Aims

Matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (Timps) are involved in extracellular matrix (ECM) modelling and regulation of basement membrane integrity. A chronic disbalance in the expression and activity of these proteinases and their inhibitors in favor of MMPs can lead to excessive degradation of the ECM, dysfunction of the blood-brain barrier (BBB) and subsequent (micro)bleeds. Cerebral amyloid angiopathy (CAA) is characterized by accumulation of the amyloid-beta protein in the cerebral vessel walls and decreased BBB integrity and may lead to (micro)bleeds. We hypothesized that a disbalance in MMP/Timp expression is associated with CAA and is reflected in the cerebrospinal fluid (CSF). We aimed to quantify the levels of MMPs and Timps in CSF to gain more insight in the pathophysiology of CAA and to establish their potential biomarker value for CAA.

Methods

We used immunoassays to quantify MMP-14 and Timp-1 in CSF. We used a cohort of controls (n=40) and CAA patients (n=28) to study the levels of MMPs and Timps in CSF.

Results

We developed and validated immunoassays for the reliable quantification of MMP-14 and Timp-1 in CSF. Our results showed that the ratio of MMP-14/Timp-1 CSF levels is decreased in CAA patients (0.016 ± 0.005) compared to controls (0.022 ± 0.008; p=0.0007).

Conclusions

Our results suggest that a disbalance in the expression of MMP-14 and Timp-1 in CSF is associated with CAA, and may be a promising biomarker both to identify CAA as well as to obtain more insight into the pathophysiological processes in CAA.

Aims

Current diagnostic criteria for cerebral amyloid angiopathy (CAA) are predominantly based on radiological identification of end-stage evidence of CAA (micro- or macrobleeds. Shotgun proteomics analysis of CAA type I transgenic rodent model tissue, compared to wild-type rats, has presented urokinase plasminogen activator (uPA) as a potential biomarker in CAA. We investigated the potential of uPA in a pilot biomarker discovery study for the diagnosis of CAA by analyzing uPA concentrations in cerebrospinal fluid (CSF) of patients with CAA compared to control patients.

Methods

CSF was obtained from patients with possible or probable CAA (according to the current diagnostic imaging tool, the Boston Criteria) (n=28), and control subjects (n=40). uPA levels in CSF were determined using a uPA Quantikine ELISA (R&D Systems).

Results

The concentration of uPA was significantly increased in the CSF of CAA patients compared to controls with median concentrations (IQR) of 92.0 (76.1 – 109.0) vs. 68.5 (56.6 – 78.4) pg/mL; p<0.001). CSF uPA levels moderately correlated with age at lumbar puncture (r_SP=0.52, p<0.001) and total protein content (r_SP=0.47, p<0.001). Diagnostic accuracy of uPA for CAA yielded an AUC of 0.76. Combining uPA with Aβ40 increased diagnostic accuracy to an AUC of 0.87.

Conclusions

Comparison of CSF levels of uPA shows a significant elevation in CAA patients compared against controls. Diagnostic accuracy of uPA proved relatively high and together with Aβ40 raised accuracy to even higher levels. This reinforces the potential of uPA as a potential CSF biomarker for the diagnosis of CAA patients and incentivizes further research.

Aims

Cerebrovascular dysfunction is now considered an early event in many cases of Alzheimer’s disease (AD). Furthermore, microglia-mediated neuroinflammation is implicated as a key driver of disease progression. Previous work in our lab utilizing a panel of wild-derived mouse strains expressing human mutant forms of APP and PS1 determined inclusion of genetic diversity provides greater human relevance to model AD complexity. One strain, WSB.APP/PS1, exhibited lower numbers of parenchymal amyloid plaques, but significant vascular associated plaque via cerebral amyloid angiopathy (CAA), partnered with robust fibrin leakage.

Methods

Bulk brain and single-cell microglia RNA-sequencing was performed on our wild-derived AD mouse panel at 8 months. Differences in vascular integrity and microglia subtypes were validated using immunofluorescence at multiple ages.

Results

Transcriptional profiling of bulk brain tissue revealed WSB.APP/PS1-specific enrichment of genes relating to vascular health including basement membrane (Col4a2, Col6a4, Mmp2), endothelial function (Ednra), and blood vessel formation (Vegfr and Pdgfr8). Single-cell RNAseq uncovered differences in the proportion and expression profiles of previously identified microglia subclusters. Unlike other strains, WSB.APP/PS1 did not demonstrate significant increases in disease-associated or interferon-responding microglia compared to wild type (WT) counterparts. Interestingly, a population of homeostatic microglia enriched for expression of P2ry12 was only present in WSB but not WSB.APP/PS1 or other WT or AD strains. P2RY12-mediated chemotaxis is critical for closure of the blood brain barrier after injury.

Conclusions

These data suggest the WSB genetic background is ideal to dissect the role of vascular dysfunction and neuroinflammation in CAA in AD and neurodegenerative diseases.

Aims

We aimed to investigate the association of serum vitamin B12, folate, and sulfur amino-acids with cognitive decline in a large sample of community dwelling older adults.

Methods

From the Swedish National Study of Aging and Care in Kungsholmen (SNAC-K), 2900 dementia-free individuals aged 60-102 years at baseline with comprehensive ssessments ncluding Mini-mental state examination (MMSE) scores were recruited. A total of 2202 participants underwent repeated MMSE assessments up to 5 occasions over 15-years. The association of baseline B12, holotranscobalamin, folate, homocysteine, methionine, cystathionine, cysteine, glutathione, and methylation status (defined as methionine/homocysteine ratio) with rate of cognitive decline was examined using linear-mixed-models.

Results

After adjusting for age, sex, education, creatinine, albumin, smoking status, systolic blood pressure, and APOEε4 status, raised baseline homocysteine and cysteine values were associated with faster rate of cognitive decline: for the highest quartile compared with the lowest, β coefficient and standard error (SE) were -0.0058 (0.002) for homocysteine (p=0.004) and -0.0051 (0.002) for cysteine (p=0.014) In contrast, a better methylation status was associated with less MMSE decline over 15 years: β (SE) was 0.0058 (0.002) for the highest quartile compared with the lowest (p=0.004). Furthermore, elevated methionine values tended to slow the rate of cognitive decline (p=0.079). No relationships were found for other sulfur amino acids, vitamin B12 or folate.

Conclusions

Markers of methylation status and raised cysteine values were related to more rapid cognitive decline over 15 years, suggesting that optimizing homocysteine, cysteine, and methionine values may be important in reducing the risk of cognitive decline.

Aims

Alzheimer disease (AD) is the most common cause of dementia. Upon autopsy, most dementia patients show the hallmark lesions of AD: neurofibrillary tangles (NFTs) and neuritic plaques (NPs). Often, patients display other non-AD lesions: Lewy bodies (LBs), diffuse plaques (DPs), cerebral amyloid angiopathy (CAA), arteriolosclerosis, hippocampal sclerosis (HS), and vascular brain injury (VBI). We report here an assessment of lesion contribution to severity of dementia.

Methods

We used the National Alzheimer’s Coordinating Center (NACC) database, identifying 5,272 individuals with a neuropathology assessment. Each lesion was scored as none/mild, moderate, or severe. Scores across all lesions were summed to construct a neuropathology composite score (NPCS). Individual lesions and NPCS were tested for association with cognitive impairment measures including the Mini-Mental State Examination (MMSE) and the sum of the Clinical Dementia Rating (CDR) using linear regression models.

Results

Full models showed most lesions were significantly associated with cognitive function (p<0.05). NFTs, HS, and NPs showed the highest effect sizes while LBs and VBI showed the lowest. When modeling with NPCS, associations with MMSE and CDR were highly significant (p<2E-16).

Conclusions

These data suggest that severity of lesions and comorbidity are highly correlated with dementia severity, even among the non-AD lesions. These data also show that, among the NACC dataset, LBs and VBI are less correlated with cognitive impairment in the presence of other lesions. Together these results confirm the utility of neuropathological collections and suggest mixed-type pathologies may be a major contributor to dementia.

Aims

To explore the associations between kidney disease and dementia, overall and by dementia subtype.

Methods

Our study cohort included 4,313,191 persons ≥40 years of age living in Denmark at some point during 1996-2017. We compared the risks of dementia in persons with and without a history of kidney disease using Cox regression with age as the underlying time scale.

Results

Compared with persons without kidney disease, persons with any kidney disease had a higher risk of dementia overall (hazard ratio [HR] 1.22, 95% confidence interval [CI] 1.19-1.25) and of early-onset (<65 years) dementia in particular (HR 1.67, 95% CI 1.50-1.85). Stronger associations were observed for acute kidney disease (dementia overall: HR 1.44, 95% CI 1.37-1.51; early-onset dementia: HR 2.07, 95% CI 1.71-2.52; late-onset dementia: HR 1.40, 95% CI 1.34-1.48) than for chronic kidney disease (dementia overall: HR 1.13, 95% CI 1.10-1.17; early-onset dementia: HR 1.61, 95% CI 1.37-1.89; late-onset dementia: HR 1.12, 95% CI 1.08- 1.16). Persons with kidney disease had increased risks of vascular dementia (HR 1.43, 95% CI 1.33-1.54) and other/unspecified dementia (HR 1.30, 95% CI 1.26-1.34); in contrast, kidney disease was not associated with the risk of Alzheimer’s disease (HR 0.98, 95% CI 0.94-1.03). Acute kidney disease was particularly strongly associated with the risk of early-onset vascular dementia (HR 3.09, 95% CI 1.65-5.81).

Conclusions

Kidney disease, especially acute kidney disease, was associated with increased risks of dementia, early-onset dementia in particular. Kidney disease was associated with increased risks of vascular dementia but not of Alzheimer’s disease.

Aims

Early detection of cognitive disorders needs reliable markers for the identification of patients with high risk at the preclinical stage of the disease. Coronary artery disease (CAD) is commonly associated with increased risk of mild cognitive impairment (MCI). A promising method of EEG rhythm desynchronization/synchronization can be used to study the neurophysiological basis of cognitive disorders in CAD patients with MCI.

Methods

We analyzed the event-related desynchronization/synchronization of theta- and alpha-rhythms during a modified odd-ball test in 47 male patients with stable CAD aged 45–69 years, in 24 of whom was MCI according to the Petersen’s criteria.

Results

It was found that MCI patients had a reduced theta-desynchronization in response to the target stimulus in the time interval 200–400 ms after the task presentation than the patients without cognitive impairment. These differences were only for the left hemisphere. There were no significant differences in the α-range detected between patients with and without MCI.

Conclusions

It was assumed that cognitive deficits in CAD patients are primarily accompanied by top-down attention control disorders. The changes in the alpha desynchronization/synchronization were common for all CAD patients and apparently reflected selective attention processes.

The reported study was funded by RFBR and Kemerovskaya region, project number 20-415-420005.