Depression is the most common psychiatric condition and regarded as a major cause of disability worldwide. In Japan, the life time prevalence of major depression has been estimated at 3-7% [1]. The presence of depressive symptoms considerably increases the risk of major depression. Among many putative risk factors, folate deficiency has been recognized since the 1960s as an important contributor to the depression [2, 3]. A lot of studies examining the risk of depression in the presence of low folate were performed [4–8], including in Japanese subjects. In 2006, Gilbody reported that there is accumulating evidence that low folate status is associated with depression by meta-analysis [9]. Contrast to that in these observational studies, the effects of folate supplementation on depressive symptomatology obtained from several intervention studies were inconsistent, Taylor et al. performed a systematic review and meta-analysis then suggested that folate may have a potential role as a supplement to other treatment for depression [10], although another study failed to provide evidence for the potentiation of antidepressant medication by folate + vitamin B₁₂ supplementation [11].

Another risk factor of depression is low socioeconomic status (SES) [12, 13]. Low-SES groups are often exposed to an accumulation of life event stressors and chronic problems related to their low-SES state, such as poor education, poor labor circumstances and unemployment, financial strain, inadequate housing, or neighbor violence. In 2003, V. Lorant reported socioeconomic inequality in depression by meta-analysis [14]. In their report, 51 studies reported an odds ratio greater than 1, of which 35 were statistically significant.

Workplace mental health has also garnered increasing attention over the past decade because depressive disorders are highly prevalent in the workplace and have an enormously negative impact on performance, productivity, absenteeism and disability costs. Several studies have found significant associations between sources of perceived stress and major depression onset [15–17], suggesting that work stress is an independent risk factor for the development of depression. The Job Demand-Control -Support (JDCS) model has dominant research on occupational stress in which three major components are used to describe workplace qualities: demands, control, and support [18]. Sanne et al. showed that high demands, low control and low support individually, but particularly combined, are risk factors for anxiety and depression [19].

Folate has been implicated to have negative association with depression, but to our knowledge, the contribution of the SES or job stress factors is still unclear. Recently, the Japanese study of Health, Occupation and Psychosocial factors related Equity (J-HOPE study) was performed to develop and expand research to elucidate mechanisms underlying the social disparity in health and establishment of measures to control over it. The current study is a part of the J-HOPE, and our objective of this study was to clarify the association between folate intake and depressive symptoms in Japanese general workers by using large-scale samples, together with the consideration of SES and job stress factors.

As to SES factors, proportion of management positions, years of continuous employment, and unadjusted annual household income were found to be significantly lower in the subjects with depressive symptom. Also the daily folate intake levels adjusted by total energy intake were significantly different between two groups (155.8 ± 54.7 vs 165.0 ± 57.4 μg/1000 kcal·day, P = 0.001), although the unadjusted folate intake were not different (P = 0.162).

For job stressor scores, notable differences were found between subjects with or without depressive symptom in job demands, control, strain and worksite supports. All of the P values were less than 0.001.

When the subjects were classified into tertiles according to energy-adjusted folate intake level, mean K6 score (±standard error of the mean [SEM]) of each subgroup adjusted for age, sex, job strain, worksite support, years of education, proportion of management position and annual household income was calculated and compared. The adjusted mean values were 5.5 ± 0.2, 5.2 ± 0.2 and 4.8 ± 0.2, respectively. P value for the trend was less than 0.001, indicating a significant linear reduction of K6 score with the increasing of folate intake (Figure 1).

We also considered the gender difference on this association. By the separate linear regression analysis performed in two genders, the association between energy-adjusted folate intake and K6 score remained in male subjects (β = −0.050, P = 0.021), but disappeared in females (β = −0.054, P = 0.412).

In the current study, we confirmed an inverse, independent linear relation of K6 score with the energy-adjusted folate intake level in Japanese workers. These results are in agreement with previous study of Japanese [5, 29]. It is worthy of note that comparing with the small sample size of their studies (517 in Ref. 5 and 530 in Ref. 29), our study was a large-scale investigation with more than 2,200 participants, providing more convincing evidence for this association. Furthermore, we took various job stress scores into account as potential relative factors at the first time. When the job stress scores were not added and the analysis was adjusted for age, sex and SES factors only, the folate intake was also associated with K6 score by linear regression (standard β was −0.059; P value for trend was 0.006) and prevalence of depressive symptom by logistic regression (ORs [95% CI]: 0.813 [0.671-0.985]; P value was 0.035). After further adjusted for the job stress scores, the relations of folate intake to K6 score and prevalence of depressive symptom remained (P values are 0.010 and 0.044, respectively), and the stress factors also showed association with depressive symptom, providing inspiration to understand thoroughly the relationships among depression and relative factors.

The results that the association between folate intake and depressive mood was found in men only, were consistent with that of Murakami et al. [5] and Nanri et al. [29], who found that higher folate intake was associated with lower prevalence of depressive symptoms in Japanese men but not in women, although in another study the association was found only in USA women [7]. The possible explanation for the gender difference is the relatively sufficient folate intake in women. In the present study, mean (±SD) values of adjusted intake of men and women were 159.4 ± 54.1 and 192.9 ± 69.9 μg/1000 kcal·day, respectively, therefore the higher intakes of female subjects maybe mask the effect of folate intake on depression. However, because of the relative small size of female subjects in our study and the importance of folate intake in pregnant women, whether the dietary folate intake affect mental health should be further investigated in larger scale and well-designed studies.

The mechanism of folate intake relate to depressive symptom was thought to be that a decreased folate intake results in an accumulation of homocysteine and then elevated homocysteine levels, which cause vascular disease of the brain, and/or transmitter alterations, could cause depression [30]. Higher total plasma homocysteine concentrations were associated with older age, male gender [31] and dietary deficiency of folate, vitamins B12 and B6 [32]. It was described that the lesions of atherosclerosis could be induced by administering homocysteic acid by animal experiment [33]. Early studies suggested a relationship between depression and cerebrovascular and cardiovascular disease [34], and deficiency of folate and elevated homocysteine levels was considered to increase the risk for vascular disease of the brain [35–37]. However, more recent evidences strongly supported a different role of folate. Folate plays a role in the 1-carbon cycle, and eventually produces S-adenosyl-_L-methionine (SAMe), which is an important methyl donor and thought to be involved in the synthesis of three key neurotransmitters in the brain-dopamine, serotonin and norepinephrine [38]. Homocysteine is a critical branch point metabolite that can influence cellular levels of SAMe and S-adenosylhomocysteine (SAH), which regulate the activity of methyltransferases important in posttranslational modification of proteins and synthesis of nucleic acids, phospholipids, and neurotransmitters [39]. Thus, low folate intakes, which result in elevated homocysteine levels, lead to a deficiency of these neurotransmitters and then cause depression. The homocysteine hypothesis of depression reviewed by Folstein et al. [30] centering around hyperhomocysteinemia leading to an increased risk of stroke, heart disease, neurotransmitter imbalances and thereby depression was further extended by a brief comment in the American Journal of Psychiatry in 2007 [40]. The authors herein stated that homocysteine is metabolized to SAMe, then influence DNA methylation. Secondly, homocysteine itself was shown to affect global and gene promoter DNA methylation, and the administration of acute homocysteine usually leads to demethylation of promoter DNA with a subsequent increase in gene expression [41]. A further study supported the hyperhomocysteine hypothesis of depression, reporting significantly higher homocysteine levels in patients with moderate depressive symptoms and eating disorder diagnoses [42]. Our observations epidemiologically confirmed with this hypothesis.

Some limitations of the present study are worth mentioning. Firstly, the cross-sectional nature of the study does not permit the assessment of causality owing to the uncertain temporality of the association. Depressed persons may eat less or more and the intakes of nutrition are influenced by the total amount of intake. In order to remove this effect, we used the total energy adjusted folate intake level (μg/1000 kcal per day). Our data showed energy adjusted low folate intake was significantly related to depressive symptoms, and the probability of the reverse causation seems slim. Secondly, our subjects were workers in one large company and not a random sample of Japanese workers, and thus the results may not apply to the general Japanese population. However, the workers were recruited from 12 offices all over the Japan (From Hokkaido to Kyushu). So the geographical deviation was reasonably diluted, but it is noteworthy fact that our result is the result of a large company, not a small company. Finally, dietary data were obtained from a self-administered semi-quantitative dietary assessment questionnaire [25]. Because the actual dietary habits were not observed, the results should be interpreted with caution, although the validity of this questionnaire appears reasonable [26–28].

In conclusion, low folate intake was independently related to higher depression scores and increased prevalence of depressive symptoms in Japanese workers, adjusting for job strains, supports and various SES factors. Further prospective or intervention studies are warranted to determine this association as well as the roles of SES and stress effect on mental health outcomes.