Dietary Habit-Induced Gynecologic Disorders in Young Female Students – Lessons from Rodent Experiments

Abstract

Currently, dieting and breakfast skipping is increasing among young women in Japan. We found that breakfast skipping among female students was accompanied by menstrual disorders, while students who had dieted in the past experienced deterioration in menstrual pains, warning that abnormal eating in young women may induce obstetric and gynecological disorders in the future. We named this concept “adolescent dietary habit-induced obstetric and gynecologic disease (ADHOGD)”. A questionnaire survey showed that pregnant women who had menstrual pain in their youth were at high risk of hypertensive disorders during pregnancy. In rodents, ovulation was suppressed in young female rats whose feeding was limited to the non-active (light) phase. In female mice, feeding stimulation directly regulated the uterine clock gene rhythm. Furthermore, in conditional knockout mice of uterine Bmal1, the fetuses died before delivery, indicating that abnormal uterine clock function cannot maintain fetal development. These findings suggest a mechanism of ADHOGD, in which hunger stress due to inappropriate eating habits during adolescence and young adulthood affects uterine function via clock gene abnormalities, causing placental dysfunction and fetal growth failure during pregnancy. Thus, valid and appropriate rodent experiments are effective to analyze ADHOGD, especially from the aspect of circadian rhythms.

1. Introduction

Currently, both dieting and breakfast skipping are increasing among young women in Japan. We found that breakfast skipping among female students was accompanied by menstrual disorders, while students who had dieted in the past experienced deterioration in menstrual pains after recovery [1, 2], warning that abnormal eating rhythm in young women may induce obstetric and gynecological disorders in the future [3, 4]. Subsequently, a correlation between menstrual pain and breakfast skipping in young women was confirmed worldwide [5, 6]. Therefore, we considered menstrual pain associated with this eating rhythm abnormality as a universal risk for obstetric and gynecologic diseases across races and proposed the concept that abnormal eating habits in youth can induce obstetric and gynecologic diseases [7].

The concept of developmental origins of health and disease (DOHaD) is currently becoming a great concern in the world [8]. This hypothesis was initially called ‘fetal origins of adult disease’ [9] and proposed that exposure to a poor nutritional environment during critical periods of development and growth may determine the later onset of human diseases in adulthood [10]. In DOHaD theory, prenatal and perinatal stages were proposed as the responsible windows when predictive adaptation to environmental influences can occur [11]. On the other hand, reproductive organs markedly develop and mature during adolescence and young adulthood. Consequently, we proposed that inadequate dietary habits such as dieting and breakfast skipping during adolescence and young adulthood impair the development and maturation of reproductive functions, which induces the latent progression of obstetrics and gynecologic disorders and leads to the latter onset of obstetrics and gynecologic diseases [3, 12]. Recently, we named this concept “adolescent dietary habit-induced obstetric and gynecologic disease (ADHOGD)” (Figure 1) [7].

Figure 1.

Concept of ADHOGD. 1) Adverse dietary habits such as diet and meal skipping during adolescence and young adulthood impair the development and maturation of the reproductive function, 2) which induces the latent progression of obstetric and gynecologic disorders. Although 3) recovery is achieved after correcting adverse eating habits, 4) the reproductive function is precociously declined, 5) which later leads to the onset of obstetrics and gynecologic diseases, especially during pregnancy.

Although it has been known that abnormalities in the light–dark cycle, such as dieting and night work, suppress the hypothalamic system in humans, the effects of feeding rhythms on reproductive function were unknown. Therefore, when young female rats were restricted to feeding only during the active (dark) or inactive (light) phase, suppression of ovulation was observed in the inactive feeding group [13]. In addition to the above, the following important discoveries regarding the mechanism of ADHOGD and its effects on the uterus have been made by the authors. First, Ono et al. showed in a questionnaire survey of pregnant women that women who had menstrual pain in their youth were at high risk of developing hypertensive disorders of pregnancy (HDP) even if their pain improved [14]. Gestational hypertension is believed to be caused by the inadequate reconstruction of maternal blood vessels during placentation, and these results indicate that the uterus may “remember” the abnormal uterine function associated with menstrual pain in young adulthood. On the other hand, Ando et al. observed that feeding stimulation can directly regulate the uterine clock gene expression rhythm in mice. This indicates that abnormal feeding rhythm may directly disrupt uterine function without involving the ovary [15]. Furthermore, Daikoku et al. created mice deficient in the uterus-specific clock gene Bmal1 and observed that the fetuses died before delivery, indicating that abnormal uterine clock function cannot maintain fetal development. The cause of this phenomenon was elucidated to be abnormal placentation accompanied by a failure in the construction of maternal blood vessels, indicating the possibility that reproductive and perinatal diseases are induced by abnormal uterine clock [16].

Based on this background, in this chapter, we precisely introduce a new concept of dietary habit-induced gynecologic disorders in female students and the rodent experiments that provided current evidence to support this concept from the aspect of circadian rhythms.

2. Mechanism of dietary habit-induced gynecological disorders

2.1 Relationship between inadequate dietary habits and gynecological disorders

Skipping meals and dieting for beauty purposes are spreading among young Japanese women. In parallel with this, the increase in gynecological diseases such as endometriosis, which causes dysmenorrhea, has become a big problem. Although there has been little evidence about the relationship between menstrual pain and eating habits, we previously reported that breakfast deprivation among female students was accompanied by menstrual cramps. For female students, menstrual pain is important medical information because it is a symptom suggesting organic gynecological diseases.

Since there is no appropriate medical terminology for females (ages 18-22) who are in the process of completing their reproductive functions after puberty, preventive dietary guidance and indicators that target this generation were not sufficient. Therefore, we position this period as the "maturation age of reproductive function," and have conducted a fact-finding survey of female students for about 20 years with the aim of clarifying the relationship between dietary habits and reproductive dysfunction. As a result, we found that students who have dieted in the past have strong menstrual cramps, and this led to a warning of the danger of “development of organic gynecological diseases after ending the diet” [2, 4].

On the other hand, skipping breakfast was associated with menstrual pain in female students, although the total daily food intake did not decrease [1]. Later, a large-scale Palestinian study confirmed that skipping breakfast is associated with menstrual pain [5]. Recently, it was proposed that the disruption of the circadian rhythm can cause poor reproductive outcomes [17]. Focusing on the fact that breakfast corresponds to the start of the circadian rhythm, we proposed the hypothesis that “skipping breakfast interferes with the circadian rhythm and adversely affects reproductive function” [3, 7].

2.2 The effects of meal timing during the circadian cycle on female rat reproductive function

To investigate the effects of meal timing during the circadian cycle on ovarian function, we explored this issue using young female rats [13]. Considering that rats are active at night, 8-week-old female Wistar rats were classified into three groups: day-only feeding group (inactive phase), night-only feeding group (active phase), and control group (no time or calorie restriction). During a 4-week feeding restriction, body weight in each group was measured by weighing scales and ovulatory frequency was assessed by vaginal smears. After the dietary restriction, ovaries were removed and the numbers of growing follicles and corpus luteum were evaluated by examining hematoxylin- and eosin-stained tissue sections.

As a result, in the daytime feeding group, the ovulatory number and frequency were significantly reduced compared to the control group. At the same time, the amount of daily food intake was reduced by approximately 20% and body weight gain was suppressed. In contrast, in the night-time-fed group, there were no differences in the frequency of ovulation as compared with the control group, indicating that the defect of food intake in the non-active phase did not affect the estrus cycle. These findings led us to conclude that feeding defect in the active phase is critical for young female rats and that the timing of food intake during the active or non-active phase is a crucial factor that influences female reproductive function [13].

Furthermore, even when 8-week-old female rats were fed only at night for 4 weeks with 20% less food in the control group (no restriction), their estrus cyclicity did not change despite significant reductions in weight gain and food intake as compared with the control group. These findings also support the above speculation that reproductive dysfunction was more intensely induced by the differences in the timing of food intake during the active or non-active phase than by insufficient calorie intake [13].

In humans, it was well known that shift workers increase the incidence of reproductive disorders such as menstrual troubles, endometriosis, infertility, miscarriage, or pre-term delivery [18, 19, 20]. Food intake is one of the important regulators that can reset the rhythm of the central clock in the brain and peripheral clocks in the digestive organs [21]. It should be noted that in shift workers, the timing of food intake is synchronous with their active behaviors. In this regard, the rat model in which the feeding is inhibited during the active phase may be more stressful than shift workers. Our findings showed that feeding defect in the active phase of the post-pubertal female rat impairs ovarian function, indicating that the timing of feeding during the circadian cycle can interfere with female reproductive function. However, in adult mice, it was reported that although the birth of the first litter was significantly delayed and total reproductive output was significantly reduced in the light-fed group, estrous cycling and pregnancy maintenance did not differ between the light-fed and dark-fed groups, concluding that mistimed feeding inhibits reproduction in mice by reducing successful mating behavior [22].

2.3 Feeding directly regulates the uterine clock rhythm in mice

To examine the relationship between food intake and the uterine clock rhythm, Ando et al. investigated the effects of the first meal occasion in the active phase on the uterine clock expression [15]. In this study, Zeitgeber time (ZT) was defined as ZT 0 (8:45) with lights on and ZT 12 (20:45) with lights off. Eight-week-aged young female mice were divided into three groups: group I (ad-libitum feeding), group II (time-restricted feeding during ZT12-16, initial 4 hours of the active period), and group III (time-restricted feeding during ZT20-24, last 4 hours of the active period, a breakfast-skipping model). After two weeks of dietary restriction, mice in each group were sacrificed at 4-hour intervals and the expression profiles of uterine clock genes, Bmal1, Per1, Per2, and Cry1, were examined.

In accordance with previous reports that the rat uterus has circadian rhythms of clock genes [23, 24], both mRNA and protein expressions of Bmal1, Per1, Per2, and Cry1 in the murine uterus were demonstrated to create a circadian clock rhythm. In addition, immunohistochemical analysis showed that Bmal1 protein expression was synchronized among the endometrium and myometrium. Importantly, in groups I and II, both mRNA and protein expressions of Bmal1 were elevated after ZT12 at the start of the active phase, whereas Bmal1 expression was elevated just after ZT20 in group III, indicating that the uterine clock rhythm had shifted 8 hours backward. A similarly delayed time lag was also observed for Cry1 mRNA expression (Figure 2). This indicates that time-restricted feeding can reset the circadian rhythm of the uterine clock gene expressions and suggests that the uterine peripheral clock is more intensely regulated by diet than by the light/dark cycle, which is a main regulator of the central clock [25, 26].

Figure 2.

Circadian rhythms of uterine clock gene Cry1. The mRNA expressions of Cry1 in the uterus showed circadian cycles. Relative expressions are presented as fold units per minimal values. Cry1 expression in the group (feeding during ZT12-16) was elevated on ZT12 at the start of the active phase and food intake. In contrast, its expression in the group (feeding during ZT20-24) was elevated on ZT20 at the start of food intake, showing an 8-hour backward shift in circadian rhythms. This suggests that the uterine peripheral clock is more strongly regulated by diet than by the light/dark cycle.

Dietary intake was reported to regulate the rhythm of the hepatic circadian clock in the liver [27]. However, in contrast to these vital organs for energy metabolism, the physiological significance of dietary regulation of the uterine peripheral clock is unclear. In pregnant mice, clock gene oscillation was reported to be present in the uterus, placenta, and fetal membranes [28]. For the fetus that cannot receive light stimulation in the uterus, transplacental glucose transport from maternal blood after food intake can be a direct signal to detect the maternal circadian rhythm during pregnancy [29]. A recent study reported that the trophoblast controls transport across the placenta in mice during pregnancy by regulating the circadian expression and activity of the xenobiotic efflux pump, ATP-binding cassette sub-family B member 1 (ABCB1) [30]. Accordingly, we speculate that the uterus is prepared to synchronize its function with food intake to provide an adequate environment for the fetus.

As we demonstrated in the above section, ovulations were impaired in daytime-fed young female rats, indicating that the timing of food intake is an important factor regulating the hypothalamic-pituitary-ovarian axis [13]. Both dietary and light/dark cycles mainly regulate circadian rhythms in the brain [25, 26]. Recent studies suggest that reproductive rhythms are disrupted by circadian rhythms [20, 31]. Since meal skipping at the start of the active phase can interfere with the central clock system [32], it is speculated that breakfast-skipping influences reproductive rhythms [3, 13]. In addition, taking into consideration that the timing of food intake can reset the uterine clock rhythm, we can elicit another mechanism that meal skipping directly disturbs the circadian rhythm of the uterine clock system and causes uterine dysfunction [7].

2.4 Uterine deletion of Bmal1 induces intrauterine fetal death in mice

Systemic Bmal1 knockout (KO) female mice were reported to be infertile, showing multiple organ disorders [33]. However, despite the circadian expression of uterine clock genes such as Per1-3, Cry1-2, Bmal1, and Clock, their precise roles in reproductive functions remain unclear [34, 35]. Later, Ls et al. reported that the uterine expression of BMAL1 was decreased in women suffering from recurrent spontaneous abortion [36]. They also showed that BMAL1-depleted decidual cells inhibited trophoblast invasion, proposing the important role of the uterine clock in the establishment of human pregnancy [36]. A recent report reported that mRNA expression of Clock was increased in the human placenta during preeclampsia, suggesting the involvement of Clock in the pathogenesis of preeclampsia [37].

Based on this background, to investigate the pathological roles of uterine clock genes during pregnancy, we produced conditional deletion of uterine Bmal1 (cKO) mice [16] using Bmal1-loxP [38] and progesterone receptor-cre mice as reported previously [39, 40]. As a result, cKO mice could receive embryo implantation, but not sustain the pregnancy. By histological examinations, abnormal structures of the placenta were observed in cKO mice. The main feature of placental abnormality in cKO mice was the poor development of maternal vascularity. In cKO mice, it was speculated this structural abnormality caused the narrowing of the maternal vascular spaces and the reduction of maternal blood flow into the intervillous spaces in the labyrinth layer, leading to impaired maternal-fetal nutritional exchanges through this layer.

Next, we performed microarray analysis to detect the changes in gene expression in the uterine tissues in cKO mice. Gene ontology analysis of microarray revealed that several genes related to immune response were down-regulated in the cKO uterus, suggesting that immune dysfunction is involved in the poor formation of placenta. It is well known that the cellular immune responses by regulatory T cells, effector T cells, NK cells, and monocytes are related to the pathogenesis of gestational hypertension [41]. Among the down-regulated genes, we paid an attention to GO:0045954 “positive regulation of natural killer cell-mediated cytotoxicity”, because NK cells were shown to directly interact with trophoblast and regulate their invasion and vascular reconstruction [42]. Consequently, we examined the distribution of subtypes of uterine NK (uNK) in the embryo implantation sites and the placenta from early pregnancy to mid-pregnancy.

In cKO mice, the numbers of NK cells expressing PAS(+)/DAB(-) were decreased in the spongiotrophoblast layer of the placenta. In addition, uNK cells in cKO mice hardly expressed CD161 which is an immunosuppressive molecule [43], indicating that their subtypes of uNK cells are different from those in WT mice. This also suggests that the deletion of clock genes induced functional changes in uNK cells, which directly interact with fetal trophoblast. Since CD161 is an immunosuppressive NK marker [43], the decrease of uNK cells in the spongiotrophoblast layer may interrupt trophoblast function to sustain maternal-fetal immune tolerance. From these findings, we concluded that the murine uterine clock system is critical for the placental function to maintain pregnancy after embryo implantation and proposed that the disorder of a uterine clock system can be a candidate to induce uterine dysfunction during pregnancy in ADHOGD.

Intriguingly, progesterone supplementation recovered pregnancy outcomes in cKO mice, showing that some cKO mice sustained pregnancy until the term [16]. Although the structural abnormalities in the placenta were not improved, the number of CD161-positive NK cells in the spongiotrophoblast layer was increased in cKO mice. Clinically, it has been accepted that progesterone supplement prevents miscarriage in women with recurrent miscarriage of unclear etiology [44] and protects against the onset of HDP [45]. Since a recent study reported that progesterone supplement therapy also improves the functional failure of the placenta with structural abnormalities [46], this cKO model will provide supporting experimental evidence of the role of progesterone in the treatment of perinatal complications.

2.5 HDP is associated with dysmenorrhea in early adulthood

During human placentation, the trophoblast invades the maternal decidua and reconstructs maternal spiral arteries. This remodeling reduces arterial contractility and supplies adequate maternal blood flow into the intervillous spaces [47]. If this process is disrupted, the poor blood supply to the intervillous space will induce placental dysfunction and various pregnancy complications, such as HDP, fetal growth restriction, and stillbirth in the late stage of pregnancy [42]. The impaired vascular reconstruction associated with abnormal placental formation observed in our cKO mouse model may be corresponding to the pathogenesis of human placental dysfunction in HDP.

Recently, by a prospective cohort questionnaire study, we found that pregnant women who experienced dysmenorrhea at a younger age in the past had a higher incidence of developing HDP [14]. We recruited 193 pregnant participants and collected valid data from 190 women concerning characteristics, menstrual abnormalities, and lifestyle factors. We also used medical records to examine the relationship between menstrual abnormalities and the onset of HDP. As a result, a total of 26 patients developed HDP, 10 had early onset and 16 had late onset. The HDP group was significantly older than the non-HDP group. Although no significant association was observed between HDP and dysmenorrhea just prior to pregnancy, there was a significant increase in the incidence of HDP when patients had dysmenorrhea around the age of 20 years.

Since the reproductive organs develop and become matured during young adulthood, this period is important for establishing female reproductive functions. As described in the above section, we reported that the incidence of dysmenorrhea is high in female college students with a history of dieting and proposed that dysmenorrhea and poor diet in young adulthood induce the later development of perinatal and gynecologic diseases [7]. Consequently, this cohort study suggests one of the progressing processes of ADHOGD whereby although dysmenorrhea in young adulthood is improved, it will become manifested as ADHOGD when physical stresses such as pregnancy are imposed (Figure 1).

3. Conclusion

The epidemiological data and animal experimental findings suggest that irregular competition between light and feeding rhythm can induce uterine clock dysfunction. These findings also suggest a mechanism of ADHOGD, in which hunger stress due to inappropriate eating habits during adolescence and young adulthood affects uterine function via clock gene abnormalities and other factors, causing placental dysfunction and fetal growth failure during pregnancy, which in turn affects the next generation. Thus, rodent experiments are needed to analyze the new concept of dietary habit-induced gynecologic disorders, especially from the aspect of circadian rhythms, and valid and appropriate experimental models will provide current evidence in support of this concept.

Acknowledgments

This work was supported in part by Grants-in-Aid for Scientific Research (nos. 19H01617, 19K09776, 21K18297, and 22K09556) and a Health and Labour Sciences Research Grant (no. 19216689).

Conflict of interest

The authors declare no conflict of interest.

Abbreviations

ADHOGD	adolescent dietary habit-induced obstetric and gynecologic disease
DOHaD	developmental origins of health and disease
HDP	hypertensive disorders of pregnancy
KO	knockout
uNK	uterine NK
ZT	Zeitgeber time