Skip to main content
Download PDF

Gut microbiota, circulating cytokines and dementia: a Mendelian randomization study

Journal of Neuroinflammation volume 21, Article number: 2 (2024) Cite this article

Abstract

Background

Some studies have shown that gut microbiota may be associated with dementia. However, the causal effects between gut microbiota and different types of dementia and whether cytokines act as a mediator remain unclear.

Methods

Gut microbiota, cytokines, and five dementia types, including Alzheimer’s disease (AD), frontotemporal dementia (FTD), dementia with Lewy body (DLB), vascular dementia (VD), and Parkinson’s disease dementia (PDD) were identified from large-scale genome-wide association studies (GWAS) summary data. We used Mendelian randomization (MR) to investigate the causal relationships between gut microbiota, cytokines, and five types of dementia. Inverse variance weighting (IVW) was used as the main statistical method. In addition, we explored whether cytokines act as a mediating factor in the pathway from gut microbiota to dementia.

Results

There were 20 positive and 16 negative causal effects between genetic liability in the gut microbiota and dementia. Also, there were five positive and four negative causal effects between cytokines and dementias. Cytokines did not act as mediating factors.

Conclusions

Gut microbiota and cytokines were causally associated with five types of dementia, and cytokines seemed not to be the mediating factors in the pathway from gut microbiota to dementia.

Introduction

Dementia is a syndrome characterized by cognitive and memory impairment. Its subtypes are Alzheimer’s disease (AD), frontotemporal dementia (FTD), dementia with Lewy body (DLB), and vascular dementia (VD) [1]. In addition, the full spectrum of cognitive impairment occurs in individuals with Parkinson’s disease (PD), from subjective cognitive decline and mild cognitive impairment to Parkinson’s disease dementia (PDD) [2].

The ‘gut microbiota’ can be defined as all the species within the ecosystem and are considered the largest reservoir of microbes in the human body, containing about 1014 microbes [3]. The human gut contains approximately 1.5 kg of cells, comprising Archaea and Eukaryotes, but are predominantly bacterial [4]. These gastrointestinal microbiotas play important roles in physiological homeostasis and metabolism, including immune system development, vitamin production, and nutrient absorption [5]. Studies have shown that gut microbiotas were associated with dementia [3, 6]. In addition, gut microbiota plays an important role in regulating cytokines [7, 8].

Inflammation is a risk factor for mild cognitive impairment and AD [9, 10]. Epidemiological studies have found that elevated levels of systemic inflammation were associated with cognitive decline [11]. It seemed that both gut microbiota and cytokines can affect the development of dementia. We assumed that cytokines may be mediating factors in the pathway from gut microbiota to dementia.

Although randomized controlled trials could help establish causal relationships between gut microbiota or cytokines and dementia, they are difficult to perform in humans due to the limitation of objective conditions, such as the screening of strains and lowering the levels of cytokines. As a result, most of the current research conclusions are based on observations of the composition and changes in the gut microbiota in dementia patients' feces [6, 12], or indirect interventions such as probiotic supplementation [13], and aspirin or other non-steroidal anti-inflammatory drugs [14].

Genome-wide association studies (GWAS) have tested millions of genetic variants in many individual genomes to identify genotype–phenotype associations and have revolutionized the field of complex disease genetics in the past decade [15].

Mendelian randomization (MR) is a genetic epidemiological method. In MR, genetic variants are used as instrumental variables (IVs) for assessing the causal effect between exposure and outcome [16,17,18,19]. Genetic variants have been determined at the time of conception and therefore MR is less susceptible to environmental confounding factors and reverse causality compared with observational studies [16, 17]. A one sample MR analysis requires the exposure and outcome from the same individual whereas a two-sample MR analysis requires them from different GWAS summary databases. The two-sample MR method is greater statistical power to obtain the causal effects between "the exposure factors" and "the outcome" by taking advantage of published summary estimates from large scale different GWAS [20, 21]. Large-scale summary statistics were available to analyze the relationships between gut microbiota or cytokines and dementia, which improved the statistical power of two-sample MR analysis.

In this study, we conducted a comprehensive MR analysis to explore the causal effects between the gut microbiome, cytokines, and multiple dementia types (including AD, FTD, DLB, VD, and PDD). Then we explored whether cytokines as mediators in the pathway from gut microbiota to dementia. In addition, through reverse causality analysis, we can also examine whether genetic predisposition to dementia risk affects gut microbiota and cytokines.

Methods

Study design

This study has three main components as outlined in Fig. 1: analysis of causal effects of 211 gut microbiota on five dementias (step 1A); analysis of causal effects of 41 cytokines on five dementias (step 2A); and mediation analysis of cytokines in the pathway from gut microbiota to dementias (step 3). We defined single-nucleotide polymorphisms (SNPs) as IVs. Mendelian randomization is based on three core assumptions: (1) the IVs are closely associated with the exposure factors; (2) IVs are not associated with confounding factors; (3) IVs do not affect the outcome directly, and it can only affect outcome via the exposure [22].

Fig. 1
figure 1

Study overview. Step 1A represents the causal effects of gut microbiota on dementia. Step 1B represents the bi‑directional causal effects between gut microbiota and dementia. Step 2A represents the causal effects of cytokines on dementia. Step 2B represents the bi‑directional causal effects between cytokines and dementia. Step 3 represents the mediating analysis of cytokines in the pathway from the gut microbiota to dementia: path c was the total effect of gut microbiota on dementia; path b was the causal effect of cytokines on dementia; path a was the causal effect of gut microbiota on cytokines

Full size image

Data source

The genetic data for the gut microbiome came from the latest GWAS summary data, in which the MiBioGen consortium curated and analyzed genome-wide genotypes and 16S fecal microbiome data from 18,340 individuals (24 cohorts) [23]. The GWAS summary data included a total of 211 gut microbiota taxa (131 genera, 35 families, 20 orders, 16 classes, and 9 phyla). The genetic data for cytokines came from the previously GWAS (8,337 individuals), including 41 cytokines [24].

The GWAS summary data of AD, FTD, VD, and PDD were derived from the eighth version of the Finngen consortium (https://r8.risteys.finngen.fi/). It was a prospective cohort study that patients were screened using International Classification of Diseases (ICD) diagnosis codes for four subtypes of dementia. We downloaded genetic data of four subtypes of dementia from the Finngen database, respectively. The GWAS summary data of DLB were derived from the study by chia et al. [25], and included in the IEU Open GWAS database (https://gwas.mrcieu.ac.uk/). Participants were recruited across 44 institutions/consortia and were diagnosed according to established consensus criteria. Detailed information is provided in Additional file 2: Table S1.

The present study is a secondary analysis of publicly available GWAS summary statistics. Ethical approval was granted for each of the original GWAS studies. In addition, no individual-level data was used in this study. Therefore, no new ethical review board approval was required.

Instrumental variables selection

First, we selected the SNPs with significant associations for gut microbiota (P < 1 × 10–5). To maximize the number of available instruments for each cytokine, we selected the SNPs with a P-value of 5 × 10–6 as the threshold. Then we excluded the SNPs with linkage disequilibrium (LD) in the analysis. The LD of chosen SNPs strongly related to gut microbiota should meet the condition that r2 < 0.001 and distance > 10,000 kb [26]. An important step in MR analysis is to ensure that the effects of SNPs on exposure correspond to the same allele as the effects on outcome. After matching the outcome, we removed palindromic SNPs. (A palindromic SNP is an SNP with the A/T or G/C allele.)

We extracted the relevant information: chromosome, effect allele (EA), other allele (OA), effect allele frequency (EAF), effect sizes (β), standard error (SE), and P-value. Last, we calculated the explained variance (R2) and F-statistic parameters to determine whether the identified IVs were strongly associated with exposure. Generally, SNPs with F-statistic parameters < 10 are considered weak instruments [27]. In this study, R2 = 2 × EAF × (1-EAF) × β2 / (2 × EAF × (1-EAF) × β2 + 2 × EAF × (1-EAF) × N × SE2), where N is the sample size of the GWAS for FI, and F = R2 × (N-2)/(1-R2) [28].

MR analysis

Primary analysis

To estimate the causal effects of gut microbiota and cytokines on dementia, we performed two-sample MR analysis, respectively (step 1A and step 2A in Fig. 1). The inverse variance weighted (IVW) approach was as the essential analysis method and the Wald ratios test for features containing only one IV [29]. MR results were expressed as odds ratios (ORs) with the corresponding 95% confidence intervals (CI). The results were statistically significant when P-value of IVW were less than 0.05 and the direction of IVW and MR-Egger were consistent. A two-sided P-value that passed the Bonferroni correction P-value was defined as statistically significant, which is 0.0012 (0.05/41) for cytokines. A P < 0.05, but above the Bonferroni-corrected threshold, was considered as suggestive for association.

Mediation analysis

By the two-sample analysis (step 1A and step 2A in Fig. 1), the gut microbiota and cytokines with significant causal effects on dementias were included in the mediation analysis. We explored whether gut microbiota had a causal effect on cytokines (step 3, path a, in Fig. 1), and if so, we would perform multiple MR analysis to explore whether cytokines were the mediation factors in the pathway from gut microbiota to dementia.

Bi‑directional causality analysis

To evaluate bi-directional causation effects between gut microbiota, cytokines, and dementias, we used dementias as “exposure” and gut microbiota or cytokines associated with dementias as “outcome” (step 1B and step 2B in Fig. 1). We selected the SNPs significantly associated with dementia (P < 5 × 10–8) as IVs.

Sensitivity analysis

We performed Cochran's Q test to evaluate the heterogeneity of each SNP [30] and generated scatter plots of SNP–exposure associations and SNP–outcome associations to visualize MR results. Leave-one-out analysis was performed to evaluate if each SNP could affect the results (by excluding each SNP at a time sequentially and an IVW method was performed on the remaining SNPs to assess the potential influence of a particular variant on the estimates) [31]. In addition, we used MR-PRESSO and MR-Egger regression to test the potential horizontal pleiotropy effect. MR-PRESSO was used to detect significant outliers and to correct the horizontal plural effect by removing outliers [32].

All analyses were carried out using R (v4.2.1) statistical software. MR analysis was performed using the R-based package “TwoSampleMR”. The “MR_PRESSO” package was used for multiplicity tests [33].

Results

Instrumental variable selection

Initially, we identified 224, 478, 1667, 280, and 125 SNPs associated with 210 gut microbiotas at the class, family, genus, order, and phylum levels, respectively, at a level of P < 1 × 10–5 (one gut microbiota was excluded due to no eligible SNPs). These 2774 SNPs were selected as IVs for the 210 gut microbiota taxa (Additional file 3: Table S2). Then, we identified 451 SNPs associated with 41 cytokines at a level of P < 5 × 10–6 (Additional file 4: Table S3).

Causal effects of gut microbiota and cytokines on multiple dementia types

AD

A total of six gut microbiotas (including one family, four genera, and one order) were associated with AD (Additional file 5: Table S4, Fig. 2). Detailed 61 SNPs information for six gut microbiotas is shown in Additional file 6: Table S5.

Fig. 2
figure 2

Mendelian randomization results of causal effects between gut microbiotas and five subtypes of dementia

Full size image

As shown in Fig. 2, MR analysis suggested that genetic prediction of three gut microbiotas (genus Allisonella, genus Lachnospiraceae FCS020 group, and genus Sellimonas) was associated with an increased risk of AD. The genus Allisonella (OR = 1.235, 95%CI = 1.058 ~ 1.441, P = 0.007), genus Lachnospiraceae FCS020 group (OR = 1.374, 95%CI = 1.109 ~ 1.702, P = 0.004) significantly increased the risk of AD.

Genetic prediction of three gut microbiotas (family Defluviitaleaceae, genus Anaerotruncus, and order Bacillales) was associated with a decreased risk of AD. The family Defluviitaleaceae (OR = 0.771, 95%CI = 0.628 ~ 0.946, P = 0.013), and order Bacillales (OR = 0.786, 95%CI = 0.688 ~ 0.899, P < 0.001) significantly decreased the risk of AD.

As shown in Fig. 3, macrophage migration inhibitory factor (MIF) (OR = 1.322, 95%CI = 1.141 ~ 1.532, P < 0.001) and basic fibroblast growth factor (FGFBasic) (OR = 1.538, 95%CI = 1.202 ~ 1.969, P = 0.001) significantly increased the incidence of AD (Additional file 7: Table S6).

Fig. 3
figure 3

Mendelian randomization results of causal effects between cytokines and five subtypes of dementia

Full size image

FTD

A total of eight gut microbiota (including one class, one family, five genera, and one order) were associated with FTD (Additional file 5: Table S4, Fig. 2). Detailed 87 SNPs information for the eight gut microbiotas is shown in Additional file 6: Table S5.

Figure 2 shows that genetic prediction of seven gut microbiotas (class Melainabacteria, family Rhodospirillaceae, genus Eubacterium fissicatena group, genus Phascolarctobacterium, unknown genus id.2041, unknown genus id.826, and order Rhodospirillales) was associated with an increased risk of FTD. The class Melainabacteria (OR = 3.313, 95%CI = 1.185 ~ 9.264, P = 0.022), unknown genus id.826 (OR = 6.626, 95%CI = 1.923 ~ 22.837, P = 0.003), and order Rhodospirillales (OR = 3.863, 95%CI = 1.156 ~ 12.914, P = 0.028) significantly increased the risk of FTD.

The genus Desulfovibrio (OR = 0.226, 95%CI = 0.057 ~ 0.902, P = 0.035) was associated with a decreased risk of FTD.

Figure 3 shows that tumor necrosis factor-alpha (TNF-α) and interleukin-16 (IL-16) were associated with FTD (Additional file 7: Table S6). TNF-α (OR = 3.108, 95%CI = 1.137 ~ 8.493, P = 0.027) significantly increased the incidence of FTD.

DLB

A total of nine gut microbiotas (including one class, six genera, one order, and one phylum) were associated with DLB (Additional file 5: Table S4, Fig. 2). Detailed 69 SNPs information for the nine gut microbiotas is shown in Additional file 6: Table S5.

As shown in Fig. 2, genetic prediction of four gut microbiotas (class Alphaproteobacteria, genus Bifidobacterium, genus Lachnospiraceae UCG001, and order Bacillales) was associated with an increased risk of DLB. The class Alphaproteobacteria (OR = 1.970, 95%CI = 1.320 ~ 2.940, P = 0.001) and order Bacillales (OR = 1.378, 95%CI = 1.116 ~ 1.703, P = 0.003) significantly increased the risk of DLB.

Genetic prediction of five gut microbiotas (genus Flavonifractor, genus Lachnospira, genus Ruminococcus gnavus group, genus Victivallis, and phylum Cyanobacteria) was associated with a decreased risk of DLB. The genus Lachnospira (OR = 0.132, 95%CI = 0.033 ~ 0.538, P = 0.005), genus Ruminococcus gnavus group (OR = 0.678, 95%CI = 0.523 ~ 0.878, P = 0.003), and phylum Cyanobacteria (OR = 0.619, 95%CI = 0.42 ~ 0.913, P = 0.016) significantly decreased the risk of DLB.

As shown in Fig. 3, MIF (OR = 0.737, 95%CI = 0.584 ~ 0.929, P = 0.010) seemed to be a protective factor for DLB (Additional file 7: Table S6).

VD

A total of six gut microbiotas (including five genera and one phylum) were associated with VD (Additional file 5: Table S4, Fig. 2). Detailed 57 SNPs information for the six gut microbiotas is shown in Additional file 6: Table S5.

As shown in Fig. 2, genetic prediction of four gut microbiotas (genus Eubacterium nodatum group, genus Ruminococcus gauvreauii group, genus Slackia, and genus Veillonella) was associated with an increased risk of VD. The genus Ruminococcus gauvreauii group (OR = 1.534, 95%CI = 1.112 ~ 2.117, P = 0.009), genus Slackia (OR = 1.533, 95%CI = 1.114 ~ 2.109, P = 0.009), and genus Veillonella (OR = 1.636, 95%CI = 1.083 ~ 2.472, P = 0.019) significantly increased risk of VD.

Genetic prediction of two gut microbiotas was associated with a decreased risk of VD. The genus Prevotella9 (OR = 0.772, 95%CI = 0.615 ~ 0.970, P = 0.026) and phylum Lentisphaerae (OR = 0.755, 95%CI = 0.600 ~ 0.951, P = 0.017) decreased the risk of VD.

FGFBasic (OR = 1.548, 95%CI = 1.086 ~ 2.207, P = 0.016) was associated with VD (Fig. 3, Additional file 7: Table S6).

PDD

A total of nine gut microbiota (including two classes, one family, three genera, two orders and one phylum) were associated with PDD (Additional file 5: Table S4, Fig. 2). Detailed 103 SNPs information for nine gut microbiotas is shown in Additional file 6: Table S5.

As shown in Fig. 2, genetic prediction of two gut microbiotas was associated with an increased risk of PDD. The genus Romboutsia (OR = 2.475, 95%CI = 1.304 ~ 4.695, P = 0.006) and genus Roseburia (OR = 2.198, 95%CI = 1.067 ~ 4.527, P = 0.033) increased the risk of PDD.

Genetic prediction of seven gut microbiotas was associated with a decreased risk of PDD. Notably, the class Erysipelotrichia, family Erysipelotrichaceae, and order Erysipelotrichales were the same gut microbiota. The class Lentisphaeria (OR = 0.463, 95%CI = 0.255 ~ 0.840, P = 0.011), genus Lachnoclostridium (OR = 0.440, 95%CI = 0.216 ~ 0.897, P = 0.024), order Victivallales (OR = 0.463, 95%CI = 0.255 ~ 0.840, P = 0.011), and phylum Lentisphaerae (OR = 0.474,95%CI = 0.281 ~ 0.800, P = 0.005) significantly decreased the risk of PDD.

A total of three cytokines were associated with PDD, including platelet-derived growth factor BB (PDGFbb), monokine induced by interferon-gamma (MIG), and interleukin-17 (IL-17) (Fig. 3, Additional file 7: Table S6). PDGFbb (OR = 0.625,95%CI = 0.454 ~ 0.861, P = 0.004) had a protective causal effect on PD. MIG (OR = 1.388,95%CI = 1.041 ~ 1.850, P = 0.026) was a risk factor for PDD.

Sensitivity analyses

According to MR-Egger regression intercept approach, genetic pleiotropy did not bias the results, and MR-PRESSO analysis proved that there was no horizontal pleiotropy in the MR study (P > 0.05, Additional file 8: Table S7). The Cochran’s Q tests showed no significant heterogeneity (P > 0.05, Additional file 8: Table S7).

The results of “leave-one-out” analysis proved that MR analysis turned out to be reliable. (The null line is not within the total confidence interval of the SNPs, Additional file 1: Figure S1-5.) The scatter plots showed the overall effect of gut microbiota on dementia (Additional file 1: Figure S6-10). In addition, the forest plots indicated the causal associations between gut microbiota and dementia (Additional file 1: Figure S11-15).

Bi‑directional causal effects of dementias on gut microbiota and cytokines

As shown in Additional file 9: Table S8, there was no reverse effect between gut microbiota, cytokines, and AD. After matching FTD and gut microbiota or cytokines, no SNP can be used as IV. DLB had causal effects on genus Lachnospira (OR = 0.833, 95%CI = 0.755 ~ 0.919, P < 0.001) and order Bacillales (OR = 1.075, 95%CI = 1.005 ~ 1.150, P = 0.034). VD had causal effects on genus Veillonella (OR = 1.071, 95%CI = 1.003 ~ 1.143, P = 0.040) and FGFBasic (OR = 1.085, 95%CI = 1.016 ~ 1.157, P = 0.014). PDD had a causal effect on Erysipelotrichia (OR = 1.070, 95%CI = 1.005 ~ 1.139, P = 0.034).

Mediation analysis

In this study, gut microbiota and cytokines all had causal effects on dementias. It seemed that cytokines played a mediating effect in the pathway from gut microbiota and dementia. One of the requirements for mediating effect is that gut microbiota was significantly associated with cytokines. However, our results revealed that there were no causal effects between gut microbiotas associated with dementias and cytokines associated with dementias (step 3a in Fig. 1; Additional file 10: Table S9), which indicated that cytokines did not act as a mediator in the pathway from gut microbiotas and dementias.

Discussion

The gut microbiome assists in many daily functions of the brain, including regulating the activation state of the hypothalamic–pituitary–adrenal (HPA) axis and activating the vagus and adrenergic nerves; in addition, intestinal bacteria themselves can synthesize and release many neurotransmitters and neuromodulators, or stimulate intestinal endocrine cells to synthesize and release neuropeptides [3].

The maladjustment of the brain–intestine–microbiome axis may lead to the dysfunction of the intestinal epithelial barrier, which in turn promotes the invasion of neuroactive substances, including neurotropic viruses and so on [34]. The gut microbiome may also play a role in metabolic diseases, such as insulin resistance and fatty liver disease [35]. Studies have shown that cholesterol metabolism is related to the mechanism of dementia development [36].

Studies have indicated that gut microbiota might affect dementia development. Naoki performed a cross-sectional study revealing that the number of Bacteroides (enterotype I) was lower in demented than non-demented patients [6]. However, due to the different types of dementia and the complexity of the gut microbiota, it was difficult to adequately summarize the gut microbiota affecting dementia through observational studies.

In this study, we used an MR study to explore the potential causal effects between gut microbiota and dementia. We analyzed the relationships between 210 common gut microbiota abundance and five types of dementia (AD, FTD, DLB, VD, PDD). The results showed that some gut microbiotas were risk factors, and some were protective factors for each dementia subtype.

High abundance of Allisonella, Lachnospiraceae FCS020 group, and Sellimonas could increase the risk of AD. Allisonella was associated with high level of inflammation [37], which provided a hypothesis on how Allisonella increased the risk of AD. A high abundance of Defluviitaleaceae, Anaerotruncus, and Bacillale could decrease the risk of AD.

Few studies explored the association between gut microbiota and FTD. In this study, Melainabacteria, Rhodospirillaceae, Eubacterium fissicatena group, Phascolarctobacterium, unknown genus id.2041, unknown genus id.826, and Rhodospirillales may be risk factors for FTD, and Desulfovibrio seemed to be a protective factor for FTD. Further studies were necessary considering there were only 103 cases of FTD.

In DLB, genus Ruminococcus may mitigate neuroinflammation in the substantia nigra by increasing secondary bile acids [38], which may explain why Ruminococcus decreased the risk of DLB. In addition, Flavonifractor, Lachnospira, Victivallis, and Cyanobacteria were also protective factors for DLB. Alphaproteobacteria, Bifidobacterium, Lachnospiraceae UCG001, and Bacillales may increase the incidence of DLB.

The evidence regarding whether the specific gut microbiota affected VD remained unclear. By MR analysis, we found Eubacterium nodatum group, Ruminococcus gauvreauii group, Slackia, and Veillonella may be the risk factors for VD; Prevotella9 and Lentisphaerae may decrease the incidence of VD.

Xie et al. reported that lower levels of Romboutsia and Roseburia were related to depressive symptoms in PD patients [39]. It seemed that a high abundance of Romboutsia and Roseburia could improve the symptom of PD patients. Contrary to their conclusions, we found that Romboutsia and Roseburia were associated with an increased risk of PDD. In terms of protective factors for PDD, our findings were similar to those of recent studies that in PD patients, Erysipelotrichaceae were markedly lowered, which proved that a higher abundance of Erysipelotrichaceae decreased the risk of PDD [40].

This study determined whether gut microbes were "helpful" or "harmful" to dementia by their relative abundance expression. However, the exact mechanism by which the gut microbiota causes dementia has not been determined. We assumed that cytokines may be mediating factors between gut microbiota and dementias.

According to MR analysis, we found that MIF and FGFBasic significantly increased the risk of AD. MIF is a pro-inflammatory cytokine. Previous studies have shown that increased MIF level could be a potential AD biomarker [41]. However, it seemed to be a negative correlation between MIF and DLB. TNF-α was significantly associated with FTD, which may be related to central degeneration [42].

In addition, dementia itself may affect changes in gut microbiota and cytokines. Therefore, we explored the causal effects of five dementia subtypes on gut microbiota and cytokines. The results showed that DLB had bi-directional causal effects on Lachnospira and Bacillales; VD had a bi-directional causal effect on Veillonella and FGFBasic; PDD had a bi-directional causal effect on Erysipelotrichia.

This was the first study to conduct a large-scale MR analysis of the causal relationships between the gut microbiome, cytokines, and several dementia subtypes. Our study had some limitations. First, the cases of dementia subtypes, especially FTD and PDD, were insufficient. Second, our study only analyzed the European population. Third, the 41 cytokines were derived from the blood, not the cerebrospinal fluid. Last, though we explored the mediating effects of cytokines between the abundance of different gut microbiota and dementias, the mechanisms how gut microbiota affected the onset of dementias remained to be studied considering that cytokines did not act as a mediating factor.

Conclusion

In this study, we comprehensively explored the causal effects between gut microbiota, cytokines, and dementias. There were 20 positive and 16 negative causal effects between genetic liability in the gut microbiota and dementias. There were five positive correlations and four negative causal effects between cytokines and dementias. In addition, we found four bi-directional causal effects between the gut microbiota and dementias, and one between cytokines and dementias. Cytokines seemed not to act as a mediating factor in the pathway from gut microbiota to dementias.

Availability of data and materials

All data used in the present study were obtained from genome‑wide association study summary statistics which were publicly released by genetic consortia.

Abbreviations

AD:

Alzheimer’s disease

FTD:

Frontotemporal dementia

DLB:

Dementia with Lewy body

VD:

Vascular dementia

PD:

Parkinson’s disease

PDD:

Parkinson’s disease dementia

GWAS:

Genome-wide association studies

MR:

Mendelian randomization

IVs:

Instrumental variables

SNP:

Single-nucleotide polymorphism

ICD:

International Classification of Diseases

LD:

Linkage disequilibrium

EA:

Effect allele

OA:

Other allele

EAF:

Effect allele frequency

β:

Effect sizes

SE:

Standard error

IVW:

Inverse variance weighted

OR:

Odds ratio

CI:

Confidence interval

References

  1. Maclin JMA, Wang T, Xiao S. Biomarkers for the diagnosis of Alzheimer’s disease, dementia Lewy body, frontotemporal dementia and vascular dementia. Gen Psychiatr. 2019;32(1): e100054.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Aarsland D, Batzu L, Halliday GM, Geurtsen GJ, Ballard C, Ray Chaudhuri K, et al. Parkinson disease-associated cognitive impairment. Nat Rev Dis Primers. 2021;7(1):47.

    Article  PubMed  Google Scholar 

  3. Alkasir R, Li J, Li X, Jin M, Zhu B. Human gut microbiota: the links with dementia development. Protein Cell. 2017;8(2):90–102.

    Article  CAS  PubMed  Google Scholar 

  4. Spanogiannopoulos P, Bess EN, Carmody RN, Turnbaugh PJ. The microbial pharmacists within us: a metagenomic view of xenobiotic metabolism. Nat Rev Microbiol. 2016;14(5):273–87.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. O’Hara AM, Shanahan F. The gut flora as a forgotten organ. EMBO Rep. 2006;7(7):688–93.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Saji N, Niida S, Murotani K, Hisada T, Tsuduki T, Sugimoto T, et al. Analysis of the relationship between the gut microbiome and dementia: a cross-sectional study conducted in Japan. Sci Rep. 2019;9(1):1008.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Ponziani FR, Nesci A, Caputo C, Salvatore L, Picca A, Del Chierico F, et al. High prevalence of lower limb atherosclerosis is linked with the gut–liver axis in patients with primary biliary cholangitis. Liver Int. 2023;43(2):370–80.

    Article  CAS  PubMed  Google Scholar 

  8. Kamada N, Seo SU, Chen GY, Núñez G. Role of the gut microbiota in immunity and inflammatory disease. Nat Rev Immunol. 2013;13(5):321–35.

    Article  CAS  PubMed  Google Scholar 

  9. Shen XN, Niu LD, Wang YJ, Cao XP, Liu Q, Tan L, et al. Inflammatory markers in Alzheimer’s disease and mild cognitive impairment: a meta-analysis and systematic review of 170 studies. J Neurol Neurosurg Psychiatry. 2019;90(5):590–8.

    Article  PubMed  Google Scholar 

  10. Darweesh SKL, Wolters FJ, Ikram MA, de Wolf F, Bos D, Hofman A. Inflammatory markers and the risk of dementia and Alzheimer’s disease: a meta-analysis. Alzheimers Dement. 2018;14(11):1450–9.

    Article  PubMed  Google Scholar 

  11. Walker KA, Gottesman RF, Wu A, Knopman DS, Gross AL, Mosley TH Jr, et al. Systemic inflammation during midlife and cognitive change over 20 years: the ARIC study. Neurology. 2019;92(11):e1256–67.

    CAS  PubMed  PubMed Central  Google Scholar 

  12. Nishiwaki H, Ueyama J, Kashihara K, Ito M, Hamaguchi T, Maeda T, et al. Gut microbiota in dementia with Lewy bodies. NPJ Parkinsons Dis. 2022;8(1):169.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  13. Meng HYH, Mak CCH, Mak WY, Zuo T, Ko H, Chan FKL. Probiotic supplementation demonstrates therapeutic potential in treating gut dysbiosis and improving neurocognitive function in age-related dementia. Eur J Nutr. 2022;61(4):1701–34.

    Article  CAS  PubMed  Google Scholar 

  14. Jordan F, Quinn TJ, McGuinness B, Passmore P, Kelly JP, Tudur Smith C, et al. Aspirin and other non-steroidal anti-inflammatory drugs for the prevention of dementia. Cochrane Database Syst Rev. 2020;4(4):Cd011459.

    PubMed  Google Scholar 

  15. Visscher PM, Wray NR, Zhang Q, Sklar P, McCarthy MI, Brown MA, et al. 10 years of GWAS discovery: biology, function, and translation. Am J Hum Genet. 2017;101(1):5–22.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Lawlor DA, Harbord RM, Sterne JA, Timpson N, Davey SG. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med. 2008;27(8):1133–63.

    Article  PubMed  Google Scholar 

  17. Smith GD, Ebrahim S. ‘Mendelian randomization’: can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol. 2003;32(1):1–22.

    Article  PubMed  Google Scholar 

  18. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Davies NM, Holmes MV, Davey SG. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;362: k601.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Pierce BL, Burgess S. Efficient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol. 2013;178(7):1177–84.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Magnus MC, Miliku K, Bauer A, Engel SM, Felix JF, Jaddoe VWV, et al. Vitamin D and risk of pregnancy related hypertensive disorders: mendelian randomisation study. BMJ. 2018;361: k2167.

    Article  PubMed  PubMed Central  Google Scholar 

  22. Bowden J, Holmes MV. Meta-analysis and Mendelian randomization: a review. Res Synth Methods. 2019;10(4):486–96.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Kurilshikov A, Medina-Gomez C, Bacigalupe R, Radjabzadeh D, Wang J, Demirkan A, et al. Large-scale association analyses identify host factors influencing human gut microbiome composition. Nat Genet. 2021;53(2):156–65.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  24. Ahola-Olli AV, Würtz P, Havulinna AS, Aalto K, Pitkänen N, Lehtimäki T, et al. Genome-wide association study identifies 27 loci influencing concentrations of circulating cytokines and growth factors. Am J Hum Genet. 2017;100(1):40–50.

    Article  CAS  PubMed  Google Scholar 

  25. Chia R, Sabir MS, Bandres-Ciga S, Saez-Atienzar S, Reynolds RH, Gustavsson E, et al. Genome sequencing analysis identifies new loci associated with Lewy body dementia and provides insights into its genetic architecture. Nat Genet. 2021;53(3):294–303.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  26. Myers TA, Chanock SJ, Machiela MJ. LDlinkR: an R package for rapidly calculating linkage disequilibrium statistics in diverse populations. Front Genet. 2020;11:157.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Burgess S, Small DS, Thompson SG. A review of instrumental variable estimators for Mendelian randomization. Stat Methods Med Res. 2017;26(5):2333–55.

    Article  PubMed  Google Scholar 

  28. Papadimitriou N, Dimou N, Tsilidis KK, Banbury B, Martin RM, Lewis SJ, et al. Physical activity and risks of breast and colorectal cancer: a Mendelian randomisation analysis. Nat Commun. 2020;11(1):597.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. 2013;37(7):658–65.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Cohen JF, Chalumeau M, Cohen R, Korevaar DA, Khoshnood B, Bossuyt PM. Cochran’s Q test was useful to assess heterogeneity in likelihood ratios in studies of diagnostic accuracy. J Clin Epidemiol. 2015;68(3):299–306.

    Article  PubMed  Google Scholar 

  31. Burgess S, Thompson SG. Interpreting findings from Mendelian randomization using the MR-Egger method. Eur J Epidemiol. 2017;32(5):377–89.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Ong JS, MacGregor S. Implementing MR-PRESSO and GCTA-GSMR for pleiotropy assessment in Mendelian randomization studies from a practitioner’s perspective. Genet Epidemiol. 2019;43(6):609–16.

    Article  PubMed  PubMed Central  Google Scholar 

  34. Hawkes CH, Del Tredici K, Braak H. Parkinson’s disease: a dual-hit hypothesis. Neuropathol Appl Neurobiol. 2007;33(6):599–614.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Dumas ME, Barton RH, Toye A, Cloarec O, Blancher C, Rothwell A, et al. Metabolic profiling reveals a contribution of gut microbiota to fatty liver phenotype in insulin-resistant mice. Proc Natl Acad Sci USA. 2006;103(33):12511–6.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  36. Martins IJ, Hone E, Foster JK, Sünram-Lea SI, Gnjec A, Fuller SJ, et al. Apolipoprotein E, cholesterol metabolism, diabetes, and the convergence of risk factors for Alzheimer’s disease and cardiovascular disease. Mol Psychiatry. 2006;11(8):721–36.

    Article  CAS  PubMed  Google Scholar 

  37. Aranaz P, Ramos-Lopez O, Cuevas-Sierra A, Martinez JA, Milagro FI, Riezu-Boj JI. A predictive regression model of the obesity-related inflammatory status based on gut microbiota composition. Int J Obes. 2021;45(10):2261–8.

    Article  CAS  Google Scholar 

  38. Hirayama M, Nishiwaki H, Hamaguchi T, Ohno K. Gastrointestinal disorders in Parkinson’s disease and other Lewy body diseases. NPJ Parkinsons Dis. 2023;9(1):71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Xie A, Ensink E, Li P, Gordevičius J, Marshall LL, George S, et al. Bacterial butyrate in Parkinson’s disease is linked to epigenetic changes and depressive symptoms. Mov Disord. 2022;37(8):1644–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Bedarf JR, Hildebrand F, Coelho LP, Sunagawa S, Bahram M, Goeser F, et al. Functional implications of microbial and viral gut metagenome changes in early stage L-DOPA-naïve Parkinson’s disease patients. Genome Med. 2017;9(1):39.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Zhang S, Zhao J, Zhang Y, Zhang Y, Cai F, Wang L, et al. Upregulation of MIF as a defense mechanism and a biomarker of Alzheimer’s disease. Alzheimers Res Ther. 2019;11(1):54.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  42. Chu M, Wen L, Jiang D, Liu L, Nan H, Yue A, et al. Peripheral inflammation in behavioural variant frontotemporal dementia: associations with central degeneration and clinical measures. J Neuroinflammation. 2023;20(1):65.

    Article  PubMed  PubMed Central  Google Scholar 

Download references

Acknowledgements

The authors thank the investigators of the original studies for sharing the GWAS summary statistics.

Funding

The study was supported by the National Natural Science Foundation of China [grant number 82270644] and the Postgraduate Innovation Research and Practice Program of Anhui Medical University [grant number YJS20230017].

Author information

Author notes

  1. Dong Ji and Wen-Zhu Chen contributed equally to this work and should be considered co-first authors.

Authors and Affiliations

  1. Department of Anesthesiology, The First Affiliated Hospital of Anhui Medical University, No.218 Jixi Road, Shushan District, Hefei, Anhui, China

    Dong Ji, Wen-Zhu Chen, Lei Zhang & Li-Jian Chen

  2. Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, Hefei, Anhui, China

    Zhi-Hua Zhang

Authors
  1. Dong Ji
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Wen-Zhu Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lei Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Zhi-Hua Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Li-Jian Chen
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

Conceptualization: DJ, WZC, and LJC. Acquisition, analysis, or interpretation of data: all authors. Drafting of the manuscript: DJ, WZC, and LZ. Critical revision of the manuscript for important intellectual content: DJ, LZ and LJC. Statistical analysis: DJ, ZHZ, and LJC. Obtained funding: DJ and LJC. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Li-Jian Chen.

Ethics declarations

Ethics approval and consent to participate

The present study is a secondary analysis of publicly available data. Ethical approval was granted for each of the original GWAS studies. In addition, no individual-level data were used in this study. Therefore, no new ethical review board approval was required.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

The plots of MR analysis results.

Additional file 2: Table S1.

Overview of the source of dementia data.

Additional file 3: Table S2.

2774 SNPs for the 210 gut microbiota taxa.

Additional file 4: Table S3.

451 SNPs for the 41 cytokines.

Additional file 5: Table S4.

The causal effects of gut microbiota on dementia. 

Additional file 6: Table S5.

The characteristics of SNPs analyzing the causal effects of gut microbiota on dementia.

Additional file 7: Table S6.

The causal effects of cytokines on dementia. 

Additional file 8: Table S7.

Mendelian randomization Sensitivity analysis. 

Additional file 9: Table S8.

The causal effects of dementia on gut microbiota and cytokines. 

Additional file 10: Table S9.

Causal effects of gut microbiota associated with dementia on cytokines associated with dementia. 

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Ji, D., Chen, WZ., Zhang, L. et al. Gut microbiota, circulating cytokines and dementia: a Mendelian randomization study. J Neuroinflammation 21, 2 (2024). https://doi.org/10.1186/s12974-023-02999-0

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s12974-023-02999-0

Keywords

Journal of Neuroinflammation

ISSN: 1742-2094