Work package 2

This Work Package explored a wide range of metabolites to discover novel pathways underlying the co-occurrence of stroke and Alzheimer's disease with the ultimate goals to identify persons with a potentially modifiable metabolic aetiology. Work Package 2 applied untargeted metabolomics to genes that relate to both stroke and AD to fully cover the physicochemical and biochemical metabolite space.

Identification of blood based metabolic pathways related to genes implicated in stroke and AD

To understand the question how vascular dysfunction and metabolic dysregulation contribute to impairment in cognitive performance, we performed a comprehensive metabolic analyses of circulating metabolites within the Rotterdam Study and Erasmus Rucphen Family study (ERF). Both studies are deeply characterized from cognitive function, dementia including Alzheimer disease and stroke.

We identified 18 metabolites that were significantly associated with general cognitive function. Of the 18 metabolites, 17 were tested for replication in independent cohorts and out of these, we found 15 to be associated with general cognitive ability in the replication cohorts. Thirteen metabolites surpassed the more stringent Bonferroni corrected threshold for significance in the replication.

Combining discovery and replication data, 12 metabolites were associated with higher general cognitive ability and 3 were associated with lower general cognitive ability. The metabolites associated with increased higher cognitive ability include 11 HDL sub-fractions, the most significant being free cholesterol in HDL and docosahexaenoic acid (DHA), an omega-3-fatty acid.

The three metabolites that were associated with lower general cognitive ability include glycoprotein acetyls, glutamine and ornithine.

Next, we examined whether the 15 metabolites associated with general cognitive ability were associated with dementia. We compared 1,949 dementia patients, of whom 1,356 were AD cases, with 23,827 controls. Six metabolites were associated with dementia and three of these were also associated with AD. Free cholesterol in small HDL associated most significantly with a lower risk of dementia. Other metabolites associated with a lower dementia risk were DHA and sub-fractions of medium size HDL particles. Higher glutamine levels were associated with an increased risk of dementia. The association of free cholesterol in small HDL and DHA surpassed the more stringent Bonferroni corrected threshold for significance.

HDL lipoprotein sub-fractions
The specific lipoprotein sub-fractions could point to the specific functions of lipoprotein sub-fractions. HDL fractions are well known to be individually tasked for different functions across lipid metabolism, inflammation, anti-oxidation, and host defence. Additionally specific protein pairs on specific HDL subspecies exist that maintain stable compositions. Previous research reported links between HDL cholesterol profiles and changes in vascular health with plaque accumulation in arteries of the brain, damage to the blood brain barrier  and occurrence of thrombosis. All are possibly leading to progressive vascular brain damage resulting in loss of white matter microstructural organization.
Docosahexaenoic acid (DHA)
DHA levels in blood are highly associated with omega-3 fatty acid intake through diet. It cannot be de novo synthesized in the brain and is therefore actively transported over the blood brain barrier through the MSFD2A. DHA is essential for normal brain development in early life and is frequently associated with cognition. High intake might also be beneficial in late life as DHA and fish oil intake are associated with less Alzheimer disease pathology. The evidence of the attributed beneficial effects of DHA on the brain in literature is inconsistent.
In the brain, glutamine is not only used for energy production and protein synthesis, as in other cells, but is also an essential precursor for biosynthesis of amino acid neurotransmitters. It is involved in the glutamine-glutamate/GABA cycle, a well-studied concept in excitatory signalling in the brain. The cycle involves transfer of glutamine from astrocytes to neurons and neurotransmitter glutamate or GABA from neurons to astrocytes. The leading opinion in the field is that in the brain an excess of glutamate, excitotoxicity, is seen as detrimental and glutamine in the brain as beneficial.
Glycoprotein Acetyls
The measured glycoprotein acetyls is mainly alpha-1-acid glycoprotein (AGP), also called orosomucoid, is an acute phase plasma alpha-globulin glycoprotein. The protein is widely studied and has previously been found to predict 10-year mortality. Increased plasma levels of Glycoprotein acetyls as reaction to various diseases (cancer and inflammatory diseases) or following trauma (surgery) might explain the association with increased mortality, and could partially explain the association with general cognitive ability as chronic diseases decrease cognitive abilities. Another function of AGP is to carry mainly neutrally charged medications in blood, for example anti-depressants. The plasma concentration of AGP is relatively low and there is only one drug-binding site in each AGP molecule [68], leading to lower antidepressant response in higher AGP concentrations.
Ornithine as a non-proteinogenic amino acid is an important intermediate product in arginine degradation and urea cycle. Hyperornithinemia is also the biochemical hallmark of an inherited metabolic disease, hyperornithinemia–hyperammonemia–homocitrullinuria (HHH) syndrome. This disease is clinically characterized by mental retardation whose pathogenesis is still poorly known.

Identification of blood based metabolic pathways related to stroke and AD as clinical outcomes

For the identification of metabolites associated with incident stroke samples from the Rotterdam study were analysed using MS-based metabolomics: three xenobiotics, a metabolite of vitamin C (threonate) and an amino acid showed significant association with incident stroke.

Using NMR-based metabolomics (Brainshake platform) we could identify ten metabolites associated with the risk of stroke. These include amino acid histidine, glycolysis-related metabolite pyruvate, inflammation marker glycoprotein acetyls, cholesterol in HDL2, and lipoprotein subfractions such as cholesterol in medium HDL and triglycerides in medium and large LDL particles which showed association with incident stroke events. Amino acid phenylalanine and HDL subfractions including cholesterol and free cholesterol in large HDL were associated with ischemic incident stroke.

For the association of blood-based metabolites with Alzheimer some metabolites showed interesting changes, but no metabolites were significant.

For the association of blood-based metabolites with dementia, the strongest evidence for association was for 3-methyl catechol sulphate.

Complementary to the previously mentioned platforms we developed an LC-MS/MS method for the analysis of signalling and bioactive lipids. This platform is rather unique, informs about inflammation, immune response, vascular function, oxidative stress and various signaling pathways. The platform is not offered by a commercial metabolomics provider so far, and will allow Leiden University to participate in more relevant studies on AD and stroke studies, and cardiovascular and neurodegenerative diseases in general. The design of the hanging droplet evaporator has been finalized, and the system used for injection into LC-MS. The hanging droplet evaporator allowed to enrich metabolites at least by a factor of 5, and to switch the solvent composition prior to injection allowing to obtain a broader coverage of analytes using one injection and better peak shapes. The variation of the quantification of metabolites using the hanging droplet evaporator as injector for LC-MS/MS was for signaling lipids 3-5%.

These metabolites were chosen based on their (patho)physiological role in (neuro)inflammation, immunology, vascular function and cell differentiation and survival. Various LPAs and oxylipins showed a significant association with AD.

After FDR correction, LPA 16:1 showed significantly elevated levels in AD patients. This metabolite, we have found earlier to predict the conversion from mild cognitive impairment to AD in a Spanish population. Multiple metabolites of oxylipins also showed association with stroke, even in small sample size. Of note is that 2 metabolites, 12-HETE and cLPA 18(1), show a borderline significant association with AD and stroke. Bile acids showed association with general cognition, some of which were also reported in earlier studies.

Signaling lipids were profiled using LC-MS. We found association of increased levels of LPAs, secondary bile acid and LPEs with higher incidence of AD. We have shown that higher levels of LPAs are associated with increased AD biomarkers in CSF. Association of DCA is interesting because several studies have reported the role of gut-microbiota in modifying the levels of this metabolite in blood and its role in cholesterol clearance pathway.

In the association of signaling lipids with magnetic resonance imaging markers, we observed association of increased levels of oxidative stress marker with decreased levels of total hippocampal volume and total brain volume. Moreover, association of increased levels of LPAs with decreased levels total hippocampal volume and total brain volume is consistent with their role in higher AD

Improvement sensitivity of signaling and bioactive lipid platform

For the analysis of metabolites in organ-on-chip models the sensitivity was improved by optimizing the sample preparation and injection. The signal could be increased by a factor of 5-10 of metabolites. This method has been validated and used for the first analysis to study the effect of blood of patients with high degree of inflammation perfused through bloodvessels-on-a-chip.

Identification of common metabolic pathways in stroke patients that associate with increased risk of AD

We compared the metabolites associated with AD genes and those associated with stroke. We first compared the metabolites measured using MS-based metabolomics associating with AD, dementia and cognition and stroke. Glutaconate, quinate and SDMA+ADMA were found to associate with AD and stroke. SDMA+ADMA could relate to a role of endothelial cells in both AD and stroke. 

For the common metabolites using NMR metabolomics associating with AD genes and stroke a few especially HDL fractions were found to overlap.

When comparing metabolites correlating with cognition/dementia and stroke as outcome, and using NMR metabolomics, glycoprotein acetyls, large LDL triglycerides and medium LDL triglycerides were found in common.

Identification of metabolic pathways using existing MRI markers

These analyses were conducted in collaboration with WP3. We used a nuclear magnetic resonance platform to determine concentrations of 143 plasma-based metabolites. Brain and hippocampal size and WMH burden were estimated using visual magnetic resonance imaging ratings or volumetric MRI measures. We meta-analysed the results of linear regression analyses using inverse variance-weighted fixed-effects to determine the association of metabolites with MRI measures.

Glucose levels were found to be associated with smaller brain size, while high levels of two small high-density lipoprotein particles were associated with larger brain size. We found no significant associations for metabolites with hippocampal size. Thirty-four lipids, mainly low-density lipoprotein (LDL) particles, were associated with less WMH in one model, however these associations lost significance the other model we tested. 

Significant results

In summary, a novel platform for metabolites, the signalling and bioactive lipid platform, was developed that covers highly relevant pathways for AD and stroke. First results were already promising, and we expect that this platform will continue to deliver metabolic pathways additional to the once already reported.

  • NMR-based metabolomics have delivered promising metabolic markers and pathways for stroke, Alzheimer, and also common pathways for stroke and Alzheimer.
  • A miniaturized metabolomics platform to analyse blood based metabolic pathways across the blood brain barrier.
  • Identified signaling lipids associated with risk of Alzheimer’s disease
  • Identified signaling lipids associated with with general cognition
  • Identified signaling lipids associated with magnetic resonance imaging markers
  • Finalised the design of the hanging droplet evaporator, and incorporated the system for injection into the LC-MS pipeline for the analysis of signaling and bioactive lipids