Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 7, 318

Secretory Products of the Human GI Tract Microbiome and Their Potential Impact on Alzheimer's Disease (AD): Detection of Lipopolysaccharide (LPS) in AD Hippocampus


Secretory Products of the Human GI Tract Microbiome and Their Potential Impact on Alzheimer's Disease (AD): Detection of Lipopolysaccharide (LPS) in AD Hippocampus

Yuhai Zhao et al. Front Cell Infect Microbiol.


Although the potential contribution of the human gastrointestinal (GI) tract microbiome to human health, aging, and disease is becoming increasingly acknowledged, the molecular mechanics and signaling pathways of just how this is accomplished is not well-understood. Major bacterial species of the GI tract, such as the abundant Gram-negative bacilli Bacteroides fragilis (B. fragilis) and Escherichia coli (E. coli), secrete a remarkably complex array of pro-inflammatory neurotoxins which, when released from the confines of the healthy GI tract, are pathogenic and highly detrimental to the homeostatic function of neurons in the central nervous system (CNS). For the first time here we report the presence of bacterial lipopolysaccharide (LPS) in brain lysates from the hippocampus and superior temporal lobe neocortex of Alzheimer's disease (AD) brains. Mean LPS levels varied from two-fold increases in the neocortex to three-fold increases in the hippocampus, AD over age-matched controls, however some samples from advanced AD hippocampal cases exhibited up to a 26-fold increase in LPS over age-matched controls. This "Perspectives" paper will further highlight some very recent research on GI tract microbiome signaling to the human CNS, and will update current findings that implicate GI tract microbiome-derived LPS as an important internal contributor to inflammatory degeneration in the CNS.

Keywords: 42 amino acid amyloid-beta (Aβ42) peptide; Alzheimer's disease (AD); Bacteriodetes fragilis (B. fragilis); Escherichia coli (E. coli); lipopolysaccharide (LPS); microbiome; small non-coding RNAs (sncRNAs); thanatomicrobiome.


Figure 1
Figure 1
Like other Gram-negative bacilli, the gastrointestinal (GI) tract abundant Bacteroides fragilis (micrograph of B. fragilis shown; original photo courtesy of Rosa Rubicondior; ( is capable, when stressed, of releasing a broad spectrum of highly neurotoxic, pro-inflammatory and potentially pathogenic molecules; these comprise five major classes of secreted molecules and include endotoxins, exotoxins, lipooligosacahride (LOS) and lipopolysaccharide (LPS), amyloids, and small non-coding RNAs (sncRNA). For example, the human GI tract-abundant B. fragilis secretes the endotoxin fragilysin and B. fragilis LPS (BF-LPS) both of which have been shown recently to be strongly pro-inflammatory and extremely neurotoxic toward human CNS neurons in primary culture (Li et al., ; Lukiw, 2016). While the phyla Bacteriodetes (~20% of all GI tract bacteria), Firmicutes (~80% of all GI tract bacteria), Actinobacteria, Proteobacteria, and Verrumicrobia (together, typically ~4% of all GI tract bacteria), are the most common microbes in the human GI tract microbiome it should be kept in mind that other microbes including fungus, protozoa, viruses, and other commensal microorganisms may also contribute neurotoxic exudates which are highly toxic and detrimental to the homeostasis of CNS neurons.
Figure 2
Figure 2
(A) human brain temporal lobe neocortex [N = 6 control and 6 sporadic AD cases; quantified in (B)]; and (C) hippocampus (N = 2 control and N = 4 AD cases; quantified in (D)] were analyzed for LPS against β-actin abundance in the same sample (using anti-E. coli LPS; cat # ab35654 from Abcam, Cambridge UK and anti-β-actin cat # 3700 from Cell Signaling, Danvers MA, USA) using Western analysis as previously described by our group (Bhattacharjee et al., ; Zhao et al., 2016); all AD and control tissues were analyzed in a RNA-analysis clean room facility; all control and AD tissues were age- and gender-matched; there were no significant differences between the age (control 72.9 ± 8.1 years, AD 74.2 ± 9.1 years), gender (all female), PMI (all tissues 3.5 h post-mortem or less), RNA quality or RNA yield between each of the two groups; LPS abundance was found to be on average over two-fold as abundant in AD when compared to age-, gender, and PMI-matched control neocortex in 6 of 6 cases; LPS was found to be on average three-fold as abundant in AD when compared to age-, gender, and PMI-matched control hippocampus in 3 of 4 cases; some advanced AD hippocampal samples exhibited up to a 26-fold increase in LPS over age-matched controls (C, LPS in control lane 2 vs. AD lane 5); because one major source of LPS are Gram-negative bacteria of the human GI tract (predominantly B. fragilis and E. coli), this suggests that in vivo intensely pro-inflammatory LPS species may be able to “leak” through at least two major biophysiological barriers—the GI tract barrier and the BBB—to access brain compartments (see Devier et al., ; Halmer et al., ; Choi et al., ; Minter et al., ; Montagne et al., ; Richards et al., ; Soenen et al., ; van de Haar et al., ; Zhan and Davies, ; Zhao et al., ; Varatharaj and Galea, 2017). Unpublished work from this laboratory further indicates the positive detection of LPS in 36 of 36 AD tissues sampled from the superior temporal lobe neocortex in aged individuals (age range 66–79 yr; see Table 1 in Cui et al., 2010). Another recent investigation reports the finding of LPS in gray matter (temporal lobe) and white matter (frontal lobe) of the AD brain (Zhan et al., 2016). Together these data also suggest that neurotoxic cocktails secreted by multiple GI tract microbes or other microbial species (Figure 1) may have considerable potential to support intense pro-inflammatory signaling within the CNS especially over the course of aging when barriers become more “leaky” (Hill and Lukiw, ; Keaney and Campbell, ; Montagne et al., ; Choi et al., ; Köhler et al., ; Minter et al., ,; Richards et al., ; van de Haar et al., ; Zhan and Davies, ; Varatharaj and Galea, 2017); (B) and (D) represent the mean plus one standard deviation of that mean; *p < 0.05, **p < 0.01 ANOVA; NC, negative control using a control murine brain extract (strain C57BL/6J); in (B) and (D) a dashed horizontal line at 100 is included for ease of comparison.

Similar articles

See all similar articles

Cited by 39 PubMed Central articles

See all "Cited by" articles


    1. Alan G., Sarah J. P. (2012). Microbes as forensic indicators. Trop. Biomed. 29, 311–330. - PubMed
    1. Alkasir R., Li J., Li X., Jin M., Zhu B. (2016). Human gut microbiota: the links with dementia development. Protein Cell 8, 90–102. 10.1007/s13238-016-0338-6 - DOI - PMC - PubMed
    1. Alonso R., Pisa D., Marina A. I., Morato E., Rábano A., Carrasco L. (2014). Fungal infection in patients with Alzheimer's disease. J. Alzheimers Dis. 41, 301–311. 10.3233/JAD-132681 - DOI - PubMed
    1. Andreeva T. V., Lukiw W. J., Rogaev E. I. (2017). Biological basis for amyloidogenesis in Alzheimer's disease. Biochemistry 82, 122–139. 10.1134/S0006297917020043 - DOI - PubMed
    1. Asti A., Gioglio L. (2014). Can a bacterial endotoxin be a key factor in the kinetics of amyloid fibril formation? J. Alzheimers Dis. 39, 169–179. 10.3233/JAD-131394 - DOI - PubMed

Publication types

MeSH terms