Background
The gut microbiota (GM) represents the most densely populated bacterial community colonizing the human body and covers a wide range of functions critical to several aspects of human health. The GM is a key driver of the innate immune system [
1,
2] and is involved in the degradation of macronutrients and production of metabolites [
3]. Intestinal bacteria and their products modulate endothelial cell function and, at the same time, intestinal epithelial cells influence immune responses and shape the microbial composition [
4] and generate a barrier preventing the passage of antigens and bacteria from the gut into the bloodstream [
5].
Alzheimer’s disease (AD) pathology is characterized by the extracellular accumulation of β-amyloid (A), thought to facilitate intracellular cortical deposition and spreading of hyper-phosphorylated tau (T) which in turn drives progressive neurodegeneration (N), ultimately leading to cognitive impairment (CI) [
6]. Clinical and preclinical evidence supports GM involvement in promoting AD onset and progression. Individuals with AD dementia and in the AD preclinical state reported alterations in the GM composition compared with healthy controls [
7‐
11], and amyloid deposition was associated with pro-inflammatory bacterial species in the gut and pro-inflammatory cytokines in the blood [
12]. Preclinical studies showed that GM is necessary for brain amyloid deposition [
13‐
15] and is associated with neurodegeneration [
16‐
18]. While growing evidence suggests that GM may impact cognitive impairment via signaling molecules of the microbiota-gut-brain axis (MGBA) [
19‐
21], few studies in humans have assessed their relationship with markers of the amyloid cascade (ATN scheme) and CI. Furthermore, to the best of our knowledge, the simultaneous evaluation of a large panel of immune and endothelial mediators and their association with GM in AD patients has not been described previously.
We hypothesized that the fecal microbial alterations of patients with CI due to AD are associated with a specific profile of immune and endothelial MGBA mediators in the blood and that the latter is linked to markers of the amyloid cascade. To evaluate the existence and specificity of such a peripheral signature of sporadic AD, we investigated the association of fecal bacterial genera, MGBA mediators, and amyloid cascade markers in cognitively unimpaired persons (CU), patients with CI not due to AD (CI-NAD) and patients with CI due to AD (CI-AD).
Discussion
This study investigated the association of GM with immune and endothelial MGBA mediators and of the latter with markers related to the amyloid cascade in a cohort of cognitively unimpaired persons, patients with cognitive impairment due to AD, and patients with cognitive impairment not due to AD. We report multiple interactions between circulating molecules involved in the MGBA and amyloid aggregation, tau phosphorylation, and neurodegeneration in CI-AD but not in CI-NAD. Our results suggest that peripheral mediators belonging to the MGBA might represent a peripheral signature of AD. Moreover, they strongly support the key role of peripheral mediators in the pathophysiology of AD although the design of the study cannot provide information on the causal chain.
To the best of our knowledge, 10 reports have previously addressed changes in the whole GM in human AD [
7‐
10,
35‐
40] and MCI patients [
9,
10,
35,
38,
39], with often conflicting results. For example,
Blautia was found both decreased (here and [
9,
39]) as well as increased [
7,
35] in AD patients compared to controls. Similarly,
Bifidobacterium was found decreased [
7] or increased (here and [
35,
40]) depending on the studies. Explanations of these discrepancies can be found in the use of different methods for studying the microbiota profile and to the different criteria for defining Alzheimer’s dementia applied in the studies (Additional file
5). Of the 10 published reports, only 3 reported information on ATN markers [
7,
10,
36], and only one used the amyloid positivity finding for the inclusion of patients as in our study [
36]. Moreover, sample collection and storage methods, different bioinformatics pipelines, or even the same pipeline on different operating systems have been reported to impact the relative abundance of the dominant bacterial phyla and genera [
41,
42]. DNA extraction methods and laboratory locations have been shown to lead to up ten-fold and two-fold differences respectively in the relative abundances of specific bacterial genera, respectively [
43].
We confirmed previous human findings demonstrating that blood levels of LPS [
44], soluble CAMs (i.e., VCAM-1, PECAM-1, P-Selectin, E-Selectin, NCAM, ICAM-1) [
27,
45‐
49], IL1β, IL6, TNFα [
30] are increased in AD patients compared with controls. LPS has been reported to induce amyloid and tau aggregation [
50,
51], tau phosphorylation [
51], neurodegeneration [
52], to reduce synaptic plasticity [
53] and to increase microglia density [
50] in mouse and rat brains. Our results are consistent with the notion that LPS translocates from the gut to the bloodstream as consequence of increased intestinal permeability [
54] and co-localizes with amyloid plaques [
55]. The increase in sCAMs reflecting endothelial and vascular damage indicated a general dysfunction of the blood-tissue barriers, including the blood–brain barrier. Furthermore, the upregulation of CAMs involved in the response to infection and controlling leukocyte trafficking is a further indication of widespread vascular inflammation. The latter is thought to promote detrimental processes as it (i) disturbs amyloid-β homeostasis [
56] and (ii) facilitates the passage of antigens and bacteria from the gut into the bloodstream [
5]. In line with previous findings [
57‐
59], AD patients showed increased pTau-181 and both cognitively impaired groups reported increased NfL and performed worse on the MMSE and ADAScog compared to cognitively normal participants.
Communication between the compartments
The next step was to integrate the data from the 3 systems and to validate our hypothesis by using correlation analysis. According to the hypothesis of the study, we provide evidence revealing that a specific GM community contributes to AD pathology and cognitive impairment by increasing intestinal permeability and via bacteria products and inflammatory mediators. Although a multitude of animal and in-vitro studies established associations between the immune and endothelial MGBA mediators and AD features [
18,
60‐
67], this relationship in human AD patients is still under-studied. Our analysis revealed that, in patients with cognitive impairment due to AD, (i) sPECAM-1, sP-, sE-Selectins, and IL6 were associated with amyloid and tau pathology; (ii) sP-Selectin, sNCAM, and the decrease of IL10 with neurodegeneration; and (iii) LPS and IL1β with cognitive impairment. Conversely, in patients with cognitive impairment not due to AD, no immune and endothelial MGBA mediators were associated with amyloid and tau pathology and neurodegeneration and many of them were associated with cognitive performance (sPECAM-1, sP-Selectin, IL1β, and decreased of IL10), suggesting that the link between GM alterations and cognitive impairment in non-AD patients likely involved different pathways than AD. While the CI-NAD is a clinically and biologically heterogeneous group, we observed significant associations between GM and MGBA vascular mediators. This, together with the higher occurrence of hypertension within this group, suggests that our CI-NAD might be enriched by patients with vascular disease.
Limitation
Although this study is one of the few that includes the amyloid marker for subject inclusion, it has also limitations. First of all, this is a small observational study and should be considered with caution due to participant selection and confounding biases. Replication of the results in an independent, larger cohort is needed, which would allow the evaluation of a broader panel of MGBA markers and the use of more complex statistical models. Although our analysis is limited to selected innate immunity and endothelial mediators, gut-brain communication is much more complex and involves, among others, mediators of cellular immunity and circulating molecules directly produced by the gut microbiota such as gut peptides, neurotransmitters, and metabolites. Of note, despite evidence suggesting that the bacterial metabolites’ short-chain fatty acids (SCFAs) have neuroprotective effects, recent studies showed that SCFAs supplementation could increase AD pathology in several AD animal models, possibly through immune cells activation [
18,
61,
68]. This suggests that the involvement of SCFAs in microbiota-gut-brain interactions is much more complex than initially thought and that further studies aimed at establishing their relationship are necessary. Furthermore, here we used non-parametric correlations to study the association of markers belonging to the 3 compartments. This, together with the cross-sectional design, prevents us from drawing causal inferences between GM and MGBA marker alterations and AD pathological changes and symptoms onset. Longitudinal studies investigating also the earlier stages of the disease (i.e., amyloid positive and cognitive intact persons) are required to elucidate whether the AD-related microbiota alterations are upstream or downstream to brain AD pathological changes. Last, some of the studied markers were not available for all study participants (Additional file
6).
Conclusion
This study confirms the presence of a peripheral signature of Alzheimer’s disease featuring microbiota-gut-brain axis markers. The results suggest that the gut microbiota exerts its action on the brain at least in part by modulating endothelial cell function and the levels of circulating inflammatory and microbial products. LPS in the blood and the upregulation of soluble CAMs involved in endothelial damage and vascular changes suggest the presence of a more permeable intestinal barrier in AD. It is therefore plausible that GM products and inflammatory modulators (not only those measured here but also bacteria, fungi, and immune cells) translocate into the bloodstream, reach the brain and trigger the amyloid cascade. The association of LPS and several sCAMs with AD pathology and neurodegeneration provides a new line of evidence for a possible direct link between MGBA and brain pathological changes and offers new biomarkers and treatment targets for AD.
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.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.