Journal of Clinical Chemistry and Laboratory Medicine
Open Access

Gut Microbiota and Central Nervous System: A Bidirectional Two-Sample Mendelian Randomized Analysis

Karl Krupp^* and Qingfeng Wang ^*Correspondence: Karl Krupp, Department of Medicine, University of Arizona Cancer Center, Tucson, Arizona, USA, Email:

Author info »

Introduction

Anoxic Brain Injury (ABI), such as traumatic brain injury, stroke, cardiac arrest, asphyxia and neonatal ischemic hypoxic encephalopathy, is a common clinical cause of central nervous system injury [1-5]. It can occur in various age groups with poor prognosis. In severe cases may have permanent mental and cognitive dysfunction [1,6,7]. Bacterial Meningitis (BM) is an infectious disease of the CNS that commonly affects adults and children and with high morbidity, mortality and sequelae characteristics. Currently, the conservative treatments of ABI and BM have not yielded satisfactory therapeutic results.

The balance and stability of the gut microbiota is crucial for host healthy. In contrast, conditions such as hypoxia and infection will be disordering gut microbiota, disrupting the bidirectional balance between the gut and brain in Table 1 and causing cognitive and motor deficits [8,9]. Numerous studies have shown that the gut microbiota was involved in the regulation of cellular and molecular mechanisms of the brain injury process, found that a decreased diversity of intestinal flora in patients with CNS diseases, mainly in the abundance of Clostridium, Anaerostipes and Lachnobacterium [10]. In addition, research showed the correlation between the changed gut microbiota and clinical phenotype [11]. Meanwhile, it has been showed that changed flora are associated with the onset and progression of ABI and BM [5,10].

Table 1: Association of genetically predicted the causal effect between the gut microbiota and ABI and BM by four different MR methods: IVW, MR Egger, weighted median, weighted mode.

Seki, et al. found Klebsiella may as an outstanding predictor indicator of brain injury in preterm infants [12]. Animal study have shown that the prognosis of ABI in rats can be improved by intervening gut microbiota (such as butyrate-producing bacteria) [5,13]. However, the causal relationship still unclear.

The intestinal flora is the largest immune organ in the human body and its metabolites such as Short-Chain Fatty Acids (SCFA), tryptophan and kynurenine are involved in immune regulation in the body and exert anti-inflammatory effects [14,15]. In fact, due to the numerous confounding factors in clinical and animal studies, it is still difficult to clarify how gut microbiota connects with CNS diseases. The MR analysis method can exclude the interference of confounding factors and avoid the influence of reverse confounding factors, which makes the results more rigorous and reliable.

This study first to use a two-sample bidirectional MR method to analyze the potential causal relationship between the gut microbiota and two different CNS disorders (ABI and BM), as well as to investigate the genetic relationship between them.

Materials and Methods

Study design

We conducted a two-sample bidirectional MR study to investigate the causal relationship between the gut microbiome and ABI and BM. To ensure valid Instrumental Variables (IVs) were obtained, MR was designed must base on three basic assumptions as follows: (1) Single Nucleotide Polymorphisms (SNPs) were robustly associated with exposure factors; (2) SNPs must be independent of any conventional and unknown confounders; (3) SNPs must be associated with outcomes only through exposure factors.

Source of datasets

Datasets of gut microbiota, ABI and BM were obtained from the GWAS database. The gut microbiota data was from a large GWAS analysis of 24 cohorts (18,340 individuals) conducted by MiBioGen consortium [8], which included genome-wide genotyping and 16sRNA sequencing data. ABI dataset (contains 191 cases and 205,799 controls) and BM dataset (contains 574 cases and 217,485 controls) were conducted by FinnGen consortium derived from European descent groups (Table 2).

Table 2: Characteristics of the study used for primary MR analysis.

This study performed the secondary analysis using public GWAS datasets and the Institutional Review Board review was not required.

Instrumental variables

SNPs were selected as IVs at a threshold of P<1 × 10-5. Meanwhile, we only selected independent genetic variants which are not in the linkage disequilibrium (defined as R2<0.001, Kb=10000). Then, SNPs that did not have A/T or C/G polymorphisms were excluded from the pool of SNPs based on the principle that the effects of selected SNPs on exposure and outcome were caused by having the same allele. We also calculated F-statistics for the SNPs to assess their instrumental strengths. F-statistic less than 10 were removed. Ultimately, 2037 SNPs were identified that were associated with 195 microbiota traits (9 phyla, 16 classes, 20 orders, 31 families, 119 genera). The study used two-sample MR analysis for causal analysis.

Mendelian randomization analysis

In this study, we choose Inverse Variance Weighted (IVW), MR-Egger regression, weighted median estimator and weighted mode for MR analysis. Their characteristics have been described by several studies [16-18].

In addition, outliers can be detected for pleiotropy bias through MR of pleiotropy residuals and Cochran’s Q test to quantify the heterogeneity among the selected SNPs (P<0.05 was considered as possible heterogeneity in IVs) [19]. A leave-one-out sensitivity analysis was performed on the results by observing whether there was a statistical difference before and after removing each SNP. If there is little change in the results removing the SNP, which indicates that the SNP would not have a nonspecific effect on the effect estimate?

Evaluation of horizontal multidirectional and heterogeneity

The intercept term of MR-Egger regression detects the presence of directional heterogeneity, when the ending term egger intercept is close to zero; it represents no heterogeneity in the IVs. Analyses were conducted by the “TwoSampleMR” and “MendelR” packages [20,21]. Results were presented as Odds Ratios (OR) with respective 95% CI. All presented P-values were two-sided and statistical significance was set at the 5% level.

Results

Genetic instruments for gut microbiome

There were 195 bacteria traits containing five biological levels in our study. The detailed information of the SNPs for each bacteria trait.

Gut microbiota exposure was obtained from 24 cohort studies in the United States, Canada, Israel, South Korea, Germany, Denmark, The Netherlands, Belgium, Sweden, Finland and the United Kingdom. After removing linkage disequilibrium, a total of 2037 SNPs were enrolled. In addition, we collected additional information about the SNPs, such as effector alleles, beta, se and P values.

Mendelian randomization analysis of gut microbiota and ABI and BM

Based on several different MR methods, we observed a potential causal relationship between the gut microbiota and ABI and BM. With the result in IVW, we found two gut taxa positively associated with risk of ABI (family_Lachnospiraceae: OR 5.13, 95%CI 1.13-23.32; genus_Butyricicoccus: OR 6.53,95%CI 1.47-29.01). While three gut taxa positively associated with BM risk (genus_Lactococcus: OR 1.50, 95%CI 1.02-2.20; genus_Ruminococcus gauvreauii: OR 2.85, 95%CI 1.75-5.18; genus_Desulfovibrionales: OR 2.06, 95% CI 1.02-4.16) and three gut taxa negatively associated with BM risk (genus_Eubacterium ventriosum: OR 0.46, 95%CI 0.25-0.85; genus_Erysipelatoclostridium: OR 0.48，95%CI 0.31-0.76; order_NB1n: ORn 0.56, 95%CI 0.39-0.80). Other algorithms had similar results (Table 1).

No heterogeneity effect found by Cochran's Q and the P>0.05 in MR-Egger interprets, showing the absence of horizontal pleiotropy (Table 3). Firstly, we visually examined forest plot and funnel plot. Leave-one-out analysis also revealed the robustness of our main results. Finally, four methods were employed to assess the results of MR analysis and the scatter plot was generated for BM and ABI.

Table 3: MR results of causal links between gut microbiota and ABI and BM.

Genetic instruments for ABI and BM

Additional information on SNPs, such as effector alleles, beta, se and P-values, was similarly collected when ABI and BM were used as exposure factors.

We found that increased abundance of the Veillonellaceae family, Lachnospiraceae NC2004 group and Eisenbergiella genus was positivity associated with ABI risk, while the decreased abundance of the Oscillibacter genus negative associate with ABI risk, which suggest that Oscillibacter genus act as a protective infector in ABI. The risk of BM was positivity associated with the increased abundance of Clostridiales vadin BB60 group family, while the decreased abundance of Eubacterium hallii group genus, Eubacterium ventriosum group genus and Erysipelatoclostridium genus was negativity associated with BM risk. Other algorithms yielded similar results, while significant differences were only observed in IVW (Table 4). No heterogeneity was found by Cochran's Q (P>0.05). We examined forest plot and funnel plot. Leave-one-out analysis also revealed the robustness of our main results. Finally, four methods were employed to assess the results of MR analysis, and the scatter plot was generated for gut flora.

Table 4: Association of genetically predicted the causal effect between ABI, BM and the gut microbiota by four different MR methods: IVW, MR Egger, weighted median and weighted mode.

Discussion

Recently the gut microbiome is recognized as a key regulator of host healthy. The gut microbiota effect on the host may through the metabolome, transcriptome, and epigenome pathways [22,23]. With the unveiling of the "gut-brain axis", it has been found that the connection can be through: (1) Bacterial components such as lipopolysaccharides stimulate the immune system to produce systemic or CNS inflammation [24]; (2) Bacterial proteins cross over with antigens to stimulate dysfunction in adaptive immunity [24]; (3) Bacterial enzymes produce neurotoxins and neurological metabolites[25]; (4) Gut microbiota produce hormones and neurotransmitters [26,27]; (6) Intestinal bacteria directly stimulate adaptive immunity [28].

In this bidirectional MR study, we found a causal relationship between gut microbiota and ABI and BM. The Lachnospiraceae families and Butyricicoccus genus are all Firmicutes and positively associated with the risk of ABI. In a reverse causality test, ABI altered the gut microbiota, with increased abundance of the Lachnospiraceae families. Butyricoccus spp. is related to butyrate production a Short-Chain Fatty Acids (SCFA), so we speculate that butyrate may act as a risk factor in ABI. Animal studies have found that Butyricoccus genus was increased significantly in the infected mice that may associated with an upregulation of inflammation response in the intestines [29]. However, the results of this MR analysis are contrary to other brain-gut axis diseases [5,30]. Previous studies have demonstrated the butyrate can promote the process of renewal and repair to intestinal cell, as well as enhance the function of immune cell. However, there are little clinical studies about the state of the intestines in patients with ABI. It’s still not clear whether this paralleled level of Butyricoccus spp. to intestinal inflammation exists in ABI patients. Only it can be confirmed by numerous of subsequent clinical and animal experiments. Reversely, the decreased abundance of Lachnospiraceae family is negatively associated with severity in other CNS diseases, such as depression and Parkinson [30,31]. Meanwhile, previous studies of hypoxic or ischemic-hypoxic brain injury have shown a decrease in the proportion of Firmicutes [5]. It’s difficult to conclude they are protective or risk factor, because of Lachnospiraceae family and Butyricoccus genus as a member of the Firmicutes. Patnala’ team shown that butyrate regulates H3K9ac and enhances neuroprotection of microglia cell at the gene level during stroke [32]. However, this MR research did not involve fecal metabolites, so it is difficult to determine the role of butyrate in ABI, which needs to be confirmed by further animal and clinical studies.

The increased abundance of Lactococcus genus is positively associated with the risk of BM. Since the 1990s, there have been successive case reports of blood-borne infections caused by the Lactococcus genus [33,34]. The Ruminococcus gauvreauii genus is a bacterium isolated from human bile. Djawad, et al. found that Ruminococcus gauvreauii genus is involved in the synthesis of glutamate, butyric acid, 5-hydroxytryptamine, all these neurotransmitters are associated with depression [35]. The Desulfovibrionales genus, one of the Aspergillus, can use lactic acid, pyruvic acid and ethanol as carbon sources to reduce sulphate to hydrogen sulphide, the latter of which is closely associated with inflammatory responses in the host and as a risk factor for inflammation. Clinical studies have shown that the abundance of Desulfovibrionales genus correlates with the severity of Parkinson's and the mechanism may be related to the fact that Desulfovibrionales genus can produce hydrogen sulphide and lipopolysaccharide or induce oligomerisation of a-synuclein [36,37].

Conclusion

In summary, defining the characteristics of a bacterium requires placing it in the context of the entire microecosystem and discerning its abundance threshold and functional characteristics through the composition and metabolism of the entire ecological flora. Although the abundance of some of the bacteria in our study served as a disease risk factor, which contradicts of other studies, we believe that the gut flora acts as a dynamic equilibrium to influence the host and the pattern of response varies with different diseases. It’s still difficult to define clearly link between the gut microbiota and the host by the results of the current clinical and animal studies.

Despite the rigorous statistical methods used in this study, limitations still exist in this study. Although we used linear MR analyses, the lack of specific sample information, it was not possible to conduct further observational analyses on the age and sex of the exposure and outcome populations. Furthermore, our IVs were screened for P<1 × 10⁵, and the results may have been affected by weak instrumental bias.

Using a two-sample bidirectional MR analysis, our study confirmed a bidirectional causal relationship between gut microbial abundance and the risk of ABI and BM. Among this, a strong association between elevated abundance of Lachnospiraceae family and the risk of ABI, which may serve as a potential target for the treatment of ABI?

Consent for Publication

Informed consent was obtained from all subjects involved in the study.

Availability of Data and Materials

The data presented in this study are available on request from corresponding author.

Funding

This work was supported by The National Natural Science Foundation of China (81970265,82000303).

Author’s Contributions

Conceptualization, Jia An, Mingtang Ye; data collection, Jia An; data analysis and interpretation, Di Yu, Qingfeng Wang; software and statistical analysis, Qiang Wang, Kede Wu; writing-original draft preparation, Jia An, Zhaocong Yang, Xuming Mo. All authors contributed to the final version of the manuscript and agreed to the published version of the manuscript.

References

1. Shavelle RM, Brooks JC, Strauss DJ. An update on survival after anoxic brain injury in adolescents and young adults. Brain Inj. 2018;32(13-14):1879.

   [Crossref] [Google Scholar] [PubMed]

2. Hu W, Kong X, Wang H, Li Y, Luo Y. Ischemic stroke and intestinal flora: an insight into brain–gut axis. Eur J Med Res. 2022;27(1):1-3.

   [Crossref] [Google Scholar] [PubMed]

3. Khan MZ, Khan MU, Patel K, Khan SU, Valavoor S, Osman M, et al. Trends, predictors and outcomes after utilization of targeted temperature management in cardiac arrest patients with anoxic brain injury. Am J Med Sci. 2020;360(4):363-371.

   [Crossref] [Google Scholar] [PubMed]

4. Kriel RL, Krach LE, Luxenberg MG, Jones-Saete C, Sanchez J. Outcome of severe anoxic/ischemic brain injury in children. Pediatric neurol. 1994;10(3):207-212.

   [Crossref] [Google Scholar] [PubMed]

5. He X, Zhang T, Zeng Y, Pei P, Liu Y, Jia W, et al. Sodium butyrate mediates histone crotonylation and alleviated neonatal rats hypoxic–ischemic brain injury through gut–brain axis. Front Microbiol. 2022;13:993146.

   [Crossref] [Google Scholar] [PubMed]

6. Garcia-Molina A, Roig-Rovira T, Ensenat-Cantallops A, Sanchez-Carrion R, Pico-Azanza N, Bernabeu M, et al. Neuropsychological profile of persons with anoxic brain injury: Differences regarding physiopathological mechanism. Brain inj. 2006;20(11):1139-1145.

   [Crossref] [Google Scholar] [PubMed]

7. Moulaert VR, Verbunt JA, van Heugten CM, Wade DT. Cognitive impairments in survivors of out-of-hospital cardiac arrest: a systematic review. Resuscitation. 2009;80(3):297-305.

   [Crossref] [Google Scholar] [PubMed]

8. 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):156-165.

   [Crossref] [Google Scholar] [PubMed]

9. Yuan B, Lu XJ, Wu Q. Gut microbiota and acute central nervous system injury: a new target for therapeutic intervention. Front Immunol. 2021;12:800796.

   [Crossref] [Google Scholar] [PubMed]

10. Grochowska M, Laskus T, Paciorek M, Pollak A, Lechowicz U, Makowiecki M, et al. Patients with Infections of The Central Nervous System Have Lowered Gut Microbiota Alpha Diversity. Curr Issues Mol Biol. 2022;44(7):2903-2914.

    [Crossref] [Google Scholar] [PubMed]

11. Li H, Zhang L, Zhang K, Huang Y, Liu Y, Lu X, et al. Gut microbiota associated with cryptococcal meningitis and dysbiosis caused by anti-fungal treatment. Front Microbiol. 2023;13:1086239.

    [Crossref] [Google Scholar] [PubMed]

12. Seki D, Mayer M, Hausmann B, Pjevac P, Giordano V, Goeral K, et al. Aberrant gut-microbiota-immune-brain axis development in premature neonates with brain damage. Cell Host Microbe. 2021;29(10):1558-1572.

    [Crossref] [Google Scholar] [PubMed]

13. Olson CA, Iñiguez AJ, Yang GE, Fang P, Pronovost GN, Jameson KG, et al. Alterations in the gut microbiota contribute to cognitive impairment induced by the ketogenic diet and hypoxia. Cell Host Microbe. 2021;29(9):1378-1392.

    [Crossref] [Google Scholar] [PubMed]

14. Sarkar A, Lehto SM, Harty S, Dinan TG, Cryan JF, Burnet PW. Psychobiotics and the manipulation of bacteria–gut–brain signals. Trends Neurosci. 2016;39(11):763-781.

    [Crossref] [Google Scholar] [PubMed]

15. Hsiao EY, McBride SW, Hsien S, Sharon G, Hyde ER, McCue T, et al. Microbiota modulate behavioral and physiological abnormalities associated with neurodevelopmental disorders. Cell. 2013;155(7):1451-1463.

    [Crossref] [Google Scholar] [PubMed]

16. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol. 2015;44(2):512-525.

    [Crossref] [Google Scholar] [PubMed]

17. Bowden J, Del Greco M F, Minelli C, Davey Smith G, Sheehan N, Thompson J. A framework for the investigation of pleiotropy in two‐sample summary data Mendelian randomization. Stat Med. 2017;36(11):1783-1802.

    [Crossref] [Google Scholar] [PubMed]

18. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304-314.

    [Crossref] [Google Scholar] [PubMed]

19. Xiang K, Wang P, Xu Z, Hu YQ, He YS, Chen Y, et al. Causal effects of gut microbiome on systemic lupus erythematosus: a two-sample mendelian randomization study. Front Immunol. 2021;12:667097.

    [Crossref] [Google Scholar] [PubMed]

20. Yavorska OO, Burgess S. MendelianRandomization: an R package for performing Mendelian randomization analyses using summarized data. Int J Epidemiol. 2017;46(6):1734-1739.

    [Crossref] [Google Scholar] [PubMed]

21. Hemani G, Zheng J, Elsworth B, Wade KH, Haberland V, Baird D, et al. The MR-Base platform supports systematic causal inference across the human phenome. Elife. 2018;7:e34408.

    [Crossref] [Google Scholar] [PubMed]

22. Sorboni SG, Moghaddam HS, Jafarzadeh-Esfehani R, Soleimanpour S. A comprehensive review on the role of the gut microbiome in human neurological disorders. Clin Microbiol Rev. 2022;35(1):e00338-20.

    [Crossref] [Google Scholar] [PubMed]

23. Cresci GA, Bawden E. Gut microbiome: what we do and don't know. Nutr Clin Pract. 2015;30(6):734-746.

    [Crossref] [Google Scholar] [PubMed]

24. Hao MM, Stamp LA. The many means of conversation between the brain and the gut. Nat Rev Gastroenterol Hepatol. 2023;20(2):73-74.

    [Crossref] [Google Scholar] [PubMed]

25. Teng Y, Mu J, Xu F, Zhang X, Sriwastva MK, Liu QM, et al. Gut bacterial isoamylamine promotes age-related cognitive dysfunction by promoting microglial cell death. Cell Host Microbe. 2022;30(7):944-960.

    [Crossref] [Google Scholar] [PubMed]

26. Ikeda T, Nishida A, Yamano M, Kimura I. Short-chain fatty acid receptors and gut microbiota as therapeutic targets in metabolic, immune, and neurological diseases. Pharmacol Ther. 2022;239:108273.

    [Crossref] [Google Scholar] [PubMed]

27. Bosch JA, Nieuwdorp M, Zwinderman AH, Deschasaux M, Radjabzadeh D, Kraaij R, et al. The gut microbiota and depressive symptoms across ethnic groups. Nat Commun. 2022;13(1):7129.

    [Crossref] [Google Scholar] [PubMed]

28. Dohnalová L, Lundgren P, Carty JR, Goldstein N, Wenski SL, Nanudorn P, et al. A microbiome-dependent gut–brain pathway regulates motivation for exercise. Nature. 2022;612(7941):739-747.

    [Crossref] [Google Scholar] [PubMed]

29. Alam MS, Gangiredla J, Hasan NA, Barnaba T, Tartera C. Aging-induced dysbiosis of gut microbiota as a risk factor for increased listeria monocytogenes infection. Front Immunol. 2021;12:672353.

    [Crossref] [Google Scholar] [PubMed]

30. Ye X, Wang D, Zhu H, Wang D, Li J, Tang Y, et al. Gut microbiota changes in patients with major depressive disorder treated with vortioxetine. Front Psychiatry. 2021;12:641491.

    [Crossref] [Google Scholar] [PubMed]

31. Chen W, Bi Z, Zhu Q, Gao H, Fan Y, Zhang C, et al. An analysis of the characteristics of the intestinal flora in patients with Parkinson’s disease complicated with constipation. Am J Transl Res. 2021;13(12):13710.

    [Google Scholar] [PubMed]

32. Patnala R, Arumugam TV, Gupta N, Dheen ST. HDAC inhibitor sodium butyrate-mediated epigenetic regulation enhances neuroprotective function of microglia during ischemic stroke. Mol Neurobiol. 2017;54:6391-6411.

    [Crossref] [Google Scholar] [PubMed]

33. Durand JM, Rousseau MC, Gandois JM, Kaplanski G, Mallet MN, Soubeyrand J. Streptococcus lactis septicemia in a patient with chronic lymphocytic leukemia. Am J Hematol. 1995;50(1):64-65.

    [Crossref] [Google Scholar] [PubMed]

34. Fefer JJ, Ratzan KR, Sharp SE, Saiz E. Lactococcus garvieae endocarditis: report of a case and review of the literature. Diagn Microbiol Infect Dis. 1998;32(2):127-130.

    [Crossref] [Google Scholar] [PubMed]

35. Radjabzadeh D, Bosch JA, Uitterlinden AG, Zwinderman AH, Ikram MA, van Meurs JB, et al. Gut microbiome-wide association study of depressive symptoms. Nat Commun. 2022;13(1):7128.

    [Crossref] [Google Scholar] [PubMed]

36. Hu H, Shao W, Liu Q, Liu N, Wang Q, Xu J, et al. Gut microbiota promotes cholesterol gallstone formation by modulating bile acid composition and biliary cholesterol secretion. Nat Commun. 2022;13(1):252.

    [Crossref] [Google Scholar] [PubMed]

37. Murros KE, Huynh VA, Takala TM, Saris PE. Desulfovibrio bacteria are associated with Parkinson’s disease. Front Cell Infect Microbiol. 2021;11:652617.

    [Crossref] [Google Scholar] [PubMed]