Olfactory Dysfunction in Aging and Neurodegenerative Disease

Session 1: Olfaction and the Aging Brain in Health and Disease

Dr. Pamela Dalton, the session chair, addressed the need for increased research related to changes associated with “normal” aging to better understand the link between olfactory dysfunction and development of AD and related dementias (ADRD). She described her team’s work in measuring the effects of chronic exposure to pollutants on different aspects of olfaction and their role in exacerbating age-associated olfactory dysfunction. Dr. Holbrook discussed his work in characterizing changes associated with age in the peripheral olfactory system using human olfactory tissue autopsy samples, which identified age-related decline in the olfactory area and the olfactory sensory neuron population. Noting that most of our knowledge of the olfactory system comes from research on mice, Dr. Gottfried described his work comparing differences between mice and humans in how odor information at the olfactory epithelium is packaged and presented to the olfactory bulb. Dr. Larsson described results from a large Swedish longitudinal study which provides a window into changes in olfaction with age and in relation to dementia, bolstering the idea discussed throughout these sessions that loss of smell could serve as an early warning sign of cognitive decline and age-related health changes. Dr. Chen summarized recent findings from the Health, Aging, and Body Composition (Health ABC) study, which provided some of the first epidemiological evidence that poor olfaction is associated with multiple adverse health outcomes besides neurodegenerative diseases.

Session 2: Clinical Utility of Olfactory Biomarkers for Neurodegenerative Disease

The session chair, Dr. Mark Albers, noted the heterogeneity of AD and the clinical need for molecularly agnostic functional tests for use in screening to identify AD development before individuals exhibit mild cognitive impairment. He described ongoing work in developing and testing an at-home screening tool that could be used to flag risk and make screening more accessible by identifying those who could benefit from more expensive and invasive testing. Dr. Tian presented data from the NIA Intramural Research Program’s Baltimore Longitudinal Study of Aging, which supported the conclusion that olfactory decline is related to worsening of AD pathology, brain atrophy, and cognition. Future studies with longer follow-up are needed to understand whether reduced olfactory function precedes cognitive changes and whether these associations are mediated through brain atrophy. Dr. Devanand described research showing that impaired odor identification and impaired global cognition are associated with an increased likelihood for transition to dementia. Dr. Devanand’s data illustrated the predictive value of pairing odor identification testing with global cognitive screening. Focusing on environment as a risk factor for olfactory impairment and risk of neurodegenerative disease, Dr. Pinto discussed data from several large, diverse longitudinal cohort studies that have illustrated the deleterious effects of pollution exposure on the olfactory system and how they relate to neuropathology and dementia. Dr. Bankova described research on COVID-19-related anosmia. While results are preliminary and based on small samples, early results indicate there is an inflammatory signature that merits further exploration in a larger, more controlled study.

Session 3: Candidate Mechanisms Underlying Olfactory Dysfunction in Aging and Neurodegenerative Disease

Dr. McClintock described his work using mouse models to explore what the olfactory epithelium can reveal about the link between the APOE Ɛ4 allele and late onset AD. Data suggest a progression of worsening events in the epithelium as mice age, including decreased glucose metabolism, possible mitochondrial dysfunction, and increased neuronal loss, which mimics the progression of the disease in humans. Dr. Chandra described a mouse study of olfactory deficit in PD, which identified α-synuclein pathology along the entire central olfactory pathway, primarily affecting projection neurons. Dysfunction included a reduction in olfactory bulb (OB) neurogenesis and changes in synaptic vesicular transport affecting synaptic function, both of which are likely contributing to olfactory behavioral deficits. Dr. Schoonover discussed the role of the piriform cortex on impaired recall of olfactory memories. He described unpublished work in exploring the stability of odor responses in the piriform over time and testing whether passive olfactory experiences alter the synaptic connectivity of the piriform network. Dr. Wesson, the session chair, described his team’s work in assaying mouse olfactory perception by monitoring changes in respiratory dynamics during sniffing, which they used to investigate changes in odor perception. Initial results indicate synuclein is a component in the pathogenesis of PD that alters physiology of the olfactory system in ways that are important for changing odor perception. Dr. Kay discussed studies related to persistent COVID-19 olfactory dysfunction, exploring whether dysfunction driven by olfactory bulb damage could suggest a wave of post-COVID dementia in the coming decades. People who have had even mild cases of COVID-19 show signs of degeneration in cortical areas connected with the olfactory system. Given the influence of the OB and its pathology on systemwide activity, more studies are needed using human and animal models to investigate changes over time in OB-associated systemwide connection patterns.

Session 4: Leveraging Emerging Technologies and Methods to Explore Mechanisms of Olfactory Dysfunction

Noting the importance of developing sequencing-based biomarkers for AD, Dr. Datta described his work to identify transcriptional biomarkers for olfactory function. His research team is sampling the olfactory epithelium (OE) and performing sequencing to identify transcriptional signatures that reflect the underlying function of the olfactory system in the periphery. Given the key role basal stem cells play in the maintenance and regeneration of the OE, session chair Dr. Bradley Goldstein discussed his research to better understand how damage and repair occur and how damage impacts maintenance of the olfactory system in the context of aging or disease. Applying a brush biopsy-based single-cell RNA sequencing approach, researchers can study the neuronal lineage cell clusters and compare AD versus control olfactory epithelial cell types. Dr. Frank-Ito described the effects of nasal vestibule morphological variations on airflow and odorant transport to the olfactory cleft. Computer modeling studies have demonstrated differences in odorant transport that may have implications for individual differences in olfaction and can inform intranasal spray and drug design research to better target the olfactory airspace. Dr. Restrepo’s team set out to determine whether varicella zoster virus and herpes simplex virus type 1 infection of the olfactory system plays a role in cognitive decline in AD. Research using tissue samples from controls and patients with familial AD (FAD), suggests that viral infection and associated inflammation and dysregulation of myelination of the olfactory system may disrupt hippocampal function, contributing to acceleration of FAD progression. Dr. Hua brought a biomedical engineering perspective to the discussion, explaining challenges to capturing complete magnetic resonance imaging (MRI) of the olfactory regions. Describing early results of his ongoing work exploring olfactory functional MRI of the OB in subjects with early PD, his results indicate that improved methodological approaches and new advancements in MRI technology can capture physiological changes in the olfactory region that may be important factors in the development of neurogenerative diseases.

Overall Discussion

There was a general acknowledgement that olfactory dysfunction is not a sole predictor of AD risk but is a potentially useful part of the toolkit for stratifying risk and identifying patients at an early stage for intervention. The workshop concluded with a discussion highlighting priorities for olfactory research. Primary topics included clinical application of olfaction testing, the need for longitudinal studies, and the promise of emerging technologies in research. Current smell tests are helpful in early diagnosis of ADRD, yet are not in wide use in clinics and practitioners’ offices. To promote wider use of olfaction testing, smell tests should have formal score classifications like blood and amyloid testing. The olfaction field must develop AD biomarkers to facilitate widespread use, develop compelling evidence that an olfactory test and/or biomarkers will help physicians treat patients, and provide training for medical practitioners on olfactory testing. Longitudinal studies with diverse cohorts that track the same people over long periods of time is the only way to thoroughly assess change within individuals and rate of conversion to dementia. Researchers need to collaborate and validate each other’s results across countries, which would allow for combination of data sets where study samples are small. Recent technological advances open new possibilities for studying the olfactory system on both the macro and micro levels. In addition to gaining a better understanding of what constitutes “normal” aging, the olfaction field needs to expand to encompass general physiological assessments of central olfactory structures in human subjects and animal models, including how the system processes odors, how experiences and internal states modify the system, and how the system is uniquely vulnerable to age. Because the olfactory system directly interfaces with the environment, olfactory research also provides an ideal opportunity to understand environmental contributors to neurodegeneration, including the effects of differential exposure and the contributions of social determinants of health.

Closing Remarks

Dr. Sullivan thanked participants for sharing their work, insights, and collective enthusiasm for understanding the relationship between olfactory function and neurogenerative diseases. The timing is right to make substantial progress based on the promising and innovative ideas surrounding biomarkers, whether they be imaging, molecular, or psychophysical. She also cited opportunities for understanding the underlying disease mechanisms and expressed hope that the workshop will promote discussions on how NIA and NIDCD can continue to work together.

Welcoming Remarks

Eliezer Masliah, M.D., NIA; Merav Sabri, Ph.D., NIDCD; Coryse St. Hillaire-Clarke, Ph.D., NIA

Dr. St. Hillaire-Clarke provided introductory comments and briefed the participants on meeting logistics. Dr. Sabri welcomed invited participants and staff from the National Institute on Aging (NIA) and the National Institute on Deafness and Other Communication Disorders (NIDCD). The goal of the workshop was to convene leaders in research on olfaction and brain aging to review the state of the science, identify gaps in current knowledge, and explore new opportunities and challenges for elucidating the mechanisms underlying the association between smell loss and risk of cognitive decline and neurodegenerative disease.

Dr. Masliah introduced NIA’s Division of Neuroscience to provide additional context for those joining the meeting who were outside of the aging or Alzheimer’s fields. NIA is currently the third largest Institute in the National Institutes of Health, and the Division of Neuroscience is primarily focused on fostering and supporting research and training in the areas of brain aging and dementias of old age such as Alzheimer’s disease (AD) and AD related dementias (ADRD). Research supported by NIA focuses on the full spectrum of brain aging, from understanding the basic neurobiology to mechanisms underlying changes in sensory, motor, cognitive and behavioral processes to genetics, populations, and translation of research discoveries into practical applications that improve care and outcomes for individuals with AD/ADRD. The Division has a robust program with more than 400 clinical trials underway, a number of new compounds for Alzheimer’s disease (AD) in the pipeline, and very robust basic science and cognitive behavioral science programs. Dr. Masliah noted the increased attention to olfactory disorders in the wake of the COVID-19 pandemic, which has introduced new researchers with a range of approaches and perspectives to ongoing NIA work and created new opportunities for research partnerships. The workshop allowed researchers to share their work and make connections with other researchers working in this area. Introductions of the in-person group noted a range of backgrounds and expertise, including neuropsychology, neuroscience, biomedical engineering, epidemiology, gerontology, and allergy. Attendees’ work ranged from basic science through clinical trials, translational research, and clinical care.

Keynote Address: Olfactory Dysfunction in Alzheimer’s Disease: Current Research and Future Directions

Claire Murphy, Ph.D., San Diego State University

AD has rapidly increased in the present century, with profound consequences for individuals and society. An estimated 6.7 million Americans are living with AD and prevalence doubles for every 5-year increase between ages 65 and 85. Olfactory function is early and profoundly affected in AD, raising the possibility that olfactory dysfunction, in combination with markers such as neuroimaging, may provide an accessible biomarker for early AD. Identifying the mechanisms underlying the particular vulnerability of the olfactory system in AD may elucidate critical elements of the neurodegenerative disease process and potential targets for intervention.

Dr. Murphy briefly covered the history of AD research, including its first identification in 1906, the post-mortem studies by Braak & Braak in the 1990s1 that established the cascade of events within the brain as AD develops, and work that linked the apolipoprotein E (ApoE) ε4 allele to increased AD risk. Changes associated with AD occur first in regions central to olfactory processing: the entorhinal and transentorhinal cortices, the anterior olfactory nucleus, and the olfactory bulb. Accurate odor identification is a complex process, involving sensory, cognitive, and semantic abilities. Clinical and longitudinal research has established the link between worse odor identification and markers of neurodegeneration, including decreased hippocampal volume, thinner entorhinal cortex, and worse episodic memory.
Predicting who will develop the disease at the earliest possible timepoint is critical for intervention before the neuropathological processes of AD ravage the brain and impair neurological integrity and cognitive function. Biomarkers can be used to identify candidates for clinical trials and, more importantly, those most likely to benefit from drugs in development as they become available. Approximately 85-90% of AD patients show olfactory impairment and a number of tests are available to measure olfactory dysfunction, including measures of odor threshold, odor identification, odor recognition memory, and odor remote memory. Sensitivity and specificity for discriminating AD patients from controls varies, with an average 85% rate of correct classification. Because different olfactory tasks reflect different underlying neuropathology, combining olfactory tasks that reflect neuropathology in different areas increases sensitivity and specificity.

The COVID-19 pandemic has created increased awareness of olfactory impairment and driven research into mechanisms and interventions. However, much of this research has not included older adults and the most common treatments for olfactory dysfunction, including oral or nasal steroids and smell training, have poor efficacy in older adults. There is a pressing need for research to determine how long COVID, multiple bouts with acute olfactory dysfunction from COVID, or COVID superimposed on partial age-related olfactory impairment affects long term olfactory dysfunction and susceptibility to dementia.
Both basic and clinical research is needed to enhance understanding of normal and impaired function at the peripheral, central, clinical, and population levels. Dr. Murphy also highlighted the value of developing partnerships with patient groups, clinicians, and basic scientists to facilitate patient-centered research, education, discovery, and therapeutics.

Keynote Q&A

Audience members noted lack of data on repeated testing over years, which would be needed to increase our understanding of the predictive ability of olfactory dysfunction, as well as the need for control groups followed longitudinally. There was a general acknowledgement that olfactory dysfunction is not a sole predictor of AD risk but is a potentially useful part of the toolkit for stratifying risk and identifying patients at an early stage for intervention. The group agreed that any biomarker that can contribute to better understanding would be valuable.

The paucity of data on ethnic and racial differences in the link between olfactory dysfunction and AD risk was discussed. Dr. Murphy agreed, citing a project she’s working on currently that incorporates some of the work described in the keynote into an ongoing study of Hispanic adults. Multiple commentators noted the need for more work into racial/ethnic differences and influences on the course of Alzheimer’s development.

Commentators highlighted Dr. Murphy’s point about the need to incorporate multiple measures of olfactory dysfunction and noted that this could be done with minimal additional effort. For example, in addition to being asked if they can identify what a smell is, patients could be asked how familiar it is, how pleasant it is, and how strong it is, potentially assessing neuropathology in different areas. Audience members further noted that questions regarding sense of smell are not always routinely incorporated into primary care exams of older adults, which misses an opportunity for early recognition. The importance of including the patient voice in guiding interventions was also discussed, as was the importance of including older people in clinical trials and conducting sub analyses by age in studies of olfactory dysfunction.

The Aging Olfactory System: What is Normal?

Pamela Dalton, Ph.D., Monell Chemical Senses Center, Session Chair

Olfactory deficits in normal aging result in part from the olfactory system’s direct exposure to elements such as air pollutants and viral and sinus infections. Simultaneously, aging is associated with a decrease in both olfactory receptor cells and olfactory bulb volume. However, changes associated with “normal” aging are not well-defined, which limits our ability to understand the link between olfactory dysfunction and development of ADRD.
In a recent paper, Japanese researchers looked at the amount and composition of olfactory mucus in young and elderly subjects, finding reductions in olfactory mucus and respiratory mucus as a function of age. Additionally, there are age-related changes in the clearance of chemical molecules from the physical olfactory environment and desensitization to repeated exposure. An experiment described in a 1989 publication by Stevens and colleagues demonstrated people lose sensitivity to odors after exposure more quickly and recover more slowly as they age.

Dr. Dalton and her team have decades of work in studying individual exposure to different kinds of pollutants. Measuring the effects of occupational exposure is important at all ages, but these insults can build up as we age and exacerbate age-associated olfactory dysfunction. Dr. Dalton described longitudinal cohort studies involving individuals exposed during the World Trade Center disaster and studies of individuals of different ages and lengths of involvement in the fire service that demonstrated the role of exposure to chemicals and pollutants in increasing inflammation and decreasing odor thresholds. Dr. Dalton further described current but unpublished work that provides further evidence for the association of inflammation with age and olfactory dysfunction.

Because there are multiple factors that can contribute to the olfactory dysfunction among the elderly, it will be important to distinguish between what we are calling normal olfactory aging or some of these other factors and the onset of neurodegenerative disease. Some of the longitudinal datasets discussed later in the workshop will be critical in tracking conversion to AD and identifying the factors and trajectory of olfactory cognitive dysfunction.

Age-related Changes in Olfactory Epithelium and Olfactory Bulb

Eric Holbrook, M.D., Harvard Medical School

Based on work extending back to a study by Doty and colleagues published in 1984,4 we have a collective understanding that increasing age is associated with a loss of sense of smell, with a sharp decline beginning at about age 65 years. The olfactory epithelium is capable of lifelong regeneration due to the presence of basal progenitor cells, but this capability diminishes with advancing age and contributes to a decrease in odor perception in older individuals. Dr. Holbrook described a study designed to characterize changes associated with age in the peripheral olfactory system. Using human olfactory tissue autopsy samples from 36 subjects (12 female) with an average age of 74.1 years and a comparison group of embryonic specimens, Dr. Holbrook and his team conducted an in-depth analysis of the olfactory epithelium and its neuronal projections onto the olfactory bulb as a function of age.

Although there was no available information on olfactory function of donors, 15 subjects had a medical history of dementia, with six having a specific diagnosis (five AD, one Parkinson’s disease). Use of embryonic olfactory tissue as a comparator allowed for identification of changes that may occur over time with adult tissue. After removing the mucosal membrane, the team stained the epithelium for markers of neurons. The embryonic tissue displayed a complete olfactory sheet, with no respiratory replacement in the middle of the epithelium, and a good margin of difference between respiratory and olfactory tissue. In adult autopsy subjects, the tissue had a pock-marked appearance, a rough epithelial border, and signs of age-related decline in the olfactory area and the olfactory neuron population. The reduction in olfactory neurons appears to progress in a posterior and dorsal direction possibly related to the accumulation of environmental toxin exposures over time. On a histological level, there was an age-related decrease in dividing globose basal cells which may result in the lack of typical olfactory neuron regeneration and finding of decreased neuron population seen as a function of age. Histological findings in mouse models are similar to those found in human samples and suggest an exhaustion of regenerative capacity of globose basal cells with age. However, there is some evidence that therapies designed to restore epithelial damage could provide useful neuronal synaptic function in the olfactory bulbs.
For specimens from patients with a known diagnosis of dementia, no significant relationship was found between age and olfactory epithelium area, nor was dementia found to correlate with olfactory bulb volume. However, analysis of changes with onset of dementia relied on information supplied with the autopsy specimen and under-reporting was likely. Further, without any information on donor’s olfactory function it was not possible to identify subjects in the early stages of AD for analysis or test any theories regarding the relationship between olfactory dysfunction and AD risk

Characterizing the Fine Structure and Wiring of the Human Olfactory Epithelium and Olfactory Bulb at the Level of Individual Olfactory Receptor Genes

Jay Gottfried, M.D., Ph.D., University of Pennsylvania

There has long been a general lack of appreciation for how the sense of smell contributes to our experience of the world, which has seen a shift as a result of the COVID-19 pandemic. However, most of our knowledge of the olfactory system comes from research on mice. That research shows that olfactory sensory neurons expressing the same receptor converge onto just one or two glomeruli in the olfactory bulb. Questions remain, however, about the fine-grained wiring and projection patterns in the human olfactory system as well as the fine-grained cellular and molecular features of the human olfactory bulb.

A paradigm shift in olfactory research occurred when a study by Maresh et al. (2008) found that the average number of human olfactory bulb glomeruli is >5,500, yielding a convergence ratio of about 16:1. Data suggested that the initial coding of odor information in the human olfactory bulb may differ from the models developed for rodents, and the recruitment of additional glomeruli for olfactory receptor subpopulations may contribute to more robust odor representations. Dr. Gottfried described his recent, unpublished work comparing differences between mice and humans in how odor information at the olfactory epithelium is packaged and presented to the olfactory bulb, which has implications for shaping future research and increasing our understanding of the close connections between the olfactory system and other brain regions.

In the context of the work presented by other speakers, clinical observations that olfactory impairment often precedes other major disease symptoms and predicts severity of disease highlights the importance of better understanding human olfactory function. We know that olfactory impairments coincide with deficits in cognition, motivation, memory, and emotion, but further research in both mice and humans is needed to determine whether olfactory dysfunction has a causal relationship with neurological and psychiatric disorders or is one of the symptoms.

Smell Loss as a Marker for Cognitive Decline

Maria Larsson, Ph.D., Stockholm University

Olfaction is unique among the other senses because it has direct projections to the areas of the brain that are affected earliest in both the normal aging process and in preclinical phases of dementia. Odor identification tests also put heavy demands on other cognitive functions, such as semantic memory and perceptual speed. Imaging studies clearly show that different parts of the brain are activated depending on the type of task a subject is encountering.

Results from a large longitudinal study, the Swedish National Study on Aging and Care in Kungsholmen (SNAC-K) provide an extraordinary window into changes in olfaction with age and relation to dementia, bolstering the idea discussed throughout these sessions that loss of smell could serve as an early warning sign of cognitive decline and age-related health changes. SNAC-K includes longitudinal measures of declarative odor memory (i.e., episodic odor memory, odor identification), a variety of cognitive domains, and health factors.
In one study, researchers hypothesized that performance in the episodic odor memory test would cluster with results from the other cognitive episodic tests, including word recall and word recognition. In reality the olfactory domain was highly unique and did not cluster with the episodic or semantic memory tasks. The study showed that when assessing olfaction by means of memory function, declarative odor memory forms a unique entity in old age. Results also show that the neural organization is fundamentally different from other types of sensory modalities and olfaction provides additional information about the cortical integrity of the brain.

SNAC-K has also been used to evaluate markers of olfactory dysfunction for estimating hazard of dementia in older adults. In data from 2,473 individuals with a mean age of 70 years, olfactory dysfunction was associated with increased hazard of dementia, with the strongest association for anosmia. Results remained significant across age and sex groups even after adjusting for potential confounders. APOE ε4 carriers with anosmia had the highest hazard of dementia, indicating the importance of genetic risk of AD. Although some olfactory loss may be an inevitable part of aging, research based on SNAC-K has also highlighted the confluence of demographic, vascular, and genetic factors that contribute to rate of decline in odor identification in aging. Insights from SNAC-K into olfactory-language interactions could be harnessed to develop and optimize brief and efficient odor identification tests for older adults.

Poor Olfaction and the Health of Older Adults Beyond ADRD

Honglei Chen, M.D., Ph.D., Michigan State University

The Health, Aging and Body Composition (Health ABC) study was designed to investigate the role of body composition changes in functional decline, disability, and longevity in older adults. The study, funded by NIA, enrolled more than 3,000 community-dwelling older adults (48% male and 37% African American) without functional limitations in 1997-1998 and study participants were followed longitudinally for more than a decade. Sense of smell was assessed once, at the year 3 clinical visit in 2000 to 2001. Compared with study participants with good olfaction, those with poor olfaction had a three-fold higher risk for dementia and a five-fold higher risk for Parkinson's Disease (PD). These associations were independent of potential confounding from population demographics, lifestyle factors, and major health indicators. They also found that poor olfaction robustly predicts a higher mortality among older adults, another now well-documented finding for poor olfaction.

Interestingly, the Health ABC study also provided some of the first epidemiological evidence that only about 22% of the excess mortality in older adults associated with poor olfaction could be explained by dementia or PD, leading researchers to explore the broader health ramifications of poor olfaction besides neurodegenerative diseases. Dr. Chen presented results of several recently published findings that documented poor olfaction in older adults was associated with higher risks of pneumonia hospitalization, physical functioning decline, decline in mobility, depression, kidney function, and congestive heart failure. Findings support the notion that poor olfaction is associated with adverse health outcomes across multiple biological systems. The potential role of poor olfaction as a marker of age acceleration was discussed, reflecting underlying pathogenesis for other age-related diseases.

Further study is needed to confirm these findings and understand olfaction’s broader implications for the health of older adults beyond ADRD as a risk marker for or contributor to adverse health outcomes. Key issues moving forward are whether olfaction is a target for intervention or a red flag of evolving health issues which requires intervention from another actionable perspective (e.g., health diets, more exercise, or medical).

Panel Discussion and Q&A

Moderator: Pamela Dalton

A panel member sought to bring together multiple presentations by identifying the apparent role of inflammation as a potential mediator of changes in olfactory function. Dr. Dalton acknowledged limited work in this area, with varying relationships between type of inflammation and outcomes, as well as some neuroprotective factors. While inflammation seems to be a common theme across database studies, Dr. Gottfried noted that general inflammation was not an identified factor in either human or animal models and he expects that the role and mechanisms of inflammation will be more specific than general counts of white blood cells, macrophages, or neutrophils. Other commentors concurred, reasoning that a large cohort would be needed to understand which cells and which mediators are involved. Newer methods for sequencing are making this approach easier with samples collected through nasal swabs. Collaboration across programs could produce more robust data sets to answer some of these questions.

Integrity of the olfactory epithelium is reflected in the mucus and the amount of mucus decreases naturally with age. Discussion focused on differences in respiratory and olfactory mucosa, which secrete different types of mucous that comingle, creating challenges in determining which is the source of findings in the inflammatory milieu of a sample. Dr. Dalton remarked that collecting simultaneously from multiple sites can help researchers to infer the likely source of the mucous of interest. It was noted that the process of collection can also create mixing.

The established relationship between traumatic brain injury and dementia, in the context of any mediating role of trauma to the olfactory bulb, was also discussed. At least one study has indicated that peripheral traumatic smell loss does not translate to later dementia. Rodent studies have also indicated that cutting the olfactory nerve does not have the same relationship with depressive phenotypes as removal of the olfactory bulb. One commentator discussed his ongoing, unpublished research related to traumatic brain injury, making the point that start of injury is far easier to pinpoint in studies of traumatic brain injury, whereas it is far harder to identify when dementia, or the path to dementia, begins.

Dr. Larsson was asked about future directions of her work. She discussed looking at smell loss as part of a larger process of losing parts of the regenerative process with age, as well as noting ongoing work to replicate the findings of olfactory-mortality link in another population. Replication of these results would strengthen the research findings, but would not elucidate mechanisms or moderators. Dr. Chen suggested the value of combining signals across work in this area—which encompasses epidemiological studies, human studies, autopsy studies, and animal models—to better understand the relationship of olfactory dysfunction and aging. The relationship between olfactory dysfunction and dementia risk or mortality is evidenced across a wide range of current and emerging research. The prevailing feeling was that olfaction will likely be found not to be a target for intervention in its own right, but is more likely to be a marker for something else.
The value of olfactory tests as a means of identifying dementia risk at an early stage is tantalizing, but there is still work to do in setting clear thresholds regarding what is considered abnormal. This is problematized by the lack of research in diverse populations, apart from the Health ABC study discussed by Dr. Chen.

Probing Brain Health with Digital Accessible Remote Olfactory Related Health Assessments: Early Detection and Augmenting Clinical Trials

Session Chair: Mark Albers, M.D., Ph.D., Harvard University

The onset rate for AD is 15-20 years. Viewing Clifford Jack’s visual model of the sequence and timing of imaging and biomarker changes in AD, biomarkers appear in stage 1, while a person is still asymptomatic. The challenge is to detect the disease at this stage, before individuals exhibit mild cognitive impairment or progress to dementia. As discussed in presentations earlier in the day, it is becoming clear that smell needs to be a part of that testing.

AD is a heterogenous disease with many developmental pathways, complicating detection through biomarkers. Functional tests that are molecularly agnostic for use in screening may be an answer to this challenge. A simple at-home test could be used to flag risk and make screening more accessible by identifying who could benefit from more expensive and invasive testing. Dr. Albers described an at-home test currently in development, which grew out of research funded by NIA. As described in the session and on clinicaltrials.gov (NCT05881239) the goal is to develop a primary screen for identifying individuals at risk for developing AD and follow disease progression in individuals without overt cognitive symptoms. The test, which is embedded in several ongoing trials, includes measures of odor intensity, identification, memory, and discrimination. Results from these trials will be used to refine the at-home test, with the goal of making remote screening for preclinical AD widely available to address this unmet clinical need.

Longitudinal Findings on Olfaction and Biomarkers of Brain Aging and Alzheimer’s Disease

Qu (Teresa) Tian, Ph.D., NIA Intramural Research Program

Olfactory function declines with age and olfactory dysfunction is one of the earliest features of AD and PD, but data has primarily been cross sectional and are limited regarding the mechanisms underlying this relationship. Dr. Tian presented data from the NIA Intramural Research Program’s Baltimore Longitudinal Study of Aging (BLSA), which began in 1958. Amyloid brain imaging began in 2005 and tau imaging and olfaction were added in 2015. Results support the conclusion that olfactory decline is related to worsening of AD pathology, brain atrophy, and cognition.

In one study in which odor identification tests were administered to 364 participants at two year intervals, results revealed that lower odor identification was associated with a higher risk of developing mild cognitive impairment (MCI) in those age 60+ even after accounting for APOE Ɛ4 status. Loss of olfaction was related to faster buildup of amyloid and tau deposits over time. Dr. Tian and her team also made longitudinal assessments of data from the Baltimore study based on changes in cognition and neuroimages of brain structure done up to 14 years prior to and five years after 2015. The association among decline in olfaction, cognitive decline (memory, attention, psychomotor speed, and manual dexterity), and brain atrophy were localized to the same brain regions—orbitofrontal, insula, and temporal—indicating that olfactory decline is related to worsening of AD pathology, brain atrophy, and cognition.

Future studies with longer follow-up are needed to understand whether reduced olfactory function precedes cognitive changes and whether these associations are mediated through brain atrophy. More research is also needed to define where and when olfactory deficit occurs in the early stages of neurodegeneration and aging (visualized in relation to the Jack model), as well as to define what is meant by a meaningful decline in olfaction. Explorations of heterogeneity (e.g., sex differences) in the relationship between olfactory decline and worsening AD pathology are also needed.

Olfactory Impairment as an Early Biomarker of Alzheimer's Disease

D.P. Devanand, M.D., Columbia University

Neuronal damage to regions in the olfactory pathways likely underlies the relationship between odor identification impairment and AD, PD, and related disorders. Researchers know that the olfactory bulb is affected in early AD, and odor identification deficits during life correlate with tangles in the OB and olfactory projection areas at autopsy.

Analysis of data from the Washington Heights/Inwood Columbia Aging Project (WHICAP) by Dr. Devanand’s research team showed that impaired odor identification and impaired global cognition are associated with an increased likelihood for transition to dementia. The sample was older (average age 80) and diverse (majority Hispanic and African American; 25-30% Caucasian). Olfaction tests were not initially a part of the study when it began in the 1990s; they were added as a supplement at a later point, leading to the high average age in analysis. Impaired odor identification predicted transition to AD and odor identification test results indicated that the number of people with intact odor identification who transition to dementia is very low—3.4% for the entire sample—although the oldest intact cohorts transition to dementia more frequently. Similar trends were seen in the Health ABC study which was discussed in an earlier session.

The inverse question remained, however: Does a strong performance on both odor identification and global cognition tests predict less chance of transition to dementia or cognitive decline during long-term follow-up? Dr. Devanand’s data showed that pairing odor identification testing with global cognitive screening identified individuals at low risk for transition to dementia and had high positive predictive value. This can reduce the need for extensive testing to establish a diagnosis and may be valuable when screening participants for clinical trials. Predictive results are independent of race, unlike cognitive tests and the APOE genotype. The findings from these studies and other epidemiological studies in community cohorts need to be examined in clinical settings, where patients present with cognitive complaints.

Environmental Determinants of Alzheimer’s Disease: The Human Olfactory System, Sensor, and Portal

Jayant Pinto, M.D., University of Chicago

Both nature and nurture are involved in age-related olfactory decline: there are genetic, demographic, and behavioral inputs, as well as environmental exposures. This session focused on environment as a factor that is modifiable and could potentially be used to improve olfaction and decrease risk of neurodegenerative disease.
Dr. Pinto is part of the National Social Life, Health & Aging Project (NSHAP), a nationally-representative, community-based, diverse probability sample. All five senses were measured at baseline, and at five, 10, and 15-year follow-ups, including serial measures of olfaction that allow an examination of air pollution and olfactory function. Dr. Pinto’s team knew where research subjects resided and were able to generate validated spaciotemporal models to measure air pollution exposure, then study how it affects olfactory health outcomes. Increased particulate matter (PM) of PM_2.5 or less was associated with increased odds of olfactory dysfunction, suggesting that exposure to pollution does have deleterious effects on the olfactory system.

The NSHAP analysis led to Dr. Pinto’s involvement in AERONOSE, an NIA-funded project to study pollution effects on the olfactory system and how they relate to neuropathology and dementia. Major barriers to understanding the role of air pollution in development of AD have included inadequate assessment of cognitive decline, over-reliance on medical records and claims data to measure dementia status, sparse information on the connection of air pollution exposure to dementia pathology, and poor understanding of underlying mechanisms, including cerebrovascular and olfactory pathways as well as PM access to the brain. AERONOSE takes advantage of five large, diverse ongoing longitudinal cohort studies of neurologic aging that have collected rich clinical information on >4000 older adults: the Memory and Aging Project, the Minority Aging Research Study, the Religious Orders Study, the Latino CORE Study, and the Clinical Core. The data collected include annual assessments of cognition, dementia status, and olfaction, as well as brain specimens at autopsy (N>1600). The team is using models to estimate participants' long-term air pollution exposures and is the first study to use state-of-the-art, advanced microscopy to directly measure particulate matter in the olfactory bulb. Early results of the study were described, and quality and content of images captured were demonstrated through several examples. The project is generating pivotal new information on the contribution of air pollution to AD, with the long-term goals of reducing the burden of AD on a population scale through public health interventions.

COVID-19 Anosmia as a Window into Inflammatory Markers of Olfactory Dysfunction

Lora Bankova, M.D., Harvard Medical School

Earlier discussion considered the long-term effects of COVID-19 on olfactory dysfunction and what influence this could have on future development of neurodegenerative disease. The large number of people experiencing chronic olfactory dysfunction as a result of COVID-19 should also be considered as a potential confounder in studies of olfactory dysfunction and AD risk. In one study, 61% of people in one study of COVID-19 reported acute loss of smell and one in four of those reported only partial recovery (24%) or no recovery (3.7%) of smell.

Inspired by the COVID-19 pandemic and a desire to help her patients who were experiencing olfactory dysfunction, Dr. Bankova and her team at the Sinus Center at Brigham and Women’s Hospital began recruiting people from clinics for a study of COVID19-related anosmia. Researchers tested subjects’ sense of smell with odor identification tests and measured sense of taste using filter strips. The team then collected samples by directing subjects to swab their noses, followed by bulk RNA sequencing, flow cytometry, single cell RNA sequencing, and analysis of proteins and lipids in the nasal fluid. Dr. Bankova presented initial results of the ongoing study, with the caveat that methodology has been refined over time and results will be confirmed with a larger sample once funding is obtained. Individuals with anosmia or dysgeusia were compared with age-matched healthy controls. While there is still much to learn, early results indicate there is an inflammatory signature that merits further exploration. Whether this inflammation is related to inflammation and neuroinflammation in development of AD is as yet unknown, it will be explored further.

Session 2 Panel Discussion and Q&A

Moderator: Mark Albers

There was discussion about whether some of Dr. Bankova’s observations were related to persistent breakdown in the epithelial barrier in long COVID patients, and how these changes could potentially be modeled. The group made some observations about the initial study results and offered advice for Dr. Bankova and her team to inform a larger scale study.
Following on topics discussed in Dr. Bankova’s presentation, a question was asked about the value of measuring sense of taste in conjunction with sense of smell, in the context of AD studies. Not nearly as many studies are available regarding the relationship between taste and AD and tools for measuring taste dysfunction are lacking. Data thus far have shown that change in taste come late in the AD development spectrum and is not likely to be a helpful early marker. However, it was noted that these studies have been based on available data and the role of dysgeusia in neurodegenerative disease could be studied in a more systematic way.

In regard to Dr. Pinto’s presentation, there was additional discussion of development of models for the relationship of pollution to olfactory dysfunction. The group considered difficulties in scale, data gathering, and obtaining long-term exposure. Dr. Pinto noted that the PM_2.5 cutoff used in his study and others can be viewed as arbitrary and it ignores a lot of other fine particulates that we are breathing in. They are also beginning to look at indoor exposure—most of the work thus far has focused on outdoor exposure. Chemical toxins, emissions from gas stoves, and other exposures are present in an indoor environment. It was noted that sealing buildings for energy efficiency could have unintended consequences of increasing exposure to indoor toxins.

The panel discussed environmental exposure as a potential manifestation of social determinants of health. It was recognized that this is a key issue and Dr. Pinto’s team is examining differential exposure. Other speakers noted that low enrollment of non-White participants in many studies limits the ability to fully understand racial/ethnic differences in the relationship between olfactory dysfunction and AD risk and the role that social determinants of health play in differential rates of AD development across race and gender. Amidst discussion of different versions of smell tests, there was discussion about how tests perform with different populations, both diverse populations in the U.S. and in different countries. Tests can have wildly different results across countries and cultures and need to be adjusted for context.

Discussion ended with a conversation about how decisions are made determining what weight to give to different components of the tests when deciding whether someone has a deficit. Dr. Albers noted that his team is currently working on this issue, exploring how tests in phenotyped individuals correlate with biomarkers. If we can identify signatures that correlate with presence of pathology in development of neurodegenerative disease, results could be used to develop algorithms that could predict increased risk of AD pathology of one type or another. This would allow patients to get through the testing pathway more efficiently. We will need large datasets and computation tools to start to address these questions.

Mouse Olfactory Epithelium Is Differently Sensitive to Human Apolipoprotein E Alleles

Timothy McClintock, Ph.D., University of Kentucky

Familial AD due to causative genes represents less than 5% of AD patients. Late onset AD accounts for the other 95%, and the greatest risk factor is the APOE Ɛ4 allele. Researchers poorly understand how APOE connects to the factors of AD pathogenesis. Dr. McClintock’s research team set out to explore what the olfactory epithelium (OE) can reveal about those links. The team studied APOE mice engineered to express the three most common human APOE alleles—Ɛ2, Ɛ3, and Ɛ4.

Published preliminary data showed glucose hypometabolism at age six months, which is strongly linked to the APOE Ɛ4 genotype. The APOE Ɛ4 samples showed less glucose uptake in the OE relative to Ɛ3 samples. Results also showed transcriptome differences at age six months between Ɛ3 and Ɛ4 mice. One of these transcripts—asparagine synthetase—is glucose-sensitive and inversely related to glucose uptake. Less glucose uptake in Ɛ4 mice correlates with increased amounts of this mRNA. Researchers also detected age-dependent increase in loss of olfactory sensory neurons and possible mitochondrial dysfunction in the OE of Ɛ4 mice as compared to Ɛ3 mice. Data suggest a progression of worsening events in the epithelium as the mouse ages, which mimics the progression of the disease in human populations.

Ongoing studies of the OE tissue will include observations of transcriptome and lipidome, the immune response, glucose hypometabolism at different ages, and olfactory sensory neuron-specific functions (e.g., synapse density and turnover). Team members have also developed a new mouse model which will allow them to study the effects of timing of the OE switch, either early in life or later in life, which could provide insight into the reversibility of Ɛ4 effects.

Olfactory Deficits in Parkinson’s Disease

Sreeganga Chandra, Ph.D., Yale University

PD, like AD, has a long prodromal phase before symptoms manifest. Because the disease lacks an imaging biomarker to identify a prodromal patient, interest is growing in rapid eye movement (REM) sleep behavior disorder (RBD), where patients act out their dreams because of muscle atonia loss. Thirty percent of all PD patients have REM RBD. Seventy-five percent of these REM RBD patients phenoconvert to PD within 10 years, 15% develop Lewy Body dementia, and the remaining patients develop a synuclein-related disease called multiple systems atrophy. Researchers are interested in how REM RBD phenoconverts to PD. Data indicate that olfactory function and conversion to PD are inversely correlated.
Dr. Chandra described a mouse study published in 2023 that used wildtype and α-synuclein transgenic (α-syn-Tg) mice to study olfactory deficit in PD. Results showed that mice performed normally at age 6-7 months on a buried food test. By 12-14 months, representing late-stage motor progression, the α-syn-Tg mice took much longer to complete the test and had Parkinsonian-like phenotypes. Transgenic expression caused α-syn pathology along the entire central olfactory pathway, primarily affecting projection neurons. It was noted that α-syn pathology did not cause structural and morphologic abnormalities in the olfactory system. Dysfunction included a reduction in OB neurogenesis and changes in synaptic vesicular transport affecting synaptic function, both of which are likely contributing to olfactory behavioral deficits. Functional fluoroscopic in vivo imaging of the mice was stalled by the COVID-19 pandemic, and Dr. Chandra’s team is currently attempting to resume these experiments.

Learning and Forgetting in the Olfactory Cortex

Carl Schoonover, Ph.D., Columbia University

Odor identification becomes increasingly difficult as people progress from cognitively normal to possible/probable AD. Part of this progression is due to impaired recall of olfactory memories, possibly indicating that olfactory memories are being erased. One way to erase a memory is through overwriting, known in psychology as retroactive interference. Even healthy humans and animals are constantly losing memories by overwriting the old as they form the new. Dr. Schoonover discussed the role of the piriform cortex (PC), the largest olfactory region, which has been proposed to support, odor identification, associative learning, and pattern completion. Unstable odor responses would present a challenge in identifying scent. He is exploring the possibility that these processes are a possible explanation, at least in part, for the deficits in odor identification that are being observed clinically in early phases of MCI.

Dr. Schoonover’s and his colleagues looked at whether odor responses in the PC are stable over time in this non-topographic network. To do so, researchers presented mice with neutral odorants every eight days, using a chronically implanted silicon probe to record neurons. The neurons that preserved responses represented only 3%, with the other 97% of neuron responses changing over time. The team studied whether the PC retains and stabilizes responses to odors only when they are salient to the animal by testing neuronal response after an odor/shock pairing, comparing these to an unpaired odor, and to a neutral odor. While results indicated that the animal learned and retained knowledge over time, physiological change in response to these same odors exhibited drift at about the same rate for all three odor types. The performance of a linear classifier trained on the first recording day approached chance levels after 32 days. Despite this drift, the animals recognized the paired odor and responded appropriately. Researchers also found that if an animal receives a daily repeated experience of an odor, that dramatically stabilizes the animal’s response to it. Dr. Schoonover then described unpublished work designed to test whether passive olfactory experiences alter the synaptic connectivity of piriform network.

Drift could simply be the result of a system that is continuously learning and continuously overwriting itself over the lifetime of the animal. A century of human and animal experimental psychology indicates that memories do not get overwritten randomly. The process seems to be a well-regulated operation. Whatever regulates this orderly overriding of odor memories may have gone awry in AD. Mouse models can be used to study memory overwriting in the piriform cortex through long term, longitudinal observations. These will produce two yardsticks: drift rate and direction, and alterations in synaptic connectivity. These can probe the phenomenology of overwriting, its mechanism, and how it may go wrong in AD.

Early Events in Olfactory Circuit and Perceptual Dysfunction in Parkinson’s Model Mice

Dan Wesson, Ph.D., University of Florida, Session Chair

Changes in olfactory function may serve as a magnifying glass to examine disease pathogenesis in PD and possible reciprocal relationships among pathologies in AD, PD, and other neurodegenerative conditions. As part of an investigation into whether pathological α-syn in the OB impacts odor perception, Dr. Wesson’s team built a system to assay mouse olfactory perception by monitoring changes in the respiratory dynamics during sniffing. Researchers found no changes in basic respiration and no pathology in the cerebellum or respiratory pons that accompanied reduced smell perception in mice, indicating preservation of the ability to coordinate respiration.

Deficits in olfactory perception and aggregation of α-synuclein fibrils (α-syn) in the olfactory bulb (OB) are observed during early stages of PD and have been associated with the PD prodrome, before onset of the classic motor deficits. To further investigate changes in odor perception, Dr. Wesson’s team seeded recombinant pre-formed fibrils of α-syn in the OB, aged mice for one to three months, and measured the cortical field potential. The team found that even one month after injection, the seeding of α-syn in the OB drove a high amount of aberrant beta oscillations in the bulb. Elevated amounts of beta oscillations were seen across many odors and were consistent across animals. This was the first exploration of in vivo changes in nerve physiology using this model in PD. Since α-syn seeding in the OB disrupts neural activity, and beta oscillations are linked to odor perception and learning, this may explain perceptual deficits. Results so far provide evidence that synuclein is a component in the pathogenesis of PD that alters physiology of the olfactory system in ways that are important for changing odor perception.

Dr. Wesson’s team has also been investigating whether changes in beta oscillations alter the structure of odor-evoked action potential firing and is now investigating how these changes are altering the synchrony of action potentials in a control animal vs. a pre-formed fibrils-treated animal. Remaining knowledge gaps include which cells are functionally vulnerable to α-syn pathology and why, the link between cellular dysfunction (e.g., neuroinflammation, oxidative stress) during pathogenesis and changes in olfactory function, the role of glial pathology and how it affects olfactory neural activity, and the relationship between pathological burden, network dysfunction, and clinical manifestation. Determining whether the olfactory bulb could be a starting point of pathology and pathological spread is crucial to understanding how neurodegenerative diseases evolve.

COVID-19 as an Olfactory Route to Neurodegeneration

Leslie Kay, Ph.D., University of Chicago

Dr. Kay, whose work has been cited repeatedly over the preceding sessions, discussed her studies related to persistent COVID-19 olfactory dysfunction. If this dysfunction is driven by olfactory bulb damage, she reasoned, it could suggest a wave of post-COVID dementia in the coming decades. SARS-CoV-2 can cause inflammation and viral invasion of the olfactory bulb, initiating a cascade of degeneration similar to AD and Lewy body disease. People who have had even mild cases of COVID-19 show signs of degeneration in cortical areas connected with the olfactory system.

The olfactory bulb is a major part of the limbic system and is involved in its cognitive changes. When the OB is removed completely, it results in anhedonia, which is why an olfactory bulbectomy model is often used in mouse models for treatment of depression. Research in the Kay lab and others suggests that these effects are obtained even with OB silencing. This underscores the OB’s constant conditioning of the limbic system, affecting how the system operates moment-to-moment.

Various pathways besides the olfactory sensory neurons can deliver viruses and the effects of viruses (e.g., inflammatory molecules) to the brain. COVID-19 does not enter the sensory neurons, but other possibilities for entry of COVID infection into the olfactory bulb include ensheathing cells, cerebral spinal fluid, the trigeminal or terminal nerves, or the shared capillary system between the OB and sensory epithelium. There is a pressing need for more research on treatments for olfactory dysfunction and longitudinal studies including cognitive and olfactory function from patients who have recovered from even mild COVID-19. Small studies have shown correlation between persistent smell loss in COVID-19 and anxiety and general cognitive impairment. The UK longitudinal PROTECT study, which included 3,142 older adults with a mean age of 67.5 years, looked at changes in cognition over the first 2 years of the COVID-19 pandemic, in comparison to year before COVID, although changes in sense of smell were not tested. People who contracted COVID and those who did not began at similar levels of working memory and executive function in 2017-2018, but those who contracted COVID experienced sharper declines in both working memory and executive function through 2021-2022 compared to those who did not contract COVID-19. Subjects with Mild Cognitive Impairment at the beginning of the pandemic declined more than the other groups. Decline was also associated with decreased levels of exercise and increased levels of loneliness and depression in the COVID group. Given the influence of the OB and its pathology on systemwide activity, more studies are needed using human and animal models to investigate changes in systemwide connection patterns over time that are associated with the OB.

Session 3 Panel Discussion and Q&A

Discussion on the role of the olfactory system resulted in agreement that more study is needed on the non-sensory roles of the OB. For example, while the OB has a robust immune system, it also handles glucose and may be involved in spatial cognition. The OB projects to areas such as the tubercle, cortex, and PC that are known to be involved in the association of odor with emotions, memories, context and meaning, and action. More study is needed for an expanded view of the interactions of these areas and the roles they play beyond odor processing.

Discussing Dr. Schoonover’s work related to AD and OB drift, the group agreed that more understanding of this relationship is needed. One example would be an experiment in which human subjects are presented with the same set of odors two to four weeks apart. In between those time points, participants would be presented with an entirely different set of odors. At the end of the two-to-four-week period, subjects would be asked if they remember smelling the first set of odors. Researchers would determine the extent to which interference affects the subjects’ ability to recognize the odors from weeks prior. If an increased rate of interference predicted future MCI this could serve as an additional olfaction-based test.
Commenting on Dr. Schoonover’s presentation, one participant noted that overwriting of memories may not necessarily be the explanation for problems of recall of olfactory memories in AD. It could be a problem with indexing—i.e., the memory is still there, but can’t be accessed. The site of indexing is the olfactory bulb, which could be drifting in its own way.

Workshop participants discussed the relationship of emotional attachment to odors and dementia. Dr. Kay noted that sensory perception and memory are not separate functions and work together in a nonrandom way. Still, people do not remember odors any better because of the emotional content, although they feel like they do. Dr. Larsson cited her work with autobiographical memories. People typically generated fewer memories from odor cues compared to verbal cues and the memories generated often belonged to the childhood period. Dr. Kay explained that when her team trained rats to associate an odor with something aversive, the rats persisted in the negative association, even though the rats were trained to other odors in the interim.

There was considerable discussion about the sufficiency of existing animal models and the need for more reproducibility in methods. Dr. Wesson observed that research groups have not been very careful in their papers in describing the specific steps, so we have a real benchmark for what the rigor should look like. It was agreed that reproducibility and consistency of rigor needs a lot of work.

AD is a complex disease. The path from mouse model to clinical trial falters in translation. When clinical trials are designed, many researchers are still treating AD as one disease rather than thinking about it more mechanistically. This is changing with the field’s new openness in studying AD biomarkers. There was also a call for research funding to support the study of multiple strains of mice to mimic human genetic diversity.

Meeting Summary | Day 2

Olfactory Gene Expression Programs in Health and Disease

Sandeep Robert (Bob) Datta, M.D., Ph.D., Harvard Medical School

It is almost certain that neural biomarkers will be more sensitive than cognitive and perceptual biomarkers for identifying AD development at the earliest stages, increasing the possibility of early diagnosis and prevention. Developing sequencing-based biomarkers is new to AD research, but it is the standard in cancer and other fields. This type of work also provides an opportunity to look at neurons directly in the context of disease, which is central to understanding AD. Dr. Datta’s group created an environmental state score to aid in capturing environment-dependent changes in gene expression, as described in a 2021 paper. They found that olfactory sensory neurons adaptively shape their odor responses through environment-dependent changes in gene expression as a means of separating salient olfactory cues from predictable background.

Incorporating this work in environmental state scores, Dr. Datta’s lab is working to identify transcriptional biomarkers for olfactory function. Researchers are sampling the OE and performing sequencing to identify transcriptional signatures that reflect the underlying function of the olfactory system in the periphery. His work uses gene expression in the nose as a window onto olfactory function. A huge amount of data is needed, because there is not a single olfactory sensory neuron type, there are a thousand subtypes, each characterized by olfactory receptor expressed. Dr. Datta described his ongoing, unpublished work in this area, early findings, and the promising biomarkers identified thus far. The next step is to test the use of these biomarkers in humans. Dr. Datta is collaborating with Dr. Goldstein, the following speaker. Early evidence suggests that the genes they are interested in are present in humans, but it is early in the research process and the sample is still small. Their early work suggests that for the first time we can do molecular biology in living humans and assess their olfactory function.

Approaches to Understanding Olfactory Pathobiology and Repair in Rodent Models and Human

Bradley Goldstein, M.D., Ph.D., Duke University, Session Chair

Following on the work presented by Dr. Datta, it is clear that the olfactory periphery is a complex organ that does more than just passively relay odors. The OE is highly vulnerable to damage and requires continuous renewal to maintain function. Basal stem cells have key roles in the maintenance and regeneration of the OE, and it is important to understand how damage and repair occur and how damage impacts maintenance of the olfactory populations with aging or disease. Research has shown that rodent OE supports ongoing neurogenesis through adulthood, but there has been little direct evidence to evaluate how well human olfactory neurogenesis persists across a much longer lifespan. Most of what is known in humans comes from postmortem studies, which are tremendously valuable but reflect end-stage fulminant disease.

Dr. Goldstein described his team’s work in studying the roles for Polycomb repressive complex 2 (PRC2) in cell renewal or differentiation in the OE. They found that perturbing PRC2 alters the expression of transcription factors in basal cells. PRC2 on olfactory basal cells regulate cell fate decisions. In the mouse model, PRC2 was a master regulator to control or repress non-neural gene expression and promote a normal differentiation pathway. This finding provides a mechanistic understanding of how cell repair works that may lead to therapeutic approaches for damage to the periphery and neurogenic exhaustion.

As others speakers have noted, we can consider the olfactory epithelium to be a window into the nervous system, and we have the ability to explore that at any stage of AD development—anywhere along the curve of Clifford Jack’s model of the sequence and time of biomarkers of AD development. The AD field is poised to consider translational approaches to develop human therapies. By applying a brush biopsy-based single-cell RNA sequencing approach to look at the human periphery, for example, researchers can sample what is going on in the nervous system at various disease stages. Researchers can study the neuronal lineage cell cluster and compare AD versus control olfactory sensory neurons. Biopsy approaches and downstream assays will continue to be refined as technologies evolve.

Nasal Anatomy and Respiratory Effort in the Context of Age-Related Olfactory Changes: Investigating the Conductive Component of Olfaction

Dennis Onyeka Frank-Ito, Ph.D., Duke University

Variations in the human nose are often discussed in terms of morphological differences based on race and ethnic groups. Less discussed are the effects of nasal vestibule morphological variations on airflow and odorant transport to the olfactory cleft. When viewed from the side, human nasal vestibule variations can be described as standard, notched (indentation in the airway), and elongated. Dr. Frank-Ito described a study in which computer modeling was used to produce 3D reconstructions of participants’ nasal airways, with discretization applied to produce airflow simulation for the study of odorant and drug particle transport. The study sample size included 12 subjects with standard vestibules and eight with notched vestibules. Odorant transport was simulated for three different odorants and olfactory cleft odorant flux was computed for each simulation. Results showed that air flow in the olfactory cleft was greater in the standard phenotype compared to the notched phenotype. The impact of nasal vestibule morphological variations on odorant flux at the olfactory cleft may have implications for individual differences in olfaction, a topic that needs further investigation. Dr. Frank-Ito described ongoing, unpublished research into age-related olfactory airflow, which has implications for future intranasal spray and drug design to target the olfactory airspace.

Advanced Neurophotonics, Proteomics, and Transcriptomics Shed Light on the Potential Link Between Olfactory Dysfunction and Progression of Alzheimer’s Disease

Diego Restrepo, Ph.D., University of Colorado

Dr. Restrepo’s team set out to determine whether alpha herpes virus infection of the olfactory system plays a role in cognitive decline in AD. Varicella zoster virus (VZV) and herpes simplex virus type 1 (HSV-1) are present in about 90% of us, persisting in the body as latent infections that turn on and off over time after the primary infection. Previous epidemiological studies have shown that a dermatomal VZV infection causes a small, elevated risk of AD. In one study (in preprint) of 282,541 adults in Wales, a distinct drop in probability of new onset dementia is seen in individuals born after 1933, when the Zolstavax vaccine became available for VZV. Interestingly, the change was much larger in women then men. In a large population-based sample in Sweden, cumulative incidence of dementia was significantly higher in those with VZV who received no treatment, in comparison to those who were treated.

Advanced spatial proteomics, transcriptomics, and neurophotonics can assist investigators in unraveling the etiology of the involvement of olfaction in AD. Dr. Restrepo’s research team found that alpha herpes virus infection of the olfactory system plays a role in AD cognitive decline. The study used tissue samples from controls and patients with familial AD (FAD), all aged 45-55 years. Researchers studied OB tissue to determine the consequences of a viral OB infection from VZV or HSV-1. Proteomic analysis revealed spatial immune and AD responses in the olfactory tract and bulb of FAD samples. When examining protein expression differences between brain regions within the control and FAD groups, researchers found that FAD patients had a distinct group of proteins in the high myelin area, all involving remyelination. These findings raise the possibility that viral infection and associated inflammation and dysregulation of myelination of the olfactory system may disrupt hippocampal function, contributing to acceleration of FAD progression.

Imaging the Olfactory System Using Advanced Structural, Functional, and Physiological MRI

Jun Hua, Ph.D., Johns Hopkins University

Dr. Hua brought a biomedical engineering perspective to the discussion. The olfactory regions present a challenge to magnetic resonance imaging (MRI). A homogenous field produces the best image, but the olfactory regions are very heterogenous and have a magnetic susceptibility difference between the nasal cavity, tissue, and bone. MRI has improved, however, with new technology development over the past decade. Particularly, high field 7 Tesla (T) MRI scanners provide superior sensitivity to map small structures in the olfactory system that cannot be discerned with current MRI methods. The Food and Drug Administration approved the first high field 7 Tesla (T) MRI scanner for clinical use in 2017. Although a scan of a human OB at 7T does not show distinct layers, it begins to show smaller structures in the bulb.

Dr. Hua described development and testing of a T2-prepared blood-oxygenation-level-dependent (T2prep BOLD) fMRI to provide clearer access to regions at the interfaces of bone, soft tissue, and air. Dr. Hua’s lab now uses T2prep BOLD in its studies to detect robust signals from the OB in human subjects. He described early results of ongoing work exploring olfactory fMRI of the OB in subjects with early PD. High field 7 Tesla advanced imaging of small structures in the olfactory system shows clear advantage over the current mainstream 3T MRI scanners. The results obtained using the T2prep BOLD fMRI technique mean the human OB is no longer a blind spot for fMRI. Technical work is still needed to validate early research results suggesting that physiological changes in the olfactory region may be important factors in the development of neurogenerative diseases.

Session 4 Panel Discussion and Q&A

Moderator: Bradley Goldstein

Workshop participants discussed the possibility of using advanced imaging to support the theory that mitochondrial dysfunction due to environmental toxins may be involved in neurodegeneration. Dr. Hua said that although MRI can be used to look at content in tissue, this would be difficult in the OB.

Dr. Frank-Ito’s findings on the morphology of the nose sparked discussion of how this information might be applied to improve smell tests. These tests might be more efficient, for example, if the odorant is sprayed into the nostril rather than sniffed, which delivers only a small fraction of the emanated odor. The morphology of the nose may also cause variations in the way people sniff as they make physical compensations to change the shape of their noses to improve their sense of smell. It may be possible to conduct a multivariable analysis of the sniff, measure respiration, and use that information to help people more easily take a smell test. Dr. Datta also commented that as animals age, the way they engage with their environment and their characteristic sniffing patterns change. The chain of events that lead a person to explore an odor are unknown. More study is needed on the relationship between sniffing and related neural architecture.

Participants explored the new avenues of investigation opened by Dr. Datta’s work with gene expression. The group discussed the need for further information on how the brain processes the widescale change in gene expression and whether neurons regenerated after profound damage to the OE have the same transcriptional signature that allows them to upregulate the environmental state score. Dr. Datta noted that new cells take a trajectory in which environmental state scores are initially high because the olfactory sensors do not work well, then settle down to a normal dynamic range.

Dr. Goldstein noted that as the OE naturally changes with age, there is a patchy replacement of healthy olfactory epithelium with not-perfectly-normal epithelium. Researchers do not understand how and why this respiratory-like metaplasia happens. The result is that at some point, there is enough neural loss to effect function. Workshop participants discussed whether the biopsy-based processes discussed by Dr. Goldstein could be scaled to measure large numbers of people for an epidemiological study. He explained that scaling would require gaining experience with performing the brushing. Getting a reasonable yield of olfactory neurons takes experience. The protocols for handling and processing specimens are rapidly improving.

Clinical Application of Olfaction Testing

Current smell tests are helpful in early diagnosis of ADRD, yet those tests are not in wide use in clinics and practitioners’ offices. This is despite research results confirming that olfaction testing reveals a significant difference between clinically diagnosed AD patients and controls. Conversely, when a person does well on an olfaction test, it indicates that further brain imaging may not be needed. A variety of barriers prevent the wide use of olfaction testing:

• Smell tests do not have formal score classifications like blood and amyloid testing. For example, the widely used University of Pennsylvania Smell Identification Test (UPSIT) is scored out of 40, where a higher score indicates better olfaction.38 If patients score 28-30, they are not in danger of developing dementia. If they score 15-16, they are likely to have dementia or some other serious condition. Many older people score somewhere in between, and there is no formal explanation of what that means. Other tests, such as blood pressure and hemoglobin A1C, produce concrete numbers that people understand. To be useful, olfaction tests need not produce one outcome, positive or negative. There can be a range, as with prostate antigens39. The AD research field has not standardized and harmonized smell tests. Before amyloid was accepted as a biomarker, advocates worked for 10 years and held numerous international conferences. Coming to agreement was difficult. The olfactory research field has this same problem.
• The olfaction field needs to develop compelling evidence that an olfactory test and/or biomarkers will help physicians treat patients. The case could then be made to Medicare that instead of ordering six different tests, there is a much more efficient way to conduct the workflow. Once smell tests have a billing code, they will be available to those who need them. Doctors will order olfactory testing when it makes a difference in how they care for patients.
• Medical practitioners, including students, need training on olfactory testing.
• A consortium should be created to conduct biomarker research. Patients need to be part of the development of that process and can educate researchers about symptoms as the investigators put together studies. This will help increase the number of people who participate in clinical trials, especially in the wake of COVID-19. Patient involvement will also help with articulation of issues to policymakers. The center should be a coalition of clinicians, basic scientists, and patients.
• Researchers doing epidemiologic studies need encouragement to include smell tests in their protocols. Many investigations collect a huge sweep of health effects end points, and researchers do not think to include olfactory testing. NIH spearheaded a toolbox with a suite of tools that can be used by nonexperts. The NIH toolbox odor identification test costs about $2.95 and tracks the most common odor identification tests (UPSIT and B-SIT) well, but it is not well known, and people are not taking advantage of it.

Longitudinal Studies

• Tracking the same people over long periods of time with longitudinal studies is the only way to thoroughly assess change within those individuals. One example is studying older people who were severely affected by the COVID-19 virus to determine whether they are also carriers of the APOE4 e4 allele. Researchers could study the subjects’ rates of conversion to dementia and accelerated aging.
• Researchers need to collaborate and validate each other’s results across countries. This would allow investigators to combine data sets when study samples are small.
• Investigators need to broaden studies to include diverse cohorts and more brain-related outcomes, such as glucose metabolism changes, immunological changes, blood flow changes, capillary leakages, passage of pathogens between the capillary bed that feeds the epithelium and the OB, differences in nutritional states, circadian times, and sex habits. These will allow investigators to determine the causes of neuronal changes.

Imaging and the Expansion of Mechanistic Research

• Technology opens new possibilities for studying the olfactory system on both the macro and micro levels. Emerging technologies should be further developed and applied to studying the olfactory system at multiple levels spanning from subcellular imaging to neuronal circuitry mapping.
• The olfaction field needs to expand to encompass general physiological assessments of central olfactory structures in human subjects and animal models. This includes how the system processes odors, how experience and internal state modify the system, and how the system is uniquely vulnerable to age. Investigators also need a better understanding of what constitutes “normal” aging and how the system changes in the context of neurodegenerative disease.
• Because the olfactory system directly interfaces with the environment, olfactory research provides a golden opportunity to understand how environmental contributors impact neurodegenerative processes. For example, advanced imaging allows researchers to measure metal concentration in mitochondria or observe the OE in vivo to assess changes. More data is also needed on viruses and air pollutants that cause olfactory loss, then move further into the brain to trigger other neurological changes.
  

Closing Remarks

Susan Sullivan, Ph.D., NIDCD and Coryse St. Hillaire-Clarke, Ph.D., NIA

Dr. Sullivan thanked participants for coming from across the country and beyond to share their work, insights, and collective enthusiasm for understanding the relationship between olfactory function and neurogenerative diseases. The timing is right for making substantial progress based on the promising and innovative ideas surrounding biomarkers, whether they be imaging, molecular, or psychophysical. She also cited opportunities for understanding the underlying disease mechanisms and expressed hope that the workshop will promote discussions on how NIA and NIDCD can continue to work together. She encouraged participants to widen their impact by writing a white paper and emphasized the importance of the final workshop discussion as a high-level overview of next steps and goal-setting for what can be accomplished over the next five years. Dr. St. Hillaire-Clarke thanked the participants as well as members of the meeting planning team before closing the meeting.

Abbreviations

α-syn: α-synuclein
A1c: glycated hemoglobin
AD: Alzheimer’s disease
ADRD: Alzheimer’s disease and related dementias
ApoE: apolipoprotein E
BLSA: Baltimore Longitudinal Study of Aging
COVID: coronavirus disease
ES: environmental state
FAD: familial Alzheimer’s disease
Health ABC: Health, Aging, and Body Composition study
HSV-1: herpes simplex virus type 1
fMRI: functional magnetic resonance imaging
MCI: mild cognitive impairment
MRI: magnetic resonance imaging
NIA : National Institute on Aging
NIDCD: National Institute on Deafness and Other Communication Disorders
NSHAP: National Social Life, Health & Aging Project
OB: olfactory bulb
OE: olfactory epithelium
PC: piriform cortex
PD: Parkinson’s disease
PM: particulate matter
PRC2: polycomb repressive complex 2
RBD: rapid eye movement sleep disorder
REM: rapid eye movement
RNA: ribonucleic acid
SARS: severe acute respiratory syndrome
SNAC-K: Swedish National Study on Aging and Care in Kungsholmen
VZV: varicella zoster virus
WHICAP: Washington Heights/Inwood Columbia Aging Project