Date of Degree


Document Type


Degree Name





Maria Figueiredo-Pereira

Committee Members

Patrica Rockwell

Thomas Schmidt-Glenewinkel

Jared Fine

Giovanni Manfredi

Maria Figueiredo-Pereira

Subject Categories



Alzheimer's Disease, Inflammation, in vivo AD models, prostanglanin J2, in vitro AD model, ubiquitin-proteasome system in AD


Alzheimer’s disease (AD) is a neurodegenerative disorder characterized by progressive neuronal loss that over the years spreads from the hippocampus to the neural cortex and impairs memory and cognitive functions. At the cellular level AD is linked to the presence of β-amyloid plaques and neurofibrillary tangles but despite decades of research little is known about their contribution to neurodegeneration and whether they are a cause or rather a result of the disease. It is well established that proteasome activity is impaired in AD brains and some studies suggest that this could be one of the initial factors leading to development of this disorder. Defective ubiquitin proteasome pathway (UPP) leads to accumulation of truncated or misfolded, ubiquitinated proteins (Ub-proteins) that can further form insoluble aggregates such as β-amyloid plaques and neurofibrillary tangles. The UPP involves a multitude of components and steps. It starts with the ATP-dependent E1, E2 and E3 enzymatic cascade that results in tagging targeted proteins with polyubiquitin chains. Subsequently, shuttling factors recognize and deliver tagged proteins to the proteasome for degradation. Once at the proteasome, deubiquitinating enzymes (DUBs) remove the polyubiquitin chain and still other proteasome subunits are involved in unfolding and translocating the targeted protein into the degradation chamber where it is cleaved into small peptide fragments. Several components of this pathway were shown to be dysregulated in neurodegenerative disorders. It has been recognized that enhancing UPP activity could be of therapeutic benefit as it would prevent or diminish accumulation of toxic proteins and possibly prevent neuronal death.

The major GOAL of our studies was to investigate two new potential therapeutic targets for AD related to UPP function:

a) The Usp14 deubiquitinating enzyme – we investigated whether its inhibition increases proteasome-dependent degradation and decreases toxicity induced by the endogenous product of inflammation PGJ2.

b) The product of inflammation PGJ2 that impairs different steps of the UPP – we investigated whether PGJ2 induces AD-like neuronal and behavioral pathology in vivo, by microinfusing PGJ2 into hippocampi of young and old mice. In addition, we determined the potential of PACAP to overcome the deleterious effects of PGJ2.

a) Studies with Usp14: Usp14 is a deubiquitinating enzyme that decreases proteasome-dependent degradation rates of certain substrates by removing their polyubiquitin chain before the substrate is committed to degradation. This leads to premature dissociation of the substrate from the proteasome thus escaping degradation altogether. A recent study showed that downregulating Usp14 leads to increased degradation of known proteasome substrates implicated in neurodegenerative diseases, such as Tau, TDP-43 and ataxin-3. In addition, IU1 (1-[1-(4-Fluoro-phenyl)-2,5-dimethyl-1H-pyrrol-3-yl]-2-pyrrolidin-1-yl-ethanone) is a selective Usp14 inhibitor that was shown to increase proteasome-dependent degradation rates and could serve as a potential therapeutic. These studies were conducted with the HEK (human embryonic kidney) cell line and murine embryonic fibroblasts (MEFs), and not with primary neurons.

We investigated the therapeutic potential of targeting Usp14 with IU1 in a more relevant cell model for AD, which is rat E18 cerebral cortical neuronal cultures. The results of our studies were most unexpected. We established that IU1 treatment indeed diminishes the levels of polyubiquitinated proteins but not due to enhanced proteasome-dependent degradation. Instead IU1 prevented the formation of polyubiquitin chains. IU1 blocked mitochondrial complex 1 which resulted in mitochondrial impairment and drastically lowered ATP levels. Such low levels of ATP are insufficient to activate the E1 enzyme of the ubiquitination cascade. E1 carries-out the first step that is responsible for activating ubiquitin. By assessing E1~Ub thioester formation, we confirmed that IU1-treatment lowers E1 activity resulting in low levels of Ub-proteins. Initially it may have appeared that IU1 treatment indeed sped up protein clearance by the proteasome. However, our observation that IU1-treatment lowered 26S proteasome levels with a concomitant increase in 20S proteasomes, suggested that there was not enough ATP in the cell to assemble 26S proteasomes and ensure their optimal activity.

Overall, we showed that IU1, a potential therapeutic that was intended to prevent the accumulation of ubiquinated proteins, did not increase proteasome-dependent degradation. Instead, IU1-treatment blocked the ubiquitination cascade by inhibiting the mitochondrial complex 1 and depleting ATP levels.

Our additional studies demonstrated that downregulating Usp14 with siRNA in rat cortical cultures, or Usp14 loss in cortical cultures from Usp14axJmice, did not improve the rates of Ub-protein degradation. Our results are of high importance to the development of potential therapeutics that target the UPP because of its relevance to AD.

b) Studies with the product of inflammation PGJ2: Although UPP impairment has been linked to the development of AD, it is still not known what triggers it. Studies point to chronic neuroinflammation as a major contributing factor. Traumatic brain injury (TBI) and stroke both induce neuroinflammation and increase the probability of developing AD. Chronic neuroinflammation, as opposed to its acute version, wreaks havoc in the cells, perpetuating the cycle of synthesis of toxic prostaglandins and continual activation of pathways that lead to neuronal demise. Prostaglandin J2 (PGJ2) is an endogenous product of inflammation and is released by activated microglia, astrocytes and neurons as a part of the immune response. Previous work from our lab confirmed that PGJ2 is the most toxic of the prostaglandins tested and induces AD-relevant neuropathological changes at the molecular and cellular levels. Firstly, PGJ2 activates caspase-3 and leads to caspase-3-dependent Tau cleavage at Asp421, which leads to formation of aggregation prone ΔTau, the major component of neurofibrillary tangles. Secondly, PGJ2 leads to the accumulation and aggregation of ubiquitinated proteins in plaques and tangles, which are AD hallmarks. Thirdly, PGJ2 perturbs the UPP (a) by causing 26S proteasome dissociation of the regulatory and core particles due to carbonylation of the Rpt5 subunit, thus decreasing the levels of 26S proteasomes, and (b) by inhibiting DUBs such UCHL-1. PGJ2 can exert its function via two mechanisms. The first mechanism involves covalent modification of exposed cysteine residues in proteins leading to the formation of Michael adducts. The second mechanism is mediated by receptor binding, including the DP2 and PPARγ receptors.

PGJ2 is a product of the pro-inflammatory enzymes cyclooxygenases, including cyclooxygenase-2 (COX-2), the levels of which increase in AD brains and negatively affect neuronal function. COX-2 catalyzes the synthesis of prostaglandins, some of which are neuroprotective while others are neurotoxic. Treatment with non-steroidal anti-inflammatory drugs (NSAIDs), which target cycloxygenases, is one of the therapeutic strategies aimed at minimizing neuroinflammation. However, this anti-inflammatory strategy can produce adverse effects, including renal failure, heart attack and stroke, as well as preventing the synthesis of neuroprotective prostaglandins. Therefore, identifying a therapeutic target that is downstream of cyclooxygenases would be beneficial to prevent the effects of neurotoxic prostaglandins, without interfering with the neuroprotective ones. Targeting PGJ2, shown to induce AD-like pathology in cerebral cortical neuronal cultures, could offer a better strategy to prevent inflammation-dependent neurotoxicity.

We investigated whether or not PGJ2 elicits AD-like pathology in vivo by injecting it into the CA1 hippocampal brain region in mice. Since aging is a major risk factor in sporadic AD (sAD) we treated old (52 weeks of age) and young (12 weeks of age) mice to determine if advanced age can make them more susceptible to PGJ2-linked toxicity. Our studies revealed that PGJ2 impairs learning and memory retention in old but not in young mice as assessed with the radial 8-arm maze (RAM) test. Additionally, PGJ2 induced neurodegeneration that spanned from the CA1 to the CA3 hippocampal region in old but not young mice. Lastly we investigated the levels of dendritic spines as their increased numbers are associated with learning and memory. However, the decrease in dendritic spine levels is linked to memory impairment associated with neurodegeneration. We showed that PGJ2-treatment induced dendritic spine changes in old but not in young mice. Compared to young mice, old mice displayed significantly less plastic, immature spines associated with learning, which could explain their poor performance during the RAM test. Therefore, our in vivo studies showed that PGJ2 hippocampal injections induced AD-like symptoms and pathology in the mice.

Lastly, we co-injected the group of old mice (53 weeks of age) with PGJ2 and PACAP27 to determine if some of the PGJ2-induced pathological symptoms could be prevented. PACAP is a neuroprotective peptide that acts through activation of the cAMP/PKA pathway. PACAP is abundant in the brain and its levels decrease during neurodegeneration. We subjected the mice to RAM training and found learning improvement in mice co-administered PGJ2+PACAP compared to mice treated with PGJ2 alone, although this change was not statistically significant. These results indicate that PGJ2 is indeed an effective therapeutic target, and that targeting PGJ2 to prevent/treat AD could potentially be more efficient than targeting COX-2.

In conclusion, we investigated two potential therapeutic approaches for AD: a) increasing UPP activity, and b) preventing inflammation-based toxicity by focusing on PGJ2, a product of cyclooxygenases. Our studies showed that IU1 is not an effective therapeutic for neurons due to its off-targeting effects on mitochondrial complex 1. Moreover, genetically downregulating Usp14 is not sufficient to induce significant changes in proteasome-dependent degradation rates. Perhaps combining this strategy with another neuroprotective therapeutic or peptide, such as PACAP, would improve the results. Lastly, we showed that PGJ2 by itself induces AD-like pathology in vivo that could be partially prevented by PACAP. PGJ2 could be a better and more directed therapeutic target than COX-2, to halt the deleterious effects of neuroinflammation with fewer side effects.

Included in

Biochemistry Commons



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.