Gut Microbes

ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/kgmi20

Indole derivatives ameliorated the methamphetamine-

induced depression and anxiety via aryl hydrocarbon
receptor along “microbiota-brain” axis

Xi Wang, Miaoyang Hu, Weilan Wu, Xinyu Lou, Rong Gao, Tengfei Ma, S
Thameem Dheen, Jie Cheng, Jianping Xiong, Xufeng Chen & Jun Wang

To cite this article: Xi Wang, Miaoyang Hu, Weilan Wu, Xinyu Lou, Rong Gao, Tengfei Ma,
S Thameem Dheen, Jie Cheng, Jianping Xiong, Xufeng Chen & Jun Wang (2025) Indole
derivatives ameliorated the methamphetamine-induced depression and anxiety via aryl
hydrocarbon receptor along “microbiota-brain” axis, Gut Microbes, 17:1, 2470386, DOI:
10.1080/19490976.2025.2470386

To link to this article: https://doi.org/10.1080/19490976.2025.2470386

© 2025 The Author(s). Published with

license by Taylor & Francis Group, LLC.

View supplementary material

Published online: 25 Feb 2025.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=kgmi20
GUT MICROBES
2025, VOL. 17, NO. 1, 2470386
https://doi.org/10.1080/19490976.2025.2470386

RESEARCH PAPER

Indole derivatives ameliorated the methamphetamine-induced depression and

anxiety via aryl hydrocarbon receptor along “microbiota-brain” axis
Xi Wanga,b*, Miaoyang Hua*, Weilan Wuc, Xinyu Loua, Rong Gaoc, Tengfei Mad, S Thameem Dheenb, Jie Chenga,
Jianping Xionga, Xufeng Chene, and Jun Wanga,e
a
Center for Global Health, The Key Laboratory of Modern Toxicology, Ministry of Education, Department of Toxicology, School of Public Health,
Nanjing Medical University, Nanjing, China; bDepartment of Anatomy, Yong Loo Lin School of Medicine, National University of Singapore,
Singapore, Singapore; cDepartment of Hygienic Analysis and Detection, The Key Laboratory of Modern Toxicology, Ministry of Education,
School of Public Health, Nanjing Medical University, Nanjing, China; dStem Cell and Neural Regeneration and Key Laboratory of Cardiovascular
& Cerebrovascular Medicine, School of Pharmacy, Nanjing Medical University, Nanjing, China; eDepartment of Emergency Medicine, The First
Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China

ABSTRACT ARTICLE HISTORY

In addition to the high neurotoxicity, depression, and anxiety are the most prominent character­ Received 28 October 2024
istics of methamphetamine (Meth) withdrawal. Studies to date on the issue of Meth-associated Revised 24 January 2025
depression and anxiety are focused on the brain, however, whether peripheral homeostasis, Accepted 17 February 2025
especially the “microbiota-gut” axis participates in these adverse outcomes, remains poorly under­ KEYWORDS
stood. In the current study, with the fecal microbiota transplantation (FMT) assay, the mice received Depression and anxiety;
microbiota from Meth withdrawal mice displayed marked depression and anxiety behaviors. The methamphetamine; indole
16S rRNA sequencing results showed that Meth withdrawal contributed to a striking reduction of derivatives; aryl hydrocarbon
Akkermansia, Bacteroides, Faecalibaculum, Desulfovibrio, and Anaerostipes, which are known to be receptor
associated with tryptophan (TRP) metabolism. Noteworthily, the substantial decreases of the
indole derivatives from the TRP metabolic pathway, including IAA, IPA, ILA, IET, IArA, IAld, and
TRM were observed in the serum of both Meth abusing humans and mice during Meth withdrawal
with the UHPLC-MS/MS analysis. Combining the high and low TRP diet mouse model, the mice
with high TRP diet obviously impeded Meth-associated depression and anxiety behaviors, and
these results were further strengthened by the evidence that administration of IPA, IAA, and indole
dramatically ameliorated the Meth induced aberrant behaviors. Importantly, these protective
effects were remarkably counteracted in aryl hydrocarbon receptor knockout (AhR KO) mice,
underlining the key roles of microbiota-indoles-AhR signaling in Meth-associated depression and
anxiety. Collectively, the important contribution of the present work is that we provide the first
evidence that peripheral gut homeostasis disturbance but not limited to the brain, plays a key role
in driving the Meth-induced depression and anxiety in the periods of withdrawal, especially the
microbiota and the indole metabolic disturbance. Therefore, targeting AhR may provide novel
insight into the therapeutic strategies for Meth-associated psychological disorders.

1. Introduction
Meth abuse is more prone to causing severe
According to the 2023 World Drug Report, the psychological disorders, particularly depression
number of individuals who used amphetamine- and anxiety, and the depressive symptoms often
type stimulants reached as high as 36 million,1 emerge during Meth withdrawal period.2
and the number of Meth abusers is still increas­ A cross-sectional study reported a depression
ing. As a typical representative of amphetamine- prevalence rate of 39.5% among 1,277
type drugs, Meth (commonly known as “ice”) is participants,3 while another cohort study indi­
highly addictive and neurotoxic, with an average cated that 157 out of 243 participants (64.6%)
elimination half-life approximately to 10 hours in exhibited depressive symptoms during Meth
individuals. Compared to the traditional drugs, withdrawal, with 74 (30.5%) experiencing

CONTACT Xufeng Chen [email protected] Department of Emergency Medicine, the First Affiliated Hospital of Nanjing Medical University, 300
Guangzhou Road, Nanjing, Jiangsu 210029, China; Jun Wang [email protected] The Center for Global Health, the Key Laboratory of Modern
Toxicology, Ministry of Education, Department of Toxicology, School of Public Health, School of Public Health, Nanjing Medical University, Nanjing 211166, China
*Xi Wang and Miaoyang Hu contributed equally to this work.
Supplemental data for this article can be accessed online at https://doi.org/10.1080/19490976.2025.2470386
© 2025 The Author(s). Published with license by Taylor & Francis Group, LLC.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted
Manuscript in a repository by the author(s) or with their consent.
2 X. WANG ET AL.

moderate and 83 (34.1%) experiencing severe which has been demonstrated to inhibit CNS
depression.4 Meth abuse inflicts severe physical inflammation.13 However, whether the microbiota
and psychological harm, such as self-harm and and (or) the corresponding metabolites partici­
suicidal behaviors, posing significant challenges pated in driving the Meth-induced depression
to public health. Therefore, alleviating Meth and anxiety, to date, remain largely unknown.
withdrawal-related depressive symptoms and Aryl hydrocarbon receptor (AhR), a transcriptional
deciphering the underlying mechanisms is factor expressed in immune cells, microglia, intestinal
imperative. epithelial cells, etc., can be activated by specific small
For Meth-induced depression and anxiety, more molecules from diet, microbial metabolites, and envir­
attention is paid to the central nervous system onmental pollutants.14,15 Upon binding with its
(CNS), while few studies have tried to unravel the ligands, AhR translocates to the nucleus and forms
cellular basis of Meth-induced psychological disor­ a heterodimer with ARNT to regulate CYP1A1,
ders from peripheral systems. As a “second brain,” CYP1A2, and CYP1B1 expression,16 a negative feed­
the gut and its interactions with microbiota are back loop that is crucial for indole and its derivatives
implicated in CNS activities, including mood, degradation. A study reported that by colonizing
stress response, and cognitive function through Escherichia coli modified for indole metabolism and
neural, endocrine, immune, and metabolic TRP metabolism-deficient Escherichia coli, indole,
pathways,5–7 therefore, targeting gut microbiota a TRP metabolite of intestinal flora, and its derivatives
for the treatment of psychological disorders might can activate neurogenesis through the AhR
be a promising therapeutic approach. The gut- pathway.17 Additionally, intermittent fasting may
brain axis is a complex bidirectional communica­ promote axonal regeneration after sciatic nerve
tion network that links the gut microbiota to the crush injury in mice by enhancing IPA production
CNS. Tryptophan (TRP), one of the nine essential by gut Gram-positive bacteria.18 Indole has been
amino acids, plays a crucial role in coordinating found to harbor the neuroprotective properties both
gastrointestinal function and CNS activities. TRP is in vitro and in vivo,19 since supplementation with
primarily metabolized through three pathways: the indole and its metabolites such as IS, IPA and IAld
kynurenine pathway, the serotonin pathway, and significantly reduces CNS inflammation in germ-free
the microbial metabolic pathway (Indole metabolic mice.13 Nonetheless, whether the indole metabolic
pathway).8 For the microbial metabolic pathway, pathway is involved in Meth-induced depression
the gut microbiota metabolizes TRP into several and anxiety remains ambiguous. Thus, deciphering
indole derivatives, such as indole-3-aldehyde the roles of indole derivatives and AhR signaling in
(IAld), indole-3-acetic acid (IAA), indole-3-pro­ Meth-induced neurobehavioral changes is of signifi­
pionic acid (IPA), indole-3-pyruvate (IPYA), etc.9 cant importance.
In recent years, accumulating reports pointed to Therefore, in the present study, the Meth with­
the influence of the TRP metabolic pathway far drawal mouse model was established to evaluate
beyond the traditional focus on the study of the the effects of Meth on gut microbiota composition,
serotonin and the kynurenine pathways, the TRP TRP metabolic pathways, and the interaction with
microbial metabolic pathway has gained more and AhR. Additionally, a Meth mouse model supple­
more attention since the metabolites indole and its mented with indole derivatives and AhR agonist, in
derivatives are recognized as the essential signaling combination with AhR KO mice were used, aiming
molecules within the microbiota-gut-brain axis, to elucidate the critical roles of the AhR, which may
capable of influencing brain function and serve as a bridge between the gut microbiota and
behaviors.10 In depression patients, the levels of the CNS, particularly in the regulation of microglial
indole and its derivatives have been shown to cor­ morphology and neurogenesis. Moreover, the
relate closely with the severity of depressive and levels of the indole metabolites in the serum of
anxiety symptoms.11 For instance, Bifidobacterium the Meth withdrawal individuals were also evalu­
breve CCFM1025 effectively alleviates depression ated, which may provide solid support for the dis­
and related gastrointestinal disorders in patients, turbance of the indole metabolic pathway in Meth-
concomitant with the increased levels of IPA,12 induced depression and anxiety.
 GUT MICROBES 3

2. Materials and methods 2.2.2. Fecal microbiota transplantation (FMT)

treatment
2.1. Reagents and antibodies
A gut microbiota depletion mouse model was con­
Methamphetamine (purity over 99%) was provided structed with a one-week antibiotic cocktail (ABX)
by the National Institutes for Food and Drug Control treatment, consisting of vancomycin (50 mg/kg/
(Beijing, China). Standard substances IAA (CAS: 87- day), neomycin (100 mg/kg/day), metronidazole
51-4), IPA (CAS: 830-96-6), IArA (CAS: 29953-71-7), (100 mg/kg/day), ampicillin (100 mg/kg/day), and
IAM (CAS: 879-37-8), IPYA (CAS: 392-12-1), IAld amphotericin B (1 mg/kg/day). All antibiotics were
(CAS: 487-89-9), TRM (CAS: 61-54-1), IET (CAS: purchased from Solarbio Science & Technology
526-55-6), and IND (CAS: 120-72-9) were purchased Co., Ltd. (Beijing, China). The depletion of endo­
from Sigma-Aldrich (Darmstadt, Germany). ILA genous gut microbiota was assessed by aerobic/
(CAS: 1821-52-9) was purchased from Macklin anaerobic culturing on TSA plates. Fresh fecal sam­
(Shanghai, China). Internal standards Trp-D5 and ples from Meth-withdrawal donor mice were col­
IAA-D5 were obtained from Toronto Research lected, weighed, and suspended in sterile PBS at
Chemicals (Toronto, Canada). HPLC-grade metha­ a 1:10 ratio (w/v, feces/PBS). The suspension was
nol, acetonitrile, formic acid, and 6-Formylindolo vortexed, then centrifuged at 1000 rpm for 5
[3,2-b] carbazole (Ficz) were purchased from Sigma- minutes at 4°C. The supernatant was collected
Aldrich (Darmstadt, Germany). Anti-DCX (4604S) and centrifuged at 8000 rpm for 5 minutes at 4°C
was purchased from Cell Signaling Technology to remove bacterial pellets. The final mixture was
(MA, USA). Anti-IBA1 (019–19741) was purchased resuspended in sterile PBS for transplantation.
from WAKO (Tokyo, Japan). Anti-IBA1 (ab5076) Recipient mice (including FMT-Control, Saline,
was obtained from Abcam (Cambridge, UK). Alexa and FMT-WD 14d groups) were administered
Fluor™ 488 goat anti-rabbit IgG (A-11008) was pur­ 200 μL fecal suspension via oral gavage every two
chased from Thermo Fisher Scientific (Waltham, MA, days for 14 consecutive days.
USA). The HRP-conjugated secondary antibodies
were from Biosharp (Nanjing, China). 2.2.3. Dietary TRP supplementation
The TRP feed was purchased from TROPHIC Animal
Feed High-Tech Co., Ltd (Nantong, China) and the
2.2. Animals and treatment mice were randomly divided into four groups:
Control (Saline), Low TRP (0.09% TRP), Meth with­
Male wild-type (WT) C57BL/6 mice (6 weeks old, drawal (WD 14d), and Meth withdrawal with high
15–20 g) were purchased from the Experimental TRP diet (WD 14d+High TRP, 1.19% TRP). Mice in
Animal Center of Nanjing Medical University and the control and Meth withdrawal groups were fed
maintained in an SPF environment with a controlled a control diet (AIN93M standard diet for rodents).
room temperature (RT, 21–23°C), humidity (50–­
60%) and 12 h/12 h light-dark cycle. AhR−/− C57BL/ 2.2.4. Indoles supplementation and Ficz treatment
6 mice were obtained from GemPharmatech Co., Ltd The indoles solution (a mixture of IAA, IPA, and
(Nanjing, China). All animal experiments in this indole) was prepared with a final concentration of
study were approved by the Experimental Animal 300 μM before being dispensed into drinking bottles.
Welfare Ethics Review Committee of Nanjing The AhR agonist, Ficz (Sigma-Aldrich; USA) was
Medical University (approval No. IACUC-2302024). dissolved by DMSO (Sigma-Aldrich, USA) and sub­
sequently diluted with olive oil (Sigma-Aldrich). The
2.2.1. Meth administration solution was administered via intraperitoneal injec­
After 1 week of acclimatization, the mice were ran­ tion (i.p.) at a dosage of 0.04 mg/kg/day, thrice weekly.
domly divided into control and Meth withdrawal
groups (5 mg/kg, i.p., once daily for 10 consecutive
2.3. Population samples
days, followed by a 14-day withdrawal, after that
named “WD 14d”). The control group received This study employed a rigorous cross-sectional
saline injections. design, recruiting a total of 78 Meth abusers and
4 X. WANG ET AL.

79 healthy controls. The sample size was deter­ were sequenced on an Illumina MiSeq platform
mined using G Power20 based on an a priori (Illumina Inc., San Diego, CA, USA) to generate
power analysis, which calculated the required sam­ paired-end reads. Sequencing data were processed
ple size to achieve a statistical power of 0.8 with and analyzed using BMKCloud (www.biocloud.
a medium effect size (Cohen’s d = 0.5) and net), where bioinformatics pipelines including
a significance level of 0.05 (Figure S5). All partici­ quality filtering, operational taxonomic unit
pants provided informed consent for sample col­ (OTU) clustering, and taxonomic classification
lection and participation in the study. The study were applied to assess microbial diversity and com­
protocol was approved by the Medical Ethics position across different experimental groups.
Committee of Changshu First People’s Hospital,
affiliated with Suzhou University (Ethical
2.6. UHPLC-MS/MS analysis of indole derivatives
Application and Approval 2016–01). Serum sam­
ples from individuals with a history of Meth abuse The sample preparation of serum and fecal samples
within 1 month and the serum of Meth abusers and was optimized based on previous studies.21 In brief,
healthy controls were collected for analysis. after preparing the serum and fecal samples, the
samples were analyzed using an ExionLC™ AC
HPLC system coupled to a SCIEX Triple Quad
2.4. Assessment of depression- and anxiety-like
5500 Plus mass spectrometer (AB SCIEX, USA).
behaviors
The mass spectrometer was operated in positive
Mouse depression-like behaviors were assessed by electrospray ionization mode and multiple reaction
the tail suspension test (TST) and forced swim test monitoring (MRM) mode. Data were acquired by
(FST), while anxiety-like behaviors were evaluated Analyst® software (AB SCIEX, USA) and analyzed
by the open field test (OFT) and elevated plus maze by MultiQuant™ software (AB SCIEX, USA).
(EPM) test. For a detailed protocol, please refer to
the Supplementary Materials.
2.7. Immunofluorescence and microglial
morphometric analysis
2.5. 16S rRNA sequencing and analysis
Mice were perfused with 4% paraformaldehyde,
Fecal samples were collected from mice at 8:00–­ and the brain tissue was collected and fixed in 4%
10:00 AM on WD 0d and WD 14d. The samples paraformaldehyde. After dehydration, the tissue
were placed into sterile tubes, and immediately was embedded in an optimal cutting temperature
frozen in liquid nitrogen. Total microbial DNA (OCT) compound and cryosectioned in liquid
was extracted using the QIAamp DNA Stool Mini nitrogen after freezing. Coronal brain sections of
Kit (Qiagen, Hilden, Germany) according to the the hippocampus were obtained using a cryostat
manufacturer’s instructions. For bacterial 16S maintained at −15°C to −20°C, with a section
rRNA gene sequencing, the V3-V4 hypervariable thickness of 25–30 μm. The sections were blocked
regions of the 16S rRNA gene were amplified using with a blocking solution containing 5% goat serum,
Premix Ex Taq™ Hot Start Version (Takara, Dalian, then permeabilized with 0.3% Triton X-100.
China) with the following universal primers: Afterward, the primary antibodies were incubated
319F (5′-ACTCCTACGGGAGGCAGCAG-3′) overnight, followed by incubation with secondary
and 806 R (5′-GGACTACHVGGGTWTCTAAT antibodies. After washing, the sections were
-3′). Amplicons were purified using the AxyPrep mounted with DAPI (ab104139, Abcam).
DNA Gel Extraction Kit (Axygen Biosciences, Photomicrographs were acquired using a Zeiss
Union City, CA, USA) and quantified with LSM900 scanning laser confocal microscope
QuantiFluor-ST (Promega, USA). The purified (Zeiss, Oberkochen, Germany). The 3D recon­
amplicons were then pooled in equimolar concen­ struction of microglia was imaged on the confocal
trations and subjected to library preparation. The laser scanning microscope with Z-stacks completed
quality and quantity of the amplicon library were in the z-axis direction with a step size of 1.0 μm.
assessed before sequencing. The prepared libraries Microglial morphology was recorded and analyzed
 GUT MICROBES 5

using IMARIS software (version 9.0.1, Bitplane, an RPMI digestion solution containing 1 mg/mL type
Switzerland). 5–6 cells were reconstructed for II collagenase (Sigma, USA) and 1% DNAse (Sigma,
each mouse, with 3–4 mice per group. USA), followed by incubation at 37°C for 45 minutes.
After digestion, myelin was removed using a 30%
Percoll solution in RPMI. The isolated cells were
2.8. Quantitative RT-PCR
washed, incubated with an Fc receptor-blocking solu­
Total RNA was extracted from brain tissue lysates tion for 5 minutes, and then incubated with antibo­
using the TRIzol reagent (343903, Life Technologies, dies. Microglia were identified as CD45midCD11b+
USA). The concentration and purity of the extracted cells. All antibodies were purchased from
RNA were assessed using a NanoDrop 2000 spectro­ eBioscience (Thermo Fisher Scientific, USA).
photometer (Thermo, USA). SYBR Green Supermix
(11201ES08, Yeasen Biotech Co., Ltd., China) was
used to measure relative mRNA levels. Fluorescence 2.12. Statistical analysis
changes were tracked using a LightCycler 480 system Statistical analyses were performed using GraphPad
(Roche, Switzerland), with GAPDH serving as the Prism 8.0 (San Diego, CA, USA). Most data are presented
reference gene. The primer sequences of RT-PCR as mean ± SEM, except for box plots, which display max­
are detailed in Supplementary Materials Table S1. imum and minimum values. Statistical significance was
defined as p < 0.05. The Shapiro-Wilk test and Levene’s
2.9. Histological analysis test were used to assess normality and homogeneity of
variance, respectively. For comparisons between two
Colon tissues were fixed in formalin, embedded in independent groups that met these assumptions, an
paraffin, and sectioned into 4 μm-thick slices for his­ unpaired Student’s t-test was employed, while the Mann-
topathological analysis. These sections were stained Whitney U test was used for non-normally distributed
with hematoxylin and eosin (H&E), then scanned data. For multiple group comparisons, one-way or two-
using a Pannoramic scanner (3DHISTECH, way analysis of variance (ANOVA) was applied as appro­
Germany), and histological scores were determined priate, followed by Tukey’s post hoc test for pairwise
according to previous methods.22 comparisons due to its high efficiency in analyzing data­
sets with equal or similar sample sizes. For 16S rRNA
2.10. Microscale thermophoresis (MST) sequencing data, principal coordinate analysis (PCoA)
was used to evaluate the main compositional features of
The MST assay was performed based on the previous the samples. Differential microbial taxa between the two
study.23 IPA, IAA and indole were diluted to eight groups were identified using the Wilcoxon rank-sum test
different concentrations and mixed in equal amounts, due to the non-normal distribution of the data. For multi­
and the concentration of GFP-AhR was determined ple-group comparisons, the Kruskal-Wallis test followed
using a GFP quantification kit (abcam, ab235672, by the Benjamini-Hochberg procedure for false discovery
USA). The mixed samples were loaded into capillaries rate correction was employed. Receiver operating charac­
(Monolith NT.115 Capillary, NanoTemper, teristic (ROC) curve analysis was performed using SPSS
Germany) and the thermophoresis signals were mea­ 22.0 (Chicago, IL, USA) to differentiate Meth abusers
sured using a Monolith NT.115 instrument from healthy controls based on indole derivative levels.
(NanoTemper, Germany) following the manufac­
turer’s standard protocol. Fluorescence thermophor­
esis and changes in kd values were analyzed using 3. Results
Nano Temper Analysis software (Germany).
3.1. Meth-induced depression and anxiety
behaviors and microbiota disturbance in mice
2.11. Flow cytometry
Since microbiota dysbiosis influences depression
Mice were anesthetized and perfused with ice-cold and anxiety behaviors,24 it is logical to explore
PBS to remove circulating leukocytes. After perfusion, whether Meth exposure impacts microbiota and
the hippocampus was gently dissociated and placed in the behaviors. To this end, the microbiota changes
6 X. WANG ET AL.

and the psychological behaviors induced by Meth both of the TST and FST. Additionally, the mice
were examined. With the daily intraperitoneal spent less time in the center arena in OFT and open
injection of Meth (5 mg/kg) for 10 days and the arms in the EPM test, characterized as anxiety
following 14-day withdrawal, the mouse fecal sam­ behaviors (Figure 1(c–f)). To gain a more compre­
ples were collected on day 10 and day 24, then by hensive knowledge of the behavioral changes
behavioral testing (Figure 1(a)). During the Meth induced by Meth, the composition of the gut
exposure and withdrawal periods, mice exhibited microbiota was analyzed. The alpha diversity was
a significant decrease in body weight compared to assessed by calculating the Ace, Chao, and
the control group (Figure 1(b)). In parallel, the Shannon indices at the OTU level. It showed that
withdrawal mice displayed significant depression the Ace and Chao indices of the WD 0d and WD
behaviors, with increased immobility states in 14d groups were significantly higher than those of

Figure 1. Effects of Meth on mouse behaviors and gut microbiota. (a) Experimental design and timeline. (b) Body weight change
(n=8–10 per group). (c,d) Immobility time (%) in the FST and TST; time spent in the open arms (%) in the EPM test; time spent in the
center (s) and total track length (m) in the OFT (n=8 per group). (e,f) The representative trajectory diagrams in the OFT and EPM. (g)
Ace, Chao, and Shannon index of α diversity. (h) Principal coordinate analysis of β diversity based on Bray-Curtis distances among the
control, saline, WD 0d, and WD 14d groups. (i) Stacked chart of gut microbiota composition at the phylum level. (j) The ratio of phylum
Firmicutes/Bacteroidetes. (k) Genomic functional prediction of gut microbiota community by Tax4Fun. (l) Random forest analysis of gut
microbiota importance at the genus level on WD 14d and control, saline groups. Each dot indicates an individual mouse. Data are
expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001.
 GUT MICROBES 7

the control group, and the Shannon index of the was also displayed in humans, the levels of indole
WD 14d group was higher than that of the control metabolites in the serum of human healthy controls
groups albeit the WD 0d group exhibited no (HC) and Meth abusers were examined
obvious changes (Figure 1(g)). According to prin­ (Figure 2(d)). Consistently, the levels of indole deri­
cipal coordinate analysis (PCoA), the beta diversity vatives, including IAA, IPA, ILA, IArA, IAld, TRM
of the gut microbiota differed among the control, and IET were strikingly decreased in Meth abusers
WD 0d, and WD 14d groups (Figure 1(h)). For (Figure 2(e)). Noteworthily, the receiver operating
microbiota abundance, the relative abundance of characteristic (ROC) curve displayed a high capa­
Firmicutes significantly increased whereas the city in distinguishing healthy controls from Meth
abundance of Verrucomicrobia dramatically abusers, with an AUC value of 89.6803
decreased in the WD 0d and WD 14d groups (Figure 2(f)). Interestingly, with the 16S RNA
(Figure 1(i)). Additionally, the ratio of Firmicutes sequence assay, the microbiota known to metabo­
to Bacteroidetes (Figure 1(j)) obviously increased in lize TRP into indole derivatives, including
WD 0d group, even lasting 14 days post Meth Akkermansia, Bacteroides, Faecalibaculum,
treatment, reflecting a stable state of dysbiosis. Desulfovibrio, and Anaerostipes were markedly
Furthermore, the Random Forest analysis ranked reduced in the WD 14d group (Figure 2(g)).
the importance of the microbiota on WD 14d and Among these taxa, Akkermansia was significantly
the control group at the genus level. Tax4Fun was and consistently reduced following Meth exposure,
employed to evaluate the functional changes in the prompting further investigation into its functional
gut microbiota. The analysis revealed that the phe­ role in TRP metabolism. To validate the contribu­
nylalanine, tyrosine, and tryptophan biosynthesis, tion of A. muciniphila to indole metabolism, we
as well as the tryptophan metabolism (highlighted assessed its ability to metabolize TRP into indole
in red), exhibited significant differences between derivatives. Using the A. muciniphila MucT (ATCC
the WD14d and control groups, with a notable BAA-835) strain, we demonstrated that
reduction in activity observed in the WD14d A. muciniphila could metabolize TRP into multiple
group. Additionally, pathways such as ABC trans­ indole derivatives, including IPA, IAld, IArA, and
porters, other glycan degradation, and degradation TRM, in a concentration-dependent manner
of aromatic compounds also demonstrated signifi­ (Figure S1). These findings provide direct evidence
cant variations between the two groups linking the reduction of A. muciniphila to the
(Figure 1(k)). Based on the mean decrease in the decrease in indole derivatives observed during
Gini coefficient, the top six genera contributing to Meth withdrawal. Collectively, these results suggest
group differences were Bacteroides, Ileibacterium, that a pronounced blockage of indole metabolism
Faecalibaculum, Anaerostipes, Akkermansia, and via the reduced indole-producing microbiota by
Desulfovibrio (Figure 1(l)). Meth.

3.2. Meth withdrawal dramatically decreased 3.3. Transplantation of feces from

indole derivatives metabolized by the gut Meth-withdrawal mice resulted in depression- and
microbiota anxiety-like behaviors in recipient mice

To further tie together the microbiota dysbiosis and Having determined the Meth induced microbiota
the depression and anxiety behaviors induced by dysbiosis, we hypothesized whether microbiota
Meth, the indole metabolic pathway (Figure 2(a)) dysbiosis can induce anxiety- and depression-like
was detected in mouse feces and serum. Of note, the changes. Herein, the fecal microbiota transplanta­
levels of most indole metabolites, including IAA, tion (FMT) experiment was performed. Figure 3(a)
IPA, ILA, IArA, IAM, IPYA, IAld, TRM and IET presents the experimental timeline, dividing the
were significantly reduced in the WD 14d group mice into two groups: Donor and Recipient mice.
compared with the control group, implying The donor mice were treated with either Meth or
a blockage in indole metabolism (Figure 2(b,c)). saline, followed by behavioral testing and fecal
To verify whether the aberrant indole metabolism sample collection on the last day of withdrawal
8 X. WANG ET AL.

Figure 2. Indole metabolism in Meth withdrawal mice and humans. (a) Schematic representation of the TRP microbial metabolic
pathway, including IAA, IPA, ILA, IArA, IAM, IPYA, IAld, TRM and IET. (b) Analysis of indole derivatives in feces of WD 14d mice
compared to control mice (n = 8 per group). (c) Analysis of indole derivatives in serum samples in mice. (d) Schematic of sample
collection, pre-treatment, and mass spectrometry detection and analysis in healthy controls and Meth abusers. (e) Comparison of
indole derivatives in the serum of Meth abusers versus healthy controls (HC) (n = 78–80 per group). (f) ROC curve for distinguishing
Meth abusers from healthy controls based on indole derivative levels. (g) Relative abundance of gut microbiota genera in WD 14d and
control mice. Each dot indicates an individual mouse. Data are expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p <
0.0001.

(WD14d). For recipient mice, the gut microbiota gut dysbiosis, we conducted supplementary experi­
was depleted with an antibiotic cocktail (ABX), ments focusing on intestinal barrier integrity and
then received FMT from the donor mice. To mucus production. Our results revealed that Meth
address the potential concern regarding residual withdrawal (WD 14d) significantly increased
Meth or its metabolites in donor fecal material, serum LPS levels, indicating compromised gut bar­
we performed UHPLC-MS/MS analysis to detect rier function (Figure S4(a)). Tight junction-related
Meth levels in fecal samples at different time points genes (ZO1, Occludin, Claudin5) were significantly
post-injection. Meth was detectable in the feces downregulated in WD 14d mice, consistent with
2 hours after injection. However, no residual barrier dysfunction (Figure S4(b)). Additionally,
Meth was detected in feces collected at 1 day, 3 goblet cell numbers were markedly reduced
days, or 14 days post-injection. These results con­ (Figure S4(c)), accompanied by a significant
firm that the effects observed in FMT-WD 14d decrease in MUC2 expression and mucus layer
mice were not attributable to residual Meth in thickness, as demonstrated by immunofluores­
donor fecal material but rather to changes in the cence and Alcian blue staining (Figures S4(d) and
gut microbiota composition (Figure S3). To further S4(e)). These disruptions in the mucus layer are
explore the mechanisms underlying Meth-induced particularly relevant to the observed reduction in
 GUT MICROBES 9

Figure 3. Roles of the microbiota in mediating Meth-induced anxiety and depression-like behavior in mice. (a) Experimental design
and timeline. (b,c) Time spent in the open arms (%) in the EPM test; and time spent in the center (s) and total track length (m) in the
OFT; immobility time (%) in the FST and TST. (d)The representative trajectory diagrams in the OFT and EPM. (e,f) Levels of indole
derivatives in the serum and feces of FMT recipient mice. (g-h) H&E staining of the colons (scale bar 100 μm) and the histological score
analysis. (i) Villus length analysis. Data are expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001; #p < 0.05,
##
p < 0.01, ###p < 0.001, ####p < 0.0001.

Akkermansia abundance, as this bacterium ecological niche, leading to its decline during
depends on mucin as its primary nutrient source. Meth withdrawal. This loss of Akkermansia, in
The reduction in mucus production and goblet cell turn, may exacerbate gut barrier dysfunction and
numbers likely deprives Akkermansia of its further contribute to gut dysbiosis. Behavioral
10 X. WANG ET AL.

assessments indicated that the WD 14d donor mice (Figure 4(b)), in parallel, it retarded the Meth-
and the recipient mice (FMT-WD 14d) exhibited induced decreased levels of IAA, IPA, ILA, IArA,
significant anxiety- and depression-like behaviors, IET, IAM, IPYA, IAld and TRM. Particularly the
characterized by reduced time spent in the open levels of IPA, IAM, IPYA, and TRM (Figure 4(c)),
arms of the EPM and the center of the OFT, suggesting a salutary role of TRP metabolites in
decreased total distance traveled in the OFT, and alleviating Meth-induced anxiety and depression-
increased immobility in the FST and TST (Figure 3 like behaviors. To further substantiate the causal
(b–d)). Furthermore, the indole derivatives in the link between the reduction in indole derivatives
serum and feces of recipient mice significantly and AhR pathway activity, we performed qPCR
reduced (Figure 3(e,f)), indicating that Meth- analysis of AhR-related genes. Our results revealed
induced behavioral changes might be partially that both the low TRP diet group and the WD 14d
ascribed to the microbiota dysbiosis. Since the group exhibited significant downregulation of Ahr,
impairments of the intestinal integrity were linked Cyp1a1, and Cyp1b1, indicating suppression of the
to the psychiatric disorders, the histological analy­ AhR pathway. Importantly, supplementation with
sis of the intestinal structure in both donor and the high TRP diet reversed this suppression, restor­
recipient mice was conducted. It showed that the ing gene expression levels to those comparable to
pathological scores were significantly increased the control group (Figure S2). These findings con­
and the villus length was notably reduced in the firm that the reduction in indole derivatives
WD 14d and FMT-WD 14d mice (Figure 3(g–i)), observed in vivo contributes to AhR pathway dys­
further supporting the critical role of gut micro­ regulation and suggest that dietary TRP supple­
biota and the corresponding metabolites in Meth- mentation mitigates these effects by restoring
induced anxiety and depression-like behaviors. indole derivative levels and AhR signaling activity.

3.4. Dietary TRP improves the aberrant behaviors 3.5. Meth withdrawal causes microglial dystrophy
and metabolism in Meth withdrawal mice and reduces neurogenesis in the hippocampus
Having determined the reduction of indole meta­ To further decipher the adverse impact of Meth on
bolites and the corresponding producing micro­ mouse behaviors, the microglial morphology and
biota, then we explored whether dietary TRP neurogenesis in the hippocampus were analyzed.
supplementation could reverse the Meth-induced As depicted in Figure 5(a), the microglia undergo
aberrant behaviors and indole metabolic changes. dramatical morphological changes after the Meth
Figure 4(a) shows the timeline of the experimental challenge, the WD 0d microglia exhibited an
design. The high TRP diet (1.9% TRP) was pre­ increased cell volume, shorter and thicker pro­
treated and maintained until Meth withdrawal for cesses, and reduced branching. These features indi­
14 days, and the low TRP diet (0.09% TRP) was cated the activated microglia in the early stage of
applied here for positive control aiming to decipher Meth exposure. The presentation of the 3D recon­
the roles of the reduced indole metabolites in structions of these microglia (Figure 5(b)), visually
affecting behaviors. It showed that mice in the illustrated these morphological changes. However,
low TRP diet group and the WD 14d group exhib­ by WD 14d, microglial processes were significantly
ited significant anxiety-like changes, manifesting as reduced, characterized by the dystrophic cells,
the reduction of open-arm residence time in the including the retracted and fragmented branches.
EPM, and a decreased residence time in the center Sholl analysis quantitatively validated these
area accompanied by a decreased total trajectory changes, which showed a gradual decrease in
length in the OFT test. Meanwhile, the low TRP microglial branches from WD 0d to WD 14d
diet stimulated pronounced depression like (Figure 5(c–d)). Furthermore, the morphological
changes due to the increased immobility time in parameters of microglia were quantified, including
the FST and TST. Agree with our assumption, the process length, branch, and terminal points. It
high TRP diet substantially ameliorated the Meth showed a pronounced reduction in microglial
induced anxiety and depression-like behaviors branches after Meth exposure, and these changes
 GUT MICROBES 11

Figure 4. Effects of dietary TRP supplementation on Meth-induced behavioral changes and levels of indole derivatives in mice. (a)
Experimental design and timeline. (b) Time spent in the open arms (%) in the EPM test; time spent in the center (s) and total track
length (m) in the OFT; immobility time (%) in the FST and TST. (c) Levels of indole derivatives in fecal samples from different groups.
Data are expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; #p < 0.05, ##p < 0.01.

did not store even after 14 days of withdrawal a marker of the immature neuron and neuronal
(Figure 5(g–i)). The number of IBA1+ cells development, was used for assessing neurogenesis.
increased in the WD 0d group, along with enlarged As depicted in Figure 5(m), a significantly
cell volume, but significantly decreased in number decreased number of DCX+ cells in the DG region
and volume by WD 14d (Figure 5(e–f)). The flow of both WD 0d and WD 14d mice was observed,
cytometry and Western blot analyses further con­ indicating neurogenesis defects. These results sug­
firmed the reduction in microglial numbers in the gested that Meth exposure dynamically modulated
WD 14d group (Figure 5(j–l)). Additionally, DCX, microglial outcome over time and disturbed
12 X. WANG ET AL.

Figure 5. Effects of Meth on microglial morphology and neurogenesis in the hippocampus. (a) Representative immunofluorescence
images of IBA1+ microglia (green) in the hippocampus of control, saline, WD 0d, and WD 14d groups. DAPI (blue) was used for nuclei
staining (scale bar 50 μm). (b) 3D reconstructions of microglial cells from the hippocampal regions of each group. (c) Sholl analysis
quantifies the number of microglial intersections as a function of the distance from the cell soma. (d-i) Quantitative analyses of
microglial morphology: (d) area under the curve (AUC) from Sholl analysis, (e) number of IBA1+ cells, (f) volume of IBA1+ cells, (g) total
 GUT MICROBES 13

neurogenesis in the hippocampus, which might be Additionally, the downstream signaling of the
involved in Meth-induced anxiety and depression- AhR was examined as well. Compared with the
like behaviors. control group, the mRNA levels of the Ahr,
Cyp1a1, and Cyp1b1 were significantly reduced in
the WD 14d group, an effect that was substantially
3.6. Indole derivatives drive AhR activation and rescued by both Ficz and indole derivatives supple­
alleviate Meth-induced anxiety and depression-like mentation, implying the pivotal roles of the AhR
behaviors pathway in alleviating Meth-induced behavioral
abnormalities (Figure 6(h)).
To ascertain whether the salutary effects of TRP
were ascribed to its metabolites, the interaction
between indole derivatives and AhR was examined.
3.7. The protective effects of indole derivatives and
Since some indole derivatives are the natural
AhR against Meth-induced microglial
ligands for AhR, the binding affinity of IPA, IAA,
morphological dystrophy and neuronal damage
and IND to AhR was determined by the MST assay,
which was based on the dramatical reduction of Having confirmed the effects of indole derivatives
these indole derivatives in serum and feces in Meth and the AhR signaling pathway on Meth-induced
treated mice. As shown in Figure 6(a), IPA, IAA, behavioral abnormalities, then we sought to inves­
and IND exhibited different binding affinities to tigate the impact of these molecules on microglial
AhR, with Kd values of 0.674 µM, 2.967 µM, and morphology and neurogenesis, which are closely
0.872 µM, respectively, indicating that these indole associated with depression and anxiety. As shown
derivatives can effectively bind to AhR in Figure 7(a), the microglial morphology under­
(Figure 6(b)). Then we moved to address whether went dramatical changes after Meth withdrawal,
the salutary effects of TRP were ascribed to indole including decreased process length and branch
derivatives and AhR interaction. Figure 6(c) pre­ numbers, and these phenomena were obviously
sents the experimental timeline, which included restored after administration of Ficz and indole
four groups: Control, WD 14d, WD 14d + Ficz, derivatives. The 3D reconstructed images in
and WD 14d + indoles. Consistent with our Figure 7(b) visually depicted these morphological
assumption, exogenous administration of indole changes, emphasizing the remodeling effects of
derivatives (IPA, IAA, and IND) markedly alle­ Ficz and indole derivatives on microglia.
viated the Meth induced depression- and anxiety- Meanwhile, the Sholl analysis indicated
like behaviors, evidenced by reduction of the a significant reduction in microglial branching
immobility time in FST and TST, paralleled with complexity in the WD 14d group, while supple­
increased time spent in the open arms of the EPM mentation with Ficz or indole derivatives obviously
and in the center area of the OFT (Figure 6(d–j)). restored this complexity (Figure 7(c)). Further
For addressing the roles of AhR in Meth induced quantification of microglial morphological para­
depression- and anxiety-like changes, Ficz, one meters showed that in the WD 14d group, the
potent agonist of AhR, was administrated. area under the curve (AUC, Figure 7(d)), IBA1+
Consistent with the results presented in indole cell number (Figure 7(e)), cell volume (Figure 7(f)),
derivatives administration, Ficz treatment process length (Figure 7(g)), terminal points
obviously mitigated the Meth induced depression- (Figure 7(h)), and branch points (Figure 7(i))
and anxiety-like changes, unraveling the critical were all significantly reduced, indicating the
roles of AhR in this process. marked damages in microglial morphology and

process length, (h) number of terminal points, and (i) number of branch points. (j) Flow cytometry gating strategy for identifying
microglial cells (CD11b+ CD45mid) in the hippocampus. (k) Flow cytometry results quantifying the percentage of CD11b+ CD45mid
microglia in the control, saline, and WD 14d groups. (l) Western blot analysis of IBA1 protein expression in hippocampal lysates. (m)
Immunofluorescence staining for doublecortin (DCX, green) in the hippocampus (scale bar 100 μm), with zoomed-in images
highlighting neurogenesis. Data are expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.
14 X. WANG ET AL.

Figure 6. Indole derivatives and AhR agonist Ficz alleviated Meth-induced anxiety and depression-like behaviors. (a) Schematic
diagram of the MST experimental design to measure the binding of IPA, IAA, and IND to AhR. (b) The binding affinity of IPA, IND, and
IAA to AhR. (c) Experimental design and timeline. (d,e) Time spent in the open arms (%) in the EPM test; time spent in the center (s)
and total track length (m) in the OFT; immobility time (%) in the FST and TST (n=5 per group). (f,g) the representative trajectory
diagrams in the OFT and EPM. (h) mRNA expression of AhR pathway-related genes (Ahr, Cyp1a1, and Cyp1b1) in the hippocampus.
Data are expressed as mean ± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

function. Of note, Ficz and indole derivatives 3.8. AhR KO impedes the salutary effects of indole
markedly reversed the aforementioned damages. derivatives against the Meth induced depression
Flow cytometry analysis indicated that Ficz and and anxiety-like behaviors
indole derivatives significantly increased the pro­
To confirm the underlying roles of the AhR in the
portion of CD11b+CD45mid microglia in the WD
anti-depression and anti-anxiety of indole deriva­
14d group, supporting the restorative effects of Ficz
tives in the Meth withdrawal mouse, the AhR
and indole derivatives on microglia (Figure 7(j)).
knockout (AhR KO) mouse was applied.
Similarly, Ficz and indole derivatives notably ame­ Figure 8(a) presents the experimental timeline of
liorated the Meth induced impaired neurogenesis, Meth exposure and indole supplementation in both
evidenced by the increased DCX+ cells wild-type (WT) and AhR KO mice and the beha­
(Figure 7(k)). vioral tests were conducted after the withdrawal
 GUT MICROBES 15

Figure 7. Protective effects of indole derivatives and Ficz on Meth-induced microglial morphological changes and neurogenesis
defects in the hippocampus. (a) Representative immunofluorescence images of IBA1+ microglia (green) in the hippocampus of control,
saline, WD 14d, WD 14d+Ficz, and WD 14d+indoles groups. DAPI (blue) was used for nuclei staining (scale bar 50 μm). (b) 3D
reconstructions of microglial cells in the hippocampus. (c) Sholl analysis quantifies the number of microglial intersections as a function
16 X. WANG ET AL.

Figure 8. Roles of AhR KO in Meth-induced depressive and anxiety-like behaviors. (a) Experimental design and timeline. (b) Gel image
showing PCR results confirming the AhR KO in mice. (c) Immobility time (%) in the FST and TST. (d,e) The representative trajectory
diagrams in the OFT and EPM. (f) Time spent in the open arms (%) in the EPM test; time spent in the center (s) and total track length
(m) in the OFT. (g) mRNA expression of AhR-related gene (Ahr, Cyp1a1, and Cyp1b1) in the hippocampus. Data are expressed as mean
± SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001; #p < 0.05, ###p < 0.001.

period. The confirmation of AhR KO mice was indicated by increased immobility time in the FST
confirmed by PCR, which showed the absence of and TST, and these effects were obviously alle­
the AhR gene (Figure 8(a)). The behavioral analysis viated by indole derivatives, noteworthily, the pro­
revealed that in the WD 14d group, WT mice tective effects of indole derivatives were abolished
exhibited significant depression-like behaviors, as in AhR KO mice (Figure 8(c)). Similarly, the

of the distance from the cell soma. (d-i) quantitative analyses of microglial morphology: (d) AUC from Sholl analysis, (E) number of
IBA1+ cells, (f) volume of IBA1+ cells, (g) total process length, (h) number of terminal points, and (i) number of branch points. (j) Flow
cytometry quantifies the percentage of CD11b+ CD45mid microglia in each group. (k) Immunofluorescence staining for DCX (green) in
the hippocampus of each group, with zoomed-in images highlighting neurogenesis (scale bar 100 μm). Data are expressed as mean ±
SEM, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.
 GUT MICROBES 17

beneficial effects of indole derivatives on anxiety neurogenesis in Meth-treated mice, unraveling

were notably diminished in AhR KO mice as well the neuroprotective effects of indole derivatives
(Figure 8(d–f)). mRNA analysis further supported via AhR.
the activation of AhR signaling by indole deriva­
tives, since indole derivatives significantly restored
4. Discussion
the decreased levels of Cyp1a1, and Cyp1b1 genes
induced by Meth, and these effects were ablated in Epidemic studies have found that Meth withdrawal
AhR KO mice. These results, taken together, can trigger depression,25–27 and similar results were
underscore the critical roles of AhR in indole deri­ validated in Meth-withdrawal mice.28–30 The dis­
vative-mediated behaviors and molecular signal­ tribution in distinct tissues of Meth post-exposure
ings (Figure 8(g)). in mice was examined and its levels in the blood
and brain approached zero at 4 and 8 hours,
respectively.2 Thus, the inflammatory effects and
depressive behaviors induced by Meth withdrawal
3.9. AhR KO hampers the protective effects of
may represent a continuation of the adverse effects
indole derivatives against Meth induced microglial
on distinct tissues and organs after initial exposure.
dystrophy and neurogenesis defects
The toxicity of Meth to date has predominantly
Since the critical roles of the indole derivative in concentrated on the central nervous, circulatory,
the modulation of the microglial morphology and and respiratory systems, while less attention has
neurogenesis aforementioned, then we investigated been paid to the digestive system. As the “second
whether the effects were dependent on the AhR. As brain” and the most predominant immunol system,
shown in Figure 9, the morphology of microglia the intestinal system may be highly susceptible to
(Figure 9(a)) and 3D reconstructed images Meth attack. Therefore, the psychiatric disorders
(Figure 9(b)) in the hippocampal region were sig­ induced by Meth may not only limit to its direct
nificantly changed in WT and AhR KO mice. impact on the brain, and the microbiota-gut axis
A significant reduction in branching complexity may presumably be the major target. Studies have
was observed in the WD 14d group, which was shown that Meth exposure alters the α and β diver­
obviously restored following indole derivative sup­ sity of gut microbiota in mice, increasing the relative
plementation. These protective effects, however, abundance of pathogenic microbiota and recipro­
were markedly attenuated in AhR KO mice cally decreasing the beneficial ones.31 Additionally,
(Figure 9(c)), indicating the crucial role of AhR microbiota abundance changes in Meth-exposed
signaling in microglial morphological remodeling. mice were significantly correlated with depression-
Further quantification of morphological para­ like behaviors and neurotoxicity.32 In recent years,
meters revealed pronounced decreases of AUC, the roles of gut microbiota and the corresponding
IBA1+ cell number, cell volume, process length, metabolites in regulating host behaviors and brain
terminal branch number, and branch points in function have garnered considerable attention, par­
the WD 14d group, whereas indole derivative treat­ ticularly the TRP microbial metabolites, which are
ment significantly restored these deteriorated considered crucial to the regulation of the gut-brain
impacts. Specifically, the beneficial effects were axis. Thus, the current study aims to address the
abolished in AhR KO mice (Figure 9(d–i)), sup­ alterations of gut microbiota and the corresponding
porting the importance of AhR signaling in the metabolites, especially the potential roles of indole
regulation of microglial morphology. Similarly, derivatives in Meth-induced anxiety and depres­
the protective effects of the indole derivatives sion-like behaviors in mice. In line with our assump­
against the Meth induced neurogenesis defects tion, Meth withdrawal significantly altered the
were restrained in AhR KO mice, since the DCX+ microbiota composition in mice, and transplanta­
cells in the Meth+indole derivative were markedly tion of the gut microbiota from Meth-exposed mice
decreased in AhR KO mice (Figure 9(k)). These to recipient mice induced anxiety- and depression-
findings underscore the pivotal roles of AhR in the like behaviors, providing evidence for the involve­
remodeling of microglial morphology and ment of microbiota in the pathology. Intriguingly,
18 X. WANG ET AL.

Figure 9. AhR KO alleviated the protective effects of indole derivatives against Meth-induced microglial dystrophy and neurogenesis
defects. (a) Representative immunofluorescence images of IBA1+ microglia (green) in the hippocampus. DAPI (blue) was used for
nuclei staining (scale bar 50 μm). (b) 3D reconstructions of microglial cells from the hippocampal regions of each group. (c) Sholl
analysis quantifies the number of microglial intersections as a function of the distance from the cell soma. (d–i) Quantitative analyses
of microglial morphology: (d) area under the curve (AUC) from Sholl analysis, (e) number of IBA1+ cells, (f) volume of IBA1+ cells, (g)
total process length, (h) number of terminal points, and (i) number of branch points. (j) Immunofluorescence staining for DCX (green)
in the hippocampus, with zoomed-in images highlighting neurogenesis (scale bar 100 μm). Data are expressed as mean ± SEM, *p <
0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Meth withdrawal resulted in a marked reduction in and depression-like behaviors in Meth withdrawal
the levels of the indole metabolite-producing micro­ mice. Importantly, we corroborated these findings in
biota, thereby inhibiting the TRP microbial meta­ a population of Meth abusers, whose levels of indole
bolic pathway and decreasing the indole derivative derivatives were obviously reduced, suggesting the
levels in serum and feces. Reciprocally, a high-TRP involvement of the gut microbiota and its indole
diet, which facilitates the restoration of indole deri­ metabolites in the Meth withdrawal period in
vative levels, significantly ameliorated the anxiety- humans.
 GUT MICROBES 19

TRP, an essential amino acid and the only amino study that decreased survival microglia in the hip­
acid containing the indole structure, is primarily pocampus induced by Meth.42
obtained from diet. Numerous in vivo and in vitro After Meth exposure, the reduction of indole
studies have demonstrated that Akkermansia influ­ metabolites profoundly impacts the gut-
ences gut health and modulates immune responses microbiota-brain axis, a complex bidirectional
by metabolizing TRP into indole derivatives, communication system that relies on the AhR to
including indole, IAA, IArA, and ILA.33–35 transmit TRP metabolic signaling to the brain via
Similarly, Bacteroides species are particularly active immune response, peripheral metabolic circula­
in the distal colon, where they ferment TRP to tion, and the vagus nerve. Noteworthily, the micro­
produce indole derivatives, for example, bial metabolites derived from the TRP metabolic
Bacteroides thetaiotaomicron, is known for indole pathway can cross the blood-brain barrier(BBB),
and its derivatives production, such as IAA and facilitating tissue-specific transport and gut-brain
ILA.36 Besides, Anaerostipes is well-known for pro­ interactions.43 Similar studies have analyzed the
ducing both short-chain fatty acids (SCFA) and decreased levels of IAA in the blood and cerebrosp­
facilitating TRP metabolism, generating ILA.9,37,38 inal fluid in Alzheimer’s disease (AD) patients that
Noteworthily, in the current study, Meth withdra­ the declined IAA decreases the combination of
wal resulted in a significant reduction of indole AhR, which promotes the NLRP3 inflammasome
metabolite-producing microbiota, particularly the accumulation via NF-κB activation, thereby stimu­
decreased abundance of Akkermansia, Bacteroides, lating neuroinflammation.44 On the contrary, sup­
and Anaerostipes species, which presumably facili­ plementation with indole, IS, IPA, and IAld
tated the reduced production of indole derivatives, significantly reduced brain inflammation in germ-
contributing to the development of anxiety and free mice.13 Similarly, in SPF mice, indole and its
depression-like behaviors. derivatives promote neurogenesis via the AhR,
Recently, a single-nucleus RNA-sequencing and while other metabolic product of TRP, such as
spatial transcriptomics in female cynomolgus kynurenine, exhibits no neurogenic effects.17
macaques depression model revealed that the Moreover, in a mouse model of irritable bowel
gene expression changes associated with depres­ syndrome (IBS) induced by Citrobacter rodentium
sive-like behaviors mostly in microglia;39 In infection, enhanced microbial TRP metabolism by
a chronic unpredictable stress (CUS) mouse Lactobacillus reuteri restored colonic sensitivity
model, the microglial cells undergo dramatical and reduced anxiety-like behavior by upregulating
dynamic changes from initial proliferation and IL-22 levels in the colon.45 These results highly
activation to dystrophy and apoptosis. These supported the Meth-induced depression and anxi­
changes were particularly pronounced in the ety-like behaviors involving indole metabolite-AhR
microglia and were closely linked to depression- signaling defects, which may provide promising
like behaviors.40 Similarly, during the Meth expo­ therapeutic targets. Compared to other studies,
sure period, microglial cells displayed a persistent our research for the first time, revealed the micro­
activated state, characterized by enlarged cell biota in Meth treated mice but not limited to the
bodies and shortened, thickened dendrites, then neural injury can ignite depression and anxiety-like
the microglia transitioned into a dystrophic mor­ changes. More specifically, microbiota metabolites,
phology, marked by reduced cell size, decreased such as IPA and IAA, exerted salutary effects on the
branches, and even fragmentation, deciphering Meth induced neurobehavioral abnormalities,
a shifting from an initial “surveillance state” to which provide a novel interventive strategy.
a “dystrophic” state. Consistent with our findings, To ascertain the critical roles of the AhR in Meth
a study on Meth-exposed rats exhibited significant induced neurobehavioral disorders, supplementa­
morphological changes in microglia within the tion with indole derivatives and the AhR agonist
nucleus accumbens core 31 days after Meth with­ Ficz in Meth-withdrawal mice significantly
drawal, characterized by a reduction in microglial restored microglial morphology and improved
cell numbers and decreased branch lengths,41 simi­ neurogenesis. To further explore the specific role
larly, the microglia injury is confirmed by another of AhR in the neuroprotective effects of indole
20 X. WANG ET AL.

derivatives, the AhR KO mice were exploited. and depression symptoms associated with Meth
Intriguingly, the protective effects of indole deriva­ withdrawal and offer potential targets and stra­
tives against Meth induced microglial dystrophy tegies for Meth abuse. Restoring the balance of
and neurogenesis defects were ablated in AhR KO the gut microbiota, particularly by promoting
mice, underpinning the specific salutary roles of the synthesis of TRP metabolites such as indole
indole derivatives in the regulation of microglial derivatives, may serve as an effective interven­
activity and neurogenesis via AhR, placing tion to alleviate Meth withdrawal symptoms. For
a unique role of AhR in Meth induced neurobeha­ example, the use of probiotics, dietary adjust­
vioral abnormalities. ments (e.g., high-TRP diets), or microbiota-
In conclusion, this study, for the first time, targeting strategies to restore gut function may
deciphered the peripheral mechanisms by unrec­ help reduce anxiety and depression symptoms in
ognized Meth-induced anxiety and depression- Meth-addicted individuals. Therefore, future
like behaviors via the disturbance of microbiota drug interventions may be developed based on
and the corresponding metabolites, particularly the regulation of the TRP microbiota metabo­
the indole derivatives. In this process, Meth lism and AhR receptors, like AhR agonists,
withdrawal induced decrease of indole metabo­ which may provide innovative strategies for
lite-producing microbiota, contributing to the Meth associated neurotoxicity.
striking reduction of indole metabolites, which
in turn, decreased AhR expression and impacted
5. Limitation
microglial morphology and neurogenesis, and
finalized the anxiety and depression-like changes The present study still has some limitations.
in mice (Figure 10). Therefore, the present study Firstly, due to certain constraints, we were unable
provide novel biological insights into the anxiety to conduct systematic anxiety and depression

Figure 10. A schematic depicting the peripheral mechanism underlying Meth-induced anxiety and depression-like behaviors,
particularly highlighting the role of microbiota and their indole metabolites. During Meth withdrawal, a reduction of indole
metabolite-producing microbiota decreases the levels of indole metabolites, which impedes AhR expression and causes microglial
dystrophy and neurogenesis defects and finalizes anxiety and depression-like behaviors in mice. Noteworthily, supplementing with
indole metabolites, TRP diet, or targeting AhR could offer novel strategies for Meth-induced anxiety and depression.
 GUT MICROBES 21

assessments in Meth-abusing individuals, nor 82072159] and the Major Projects of Natural Science
were we able to obtain fecal samples from this Foundation in Jiangsu Higher Education Institutions of
population for further analysis of indole metabo­ China [No. 22KJA330001], the Young and Middle-aged
Academic Leaders of “Blue Project” in Jiangsu Province
lism. Secondly, although our study demonstrates [NO.2022-29].
that indole derivatives confer neuroprotective
effects through AhR activation, the precise
mechanisms of interaction among these mole­ Author contributions
cules remain unclear. AhR signaling is highly
complex and context-dependent, with its effects Xi Wang and Jun Wang conceived and designed the over­
varying across tissues and cell types. Further all study; Xi Wang, Miaoyang Hu, Weilan Wu, and Xinyu
Lou performed all the experiments; Weilan Wu, Jie Cheng,
research is needed to elucidate the cell-specific
Jianping Xiong, and Xufeng Chen performed human sub­
and tissue-specific roles of AhR activation in the ject recruitment and clinical data analysis; S Thameem
gut-brain axis. Thirdly, our experiments primarily Dheen and Tengfei Ma contributed to the methodology;
relied on a mouse model to investigate Meth Xi Wang wrote the manuscript; Rong Gao, Xufeng Chen,
withdrawal and its effects on the gut microbiota and Jun Wang revised the manuscript. All authors
and behaviors, however, they cannot totally reviewed and approved the final version of the
manuscript.
represent the toxic effects of Meth abuse and
the characteristics of mental disorders in humans.
Translational studies in non-human primates or
Data availability statement
Meth-abusing human populations will be essen­
tial to confirm the clinical applicability of our All data generated or analyzed during this study are included
findings. Fourthly, although a reduction in in this published article and its supplementary information
Akkermansia and its potential link to mucus files. The datasets presented in this study can be found in
online repositories. The names of the repository/repositories
layer disruption were confirmed, we did not
and accession number(s) can be found below: https://www.
directly demonstrate the functional contribution ncbi.nlm.nih.gov/, PRJNA1177660.
of this bacterial species to Meth-induced gut dys­
biosis and behavioral abnormalities. Future stu­
dies using approaches such as A. muciniphila References
supplementation or depletion experiments are 1. World Drug Report 2023. UNODC; 2023.
warranted to establish its causal role in this pro­ 2. Cruickshank CC, Dyer KR. A review of the clinical
cess. Lastly, our study focused on the role of pharmacology of methamphetamine. Addiction.
indole derivatives and AhR activation in Meth 2009;104(7):1085–1099. doi:10.1111/j.1360-0443.2009.
withdrawal-related anxiety and depression. 02564.x.
3. Ma J, Sun X-J, Wang R-J, Wang T-Y, Su M-F, Liu M-X,
However, the development of these behaviors is
Li S-X, Han Y, Meng S-Q, Wu P, et al. Profile of psy­
multifactorial, involving other neurotransmitter chiatric symptoms in methamphetamine users in China:
systems, immune pathways, and environmental greater risk of psychiatric symptoms with a longer dura­
factors. Integrating these additional variables tion of use. Psychiat Res. 2018;262:184–192. doi:10.1016/
into future research will provide a more holistic j.psychres.2018.02.017.
understanding of Meth-induced neurobehavioral 4. Zhang J, Xie Y, Su H, Tao J, Sun Y, Li L, Liang H, He R,
Han B, Lu Y, et al. Prevalence and correlates of depres­
changes.
sive symptoms during early methamphetamine with­
drawal in Han Chinese population. Drug Alcohol
Depen. 2014;142:191–196. doi:10.1016/j.drugalcdep.
Disclosure statement
2014.06.021.
No potential conflict of interest was reported by the author(s). 5. Cryan JF, Dinan TG. Mind-altering microorganisms:
the impact of the gut microbiota on brain and
behaviour. Nat Rev Neurosci. 2012;13(10):701–712.
Funding doi:10.1038/nrn3346.
6. Heijtz RD, Wang S, Anuar F, Qian Y, Björkholm B,
This study was supported by the National Natural Science Samuelsson A, Hibberd ML, Forssberg H, Pettersson S.
Foundation of China [82473664, 82472240, 82073584, Normal gut microbiota modulates brain development
22 X. WANG ET AL.

and behavior. Proc Natl Acad Sci USA. 2011;108 17. Wei GZ, Martin KA, Xing PY, Agrawal R, Whiley L,
(7):3047–3052. doi:10.1073/pnas.1010529108. Wood TK, Hejndorf S, Ng YZ, Low JZY, Rossant J, et al.
7. Hsiao EY, McBride SW, Hsien S, Sharon G, Hyde ER, Tryptophan-metabolizing gut microbes regulate adult
McCue T, Codelli JA, Chow J, Reisman SE, neurogenesis via the aryl hydrocarbon receptor. Proc
Petrosino JF, et al. The microbiota modulates gut phy­ Nat Acad Sci. 2021;118:e2021091118.
siology and behavioral abnormalities associated with 18. Serger E, Luengo-Gutierrez L, Chadwick JS, Kong G,
autism. Cell. 2013;155(7):1451–1463. doi:10.1016/j.cell. Zhou L, Crawford G, Danzi MC, Myridakis A,
2013.11.024. Brandis A, Bello AT, et al. The gut metabolite
8. Li D, Yu S, Long Y, Shi A, Deng J, Ma Y, Wen J, Li X, indole-3 propionate promotes nerve regeneration and
Liu S, Zhang Y, et al. Tryptophan metabolism: repair. Nature. 2022;607(7919):585–592. doi:10.1038/
mechanism-oriented therapy for neurological and psy­ s41586-022-04884-x.
chiatric disorders. Front Immunol. 2022;13:985378. 19. Yin J, Zhang Y, Liu X, Li W, Hu Y, Zhang B, Wang S.
doi:10.3389/fimmu.2022.985378. Gut microbiota-derived indole derivatives alleviate
9. Roager HM, Licht TR. Microbial tryptophan catabolites neurodegeneration in aging through activating
in health and disease. Nat Commun. 2018;9(1):3294. GPR30/AMPK/SIRT1 pathway. Mol Nutr Food Res.
doi:10.1038/s41467-018-05470-4. 2023;67(9):2200739. doi:10.1002/mnfr.202200739.
10. Pappolla MA, Perry G, Fang X, Zagorski M, 20. Faul F, Erdfelder E, Buchner A, Lang A-G. Statistical
Sambamurti K, Poeggeler B. Indoles as essential med­ power analyses using G*power 3.1: tests for correlation
iators in the gut-brain axis. Their role in alzheimer’s and regression analyses. Behav Res Methods. 2009;41
disease. Neurobiol Dis. 2021;156:105403. doi:10.1016/j. (4):1149–1160. doi:10.3758/BRM.41.4.1149.
nbd.2021.105403. 21. Yao J, Wang J, Wu L, Lu H, Wang Z, Yu P, Xiao H,
11. Brydges CR, Fiehn O, Mayberg HS, Schreiber H, Gao R, Yu J. Perinatal exposure to bisphenol a causes
Dehkordi SM, Bhattacharyya S, Cha J, Choi KS, a disturbance of neurotransmitter metabolic pathways in
Craighead WE, Krishnan RR, et al. Indoxyl sulfate, female mouse offspring: a focus on the tryptophan and
a gut microbiome-derived uremic toxin, is associated dopamine pathways. Chemosphere. 2020;254:126715.
with psychic anxiety and its functional magnetic reso­ doi:10.1016/j.chemosphere.2020.126715.
nance imaging-based neurologic signature. Sci Rep-Uk. 22. Araki A, Kanai T, Ishikura T, Makita S, Uraushihara K,
2021;11(1):21011. doi:10.1038/s41598-021-99845-1. Iiyama R, Totsuka T, Takeda K, Akira S, Watanabe M.
12. Tian P, Chen Y, Zhu H, Wang L, Qian X, Zou R, Zhao J, MyD88-deficient mice develop severe intestinal inflam­
Zhang H, Qian L, Wang Q, et al. Bifidobacterium breve mation in dextran sodium sulfate colitis. J Gastroenterol.
2005;40(1):16–23. doi:10.1007/s00535-004-1492-9.
CCFM1025 attenuates major depression disorder via
23. Wang X, Hu M, Li M, Huan F, Gao R, Wang J. Effects
regulating gut microbiome and tryptophan metabo­
of exposure to 3,6-DBCZ on neurotoxicity and AhR
lism: a randomized clinical trial. Brain Behav Immun. pathway during early life stages of zebrafish (Danio
2022;100:233–241. doi:10.1016/j.bbi.2021.11.023. rerio). Ecotox Environ Safe. 2024;270:115892. doi:10.
13. Rothhammer V, Mascanfroni ID, Bunse L, 1016/j.ecoenv.2023.115892.
Takenaka MC, Kenison JE, Mayo L, Chao C-C, 24. Foster JA, Neufeld K-A. Gut–brain axis: how the micro­
Patel B, Yan R, Blain M, et al. Type I interferons and biome influences anxiety and depression. Trends
microbial metabolites of tryptophan modulate astrocyte Neurosci. 2013;36(5):305–312. doi:10.1016/j.tins.2013.
activity and central nervous system inflammation via 01.005.
the aryl hydrocarbon receptor. Nat Med. 2016;22 25. Casaletto KB, Obermeit L, Morgan EE, Weber E,
(6):586–597. doi:10.1038/nm.4106. Franklin DR, Grant I, Woods SP, Translational
14. Gutiérrez-Vázquez C, Quintana FJ. Regulation of the Methamphetamine AIDS Research Center (TMARC)
immune response by the aryl hydrocarbon receptor. Group. Depression and executive dysfunction contri­
Immunity. 2018;48(1):19–33. doi:10.1016/j.immuni. bute to a metamemory deficit among individuals with
2017.12.012. methamphetamine use disorders. Addict Behav.
15. Qiu J, Heller JJ, Guo X, Chen ZE, Fish K, Fu Y-X, 2015;40:45–50. doi:10.1016/j.addbeh.2014.08.007.
Zhou L. The aryl hydrocarbon receptor regulates gut 26. Luan X, Chen H, Qiu H, Shen H, Zhao K, Ren W, Gu Y,
immunity through modulation of innate lymphoid Su H, Zhang J, Lv D, et al. Association between serum
cells. Immunity. 2012;36(1):92–104. doi:10.1016/j. malondialdehyde levels and depression during early
immuni.2011.11.011. methamphetamine withdrawal. Neurosci Lett.
16. Sondermann NC, Faßbender S, Hartung F, Hätälä AM, 2018;687:22–25. doi:10.1016/j.neulet.2018.09.021.
Rolfes KM, Vogel CFA, Haarmann-Stemmann T. 27. Ren W, Luan X, Zhang J, Gutteea P, Cai Y, Zhao J,
Functions of the aryl hydrocarbon receptor (AHR) Gu Y, Wu C, Su H, Tao J, et al. Brain-derived neuro­
beyond the canonical AHR/ARNT signaling pathway. trophic factor levels and depression during metham­
Biochem Pharmacol. 2023;208:115371. doi:10.1016/j. phetamine withdrawal. J Affect Disord.
bcp.2022.115371. 2017;221:165–171. doi:10.1016/j.jad.2017.06.017.
 GUT MICROBES 23

28. Silva CD, Neves AF, Dias AI, Freitas HJ, Mendes SM, axis and potential therapeutic targets: a focus on
Pita I, Viana SD, de Oliveira PA, Cunha RA, Fontes human neurological and neuropsychiatric diseases.
Ribeiro CA, et al. A single neurotoxic dose of metham­ Neuropharmacology. 2023;239:109690. doi:10.1016/j.
phetamine induces a long-lasting depressive-like beha­ neuropharm.2023.109690.
viour in mice. Neurotox Res. 2014;25(3):295–304. 38. Su X, Gao Y, Yang R. Gut microbiota-derived trypto­
doi:10.1007/s12640-013-9423-2. phan metabolites maintain gut and systemic
29. Iijima M, Koike H, Chaki S. Effect of an mGlu2/3 homeostasis. Cells-Basel. 2022;11(15):2296. doi:10.
receptor antagonist on depressive behavior induced by 3390/cells11152296.
withdrawal from chronic treatment with 39. Wu J, Li Y, Huang Y, Liu L, Zhang H, Nagy C, Tan X,
methamphetamine. Behav Brain Res. 2013;246:24–28. Cheng K, Liu Y, Pu J, et al. Integrating spatial and
doi:10.1016/j.bbr.2013.02.039. single-nucleus transcriptomic data elucidates
30. Wang X, Tong B, Hui R, Hou C, Zhang Z, Zhang L, microglial-specific responses in female cynomolgus
Xie B, Ni Z, Cong B, Ma C, et al. The role of hyperther­ macaques with depressive-like behaviors. Nat
mia in methamphetamine-induced depression-like Neurosci. 2023;26(8):1352–1364. doi:10.1038/s41593-
behaviors: protective effects of coral calcium hydride. 023-01379-4.
Front Mol Neurosci. 2021;14:808807. doi:10.3389/ 40. Kreisel T, Frank MG, Licht T, Reshef R, Ben-
fnmol.2021.808807. Menachem-Zidon O, Baratta MV, Maier SF,
31. Chen L-J, Zhi X, Zhang K-K, Wang L-B, Li J-H, Liu J-L, Yirmiya R. Dynamic microglial alterations underlie
Xu L-L, Yoshida JS, Xie X-L, Wang Q. Escalating stress-induced depressive-like behavior and suppressed
dose-multiple binge methamphetamine treatment eli­ neurogenesis. Mol Psychiatr. 2014;19(6):699–709.
cits neurotoxicity, altering gut microbiota and fecal doi:10.1038/mp.2013.155.
metabolites in mice. Food Chem Toxicol. 41. Yu Z, Chen W, Zhang L, Chen Y, Chen W, Meng S,
2021;148:111946. doi:10.1016/j.fct.2020.111946. Lu L, Han Y, Shi J. Gut-derived bacterial LPS attenuates
32. Yang C, Fu X, Hao W, Xiang X, Liu T, Yang B-Z, incubation of methamphetamine craving via modulat­
Zhang X. Gut dysbiosis associated with the rats’ ing microglia. Brain Behav Immun. 2023;111:101–115.
responses in methamphetamine-induced conditioned doi:10.1016/j.bbi.2023.03.027.
place preference. Addict Biol. 2021;26:e12975. doi:10. 42. Zhang Y, Shen K, Bai Y, Lv X, Huang R, Zhang W,
1111/adb.12975. Chao J, Nguyen LK, Hua J, Gan G, et al. Mir143-BBC3
33. Gu Z, Pei W, Shen Y, Wang L, Zhu J, Zhang Y, Fan S,
cascade reduces microglial survival via interplay
Wu Q, Li L, Zhang Z. Akkermansia muciniphila and its
between apoptosis and autophagy: implications for
outer protein Amuc_1100 regulates tryptophan metabo­
methamphetamine-mediated neurotoxicity. Autophagy.
lism in colitis. Food Funct. 2021;12(20):10184–10195.
2016;12(9):1538–1559. doi:10.1080/15548627.2016.
doi:10.1039/D1FO02172A.
1191723.
34. Yin J, Song Y, Hu Y, Wang Y, Zhang B, Wang J, Ji X,
43. Lai Y, Liu C-W, Yang Y, Hsiao Y-C, Ru H, Lu K. High-
Wang S. Dose-dependent beneficial effects of tryptophan
coverage metabolomics uncovers microbiota-driven
and its derived metabolites on akkermansia in vitro:
a preliminary prospective study. Microorganisms. biochemical landscape of interorgan transport and
2021;9(7):1511. doi:10.3390/microorganisms9071511. gut-brain communication in mice. Nat Commun.
35. Lin X, Tawch S, Singh A, Morgun A, Shulzhenko N, 2021;12(1):6000. doi:10.1038/s41467-021-26209-8.
Kumar P. Akkermansia muciniphila induces Th17 cells 44. Guo X, Li C, Zhang J, Sun M, Xu J, Xu C, Kuang H,
by mediating tryptophan metabolism and exacerbates Xu L. Chiral nanoparticle-remodeled gut microbiota
experimental autoimmune encephalomyelitis (EAE). alleviates neurodegeneration via the gut–brain axis.
J Immunol. 2021;206(1_Supplement):105.09. doi:10. Nat Aging. 2023;3(11):1415–1429. doi:10.1038/s43587-
4049/jimmunol.206.Supp.105.09. 023-00516-9.
36. Hyland NP, Cavanaugh CR, Hornby PJ. Emerging effects 45. Meynier M, Baudu E, Rolhion N, Defaye M, Straube M,
of tryptophan pathway metabolites and intestinal micro­ Daugey V, Modoux M, Wawrzyniak I, Delbac F,
biota on metabolism and intestinal function. Amino Acids. Villéger R, et al. AhR/IL-22 pathway as new target for
2022;54(1):57–70. doi:10.1007/s00726-022-03123-x. the treatment of post-infectious irritable bowel syn­
37. Zhou Y, Chen Y, He H, Peng M, Zeng M, Sun H. drome symptoms. Gut Microbes. 2022;14(1):2022997.
The role of the indoles in microbiota-gut-brain doi:10.1080/19490976.2021.2022997.