Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Oct 1.
Published in final edited form as: Structure. 2019 Aug 26;27(10):1570–1580.e4. doi: 10.1016/j.str.2019.08.003

Structure of Membrane-Bound Huntingtin Exon 1 Reveals Membrane Interaction and Aggregation Mechanisms

Meixin Tao 1, Nitin K Pandey 1, Ryan Barnes 2, Songi Han 2, Ralf Langen 1,3,*
PMCID: PMC6774876  NIHMSID: NIHMS1538879  PMID: 31466833

SUMMARY

Huntington’s Disease is caused by a polyQ expansion in the first exon of huntingtin (Httex1). Membrane interaction of huntingtin is of physiological and pathological relevance. Using EPR and ODNP, we find that the N-terminal residues 3-13 of wild-type Httex1(Q25) form a membrane-bound, amphipathic α-helix. This helix is positioned in the interfacial region, where it is sensitive to membrane curvature and electrostatic interactions with headgroup charges. Residues 14-22, which contain the first five residues of the polyQ region, are in a transition region that remains in the interfacial region without taking up a stable, α-helical structure. The remaining C-terminal portion is solvent-exposed. The phosphomimetic S13D/S16D mutations, which are known to protect from toxicity, inhibit membrane binding and attenuate membrane-mediated aggregation of mutant Httex1(Q46) due to electrostatic repulsion. Targeting the N-terminal membrane anchor using post-translational modifications or specific binders could be a potential means to reduce aggregation and toxicity in vivo.

Keywords: Huntington’s Disease, Huntingtin Exon 1, Polyglutamine Expansion, Protein-Membrane Interaction, Membrane-Mediated Aggregation, Protein Misfolding, EPR, ODNP

Graphical Abstract

graphic file with name nihms-1538879-f0001.jpg

IN BRIEF:

Membrane interaction is relevant to huntingtin toxicity and misfolding in Huntington’s disease. Using EPR and ODNP, Tao et al. show structural details of the physiologically relevant huntingtin exon 1 in its membrane-bound state. The structural information together with binding studies revealed the factors that control membrane interaction and membrane-mediated aggregation.

INTRODUCTION

Huntington’s Disease (HD) is a fatal neurodegenerative disorder (Gomez-Tortosa et al., 1998; Kirkwood et al., 2001; Paulsen, 2011; Smith et al., 2000) with no existing cure. The disease is caused by the expression of mutant huntingtin protein with long, expanded polyglutamine stretches (>36Q) that forms amyloid-like aggregates, which can be found in HD patient tissue samples as well as cell and animal models (luchi et al., 2003; Scherzinger et al., 1997). A number of membrane-related functions have been reported for huntingtin, including vesicular trafficking (Gauthier et al., 2004; Gunawardena et al., 2003; Lee et al., 2004; Pal et al., 2006; Velier et al., 1998) and autophagy (Gelman et al., 2015; Martin et al., 2015; Ochaba et al., 2014). In addition to these physiological roles, membrane interaction is likely also of relevance in disease. In vitro, membrane interaction can potently accelerate aggregation and fibril formation of mutant huntingtin, in a misfolding process that can cause membrane damage and the formation of potentially toxic misfolded species (Pandey et al., 2018). In vivo, mutant huntingtin or its naturally occurring N-terminal fragments are present in membrane fractions (Kim et al., 2001; Suopanki et al., 2006; Velier et al., 1998). Moreover, mutant huntingtin can alter the fluidity of cell membranes (Sameni et al., 2018) and disrupt the morphology of membranous organelles (Liu et al., 2015; Squitieri et al., 2006). Interestingly, perturbations in lipid homeostasis and biosynthesis have also been reported in HD (Aditi et al., 2016; Block et al., 2010; Kreilaus et al., 2016; Valenza et al., 2005). These dysregulations and perturbations could contribute to neuronal toxicity and loss (Albin et al., 1992; Davies et al., 1997; Reiner et al., 1988; Richfield et al., 1995).

A frequently studied and biologically relevant fragment of the huntingtin protein is its exon 1 (Httex1). Full-length huntingtin is a large protein with 67 exons, but among these exons, Httex1 is of importance for a number of reasons. Httex1 contains the disease-causing polyQ region (Figure S1A) and forms amyloid-like fibrils in a Q-length dependent fashion. It is naturally generated in vivo, as Httex1 or fragments of similar size arise from aberrant splicing or proteolysis (Gafni et al., 2004; Graham et al., 2006; Neueder et al., 2017; Sathasivam et al., 2013; Wellington et al., 1998; Wellington et al., 2000). Importantly, multiple cell and animal models revealed that mutant Httex1 is sufficient to induce toxicity and mimic disease pathology (Chan et al., 2014; Duennwald and Lindquist, 2008; Mangiarini et al., 1996; Yang et al., 2010). Httex1 contains an important membrane-binding region at its N-terminus (N17 domain) that is thought to be a membrane anchor for interaction with a wide range of intracellular membranes, including those of ER, Golgi and mitochondria (Atwal et al., 2007; Rockabrand et al., 2007). The N-terminus also contains a nuclear export signal and is the target of multiple post-translational modifications (Atwal and Truant, 2008; Atwal et al., 2007; DiGiovanni et al., 2016; Lee et al., 2013; Zheng et al., 2013). Of note are the phosphorylation sites at residues S13 and S16, as phosphomimetic (S13D/S16D) mutations have been protective against mutant huntingtin toxicity in both cell and animal models (Di Pardo et al., 2012; Gu et al., 2009; Mishra et al., 2012). Bordering the polyQ domain at the C-terminus is a 50 amino acid long proline rich domain (PRD) that mediates the interactions of Httex1 with its protein binding partners containing SH3, WW or EVH1 domains (Gao et al., 2006; Kay et al., 2000; Zarrinpar et al., 2003).

In fibrils of Httex1, each of the three domains takes up very different structures (Bugg et al., 2012; Isas et al., 2017; Sivanandam et al., 2011). The polyQ region forms the β-sheet core of the fibrils. While the N17 also becomes ordered and tightly packed, it contains mainly α-helical structure. The C-terminal PRD takes up a mixture of polyproline II helical and random coil structure, and according to the bottle brush model of Httex1 fibrils (Bugg et al., 2012; Isas et al., 2017), these regions form bristles that face outward away from the fibril core. The formation of Httex1 fibrils can be promoted by the presence of membranes (Pandey et al., 2018). Unfortunately, no detailed structural information for membrane-bound Httex1 is currently available. Most of the existing structural information has been obtained from studies of smaller, Httex1-mimicking peptides with various membrane-mimetic molecules (Ceccon et al., 2018a; Ceccon et al., 2018b; Levy et al., 2018; Michalek et al., 2013a; Michalek et al., 2013b). These studies concluded that the N17 can adopt α-helical structure in its membrane-bound state, but there is little consensus on the length of the helix and the residues which partake in helix formation (Ceccon et al., 2018b; Levy et al., 2018; Michalek et al., 2013b). The structural differences in prior studies could have been caused by the different peptide lengths or the different membrane mimetic reagents that were used in those studies. The precise peptide length could become important if regions outside the N17 modulate membrane interaction. In fact, it has been suggested that the C-terminal PRD might also influence membrane interaction (Burke et al., 2013). The use of different lipid or membrane-mimetic conditions could have further contributed to the structural difference in prior studies. Such effects have already been reported for other amyloidogenic proteins, such as α-synuclein (Bortolus et al., 2008; Drescher et al., 2008; Georgieva et al., 2008; Jao et al., 2008; Ulmer et al., 2005; Varkey et al., 2013).

The main goals of the present study were, therefore, to (1) determine how the entire Httex1 protein binds to intact phospholipid bilayers (rather than membrane mimetics), (2) measure the depth of Httex1 insertion into membranes, and (3) determine mechanistically which forces promote membrane binding, membrane-mediated aggregation and fibril formation. Toward this end, we used a combination of biophysical techniques, including transmission electron microscopy (TEM), circular dichroism (CD), continuous-wave electron paramagnetic resonance (CW-EPR) and Overhauser dynamic nuclear polarization (ODNP).

RESULTS

Httex1(Q25) membrane interaction is mediated by its N-terminal region

In order to investigate the membrane-binding induced conformational changes by EPR, we generated tag-free, singly R1-labeled Httex1(Q25) derivatives and recorded their EPR spectra in the absence and presence of small unilamellar vesicles (SUVs) containing 25% POPS and 75% POPC. Under these conditions, SUVs remained visible by transmission electron microscopy over a period of at least seven hours regardless of whether Httex1 was absent or present (Figure S1B and S1C). As illustrated with three sites for each domain (N17, polyQ and PRD), all spectra of R1-labeled derivatives of monomeric Httex1(Q25) gave rise to CW-EPR spectra with relatively sharp line shapes (Figure 1, black spectra), consistent with a significant degree of structural dynamics and fast rotational tumbling in the absence of membranes (Bravo-Arredondo et al., 2018; Newcombe et al., 2018). In contrast, the Httex1(Q25) derivatives with spin labels introduced within the N17 domain gave rise to line broadening and pronounced reduction in amplitude in the presence of lipid vesicles (Figure 1, red spectra), indicating that the N17 domain becomes less dynamic upon membrane interaction. This ordering also affected an N-terminal site in the polyQ region (21R1), while the more C-terminal sites (35R1, 42R1, 55R1, 80R1, 90R1) did not reveal any significant spectral changes upon membrane interaction. To further verify whether membrane-induced structural changes were localized to the N-terminal region of Httex1(Q25), we generated 20 additional R1-labeled derivatives (Figure S1A) and recorded their EPR spectra in the presence of membranes (Figure S2). The mobility information contained in all spectra was summarized by the inverse central line width (ΔH0−1), a commonly employed mobility parameter (Figure 2A, red circles). In the presence of membranes, the lowest ΔH0−1 values were obtained for the first 22 residues while the values in the C-terminal region were much larger, indicating much higher mobility in the latter regions. The result indicated that most membrane-induced ordering occurred in the first ~ 22 amino acids, thus including a region that encompasses the N17 and the first five N-terminal residues in the polyQ region. From residue 23 onwards, mobility increased quickly, further supporting the notion that the effect of membrane interaction was less pronounced for more C-terminal residues. While the EPR spectra for the first 22 amino acids indicated structural ordering, they also revealed a lack of tertiary or quaternary packing interactions (Figure S2). Such interactions would have led to more strongly immobilized spectral components and larger line widths, neither of which were observed here. Collectively, the line shapes suggested that no significant oligomerization or aggregation of the proteins occurred on the lipid membrane surface for Httex1(Q25). Under the present conditions, we also found no evidence for aggregation of the Httex1(Q25) over a period of seven hours by TEM (Figure S1C). Moreover, we monitored the EPR spectrum of Httex1(Q25) labeled at position 35 (35R1) (Figure S1D) during the same time period in the presence of membranes. Residue 35 was chosen because it is located within the center of the polyQ domain, a region that adopts a β-sheet structure in fibrils (Isas et al., 2017). The EPR spectral lines remained consistently sharp, indicating that the site resides in a highly dynamic, freely moving region, further supporting the lack of aggregation under the present conditions. The finding that this non-pathogenic variant (i.e. short Q-length containing Httex1(Q25)) does not aggregate under these conditions is important, as we previously found that membranes can potently accelerate fibril formation of Httex1 with longer, pathogenic Q-lengths (Pandey et al., 2018) (also see below).

Figure 1. CW-EPR spectra of singly R1-labeled Httex1(Q25) derivatives.

Figure 1.

Open in a new tab

The X-band EPR spectra of Httex1(Q25) derivatives with spin labels in N17 (4R1, 9R1, 17R1), polyQ (21R1, 35R1, 42R1) or PRD (55R1, 80R1, 90R1) in the absence of lipid vesicles (black) and in the presence of lipid vesicles (red). Scan width was 100 G and modulation amplitude was 1.5 G. All spectra were normalized to have the same central line amplitude for protein in the absence of lipid vesicles. EPR central line amplitude reduction and line broadening upon membrane binding are visible for 4R1, 9R1, 17R1, and 21R1, but not for the more C-terminal sites. See also Figure S1 and S2.

Figure 2. EPR accessibility measurements and CD indicate the formation of helical structure upon membrane binding.

Figure 2.

Open in a new tab

(A) Depth parameter Φ (blue) and inverse central line width ΔH0−1 (red) are plotted for the various labeling positions. Error bars show ± standard deviation of independent repeats (n = 3). (B) Helical wheel representation of Httex1(Q25) amphipathic helix (residues 3 to 13). Residues with Φ values > 0.8 (more membrane exposed) are colored in grey and fall onto one surface of the helical wheel. (C) Normalized EPR central line amplitudes of 15 μM Httex1(Q25) 21R1 in the presence of increasing concentrations of 25% POPS and 75% POPC SUV lipids (0, 0.15, 0.75, 1.5, 3, 10 mM). All spectra were normalized and the spectral amplitude of the sample in the absence of lipid vesicles was set to 1. The gradually stabilized curve indicates that membrane binding saturates between 1 and 2 mM of lipid. (D) CD spectra of Httex1(Q25) 21R1 in the absence and presence of 3 mM 25% POPS and 75% POPC SUV lipid. The spectra are represented as mean residual ellipticities [θ]MRE222nm plotted as function of wavelength (200 nm to 260 nm). (E) The difference spectrum between the membrane-bound spectrum and the solution spectrum shown in (D) had minima near ~ 208 nm and 222 nm, suggesting increased α-helicity upon membrane interactions. See also Figure S2 and S3.

Accessibility measurements reveal formation of a short amphipathic α-helix in N17 with shallow membrane penetration.

To characterize the structure and membrane immersion depth of Httex1(Q25) in more details, we measured the O2 and NiEDDA accessibilities (ΠO2 and ΠNiEDDA) of the R1-labeled derivatives. This method is based on the preferential partitioning of O2 into the hydrophobic membrane environment, whereas NiEDDA is predominantly located in the hydrophilic aqueous environment (Altenbach et al., 1994). It is well established that the ratio of the two accessibilities is a measure of membrane immersion depth (Altenbach et al., 1994). The natural log of this ratio represents the immersion depth parameter Φ (Φ = In(ΠO2/ΠNiEDDA)). This measure is directly proportional to the immersion depth with larger Φ values corresponding to deeper membrane insertion. As shown in Figure 2A (blue circles), the Φ values are generally elevated in the first 22 residues, indicating that this region, which also experienced the most pronounced membrane-mediated structuring, is in contact with the membrane.

While the Φ values were generally high in the first 22 amino acids, some fluctuations could still be detected. These were particularly pronounced for residues 3 to 13 for which the Φ values underwent a periodic oscillation where residues 3, 4, 6, 7, 8, 10 and 11 had the highest values (more membrane exposed), while residues 5, 9, 12, and 13 are local minima (more solvent exposed). As illustrated with the helical wheel representation in Figure 2B, the two groups of residues fall onto opposing surfaces of an amphipathic helix. In agreement with the notion of an asymmetrically solvated, amphipathic helix, we also find that the O2 and NiEDDA accessibilities for this region are out-of-phase, with a periodicity of an α-helix (~ 3.6 amino acids/turn) (Figure S3). Out-of-phase periodicity is typically observed for asymmetrically solvated helices as residues facing toward the acyl chain region experience the highest O2 and the lowest NiEDDA accessibilities, while the opposite is the case for residues on the opposing, solvent-facing surface (Hubbell et al., 1998). To further verify that membrane interaction induces α-helical structure in Httex1, we performed circular dichroism (CD) experiments. Since the high UV light absorbance at the highest lipid concentrations used in the EPR experiments (up to 20 mM) complicated the CD experiments, we first used EPR to ensure that Httex1 was still membrane-bound at lower vesicle concentrations. Using the EPR amplitude change of Httex1(Q25) 21R1, we found that vesicle binding saturates at lipid concentrations between 1 and 2 mM (Figure 2C). Using CD, we were able to record spectra with up to 3 mM lipid and obtained reasonable signal-to-noise for a wavelength range between 200 to 260 nm. Indeed, the addition of lipid vesicles induced a shift in the CD spectra (Figure 2D). The difference spectrum obtained for the protein in the absence and presence of 3 mM lipid had the typical shape expected for α-helix formation (Figure 2E). These data further support the notion that membrane interaction increases α-helical structure in Httex1(Q25).

In order to determine the degree of membrane immersion for the membrane-bound regions of Httex1(Q25), we further analyzed the EPR data by converting the Φ values of membrane-exposed residues into immersion depth using calibration standards (see STAR METHODS). Based on this calibration, we find that the side chains of the membrane-exposed residues in the helix are located just below the level of the phosphate group (3R1, 4R1, 7R1 and 11R1 at − 1, − 1, − 1 and − 5 Å, respectively). Here, the negative values indicate a location in the hydrophobic portion of the bilayer relative to the phosphate level which is defined as zero. Considering that the membrane-exposed side chains are the most deeply inserted part of an amphipathic helix, we can estimate that the center of the helix is in the interfacial region, above the phosphate level of the lipid headgroups.

ODNP measurements indicate solvent-exposed, C-terminal region that does not interact with the membrane.

Due to the shallow insertion of Httex1 into the membrane, O2 and NiEDDA accessibility measurements could not give reliable membrane proximity estimates for much of the Httex1(Q25) protein. In order to extend the range for membrane proximity measurements into the aqueous solution, we performed ODNP measurements that are more sensitive to distances above the phosphate level (Fisette et al., 2016). ODNP measures water accessibility and dynamics, which is quantified using the electron-proton spin cross relaxivity, kσ, between the unpaired electron spin of the R1 label and the proton spin of water. We then compute the kσ retardation factor as determined by the ratio of kσ of bulk water over kσ of surface hydration water, where a value of 1 corresponds to bulk water diffusivity and increasing kσ retardation values to slower water diffusivity. Two key factors have been established that allow us to deduce from the kσ retardation factor information about the distance of the R1 label from the lipid membrane phosphate level. On the one hand, the kσ retardation factor has been shown to have an approximately linear relationship with the retardation in surface water diffusivity in the retardation factor range up to 6, and still monotonically increase with water diffusivity retardation beyond this range (Barnes et al., 2017). On the other hand, it has been established previously that the retardation in surface water diffusivity inversely scales with the distance of the R1 label to the phosphate level on lipid membrane surfaces, with water diffusivity monotonically approaching bulk water diffusivity values at greater distance from the membrane surface (Cheng et al., 2013; Fisette et al., 2016). Consequently, greater kσ retardation corresponds to distances closer to the lipid membrane. The kσ retardation factors were measured for six singly R1-labeled Httex1(Q25) derivatives (11R1, 22R1, 30R1, 42R1, 60R1 and 90R1) (Figure 3A) and the ODNP data were summarized in Figure S4. The data follow a qualitatively similar trend as the EPR-based mobility and accessibility studies. The strongest water diffusion retardation was seen for residue in the N-terminal region (11R1) and the values decreased toward the C-terminal region, with residues 20-30 being in a transition region (22R1) and the smallest retardation values observed from residue 30 onward. When converted into distances using an empirical calibration (Fisette et al., 2016), we find that the 11R1 position penetrates ~ 1 Å below the phosphate level of the lipid headgroup (Figure 3B). This result is consistent with the shallow membrane penetration obtained from EPR power saturation. In addition, residue 22 was estimated to be ~ 5 Å above the phosphate level. The further C-terminal residues (30, 42, 60 and 90) were at distances more than 10 Å above the phosphate level, i.e. farther enough from the level of the phosphate to approach the dynamics of the bulk water environment. Collectively, the ODNP and the EPR data suggest a simple model of membrane-bound Httex1 in which residues 3 to 13 form an α-helical structure, the ordered secondary structure is then gradually lost in the transition region from residue 14 to 22, which are located still within the interfacial membrane contact region, while residues 23 to 92 are in the bulk aqueous environment (Figure 6A).

Figure 3. Httex1(Q25) ODNP Measurement.

Figure 3.

Open in a new tab

(A) kσ retardation factors of Httex1(Q25) versus residue number. (B) Distance to phosphate level versus residue number, where 0 is at, − is below and + is above phosphate level of the lipid headgroup. kσ retardation factors of Annexin B12 are taken from a previous publication (Cheng et al., 2013) as the reference to calibrate the distance of Httex1(Q25) (error bars ± standard deviations, n = 3). See also Figure S4.

Figure 6. Schematic illustration of Httex1 membrane interaction.

Figure 6.

Open in a new tab

(A) The interaction of Httex1(Q25) with curved membranes is shown. PolyQ and PRD domains are colored brown and green respectively. N17 is blue with some residues highlighted. Selected solvent-exposed residues (E5 and K9) are highlighted in cyan, while selected membrane-exposed residues (L4, L7 and L11) are in red, phosphorylation residues (S13 and S16) are in yellow and Q22 is in magenta. The α-helical structure for residues 3-13 is shown via ribbon representation. The C-terminus of N17 and the N-terminus of polyQ domain comprise the transition region (14-22) and the rest of the protein (residue 23 onward) are completely solvent-exposed. (B) Illustration of how the N-terminal Httex1 helix (red circle) preferentially interacts with curved membranes. Vesicles with positive membrane curvature have greater lipid packing defects in the headgroup region, allowing the helix to more readily bind to membranes. (C) Membrane-mediated aggregation of long Q-length Httex1 illustrated with Httex1(Q46). Monomeric Httex1(Q46) collisionally encounter each other on the membrane, the polyQ regions from different molecules come into contact and initiate intermolecular β-sheet formation, ultimately leading to the formation of fibrils and aggregates (color code of domains is as in A).

Httex1(Q25) membrane binding depends on vesicle size and membrane curvature

The relatively shallow insertion of Httex1(Q25) into the membrane could have implications for the membrane curvature dependence of membrane binding. That is, by being located up in the headgroup region, one might expect Httex1 to have stronger binding to membranes with positive curvature as such membrane can more easily accommodate additional protein mass in their headgroups (Figure 6B). To test this model, we incubated Httex1(Q25) 21R1 with vesicles of increasing diameter (i.e. decreasing positive curvature). For all experiments, we used 1.5 mM lipid, the concentration at which binding was close to saturation in the prior experiments (Figure 2C). Again, we used the EPR amplitude changes as a measure of membrane binding. Using this readout, we found that binding to the highly curved SUVs was much more potent than to larger vesicles with 200 nm or 1000 nm diameters (Figure 4A, for EPR spectra see Figure S5A). As described in the STAR METHODS section, it was possible to convert the spectral and amplitude changes into percent binding. This yielded 91, 32 and 23% binding for SUVs, 200 nm and 1000 nm vesicles, respectively. Thus, vesicle size and membrane curvature strongly influence membrane binding affinity. This is consistent with prior observations made on Httex1-mimicking peptides interacting with lipids tethered to different solid supports (Chaibva et al., 2014).

Figure 4. Membrane curvature, ionic strength and charge modulate Httex1(Q25) membrane binding.

Figure 4.

Open in a new tab

Percent membrane binding of Httex1(Q25) 21R1 was obtained from the EPR spectra under varying conditions. Panel (A) shows the effect of decreasing membrane curvature by using SUVs, 200 nm LUVs, 1000 nm LUVs in 20 mM Hepes, 100 mM NaCl, pH 7.4 buffer. In panel (B) increasing concentrations of NaCl (0, 100, 500, 1000 mM in 20 mM Hepes, pH 7.4) and SUVs with 25% POPS and 75% POPC are used. In panel (C) SUVs with increasing molar percentages of negatively-charged POPS lipids (0, 5, 10, 25, 50%) are used in 20 mM Hepes, 100 mM NaCl, pH 7.4 buffer. One-way ANOVA analyses with Dunnett’s multiple comparisons test were performed relative to the experiment group using 25% POPS and 75% POPC SUVs in 20 mM Hepes, 100 mM NaCl, pH 7.4. Statistically significant differences are shown as **** adjusted P < 0.0001 and ** 0.001 < adjusted P < 0.01. Error bars show ± standard deviation of independent repeats (n = 3). All data analyses were performed in GraphPad Prism 7.0. For EPR spectra, see Figure S5.

Electrostatic interactions contribute to Httex1 membrane interaction

According to the just outlined structural model (Figure 6A), the N17 comes into close contact with the membrane. This region of Httex1(Q25) has a net positive charge (3 Lys residues and 2 Glu residues), making it possible that charge interactions with negatively charged membranes could promote binding. In order to test for electrostatic components that could promote membrane interactions, we investigated the ionic strength and lipid charge dependence of Httex1 membrane interaction. To test the effect of ionic strength, we incubated the R1-labeled protein Httex1(Q25) 21R1 with SUVs containing 25% POPS and 75% POPC at different ionic strengths (i.e. NaCl concentrations), and monitored the resulting changes in EPR amplitude. Increasing ionic strength attenuated the membrane-dependent drop in signal amplitude in a relatively small, but clearly detectable manner (Figure 4B), indicating less membrane binding under those conditions (for EPR spectra see Figure S5B). This suggests that electrostatic interactions modulate the binding event. Next, we evaluated how membrane binding depends on the molar fraction of the negatively charged POPS lipids. Increasing amounts of POPS strongly enhanced membrane binding, but it should be noted that clearly detectable (40%) binding was also observed for vesicles containing the net neutral 100% POPC (Figure 4C, for EPR spectra see Figure S5C). Collectively, these experiments show that charge interactions significantly contribute to the membrane interaction of Httex1, but that there are additional energy contributions that do not require a net negative membrane surface charge.

Phosphomimetic mutations in N17 potently decrease the membrane binding affinity of Httex1(Q25) and protect Httex1(Q46) from membrane-mediated aggregation.

Having found that charge interactions contribute to the membrane interaction of Httex1(Q25), we next wanted to know how the introduction of negative charges in the N17 modulates membrane interactions. A well-studied alteration of N17 charge is phosphorylation at positions S13 and S16. This modification has frequently been mimicked by S13D/S16D mutations, which introduce two additional negative charges into the N17. Phosphomimetic (S13D/S16D) mutations have previously been shown to reduce huntingtin toxicity in cell and animal models of HD (Di Pardo et al., 2012; Gu et al., 2009; Mishra et al., 2012). To test whether the additional negative charges at positions 13 and 16 affected membrane interaction, we performed lipid titration experiments where binding was monitored using EPR and CD as described above. In these experiments, the 21R1 derivatives of Httex1(Q25) or Httex1(Q25) S13D/S16D were incubated with SUVs at varying lipid concentrations (up to 20 mM). As shown in Figure 5A (for EPR spectra see Figure S6A and S6B), the phosphomimetic mutations reduced membrane binding affinity, as indicated by the ~ 15 times higher amounts of lipid required for saturation of membrane binding. The reduced membrane binding propensity of the S13D/S16D mutant was further verified by CD, where lipid addition produced much smaller changes in the mean residual ellipticity Δ[θ]MRE222nm (Figure 5B, for CD spectra see Figure S6C and S6D). Thus, the EPR and CD data support the notion that the S13D/S16D mutations reduced membrane binding.

Figure 5. Phosphomimetic (S13D/S16D) mutations decrease membrane binding affinity of Httex1(Q25) and protect Httex1(Q46) from membrane-mediated aggregation.

Figure 5.

Open in a new tab

(A) The central line amplitudes of the EPR spectra for the 21R1 derivative at 15 μM are shown as function of lipid concentration. Data from proteins containing S13/S16 are shown in black, while those from S13D/S16D are shown in red. All spectral amplitudes were normalized relative to the spectrum in the absence of lipid, which was set to 1. For EPR spectra, see Figure S6A and S6B. (B) Changes in CD mean residual ellipticity Δ[θ]MRE222nm were plotted against lipid concentration, colored as in (A). For CD spectra, see Figure S6C and S6D. (C) For monitoring aggregation kinetics of Httex1(Q46), EPR spectra were recorded for 15 μM Httex1(Q46) 35R1 with or without S13D/S16D mutations in the presence or absence of 375 μM 25% POPS and 75% POPC SUV lipids. The central line amplitudes were normalized by setting the t0 amplitude to 1 and the resulting values from 1 hour and 2 hours post mixing were plotted for comparison (S13/S16 in black and S13D/S16D in red). Unlike the Httex1(Q46) 35R1 group, no statistically significant membrane-mediated aggregation was observed for Httex1(Q46) 35R1 with S13D/S16D mutations (t-tests, ** p < 0.01, n = 3). Error bars show standard deviation of independent repeats (n = 3). All data analyses were performed in GraphPad Prism 7.0. For a more complete time course, see Figure S6E.

For the non-pathogenic Httex1(Q25) discussed thus far, we find no evidence for oligomerization or aggregation. However, we recently reported that interaction with negatively charged membranes promotes aggregation and fibril formation in case of the pathogenic, long Q-length containing Httex1(Q46) (Pandey et al., 2018). Inasmuch the S13D/S16D mutations attenuate membrane interaction of Httex1(Q25), we sought to test how these mutations affect membrane-mediated aggregation. In order to monitor misfolding of Httex1(Q46) in the presence of membranes, we again used the previously developed EPR-based readout, which monitors the β-sheet formation of the polyQ region by tracking the spectral amplitude decrease of 35R1. A faster reduction in the signal amplitude corresponds to faster rates of aggregation. First, we verified that the vesicles used in the present study (SUVs) could potently accelerate the aggregation of Httex1 for its long Q-length variant (Q46). In the presence of SUVs, we found that the EPR amplitudes decayed more rapidly than in the absence of SUVs (Figure 5C, for a more complete time course see Figure S6E). In contrast, the aggregation of 35R1 Httex1(Q46) with the S13D/S16D mutations was less sensitive to the addition of liposomes. After 1 or 2 hours, the EPR amplitudes were very similar regardless of whether lipids were added or not. It should also be noted that the S13D/S16D mutations slowed down the aggregation of Httex1(Q46) in solution, which is in agreement with previous studies (Gu et al., 2009; Mishra et al., 2012). The minor effect of membranes on promoting the aggregation kinetics of Httex1(Q46) with S13D/S16D mutations compared to Httex1(Q46) was consistent with the aforementioned attenuation of membrane interaction caused by the S13D/S16D mutation. Together these data indicate that the S13D/S16D mutations are generally inhibitory for solution or membrane-mediated aggregation pathways.

DISCUSSION

Here we investigated the structure of Httex1 stably-bound to intact phospholipid membranes using a combination of biophysical techniques. We found that negative membrane charge and strong membrane curvature (SUV) led to strong binding. Inasmuch as Httex1(Q25) was stably bound without aggregating into misfolded species, it was possible to systematically investigate its structural features, including the local secondary structure and the membrane proximity of the protein. As schematically illustrated in Figure 6A, we found that the N-terminal portion of the N17 forms a short amphipathic α-helix extending approximately from residue 3 to 13. This α-helix is largely located in the headgroup region. This region was C-terminally flanked by a transition region that extends from residue 14 to 22. While this region did not exhibit the characteristic helical structure of the preceding region, it nonetheless became less dynamic (presumably partially structured) and remained in the proximity of the headgroup region of the lipid membrane. For the more C-terminal residues, the EPR spectra of R1-labeled Httex1(Q25) in solution and bound to membranes became more and more superimposable, suggesting that these regions did not experience a significant structural change upon membrane interaction. Altogether we found that the changes in structure and dynamics of Httex1 were predominantly within the first 22 residues. This region encompasses not only the N17, but also the first five Gln residues. This perhaps somewhat unexpected involvement of residues in the polyQ region shows that shorter peptides, which only contain the N17, may not be ideal models for studying Httex1 membrane interaction.

Our data on Httex1(Q25) also indicated that negative membrane charge and positive membrane curvature (outside surface of highly curved SUVs) significantly promote Httex1 membrane interaction (Figure 4). This binding behavior can be rationalized by the structure and the increased headgroup spacing in case of positive curvature (Figure 6B). Highly curved membranes, such as those in an SUV have a much higher outer than inner surface area. This means, a headgroup for a given lipid in the outer leaflet has to cover a much larger surface area than the acyl chain tails. This reduced packing density in the outer leaflet increases progressively with increasing distance away from the membrane interior. As a consequence, positively curved membranes are well known to have packing defects in the headgroup region, which can cause water penetration into the acyl chain region. By binding higher up in the headgroup region, the Httex1(Q25) amphipathic helix is well-situated to fill the voids in the headgroup region, thereby reducing packing defects. This idea is furthermore consistent with computational studies showing that a headgroup location of an amphipathic helix favors positive rather than negative membrane curvature (Campelo et al., 2008).

The importance of negatively charged lipids in the membrane interaction of Httex1 is consistent with the predominantly positively charged characteristics of the N17. The N17 is overall positively charged, containing three lysines (K6, K9, and K15) and two glutamic acids (E5 and E12). The location of the charged residues also likely impacts the structure of the membrane-bound state. K6 and K9 as well as E5 and E12 are located within the amphipathic helix. As is commonly observed for membrane-bound amphipathic helices, all these residues are located on the hydrophilic surface. In contrast, K15 is just flanking the helical region. Interestingly, K15 would be located on the hydrophobic surface of the helix if the helix were to be further extended. This would be a very uncommon position for a Lys residue and could explain why the helical region does not extend all the way to K15. Although K15 is outside the helix, it is nonetheless in the interfacial region, where it can interact with the negatively charged lipid headgroup and promote membrane interaction. The notion that positive charges on the N17 promote membrane interaction is further supported by the observation that acetylation of the N17 Lys residues reduces membrane-binding affinity (Chaibva et al., 2016). Conversely, N17 phosphorylation at residues S13 and S16 would introduce two negatively charged moieties into the interfacial region. While this location avoids exposure to the non-polar region of the membrane, bringing additional negative charges into the net negatively charged interfacial region would nonetheless be expected to be unfavorable. Here we mimicked phosphorylation by introducing S13D/S16D mutations, which are well characterized, and which have been shown to block toxicity in cell and animal models (Di Pardo et al., 2012; Gu et al., 2009; Mishra et al., 2012). As expected from their locations, the mutations potently inhibited membrane binding by an order of magnitude. This effect is likely even more significant in case of actual phosphorylation. While phosphomimetic mutations as well as phosphorylation introduce negative charge, there are some chemical differences. Depending on the pH, a phosphorylated side-chain can have up to two charges, whereas an Asp side chain only has one. The larger number of negative charges introduced with phosphorylation could therefore reduce membrane affinity even more. One way of accommodating an acidic side chain upon membrane interaction is to protonate it. While one of the negative charges on a phosphate can be protonated at mildly acidic pH, the second charge has a pKa that is much lower than that of Asp. Thus, it would be more difficult to protonate the second charge of a phosphate as compared to that of an Asp. While we did not directly compare our results to the phosphorylated Httex1, a recent study found that phosphorylation of S13 and S16 was indeed more potent at inhibiting membrane interaction than the phosphomimetic mutants for N-terminal peptides (Deguire et al., 2018). Despite the significant importance of electrostatic interactions in membrane binding of Httex1, other factors are likely to contribute as well. We conclude this from the fact that interaction with uncharged membranes is still possible (albeit strongly reduced). This residual binding is likely driven by hydrophobic residues such as L4, L7, and F11, which are facing toward the membrane.

A common feature in all the EPR spectra of membrane-bound Httex1(Q25) was the lack of strongly immobilized components (Figure S2), which would have been expected in case of tertiary or quaternary contacts. This means that Httex1(Q25) is largely monomeric under the present conditions, and that it does not contain any globular structures. This behavior was reminiscent of what we observed in the early spectra of membrane-bound Httex1(Q46) just prior to aggregation and fibril formation, as these spectra also lacked strongly immobilized components immediately after the addition of lipids (Pandey et al., 2018). This means that membrane binding did not immediately induce an oligomeric state, but that oligomerization was a subsequent step. Oligomerization was found to be polyQ-length dependent, as shorter Q-lengths, like the Httex1(Q25) studied here, did not exhibit any evidence of oligomerization or aggregation. How then could the largely monomeric, membrane-bound state facilitate aggregation of Httex1(Q46)? Compared to the inefficient, three-dimensional diffusional encounter of monomeric proteins in solution, Httex1 molecules can more readily find each other when diffusing in two-dimensions on the membrane surface. This reduced dimensionality is thought to be a major driver of membrane-mediated aggregation of various amyloidogenic proteins (Jayasinghe and Langen, 2007; Rawat et al., 2018). For Httex1, the N-terminal helix is the primary membrane anchor that promotes two-dimensional diffusion. The contacts between multiple Httex1 molecules could then initiate intermolecular β-sheet formation in the polyQ region, ultimately leading to fibril formation (Figure 6C). It is also possible that membrane binding causes a conformational change in the monomer structure, especially in the transition region. In addition, formation of β-sheet structure in the polyQ region during aggregation could be further promoted by the relatively low dielectric constant of the interfacial region, which can favor the formation of secondary structure. It is also interesting to note that the membrane-bound α-helix does not have to be converted into β-sheet structure upon fibril formation. In fact, the helical region in the membrane-bound state (residues 3 to 13) closely corresponds to the α-helical region found in fibrils (residues 4 to 11) (Sivanandam et al., 2011).

Our prior study identified parallel pathways (in solution and on membranes), by which Httex1 with expanded Q-lengths can aggregate into fibrils. It is therefore possible that an effective therapeutic strategy might have to take all of these pathways into account. The phosphomimetic mutations have previously been shown to be highly protective against huntingtin toxicity (Di Pardo et al., 2012; Gu et al., 2009; Mishra et al., 2012) and they can block aggregation in solution as well as on membranes. Considering the importance of the N17 for aggregation in both cases, targeting the N17 by binders or post-translational modifications might therefore be a powerful strategy for blocking multiple aggregation pathways in the cell.

STAR METHODS

LEAD CONTACT AND MATERIALS AVAILABILITY

Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Ralf Langen (langen@usc.edu).

EXPERIMENTAL MODEL AND SUBJECT DETAILS

Bacterial Culture

Httex1(Q25) and Httex1(Q46) proteins were expressed and purified from Escherichia coli BL21(DE3) competent cells (Agilent) in LB media, by IPTG induction.

METHOD DETAILS

Preparation of R1-labeled Httex1 Derivatives

The native amino acid sequence of human Httex1 fragment contains no cysteine. Cysteine mutations as well as phosphomimetic (S13D/S16D) mutations were introduced either by Q5® Site-Directed Mutagenesis Kit (E0554S, NEB) or NEBuilder® HiFi DNA Assembly Cloning Kit (E5520S, NEB) based on templates previously synthesized by GenScript. All plasmids were verified by DNA sequencing. Monomeric Httex1 fragments were expressed from pET-28b(+) plasmid as a Trx-Httex1 fusion protein, which had a thioredoxin (Trx) fused to the N-terminus of Httex1. This Trx-Httex1 fusion protein was purified, spin-labeled and cleaved to separate Httex1 from Trx using a previously described protocol (Pandey et al., 2018). The only modification was that the present Trx-Httex1 contained a 6×His-tag at the N-terminus of the Trx tag rather than the C-terminus of Httex1. This was done to ensure that the 6xHis-tag could later be cleaved off together with the thioredoxin fusion partner (Pandey et al., 2018). As in the prior study, the resulting Httex1 was lyophilized after the final purification step, dissolved using 0.5% TFA (v/v) in methanol and dried completely under gentle nitrogen flow. The resulting protein film was solubilized either directly with buffer or by SUVs prepared in the same buffer.

Preparation of Small and Large Unilamellar Vesicles

1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylcholine (POPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-L-serine] (POPS) were purchased as chloroform solution from Avanti Polar Lipids, Inc (#850457C, #840034C). SUVs and LUVs of varying POPS to POPC ratios were prepared in the respective buffers used for studying Httex1 interaction. For all vesicle preparations, lipids were first mixed and dried under a gentle nitrogen flow to form a thin layer of lipid film and placed at room temperature in a desiccator overnight. For preparation of SUVs, lipid films were resuspended in buffer and sonicated until clear. LUVs (25% POPS and 75% POPC) were made by freeze-thawing the solubilized lipid suspension for 10 cycles, followed by 21 times of back-and-forth extrusions through either 200 nm or 1000 nm Nucleopore Track-Etch Membrane (Whatman® #800281, #800319) using glass syringes. For testing the effect of negatively charged lipids on Httex1 membrane binding, SUVs were prepared containing 0, 5, 10, 25 and 50% POPS and the remaining lipids were POPC. For all other experiments, 25% POPS and 75% POPC were used. The buffer was 20 mM Hepes, 100 mM NaCl, pH 7.4 except in the following cases: For NiEDDA accessibility measurement, SUVs were prepared with an additional 3 mM NiEDDA. For testing the effect of ionic strength on Httex1 membrane binding, SUVs were prepared in 20 mM Hepes, pH 7.4 containing increasing concentrations of NaCl (0, 100, 500, 1000 mM). SUVs for circular dichroism were prepared in 20 mM sodium phosphate, pH 7.0. No additional salt was added to avoid high absorbance in the UV. For consistency, the same SUVs were used for EPR lipid titration and EPR kinetics experiments.

Transmission Electron Microscopy

Electron microscopy was used to verify SUV size and shape and to detect potential protein aggregates. 10 mM SUVs were pre-spun for 10 minutes at 15,000 rpm. The vesicle suspension (supernatant) was used to solubilize Httex1(Q25) to a concentration of 7.5 μM. SUVs alone as well as the SUV/Httex1(Q25) mixture were then incubated at room temperature. After seven hours, the samples were transferred to 150 mesh carbon coated copper EM grids (#FCF-150-Cu, Electron Microscopy Sciences). The protein-containing sample was incubated for 10 minutes while the SUV only sample was incubated for 1 hour to compensate for the poor absorption in the absence of proteins. Both samples were subsequently stained with 2% uranyl acetate for 1 minute. The grids were then examined under JEOL JEM-1400 transmission electron microscope operated at 100 kV and detected using a Gatan ORIUS™ CCD camera.

EPR Measurements

X-band continuous-wave EPR spectra for all R1-labeled Httex1(Q25) derivatives were recorded in a Bruker EMX spectrophotometer fitted with a Bruker ER4119HS resonator. For spectra of Httex1 in the absence of lipids, 15 μM protein was solubilized in buffer (20 mM Hepes, 100 mM NaCl, pH 7.4). For spectra of Httex1 in the presence of lipids, Httex1 film was solubilized directly in 10 mM SUVs to a final protein concentration of 7.5 μM. The vesicle-bound proteins were then harvested by ultracentrifugation at 60,000 rpm, 25 °C for 30 minutes (Beckman™ TLX Ultracentrifuge). Samples were then loaded into round borosilicate glass capillaries (#CV6084-B-100, VitroCom Inc.) and the respective X-band EPR spectra were recorded at room temperature using a scan width of 100 G and a modulation amplitude of 1.5 G.

EPR was also used to quantify Httex1 membrane binding based on the distinctively different spectra for the bound and unbound states of the 21R1 Httex1(Q25) derivative. For these experiments, Httex1 and membranes were not harvested by ultracentrifugation. Rather, the R1-labeled proteins were directly solubilized in the indicated vesicles using Httex1 concentrations of 15 μM. All spectra were normalized by double integration to the same number of spins. The assignment of percent binding was calibrated using the normalized spectra of the fully bound and fully unbound states as reference. This was possible because all spectra were the composite mixture of varying amounts of the same spectral components arising from bound and unbound protein. Binding percentage was then estimated based on the linear amplitude decay of the normalized EPR spectra that occurs as membrane binding immobilizes 21R1. Here the spectral amplitude of the free protein (in the absence of lipid) corresponded to 0% membrane binding while the spectral amplitude of the fully bound form corresponded to 100% binding.

O2 and NiEDDA accessibility measurements were performed using EPR power saturation. Samples were prepared using ultracentrifugation as described above to remove any potential unbound protein and loaded into a TPX capillary. EPR spectra at different incident powers were recorded using a Bruker EMX spectrophotometer fitted with a Bruker dielectric resonator (ER4123D) at room temperature (Altenbach et al., 1994). For O2 accessibility measurements, the power dependence of the EPR spectra was first determined in the presence of O2 from air. Then the same sample was exposed to a continuous flow of N2 and the EPR experiments were repeated. For NiEDDA accessibility measurements, power saturation experiments were performed in the presence of 3 mM NiEDDA and N2 flow. The respective power saturation data were converted into O2 and NiEDDA accessibility parameters, and Φ values according to previously published methods (Altenbach et al., 1994).

To convert the Φ values into immersion depth, calibration experiments were performed using SUVs containing 1% spin-labeled lipids (Avanti Polar Lipids): 1-palmitoyl-2-oleoyl-sn-glycero-3-phospho(tempo)choline (#810609C), 1-palmitoyl-2-stearoyl-(n-DOXYL)-sn-glycero-3-phosphocholine (n = 5, 7, 10; #810601C, #810602C, #810603C). The relationship between membrane immersion depth and the depth parameter Φ was established previously (Altenbach et al., 1994) as d[Å] = aΦ + b. We found that under the current experiment conditions a = 9.6 and b = − 9.7.

EPR kinetics experiments were performed by solubilizing 15 μM 35R1 derivative of Httex1(Q46) with or without S13D/S16D mutations either directly with buffer or with 375 μM SUV lipids. The EPR spectra were then recorded at a time interval of 15 minutes for a period of 15 hours. The spectral amplitudes were normalized to that of the initial t = 0 spectrum, which was set to 1.

ODNP Measurements

Six singly R1-labeled Httex1(Q25) derivatives (11R1, 22R1, 30R1, 42R1, 60R1 and 90R1) were selected for ODNP measurement. 7.5 μM protein was mixed with 10 mM SUVs and then concentrated using 100 kDa centrifugal filters (Amicon Ultra #UFC510096) followed by buffer washing (× 3 times) to ensure removal of potential unbound protein. Samples of 3.5 μl volume were filled into a 0.6 mm i.d and 0.84 mm o.d. quartz capillary to acquire ODNP data, as described previously (Franck et al., 2013; Kaminker et al., 2015). A “pass through” NMR probe that fits inside a 3 mm i.d., 6 mm o.d. quartz tube was used for the measurements. The quartz tube along with the NMR probe and sample was inserted into a microwave cavity (ER 4119HS-LC, Bruker Biospin), while ensuring that the position of the sample within the cavity was reproducible between measurements. The ODNP experiments were performed using a Bruker EMX CW-EPR and a Bruker Avance III NMR console. The samples were sealed in a capillary with Critoseal on top and beeswax on the bottom. The sample was irradiated with up to 6 W of microwaves at the EPR resonant frequency of the R1 label (Armstrong et al., 2008; Franck et al., 2013). The power to the microwave resonator was sampled by a 20 dB directional coupler (Narda 4015C-20) and further attenuated by a 10 dB attenuator (Narda 4778-10) and Coax microwave cable which was approximated to have a loss of 2 dB. The sampled power was measured with a Gigatronics 8541C power meter and 80401A power sensor. The magnetic field was set on resonance at the central electron hyperfine transition, here at 9.8 GHz. The R1 label concentration of each sample was determined from the double integral of its CW EPR spectrum. The ODNP-derived electron-proton spin cross relaxivity, kσ, was normalized to the sample concentration derived from spin counting per integration of the CW-EPR spectrum. The ODNP data yielding the kσ values are shown in the table of Figure S4B. All ODNP data were measured in triplicates, constituting the error bars in the ODNP data derived kσ values. The theory behind the physical basis of kσ values, as well as the data analysis to obtain kσ values and retardation factors from ODNP data is described in the literature (Barnes et al., 2017).

Circular Dichroism

Circular dichroism (CD) measurements were performed at room temperature using a Jasco J-810 spectropolarimeter in a 0.1 mm quartz cuvette every 0.5 nm at a 50-nm/min scan rate. The scan wavelength range was from 200 nm to 260 nm. Experiments were done in triplicates and all spectra were baseline subtracted. Data were converted into mean residual ellipticity [θ]MRE222nm.

QUANTIFICATION AND STATISTICAL ANALYSIS

The statistical analysis and the software used can be found in the relevant sections of the methods and the figure legends.

DATA AND CODE AVAILABILITY

Spectra for EPR and CD as well as ODNP data are available as Supplemental Information. This study did not generate new software.

Supplementary Material

2

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER
Bacterial and Virus Strains
E.Coli BL21(DE3) Agilent Cat#200131
Chemicals, Peptides, and Recombinant Proteins
1-Palmitoyl-2-oleoyl-sn-glycero-3-phosphatidylcholine (POPC) Avanti Polar Lipids Cat#850457C
1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-L-serine] (POPS) Avanti Polar Lipids Cat#840034C
1-palmitoyl-2-oleoyl-sn-glycero-3-phospho(tempo)choline Avanti Polar Lipids Cat#810609C
1-palmitoyl-2-stearoyl-(5-DOXYL)-sn-glycero-3-phosphocholine Avanti Polar Lipids Cat#810601C
1-palmitoyl-2-stearoyl-(7-DOXYL)-sn-glycero-3-phosphocholine Avanti Polar Lipids Cat#810602C
1-palmitoyl-2-stearoyl-(10-DOXYL)-sn-glycero-3-phosphocholine Avanti Polar Lipids Cat#810603C
MTSL Toronto Research Chemicals Cat#O875000
IPTG Biopioneer Cat#C0012
LB Broth Powder Biopioneer Cat#CMLP
Imidazole Sigma-Adrich Cat#56750
Tris Base Fisher BioReagents Cat#BP152-5
DTT Biopioneer Cat#C0040
Sodium Chloride Fisher BioReagents Cat#BP358-10
HEPES, free acid Biopioneer Cat#C0113
Sodium Phosphate Monobasic Anhydrous Sigma-Adrich Cat#S3139
Sodium Phosphate Dibasic Heptahydrate Sigma-Adrich Cat#S9390
Urea Sigma-Adrich Cat#U5128
Trifluoroacetic acid Sigma-Adrich Cat#T6508
Methanol Sigma-Adrich Cat#322415
Acetonitrile Fisher Chemical Cat#A998
Kanamycin Sulfate Biopioneer Cat#C0031
EKMax™ Enterokinase Invitrogen Cat#E18002
Uranyl Acetate Electron Microscopy Sciences Cat#541-09-3
Nickel (II) hydroxide Sigma-Adrich Cat#283622
EDDA Sigma-Adrich Cat#158186
Critical Commercial Assays
Q5® Site-Directed Mutagenesis Kit New England Biolabs Cat#E0554S
NEBuilder® HiFi DNA Assembly Cloning Kit New England Biolabs Cat#E5520S
Recombinant DNA
N6His-Trx-Httex1(Q25)-C2-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C3-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C4-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C5-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C6-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C7-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C8-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C9-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C10-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C11-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C12-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C13-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C14-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C15-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C16-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C17-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C19-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C21-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C21-S13DS16D-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C22-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C23-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C24-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C25-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C30-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C35-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C42-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C55-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C60-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C80-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q25)-C90-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q46)-C35-pET28b(+) This paper N/A
N6His-Trx-Httex1(Q46)-C35-S13DS16D-pET28b(+) This paper N/A
Software and Algorithms
Chimera UCSF Chimera https://www.cgl.ucsf.edu/chimera/
Prism 7 Graphpad https://www.graphpad.com/scientific-software/prism/
Other
His60 Ni Superflow Resin Takara Cat#635662
HiTrap Q XL Column GE Healthcare Cat#17515901
Jupiter C4 LC Column Phenomenex Cat#00G-4169-E0
200 nm Nucleopore Track-Etch Membrane Whatman Cat#800281
1000 nm Nucleopore Track-Etch Membrane Whatman Cat#800319
100 kDa Centrifugal Filters Amicon Ultra Cat#UFC510096
150 Mesh Carbon Coated Copper EM Grids Electron Microscopy Sciences Cat#FCF-150-Cu
Open in a new tab

HIGHLIGHTS.

  • Httex1 forms a short, N-terminal amphipathic α-helix upon membrane interaction

  • The α-helix is sensitive to membrane curvature and electrostatic interactions

  • Membrane-mediated structuring extends into the N-terminus of the polyQ domain

  • Phosphomimetic mutations inhibit membrane binding and membrane-mediated aggregation

ACKNOWLEDGEMENTS

This study was supported by the National Institutes of Health (NIH) (R01NS084345 to R.L.) and (R01GM116128 to S.H.). This work made use of shared facilities of the UCSB MRSEC (NSF DMR 1720256), a member of the Materials Research Facilities Network (www.mrfn.org). Support for the ODNP studies was provided by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany’s Excellence Strategy – EXC-2033 – Project number 390677874.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

DECLARATION OF INTERESTS

The authors declare no competing interests.

REFERENCES

  1. Aditi K, Shakarad MN, and Agrawal N (2016). Altered lipid metabolism in Drosophila model of Huntington's disease. Sci Rep 6, 31411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Albin RL, Reiner A, Anderson KD, Dure L.S.t., Handelin B, Balfour R, Whetsell WO Jr., Penney JB, and Young AB (1992). Preferential loss of striato-external pallidal projection neurons in presymptomatic Huntington's disease. Ann Neurol 31, 425–430. [DOI] [PubMed] [Google Scholar]
  3. Altenbach C, Greenhalgh DA, Khorana HG, and Hubbell WL (1994). A collision gradient method to determine the immersion depth of nitroxides in lipid bilayers: application to spin-labeled mutants of bacteriorhodopsin. Proc Natl Acad Sci U S A 91, 1667–1671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Armstrong BD, Lingwood MD, McCarney ER, Brown ER, Blumler P, and Han S (2008). Portable X-band system for solution state dynamic nuclear polarization. J Magn Reson 191, 273–281. [DOI] [PubMed] [Google Scholar]
  5. Atwal RS, and Truant R (2008). A stress sensitive ER membrane-association domain in Huntingtin protein defines a potential role for Huntingtin in the regulation of autophagy. Autophagy 4, 91–93. [DOI] [PubMed] [Google Scholar]
  6. Atwal RS, Xia J, Pinchev D, Taylor J, Epand RM, and Truant R (2007). Huntingtin has a membrane association signal that can modulate huntingtin aggregation, nuclear entry and toxicity. Hum Mol Genet 16, 2600–2615. [DOI] [PubMed] [Google Scholar]
  7. Barnes R, Sun S, Fichou Y, Dahlquist FW, Heyden M, and Han S (2017). Spatially Heterogeneous Surface Water Diffusivity around Structured Protein Surfaces at Equilibrium. J Am Chem Soc 139, 17890–17901. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Block RC, Dorsey ER, Beck CA, Brenna JT, and Shoulson I (2010). Altered cholesterol and fatty acid metabolism in Huntington disease. J Clin Lipidol 4, 17–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Bortolus M, Tombolato F, Tessari I, Bisaglia M, Mammi S, Bubacco L, Ferrarini A, and Maniero AL (2008). Broken helix in vesicle and micelle-bound alpha-synuclein: insights from site-directed spin labeling-EPR experiments and MD simulations. J Am Chem Soc 130, 6690–6691. [DOI] [PubMed] [Google Scholar]
  10. Bravo-Arredondo JM, Kegulian NC, Schmidt T, Pandey NK, Situ AJ, Ulmer TS, and Langen R (2018). The Folding Equilibrium of Huntingtin Exon-1 Monomer Depends on its Polyglutamine Tract. J Biol Chem. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Bugg CW, Isas JM, Fischer T, Patterson PH, and Langen R (2012). Structural features and domain organization of huntingtin fibrils. J Biol Chem 287, 31739–31746. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Burke KA, Kauffman KJ, Umbaugh CS, Frey SL, and Legleiter J (2013). The interaction of polyglutamine peptides with lipid membranes is regulated by flanking sequences associated with huntingtin. J Biol Chem 288, 14993–15005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Campelo F, McMahon HT, and Kozlov MM (2008). The hydrophobic insertion mechanism of membrane curvature generation by proteins. Biophys J 95, 2325–2339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Ceccon A, Clore GM, and Tugarinov V (2018a). Decorrelating Kinetic and Relaxation Parameters in Exchange Saturation Transfer NMR: A Case Study of N-Terminal Huntingtin Peptides Binding to Unilamellar Lipid Vesicles. J Phys Chem B. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Ceccon A, Schmidt T, Tugarinov V, Kotler SA, Schwieters CD, and Clore GM (2018b). Interaction of Huntingtin Exon-1 peptides with lipid-based micellar nanoparticles probed by solution NMR and Q-band pulsed EPR. J Am Chem Soc. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Chaibva M, Burke KA, and Legleiter J (2014). Curvature enhances binding and aggregation of huntingtin at lipid membranes. Biochemistry 53, 2355–2365. [DOI] [PubMed] [Google Scholar]
  17. Chaibva M, Jawahery S, Pilkington A.W.t., Arndt JR, Sarver O, Valentine S, Matysiak S, and Legleiter J (2016). Acetylation within the First 17 Residues of Huntingtin Exon 1 Alters Aggregation and Lipid Binding. Biophys J 111, 349–362. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Chan AW, Xu Y, Jiang J, Rahim T, Zhao D, Kocerha J, Chi T, Moran S, Engelhardt H, Larkin K et al. (2014). A two years longitudinal study of a transgenic Huntington disease monkey. BMC Neurosci 15, 36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Cheng CY, Varkey J, Ambroso MR, Langen R, and Han S (2013). Hydration dynamics as an intrinsic ruler for refining protein structure at lipid membrane interfaces. Proc Natl Acad Sci U S A 110, 16838–16843. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Davies SW, Turmaine M, Cozens BA, DiFiglia M, Sharp AH, Ross CA, Scherzinger E, Wanker EE, Mangiarini L, and Bates GP (1997). Formation of neuronal intranuclear inclusions underlies the neurological dysfunction in mice transgenic for the HD mutation. Cell 90, 537–548. [DOI] [PubMed] [Google Scholar]
  21. Deguire SM, Ruggeri FS, Fares MB, Chiki A, Cendrowska U, Dietler G, and Lashuel HA (2018). N-terminal Huntingtin (Htt) phosphorylation is a molecular switch regulating Htt aggregation, helical conformation, internalization, and nuclear targeting. J Biol Chem. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Di Pardo A, Maglione V, Alpaugh M, Horkey M, Atwal RS, Sassone J, Ciammola A, Steffan JS, Fouad K, Truant R et al. (2012). Ganglioside GM1 induces phosphorylation of mutant huntingtin and restores normal motor behavior in Huntington disease mice. Proc Natl Acad Sci U S A 109, 3528–3533. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. DiGiovanni LF, Mocle AJ, Xia J, and Truant R (2016). Huntingtin N17 domain is a reactive oxygen species sensor regulating huntingtin phosphorylation and localization. Hum Mol Genet 25, 3937–3945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Drescher M, Veldhuis G, van Rooijen BD, Milikisyants S, Subramaniam V, and Huber M (2008). Antiparallel arrangement of the helices of vesicle-bound alpha-synuclein. J Am Chem Soc 130, 7796–7797. [DOI] [PubMed] [Google Scholar]
  25. Duennwald ML, and Lindquist S (2008). Impaired ERAD and ER stress are early and specific events in polyglutamine toxicity. Genes Dev 22, 3308–3319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Fisette O, Paslack C, Barnes R, Isas JM, Langen R, Heyden M, Han S, and Schafer LV (2016). Hydration Dynamics of a Peripheral Membrane Protein. J Am Chem Soc 138, 11526–11535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Franck JM, Pavlova A, Scott JA, and Han S (2013). Quantitative cw Overhauser effect dynamic nuclear polarization for the analysis of local water dynamics. Prog Nucl Magn Reson Spectrosc 74, 33–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Gafni J, Hermel E, Young JE, Wellington CL, Hayden MR, and Ellerby LM (2004). Inhibition of calpain cleavage of huntingtin reduces toxicity: accumulation of calpain/caspase fragments in the nucleus. J Biol Chem 279, 20211–20220. [DOI] [PubMed] [Google Scholar]
  29. Gao YG, Yan XZ, Song AX, Chang YG, Gao XC, Jiang N, Zhang Q, and Hu HY (2006). Structural insights into the specific binding of huntingtin proline-rich region with the SH3 and WW domains. Structure 14, 1755–1765. [DOI] [PubMed] [Google Scholar]
  30. Gauthier LR, Charrin BC, Borrell-Pages M, Dompierre JP, Rangone H, Cordelieres FP, De Mey J, MacDonald ME, Lessmann V, Humbert S et al. (2004). Huntingtin controls neurotrophic support and survival of neurons by enhancing BDNF vesicular transport along microtubules. Cell 118, 127–138. [DOI] [PubMed] [Google Scholar]
  31. Gelman A, Rawet-Slobodkin M, and Elazar Z (2015). Huntingtin facilitates selective autophagy. Nat Cell Biol 17, 214–215. [DOI] [PubMed] [Google Scholar]
  32. Georgieva ER, Ramlall TF, Borbat PP, Freed JH, and Eliezer D (2008). Membrane-bound alpha-synuclein forms an extended helix: long-distance pulsed ESR measurements using vesicles, bicelles, and rodlike micelles. J Am Chem Soc 130, 12856–12857. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Gomez-Tortosa E, del Barrio A, Garcia Ruiz PJ, Pernaute RS, Benitez J, Barroso A, Jimenez FJ, and Garcia Yebenes J (1998). Severity of cognitive impairment in juvenile and late-onset Huntington disease. Arch Neurol 55, 835–843. [DOI] [PubMed] [Google Scholar]
  34. Graham RK, Deng Y, Slow EJ, Haigh B, Bissada N, Lu G, Pearson J, Shehadeh J, Bertram L, Murphy Z, et al. (2006). Cleavage at the caspase-6 site is required for neuronal dysfunction and degeneration due to mutant huntingtin. Cell 125, 1179–1191. [DOI] [PubMed] [Google Scholar]
  35. Gu X, Greiner ER, Mishra R, Kodali R, Osmand A, Finkbeiner S, Steffan JS, Thompson LM, Wetzel R, and Yang XW (2009). Serines 13 and 16 are critical determinants of full-length human mutant huntingtin induced disease pathogenesis in HD mice. Neuron 64, 828–840. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Gunawardena S, Her LS, Brusch RG, Laymon RA, Niesman IR, Gordesky-Gold B, Sintasath L, Bonini NM, and Goldstein LS (2003). Disruption of axonal transport by loss of huntingtin or expression of pathogenic polyQ proteins in Drosophila. Neuron 40, 25–40. [DOI] [PubMed] [Google Scholar]
  37. Hubbell WL, Gross A, Langen R, and Lietzow MA (1998). Recent advances in site-directed spin labeling of proteins. Curr Opin Struct Biol 8, 649–656. [DOI] [PubMed] [Google Scholar]
  38. Isas JM, Langen A, Isas MC, Pandey NK, and Siemer AB (2017). Formation and Structure of Wild Type Huntingtin Exon-1 Fibrils. Biochemistry 56, 3579–3586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Iuchi S, Hoffner G, Verbeke P, Djian P, and Green H (2003). Oligomeric and polymeric aggregates formed by proteins containing expanded polyglutamine. Proc Natl Acad Sci U S A 100, 2409–2414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Jao CC, Hegde BG, Chen J, Haworth IS, and Langen R (2008). Structure of membrane-bound alpha-synuclein from site-directed spin labeling and computational refinement. Proc Natl Acad Sci U S A 105, 19666–19671. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Jayasinghe SA, and Langen R (2007). Membrane interaction of islet amyloid polypeptide. Biochim Biophys Acta 1768, 2002–2009. [DOI] [PubMed] [Google Scholar]
  42. Kaminker I, Barnes R, and Han S (2015). Overhauser Dynamic Nuclear Polarization Studies on Local Water Dynamics. Methods Enzymol 564, 457–483. [DOI] [PubMed] [Google Scholar]
  43. Kay BK, Williamson MP, and Sudol M (2000). The importance of being proline: the interaction of proline-rich motifs in signaling proteins with their cognate domains. FASEB J 14, 231–241. [PubMed] [Google Scholar]
  44. Kim YJ, Yi Y, Sapp E, Wang Y, Cuiffo B, Kegel KB, Qin ZH, Aronin N, and DiFiglia M (2001). Caspase 3-cleaved N-terminal fragments of wild-type and mutant huntingtin are present in normal and Huntington's disease brains, associate with membranes, and undergo calpain-dependent proteolysis. Proc Natl Acad Sci U S A 98, 12784–12789. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kirkwood SC, Su JL, Conneally P, and Foroud T (2001). Progression of symptoms in the early and middle stages of Huntington disease. Arch Neurol 58, 273–278. [DOI] [PubMed] [Google Scholar]
  46. Kreilaus F, Spiro AS, McLean CA, Garner B, and Jenner AM (2016). Evidence for altered cholesterol metabolism in Huntington's disease post mortem brain tissue. Neuropathol Appl Neurobiol 42, 535–546. [DOI] [PubMed] [Google Scholar]
  47. Lee CY, Cantle JP, and Yang XW (2013). Genetic manipulations of mutant huntingtin in mice: new insights into Huntington's disease pathogenesis. FEBS J 280, 4382–4394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Lee WC, Yoshihara M, and Littleton JT (2004). Cytoplasmic aggregates trap polyglutamine-containing proteins and block axonal transport in a Drosophila model of Huntington's disease. Proc Natl Acad Sci U S A 101, 3224–3229. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Levy GR, Shen K, Gavrilov Y, Smith PES, Levy Y, Chan R, Frydman J, and Frydman L (2018). Huntingtin's N-Terminus Rearrangements in the Presence of Membranes: A Joint Spectroscopic and Computational Perspective. ACS Chem Neurosci. [DOI] [PubMed] [Google Scholar]
  50. Liu KY, Shyu YC, Barbaro BA, Lin YT, Chern Y, Thompson LM, James Shen CK, and Marsh JL (2015). Disruption of the nuclear membrane by perinuclear inclusions of mutant huntingtin causes cell-cycle re-entry and striatal cell death in mouse and cell models of Huntington's disease. Hum Mol Genet 24, 1602–1616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Mangiarini L, Sathasivam K, Seller M, Cozens B, Harper A, Hetherington C, Lawton M, Trottier Y, Lehrach H, Davies SW, et al. (1996). Exon 1 of the HD gene with an expanded CAG repeat is sufficient to cause a progressive neurological phenotype in transgenic mice. Cell 87, 493–506. [DOI] [PubMed] [Google Scholar]
  52. Martin DD, Ladha S, Ehrnhoefer DE, and Hayden MR (2015). Autophagy in Huntington disease and huntingtin in autophagy. Trends Neurosci 38, 26–35. [DOI] [PubMed] [Google Scholar]
  53. Michalek M, Salnikov ES, and Bechinger B (2013a). Structure and topology of the huntingtin 1-17 membrane anchor by a combined solution and solid-state NMR approach. Biophys J 105, 699–710. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Michalek M, Salnikov ES, Werten S, and Bechinger B (2013b). Membrane interactions of the amphipathic amino terminus of huntingtin. Biochemistry 52, 847–858. [DOI] [PubMed] [Google Scholar]
  55. Mishra R, Hoop CL, Kodali R, Sahoo B, van der Wel PC, and Wetzel R (2012). Serine phosphorylation suppresses huntingtin amyloid accumulation by altering protein aggregation properties. J Mol Biol 424, 1–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Neueder A, Landles C, Ghosh R, Howland D, Myers RH, Faull RLM, Tabrizi SJ, and Bates GP (2017). The pathogenic exon 1 HTT protein is produced by incomplete splicing in Huntington's disease patients. Sci Rep 7, 1307. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Newcombe EA, Ruff KM, Sethi A, Ormsby AR, Ramdzan YM, Fox A, Purcell AW, Gooley PR, Pappu RV, and Hatters DM (2018). Tadpole-like Conformations of Huntingtin Exon 1 Are Characterized by Conformational Heterogeneity that Persists regardless of Polyglutamine Length. J Mol Biol. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Ochaba J, Lukacsovich T, Csikos G, Zheng S, Margulis J, Salazar L, Mao K, Lau AL, Yeung SY, Humbert S et al. (2014). Potential function for the Huntingtin protein as a scaffold for selective autophagy. Proc Natl Acad Sci U S A 111, 16889–16894. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Pal A, Severin F, Lommer B, Shevchenko A, and Zerial M (2006). Huntingtin-HAP40 complex is a novel Rab5 effector that regulates early endosome motility and is up-regulated in Huntington's disease. J Cell Biol 172, 605–618. [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Pandey NK, Isas JM, Rawat A, Lee RV, Langen J, Pandey P, and Langen R (2018). The 17-residue-long N terminus in huntingtin controls stepwise aggregation in solution and on membranes via different mechanisms. J Biol Chem 293, 2597–2605. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Paulsen JS (2011). Cognitive impairment in Huntington disease: diagnosis and treatment. Curr Neurol Neurosci Rep 11, 474–483. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Rawat A, Langen R, and Varkey J (2018). Membranes as modulators of amyloid protein misfolding and target of toxicity. Biochim Biophys Acta Biomembr. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Reiner A, Albin RL, Anderson KD, D'Amato CJ, Penney JB, and Young AB (1988). Differential loss of striatal projection neurons in Huntington disease. Proc Natl Acad Sci U S A 85, 5733–5737. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Richfield EK, Maguire-Zeiss KA, Vonkeman HE, and Voorn P (1995). Preferential loss of preproenkephalin versus preprotachykinin neurons from the striatum of Huntington's disease patients. Ann Neurol 38, 852–861. [DOI] [PubMed] [Google Scholar]
  65. Rockabrand E, Slepko N, Pantalone A, Nukala VN, Kazantsev A, Marsh JL, Sullivan PG, Steffan JS, Sensi SL, and Thompson LM (2007). The first 17 amino acids of Huntingtin modulate its sub-cellular localization, aggregation and effects on calcium homeostasis. Hum Mol Genet 16, 61–77. [DOI] [PubMed] [Google Scholar]
  66. Sameni S, Malacrida L, Tan Z, and Digman MA (2018). Alteration in Fluidity of Cell Plasma Membrane in Huntington Disease Revealed by Spectral Phasor Analysis. Sci Rep 8, 734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Sathasivam K, Neueder A, Gipson TA, Landles C, Benjamin AC, Bondulich MK, Smith DL, Faull RL, Roos RA, Howland D et al. (2013). Aberrant splicing of HTT generates the pathogenic exon 1 protein in Huntington disease. Proc Natl Acad Sci U S A 110, 2366–2370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  68. Scherzinger E, Lurz R, Turmaine M, Mangiarini L, Hollenbach B, Hasenbank R, Bates GP, Davies SW, Lehrach H, and Wanker EE (1997). Huntingtin-encoded polyglutamine expansions form amyloid-like protein aggregates in vitro and in vivo. Cell 90, 549–558. [DOI] [PubMed] [Google Scholar]
  69. Sivanandam VN, Jayaraman M, Hoop CL, Kodali R, Wetzel R, and van der Wel PC (2011). The aggregation-enhancing huntingtin N-terminus is helical in amyloid fibrils. J Am Chem Soc 133, 4558–4566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  70. Smith MA, Brandt J, and Shadmehr R (2000). Motor disorder in Huntington's disease begins as a dysfunction in error feedback control. Nature 403, 544–549. [DOI] [PMC free article] [PubMed] [Google Scholar]
  71. Squitieri F, Cannella M, Sgarbi G, Maglione V, Falleni A, Lenzi P, Baracca A, Cislaghi G, Saft C, Ragona G et al. (2006). Severe ultrastructural mitochondrial changes in lymphoblasts homozygous for Huntington disease mutation. Mech Ageing Dev 127, 217–220. [DOI] [PubMed] [Google Scholar]
  72. Suopanki J, Gotz C, Lutsch G, Schiller J, Harjes P, Herrmann A, and Wanker EE (2006). Interaction of huntingtin fragments with brain membranes--clues to early dysfunction in Huntington's disease. J Neurochem 96, 870–884. [DOI] [PubMed] [Google Scholar]
  73. Ulmer TS, Bax A, Cole NB, and Nussbaum RL (2005). Structure and dynamics of micelle-bound human alpha-synuclein. J Biol Chem 280, 9595–9603. [DOI] [PubMed] [Google Scholar]
  74. Valenza M, Rigamonti D, Goffredo D, Zuccato C, Fenu S, Jamot L, Strand A, Tarditi A, Woodman B, Racchi M et al. (2005). Dysfunction of the cholesterol biosynthetic pathway in Huntington's disease. J Neurosci 25, 9932–9939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  75. Varkey J, Mizuno N, Hegde BG, Cheng N, Steven AC, and Langen R (2013). alpha-Synuclein oligomers with broken helical conformation form lipoprotein nanoparticles. J Biol Chem 288, 17620–17630. [DOI] [PMC free article] [PubMed] [Google Scholar]
  76. Velier J, Kim M, Schwarz C, Kim TW, Sapp E, Chase K, Aronin N, and DiFiglia M (1998). Wild-type and mutant huntingtins function in vesicle trafficking in the secretory and endocytic pathways. Exp Neurol 152, 34–40. [DOI] [PubMed] [Google Scholar]
  77. Wellington CL, Ellerby LM, Hackam AS, Margolis RL, Trifiro MA, Singaraja R, McCutcheon K, Salvesen GS, Propp SS, Bromm M et al. (1998). Caspase cleavage of gene products associated with triplet expansion disorders generates truncated fragments containing the polyglutamine tract. J Biol Chem 273, 9158–9167. [DOI] [PubMed] [Google Scholar]
  78. Wellington CL, Singaraja R, Ellerby L, Savill J, Roy S, Leavitt B, Cattaneo E, Hackam A, Sharp A, Thornberry N et al. (2000). Inhibiting caspase cleavage of huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem 275, 19831–19838. [DOI] [PubMed] [Google Scholar]
  79. Yang D, Wang CE, Zhao B, Li W, Ouyang Z, Liu Z, Yang H, Fan P, O'Neill A, Gu W et al. (2010). Expression of Huntington's disease protein results in apoptotic neurons in the brains of cloned transgenic pigs. Hum Mol Genet 19, 3983–3994. [DOI] [PMC free article] [PubMed] [Google Scholar]
  80. Zarrinpar A, Bhattacharyya RP, and Lim WA (2003). The structure and function of proline recognition domains. Sci STKE 2003, RE8. [DOI] [PubMed] [Google Scholar]
  81. Zheng Z, Li A, Holmes BB, Marasa JC, and Diamond MI (2013). An N-terminal nuclear export signal regulates trafficking and aggregation of Huntingtin (Htt) protein exon 1. J Biol Chem 288, 6063–6071. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

2
NIHMS1538879-supplement-2.pdf (1.7MB, pdf)

Data Availability Statement

Spectra for EPR and CD as well as ODNP data are available as Supplemental Information. This study did not generate new software.

RESOURCES