Partner Research Centers:


Institute on Aging

IOA Research

Some of the Latest Aging-Related Research at Penn


Newly Discovered Gene Variants Link Innate Immunity and Alzheimer's Disease

Findings give neurologists fresh ideas for enlisting immune system to fight Alzheimer's 

PHILADELPHIA — Three new gene variants, found in a genome wide association study of Alzheimer's disease (AD), point to the brain's immune cells in the onset of the disorder. These genes encode three proteins that are found in microglia, cells that are part of the brain's injury response system. The study is an international collaboration of four AD research consortia that analyzed DNA from 85,000 subjects. The results are reported online this week in Nature Genetics.

Studies of this type focus on identifying new therapeutics targets for treatment or prevention of AD, a goal of researchers world-wide. Genetic variation of the type described in this paper are "experiments of nature," of a sort, that reveal when a specific gene is altered, disease risk can be affected. 

"This is direct evidence that if drugs can be designed to target these proteins, we have a chance to alter disease risk in people," said senior author Gerard Schellenberg, PhD, a professor of Pathology and Laboratory Medicine, and director of the Alzheimer Disease Genetics Consortium (ADGC) at the Perelman School of Medicine at the University of Pennsylvania. "It's been known for decades that microglia — a first-line-of-defense cell we are born with — surround amyloid plaque deposits associated with Alzheimer's. These multiple gene 'hits' all originating from microglia are the clearest demonstration that these cells are part of Alzheimer's pathology and, more importantly, provide clear protein targets where we can start to intervene with drugs."

The ADGC, supported by the National Institute on Aging (NIA) at the National Institutes of Health, is one of the four consortia of the International Genomics of Alzheimer's Project on this study. The others are Cohorts for Heart and Aging in Genomic Epidemiology (CHARGE), European Alzheimer’s Disease Initiative (EADI), and Genetic and Environmental Risk in Alzheimer’s Disease (GERAD). 

The variants the team found — PLCG2, ABI3, and TREM2 — are all protein-coding mutations in genes that are highly expressed in microglia and are part of an immune cell protein network where multiple components contribute to AD risk. One of the genes, PLCG2, is an enzyme that is a potential drug target. 

Key questions remain in how microglia should be targeted and whether the injury response should be inhibited or activated and at what stage of disease. “Since prevention is a key goal of therapy, influencing microglial cells before onset of cognitive changes needs to be explored,” Schellenberg said.

The three variants they identified are fairly rare and he accounts for their success in finding them to their three-stage study. In the first stage, the entire protein coding regions of 34,290 samples were sequenced. In the second and third stages, the team further refined the sequences of variants and verified the significant hits against untested samples from AD patients. 

“Our findings show that microglia and the innate immune system -- via microglia -- directly contribute to susceptibility of late-onset Alzheimer’s disease, and are not just a down-stream ‘after-the-fact’ consequence of damage to the brain,” Schellenberg said.  

Work at Penn is funded through the NIA (UO1AG032984). Penn coauthors are Amanda Kuzma, Otto Valladares, Liming Qu, Yi Zhao, John Malamon, Beth Dombroski, Laura B. Cantwell, Adam C. Naj, Steven D. Arnold, John Q. Trojanowski, and Vivianna M. Van Deerlin. The ADGC members are Schellenberg and coauthor Li-San Wang at Penn, Richard Mayeux at Columbia, Margaret Pericak-Vance at the University of Miami, , Lindsay Farrer at Boston University, and Jonathan Haines at Case Western Reserve University. 

Full Publication

Close section


Evidence of Alzheimer's in Patients with Lewy Body Disease Tracks with Course of Dementia

Retrospective Penn Study could help guide design of trials for emerging therapies

PHILADELPHIA – Patients who had a diagnosis of Parkinson’s disease (PD) with dementia (PDD) or dementia with Lewy bodies (DLB) and had higher levels  of Alzheimer’s disease (AD) pathology in their donated post-mortem brains also had more severe symptoms of these Lewy body diseases (LBD) during their lives, compared to those whose brains had less AD pathology, according to research from the Perelman School of Medicine at the University of Pennsylvania. In particular, the degree of abnormal tau protein aggregations, indicative of AD, most strongly matched the clinical course of the LBD patients who showed evidence of dementia prior to their deaths, the team reports in The Lancet Neurology First Online, ahead of the January print edition.

The team used post-mortem brain tissue donated by 213 patients with LBD and associated dementia, which was confirmed during autopsies to have alpha-synuclein pathology. They paired the tissue analysis with the patients’ detailed medical records. This unique study combined data from eight academic memory or movement disorder centers, including the Penn Alzheimer’s Disease Core Center (ADCC) and the Udall Center for Parkinson’s Disease Research.

LBD is a family of related brain disorders made up of the clinical syndromes of PD, without or with dementia or DLB. LBD is associated with clumps of misshapened alpha-synuclein proteins. On the other hand, AD pathology is made up of clusters of the protein beta-amyloid called plaques and twisted strands of the protein tau, called tangles. Patients with LBD may have varying amounts of AD pathology, in addition to alpha-synuclein pathology.

Treatments directed at tau and amyloid-beta proteins are currently being tested in patients with Alzheimer’s disease. This study could help in selecting appropriate patients for trials of emerging therapies targeting these proteins singly or in combination with emerging therapies targeting alpha-synuclein protein in LBD. 

The study, led by David Irwin, MD, an assistant professor of Neurology at Penn and an attending cognitive neurologist in the Penn Frontotemporal Degeneration Center and the Center for Neurodegenerative Disease Research, suggests that Lewy body pathology is the primary driver of disease seen in the patients; whereas, AD pathology has an impact on the overall course of disease.

“We are excited with the results of this collaborative study that points to tau as a major correlate of dementia since therapies targeting tau in AD are advancing and they could be as relevant to AD as to LBD with co-occurring AD-like tau pathology.” Irwin said. “In addition, clinical trials for other synuclein-related brain disorders may be improved by taking into account biomarkers of AD pathology.”

“This study is important for many reasons, not the least of which it illustrates the power of research that harnesses the resources of all of these collaborating centers,” said senior author John Q. Trojanowski, MD, PhD, director of the Penn ADCC and Udall Center and a professor of Pathology and Laboratory Medicine.

None of the LBD patients had a clinical diagnosis of AD, but their post-mortem brain tissue revealed varying amounts of AD neuropathology. Post-mortem analysis of five brain regions per patient showed that they fell into one of four categories of AD pathology: 23 percent negligible or no AD, 26 percent had low-level, 21 percent intermediate, and 30 percent had high-level.

Increasing severity of AD pathology correlated with a shortened time from motor symptoms to the onset of dementia and death, with the most significant trends seen in the intermediate- and high-level AD groups compared to the low-level and no AD groups. Tau pathology, in particular, was the strongest predictor of a shorter time to dementia and death. AD pathology was also higher in patients who were older at the time of onset of motor symptoms and dementia.

“We found that patients with a higher burden of Alzheimer’s pathology also had a higher burden of alpha-synuclein pathology in their brain,” Irwin said. “From this, we inferred a potential synergism between the deleterious processes in AD and DLB.” Trojanowski added there is experimental evidence for synergies between these pathologies in animal models. 

The team also found that two relevant genetic variants in sequences of the patients’ DNA samples correlated with the amount of AD pathology. The frequency of a genetic variant in a gene coding for a protein involved in cholesterol metabolism (APOE, the most common risk factor for AD) was more frequent in patients who were in the intermediate or high AD pathology group compared to those in the low-level or no AD group. Interestingly, a variation in the gene for the protein GBA (a risk factor for LBD) was more frequent in patients without significant AD pathology. This gene is associated with LBD overall but not the subgroup with AD pathology.

In the brain, the enzyme GBA normally aids in the breakdown of worn out and misshapened proteins, such as alpha-synuclein. Together these findings suggest that genetic risk factors could influence the amount of AD pathology in LBD. Further understanding of the relationships between genetic risk factors and AD and alpha-synuclein pathology will help improve treatments for these disorders.  

In the near future, using both post-mortem brain tissue and imaging in living patients, researchers in the Udall Center and ADCC will study how declines in cognitive ability relate to AD pathology in LBD. They hope this collaborative approach will help improve the diagnosis of these co-occurring indicators as early as possible in living patients.

This study was funded by the National Institutes of Health (K23 NS088341, P50 NS053488, P30 AG010124, P50 AG005133, P30 AG028383,P30 AG008017, P50 NS062684, P50 AG005136, P50 NS062684, R01 NS48595, U01 AG006781, R01 NS065070), as well as the Veterans Affairs Geriatric Research Education and the Clinical Center at the VA Puget Sound Health Care System.

Other coauthors are Murray Grossman, Daniel Weintraub, Howard I. Hurtig, John E Duda, Sharon X. Xie, Edward B. Lee, Vivianna M. Van Deerlin, Oscar L. Lopez, Julia K. Kofler, Peter T. Nelson, Gregory A. Jicha, Randy Woltjer, Joseph F. Quinn, Jeffery Kaye, James B. Leverenz, Debby Tsuang, Katelan Longfellow, Dora Yearout, Walter Kukull, C. Dirk Keene, Thomas J. Montine, and Cyrus P. Zabetian. 

Full Penn Medicine News Release

Close section


More Human-like Model of Alzheimer's Better Mirrors Tangles in the Brain

Penn Study Describing Improved Animal Model Sheds Light on Pathways of Alzheimer's and other Tauopathies 

Tangled up brain fibrils made up of a rogue protein known as tau are the hallmark of Alzheimer’s disease (AD) that likely hold the key to treatments, making them of great interest to researchers. Mimicking the formation and spread of these tangles in animal models with greater accuracy allows scientists to better investigate new therapies to stop or slow their spread.

A new animal model developed at Penn Medicine using tau tangles isolated from the brains of Alzheimer’s patients rather than synthetic tau tangles paints a closer picture of the tau pathology in the AD brain, researchers from the Center for Neurodegenerative Disease Research(CNDR) at the Perelman School of Medicine at the University of Pennsylvania reported in the print issue of the Journal of Experimental Medicine.  Seeding normal, wildtype mice with the highly potent Alzheimer’s brain-tau (AD-tau) protein induced damaging tangles in their brains for study, mirroring a more realistic progression of tau tangles seen in AD patients’ brains.

Importantly, the mice used in the experiment were non-transgenic, meaning they did not overexpress tau protein. Past animal studies, including research from Penn Medicine, using synthetic tau fibrils provided an explanation for the progression of Alzheimer’s and other related tauopathies by implicating the cell-to-cell transmission of pathological tau. However, this phenomenon was only demonstrated in models overexpressing tau. But increased tau expression is not a cause of Alzheimer’s or other conditions involving misshapened tau protein. 

“Alzheimer’s patients don’t generally overexpress tau or harbor tau mutations, so it was important to develop a model that can recreate the pathology in a setting that more closely resembles what’s happening in patients,” said senior author Virginia M.-Y. Lee, PhD, MBA, CNDR director and a professor of Pathology and Laboratory Medicine.  “This model will open up many new directions and opportunities for not just Alzheimer’s, but also for other pathological tau disorders, such as corticobasal degeneration [CBD] and progressive supranuclear palsy [PSP], conditions which cause Parkinson-like symptoms, but CBD and PSP also are associated with cognitive impairments that mimic Alzheimer’s.”

Normal tau keeps nerve cells functioning properly, but pathological tau can cause the protein to go rogue, or misfold, triggering the formation of protein clumps known as neurofibrillary tangles, which are tightly linked to AD. It has been shown to move from cell to cell to form tangles in the brain, first in the areas that make memories, and then outward to areas associated with remembering.

Synthetic tau tangles have been used in transgenic mice to pattern this spread by Penn and others in the field, but until now, it couldn’t produce enough misshapened tau protein in normal mice to fully support this hypothesis.  

This is the first time, to the authors’ knowledge, researchers have shown abundant amounts of tau tangles convincingly induced in multiple brain regions within a few months after inoculation of AD-tau.  This observation provides the strongest support thus far for the physiological relevance of cell-to-cell transmission of pathological tau in human tauopathies, the authors said.

“This relevant mouse model will allow us to better study the architecture of tau and its physiological consequence in mechanistic and therapeutic investigations,” Lee said, “but it also provides an experimental paradigm to examine how other factors, such as amyloid plaques, another hallmark of AD, contribute to the spreading, and how tau spreads in other diseases, like CBD and PSP, among other important questions that need to be answered.”

Penn co-authors include Jing L. Guo, Sneha Narasimhan, Lakshmi Changolkar, Zhuohao He, Anna Stieber, Bin Zhang, Ronald J. Gathagan, Michiyo Iba, Jennifer D. McBride, and John Q. Trojanowski. 

This work was funded by the National Institutes of Health (AG10124 and AG17586), CurePSP, the Woods Foundation, and the BrightFocus Foundation (A2014005F).

Penn Medicine News Release

Full Journal of Experimental Medicine publication

Close section


Rhythm experience and Africana culture trial (REACT!): A culturally salient intervention to promote neurocognitive health, mood, and well-being in older African Americans

Abstract: The Rhythm Experience and Africana Culture Trial (REACT!) is a multi-site randomized controlled intervention study designed to examine the efficacy of using African Dance as a form of moderate-intensity physical activity to improve cognitive function in older African Americans. African Americans are almost two times more likely than Caucasians to experience cognitive impairment in late adulthood. This increased risk may be attributed to lower level and quality of education, lower socioeconomic status, and higher prevalence of vascular diseases, type 2 diabetes, hypertension, and obesity, all of which are recognized as risk factors for dementia. Fortunately, interventions targeting cardiovascular health (i.e., physical activity) are associated with improved neurocognitive function and a reduced risk for dementia, so African Americans may be particularly suited for interventions targeting cardiovascular health and cognitive function. Here, we describe a randomized intervention protocol for increasing physical activity in older (65–75 years) African Americans. Participants (n = 80) at two study locations will be randomized into one of two groups. The treatment group will participate in African Dance three times per week for six months and the control group will receive educational training on Africana history and culture, as well as information about health behaviors, three times per week for six months. If successful, the REACT! study may transform community interventions and serve as a platform and model for testing other populations, age groups, and health outcomes, potentially identifying novel and creative methods for reducing or eliminating health disparities.

Full publication via Contemporary Clinical Trials

Learn more

Close section


Antipsychotic Drugs Linked to Increased Mortality Among Parkinson's Disease Patients

At least half of Parkinson’s disease patients experience psychosis at some point during the course of their illness, and physicians commonly prescribe antipsychotic drugs, such as quetiapine, to treat the condition. However, a new study by researchers at the Perelman School of Medicine at the University of Pennsylvania, the University of Michigan Medical School, and the Philadelphia and Ann Arbor Veterans Affairs (VA) Medical Centers and suggests that these drugs may do significantly more harm in a subset of patients.

The researchers’ analysis of about 15,000 patient records in a VA database found that Parkinson’s patients who began using antipsychotic drugs were more than twice as likely to die during the following six months, compared to a matched set of Parkinson’s patients who did not use such drugs.

“I think that antipsychotic drugs should not be prescribed to Parkinson’s patients without careful consideration,” said first author Daniel Weintraub, MD, who is an associate professor of Psychiatry and Neurology at Penn Medicine and a fellow in Penn’s Institute on Aging. Senior author Helen C. Kales, MD, professor of Psychiatry at University of Michigan and a Research Investigator at the VA Center for Clinical Management Research added, “Treatment with antipsychotics should be reserved for those cases where the benefits exceed the risks.”

These findings are not the first to link antipsychotic drugs to increased mortality. Studies dating back to the early 2000s, including a number from Dr. Kales’ group, have found increased mortality with antipsychotic use among patients who have dementia in the general population. Since 2005 the FDA has mandated “black box” warnings on antipsychotic drug packaging, noting the apparently increased risk of death when these drugs are used in dementia patients.

Although most dementia cases are accounted for by Alzheimer’s disease, there are other forms of dementia, including one that eventually emerges in about 80 percent of Parkinson’s patients, usually many years after their Parkinson’s diagnosis. However, a study by Weintraub, Kales and colleagues in 2011 found that the FDA warnings had done little to curb antipsychotic prescriptions for Parkinson’s dementia patients.

For the new study, Weintraub, Kales and colleagues examined the possibility that antipsychotic drug use is associated with higher mortality not just in Parkinson’s dementia patients, but in all Parkinson’s disease patients. Psychosis in Parkinson’s, although it is associated with dementia and later-stage disease, can occur even in the early stages of illness and in the absence of dementia. “It happens not uncommonly earlier in the course of the illness,” Weintraub said.

The underlying causes of psychosis in Parkinson’s are not well understood, but are thought to include the spread of the neurodegenerative disease process to certain brain areas, as well as particular or higher doses of Parkinson’s drugs that enhance dopamine function.

For the study, the researchers examined records from a large Veterans Affairs database, comparing a group of 7,877 Parkinson’s patients who were prescribed antipsychotic drugs at any time during 1999-2010 to an equal-sized “control group” of Parkinson’s patients who did not use antipsychotic drugs. To reduce differences between the groups that could bias the comparison, the investigators paired each patient in the antipsychotic group with a control patient who was matched for age, gender, race, years since diagnosis, presence of dementia, and other relevant factors.

The analysis revealed that in the 180 days after they first took antipsychotic drugs, patients in the first group died in much larger numbers, compared with the matched control patients during the same periods. Overall the Parkinson’s patients who used antipsychotics had 2.35 times the mortality of the non-users.

The relative risk seemed to vary by the specific drug—for example, 2.16 times higher for quetiapine fumarate compared with non-treatment, 2.46 for risperidone, 2.79 for olanzapine, and 5.08 for haloperidol. First-generation or “typical” antipsychotics, which include haloperidol, collectively were associated with about 50 percent greater relative mortality risk, compared to more recently developed “atypical” antipsychotics such as risperidone and quetiapine.

Antipsychotic drugs have a variety of potential side-effects, including reduced alertness, increased risks of diabetes and heart disease, decreased blood pressure, and—with longer-term use—movement disorders that can resemble those seen in Parkinson’s. The initial FDA warnings were based on findings of increased strokes among antipsychotic users. But researchers still do not fully understand why these drugs are linked to higher mortality in certain patient groups. “In this study we looked at the dataset for clues,” said Weintraub, “but the most common cause of death listed was ‘Parkinson’s disease’—so there really wasn’t anything that pointed to a specific cause or mechanism.“

Weintraub, Kales and colleagues are now conducting a follow-up study that might shed more light on that mechanism. They will examine the same VA database, looking not at mortality but at “morbidity”—disease diagnoses, injuries and other new episodes of ill-health—among Parkinson’s patients taking antipsychotic drugs, comparing them with the same matched controls.

For the present, Weintraub and Kales suggest that neurologists, psychiatrists and other physicians should prescribe antipsychotics to Parkinson’s patients only after looking for other possible solutions, such as treating any co-morbid medical conditions associated with psychosis, reducing the dosage of dopamine replacement therapies, and simply managing the psychosis without antipsychotics.

Kales notes, “Clinicians really need to assess whether the psychosis is actually ‘threatening’ the health/safety of the person with Parkinson’s or those around them. If not, use other modalities including family education and behavioral and environmental modifications. ” Further, “patients should not be left on these drugs long-term without re-evaluation,” Weintraub said.

Other co-authors of the study were Jayne Wilkinson  and Eugenia Mamikonyan  of Penn and the Philadelphia VA; Claire Chiang, Hyungjin Myra Kim, and Barbara Stanislawski, the University of Michigan and the VA Ann Arbor Healthcare System; and Connie Marras of the University of Toronto.

Funding was provided by the Veterans Health Administration (IIR 12-144-2).

Penn Medicine News Release  

Full JAMA Neurology Publication  

Close section


Penn Study Identifies Enzyme Key to Link Between Age-Related Inflammation and Cancer

For the first time, researchers have shown that an enzyme key to regulating gene expression -- and also an oncogene when mutated -- is critical for the expression of numerous inflammatory compounds that have been implicated in age-related increases in cancer and tissue degeneration, according to new research from the Perelman School of Medicine at the University of Pennsylvania. Inhibitors of the enzyme are being developed as a new anti-cancer target.

Aged and damaged cells frequently undergo a form of proliferation arrest called cellular senescence. These fading cells increase in human tissues with aging and are thought to contribute to age-related increases in both cancer and inflammation. The secretion of such inflammatory compounds as cytokines, growth factors, and proteases is called the senescence-associated secretory phenotype, or SASP. 

In a study published this week in Genes & Development, genetic and pharmacological inhibition of the enzyme, called MLL1, in both human cells and mice prevents the deleterious activation of the DNA damage response, which causes SASP expression. 

“Since tumor-promoting inflammation is one of the hallmarks of cancer, these findings suggest that MLL1 inhibitors may be highly potent anti-cancer drugs through both direct epigenetic effects on proliferation-promoting genes, as well as through the inhibition of inflammation in the tumor microenvironment,” says first author Brian Capell, MD, PhD, a medical fellow in the lab of Shelley Berger, PhD, the Daniel S. Och University Professor in the Departments of Cell & Developmental Biology, Genetics , and Biology.

Berger is also the director of the Penn Epigenetics Program. Capell is an instructor and attending physician in the Department of Dermatology and is a postdoctoral fellow in the Berger lab. 

The mechanism of this inhibition is through the direct epigenetic regulation by MLL1 of critical proliferation-promoting cell cycle genes that are required for triggering the DNA damage response in the body. MLL1 is an enzyme that adds methyl groups to loosen chromatin, the proteins around which DNA winds, so that part of the genome can be “read” and translated into proteins – its epigenetic role. However, MLL1 is also commonly mutated in numerous human cancers, particularly in pediatric and adult blood cancers. 

“We show that MLL1 inhibition blocks the expression of inflammatory genes in both senescent and cancerous human cells, including those derived from human breast cancer” Capell said. 

Knowing that MLL1 has been implicated in cell-cycle regulation, when the researchers inhibited MLL1, proliferation-promoting genes were shut down and the DNA damage response and resulting inflammation was suppressed. Indeed, in the case of applying this result to fighting cancer, this is a desired effect, since an increase in inflammation can promote both the development and progression of cancer. 

“In cancer, this could be a potent one-two punch, by blocking both proliferation-promoting genes as well as the cancerous inflammation,” Capell explained. “One could imagine taking an MLL1 inhibitor as a primary treatment, but also as an adjuvant therapy to tamp down the rampant inflammation caused by drugs like chemotherapies. More speculatively, given that the SASP has been implicated in numerous other age-related disorders, it will be worth testing the effects of MLL1 inhibition in other aging and inflammatory disease models.”

The research was supported in part by the National Institute of Aging (P01AG031862), the Dermatology Foundation, the American Skin Association, and the Melanoma Research Foundation.

Penn Medicine News Release.

Publication in Genes & Development.

Close section


Adults with OCD Can Benefit from Exposure Therapy When Common Drug Treatment Options Fail, Penn Study Finds

Researchers first to test therapy next to drug treatment.

PHILADELPHIA – Patients with Obsessive-Compulsive Disorder (OCD) can improve their symptoms significantly by adding exposure and response prevention therapy to their treatment regimen when common drug treatment options have failed, according to new research from psychiatrists at the Perelman School of Medicine at the University of Pennsylvania. Exposure and response prevention therapy is a type of cognitive behavior therapy in which the patient is asked to confront triggers that give rise to their obsessions in order to refrain from performing the rituals in response to these obsessions. The study is published in the Journal of Clinical Psychiatry.

OCD is marked by the performing of “rituals” to decrease distress related to one’s obsessions—such as excessive hand-washing to cope with a fixation on hand hygiene, for example.  

“We know that exposure and response prevention therapy (EX/RP) can benefit these patients,” said lead author, Carmen McLean, PhD, an assistant professor of clinical psychology in the department of Psychiatry at the Center for the Treatment and Study of Anxiety at Penn. “But this study showed that EX/RP is also effective for OCD sufferers who do not benefit sufficiently from common drug treatments for OCD.”

A previous study compared the effects of adding risperidone, pill placebo, and up to 17 twice-weekly therapist-led sessions of EX/RP to medication for OCD. “We found compared to patients who received medication or placebo, those who received EX/RP showed significantly more reductions in OCD symptoms and depression, as well as significantly more increases in insight, quality of life, and social function after only eight weeks,” McLean said. 

The current study included 32 patients who crossed over to receive 17 weeks of EX/RP treatment after not benefitting sufficiently from risperidone. Evaluation at 12 and 16 weeks showed significant symptom improvement, with 25 (78 percent) of patients completing treatment; 17 (53 percent) of them were classified as treatment responders and 11 (34 percent) classified as excellent responders at a 32-week follow-up evaluation. The remaining patients required medication changes during the follow-up period, which enabled them to shift to excellent-responder status.  

This study adds to the large body of research that shows the benefits of exposure therapy for patients with OCD. “We want patients to know that there is another option, if common drug treatments have failed them,” explained senior author, Edna Foa, PhD, professor of Clinical Psychology in the department of Psychiatry and director of the Center for the Treatment and Study of Anxiety at Penn and the creator of exposure therapy. “The therapy can be life-saving, if patients are aware of it.” 

Additional Penn authors include Laurie J. Zandberg, PsyD; and Joseph K. Carpenter, BA. 

This research was supported by the National Institute of Mental Health (R01 MH45404) and (R01 MH045436).

Close section


Brain's Hippocampus is Essential Structure for All Aspects of Recognition Memory, Penn Medicine Researchers Find

PHILADELPHIA – The hippocampus, a brain structure known to play a role in memory and spatial navigation, is essential to one’s ability to recognize previously encountered events, objects, or people – a phenomenon known as recognition memory – according to new research from the departments of Neurosurgery and Psychology in the Perelman School of Medicine at the University of Pennsylvania and the Penn School of Arts and Sciences. Their work is published in PNAS.

Recognition memory is composed of two processes: recollection, or recognizing something along with vivid details of the initial encounter; and familiarity, a general sense of having previously encountered something. These processes often break down as a result of aging, neurodegenerative disorders (e.g. Alzheimer's disease), or traumatic brain injury, and the new findings provide a roadmap to examine strategies to improve these functions. 

“There has been a longstanding debate in the field of recognition memory about how the human hippocampus contributes to our ability to recognize,” said lead author Maxwell Merkow, MD, NeurosurgeryChief Resident at the Hospital of the University of Pennsylvania. “One segment of the scientific literature contends that neural activity in the hippocampus only contributes to recollection, whereas some believe hippocampal activity supports both recollection and familiarity. Our study aimed to get to the bottom of this.” 

The Penn team, led by Michael Kahana, PhD, director of the Computational Memory Lab, hypothesized that the hippocampus supported both recollection and familiarity, the twin processes believed to underlie recognition memory. Showing a clear link between hippocampal activity and recognition memory performance in general has previously proven elusive, having been documented in just a few earlier studies. This paper is the first to also record a link between hippocampal activity and both the processes of recollection and familiarity.

Merkow and colleagues studied 66 patients who were already undergoing intracranial monitoring of their hippocampus for epilepsy. Using these direct electrical recordings, the team was able to test the level of high frequency neuronal activity (a marker of neurons firing) in this region, a very precise measure which captures activity tied to cognition processes lasting mere hundreds of milliseconds. 

The team administered a memory task in which participants were shown and asked to remember a series of words. Patients were then tested by being shown a second series of words, some of which they had seen before, and some that were new. Patients had to determine whether or not each word had been part of the group they had learned initially. While all of this was going on, the team recorded electrical data directly from the patient's hippocampus. 

They found elevated high frequency activity during those trials in which the patient correctly identified a word they had previously seen. This was opposed to lower activity during trials where they either failed to recognize an old word or in which they saw a new word, whether or not they correctly identified it as new. 

Another major finding was that the strength of hippocampal activity predicted behavioral performance, thereby directly linking the hippocampus to recognition memory. Crucially, both the recollection and familiarity components of recognition correlated with hippocampal activity. These data show that the cognitive processes we use for recognition memory are both supported by actions within the hippocampus.

“This work directly addresses the issue of where in the brain recognition takes place,” Merkow said. “We now need to focus our efforts on how these processes occur.” The team plans to use the same high frequency recordings from smaller electrodes to answer this question. This work brings science one step closer to understanding how brain activity supports memory and potentially improving memory through future interventions. 

Additional Penn authors include John F. Burke.

This work was supported by the National Institutes of Health (MH055687)

Close section


Penn Bioethicist Calls on Researchers for More Evidence-based End-of-Life Care Programs 

End-of-Life Program Approvals Should More Closely Mirror Drug Approval Process, Author Says.

PHILADELPHIA – Although the public and private sectors are currently engaged in an unprecedented array of efforts to improve end-of-life care, too many of these programs are not evidence-based, according to a scholar from the Perelman School of Medicine at the University of Pennsylvania. Writing in the New England Journal of MedicineScott Halpern, MD, PhD, associate professor of Medicine, Epidemiology, and Medical Ethics and Health Policy, says that despite recent federal decisions that signal a renewed interest in improving end-of-life care, investigators and research sponsors must be more involved to “identify, develop and rigorously test interventions so they can offer guidance” on implementing programs that work among the terminally ill.

In his commentary, Halpern says if end-of-life care policies were approached in the same way the United States adopts new drug policies, the long-term interests of patients, health systems, insurers, and the government would be better served.  

“In July 2015, the Centers for Medicare and Medicaid Services (CMS) announced its plans to reimburse physicians for engaging their patients in advance care planning discussions,” Halpern writes. Although he notes that the decision was based on the “valid premise” that communication among all patients and clinicians is an important way to improve the quality of end-of-life care, Halpern says the problem is that “no current policy or practice designed to improve care… is backed by a fraction of the evidence” required for drug approvals in the United States. 

Halpern suggests four developments that he calls “attainable,” and describes how achieving these goals will help achieve evidenced-based end-of life-care:

  • Increased use of large randomized trials and experimental studies that help determine whether current and novel interventions improve outcomes that are important to patients and society. For example, Halpern describes the literature on advance care planning and completing advance directives as “provocative” but “insufficient” to determine whether these widely advocated practices actually improve patient care and reduce costs.
  • Better measures used in studies to quantify the effectiveness of end-of-life interventions. Technological advances that allow for processing of electronic medical record data make it easier to evaluate measures of care that matter to patients and their families. Implementation of these technologies will make it easier to develop large-scale, low-cost assessments of which interventions improve patient and family goals.
  • Development of interventions that more accurately show how patients, their families and care teams make decisions about which care plan to pursue. Novel insights into the decision-making process may bring to light options that better serve patients.
  • Health systems, insurers, and other entities must be more open to experimentation. Instead of simply being motivated to “do something,” opportunities abound for rigorously testing new initiatives, thereby benefiting the long-term interests of both the organizations and the patients they serve. By contrast, Halpern notes that implementing change absent a rigorous evaluation plan crowds out opportunities for learning.

Though encouraged by the enthusiasm of federal organization and their willingness to intervene in end-of-life care programs, Halpern says the central challenge today is to “avoid complacency regarding plausibly useful but non-evidence-based initiatives. Researchers, research sponsors, and large insurers, employers and health systems can work together to advance knowledge about what works best for whom.”

Close section


Penn Team Pinpoints Developmental Gene that Regulates Repair and Regeneration in Adult Lungs

New role for hedgehog gene offers better understanding of lung disease.

PHILADELPHIA - The whimsically named sonic hedgehog gene, best known for controlling embryonic development, also maintains the normal physiological state and repair process of an adult healthy lung, if damaged, according to new research from the Perelman School of Medicine at the University of Pennsylvania published online in Nature in advance of the print edition.

Tissues are not all created equal in their ability to regenerate.  Skin and blood cells are continually turning over, making entirely new populations of cells every few days. At the other end of the spectrum, heart and brain cells regenerate slowly, if at all, after injury. Between these two extremes are tissues such as the lung and liver, which have little cellular turnover in normal adults, but can regenerate extensively after injury. Such tissues, overall, are thought to be quiescent. 

This inactive state was previously thought to be the default mode of many tissues, including the lung, in the absence of a proliferative stimulus such as injury. However, it has remained unclear how quiescence is maintained in organs such as the lung that display a low level of cell turnover. 

“We demonstrated that quiescence in the adult lung in an animal model is an actively maintained state and is regulated by hedgehog signaling,” said senior author Ed Morrisey, PhD, the Robinette Foundation Professor of Medicine and a professor of Cell and Developmental Biology. Morrisey is also director of the Penn Center for Pulmonary Biology and scientific director of the Penn Institute for Regenerative Medicine

“We were surprised,” Morrisey recalled. “This was the exact opposite of what other researchers had suggested and pretty much the opposite of what happens during development. We scratched our heads for a long time.”
First author Tien Peng, MD and other members of the Morrisey lab used multiple approaches to determine what hedgehog was doing in the adult lung. First, they deleted the gene sonic hedgehog in airway epithelial cells of the adult lung. The protein made from sonic hedgehog is secreted from airway epithelial cells and acts on the adjoining cells surrounding the airways called mesenchymal cells. 

The team observed that after the loss of sonic hedgehog expression,mesenchymalcells began to spontaneously proliferate. This also occurred when they directly inactivated hedgehog signaling in mesenchymal cells themselves.

To determine what occurred after lung injury, the researchers performed multiple different injuries to lung tissue and found that in contrast to previous reports,hedgehogsignaling decreased. This decline correlated with the loss of the cells that normally express the sonic hedgehog gene, which were destroyed as a result of the injury. 

With this new concept of what sonic hedgehog is doing in the adult lung, the researchers then asked what would happen if they turned on the hedgehog pathway after lung injury. Consistent with their other observations, the Morrisey team found that activation of the hedgehog pathway inhibited proliferation of the mesenchymal cells surrounding the lung airways.

Overall, they found that activation of hedgehog during an injury to epithelial cells weakens replication of mesenchymal cells, whereas inactivation of hedgehog signaling prevents the restoration of quiescence after an injury. Finally, they showed that hedgehog signaling in mesenchymal cells also regulates epithelial quiescence and loss of this regulation leads to abnormal epithelial regeneration after injury. Loss of hedgehog in the adult lung leads to too many epithelial cells lining the airways after injury whereas increased hedgehog signaling blocks regeneration of the airway epithelium. Such results suggest that increased hedgehog signaling causes a breakdown of the normal regenerative properties of the lung airways, leading to degenerative disease states. 

“These results demonstrate that epithelial-mesenchymal interactions coordinated by hedgehog maintains the normal state of a healthy lung, and turning off hedgehog during injury can lead to abnormal cell repair and regeneration in the lung,” Morrisey said. “We think that mistakes in the hedgehog feedback loop could contribute to many adult lung diseases characterized by a chronic injury and regeneration cycle.” 

Indeed, the hedgehog pathway has been implicated in previous genome-wide association studies of adult lung disease. “We now have a better idea of what is going on in lung disease – in adults, hedgehog suppresses proliferation and maintains cell quiescence, as opposed to its opposite role in embryo development,” Morrisey explained. “These surprising findings suggest that researchers have to be careful in predicting the function of developmental pathways in adult organs. We need to remember that they will not always function in the same manner.” 

These studies were supported by funds from the National Heart, Lung & Blood Institute (HL110942, HL100405, HL087825, K08-HL121146), an American Heart Association Fellow-to-Faculty Transition Award, and an Actelion ENTELLIGENCE Award. 

Coauthors are Tien Peng, David B. Frank, Rachel S. Kadzik, Michael P. Morley, Komal S.Rathi, Tao Wang, Su Zhou, Lan Cheng, and Min Min Lu, all from Penn.

Close section


Higher-level Occupations May Increase Survival in Patients with a Common Form of Early-onset Dementia, Finds New Penn Medicine Research

A high level of mental activity earlier in life may buffer against disease.

PHILADELPHIA - Doctors, lawyers and other "high level" professionals may have a built-in survival edge if they're diagnosed with the disease frontotemporal dementia (FTD), according to new research from the Perelman School of Medicine at the University of Pennsylvania. Their work is published in Neurology.

FTD is a family of devastating disorders of the brain that lead to the progressive loss of brain cells (neurons) in the frontal and temporal regions of the brain, most commonly in patients between ages 50 and 65 and often causing symptoms ranging from behavioral impairments to language difficulty. Nearly 10,000 patients are diagnosed with the disease each year. As the disease progresses, it can slowly deprive an individual of their cognitive abilities, personality and eventually their independence.

"There is a notion that ones 'cognitive reserve' is built up over the course of a lifetime through experiences such as education, occupation and mental engagement," said Lauren Massimo, PhD, CRNP, a post-doctoral fellow in the department of Neurology in Penn's Frontotemporal Degeneration Center. "We believe that those with higher occupational levels are able to build up an additional level of defense against the disease through rich neural connectivity and this could contribute to longer survival."

Massimo and colleagues retrospectively examined the autopsy reports of 83 patients in the Center for Neurodegenerative Disease Research at the University of Pennsylvania, 34 of whom had confirmed FTD and 49 with autopsy-confirmed Alzheimer's disease (AD).

Each patient's occupation was recorded and ranked according to U.S. Census categories, with jobs such as factory workers and service workers in the lowest level; jobs such as tradesworkers and sales people in the next level; and professional and technical workers, such as lawyers and engineers, in the highest level. Education level was also measured in years of schooling completed.

Their analysis showed that median survival for patients with FTD was six years and nine months, and just shy of eight years for those with AD, with survival defined as the time from symptom onset until death.

Further analysis showed that patients with FTD in the highest occupation level survived an average of nine years, while people in the lower occupation group survived an average of six years, suggesting that higher occupation level is associated with longer survival in patients with FTD. Occupational level was not associated with longer survival time for patients with Alzheimer's disease. Interestingly, the team found that years of education were not associated with survival time for either group.

"These results provide support for the protective effects of occupation in FTD," Massimo said. "There may be other factors at work here such socioeconomic factors tied to occupational status that contributes to the longevity of this group. Further studies might also want to expand the sample size and occupations characterized, as ours left no room for occupations such as 'homemaker' or those outside traditional lines of work."

Other Penn researchers include Jarcy Zee, PhD; Sharon X. Xie, PhD; Corey T. McMillan, PhD; Katya Rascovsky, PhD; David I. Irwin, MD; Murray Grossman, MD, MedD.

This research was funded by the U.S. Public Health Service (F32NR014777, AG017586, AG015116, AG010124, AG043503, NS053488 and NS044266) and the Wyncote Foundation.

Close section


Additional Press Releases

Research Video

 


Funding Opportunities
Clinical Trails