Lithium ions display weak interaction with amyloid-beta (Aβ) peptides and have minor effects on their aggregation 1 Lithium ions display weak interaction with amyloid-beta (Aβ) peptides and have minor effects on their aggregation Elina Berntsson1,2, Suman Paul1, Faraz Vosough1, Sabrina B. Sholts3, Jüri Jarvet1,4, Per M. Roos5,6, Andreas Barth1, Astrid Gräslund1, Sebastian K. T. S. Wärmländer1,* 1 Department of Biochemistry and Biophysics, Stockholm University, Sweden. 2 Department of Chemistry and Biotechnology, Tallinn University of Technology, Estonia; 3 Department of Anthropology, National Museum of Natural History, Smithsonian Institution, Washington, DC, USA. 4 The National Institute of Chemical Physics and Biophysics, Tallinn, Estonia. 5 Institute of Environmental Medicine, Karolinska Institutet, Stockholm, Sweden. 6 Department of Clinical Physiology, Capio St. Göran Hospital, Stockholm, Sweden. * Correspondence: seb@dbb.su.se; Tel.: +46 8 162444 Abstract: Alzheimer’s disease (AD) is an incurable disease and the main cause of age- related dementia worldwide, despite decades of research. Treatment of AD with lithium (Li) has showed promising results, but the underlying mechanism is unclear. The pathological hallmark of AD brains is deposition of amyloid plaques, consisting mainly of amyloid-β (Aβ) peptides aggregated into amyloid fibrils. The plaques contain also metal ions of e.g. Cu, Fe, and Zn, and such ions are known to interact with Aβ peptides and modulate their aggregation and toxicity. The interactions between Aβ peptides and Li+ ions have however not been well investigated. Here, we use a range of biophysical techniques to characterize in vitro interactions between Aβ peptides and Li+ ions. We show that Li+ ions display weak and non- specific interactions with Aβ peptides, and have minor effects on Aβ aggregation. These results indicate that possible beneficial effects of Li on AD pathology are not likely caused by direct interactions between Aβ peptides and Li+ ions. Key Words: Alzheimer’s disease; protein aggregation; Metal-protein binding; Neurodegeneration; Pharmaceutics Running Title: Li+ ions have minor effects on Aβ aggregation .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 2 INTRODUCTION Alzheimer’s disease (AD) is still an incurable disease and the main cause of age-related dementia worldwide (Querfurth & LaFerla, 2010; Prince et al., 2015; Frozza et al., 2018), despite decades of research on putative drugs (Luo et al., 2013; Wärmländer et al., 2013; Decker & Munoz-Torrero, 2016; Kisby et al., 2019). In addition to signs of neuroinflammation and oxidative stress (Agostinho et al., 2010; Al-Hilaly et al., 2013; Wang et al., 2014; Heppner et al., 2015; Regen et al., 2017), AD brains display characteristic lesions in the form of intracellular neurofibrillary tangles, consisting of aggregated hyperphosphorylated tau proteins (Goedert, 2018; Gibbons et al., 2019), and extracellular amyloid plaques, consisting mainly of insoluble fibrillar aggregates of amyloid-β (Aβ) peptides (Glenner & Wong, 1984; Querfurth & LaFerla, 2010). These Aβ fibrils and plaques are the end-product of an aggregation process (Querfurth & LaFerla, 2010; Luo et al., 2016; Selkoe & Hardy, 2016) that involves extra- and/or intracellular formation of intermediate, soluble, and likely neurotoxic Aβ oligomers (Luo et al., 2014; Selkoe & Hardy, 2016; Sengupta et al., 2016; Lee et al., 2017) that can spread from neuron to neuron via exosomes (Nath et al., 2012; Sardar Sinha et al., 2018). The Aβ peptides comprise 37-43 residues and are intrinsically disordered in aqueous solution. They have limited solubility in water due to the hydrophobicity of the central and C- terminal Aβ segments, which may fold into a hairpin conformation upon aggregation (Abelein et al., 2014; Baronio et al., 2019). The charged N-terminal segment is hydrophilic and readily interacts with cationic molecules and metal ions (Luo et al., 2013; Luo et al., 2014; Tiiman et al., 2016; Wallin et al., 2016; Wallin et al., 2017; Owen et al., 2019; Wallin et al., 2020), while the hydrophobic C-terminal segment can interact with membranes where Aβ may exert its toxicity (Österlund et al., 2018; Wärmländer et al., 2019). The interactions between Aβ and metal ions are of particular interest (Duce et al., 2011; Wärmländer et al., .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 3 2013; Mital et al., 2015; Wärmländer et al., 2019; Wallin et al., 2020), as altered metal concentrations indicative of metal dyshomeostasis are a prominent feature in the brains and fluids of AD patients (Wang et al., 2015; Szabo et al., 2016), and because AD plaques contain elevated amounts of metal ions of e.g. Cu, Fe, and Zn (Beauchemin & Kisilevsky, 1998; Lovell et al., 1998; Miller et al., 2006). Interestingly, although the role of metal ions in AD pathogenesis remains debated (Duce et al., 2011; Modgil et al., 2014; Chin-Chan et al., 2015; Mital et al., 2015; Adlard & Bush, 2018; Huat et al., 2019; Wärmländer et al., 2019), monovalent ions of the alkali metal lithium [i.e., Li+ ions] may provide beneficial effects to patients with neurodegenerative disorders such as amyotrophic lateral sclerosis (ALS) (Fornai et al., 2008; Morrison et al., 2013) or AD (Engel et al., 2008; Mauer et al., 2014; Sutherland & Duthie, 2015; Decker & Munoz- Torrero, 2016; Donix & Bauer, 2016; Morris & Berk, 2016; Kerr et al., 2018; Hampel et al., 2019; Kisby et al., 2019; Priebe & Kanzawa, 2020). Lithium salts are commonly used in psychiatric medication, even though it is not understood how the Li+ ions affect the molecular mechanisms underlying the psychiatric disorders (Dell'Osso et al., 2016). Unlike other pharmaceuticals, Li+ is widely non-selective in its biochemical effects, possibly due to its general propensity to inhibit the many enzymes that have magnesium as a cofactor (Ge & Jakobsson, 2018). Cell and animal studies have provided clues regarding how Li+ ions may affect the AD disease pathology (Nery et al., 2014; Sofola-Adesakin et al., 2014; Zhao et al., 2014; Budni et al., 2017; Habib et al., 2017; Cardillo et al., 2018; Kerr et al., 2018; Habib et al., 2019; Rocha et al., 2020; Wilson et al., 2020). Due to its ability to down-regulate translation, Li+ caused a reduction in protein synthesis and thus Aβ42 levels in an adult-onset Drosophila model of AD (Sofola-Adesakin et al., 2014). Li+ reduces Aβ production by affecting the processing/cleavage of the amyloid-β precursor protein (AβPP) in cells and mice, presumably .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 4 by down-regulating the levels of phosphorylated APP. A main target of Li+ is the glycogen synthase kinase 3-beta (GSK-3β) (Ryves & Harwood, 2001) which is implicated in AD pathogenesis (Caccamo et al., 2007; Forlenza et al., 2014). In AβPP-transgenic mice, reduced activation of the GSK-3β enzyme was associated with decreased levels of APP phosphorylation that resulted in decreased Aβ production (Rockenstein et al., 2007). One study on mice with traumatic brain injury reported that Li+-treatment improved spatial learning and reduced Aβ production, possibly by reducing the levels of both AβPP and the AβPP-cleaving enzyme BACE1 (Yu et al., 2012). More recent mice studies have reported that treatment with Li+ ions improved Aβ clearance from the brain (Pan et al., 2018), reduced oxidative stress levels (Xiang et al., 2020), improved spatial memory (Habib et al., 2019), and reduced the amounts of Aβ plaques and phosphorylated tau while also improving spatial memory (Liu et al., 2020). Only a few studies have however investigated how Li+ ions could affect the molecular events that appear to underlie AD pathology, such as Aβ aggregation. One study showed that increased ionic strength, i.e. 150 mM of NaF, NaCl, or LiCl, significantly accelerated the kinetics of Aβ amyloid formation, by promoting surface-catalyzed secondary nucleation reactions (Abelein et al., 2016). Another recent study used molecular dynamics simulations to find small but distinct differences in how the three monovalent Li+, N+, and K+ ions interact with Aβ oligomers (Huraskin & Horn, 2019). The therapeutic effect of Li+ on Aβ plaque quality and toxicity has been reported in mice, where Li+ treatment before pathology onset induced smaller plaques with higher Aβ compaction, reduced oligomeric-positive halo, and attenuated capacity to induce neuronal damage (Trujillo-Estrada et al., 2013). One hypothesis is that these neuroprotective effects of Li+ could be mediated by modifications of the plaque toxicity through the astrocytic release of heat shock proteins (Trujillo-Estrada et al., 2013). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 5 Here, we use a range of biophysical techniques to characterize the in vitro interactions between Li+ ions and Aβ peptides, and how such interactions affect the Aβ amyloid aggregation processes and fibril formation. MATERIALS AND METHODS Sample preparation Recombinant Aβ40 peptides were purchased from AlexoTech AB (Umeå, Sweden) in either unlabeled or uniformly 15N-labeled form. The lyophilized peptides were stored at -80 °C. Samples were dissolved to monomeric form immediately before each measurement. The peptides were first dissolved in 10 mM NaOH, and then sonicated in an ice-bath to avoid having pre-formed aggregates in the sample solutions. Next, the samples were diluted in 20 mM buffer of either sodium phosphate or MES (2-[N-morpholino]ethanesulfonic acid). All preparation steps were performed on a bed of ice, and the peptide concentration was determined by weight. LiCl salt was purchased from Merck & Co. Inc. (USA), and MES hydrate was purchased from Sigma-Aldrich (USA). Synthetic Aβ42 peptides were purchased from JPT Peptide Technologies (Germany) and used to prepare monomeric solutions via size exclusion chromatography. 1 mg of lyophilized Aβ42 powder was dissolved in 250 mL DMSO. A Sephadex G-250 HiTrap desalting column (GE Healthcare, Uppsala) was equilibrated with 5 mM NaOH solution (pH=12.3), and washed with 10-15 mL of 5 mM NaOD, pD=12.7 (Glasoe & Long, 1960) solution. The peptide solution in DMSO was applied to the column, followed by injection of 1.25 mL of 5 mM NaOD. Collection of peptide fractions in 5 mM NaOD on ice was started at a 1 mg/mL flow rate. Ten fractions of 1 mL volumes were collected in 1.5 mL Eppendorf tubes. The absorbance for each fraction at 280 nm was measured with a NanoDrop instrument (Eppendorf, Germany), and peptide concentrations were determined using a molar extinction .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 6 coefficient of 1280 M-1cm-1 for the single Tyr in Aβ42 (Edelhoch, 1967). The peptide fractions were flash-frozen in liquid nitrogen, covered with argon gas on top in 1.5 mL Eppendorf tubes, and stored at -80°C until used. Sodium dodecyl sulfate (SDS)-stabilized Aβ42 oligomers of two well-defined sizes (approximately tetramers and dodecamers) were prepared according to a previously published protocol (Barghorn et al., 2005), but in D2O, at 4-fold lower peptide concentration and without the original dilution step (Vosough & Barth, manuscript). The reaction mixtures (100 µM Aβ42 in PBS and containing 0.05 % or 0.2% SDS) were incubated together with 0-10 mM LiCl at 37 °C for 24 hours, and then flash- frozen in liquid nitrogen and stored at -20°C for later analysis. Thioflavin T kinetics A FLUOstar Omega microplate reader (BMG LABTECH, Germany) was used to monitor the effect of Li+ ions on Aβ aggregation kinetics, 20 µM monomeric Aβ40 peptides were incubated in 20 mM MES buffer, pH 7.35, together with different concentrations of LiCl (0, 20 μM, 200 μM, 2000 μM) and 50 μM Thioflavin T (ThT). ThT is a fluorescent benzothiazole dye, and its fluorescence intensity increases when bound to amyloid aggregates (Gade Malmos et al., 2017). Samples were placed in a 96-well plate where the sample volume in each well was 100 µL, four replicates per Li+ concentration were measured, the temperature was +37 °C, excitation of the ThT dye was at 440 nm, the ThT fluorescence emission at 480 nm was measured every five minutes, each five-minute cycle involved 140 seconds of shaking at 200 rpm, the samples were incubated for a total of 15 hours, and the assay was repeated three times. To derive parameters for the aggregation kinetics, the ThT fluorescence curves were fitted to the sigmoidal equation 1: (Eq. 1) .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 7 where F0 and F∞ are the intercepts of the initial and final fluorescence intensity baselines, m0 and m∞ are the slopes of the initial and final baselines, t½ is the time needed to reach halfway through the elongation phase (i.e., aggregation half-time), and τ is the elongation time constant (Gade Malmos et al., 2017). The apparent maximum rate constant, rmax, for the growth of fibrils is given by 1/τ. Tyrosine fluorescence quenching The binding affinity between Aβ40 peptides and Li + ions was evaluated from Cu2+/ Li+ binding competition experiments (Wallin et al., 2020). The affinity of the Cu2+·Aβ40 complex was measured via the quenching effect of Cu2+ ions on the intrinsic fluorescence of Y10, which is the only fluorophore in native Aβ peptides. The fluorescence emission intensity at 305 nm (excitation wavelength 276 nm) was recorded at 20 °C using a Jobin Yvon Horiba Fluorolog 3 fluorescence spectrophotometer (Longjumeau, France). The titrations were carried out by consecutive additions of 0.8 – 3.2 µL aliquots of either 2, 10, or 50 mM stock solutions of CuCl2 to 800 µL of 10 µM Aβ40 in 20 mM MES buffer, pH 7.35, in a quartz cuvette with 4 mm path length. After each addition of CuCl2 the solution was stirred for 30 seconds before recording fluorescence emission spectra. Copper titrations were conducted for Aβ40 samples both in the absence and the presence of 1 mM LiCl. The dissociation constant of the Cu2+·Aβ40 complex was determined by fitting the Cu 2+ titration data to equation 2: (Eq. 2) .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 8 where I0 is the initial fluorescence intensity without Cu 2+ ions, I∞ is the steady-state (saturated) intensity at the end of the titration series, [Aβ] is the peptide concentration, [Cu] is the concentration of added Cu2+ ions, KD is the dissociation constant of the Cu 2+·Aβ40 complex, and k is a constant accounting for the concentration-dependent quenching effect induced by free (non-bound) Cu2+ ions that may collide with the Y10 residue (Lindgren et al., 2013). This model assumes a single binding site. As no corrections for buffer conditions are made, i.e. in terms of possible interactions between the metal ions and the buffer, the calculated dissociation constant should be considered to be apparent. Atomic force microscopy imaging Samples of 20 µM Aβ40 in 5 mM MES buffer (total volume 100 µL, pH 7.35) with either 0, 20 µM, 200 µM or 2 mM LiCl were put in small Eppendorf tubes and incubated for 72 hours at 37 oC under continuous shaking at 300 rpm. A droplet (1 µL) of incubated solution was then placed on a fresh silicon wafer (Siegert Wafer GmbH, Germany) and left to dry for 2 minutes. Next, 10 uL of Milli-Q H2O was carefully added to the semi-dried sample droplet and soaked immediately with a lint-free wipe, to remove excess salts in a mild manner. The wafer was left to dry in a covered container to protect it from dust, and atomic force microscopy (AFM) images were recorded on the same day. A neaSNOM scattering- type near-field optical instrument (Neaspec GmbH, Germany) was used to collect the AFM images under tapping mode (Ω: 280 kHz, tapping amplitude 50-55 nm) using Pt/Ir-coated monolithic ARROW-NCPt Si tips (NanoAndMore GmbH, Germany) with tip radius <10 nm. Images were acquired on 2.5 x 2.5 µm scan-areas (200 x 200-pixel size) under optimal scan- speed (i.e., 2.5 ms/pixel). The recorded images were minimally processed using the Gwyddion software where a basic plane leveling was performed (Nečas & Klapetek, 2012). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 9 Nuclear magnetic resonance spectroscopy An Avance 700 MHz nuclear magnetic resonance (NMR) spectrometer (Bruker Inc., USA) equipped with a cryoprobe was used to investigate possible interactions between Li+ ions and monomeric Aβ40 peptides at the atomic level. 2D 1H-15N-HSQC spectra of 92.4 μM monomeric 15N-labeled Aβ40 peptides were recorded at 5 °C with 90/10 H2O/D2O, either in 20 mM MES buffer at pH 7.35 or in 1x PBS buffer (137 mM NaCl, 2.7 mM KCl, and 10 mM phosphate pH 7.4), before and after additions with LiCl. Diffusion measurements were performed on a sample of 55 μM unlabeled monomeric Aβ40 peptide in 20 mM sodium phosphate buffer, 100 % D2O, pD 7.5, at 5 °C, before and after additions with LiCl dissolved in D2O. The diffusion experiments employed pulsed field gradients (PFG:s) according to previously described methods (Danielsson et al., 2002), and methyl group signals between 0.7-0.4 ppm were integrated, evaluated, and corrected for the viscosity of D2O at 5 °C (Cho et al., 1999). All NMR data was processed with the Topspin version 3.6.2 software, and the HSQC crosspeak assignment for Aβ40 in buffer is known from previous studies (Danielsson et al., 2006). Circular dichroism spectroscopy Circular dichroism (CD) spectra of 20 μM Aβ40 peptides in 20 mM sodium phosphate buffer, pH 7.35, were recorded at 20 °C using a Chirascan CD spectrometer (Applied Photophysics, UK) and a quartz cuvette with an optical path length of 2 mm. Measurements were done between 190 − 250 nm, with a step size of 1 nm and a sampling time of 4 s per data point. First, a spectrum was recorded for Aβ40 alone. Next, micelles of 50 mM SDS were added to create a membrane-mimicking environment. Finally, LiCl was titrated to the sample in steps up to a concentration of 512 µM. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 10 Blue native polyacrylamide gel electrophoresis Homogeneous solutions of 100 µM Aβ42 oligomers prepared in presence and absence of 0 – 10 mM Li+ ions were analyzed with blue native polyacrylamide gel electrophoresis (BN- PAGE) using the Invitrogen system. 4-16% Bis-Tris Novex gels (ThermoFisher Scientific, USA) were loaded with 10 µL of Aβ42 oligomer samples alongside the Amersham High Molecular Weight calibration kit for native electrophoresis (GE Healthcare, USA). The gels were run at 4 °C using the electrophoresis system according to the Invitrogen instructions (ThermoFisher Scientific, USA), and then stained using the Pierce Silver Staining kit according to the instructions (ThermoFisher Scientific, USA). Infrared spectroscopy Fourier-transformed infrared (FTIR) spectra of Aβ42 oligomers were recorded in transmission mode on a Tensor 37 FTIR spectrometer (Bruker Optics, Germany) equipped with a sample shutter and a liquid nitrogen-cooled MCT detector. The unit was continuously purged with dry air during the measurements. 8-10 µL of the 80 µM Aβ42 oligomer samples, containing 0 – 10 mM LiCl, were put between two flat CaF2 discs separated by a 50 µm plastic spacer covered with vacuum grease at the periphery. The assembled discs were mounted in a holder inside the instrument’s sample chamber. The samples were allowed to sit for at least 15 minutes after closing the chamber lid, to avoid interference from CO2 and H2O vapor. FTIR spectra were recorded at room temperature in the 1900-800 cm-1 range, with 300 scans for both background and sample spectra, using a 6 mm aperture and a resolution of 2 cm-1. The light intensities above 2200 cm-1 and below 1500 cm-1 were blocked with respectively a germanium filter and a cellulose membrane (Baldassarre & Barth, 2014). The spectra were analyzed and plotted with the OPUS 5.5 software, and second derivatives were calculated with a 17 cm-1 smoothing range. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 11 RESULTS ThT fluorescence: influence of Li+ ions on Aβ40 aggregation The fluorescence intensity of the amyloid-marker molecule ThT was measured when 20 µM Aβ40 samples were incubated for 15 hours together with different concentrations of LiCl (Fig. 1). Fitting Eq. 1 to the ThT fluorescence curves yielded the kinetic parameters t1/2 (aggregation half-time) and rmax (maximum aggregation rate) (Fig. 1; Table 1). For 20 µM Aβ40 alone, the aggregation half-time is approximately 3.7 hours under the experimental conditions used, and the maximum aggregation rate is 0.5 hours-1 (Table 1). These kinetic parameters are not much affected by addition of LiCl in 1:1 or 10:1 Li+:Aβ ratios. At the Li+:Aβ ratio of 100:1, the rmax value remains largely unaffected while the aggregation half- time is increased to almost 5 hours (Fig 1; Table 1). The observation that a Li+:Aβ ratio of 100:1 is required to shift the ThT curve clearly shows that Li+ ions do not have a strong effect on the Aβ40 aggregation kinetics. AFM imaging: effects of Li+ ions on the morphology of Aβ40 aggregates AFM images (Fig. 2) were recorded for the aggregation products of 20 µM Aβ40 peptide, incubated for three days without or with LiCl. The control sample without Li+ displays long (> 2 µm) amyloid fibrils that are around 6 nm thick, together with small (< 2 nm) aggregate particles that may be protofibrils (Fig. 2A). The distribution and sizes of these aggregates are rather typical for Aβ40 aggregates formed in vitro (Luo et al., 2014). The Aβ40 samples incubated in the presence of different concentrations of Li+ ions display amyloid fibrils of similar size and shape, although these fibrils are more densely packed and they appear to be more numerous (Figs. 1B, 1C, 1D). Compared to the control sample, there are fewer small (< 2 nm) aggregate particles in the samples incubated together with Li+ ions in 10:1 and 100:1 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 12 Li+:Aβ ratios. This suggests that Li+ ions may induce some differences in the Aβ40 aggregation process. NMR spectroscopy: interactions between Li+ ions and Aβ40 monomers High-resolution liquid phase NMR experiments were conducted to investigate if residue- specific molecular interactions could be observed between Li+ ions and monomeric Aβ40 peptides. 2D 1H-15N-HSQC spectra showing the amide crosspeak region for 92.4 μM monomeric 15N-labeled Aβ40 peptides are presented in Fig. 3A, before and after addition of LiCl in 1:1, 1:10, and 1:100 Aβ:Li+ ratios in 20 mM MES buffer, 7.35. Addition of Li+ ions induces loss of signal intensity mainly for amide crosspeaks corresponding to residues in the N-terminal half of the peptide, indicating selective Li+ interactions in this region (Fig. 3B). The effects are clearly concentration-dependent. Because Li+ ions are not paramagnetic, this loss of signal intensity is arguably caused by chemical exchange related to structural rearrangements induced by the Li+ ions. As no chemical shift changes are observed for the crosspeak position (Fig. 3A), these Li+-induced secondary structures appear to be short-lived. Figs. 3C and 3D show the results of similar experiments carried out in 1x PBS buffer, i.e. 137 mM NaCl, 2.7 mM KCl, and 10 mM phosphate pH 7.4. Here, the Li+ ions induce virtually no changes in the crosspeak intensities, showing that the weak Li+/Aβ40 interactions observed in pure MES buffer (Figs. 3A,B) disappear when the buffer and ionic strength correspond to physiological conditions. Diffusion measurements were carried out for 55 μM Aβ40 peptides in D2O, before and after addition of LiCl in 1:1, 20:1, and 100:1 Li+:Aβ ratios. Addition of 1:1 Li+ produces an increase in the Aβ40 diffusion rate by around 4%, i.e. from 5.97·10 -11 m2/s to 6.23·10-11 m2/s (Figs. 4A and 4B). This somewhat faster diffusion is likely caused by the Aβ40 peptide adopting a slightly more compact structure in the presence of Li+ ions, an effect similar to .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 13 that previously reported for zinc ions (Abelein et al., 2015). Addition of even higher Li+ concentrations – 20 and 100 times the Aβ40 concentration – produces diffusion rates that are similar but a little bit lower than the diffusion rate measured for 1:1 Li+:Aβ40 ratio, i.e. respectively 6.19·10-11 m2/s and 6.15·10-11 m2/s (Figs. 4C and 4D), indicating that the effect of Li+ on the Aβ40 secondary structure and diffusion has been saturated. Fluorescence spectroscopy: Li+ binding affinity to the Aβ40 monomer Binding affinities for metal ions to Aβ peptides can often be measured via the quenching effect on the intrinsic fluorescence of Y10, the only fluorophore in native Aβ peptides. However, not all metal ions interfere with tyrosine fluorescence, and initial experiments showed that addition of Li+ ions does not affect the Aβ40 fluorescence. The binding affinity of Li+ ions to Aβ40 was therefore evaluated from binding competition experiments with Cu 2+ ions (Danielsson et al., 2007; Wallin et al., 2020), which induce much stronger tyrosine fluorescence quenching when bound to the peptide than when free in the solution (Lindgren et al., 2013). Fig. 5 shows the results of titrating CuCl2 to Aβ40, both in the absence (red circles) and in the presence (blue triangles) of 1 mM LiCl. Three titrations were carried out for each condition, producing apparent KD values for the Cu2+·Aβ40 complex of respectively 3.1 µM, 2.1 µM, and 5.1 µM without LiCl, i.e. on average 3.4 ± 1.6 µM, and 2.1 µM, 0.9 µM, and 0.8 µM with LiCl present, i.e. on average 1.3 ± 0.8 µM. The obtained values are in line with earlier fluorescence measurements of the Cu2+ binding affinity to the Aβ40 peptide, although this affinity is known to vary with the pH, the buffer, and other experimental conditions (Ghalebani et al., 2012; Alies et al., 2013). The difference between the average measured KD values is not significant at the 5% level with a two-tailed t-test, which shows that Li+ ions are not able to compete with Cu2+ for binding to Aβ. Thus, the Li+ binding affinity for Aβ40 is likely in the millimolar range, or weaker. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 14 CD spectroscopy: effects of Li+ ions on Aβ40 structure in SDS Although Aβ peptides are generally disordered in aqueous solutions, they adopt an α- helical secondary structure in membranes and membrane-mimicking environments such as SDS micelles (Tiiman et al., 2016; Österlund et al., 2018). Thus, the CD spectrum for Aβ40 in sodium phosphate buffer displays the characteristic minimum for random coil structure at 198 nm (Fig. 6). Addition of 50 mM SDS, which is well above the critical concentration for micelle formation (Österlund et al., 2018), induces an alpha-helical structure with characteristic minima around 208 and 222 nm. Titrating LiCl in concentrations up to 512 µM to the Aβ40 sample slightly increases the general CD intensity, but does not change the overall spectral shape – the minima remain at their respective positions. The intensity changes are not caused by dilution of the sample during the titration, as the added volumes are very small, and as dilution would not increase but rather decrease the CD intensity. The observed changes in CD intensity therefore suggest a small but distinct binding effect of LiCl ions. This binding effect appears to be much weaker than the structural rearrangements and Aβ coil-coil-interactions previously reported to be induced by Cu2+ ions (Tiiman et al., 2016). BN-PAGE: effects of Li+ ions on Aβ42 oligomer formation and stability Well-defined and SDS-stabilized Aβ42 oligomers were prepared in the presence of different amounts of LiCl. SDS treatment of Aβ42 peptides at low concentrations (≤ 7 mM) leads to formation of stable and homogeneous Aβ42 oligomers of certain sizes and conformations (Barghorn et al., 2005; Rangachari et al., 2007). As shown in Fig. 7, two sizes of Aβ42 oligomers are formed in presence of the two SDS concentrations. In 0.2% (6.9 mM) SDS, small oligomers with a molecular weight (MW) around 16-20 kDa are formed. These oligomers appear to contain a large fraction of tetramers (Vosough & Barth, manuscript). In .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 15 0.05% (1.7 mM) SDS, larger oligomers with MWs around 55-60 kD are formed (Barghorn et al., 2005). These larger oligomers, which most likely contain twelve Aβ42 monomers, display a globular morphology and are therefore sometimes called globulomers (Barghorn et al., 2005). All oligomers were analyzed by BN-PAGE instead of by SDS-PAGE to avoid disruption of the non-cross linked Aβ42 oligomers by the high (>1%) SDS concentrations used in SDS- PAGE (Bitan et al., 2005). As shown in lanes 2-5 and 6-9 of Fig. 7, increasing LiCl concentrations have weak or no effects on the size or homogeneity of the formed Aβ42 oligomers, as the bands retain their shape and intensity. Only for the globulomers subjected to the highest LiCl concentration (10 mM) is the intensity of the BN-PAGE band slightly reduced (lane 5, Fig 7). FTIR spectroscopy: effects of Li+ ions on Aβ42 oligomer structure The secondary structures of Aβ42 oligomers formed with different Li + concentrations were studied with FTIR spectroscopy, where the amide I region (1700-1600 cm-1) is very sensitive to changes in the protein backbone conformation. The technique is useful also in amyloid research, given its capacity to characterize β-sheets (Barth, 2007; Sarroukh et al., 2013). Fig. 8 shows second derivative IR spectra for the amide I region of Aβ42 globulomers (Fig. 8A) and smaller oligomers (Fig. 8B), prepared with different concentrations of Li+ ions. Monomeric Aβ42 displayed a relatively broad band at 1639-1640 cm -1, which is in agreement with the position of the band for disordered (random coil) polypeptides measured in D2O (Barth, 2007). For both types of Aβ42 oligomers, this main band is much narrower and downshifted by about 10 cm-1, while a second smaller band appears around 1685 cm-1. This split band pattern is indicative of an anti-parallel β-sheet conformation (Cerf et al., 2009). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 16 Earlier studies in our laboratory have shown a relationship between the position and width of this main band, and the size and homogeneity of the Aβ42 oligomers (Vosough & Barth, manuscript). The lower band position of the larger oligomers is in line with this relationship and our previous results, and confirms the different sizes of the oligomers produced at the two SDS concentrations. We have recently observed that a number of transition metal ions induce significant effects on the main band position for Aβ42 oligomers (manuscript in preparation). Because the spectra for Aβ42 oligomers formed with different amounts of LiCl generally superimpose on the IR spectra for the Li+-free oligomers, with no shifts observed for the main band, it appears that Li+ ions have no significant effect on the oligomers’ size or secondary structure. DISCUSSION Lithium as a therapeutic agent Lithium has no known biological functions in the human body. Li+ ions readily pass biological membranes, and are evenly distributed in tissues and easily eliminated via the kidneys (Nordberg et al., 2015). Li+ ions are however far from inert, and several well-defined medical conditions related to abnormal Li+ concentrations exist. In low blood concentrations, Li+ is used as a medication for bipolar and schizoaffective disorders (Machado-Vieira et al., 2009), but at higher concentrations Li+ ions are neurotoxic (Sellers et al., 1982; Emilien & Maloteaux, 1996; Nordberg et al., 2015; Wen et al., 2019). This leaves a narrow therapeutic window of 0.6 -1.2 mM that has to be closely monitored in order to prevent Li+ intoxication, which is easily recognized by EEG (Mignarri et al., 2013) and treatable by reducing the therapeutic dose. Li+ intoxication (>1.5 mM) presents as apathy, vertigo, tremor and gastrointestinal symptoms, in more severe cases confusion, psychosis, myoclonus and cardiac .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 17 arrhythmias (Nordberg et al., 2015). Li+ intoxication affects also the kidneys with polyuria and elevated U-albumin although overt renal failure is rare (Nordberg et al., 2015). Treatment of bipolar and schizoaffective disorders with Li+ has generated some knowledge about Li+ metabolism in the human body (Wen et al., 2019; Medic et al., 2020). Li+ accumulates to some extent in bone (Birch, 1974), and chronic Li+ effects are implicated in osteomalacia and severe osteoporosis (Roos, 2014). Patients treated with Li+ also show an increased frequency of hypothyroidism and goitre, and widespread effects on several facets of the endocrine system have been noted (Salata & Klein, 1987). The negative effects of Li+ on thyroid function have been clearly demonstrated in a study on populations in the Andean Mountains, where natural exposure to Li+ is high, and where urinary Li+ was found to correlate negatively with free thyroxine (T4) but correlate positively with the pituitary gland hormone thyrotropin (Broberg et al., 2011). The toxicity of Li+ is further emphasized by studies from regions with naturally elevated concentrations of Li+ in potable water, where reduced fetal size has been noted to correlate linearly with increases in blood Li+ (Harari et al., 2015). To what extent Li+ treatment reduces the development of AD symptoms is unclear (Engel et al., 2008; Mauer et al., 2014; Nordberg et al., 2015; Sutherland & Duthie, 2015). Bipolar disorder increases the risk of AD when compared to the general population, and Li+ treatment seems to reduce this risk (Velosa et al., 2020), but the mechanisms mediating this effect are far from elucidated (Kerr et al., 2018). In rare cases even regular-dose long-time Li+ therapy may cause severe intoxication of the central nervous system, characterized by cerebellar dysfunction and cognitive decline (Emilien & Maloteaux, 1996). Lithium interactions with the Aβ40 peptide .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 18 The NMR (Figs. 3 and 4), fluorescence quenching (Fig. 5), and CD (Fig. 6) experiments show that Li+ ions display weak interaction with the Aβ40 peptide, where the binding affinity for the Li+·Aβ40 complex may be in the millimolar range. The IR and CD results show minor or no effects of Li+ ions on the secondary structures of Aβ40 monomers (Fig. 6) and Aβ42 oligomers (Fig. 8). The Li+ ions may have a small effect on Aβ aggregation, with minor perturbations on the morphology of aggregated Aβ40 fibrils (Fig. 2), and effects on the Aβ40 aggregation kinetics (Fig. 1; Table 1) and Aβ42 oligomer stability (Fig. 7) only at very high Li+ concentrations. These results are in line with previous computer modeling results, which suggest small differences between how the monovalent K+, Li+, and Na+ alkali ions affect Aβ oligomerization (Huraskin & Horn, 2019). As Aβ40 and Aβ42 have identical N-terminal sequences, the two peptide variants should interact very similarly with Li+ ions, which were found to bind to the N-terminal Aβ region (Fig. 3B). The weak affinity between Aβ40 and Li + ions, and the fact that Li+ does not efficiently compete with Cu2+ ions for Aβ binding (Fig. 4), suggest that Li+ ions are not coordinated by specific binding ligands. Instead, Li+ likely engages in non-specific electrostatic interactions with the negatively charged Aβ residues, i.e. D1, E3, D7, E11, E22, and D23 (which are located in the N-terminal and central regions). The weak binding affinity to Aβ40 peptides is not caused by Li + ions being monovalent, as e.g. monovalent Ag+ ions display rather strong and specific binding to Aβ peptides (Wallin et al., 2020). Moreover, divalent Pb2+ and trivalent Cr3+ ions do not bind strongly to Aβ, while divalent Cu2+, Mn2+ and Zn2+ as well as tetravalent Pb4+ ions do (Faller, 2009; Abelein et al., 2015; Tiiman et al., 2016; Wallin et al., 2016; Wallin et al., 2017). Thus, Aβ/metal interactions are not governed by the charge of the metal ion, but rather by its specific properties, such as ionic radius and electron configuration (1s2 for Li+). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 19 It is illustrative to compare the Aβ interactions with Li+ ions to the well-studied interactions with Cu2+ and Zn2+ ions. These two divalent ions display residue-specific interactions with Aβ peptides, displaying binding affinities in the micromolar-nanomolar range and strong effects on Aβ secondary structure, aggregation, and diffusion (Danielsson et al., 2007; Faller, 2009; Lindgren et al., 2013; Abelein et al., 2015; Tiiman et al., 2016; Owen et al., 2019). Aβ binding to Cu2+ and Zn2+ is coordinated mainly by residue-specific interactions with the N-terminal His residues, i.e. H6, H13, and H14 (Faller, 2009; Lindgren et al., 2013; Abelein et al., 2015; Tiiman et al., 2016). The biological relevance of Cu2+ and Zn2+ ions in AD pathology is demonstrated by their dysregulation in AD patients (Wang et al., 2015; Szabo et al., 2016), and by them being accumulated in plaques of Aβ aggregates in AD brains (Beauchemin & Kisilevsky, 1998; Lovell et al., 1998; Miller et al., 2006). During neuronal signaling Cu2+ and Zn2+ ions are released into the synaptic clefts (Ayton et al., 2013), where they may interact with Aβ peptides to initiate Aβ aggregation (Branch et al., 2017), or modulate the formation and toxicity of Aβ oligomers (Stefaniak & Bal, 2019; Wärmländer et al., 2019). The current results indicate that Li+ ions are not able to compete with Cu2+ or Zn2+ ions for binding to Aβ peptides, and should therefore not be able to influence the in vivo effects of Cu2+ and Zn2+ ions on Aβ aggregation and toxicity. Although high concentrations of Li+ showed some effects on Aβ aggregation (Figs. 1-3;7), these effects are likely at least partly related to ionic strength effects (Abelein et al., 2016). Under physiological ionic strength, no specific interactions are observed between Aβ40 monomers and Li + ions (Fig. 3C,D). Thus, we conclude that the previously reported possible beneficial effects of Li+ on Alzheimer’s disease progression (Mauer et al., 2014; Sutherland & Duthie, 2015; Kerr et al., 2018; Hampel et al., 2019; Velosa et al., 2020) seem not to be caused by direct interactions between Li+ ions and Aβ peptides. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 20 ACKNOWLEDGMENTS: We thank Elizabeth (Li) Wang for helpful discussions. This work was supported by grants from the Swedish Alzheimer Foundation and the Swedish Research Council to AG, the Swedish Brain Foundation to AG and AB, the Magnus Bergvall Foundation to SW and PR, the Ulla-Carin Lindquist ALS Foundation to PR, and from Olle Engkvist's Foundation, the Stockholm Region, and Knut and Alice Wallenberg Foundation to AB. CONFICT OF INTEREST: The authors declare no conflict of interest. REFERENCES Abelein A, Abrahams JP, Danielsson J, Gräslund A, Jarvet J, Luo J, Tiiman A, Wärmländer SK (2014). The hairpin conformation of the amyloid beta peptide is an important structural motif along the aggregation pathway. J Biol Inorg Chem 19(4-5): 623-634. doi: 10.1007/s00775-014- 1131-8. Abelein A, Gräslund A, Danielsson J (2015). Zinc as chaperone-mimicking agent for retardation of amyloid beta peptide fibril formation. Proc Natl Acad Sci U S A 112(17): 5407-5412. doi: 10.1073/pnas.1421961112. Abelein A, Jarvet J, Barth A, Gräslund A, Danielsson J (2016). Ionic Strength Modulation of the Free Energy Landscape of Abeta40 Peptide Fibril Formation. J Am Chem Soc 138(21): 6893-6902. doi: 10.1021/jacs.6b04511. Adlard PA, Bush AI (2018). Metals and Alzheimer's Disease: How Far Have We Come in the Clinic? J Alzheimers Dis 62(3): 1369-1379. doi: 10.3233/JAD-170662. Agostinho P, Cunha RA, Oliveira C (2010). Neuroinflammation, oxidative stress and the pathogenesis of Alzheimer's disease. Curr Pharm Des 16(25): 2766-2778. doi: 10.2174/138161210793176572. Al-Hilaly YK, Williams TL, Stewart-Parker M, Ford L, Skaria E, Cole M, Bucher WG, Morris KL, Sada AA, Thorpe JR, Serpell LC (2013). A central role for dityrosine crosslinking of Amyloid-beta in Alzheimer's disease. Acta Neuropathol Commun 1: 83. doi: 10.1186/2051-5960-1-83. Alies B, Renaglia E, Rozga M, Bal W, Faller P, Hureau C (2013). Cu(II) affinity for the Alzheimer's peptide: tyrosine fluorescence studies revisited. Anal Chem 85(3): 1501-1508. doi: 10.1021/ac302629u. Ayton S, Lei P, Bush AI (2013). Metallostasis in Alzheimer's disease. Free Radic Biol Med 62: 76-89. doi: 10.1016/j.freeradbiomed.2012.10.558. Baldassarre M, Barth A (2014). Pushing the detection limit of infrared spectroscopy for structural analysis of dilute protein samples. Analyst 139(21): 5393-5399. doi: 10.1039/c4an00918e. Barghorn S, Nimmrich V, Striebinger A, Krantz C, P K, Janson B, Bahr M, Schmidt M, Bitner RS, Harlan J, Barlow E, Ebert U, Hillen H (2005). Globular amyloid β-peptide1-42 oligomer – a homogenous and stable neuropathological protein in Alzheimer’s disease. J Neurochem 95: 834–847. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 21 Baronio CM, Baldassarre M, Barth A (2019). Insight into the internal structure of amyloid-beta oligomers by isotope-edited Fourier transform infrared spectroscopy. Phys Chem Chem Phys 21(16): 8587-8597. doi: 10.1039/c9cp00717b. Barth A (2007). Infrared spectroscopy of proteins. Biochim Biophys Acta 1767(9): 1073-1101. doi: 10.1016/j.bbabio.2007.06.004. Beauchemin D, Kisilevsky R (1998). A method based on ICP-MS for the analysis of Alzheimer's amyloid plaques. Anal Chem 70(5): 1026-1029. Birch NJ (1974). Lithium accumulation in bone after oral administration in rat and in man. Clin Sci Mol Med 46(3): 409-413. doi: 10.1042/cs0460409. Bitan G, Fradinger EA, Spring SM, Teplow DB (2005). Neurotoxic protein oligomers--what you see is not always what you get. Amyloid 12(2): 88-95. doi: 10.1080/13506120500106958. Branch T, Barahona M, Dodson CA, Ying L (2017). Kinetic Analysis Reveals the Identity of Abeta- Metal Complex Responsible for the Initial Aggregation of Abeta in the Synapse. ACS Chem Neurosci 8(9): 1970-1979. doi: 10.1021/acschemneuro.7b00121. Broberg K, Concha G, Engstrom K, Lindvall M, Grander M, Vahter M (2011). Lithium in drinking water and thyroid function. Environ Health Perspect 119(6): 827-830. doi: 10.1289/ehp.1002678. Budni J, Feijo DP, Batista-Silva H, Garcez ML, Mina F, Belletini-Santos T, Krasilchik LR, Luz AP, Schiavo GL, Quevedo J (2017). Lithium and memantine improve spatial memory impairment and neuroinflammation induced by beta-amyloid 1-42 oligomers in rats. Neurobiol Learn Mem 141: 84-92. doi: 10.1016/j.nlm.2017.03.017. Caccamo A, Oddo S, Tran LX, LaFerla FM (2007). Lithium reduces tau phosphorylation but not A beta or working memory deficits in a transgenic model with both plaques and tangles. Am J Pathol 170(5): 1669-1675. doi: 10.2353/ajpath.2007.061178. Cardillo GM, De-Paula VJR, Ikenaga EH, Costa LR, Catanozi S, Schaeffer EL, Gattaz WF, Kerr DS, Forlenza OV (2018). Chronic Lithium Treatment Increases Telomere Length in Parietal Cortex and Hippocampus of Triple-Transgenic Alzheimer's Disease Mice. J Alzheimers Dis 63(1): 93- 101. doi: 10.3233/JAD-170838. Cerf E, Sarroukh R, Tamamizu-Kato S, Breydo L, Derclaye S, Dufrene YF, Narayanaswami V, Goormaghtigh E, Ruysschaert JM, Raussens V (2009). Antiparallel beta-sheet: a signature structure of the oligomeric amyloid beta-peptide. Biochem J 421(3): 415-423. doi: 10.1042/BJ20090379. Chin-Chan M, Navarro-Yepes J, Quintanilla-Vega B (2015). Environmental pollutants as risk factors for neurodegenerative disorders: Alzheimer and Parkinson diseases. Front Cell Neurosci 9: 124. doi: 10.3389/fncel.2015.00124. Cho CH, Urquidi J, Singh S, Wilse Robinson G (1999). Thermal Offset Viscosities of Liquid H2O, D2O, and T2O. J. Phys. Chem. B 103(11): 1991-1994. Danielsson J, Andersson A, Jarvet J, Gräslund A (2006). 15N relaxation study of the amyloid beta- peptide: structural propensities and persistence length. Magn Reson Chem 44 Spec No: S114-121. doi: 10.1002/mrc.1814. Danielsson J, Jarvet J, Damberg P, Gräslund A (2002). Translational diffusion measured by PFG‐NMR on full length and fragments of the Alzheimer Aβ(1–40) peptide. Determination of hydrodynamic radii of random coil peptides of varying length. Magnetic Resonance in Chemistry 40(13): S89-S97. Danielsson J, Pierattelli R, Banci L, Gräslund A (2007). High-resolution NMR studies of the zinc- binding site of the Alzheimer's amyloid beta-peptide. FEBS J 274(1): 46-59. doi: 10.1111/j.1742-4658.2006.05563.x. Decker M, Munoz-Torrero D (2016). Special Issue: "Molecules against Alzheimer". Molecules 21(12) doi: 10.3390/molecules21121736. Dell'Osso L, Del Grande C, Gesi C, Carmassi C, Musetti L (2016). A new look at an old drug: neuroprotective effects and therapeutic potentials of lithium salts. Neuropsychiatr Dis Treat 12: 1687-1703. doi: 10.2147/NDT.S106479. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 22 Donix M, Bauer M (2016). Population Studies of Association Between Lithium and Risk of Neurodegenerative Disorders. Curr Alzheimer Res 13(8): 873-878. doi: 10.2174/1567205013666160219112957. Duce JA, Bush AI, Adlard PA (2011). Role of amyloid-β–metal interactions in Alzheimer’s disease. Future Neurol 6(5): 641–659. Edelhoch H (1967). Spectroscopic Determination of Tryptophan and Tyrosine in Proteins. Biochemistry 6(7): 1948–1954. Emilien G, Maloteaux JM (1996). Lithium neurotoxicity at low therapeutic doses Hypotheses for causes and mechanism of action following a retrospective analysis of published case reports. Acta Neurol Belg 96(4): 281-293. Engel T, Goni-Oliver P, Gomez de Barreda E, Lucas JJ, Hernandez F, Avila J (2008). Lithium, a potential protective drug in Alzheimer's disease. Neurodegener Dis 5(3-4): 247-249. doi: 10.1159/000113715. Faller P (2009). Copper and zinc binding to amyloid-beta: coordination, dynamics, aggregation, reactivity and metal-ion transfer. Chembiochem 10(18): 2837-2845. doi: 10.1002/cbic.200900321. Forlenza OV, De-Paula VJ, Diniz BS (2014). Neuroprotective effects of lithium: implications for the treatment of Alzheimer's disease and related neurodegenerative disorders. ACS Chem Neurosci 5(6): 443-450. doi: 10.1021/cn5000309. Fornai F, Longone P, Cafaro L, Kastsiuchenka O, Ferrucci M, Manca ML, Lazzeri G, Spalloni A, Bellio N, Lenzi P, Modugno N, Siciliano G, Isidoro C, Murri L, Ruggieri S, Paparelli A (2008). Lithium delays progression of amyotrophic lateral sclerosis. Proc Natl Acad Sci U S A 105(6): 2052- 2057. doi: 10.1073/pnas.0708022105. Frozza RL, Lourenco MV, De Felice FG (2018). Challenges for Alzheimer's Disease Therapy: Insights from Novel Mechanisms Beyond Memory Defects. Front Neurosci 12: 37. doi: 10.3389/fnins.2018.00037. Gade Malmos K, Blancas-Mejia LM, Weber B, Buchner J, Ramirez-Alvarado M, Naiki H, Otzen D (2017). ThT 101: a primer on the use of thioflavin T to investigate amyloid formation. Amyloid 24(1): 1-16. doi: 10.1080/13506129.2017.1304905. Ge W, Jakobsson E (2018). Systems Biology Understanding of the Effects of Lithium on Affective and Neurodegenerative Disorders. Front Neurosci 12: 933. doi: 10.3389/fnins.2018.00933. Ghalebani L, Wahlström A, Danielsson J, Wärmländer SK, Gräslund A (2012). pH-dependence of the specific binding of Cu(II) and Zn(II) ions to the amyloid-beta peptide. Biochem Biophys Res Commun 421(3): 554-560. doi: 10.1016/j.bbrc.2012.04.043. Gibbons GS, Lee VMY, Trojanowski JQ (2019). Mechanisms of Cell-to-Cell Transmission of Pathological Tau: A Review. JAMA Neurol 76(1): 101-108. doi: 10.1001/jamaneurol.2018.2505. Glasoe PK, Long FA (1960). Use of glass electrodes to measure acidities in deuterium oxide. J Phys Chem 64: 88–90. Glenner GG, Wong CW (1984). Alzheimer's disease: initial report of the purification and characterization of a novel cerebrovascular amyloid protein. Biochem Biophys Res Commun 120(3): 885-890. Goedert M (2018). Tau filaments in neurodegenerative diseases. FEBS Lett 592(14): 2383-2391. doi: 10.1002/1873-3468.13108. Habib A, Sawmiller D, Li S, Xiang Y, Rongo D, Tian J, Hou H, Zeng J, Smith A, Fan S, Giunta B, Mori T, Currier G, Shytle DR, Tan J (2017). LISPRO mitigates beta-amyloid and associated pathologies in Alzheimer's mice. Cell Death Dis 8(6): e2880. doi: 10.1038/cddis.2017.279. Habib A, Shytle RD, Sawmiller D, Koilraj S, Munna SA, Rongo D, Hou H, Borlongan CV, Currier G, Tan J (2019). Comparing the effect of the novel ionic cocrystal of lithium salicylate proline (LISPRO) with lithium carbonate and lithium salicylate on memory and behavior in female APPswe/PS1dE9 Alzheimer's mice. J Neurosci Res 97(9): 1066-1080. doi: 10.1002/jnr.24438. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 23 Hampel H, Lista S, Mango D, Nistico R, Perry G, Avila J, Hernandez F, Geerts H, Vergallo A, Alzheimer Precision Medicine I (2019). Lithium as a Treatment for Alzheimer's Disease: The Systems Pharmacology Perspective. J Alzheimers Dis 69(3): 615-629. doi: 10.3233/JAD-190197. Harari F, Langeen M, Casimiro E, Bottai M, Palm B, Nordqvist H, Vahter M (2015). Environmental exposure to lithium during pregnancy and fetal size: a longitudinal study in the Argentinean Andes. Environ Int 77: 48-54. doi: 10.1016/j.envint.2015.01.011. Heppner FL, Ransohoff RM, Becher B (2015). Immune attack: the role of inflammation in Alzheimer disease. Nat Rev Neurosci 16(6): 358-372. doi: 10.1038/nrn3880. Huat TJ, Camats-Perna J, Newcombe EA, Valmas N, Kitazawa M, Medeiros R (2019). Metal Toxicity Links to Alzheimer's Disease and Neuroinflammation. J Mol Biol 431(9): 1843-1868. doi: 10.1016/j.jmb.2019.01.018. Huraskin D, Horn AHC (2019). Alkali ion influence on structure and stability of fibrillar amyloid-beta oligomers. J Mol Model 25(2): 37. doi: 10.1007/s00894-018-3920-4. Kerr F, Bjedov I, Sofola-Adesakin O (2018). Molecular Mechanisms of Lithium Action: Switching the Light on Multiple Targets for Dementia Using Animal Models. Front Mol Neurosci 11: 297. doi: 10.3389/fnmol.2018.00297. Kisby B, Jarrell JT, Agar ME, Cohen DS, Rosin ER, Cahill CM, Rogers JT, Huang X (2019). Alzheimer's Disease and Its Potential Alternative Therapeutics. J Alzheimers Dis Parkinsonism 9(5) doi: 10.4172/2161-0460.1000477. Lee SJ, Nam E, Lee HJ, Savelieff MG, Lim MH (2017). Towards an understanding of amyloid-beta oligomers: characterization, toxicity mechanisms, and inhibitors. Chem Soc Rev 46(2): 310- 323. doi: 10.1039/c6cs00731g. Lindgren J, Segerfeldt P, Sholts SB, Gräslund A, Karlström AE, Wärmländer SK (2013). Engineered non-fluorescent Affibody molecules facilitate studies of the amyloid-beta (Abeta) peptide in monomeric form: low pH was found to reduce Abeta/Cu(II) binding affinity. J Inorg Biochem 120: 18-23. doi: 10.1016/j.jinorgbio.2012.11.005. Liu M, Qian T, Zhou W, Tao X, Sang S, Zhao L (2020). Beneficial effects of low-dose lithium on cognitive ability and pathological alteration of Alzheimer's disease transgenic mice model. Neuroreport 31(13): 943-951. doi: 10.1097/WNR.0000000000001499. Lovell MA, Robertson JD, Teesdale WJ, Campbell JL, Markesbery WR (1998). Copper, iron and zinc in Alzheimer's disease senile plaques. J Neurol Sci 158(1): 47-52. Luo J, Mohammed I, Wärmländer SK, Hiruma Y, Gräslund A, Abrahams JP (2014). Endogenous polyamines reduce the toxicity of soluble abeta peptide aggregates associated with Alzheimer's disease. Biomacromolecules 15(6): 1985-1991. doi: 10.1021/bm401874j. Luo J, Otero JM, Yu CH, Wärmländer SK, Gräslund A, Overhand M, Abrahams JP (2013). Inhibiting and reversing amyloid-beta peptide (1-40) fibril formation with gramicidin S and engineered analogues. Chemistry 19(51): 17338-17348. doi: 10.1002/chem.201301535. Luo J, Wärmländer SK, Gräslund A, Abrahams JP (2014). Alzheimer peptides aggregate into transient nanoglobules that nucleate fibrils. Biochemistry 53(40): 6302-6308. doi: 10.1021/bi5003579. Luo J, Wärmländer SK, Gräslund A, Abrahams JP (2016). Cross-interactions between the Alzheimer Disease Amyloid-beta Peptide and Other Amyloid Proteins: A Further Aspect of the Amyloid Cascade Hypothesis. J Biol Chem 291(32): 16485-16493. doi: 10.1074/jbc.R116.714576. Luo J, Yu CH, Yu H, Borstnar R, Kamerlin SC, Gräslund A, Abrahams JP, Wärmländer SK (2013). Cellular polyamines promote amyloid-beta (Abeta) peptide fibrillation and modulate the aggregation pathways. ACS Chem Neurosci 4(3): 454-462. doi: 10.1021/cn300170x. Machado-Vieira R, Manji HK, Zarate CA, Jr. (2009). The role of lithium in the treatment of bipolar disorder: convergent evidence for neurotrophic effects as a unifying hypothesis. Bipolar Disord 11 Suppl 2: 92-109. doi: 10.1111/j.1399-5618.2009.00714.x. Mauer S, Vergne D, Ghaemi SN (2014). Standard and trace-dose lithium: a systematic review of dementia prevention and other behavioral benefits. Aust N Z J Psychiatry 48(9): 809-818. doi: 10.1177/0004867414536932. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 24 Medic B, Stojanovic M, Stimec BV, Divac N, Vujovic KS, Stojanovic R, Colovic M, Krstic D, Prostran M (2020). Lithium - Pharmacological and Toxicological Aspects: The Current State of the Art. Curr Med Chem 27(3): 337-351. doi: 10.2174/0929867325666180904124733. Mignarri A, Chini E, Rufa A, Rocchi R, Federico A, Dotti MT (2013). Lithium neurotoxicity mimicking rapidly progressive dementia. J Neurol 260(4): 1152-1154. doi: 10.1007/s00415-012-6820-z. Miller LM, Wang Q, Telivala TP, Smith RJ, Lanzirotti A, Miklossy J (2006). Synchrotron-based infrared and X-ray imaging shows focalized accumulation of Cu and Zn co-localized with beta-amyloid deposits in Alzheimer's disease. J Struct Biol 155(1): 30-37. doi: 10.1016/j.jsb.2005.09.004. Mital M, Wezynfeld NE, Fraczyk T, Wiloch MZ, Wawrzyniak UE, Bonna A, Tumpach C, Barnham KJ, Haigh CL, Bal W, Drew SC (2015). A Functional Role for Abeta in Metal Homeostasis? N- Truncation and High-Affinity Copper Binding. Angew Chem Int Ed Engl 54(36): 10460-10464. doi: 10.1002/anie.201502644. Modgil S, Lahiri DK, Sharma VL, Anand A (2014). Role of early life exposure and environment on neurodegeneration: implications on brain disorders. Transl Neurodegener 3: 9. doi: 10.1186/2047-9158-3-9. Morris G, Berk M (2016). The Putative Use of Lithium in Alzheimer's Disease. Curr Alzheimer Res 13(8): 853-861. doi: 10.2174/1567205013666160219113112. Morrison KE, Dhariwal S, Hornabrook R, Savage L, Burn DJ, Khoo TK, Kelly J, Murphy CL, Al-Chalabi A, Dougherty A, Leigh PN, Wijesekera L, Thornhill M, Ellis CM, O'Hanlon K, Panicker J, Pate L, Ray P, Wyatt L, Young CA, Copeland L, Ealing J, Hamdalla H, Leroi I, Murphy C, O'Keeffe F, Oughton E, Partington L, Paterson P, Rog D, Sathish A, Sexton D, Smith J, Vanek H, Dodds S, Williams TL, Steen IN, Clarke J, Eziefula C, Howard R, Orrell R, Sidle K, Sylvester R, Barrett W, Merritt C, Talbot K, Turner MR, Whatley C, Williams C, Williams J, Cosby C, Hanemann CO, Iman I, Philips C, Timings L, Crawford SE, Hewamadduma C, Hibberd R, Hollinger H, McDermott C, Mils G, Rafiq M, Shaw PJ, Taylor A, Waines E, Walsh T, Addison-Jones R, Birt J, Hare M, Majid T (2013). Lithium in patients with amyotrophic lateral sclerosis (LiCALS): a phase 3 multicentre, randomised, double-blind, placebo-controlled trial. Lancet Neurol 12(4): 339-345. doi: 10.1016/S1474-4422(13)70037-1. Nath S, Agholme L, Kurudenkandy FR, Granseth B, Marcusson J, Hallbeck M (2012). Spreading of neurodegenerative pathology via neuron-to-neuron transmission of beta-amyloid. J Neurosci 32(26): 8767-8777. doi: 10.1523/JNEUROSCI.0615-12.2012. Nečas D, Klapetek P (2012). Gwyddion: an open-source software for SPM data analysis. Central European Journal of Physics 10: 181-188. doi: https://doi.org/10.2478. Nery LR, Eltz NS, Hackman C, Fonseca R, Altenhofen S, Guerra HN, Freitas VM, Bonan CD, Vianna MR (2014). Brain intraventricular injection of amyloid-beta in zebrafish embryo impairs cognition and increases tau phosphorylation, effects reversed by lithium. PLoS One 9(9): e105862. doi: 10.1371/journal.pone.0105862. Nordberg G, Fowler B, Nordberg M, (eds). (2015). Handbook on the Toxicology of Metals, Elsevier. Owen MC, Gnutt D, Gao M, Wärmländer SKTS, Jarvet J, Gräslund A, Winter R, Ebbinghaus S, Strodel B (2019). Effects of in vivo conditions on amyloid aggregation. Chem Soc Rev 48(14): 3946- 3996. doi: 10.1039/c8cs00034d. Pan Y, Short JL, Newman SA, Choy KHC, Tiwari D, Yap C, Senyschyn D, Banks WA, Nicolazzo JA (2018). Cognitive benefits of lithium chloride in APP/PS1 mice are associated with enhanced brain clearance of beta-amyloid. Brain Behav Immun 70: 36-47. doi: 10.1016/j.bbi.2018.03.007. Priebe GA, Kanzawa MM (2020). Reducing the progression of Alzheimer's disease in Down syndrome patients with micro-dose lithium. Med Hypotheses 137: 109573. doi: 10.1016/j.mehy.2020.109573. Prince M, Wimo A, Guerchet M, Ali G-C, Wu Y-T, Prina M (2015). World Alzheimer Report 2015 - The Global Impact of Dementia. London, UK. Querfurth HW, LaFerla FM (2010). Alzheimer's disease. N Engl J Med 362(4): 329-344. doi: 10.1056/NEJMra0909142. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.2478 https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 25 Rangachari V, Moore BD, Reed DK, Sonoda LK, Bridges AW, Conboy E, Hartigan D, Rosenberry TL (2007). Amyloid-beta(1-42) rapidly forms protofibrils and oligomers by distinct pathways in low concentrations of sodium dodecylsulfate. Biochemistry 46(43): 12451-12462. doi: 10.1021/bi701213s. Regen F, Hellmann-Regen J, Costantini E, Reale M (2017). Neuroinflammation and Alzheimer's Disease: Implications for Microglial Activation. Curr Alzheimer Res 14(11): 1140-1148. doi: 10.2174/1567205014666170203141717. Rocha NKR, Themoteo R, Brentani H, Forlenza OV, De Paula VJR (2020). Neuronal-Glial Interaction in a Triple-Transgenic Mouse Model of Alzheimer's Disease: Gene Ontology and Lithium Pathways. Front Neurosci 14: 579984. doi: 10.3389/fnins.2020.579984. Rockenstein E, Torrance M, Adame A, Mante M, Bar-on P, Rose JB, Crews L, Masliah E (2007). Neuroprotective effects of regulators of the glycogen synthase kinase-3beta signaling pathway in a transgenic model of Alzheimer's disease are associated with reduced amyloid precursor protein phosphorylation. J Neurosci 27(8): 1981-1991. doi: 10.1523/JNEUROSCI.4321-06.2007. Roos PM (2014). Osteoporosis in neurodegeneration. J Trace Elem Med Biol 28(4): 418-421. doi: 10.1016/j.jtemb.2014.08.010. Ryves WJ, Harwood AJ (2001). Lithium inhibits glycogen synthase kinase-3 by competition for magnesium. Biochem Biophys Res Commun 280(3): 720-725. doi: 10.1006/bbrc.2000.4169. Salata R, Klein I (1987). Effects of lithium on the endocrine system: a review. J Lab Clin Med 110(2): 130-136. Sardar Sinha M, Ansell-Schultz A, Civitelli L, Hildesjo C, Larsson M, Lannfelt L, Ingelsson M, Hallbeck M (2018). Alzheimer's disease pathology propagation by exosomes containing toxic amyloid- beta oligomers. Acta Neuropathol 136(1): 41-56. doi: 10.1007/s00401-018-1868-1. Sarroukh R, Goormaghtigh E, Ruysschaert JM, Raussens V (2013). ATR-FTIR: a "rejuvenated" tool to investigate amyloid proteins. Biochim Biophys Acta 1828(10): 2328-2338. doi: 10.1016/j.bbamem.2013.04.012. Selkoe DJ, Hardy J (2016). The amyloid hypothesis of Alzheimer's disease at 25 years. EMBO Mol Med 8(6): 595-608. doi: 10.15252/emmm.201606210. Sellers J, Tyrer P, Whiteley A, Banks DC, Barer DH (1982). Neurotoxic effects of lithium with delayed rise in serum lithium levels. Br J Psychiatry 140: 623-625. doi: 10.1192/bjp.140.6.623. Sengupta U, Nilson AN, Kayed R (2016). The Role of Amyloid-beta Oligomers in Toxicity, Propagation, and Immunotherapy. EBioMedicine 6: 42-49. doi: 10.1016/j.ebiom.2016.03.035. Sofola-Adesakin O, Castillo-Quan JI, Rallis C, Tain LS, Bjedov I, Rogers I, Li L, Martinez P, Khericha M, Cabecinha M, Bahler J, Partridge L (2014). Lithium suppresses Abeta pathology by inhibiting translation in an adult Drosophila model of Alzheimer's disease. Front Aging Neurosci 6: 190. doi: 10.3389/fnagi.2014.00190. Stefaniak E, Bal W (2019). Cu(II) Binding Properties of N-Truncated Abeta Peptides: In Search of Biological Function. Inorg Chem 58(20): 13561-13577. doi: 10.1021/acs.inorgchem.9b01399. Sutherland C, Duthie AC (2015). Invited commentary on ... Lithium treatment and risk for dementia in adults with bipolar disorder. Br J Psychiatry 207(1): 52-54. doi: 10.1192/bjp.bp.114.161729. Szabo ST, Harry GJ, Hayden KM, Szabo DT, Birnbaum L (2016). Comparison of Metal Levels between Postmortem Brain and Ventricular Fluid in Alzheimer's Disease and Nondemented Elderly Controls. Toxicol Sci 150(2): 292-300. doi: 10.1093/toxsci/kfv325. Tiiman A, Luo J, Wallin C, Olsson L, Lindgren J, Jarvet J, Roos PM, Sholts SB, Rahimipour S, Abrahams JP, Karlström AE, Gräslund A, Wärmländer SKTS (2016). Specific Binding of Cu(II) Ions to Amyloid-Beta Peptides Bound to Aggregation-Inhibiting Molecules or SDS Micelles Creates Complexes that Generate Radical Oxygen Species. J Alzheimers Dis 54(3): 971-982. doi: 10.3233/JAD-160427. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 26 Trujillo-Estrada L, Jimenez S, De Castro V, Torres M, Baglietto-Vargas D, Moreno-Gonzalez I, Navarro V, Sanchez-Varo R, Sanchez-Mejias E, Davila JC, Vizuete M, Gutierrez A, Vitorica J (2013). In vivo modification of Abeta plaque toxicity as a novel neuroprotective lithium-mediated therapy for Alzheimer's disease pathology. Acta Neuropathol Commun 1: 73. doi: 10.1186/2051-5960-1-73. Velosa J, Delgado A, Finger E, Berk M, Kapczinski F, de Azevedo Cardoso T (2020). Risk of dementia in bipolar disorder and the interplay of lithium: a systematic review and meta-analyses. Acta Psychiatr Scand doi: 10.1111/acps.13153. Vosough F, Barth A (manuscript). Characterization of homogeneous and heterogeneous amyloid-β42 oligomer preparations with biochemical methods and infrared spectroscopy reveals a correlation between infrared spectrum and oligomer size. Wallin C, Friedemann M, Sholts SB, Noormagi A, Svantesson T, Jarvet J, Roos PM, Palumaa P, Gräslund A, Wärmländer SKTS (2020). Mercury and Alzheimer's Disease: Hg(II) Ions Display Specific Binding to the Amyloid-beta Peptide and Hinder Its Fibrillization. Biomolecules 10(1): 44. doi: 10.3390/biom10010044. Wallin C, Jarvet J, Biverstål H, Wärmländer S, Danielsson J, Gräslund A, Abelein A (2020). Metal ion coordination delays amyloid-beta peptide self-assembly by forming an aggregation-inert complex. J Biol Chem 295(21): 7224-7234. doi: 10.1074/jbc.RA120.012738. Wallin C, Kulkarni YS, Abelein A, Jarvet J, Liao Q, Strodel B, Olsson L, Luo J, Abrahams JP, Sholts SB, Roos PM, Kamerlin SC, Gräslund A, Wärmländer SK (2016). Characterization of Mn(II) ion binding to the amyloid-beta peptide in Alzheimer's disease. J Trace Elem Med Biol 38: 183- 193. doi: 10.1016/j.jtemb.2016.03.009. Wallin C, Sholts SB, Österlund N, Luo J, Jarvet J, Roos PM, Ilag L, Gräslund A, Wärmländer S (2017). Alzheimer's disease and cigarette smoke components: effects of nicotine, PAHs, and Cd(II), Cr(III), Pb(II), Pb(IV) ions on amyloid-beta peptide aggregation. Sci Rep 7(1): 14423. doi: 10.1038/s41598-017-13759-5. Wang X, Wang W, Li L, Perry G, Lee HG, Zhu X (2014). Oxidative stress and mitochondrial dysfunction in Alzheimer's disease. Biochim Biophys Acta 1842(8): 1240-1247. doi: 10.1016/j.bbadis.2013.10.015. Wang ZX, Tan L, Wang HF, Ma J, Liu J, Tan MS, Sun JH, Zhu XC, Jiang T, Yu JT (2015). Serum Iron, Zinc, and Copper Levels in Patients with Alzheimer's Disease: A Replication Study and Meta- Analyses. J Alzheimers Dis 47(3): 565-581. doi: 10.3233/JAD-143108. Wen J, Sawmiller D, Wheeldon B, Tan J (2019). A Review for Lithium: Pharmacokinetics, Drug Design, and Toxicity. CNS Neurol Disord Drug Targets 18(10): 769-778. doi: 10.2174/1871527318666191114095249. Wilson EN, Do Carmo S, Welikovitch LA, Hall H, Aguilar LF, Foret MK, Iulita MF, Jia DT, Marks AR, Allard S, Emmerson JT, Ducatenzeiler A, Cuello AC (2020). NP03, a Microdose Lithium Formulation, Blunts Early Amyloid Post-Plaque Neuropathology in McGill-R-Thy1-APP Alzheimer-Like Transgenic Rats. J Alzheimers Dis 73(2): 723-739. doi: 10.3233/JAD-190862. Wärmländer S, Tiiman A, Abelein A, Luo J, Jarvet J, Söderberg KL, Danielsson J, Gräslund A (2013). Biophysical studies of the amyloid beta-peptide: interactions with metal ions and small molecules. Chembiochem 14(14): 1692-1704. doi: 10.1002/cbic.201300262. Wärmländer SKTS, Österlund N, Wallin C, Wu J, Luo J, Tiiman A, Jarvet J, Gräslund A (2019). Metal binding to the Amyloid-β peptides in the presence of biomembranes: potential mechanisms of cell toxicity. Journal of Biological Inorganic Chemistry 24: 1189–1196 Xiang J, Cao K, Dong YT, Xu Y, Li Y, Song H, Zeng XX, Ran LY, Hong W, Guan ZZ (2020). Lithium chloride reduced the level of oxidative stress in brains and serums of APP/PS1 double transgenic mice via the regulation of GSK3beta/Nrf2/HO-1 pathway. Int J Neurosci 130(6): 564-573. doi: 10.1080/00207454.2019.1688808. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 27 Yu F, Zhang Y, Chuang DM (2012). Lithium reduces BACE1 overexpression, beta amyloid accumulation, and spatial learning deficits in mice with traumatic brain injury. J Neurotrauma 29(13): 2342-2351. doi: 10.1089/neu.2012.2449. Zhao L, Gong N, Liu M, Pan X, Sang S, Sun X, Yu Z, Fang Q, Zhao N, Fei G, Jin L, Zhong C, Xu T (2014). Beneficial synergistic effects of microdose lithium with pyrroloquinoline quinone in an Alzheimer's disease mouse model. Neurobiol Aging 35(12): 2736-2745. doi: 10.1016/j.neurobiolaging.2014.06.003. Österlund N, Kulkarni YS, Misiaszek AD, Wallin C, Krüger DM, Liao Q, Mashayekhy Rad F, Jarvet J, Strodel B, Wärmländer SKTS, Ilag LL, Kamerlin SCL, Gräslund A (2018). Amyloid-beta Peptide Interactions with Amphiphilic Surfactants: Electrostatic and Hydrophobic Effects. ACS Chem Neurosci 9(7): 1680-1692. doi: 10.1021/acschemneuro.8b00065. 20 μM Aβ40 1:1 Li +:Aβ 10:1 Li+:Aβ 100:1 Li+:Aβ t1/2 [hours] 3.7 ± 0.7 3.6 ± 2.1 3.8 ± 0.6 4.9 ± 1.3 rmax [hours -1] 0.5 ± 0.1 0.4 ± 0.2 0.5 ± 0.2 0.4 ± 0.1 Table 1. Kinetic parameters for Aβ40 fibril formation, i.e. aggregation half-time (t1/2) and maximum aggregation rate (rmax), derived from fitting the curves in Fig. 1 to Eq. 1. FIGURES .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 28 Fig. 1. Amyloid fibril formation monitored by ThT aggregation. Samples of 20 μM Aβ40 peptides in 20 mM MES buffer, pH 7.35, were incubated at +37 °C together with 50 μM Thioflavin-T and different concentrations of LiCl: 0 μM – black; 20 μM – red; 200 μM – green; 2000 μM – blue. The circles represent average data points for four replicates, while the solid lines are derived from fitting to Eq. 1. Fig. 2. Solid state AFM images (A1-D1) of aggregates of 20 µM Aβ40, incubated in 5 mM MES buffer, pH 7.35, for 72 hours at +37 °C with 300 rpm shaking, together with different concentrations of LiCl. A. control sample - no LiCl; B. 20 µM LiCl; C. 200 µM LiCl; D. 2 mM LiCl. The height profile graphs (A2-D2) below the AFM images correspond to the cross- sections of Aβ40 fibrils shown as white lines in the AFM images. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 29 Fig. 3. NMR experiments for interactions between Aβ40 monomers and Li + ions. (A) 2D 1H- 15N-HSQC spectra of 92.4 μM 15N-labeled Aβ40 peptides in 20 mM MES buffer, pH 7.35 at +5 °C, recorded for Aβ40 peptides alone (dark sky blue) and in the presence of either 924 μM LiCl (1:10 Aβ:Li ratio; passion red) or 9.24 mM LiCl (1:100 Aβ:Li ratio; Robin egg blue). (B) Relative intensities of Aβ40 residue crosspeaks shown in (A), after addition of LiCl in 1:1, 1:10, and 1:100 Aβ:Li ratios. (C and D) similar experiments as in A and B, but carried out in the presence of 1x PBS buffer, and for Aβ:Li ratios of 1:1, 1:5, and 1:50. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 30 Fig. 4. NMR diffusion data for 55 μM Aβ40 peptides in sodium phosphate buffer, pH 7.35 at +5 °C, recorded both in absence (A) and presence of different Li+ concentrations, i.e. 55 μM (B), 1.1 mM (C), and 5.5 mM (D). Fig. 5. Binding curves for the Cu2+·Aβ40 complex, obtained from the quenching effect of Cu2+ ions on the intrinsic fluorescence of Aβ residue Y10. CuCl2 was titrated to 10 µM Aβ40 in 20 mM MES buffer, pH 7.35 at 20 °C, both in the absence (red dots) and the presence (blue triangles) of 1 mM LiCl. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 31 Fig. 6. CD spectra of 20 μM Aβ40 peptides at 20 °C in 20 mM sodium phosphate buffer, pH 7.35. Spectra were recorded for Aβ in buffer only (black), after addition of 50 mM micellar SDS (brown), and after subsequent addition of between 2 µM (blue) and 512 µM (gray) of LiCl. The inset figure shows a close-up of the CD signals for the LiCl titration in the 210-230 nm range. Fig. 7. BN-PAGE gel showing the effects of different concentrations of Li+ ions on the formation of SDS-stabilized Aβ42 oligomers. Lane 1: monomers prepared in 5 mM NaOD. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/ 32 Lanes 2-5: Aβ42 globulomers formed after 24 hours of incubation with 0.05% SDS and different LiCl concentrations. Lanes 6-9: Aβ42 oligomers formed after 24 hours of incubation with 0.2% SDS and different LiCl concentrations. Fig. 8. Second derivatives of infrared absorbance spectra for 100 µM Aβ42 monomers (black) and 80 µM SDS-stabilized Aβ42 oligomers formed in absence (blue) and presence of 0.1 mM (red), 1 mM (purple), and 10 mM (green) of LiCl. The results are shown for Aβ42 globulomers at 0.05% SDS (A) and smaller oligomers at 0.2% SDS (B). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted January 4, 2021. ; https://doi.org/10.1101/2021.01.03.425155doi: bioRxiv preprint https://doi.org/10.1101/2021.01.03.425155 http://creativecommons.org/licenses/by-nc-nd/4.0/