Theses Doctoral

Using selective autophagy to determine protein aggregation's pathogenic contribution to neurodegenerative disease

Croce, Katherine Rose

The aberrant accumulation of aggregated proteins is a pathologic hallmark across adult-onset neurodegenerative diseases, the majority of which have no effective treatment. Although the relative importance of these structures to pathogenesis has been proposed in several diseases, there is little understanding of how we might accelerate the turnover of aggregated proteins, and in turn, a lack of consensus about whether targeting them would provide any therapeutic benefit.

The overarching goal of my dissertation is to address both of these questions by focusing on how the pathway macroautophagy might handle protein aggregates in the adult brain. Aggregation-prone proteins are preferentially degraded through the lysosome-mediated degradation pathway macroautophagy (referred to hereafter as autophagy) (Ravikumar 2002; Iwata 2005; Yamamoto 2006). Although studies suggest that aggregates are degraded in bulk by autophagy (Ravikumar 2002; Iwata 2005), studies show that they are more likely degraded in an adaptor-protein dependent manner (Lemasters, 2005; Kraft, 2008; Hanna, 2012; Isakon, 2012; Filimonenko, 2010).

In the Yamamoto lab, we have found that the adaptor, the Autophagy-linked FYVE protein (Alfy/WDFY3), is required for the degradation of detergent-insoluble aggregated proteins through selective autophagy in cell-based systems and the adult brain (Simonsen, 2004; Eenjes, 2016; Filimonenko, 2010; Fox, 2020). Through immunohistochemical and loss-of-function studies, Alfy has been implicated in the turnover of disease-relevant protein aggregates including mHtt, α-synuclein, SOD1, and TDP-43, as well as protein complexes such as the midbody ring (Filimonenko, 2010; Clausen, 2010; Han, 2014; Hocking, 2010; Isakson, 2013; Kadir, 2016).

Here, I present a potential strategy to suppress disease progression across neurodegenerative disorders by increasing the levels, and thereby the function, of Alfy. I hypothesized that genetically augmenting Alfy levels in the brain will be sufficient to alleviate aggregate burden and delay the onset of proteotoxic stress in different mouse models of neurodegeneration. Using biochemical and genetic approaches, I conducted an extensive in vivo study, demonstrating that augmenting Alfy expression levels in mice can be neuroprotective, and that Alfy may be an influential genetic modifier of neurodegenerative disease.

Using two independent genetic approaches that upregulate Alfy expression, I found that they both dramatically delay the onset of disease phenotypes in mouse models of Huntington’s disease, synucleinopathy and TDP-43 proteinopathy. First, I found that ectopic overexpression of Alfy has a pronounced, neuroprotective effect on reducing aggregation, improving motor function, and extending survival in disease models. In parallel, I used mouse genetics to verify the potency of a rare Alfy variant identified in a large Venezuelan cohort of Huntington’s disease that correlated with delayed onset in Huntington’s disease by 10-20 years.

Excitingly, in support of our hypothesis, I found that the presence of this single nucleic acid polymorphism led to an increase in steady state levels of Alfy in both humans and in mice, and it was sufficient to recapitulate the benefits of ectopic Alfy overexpression. Taken together, these studies demonstrate that increasing Alfy levels in the brain are sufficient to augment the turnover of aggregated proteins, and may be an effective therapeutic strategy that can be beneficial across neurodegenerative diseases.


  • thumnail for Croce_columbia_0054D_17370.pdf Croce_columbia_0054D_17370.pdf application/pdf 10.1 MB Download File

More About This Work

Academic Units
Pathobiology and Molecular Medicine
Thesis Advisors
Yamamoto, Ai
Ph.D., Columbia University
Published Here
July 27, 2022