Abstract
Background
Intestinal bacteria-derived molecules, such as lipopolysaccharide (LPS) produced by Gram-negative bacteria, can translocate into the bloodstream through the gut wall, contributing to inflammation and neurodegeneration via oxidative stress mechanisms. NADPH oxidase-2 activation and superoxide anion production play a key role in this process, particularly in conditions like Alzheimer's disease (AD), where gut permeability is often altered. This study hypothesized that modulating gut microbiota with the probiotic Escherichia coli Nissle 1917 (ECN) could mitigate LPS translocation and its associated inflammatory effects.
Objective
To evaluate the effect of daily ECN administration on serum LPS levels in elderly AD patients.
Methods
A randomized, double-blind, placebo-controlled trial was conducted with 40 mild AD patients, with 39 completing the study (20 ECN, 19 placebo). Participants received ECN (2.5–25 × 10^9 CFU/capsule) or placebo for six weeks. The serum activity of soluble NOX2-dp (sNOX2-dp), hydrogen peroxide (H2O2) production, tumor necrosis factor (TNF)-α levels and LPS was evaluated, while serum zonulin levels were measured to assess gut permeability.
Results
The ECN group showed significant reductions in sNOX2-dp (−21%), H2O2 (−27%), TNF-α (−18%), LPS (−15%), and zonulin (−35%), along with improved Mini-Mental State Examination (MMSE) scores. No significant changes were seen in the placebo group. Mixed ANOVA showed significant time-by-treatment interactions for zonulin (p = 0.04) and MMSE (p < 0.001). Changes in LPS correlated with changes in zonulin (Rs = 0.408, p = 0.011).
Conclusions
ECN may strengthen gut barrier function, reduce endotoxemia, and attenuate inflammation in AD, though larger studies are needed to confirm these findings.
Keywords
Introduction
Alzheimer's disease (AD) is one of the primary causes of dementia worldwide, accounting for 60–70% of all cases and affecting over 55 million people globally, with numbers rising steadily in industrialized countries. 1 However, despite its prevalence, the precise pathogenic mechanisms of AD remain only partially understood.
In recent years, research has focused on the role of inflammatory processes and oxidative stress as key elements in the progression of neurodegenerative diseases, including AD. 2 At the same time, there is growing attention to the role of the gut microbiota in modulating neuroinflammation and its influence on neurodegenerative diseases.3,4 This perspective broadens our understanding of AD and suggests new pathways for treatment and prevention of this debilitating condition.
The so-called “gut-brain-microbiota axis” describes the complex, bidirectional interaction between the central nervous system (CNS) and the gut microbiota. 5 Numerous studies have confirmed that this connection affects not only metabolic and immune health but also the nervous system, facilitating communication and regulation of inflammatory and oxidative processes involved in neurodegenerative diseases. 5
Recent studies have highlighted how alterations in gut microbiota can contribute to neuroinflammation, a recurrent feature in neurodegenerative diseases like AD.3,5 Current hypotheses suggest that molecules derived from intestinal bacteria, such as lipopolysaccharide (LPS) produced by Gram-negative bacteria, can enter the bloodstream through the gut wall (especially in the case of “leaky gut,” or increased intestinal permeability) and contribute to inflammatory responses in the brain, triggering neurodegenerative processes. 5 Specifically, LPS is associated with increased oxidative stress and neuroinflammation, factors that can accelerate AD progression. 6
The activity of the enzyme NOX2, responsible for the production of reactive oxygen species (ROS), has been linked to various forms of neuroinflammation and cognitive decline, especially in experimental models of AD and Parkinson's disease. 7 Studies in murine models of AD have shown that increased NOX2 activation is associated not only with higher oxidative stress but also with the formation of amyloid plaques, a hallmark of the disease.7,8 Likewise, NOX2 activation has been associated with elevated amyloid-β (Aβ) levels and neuronal degeneration, contributing to the progressive loss of cognitive functions.7–9 Furthermore, our research group demonstrated that LPS and NOX-2 activity is increased in AD. 3
The growing understanding of the link between gut microbiota and neurodegeneration has opened new therapeutic perspectives, particularly through the use of probiotics and dietary interventions to modulate gut microbiota composition.10,11 Probiotics, beneficial bacteria administered to improve gut microbiota balance, have been associated with reduced systemic inflammation and improved markers of oxidative stress, with positive effects on cognitive health. 11 A recent meta-analysis confirmed that regular probiotic consumption may improve cognitive performance in patients with AD, indicating that their antioxidant and anti-inflammatory actions may slow the progression of neurodegeneration. 12
In a murine model of autoimmune encephalitis, oral administration of the Escherichia coli Nissle 1917 (ECN) probiotic demonstrated a reduction in neuroinflammation, suggesting a potential therapeutic role for this probiotic strain in conditions associated with inflammatory diseases of the CNS or other immunological disorders linked to leaky gut. 13 The probiotic ECN contains purified LPS with a shortened carbohydrate chain, which some studies suggest may be responsible for its anti-inflammatory effects.14,15
We hypothesized that modulating the gut microbiota through the oral administration of the single-strain ECN probiotic could help reduce LPS translocation into the bloodstream in the presence of altered intestinal permeability, a condition frequently observed in elderly individuals. This effect could decrease systemic NOX-2 enzyme activation and lower oxidative stress levels, which contribute to neuroinflammation. Therefore, we conducted a randomized, double-blind, controlled clinical trial to investigate the effects of ECN on serum LPS levels and systemic oxidative activity in a population of elderly subjects with AD.
Study objectives
The primary objective of this study is to assess the effects of daily administration of ECN, a single-species probiotic, on baseline LPS serum levels in a population of elderly patients with AD.
Secondary objectives
To analyze the effect of ECN probiotic treatment on baseline serum levels of sNOX2-dp.
To evaluate the effect of ECN probiotic treatment on baseline serum levels of zonulin, a molecule that modulates tight junctions, with its overexpression being associated with increased intestinal permeability.
To investigate the effect of ECN probiotic treatment on the following clinical outcomes: cognitive performance measured by the Mini-Mental State Examination (MMSE).
Methods
Study design
We conducted a randomized, double-blind, placebo-controlled clinical trial with the ECN probiotic containing Escherichia coli strain Nissle 1917 (2.5–25 × 10^9 CFU/capsule) versus placebo (42% maltodextrin, 12% talc, 17% gelatin, 3% water), both pre-randomized by the nutraceutical company Ca.Di. Group S.p.A. The intervention lasted six weeks.
The study was divided into two phases. In the first phase, patients were recruited, data were collected before and after the six-week intervention, and the probiotic was distributed according to a consecutive randomization list. During the second phase, the Sapienza University Research Laboratory, affiliated with the Department of Clinical, Internal, Anesthesiological, and Cardiovascular Sciences, performed blinded blood analyses for both the probiotic and placebo groups. Finally, after unblinding the randomization list, the statistical analysis of the variables was performed.
Participants
The study included patients aged 60–90 years, diagnosed with Alzheimer's dementia, who were referred to the Geriatric Evaluation Unit, Dementia and Cognitive Disorders Center at Policlinico Umberto I in Rome, between March and September 2023. All participants had a diagnosis of probable Alzheimer's dementia based on NIA-AA criteria, with mild severity according to the Clinical Dementia Rating (CDR) scale.
Exclusion criteria were: MMSE < 20/30; CDR > 1; dysphagia; severe hepatic or renal failure; uncontrolled diabetes mellitus; chronic infections; active oncological disease; active chronic inflammatory disease; acute illness; body mass index (BMI) < 22 or > 30; use of antioxidant supplements or probiotics in the six weeks prior to the study; consumption of yogurt, kefir, or other fermented foods; and use of antibiotics.
Intervention
Participants were divided into two groups (probiotic and placebo) according to a randomized sequence, in a double-blind model (see flow-chart in the supplementary data). Each patient was asked to take the ECN probiotic (or placebo) daily in capsule form, with a dosage of 640 mg/day, divided into one capsule before lunch and one capsule before dinner. The placebo was indistinguishable from the probiotic in packaging, shape, and appearance. The capsules containing ECN or placebo were prepared by CADI Group Company (Rome, Italy) and provided to the study's medical staff in a blinded manner, with coded labeling unknown to both the doctors and patients involved in the study.
At both baseline and after six weeks of intervention, all participants underwent a comprehensive geriatric assessment, including an accurate medical history and physical examination. The following parameters were collected: anthropometric and sociodemographic data (age, sex, weight, height, body mass index, blood pressure, education), dietary habits and bowel movements, medication use, cardiovascular risk factors, and comorbidities. Additionally, neurocognitive, emotional-behavioral, and functional independence profiles were assessed using the following tests: MMSE, 16 Activity Daily Living (ADL), 17 and Instrumental Activity Daily Living (IADL), 18 Severity Index (SI). 19
Lastly, patients (and their caregivers) were asked to keep a food diary, noting the patient's diet for three randomly chosen days each week during the first, third, and sixth weeks of intervention. They were also requested to report any antibiotic use.
The medical staff monitored patient adherence through pill/dose count, supported by caregivers. No patients were hospitalized during the study period. The pill/dose count adherence rate exceeded 90%. All patients and caregivers were instructed to take the prescribed capsules daily and return any remaining pills to the clinic at their follow-up visit.
Blood samples
The blood samples, collected from each patient, were centrifuged at 300 g for 10 min, and then used for measuring LPS, zonulin, sNOX2-dp, hydrogen peroxide (H2O2), and tumor necrosis factor (TNF)-α using an ELISA method.
Serum LPS levels were determined using an ELISA kit from (Cusabio; CSB-E09945 h), as previously described. 20
Serum zonulin levels were quantified using an ELISA kit (Elabscience; E-EL-H5560) as previously described. 20
Specifically, serum soluble NOX2-derived peptide (sNOX-dp), a marker of NADPH oxidase activity, was performed using a laboratory-patented ELISA method. 21
H2O2 was measured by a colorimetric assay according to manufacturer's instruction (Abcam; ab102500). The values were expressed as μM and intra- and inter-assay CV were both <10%. TNF-α concentration was measured using a commercially ELISA kit (Abcam; ab181421) values were expressed as pg/ml and the intra-assay and inter-assay coefficients of variation were <10%.
Ethical considerations
This study protocol was designed and conducted in compliance with the principles outlined in the Good Clinical Practice (GCP) guidelines, where applicable, and in accordance with the Declaration of Helsinki and current guidelines for interventional studies. The research was approved by the Ethics Committee of Sapienza University, Rome (Rif. 6808 Prot. 0528/2022).
Informed consent
All enrolled subjects (and/or their caregivers) were informed and provided written informed consent for the processing of personal data for scientific research purposes.
Sample size determination
The sample size was calculated based on a two-tailed Student's t-test for independent groups, considering the following:
A difference (δ) of >6 pg/ml in LPS levels between the probiotic and placebo groups, with a standard deviation (SD) of 5.7 pg/ml.
Sample size calculation was computed to achieve a 90% chance of detecting a significant reduction in the primary outcome measure, deemed significant at a 5% level, from 34 pg/ml in the control group to 28 pg/ml in the experimental group.
The error probability (α) was set at 0.05, with a power of 0.90 (1-β). The minimum sample size was n = 19 patients/group.
Randomization
Sealed Envelope Ltd. 2024. Create a blocked randomization list. [Online] Available from: https://www.sealedenvelope.com/simple-randomiser/v1/lists.
Statistical analysis
Statistical analyses were performed using SPSS 25.0 software for Windows (SPSS, Chicago, IL, USA). Data are presented as means ± standard deviations (SD) for normally distributed variables and as frequencies and percentages for categorical variables.
The study data were analyzed to evaluate the treatment and time effects using mixed ANOVA for each variable of interest. A between-group factor (probiotic treatment versus placebo) and a within-group factor (pre-treatment versus post-treatment) were considered. Paired comparisons were adjusted using the Bonferroni correction and replicated by non-parametric test (Wilcoxon test). The Spearman correlation test was used for bivariate analysis.
Results
Forty patients with mild AD were consecutively recruited (see Supplemental Figure 1). One patient dropped out of the study for personal reasons, so a total of 39 out of 40 patients completed the trial (20 in the probiotic group and 19 in the placebo group). No significant adverse reactions were observed during the study period.
The clinical characteristics of patients in the probiotic and placebo groups are reported in Table 1. No significant differences were found between the two groups in terms of age, gender, BMI, education level, smoking, polypharmacy, or comorbidity degree (Table 1).
Characteristics of participants in the probiotic and placebo groups.
SD: standard deviation; ADL: basic activities of daily living; IADL: instrumental activities of daily living; CIRS: Cumulative Illness Rating Scale; SI: Severity Index; CI: Comorbidity Index.
Pairwise comparisons demonstrated that, compared to baseline values, after 6 weeks of ECN assumption, sNOX2-dp levels significantly decreased from 30.0 ± 7.7 to 23.6 ± 6.4 pg/ml (−21%, p = 0.01), H2O2 from 36.6 ± 10.8 to 26.7 ± 8.1 µM/L (−27%, p = 0.001), TNF-α from 37.4 ± 6.1 to 30.5 ± 8.5 pg/ml (−18%, p = 0.004), LPS from 34.3 ± 6.8 to 28.9 ± 4 pg/ml (−15%, p = 0.01), and zonulin levels decreased from 2.0 ± 0.7 to 1.3 ± 0.4 ng/ml (−35%, p = 0.001)) (Figure 1). Additionally, MMSE increased after ECN assumption from 22.5 ± 1.4 to 23.8 ± 1.3 (p = 0.03) (Figure 1).

Effects of 6-week treatment (mean ± Standard Error) with placebo (dashed line) or ECN probiotic (solid line) on sNOX2-dp (A), LPS (B), zonulin (C), and Mini-Mental State Examination (MMSE, D) in patients with Alzheimer's disease. °Significant interaction at mixed ANOVA analysis.

In patients with Alzheimer's disease, there is an increased translocation of LPS (lipopolysaccharides) from the intestine into the systemic circulation. LPS binds to the TLR4 receptor, activating NOX2, which leads to an increase in free radicals, endothelial dysfunction, and platelet activation. Endothelial dysfunction at the level of the blood-brain barrier may facilitate the passage of LPS, which can activate microglia and promote the entry of activated platelets, thereby enhancing amyloid-β deposition. This process may contribute to neuroinflammation and neurodegeneration.
No significant effect was observed after 6 weeks of placebo assumption on sNOX2-dp from (29.2 ± 6.2 to 28.1 ± 9.0 pg/ml), LPS (33.7 ± 6.5 to 31.3 ± 6.2 pg/dl), zonulin (from 1.99 ± 0.6 to 1.83 ± 0.65 ng/ml), and MMSE (from 22.9 ± 1.3 to 22.5 ± 1.6).
Mixed ANOVA analysis revealed a significant interaction between time and type of administration between the treatments was found for zonulin (F = 4.37, p = 0.04), H2O2 (F = 7.42, p = 0.01), TNF-α (F = 5.59, p = 0.02), and MMSE (F = 22.0, p < 0.001). Conversely, no significant interaction was found between time and type of administration for LPS (F = 0.969, p = 0.332) and NOX-2 (F = 2.3, p = 0.135).
Simple linear regression analysis showed in the overall population that Δ of LPS was significantly associated with Δ of zonulin (Rs: 0.408, p = 0.011).
Discussion
This study showed that ECN decreases serum LPS and NOX2 levels in AD patients (Figure 2).
Previous investigations have identified gut dysbiosis in patients with AD, introducing the concept of the “gut-brain axis”, a sort of neuro-inflammatory process derived from the gut.22,23 Gram-negative bacteria in the gastrointestinal tract, such as Bacteroides fragilis and Escherichia coli (EC), release LPS, which exert pro-inflammatory effects on neurons. 6 Animal studies have demonstrated that systemic administration of LPS exacerbates neuroinflammation and promotes progressive neurodegeneration.6,24 Thus, lowering LPS may represent an interesting therapeutic perspective for AD treatment. In the present study we tested the hypothesis that a probiotic treatment counteracting EC could improve gut permeability and results in LPS lowering.
The data obtained from our study indicate a significant decrease in LPS levels after assumption of ECN in AD patients. The mechanism through which the reduction in LPS could be closely tied to an improvement in intestinal permeability, suggesting the strengthening of the tight junctions in the intestinal barrier. 25 Zonulin plays a key role in this process: it is a regulatory protein that modulates the tight junctions between intestinal epithelial cells.26,27 Elevated zonulin levels increase intestinal permeability, allowing molecules like LPS to pass into the bloodstream.26,27 Lowering zonulin levels is thus associated with a more intact and functional intestinal barrier.26,28
The reduction in LPS explains a decrease in oxidative stress, as, upon binding to its receptor TLR4, LPS increases Nox2 activity so enhancing the formation of ROS.25,29–31
In this context, it is interesting it is noteworthy that neuronal microglia express Toll-like receptor 4 (TLR4), a key regulator of inflammation.32,33 This recognition plays a critical role in the pathophysiological development of AD.32,33 Activation of TLR4 by LPS has been shown to stimulate microglial activation, leading to increased neuronal damage in the APP/PS1 transgenic mouse model. 32 Another compelling hypothesis could be linked to the activation of NOX2 in platelets and endothelial cells. Endothelial dysfunction induced by NOX2 activation may lead to blood-brain barrier impairment, facilitating neuroinflammation through increased platelet infiltration. Platelets play a critical role as they can produce Aβ fragments and deposit them in brain tissue, thereby contributing to neuroinflammation as well as the development and progression of AD. 34 In this regard, ex vivo studies on platelets from human subjects have demonstrated that platelets, when stimulated with LPS, are capable of releasing Aβ. 35 Furthermore, studies in murine models have shown that NOX2-derived ROS contribute to vascular dysfunction induced by amyloid-β protein precursor overexpression and to the associated behavioral decline. 36
Using an assay that allows to measure Nox2 in the serum, 21 we could demonstrate a significant reduction of Nox2 activity after ECN intake, that paralleled the reduction of LPS, so suggesting that LPS lowering may limit Nox2-derived ROS production, potentially alleviating oxidative damage in microglia and brain tissue in AD patients. Further studies are required to validate this pathophysiological hypothesis.
In accordance with previous studies, we observed that treatment with the ECN probiotic improved cognitive status as assessed by the MMSE.37–39 However, the low number of treated patients and the slight, albeit non-significant, decrease in MMSE scores observed in the control group may have amplified the difference with the increased MMSE scores seen in the group of patients treated with the ECN probiotic. Therefore, we believe that further studies with a larger sample size are necessary to evaluate this potentially interesting effect.
The study has limitations and implications. Firstly, we did not verify at the cellular level whether the TLR4 receptor is upregulated or whether there are significant changes of its expression. Secondly, it was not determined whether Escherichia coli Nissle directly influenced the composition of the gut microbiota and eventually reduced gut dysbiosis.
Additionally, ECN contains a purified form of LPS with anti-inflammatory properties. If degraded to monophosphoryl form, LPS may still bind to TLR4 but acts as a TLR4 antagonist. 25 Therefore, we cannot exclude the possibility that, at the intestinal level, TLR4 inhibition by ECN may be a ‘primum movens’ of a sequence of events leading to reduced LPS translocation into the circulatory system.
Finally, the direct effect at the brain level was not evaluated; instead, only the systemic indirect effect of low circulating levels of LPS on NOX2 activation was assessed.
In conclusion, ECN appears to improve intestinal permeability and reduce low grade endotoxemia and systemic NOX2 activation; thereby, it could decrease the risk of potential inflammatory insults, even at the cerebral level. While preliminary data support this hypothesis, the limited sample size prevents any definitive conclusions, and further research is required.
Supplemental Material
sj-pdf-1-alz-10.1177_13872877251345157 - Supplemental material for Effect of the probiotic Escherichia coli Nissle 1917 on serum levels of NADPH oxidase-2 and lipopolysaccharide in patients with Alzheimer's disease
Supplemental material, sj-pdf-1-alz-10.1177_13872877251345157 for Effect of the probiotic Escherichia coli Nissle 1917 on serum levels of NADPH oxidase-2 and lipopolysaccharide in patients with Alzheimer's disease by Lorenzo Loffredo, Alba Rosa Alfano, Evaristo Ettorre, Giovambattista Desideri, Roberto Carnevale, Maurizio Forte, Vittorio Maglione, Simona Bartimoccia, Valentina Castellani, Chiara Maria Totè, Pasquale Pignatelli, Francesco Violi6 and Neurodegenerative Disease study group in Journal of Alzheimer's Disease
Footnotes
ORCID iDs
Ethical considerations
This study was approved by the Ethics Committee of Sapienza University, Rome (Rif. 6808 Prot. 0528/2022).
Consent to participate
All enrolled subjects (and/or their caregivers) were informed and provided written informed consent for the processing of personal data for scientific research purposes.
Author contributions
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research leading to these results has received funding from the European Union— NextGenerationEU through the Italian Ministry of University and Research under PNRR—M4C2- I1.3 Project PE_00000019 “HEAL ITALIA” to Roberto Carnevale CUP B23D22000580004, IRCCS Neuromed Pozzilli (IS). The views and opinions expressed are those of the authors only and do not necessarily reflect those of the European Union or the European Commission. Neither the European Union nor the European Commission can be held responsible for them.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data availability statement
Data will be made available on request.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
