Pregnant women inevitably experience uneasy emotions and depressive symptoms when they are about to face delivery in the third trimester of pregnancy. Prenatal depression is a psychological symptom characterized by a series of symptoms such as depression, crying, sadness, irritability, or suicide, prenatal depression is one of the most common mental disorders during pregnancy, with a global incidence of 20.7%
[1]. It not only affects the health of pregnant women, but also causes many adverse health effects in the infant, such as intrauterine growth restriction, low birth weight, preterm birth, and cognitive-behavioral problems
[2-3].
Although the pathogenesis of prenatal depression is currently unclear, multiple studies have demonstrated that unbalanced gut microbiota and disruption of the gut-brain axis (GBA), possibly via neural, endocrine, and immune pathways, influence brain function and behaviors that contribute to mental disorders
[4-5], the way gut microbiota affects this may be related to changes in tryptophan metabolism, among other aspects. It has been found that pregnancy can lead to drastic alterations in the gut microbiota composition. During pregnancy, community richness decreases and community composition shows significant changes from early to late gestation, with an increase in lactate-producing bacteria, a decrease in butyrate-producing bacteria, and a significant reduction in the family
Lachnospiraceae and genus
Ruminococcus[6-7]. In animal experiments, it was found that stress was a predictor of gut microbiota changes during pregnancy. It decreased the β-diversity of gut microbiota, and members of the phylum
Firmicutes and the order
Clostridium (
Lachnospiraceae and
Ruminococcus) increased, paralleling a decrease in the phylum
Bacteroidetes during late pregnancy in humans and animals
[8]. Among pregnant women with prenatal depression, those with scores over 9 on the Edinburgh Postpartum Depression Scale (EPDS) in the second trimester had increased abundance of
Faecalibacterium and decreased levels of
Ruminococcus[9-10].
Regarding the GBA, 5-hydroxytryptamine (5-HT) is an important neurotransmitter in the central and peripheral nervous systems and is closely related to the occurrence and development of depression
[11]. However, this neurotransmitter cannot cross the blood-brain barrier (BBB); the 5-HT found in the brain is mainly derived from metabolized tryptophan that has crossed the BBB
[12]. The BBB, which consists of tightly junctional transmembrane proteins, tightly restricts the diffusion of water-soluble substances in the blood into the brain, such as 5-HT
[13]. Since tryptophan can cross the BBB and promote 5-HT synthesis in the brain, it is suggested that tryptophan in the blood is a predictive factor of central nervous system dysregulations. Tryptophan is the only precursor of 5-HT, and when ingested in a sufficient amount, the amino acid concentration can be used to approximate the amount of 5-HT synthesis, which can directly impact the occurrence of depression
[11]. During pregnancy, the activity of tryptophan catabolites (TRYCAT) pathway will change specifically, which may be related to hormonal changes and immune system, and whether it is related to gut microbiota during pregnancy has not been reported
[14]. Several studies
[12, 15] in animal models or human populations have suggested that the level of tryptophan in the blood may be controlled by the gut microbiota, either directly or indirectly, and the influence of microorganisms on tryptophan metabolism and the 5-HT system may be an important node in this regulation.
Although it was known that the gut microbiota was associated with depression, and tryptophan metabolism was one of the potential mechanisms, the way in which the gut microbiota during pregnancy affected gestational depression has not been fully elucidated. Most of the relevant studies have focused on postnatal depression, while there are relatively few empirical studies on the “gut microbiota-tryptophan-brain axis” in relation to prenatal depression. Furthermore, there is a paucity of research on how gut microbiota changes may be specifically mediated via tryptophan metabolism. Therefore, our study aimed to explore the impact of changes in the gut microbiota during pregnancy on the development of prenatal depression and to evaluate the mediating role of maternal plasma tryptophan in the 5-HT pathway of the GBA.
1 Materials and methods
1.1 Ethics statement
The study protocol was approved by the Medical Ethics Committee of School of Medicine of Wuhan University (2019YF2019). The participants agreed to join the study and signed informed consent.
1.2 Participant recruitment
This was a case-control study conducted in China. In total, we recruited 73 pregnant women in the third trimester of pregnancy from October 2020 to October 2021. Questionnaires, blood samples, and fecal samples were collected in the third trimester.
The inclusion criteria were as follows: 1) Over aged 20; 2) no pregnancy complications; 3) residence in the same place for a long time; and 4) self-reported no cognitive dysfunction. The exclusion criteria were as follows: 1) Self-reported smoking and alcohol abuse; 2) body mass index (BMI)≤30.0 kg/m2 before pregnancy; 3) history of mental illness before pregnancy; 4) a fever within the week prior to sample collection, 5) gastrointestinal symptoms (such as diarrhea, constipation, etc.) within one week prior to sample collection; 6) used medications within 30 days prior to sample collection that could affect the gut microbiota, such as probiotics, dilators, antibiotics and antivirals, etc.
1.3 Prenatal depression measurement
The EPDS is a self-report tool mainly used to screen for prenatal and postpartum depression
[16-17]. The EPDS (Chinese Mainland version) had good reliability, the Cronbach’s alpha was 0.79
[18]. The EPDS included 10 items, with a total score of 0 to 30 points. Higher scores indicate more severe depression. Participants with an EPDS score equal to or greater than 10 were assigned to a prenatal depression group, while those with an EPDS score less than 10 were assigned to a comparison group.
1.4 Fecal DNA extraction and sequencing
The gut microbiota was profiled using 16S ribosomal RNA (rRNA) sequence. Detailed protocols for DNA extraction, polymerase chain reaction (PCR) amplification, and 16S rRNA sequence have been described in previously published articles
[19].
1.5 Plasma and fecal amino acid measurements
Each participant collected 5 mL venous blood by a professional nurse using a sodium citrate (anticoagulant 1꞉9) vacuum anticoagulant tube for determination of plasma amino acids. Samples were processed immediately after collection. After centrifugation, 200 µL/tube (2 tubes) of plasma was collected, labeled with the sample identifier as the participants’ plasma samples, and stored at -80 ℃. Fecal samples are also stored at -80 ℃. Repeated freezing and thawing were avoided.
A targeted ultrahigh performance liquid chromatography-electrospray ionization tandem mass spectrometry (UHPLC-ESI-MS/MS) method has been developed for the quantification of amino acid
[20-21]. The experimental parameters were analyzed on an UHPLC-ESI-MS/MS system, consisting of a Shimadzu MS-8050 mass spectrometer (Kyoto, Japan) and an electrospray ionization source (Turbo Ion-spray), a Shimadzu LC-30AD UHPLC system (Kyoto, Japan), 230AD pumps, a SIL-30AC automatic sampler, a CTO-30A thermostatic column, and a DGU-20A5R degassing device.
1.6 Statistical analysis
The data was analyzed using SPSS 26.0 and R software. For general demographic characteristics analysis, the quantitative data were described with mean±standard deviation (M±SD), and qualitative data were described with frequency and percentage statistically. For gut microbiota analysis, the operational taxonomic unit (OTU) was flattened according to the minimum number of sample sequences to ensure the uniform distribution of gut microbiota. Core OTU analysis was used to analyze whether the sample size was sufficient. Community bar charts were used to depict the species composition of 2 groups of gut microbiota at the phylum and genus levels. The Wilcoxon rank-sum test was used to test for significant differences between 2 groups, to compare the species between the 2 groups at the genus level, and to obtain significantly different species between groups. The α-diversity measures of gut microbiota were used to determine the community richness and diversity of microbial communities. The principal co-ordinates analysis (PCoA) was performed using unweighted-UniFrac distance to determine the β-diversity of gut microbiota and compare the community composition between the 2 groups of samples. In addition, partial least squares discriminant analysis (PLS-DA) was also performed to compare gut microbiota composition between the 2 groups.
Comparisons of the α-diversity and amino acid measures between the 2 groups were performed using the independent samples t-test and the Wilcoxon rank-sum test. Spearman correlation analysis was performed: 1) Between plasma tryptophan and the top 200 most abundant species on genus level; 2) between the flattened OTU-normalized data for 2 groups of species that differ on the genus level and the EPDS scores. Simple linear regression was performed with plasma tryptophan concentration as the dependent variable for genera related to plasma tryptophan concentration. Binary Logistic regression was performed with gut microbiota that were different from the 2 groups. The mediation model was investigated by using model 4 in the PROCESS procedure for SPSS version 4.0.
2 Results
2.1 Demographic characteristics
There were 34 pregnant women in the prenatal depression group and 39 in the comparison group. All participants were married. Most of the pregnant women in the prenatal depression group (88.2%) and the comparison group (71.8%) had a bachelor’s degree or above. There were 31 (91.2%) women in the prenatal depression group and 35 (89.7%) women in the comparison group living in the city. Demographic information is listed in
Table 1. Specific differences in the frequency of food consumption over a one-week timeframe between 2 groups are presented in Supplemental
Table 1 (
https://doi.org/10.57760/sciencedb. 36710).
The vast majority of pregnant women conceived naturally and had planned pregnancies, of which 32 (94.1%) from the prenatal depression group and 35 (89.7%) from the comparison group were naturally conceived, and 21 (61.8%) from the prenatal depression group and 30 (76.9%) from the comparison group had planned pregnancies. All were singleton pregnancies without abnormalities. Normality testing results of quantitative variables are presented in Supplemental
Table 2 (
https://doi. org/10.57760/sciencedb.36710).
2.2 Gut microbiota α- and β-diversity differences
The core OTU analysis meant the common number of OTUs decreased with the number of samples until flat, indicating the sequencing quantity was sufficient, shown in Supplemental
Figure 1 (
https://doi.org/10. 57760/sciencedb.36710) . The community richness of the comparison group was higher than that of the prenatal depression group as measured by Sobs (
P=0.730), Chao (
P=0.754), and Ace (
P=0.820), but the community diversity of the prenatal depression group was higher, as indicated by Shannon (
P=0.835) and Simpson (
P=0.625), as well as coverage (
P=0.673). Gut microbiota α-diversity was not significantly different between the 2 groups (
P>0.05,
Figure 1). Unweighted UniFrac PCoA on OTU level was used and indicated no obvious separation (
r=0.008 6,
P=0.271) between the 2 groups (
Figure 2A). However, PLS-DA results showed the overall structures of maternal gut microbiota on OTU level were significantly different between the 2 groups (
Figure 2B). In PCoA of the PC1 axis, the community composition of the prenatal depression group was more dispersed along the PC1 axis, and its β diversity was lower than that of the comparison group, although the difference was not statistically significant (
P>0.05,
Figure 2C).
2.3 Gut microbiota composition differences
At the phylum level, the community composition of the gut microbiota was the same phyla. The proportions of
Firmicutes were 75.6% in the prenatal depression group and 73.5% in the comparison group, and the proportions of
Actinobacteriota were 13.8% in the prenatal depression group and 14.8% in the comparison group (
Figure 3A).
In the prenatal depression group, 18 genera accounted for over 1% of gut microbiota composition, while there were 23 genera in the comparison group. The dominant genera in both were
Blautia,
Faecalibacterium, and
Bifidobacterium, which accounted for 14.6%, 13.1%, and 11.3% of the prenatal depression group, and 14.5%, 9.20% and 10.9% of the comparison group (
Figure 3B).
On the genus level,
Butyricicoccus and three genera (
unclassified_f__Lachnospiraceae,
Lachnospiraceae_UCG-001, and
Lachnospiraceae_NC2004_group) belonging to family
Lachnospiraceae were significantly more abundant in the prenatal depression group, while
Enterococcus,
Candidatus_Soleaferrea, and
Prevotella were significantly more abundant in the comparison group (
Figure 4).
2.4 Differences in plasma and fecal amino acids concentration
In the comparison group, the concentration of plasma serine was higher than in those with prenatal depression. The concentration of plasma 5-HT (
t=2.070,
P=0.042) and tryptophan (
t=-2.964,
P=0.004) was significantly different between the 2 groups (
Figure 5). There was no significant difference in the TRP/KYN between the prenatal depression group and the comparison group (0.088±0.025 vs 0.079±0.016,
P>0.05). For fecal amino acids concentrations, the concentration of kynurenine was not detected in 2 groups, and there was no difference in the fecal amino acid concentrations between 2 groups (
P>0.05,
Table 2).
2.5 Relationship between maternal gut microbiota and plasma tryptophan
Butyricimonas,
Subdoligranulum,
Faecalibacterium, and other 10 genera had a positive correlation relationship with tryptophan, while
Enterococcus,
Escherichia-Shigella,
[Ruminococcus]_gnavus_group, and
Erysipelatoclostridium were negatively correlated with tryptophan (all
P<0.05,
Figure 6). Using plasma tryptophan concentration as the dependent variable, simple linear regression was performed on tryptophan concentration-related genera. Significance was only found in the negative correlation between
[Ruminococcus]_gnavus_group (
β=-246.942, 95%
CI -402.491 to -91.392),
Enterococcus (
β=-331.150, 95%
CI -544.617 to -117.684) and tryptophan (both
P<0.05,
Table 3).
2.6 Mediating effect of plasma tryptophan between maternal gut microbiota and prenatal depression
Correlation heatmap was performed between the genera differentially expressed in 2 groups and EPDS scores (
Figure 6). We find that
Lachnospiraceae_NC2004_group (
β=5.870,
OR=354.354, 95%
CI 1.24 895 to 100 619.527,
P=0.042) was higher in the prenatal depression group, and
Candidatus_Soleaferrea (
β=-19.945,
OR<0.001, 95%
CI <0.001 to 0.002,
P=0.004) and
Enterococcus (
β=-9.074,
OR<0.001, 95%
CI <0.001 to 0.183,
P=0.016) were higher in the comparison group (
Table 4).
Enterococcus was the only gut microbiota related to plasma tryptophan concentration and prenatal depression after controlling covariates. The mediation effect model showed that plasma tryptophan played a partially intermediate role between maternal gut microbiota and prenatal depression (
Figure 7 and
Table 5). The detailed results obtained by model 4 are shown in Supplemental
Table 3 (
https://doi.org/10. 57760/sciencedb.36710).
3 Discussion
This study explored the mediating role of tryptophan in the relationship between prenatal depression and maternal gut microbiota. We found that the abundances of Candidatus_Soleaferrea and Enterococcus were negatively correlated with prenatal depression, while the abundance of Lachnospiraceae_NC2004_group was positively correlated with prenatal depression. We also discovered a significant difference in plasma tryptophan in pregnant women between the 2 groups. Therefore, it is plausible that plasma tryptophan was a key component in the GBA 5-HT pathway affecting the occurrence of prenatal depression.
Maternal gut microbiota changes dramatically during pregnancy. The GBA theory holds that the composition of the gut microbiota is closely related to negative emotions such as anxiety and depression. The α- and β-diversity can measure the richness and evenness of gut microbiota, and can reflect the stability of gut microbiota. There was no significant difference in the α- and β-diversity of gut microbiota between the prenatal depression group and the comparison group, which was consistent with previous studies
[19, 22-23]. Although the diversity of most studies has remained the same, changes in the abundance of specific bacterial genera may also affect the occurrence of prenatal depression.
Our results indicated there were significant differences in the abundance of
Butyricicoccus,
Enterococcus,
Lachnospiraceae_NC2004_group,
Family_XIII_UCG-001, and
Candidatus_Soleaferrea between the prenatal depression group and the comparison group. Consistent with our findings, previous studies
[24-25] found that those with depression symptoms had higher abundance of the family
Lachnospiraceae and genus
Candidatus_Soleaferrea, and lower abundance of the
Enterococcus compared with those without depression symptoms. The significant differences we found in
Butyricicoccus and
Family_XIII_UCG-001 were consistent with previous findings
[26] in mouse models. In addition to increased chronic social defeat stress that induced depressive behavior, higher abundance of
Family_XIII_UCG-001 was also associated with more serious depression-like behavior
[27].
Since the reduced availability of 5-HT in the brain was a key feature of depression incidence, it can be posited that tryptophan as the sole precursor of 5-HT, plays a significant role in the development of depression
[15, 28]. It was found that 5-HT in blood had an effect on emotional and psychological stress, as the level of 5-HT in the central nervous system of depressed patients was less than that of non-depressed patients
[29]. This study has revealed the decrease of plasma 5-HT was associated with an increased risk of prenatal depression, but 5-HT did not show a mediating or regulatory effect between the gut microbiota and prenatal depression
[30]. We found that plasma tryptophan concentration in those with prenatal depression was significantly different from comparison group, with plasma tryptophan, kynurenine and KYN/TRP higher in the prenatal depression group. Previous study
[31] have shown that the tryptophan concentration in the participants with current major depressive disorder was higher than that in the control group, which was consistent with our research results. A possible explanation is that in the tryptophan metabolic pathway, conversion to 5-HT was inhibited, leading to increased indoleamine-2,3-dioxygenase (IDO) activity, while conversion to kynurenine was enhanced. Therefore, physiological changes that lead to altered tryptophan metabolism may lead to the occurrence of depression
[32-34]. In addition, this study also demonstrated the association between the concentration of plasma tryptophan and the abundance of
Enterococcus. Prior research
[35] has indicated that the abundance of
Enterococcus can influence the metabolism of tryptophan, consistent with our results. Gut microbiota can directly utilize tryptophan, potentially limiting the tryptophan’s bioavailability, reducing the production of 5-HT, and gut microbiota also can indirectly impact the metabolism of tryptophan in blood by disrupting the activity of IDO, thus affecting the downstream synthesis of 5-HT
[15, 36].
This study revealed that
Enterococcus was enriched in the comparison group, and tryptophan was a mediator between
Enterococcus and prenatal depression symptoms. Moreover, the concentration of plasma tryptophan was positively related to the abundance of
Enterococcus, consistent with prior animal studies
[35, 37-38].
Enterococcus faecalis prevented colitis-induced depressive-like behavior through the GBA in mice
[39]. Intervention with probiotics (
Bifidobacterium,
Lactobacillus,
Enterococcus and
Bacillus cereus) can also alleviate depressive symptoms
[40].
Enterococcus has been shown to affect depressive behavior in the brain-gut axis through the inflammatory pathway
[40-43]. A previous publication
[44] suggests a potential mechanistic framework, reporting a significant three-way interaction between prenatal tryptophan levels, IL-6, and EPDS scores. Thus, the mechanism by which
Enterococcus affects prenatal depression may be through its effect on inflammatory factors, thereby changing tryptophan metabolism and ultimately affecting 5-HT production. According to the monoamine hypothesis, 5-HT is one of the major neurotransmitters in depression, and there is a significant decrease in 5-HT levels in depressed individuals
[45].
This study also found that
Candidatus_Soleaferrea was enriched in the comparison group, while
Lachnospiraceae_NC2004_group was enriched in the prenatal depression group.
Candidatus_Soleaferrea belongs to the family
Ruminococcaceae. In an animal model, supplementation with
Enterococcus altered gut microbial composition by increasing the abundance of
Candidatus_Soleaferrea[35]. There is evidence to suggest that the family
Ruminococcaceae is related to tryptophan metabolism and depression, concurrent with the improvement of depression symptoms, the concentration of tryptophan decreased while the abundance of
Ruminococcaceae increased
[46].
Lachnospiraceae_NC2004_group belongs to the family
Lachnospiraceae. Prior study
[47] had linked the family
Lachnospiraceae to depression. As there are few studies on these genera, the role of
Candidatus_Soleaferrea and
Lachnospiraceae_NC2004_group in prenatal depression can be investigated further.
We explored the potential association between “gut microbiota-tryptophan- prenatal depression” in the Chinese pregnant women and found that the specific genus, such as Enterococcus and Candidatus_Soleaferrea had a protective effect against prenatal depression. There are several limitations in our study. Firstly, although we used a self-made 10-item diet scale to investigate diet, diet as one of the important factors affecting gut microbiota, still needed to be carefully considered. Secondly, all the participants were highly educated pregnant women from the same region, and some of the research data were self-reported, which may lead to certain selection bias and information bias. In addition, shotgun metagenomic sequencing is more capable of analyzing the complex composition and functions of gut microbiota. Compared with shotgun metagenomic sequencing, the 16S rRNA sequence has some limitations. Blood and fecal samples were only collected in the third trimester of pregnancy, making it difficult to detect the changes in the gut microbiota and plasma tryptophan throughout pregnancy. Although demographic and dietary factors that could cause confounding were controlled, endocrine indicators (such as hormone levels) and psychological-social factors (such as stress, family support, etc.) that might have an impact were not included and analyzed. These variables will be included in the consideration of covariates in future research. Finally, the small sample size resulted in limited generalizability.
In summary, prenatal depression is related to plasma tryptophan concentration. The abundances of Candidatus_Soleaferrea and Enterococcus were negatively associated with prenatal depression, while the abundance of Lachnospiraceae_NC2004_group was positively associated with prenatal depression.
the National Natural Science Foundation of China(81903334)
the National Natural Science Foundation of China(82473647)
©Journal of Central South University (Medical Science). All rights reserved.
京公网安备11010202008535号
PDF
(2871KB)
