ANALYSIS OF ATP-DEPENDENT CHAPERONE PROTEIN INTERACTIONS WITH MISFOLDED PROTEINS, AMYLOIDS, AND AGGREGATED PROTEINS

Protein aggregation occurs when proteins adopt non-native conformations, exposing hydrophobic surfaces due to misfolding. The exposed regions of two or more proteins associate to form amorphous deposits or highly ordered, stable fibrillar structures called amyloid aggregates. Within the cell, a robust network of proteins safeguard against protein misfolding and aggregation. This network is referred to as the proteostasis network and includes molecular chaperone proteins, co-chaperone proteins, and the ubiquitin proteasome system. Molecular chaperone proteins function in various cellular processes including intracellular transport, oligomeric assembly, and efficient protein folding. Moreover, molecular chaperones are required for folding denatured, misfolded, and de novo proteins into their native conformation. It is thought that neurodegenerative diseases may be the result of a derailed proteostasis network in response to aging, mutations, and environmental stress, among other factors that contribute to protein aggregation. The neurodegenerative diseases Alzheimer’s disease, Huntington’s disease, and Parkinson’s disease are characterized by protein misfolding and accumulation into aggregates composed of amyloid fibrils. In each of these protein misfolding diseases, the roles of chaperone proteins are complex and not well understood. Given that these diseases share a common theme of protein misfolding and aggregation, researchers have questioned whether molecular chaperone proteins are involved in disease pathology, which has led to investigations into the possible use of chaperone-based strategies as treatment options. It is thought that protein aggregation in neurodegenerative diseases results from disturbances in pathways that regulate protein quality control. This thesis investigates the role of ATP-dependent chaperone proteins in disaggregating and resolubilizing protein aggregates, including model aggregates, amyloids of Sup35 in yeast and hyperphosphorylated tau in human cells. Here, we investigated the biochemical properties of the Hsp100/Clp chaperone protein family, which couples ATP binding and hydrolysis to unfold and reactivate aggregated and misfolded polypeptides. We studied the role of molecular chaperone proteins in the progression of protein aggregation in a model of Alzheimer’s disease and developed novel chaperone tools for targeting amyloid proteins for protein clearance. Furthermore, we investigated the mechanism of substrate recognition and amyloid disassembly by Hsp104 in a yeast [PSI+] prion model of amyloid assembly. Moreover, we identified a novel function of the chaperone protein ClpX in protein disaggregation in vitro and in vivo. We observed that ClpX can bind and reactivate native and engineered protein aggregates in the absence of ATP. Lastly, in an Alzheimer’s disease model of hyperphosphorylated tau, we monitored changes in chaperone protein levels in response to inhibition of protein phosphatases. Research from this dissertation will contribute to further understanding Alzheimer’s disease pathogenesis, and, more broadly, the breakdown of protein homeostasis in neurodegenerative disease, and roles for chaperone proteins in managing proteotoxic aggregates in possible future therapies.


Introduction to Protein Conformational Disorders and Neurodegeneration
Alzheimer's disease (AD), Parkinson's disease (PD), Huntington's disease (HD), and transmissible spongiform encephalopathies such as Creutzfeldt-Jakob disease are a group of progressive, fatal neurodegenerative diseases that share the common pathological feature of protein aggregation and amyloid deposits [1]. Amyloid deposits are composed of protein aggregates that result from proteins that adopt an aberrant, non-native conformation through a misfolding event [2]. Neurodegenerative diseases that share the common feature of protein misfolding are also referred to as protein conformational disorders [3].
Neurodegenerative diseases have a devastating impact on family members, caregivers, and society. Most diseases involve processes that impact the quality of life of an individual, from classic involuntary muscle movements witnessed in both PD and HD, to cognitive decline that occurs in individuals with AD. Neurodegenerative diseases also have a major impact on the economy and caregivers. In the case of AD, in 2015, American caregivers are estimated to have contributed over 18.2 billion hours of unpaid assistance to people with Alzheimer's dementia and other dementias, which amounts to approximately $230.1 billion in unpaid care [4]. Incidence rates of these age-related neurodegenerative diseases are expected to rise as the population ages. As of 2017, the Alzheimer's Association reports that an estimated 5.5 million people in the United States have Alzheimer's disease and by 2050, this number may increase to 16 million [4].
Understanding the biochemical, molecular, and cellular pathways and mechanisms involved in Alzheimer's disease pathology and progression is important for developing more efficient disease-modifying therapeutic strategies. This chapter will discuss in vitro technologies commonly used to isolate, visualize, and characterize amyloidogenic protein aggregates and in vivo methods of identifying protein aggregates in the brains of patients with Alzheimer's disease. We will also provide examples of agents that have been shown to promote aggregation, and discuss different approaches that may inhibit or stall amyloid formation.

Protein Quality Control
Protein function is dependent on proper folding of a protein into a native, thermodynamically stable and biologically active conformation, with a three-dimensional structure. To ensure that proteins fold correctly there is a protein quality control network in cells, which consists of a group of molecular chaperone proteins and proteases.
Stress and subsequent damage to functional, folded proteins may be caused by environmental changes (for example, shifts in temperature or hydration), free radicals, heavy metals, or tissue injury. These stresses can also modify the expression levels of components of the protein quality control machinery within the cell to manage the disruption to protein homeostasis [5,6] [7]. Thermotolerance studies in cells have led to the identification of many proteins that are highly expressed during exposure to heat stress. Therefore, these proteins were given the name "heat shock proteins". In eukaryotes, the heat shock response, which includes the increased expression of heat shock proteins in response to physiological stressors, is stimulated by activation of heat shock transcription factor (HSF-1) and transcription of chaperone genes, such as genes encoding chaperone proteins Hsp70 and Hsp40 [8][9][10] Heat shock proteins are highly conserved and ubiquitous proteins that represent a component of the cellular protein quality control network. Heat shock proteins, or molecular chaperone proteins, are essential for managing protein folding and maintaining protein homeostasis in both normal conditions and conditions of physiological cell stress [11]. Misfolded proteins pose a threat to cell viability by interfering with protein-protein interactions, protein-membrane interactions, promoting aberrant protein interactions and inducing aggregation. Protein misfolding also increases the demands on the protein quality control network, which is thought to lead to further misfolding and aggregation events [12,13]. During aging, the cell's ability to mount a heat shock response in response to stress conditions decreases [14] . Expression of ATP-dependent chaperone proteins, such as Hsp90 and Hsp70, declines with age and this decrease in expression levels is even greater in Alzheimer's disease and Huntington's disease [15].
Another component of the quality control network within the cell is the 26S proteasome, a large protein degradation machine that promotes the clearance of misfolded proteins and proteins with an ubiquitin tag. In some cases, misfolded and dysfunctional proteins are tagged with multiple ubiquitin domains by ubiquitin ligases and directed to the 26S proteasome for degradation. Protein degradation may also play a role in age-associated neurodegenerative diseases, including AD and PD, where protein aggregates have been shown to contain high levels of ubiquitin, which may contribute to neurodegeneration and neuronal cell death [16] . Research suggests that proteasome activity declines and may be impaired in the aging brain [17,18]. Together, the molecular chaperones and the 26S proteasome maintain steady-state levels of healthy proteins within the cell, creating a balance of active and degraded proteins, which becomes more fragile as the proteostasis network weakens or is overburdened.

Protein aggregation and amyloids
When proteostasis is disrupted due to protein folding inefficiency or instability, proteins become insoluble and aggregate. Several neurodegenerative diseases contain a highly specialized and thermodynamically stable aggregate, called an amyloid.
Amyloid deposits are composed of bundled fibrillar aggregates rich in stacked betasheets that are oriented perpendicular to the long axis of the amyloid fiber [19,20]. A network of hydrogen bonding interactions within and between beta sheets stabilize the highly ordered beta-rich structure of amyloids [21]. Amyloid fibrils have common betarich structures, fibril morphologies, and chemical properties. Generally, the mature amyloid fibril is approximately 100 Angstroms in diameter, twisted, and unbranched [22].
Amyloids are insoluble and resistant to detergents and many other chemicals known to disrupt protein folding and proteolysis [23,24]. A unifying feature of amyloids is their distinctive cross-beta pattern observed in X-ray diffraction studies. The amyloid fibril structure is characterized by stacked beta-sheets that have an approximate intersheet distance of 11 Angtstroms and interstrand distances of 4.8 Angstroms [22] .
It is thought that amyloids arise from a nucleation-dependent polymerization event called fibrillization ( Figure 1). By contrast, intermediates in the amyloid assembly pathway, such as pre-amyloid oligomers, are soluble in solution and have diverse sets of conformational states [25]. How proteins progress through conformational stages of the amyloid assembly pathway(s) from oligomers to protofibrils and finally, to amyloid fibrils, is not well understood, but novel methods using molecular probes and optical spectroscopy may provide important information about these intermediate oligomeric species [26]. Although the precise molecular events of amyloid assembly are not known, in vitro experiments suggest that small oligomers assemble and polymerize to form an intermediate protofilament, which further aggregates to become a mature fibril [25]. The fibrillar aggregates accumulate into a deposit called an amyloid plaque, which is thought to be irreversible. In vivo preformed amyloids can act as "seeds" to promote mature amyloid formation in mouse models of amyloid assembly [27]. Although it is unknown how amyloids propagate, a prion-like mechanism has been suggested, where a misfolded protein (i.e., an amyloid) induces conformational changes in folded protein by templated conformational conversion [28] . In this model, the amyloid acts as a seed or template to spread the beta-rich structure to other proteins, promoting fibrillization and thus seeding aggregation.
The pathogenesis of several neurodegenerative diseases is linked to the formation of amyloids that accumulate within specific tissue regions. In AD, amyloid beta plaques are formed from the misprocessing and overproduction of the amyloid precursor protein (APP), and plaques are typically found extracellular to neurons in the frontal cortex and hippocampus of patients ( Figure 1B). A second hallmark of AD pathology is known as the tau tangle. Tau tangles, or neurofibrillary tangles (NFTs) are intracellular aggregates formed by the polymerization of paired helical filaments ( Figure 1B) [29][30][31] [32]. In PD, misfolded, aggregated alpha-synuclein (α-synuclein) has been shown to be the major constituent of both Lewy bodies and Lewy neurite pathology in the brains of patients with PD patients [33]. HD is predominantly genetic, and involves duplications in the Huntingtin gene (Htt). Individuals with htt mutations develop aggregates of Htt in amyloid-containing inclusion bodies, which are typically found in the striatum, cortex, and the spinal cord [34]. Although these neurodegenerative diseases share the common feature of amyloids deposits, they also exhibit differences. The neuronal cell populations that are lost in each of the diseases varies, the disease symptoms are heterogeneous, and the pathological proteins that misfold into amyloid in each disease share few similarities in sequence and structure. In addition, amyloid load is poorly correlated to cognitive decline in neurodegenerative diseases and accumulating research points to prefibrillar oligomers as the toxic species responsible for neurodegeneration and behavioral symptoms [35].
Besides age-related neurodegenerative diseases, transmissible spongiform encephalopathies, also referred to as prion disorders, provide further examples of pathological alterations in protein conformation that are associated with neurodegeneration and lead to death in humans [1]. In these disorders, a soluble prion protein (PrP) adopts a new conformation, or protein fold. This new conformer of the prion protein is an amyloid and it has self-templating and infectious properties. In the prion conformation, the protein can then bind to another native protein and cause it to adopt the aberrant, alternative conformation. In this way, the abnormal conformer can selftemplate and propagate the disease-associated protein conformation [36]. Diseases that fall within the TSE category include scrapie in sheep, bovine spongiform encephalopathy ("mad cow" disease), chronic wasting disease in deer and elk, and Creutzfeldt-Jakob disease in humans [37] [38].

Monitoring the Kinetics of Amyloid Formation in vitro and Quantifying Aggregation
It is hypothesized that the mechanism of amyloid assembly is a nucleationdependent fibrillization characterized by a rate-limiting, slow lag phase required to form a critical nucleus followed by the polymerization of fibrils at an exponential rate [39,40]. For a nucleus to form (termed "nucleation"), a set of high-energy oligomeric intermediates must be stabilized by a molecular event, such as a protein conformational change. The factors that drive nucleation and the structures of the high-energy conformational states that compose the nucleus are not well understood [41]. Although amyloids are stable structures in general, the fibrils can break into smaller fragments that may seed additional amyloids. Multiple detection and quantification methods for amyloids rely on the optical qualities of amyloids and the kinetics of amyloidogenesis. Congo Red and Thioflavin T (ThT) are small molecules that bind selectively to amyloids, inducing changes in optical features such as fluorescence quantum efficiency [26]. Light scattering or fluorescence assays are used to monitor the kinetics of amyloid fibril assembly over time [38]. Traditional methods for screening amyloid aggregates include biophysical, optical, and microscopic techniques that visualize small molecule binding to amyloids, such as Congo red (CR) and ThioflavinT (ThT) dye binding [42]. The most commonly used techniques for reporting and quantifying amyloids, in addition to novel methods for identifying different conformational states of amyloids, are explained below.

Detection of Amyloids by Molecular Probes and Fluorescence Spectroscopy
Fluorescence spectroscopy is used to identify and characterize protein conformations of amyloids in vitro under both steady state and kinetic conditions, allowing researchers to investigate amyloid assembly and visualize amyloid fibrils. The most commonly used fluorescent dyes that bind selectively to amyloid are Congo Red [43] and ThioflavinT [44]. ThT is a benzothiazole-based fluorescent dye that specifically binds to the surface of cross-β sheets of amyloid fibrils [45]. Vassar and Culling demonstrated that ThT showed highly specific binding to amyloid deposits in tissue sections and this binding was associated with an enhanced fluorescence signal from ThT [23]. When ThT binds amyloid fibril in vitro or in situ, it shows enhanced fluorescence that can be monitored by fluorescent spectroscopy [46]. In vitro binding of amyloid fibrils to ThT was first described by Naiki and coworkers [44]. When ThT was added to samples containing up to 2.0 µg/mL of amyloid fibrils, researchers observed an intense fluorescence at an excitation maximum of 450 nm and an emission maximum of 482 nm [44]. The molecular probe ThT has also been used to monitor amyloid fibril assembly in real-time [47,48].
There are limitations and caveats to quantifying amyloid fibrils based on ThT fluorescence assays. The selectivity of ThT binding to amyloid fibrils is unclear. ThT fluorescence emission can be affected by different amyloid fibril morphologies and pH; at a basic pH, the ThT molecule becomes hydroxylated [22]. ThT can also bind to other fibrils besides amyloids, such as keratin and elastin [49] . Moreover, research shows that ThT can bind to nucleic acids of DNA and this binding is associated with enhanced ThT fluorescence; [50] however, reducing the pH lowers the affinity of ThT for nucleic acids [51]. The presence of polyphenols such as curcumin, quercetin, and resveratrol in samples containing fibrillar amyloid-beta interferes with ThT fluorescence readings in both in situ ThT assays and single time-point ThT assays [52]. Considering that factors such as ionic strength, pH, fibril morphology, and concentration of both ThT and protein fibril can modify the fluorescence emission of ThT, it may be more useful to combine other methods of measuring amyloid fibril kinetics with ThT fluorescence assays [53].
Congo Red is a crimson-hued, lipophilic diazo dye that selectively binds with high specificity to amyloid fibrils in vitro, in situ, and ex vivo [46]. . The Congo Red birefringence assay can be used to identify amyloid fibrils. Using a polarized light microscope, the presence of amyloid can be investigated by determining if the sample stained with Congo Red shows apple-green birefringence under crossed polarizers; if no birefringence is observed, then it can be concluded that no amyloid is present [46]. The characteristic apple-green birefringence can be difficult to detect and other problems with this technique include high levels of background staining and challenges to reproducibility [53,46]. Research supports that Congo red does not bind to different amyloid fibrillar deposits by the same mechanism and may have relatively low sensitivity [54][55][56] A spectroscopic assay that examined Congo Red dye binding to fibrillar betapleated sheet insulin, a representative amyloid, showed that the absorbance spectrum of Congo Red changes upon binding to amyloid [43]. Klunk and coworkers further developed and extended the Congo Red-insulin fibril spectroscopic assay to quantify amyloid fibrils by examining amyloid-beta aggregation with Congo Red. Briefly, the absorbance of amyloid samples stained Congo Red and amyloid samples alone must be measured at the wavelengths 403 nm and 541 nm before performing necessary calculations to quantify the aggregated amyloid beta concentration in the samples [57].
When Congo Red binds to amyloid fibrils, it exhibits a red shift in absorbance maxima from 490 nm to 540 nm [58]. Congo red dye binding to amyloid fibrils causes Congo red to change in color from orange-red to rose and because of this, a spectral shift in the absorbance spectrum of Congo Red can be measured using a UV-Vis spectrophotometer [59].
Congo Red and Thioflavin T dyes do not bind to pre-amyloid conformations in the amyloid assembly pathway. The development and use of fluorescent dyes 1anilinonapthalene-8-sulfonate (ANS), 4-4-bis-1-phenylamine-8-napthalene sulfonate (Bis_ANS), and 4-(dicyanovinyl)-julolidine (DCVJ) allows for identification and monitoring of early amyloid protein conformations [60]. Bis-ANS has been shown to bind to prefibrillar protein conformations associated with the early stages of the amyloid assembly pathway. When Bis-ANS binds to the hydrophobic residues on the surface of proteins and is excited at 360 nm, it emits a 530 nm wavelength [48]. DCVJ was shown to bind selectively to prefibrillar oligomers, making it a useful tool for studying the early stages of amyloid assembly [60].
Moreover, A drawback of Congo Red and Thioflavin T dyes is that they cannot be used in vivo. The development of BoDipy-Oligomer (BD-Oligo) by the research group of Teoh et al. may have bridged this gap in the field. BD-Oligo is a fluorescent probe selective for oligomeric intermediates in amyloid formation [61]. BoDipy-Oligomer can cross the blood brain barrier and can label disease-associated oligomers in real-time without using radiation, making it an attractive candidate for clinical screens of presymptomatic patients of AD [61] [62]. A novel stain that has recently begun to be employed for the detection of Aβ plaques is Amylo-Glo. While this has recently been shown to bind to Aβ plaques in 11 month APP/PS1 transgenic mice, the specificity of the marker for Aβ as opposed to other aggregated proteins has not been predetermined [63]. Unlike Congo Red and Thioflavin T, Amylo-glo has a blue UV excitable stain, making it ideal for co-labeling and staining studies, it is also much brighter than these traditional dyes and appears more sensitive [64].

Protein Misfolding Cyclic Amplification
Protein Misfolding Cyclic Amplification (PMCA) is a technique that was developed in 2001 for the detection and amplification of prion proteins [65]. The first phase of PMCA involves utilizing a trace amount of infectious prion purified from either brain homogenates or cell lysates to induce the conversion of the normal protein into the misfolded infectious aggregate. In the second phase, the sample is sonicated, which fragments the aggregate and therefore frees up the template to polymerize again.
Therefore, by PMCA the fibril concentration increases exponentially [66]. A concern of the PMCA reaction and technique is its ability to detect low levels of prion protein in a sample, as well as the efficiency of the reaction. PMCA may vary depending on experimental conditions, such as the reaction components and sonication. A modification of the experimental design to include Teflon beads (PMCAb) increased the efficiency of the PMCA reaction and the sensitivity of prion detection [67].
Quantitative 'qPMCA' was developed to quantify the prion concentration in a sample since the amount of cycles of PMCA needed to detect the prion is directly related to the amount present in the original sample [68,69]. Briefly, the procedure involves the purification of prion from the original lysate, followed by PMCA reactions containing multiple dilutions of prion. Then the number of rounds of PMCA required to produce a detectable signal are determined by Western blot analysis. From this, a curve is generated of prion concentration versus PMCA detection round. Extrapolating from this curve would provide the original prion levels in the sample [68,69]. Although developed for prions associated with TSEs, PCMA has recently been applied to the detection of Aβ oligomers in cerebrospinal fluid of patients with AD [70].

Quaking Induced Conversion Assay and the Amyloid Seeding Assay (ASA)
The amyloid seeding assay (ASA) is a method in which rapid amyloid formation or seeding occurs when amyloid fibrils are added to a fresh pool of soluble proteins. The ASA is useful tool to screen for agents that may be aggregation inhibitors, and for detecting prions in biological samples. The formation of these aggregates can be observed by monitoring Thioflavin T (ThT) binding to amyloid fibers over time [71].
Similar to PMCA, the Quaking Induced Conversion Assay (QuIC) was developed to address technical difficulties associated with PMCA. In QuIC automated orbital shaking is utilized instead of sonication, which is used in PMCA. Both the ASA and QuIC assays are limited in their ability to accurately quantify protein aggregates and are considered less reliable than the new real-time quaking induced conversion assay (RT-QuICK) [72]. RT-QuICK utilizes some components of the QuIC assay, such as intermittent shaking but also incorporates ThT to monitor the formation of fibrils, as seen in ASA's [73].Furthermore, prion and fibril concentration is accurately determined by using RT-QuICK if it is coupled with end-point dilution analysis. In this method, homogenates are serially diluted. Then the samples are analyzed by RT-Quick for each dilution, and then ThT fluorescence levels are measured to establish a baseline. The advantage of utilizing RT-QuICK is that it is a rapid, sensitive, cost-effective assay as compared to alternative techniques [72].

Monitoring Amyloids in vivo
The ability to develop in vivo detection and monitoring methods for amyloid assembly remains difficult for most of the amyloidogenic diseases, especially in HD and PD. In HD, researchers have had success utilizing positron emission topography (PET) to monitor disease progression in individuals diagnosed with the disease relative to healthy individuals. However, in both PD and HD there is still no tracer that can both bind to aggregates and be simultaneously monitored by PET scans. In contrast, there have been more successful developments of radioligands selective for beta-amyloid and paired helical filament tau protein that enable in vivo monitoring and visualization of ADrelated amyloid plaque and neurofibrillary tau protein aggregate accumulation.
The PET imaging agents that are selective for fibrillar amyloid-beta plaques include 11-C-labeled Pittsburgh compound B and florbetapir-F18 [74]. The first PET imaging probe developed that could bind to amyloid beta aggregates was Pittsburgh Compound B (PiB-C11) [75]. PiB-C11 effectively binds to amyloid plaques in brain regions of people with Alzheimer's disease that correlate with the same brain regions showing amyloid accumulation from post-mortem histopathological analysis of brain tissue [76]. However, PiB-C11 has a short half-life of approximately 20 minutes, which limits its efficiency [77]. The first FDA-approved PET tracer for clinical imaging of amyloid-beta deposits is florbetapir-F18 (Amyvid TM ) [76]. The PET ligand florbetapir-F18 ( 18 F-AV-45) binds to amyloid beta plaques with similar specificity and regional uptake as PiB-C11 and has a half-life of approximately 110 minutes [77]. Recent studies have shown that florbetapir-F18 is both sensitive and specific to the detection of Aβ and can be used to clinically distinguish individuals [78]. Moreover, in vitro analysis of the of florbetapir-F18 binding in human postmortem brain tissue sections confirmed that the labeling intensity of florbetapir-F18 correlates with amyloid plaque density and the PET tracer did not bind to neurofibrillary tangles [74].
Research findings from postmortem brain analyses show that neurofibrillary tangles correlate with cognitive decline and neuronal death whereas amyloid plaques do not strongly correlate with dementia severity [79,80]. Therefore, there development of PET tracers that bind to paired helical filaments, which are composed of pathologically hyperphosphorylated tau, could be used to diagnose and monitor treatment outcomes in vivo [81]. The PET ligand 18 F-AV-1451 binds selectively and with high affinity to hyperphosphorylated tau in paired helical filaments [82,83]. A study that investigated 18 F-AV-1451 binding in healthy elderly people found that uptake and retention of AV-1451 correlated with age and amyloid plaque accumulation [84]. AV-451 uptake in the medial temporal lobe was also found to be significantly associated with cognitive decline and impaired episodic memory [84]. The development of amyloid beta and tau PET-imaging probes provide a new tool for measuring pathological amyloid beta and neurofibrillary tangle deposits in vivo, which could be useful in the clinic for diagnosis of Alzheimer's disease.

Cytotoxicity assays
Amyloid assembly intermediates are widely considered to be more toxic species in the amyloidogenesis pathway, rather than the mature amyloid fiber present in plaques, yet there is still ambiguity about whether protofibrils, oligomer species, or a combination of both causes toxicity. Research shows that oligomers from the selfassembly of amyloid beta peptide are 10-fold more toxic to neuronal cell cultures than fibrillar amyloid-beta  peptide. Amyloid-beta peptide  is toxic to cells at micromolar concentrations whereas oligomeric species are toxic at nanomolar concentrations [85]. Reliable cytotoxicity assays provide important information about cellular health and viability following exposure to a toxic substance and are useful for investigating the correlation between protein aggregation and cytotoxicity, which may provide information important for understanding the pathogenesis of protein conformational diseases.
In vitro cytotoxicity assays are used to determine the viability of a cell following exposure to a toxic substance, such as a misfolded protein conformer or mature amyloid. The in vitro assays rely on the markers of cell death, whether that is the compromised integrity of the cell membrane, the release of enzymes into the extracellular space that are normally compartmentalized, etc., as direct and indirect measurements of cell viability. However, assays to monitor long-term cytotoxicity are lacking. Colorimetric in vitro assays are widely used pharmacology and ecotoxicology studies to determine cell death following exposure to a toxicant or a suspected toxic species [86]. Often several cytotoxicity assays are employed to accurately determine cell fitness and survival. Cytotoxicity assays have also shown the correlation that as the order of oligomer species increases, the degree of toxicity of the oligomer also increases providing valuable information about the identity of most toxic soluble oligomer species, which has implications for drug target research [87].
Below are examples of commonly used and robust assays for measuring the number of living cells (cell viability) and determining the number of dead cells (cytotoxicity). The LDH assay and MTT assay are spectrophotometry-based assays for measuring cytotoxicity in vitro. Despite the ease of use of these inexpensive assays and their ability to be used in high-throughput screening, there are several limitations associated with their use. Notably, spectrophometric assays that indirectly measure plasma membrane breakdown do not distinguish among different cell death modalities [88]. Vital dye assays represent another method for measuring cell viability in response to cytotoxic agents in vitro. However, vital dyes are also limited by their inability to distinguish different modes of cell death [89].

LDH Leakage Assay for Compromised Membrane Integrity
Lactate dehydrogenase (LDH) is a cytosolic enzyme that can be used as a biomarker of cell death and cell lysis [90]. The lactate dehydrogenase assay is an example of an intracellular protein release assay. After cell death, the cell membrane loses its structural integrity, which results in the release of normally intracellular enzymes into the cell culture medium [91]. Colorimetric and fluorometric detection of LDH using commercially available kits allows researchers to monitor cytotoxicity in vitro [88] . LDH is catalyzes the oxidization of lactate to pyruvate and generates NADH; NADH strongly absorbs 340 nm light. Monitoring absorbance values using a 96-well plate provides an estimation of cytotoxicity [90]. The LDH assay is widely used to investigate cytotoxic affects and was recently applied to detect cytotoxicity of amyloids in rat embryonic neuronal cell culture [92,93]. However, disadvantages of the LDH assay include low sensitivity and a high degree of variability [86]. Moreover, LDH enzymatic activity can be affected by cell culture medium, such as pH and components within the medium, leading to an underestimation of cell death [88] .

MTT Assay
The 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) reduction inhibition assay, commonly referred to as the MTT assay, is a robust colorimetric assay for assessing cell viability in a variety of cultured cell types [90]. The MTT assay of cell viability was first described by Mosmann [94]. MTT is a water-soluble tetrazolium salt that converts toformazan, which is purple and water insoluble, by reduction of the tetrazolium ring mediated by mitochondrial dehydrogenase enzymatic activity [95]. Only viable, metabolically active cells will convert MTT to formazan. Thus, formazen will accumulate as as a precipitate in viable, healthy cells. Spectrophotometric quantification of relative levels of formazan allows researchers to measure MTT conversion and overall cell viability [90]. The MTT assay is versatile and linear over a broad range of densities. However, the MTT assay may be more useful in cells with high metabolic activity [86].

Vital Dyes
Besides the LDH and MTT assay, vital dyes are frequently used by researchers to distinguish between living and dead cells in vitro. Vital dyes are fluorescent or colored molecules that can be used in membrane permeability assays to measure cell death [89] . Exclusion dyes, such as Trypan blue, cannot cross plasma membranes of healthy cells, and can be used to measure cell viability. Other commonly used exclusion dyes include propidium iodide and 4',6-diamidino-2-phenylindole, which label dead cells that no longer have intact plasma membranes [96]. Researchers determine the percentage of viable cells in response to a cytotoxic agent by staining the cells with an exclusion dye and counting the number of viable cells using a light microscope and a cell counting instrument, such as a hemocytometer. The percentage of viable cells is the ratio of viable cells to the total number of cells [89] .
Dyes can also be used to label living cells, such as calcein acetoxymethyl ester (calcein-AM). The vital dye calcein-AM has been used effectively in cytotoxicity assays [97,98]). Calcein-AM is lipid soluble and easily crosses cell membranes. Inside the cell, is hydrolyzed by cytosolic esterases to yield calcein, a fluorescent and membrane impermeable product [89] . Cells with intact plasma membranes will retain fluorescent calcein. Dying or dead cells with damaged plasma membranes release calcein, which has a strong green fluorescence signal at 530 nm [99]. Therefore, cytotoxicity can be determined by measuring the release of calcein from lysed cells using a fluorimeter. A drawback of calcein-AM is that esterase enzymatic activity may be affected by events not related to cell death, which may lead the researcher to underestimate levels of cytotoxicity [89].

Pharmacological Agents and Drug Therapies for Protein Misfolding Diseases
Aggregation can be promoted by intrinsic and extrinsic factors, including the amino acid sequence of a polypeptide, pH, temperature, chaperone protein levels, and protein concentration [38]. Aggregation can be inhibited in vitro by different classes of compounds, notably anthracycline, anionic sulphonates, Congo red, and rifampicin, which prevent the formation of oligomers, as well as by other proteins. Prefibrillar oligomers are the most toxic protein conformation, not the mature amyloid fibril [100], therefore many pharmacological studies have explored different avenues for abrogating protein aggregation and its associated toxicity. Strategies explored so far include increasing the activity of the protein quality control network, stabilizing the soluble disease-associated protein, destabilizing the insoluble misfolded protein, preventing aggregation at the early stages of amyloid formation, and changing the protein folding landscape to favor native protein folding.

Metals
It has been shown that metals may have a role in inducing protein aggregation.
Specifically, divalent Cd 2+, but not Mn 2+ impairs proteosomal activity in mouse neuronal cells expressing prion protein. This impaired proteasome system lead to an accumulation of high molecular weight ubiquitinated proteins in the cell, and ultimately cell toxicity and death [101]. These observations were also replicated in yeast [102].
Arsenite, As (III) was also found to interfere with chaperone activity and with the folding of peptides in yeast. As (III) exposure was shown to induce the formation of over 140 protein aggregates, which can further induce misfolding and aggregation of other proteins [102]. A number of other metals have also been identified as promoting protein aggregation, and inhibiting of protein refolding both in vivo and in vitro, including Hg 2+ and Pb 2+ [103,104].

Therapeutic Intervention Methods Targeting Aggregates
A number of therapeutic strategies have been implemented as a means to target amyloid and aggregate formation due to the inherent toxicity of both the fibrils and aggregates [105,106]. Some techniques include inhibiting the formation of the β-sheet secondary structure by using short peptides, or small inhibitory molecules. Others include bolstering the natural defense towards aggregates, such as heat shock proteins, or agents that induce chaperone activity, to enhance clearance mechanisms.

β-sheet Breakers
β-sheet breakers are short peptides that recognize and inhibit the amyloid conformation. Specifically, this is commonly done using short peptides, that recognize the hydrophobic or "central core" of the beta amyloid (Aβ) peptide [100]. This class of peptides can bind to oligomers and fibrils to destabilize the β-sheet structure that is a common trait of amyloids. The unique property of these peptides is their specificity for abnormally folded proteins; they destabilize only non-native protein conformers without affecting the soluble protein conformer [100]. Thus, this class of drugs can deter amyloid fibrillogenesis and downstream aggregation [107].Two β-sheet breaker peptides, KLVFF and LPFFD , have been shown to prevent the oligomerization of oligomers into amyloid fibrils, thus preventing the formation of amyloid plaques [108,109]. Research shows that the LPFFD peptide can better halt amyloid fibrillogenesis than KLVFF due to its lower hydrophobicity [110].

Antibodies
Several monoclonal antibodies have been developed that specifically target amyloid-beta monomers and oligomers, as well as some polyglutamine aggregates [105]. Two antibodies have been specifically examined for their role in mitigating Aβ pathology: bapineuzumab and solanezumab. Bapineuzumab targets the N-terminus of amyloid-beta whereas solanezumab binds to the central region. Both monoclonal antibodies have been tested in clinical trials in patients diagnosed with mild to moderate AD. However, despite success in the laboratory, monoclonal antibodies have not yet shown significant clinical changes or altered disease endpoints in patients [111].
A conformation specific peptide, which mimics a pathological conformation of amyloids, was developed using polymerized British amyloidosis (pABri) peptide.
Although the peptide does not have sequence homology to amyloid-beta or any other human protein, pABri is synthesized in the beta-sheet form and elicits an immune response against the pABri peptide, as well as other amyloids. This technique been used successfully in APP transgenic mice to direct the immune system against amyloid formation, which led to a reduction in amyloid plaque burden and improved cognition [112].

Natural Products as Putative Therapeutic Agents
Pharmacological studies have identified many diverse compounds associated with increasing molecular chaperone expression, restoring proper protein folding, decreasing the levels of pathologic misfolded proteins and inhibiting amyloid formation.
For example, polyphenols have strong anti-amyloidogenic activity and act by promoting clearance of amyloid from the system. Most polyphenols have been shown to disrupt aggregation of α-synuclein [106]. A polyphenol that is commonly found in green tea, (−)epigallocatechin-gallate has been shown to inhibit Huntington aggregation [113]. Other polyphenols such as curcumin have also been shown to aid in reducing protein aggregation through the induction of heat shock proteins [114]. Other examples include natural products that may interrupt or modulate the processing of amyloid in the brain, specifically pomegranate extract has been shown to disrupt Aβ in transgenic mice favoring the Aβ40 species over Aβ42 [115].
Drugs that inhibit amyloid fibrillization are also under investigation as potential therapies for protein conformational disorders. For example, cyclohexanehexol stereoisomers exhibit anti Aβ aggregation activity and are naturally occurring compounds with high absorption levels and no reported adverse side effects [100].
Additional research is focused on applications of chemical chaperones, ligands that bind to native proteins and osmolytes to stabilize the native protein state and destabilize the non-native, aberrantly folded state [38].

Molecular Chaperones
Molecular chaperone proteins, also referred to as heat shock proteins, are activated during cellular stress, and recognize and bind to surfaces of non-native or misfolded proteins to prevent or alleviate protein aggregation [116]. Biochemical assays have shown that several classes of molecular chaperones can modulate amyloid levels and inhibit amyloid fibrillization. Hsp70 and the co-chaperone Hsp40 were reported to block the self-assembly of polyglutamine [117]. Therefore, heat shock proteins represent a potential therapeutic avenue for future development of drugs to alleviate amyloid burden.
Another class of possible therapeutic agents that may be effective against neurodegeneration includes small molecules that target the heat shock response or selectively alter the activity of chaperone proteins. Hsp90 has emerged as a possible target for neurodegeneration and cancer therapy because of its role as a robust protein chaperone that assists in stabilizing and activating several disease-related proteins.  [124]. Yeast protein Hsp104 works with Hsp70/40 system to disassemble and rescue aggregated proteins has also been shown to disassemble amyloid fibers in vitro [125] . Overexpression of Hsp70 in a Drosophila model of Huntington's Disease mitigated neurodegeneration and increased the lifespan of the fly [126].

Chemical chaperones
Chemical chaperones are a unique class of drugs that can alter the balance of folded and misfolded proteins within the cell. Chemical chaperones are low molecular mass and can stabilize proteins against misfolding and aggregation, caused by thermal and chemical denaturing events. Some promising groups of compounds that target prion diseases include the anthracyclines, porphyrins, and diazo dyes. Chemical chaperones can alter protein conformation by blocking the conversion of soluble, biologically active prion protein to the insoluble, transmissible prion conformer [127,128]. The clinically approved anti-malarial compound quinacrine, binds specifically to the C-terminal helix of the normal isoform of the prion protein and stabilizes the normal protein conformation [129]. By stabilizing the normal, cellular non-prion protein, whose function is not known, conversion to the misfolded, transmissible prion isoform is blocked. However, quinacrine cannot pass through the blood-brain barrier so it has limited therapeutic applications in animal models [49]. Therapeutic research for prion diseases now includes an array of compounds such as the polycationic compounds (i.e., dendritic polyamines), polysulphate polyanions/glycosaminoglycans, tetracyclic compounds, and tetrapyrrolic compounds that have anti-prion activities [130].

Conclusion
Much progress has been made in characterizing and understanding the structure and assembly of amyloid. The most toxic species in amyloid disorders are widely considered to be the oligomers preceding the formation of mature amyloid fibrils and plaque deposits. In addition, the conformational conversion process itself may cause downstream events that contribute to neurodegeneration. There are many pressing questions left to address in the field of amyloid proteins and neurodegenerative disease.
It has been challenging to determine the structural identity and associated toxicity of various oligomers. However, the heterogeneous nature of oligomers and amyloids could  This study provides novel insight into chaperone-mediated recognition and disassembly of amyloids, which are expressed in yeast as prions and are pervasive in human neurodegenerative disorders, such as Alzheimer's disease, Parkinson's disease, and Huntington's disease.

Introduction
Prions are amyloidogenic protein conformations that propagate from cell to cell in either a functional or pathogenic capacity (2,3,4). Prions are non-genetic elements that self-template and are inherited in a non-Mendelian manner; reversion of the prion fold to the native fold, also referred to as prion curing, is reversible (5).
Budding yeast have been shown to contain as many as eight propagating prions (4).
One of the most extensively studied prions in yeast is Sup35, which confers the Sup35 is a translation termination protein that exists in at least two conformations: the native conformation and the prion conformation (6 (13). Mutations in the middle domain of Hsp104 also affect prion propagation in yeast and yield differential [PSI + ] phenotypes that range in color from pink to red yeast colonies (14).
Hsp104 is an ATP-dependent molecular chaperone protein that functions to remodel the Sup35 amyloid and segregate it into daughter cells during division.
Hsp104 is a ring-shaped hexameric protein that belongs to the AAA+ (ATPases associated with diverse cellular activities) ATPase protein superfamily (15,16).
Common to all AAA+ proteins are highly conserved nucleotide binding domains that contain Walker motifs (17). Several proteins in the AAA+ superfamily use mechanical force to remodel substrate proteins by hydrolyzing ATP (18). Hsp104 contains two nucleotide-binding domains per subunit that function cooperatively (19,20). The flexible pore loops of Hsp100 proteins are critical for substrate translocation and protein remodeling. Mutations targeting the pore loops of Hsp100 proteins, such as HslU and ClpX, impair substrate engagement and translocation of polypeptides through the axial channel (21,22). The mechanism for Hsp104-mediated disaggregation is thought to rely on the threading of polypeptides through an axial channel. Cryoelectron microscopy maps of Hsp104 hexamers bound to different nucleotides suggests that the two nucleotide binding domains of Hsp104 form stacked rings and a central cavity (23). Research suggests that substrates are disaggregated by Hsp104 by threading and translocation of substrate through the axial channel into the central cavity of Hsp104 due to substrate binding of the NBD1 and NBD2 rings coupled to sequential ATP hydrolysis (23). Conserved tyrosine residues (pore-loop) located in the central channel of Hsp104 bind to and grip polypeptides (24 -28). The mechanism of disaggregation by substrate threading and translocation is conserved in ClpB, the bacterial homolog of Hsp104 (26,28,29,30). ATP hydrolysis cycles cause changes in the conformation of Hsp104, thus providing the mechanical force required to unfold and translocate polypeptides through the axial pore of Hsp104 (23,26).
Hsp104 remodels heterogenous aggregates, including amorphous as well as highly ordered amyloid aggregates (31 -33). Hsp104 reactivates misfolded proteins and is necessary for prion inheritance in Saccharomyces cerevisiae (34,35). It is thought that Hsp104 fragments prions and generates seeds for prion assembly, which catalyzes the propagation of prions in yeast (36 (44,45). Hsp70 acts upstream of Hsp104 and regulates Hsp104 binding to protein aggregates (46). Hsp70 is required for speciesspecific targeting of client proteins to ClpB and Hsp104, suggesting a conserved mechanism of substrate binding and disaggregation by the Hsp70-Hsp100 chaperone system at the surface of protein aggregates (32). Hsp70 binds to the middle domain of Hsp104, which may facilitate threading of protein substrates through the Hsp104 axial channel (47,48).

Curing of [PSI + ] by Hsp104 overexpression detected by fluorescence
Although the [PSI + ] colorimetric conversion assay has been widely used to qualitatively assess amyloid abundance and prion inheritance, it does not allow for To determine if red pigment accumulation is quantifiable in cell lysates by fluorescence, we cultured [psi -] and [PSI + ] yeast on agar plates for five days until red colonies were visible for the [psi -] strain ( Figure 1A). We harvested cells from solid media, lysed the cells with glass beads and measured protein concentrations of the soluble lysate. We monitored fluorescence emission spectra of soluble cell extracts, normalized to total protein, following excitation at 488 nm. Fluorescence has also previously been used to separate mixed populations of red and white yeast cells by fluorescence-activated cell sorting (FACS) with an excitation at 488 nm (52). We observed that [psi -] cell lysates produce a large, broad emission peak at 565 nm; however, a much lower emission peak was detected for [PSI + ] cell lysates ( Figure   1B).
To determine if the amplitude of the emission peak is dependent on the protein concentration in [psi -] cell lysate, we compared emission spectra over a range of concentrations (0.05 to 0.8 mg ml -1 ) ( Figure 1C). We observed that as protein concentration increased, the amplitude of the emission peak at 565 nm also  Figure 1E). However, under the same visualization conditions, we did not observe endogenous fluorescence in [PSI + ] cells ( Figure 1E). which appeared much weaker in intensity than the fluorescence associated with S.

Saccharomyces cerevisiae
To identify which amino acids in the Hsp104 N-domain are important for amyloid recognition and curing of the [PSI + ] prion phenotype, we performed random mutagenesis and phenotypic screening to monitor S. cerevisiae cells for colony color conversion in cells expressing Hsp104 mutant proteins in the plasmid pYS104 ( Figure   3A). Using error-prone PCR amplification, we generated random mutations in hsp104 in the region spanning the N-domain (amino acid residues 1 -307). Next, the mutagenized fragments were used as primers to amplify a plasmid containing full-    Figure 4A).

Functional analyses of Hsp104 N-domain mutants in vivo
The  (55) containing mutations shows that the following mutations are near the surface: L54P, V74A, P81L, A133D, and E138V ( Figure 4B). Thus, it is possible that these residues may contact amyloid substrate and participate in substrate binding.
Yeast strains carrying mutations in hsp104 show maximal differences in fluorescence intensity at 565 nm when excited at 488 nm. One-way ANOVA of the fluorescence intensities was used to determine that the means of the fluorescence intensity for each strain at the emission wavelength 565 nm is significantly different (p < 0.0001) ( Figure 4C). Confocal microscopy was used to visualize the endogenous fluorescence of S. cerevisiae carrying the Hsp104(V69F) mutation, which is associated with loss of prion curing ( Figure 4D). We observed reduced endogenous fluorescence in this yeast strain ( Figure 4D), as expected from the conversion of colony color from red to white on SD-URA solid media.

Characterization of Hsp104 N-domain variants by thermotolerance assays
Hsp104 is required for thermotolerance in yeast. To determine if the mutations that we identified in the N-domain are impaired for thermotolerance, we carried out thermotolerance assays with different strains of yeast containing mutations in hsp104.
Yeast strains were grown to mid-logarithmic phase of growth and Hsp104 expression was induced by mild heat shock at 37°C for 30 minutes. Yeast cells were exposed to extreme heat shock at 52°C at 0, 5, and 10 minutes and serially diluted fivefold on SD-URA solid media. Heat shock at 52°C for 15 minutes and 30 minutes resulted in no colony growth on SD-URA plates (data not shown). We assessed the cell viability  Figure S1). We observed that yeast containing the mutation Hsp104(L54P) had a 1.8-fold increase in thermal aggregate accumulation compared to yeast cells overexpressing wild-type Hsp104 ( Figure 6). The mutation Hsp104(V74A) was associated with a1.8-fold increase in thermal aggregates and yeast containing Hsp104(P81L) had a 1.9-fold increase in thermal aggregate load compared to wild-type Hsp104 ( Figure 6). Therefore, the hsp104 mutations L54P, V74A, and P81L are associated with impaired reactivation of protein aggregates.

Discussion
The amino acid residues 1 -163 in the N-domain of ClpB are thought to be important for early recognition of polypeptide substrates targeted for disaggregation (56 -58 thermophilus ClpB shows that the following residues are surface-exposed and are likely involved in direct substrate recognition of aggregated proteins: T7, L91, L14, D103, and E109 ( Figure 7A).  Figure 7C). Our results suggest that surface-exposed Hsp104 residues, including A133D, bind to Sup35 amyloid. The Hsp104 residues P81L, V74A, and L54P are associated with increased thermal aggregate load after extreme heat shock, suggesting a role for these residues in protein disaggregation. Thus, these residues may be involved in the recognition of amyloid fibrils. Our results suggest that conserved residues H34 and A133 in the N- Site-directed mutagenesis of hsp104 on the vectorpYS104 (54) was carried out using the QuikChange Mutagenesis kit (Agilent Genomics). Primers were designed to introduce the following substitutions into hsp104: A99T, V74A, H34D, and E138V (Table 3). Yeast transformants were grown on synthetic dextrose plates lacking uracil and were replica plated.

Yeast Whole Cell Extraction and Western Blot
Yeast whole cell lysates were extracted under non-denaturing conditions as previously described. Yeast cells were grown on synthetic dextrose solid media lacking uracil and supplemented with 2 % galactose at 30°C. After five days of growth, yeast cells were scraped off the solid media and pelleted at 3,000

Fluorescence [PSI + ] Assay
Yeast whole cell lysates were normalized to a final protein concentration of 0.5 mg/mL. Using a quartz cuvette, 80 μL of lysate was excited at 488 nm at PMT 750 V.
Emission wavelengths were captured between 500 and 800 nm. The excitation slit width was 10 nm and the emission wavelength was 20 nm.

Thermotolerance Assay
Thermotolerance experiments were conducted with yeast strains as previously described (69,70). Briefly, yeast cells were grown in SD-URA media to midlogarithmic phase (OD600 of 0.400 -0.600) in SD-URA media.

Statistical analyses
Differences in mean fluorescence intensities were measured using an unpaired, onetailed t-test with alpha level equal to 0.05. Analysis of variance tests and post-hoc Tukey's multiple comparison tests were conducted to determine differences in fluorescence intensities across more than two yeast strains. Statistical analyses were carried out using Graphpad Prism version 6.0, GraphPad Software, La Jolla, California, USA (www.graphpad.com).

Pairwise protein alignment
Protein sequences were globally aligned using the EMBOSS Needle tool (66) from the European Bioinformatics Institute (EMBL-EBI) (65). The default settings in EMBOSS Needle were used for protein alignment.

Disclosure of Potential Conflicts of Interest
The authors do not have any conflict of interest or financial disclosures.      In this study, we use engineered and native substrates to investigate the role of ClpX and ClpXP in the disassembly and degradation of protein aggregates that bear specific ClpX recognition signals. We observed that ClpX, with and without ClpP, destabilizes Gfp-ssrA aggregates in vitro. The native ClpXP substrate FtsZ forms several discrete conformations, including linear ordered polymers and also heat-78 induced aggregates. Our results show that ClpXP disassembles both heat-induced and linear polymers containing FtsZ. Finally, we also demonstrate that thermal stress promotes aggregation of FtsZ, which is exacerbated in cells deleted for clpX or clpP.
Together, these results show bona fide chaperone activity for ClpX in vitro and suggest that ClpX, with or without ClpP, may play a broader role in rescue and disassembly of protein aggregates.

Bacterial strains and plasmids
E. coli strains and plasmids used in this study are described in Table 1

Expression and purification of proteins
Gfp-ssrA was purified as previously described (Yakhnin et al., 1998). ClpX, ClpP, FtsZ, and FtsZ(ΔC67) were each overexpressed in E. coli BL21 (λDE3) and Heat-induced aggregation of FtsZ with time was monitored by 90º-angle light scatter with the temperature of the cuvette holder set to 65 ºC using a circulating water bath. intensities were analyzed by densitometry (NIH ImageJ), normalized to the intensity of the average of the 'no heat' sample, and evaluated for significance by the Mann-Whitney test. Where indicated, to test a mild heat shock condition, cells were incubated in a water bath at 42 °C for 30 minutes, followed by recovery at 30 °C for 35 minutes, and analyzed as described.

ClpXP degrades aggregates in vitro
To determine if ClpX can remodel protein substrates from the aggregated state, we used the fusion protein, Gfp-ssrA, which forms aggregates upon heat treatment (Zietkiewicz et al., 2004;Zietkiewicz et al., 2006). Gfp-ssrA is rapidly degraded by ClpXP and has been extensively studied to understand substrate targeting by ClpXP. The Gfp moiety is widely used in protein disaggregation assays because it forms non-fluorescent aggregates when heated, but is disaggregated and reactivated by several chaperone systems (Zietkiewicz et al., 2004;Zietkiewicz et al., 2006). Therefore, we heated Gfp-ssrA at 85 °C for 15 minutes to induce aggregation (aggGfp-ssrA), resulting in an 86% loss of fluorescence emitted ( Figure 1A). Next, to measure the distribution of aggregates by size after heating, we performed dynamic light scattering (DLS) of untreated and heat-denatured Gfp-ssrA. We observed that without heating, the particle sizes are uniform with an average hydrodynamic diameter of 8-10 nm ( Figure 1B). After heating, aggregates are approximately 500-600 nm, and there is a narrow distribution of particle sizes and no small particles (i.e., less than 100 nm) ( Figure 1C). Upon heat-treatment, aggregation of Gfp-ssrA (1.5 μM) occurs rapidly and plateaus by 10 minutes by 90°-angle light scattering ( Figure  1D). The heat inactivation is irreversible since incubation of aggregated Gfp-ssrA (aggGfp-ssrA) alone does not lead to appreciable fluorescence reactivation, which is consistent with previous reports using Gfp ( Figure S1) (Zietkiewicz et al., 2004). To determine if ClpXP can bind to aggregates and degrade them, we incubated aggGfp-ssrA with ClpXP and monitored turbidity by 90°-angle light scattering. Incubation of aggGfp-ssrA with ClpXP led to a 35% loss of turbidity in 2 hours ( Figure 1E).
However, when ClpXP was omitted from the reaction, there was very little change in turbidity over time (5% loss in 2 hours) ( Figure 1E). This suggests that ClpXP targets aggregated substrates for degradation. To determine if degradation is required to reduce turbidity, we omitted ClpP and observed that ClpX is capable of reducing sample turbidity by 15% in 2 hours ( Figure 1E). Finally, when ATP was omitted from the reaction containing ClpXP, we observed a less than 10% reduction in the turbidity of the reaction ( Figure 1E). To confirm that ClpXP degrades aggGfp-ssrA, we incubated aggGfp-ssrA with combinations of ClpX, ClpP and ATP, and sampled degradation reactions after 2 hours. We observed that in the presence of ClpXP, aggGfp-ssrA is degraded, but not when ClpP or ATP was omitted ( Figure 1F).
Together, these results demonstrate that ClpXP targets aggregates for ATPdependent degradation and that ClpX is also capable of promoting disassembly in the absence of ClpP.
FtsZ is a well-characterized ClpXP substrate that is essential for cell division and forms linear polymers in vitro in the presence of GTP (Erickson et al., 2010). We previously showed that ClpXP binds to GTP-stimulated FtsZ polymers and promotes FtsZ degradation (Camberg et al., 2009). ClpXP also recognizes and degrades nonpolymerized FtsZ, but less efficiently than polymerized FtsZ (Camberg et al., 2009). In vitro, FtsZ rapidly aggregates when heated at 65 °C and this aggregation is associated with an increase in overall light scatter and a 97% loss of GTPase activity (Figure 2A and 2B). FtsZ, which purifies as a mixture of monomers (40.4 kDa) and dimers (80.8 kDa), has an average hydrodynamic diameter of 10-15 nm by DLS ( Figure 2C). Heat treatment of FtsZ (5 μM) at 65 °C produces several particle sizes, including small (30-40 nm) and large aggregates (>300 nm) ( Figure 2D). To determine if ClpXP reduces the turbidity associated with aggregated FtsZ (aggFtsZ), we incubated aggFtsZ with ClpXP and ATP and observed a 40% loss of turbidity after incubation with ClpXP for 2 hours ( Figure 2E). However, in the absence of ClpXP, the light scatter signal remained stable for aggFtsZ ( Figure 2E). Incubation of ClpX with aggFtsZ also resulted in a 25% loss in light scatter, suggesting that ClpX also promotes disassembly of aggregates similar to what we observed for aggGfp-ssrA ( Figure 2E and 1E).
Next, to confirm that aggFtsZ is degraded by ClpXP, we assembled reactions containing combinations of aggFtsZ, ClpX, ClpP and ATP and sampled these reactions at 0 and 120 minutes for analysis by SDS-PAGE. We observed that in the presence of ClpXP and ATP, 50% of the total aggFtsZ in the reaction is lost to degradation after 120 minutes ( Figure 2F). Omission of either ClpP or ATP from the reaction prevents loss of aggFtsZ ( Figure 2F). These results indicate that ClpXP degrades aggFtsZ. Furthermore, the amount of aggFtsZ after incubation with ClpX is unchanged despite the decrease in light scatter detected, suggesting that ClpX can disaggregate aggFtsZ ( Figure 2E and 2F).
In addition to forming aggregates upon heating, FtsZ also assembles into a linear head-to-tail polymer, which is a native, ordered aggregate, and distinct from the disordered aggregates which are induced by heating (aggFtsZ). We compared the loss of aggFtsZ by ClpXP to a similar reaction monitoring loss of native polymerized FtsZ, which is a known substrate of ClpXP. Like aggFtsZ, we also observed a ~50% loss of polymeric FtsZ, stabilized by the GTP analog GMPCPP, after 120 minutes in reactions containing ClpXP and ATP ( Figure 2F). GMPCPP promotes the assembly of stable polymers that are far less dynamic than polymers assembled with GTP (Lu et al., 2000). To test if ClpXP disassembles GMPCPP-stabilized FtsZ polymers, we incubated pre-assembled polymers with ClpXP and ATP. Then, we collected polymers by high-speed centrifugation. In these assays, we used active fluorescent containing ClpXP (26% of the total FtsZ was recovered in the reaction containing 1 μM ClpXP), indicating that ClpXP is highly effective at promoting the disassembly of GMPCPP-stabilized FtsZ polymers ( Figure 2G).

ClpX reactivates heat-aggregated Gfp-ssrA
Incubation of ClpX with aggGfp-ssrA resulted in loss of turbidity, suggesting that ClpX may function independently of ClpP to reactivate substrates ( Figure 1E).

Reactivation of misfolded proteins may occur through binding and stabilization of
intermediates enabling proteins to adopt the native folded conformation, or through ATP-dependent chaperone-assisted unfolding. To determine if ClpX, which recognizes the ssrA amino acid sequence, is able to reactivate aggGfp-ssrA, we ClpX catalyzes ATP-dependent unfolding of substrates Singh et al., 2000). To determine if ATP is essential for reactivation, we incubated aggGfp-ssrA with ClpX under various nucleotide conditions including with ATP, the ATP analog ATPɣS, ADP and omission of nucleotide. We observed an 82% slower rate of fluorescence reactivation when ClpX and aggGfp-ssrA were incubated with ATPɣS than with ATP (0.02 AU min -1 and 0.11 AU min -1 , respectively), and no recovery over background with ADP or without nucleotide ( Figure 3B). Reactivation by ClpX and ATP is prevented in the presence of ClpP, and the residual fluorescence after heat treatment is lost upon degradation ( Figure S2). Together, these results indicate that ClpX requires ATP to reactivate Gfp-ssrA and, surprisingly, that ATPɣS is also capable of promoting reactivation, although at a much slower rate than ATP ( Figure   3B).

Reactivation and disaggregation by ClpX requires a specific recognition sequence
Next, we determined if a ClpX recognition motif is important for efficient recognition of aggregated substrates by ClpX. We compared reactivation of aggGfp-ssrA with heat-aggregated Gfp (aggGfp) without an ssrA tag. We observed that after incubation with ClpX and ATP for 60 minutes, approximately 30 units of fluorescence were recovered, which is 8% of the initial pre-heat fluorescence, indicating that aggGfp is a poor substrate for reactivation by ClpX ( Figure 4A). In contrast, aggGfp-ssrA recovered 33% (>100 units) of the initial pre-heat fluorescence after incubation with ClpX ( Figure 4A). in an 84% loss of GTP hydrolysis activity and an increase in light scatter, which is stable over time ( Figure 4B and 4C). In the presence of ClpXP, we observed no decrease in light scatter for aggFtsZ(ΔC67) after incubation for 120 minutes ( Figure   4C), which is expected since FtsZ(ΔC67) is a poor substrate for ClpXP degradation ( Figure S3). Together, these results demonstrate that for ClpX to recognize aggregates and promote disaggregation, disassembly and/or reactivation, a ClpX recognition motif is required.

Impaired reactivation by ClpX(E185Q)
ATP is required for reactivation of aggGfp-ssrA, however, it is unknown if this event requires ATP-hydrolysis and substrate unfolding. Therefore, we used the ClpX mutant protein ClpX(E185Q), which has a mutation in the Walker B motif and is defective for ATP-hydrolysis, but interacts with substrates (Hersch et al., 2005;Camberg et al., 2014). We observed that ClpX(E185Q) is defective for disaggregation of aggGfp-ssrA by monitoring turbidity by 90°-angle light scatter of reactions containing aggGfp-ssrA, ClpX(E185Q) and ATP ( Figure 5A). We also tested if aggFtsZ is disassembled by ClpX(E185Q), and observed no reduction in light scatter in reactions containing aggFtsZ, ClpX(E185Q) and ATP after 120 minutes compared to ClpX ( Figure 5B). Finally, we tested if reactivation of aggGfp-ssrA requires ATP hydrolysis using ClpX(E185Q) instead of ClpX. We observed that ClpX(E185Q) promotes a small amount of reactivation of aggGfp-ssrA and restores fluorescence, but to a much lesser extent than the level observed for wild type ClpX ( Figure 5C). These results suggest that ATP hydrolysis by ClpX is required to promote efficient reactivation of aggGfp-ssrA and disassembly of large complexes containing aggFtsZ or aggGfp-ssrA ( Figure 5A, 5B and 5C).

ClpXP prevents accumulation of FtsZ aggregates in vivo under extreme thermal stress
ClpX and ClpP were previously reported to localize to protein aggregates in E.
coli, suggesting that ClpXP may target aggregates in vivo for direct degradation . We used the native ClpXP substrate FtsZ, which aggregates upon heat treatment, to determine if ClpX and/or ClpXP modulates FtsZ aggregate accumulation after thermal stress by comparing the levels of FtsZ present in insoluble cell fractions (Figure 2A and 6A). Wild type cells and cells deleted for clpX, clpP, clpB, clpA, dnaK, lon, hslU and hslV were exposed to heat shock and insoluble protein fractions were collected and analyzed by immunoblot. We observed that FtsZ was present in the insoluble fraction of wild type cells (BW25113), and this amount was 42% higher in cells exposed to heat shock at 50 °C ( Figure 6A and S4A). However, FtsZ levels were even higher in the insoluble fractions of ΔclpX and ΔclpP strains compared to the parental strain (2.4-fold and 2.3-fold, respectively), although the amount of total protein was similar to the wild type strain exposed to heat shock ( Figure S4B). We detected less protein overall in the ΔdnaK strain after recovery, but this strain also had poor viability after heat shock and recovery ( Figure S4C). In addition, we also detected ClpX in the insoluble fraction in all strains except the clpX deletion strain ( Figure S4A). Next, we conducted a mild heat shock, 42 °C for 30 minutes, followed by recovery, and observed that deletion of clpB had a larger effect on the accumulation of insoluble FtsZ than deletion of clpX ( Figure S4D). To determine the relative contributions of either clpB or clpX during a 40 minute recovery period after incubation at 50 °C, we analyzed insoluble FtsZ levels at 20 minute time intervals during recovery (Fig. 6B). Notably, we observed that in cells deleted for clpX, insoluble FtsZ was present immediately after heat treatment and continued to accumulate throughout the recovery period to a greater extent than in wild type or clpB deletion cells. These results suggest that ClpXP prevents accumulation of FtsZ aggregates in cells exposed to extreme thermal stress. Since we observed that insoluble FtsZ levels were elevated in ΔclpB strains exposed to mild heat shock ( Figure S4D), we repeated the recovery time course in clpX and clpB deletion strains after mild heat shock, 42 °C for 30 minutes, to monitor insoluble FtsZ levels ( Figure   S4E). We observed that insoluble FtsZ accumulates during the recovery period in clpB deletion strains after mild heat shock ( Figure S4E).
Finally, if ClpXP is active in cells after severe heat shock, then it should not be a thermolabile protein. To determine if ClpXP remains active after exposure to 50 °C in vitro, we incubated ClpXP in buffer at 50 °C for one hour, and then measured activity after addition of Gfp-ssrA by monitoring the loss of Gfp-ssrA fluorescence. We observed that ClpXP remained active for unfolding and degradation of Gfp-ssrA after incubation at 50 °C for one hour ( Figure S4F). As a control, ClpXP was also incubated in buffer at 30 °C for one hour and then assayed for activity. We observed that ClpXP incubated at 30 °C was more active than ClpXP incubated at 50 °C, suggesting that a partial loss of activity had occurred at high temperature ( Figure S4F). However, this assay was performed in the complete absence of other cellular chaperones or substrates and suggests that some ClpXP likely continues to retain activity after exposure to heat stress, while some may become inactivated.

Discussion
Here, using both a native and an engineered aggregated substrate, we demonstrate that ClpXP has the operational capacity to disassemble and degrade large aggregates that have ClpX degrons. In this study, FtsZ, a native substrate of ClpXP in E. coli, was aggregated in vitro by thermal stress, and we further show that FtsZ also aggregates in vivo when cells are exposed to high temperature (Figure 2A and 6A). The observation that FtsZ is aggregation prone is in agreement with a prior study reporting the presence of FtsZ in intracellular aggregates of ΔrpoH cells  Figure 1E). In addition, the Walker B mutation in ClpX, E185Q, which impairs ATP hydrolysis, also impairs disaggregation of aggGfp-ssrA and, to a lesser extent, aggFtsZ. Aggregate disassembly and resolubilization by ClpX was previously described using the substrate lambda O protein, and here we show disassembly of aggregates and kinetic monitoring using two additional substrates, as well as reactivation of Gfp-ssrA fluorescence (Wawrzynow et al., 1995). Reactivation of Gfp-ssrA is largely dependent on ATP hydrolysis ( Figure 3B Therefore, it is likely that large aggregates contain loosely associated unfolded proteins, which can be removed and reactivated by ClpX and, in the case of Gfp-ssrA, allowed to spontaneously refold. As expected, recognition by ClpX is highly specific, as Gfp without an ssrA-tag is not reactivated ( Figure 4A).
We also detected partial disaggregation of aggFtsZ by ClpX, but not by ClpX(E185Q) ( Figure 5B). Aggregation of FtsZ is induced at 65 °C, but the aggregates formed by FtsZ are smaller than those formed by Gfp-ssrA (30 nm and 600 nm, respectively) ( Figure 1C and 2D). FtsZ aggregates likely contain 8-10 monomers, based on the average size of a folded FtsZ monomer, which is approximately 40 Å in diameter ( Figure 2D) (Oliva et al., 2004). In contrast, Gfp aggregates in this study likely contain more than 120 subunits, based on an average size of a folded Gfp monomer, which is approximately 50 Å across the long axis (van Thor et al., 2005). The small size of the FtsZ aggregate may allow it to be more susceptible to disassembly by ClpX than a larger aggregate.
In the model for disassembly of aggregates by ClpXP, ClpX binds to exposed recognition tags on the surface of the aggregate and promotes removal, unfolding and degradation of protomers from within the aggregate ( Figure 7A). Removal of protomers eventually leads to destabilization and fragmentation of the aggregate as well as degradation ( Figure 1F and 2F). Although this process does not require ClpP, it occurs more robustly when ClpP is present than when ClpP is omitted ( Figure 1E and 2E).
For aggregated substrate reactivation, ClpX likely engages unfolded protomers from the aggregate, which may be internal or loosely bound to the exterior of the aggregate, unfolds and release them. For small aggregates, this activity may be sufficient to lead to fragmentation and capable of promoting reactivation of substrates such as Gfp-ssrA ( Figure 7B).
Finally, we observed large increases in insoluble FtsZ when cells were exposed to two different temperatures, 50 °C, which represents extreme heat shock, and 42 °C, which represents a mild heat shock ( Figure 6A, 6B and S4D). At 42 °C, deletion of clpB was associated with a large accumulation of insoluble FtsZ, suggesting that under mild heat stress, ClpB is the major factor that ensures FtsZ solubility ( Figure S4D and S4E). However, we observed a remarkably different result after heat shock at 50 °C and throughout the recovery period. Specifically, in a clpX deletion strain, large amounts of insoluble FtsZ accumulate during the recovery period to a greater extent than in a clpB deletion strain ( Figure 6A and 6B). It is unknown if ClpXP and ClpB are processing FtsZ aggregates directly in vivo, because we did not observe a reduction of aggregated FtsZ during the recovery period for any strain. FtsZ is typically present at very high levels (5,000 to 20,000 copies per cell) and is essential for cell division in E. coli (Bramhill, 1997). Interestingly, FtsZ also forms linear polymers as part of its normal biological function to promote cell division, and polymers are efficiently recognized, disassembled and degraded by ClpXP ( Figure 2F and

In vitro tau phosphorylation by okadaic acid
Okadaic acid is a lipophilic polyether toxin produced by marine dinoflagellates of the Prorocentrum and Dinophysis genera that accumulates in filter-feeding shellfish; consumption of mollusks containing okadaic acid causes diarrheic shellfish poisoning in humans, which causes damage to the gastrointestinal tract and can be fatal (Dickey et al., 1990;Cembella et al., 1990;Ito et al., 1994;Edebo et al., 1988).

Cell Culture Growth and Differentiation
Human neuroblastoma SH-SY5Y cells were grown as monolayer cultures in Membranes were incubated in HRP-conjugated secondary antibody dilutions for 1 hour at room temperature before developing the blots with chemiluminescent West Pico Plus reagent (Thermo Fisher Scientific).

Cell Death and Cell Viability Assay
To determine the effects of okadaic acid on differentiated cell morphology, cells in 6-well plates were visualized with phase-contrast microscopy using EVOS FL

Statistical Analyses
Statistical analyses were conducted with Graphpad Prism version 6.0, GraphPad Software, La Jolla, California, USA (www.graphpad.com). Error bars represent mean +/-standard error mean (SEM). One-way analysis of variance (ANOVA) followed by Tukey's post hoc multiple comparisons test was used to analyze data sets containing more than two groups. P-values less than 0.05 were considered significant.

Phosphatase inhibition by okadaic acid in undifferentiated neuronal cells
To monitor the effects of increased phosphorylation of tau on the physiology and chaperone protein expression of neuroblastoma cells, we first treated undifferentiated SH-SY5Y cells with okadaic acid, which inhibits PP2A phosphatase leading to tau hyperphosphorylation (schematic in Figure 1C) (Bialojan and Takai, 1988). As shown in Figure  Consistent with this, we also observed that after 24 h exposure to low doses of okadaic acid cells, cells became rounded and detached (Supplemental Figure S1A).
Viability continued to decrease across a range of okadaic acid concentrations during prolonged incubations (42 h) (Supplemental Figure S1B and S1C).
Since differentiated SH-SY5Y cells displayed phenotypically normal morphologies and no significant loss of viability after low-dose (10 nM and 25 nM) short-term (3 hr) treatment with okadaic acid ( Figure 3A, 3C, 3G and S1A), we  Figure 4A). However, we observed no increase in Tau phosphorylated at Thr181, indicating that not all sites are uniformly affected by okadaic acid. Interestingly, Hsp70 protein levels decrease dramatically after exposure to 25 nM okadaic acid for 3 h, and we also observed slightly less Hsp90 after treatment ( Figure 4A).
Next, to determine if changes in protein levels that we detected get more pronounced with extended incubation with okadaic acid, we harvested cells at 6 h and probed lysates for Hsp70, Hsp90 and phosphorylated-tau (Thr212) ( Figure 4B). We observed that levels of Hsp70 and Hsp90 continued to decrease over time, indicating that the reductions in levels were both dose-dependent and time-dependent.
Moreover, we continued to observe elevated levels of tau phosphorylated at Thr212.
Finally, long-term exposure to okadaic acid does not impact overall tau levels as detected by total tau antibody, which reacts with multiple tau isoforms (Supplemental Figure S2A and S2B). Together, these results demonstrate that inhibition of protein phosphatases leads to enhanced phosphorylation of tau and correlates with reduced function of ATP-dependent chaperones in differentiated SH-SY5Y cells.

Discussion
Protein aggregation is observed throughout several neurodegenerative diseases, such as Alzheimer's disease (AD), Parkinson's disease, and Huntington's disease (Pruisner, 2013   Here, we used okadaic acid as it has previously been reported to alter tau     A.) SH-SY5Y cells were exposed to okadaic acid for 3 hours and 25 µg of cell lysate was run on SDS-PAGE and immunoblotted.
B.) SH-SY5Y cells were exposed to okadaic acid for 6 hours and 15 µg of cell lysate was run on SDS-PAGE and immunoblotted.     Table S1. Okadaic acid doses required to observe in vitro changes in tau phosphorylation levels.
In the cell, protein homeostasis is regulated by chaperone proteins that can remodel and degrade misfolded, unfolded, and aggregate proteins. ClpXP and Hsp104 are chaperone proteins that belong to the AAA+ superfamily of ATPases (5).
The AAA+ (ATPases associated with diverse cellular activities) superfamily of ATPdependent enzymes have highly conserved domains and function as chaperones and/or proteases to manage protein folding and multiprotein complexes (5). The Clp/Hsp100 family of ATP-dependent enzymes within the AAA+ protein superfamily includes the cytosolic proteins ClpX, ClpA, and Hsp104. Members of the Clp/Hsp100 family refold or degrade protein assemblies, misfolded proteins, and aggregated proteins (6). Within the Clp/Hsp100 family, ATPases (ClpX) partner with a peptidase (ClpP) to form stacked, barrel-like structures that functions as chaperone proteases (6). The chaperone ATPase component of this complex threads polypeptides through the central axial channel of the chaperone protease complex into the peptidase chamber for degradation (7).
ClpXP is an ATP-dependent chaperone protease that binds to and degrades an estimated 50 substrate proteins based on proteomic analysis (8). ClpX forms a hexameric ring that binds to ClpP, which forms stacked heptameric rings; thus, there is asymmetrical binding between ClpX and ClpP (9, 10). ClpX contains an N-domain, a large nucleotide-binding domain (AAA+ domain), and a small AAA+ domain (11).
The N-domain of ClpX forms dimers that are connected to the large AAA+ domain by a flexible linker region and functions to bind adaptors, client proteins, and stabilize the hexameric structure of ClpX (12). Substrate recognition by ClpX requires the presence of short peptide sequences, or degradation tags, on client proteins, on the C-terminus or N-terminus of client proteins (5,83). An example of a C-terminal degradation tag recognized by ClpX is the ssrA tag, an 11-amino acid peptide basal levels of Hsp104 "cures" yeast of the [PSI + ] prion phenotype (27,28). The Ndomain of Hsp104 (amino acid residues 1 -147) is required for [PSI + ] prion curing in S. cerevisiae, but is not required for thermotolerance or [PSI + ] prion propagation . Overexpression of Hsp104 is neuroprotective in models of Parkinson's disease, suggesting that Hsp104 can also bind to human prion proteins, such as alpha-synuclein (30).  (Table 1). Site-directed mutagenesis was carried out using the QuikChange II XL Site-directed Mutagenesis kit. To express the ClpX sequence without the N-terminal domain (ClpXΔNTD), both pYES2 and pYES2+ClpX were double digested using the restriction enzymes SacI and NotI (New England Biolabs), resulting in linearized DNA that was subsequently ligated overnight at 16°C.
The ligation products were heat inactivated at 65°C and ligation products were transformed into XL Blue chemically competent Escherichia coli cells (Agilent).
Sanger sequencing of the construct confirmed that the ClpXΔNTD sequence (1092 base pairs) was successfully cloned into the pYES2 vector.
In order to express the Hsp104 N-terminal domain (amino acid residues 1 - The chimeric protein Hsp104 + ClpXΔNTD + ClpP has an N-terminal domain from Hsp104, and ATPase domain from ClpX, and the ClpP full-length sequence ( Figure 1B). The chimeric protein Hsp104 + ClpXΔNTD + ClpP was transformed into [PSI + ] yeast and grown on SD-URA plates for five days at 30°C. The chimeric protein did not alter [PSI + ] phenotype in yeast ( Figure 1C). To determine if ClpX and ClpP protein were expressed in yeast, lysates were extracted, normalized, and immunoblotted. We found that yeast containing the chimeric protein did express ClpX and ClpP (Figure 2A). Moreover, spectrophotometric analysis of yeast lysates using a novel [PSI + ] assay that we developed showed that level of red pigmentation, which accumulates in [psi -] yeast, is reduced in yeast expressing the chimeric proteins