Dear MMSDG members,

This Tuesday, we have a very exciting program for student night. I received 12 abstracts, and in order to accommodate everyone we will start earlier at 18h45 and all presentations will for 6 minutes. Please see the order of presentation below.

Sincerely,

The MMSDG organizing committee.

 

Time	Name	Tittle
18h45	Arianna Cirillo	A drop’s journey: oxylipin profiling using blood microsampling devices
18h51	Danielle Simons	Novel roles of the mitochondrial processing peptidase beyond protein import
18h57	Donald Bimpong	Chemical leachables from microplastics from single-use food contact materials: a non-targeted study under consumer-use conditions
19h03	Elyssa Baker	Absolute Quantitation of the Whole-brain Proteome in Preclinical Alzheimer’s Disease Mouse Models Utilizing the SysQuan Platform
19h09	Clanephtha Gertilus	Detection method for piperazine and its salts concentrations in the air by LC-MS/MS in accordance with the ACGIH® inhalable fraction and vapor (IFV) notation
19h15	Justine Gatein	Diol-Specific Boronic Acid Mass Tags for the Detection and Imaging of Glucose by MALDI MSI
19h21	Léa Christophe	Enhancing the Separation of Isomeric Antigens in Acute Myeloid Leukemia Using Alternative Fragmentation Methods and Differential Ion Mobility
19h27	Louis Thomas	Non-targeted strategies for assessment of chemical contaminants released from reusable food bags under consumer usage
19h33	Rachel S. Pryce	High-Sensitivity Mass Tags for On-Tissue Steroid Imaging via MALDI MSI
19h39	Sai Ittoo	An LC-MS/MS method to detect metabolites involved in cysteine metabolism and oxidative stress
19h45	Samira Norouzi	Development and Evaluation of Machine Learning Models to Screen Environmental Risks of Chemical Features Based on Information Gained from Non-targeted Chromatography-Mass Spectrometry Analysis
19h51	Vasudevan Karanghat	Mass Confusion: Resolving Structural Isomers of 3,5-Dihydroxybenzoic acid and 3,5-Dihydroxyphenylpropanoic acid

 

Abstracts

 

A drop’s journey: oxylipin profiling using blood microsampling devices

Arianna Cirillo¹, Dajana Vuckovic¹
1. Department of Chemistry and Biochemistry, Concordia, University, 7141
Sherbrooke Street West, Montreal, QC H4B 1R6, Canada

Oxylipins are bioactive lipid mediators derived from the oxidation of polyunsaturated fatty
acids and play pivotal roles in inflammation, vascular function, and disease progression.
As emerging biomarkers, their accurate and precise quantification is essential for
advancing clinical and translational metabolomics. Microsampling strategies reduce
invasiveness of sample collection and simplify logistics, thus representing an attractive
alternative to conventional venous blood collection. However, the reliable measurement of
oxylipins from blood microsampling devices remains analytically challenging due to their
low abundance and chemical instability.
In this study, we systematically evaluated and compared multiple blood microsampling
devices including dried blood spots (DBS), volumetric dried blood spots collected using
Capitainer® microsampling cards and volumetric absorptive microsampling (VAMS)
devices, which use an absorptive polymer tip to collect a fixed volume of blood, for
targeted oxylipin analysis. Several extraction protocols and solvent compositions were
investigated to identify the most effective workflow in terms of analyte recovery, matrix
effects, repeatability, and compound coverage. Method performance was assessed using a
targeted C18 LC-MS/MS approach on a triple quadrupole system.
Our results demonstrate that microsampling device selection and extraction strategy
critically influence oxylipin recovery and analytical repeatability. From a single drop of
capillary blood, the optimized workflow provides sensitive, stable, and reproducible
oxylipin profiles across microsampling platforms. This work supports the integration of
microsampling approaches into large-scale population and translational metabolomics
studies, facilitating real-world biomarker discovery while minimizing patient burden

 

Novel roles of the mitochondrial processing peptidase beyond protein import

Danielle Simons, Jean-Francois Trempe

Mitochondria are multifaceted organelles that regulate energy metabolism, apoptosis, immunity, and other cellular processes. Despite having their own genome, most mitochondrial proteins are nuclear encoded, synthesized in the cytosol, and enter mitochondria via N-terminal mitochondrial targeting sequences (MTSs). The mitochondrial processing peptidase (MPP) cleaves these MTSs in the matrix to enable proper protein folding and localization. MPP functions as a heterodimer, with MPPα responsible for substrate recognition and binding and MPPβ acting as a Zn²⁺ metalloprotease that cleaves MTSs. Recent proximity-based BioID studies suggest that MPPα and MPPβ may selectively interact with other proteins as part of novel MPP subcomplexes. Specifically, MPPα was selectively enriched with NUDT8, a coenzyme A (CoA) hydrolase within the mitochondrial matrix whose broader function remains unknown. In addition, AlphaFold3 predicts a high-confidence model for the MPPα-NUDT8 dimer, suggesting this subcomplex assembles directly in cells. Based on these findings, we hypothesize that MPP participates in pathways beyond MTS processing, potentially regulating mitochondrial metabolism or responses to stress. To date, we have developed a recombinant expression system for the purification of human NUDT8 in E. coli, which revealed that MPPα-NUDT8 forms a stable complex with MPPα acting as a NUDT8-specific chaperone, permitting proper folding. Furthermore, through affinity purification-mass spectrometry (AP-MS) we identified each as a top interactor for the other. Our current work examines the proteomic and metabolomic consequences of NUDT8 knockout in cells, with preliminary findings indicating disruptions in energy metabolism and metabolic demands. Taken together, we seek to redefine our understanding of MPP beyond protein import as a dynamic regulator of mitochondrial quality control.

 

CHEMICAL LEACHABLES FROM MICROPLASTICS FROM SINGLE-USE FOOD CONTACT MATERIALS: A NON-TARGETED STUDY UNDER CONSUMER-USE CONDITIONS

Donald Bimpong¹, Zhi Hao Chi1, Nathalie Tufenkji², Stéphane Bayen^1*
1 Department of Food Science and Agricultural Chemistry, McGill University, Canada. 2 Department of Chemistry, Université de Montréal, Canada.
2 Department of Chemical Engineering, McGill University, Canada
Corresponding author E-mail: stephane.bayen@mcgill.ca

Single-use plastics are widely used as food contact materials, yet the identity and extent of chemicals migrating from these materials into food remain poorly characterized. Microplastics generated through consumer use may act as additional sources of chemical release. This study applied non-targeted analysis using liquid chromatography quadruple time-of-flight mass spectrometry (Agilent Technologies) to characterize organic leachables from MPs derived from take-out containers under simulated consumer-use conditions. Single-use plastic containers were subjected to two treatments: (i) scratched to mimic cutlery abrasion, generating surface damage and microplastic fragments, and (ii) non-scratched controls. Leachates were prepared using 20 mL of 3% acetic acid over 10 days to simulate acidic food contact. Feature extraction using Profinder (Agilent Technologies) detected 4,461 molecular features across treatment samples. Statistical analysis using Mass Profiler Professional (Agilent Technologies) of overall chemical profiles revealed chemical features of interest between scratched and non-scratched containers. Furthermore, MS/MS data for selected suspects were acquired using a targeted MS/MS method, followed by structure elucidation with SIRIUS/CSI: Finger ID. Several additives and potential non-intentionally added substances (NIAS) of toxicological concern were confirmed using pure analytical standards. This work highlights the influence of consumer-use conditions on chemical migration from Single-use plastics and demonstrates the application of NTA to characterize complex chemical leachable mixtures.

 

Absolute Quantitation of the Whole-brain Proteome in Preclinical Alzheimer’s Disease Mouse Models Utilizing the SysQuan Platform

Elyssa Baker

Comprehensive characterization of the brain proteome is essential for understanding early molecular mechanisms of Alzheimer’s disease (AD) and evaluating therapeutic effects in preclinical models. While proteomics enables broad interrogation of disease-associated pathways, most workflows rely on relative quantitation, limiting cross-study comparability and sensitivity to subtle biological changes. Here, we implement SysQuan, a stable-isotope tissue-based strategy, for absolute quantitation of the whole-brain proteome in mouse models of AD.

SysQuan uses metabolically 13C-lysine-labelled mouse tissue as an internal standard. Absolute protein concentrations are obtained by calibrating labelled-to-endogenous peptide ratios with synthesized natural (NAT) peptide standards. Whole-brain homogenates were processed using both high-pH reversed-phase fractionation to maximize depth, and streamlined unfractionated workflows to assess scalability. Samples were analyzed by DDA- and DIA-PASEF on a Bruker TimsTOF HT coupled to Evosep One LC, and data were processed using MSFragger, FragPipe, and Skyline. Regional brain analyses were also performed to evaluate anatomical consistency.

Fractionated analyses identified >9,800 protein groups, while unfractionated workflows quantified 5,450 proteins with reduced processing time, supporting higher-throughput applications. Key AD-associated proteins, including MAPT, APP, ApoE, GFAP, NfL, and SNAP-25 were consistently detected across workflows. To support upcoming absolute quantitation, 614 brain-associated NAT peptide standards have been synthesized and optimized to date. Regional analyses quantified 6,200-6,700 proteins per brain subsection. Together, these results demonstrate the feasibility and scalability of SysQuan for quantitative whole-brain proteomics and establish a foundation for future absolute measurements in preclinical AD studies.

 

Detection method for piperazine and its salts concentrations in the air by LC-MS/MS in
accordance with the ACGIH® inhalable fraction and vapor (IFV) notation

Clanephtha Gertilus

Piperazine is a diamine that is commonly found in workplaces in the form of free
base and in the form of salts. These molecules can be inhaled by the workers handling
them. In 2014, the American Conference of Governmental Industrial Hygienist (ACGIH) has
set a new threshold limit value time-weighted average of 8 hours (TLV-TWA) of 0,1 mg/m³ for
the piperazine and its salts with the Inhalable Fraction and Vapor notation (IFV). This new
limit value replaces the previous TLV-TWA of 5 mg/m³ for piperazine dihydrochloride. The
notation IFV implies that the method must be effective in sampling both the inhalable
(particles) and the vapor. Analytical methods for measuring airborne piperazine
dihydrochloride concentrations are inadequate for IFV notation, and no method currently
exists for measuring airborne piperazine and its salt concentrations, complicating exposure
assessments in workplaces using these compounds. The proposed research aims to
develop and validate a method for assessing airborne piperazine and its salts according to
the ACGIH IFV notation and to apply it in real-world workplaces. The impregnated filter
approach, known to be effective for collecting particles and the vapor phase, is used with an
IOM sampling device. Two derivatization reagents were tested: 1-naphthysothiocyanate
(NIT) and dansyl chloride (DNS-Cl). The IOM sampler was adapted and validated using ISO
23861 as a guide. Additional tests were carried out on the sampler to ensure the efficient
capture of piperazine in suspension and to define the sampling parameters that allows the
optimal determination of the concentration of piperazine and its salts.

 

Diol-Specific Boronic Acid Mass Tags for the Detection and Imaging of Glucose by MALDI MSI

Justine Gatein, Nassim Maarouf Mesli, William D. Lubell and Pierre Chaurand

The advancement of on-tissue chemical derivatization (OTCD) techniques for matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) of endogenous metabolites in tissues has attracted significant interest due to their benefits in enhancing detection sensitivity and ionization efficiency for poorly ionizable or/and low-abundance metabolites. Amongst these, glucose is of high interest because of its implication in numerous diseases. OTCD with boronic acids enables the selective targeting of diols while enhancing detection sensitivity. A new boronic acid mass tags containing a positively charged amine function was developed to compensate for the low ionization efficiency of glucose to enhance its detection in MSI. OTCD was performed by spray deposition of the mass tag (5 ml/ml in 50%MeOH) followed by 2′,6′dihydroxyacetophenone matrix. MALDI MSI was acquired in positive ion mode. This method was compared to direct MALDI MSI for glucose using N-(1-naphthyl) ethylenediamine dinitrate matrix doped with 20 mM KCl. Glucose was successfully detected in kidney, ovaries and brain tissue sections (12 µm thick) by MALDI MSI at a spatial resolution of 25 µm with both methods. OTCD proved to be more sensitive by ~4-fold for brain tissue sections, while for ovaries and kidney tissue, the detection improvements were ~4.4-fold and ~3.4-fold respectively. High spatial resolution images were also obtained at 10 and 5 µm on brain tissue section displaying the expected glucose distribution and without delocalization or edge effects. By providing spatially resolved metabolic information, this method will offer valuable insights into disease-specific metabolic reprogramming and may contribute to the development of more effective, targeted therapeutic interventions.

 

Enhancing the Separation of Isomeric Antigens in Acute Myeloid Leukemia
Using Alternative Fragmentation Methods and Differential Ion Mobility

Léa Christophe^1,2, Eric Bonneil¹, Chantal Durette¹, Cristina Mirela Pascariu¹, Joel
Lanoix¹, Marie-Pierre Hardy¹, Krystel Vincent¹, Claude Perreault^1,3 and Pierre Thibault^1,2

1. Institute of Research in Immunology and Cancer, Université de Montréal, Montréal, Canada.
2. Department of Chemistry, Université de Montréal, Montréal, Canada.
3. Department of Medicine, Université de Montréal, Montréal, Canada.

Despite advances in chemotherapy, acute myeloid leukemia (AML) remains associated with
poor clinical outcomes. The identification of tumor‑specific antigens (TSAs) offers a promising
avenue for targeted cancer immunotherapies, but robust TSA profiling remains technically
demanding. In particular, the immunopeptidome is enriched with MHC-associated peptides
(MAPs) that have isomeric variant(s) giving rise to co-eluting peptides with similar
fragmentation patterns. Conventional LC-MS/MS workflows often lack the resolving power to
separate these isomers, leading to ambiguous assignments and reduced confidence in peptide
identification.
To address this challenge, we developed an advanced mass spectrometry workflow that
integrates liquid chromatography (LC), high-field asymmetric waveform ion mobility
spectrometry (FAIMS), MS³-based fragmentation using collision-induced dissociation (CID)
and electron-transfer/higher-energy collision dissociation (EThcD). This combined strategy
improves the separation and sequencing of isomers that are otherwise difficult to distinguish.
We benchmarked and optimized the method using a library of synthetic peptides comprising
AML‑associated MAPs, their isomeric variants, and peptides containing one or more leucine
or isoleucine residues positioned at distinct sites along the backbone. The optimized method
was then tested using targeted PRM analyses on extracts from 2 million BLCL‑4 cells and 100
million AML cells (MOLM13, SKM1, and THP1). In total, 232 Ile/Leu residues from 112
synthetic MAPs and 189 Ile/Leu residues from the 100 most abundant MAPs identified in
BLCL-4 cells were correctly assigned. Notably, the complementarity of EThcD and MS³-CID
enabled unambiguous discrimination of the isomeric pair EILELLNQR/EILEILQNR in SKM1 cells.
This integrated method resolves key sources of ambiguity in immunopeptidomics and enable
reliable discrimination of isomeric MAPs, supporting more accurate TSA identification in AML
cells.

 

NON-TARGETED STRATEGIES FOR ASSESSMENT OF CHEMICAL CONTAMINANTS RELEASED
FROM REUSABLE FOOD BAGS UNDER CONSUMER USAGE

Louis Thomas

Reusable plastic food bags are intended to reduce plastic waste generated by single-use
alternatives. However, these reusable food and beverage containers may serve as under
recognized sources of chemical residues and contaminants, potentially leading to consumer
exposure. Throughout their lifecycle, these bags undergo multiples weathering steps such as
washing, scratching, microwave heating or freezing storage. Due to these processes, IAS or NIAS
(Intentionally or Non-Intentionally Added Substances) contaminants coming from the plastic
formulation may subsequently migrate into food matrices. As these products become more
popular in North America, concerns are mounting on whether these bags represent a truly
economical and safe alternative.
In this study, critical conditions have been determined on one brand using 3% Acetic Acid as food simulant mimicking the potential transfer to food and applying dishwashing and scratching conditions. Food simulant was then analyzed using liquid chromatography coupled with quadrupole time-of-flight mass spectrometry (LC-QTOF) and using non-targeted strategies. Data treatment processes were focused on the comparison of chemical profiles between weathering conditions applied and time of migration. Some contaminants are only seen in one condition applied (Dishwashing 10 cycles, Dishwashing 5 cycles, none) while other contaminants are present in all of them. The usage of dishwashing and the number of washing cycles quantitatively impacted the release of contaminants. Dishwashing 5 cycles has been determined as the most critical condition in our testing. MS/MS fragmentation has been performed on the features of interest resulting in the detection of contaminants of concern.

 

High-Sensitivity Mass Tags for On-Tissue Steroid Imaging via MALDI MSI

Rachel S. Pryce, Nassim Maarouf, William D. Lubell, Pierre Chaurand
Université de Montréal, Montréal, QC, Canada

Introduction:
Mass Spectrometry Imaging (MSI) is a powerful technique for determining the localization and relative abundances of compounds in tissue. Steroids are of interest in the investigation of disease, as they have been linked to various cancers, cardiovascular and autoimmune disorders. However, steroids are challenging to detect using standard MSI protocols due to low ionization efficiency and low abundance. While on-tissue chemical derivatization can enhance signal, it often lacks sensitivity. In this study, we present a high-throughput platform for rapidly screening different potential mass tags to develop molecules tailored to MALDI MSI, which enables more robust and sensitive steroid imaging.

Methods:
In this study, ketone-containing steroids were targeted using a modular reactive tag design, combining a hydrazine moiety with a charged UV absorbing modules. To facilitate rapid screening of several potential structures, the charged modules were bond to cholesterol, forming stable adducts that were tested for detectability without having to optimize reactivity. The best adducts were selected and synthesized into mass tags. These mass tags were then tested on rabbit adrenal gland for the detection of steroids.

Results:
Results from screening the adducts showed a potential of a 10-fold to 30-fold increase in sensitivity from the commercially available Girard’s reagent P (GP), that is widely used for steroid detection in mass spectrometry. After synthesizing the mass tags, several continued to show significantly higher signal when compared to GP, with one of the tags showing a 3-fold increase in corticosterone, 12-fold in 11-deoxycorticosterone, and nearly 20-fold in estrone detection. These tags greatly outperform GP for the detection and mapping of ketone steroids.

 

An LC-MS/MS method to detect metabolites involved in cysteine metabolism and oxidative stress

Sai Ittoo, Nathan Ghafari, Lekha Sleno
Université du Québec à Montréal, Department of Chemistry

Numerous studies have shown that cysteine metabolism and oxidative stress are impacted by several pathologies. Being able to quantify these metabolites can thus allow the monitoring of perturbations in metabolism. However, cysteine and its metabolites present numerous analytical challenges, particularly because they are very polar small molecules and often include oxidizable sites, making them potentially unstable. As part of this project, we developed an approach using high-resolution tandem mass spectrometry coupled with liquid chromatography. To enable the detection of 40 metabolites of interest, the analysis was performed with or without derivatization using two complementary chromatographic methods. This method has been applied to studying perturbations in metabolism in yeast mutants.

 

Development and Evaluation of Machine Learning Models to Screen
Environmental Risks of Chemical Features Based on Information
Gained from Non-targeted Chromatography-Mass Spectrometry Analysis

Samira Norouzi^1,2, Paula Gomez^1,2, Xianming Zhang^1,2*
1 Department of Chemistry and Biochemistry, Concordia University, Montréal, Québec H4B 1R6,
Canada
2 Centre for Research in Molecular Modelling (CERMM), Concordia University, Montréal,
Québec H4B 1R6, Canada

Mass spectrometry (MS) is a powerful tool for detecting and analyzing chemicals in complex
environmental samples, but interpreting raw mass spectra for environmental risk assessments
remains challenging. In this study, we developed and evaluated classification models to predict
bioaccumulation, toxicity, and persistence to connect MS data to environmental chemical
endpoints. PyCaret, a Python-based ML package, was sued to streamline the development and
evaluation of predictive models. Through hyperparameter tuning and 10-fold cross-validation, the
models demonstrated strong predictive performance, achieving an accuracy of over 0.85 on test
sets. For bioaccumulation and air-persistence classification, Extra Trees Classifier (ET) model
exhibited significantly superior predictive performance, with accuracies of 0.90 and 0.93, and
AUC values of 0.96 and 0.98, respectively. Catboost Classifier (CAT) model outperformed others
in water-persistence and toxicity classification, with accuracy and AUC of 0.92 and 0.97 for waterpersistence models, and 0.93 and 0.98 for toxicity classification. No significant difference (< 5 %)
between the test and training results suggest that the models are well-generalized, with no evidence
of overfitting or underfitting during the training process. For all models, several fractions
(features) were identified as pivotal in models’ predictive performance. The fractions
corresponding to the mass-to-charge ratios (m/z) of M+2, M+4, M+5, M-68, and M-70 were found
to have a significant impact on the model’s ability to classify and predict outcomes accurately.
Some of these fractions may be linked to halogens like chlorine (Cl) or bromine (Br), which create
specific isotopic patterns in mass spectrometry. This finding matches the known environmental
toxicity, bioaccumulation, and persistence of halogenated compounds. This work highlights the
ability of ML models to bridge the gap between MS data and chemical risk assessments, which
provides an approach for predicting ecological impact even in the absence of complete structural information.

 

Mass Confusion: Resolving Structural Isomers of 3,5-Dihydroxybenzoic acid and 3,5-Dihydroxyphenylpropanoic acid

Vasudevan Karanghat, Concordia University

Alkylresorcinols are phenolic lipids predominantly found in whole-grain and gluten-containing foods such as wheat and rye. Following ingestion, they are metabolized to 3,5-dihydroxyphenylpropanoic acid (3,5-DHPPA) and subsequently to 3,5-dihydroxybenzoic acid (3,5-DHBA). These two compounds serve as biomarkers of whole-grain, gluten, and dietary fiber intake, and thus, their levels in human body can reliably point to their dietary habits. Urine is a preferred biological matrix for this assessment due to its non-invasive collection and relatively high metabolite concentrations. Accurate quantification of these analytes, however, is complicated by the presence of multiple endogenous structural isomers that can co-elute under conventional chromatographic conditions. In this study, a selective liquid chromatography–high resolution mass spectrometry (LC-HRMS) method was developed to resolve the isomers and quantify the 3,5-DHBA and 3,5-DHPPA in human urine. The chromatographic separation of isomers from the analytes of interest was optimized using a reversed-phase C18 column, followed by comparison of three sample preparation methods (dilution, liquid-liquid extraction, and solid-phase extraction). The resulting method enhances isomer discrimination, providing a reliable platform for dietary biomarker assessment and supporting applications in precision nutrition and clinical research.

 

Date
Date(s) - February 17, 2026
6:00 pm - 8:00 pm

Emplacement / Location
Université de Montréal - Campus MIL (Beer and pizza at 18h, conference at 19h in A-4502)