MSACL 2023 Abstract

INTRODUCTION: Frontotemporal dementia (FTD) is a form of dementia leading to language and behavioral dysfunction. In approximately half of the cases of FTD, the underlying pathology is defined by abnormal protein aggregates of transactive-response DNA binding protein 43 (TDP-43) in the frontal and temporal lobe; termed frontotemporal lobar degeneration with TDP-43 pathology (FTLD-TDP). To date, no diagnostic test for FTLD-TDP is available as we do not have fluid or imaging biomarkers that are specific for TDP-43 pathology. TDP-43 itself is not an ideal biomarker in biofluids as TDP-43 is expressed in almost all human tissues. Further, attempts at quantifying “total TDP-43” in biofluids has yielded no discriminatory power for FTLD-TDP. As such, we focused on post-translational modifications of TDP-43 with the aim of identifying disease-specific TDP-43 proteoforms.

OBJECTIVES: Our objective was to identify, characterize and quantify TDP-43 proteoforms that differentiated FTLD-TDP from relevant controls.

METHODS: To complete this objective we performed mass spectrometry analysis of brain tissue from cases with and without TDP-43 pathology. In the discovery phase, TDP-43 proteoforms were investigated via a Brucker Impact II quadrupole time-of-flight high-resolution mass spectrometer. We assessed the sarkosyl-insoluble fraction of frontal lobe brain tissue that is associated with misfolded proteins from immunohistochemically confirmed FTLD-TDP cases (n=13) and controls (n=13). Control tissues included other dementias and neuropathologically unaffected cases. Insoluble brain tissue was fractioned by electrophoresis, to isolate TDP-43 proteoforms corresponding to intact TDP-43 and TDP-43 fragments. In the verification phase, the best performing biomarker was investigated further in a larger cohort (n=51). This was done via a quantitative multiple reaction monitoring (MRM) assay on a SCIEX 6500 triple quadrupole mass spectrometer using heavy-labelled TDP-43 peptide internal standards to obtain absolute peptide concentrations.
RESULTS: In our search for disease-specific TDP-43 proteoforms we found that C-terminal peptides from truncated TDP-43 proteoforms were significantly increased in FTLD-TDP cases compared to controls. The concentration of TDP-43 fragments differentiated FTLD-TDP cases from related dementias and unaffected controls with 78% sensitivity and 100% specificity. In the larger verification cohort and using our quantitative MRM method, the C-terminal proteoforms differentiated cases with and without TDP-43 pathology with 83% sensitivity and 89% specificity.

CONCLUSION: Herein we demonstrated the value of a combined pathology- and proteoform-focused biomarker discovery effort in the identification of biomarkers for FTLD-TDP pathology. Moreover, we were able to swiftly move through the verification stage with our previous establishment of a quantitative TDP-43 proteoform assay, which required only minor modification in order to evaluate new proteoform targets identified during discovery. This work represents the largest proteomic study to date of pathology-confirmed FTLD-TDP and the first to move beyond the discovery stage.