Murray Grossman, M.D., Ed.D. - US grants

Naming, sentence comprehension, and speech fluency are frequently impaired in patients with stroke, probable dementia of the Alzheimer's type, and Parkinson's disease. The purpose of this project is to decompose complex language skills such as these into more fundamental, linguistically- motivated processes, and to begin to establish the cerebral basis for impairments in these processes using modern neurophysiologic techniques. Specifically, complex language skills such as naming will be fractionated into more basic linguistic processes, and the nature of these processes will be validated within and across groups of subjects using paper-and- pencil testing techniques. The cerebral localization of these processes will be determined by several measures of regional cerebral metabolism using positron emission tomography. The contribution of neurochemical indices to language functioning will be established by direct measurements of neurotransmitter levels in the cerebrospinal fluid. The results of the proposed studies are expected to show that complex language skills can be decomposed into more basic processes. Impairments of qualitatively similar processes will be observed across several patient groups. Each group of subjects is expected to manifest a unique clustering of these processing impairments. A deficit in performing a specific linguistic process is expected to be related to reduced metabolism in a particular cerebral region. Networks of brain regions will be seen to work together in order to support the performance of a complex skill like naming. Moreover, changes in neurotransmitter receptor densities and cerebrospinal fluid levels are expected to be correlated with specific language processing impairments. These studies will contribute to our understanding of the cerebral basis for our language capacity, and will provide a rational basis for the development of therapeutic strategies to help individuals with language impairments.

DESCRIPTION: (Adapted from the applicant's abstract) Cognitive and language impairments are frequently observed in the elderly and in patients with Alzheimer's disease. The purpose of the proposed project is to clarify the nature and significance of these deficits, and to investigate their physiologic basis. Specifically, complex skills like picture naming and sentence comprehension will be decomposed into more fundamental, linguistically-and cognitively-motivated components using paper-and-pencil testing techniques which can be equally applied to mildly and severely demented patients. Deficits in performing these specific processes will be characterized in longitudinal and cross-sectional designs, allowing the development of rich profiles encompassing the full breadth of these complex cognitive skills. Positron emission tomography (PET) will be used to monitor cerebral perfusion at baseline and during cognitive activation with discrete intellectual tasks. Cognitive correlates of regional cerebral functioning will be defined, and a physiologically-based method for staging pDAT will be developed. Detailed histopathologic abnormalities in specific, language-related cerebral regions will be studied postmortem and related to end-stage language performance and in vivo PET. The results are hypothesized to show that complex language skills like naming and sentence comprehension can be decomposed into more fundamental linguistic components. Unique profiles of processing deficits for particular components will be seen in clusters of individuals, and we hypothesize that these are expected to change quantitatively as well as qualitatively during the natural history of Alzheimer's disease. Specific brain-behavior relations are hypothesized between regional PET defects evident on baseline scans and discrete cognitive difficulties in aging and dementia. Patterns of regional cerebral activity during PET activation studies will correlate with performance accuracy and lead to a physiologically-based in vivo staging algorithm for Alzheimer's disease. An end-stage deficit in performing specific linguistic skills is expected to be related to the topographic and cytoarchitectonic location and density of neurofibrillary tangles and senile plaques. These studies are expected to provide important guidelines for mapping discrete cognitive processes on to specific cerebral regions, to improve diagnostic and prognostic capabilities, to augment our knowledge of the biology of Alzheimer's disease and the normal aging process, and to help develop rational strategies for remediating well-defined language deficits in the dementias.

To understand the basis for memory and cognitive decline in the healthy elderly and in minor depression. To establish the efficacy of behavioral and pharmacologic treatment programs for memory complaints in the healthy elderly and in patients with minor depression.

DESCRIPTION (adapted from investigator's abstract): The purpose of this study is to investigate the cognitive and physiological basis for cognitive resources such as information processing speed and the selective distribution of attention. Non-demented patients with Parkinson's disease (PD) have cognitive deficits that have been attributed to limitations in cognitive resources, or to modular deficits associated with domain-specific processes. The first Specific Aim investigates these alternatives with on-line measures of single word and sentence processing because language is a familiar domain with high face validity. Group-wide and individual patient analyses are hypothesized to show that PD patients have limitations in information processing speed and directed attention that selectively compromise specific aspects of sentence processing. These findings will contribute to cognitive neuroscience by defining the role of cognitive resources in intellectual functioning from a unique perspective. The second Specific Aim takes advantage of the well-known neurotransmitter and neuroanatomical basis for PD to advance our knowledge of the physiology of cognition. We propose to define the role of dopamine in modular and resource-based aspects of cognition with on-line studies of sentence processing in PD patient who are "on" and "off" the dopamine supplementation. Cerebral recruitment patterns for modular and resource-based processing overlap considerably, and the applicants propose to use functional magnetic resonance imaging (fMRI) to dissociate the anatomic distributions of brain activation for cognitive resources and modular processes during sentence processing challenges in healthy elderly subjects and PD patients. We expect that the dopamine status of PD patients will affect limitations in information processing speed and directed attention that contribute to cognition. The findings will suggest an important pharmacologic strategy for the treatment of a specific cognitive impairment that is common in the elderly. Healthy subjects are expected to recruit partially distinct brain regions for cognitive resources and modular processes during sentence comprehension, including dopaminergic projection zones such as frontal cortex and the striatum of the left hemisphere. PD patients are expected to have limited recruitment of the portion of the normal neural network that supports cognitive resources since PD patients have sentence comprehension difficulty due to their impaired cognitive resources. These findings will contribute to cognitive neuroscience by defining the anatomic basis for cognitive resources and the modular processes that underlie a distributed cerebral network supporting a complex cognitive process.

DESCRIPTION (adapted from investigator's abstract): Our conceptual knowledge includes the long-term representation and processing of word meanings and basic information about objects and actions. The investigators propose behavioral and functional neuroimaging studies of conceptual representations and processing in neurologically intact subjects and patients with Alzheimer's disease (AD) who have impaired semantic memory. Their cognitive model of such processing assumes that there are two modes of categorization: similarity-based categorization, in which one assesses the similarity of a test item to prototypes or exemplars of various categories; and rule-based categorization, in which one assesses if a test item meets the critical conditions specified in a rule that determines category membership. Their neural model of concepts focuses on support for rule-based processing in left dorsolateral frontal cortex (DLFC), support for similarity-based processing in left posterior heteromodal association cortex (PHAC), and the representation of information associated with a concept in modality-specific association cortex such as the ventral temporal lobe. Specific Aim 1 investigates how similarity-based and rule-based categorization are affected in AD with behavioral studies, and examines the neural basis for these processes with fMRI. Specific Aim 2 focuses on the use of categories to guide inductive inference, and investigates different types of category-based induction in AD. They hypothesize that AD patients are impaired at using similarity-based and rule-based processes for categorization and inductive reasoning, and that fMRI studies will associate these limitations with left PHAC and left DLFC, respectively. Specific Aim 3 is concerned not with the processes that operate on conceptual representations, but with the representations themselves. Here they ask whether such representations distinguish between appearance and action features for natural kind and manufactured artifact categories using tasks that require categorization or feature verification, and they investigate the neural basis for these decisions. They hypothesize that AD patients have a deficit with specific features contributing to a category, and this will be associated with interruption of a network involving visual association cortex in the left temporal lobe and left DLFC. Specific Aim 4 involves a comparison of object and action conceptual representations in AD, and investigates the neural basis for this knowledge with fMRI. They hypothesize that the mental organization of action concepts is more susceptible to impairment in AD, and this will be associated with left DLFC. This research program will expand our understanding of impaired semantic memory in AD, and provide important information for cognitive neuroscience concerning the cognitive and neural basis for normal concept processing and representation.

The previous grant period developed clinical and imaging criteria for frontotemporal dementia (FTD) subgroups, and tested hypotheses about the cognitive and neural basis for semantic and grammatical aspects of language. In the current grant period, we propose to continue hypothesis-driven studies of progressive aphasia and systematically investigate the behavioral and neural basis for the social disorder in FTD. SPECIFIC AIM 1: We will examine clinically asymptomatic family members of patients with a tau mutation to define a prodromal form of FTD. We will also pursue longitudinal behavioral and imaging studies to characterize FTD patients with a social disorder, paralleling our work with progressive aphasia. SPECIFIC AIM 2: Semantic deficits in Semantic Dementia (SD) have been related to degraded object knowledge, but the pattern of lost object knowledge is inconsistent. We will test the hypothesis that the diagnostic value of object features is impaired in SD. Cognitive and fMRI data will determine the basis for this disorder, and test whether this is related to left ventral temporal disease. SPECIFIC AIM 3: Quantitative acoustic speech analyses and grammatical agreements, combined with fMRI, will test the hypothesis that there are 2 forms of Progressive Non-fluent Aphasia (PNFA) - impaired grammar in comprehension and expression related to left ventral inferior frontal (BA 45/47) disease, and a dysarthric articulatory disorder related to left frontal opercular and insula disease. SPECIFIC AIM 4: Our two-component model of social functioning includes features of social knowledge, and executive resources that mediate implementing this knowledge in a social context. We will test the hypothesis that a social disorder in FTD is due to poor inhibitory control and limited working memory that govern factors such as the rules associated with social knowledge. We will use cognitive and fMRI data to test the hypothesis that a social disorder is due to interruption of a large-scale neural network that includes right medial orbital frontal cortex and anterior prefrontal cortex.

DESCRIPTION (provided by applicant): We seek converging evidence from cognitive studies of non-demented patients with Parkinson's disease (PD), electrocortical event-related potentials (CEPs), and functional magnetic resonance imaging (fMRI) to test our interactive neurocognitive model of core cognitive processes and executive resources in comprehension. Specific Aim 1 manipulates executive resources (working memory, strategic planning, inhibitory control) in ambiguous sentences. PD patients' impaired sentence comprehension will be related to limitations in specific executive resources. Resource-related slowing of CEPs will be seen in PD for the same material. fMRI in young subjects with this material will recruit interactive neural networks for sentence processing: left ventral inferior frontal cortex (vIFC) and left posterolateral temporal cortex (PLTC) for core language processes, and specific cognitive resources in left dorsal IFC (dIFC), prefrontal cortex, striatum, and right PLTC. To compensate for age- and disease-related resource limitations, healthy seniors and PD patients will up-regulate resource-related networks, but we expect no change in the core sentence processing network. Specific Aim 2 tests a material-neutral deficit for rules that depends on implicit memory. We examine regular and irregular morphology in verbs and nouns, and assess non-linguistic concept acquisition mediated by implicit- or rule-based learning. PD patients will show a material-specific deficit for rules in verbs. fMRI in young subjects will recruit left vIFC only for regular verb morphology, and dIFC for decision-making resources. d]FC will be up-regulated in aging and PD. Specific Aim 3 assesses the generalizeability of our model to prosody comprehension. PD patients judge acoustically simple and complex prosody stimuli at baseline and during a secondary task. Restricted resources will limit PD patients' comprehension of complex prosody. fMRI in young subjects will recruit orbital frontal and dIFC only for complex prosody, and dIFC will be up-regulated in aging and PD. Our data support a componential neurocognitive architecture consisting of dynamically interactive networks modified to process sentences depending on available resources and relative demand.

[unreadable] DESCRIPTION (provided by applicant): This competitive renewal studies the cognitive and neural basis for semantic memory deficits in Alzheimer's disease (AD). Our cognitive model includes distributed feature knowledge and categorization processes that integrate this knowledge. This maps onto a large-scale neural network where feature knowledge is biased to a particular anatomic distribution, and categorization processes are supported by prefrontal cortex (PFC) and posterior temporal-parietal heteromodal association cortex (PHAC). We found in the prior grant cycle that AD patients have impaired rule-based categorization, and fMRI work related this in part to PFC. Specific Aim 1 examines the role of specific executive resources (selective attention/inhibitory control, working memory, and task-switching) during rule-based categorization in AD, and assess the neural substrate for this with fMRI. We expect poor rule-based categorization in AD that is due to limited and dissociable executive resources, fMRI will relate each executive resource to different portions of PFC. We also found in the prior grant cycle that AD patients have limited similarity-based categorization under specific circumstances, and fMRI related this in part to PHAC. Specific Aim 2 assesses factors that affect the organization of features during similarity-based categorization and the comparisons of test stimuli with previously encountered exemplars and prototypes. We expect that AD patients have poor similarity-based categorization due to difficulty organizing large feature sets and comparing these with previously encountered exemplars, fMRI will relate this to PHAC and anterior prefrontal cortex. Our previous work showed reduced activation of PFC and PHAC in AD during semantic memory challenges for multiple categories of knowledge, and we will assess whether this is due to impaired categorization processes in AD with fMRI. We expect that limited rule-based categorization is related to reduced PFC activation in AD, and poor similarity-based categorization is related in part to limited PHAC activation. Work in the previous grant cycle also showed that degraded knowledge in semantic memory is not necessarily category-specific. Specific Aim 3 will study the status of diagnostic feature knowledge and feature redundancy in semantic networks. We expect the diagnostic status of feature knowledge to be impaired in AD, and to see greater reliance on feature redundancy in AD patients' semantic comprehension, fMRI will relate feature diagnosticity to PFC across multiple categories of knowledge. [unreadable] [unreadable]

[unreadable] DESCRIPTION (provided by applicant): Semantic memory plays a critical role in communication and thought, but advances in the cognitive and neural basis for semantic memory have proven elusive. We propose a novel approach to semantic memory by focusing on generalized quantifiers like "some of the apples" and "most of the oranges." Our model of generalized quantifier comprehension includes at least a finite set of simple circuits that represent number knowledge, a working memory device the retains the results of intermediate computations, and a small set of combinatorial operations that manipulate information held in working memory. We map this semantic domain onto a large-scale neural network that includes at least parietal cortex (for number knowledge), inferior frontal cortex (for working memory), and dorsolateral prefrontal cortex (for executive switching). Preliminary computational work, behavioral studies in healthy seniors and patients with focal neurodegenerative diseases, and functional neuroimaging work support these distinctions. Patients with corticobasal degeneration have a fundamental deficit in number knowledge but little other lexical comprehension difficulty, related to parietal atrophy. They are particularly impaired in their comprehension of "first-order quantifiers" like "some of the X" that depend crucially on number knowledge; fMRI work in healthy adults relates comprehension of first-order quantifiers to parietal cortex. Frontotemporal dementia and Alzheimer's disease patients have limited working memory and poor switching that are associated with frontal atrophy. They are impaired in their comprehension of "higher-order quantifiers" like "most of the X" that, beyond number knowledge, also require working memory and the manipulation of information retained in working memory, fMRI work in healthy adults relates comprehension of higher-order quantifiers to frontal cortex. We propose additional modeling, behavioral, and functional neuroimaging studies in healthy seniors and patients with these neurodegenerative diseases to establish the cognitive and neural contribution of number knowledge, working memory, and executive switching during generalized quantifier comprehension. The results of these studies will provide a novel and important perspective on the neural basis for semantic memory, and will improve diagnostic and prognostic accuracy in patients suffering from dementia. [unreadable] [unreadable]

This project investigates the cognitive and neural basis of impaired cognition in patients with mild Parkinson's disease dementia (PDD) and Dementia with Lewy bodies (DLB). We propose a detailed characterization of limited executive resources in PDD/DLB. By focusing on three domains of daily functioning that depend on planning and mental organization, we will test the hypothesis that limited organization in PDD/DLB is due to a breakdown of hierarchical, top-down, goal-oriented planning. Our powerful method links modern approaches to neurodegenerative diseases and cognitive neuroscience, while providing converging evidence from multiple sources to advance our appreciation of both areas with greater validity and reliability. Specific Aim 1 examines comprehension of scripts or routine activities such as "going to a restaurant." We hypothesize that mild PDD/DLB patients have difficulty understanding scripts because of impaired script organization. We will correlate cognitive deficits with MRI structural atrophy in PDD/DLB, and perform fMRI studies of healthy adults and PDD/DLB patients. This work will relate script organization to dorsolateral prefrontal cortex (dIPFC). Specific Aim 2 examines executive resources like planning that contribute to the comprehension of sentences with a temporary structural ambiguity (e.g. "While the mother walked the baby was crying"). We hypothesize that disambiguating these sentences is difficult in mild PDD/DLB because of limited planning, and imaging studies will relate this to dIPFC. Specific Aim 3 assesses PDD/DLB patients' organization of a narrative telling the wordless children's picture story "Frog, Where Are You?". We expect that mild PDD/DLB patients will be able to describe each story event, but will have difficulty synthesizing these events into an organized narrative. Imaging studies will relate this to dIPFC. Our unique approach will improve diagnostic accuracy through converging evidence from multiple sources that show a deficit in planning and mental organization in PDD/DLB, and will underline the critical role of dIPFC in this impairment. Using a within-patient design, we will compare performance across studies to assess the various forms of planning. We will investigate the role of dementia in mild PDD/DLB through comparative work with non- demented PD patients. This work will lead to novel strategies for.improving lADLs in PDD/DLB by augmenting top-down planning and goal-oriented organization in these patients. We will also contribute to unraveling an elusive element of cognitive neuroscience - the contribution of dIPFC to planning and higher cognition.

DESCRIPTION (provided by applicant): This competing renewal investigates the role of number knowledge in word meaning. In the prior funding period, we discovered that non-aphasic patients with corticobasal degeneration (CBD) have difficulty with number knowledge, and that this interferes with their comprehension of number-based quantifiers like "at least half" and "most." MRI and pathology studies associated this with parietal disease. Converging evidence related quantifier comprehension to parietal activation during fMRI studies of healthy controls. We propose to show that the quantity component of number knowledge is responsible for a specific aspect of number-based quantifier meaning, and studies of CBD and fMRI work in healthy adults will relate this to the intraparietal sulcus. This collaborates with another, pragmatic component of quantifier meaning, and studies of non-aphasic patients with frontal disease due to frontotemporal lobar degeneration (FTLD) and fMRI studies of healthy controls will relate this aspect of quantifier comprehension to prefrontal cortex. Quantifiers can be ambiguous, and a neuroeconomic approach to decision-making will be applied to the interpretation of quantifiers with ambiguous meaning. We will show that evaluating the probability of a specific interpretation is mediated by prefrontal cortex, that appreciating the value of unambiguous communication is related to orbital frontal cortex, and that the integration of these decision-making components is supported by inferior parietal cortex. PUBLIC HEALTH RELEVANCE: In this competing renewal, we propose to show that the quantity component of number knowledge contributes to the representation of number-based quantifiers like "at least half" and "most." Studies of patients with corticobasal degeneration and fMRI activation studies of healthy controls will relate this to inferior parietal cortex. This collaborates with another, pragmatic component of quantifiers, and studies of patients with frontotemporal lobar degeneration and fMRI studies of healthy adults will relate this to prefrontal cortex. Quantifiers can be ambiguous, and we will adopt a neuroeconomic approach to decision-making to help understand quantifier interpretation. This process will be mediated by a large-scale neural network including several frontal and parietal regions. The proposed work will provide important insights into the cognitive neuroscience of language and number while improving diagnostic accuracy in an elusive neurodegenerative condition.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. This project utilized behavioral testing in neuroimaging to characterize cognitive impairment in frontotemporal dementia (FTD), with a specific focus on the neural basis for semantic and grammatical aspects of language as it applies to progressive aphasia that is commonly seen in FTD.

PROJECT SUMMARY: Our prior work demonstrated hypothesis-driven characterizations of the phenotypes associated with frontotemporal lobar degeneration (FTLD), including progressive non-fluent aphasia (PNFA), semantic dementia (SD), and behavioral variant patients with a social and executive disorder (bvFTD). We also began to relate these syndromes to the major pathologies underlying FTLD. However, little work has determined the value of clinical features for predicting pathology. We propose novel studies that optimize features we can use to characterize PNFA, SD and bvFTD. We will obtain converging evidence for these observations from novel MRI techniques integrating grey matter (GM) atrophy and white matter (WM) tract defects, and from functional MRI (fMRI) studies of healthy adults using the same materials. Finally, we will relate FTLD phenotypes to pathology, and determine how these associations are modulated by genetic factors. Aim 1 addresses the basis for non-fluent speech in PNFA. We propose studies of conversational discourse in PNFA to highlight the combined role of grammatical and executive deficits in their language disorder. Integrated GM-WM imaging will relate this to left frontal disease, and converging evidence for an fMRI study of healthy adults will confirm this. Aim 2 will examine the role of degraded perceptual feature knowledge in the semantic deficit of SD patients. Imaging studies will, relate poor knowledge of visual object features to visual association cortex in the ventral temporal lobe and poor auditory feature knowledge to dorsal temporal regions. An fMRI study in healthy adults will provide converging evidence for the anatomic relationships of visual and auditory feature knowledge of object concepts. Aim 3 will demonstrate the role executive limitations in the most devastating social disorders of bvFTD patients. We will identify two forms of apathy - a vegetative form manifested as poor motivation, and an executive form that relates apathy to difficulty with multi-step tasks. Using integrated GM-WM imaging, we expect these two forms of apathy to be associated with disease in different frontal regions. Different forms of disinhibition also will be identified. One form will be related in part to impoverished social knowledge and right temporal disease, and a second form related to executive dysfunction (e.g. mental switching difficulty) involving right ventral prefrontal disease. Aim 4 will test the hypothesis that non-fluent speech due to grammatical difficulty in PNFA is associated with FTLD-Tau, and this association will be strengthened by tau haplotype. Degraded feature knowledge for object concepts in SD will be associated with FTLD-TDP, and this will be strengthened by TMEM106B. We will also test the hypothesis that the vegetative form of apathy in bvFTD is related to FTLD-Tau, while disinhibition due to impaired social knowledge is related to FTLD-TDP. This work will advance cognitive neuroscience in the areas of language processing, semantic memory, and social cognition. These findings will contribute crucially to diagnostic accuracy, clarify specific clinical features that are most likely to refiect underlying pathology, and provide meaningful endpoints for treatment trials.

CLINICAL CORE: Project Summary/Abstract Frontotemporal lobar degeneration (FTLD) is a young-onset neurodegenerative condition. The major pathologies causing FTLD include misfolded forms of tau known as FTLD-Tau, TDP-43 proteinopathy known as FTLD-TDP, and other very rare FTLD variants. It is crucial to find features identifying these pathologic forms of FTLD during life to develop reliable diagnostic markers and establish eligibility for disease-modifying therapy trials. Mutations associated with these pathologies are found in 15% of cases. The vast majority of FTLD patients have sporadic disease, but clinical features do not reliably predict FTLD-Tau or FTLD-TDP histopathology, and up to 25% of patients who present with clinical frontotemporal degeneration (FTD) are atypical forms of autopsy-confirmed Alzheimer?s disease (AD). It is thus necessary to collect additional data that reflect the pathologic form of FTLD during life more accurately. This includes clinical, neuropsychological, neuroimaging, genetic markers including single nucleotide polymorphisms (SNPs), and cerebrospinal fluid (CSF) biomarkers. Moreover, disease-modifying treatment trials require well-characterized natural histories of FTLD-related phenotypes to identify markers of a beneficial treatment response. The Clinical Core will work with other cores and projects of this Program Project Grant (PPG) to improve our understanding of disease mechanisms. To this end, we will continue to recruit patients with sporadic clinical FTD who have a high likelihood of having FTLD spectrum pathology, and asymptomatic and symptomatic carriers of mutations associated with FTLD. We will collect multimodal data to help identify the specific form of FTLD pathology to improve diagnostic accuracy in sporadic cases and characterize longitudinal progression, and we will recruit these patients for autopsy. We propose four Specific Aims to achieve these goals.

The initial funding period showed that patients with Parkinson's disease dementia (PDD) and dementia with Lewy bodies (DLB) have impaired executive control that interferes with processing crucial aspects of language such as narrative conversation and sentence ambiguity. These fine-grained measures showed greater deficits in DLB than PDD, and performance was related to gray matter (GM) atrophy of specific frontal regions, overlapping relevant measures of executive control. In addition to diagnostic specificity that promises to help differentiate DLB from PDD, these language measures have face validity since they are correlated with functional measures of daily living (ADLs). In this competing renewal, we build on initial findings with cognitive, imaging, biomarker and clinical-pathological studies that will lead to improved diagnostic accuracy. Specific Aim 1 will assess the complex cognitive and neural basis for conversational narrative in PDD/DLB. We focus on coordination, or the ability to adjust a conversational narrative to optimize communication with a conversational partner. Specific Aim 2 will assess the cognitive and neural basis for processing lexical ambiguity in PDD/DLB. We will evaluate anaphora, or the assignment of a referent to a pronoun, and homonym meaning. We expect significant deficits in these executive-mediated aspects of language, with worse performance in DLB than PDD, and both worse than PD. Regression analyses will relate these deficits to executive measures such as mental flexibility. Theory of Mind, and decision-making. Novel MRI analyses of GM atrophy and diffusion tensor imaging tractography studies of white matter (WM) disease will relate these deficits to interruption of a large-scale neural network involving several prefrontal GM regions and associated WM frontal-striatal projections. We will relate performance to functional status from Project 1, to cognitive measures from Core B, biofluid biomarkers from Core C, and novel lysates of alpha-synuclein (AS) strains from Projects 3 and 4. Specific Aim 3 will assess the pathologic basis for these cognitive deficits in a comparative, clinical-pathological assessment of PDD and DLB. We expect significant prefrontal disease with neuronal loss and gliosis. Relative to PDD, denser histopathologic abnormalities involving AS, Beta-amyloid (AB) and tau will be seen in dorsolateral, ventral-orbital and medial frontal regions in DLB. This work will lead to novel behavioral and imaging markers of disease that distinguish PDD from DLB, and can potentially serve as targets for behavioral intervention while also advancing cognitive neuroscience.

DESCRIPTION (provided by applicant): Adult aging represents a balance of decline and compensation. This grant investigates the interacting effects of adult aging, age-related declines in hearing acuity, and engagement of the neural networks that underlie the ability to comprehend spoken sentences. Good progress has been made in understanding this balance at the behavioral level, but far less is known about the underlying neural substrates that carry these effects. In this grant we address the neural basis for aural sentence comprehension and how this is affected by perceptual effort due to reduced hearing acuity, and central cognitive loads such as those imposed by syntactic complexity, speech rate, working memory demands, and the semantic and syntactic ambiguities that occur in everyday discourse. We are guided by a novel model of sentence processing that implicates several dissociable neural substrates that become engaged depending on the cognitive and perceptual demands of the materials being processed. Our subjects will be healthy young and older adults with good hearing acuity and with mild-to-moderate hearing loss. We test our neurocognitive model using novel imaging methods that combine fMRI BOLD activation studies with structural imaging of gray matter thickness and white matter projections that integrate cortical areas. We hypothesize that this model of selective working memory and decision-making during sentence processing partially breaks down during aging because of limited white matter projections between brain regions, and cortical thinning in dorsolateral prefrontal cortex. We also assess potential compensatory mechanisms. Aim 1 examines effects of syntactic complexity and speech rate, Aim 2 specifically manipulates working memory load within sentences, and Aim 3 investigates decision-making needed to resolve lexical and syntactic ambiguities encountered during sentence processing. The proposed studies will close a critical gap in our knowledge of the interacting effects of cognitive aging and age-related hearing loss on everyday speech comprehension in older adults. This work may also serve as a framework for early detection of pathological change as it affects language comprehension in the aging brain.) )

PROJECT 1: Multimodal Endophenotyping of Spreading Human Tauopathy Project Leader: Murray Grossman Project Summary/Abstract Frontotemporal degeneration (FTD) is an understudied spectrum of neurodegeneration associated with pathologies involving variants of FTLD-Tau or FTLD-TDP. Preliminary cellular and animal studies investigating the cause of progression from Projects 3 and 4 are consistent with spreading cell-to-cell transmission of disease protein, although this has not been examined in humans. In this Project, we use longitudinal clinical (Clinical Core B), neuroimaging with MRI and PET-Tau (Clinical Core B), and genetic (Project 2 and Genetics Core C) information from an established longitudinal FTD cohort supported by this PPG to examine mechanisms of disease progression in humans with Biostatistics and Database Core E, and evaluate this in the context of pathological assessment with Project 4 and Neuropathology & Biomarker Core D. Aim 1 investigates multimodal biomarkers that reflect spreading FTLD pathology during life. We use novel, statistically powerful, and multivariate integration of clinical, MRI and PET-Tau measures available in Penn's multimodal dataset to chart the longitudinal anatomic progression of disease spread in retrospective and prospective cohorts of patients with likely FTLD-Tau compared to FTLD-TDP, and explore progression in FTLD-Tau subgroups with CBD and PSP. We cross-validate clinical and neuropathological staging with autopsied FTLD-Tau patients from Project 4. We hypothesize that multivariate analyses will show an anatomically-constrained imaging pattern of longitudinal disease spread in FTLD-Tau that differs from FTLD- TDP. Aim 2 studies genetic mechanisms that modulate disease progression rate in FTD. With Project 2 and Core C, we associate a risk allele in the region coding for myelin oligodendrocyte basic protein (MOBP) with more rapid progression in FTLD-Tau, and this is related to white matter integrity in imaging studies of FTLD- Tau. We will identify other genetic markers that modulate rate of disease spread, and relate these to cognition and imaging that reflect clinical decline. We hypothesize that additional genetic, biofluid and pathologic mechanisms modulating the rate of disease progression will be identified. Aim 3 studies cognitive reserve factors that modulate rate of clinical progression in FTD. With Cores B, D and E, we examine the way in which education and job attainment impact the rate of progressive clinical decline, and imaging of anatomic mechanisms that support cognitive adaptations to progressive disease. We hypothesize that the rate of longitudinal clinical decline depends in part on professional and educational attainment related to adaptive executive resources. Together with Project 2 and Cores B, C, D and E, this proposal will validate mechanisms of cell-to-cell disease spread in human FTLD based on cell/animal studies in Projects 3 and 4, and will sharpen our understanding of the rate of progression.

? DESCRIPTION (provided by applicant): Neurodegenerative disease is a major public health problem. Frontotemporal degeneration (FTD) is a clinical neurodegenerative condition that affects both gray matter (GM) and white matter (WM) and causes a network disorder. FTD is an excellent model for directly imaging the neurobiology of neurodegeneration because the associated pathology involves a monoproteinopathy in each patient - either frontotemporal lobar degeneration (FTLD) due to tau (FTLD-tau) or to TAR DNA binding protein of 43kD (FTLD-TDP). We propose a connectomic approach to identify FTLD-tau and FTLD-TDP in vivo. This is timely because of the discovery of disease-modifying agents, and pressing needs for accurate antemortem diagnosis, biomarkers to gauge response during treatment trials, and elucidation of mechanisms of disease progression even at the preclinical stage. About 25% of cases have familial FTD (fFTD) due to a small set of mutations causing one of these pathologies. The remaining 75% of cases have sporadic FTD (sFTD) with no definitive biomarkers for FTLD- tau or FTLD-TDP pathology. Preliminary data suggest that multimodal structural MRI (sMRI) of GM disease and diffusion MRI (dMRI) of WM disease can identify vulnerable networks in FTLD-tau and FTLD-TDP. We propose a five-site consortium, including Mayo Clinic, MGH/Harvard, Northwestern University, University of California in San Francisco, and University of Pennsylvania, to acquire HCP imaging in FTD. This will be linked to two NIH-funded biomarker registries, and this linkage will result in substantial cost savings. We propose three Specific Aims. In Year 01, Aim 1 will implement and validate the Human Connectome Project (HCP) Lifespan protocol for sMRI, dMRI, resting BOLD MRI (rs-fMRI), task-based functional MRI (tfMRI) and arterial spin labeling (ASL). We will acquire initial data, harmonize data between sites and with HCP, implement quality control procedures, optimize analyses using HCP and locally-developed pipelines, and implement data sharing procedures. In Year 02, Aim 2 will study presymptomatic and symptomatic fFTD associated with FTLD-tau or FTLD-TDP, and assess sFTD in specific phenotypes highly associated with FTLD-tau or FTLD-TDP. Connectomic imaging will be integrated with NIH-funded registries that acquire clinical, genetic and biofluid data. Based on histopathology showing greater WM disease in FTLD-tau than FTLD-TDP, we expect advanced HCP imaging to show partially distinct patterns in multimodal imaging of symptomatic as well as presymptomatic individuals with familial and sporadic FTLD-tau compared to FTLD-TDP. In Years 03-04, Aim 3 will acquire longitudinal data to assess competing hypotheses about mechanisms of disease spread in presymptomatic and symptomatic FTD. Consistent with animal studies, we expect that graph theoretic and multimodal network analyses will show disease spreading locally to adjacent brain regions, affecting different networks in FTLD-tau or FTLD-TDP.

A hallmark of typical amnestic Alzheimer's disease (aAD) is neurofibrillary tau pathology. Braak first described staging of tau pathology in AD, classically evident early in the medial temporal lobe (MTL) implicated in impaired episodic memory, and then spreading to neocortical regions important for language, visuospatial and executive function and to basal ganglia for motor function. However, AD pathological change can also present clinically as a focal neocortical syndrome without amnesia, known as non-amnestic Mild Cognitive Impairment (naMCI). This includes disorders of language (logopenic variant primary progressive aphasia, lvPPA), visuospatial (posterior cortical atrophy, PCA), executive (frontal variant MCI, fvMCI), and motor (corticobasal syndrome due to AD, CBS-AD) function. Rare clinical, imaging, and autopsy studies suggest that tau pathology in naMCI is greater in neocortex than MTL, challenging fundamental assumptions about the MTL origin of cerebral tau pathology in AD. However, staging tau pathology at autopsy in naMCI has not been investigated, and longitudinal in vivo imaging to validate spreading disease in naMCI is very rare and involves small groups of clinical cases using a single imaging modality. These represent major gaps in our knowledge. In Aim 1, we study stages of spreading tau pathology at autopsy in naMCI and aAD with a novel, validated, digital method. We hypothesize that stages of accumulating tau pathology are consistent with neocortical origin and spread in naMCI, and only later spread to MTL, differing from the MTL origin of tau pathology in aAD. Pathologic staging depends on inferences from cross-sectional studies at autopsy, so Aim 2 proposes to validate stages of spreading disease in vivo with novel longitudinal MRI and tau-PET imaging in naMCI and aMCI/ aAD using unique imaging and graph theory approaches. We hypothesize a cortical origin and spread of tau in naMCI, with later accumulation of tau in MTL, reversing the MTL origin of disease in aMCI/ aAD. These findings would challenge assumptions about the MTL origin of tau pathology in AD. We test the novel hypothesis that genetic factors in FTLD-tau bias the anatomic distribution of tau pathology contributing to the atypical spread of tau in naMCI. In Aim 1, we expect that stages of tau pathology in naMCI resemble that seen in primary tauopathies due to FTLD-tau, including non-fluent/agrammatic variant PPA, behavioral variant frontotemporal dementia, and corticobasal degeneration/progressive supranuclear palsy. In Aim 2, we expect that patterns of longitudinal in vivo imaging in naMCI resemble that seen in FTLD-tau. Finally, Aim 3 proposes a targeted study of FTLD- tau-related single nucleotide polymorphism (SNP) risk alleles in naMCI. Based on our preliminary findings, we hypothesize that FTLD-tau-related risk alleles are more common in naMCI than aMCI/ aAD. These findings would challenge long-held assumptions about the pathophysiologic basis for AD pathology, develop validated diagnostic criteria for naMCI to support inclusion in disease-modifying AD treatment trials, lead to validated endpoints for these trials, and provide needed prognostic information for this underserved population.

Abstract The presence of ?-synuclein (aSyn) pathology defines Lewy Body Disorders (LBD), including Parkinson's dementia (PDD) and Dementia with Lewy Bodies (DLB). However, LBD has significant clinical heterogeneity, and current, purely clinical criteria do not identify the mixed pathology found in LBD. We hypothesize two forms of human aSyn pathology that are biologically meaningful: ?pure? aSyn with minimal AD co-pathology (aSyn- AD), and aSyn modified by AD co-pathology (aSyn+AD). Moreover, these pathologies are related to pathologic maturity and distinct strains developed in model systems in Projects I/II: One aSyn strain (Strain A) is associated with less mature pathology and induces only aSyn, and we expect to find this in both aSyn-AD and aSyn+AD; a second strain (Strain B) is associated with more mature aSyn pathology and induces both aSyn and less mature tau pathology, and we expect much more Strain B in aSyn+AD than aSyn-AD. We assess mixed pathologies further using a 2X2 design and compare LBD with AD, where half of clinical AD have ?pure? AD pathology (AD-aSyn) and half have aSyn co-pathology (AD+aSyn) at autopsy. We expect aSyn strains to be minimal in AD-aSyn, but we expect Strain A in AD+aSyn. Moreover, we expect mature tau pathology in AD. Aim 1 examines digital histology of the maturity and strain-related pathology in LBD and AD. We use a validated, parametric histology (DHist) approach to study the anatomy of aSyn, tau and A? pathology, relate these to antemortem clinical features and assess monoclonal antibodies (mAbs) raised against aSyn strains in Projects I/II. In aSyn+AD, we expect more mature aSyn and less mature tau pathology in frontal and temporal regions with executive and language deficits, respectively, and less striatal pathology, related to Strain B mAb; but aSyn-AD will have less aSyn cortical pathology and less impaired cognition, related to less mature aSyn and mAb to strain A. AD will show more mature tau in a different anatomic locus since the distribution of tau is guided in part by aSyn Strain B found in aSyn+AD but not AD-aSyn, and less mature aSyn pathology related to Strain A in AD+aSyn. Aim 2 studies in vivo MRI and cerebrospinal fluid (CSF) markers of strain-related disease in LBD and AD. Using validated CSF markers of pathology to define cases, we study cross-sectional and longitudinal cognition and MRI in living patients with likely LBD (aSyn-AD, aSyn+AD) vs. AD (AD-aSyn, AD+aSyn). In LBD with CSF consistent with aSyn+AD, we expect progressive MRI atrophy in frontal and temporal regions greater than in striatum, related to declining executive and language function, respectively. MRI in LBD with CSF consistent with aSyn-AD has less cortical and cognitive decline. Since tau is guided by the anatomic locus of aSyn induced by Strain B in aSyn+AD, cognitive deficits and MRI atrophy will differ from AD. Aim 3 studies atrophy in critical medial temporal lobe (MTL) subfields and amygdala in LBD and AD. Using DHist pathology and state-of-the-art in vivo MTL subfield segmentation, we expect distinct subfield and amygdala distributions of aSyn and tau in LBD and AD.

Frontotemporal degeneration (FTD) is an understudied clinical neurodegenerative condition. Nevertheless, this is the most common dementia after Alzheimer disease (AD) in people younger than 65. The most common pathology associated with dementia in FTD is frontotemporal lobar degeneration (FTLD) due to transactive DNA binding protein of ~43 kD (TDP-43), known as FTLD-TDP. FTLD-TDP pathology is also found in ALS and many other conditions associated with aging including Limbic-predominant Age-related TDP-43 Encephalopathy (LATE). Since our discovery of FTLD-TDP in human disease, major gaps in understanding the pathophysiology and spread of human FTLD-TDP pathology and poor antemortem identification of patients with dementia due to TDP-43 pathology have been immense impediments to developing disease-modifying treatments that aim to target this pathology. In this Program Project Grant (PPG), we propose a novel and unique multidisciplinary research program focused on elucidating pathophysiologic mechanisms of human TDP-43 pathology, identifying FTLD-TDP in vivo in dementia and aging, and understanding the clinical consequences of FTLD-TDP pathology. With greater insight into mechanisms of disease associated with TDP- 43 pathology, translational work will fill major gaps in in vivo diagnosis and prognosis. Administrative Core A is designed to provide administrative support including regulatory and financial management for the scientific work proposed by the five Projects and four Cores of this PPG, coordinate the interaction of these multidisciplinary Projects and Cores through an Internal Executive Committee (IEC) consisting of the Project Leaders (PLs) and Core Leaders (CLs), ensure that annual goals are maintained with guidance from an External Advisory Committee (EAC) of experts together with NIA staff, disseminate and coordinate our work with other scientists at and beyond University of Pennsylvania (Penn), engage young scientists in research investigating TDP-43, and disseminate knowledge to patients and families who have FTLD-TDP-related disorders. We achieve these goals through seven Specific Aims.

Clinical research criteria for primary progressive aphasia (PPA) reliably identify semantic variant PPA (svPPA) and non-fluent/agrammatic variant PPA (naPPA). While statistically associated with underlying pathology, neither these criteria nor regional MRI atrophy reliably support in vivo diagnosis of frontotemporal lobar degeneration (FTLD pathology) in an individual with sporadic FTLD-TDP or FTLD-Tau. We propose to identify macroscale network metrics and ultra-high resolution 7 tesla (7T) MRI profiles that are sensitive to both regional anatomic features and cortical laminar features in sporadic PPA variants, and we validate our findings in antemortem imaging of autopsy-confirmed cases and in familial FTLD (fFTLD) mutation carriers with known pathology. We pursue these goals in three Specific Aims. Aim 1 tests the hypothesis that multimodal 3T structural MRI (sMRI) and diffusion-weighted MRI (dMRI) networks show distinct graph theoretic metrics in sporadic svPPA and naPPA that are sensitive to both regional anatomic and laminar-specific features of disease, and assesses the same features in antemortem MRI of sporadic autopsied cases and in fFTLD mutation carriers with FTLD-TDP and FTLD-Tau, respectively. We also relate degraded network features to inexpensive targeted linguistic measures that can screen for disease. This hypothesis is based on findings in our clinical-pathological series, showing that FTLD-TDP has anterior temporal and orbital frontal pathology related to sMRI atrophy in the same regions, often in svPPA with impaired naming, and that pathology is denser in superficial laminae; by comparison, FTLD-Tau has inferior frontal and midfrontal pathology related to sMRI atrophy, often in naPPA with non-fluent speech, and pathology is relatively denser in deeper laminae. In Aim 2, we test the hypothesis that partially distinct metrics of progressive disease are seen in longitudinal, multimodal 3T MRI network of sporadic svPPA and naPPA, and that these overlap in part with network metrics in antemortem longitudinal imaging of autopsied PPA and fFTLD cases associated with FTLD-TDP and FTLD- Tau, respectively. Aim 3 examines combined laminar and regional anatomic features more directly by assessing ultrahigh resolution 7T MRI in PPA and fFTLD cases. 7T MRI is expected to identify relatively distinct cortical laminar and white matter (WM) features in sporadic svPPA vs. naPPA, and in fFTLD with FTLD-TDP vs. FTLD-Tau pathology, respectively, and these will overlap in part with distinguishing network metrics found at 3T. This is based on preliminary findings that ex vivo 7T MRI of autopsied cases shows relatively distinct, pathology-determined cortical laminar and WM features: FTLD-TDP pathology is denser in superficial cortical laminae, while FTLD-Tau pathology is relatively denser in deeper laminae and has greater WM pathology. Novel, pathology-guided, cortical laminar features, combined with regional anatomic markers, will lay the groundwork for innovative methods to distinguish FTLD-TDP from FTLD-Tau in vivo, and lead to longitudinal markers for disease progression urgently needed for disease-modifying treatment trials.

Frontotemporal degeneration (FTD) is an understudied clinical neurodegenerative condition that is the most common dementia after Alzheimer disease (AD) in people younger than 65. The most common pathology associated with FTD is frontotemporal lobar degeneration due to transactive DNA/RNA binding protein of ~43 kD (TDP-43 (FTLD-TDP), and this is also the underlying pathology in the vast majority of patients who have co-occurring amyotrophic lateral sclerosis (FTD-ALS) spectrum disorders as well as a critical force in age- related disorders such as limbic-predominant age-related TDP-43 encephalopathy (LATE). Since discovering a role for TDP-43 pathology in human disease, important progress has been made in experimental cellular and animal models of disease. However, the human brain has many unique properties associated with distinctly human clinical disorders that are not easily replicated in these experimental models. Major gaps in knowledge thus constrain the development of disease-modifying treatment trials. Among these is our limited knowledge of the pathophysiologic consequences of the accumulation and progression of abnormal TDP-43 at a molecular level. At a microscopic level, a major limitation is that most patients with accumulating TDP-43 have sporadic disease that can be identified reliably only at autopsy, although ~20% of cases have familial FTLD (fFTLD) with known pathology during life. We are limited at translating this knowledge to a macroscale level where FTLD- TDP pathology is manifested in humans with heterogeneous clinical features as diverse as emotional dysregulation and impaired language both with and without a motor disorder. Moreover, there is limited knowledge of the factors contributing to the highly varying rates of disease progression. In five novel, independent but synergistic Projects and five Cores that support each of the Projects, this unique, multidisciplinary, Program Project Grant (PPG) adopts the innovative perspective of investigating the TDP-43-associated breakdown of neural networks at molecular, microscopic and macroscale levels in humans. We hypothesize that our novel, well-integrated, network perspective will fill major gaps in knowledge by elucidating mechanistic insights into the pathophysiology of abnormal TDP-43 and the associated pattern of disease progression, and offer a fresh perspective on the identification of accumulating TDP-43 pathology during life and its longitudinal course. This proposal is consistent with the highest priorities for FTD at the 2019 Alzheimer?s Disease and Related Disorders (ADRD) summit. By focusing on disruption of neural networks at molecular, microscopic and macroscale levels of brain functioning, our multidisciplinary network approach will elucidate the pathophysiology and spread of abnormal TDP-43 in humans, and examine the consequences of TDP-43 pathology for clinical disease during life in dementia and aging using fresh approaches to improve our mechanistic understanding of TDP-43 pathology in humans while maintaining the highest level of scientific rigor and contributing to urgent clinical needs such as treatment trials.