Cerebrospinal Fluid Protein Research Articles

Learning from cerebrospinal fluid drug-resistant HIV escape-associated encephalitis: a case report

BackgroundIn the era of antiretroviral therapy (ART), central nervous system (CNS) complications in patients with human immunodeficiency virus (HIV) infection are sometimes associated with cerebrospinal fluid (CSF) viral escape. Here, we reported a case of persistent CNS viral escape with recurrent symptomatic encephalitis, which had ultimate stabilization achieved by a combination of ART adjustment and corticosteroids.Case presentationA 27-year-old man with HIV infection complained of recurrent headaches during the last year. His magnetic resonance imaging (MRI) presented diffused bilateral white matter lesions, and laboratory tests confirmed elevated CSF protein level, lymphocytic pleocytosis, and detectable CSF HIV RNA (774 copies/mL). Plasma HIV RNA was well suppressed with tenofovir, lamivudine, and lopinavir/ritonavir. Prednisone 60 mg once daily was initiated to reduce intracranial inflammation, followed by a good clinical response, with CSF HIV RNA still detectable (31.1 copies/mL). During the gradual tapering of prednisone, his headache relapsed, and booming viral loads were detected in both CSF (4580 copies/mL) and plasma (340 copies/mL) with consistent drug-resistant mutations. Thereupon, prednisone was resumed and the ART regimen was switched to zidovudine, lamivudine, and dolutegravir according to drug resistance tests. Persistent clinical recovery of symptoms, neuroimaging, and laboratory abnormalities were observed in the follow-up visits.ConclusionCSF and plasma HIV RNA and further drug resistance tests should be monitored in HIV-infected patients with neurologic symptoms, as opportunistic infections or tumors can be ruled out. ART optimization using a sensitive regimen may be crucial for addressing CSF viral escape and the related encephalitis.

Localized Scleroderma Causes Peripheral Neuropathies and Persistent Elevated Cerebrospinal Fluid Protein: A Case Report

We present a case of peripheral neuropathies and persistent elevated cerebrospinal fluid protein that is attributed to localized scleroderma. This patient’s has been miss diagnosed for more than 14 years, and the diagnosis of localized scleroderma was finally set up by skin biopsy, providing further evidence for the value of skin biopsy in the clinical practice in peripheral neuropathies.

Associations Between TREML2 Gene Variants and Alzheimer's Disease: Biomarkers, Neuroimage, and Cognition.

Recent genetic research identified a protective factor against late-onset Alzheimer's disease (AD) in Caucasians, a variant called rs3747742-C in the TREML2 gene. However, the roles of other TREML2 variants in AD have not been fully explored. We conducted a focused analysis of 16 TREML2 variants, examining their connection to AD by studying their correlation with cerebrospinal fluid (CSF) proteins, neuroimage, and cognition in the Alzheimer's Disease Neuroimaging Initiative database (ADNI). A multiple linear regression model was utilized to estimate potential associations between TREML2 genotypes and various endophenotypes in the entire ADNI sample at baseline, with age, gender, years of education, and APOE ɛ4 status included as covariates. To examine changes in clinical outcomes over time, linear mixed-effects models were employed. We found that the SNP rs17328707-A was associated with higher ADNI-VS scores, smaller ventricles, and larger middle temporal volume at baseline. The SNP rs6915083-G was linked to lower CSF t-tau and p-tau levels, and higher CSF Aβ levels. The SNP rs9394766-G was associated with a smaller hippocampus and larger ventricles at baseline. In longitudinal cohorts, the rs6915083-G SNP was associated with changes in ADNI-MEM and ADNI-EF scores, as well as the rate of hippocampal and middle temporal atrophy. Our findings reveal that TREML2 gene variants have different effects on AD. Two variants are protective, while one may be a risk factor. This enhances our understanding of AD genetics and could guide future research and personalized treatments.

Implementation of an ultra-sensitive microwell-based electrochemical sensor for the detection of Alzheimer’s disease

Alzheimer's Disease (AD) is one of the most common neurodegenerative disorders in elderly people. It is diagnosed by detecting amyloid beta (Aβ) protein in cerebrospinal fluid (CSF) obtained by lumbar puncture or through expensive positron emission tomography (PET) imaging. Although blood-based diagnosis of AD offers a less invasive and cost-effective alternative, the quantification of Aβ is technically challenging due to its low abundance in peripheral blood. To address this, we developed a compact yet highly sensitive microwell-based electrochemical sensor with a densely packed microelectrode array (20 by 20) for enhancing sensitivity. Employing microwells on the working and counter electrodes minimized the leakage current from the metallic conductors into the assay medium, refining the signal fidelity. We achieved a detection limit <10 fg/mL for Aβ by elevating the signal-to-noise ratio, thus capable of AD biomarker quantification. Moreover, the microwell structure maintained the performance irrespective of variations in bead number, indicative of the sensor's robustness. The sensor's efficacy was validated through the analysis of Aβ concentrations in plasma samples from 96 subjects, revealing a significant distinction between AD patients and healthy controls with an area under the receiver operating characteristic curve (AUC) of 0.85. Consequently, our novel microwell-based electrochemical biosensor represents a highly sensitive platform for detecting scant blood-based biomarkers, including Aβ, offering substantial potential for advancing AD diagnostics.

Mass spectrometry on cerebrospinal fluid uncovers association of novel glycolysis biomarkers with Alzheimer’s disease in a complex clinical cohort

AbstractBackgroundDiagnosis of Alzheimer’s disease (AD) primarily relies on cognitive assessments supported by imaging and cerebrospinal fluid (CSF) biomarkers of amyloid and tau proteins. These limited biomarkers correlate poorly with disease progression or treatment response. Broader, more comprehensive biomarker sets together with traditional ATN biomarkers could improve the diagnosis and prognosis of AD, identify contributing pathophysiologies and comorbidities, and prove useful in measuring treatment efficacy. In this study we leverage mass‐spectrometry to simultaneously quantify hundreds of proteins in CSF from a large‐scale neurology clinic cohort, all of whom had clinical indication for diagnostic lumbar puncture. We aim to define markers that can stratify and diagnose AD in this “real‐life” clinically complex cohort.MethodCSF samples from 408 patients were collected from the Massachusetts General Hospital Neurology Service following written informed consent for research biobanking and analysis of excess CSF. The patient cohort consisted of ATN‐verified cognitively‐unimpaired (n = 81), mild‐cognitive impairment from AD (n = 116), mild‐cognitive impairment from other non‐AD causes (n = 78), dementia from AD (n = 65) and dementia from other non‐AD causes (n = 31). Samples were analyzed on an Orbitrap Fusion using a Data Independent Acquisition method and raw files were searched using Scaffold‐DIA.Result1761 unique peptide sequences belonging to 516 proteins were robustly quantified by DIA in CSF. Linear regression analyses found 572 peptides to be significantly differentially abundant across all diagnoses compared to AD (padj &lt; 0.05). Ontological analysis of these peptides indicated enrichment of glycolysis in AD, potentially indicating a shift in brain metabolism in AD. 4 peptides belonging to 2 glycolysis proteins (PKM, ALDOA) were elevated in AD compared to all non‐AD diagnoses, suggesting specificity for AD. Differential abundance of these proteins was verified through immunoassay.ConclusionHere, unbiased mass‐spectrometry on CSF highlighted novel AD specific biomarkers ALDOA and PKM, potentially indicating a metabolic shift in the brain. Further mechanistic studies should investigate dysregulation of these markers in AD and correlation with FDG‐PET.

Time of the day at sampling affects amyloid‐beta levels in CSF and plasma

AbstractBackgroundStudies suggest that cerebrospinal fluid (CSF) levels of Alzheimer’s disease (AD) biomarkers, amyloid‐β (Aβ) and tau, follow a circadian rhythm. However sustained sampling of CSF with indwelling intrathecal catheter used in most of these studies might have affected the observed fluctuations in the biomarker levels.MethodsWe included 38 individuals with either normal (N = 18) or abnormal (N = 20) CSF Aβ42/40 status. CSF and plasma were collected, at the same occasion, at two different visits separated by an average of 53 days with lumbar punctures and venepunctures performed either in the morning or evening. At the first visit, collection was performed in the morning for 17 participants and the order was reversed (evening collection first) for the remaining 21 participants. CSF and plasma samples were analyzed for Aβ42, Aβ40, GFAP, NfL, p‐tau181, p‐tau217 and p‐tau231. CSF samples were also tested for a battery of synaptic and lysosomal proteins.ResultsCSF Aβ42 (mean difference [MD] = 0.208ng/ml; p = 0.043), CSF Aβ40 (MD = 1.85ng/ml; p&lt;0.001), plasma Aβ42 (MD = 1.65pg/ml; p = 0.002) and plasma Aβ40 (MD, 0.011ng/ml, p = 0.003) levels were increased in the evening samples compared with morning samples by 4.2%‐17.0% (Figures 1‐2). However, changes between morning and evening samples were not significant for either CSF or plasma Aβ42/40. Additional proteins in CSF, APP, STX1B, NPTXR and NPTX1, were also found increased in the evening samples (MDrange, 0.087–0.243; p&lt;0.048) by 4.7%‐12.15% (Figure 3). No significant differences were found between morning and evening levels for any of the other biomarkers, including p‐tau’s, NfL and GFAP. There were no significant interactions between time of day at sample collection and either Aβ status or order in which samples were obtained.ConclusionOur findings provide further evidence for diurnal fluctuations in Aβ peptides, which is observed in both CSF and plasma. Additionally, CSF levels of some synaptic proteins followed circadian rhythm. Importantly, CSF and plasma Aβ42/40 ratios remained unaltered, suggesting that increased production or decreased clearance of Aβ peptides during daytime similarly affect the CSF and plasma levels of both Aβ42 and Aβ40. Thus, these ratios are more suitable for implementation as biomarkers of AD as they are not influenced by diurnal fluctuations.

Sex and aging signatures in human CSF proteomics and their link to neurologic diseases

AbstractBackgroundSex and age are major risk factors for several chronic diseases. Recent human and mouse studies of age‐related molecular changes in blood provided new insights into age‐related disease biology. Cerebrospinal fluid (CSF) is more closely related to brain than blood and can provide additional insights to sex and age‐related brain diseases.MethodWe measured over 7000 proteins in CSF from 998 healthy individuals representing middle and old age (43‐91) across four multiple cohorts. After stringent QC, we performed weighted gene co‐expression network analysis (WGCNA) to group CSF proteins with similar sex and age trajectories. Each cluster was then tested for aging and sex differences. Cell‐type enrichment and disease ontology enrichment was also performed.ResultCo‐expression network analysis identified 24 modules for CSF 4926 proteins. Sex difference was detected for 2949 proteins (1730 were higher in females; 1219 were higher in males). Aging signature was detected for 4480 proteins (2999 increasing; 1481 decreasing with age). The pink module (containing SERPINA3, TREML2, and additional 197 proteins) had proteins higher in males and increasing with ages and were enriched in endothelial cell (P = 0.04). Cardiovascular diseases (FDR = 2.6E‐9), ischemic stroke (FDR = 1.7E‐7) and Alzheimer’s disease (AD) (FDR = 7.3E‐3) were significantly enriched in this module. The blue module (containing APOE, APP and additional 681 proteins) had proteins higher in females and decreasing with ages, were enriched in microglia and macrophage (P = 0.01). AD (FDR = 0.043) and vascular dementia (FDR = 0.025) were enriched. The yellow module (containing SMPD1, GNS, CTSB and additional 331 proteins) had proteins higher in females and non‐linear aging trajectory (increasing until 70 years old and decreasing) were enriched in neurons (P = 4.6E‐9). Lysosomal storage diseases (FDR = 3.4E‐03) showed enrichment for them. Among the 46 plasma proteins showing conserved aging signature in mice and humans (Lehallier et al 2019), 20 proteins including TNFSF15 and PPBP were also conserved in CSF. In addition, nine of them show the sex‐specific signature conserved across all three datasets.ConclusionOur study in CSF proteomic signatures identified clusters of proteins in distinct patterns and their relation to Alzheimer’s disease and other neurodegeneration. More detailed characterization of proteomic signatures for other age‐related chronic diseases is underway.

Sex‐specific CSF proteomic profiling of amyloid/tau positivity identifies distinctive sex‐specific alternation of multiple proteins involved in Alzheimer’s disease

AbstractBackgroundSporadic Alzheimer’s disease (AD) is a highly heterogeneous multifactorial disease in which females have a higher risk. Among 5.3 million senior Americans diagnosed with AD, over 60% are female. Despite the well‐established sex differences, sex‐specific molecular findings in AD are still limited. We therefore examined the role of sex in the proteomic alterations due to AD to elucidate potential protein markers of AD heterogeneity.MethodIn the discovery cohort, 599 preclinical AD and 435 controls underwent lumbar puncture. Over 7,000 proteins in cerebrospinal fluid (CSF) were measured using SomaLogic assay. Amyloid/tau positivity (AT classification) was determined using ELISA Aβ42 and pTau181 levels in CSF. Proteins with sex‐specific effects were identified through linear models focusing on the interaction term between AT classification and sex. Results were tested in the independent cohort (N = 732) using the identical approaches. We analyzed brain proteomics and transcriptomics data to validate the replicated sex‐specific proteins. We performed pathway enrichment, interaction network using ConsensusPathDB and drug‐target prediction.ResultWe identified 594 proteins with sex‐specific effects in discovery (P&lt;0.05 and permutation FDR &lt;0.05). Among them, 36 were validated in replication (P&lt;0.05 and consistent direction). These proteins were found to be involved in neurologic disorders and highly enriched in axon regeneration (fold enrichment = 9.38 and FDR = 2.8×10−4). Among the 36 proteins identified, 13 were classified as female‐specific and 23 as male‐specific. These 36 proteins strongly predicted amyloid/tau positivity (AUC = 0.926) for all individuals as well as 13 female‐specific proteins for females (AUC = 0.929). The interaction network identified CCN2 as a hub and contained 10 additional proteins including JAG1 and DLL4 of Notch signaling pathway and CTSB, an amyloid precursor protein secretase, among others. The results of 36 proteins in brain proteomics and transcriptomics confirmed the sex‐specific effect of CCN2 and others on AD in different tissues. Drug target prediction suggests minocycline and etoposide as potential treatments for females as well amitriptyline and verteporfin for males.ConclusionWe robustly identified 36 proteins that showed significant sex‐specific alterations by preclinical AD. Several proteins involved in AD and neurodegeneration were highlighted. Our findings facilitate mechanistic understanding of sex differences for AD risk and progression.

Molecular biomarkers quantified using the Single Molecule Array in Brazilian patients with mild Alzheimer’s disease and mild cognitive impairment

AbstractBackgroundThe clinically diagnosed late‐onset Alzheimer’s disease (AD) cases has the most heterogeneous pathology. The development of more sensitive biomarkers is essential to perform an early diagnosis, especially in the prodromal phase, mild cognitive impairment (MCI). Biomarkers such as glial fibrillary acidic protein (GFAP) and neurofilament light chain (NfL) have emerged as potential markers of the pathophysiological process of AD. The objective of this study was to evaluate the concentrations of GFAP and NfL in blood serum and CSF of patients with MCI and mild AD in a Brazilian sample, never quantified using the Single Molecule Array technology (Simoa).MethodWe used samples (blood serum and cerebrospinal fluid (CSF)) from 91 participants to perform molecular analyzes. These participants were recruited in the outpatient service of the UNICAMP Clinic Hospital. All participants underwent blood collection and only the patients, lumbar puncture, to analyze GFAP and NfL levels. All patients had pathophysiological evidence of AD (low Aβ1‐42 concentrations &lt; 540pg/mL) and was considered Tau alteration: p‐Tau &gt; 36.7 pg/mL and t‐Tau &gt; 76.7 pg/mL.ResultStatistical analyzes were performed using SPSS software (version 25). We performed a multivariate analysis of covariance (MANCOVA) to compare the biomarkers levels between the groups. To compare CSF levels (Aβ42, t‐Tau, p‐Tau, GFAP and NfL) we used T‐test analyses and to correlate GFAP and NfL with CSF proteins we use partial correlations. We did not find statistical differences between the groups regarding GFAP and NfL levels, but we did find differences between Aβ1‐42 protein between MCI and mild AD (p = 0.003). We also found moderate correlations between t‐Tau, GFAP and NfL CSF levels (r = .625, p = 0.0; r = .508, p = 0.005, respectively), and p‐Tau with GFAP CSF levels (r = .373, p = 0.046).ConclusionGFAP and NfL reflect neuro‐axonal damage and recent studies have demonstrated its effectiveness in tracking disease progression (5). Although we did not find differences between individuals in relation to GFAP and NfL markers, we were able to observe correlations between these markers at the CSF level and the proteins classically related to the pathological process of the disease.

Comparison of plasma and cerebrospinal fluid protein markers and their significance for the characterization of dementia disease status

AbstractBackgroundAlthough Aβ42 and phosphorylated tau (p‐tau) in cerebrospinal fluid (CSF) are valuable to diagnose Alzheimer’s disease (AD), many patients are reluctant to accept lumbar puncture. In addition, there is lack of biomarkers for the different stages of AD dementia. This study sought to compare the plasma and CSF protein markers to characterize dementia status.MethodThis cohort study included Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants who had baseline levels of 73 proteins measured in plasma (566 total, 58 cognitively normal (CN), 396 mild cognitive impairment (MCI), and 112 AD) and in CSF (310 total, 92 CN, 149 MCI, and 69 AD). Univariate/multivariate logistic regression and least absolute shrinkage and selection operator (LASSO) were used to identify the protein candidates for the disease classification. The following three models were examined for CN vs. MCI and MCI vs. AD; model 1: proteins only, model 2: proteins adjusted for age, sex, education, and model 3: model 2 plus APOE4 status.ResultNine proteins in plasma and CSF were chosen based on their importance from LASSO. The area under the curve (AUC) for the final multivariate logistic regression including the 9 proteins was calculated for each disease group in plasma [CN vs. MCI: 0.895 (model 1), 0.899 (model 2), 0.921 (model 3), and MCI vs. AD: 0.828 (model 1), 0.842 (model 2), 0.856 (model 3)]. In addition, AUC in CSF [CN vs. MCI: 0.678 (model 1), 0.726 (model 2), 0.770 (model 3) and MCI vs. AD: 0.711 (model 1), 0.738 (model 2), 0.765 (model 3)] was also estimated. We observed that plasma proteins outperformed the CSF proteins to be associated with MCI and AD. Also, plasma proteins were significantly better than the established AD markers, namely CSF Aβ42 and p‐tau in CN vs. MCI [AUC: 0.836 (model 1), 0.842 (model 2), 0.862 (model 3)] and MCI vs. AD [AUC: 0.665 (model 1), 0.673 (model 2), 0.680 (model 3)].ConclusionThis study demonstrates that the combination of plasma proteins could be considered an effective marker for diagnosing AD and its early‐stage symptoms with better performance than the other markers.

PREVENT‐AD cohort: from normal to impaired cognition, who are declining and why? Preliminary results

AbstractBackgroundThe StoP‐Alzheimer Center follows a cohort of cognitively normal participants at risk of developing Alzheimer’s disease (AD) due to their parental history (PREVENT‐AD cohort). Cognitive impairments were detected in 17% of the cohort (66 out of 386) mostly after between 4 to 6 years of follow‐up, at an average age of 70 ± 7 years of age.MethodWe looked at various variables at baseline (demographics, blood results, medical history, cognition, genetic variants, AD pathology, mood) to describe the differences between the participants who maintained a normal cognition versus those who declined.ResultDifference in ApoE4 status was not as important as expected with 39.4% ApoE4 carriers in the decliner group compared to 38.2% in the normal group (Table 1). Despite a normal cognition at baseline, we found subtle but significant lower cognitive performance in the decliner group on the MoCA (‐1 pt) and the Repeatable Battery for the Assessment of Neuropsychological Status (‐5 pts). Systolic blood pressure and glycosylated hemoglobin were slightly more elevated in the decliner group (132 mmHg versus 128 mmHg and 5.6% versus 5.4% respectively). When asked about their past medical history, more participants were treated for hypertension, hyperlipidemia and diabetes in the decliner group. At the level of brain AD pathology measured by PET, the decliner group presented higher levels (standard uptake values) of amyloid and tau. Overall, 27% of the decliner group were amyloid AND tau positive compared to 7% in the normal group. The decliner group also showed lower smell identification performances at baseline when compared to the group who didn’t decline.ConclusionMain differences between the groups at baseline are related to AD pathology, vascular risk factors and cognitive and smell identification capacities. Other available variables such as brain structure, cerebrospinal fluid proteins, additional cognitive tests, lifestyle questionnaires, and genetics variants need to be analyzed along with trajectories and change over time. As a growing number of participants are declining cognitively and are still followed annually with multimodal evaluations, we trust that further investigations will help better understand the development of cognitive impairment.

Choroid plexus protein turnover in human choroid plexus organoids recapitulates turnover in humans measured using stable isotope labeling kinetics (SILK)

AbstractBackgroundThe choroid plexus secretes protein‐rich cerebrospinal fluid (CSF). CSF turnover is a key mechanism for protein clearance. Neurodegenerative diseases such as AD are characterised by accumulation of excess proteins that form pathological inclusions. CSF flow dysregulation could therefore be involved in the pathophysiology of neurodegenerative diseases (ND) such as Alzheimer’s Disease. A method for measuring CSF turnover in vivo may be achieved by quantitating CSF derived protein kinetics, providing a biomarker tool with to interrogate the relationship between CSF flow disequilibrium and neurodegeneration.MethodUsing untargeted proteomics we identify the most abundant proteins in a) stem‐cell derived choroid plexus organoids and b) human CSF. We then develop targeted mass spectrometry methods to quantitate protein specific peptides in both matrices. Using a stable isotope (13C6 leucine) we label a) choroid plexus organoids and b) humans with suspected NPH undergoing a 72 hour lumbar drainage protocol. We collect serial samples of the CSF‐like organoid fluid and CSF. We then measure the ratio of labelled to unlabelled peptides to quantitate the rates of protein labelling (fractional synthesis rate) and label loss (fractional clearance rate), defining the protein turnover rate.ResultHighly‐expressed proteins in human CSF from nine individuals were also present in CSF‐like fluid from CP organoids. Of these common proteins, we were able to quantitate the turnover rates of transthyretin, apolipoprotein E, cystatin C and pigment epithelium‐derived factor, and the synthesis rates of serotransferrin and albumin. We will correlate these findings with clinical measures of CSF production.ConclusionChoroid plexus protein synthesis and clearance can be measured using SILK; choroid plexus organoid fluid protein turnover recapitulates that seen in human CSF.

Multi‐cohort cerebrospinal fluid proteomics identifies robust molecular signatures for asymptomatic and symptomatic Alzheimer’s disease

AbstractBackgroundAlzheimer’s disease (AD), the most common form of dementia, is characterized by the accumulation of Amyloid‐β (Aβ42) and hyperphosphorylated Tau 181 (p‐tau181) proteins in the brain. Changes in these proteins in AD brains and cerebrospinal fluid (CSF) are detected years before AD symptoms. Studying AD proteome in CSF can reflect its diverse underlying pathophysiology and pave the way for reliable diagnostic and therapeutic advancements.MethodWe present one of the largest CSF AD proteomic profiles (7,029 protein analytes) in CSF of 3,065 individuals in three stages. Discovery was performed in 836 samples from the Knight ADRC and 618 samples from the FACE cohorts using the ATN framework (AT‐ = 680 and AT+ = 490). The identified proteins were tested in 832 individuals (AT‐ = 235 and AT+ = 358) from the ADNI and Barcelona‐1 cohorts. The proteins that passed multiple Bonferroni corrections on the meta‐analysis were utilized for AD prediction models and pathway enrichment analysis to gain mechanistic insights into AD pathophysiology.ResultIn discovery, we identified 3,565 proteins to be significantly (FDR &lt; 0.05) altered. Of these, 2,543 passed FDR in replication with a consistent direction. In the meta‐analysis, we identified 2,233 proteins to be significantly (P‐Bonferroni &lt; 0.05) altered in the AD CSF proteome. Some of the highly significant proteins included YWHAG (P‐Bonf &lt; 4.5×10−219), SMOC1 (P‐Bonf &lt; 5.8×10−208), NRGN (P‐Bonf &lt; 3.2×10−119), and NEFL (P‐Bonf &lt; 1.8×10−37). By using lasso, we identified a set of 39 proteins with high predictive power (Discovery AUC = 1.0; Replication AUC = 0.99) representing a robust and precise AD diagnostic biomarker. Enrichment analysis highlighted several neurological disorders (e.g. AD, tauopathy, and synucleinopathy) and neuronal functions (neuron projection morphogenesis, synapse assembly and organization, and neuron differentiation) as significantly enriched in the altered AD CSF proteome.ConclusionOur findings show the promising potential of AD CSF proteomics as a reliable and robust AD prediction model. We identified proteins and biological pathways that are compromised in AD, thereby, increasing our understanding of AD biology. Our findings may accelerate the development of effective intervention therapies that target the earliest molecular triggers of AD.

Proteins involved in synapse organization are associated with longitudinal loss in gray matter network connectivity in prodromal AD

AbstractBackgroundLower gray matter connectivity measures are related to faster cognitive decline in prodromal Alzheimer’s Disease (AD). We investigated the biological underpinnings responsible for this loss of brain connections using cerebrospinal fluid (CSF) proteomics.MethodData from ADNI was used to select 82 individuals with MCI and abnormal CSF amyloid (mean age±SD 74±7) and 37 controls with normal amyloid levels (mean age±SD 75±5), for whom repeated MRI‐scans (median of 6 scans over 1.5 follow‐up years) and baseline CSF proteomic data (in total 306 proteins) were available. We used an automated pipeline (Tijms, Series et al. 2012) to construct gray matter networks from 3D‐T1 sequences and calculate connectivity density and gamma – measures previously found to be most strongly related to cognitive decline (Dicks, Vermunt et al. 2020). Linear mixed models were applied to test associations between baseline CSF protein levels and repeated connectivity metrics, controlling for intracranial volume, age and sex. Proteins that demonstrated significant associations (p&lt;0.05) were identified and examined for their biological relevance according to Gene Ontology.ResultConnectivity density declined over time with ‐0.8% per year (pMCI = 0.001; figure 1), which did not differ from controls (pdif = 0.15). Gamma declined faster in MCI than in controls (bMCI = ‐0.016 vs. bCN = ‐0.007, pdif &lt;0.001; figure 1). Higher levels of 99/108 proteins were associated with faster decline in connectivity density (a.o. tau and neuroprentaxins). These proteins were involved in regulation of synapse organization (GO:0050807; pFDR = 7.00E‐07; figure 2). Higher levels of 9/108 proteins associated with less decline and were involved in chemotaxis (GO:0048247; pFDR = 1.40E‐02). Similarly, 25 out 30 proteins (83%) associated with alterations in gamma over time, showed associations of higher levels with less gamma decline and were also involved in chemotaxis (GO:0050926; pFDR = 9.42E‐05; figure 3). There were no protein‐connectivity relationships in controls.ConclusionProteins involved in synapse organization were associated with faster decline in connectivity density over time in patients, while chemotaxis was associated with less decline. This suggests that synaptic degeneration may predict disease progression while immune responses may serve as a resilience mechanism in AD.

Integration of cerebrospinal fluid proteomics and clinical scaled scores for subtyping sporadic Parkinson’s disease

AbstractBackgroundMolecular pathways and networks associated with sporadic PD (sPD) subtypes are still unclear. We applied a systems biology approach to identifying proteomic signatures associated with sPD, cognitive and motor scaled scores, brain volumetric measurements, and cerebrospinal fluid (CSF) phosphorylated‐tau (p‐tau), total‐tau (t‐tau), alpha‐synuclein (a‐syn), and amyloid‐beta (a‐beta) levels.MethodsWe used CSF proteomic data of sPD from the Parkinson’s Progression Markers Initiative (PPMI) cohort (n = 377 sPD vs. 182 controls) and built an elastic‐net logistic regression model, adjusting for age, sex, and principal components, for sPD risk prediction using glmnet[1]. We performed correlation of baseline proteome with cognitive (MoCA) and motor assessments (UPDRS‐I, II, and III), neuroimaging data (Dopamine transporter scan; DaTscan of mean bilateral caudate, putamen, and striatum thickness (mm)), and CSF biomarkers. We implemented weighted gene co‐expression network analysis (WGCNA) to group co‐expression CSF proteins for sPD and to subtype sPD by CSF biomarkers using WGCNA[2] and ConsensusClusterPlus[3].ResultsWe found a CSF proteomic signature (59 proteins) differentially expressed in sPD compared to controls linked to neurotrophin, endocytosis, and chemokine signaling. There are 82 and 68 proteins differentially expressed in UPDRS‐II and UPDRS‐III scores. Three, 13, and 11 proteins were associated with caudate, putamen, and striatum volumes. We found proteins correlated with p‐tau (n = 1394), t‐tau (n = 1577), a‐syn (n = 1427), and a‐beta (n = 847) levels. We implemented a predictive model with 33 proteins for sPD (AUC = 0.83) that outperformed the models with UPDRS‐I (AUC = 0.71), MoCA (AUC = 0.67), p‐tau (AUC = 0.56), t‐tau (AUC = 0.61), a‐syn (AUC = 0.54), and a‐beta (AUC = 0.57) but underperformed UPDRS‐III (AUC = 0.99) or putamen (AUC = 0.99). WGCNA indicates CSF biomarkers are highly correlated to six protein modules for subtyping sPD: each module was enriched for different pathways, ME1 (SNARE and MAPK signaling), ME9 (autophagy), ME7 (lysosome and MAPK), ME6 (immune response), ME10 (proteostasis and phagocytosis), and ME2 (longevity regulating and mTOR). Those modules defined two subtypes of sPD by p‐tau levels (9.93 pg/mL, AUC = 0.909). MoCA (p = 0.035), t‐tau (p = 3.2×10−60), a‐beta (p = 1.7×10−22), and a‐syn (p = 4.6×10−32) were significantly lower in the subtype with low levels of p‐tau.ConclusionsIntegrating molecular profiling and clinical scale scores improves the prediction of sPD risk and identifies sPD subtypes.

CSF proteome profiling reveals a protein panel detecting amyloidosis and progression to dementia in cognitively unimpaired individuals

AbstractBackgroundUnderstanding the earliest biochemical changes related to Alzheimer´s disease (AD) pathology is essential for the development of biomarkers and disease‐modifying therapies. Here, we aimed to map the cerebrospinal fluid (CSF) proteomic changes associated with amyloid ß (Aß) pathology in cognitively unimpaired individuals and identify a panel of CSF biomarkers detecting amyloid positivity as well as clinical progression to dementia.MethodProximity extension‐based multiplex immunoassays were used to measure 614 proteins in 297 CSF samples from cognitively unimpaired controls with and without amyloid pathology (232 CON/Aβ‐ (age: 68±8; sex: 37% female) and 65 CON/Aβ+ (age: 65±7; sex: 49% female) from the Amsterdam Dementia cohort. A total of 39 cases (13%) progressed to mild cognitive impairment (MCI; n = 19) or dementia (n = 20). An additional cognitively unimpaired control cohort from the EMIF‐twin60+ study (103 CON/Aβ‐ and 19 CON/Aβ+) was used to validate the panel. Data was analysed using nested linear models adjusted for multiple testing and penalized generalized linear modelling.ResultWe identified 110 CSF proteins with different levels in CON/Aβ+ compared to Aβ‐ (q&lt;0.05). Proteins were enriched in processes related to proteolysis, enzyme activation and immunity. Classification modelling revealed a panel of 12‐CSF markers that detect amyloid positivity with high accuracy in both the discovery (AUC: 0.93) and validation (AUC: 0.89) cohorts. This panel also predicted clinical progression to MCI and dementia with high accuracy (AUC: 0.83). Chow´s test analysis revealed a structural change in the relation between a subset of these proteins and CSF Aβ42 before the amyloid positivity threshold.ConclusionOverall, this study provides novel pathophysiological leads associated to distinct biological processes in individuals at risk of developing AD‐dementia and provides potential biomarker tools for clinical settings or trials.

Proteomics Analysis Identifying New Markers Related to Alzheimer’s Disease Pathology In Vivo

AbstractBackgroundTo better understand the pathophysiological processes in Alzheimer’s disease (AD), we need to broaden our knowledge on molecular pathways and discover molecular markers that co‐occur with, precipitate and follow the accumulation beta‐amyloid (Aβ) and tau pathology. Using high‐throughput proteomics, we investigated how a wide array of proteins relate to AD pathology in a large cohort spanning the AD continuum.MethodWe included 949 BioFINDER‐2 participants with cerebrospinal fluid (CSF) measurements of 2943 proteins (Olink Explore 3072 assay), and CSF and/or PET measurements of Aβ and tau. We compared CSF proteins between groups based on their Aβ (A) and tau (T) status, defined respectively from CSF Aβ42/40 ratio and tau‐PET RO948 in the temporal meta‐ROI (i.e. A‐T‐ vs. A+T‐ and A+T‐ vs. A+T+). We also compared A+T+ vs. non‐AD neurodegenerative diseases. Then, among the significant proteins identified across all three comparisons, we evaluated their associations with Aβ‐ (flutemetamol) and tau‐PET, both cross‐sectionally and longitudinally (based on rate of change). These latter analyses were restricted to cognitively unimpaired and patients with MCI, who all had Aβ‐ and tau‐PET available (n = 548).ResultAcross all AT comparisons, 85 proteins were significantly different between groups (pFDR&lt;0.05; Fig.1). More advanced pathology was associated with higher protein levels, except ApoE levels that were decreased. The A+T‐ vs. A+T+ comparison revealed most protein differences (n = 74; Fig.1B), compared to A‐T‐ vs. A+T‐ or A+T+ vs. non‐AD neurodegenerative diseases. In cross‐sectional analyses, 32/85 proteins (38%) were significantly associated with Aβ‐ and tau‐PET load, with the strongest ones being microtubule‐associated protein tau (MAPT), neurogranin (NRGN), ITGAM, ITGB2 and SMOC1. When putting both pathologies in the same model, all proteins remained associated with Aβ only, except for ENO2 and GPI, which remained associated with tau only (Table 1). In longitudinal analyses, 19 proteins were associated with greater tau‐PET rate of change (the top five being MAPT, NRGN, ITGB2, FABP3 and CHIT1), whereas only SMOC1 and ITGAM were associated with greater Aβ‐PET rate of change (Table 1).ConclusionThrough comprehensive proteomics analysis, we identified novel proteins associated with AD pathology, many of which are involved innate immunity, intracellular transport and extracellular matrix binding.

Deciphering the components of Amyloid‐beta (Aβ)‐driven dementia using a novel peptide‐focused global proteomics platform

AbstractBackgroundThe overlap of Alzheimer’s disease (AD) symptomology and pathology with other dementia types is a significant challenge to treatment, patient care, and new drug development. Amyloid‐beta (Aβ) peptides begin to accumulate in AD subjects over a decade prior to clinical symptoms and impair multiple cellular pathways during disease continuum. However, how the brain proteome is remodeled by Aβ burden remains an enigma. Here we compared the cerebrospinal fluid (CSF) proteomes of demented subjects with or without Aβ pathology to unveil alterations specific to Aβ‐driven dementia.MethodMass spectrometry data (PRIDE archive PXD016278) from CSF of cognitively impaired Sweden cohort subjects was reanalyzed using our de novo‐assisted workflow. Peptide raw peak area intensities were quantile normalized and log2 transformed to reduce technical variation and ensure distribution symmetry. Differentially expressed peptides were identified via analysis of covariance after adjusting for age and gender. Multiscale Embedded Gene Co‐expression Network Analysis (MEGENA) was used to explore the dysregulated network from Aβ‐driven brain proteome remodeling.ResultThe de novo‐assisted workflow significantly increased the CSF proteome depth ‐ 4,572 CSF protein groups vs 1542 protein groups reported in PXD016278 analysis. Comparison of peptide‐level CSF proteomes of demented subjects with or without Aβ pathology identified 2,014 differentially regulated peptides (q&lt;0.2). These alterations may reflect the Aβ‐driven brain proteome remodeling in CSF. MEGENA network analysis identified several hubs and network modules. The most significantly altered ALDO‐A peptide represented a hub, a highly connected node within a network module, and exhibited high correlation with AD‐linked neurosecretory protein VGF and MAPT. The top pathways enriched in MEGENA network modules, e.g. extracellular matrix organization, cell adhesion, and neurogenesis, are consistent with other pathway enrichment analyses across other cross‐sectional studies in AD and non‐AD dementias.ConclusionThis is the first study capturing the peptide‐level alterations in CSF to reflect brain proteome remodeling by Aβ burden. This data will serve as an important resource for biomarkers demarcating AD from non‐AD.

Cerebrospinal Fluid Protein Levels in Patients with Neurodegenerative Diseases Correlates with Cognitive and Motor Function

AbstractBackgroundBlood brain barrier (BBB) dysfunction is suggested to have a causal role in neurodegenerative disorders. Cerebrospinal fluid (CSF) total protein (TP) is a proxy for CSF albumin quotient, which is an indicator of BBB breakdown. Normative values for CSF TP are based on young and healthy individuals, and there is a lack of understanding of how this biomarker changes with aging and neurodegeneration. The aim of this study is to examine the CSF profile in a cohort of patients with ALS‐FTD spectrum disorders and to determine relationship between the CSF profile and cognitive and motor measures in these patients.MethodCSF from participants with ALS‐FTD spectrum disorders enrolled in the NIH protocol “Investigating Complex Neurodegenerative Disorders related to Amyotrophic Lateral Sclerosis and Frontotemporal Dementia” were included (n = 63). We described the distribution of the CSF profile in these participants and compared the findings with existing literature. We then examined the relationship between CSF measures and functional rating scales and cognitive and motor measures.ResultThe average total protein was 47.77 mg/dL (SD = 38.29). The average CSF glucose was 66.00 mg/dL (SD = 12.80). There was a negative correlation (p = 0.0102) between CSF TP and the ALS Functional Rating Score. We will further investigate the relationship between the CSF profile and measures of cognitive and motor performance.ConclusionOur findings show that CSF TP is elevated in patients with ALS‐FTD spectrum disorders and higher CSF TP level correlated with greater disease severity in ALS. We will further investigate the relationship between CSF TP and other measures of cognitive and motor function.

Identification of Prototheca from the Cerebrospinal Fluid of a Cat with Neurological Signs.

Prototheca infections are rare in cats, and they are usually associated with cutaneous or subcutaneous infections by P. wickerhamii, with no evidence of neurological signs or systemic disease. In this study, we report the identification of prototheca in the cerebrospinal fluid (CSF) of a cat with neurological symptoms. Fourteen CSF samples were gathered from cats presented with neurological disease between 2012 and 2014. The inclusion criteria for the samples were an increase in CSF protein and cell number (pleocytosis), suggestive of an infectious inflammatory status of the central nervous system (CNS). Nine samples fulfilled the inclusion criteria (inflammatory samples), while five samples, used as control, did not (non-inflammatory samples). All the samples were screened molecularly for different pathogens associated with CNS disease in cats, including prototheca. Out of 14 CSF samples, only one inflammatory sample tested positive for prototheca. Upon sequence and phylogenetic analysis of the amplicon, the strain was characterized as P. bovis. This report is the first documented evidence of prototheca in the cerebrospinal fluid of a cat with neurological signs. Prototheca should be considered in the diagnostics procedures on the CNS of cats presented with infectious diseases.