Severity of subjective forgetfulness is associated with high dietary intake of copper in Japanese senior women: A cross‐sectional study

Abstract This study investigated the relationship between subjective forgetfulness and the dietary intake of various nutrients in middle‐aged and senior women. A cross‐sectional study of the first‐visit records of 245 Japanese women aged 40 or over was performed. The severity of subjective forgetfulness was classified according to the Menopausal Health‐Related Quality of Life Questionnaire: none and mild (“unforgetful”) or moderate and severe (“forgetful”). Dietary consumption of nutrients was estimated using the brief‐type self‐administered diet history questionnaire. The associations between the severity of subjective forgetfulness and intake of 43 major nutrients were evaluated using multivariate logistic regression analysis separately performed for two age groups: middle‐aged (40–54 years, N = 166) and senior (55 years or over, N = 79). No nutrients were found to be significantly associated with subjective forgetfulness in the middle‐aged group. In senior women, a significant positive relationship between the intake of copper and forgetfulness was found (adjusted odds ratio per 10 mg/kJ increase in copper intake: 1.25; 95% confidence interval: 1.08–1.50). Thus, high copper intake is positively associated with the severity of forgetfulness in Japanese senior women. Reducing copper consumption could help improve this symptom in this population.

energy metabolism via glucose consumption; its deficiency causes peripheral and central nervous system disorders, such as the wellknown beriberi and Wernicke-Korsakoff syndromes (Bourre, 2006a).
A lack of zinc, which has neuromodulatory properties, is attributed to memory and learning impairment (Bourre, 2006a). Impaired copper (Cu) homeostasis in the brain is associated with neurodegeneration, such as Menkes disease and Wilson's disease (Scheiber, Mercer, & Dringen, 2014). Lowered consumption of tyrosine, an essential amino acid, and a precursor for catecholamine neurotransmitters including dopamine, a regulator of cognitive function (Previc, 1999), is associated with cognitive dysfunction (Hensel et al., 2019). Docosahexaenoic acid, a known n-3 fatty acid component of the myelin involved in signal transduction, has also been reported to enhance cognitive function (Cole & Frautschy, 2010). Moreover, there are several reports on the effect of nutritional supplements, including vitamin E and n-3 fatty acids, on cognitive function in elderly adults, mild cognitive impairment, and AD (Farina, Llewellyn, Isaac, & Tabet, 2017;Fraga, Carvalho, Caramelli, de Sousa, & Gomes, 2017). However, the effects of the dietary intake of nutrients on memory in the peri-and postmenopausal periods have not yet been fully elucidated.
In the present study, we investigated the associations between the severity of subjective forgetfulness and the dietary consumption of a variety of nutrients in Japanese middle-aged and senior women.

| Study population
In this cross-sectional study, we analyzed the first-visit medical records of 700 Japanese women enrolled in the Systematic Health and Questionnaire (MHR-QOL) and the brief-type self-administered diet history questionnaire (BDHQ), evaluating the severity of subjective forgetfulness and dietary habits, respectively. Thirty-nine participants whose menopausal status was unclear due to a hysterectomy (and unilateral oophorectomy), 57 participants who had received hormone therapy, and 41 participants who had received psychotropic drugs such as hypnotics, antidepressants, and anxiolytics were excluded. For the remaining 245 participants, we investigated the associations between the severity of subjective forgetfulness and their dietary consumption of various nutrients.

The research protocol was reviewed and approved by the Tokyo
Medical and Dental University Review Board, and written informed consent was obtained from all participants. All study procedures were conducted in accordance with the Declaration of Helsinki.

| Menopausal status
The women were classified as pre-or postmenopausal if they had a regular period or no menstruation in the past 12 months, respectively. Perimenopausal status was assigned to those who had a menstruation within the past 12 months but had missed periods or had had an irregular cycle during the past 3 months.

| Physical assessments
The body composition of participants, including height, weight, body mass index (BMI), fat mass, lean body mass, muscle mass, water mass, basal metabolism, and visceral fat level, was measured using a body composition analyzer (MC190-EM; Tanita). The participants' waist and hip circumferences were measured, and their waist-to-hip ratios were evaluated. Resting energy expenditure was calculated based on respiratory volume using an indirect calorimeter (Metavine-V VMB-005N; Vine). Their body temperature was measured with a thermometer. We also assessed cardiovascular parameters such as blood pressure, heart rate, cardio-ankle vascular index, and anklebrachial pressure index (VS-1000; Fukuda Denshi). Physical fitness tests for power, reaction time, and flexibility were also conducted.
Hand-grip strength was measured twice for each hand with a hand dynamometer (Yagami) to calculate average hand-grip strength (kgf) using the larger value from each hand. The ruler drop test was performed to evaluate reaction time using a wooden ruler 60 cm in length and weighing 110 g (Yagami). Seated participants, fixing their arms on a desk and outstretching their fingers from the edge of the desk, attempted to catch the ruler, whose bottom was hung between their thumbs and index fingers by an examiner, as quickly as possible when it was dropped. The scale on the ruler was used to measure where the participants gripped. After the test was repeated seven times and the largest and smallest values were omitted, the average reaction time (cm) was calculated from the remaining five values.
The sit-and-reach test as an assessment of flexibility was performed using a reach box while sitting (Yagami).

| Lifestyle characteristics
Lifestyle factors such as frequency of smoking (none, fewer than 20, more than 20 cigarettes per day), consumption of alcohol (never, sometimes, daily) and caffeinated drinks (none, once or twice, three or more times per day), and regular exercise habits (yes, no), were assessed via a medical interview.
Its 4-point Likert scale for physical and psychological symptoms is based on symptom frequency. Respondents indicate the degree of agreement or disagreement with statements about life satisfaction and social involvement on a 2-or 4-point Likert scale. According to the participants' responses, the severity of their subjective forgetfulness was rated as none, mild, moderate, or severe based on frequency (none, zero to one time per month; mild, one to two times per week; moderate, three to four times per week; severe, almost every day). In the present study, those with no or mild forgetfulness were defined as "unforgetful" and those with moderate or severe forgetfulness as "forgetful." We added up the scores for somatic symptoms (six items), vasomotor symptoms (two items), insomnia symptoms (two items), life satisfaction (five items), and social involvement (12 items) to produce a score for each subcategory, with a high score representing good health and a high level of satisfaction and social engagement.
The BDHQ was developed to assess the intake frequency over the previous month of 61 food items in the typical Japanese diet, including beverages and seasonings. Based on the information provided in response, the daily consumption of 96 nutrients, after adjustment for total calorie intake, was estimated by an ad hoc computer algorithm. The BDHQ was validated by comparison with dietary records using a semi-weighed method (Kobayashi et al., 2011(Kobayashi et al., , 2012. In the present study, we focused on 43 major nutrients with high validity (Table S2).
The Hospital Anxiety and Depression Scale (HADS), a reliable screening instrument for anxiety and depression, was developed to assess the psychological health of patients with physical symptoms (Zigmond & Snaith, 1983). The HADS contains seven items each for anxiety and depression, and participants respond to these items on a 4-point Likert scale. Participants who scored 8-10 were considered to likely be experiencing anxiety or depression, while those who scored 11-21 were considered to definitely be experiencing anxiety or depression.

| Statistical analysis
Continuous variables are presented as mean ± standard deviation.
The required sample size was estimated at 250, calculated as 10 times the number of independent variables, estimated at 10, divided by the prevalence of forgetfulness, estimated at 0.4. The Kruskal-Wallis test, Mann-Whitney test, and chi-squared test were used to evaluate the difference between groups. Multicollinearity between variables was identified using cutoff points for Pearson or Spearman correlation coefficients of |R| > .9. The relationship between the severity of subjective forgetfulness and daily consumption of nutrients was analyzed using multivariate logistic regression analysis after adjustment for age, BMI, and the background factors related to forgetfulness obtained through the analysis. p values of <.05 were considered statistically significant. All statistical analyses were performed with GraphPad Prism version 5.02 (GraphPad Software) and JMP version 12 (SAS Institute Inc).

| RE SULTS
The average age of the 245 study participants was 53.4 ± 6.8 years (mean ± standard deviation). The percentages of women with none,

Middle-aged (N = 166)
Senior ( somatic and vasomotor problems, insomnia, anxiety, and depression more frequently than the unforgetful ones, and their life satisfaction was significantly lower. We first assessed the nutrients whose intake differed significantly between the four groups of subjective forgetfulness severity, finding no nutrient to be significantly associated with forgetfulness in the whole group.
Next, we performed a post hoc analysis of the 245 participants  Tables 1 and 3,   and Table S3. The senior group had a significantly higher waist-to-hip ratio and visceral fat level and lower basal metabolism and vascular health, while the middle-aged group had more severe physical and mental symptoms and lower levels of satisfaction and social involvement (Table 3). Moreover, the senior women consumed significantly more nutrients (except the following 11: fat, animal and vegetable fat, carbohydrates, retinol, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, cholesterol, n-6 fatty acids, and alcohol) than the middle-aged women despite no difference between the groups in total energy intake (p = .174) ( There was no multicollinearity among these four nutrients. Next, we investigated the background factors that differed significantly between the four severity groups in senior women. The results showed that reaction time and the scores for somatic symptoms, life satisfaction, social involvement, anxiety, and depression were found to be related to forgetfulness. There was no multicollinearity between these background factors. We used these six background factors as independent variables. Multivariate logistic regression analysis was conducted to identify the association between the intake of selected nutrients (potassium, magnesium, Cu, and vitamin B 1 ) and the severity of subjective forgetfulness. After adjustment for the selected four nutrients (Model 1), for age and BMI (Model 2), and for age, BMI, and the related background factors (Model 3), an association between higher dietary consumption of Cu (10 mg/kJ) and the severity of subjective forgetfulness in the senior group was found (

| D ISCUSS I ON
In this cross-sectional study, severity of subjective forgetfulness was found to be significantly associated with a high Cu consumption in Japanese senior women. We also found differences in background characteristics such as body composition, physical and psychological symptoms, and vascular health between the two age groups, indicating that our findings could be affected by these distinctions.
The main dietary sources of Cu, an essential trace element, are seafood, nuts, and soybeans. Most (~95%) serum Cu is tightly bound to caeruloplasmin, a glycoprotein produced in the liver and related to Cu and iron metabolism, and small amounts of Cu comprise Cu bound to albumin, small peptides and amino acids, often called noncaeruloplasmin-bound Cu or free Cu, which can reach the brain via the blood-brain barrier because of low molecular weight. Cu plays a key role in redox reactions via Cu-dependent enzymes such as superoxide dismutase, cytochrome c oxidase, ceruloplasmin, dopamine beta-hydroxylase, and peptidylglycine α-hydroxylating monooxygenase, contributing to antioxidative defense, iron and energy metabolism, and neurotransmitter and neuropeptide synthesis. On the other hand, Cu overload leads to raised nonbound caeruloplasmin Cu, causing increased oxidative stress and cell damage (Liao et al., 2019).
It is well-known that Cu excess in Wilson's disease and Cu deficiency
Alzheimer's disease is an irreversible progressive disease characterized by neurofibrillary tangles and senile plaque (β-amyloid deposits). Significantly lowered Cu levels in the hippocampus in AD patients have been reported (Deibel, Ehmann, & Markesbery, 1996;Xu et al., 2016). A meta-analysis found Cu depletion in several brain regions in patients with AD (Schrag, Mueller, Oyoyo, Smith, & Kirsch, 2011). In contrast, several reports have found a high accumulation of Cu in senile plaque related to amyloid formation (Lovell, Robertson, Teesdale, Campbell, & Markesbery, 1998;Miller et al., 2006). Squitti et al. (2014) showed nonbound caeruloplasmin Cu to be a predictor of the development of AD in patients with mild cognitive impairment. They also reported that serum Cu levels and nonbound ceruloplasmin Cu levels in AD patients were greater than those in a healthy control group . Moreover, James et al. (2012) demonstrated that deranged Cu metabolism in AD was responsible for oxidative stress. Chen et al. (2019) also reported that Cu exposure disturbed brain function in wild-type mice and exacerbated neurodegenerative changes in AD model mice, suggesting that Cu overload could induce mitochondrial and synaptic dysfunctions, some of the early changes seen in the course of AD (Lin et al., 2016).
Similarly, there have been several reports on the association between Cu and cognitive function. A cross-sectional study of 188 Chinese people aged 65 years or over associated high plasma Cu levels with cognitive dysfunction (Smorgon et al., 2004). In a cohort study of 1,451 residents (629 men and 849 women) aged 60 years or over, inverse correlations between plasma Cu concentrations and long-and short-time recall were observed only in women (Lam et al., 2008). Salustri et al. (2010) found that serum levels of nonbound caeruloplasmin Cu were negatively correlated with cognitive function in 64 healthy women over 50 years old. Furthermore, in a prospective study of 3,718 male and female local residents aged 65 years or over, cognitive function in those with a high intake of saturated and trans fats declined rapidly via increased Cu intake (Morris et al., 2006). An animal study also demonstrated that Cu consumption induced β-amyloid accumulation and impairment of memory and learning in cholesterol-fed rabbits, suggesting that Cu obstructs β-amyloid clearance and accelerates its accumulation, caused by elevated brain cholesterol (Sparks & Schreurs, 2003). Our findings of an association between increased Cu intake and subjective forgetfulness only in senior women might indicate that changes Among our study participants, the mean daily Cu intake in the senior group was significantly higher than that in the middle-aged group, so our results might be affected by this difference.
One of the major limitations of our study was the small and narrow study population, making it difficult to generalize our findings to a wider population. Second, because of the cross-sectional nature of our study, any causal relationship remains unclear. Moreover, we In conclusion, in this study, high consumption of Cu was positively associated with the severity of subjective forgetfulness in Japanese senior women. Therefore, decreasing Cu intake could help ameliorate this symptom for this population.

ACK N OWLED G M ENTS
The funding from Kikkoman Corporation (grant number x2136) is gratefully acknowledged.

MT received an unrestricted research grant from the Kikkoman
Corporation. The other authors have no conflicts of interest to disclose.

E TH I C A L R E V I E W
This study was approved by the Tokyo Medical and Dental University Review Board.

TA B L E 5
Relationship between daily intake of copper (10 mg/kJ) acid and severity of subjective forgetfulness Abbreviations: CI, confidence interval; OR, odds ratio.

I N FO R M E D CO N S E NT
Written informed consent was obtained from all study participants.