, methyltransferase, in rats (Glaser et al., 1985), the decrease of interleukin 2 receptor mRNA expression and interleukin 2 production in leukocytes of medical students (Glaser et al., 1990), the inhibition of apoptosis in leukocytes exposed to γ irradiation in medical students (Tomei et al., 1990), and the decrease in mRNA expression of proto oncogenes, c-myc and c-myb, in leukocytes of medical students (Glaser et al., 1993). Examination stress was used as an academic stress model for the medical students. Those pioneers have provided valuable suggestions regarding stress–gene interaction; however, other pathways, particularly carcinogenic genetic alterations under natural and ordinary stress conditions, remain unclear.

Reactive oxygen species (ROSs), produced in the ordinary course of human life, have been shown to possess a potential role in the initiation, promotion, and progression of cancer (Hursting et al., 1999 and Cooke et al., 2003). One of the base modifications due to oxidative stress is 8-hydroxydeoxyguanosine (8-OH-dG, 7,8-dihydro-8-oxoguanosine) (Kasai and Nishimura, 1984). It has been reported that this oxidative DNA damage indicates both mutagenesis and carcinogenesis, and has been considered to be useful in estimating cancer risk. Indeed, chemical carcinogens, tumor-promoting agents, and lifestyles, such as smoking and drinking, have been reported to induce the formation of 8-OH-dG in various target tissues (Kasai, 1997 and Kasai, 2002), and increased levels of 8-OH-dG were also found in cancerous tissues (Olinski et al., 1992 and Inoue et al., 1998). Owing to these findings, the formation of 8-OH-dG seems to be a valuable biomarker to evaluate the possible stress–cancer linkage under natural and ordinary stress conditions. We recently provided the first evidence that human psychological stress can be associated with increased 8-OH-dG levels in peripheral leukocytes of healthy workers, although gender difference was found (Irie et al., 2001 and Irie et al., 2002). The 8-OH-dG levels in females were positively associated with negative moods, such as depression, and inadequate stress coping strategies, whereas those in males were positively related to stressful life changes and long working hours. Among these psychological factors, depressive state was most strongly linked to the 8-OH-dG levels. In addition, we suggested that depressive state is related to the 8-OH-dG levels in leukocytes of females, at least possibly via stress-related increase in the number of neutrophils and the 8-OH-dG production from the neutrophils (Irie et al., 2003). However, whether clinical depression is really associated with 8-OH-dG production remains unknown.

The present study aimed to clarify the possible association between clinical depression and the 8-OH-dG levels in peripheral leukocytes. First, we examined the 8-OH-dG levels in patients with depression, and compared them with age- and gender-matched healthy controls. Second, we examined the factors, including depression, which significantly contributed to the 8-OH-dG levels. We also examined them in view of gender difference, since we found a gender difference in the relationships between psychological factors and leukocyte 8-OH-dG levels in our previous study (Irie et al., 2001, Irie et al., 2002 and Irie et al., 2003).

2. Materials and methods

2.1. Design and subjects

After approval by the ethics committee, subjects were recruited in a certain hospital and a corporation through announcements asking for volunteers to participate in a study concerning depression and cancer risk via oxidative DNA damage. The participants in this study were 30 outpatients (20 males and 10 females) with depression, according to the Diagnostic and Statistical Manual of Mental Disorders, fourth revision (DSM-IV) (American Psychiatric Association, 1994). Sixty age- and gender-matched healthy subjects were selected as controls in consideration of the matching ratio (1:2). The controls were randomly selected from healthy workers in our previous study (Irie et al., 2001 and Irie et al., 2003). The patients, except for one, were administered antidepressants and/or minor tranquilizers. After obtaining written informed consent for the present study from these subjects, an examination was performed with a self-administered questionnaire. Items included in the questionnaire were: the Center for Epidemiologic Studies Depression scale (CES-D) (Radloff, 1977), the Profile of Mood States (POMS) (McNair et al., 1971), and lifestyles regarding sleep, smoking, drinking, exercise, and eating, including the seven health promoting practices reported by Berkman and Breslow (1983). The practices focus on hours of sleep, appropriate body weight, regularity of exercise, moderate alcohol consumption, smoking abstinence, regularity of breakfast, and moderation of between-meal snacks. The CES-D scale is a 20-item self-report measure developed by the National Institute of Mental Health Center for Epidemiological Studies to assess depressive symptomatology in the past week (Radloff, 1977). The POMS, a 65-item measure of general emotional disturbance in the past week, contains six factorially derived subscales (Tension-Anxiety, Depression-Rejection, Anger-Hostility, Vigor-Activity, Fatigue-Inertia, and Confusion-Bewilderment). In addition to the six subscale scores, the Total Mood Disturbance (TMD) score, a single global estimate of emotional state, was derived by summing the scores across all six factors, weighting Vigor negatively. Those who had any diseases, including cancer, which had been reported to be associated with increased 8-OH-dG levels (Kasai, 1997 and Kasai, 2002), were excluded from this study.

At the time of the questionnaire survey, blood samples were collected on weekday mornings after fasting overnight. The amount of 8-OH-dG was measured by high performance liquid chromatography (HPLC) coupled to an electrochemical detector (ECD). Then, examinations were carried out about the relationships between information obtained from the questionnaire and the 8-OH-dG levels in peripheral leukocytes.

2.2. Determination of 8-OH-dG levels in leukocytes

The method of analyzing the formation of 8-OH-dG has been described elsewhere (Irie et al., 2001). In brief, the extraction of nuclear DNA from peripheral leukocytes was carried out using the DNA Extractor WB Kit (Wako Biochemicals, Osaka, Japan). The extracted samples were digested with nuclease P₁ (0.8 U, Yamasa Co., Chiba, Japan) and acid phosphatase (1 U, Sigma Chemical Co.) in a solution of 1 mM EDTA and 10 mM sodium acetate (pH 4.5). This mixture was incubated at 37 °C for 30 min, and the iron exchange resin Muromac (Muromachi Kagaku, Tokyo, Japan) was added to remove the NaI. The mixture was centrifuged at 15,000 rpm for 5 min. After the supernatant was transferred to an Ultrafree Probind filter (Millipore Co., Bedford, MA), it was centrifuged at 10,000 rpm for 2 min. The filtered deoxynucleoside was injected onto the HPLC (Beckman; Ultrasphere-ODS; 5 μm, 4.6 × 250 nm) coupled to the ECD (ESA Coulochem II: guard cell; 0.35 V, detector 1, 0.15 V, detector 2, 0.30 V). Standard samples of dG (0.5 mg/ml) and 8-OH-dG (5 ng/ml) solutions (Wako Biochemicals) were compared with the samples of the subjects. The molar ratio of 8-OH-dG to dG was calculated based on the integrated peak area of authentic 8-OH-dG with an electrochemical detector and UV absorbance of dG using Millennium software (Waters, Milford, MA). The amount of 8-OH-dG is shown as the number per 10⁵ deoxyguanosine residues (/10⁵ dG).

2.3. Statistical analysis

All the statistical analyses were conducted with SPSS 11.5 for Windows. The CES-D scores, the POMS scores, the 8-OH-dG levels, and demographic and lifestyle factors were compared between the patients and the controls using Student’s t test, χ² test, or Mann–Whitney U test. The factors which showed significant associations with the 8-OH-dG levels were analyzed using non-parametric Mann–Whitney U test or Spearman’s rank correlation. A multiple regression analysis using a stepwise method was then performed to evaluate how the 8-OH-dG level as the dependent variable was determined by the independent variables, such as age, gender, BMI, smoking, drinking, exercise, the CES-D score, or the Depression-Rejection score, and whether the subjects were patients or controls, since these factors had been reported to be related to the 8-OH-dG levels in peripheral leukocytes (Kasai, 1997, Kasai, 2002 and Irie et al., 2001). The 8-OH-dG levels showed a skewed distribution, in which only a few cases were observed at higher levels and more were found at lower levels. Therefore, the logarithms of 8-OH-dG were used in the multiple regression analysis. In addition, the effect of total mood disturbance, including depression, on the 8-OH-dG levels was examined using the TMD score, instead of the CES-D score or the Depression-Rejection score, as the independent variable in the analysis. There were high correlations in the scores of the POMS, so, to avoid collinearity, the TMD score, which represents several kinds of negative aspects of the POMS, was used in the analysis. A p value less than 0.05 (two-tailed) was considered significant.

3. Results

Comparisons of demographic, lifestyle, and psychological variables and 8-OH-dG levels between depressive patients and healthy controls are shown in Table 1. Mann–Whitney U test was used to assess the differences in the number of health practices and the 8-OH-dG levels. The Tension-Anxiety, Depression-Rejection, Anger-Hostility, Fatigue-Inertia, and TMD scores of the POMS and the CES-D score of the patients were significantly higher, whereas the Vigor-Activity score of the POMS was significantly lower than those in the controls. Alcohol drinking habit was significantly frequent in the controls compared to the patients. The 8-OH-dG levels in the subjects of this study ranged from 0.08 to 2.70, with a mean of 0.37 ± 0.33 (8-OH-dG/10⁵ dG, mean ± SD). The 8-OH-dG levels in the patients were significantly higher than those in the controls.

Table 1.

Comparisons of demographic, lifestyle, and psychological variables and 8-OH-dG levels between depressive patients and healthy controls

Variables Patient (n = 30) Control (n = 60)

Age (yr) 49.4 ± 13.7 48.1 ± 11.7

Gender (% male) 66.7 66.7

Marital status (% married) 80 85

Body mass index (kg/h²) 23.1 ± 3.5 22.5 ± 2.4

Number of health practices 2.9 ± 2.0 3.4 ± 1.5

Smoking (% current) 43.3 31.7

Drinking (% current) 36.7 66.7^⁎⁎

Exercise (% current) 10 25

CES-D score 21.4 ± 11.3 13.7 ± 6.5^⁎⁎

POMS score

Tension-Anxiety 17.1 ± 7.5 11.5 ± 4.8^⁎⁎

Depression-Rejection 21.5 ± 15.3 11.9 ± 7.8^⁎⁎

Anger-Hostility 17.9 ± 11.8 12.8 ± 6.8^⁎

Vigor-Activity 10.5 ± 5.2 14.1 ± 5.6^⁎⁎

Fatigue-Inertia 14.0 ± 8.0 8.2 ± 4.6^⁎⁎

Confusion-Bewilderment 11.0 ± 6.7 8.6 ± 3.5

Total mood disturbance 71.0 ± 48.7 38.8 ± 25.5^⁎⁎

8-OH-dG level (/10⁵ dG) 0.53 ± 0.52 0.29 ± 0.12^⁎⁎

Statistical analyses were carried out using Student’s t test, χ² test, or Mann–Whitney U test (number of health practices and 8-OH-dG level). Data were expressed as mean ± SD.
^⁎ p < 0.05.
^⁎⁎ p < 0.01, significantly different between patients and controls.

Variables	Patient (n = 30)	Control (n = 60)
Age (yr)	49.4 ± 13.7	48.1 ± 11.7
Gender (% male)	66.7	66.7
Marital status (% married)	80	85
Body mass index (kg/h²)	23.1 ± 3.5	22.5 ± 2.4
Number of health practices	2.9 ± 2.0	3.4 ± 1.5
Smoking (% current)	43.3	31.7
Drinking (% current)	36.7	66.7^⁎⁎
Exercise (% current)	10	25
CES-D score	21.4 ± 11.3	13.7 ± 6.5^⁎⁎
POMS score
Tension-Anxiety	17.1 ± 7.5	11.5 ± 4.8^⁎⁎
Depression-Rejection	21.5 ± 15.3	11.9 ± 7.8^⁎⁎
Anger-Hostility	17.9 ± 11.8	12.8 ± 6.8^⁎
Vigor-Activity	10.5 ± 5.2	14.1 ± 5.6^⁎⁎
Fatigue-Inertia	14.0 ± 8.0	8.2 ± 4.6^⁎⁎
Confusion-Bewilderment	11.0 ± 6.7	8.6 ± 3.5
Total mood disturbance	71.0 ± 48.7	38.8 ± 25.5^⁎⁎
8-OH-dG level (/10⁵ dG)	0.53 ± 0.52	0.29 ± 0.12^⁎⁎

Table 2 shows the relationships between the demographic and lifestyle categorical variables and the 8-OH-dG levels in both and total groups. The difference in the 8-OH-dG levels between the patients and the controls was significant in males, but not in females. The regularly exercising controls showed significantly higher 8-OH-dG levels compared to their counterparts. Other demographic and lifestyle variables were unrelated to the 8-OH-dG levels.

Table 2.

Relationships between demographic and lifestyle categorical variables and 8-OH-dG levels in depressive patients and healthy controls

Demographic and lifestyle variables Category Patient (n = 30) Control (n = 60) Total (n = 90)

Gender Male 0.48 ± 0.08 0.31 ± 0.02^⁎⁎ 0.37 ± 0.03

Female 0.63 ± 0.24 0.27 ± 0.02 0.39 ± 0.09

Marital status Married 0.53 ± 0.12 0.32 ± 0.02 0.39 ± 0.04

Unmarried 0.53 ± 0.12 0.27 ± 0.04 0.38 ± 0.06

Smoking Current 0.57 ± 0.12 0.33 ± 0.04 0.43 ± 0.06

Non 0.50 ± 0.14 0.28 ± 0.01 0.34 ± 0.04

Drinking Current 0.43 ± 0.07 0.29 ± 0.02 0.32 ± 0.02

Non 0.58 ± 0.15 0.31 ± 0.02 0.45 ± 0.08

Exercise Current 0.30 ± 0.06 0.34 ± 0.03^⁎ 0.34 ± 0.03

Non 0.55 ± 0.10 0.27 ± 0.02 0.38 ± 0.05

^⁎ p < 0.05.
^⁎⁎ p < 0.01, significantly different between categories by Mann–Whitney U test. mean ± SE.

Demographic and lifestyle variables	Category	Patient (n = 30)	Control (n = 60)	Total (n = 90)
Gender	Male	0.48 ± 0.08	0.31 ± 0.02^⁎⁎	0.37 ± 0.03
	Female	0.63 ± 0.24	0.27 ± 0.02	0.39 ± 0.09
Marital status	Married	0.53 ± 0.12	0.32 ± 0.02	0.39 ± 0.04
	Unmarried	0.53 ± 0.12	0.27 ± 0.04	0.38 ± 0.06
Smoking	Current	0.57 ± 0.12	0.33 ± 0.04	0.43 ± 0.06
	Non	0.50 ± 0.14	0.28 ± 0.01	0.34 ± 0.04
Drinking	Current	0.43 ± 0.07	0.29 ± 0.02	0.32 ± 0.02
	Non	0.58 ± 0.15	0.31 ± 0.02	0.45 ± 0.08
Exercise	Current	0.30 ± 0.06	0.34 ± 0.03^⁎	0.34 ± 0.03
	Non	0.55 ± 0.10	0.27 ± 0.02	0.38 ± 0.05

Table 3 shows the correlations of the amount of 8-OH-dG to continuous variables in both and total groups, such as age, BMI, the number of seven health practices, the CES-D score, and the POMS scores. There was a significant positive correlation between the CES-D scores and the 8-OH-dG levels in the patients and total subjects. The Tension-Anxiety, Depression-Rejection, Anger-Hostility, Fatigue-Inertia, Confusion-Bewilderment, and TMD scores of the POMS indicated positive and significant correlations to the 8-OH-dG levels in total subjects. When the subjects were analyzed further according to their gender, female patients, but not male ones, showed positive associations between the Tension-Anxiety, Confusion-Bewilderment, and TMD scores of the POMS and the 8-OH-dG levels.

Table 3.

Spearman’s correlations between continuous variables and 8-OH-dG levels

Variables Patient (n = 30) Control (n = 60) Total (n = 90)

Female (n = 10) Male (n = 20) Total Female (n = 20) Male (n = 40) Total

Age (yr) −0.20 0.11 0.08 0.22 0.06 0.11 0.11

Body mass index (kg/h²) 0.21 −0.02 0.12 0.12 0.15 0.15 0.16

Number of health practices −0.63 −0.10 −0.31 0.11 −0.18 −0.09 −0.20

CES-D score 0.67 0.08 0.43^⁎ 0.12 −0.04 0.02 0.24^⁎

POMS score

Tension-Anxiety 0.71^⁎ −0.17 0.22 0.22 0.13 0.17 0.36^⁎⁎

Depression-Rejection 0.69 0.06 0.40 0.08 0.02 0.05 0.30^⁎⁎

Anger-Hostility 0.61 0.01 0.23 0.05 0.13 0.11 0.23^⁎

Vigor-Activity −0.26 −0.16 −0.18 −0.33 0.16 0.00 −0.15

Fatigue-Inertia 0.60 −0.04 0.31 −0.09 0.03 0.03 0.30^⁎⁎

Confusion-Bewilderment 0.76^⁎ −0.10 0.35 −0.10 0.02 0.01 0.23^⁎

Total mood disturbance 0.76^⁎ −0.16 0.25 0.14 0.05 0.08 0.28^⁎

^⁎ p < 0.05.
^⁎⁎ p < 0.01.

Variables	Patient (n = 30)	Control (n = 60)	Total (n = 90)
Age (yr)	−0.20	0.11	0.08	0.22	0.06	0.11	0.11
Body mass index (kg/h²)	0.21	−0.02	0.12	0.12	0.15	0.15	0.16
Number of health practices	−0.63	−0.10	−0.31	0.11	−0.18	−0.09	−0.20
CES-D score	0.67	0.08	0.43^⁎	0.12	−0.04	0.02	0.24^⁎
POMS score
Tension-Anxiety	0.71^⁎	−0.17	0.22	0.22	0.13	0.17	0.36^⁎⁎
Depression-Rejection	0.69	0.06	0.40	0.08	0.02	0.05	0.30^⁎⁎
Anger-Hostility	0.61	0.01	0.23	0.05	0.13	0.11	0.23^⁎
Vigor-Activity	−0.26	−0.16	−0.18	−0.33	0.16	0.00	−0.15
Fatigue-Inertia	0.60	−0.04	0.31	−0.09	0.03	0.03	0.30^⁎⁎
Confusion-Bewilderment	0.76^⁎	−0.10	0.35	−0.10	0.02	0.01	0.23^⁎
Total mood disturbance	0.76^⁎	−0.16	0.25	0.14	0.05	0.08	0.28^⁎

Multiple regression analyses in total subjects were conducted to evaluate how the 8-OH-dG level as the dependent variable were determined by several factors, such as age, gender, BMI, smoking, drinking, exercise, the CES-D score or the Depression-Rejection score, and whether the subjects were patients or controls, as the independent variables. No other factors were significantly associated with the 8-OH-dG levels according to Mann–Whitney U test or Spearman’s rank correlation, so that only these variables were selected as the independent variables. As shown in Table 4, only whether the subjects were patients or controls was a significant predictor of the 8-OH-dG levels in male or total subjects. On the other hand, only the CES-D score was a significant predictor of the 8-OH-dG levels in female subjects. The results became prominent when the Depression-Rejection score, instead of the CES-D score, was included as the independent variable in the analysis (Table 5). The outcome was comparable to that of the Depression-Rejection score when the TMD score was used in the analysis (data not shown).

Table 4.

Variables including CES-D score associated with 8-OH-dG levels by stepwise multiple regression analyses in male, female, or total subjects

Variables Partial r SE β P R²

Male (n = 56)

Subject (control/patient) 0.211 0.063 0.425 0.002 0.181

Female (n = 27)

CES-D score 0.014 0.004 0.568 0.002 0.323

Total (n = 83)

Subject (control/patient) 0.225 0.054 0.427 0.000 0.182

Note. Partial r indicates partial regression coefficient. β indicates standardized partial regression coefficient.

Variables	Partial r	SE	β	P	R²
Male (n = 56)
Subject (control/patient)	0.211	0.063	0.425	0.002	0.181
Female (n = 27)
CES-D score	0.014	0.004	0.568	0.002	0.323
Total (n = 83)
Subject (control/patient)	0.225	0.054	0.427	0.000	0.182

Table 5.
Variables including Depression-Rejection score of POMS associated with 8-OH-dG levels by stepwise multiple regression analyses in male, female, or total subjects

Variables Partial r SE β P R²

Male (n = 56)

Subject (control/patient) 0.224 0.061 0.454 0.001 0.206

Female (n = 27)

Depression-Rejection score 0.012 0.003 0.601 0.001 0.362

Total (n = 83)

Subject (control/patient) 0.233 0.053 0.444 0.000 0.198

Note. Partial r indicates partial regression coefficient. β indicates standardized partial regression coefficient.

Variables	Partial r	SE	β	P	R²
Male (n = 56)
Subject (control/patient)	0.224	0.061	0.454	0.001	0.206
Female (n = 27)
Depression-Rejection score	0.012	0.003	0.601	0.001	0.362
Total (n = 83)
Subject (control/patient)	0.233	0.053	0.444	0.000	0.198

4. Discussion

This study showed that the 8-OH-dG levels in patients with depression were significantly higher than those in age- and gender-matched healthy controls, independently of well-known 8-OH-dG-related factors or risk factors for cancer, such as age, smoking, drinking, exercise, and obesity (Kasai, 1997, Kasai, 2002 and Levi, 1999). This tendency was found in both genders; however, females, but not males, showed no significant difference in the 8-OH-dG levels between depressive patients and healthy controls, possibly due to the small number of subjects and the marked variation of the 8-OH-dG levels. The reason for the marked variation is unknown; however, in view of the results of correlation analysis, the degree of depression in female patients appears to play a greater role in the 8-OH-dG levels compared with their counterparts. In addition, a significant correlation was observed between the CES-D scores and the 8-OH-dG levels in depressive, particularly female, patients, and this association was also found in the total female participants by multiple regression analysis. The results in patients with depression are consistent with our previous findings that the CES-D score or the Depression-Rejection score was positively related to the 8-OH-dG levels in female, but not male, healthy workers (Irie et al., 2001, Irie et al., 2002 and Irie et al., 2003). Taking these findings into consideration, it seems possible to consider that depression is related to the pathogenesis of cancer via oxidative DNA damage, and the risk increases in response to the degree of depression, especially in females. This may highlight a long-lasting conflicting theme: the potential linkage between psychological stress, including depression, and cancer, from the standpoint of the genetic alterations.

There are some possible underlying mechanisms linking depression to 8-OH-dG production. First, as reported previously (Irie et al., 2003), it is suggested that stress increases neutrophil numbers and increased neutrophil produces 8-OH-dG in females, because we found significant and positive correlations of the percentage of neutrophils to the Severe depression scores of the General Health Questionnaire and the 8-OH-dG levels in female, but not male, healthy workers.

Second, it is well known that there are some repair pathways regarding 8-OH-dG, such as base excision repair, nucleotide excision repair, mismatch repair, and the sanitization of the nucleotide pool (Cooke et al., 2000 and Boiteux et al., 2002). There have been no reports which investigated the relationship between psychological stress and 8-OH-dG repair activity; however, Kiecolt-Glaser et al. (1985) reported that high-distress non-psychotic psychiatric inpatients had significantly poorer DNA repair in lymphocytes exposed to X-irradiation than low-distress patients, and such DNA repair of all patients was reliably lower than those of non-psychiatric control subjects. Thus, depression may increase the 8-OH-dG levels, not only by damaging the DNA, but also by impairing its repair activity.

Third, substantial evidence exists for a hyperactive hypothalamic–pituitary–adrenal (HPA) axis in depression that resulted in increases in adrenocorticotropic hormone (ACTH) and cortisol (Varghese and Brown, 2001). It may thus be possible to consider that cortisol or ACTH has an intervening role in the relationship between depression and 8-OH-dG levels. However, we could not find the role regarding cortisol, as well as catecholamines, in 38 healthy workers in our previous study (Irie et al., 2002).

Fourth, it has been proposed that the hypersecretion of pro-inflammatory cytokines, especially interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor alpha (TNF-α), is involved in the pathology of depression (Raison and Miller, 2003 and O’Brien et al., 2004). Moreover, it has been found that the cytokines induces a pattern of behavioral alterations called ‘sickness behavior’ which resembles the neurovegetative symptoms of depression (Raison and Miller, 2003 and O’Brien et al., 2004). Interestingly, IL-1, IL-6, and TNF-α were reported to enhance the synthesis of 8-OH-dG (Ohtaki et al., 2003, Lin et al., 2001 and Park et al., 1998). Therefore, it is possible to speculate that such pro-inflammatory cytokines may mediate the link between depression and 8-OH-dG levels.

In line with our previous findings of healthy workers (Irie et al., 2001, Irie et al., 2002 and Irie et al., 2003), gender difference in the relationships between psychological variables and 8-OH-dG levels was observed in patients with depression. Female, but not male, patients showed highly positive associations between psychological stress variables, including depression, and 8-OH-dG levels. There are some other reports suggesting gender difference in the relationship between psychological stress and cancer initiation. For example, Hagnell (1966) reported that females who had low emotional control showed a higher cancer incidence in comparison with their counterparts, whereas males did not indicate such a tendency. Kaplan and Reynolds (1988) assessed depression in a cohort of 6848 adults in 1965 and followed their cancer initiation up to 1982, and found that the age-adjusted cancer risk estimates were 0.97 for men and 1.27 for women (p > 0.05 for both). Our present and previous findings (Irie et al., 2001, Irie et al., 2002 and Irie et al., 2003) seem to support such affirmative observations, although negative association was also reported (Penninx et al., 1998). The reason for the gender difference is unclear; further investigations are needed to clarify its possible mechanism.

During exercise, oxygen consumption can increase up to 10- to 15-fold above resting levels, which can result in cellular oxidative damage (Tsai et al., 2001). Some studies have found exercise-induced DNA damage in leukocytes (Poulsen et al., 1993), but others have not (Sacheck et al., 2003). In the present study, the regularly exercising controls showed higher 8-OH-dG levels compared to their counterparts; however, the regularly exercising patients indicated the opposite tendency. As Tsai et al. (2001) proposed, exercise-related leukocyte DNA damage may vary according to the mode, intensity, and duration of the exercise protocol. The reason for our inconsistent findings remains to be elucidated.

This study contains several possible limitations. First, the power of this study to demonstrate the relationship between clinical depression and cancer risk is limited because of the small number of subjects, particularly female patients. Second, whether the 8-OH-dG levels in other organs, such as the brain, of depressive patients similarly indicate higher levels than those in healthy controls is unknown. Third, psychological factors possibly related to the production of 8-OH-dG were not limited to depression, because the POMS-TMD score also contributed to the 8-OH-dG levels. Fourth, we could not neglect the possibility that medications, such as antidepressants and/or minor tranquilizers, might have affected the results, since most of the patients were administered such medicines. Fifth, there are several kinds of antioxidants, such as vitamins A, C, and E, glutathione, ubiquinone, flavonoids, carotenoids, and others (Urso and Clarkson, 2003 and Stanner et al., 2004). Among the antioxidants, dietary intake or supplementation of vitamins C, E, or carotenoids has been found to result in a significant decrease in 8-OH-dG synthesis (Podmore et al., 1998, Yang et al., 1999 and Thompson et al., 1999), although negative findings have also been reported (Loft et al., 1992). ω − 3 Fatty acids have been suggested to protect against 8-OH-dG synthesis in rats exposed to oxidative stress (Kikugawa et al., 2003). In addition, it is well known that the state of depression and antidepressant medications alter appetite. Accordingly, dietary intake or supplementation of such properties might have modified the results, because we did not control for diet, nor did we administer a dietary recall for the days preceding the assessment. Sixth, we could not completely rule out the possibility that hormonal influences in the week prior to the depression assessment might have affected the outcome, because we did not control for menstrual cycle phase before assessing the female participants. Finally, it may be overhasty to conclude that individuals with depression are generally at an increased risk of cancer, because it is uncertain whether subjects with high levels of 8-OH-dG really tend to develop cancer. Only 20 years have passed since the detection of 8-OH-dG (Kasai and Nishimura, 1984). Furthermore, the underlying mechanisms linking depression to 8-OH-dG production are unknown.

Despite these limitations, however, the current study provides suggestion that clinical depression is a risk factor for cancer initiation due to oxidative DNA damage. In order to make progress in this field of research, further investigations are required, incorporating large-scale studies and prospective studies that allow evaluation of the effect of degree of depression and treatment of depression on the 8-OH-dG levels not only in peripheral leukocytes, but also in other samples such as urine. In addition, the differentiation of depressed mood state from overall distress is needed to further refine remaining relationships with 8-OG-dG levels.



		This Document


			SummaryPlus
			Full Text + Links
			·Full Size Images
			PDF (153 K)



		Actions


			E-mail Article

Abstract

Article Outline