About

John Q. Trojanowski, MD, PhD, is a professor in the Department of Pathology and Laboratory Medicine at the University of Pennsylvania School of Medicine. He serves as the Director of the National Institute on Aging Alzheimer's Disease Center Core and the Institute on Aging at the University of Pennsylvania, as well as the Director of the Center on Alpha-synuclein Strains in Alzheimer’s Disease & Related Dementias. His research centers on the molecular mechanisms of neuron dysfunction, degeneration, and death in normal aging and neurodegenerative diseases, including Alzheimer's disease, Parkinson's disease, frontotemporal dementias with or without parkinsonism, and motor neuron disease. Trojanowski employs morphological, genetic, molecular biological, and biochemical methods to study human CNS and PNS tissue samples, cell lines, synthetic proteins, and transgenic models to understand disease pathology and progression.

Research topics

• Medicine
• Biology
• Neuroscience
• Pathology
• Internal medicine
• Psychology
• Genetics
• Biochemistry
• Bioinformatics
• Cell biology
• Radiology
• Immunology
• Psychiatry
• Artificial Intelligence
• Computer Science
• Gerontology
• Computational biology
• Chromatography
• Oncology
• Clinical psychology
• Chemistry
• Physics
• Developmental psychology

Selected publications

• Deciphering distinct genetic risk factors for FTLD-TDP pathological subtypes via whole-genome sequencing

  Nature Communications · 2025-04-25 · 13 citations

  articleOpen access

  Frontotemporal lobar degeneration with neuronal inclusions of the TAR DNA-binding protein 43 (FTLD-TDP) is a fatal neurodegenerative disorder with only a limited number of risk loci identified. We report our comprehensive genome-wide association study as part of the International FTLD-TDP Whole-Genome Sequencing Consortium, including 985 patients and 3,153 controls compiled from 26 institutions/brain banks in North America, Europe and Australia, and meta-analysis with the Dementia-seq cohort. We confirm UNC13A as the strongest overall FTLD-TDP risk factor and identify TNIP1 as a novel FTLD-TDP risk factor. In subgroup analyzes, we further identify genome-wide significant loci specific to each of the three main FTLD-TDP pathological subtypes (A, B and C), as well as enrichment of risk loci in distinct tissues, brain regions, and neuronal subtypes, suggesting distinct disease aetiologies in each of the subtypes. Rare variant analysis confirmed TBK1 and identified C3AR1, SMG8, VIPR1, RBPJL, L3MBTL1 and ANO9, as novel subtype-specific FTLD-TDP risk genes, further highlighting the role of innate and adaptive immunity and notch signaling pathway in FTLD-TDP, with potential diagnostic and novel therapeutic implications. Here the authors identify TNIP1 as a risk factor for a fatal neurodegenerative disorder and discover specific genetic loci associated with the three main subtypes of this disorder. The findings highlight distinct disease mechanisms, emphasizing the roles of immunity and the notch signaling pathway.

  PublisherOA PDFDOI

• Traumatic brain injury or head impacts from contact sports are associated with tau astrogliopathy

  Brain · 2025-04-02 · 6 citations

  articleOpen access

  Exposure to traumatic brain injury (TBI) and/or repetitive head impacts (RHI) increase the risk of a range of neurodegenerative pathologies, including chronic traumatic encephalopathy neuropathologic change (CTE-NC). Astrocytic tau pathology reminiscent of ageing-related tau astrogliopathy is a component feature of CTE-NC in many cases. Yet the relationship between TBI/RHI exposure and wider tau astrogliopathy, beyond that of CTE-NC, remains poorly characterized. Autopsy-derived material from 556 individuals was selected to include cases with a history of moderate or severe traumatic brain injury (survival >6 months, n = 77) or a history of participation in contact sports (n = 45), for comparison with uninjured controls with (n = 397) or without (n = 37) neuropathologically confirmed neurodegenerative disease. Representative tissue sections from multiple brain regions were then immunostained for hyperphosphorylated tau (p-tau; PHF-1) and assessed in accordance with the harmonized evaluation criteria for ageing-related tau astrogliopathy. PHF-1-immunoreactive thorn-shaped astrocytes were observed more frequently in contact sports participants (75.6%) versus controls with (32.5%; P < 0.001) and without (8.1%; P < 0.001) neurodegenerative disease. In addition, although the prevalence of thorn-shaped astrocytes following moderate/severe TBI (32.5%) was similar to neurodegenerative disease controls, regression analyses demonstrated increased odds of thorn-shaped astrocytes, when adjusting for age and sex (odds ratio 2.42, 95% confidence interval 1.29-4.54). These findings were observed regardless of whether the pathognomonic lesion of CTE-NC was present in the regions examined. Intriguingly, although subpial thorn-shaped astrocytes at sulcal depths were occasionally observed in aged controls with (3.6%) and without (2.8%) neurodegenerative disease, this pathology was considerably more common following RHI/TBI (42.2%; P < 0.001). These findings support a history of RHI or TBI as an independent risk factor for the development of thorn-shaped tau astrogliopathy, over and above ageing-related tau astrogliopathy observed in ageing and wider neurodegenerative disease. Moreover, trauma might be associated with thorn-shaped astrocytes within specific distributions, including the subpial region of the cortical sulcal depths. The clinical significance of these observations will be important to determine.

  PublisherOA PDFDOI

• Automated deep learning segmentation of high-resolution 7 Tesla postmortem MRI for quantitative analysis of structure-pathology correlations in neurodegenerative diseases

  Imaging Neuroscience · 2024 · 12 citations

  • Artificial Intelligence
  • Computer Science
  • Artificial Intelligence

  (https://pulkit-khandelwal.github.io/exvivo-brain-upenn/).

  PublisherOA PDFDOI

• High‐resolution 7 tesla postmortem MRI for quantitative analysis of structure‐pathology correlations in neurodegenerative diseases

  Alzheimer s & Dementia · 2024-12-01

  articleOpen access

  Abstract Background Postmortem MRI allows brain anatomy to be examined at high‐resolution linking pathology with morphometric measurements. However, automated methods for analyzing postmortem MRI are not well developed. We present a deep learning‐based framework for automated segmentation of cortical mantle, subcortical structures (caudate, putamen, globus pallidus, and thalamus), white matter hyperintensities (WMH), and normal appearing white matter in (n=135) postmortem human brain tissue specimens (Table 1) imaged at 0.3 mm 3 T2w 7T spanning Alzheimer’s disease and related dementias. We show generalizing capabilities across unseen images acquired at 0.28 mm 3 and 0.16 mm 3 T2*w 7T FLASH sequence. We report associations between localized cortical thickness and volumetric measurements across key regions and semi‐quantitative neuropathological ratings. Method A deep learning model was trained on manually segmented images to produce automated whole‐brain hemisphere segmentations (Figure 1) with a post‐hoc topological correction step to delineate buried sulcus. We report regional patterns of association between localized cortical thickness at 16 anatomical locations and neuropathology ratings of regional measures of p‐tau, neuronal loss; global amyloid‐β, Braak staging, and CERAD ratings obtained from histology data in a subset (n=82) with AD continuum diagnoses. We correlate subcortical volumetry and regional cortical thickness with WMH burden (Figure 2) for the entire cohort (n=135). All analyses include age, sex, and postmortem interval as covariates. Result Tau pathology in Braak regions play an important role in cortical atrophy and cognitive decline in AD. Significant negative correlations (Figure 2) between p‐tau and cortical thickness were found in angular gyrus and midfrontal regions. Cortical thickness showed significant negative correlation with neuronal loss in Brodmann area (BA) 35 and entorhinal cortex (ERC), and with Braak staging in midfrontal, ERC and BA35, regions consistent with high p‐tau uptake in PET imaging with cortical thickness on MRI. High WMH volume disrupts structural and functional connectivity impacting memory. Significant negative correlation of WMH volume with thickness in posterior cingulate and superior temporal regions was observed. Conclusion Our automated postmortem MRI framework provides geometrically accurate segmentations of several key brain regions. Our analysis linking morphometry and pathology measurements demonstrated that automated segmentation and analysis of postmortem MRI can complement and inform antemortem neuroimaging studies.

  PublisherOA PDFDOI

• Developing an anatomically valid segmentation protocol for anterior regions of the medial temporal lobe

  Alzheimer s & Dementia · 2024-12-01

  articleOpen access

  Abstract Background The anterior portion of the MTL is one of the first regions targeted by pathology in sporadic Alzheimer’s disease (AD) indicating the potential for imaging metrics from this region to serve as valuable imaging biomarkers. However, most existing automated approaches for MTL segmentation do not incorporate anterior MTL subregions, and the few that do fail to account for its complex anatomical variability. Leveraging a unique postmortem dataset consisting of histology and structural MRI scans we aimed to develop an anatomically valid segmentation protocol for anterior entorhinal cortex (ERC), Brodmann Area (BA) 35, and BA36 and apply it for automated MTL segmentation of in vivo 3 tesla (T) MRI. Method We included 20 cases between 61 to 97 years of age (50% females) with and without neurodegenerative diseases (11 vs. 9 cases) to ensure broad generalizability of the developed protocol. Postmortem digitized MTL Nissl‐stained coronal histology serial sections from these cases were registered to same‐subject 0.2×0.2×0.2‐mm 3 9.4T postmortem MRI and annotated by an expert neuroanatomist. To develop the segmentation protocol, we determined the location of the histological borders of interest in relation to anatomical landmarks observable on in vivo MRI. The protocol was first applied manually to 29 3T in vivo MRI scans and then used to train an automatic segmentation method T1‐ASHS (Automatic Segmentation of Hippocampal Subfields). Intra‐rater reliability of a manual rater and five‐fold cross‐validation accuracy of T1‐ASHS were assessed with the Dice Similarity Index (DSI). Result Segmentation rules for the borders of ERC, BA35 and BA36 based on systematic analysis of inter‐landmark distances on histological sections are shown in Figure 1. Intra‐rater reliability for the manual rater applying these rules to 15 in vivo 3T MRI scans was high (Table‐1; Figure‐2). Comparing manual segmentations with the automated ones generated by T1‐ASHS showed moderate reliability, reflecting the challenging anatomy of this region. However, segmentation accuracy for the whole MTL including the newly added region was comparable to the previously reported accuracy for MTL without this region (Table‐1). Conclusion Future work will examine trhe utility of morphometric measures of anterior MTL regions enabled by this protocol for early AD.

  PublisherOA PDFDOI

• Network Proteomics of the Lewy Body Dementia Brain Reveals Presynaptic Signatures Distinct from Alzheimer’s Disease

  bioRxiv (Cold Spring Harbor Laboratory) · 2024-01-24 · 7 citations

  preprintOpen access

  Lewy body dementia (LBD), a class of disorders comprising Parkinson's disease dementia (PDD) and dementia with Lewy bodies (DLB), features substantial clinical and pathological overlap with Alzheimer's disease (AD). The identification of biomarkers unique to LBD pathophysiology could meaningfully advance its diagnosis, monitoring, and treatment. Using quantitative mass spectrometry (MS), we measured over 9,000 proteins across 138 dorsolateral prefrontal cortex (DLPFC) tissues from a University of Pennsylvania autopsy collection comprising control, Parkinson's disease (PD), PDD, and DLB diagnoses. We then analyzed co-expression network protein alterations in those with LBD, validated these disease signatures in two independent LBD datasets, and compared these findings to those observed in network analyses of AD cases. The LBD network revealed numerous groups or "modules" of co-expressed proteins significantly altered in PDD and DLB, representing synaptic, metabolic, and inflammatory pathophysiology. A comparison of validated LBD signatures to those of AD identified distinct differences between the two diseases. Notably, synuclein-associated presynaptic modules were elevated in LBD but decreased in AD relative to controls. We also found that glial-associated matrisome signatures consistently elevated in AD were more variably altered in LBD, ultimately stratifying those LBD cases with low versus high burdens of concurrent beta-amyloid deposition. In conclusion, unbiased network proteomic analysis revealed diverse pathophysiological changes in the LBD frontal cortex distinct from alterations in AD. These results highlight the LBD brain network proteome as a promising source of biomarkers that could enhance clinical recognition and management.

  PublisherOA PDFDOI

• Additional file 2 of Alzheimer’s disease tau is a prominent pathology in LRRK2 Parkinson’s disease

  Open MIND · 2024-01-01

  dataset

  Additional file 2. Table S1. Clinical and Pathological Data for Cases.

  DOI

• Pathologic burden goes with the flow: MRI perfusion and pathologic burden in frontotemporal lobar degeneration due to tau

  Imaging Neuroscience · 2024-03-01 · 1 citations

  articleOpen access

  Regional cerebral blood flow (CBF) changes quantified using arterial spin labeling (ASL) are altered in neurodegenerative disorders such as frontotemporal lobar degeneration due to tau (FTLD-tau), but the relationship between ASL CBF and pathologic burden has not been assessed. Our objective was to determine whether regional ASL CBF acquired antemortem in patients with FTLD-tau is related to pathologic burden measured at autopsy in those same regions in the same patients to directly test the imaging-pathology relationship. In this case-control study, data were acquired between 3/4/2010 and 12/16/2018. Data processing and analysis were completed in 2023. Twenty-one participants with autopsy-confirmed FTLD-tau (N = 10 women, mean[SD] age 67.9[7.56] years) along with 25 control participants (N = 15 women, age 64.7[7.53]) were recruited through the cognitive neurology clinic at the University of Pennsylvania. All participants had ASL and T1-weighted images collected antemortem. ASL images were processed to estimate CBF and T1-weighted images were processed to estimate gray matter (GM) volumes in regions corresponding to regions sampled postmortem. Digital quantification of pathologic burden was performed to find the percent area occupied (%AO) of pathologic FTLD-tau at autopsy. Regional CBF and GM volumes were both related to pathologic burden in the same regions from the same participants. Strengths of model fits of imaging measures to pathologic burden were compared. CBF in FTLD-tau and controls were compared, with results considered significant at p < 0.05 after Bonferroni correction. We found that relative to controls, FTLD-tau displayed hypoperfusion in anterior cingulate, orbitofrontal, middle frontal, and superior temporal regions, as well as angular gyrus. For patients with FTLD-tau regional CBF was significantly associated with pathologic burden (beta = -1.07, t = -4.80, p < 0.005). Models including both GM volume and CBF provided significantly better fits to pathologic burden data than single modality models (p < 0.05, Bonferroni-corrected). Our results indicate that reduced CBF measured using ASL MRI is associated with increased pathologic burden in FTLD-tau and adds complementary predictive value of pathologic burden to structural MRI.

  PublisherOA PDFDOI

• Deciphering Distinct Genetic Risk Factors for FTLD-TDP Pathological Subtypes via Whole-Genome Sequencing

  medRxiv · 2024-06-25 · 5 citations

  preprintOpen access

  Abstract Frontotemporal lobar degeneration with neuronal inclusions of the TAR DNA-binding protein 43 (FTLD-TDP) is a fatal neurodegenerative disorder with only a limited number of risk loci identified. We report our comprehensive genome-wide association study as part of the International FTLD-TDP Whole-Genome Sequencing Consortium, including 985 cases and 3,153 controls, and meta-analysis with the Dementia-seq cohort, compiled from 26 institutions/brain banks in the United States, Europe and Australia. We confirm UNC13A as the strongest overall FTLD-TDP risk factor and identify TNIP1 as a novel FTLD-TDP risk factor. In subgroup analyses, we further identify for the first time genome-wide significant loci specific to each of the three main FTLD-TDP pathological subtypes (A, B and C), as well as enrichment of risk loci in distinct tissues, brain regions, and neuronal subtypes, suggesting distinct disease aetiologies in each of the subtypes. Rare variant analysis confirmed TBK1 and identified VIPR1 , RBPJL , and L3MBTL1 as novel subtype specific FTLD-TDP risk genes, further highlighting the role of innate and adaptive immunity and notch signalling pathway in FTLD-TDP, with potential diagnostic and novel therapeutic implications.

  PublisherOA PDFDOI

• MAPT H2 haplotype and risk of Pick's disease in the Pick's disease International Consortium: a genetic association study

  The Lancet Neurology · 2024-04-15 · 23 citations

  articleOpen access

  BACKGROUND: Pick's disease is a rare and predominantly sporadic form of frontotemporal dementia that is classified as a primary tauopathy. Pick's disease is pathologically defined by the presence in the frontal and temporal lobes of Pick bodies, composed of hyperphosphorylated, three-repeat tau protein, encoded by the MAPT gene. MAPT has two distinct haplotypes, H1 and H2; the MAPT H1 haplotype is the major genetic risk factor for four-repeat tauopathies (eg, progressive supranuclear palsy and corticobasal degeneration), and the MAPT H2 haplotype is protective for these disorders. The primary aim of this study was to evaluate the association of MAPT H2 with Pick's disease risk, age at onset, and disease duration. METHODS: In this genetic association study, we used data from the Pick's disease International Consortium, which we established to enable collection of data from individuals with pathologically confirmed Pick's disease worldwide. For this analysis, we collected brain samples from individuals with pathologically confirmed Pick's disease from 35 sites (brainbanks and hospitals) in North America, Europe, and Australia between Jan 1, 2020, and Jan 31, 2023. Neurologically healthy controls were recruited from the Mayo Clinic (FL, USA, or MN, USA between March 1, 1998, and Sept 1, 2019). For the primary analysis, individuals were directly genotyped for the MAPT H1-H2 haplotype-defining variant rs8070723. In a secondary analysis, we genotyped and constructed the six-variant-defined (rs1467967-rs242557-rs3785883-rs2471738-rs8070723-rs7521) MAPT H1 subhaplotypes. Associations of MAPT variants and MAPT haplotypes with Pick's disease risk, age at onset, and disease duration were examined using logistic and linear regression models; odds ratios (ORs) and β coefficients were estimated and correspond to each additional minor allele or each additional copy of the given haplotype. FINDINGS: We obtained brain samples from 338 people with pathologically confirmed Pick's disease (205 [61%] male and 133 [39%] female; 338 [100%] White) and 1312 neurologically healthy controls (611 [47%] male and 701 [53%] female; 1312 [100%] White). The MAPT H2 haplotype was associated with increased risk of Pick's disease compared with the H1 haplotype (OR 1·35 [95% CI 1·12 to 1·64], p=0·0021). MAPT H2 was not associated with age at onset (β -0·54 [95% CI -1·94 to 0·87], p=0·45) or disease duration (β 0·05 [-0·06 to 0·16], p=0·35). Although not significant after correcting for multiple testing, associations were observed at p less than 0·05: with risk of Pick's disease for the H1f subhaplotype (OR 0·11 [0·01 to 0·99], p=0·049); with age at onset for H1b (β 2·66 [0·63 to 4·70], p=0·011), H1i (β -3·66 [-6·83 to -0·48], p=0·025), and H1u (β -5·25 [-10·42 to -0·07], p=0·048); and with disease duration for H1x (β -0·57 [-1·07 to -0·07], p=0·026). INTERPRETATION: The Pick's disease International Consortium provides an opportunity to do large studies to enhance our understanding of the pathobiology of Pick's disease. This study shows that, in contrast to the decreased risk of four-repeat tauopathies, the MAPT H2 haplotype is associated with an increased risk of Pick's disease in people of European ancestry. This finding could inform development of isoform-related therapeutics for tauopathies. FUNDING: Wellcome Trust, Rotha Abraham Trust, Brain Research UK, the Dolby Fund, Dementia Research Institute (Medical Research Council), US National Institutes of Health, and the Mayo Clinic Foundation.

  PublisherOA PDFDOI

Recent grants

• Axonal Pathology and TBI-Related Neurodegeneration (TReND)

  NIH · $5.1M · 2016–2027

• NIH Grant R01CA036245

  NIH · $1.6M · 1999

• NIH Grant P50NS053488

  NIH · $33.7M · 2018

• NIH Grant P01NS034514

  NIH · $10.1M · 2002

• NIH Grant P01AG009215

  NIH · $16.3M · 2011

Frequent coauthors

• Virginia M.‐Y. Lee

  California University of Pennsylvania

  1709 shared

• Leslie M. Shaw

  University of Pennsylvania

  1375 shared

• Michael W. Weiner

  University of California, San Francisco

  975 shared

• Douglas Galasko

  University of California, San Diego

  877 shared

• Brit Mollenhauer

  University of Göttingen

  801 shared

• Murray Grossman

  598 shared

• Kenneth Marek

  Institute for Neurodegenerative Disorders

  586 shared

• Mark Frasier

  Michael J. Fox Foundation

  579 shared

Labs

• Pathology and Laboratory MedicinePI

Awards & honors

• Director, National Institute on Aging Alzheimer's Disease Ce…
• Director, Institute on Aging, University of Pennsylvania, Sc…
• Director, Center on Alpha-synuclein Strains in Alzheimer’s D…