Joanna Bons

Individuals with Alzheimer's disease (AD) are at an increased risk of bone fracture, while osteoporosis in women is one of the earliest predictors of AD. Yet the mechanisms linking cognitive decline and skeletal deterioration remain poorly defined. Proteomic analysis of cortical bone from aged 21‐month‐old mice revealed strong enrichment of neurodegeneration‐associated proteins, including apolipoprotein E (Apoe) and amyloid precursor protein. Apoe localized specifically to osteocytes, with expression in aged female bone nearly twice that of young 4‐month‐old male bone. Because human APOE alleles confer different age‐related AD risks, we examined their roles in bone using humanized APOE2, APOE3, and APOE4 knock‐in mice and analyzed bone and hippocampus from the same animals. APOE4 produced marked sex‐specific effects on the bone transcriptome and proteome compared with APOE2 or APOE3. Strikingly, APOE4‐associated proteomic disruptions were stronger in female bone than in the hippocampus. Functionally, APOE4 caused bone fragility in females without altering cortical structure. These deficits stemmed from impaired osteocyte perilacunocanalicular remodeling. Our findings identify APOE4 as a molecular driver of early osteocyte dysfunction and reduced bone quality, disproportionately affecting females. These findings highlight osteocytes as potential targets for early diagnosis of age‐related cognitive impairment and treatment for bone fragility, in females.

Développement de méthodes analytiques par spectrométrie de masse pour une meilleure quantification et une caractérisation fine des protéomes: La protéomique quantitative est un atout considérable pour l’étude des systèmes biologiques à l’échelle protéique, comme en témoigne le large panel de stratégies disponibles. Celui-ci peut être subdivisé de la sorte : stratégies globales, stratégies ciblées et stratégies DIA qui combinent les avantages des deux précédentes. Ce travail de thèse est focalisé sur le développement de méthodes de protéomique quantitative afin d’améliorer la quantification et la caractérisation des protéomes. Le principal verrou résidant dans l’analyse des données, la première partie de mon travail a consisté à évaluer différents logiciels de traitement des données DIA. Suite aux développement et optimisations de différentes méthodes de quantification dans des contextes biologiques variés, leurs performances ont été évaluées afin de répondre au mieux aux questions biologiques posées. Enfin, différentes stratégies analytiques ont été développées afin de caractériser finement des protéomes d’entités cellulaires, à savoir le protéome mitochondrial humain, le contenu protéique d’exosomes tumoraux et d’entités biologiques encapsulées dans les microgouttelettes.

Osteoarthritis (OA) is a degenerative skeletal condition marked by the loss of articular cartilage and changes to subchondral bone homeostasis. Treatments for OA beyond full joint replacement are lacking primarily due to gaps in molecular knowledge of the biological drivers of disease. Mass Spectrometry Imaging (MSI) enables molecular spatial mapping of the proteomic landscape of tissues. Histologic sections of human tibial plateaus from knees of human OA patients and cadaveric controls were treated with collagenase III to target extracellular matrix (ECM) proteins prior to MS Imaging of bone and cartilage proteins. Spatial MS imaging of the knee identified distinct areas of joint damage to the subchondral bone underneath areas of lost cartilage. This damaged bone signature extended underneath remaining cartilage in OA joints, indicating subchondral bone remodeling could occur before full thickness cartilage loss in OA. Specific ECM peptide markers from OA-affected medial tibial plateaus were compared to their healthier lateral halves from the same patient, as well as to healthy, age-matched cadaveric knees. Overall, 31 peptide candidates from ECM proteins, including Collagen alpha-1(I), Collagen alpha-1(III), and surprisingly, Collagen alpha-1(VI) and Collagen alpha-3(VI), exhibited significantly elevated abundance in diseased tissues. Additionally, highly specific hydroxyproline-containing collagen peptides, mainly from collagen type I, dominated OA subchondral bone directly under regions of lost cartilage but not areas where cartilage remained intact. A separate analysis of synovial fluid from a second cohort of OA patients found similar regulation of collagens and ECM proteins via LC-MS/MS demonstrating that markers of subchondral bone remodeling discovered by MALDI-MS may be detectable as biomarkers in biofluid samples. The identification of specific protein markers for subchondral bone remodeling in OA advances our molecular understanding of disease progression in OA and provides potential new biomarkers for OA detection and disease grading.

Induced pluripotent stem cell (iPSC)-derived neurons and microglia are valuable human models for studying neurodegenerative diseases. Specifically, the apolipoprotein E4 ( APOE4 ) gene is a major genetic risk factor for late-onset Alzheimer’s disease. Apolipoprotein E ( APOE ) alleles E2, E3 and E4 can be beneficial, neutral, or increase the risk of Alzheimer’s disease (AD). Here, we developed a proteomic workflow using data-independent acquisitions to provide a quantitative mass spectrometric proteome analysis, and proteomic screening assays for brain-specific cell types derived from iPSC. Protein groups were quantified in APOE3 neurons and microglia, respectively, with ∼80% overlap. Cell type-specific markers and enriched pathways reflected the specialized functions of each cell type, such as synaptic signaling in neurons and immune and inflammatory responses in microglia. The neuron-specific markers included proteins APP, CALB1, CALB2, DLGs, GAP43, NEFL, MAPs; while microglial markers included proteins AIF1, CDs, MMP9, and ITGAM. Ultimately, the combination of robust iPSC differentiation and sensitive proteomic screening assays described here provides a valuable platform for probing the cellular mechanisms underlying neurological disorders. Significance: The quantification of dysregulated proteins and pathways in patient-derived neurons and microglia can provide insights into disease etiology and progression. More broadly, this DIA approach enables deep proteome profiling of unique iPSC-derived cell models, increasing their utility for investigating disease biology and therapeutic development. We focused on iPSC models from two important cell types of the brain, excitatory neurons and microglia. We integrated the proteomes of these two cell types. These tools provide robust biological and mass spectrometric screening tools for future therapeutic interventions using disease-relevant human brain cell types or brain organoid models. Graphical Highlights: Presentation of a proteomic workflow using data-independent acquisitions to monitor and screen proteomes of iPSC-derived brain cell types. Quick MS Assays to determine protein profiles of iPSC-derived neurons and microglia. Characterization of different cell type proteomes from APOE3 iPSCs. Revealing of neuron-specific markers and microglia-specific markers by mass spectrometry. Step-by-step instructions for the set-up of the DIA-MS assays

Protein post-translational modifications (PTMs) dynamically regulate essential biological and cellular processes. Lysine succinylation changes the amino acid charge, potentially affecting protein structures and functions, and dysregulation of protein succinylation may lead to metabolic disorders. Proteome-wide succinylation quantification using proteomic tools remains challenging, especially due to the low abundance of succinylated peptides and the frequent presence of isomeric PTM forms. Ion mobility spectrometry workflows that can differentiate peptidoforms with different PTM distributions represent a powerful strategy to alleviate these challenges. Recently, a new Parallel Accumulation with Mobility Aligned Fragmentation (PAMAF™) operating mode for high-resolution ion mobility-mass spectrometry (HRIM-MS) analysis based on the structures for lossless ion manipulation (SLIM) technology was introduced. Here, we first assessed the performance of PAMAF mode for protein succinylation analysis using synthetic succinylated peptides, demonstrating residue-level differentiation of co-eluting isomers and isobars and precise PTM site localization. We leveraged this novel approach to investigate succinylome remodeling in kidney tissues from wild-type and Sirtuin-5 (Sirt5) knock-out mice, a NAD ⁺ -dependent lysine de-succinylase. PAMAF acquisitions yielded ∼1,000 confidently identified and accurately quantified succinylated peptides and sites from mouse kidney. Sirt5 regulated succinylation of mitochondrial proteins involved in metabolic processes, including fatty acid oxidation, the tricarboxylic acid cycle, and propionate metabolism.

Individuals diagnosed with Alzheimer’s disease (AD) are at an increased risk of bone fractures. Conversely, a diagnosis of osteoporosis in women is the earliest known predictor for AD. However, mechanisms responsible for the coupled decline in cognitive and skeletal health remain unclear. Proteomic analysis of cortical bone from aged mice revealed neurological disease-associated proteins that are highly enriched in aged mouse bones, including apolipoprotein E (Apoe) and amyloid precursor protein. Further, Apoe localized specifically to bone-embedded osteocytes with expression twice as high in aged female bone as in young or male counterparts. In humans, APOE allele variants carry differing AD risk with age. To investigate APOE allelic roles in bone, we utilized a humanized APOE knock-in mouse model that expresses either the protective APOE2, the neutral APOE3, or the AD risk factor APOE4, and analyzed bone and hippocampus from the same mice. APOE4 exerted strong sex-specific effects on the bone transcriptome and proteome, relative to APOE2 or APOE3. Interestingly, the APOE4-associated perturbation in the female bone proteome was more pronounced than the corresponding alterations observed in the hippocampus. APOE4 protein causes bone fragility in females, but not males, even without changes in cortical bone structure. These bone quality deficits arose from suppression of osteocyte perilacunocanalicular remodeling. We find that APOE4 is a new molecular culprit capable of disrupting osteocyte maintenance of bone quality as early as midlife in a manner that disproportionately affects females. These findings highlight osteocytes as potential targets for early diagnosis of age-related cognitive impairment, and treatment for bone fragility, in females.

Mass spectrometry imaging (MSI) is a rapidly advancing technology that provides mapping of the spatial molecular landscape of tissues for a variety of analytes. Matrix-assisted laser desorption/ionization (MALDI)-MSI is commonly employed, however, confident in situ identification and accurate quantification of analytes remain challenging. We present a novel imaging methodology combining trapped ion mobility spectrometry (TIMS)-based parallel accumulation-serial fragmentation (PASEF) with MALDI ionization for targeted imaging parallel reaction monitoring (iprm-PASEF). We investigated the spatial distribution of lipids and metabolites in liver tissues from wild-type and CD38 knockout mice (CD38 ^−/− ). CD38, an enzyme involved in nicotinamide adenine dinucleotide (NAD⁺) metabolism, significantly influences liver metabolic function and contributes to age-related NAD⁺ decline. Although CD38 deletion previously was linked to improved metabolic phenotypes, the underlying spatial metabolic mechanisms are poorly understood. The spatial iprm-PASEF workflow enabled confident identification and differentiation of lipid isomers at the MS2 fragment ion level and revealed increased NAD ⁺ and decreased adenosine diphosphate ribose (ADPR), a by-product of NAD ⁺ hydrolysis, in CD38 ^−/− livers. This approach provided confident, specific, and robust MS2-based identification and quantification of fragment ions in spatial MSI experiments. Additionally, the innovative iprm-PASEF opens unprecedented opportunities for spatial metabolomics and lipidomics, offering spatially resolved insights into molecular mechanisms.