A common antigenic motif recognized by naturally occurring human VH5–51/VL4–1 anti-tau antibodies with distinct functionalities

A common antigenic motif recognized by naturally occurring human VH5–51/VL4–1 anti-tau... Misfolding and aggregation of tau protein are closely associated with the onset and progression of Alzheimer’sDisease (AD). By interrogating IgG memory B cells from asymptomatic donors with tau peptides, we have identified two somatically mutated V 5–51/V 4–1 antibodies. One of these, CBTAU-27.1, binds to the aggregation motif H L in the R3 repeat domain and blocks the aggregation of tau into paired helical filaments (PHFs) by sequestering monomeric tau. The other, CBTAU-28.1, binds to the N-terminal insert region and inhibits the spreading of tau seeds and mediates the uptake of tau aggregates into microglia by binding PHFs. Crystal structures revealed that the combination of V 5–51 and V 4–1 recognizes a common Pro-X -Lys motif driven by germline-encoded hotspot interactions while the specificity and L n thereby functionality of the antibodies are defined by the CDR3 regions. Affinity improvement led to improvement in functionality, identifying their epitopes as new targets for therapy and prevention of AD. Keywords: Alzheimer’s disease, Tau protein, Monoclonal antibody, Antigenic motif Introduction phosphorylation and dephosphorylation of several of these Intracellular neurofibrillary tangles (NFTs) consisting of has been shown to affect its interaction with tubulin and aggregated tau protein are a hallmark of Alzheimer’sdisease cytoskeleton function [4, 38]. Hyperphosphorylation of tau (AD) and other neurogenerative disorders, collectively is thought to lead to microtubule dissociation and assembly referred to as tauopathies [31]. Tau is a microtubule- of the normally disordered, highly soluble protein into β associated protein expressed predominantly in neuronal sheet-rich aggregated fibrils called paired helical filaments axons and promotes the assembly and stability of microtu- (PHFs) that make up NFTs [8, 30, 34]. While the molecular bules [9, 46]. It is expressed in the adult human brain as six mechanism of tau aggregation remains elusive, it is believed isoforms with zero, one or two N-terminal acidic inserts that its initial nucleation step is energetically unfavorable, (0N, 1N, or2 N) and either three or four microtubule- whereas the subsequent fibril growth follows an energetic- binding repeats (3R or 4R) [18]. The tau protein contains ally downhill landscape [2, 11, 23, 39]. Accumulating many potential phosphorylation sites and the regulated evidence indicates that these fibrils can transmit from cell to cell and spread tau pathology to distant brain regions by seeding the recruitment of soluble tau into de novo aggre- * Correspondence: AApetri@its.jnj.com gates [16, 21, 22, 26, 47]. Janssen Prevention Center, Janssen Pharmaceutical Companies of Johnson & Johnson, Archimedesweg 6, 2333, CN, Leiden, the Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 2 of 17 Several monoclonal antibodies that inhibit the spread- of unphosphorylated tau peptides as baits (Fig. 1a, and ing of tau fibrils have been described and are being Additional file 1: Table S1 for peptide sequences). developed for antibody-based therapies [6, 42, 48, 49]. We have recently described the isolation of a panel of antibodies with such functional activity by interrogating Materials and methods the peripheral IgG memory B cells of healthy human Human PBMC isolation blood donors for reactivity to phosphorylated tau Wholeblood from healthymaleand femaledonors was peptides [37]. To expand the arsenal of potential targets obtained from the San Diego Blood Bank (ages 18–65 years) and include epitopes present in physiological tau, we after informed consent was obtained from the donors. used the BSelex technology in combination with a pool PBMCs were isolated on Ficoll-Paque Plus (GE Healthcare) Fig. 1 (See legend on next page.) Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 3 of 17 (See figure on previous page.) Fig. 1 Recovery and structural characterization of naturally occurring monoclonal antibodies to unphosphorylated tau epitopes from asymptomatic individuals. a BSelex method used to recover tau-specific memory B cells. PBMCs were prepared from asymptomatic blood bank donors, and mature CD22 B cells were positively selected with magnetic beads. Viable cells were stained with IgG-FITC, CD19-PerCPCy5.5, and CD27-PECy7, and with a pool of 10 overlapping unphosphorylated tau peptides spanning the longest tau isoform (relative position of each peptide along 2N4R tau indicated). + + + + + All peptides were present in the pool with an APC label as well as with a PE label and CD19 ,CD27 ,IgG ,APC ,PE cells were single-cell sorted on a Beckman Coulter MoFlo XDP. Antibody heavy and light variable chain sequences were recovered from single cells, cloned and expressed as full-length IgGs. b and c Co-crystal structures of Fab CBTAU-27.1 (b) and Fab CBTAU-28.1 (c) with tau peptides A8119 and A7731, respectively. Antibodies have been plotted as molecular surface with light chain in white and heavy chain in grey. Tau peptides are shown as cartoon with interacting amino acids plotted as sticks. Proline and lysine residues are plotted in green, amino acids in between these residues are colored in yellow and the termini in grey. Only the interacting antibody loops are outlined. d Key interactions with tau of CBTAU-27.1 (upper row) and CBTAU-28.1 (lower row). Key interacting residues are plotted as sticks, polar interactions are indicated with dotted lines, and the corresponding distances are indicated in Å. In the first panel, 312 59 interactions with Pro and Pro are compared where the proline binding pockets are visualized on a molecular surface. In the second panel, 317 67 315 65 interactions with Lys and Lys are compared. In panel 3, interactions around Leu and Asp in the central region of the epitopes are shown. e Structural basis for recognition of the Pro – X – Lys epitope motif. Epitopes of CBTAU-27.1 and CBTAU-28.1 are superimposed by aligning on the proline and lysine residues. The peptides present both residues in the same spatial orientation. In the central region (yellow), hydrophobic Leu in CBTAU-27.1 is replaced by hydrophilic and charged Asp in CBTAU-28.1. f Schematic representation of tau isoform 2N4R showing the epitopes of CBTAU-27.1 and CBTAU -28.1 (bold and underlined) and the surrounding sequences. Highlighted in grey and red are the microtubule binding motifs 306 311 and the hexapeptide VQIVYK which forms the N-terminal end of the core of PHFs, respectively. N1 and N2 indicate acidic inserts, P1 and P2 indi- cate proline-rich domains, and R1-R4 indicate microtubule-binding repeat domains and cryopreserved at 50 million cells per ml in 90% FBS Peptide synthesis and 10% DMSO. All peptides were synthesized at Pepscan BV (Lelystad, The Netherlands) or New England Peptide, Inc. using routine Fmoc-based solid phase strategies on automated synthe- Single cell sorting and recovery of heavy and light variable sizers. After acidic cleavage and deprotection, peptides were light antibody genes from tau-specific memory B cells purified by reversed phase HPLC, lyophilized and stored as Tau-peptide specific B cells were sorted and antibody powders at − 80 °C. Purity and identity were confirmed by chains were recovered as previously described [37]. LC-MS. Briefly, a panel of 10 biotinylated peptides spanning the longest tau isoform, tau441 (Additional file 1: Table S1) were mixed with streptavidin-APC or streptavidin-PE Recombinant IgG expression and purification (Thermo Fisher) at a 35:1 molar ratio and free peptide Human IgG1 antibodies were constructed by cloning the removed using BioSpin 30 column (Biorad). PBMCs heavy (V )and light (V ) chain variable regions into a sin- H L corresponding to 3 donors were thawed and B cells were gle expression vector containing the wildtype IgG constant enriched by positive selection with CD22 magnetic regions. Plasmids encoding the sequences corresponding to beads (Miltenyi Biotec). B cells were subsequently human anti-tau mAbs were transiently transfected in labeled with extracellular markers IgG-FITC, CD19- human embryonic kidney 293-derived Expi293F™ cells PerCPCy5.5 and CD27-PECy7 (all from BD Biosciences) (Thermo Fisher) and 7 days post transfection, the expressed and incubated with labeled peptide baits. Doublets and antibodies were purified from the culture medium by + + dead cells were excluded and the CD19 , IgG , CD27hi, MabSelect SuRe (GE Healthcare) Protein A affinity chroma- antigen double-positive cells were collected by single cell tography. IgGs were eluted from the column with 100 mM sorting into PCR plates containing RT-PCR reaction buf- sodium citrate buffer, pH 3.5 which was immediately buffer fer and RNaseOUT (Thermo Fisher). Heavy and light exchanged into PBS, pH 7.4 using a self-packed Sephadex chain (HC/LC) antibody variable regions were recovered G-25 column (GE Healthcare). Mouse chimeric versions of using a two-step PCR approach from single sorted mem- CBTAU-27.1 and CBTAU-28.1 were generated cloning the ory B cells using a pool of leader specific (Step I) and variable light and heavy chain into an expression vector in framework specific primers (Step II PCR). Heavy and which the human Fc was replaced with murine Fc. Fab frag- light chain PCR fragments (380–400 kb) were linked via ments were obtained by papain digestion of antibodies an overlap extension PCR and subsequently cloned into (Thermo Fisher Scientific), followed by removal of the Fc a dual-CMV–based human IgG1 mammalian expression using MabSelect SuRe resin (GE Healthcare). Each antibody vector (generated in house). Finally, all recovered anti- was quality controlled by SDS-PAGE and size exclusion bodies were expressed and tested by ELISA against the chromatography coupled with multi angle light scattering tau peptides used in the sorting experiments to identify (SEC-MALS) and was further confirmed for reactivity to tau specific antibodies. cognate tau peptide by Octet biolayer interferometry. Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 4 of 17 Recombinant tau (rtau) expression and purification each in TBS-T. Peroxidase AffiniPure goat anti-human The gene encoding the human tau-441 isoform (2N4R) IgG (Fc-γ fragment specific; Jackson ImmunoResearch) extended with an N-terminal His-tag and a C-terminal was used as secondary antibody at a 1:4000 dilution in 2. C-tag and containing the C291A and C322A mutations 5% BSA in TBS-T and incubated for 1 h at room was cloned into a kanamycin resistant bacterial expres- temperature. The membrane was washed three times for sion vector and transformed into BL21 (DE3) cells. A 5 min and developed using the Supersignal West Pico 3 ml 2-YT Broth (Invitrogen) preculture containing kit (Pierce). Images were obtained on the ImageQuant 25 mg/l kanamycin was inoculated from a single bacter- LAS-4000 (GE Healthcare). ial colony for 6 h after which it was diluted to 300 ml for overnight growth in a 1-l shaker flask and subse- Qualitative association and dissociation measurements by quently diluted to 5 -l in a 10-l wave bag. Protein expres- Octet biolayer interferometry sion was induced when the culture reached an OD 1. The relative binding of the antibodies to tau peptides was 600nm 0 by addition of 2 mM IPTG. Three hours after induction, assessed by biolayer interferometry (Octet Red 384) the bacterial pellets were harvested by centrifugation, and measurements (ForteBio) [10]. Biotinylated tau peptides lysed with Bugbuster protein extraction reagent (Merck) were immobilized on Streptavidin (SA) Dip and Read supplemented with a protease inhibitor cocktail biosensors for kinetics (ForteBio). Real-time binding curves (cOmplete™ ULTRA tablets EDTA free, Roche). Purifica- were measured by applying the sensor in a solution contain- tion was performed by affinity chromatography using self- ing 100 nM antibody. To induce dissociation, the biosensor packed Ni-Sepharose (GE Healthcare) and C-Tag resin containing the antibody-tau peptide complex was immersed (Thermo Fisher Scientific) columns. in assay buffer without antibody. The immobilization of peptides to sensors, the association and the dissociation Reactivity of anti-tau human mAbs to tau peptides by ELISA steps, were followed in different ionic strength buffers Reactivity of recovered mAbs were tested against biotinyl- containing 10% FortéBio kinetics buffer as assay buffer. The ated tau peptides as previously described [37]. Briefly, tau relative association and dissociation kinetic curves were peptides were captured on streptavidin-coated plates compared to qualitatively assess the efficiency of antibody (Thermo Fisher Scientific) at 1 μg/ml in TBS and incu- binding to peptides encompassing different tau epitopes. bated for 2 h. Goat anti-human Fab (2 μg/ml, Jackson ImmunoResearch) to measure total IgG was used, and bo- Affinity measurements by Isothermal Titration vine actin (1 μg/ml TBS, Sigma) and irrelevant peptide Calorimetry (ITC) were used to confirm specificity of the purified mAbs. The affinities of antibodies for their corresponding tau ELISA plates were blocked and purified IgGs were diluted peptides were determined in solution on a MicroCal Auto- to 5 μg/ml in TBS/0.25% BSA and titrated (5-fold serial iTC200 system (Malvern). Peptides at concentrations of ~ dilutions) against peptides. Plates were subsequently 35 μM(CBTAU-27.1), ~10 μM(dmCBTAU-27.1), ~30 μM washed and goat anti-human IgG F(ab’)2 (Jackson Immu- (CBTAU-28.1) and ~ 33 μM (dmCBTAU-28.1) were titrated noResearch) was used at 1:2000 dilution as secondary. in 20 steps of 2 μl per step, except for dmCBTAU-27.1 Following incubation, plates were washed four times in where 40 steps of 1 μl were employed, in identical buffers TBS-T and developed with Sure Blue Reserve TMB containing 200 μM CBTAU-27.1, 130 μM dmCBTAU-27.1, Microwell Peroxidase Substrate (KPL) for approximately 205 μM CBTAU-28.1 and 205 μM dmCBTAU-28.1, 2 min. The reaction was stopped by the addition of TMB respectively. The thermodynamic parameters and the Stop Solution and the absorbance at 450 nm was equilibrium dissociation constants, Kd, were determined measured using an ELISA plate reader. upon fitting the ITC data to a model assuming a single set of binding sites corresponding to an antibody:tau peptide = Western blot analysis 1:2 binding model. Immunopurified PHF and sarkosyl-insoluble fraction of AD brain lysates were provided by Steven Paul at Weill Expression, crystallization, data collection, structure Cornell Medical College and prepared as described pre- determination, and refinement of CBTAU-27.1 Fab and viously [20]. Approximately 0.3 μg protein was resolved CBTAU-28.1 Fab on SDS-PAGE (4–12% Bis-Tris Novex NuPAGE gel; To express Fab fragments for crystallization, the gene Invitrogen) and subsequently transferred onto a nitrocel- sequences coding for the hinge, C 2, and C 3of human H H lulose membrane. The membrane was blocked overnight antibodies CBTAU-27.1 (IgG1κ) and CBTAU-28.1 (IgG1κ) in 1X TBS-T with 5% BSA (blocking buffer). CBTAU-27. were removed from the corresponding IgG expression 1 and CBTAU-28.1 were used at 25 μg/ml in 2.5% BSA vector and a His tag was added to the C-terminus of each in TBS-T and incubated for 2 h at room temperature. C 1. Both Fabs were produced by transient transfection of The membranes were then washed three times for 5 min FreeStyle 293-F cells and purified using a Ni-NTA column Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 5 of 17 followed by a size exclusion chromatography using a concentration of about 50 mg/ml. The Fab:peptide Superdex 75 or Superdex 200 column (GE Healthcare). complexes (dmCBTAU-27.1 Fab with tau peptide A8119 The purified CBTAU-27.1 and CBTAU-28.1 Fabs were and dmCBTAU-28.1 Fab with tau peptide A7731) were concentrated to ~ 10 mg/ml and ~ 8 mg/ml, respectively, subjected to crystallization screening by sitting drop vapor in 20 mM Tris, pH 8.0 and 150 mM NaCl for diffusion testing 2300 conditions, by mixing 0.1 μlprotein crystallization. Crystallization experiments were set up solution and 0.1 μl reservoir, and also varying the protein using the sitting drop vapor diffusion method. Initial concentration. The dmCBTAU-27.1 Fab with peptide crystallization conditions for CBTAU-27.1 Fab and its A8119 was crystallized from 0.10 M sodium citrate buffer, complex with peptide 2833–1 (HVPGGGSVQIVYKPV pH 5.0, 20.00% (w/v) PEG 8 K at a concentration of DLSKV), and CBTAU-28.1 in complex with peptide 17 mg/ml. The dmCBTAU-28.1 Fab with peptide A7731 W1805 (TEDGSEEPGSETSDAKSTPT-amide) were was crystallized from 22% (w/v) PEG5K MME, 0.10 M obtained from robotic crystallization trials using the HEPES buffer, pH 6.75, 0.20 M KCl at a concentration of robotic CrystalMation system (Rigaku) at The Scripps 50 mg/ml. For cryo-protection, crystals were briefly Research Institute. For co-crystallization, CBTAU-27.1 immersed in a solution consisting of 75% reservoir and and CBTAU-28.1 Fabs were mixed with peptides 2833–1 25% glycerol. X-ray diffraction data were collected at and W1805, respectively, in a molar ratio of 1:10 before temperature of 100 K at the Swiss Light Source. Data were screening. Crystals of apo-form CBTAU-27.1 Fab were integrated, scaled and merged using XDS [25]. The struc- obtained at 20 °C from a reservoir solution containing 0. ture was solved with MOLREP [43] and refined with 1 M Hepes, pH 7.5, 20% PEG8000, while crystals of REFMAC5 [44]. Manual model completion was carried CBTAU-27.1 Fab with 2833–1 were grown at 20 °C from out using Coot [14]. The quality of the final model was 0.1 M Hepes, pH 7.0, 0.1 M KCl, 15% PEG5000 MME. verified PROCHECK [29] and the validation tools avail- Crystals of CBTAU-28.1 with W1805 were obtained at able through Coot [14]. The dmCBTAU-27.1 diffraction 20 °C from 0.085 M Tris-HCl, pH 8.5, 0.17 M sodium data were indexed in space group C2 and the dmCBTAU- acetate, 25.5% PEG4000. Before data collection, the crys- 28.1 data in space group P2 2 2 . Data were processed 1 1 1 tals of CBTAU-27.1 Fab and CBTAU-28.1 Fab were using the programs XDS and XSCALE (see Additional file soaked in the reservoir solution supplemented with 25% 1: Table S3). The phase information necessary to deter- (v/v) and 15% (v/v) glycerol, respectively, for a few seconds mine and analyze the structure was obtained by molecular and then flash-frozen in liquid nitrogen. X-ray diffraction replacement using previously solved structure of data of CBTAU-27.1 Fab apo-form were collected to 1.9 Å dmCBTAU-27.1 and dmCBTAU-28.1 Fabs were used resolution at beamline 23ID-D at the Advanced Photon search models. Subsequent model building and refinement Source (APS). X-ray diffraction data of CBTAU-27.1 Fab was performed according to standard protocols with the with 2833–1 and CBTAU-28.1 with W1805 were collected software packages CCP4 and COOT. For calculation of R- to 2.0 and 2.1 Å resolution, respectively, at beamline 12–2 free, 6.2% of the measured reflections were excluded from at the Stanford Synchrotron Radiation Lightsource (SSRL). refinement (see Additional file 1: Table S3). TLS refine- HKL2000 (HKL Research, Inc.) was used to integrate and ment (using REFMAC5, CCP4) resulted in lower R-factors scale the diffraction data (Additional file 1: Table S2). The and higher quality of the electron density map for structures were determined by molecular replacement dmCBTAU-27.1-A8119 using five TLS groups. Waters with Phaser [35] using search models of a human antibody were added using the “Find waters” algorithm of COOT 2-23b3 Fab (PDB ID 3QOS) for CBTAU-27.1 Fab and a into Fo-Fc maps contoured at 3.0 sigma followed by re- human antibody 1-69b3 (PDB ID 3QOT) for CBTAU-28.1 finement with REFMAC5 and checking all waters with the Fab. The models were iteratively rebuilt using Coot [14] validation tool of COOT. The occupancy of side chains, and refined in Phenix [1]. Refinement parameters included which were in negative peaks in the Fo-Fc map (contoured rigid body refinement and restrained refinement including at − 3.0 σ), were set to zero and subsequently to 0.5 if a TLS refinement. Electron density for both peptides 2833– positive peak occurred after the next refinement cycle. 1 and W1805 were clear and the peptides were built in The Ramachandran plot of the final model shows 88.7% the later stages of the refinement. Final refinement statis- (dmCBTAU-27.1) and 88.1% (dmCBTAU-28.1) of all resi- tics are summarized in Additional file 1: Table S2. dues in the most favored region, 10.8% (dmCBTAU-27.1) and 11.6% (dmCBTAU-28.1) in the additionally allowed Crystallization, data collection, and structure determination region, and 0.3% (dmCBTAU-27.1) and 0.0% (dmCBTAU- of dmCBTAU-27.1 Fab and dmCBTAU-28.1 Fab 28.1) in the generously allowed region. Residue Ala51(L) For crystallization, the dmCBTAU-27.1 and dmCBTAU- is found in the disallowed region of the Ramachandran 28.1 Fab fragments (in 20 mM HEPES buffer, pH 7.5, 7. plot (Additional file 1: Table S3) for both Fabs, as is 55 mM NaCl) were incubated with 4 mM of the respective frequently found in other Fabs [3] . This was either peptide on ice overnight and concentrated to a final confirmed from the electron density map or could not be Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 6 of 17 modelled in another more favorable conformation. Statis- and ThT. Kinetic measurements were monitored at 37 °C tics of the final structure and the refinement process are in a Biotek Synergy Neo2 Multi-Mode Microplate Reader listed in Additional file 1: Table S3. (Biotek, VT, USA) by measuring ThT fluorescence at 485 nm (20 nm bandwidth) upon excitation at Affinity maturation by phage display 440 nm (20 nm bandwidth) upon continuous shaking The coding sequence for scFv directed against CBTAU- (425 cpm, 3 mm). 28.1 epitope was cloned into an inducible prokaryotic expression vector containing the phage M13 pIII gene. Atomic force microscopy Random mutations were introduced in the scFv by error For each sample, 20 μl of rtau solution was deposited onto prone PCR (Genemorph II EZClone Domain Mutagen- freshly cleaved mica surface. After 3 min incubation, the esis kit) after which the DNA was transformed into surface was washed with double-distilled water and dried TG1bacteria. The transformants were grown to mid-log with air. Samples were imaged using the Scanasyst-air phase and infected with CT helper phages that have a protocol using a MultiMode 8-HR and Scanasyst-air genome lacking the infectivity domains N1 and N2 of silicon cantilevers (Bruker Corporation, Santa Barbara, protein pIII, rendering phage particles which are only in- USA). Height images of 1024 × 1024 pixels in size and fective if they display the scFv linked to the full-length surface areas of 10 × 10 μm were acquired under pIII [28]. Phage libraries were screened using magnetic ambient environmental conditions with peak force beads coated with rtau in immunotubes. To deselect frequency of 2 KHz. nonspecific binders, the tubes were coated with a tau peptide lacking the CBTAU-28.1 epitope. To ensure Assessment of CBTAU-27.1 binding to rtau and PHFs by maturation against the correct epitope, selection was SEC-MALS continued using beads coated with the cognate A6940 15 μM monomer or aggregated rtau were incubated with peptide. Eluted phages were used to infect XL1-blue F′ dmCBTAU-27.1 in a rtau:IgG1 = 1:0.6 ratio for 15 min E. coli XL1-blue F′ which were cultured and infected and samples were subsequently centrifuged for 15 min with CT-helper (or VCSM13) phages to rescue phages at 20,000 g. Same procedure was also applied for used for subsequent selection rounds. After three rounds controls containing only monomer rtau, aggregated rtau of panning, individual phage clones were isolated and or IgG. All samples were analyzed by SEC-MALS screened in phage ELISA for binding to rtau and cognate upon fractionation on a TSKgel G3000SWxl (Tosoh CBTAU-28.1 peptide A6940. Bioscience) gel filtration column equilibrated with 150 mM sodium phosphate, 50 mM sodium chloride In vitro tau aggregation assay at pH 7.0. at a flow rate of 1 ml/min. For molar mass Stock solutions of 500 μMthioflavin T (ThT)(Sigma- determination, in-line UV (Agilent 1260 Infinity MWD, Aldrich, St Louis, MO, USA) and 55 μM heparin Agilent Technologies), refractive index (Optilab T-rEX, (Mw = 17–19 kDa; Sigma-Aldrich, St Louis, MO, USA) Wyatt Technology) and 8-angle static light scattering were prepared by dissolving the dry powders in reaction (Dawn HELEOS, Wyatt Technology) detectors were used. buffer (0.5 mM TCEP in PBS, pH 6.7), and filtered through a sterile 0.22 μm pore size PES membrane filter Immunohistochemistry (Corning, NY, USA) or a sterile 0.22 μm pore size PVDF Brain samples were obtained from The Netherlands Brain membrane filter (Merck Millipore, Tullagreen, Cork, IRL), Bank (NBB), Netherlands Institute for Neuroscience, respectively. The concentration of the ThT solution was Amsterdam. All donors had given written informed consent determined by absorption measurements at 411 nm using for brain autopsy and the use of material and clinical infor- − 1 − 1 an extinction coefficient of 22,000 M cm .The mation for research purposes. Neuropathological diagnosis huTau441 concentration was determined by absorption was assessed using histochemical stains (haematoxylin and measurements at 280 nm using an extinction coefficient eosin, Bodian and/or Gallyas silver stains [41] and immuno- − 1 − 1 of 0.31 ml mg cm . For spontaneous conversions, histochemistry for Amyloid beta, p-tau (AT8), α-synuclein, mixtures of 15 μM huTau441 in 200 μl reaction buffer TDP-43 and P62, on formalin-fixed and paraffin-embedded containing 8 μM heparin and 50 μM ThT were dispensed tissue from different parts of the brain, including the frontal in 96-well plates (Thermo Scientific, Vantaa, Finland) that cortex (F2), temporal pole cortex, parietal cortex (superior were subsequently sealed with plate sealers (R&D Systems, and inferior lobule), occipital pole cortex, amygdala and the Minneapolis, MN). For seeding experiments, preformed hippocampus, essentially CA1 and entorhinal area of the seeds were added to the wells before sealing the plate. To parahippocampal gyrus. Staging of pathology was assessed assess the effect of IgG or Fab on the conversion, according to Braak and Braak for tau pathology and Thal huTau441 and IgG or Fab were mixed and incubated for for Amyloid beta pathology [5, 40]. Formalin-fixed and par- 20 min in reaction buffer before the addition of heparin affin embedded tissue sections (5 μm thick) were mounted Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 7 of 17 on Superfrost Plus tissue slides (Menzel-Gläser, Germany) were added to the beads together with 90 μl of the 1:1 and dried overnight at 37 °C. Sections were deparaffinised antibody-brain extract mixture. Samples were incubated and subsequently immersed in 0.3% H O in phosphate- over night at 4 °C, rotating at 5 rpm. The following day, 2 2 buffered saline (PBS) for 30 min to quench endogenous the immunodepleted fractions were separated from the peroxidase activity. Sections were either treated in sodium beads by pulling down the beads with the magnet, trans- citrate buffer (10 mM sodium citrate, pH 6.0) heated by ferred to a new 96-well PCR plate and stored at − 80 °C autoclave (20 min at 130 °C) for antigen retrieval or proc- until tested. Each condition was tested in duplicate. essed without heat pretreatment. Between the subsequent Immunodepleted fractions were incubated for 10 min with incubation steps, sections were washed extensively with Lipofectamine 2000 (Invitrogen) in Opti-MEM (Gibco) in PBS. Primary antibodies were diluted in antibody diluent a 96-well cell culture plate (Greiner Bio-one) before 5.5 × (Immunologic) and incubated overnight at 4 °C. Secondary 10 HEK biosensor cells (provided by M. Diamond, EnVison™ HRP goat anti-rabbit/mouse antibody Washington University School of Medicine) were added HRP (EV-GαM , DAKO) was incubated for 30 min at room to each well. After a 2-day incubation at 37 °C, cells were temperature (RT). 3,3-Diaminobenzidine (DAB; DAKO) washed twice with PBS, detached using Trypsin/EDTA was used as chromogen. Sections were counterstained with (Gibco) and transferred to a polypropylene round bottom haematoxylin to visualize the nuclei of the cells, dehydrated plate (Costar) containing FACS buffer (Hank’s Balanced and mounted using Quick-D mounting medium (BDH Salt Solution (Sigma), 1 mM EDTA (Invitrogen), 1% FBS Laboratories Supplies, Poole, England). For the Gallyas (Biowest)). Cells were then analyzed for FRET positivity silver staining, 30 μm thick sections were rinsed in distilled by flow cytometry using a FACS Canto II (BD Bioscience). water and incubated in 5% periodic acid for 30 min at RT, Each plate contained a brain extract only condition (to followed by an incubation in silver iodide solution assess baseline FRET response) and an antibody isotype (4% sodium hydroxide, 10% potassium iodide and 0.35% control. Results are reported as normalized values, relative silver nitrate in distilled water) for 30 min at RT. Subse- to condition without antibody. quently, sections were washed in 0.5% acetic acid and de- veloped with developer working solution (10 volumes 5% Microglia assay sodium carbonate solution, 3 volumes solution 0.2% am- Aggregated recombinant 2N4R tau (rtau) was generated monium nitrate, 0.2% silver nitrate and 1% Tungstosilicic in the absence of ThT under the conditions described for acid solution, and 7 volumes 0.2% ammonium nitrate, 0. the in vitro tau aggregation assay and covalently labelled 2% silver nitrate, 1% Tungstosilicic acid and 0.3% formal- with pHrodo® Green STP Ester (Invitrogen) following dehyde solution. After color development, sections were manufacturer’s instructions. Briefly, rtau aggregates were rinsed in 0.5% acetic acid, after which sections were incu- spun down by centrifugation at 20800 rcf for 30 min and bated in 5% sodium thiosulphate and rinsed in distilled then resuspended in 0.1 M sodium bicarbonate buffer, water. Stained sections were mounted on coated glass pH 8.5, at a final concentration of 2 mg/ml. Efficiency of slides (Menzel-Gläser) and dried for at least 2 h at 37 °C. aggregation was assessed by detecting presence of tau in Subsequently sections were fixed in ethanol 70% for the supernatant using SEC-MALS. Prior to labeling, tau 10 min, counterstained with hematoxylin, dehydrated and aggregates were briefly sonicated. Ten moles of dye were mounted with Quick D mounting medium. added per mole of protein and the mixture was incubated for 45 min at room temperature, protected from light. FRET based cellular immunodepletion assay Unconjugated dye was removed using a PD10 column Cryopreserved brain tissue was acquired from the (GE Healthcare) equilibrated with 0.1 M sodium bicar- Newcastle Brain Tissue Resource biobank. Frozen brain bonate buffer pH 8.5 and eluting the protein with the tissue samples from 17 AD patients were homogenized same buffer. Eluted fractions were evaluated for their in homogenization buffer (10 mM Tris (Gibco), 150 mM protein content by BCA assay (Thermo Fisher Scientific) NaCl (Gibco) containing protease inhibitors (cOmplete™ following the manufacturer’s instructions. Protein contain- ULTRA tablets EDTA free, Roche) to obtain a 10% (w/v) ing fractions were pooled, aliquoted and stored at − 20 °C. pooled brain homogenate. Individual antibody dilutions BV-2 cells were cultured in DMEM supplemented with were prepared in PBS pH 7.4 (Sigma), mixed with brain 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin extract in a 1:1 ratio in a 96 well PCR plate (Thermo and 2 mM L-Glutamine. Cultures were maintained in Scientific), and incubated until the beads were washed. humidified atmosphere with 5% CO2 at 37 °C. In order to Protein-G DynaBeads (Life Technologies) were added in generate immunocomplexes, 250 nM aggregated rtau, a 96-well PCR plate (Thermo Scientific) and washed covalently labelled with pHrodo Green dye, was incubated twice with PBS, 0.01% Tween20 (Sigma) by pulling down with a serial dilution (12.5–150 nM) of a chimeric version the beads with a magnet (Life Technologies). Wash buffer (mouse Fc region) of CBTAU-28.1 (parental and high was removed completely and 10 μlofPBS, 0.1% Tween20 affinity mutant) or CBTAU-27.1 (parental and high affinity Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 8 of 17 mutant) in serum-free medium. Tau immunocomplexes 52–71 for CBTAU-28.1 (Additional file 1:FigureS1).The were also generated with 300 nM Fab fragments of both specificity of these antibodies for tau was confirmed by CBTAU-28.1 and CBTAU-27.1, in the parental and high Western blot (Additional file 1:Figure S2). affinity mutant format. In each experiment, a mouse IgG1 isotype control was included together with cells incubated Recognition of a structurally identical, germline encoded with only aggregated rtau. Immunocomplexes were incu- hotspot motif bated over night at 4 °C and the day after applied to BV2 Crystal structures of the Fab fragments of CBTAU-27.1 and cells for 2 h at 37 °C with 5% CO . During the incubation, CBTAU-28.1 in complex with tau peptides spanning resi- antibody-independent tau uptake was prevented by block- dues 299–318 and 52–71 were determined at 2.0 and 2.1 Å ing the Heparan Sulfate Proteoglycan Receptor with resolution, respectively (Fig. 1b and c, Additional file 1: 200 μg/ml Heparin. After incubation, cells were harvested Table S2). The structures reveal that an intriguing similarity with 0.25% trypsin-EDTA for 20 min thus simultaneously exists in the way they bind despite recognition of very removing tau bound to the extracellular membrane, cen- distinct regions on the tau protein (Fig. 1d). Both light trifuged at 400 rcf to remove medium, washed twice with chains harbor a pocket made of aromatic tyrosine or PBS, and resuspended in flow cytometry buffer (PBS 1× phenylalanine side chains that form a binding site for a plus 0.5% BSA and 2 mM EDTA). Cells were analyzed proline residue in the N-terminal region of the different with a Canto II flow cytometer (BD) gating for live single peptides. This interaction is further stabilized by a peptide cell population, as identified by forward and side scatter backbone hydrogen bond to LCDR3 Phe (CBTAU-27.1) profiles. Results are reported as geometric mean fluores- and LCDR3 Tyr (CBTAU-28.1). Similarly, the two heavy cent intensities. Each experiment was performed twice. chains interact with a lysine in the peptide C-terminal For the microscopy experiments, cells were seeded in 96- region that involves identical hydrogen bonding networks well μClear® plate (Greiner Bio-one). After incubation with with two HCDR2 aspartates and the backbone of HCDR3 97 103 the immunocomplexes, nuclei were stained with Hoechst Ala (CBTAU-27.1) and Ser (CBTAU-28.1), respectively. (Sigma) and the acidic cellular compartment with Lyso- Both lysines are flanked by HCDR1 Trp and HCDR2 Tracker Red dye (Thermo Fisher). Live-cell imaging was Tyr that align and stabilize the aliphatic part of the Lys performed using the Opera Phenix™ High Content Screen- side chains. However, the central parts of the tau epitopes ing System (PerkinElmer) with temperature set to 37 °C differ significantly (see Fig. 1d, column 3). The four-residue and in presence of 5% CO . For high quality images, a 63× central region adopts an extended structure in CBTAU-27.1 water immersion objective was used and 0.5 μmplanes and inserts Leu into a pocket formed by LCDR3, (20 per Z-stack) were acquired per imaged field. HCDR3, HCDR1 and HCDR2. In the same spatial location, the seven-residue central region of the CBTAU-28.1 epitope Results spans the same distance between the conserved proline and Identification of naturally occurring anti-tau antibodies in lysine residues by adopting a more compact helical struc- healthy donors ture (Fig. 1e). CBTAU-28.1 Asp makes a salt bridge with 6 59 102 In total, 2.6 × 10 memory B cells from nine healthy blood Arg (HCDR2) and a hydrogen bond with Tyr (LCDR3). donors aged 18–65 years were interrogated against a pool These two antibodies thus recognize a Pro – X – Lys of 10 overlapping peptides spanning the length of tau441 motif in different tau peptides, where n is from 4 up to at (Additional file 1: Table S1). Ninety-two tau-reactive B least 7 amino acids. The proline and lysine binding pockets cells were sorted and 30 heavy and light variable chain are germline-encoded and specificity towards one or other sequences were recovered and full-length IgGs were epitope arises from the CDR3 loops, which interact with cloned and expressed. Two unique tau binding antibodies, the X region (Fig. 1d). CBTAU-27.1 and CBTAU-28.1, which are both derived The CBTAU-27.1 epitope encompasses residues 310 317 from the V 5–51 and V 4–1 germline families, were iden- YKPVDLSK in the R3 domain (Fig. 1f). The R3 H L tified. Both antibodies carry high numbers of somatic mu- domain is part of the core of PHFs [12, 15], where a 306 311 tations with 38 and 28 nucleotide substitutions for the hexapeptide VQIVYK is crucial for PHF assembly heavy and light chains of CBTAU-27.1, and 19 and 16 for [12, 15, 32, 45]. Since the CBTAU-27.1 epitope overlaps the heavy and light chains of CBTAU-28.1, respectively. this hexapeptide, in particular the key Lys [32], we Since memory B cell selections were performed using a hypothesize that CBTAU-27.1 binding to tau could peptide pool, an ELISA-based binding assay with the 10 block the nucleation interface and thus prevent aggre- individual tau peptides was performed and established gation. The CBTAU-28.1 epitope encompasses residues 58 67 that CBTAU-27.1 and CBTAU-28.1 bound to peptide EPGSETSDAK in the first N-terminal insert (Fig. 1f). A6897 (residues 299–369) and peptide A6940 (residues Since the N- and C-terminal tau regions that surround 42–103), respectively. Further mapping narrowed the epi- the repeat domains have been shown to be disordered and tope regions to residues 299–318 for CBTAU-27.1 and project away from the PHF core to form a flexible fuzzy Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 9 of 17 coat [15], binding of CBTAU-28.1 is unlikely to interfere control and AD brain tissue (Fig. 3). CBTAU-27.1 did not with PHF formation, but may—like previously described show immunoreactivity in either control or AD cases, antibodies [6, 42, 48, 49]—hamper the spreading of aggre- whereas dmCBTAU-27.1 showed immunoreactivity in the gates after they are formed. However, the affinities of both cytosol of neurons of the control cases and clear recogni- antibodies, at least to their cognate tau peptides, are in the tion of aggregated tau in neuropil threads and NFTs in high nanomolar range (Additional file 1: Figure S3A and AD brains, but only after heat pretreatment which is a B), which may limit their functional activity. Therefore, we routine ‘antigen retrieval’ procedure to recover reactivity set out to generate affinity-improved mutants of CBTAU- in formalin-fixed paraffin-embedded tissue sections. No 27.1 and CBTAU-28.1 by employing a combination of immunoreactivity of CBTAU-28.1 was detected in the rational design and random mutagenesis approaches. control cases, whereas dmCBTAU-28.1 showed diffuse immunoreactivity of neurons after heat pretreatment. In Affinity-improved antibodies retain specificity AD brains, both CBTAU-28.1 and dmCBTAU-28.1 recog- For CBTAU-27.1, we used a rational structure-based nized neuropil threads and NFTs regardless of the sample approach through analysis of the co-crystal structure treatment. CBTAU-28.1 thus recognizes PHFs without (Fig. 2a). LCDR3 Thr was identified as one location heat pretreatment whereas CBTAU-27.1, even in its high where additional hydrophobic interactions could be formed affinity variant, requires heat pretreatment to recognize without affecting the structure of the tau peptide. Isoleucine pathologic tau. This is in line with the epitope of CBTAU- introduced at this position better filled the gap between 27.1 being buried within the PHFs and becoming exposed 313 315 58 Val ,Leu and the aliphatic portion of V Arg .In upon heat pretreatment. The diffuse neuronal immunore- 27D LCDR1, Ser was mutated to tyrosine to remove the activity of dmCBTAU-27.1 and dmCBTAU-28.1 observed unfavorable contact between the serine hydroxyl and the in control brain tissue after heat pretreatment shows that proline pyrrolidine sidechain and create additional these antibodies bind to physiological tau. The observed hydrophobic interactions. These two mutations improved immunoreactivity under identical conditions shows a clear theaffinitybymorethan50-fold to thelow nanomolar improvement in the detection of tau by affinity-improved range (Fig. 2c and Additional file 1:FigureS3).For CBTAU- antibodies relative to parental antibodies without affecting 28.1, analysis of the structure did not reveal potential specificity (Fig. 3). Similar results were obtained with affinity-improving mutations and, therefore, a random mu- immunohistochemical staining on post-mortem brain tagenesis strategy was employed (Fig. 2b). This approach tissue of other tauopathies like frontotemporal lobar 32 35 led to the identification of Ser ➔Arg and Glu ➔ Lys degeneration (FTLD), Pick’s disease, progressive supra- mutations in the light chain that combined led to an ~ nuclear palsy (PSP) and primary age-related tauopathy 4-fold improvement in affinity compared to the parental (PART) cases (Additional file 1: Figure S5). Both CBTAU- antibody (Fig. 2d and Additional file 1:FigureS3). 27.1 and CBTAU-28.1 recognize pathological tau struc- Co-crystal structures of the Fab fragments of the CBTAU- tures in all these diseases, but detection by CBTAU-27.1 is 27D 94 27.1 double mutant Ser ➔ Tyr / Thr ➔ Ile (from here dependent on heat pretreatment to make its epitope on referred to as dmCBTAU-27.1) and the CBTAU-28.1 accessible. Furthermore, the detection of tau is improved 32 35 double mutant Ser ➔ Arg / Glu ➔ Lys (from here on for the affinity-improved antibodies. referred to as dmCBTAU-28.1) in complex with peptides A8119 and A7731, respectively, were determined at 3.0 and 2.85 Å resolution (Additional file 1: Table S3). Alignment of Binding domain-dependent functional activities the structures of the double mutants in complex with their The observation that the epitope of CBTAU-27.1 forms tau epitopes to the corresponding parental antibody co- part of the core of the PHFs led us to consider that crystal structures showed that both dmCBTAU-27.1 and CBTAU-27.1 might be able to prevent aggregation of tau dmCBTAU-28.1 retained the binding mode of the parental by inhibiting the initial nucleation step. While the antibody (Fig. 2e and f), with RMSD values for the peptide molecular mechanism of tau aggregation is not fully Cα atoms of 0.44 Å and 0.24 Å, respectively. The similarity understood, the current paradigm is that it follows a between the double mutants and their parental antibodies nucleation-dependent polymerization (NDP) process regarding the nature of their interactions with tau was con- [2, 11, 23, 39]. An NDP mechanism is characterized by an firmed by biolayer interferometry using buffers of different initial nucleation step (nuclei formation) followed by an ionic strengths (Additional file 1:FigureS4).Furthermore, exponential growth step (fibril elongation). Nucleation, the binding of the different antibodiestosetsoftau peptides rate-limiting step of the aggregation process, is a stochastic was assessed to confirm conservation of the specificity phenomenon and refers to the formation of high energy (Additional file 1:FigureS4). nuclei. Once nuclei are formed, they rapidly recruit tau The tau specificity of the antibodies was further assessed monomer (growth step), and convert into thermodynamic- by immunohistochemical staining on post-mortem ally stable aggregates. These aggregates can undergo Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 10 of 17 312 313 Fig. 2 Generation of affinity-improved mutants of CBTAU-27.1 and CBTAU-28.1a Structure-based design of mutants around Pro (left panel) and Val 312 94 (right panel). The tau epitope is illustrated as in Fig. 1b. Antibody loops and the key residues interacting with Pro and Thr are plotted in white. Proposed 27D mutations are shown as orange sticks on top of the corresponding wild-type side chains. Ser is mutated to tyrosine (left panel) to 312 94 313 enlarge the hydrophobic pocket of Pro ,and Thr is mutated to isoleucine (right panel) to fill the empty cavity surrounding Val and Leu . By introducing both mutations, additional hydrophobic contacts between tau and the antibody loops could be formed, potentially resulting in a lower desolvation penalty and increased affinity. b Schematic representation of the CBTAU-28.1 affinity maturation process by random mutagenesis. Mutations were introduced randomly by error prone PCR in the coding sequence for the single-chain variable fragment (scFv) directed against the CBTAU-28.1 epitope. M13 phage libraries displaying the scFv were screened against rtau and peptide A6940. Affinity-matured variants were identified by phage ELISA and converted into an IgG1 format to assess binding in solution. c and d Association and dissociation profiles for parental and affinity improved CBTAU-27.1 (c) and CBTAU-28.1 (d) variants to their corresponding cognate peptides as determined by Octet biolayer interferometry. Affinities as determined by ITC (K ) are shown on the individual graphs. (e and f) Co-crystal structures of the Fabs of dmCBTAU-27.1 (e) and dmCBTAU-28.1 (f) with tau peptides A8119 and A7731, respectively. Antibodies are illustrated as molecular surfaces (colored as in panel A), together with tau epitopes as sticks with yellow carbons. The corresponding parental co-crystal structures have been aligned using their variable regions, and their tau epitopes are shown as blue mesh on top of the mutant epitopes. fragmentation generating more fibril ends that are capable to as “seeding”. To assess whether CBTAU-27.1 can inter- of recruiting tau monomers and converting them into de fere with the aggregation of tau, we have developed a robust novo fibrils. This process is in most general terms referred and highly reproducible in vitro assay that monitors the Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 11 of 17 Fig. 3 Detection of immunoreactivity in human brain tissue by CBTAU-27.1 and CBTAU-28.1 and affinity-improved variants. Immunohistochemistry was performed on 5 μm thick formalin-fixed paraffin embedded sections of the hippocampal region using a 0.1 μg/ml antibody concentration. Immunodetection using CBTAU-27.1 (a-d), dmCBTAU-27.1 (e-h), CBTAU-28.1 (i-l), dmCBTAU-28.1 (m-p) and PHF-tau-specific mouse antibody AT8 (q-t) in control and AD brain tissue without or with heat pretreatment using sodium citrate buffer. Gallyas staining for detection of NFTs and neuropil threads is shown from the same control and AD case of corresponding areas for comparison (u, v). Immunoreactivity was visualized using DAB (brown) and nuclei were counterstained with haematoxylin (blue). Representative areas of the CA1/subiculum of the hippocampus are shown. Scale bars represent 50 μm heparin-induced aggregation of full-length recombinant microscopy (AFM) (Additional file 1:FigureS6C) and were Tau441 (rtau) by thioflavin T (ThT) fluorescence. The extremely efficient in seeding de novo aggregation of tau aggregation behavior of tau in our assay fulfills the expected (Additional file 1:FigureS6D andE). features of an NDP: sigmoidal kinetic curves with a well- In agreement with our hypothesis, CBTAU-27.1 inhib- defined lag phase followed by exponential growth ending in ited tau aggregation, as reflected by longer lag phases a stationary phase (Additional file 1:FigureS6).The aggre- and lower final ThT fluorescence signal in the kinetic gation kinetics of tau were highlyreproducibleand dis- curves. This inhibitory effect was strongly enhanced played the expected concentration dependence (Additional after affinity improvement (Fig. 4a, and b, and file 1: Figure S6B). Furthermore, the obtained tau aggregates Additional file 1: Figure S7). To shed further light on the displayed PHF-like morphology as assessed by atomic force mechanism, we assessed the ability of dmCBTAU-27.1 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 12 of 17 Fig. 4 CBTAU-27.1, but not CBTAU-28.1, inhibits the aggregation of recombinant tau in vitro. Aggregation of rtau in the absence (black) or presence of CBTAU-27.1 (a), dmCBTAU-27.1 (b), Fab CBTAU-27.1 (c), Fab dmCBTAU-27.1 (d), CBTAU-28.1 (e), dmCBTAU-28.1 (f), Fab CBTAU-28.1 (g)orFab dmCBTAU-28.1 (h), as monitored continuously for 120 h by ThT fluorescence. Three different rtau-IgG (1:0.2 – red, 1:0.4 – blue, and 1: 0.6 – purple) and rtau-Fab (1:0.4 – red, 1: 0.8 – blue, and 1:1.2 – purple) stoichiometries were tested. Each condition was tested in quadruplicate and one representative curve is shown for each condition. For complete datasets, see Additional file 1: Figures S7-S10 to alter the tau conversion when added at later times sequestering monomeric tau (Additional file 1: Figure after initiating the aggregation. In all cases, our results S11). The hypothesis that CBTAU-27.1 targets mono- show that stoichiometric amounts of dmCBTAU-27.1 meric tau and does not interact with tau aggregates was can not only prevent tau aggregation, but also arrest it further confirmed by sedimentation experiments even in the exponential phase where significant amounts followed by SEC-MALS size determination, which indi- of seeds are already present, presumably by binding and cates that the antibody binds to two tau monomers Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 13 of 17 using its two Fab arms while not being able to co-sedi- ment with preformed tau aggregates (Additional file 1: Figure S12). In sum, these results confirm the initial hy- pothesis that an antibody that targets the key PHF inter- face of the monomeric tau can block its misfolding and aggregation. Both double mutant and parental CBTAU-28.1 showed comparable alterations in the aggregation kinetics (Fig. 4e,and f, and Additional file 1:Figure S8). While somewhat longer lag phases and lower end-point fluorescent signals were observed in the presence of anti- body, the effect did not seem to be dose dependent and the kinetics were strikingly irreproducible. Furthermore, visual inspection of reaction mixtures after 120 h incuba- tion revealed that CBTAU-28.1 induced formation of large polymeric structures (Additional file 1: Figure S13), sug- gesting it can crosslink tau aggregates. This notion is sup- ported by the fact that the Fabs of both parental and dmCBTAU-28.1 did not affect tau aggregation (Fig. 4g, and h, and Additional file 1: Figure S10). In contrast, CBTAU-27.1 and dmCBTAU-27.1 Fabs showed similar in- hibitory effects as their corresponding antibodies (Fig. 4c, Fig. 5 CBTAU-28.1, but not CBTAU-27.1, is capable of immunodepleting and d, and Additional file 1: Figure S9), emphasizing the seeds from AD brains. Residual seeding activity of human AD brain different mechanisms by which CBTAU-27.1 and homogenates following immunodepletion with different concentrations CBTAU-28.1 interfere with tau aggregation. of CBTAU-27.1 and dmCBTAU-27.1 (a)orCBTAU-28.1and dmCBTAU-28.1 (b) as measured by FRET signal in biosensor cells expressing the We next assessed the ability of the antibodies to bind microtubule repeat domains of tau (aa 243–375) fused either to yellow or tau aggregates and thus potentially block the propagation cyan fluorescent protein. Uptake of exogenous tau aggregates into the and spreading of tau pathology. We therefore incubated cells results in aggregation of the tau fusion proteins, which is detected human AD brain homogenate containing PHFs with the by FRET. As positive and negative controls, a human IgG1 chimeric antibodies and depleted the antibody-antigen complexes. version of murine anti-PHF antibody AT8 and anti-RSV-G antibody RSV-4.1 were taken along, respectively. For the controls, the same data The residual seeding capacity was assessed using a cell- are shown in plots A and B for visualization purposes. Error bars indicate based biosensor assay [24, 48]. In line with its inability to the SD of two independent experiments bind PHFs, CBTAU-27.1 did not reduce the seeding activ- ity of the AD brain homogenate, while dmCBTAU-27.1 showed only minor reduction and only at the highest con- CBTAU-27.1 and dmCBTAU-27.1, was confirmed by centration tested (Fig. 5a). In contrast, CBTAU-28.1, like confocal microscopy (Fig. 6b,and d, and Additional file 1: mouse anti-PHF antibody AT8, depletes seeding activ- Figure S14). ity from AD brain homogenate and this in vitro activ- ity is enhanced for dmCBTAU-28.1 (Fig. 5b). The observation that CBTAU-28.1 and dmCBTAU-28 Discussion can bind PHFs led us to explore the possibility that these Implications for therapy and vaccines antibodies may furthermore enhance the uptake of tau ag- By binding to the region that is critical for the aggrega- gregates by microglia, the resident macrophage cells of tion of tau and which forms the core of PHFs, CBTAU- central nervous system [17]. Indeed, CBTAU-28.1 and 27.1 prevents aggregation of tau in vitro. This functional dmCBTAU-28.1 promoted the uptake of aggregated rtau activity identifies its epitope as a potential target for im- into mouse microglial BV2 cells and the affinity-improved munotherapy and could perhaps allow earlier interven- antibody appeared to mediate tau uptake to a greater ex- tion than antibodies that inhibit the spreading of already tent (Fig. 6a). ThefactthatFabs of bothparentaland formed tau seeds [6, 42, 48, 49]. Evidence that interfering dmCBTAU-28.1 did not increase basal tau uptake indi- with tau aggregation through immunotherapy may be cates that the uptake is Fc mediated. As expected, possible is provided by murine antibody DC8E8 which CBTAU-27.1 and dmCBTAU-27.1 did not show activity in also targets monomeric tau, inhibits tau aggregation in this assay (Fig. 6c). Antibody-mediated tau uptake and vitro, and reduces tau pathology in a murine AD model localization of rtau aggregates in the endolysosomal com- [27]. An alternative approach could be the development of partment by CBTAU-28.1 and dmCBTAU-28.1, but not a small molecule drug targeting monomeric tau that Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 14 of 17 Fig. 6 CBTAU-28.1, but not CBTAU-27.1, enhances uptake of tau aggregates into microglial BV2 cells. a and c Aggregated recombinant tau was covalently labelled with pHrodo Green dye and incubated with chimeric versions (containing mouse instead of human Fc region) of CBTAU-28.1, dmCBTAU-28.1, CBTAU27.1, dmCBTAU-27.1, their Fab fragments, a mouse IgG1 isotype control antibody (IC), or no antibody (rtau). Immunocomplexes were subsequently incubated with BV2 cells and their uptake was assessed by flow cytometry as expressed by the geometric mean fluorescent intensity. Error bars indicatethe SD of two independent experiments. b and d Preformed pHrodo-Green labeled immunocomplexes of rtau with chimeric dmCBTAU-28.1 or dmCBTAU-27.1 (at a concentration of 150 nM) were incubated with BV-2 cells. After incubation, nuclei were stained with Hoechst (blue) and the acidic cellular compartment with LysoTracker Red dye and uptake of immunocomplexes was assessed by live-cell imaging. Images represent maximum intensity projections of a 20 planes Z-stack (0.5 μm planes) acquired with a 63× water immersion objective interacts with the CBTAU-27.1 epitope. This epitope does functional activities) to be used for different purposes at dif- not overlap with the microtubule binding motifs (Fig. 1f), ferent stages of disease. The fact that both antibodies inter- suggesting that such a drug may not interfere with the act with tau aggregates from different tauopathies normal function of physiological tau while preventing its (Additional file 1: Figure S5) suggests that they may hold aggregation. Furthermore, the epitope of CBTAU-27.1 is therapeutic potential for various related neurodegenerative specific for tau, and is not present on MAP2, a micro- diseases. Clearly, effective therapy would require binding to, tubule stabilizing protein closely related to tau [13]. Like and clearance of different tau aggregate species (e.g. previously described antibodies, CBTAU-28.1 may be able low and high molecular weight species) and assess- to inhibit the spread of tau pathology. In addition, it dem- ment of the ability of these antibodies to do so will onstrated an Fc-dependent enhancement of the uptake of be the subject of future studies. tau aggregates by microglial BV2 cells. It is probably able Both antibodies are derived from the V 5–51 and to enhance the uptake of aggregates because its epitope is V 4–1 germline families and bind their respective epi- distant from the core of the PHFs and remains accessible. topes through hotspot interactions that are remarkably In summary, CBTAU-27.1 binds an epitope crucial for alike, pointing towards a conserved structural motif in tau aggregation that becomes buried inside PHFs and there- tau that could not have been predicted from sequence fore inhibits aggregation by binding and sequestering tau. analysis alone. The motif containing proline and lysine However, it does not decrease seeding activity of previously separated by 4 to 7 amino acids, is commonly found in formed aggregates and is thus functional at earlier stages of tau, and appears nine times on 2N4R. The lack of som- tau aggregation. In contrast, CBTAU-28.1 binds PHFs, atic mutations around the hotspot proline and lysine cross-links tau aggregates and depletes seeding activity, but suggests these two key tau residues could be responsible does not affect initial tau aggregation. CBTAU-27.1 and for the initial recognition of the V 5–51 and V 4–1 H L CBTAU-28.1 (and their respective affinity-improved vari- germline combination. The same V 5–51 or V 4–1 H L ants) thus have complementary activities that may allow hotspot interactions can be separately found in other these antibodies (or drug modalities mimicking these crystallized antibody complexes [7, 19, 33, 36], but the Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 15 of 17 combination of the two germlines could be more preva- Ethics approval and consent to participate Brain samples were obtained from The Netherlands Brain Bank (NBB), lent against tau, as they seem to get triggered by the Pro Netherlands Institute for Neuroscience, Amsterdam. All donors had given – X – Lys motif abundantly present on the protein. written informed consent for brain autopsy and the use of material and clinical Identification and characterization of these antibodies information for research purposes. Whole blood from healthy male and female donors was obtained from the San Diego Blood Bank (ages 18–65 years) after may thus be exploited to develop antibody or drug regi- informed consent was obtained from the donors. mens for distinct phases of progression of tau pathology and pave the way towards a peptide-based tau vaccine, Competing interests by taking advantage of the apparent immunogenicity of The authors declare that they have no competing interests. the identified motif and presenting both hotspot residues in the right spacing and orientation. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional file Author details Janssen Prevention Center, Janssen Pharmaceutical Companies of Johnson Additional file 1: Figure S1. Peptide epitope mapping. Figure S2. & Johnson, Archimedesweg 6, 2333, CN, Leiden, the Netherlands. Janssen Reactivity of CBTAU-27.1 and CBTAU-28.1 to PHF-tau. Figure S3. Affinity Prevention Center, Janssen Pharmaceutical Companies of Johnson & of CBTAU-27.1, CBTAU-28.1 and their affinity-matured mutants for their Johnson, 3210 Merryfield Row, San Diego, CA 92121, USA. Janssen cognate tau peptides. Figure S4. Affinity-improved antibodies Neuroscience Discovery, Janssen Pharmaceutical Companies of Johnson & dmCBTAU-27.1 and dmCBTAU-28.1 retain both the nature and specificity Johnson, Turnhoutseweg 30, 2340 Beerse, Belgium. Molecular and Cellular of the interactions of the parental antibodies with tau. Figure S5. Detection Pharmacology, Discovery Sciences, Janssen Pharmaceutical Companies of of immunoreactivity in various tauopathies by CBTAU-27.1 and CBTAU-28.1 Johnson & Johnson, Turnhoutseweg 30, 2340 Beerse, Belgium. Department and affinity-improved variants. Figure S6. Set-up of an in vitro rtau of Integrative Structural and Computational Biology, The Scripps Research aggregation assay. Figure S7. Complete dataset for in vitro tau aggregation Institute, La Jolla, CA 92037, USA. Proteros Biostructures GmbH, in the absence or presence of CBTAU-27.1 and dmCBTAU-27.1. Figure S8. Bunsenstraße 7a, 82152 Planegg, Germany. Department of Pathology, Complete dataset for in vitro tau aggregation in the absence or Amsterdam Neuroscience, VU University Medical Center, De Boelelaan 1117, presence of CBTAU-28.1 and dmCBTAU-28.1. Figure S9. Complete dataset 1081, HV, Amsterdam, the Netherlands. Skaggs Institute for Chemical for in vitro tau aggregation in the absence or presence of Fab-CBTAU-27.1 Biology, The Scripps Research Institute, La Jolla, CA 92037, USA. Department and Fab-dmCBTAU-27.1. Figure S10. Complete dataset for in vitro of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington tau aggregation in the absence or presence of Fab-CBTAU-28.1 and Avenue, Boston, MA 02115, USA. Department of Neurology, Amsterdam Fab-dmCBTAU-28.1. Figure S11. In vitro tau aggregation in the presence of Neuroscience, Academic Medical Center, Meidreefberg 9, 1105, AZ, dmCBTAU-27.1 added at different time points. Figure S12. Assessment of Amsterdam, the Netherlands. Present address: Janssen R&D US, 3210 CBTAU-27.1 binding to rtau and PHFs by SEC-MALS. Figure S13. Merryfield Row, San Diego, CA 92121, USA. Janssen Vaccines and Macroscopic image of rtau fibrils generated in the absence and presence of Prevention, Janssen Pharmaceutical Companies of Johnson and Johnson, CBTAU-28.1. Figure S14. Tau aggregates are internalized by BV-2 cells and Archimedesweg 6, Leiden, CN 2333, the Netherlands. localize in cellular acidic organelles. Table S1. Names and sequences of tau peptides used in this study. The first 10 peptides listed were used as Received: 6 April 2018 Accepted: 7 May 2018 baits in the BSelex method. Table S2. Data collection and refinement statistics for CBTAU-27.1 Fab and CBTAU-28.1 Fab. Table S3. Data collection and refinement statistics for dmCBTAU-27.1 - A8119 and References dmCBTAU-28.1 - A7731 complexes. (DOCX 30952 kb) 1. Afonine PV, Grosse-Kunstleve RW, EcholsN,Headd JJ,MoriartyNW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH, Adams PD (2012) Towards automated crystallographic structure refinement with phenix.Refine. Acta Crystallogr Sect D Biol Crystallogr 68:352–367. https://doi.org/10.1107/S0907444912001308 Acknowledgements 2. Apetri AC, Vanik DL, Surewicz WK (2005) Polymorphism at residue 129 modulates We would like to thank Mohammed Drissi Saidi, Hector Quirante, Başak Kükrer, the conformational conversion of the D178N variant of human prion protein 90- Otto Diefenbach, Tariq Nahar and Imke Sprengers, for protein generation and 231. Biochemistry 44:15880–15888. https://doi.org/10.1021/bi051455+ analysis, Alberto Carpinteiro Soares and Tjado Morrema for technical assistance, 3. Arevalo JH, Stura EA, Taussig MJ, Wilson IA (1993) Three-dimensional Frederique Bard and Louis de Muynck for valuable comments and advice. structure of an anti-steroid Fab' and progesterone-Fab' complex. J Mol Biol Human brain tissue for the immunodepletion experiments performed in this 231:103–118. https://doi.org/10.1006/jmbi.1993.1260 study was provided by the Newcastle Brain Tissue Resource which is funded in 4. Billingsley ML, Kincaid RL (1997) Regulated phosphorylation and dephosphorylation part by a grant from the UK Medical Research Council (G0400074), by NIHR of tau protein: effects on microtubule interaction, intracellular trafficking Newcastle Biomedical Research Centre and Unit awarded to the Newcastle and neurodegeneration. Biochem J 323(Pt 3):577–591 upon Tyne NHS Foundation Trust and Newcastle University, and as part of the 5. Braak H, Braak E (1991) Neuropathological stageing of Alzheimer-related Brains for Dementia Research Programme jointly funded by Alzheimer’s changes. Acta Neuropathol 82:239–259 Research UK and Alzheimer’sSociety. 6. Bright J, Hussain S, Dang V, Wright S, Cooper B, Byun T, Ramos C, Singh A, Parry G, Stagliano N, Griswold-Prenner I (2015) Human secreted tau increases amyloid-beta production. Neurobiol Aging 36:693–709. https://doi.org/10. Authors’ contributions 1016/j.neurobiolaging.2014.09.007 Project design by AA, GP, JW, and JG; aggregation assay and in vitro 7. Chen X, Zhao, Y., Harlos, K., Snir, O., Sollid, L.M. (2017) Crystal structure of antibody screening by AA and RC; antibody discovery, optimization and anti-gliadin 1002-1E03 Fab fragment in complex of peptide PLQPEQPFP. protein expression and purification by JJ, GP, RJ, EK, TH, JW, HV, BS, EB, DZ, DOI102210/pdb5ijk/pdb. doi:https://doi.org/10.2210/pdb5ijk/pdb DM and AA; FRET based cellular immunodepletion assay by MB, KD, KK, and 8. Cleveland DW, Hwo SY, Kirschner MW (1977) Physical and chemical MM, aggregation, labeling and microglia assay RT, DM, RC, JA, and AA; properties of purified tau factor and the role of tau in microtubule biophysical characterization by AA, RC, BS, JA, HI, MW; immunohistochemical assembly. J Mol Biol 116:227–247 analysis by JJH and KU, X-ray work and analysis by MM, SS, HZ, XZ, WY and 9. Cleveland DW, Hwo SY, Kirschner MW (1977) Purification of tau, a microtubule- IAW; statistical analysis by MK, and manuscript written by WK, EJMS, and AA. associated protein that induces assembly of microtubules from purified tubulin. All authors read and approved the final manuscript. J Mol Biol 116:207–225 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 16 of 17 10. Concepcion J, Witte K, Wartchow C, Choo S, Yao D, Persson H, Wei J, Li P, 30. Lee VM, Balin BJ, Otvos L Jr, Trojanowski JQ (1991) A68: a major subunit of paired Heidecker B, Ma W, Varma R, Zhao LS, Perillat D, Carricato G, Recknor M, Du K, helical filaments and derivatized forms of normal tau. Science 251:675–678 Ho H, Ellis T, Gamez J, Howes M, Phi-Wilson J, Lockard S, Zuk R, Tan H (2009) 31. Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopathies. Annu Label-free detection of biomolecular interactions using BioLayer interferometry Rev Neurosci 24:1121–1159. https://doi.org/10.1146/annurev.neuro.24.1.1121 for kinetic characterization. Comb Chem High Throughput Screen 12:791–800 32. Li W, Lee VM (2006) Characterization of two VQIXXK motifs for tau fibrillization 11. Crespo R, Rocha FA, Damas AM, Martins PM (2012) A generic crystallization- in vitro. Biochemistry 45:15692–15701. https://doi.org/10.1021/bi061422+ like model that describes the kinetics of amyloid fibril formation. J Biol 33. Liao HX, Bonsignori M, Alam SM, McLellan JS, Tomaras GD, Moody MA, Chem 287:30585–30594. https://doi.org/10.1074/jbc.M112.375345 Kozink DM, Hwang KK, Chen X, Tsao CY, Liu P, Lu X, Parks RJ, Montefiori DC, 12. Daebel V, Chinnathambi S, Biernat J, Schwalbe M, Habenstein B, Loquet A, Ferrari G, Pollara J, Rao M, Peachman KK, Santra S, Letvin NL, Karasavvas N, Akoury E, Tepper K, Muller H, Baldus M, Griesinger C, Zweckstetter M, Yang ZY, Dai K, Pancera M, Gorman J, Wiehe K, Nicely NI, Rerks-Ngarm S, Mandelkow E, Vijayan V, Lange A (2012) Beta-sheet core of tau paired Nitayaphan S, Kaewkungwal J, Pitisuttithum P, Tartaglia J, Sinangil F, Kim JH, helical filaments revealed by solid-state NMR. J Am Chem Soc 134:13982–13989. Michael NL, Kepler TB, Kwong PD, Mascola JR, Nabel GJ, Pinter A, Zolla- https://doi.org/10.1021/ja305470p Pazner S, Haynes BF (2013) Vaccine induction of antibodies against a 13. Dehmelt L, Halpain S (2005) The MAP2/tau family of microtubule-associated structurally heterogeneous site of immune pressure within HIV-1 envelope proteins. Genome Biol 6:204. https://doi.org/10.1186/gb-2004-6-1-204 protein variable regions 1 and 2. Immunity 38:176–186. https://doi.org/10. 14. Emsley P, Lohkamp B, Scott WG, Cowtan K (2010) Features and 1016/j.immuni.2012.11.011 development of Coot. Acta Crystallogr Sect D Biol Crystallogr 66:486–501. 34. Mandelkow E, von Bergen M, Biernat J, Mandelkow EM (2007) Structural https://doi.org/10.1107/S0907444910007493 principles of tau and the paired helical filaments of Alzheimer's disease. 15. Fitzpatrick AWP, Falcon B, He S, Murzin AG, Murshudov G, Garringer HJ, Brain Pathol 17:83–90. https://doi.org/10.1111/j.1750-3639.2007.00053.x Crowther RA, Ghetti B, Goedert M, Scheres SHW (2017) Cryo-EM structures 35. McCoy AJ, Grosse-Kunstleve RW, Storoni LC, Read RJ (2005) Likelihood-enhanced of tau filaments from Alzheimer's disease. Nature 547:185–190. https://doi. fast translation functions. Acta Crystallogr Sect D Biol Crystallogr 61:458–464. org/10.1038/nature23002 https://doi.org/10.1107/S0907444905001617 16. Frost B, Jacks RL, Diamond MI (2009) Propagation of tau misfolding from 36. Ofek G, Zirkle B, Yang Y, Zhu Z, McKee K, Zhang B, Chuang GY, Georgiev IS, the outside to the inside of a cell. J Biol Chem 284:12845–12852. https://doi. O'Dell S, Doria-Rose N, Mascola JR, Dimitrov DS, Kwong PD (2014) Structural org/10.1074/jbc.M808759200 basis for HIV-1 neutralization by 2F5-like antibodies m66 and m66.6. J Virol 17. Ginhoux F, Lim S, Hoeffel G, Low D, Huber T (2013) Origin and 88:2426–2441. https://doi.org/10.1128/JVI.02837-13 differentiation of microglia. Front Cell Neurosci 7:45. https://doi.org/10.3389/ 37. Pascual G, Wadia JS, Zhu X, Keogh E, Kukrer B, van Ameijde J, Inganas fncel.2013.00045 H, Siregar B,PerdokG,Diefenbach O,Nahar T, SprengersI, Koldijk MH, 18. Goedert M, Spillantini MG, Jakes R, Rutherford D, Crowther RA (1989) Multiple der Linden EC, Peferoen LA, Zhang H, Yu W, Li X, Wagner M, Moreno isoforms of human microtubule-associated protein tau: sequences and V, Kim J, Costa M, West K, Fulton Z, Chammas L, Luckashenak N, localization in neurofibrillary tangles of Alzheimer's disease. Neuron 3:519–526 Fletcher L, Holland T, Arnold C, Anthony Williamson R, Hoozemans JJ, 19. Gorny MK, Sampson J, Li H, Jiang X, Totrov M, Wang XH, Williams C, O'Neal Apetri A, Bard F, Wilson IA, Koudstaal W, Goudsmit J (2017) T, Volsky B, Li L, Cardozo T, Nyambi P, Zolla-Pazner S, Kong XP (2011) Immunological memory to hyperphosphorylated tau in asymptomatic Human anti-V3 HIV-1 monoclonal antibodies encoded by the VH5-51/VL individuals. Acta Neuropathol 133:767–783. https://doi.org/10.1007/ lambda genes define a conserved antigenic structure. PLoS One 6:e27780. s00401-017-1705-y https://doi.org/10.1371/journal.pone.0027780 38. Sanchez C, Diaz-Nido J, Avila J (2000) Phosphorylation of microtubule-associated protein 2 (MAP2) and its relevance for the regulation of the neuronal 20. Greenberg SG, Davies P (1990) A preparation of Alzheimer paired helical cytoskeleton function. Prog Neurobiol 61:133–168 filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proc Natl Acad Sci U S A 87:5827–5831 39. Surewicz WK, Jones EM, Apetri AC (2006) The emerging principles of mammalian prion propagation and transmissibility barriers: insight from 21. Guo JL, Lee VM (2011) Seeding of normal tau by pathological tau studies in vitro. Acc Chem Res 39:654–662. https://doi.org/10.1021/ar050226c conformers drives pathogenesis of Alzheimer-like tangles. J Biol Chem 286: 40. Thal DR, Rub U, Orantes M, Braak H (2002) Phases of a beta-deposition in 15317–15331. https://doi.org/10.1074/jbc.M110.209296 the human brain and its relevance for the development of AD. Neurology 22. Guo JL, Narasimhan S, Changolkar L, He Z, Stieber A, Zhang B, Gathagan RJ, Iba M, 58:1791–1800 McBride JD, Trojanowski JQ, Lee VM (2016) Unique pathological tau conformers 41. Uchihara T (2007) Silver diagnosis in neuropathology: principles, practice from Alzheimer's brains transmit tau pathology in nontransgenic mice. J Exp Med and revised interpretation. Acta Neuropathol 113:483–499. https://doi.org/ 213:2635–2654. https://doi.org/10.1084/jem.20160833 10.1007/s00401-007-0200-2 23. Harper JD, Lansbury PT Jr (1997) Models of amyloid seeding in Alzheimer's disease and scrapie: mechanistic truths and physiological consequences of 42. Umeda T, Eguchi H, Kunori Y, Matsumoto Y, Taniguchi T, Mori H, Tomiyama T the time-dependent solubility of amyloid proteins. Annu Rev Biochem 66: (2015) Passive immunotherapy of tauopathy targeting pSer413-tau: a pilot study 385–407. https://doi.org/10.1146/annurev.biochem.66.1.385 in mice. Ann Clin Transl Neurol 2:241–255. https://doi.org/10.1002/acn3.171 43. Vagin A, .Teplyakov, A. (1997) MOLREP: an Automated Program for 24. Holmes BB, Furman JL, Mahan TE, Yamasaki TR, Mirbaha H, Eades WC, Belaygorod Molecular Replacement J Appl Cryst 30:1022–1025 L, Cairns NJ, Holtzman DM, Diamond MI (2014) Proteopathic tau seeding predicts 44. Vagin AA, Steiner RA, Lebedev AA, Potterton L, McNicholas S, Long F, tauopathy in vivo. Proc Natl Acad Sci U S A 111:E4376–E4385. https://doi.org/10. Murshudov GN (2004) REFMAC5 dictionary: organization of prior chemical 1073/pnas.1411649111 knowledge and guidelines for its use. Acta Crystallogr Sect D Biol 25. Kabsch W (2010) Xds. Acta Crystallogr Sect D Biol Crystallogr 66:125–132. Crystallogr 60:2184–2195. https://doi.org/10.1107/S0907444904023510 https://doi.org/10.1107/S0907444909047337 45. von Bergen M, Friedhoff P, Biernat J, Heberle J, Mandelkow EM, Mandelkow 26. Kfoury N, Holmes BB, Jiang H, Holtzman DM, Diamond MI (2012) Trans-cellular E (2000) Assembly of tau protein into Alzheimer paired helical filaments propagation of tau aggregation by fibrillar species. J Biol Chem 287:19440–19451. depends on a local sequence motif ((306)VQIVYK(311)) forming beta https://doi.org/10.1074/jbc.M112.346072 structure. Proc Natl Acad Sci U S A 97:5129–5134 27. Kontsekova E, Zilka N, Kovacech B, Skrabana R, Novak M (2014) Identification of structural determinants on tau protein essential for its 46. Weingarten MD, Lockwood AH, Hwo SY, Kirschner MW (1975) A protein factor pathological function: novel therapeutic target for tau immunotherapy essential for microtubule assembly. Proc Natl Acad Sci U S A 72:1858–1862 in Alzheimer's disease. Alzheimer's Res & Ther 6:45. https://doi.org/10. 47. Wu JW, Herman M, Liu L, Simoes S, Acker CM, Figueroa H, Steinberg JI, 1186/alzrt277 Margittai M, Kayed R, Zurzolo C, Di Paolo G, Duff KE (2013) Small misfolded tau species are internalized via bulk endocytosis and anterogradely and 28. Kramer RA, Cox F, van der Horst M, van der Oudenrijn S, Res PC, Bia J, retrogradely transported in neurons. J Biol Chem 288:1856–1870. https://doi. Logtenberg T, de Kruif J (2003) A novel helper phage that improves phage org/10.1074/jbc.M112.394528 display selection efficiency by preventing the amplification of phages 48. Yanamandra K, Kfoury N, Jiang H, Mahan TE, Ma S, Maloney SE, Wozniak DF, without recombinant protein. Nucleic Acids Res 31:e59 Diamond MI, Holtzman DM (2013) Anti-tau antibodies that block tau aggregate 29. Laskowski RA, MacArthur, M.W., Moss, D. S. & Thornton, J.M (1993) PROCHECK: seeding in vitro markedly decrease pathology and improve cognition in vivo. a program to check the stereochemicai quality of protein structures. J Appl Neuron 80:402–414. https://doi.org/10.1016/j.neuron.2013.07.046 Crystallogr 26:283–291 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 17 of 17 49. Yanamandra K, Patel TK, Jiang H, Schindler S, Ulrich JD, Boxer AL, Miller BL, Kerwin DR, Gallardo G, Stewart F, Finn MB, Cairns NJ, Verghese PB, Fogelman I, West T, Braunstein J, Robinson G, Keyser J, Roh J, Knapik SS, Hu Y, Holtzman DM (2017) Anti-tau antibody administration increases plasma tau in transgenic mice and patients with tauopathy. Sci Transl Med 9. https://doi.org/10.1126/scitranslmed.aal2029 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Acta Neuropathologica Communications Springer Journals
Loading next page...
 
/lp/springer_journal/a-common-antigenic-motif-recognized-by-naturally-occurring-human-vh5-YnTqA0iu0c
Publisher
BioMed Central
Copyright
Copyright © 2018 by The Author(s).
Subject
Biomedicine; Neurosciences; Pathology; Neurology
eISSN
2051-5960
D.O.I.
10.1186/s40478-018-0543-z
Publisher site
See Article on Publisher Site

Abstract

Misfolding and aggregation of tau protein are closely associated with the onset and progression of Alzheimer’sDisease (AD). By interrogating IgG memory B cells from asymptomatic donors with tau peptides, we have identified two somatically mutated V 5–51/V 4–1 antibodies. One of these, CBTAU-27.1, binds to the aggregation motif H L in the R3 repeat domain and blocks the aggregation of tau into paired helical filaments (PHFs) by sequestering monomeric tau. The other, CBTAU-28.1, binds to the N-terminal insert region and inhibits the spreading of tau seeds and mediates the uptake of tau aggregates into microglia by binding PHFs. Crystal structures revealed that the combination of V 5–51 and V 4–1 recognizes a common Pro-X -Lys motif driven by germline-encoded hotspot interactions while the specificity and L n thereby functionality of the antibodies are defined by the CDR3 regions. Affinity improvement led to improvement in functionality, identifying their epitopes as new targets for therapy and prevention of AD. Keywords: Alzheimer’s disease, Tau protein, Monoclonal antibody, Antigenic motif Introduction phosphorylation and dephosphorylation of several of these Intracellular neurofibrillary tangles (NFTs) consisting of has been shown to affect its interaction with tubulin and aggregated tau protein are a hallmark of Alzheimer’sdisease cytoskeleton function [4, 38]. Hyperphosphorylation of tau (AD) and other neurogenerative disorders, collectively is thought to lead to microtubule dissociation and assembly referred to as tauopathies [31]. Tau is a microtubule- of the normally disordered, highly soluble protein into β associated protein expressed predominantly in neuronal sheet-rich aggregated fibrils called paired helical filaments axons and promotes the assembly and stability of microtu- (PHFs) that make up NFTs [8, 30, 34]. While the molecular bules [9, 46]. It is expressed in the adult human brain as six mechanism of tau aggregation remains elusive, it is believed isoforms with zero, one or two N-terminal acidic inserts that its initial nucleation step is energetically unfavorable, (0N, 1N, or2 N) and either three or four microtubule- whereas the subsequent fibril growth follows an energetic- binding repeats (3R or 4R) [18]. The tau protein contains ally downhill landscape [2, 11, 23, 39]. Accumulating many potential phosphorylation sites and the regulated evidence indicates that these fibrils can transmit from cell to cell and spread tau pathology to distant brain regions by seeding the recruitment of soluble tau into de novo aggre- * Correspondence: AApetri@its.jnj.com gates [16, 21, 22, 26, 47]. Janssen Prevention Center, Janssen Pharmaceutical Companies of Johnson & Johnson, Archimedesweg 6, 2333, CN, Leiden, the Netherlands Full list of author information is available at the end of the article © The Author(s). 2018 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 2 of 17 Several monoclonal antibodies that inhibit the spread- of unphosphorylated tau peptides as baits (Fig. 1a, and ing of tau fibrils have been described and are being Additional file 1: Table S1 for peptide sequences). developed for antibody-based therapies [6, 42, 48, 49]. We have recently described the isolation of a panel of antibodies with such functional activity by interrogating Materials and methods the peripheral IgG memory B cells of healthy human Human PBMC isolation blood donors for reactivity to phosphorylated tau Wholeblood from healthymaleand femaledonors was peptides [37]. To expand the arsenal of potential targets obtained from the San Diego Blood Bank (ages 18–65 years) and include epitopes present in physiological tau, we after informed consent was obtained from the donors. used the BSelex technology in combination with a pool PBMCs were isolated on Ficoll-Paque Plus (GE Healthcare) Fig. 1 (See legend on next page.) Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 3 of 17 (See figure on previous page.) Fig. 1 Recovery and structural characterization of naturally occurring monoclonal antibodies to unphosphorylated tau epitopes from asymptomatic individuals. a BSelex method used to recover tau-specific memory B cells. PBMCs were prepared from asymptomatic blood bank donors, and mature CD22 B cells were positively selected with magnetic beads. Viable cells were stained with IgG-FITC, CD19-PerCPCy5.5, and CD27-PECy7, and with a pool of 10 overlapping unphosphorylated tau peptides spanning the longest tau isoform (relative position of each peptide along 2N4R tau indicated). + + + + + All peptides were present in the pool with an APC label as well as with a PE label and CD19 ,CD27 ,IgG ,APC ,PE cells were single-cell sorted on a Beckman Coulter MoFlo XDP. Antibody heavy and light variable chain sequences were recovered from single cells, cloned and expressed as full-length IgGs. b and c Co-crystal structures of Fab CBTAU-27.1 (b) and Fab CBTAU-28.1 (c) with tau peptides A8119 and A7731, respectively. Antibodies have been plotted as molecular surface with light chain in white and heavy chain in grey. Tau peptides are shown as cartoon with interacting amino acids plotted as sticks. Proline and lysine residues are plotted in green, amino acids in between these residues are colored in yellow and the termini in grey. Only the interacting antibody loops are outlined. d Key interactions with tau of CBTAU-27.1 (upper row) and CBTAU-28.1 (lower row). Key interacting residues are plotted as sticks, polar interactions are indicated with dotted lines, and the corresponding distances are indicated in Å. In the first panel, 312 59 interactions with Pro and Pro are compared where the proline binding pockets are visualized on a molecular surface. In the second panel, 317 67 315 65 interactions with Lys and Lys are compared. In panel 3, interactions around Leu and Asp in the central region of the epitopes are shown. e Structural basis for recognition of the Pro – X – Lys epitope motif. Epitopes of CBTAU-27.1 and CBTAU-28.1 are superimposed by aligning on the proline and lysine residues. The peptides present both residues in the same spatial orientation. In the central region (yellow), hydrophobic Leu in CBTAU-27.1 is replaced by hydrophilic and charged Asp in CBTAU-28.1. f Schematic representation of tau isoform 2N4R showing the epitopes of CBTAU-27.1 and CBTAU -28.1 (bold and underlined) and the surrounding sequences. Highlighted in grey and red are the microtubule binding motifs 306 311 and the hexapeptide VQIVYK which forms the N-terminal end of the core of PHFs, respectively. N1 and N2 indicate acidic inserts, P1 and P2 indi- cate proline-rich domains, and R1-R4 indicate microtubule-binding repeat domains and cryopreserved at 50 million cells per ml in 90% FBS Peptide synthesis and 10% DMSO. All peptides were synthesized at Pepscan BV (Lelystad, The Netherlands) or New England Peptide, Inc. using routine Fmoc-based solid phase strategies on automated synthe- Single cell sorting and recovery of heavy and light variable sizers. After acidic cleavage and deprotection, peptides were light antibody genes from tau-specific memory B cells purified by reversed phase HPLC, lyophilized and stored as Tau-peptide specific B cells were sorted and antibody powders at − 80 °C. Purity and identity were confirmed by chains were recovered as previously described [37]. LC-MS. Briefly, a panel of 10 biotinylated peptides spanning the longest tau isoform, tau441 (Additional file 1: Table S1) were mixed with streptavidin-APC or streptavidin-PE Recombinant IgG expression and purification (Thermo Fisher) at a 35:1 molar ratio and free peptide Human IgG1 antibodies were constructed by cloning the removed using BioSpin 30 column (Biorad). PBMCs heavy (V )and light (V ) chain variable regions into a sin- H L corresponding to 3 donors were thawed and B cells were gle expression vector containing the wildtype IgG constant enriched by positive selection with CD22 magnetic regions. Plasmids encoding the sequences corresponding to beads (Miltenyi Biotec). B cells were subsequently human anti-tau mAbs were transiently transfected in labeled with extracellular markers IgG-FITC, CD19- human embryonic kidney 293-derived Expi293F™ cells PerCPCy5.5 and CD27-PECy7 (all from BD Biosciences) (Thermo Fisher) and 7 days post transfection, the expressed and incubated with labeled peptide baits. Doublets and antibodies were purified from the culture medium by + + dead cells were excluded and the CD19 , IgG , CD27hi, MabSelect SuRe (GE Healthcare) Protein A affinity chroma- antigen double-positive cells were collected by single cell tography. IgGs were eluted from the column with 100 mM sorting into PCR plates containing RT-PCR reaction buf- sodium citrate buffer, pH 3.5 which was immediately buffer fer and RNaseOUT (Thermo Fisher). Heavy and light exchanged into PBS, pH 7.4 using a self-packed Sephadex chain (HC/LC) antibody variable regions were recovered G-25 column (GE Healthcare). Mouse chimeric versions of using a two-step PCR approach from single sorted mem- CBTAU-27.1 and CBTAU-28.1 were generated cloning the ory B cells using a pool of leader specific (Step I) and variable light and heavy chain into an expression vector in framework specific primers (Step II PCR). Heavy and which the human Fc was replaced with murine Fc. Fab frag- light chain PCR fragments (380–400 kb) were linked via ments were obtained by papain digestion of antibodies an overlap extension PCR and subsequently cloned into (Thermo Fisher Scientific), followed by removal of the Fc a dual-CMV–based human IgG1 mammalian expression using MabSelect SuRe resin (GE Healthcare). Each antibody vector (generated in house). Finally, all recovered anti- was quality controlled by SDS-PAGE and size exclusion bodies were expressed and tested by ELISA against the chromatography coupled with multi angle light scattering tau peptides used in the sorting experiments to identify (SEC-MALS) and was further confirmed for reactivity to tau specific antibodies. cognate tau peptide by Octet biolayer interferometry. Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 4 of 17 Recombinant tau (rtau) expression and purification each in TBS-T. Peroxidase AffiniPure goat anti-human The gene encoding the human tau-441 isoform (2N4R) IgG (Fc-γ fragment specific; Jackson ImmunoResearch) extended with an N-terminal His-tag and a C-terminal was used as secondary antibody at a 1:4000 dilution in 2. C-tag and containing the C291A and C322A mutations 5% BSA in TBS-T and incubated for 1 h at room was cloned into a kanamycin resistant bacterial expres- temperature. The membrane was washed three times for sion vector and transformed into BL21 (DE3) cells. A 5 min and developed using the Supersignal West Pico 3 ml 2-YT Broth (Invitrogen) preculture containing kit (Pierce). Images were obtained on the ImageQuant 25 mg/l kanamycin was inoculated from a single bacter- LAS-4000 (GE Healthcare). ial colony for 6 h after which it was diluted to 300 ml for overnight growth in a 1-l shaker flask and subse- Qualitative association and dissociation measurements by quently diluted to 5 -l in a 10-l wave bag. Protein expres- Octet biolayer interferometry sion was induced when the culture reached an OD 1. The relative binding of the antibodies to tau peptides was 600nm 0 by addition of 2 mM IPTG. Three hours after induction, assessed by biolayer interferometry (Octet Red 384) the bacterial pellets were harvested by centrifugation, and measurements (ForteBio) [10]. Biotinylated tau peptides lysed with Bugbuster protein extraction reagent (Merck) were immobilized on Streptavidin (SA) Dip and Read supplemented with a protease inhibitor cocktail biosensors for kinetics (ForteBio). Real-time binding curves (cOmplete™ ULTRA tablets EDTA free, Roche). Purifica- were measured by applying the sensor in a solution contain- tion was performed by affinity chromatography using self- ing 100 nM antibody. To induce dissociation, the biosensor packed Ni-Sepharose (GE Healthcare) and C-Tag resin containing the antibody-tau peptide complex was immersed (Thermo Fisher Scientific) columns. in assay buffer without antibody. The immobilization of peptides to sensors, the association and the dissociation Reactivity of anti-tau human mAbs to tau peptides by ELISA steps, were followed in different ionic strength buffers Reactivity of recovered mAbs were tested against biotinyl- containing 10% FortéBio kinetics buffer as assay buffer. The ated tau peptides as previously described [37]. Briefly, tau relative association and dissociation kinetic curves were peptides were captured on streptavidin-coated plates compared to qualitatively assess the efficiency of antibody (Thermo Fisher Scientific) at 1 μg/ml in TBS and incu- binding to peptides encompassing different tau epitopes. bated for 2 h. Goat anti-human Fab (2 μg/ml, Jackson ImmunoResearch) to measure total IgG was used, and bo- Affinity measurements by Isothermal Titration vine actin (1 μg/ml TBS, Sigma) and irrelevant peptide Calorimetry (ITC) were used to confirm specificity of the purified mAbs. The affinities of antibodies for their corresponding tau ELISA plates were blocked and purified IgGs were diluted peptides were determined in solution on a MicroCal Auto- to 5 μg/ml in TBS/0.25% BSA and titrated (5-fold serial iTC200 system (Malvern). Peptides at concentrations of ~ dilutions) against peptides. Plates were subsequently 35 μM(CBTAU-27.1), ~10 μM(dmCBTAU-27.1), ~30 μM washed and goat anti-human IgG F(ab’)2 (Jackson Immu- (CBTAU-28.1) and ~ 33 μM (dmCBTAU-28.1) were titrated noResearch) was used at 1:2000 dilution as secondary. in 20 steps of 2 μl per step, except for dmCBTAU-27.1 Following incubation, plates were washed four times in where 40 steps of 1 μl were employed, in identical buffers TBS-T and developed with Sure Blue Reserve TMB containing 200 μM CBTAU-27.1, 130 μM dmCBTAU-27.1, Microwell Peroxidase Substrate (KPL) for approximately 205 μM CBTAU-28.1 and 205 μM dmCBTAU-28.1, 2 min. The reaction was stopped by the addition of TMB respectively. The thermodynamic parameters and the Stop Solution and the absorbance at 450 nm was equilibrium dissociation constants, Kd, were determined measured using an ELISA plate reader. upon fitting the ITC data to a model assuming a single set of binding sites corresponding to an antibody:tau peptide = Western blot analysis 1:2 binding model. Immunopurified PHF and sarkosyl-insoluble fraction of AD brain lysates were provided by Steven Paul at Weill Expression, crystallization, data collection, structure Cornell Medical College and prepared as described pre- determination, and refinement of CBTAU-27.1 Fab and viously [20]. Approximately 0.3 μg protein was resolved CBTAU-28.1 Fab on SDS-PAGE (4–12% Bis-Tris Novex NuPAGE gel; To express Fab fragments for crystallization, the gene Invitrogen) and subsequently transferred onto a nitrocel- sequences coding for the hinge, C 2, and C 3of human H H lulose membrane. The membrane was blocked overnight antibodies CBTAU-27.1 (IgG1κ) and CBTAU-28.1 (IgG1κ) in 1X TBS-T with 5% BSA (blocking buffer). CBTAU-27. were removed from the corresponding IgG expression 1 and CBTAU-28.1 were used at 25 μg/ml in 2.5% BSA vector and a His tag was added to the C-terminus of each in TBS-T and incubated for 2 h at room temperature. C 1. Both Fabs were produced by transient transfection of The membranes were then washed three times for 5 min FreeStyle 293-F cells and purified using a Ni-NTA column Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 5 of 17 followed by a size exclusion chromatography using a concentration of about 50 mg/ml. The Fab:peptide Superdex 75 or Superdex 200 column (GE Healthcare). complexes (dmCBTAU-27.1 Fab with tau peptide A8119 The purified CBTAU-27.1 and CBTAU-28.1 Fabs were and dmCBTAU-28.1 Fab with tau peptide A7731) were concentrated to ~ 10 mg/ml and ~ 8 mg/ml, respectively, subjected to crystallization screening by sitting drop vapor in 20 mM Tris, pH 8.0 and 150 mM NaCl for diffusion testing 2300 conditions, by mixing 0.1 μlprotein crystallization. Crystallization experiments were set up solution and 0.1 μl reservoir, and also varying the protein using the sitting drop vapor diffusion method. Initial concentration. The dmCBTAU-27.1 Fab with peptide crystallization conditions for CBTAU-27.1 Fab and its A8119 was crystallized from 0.10 M sodium citrate buffer, complex with peptide 2833–1 (HVPGGGSVQIVYKPV pH 5.0, 20.00% (w/v) PEG 8 K at a concentration of DLSKV), and CBTAU-28.1 in complex with peptide 17 mg/ml. The dmCBTAU-28.1 Fab with peptide A7731 W1805 (TEDGSEEPGSETSDAKSTPT-amide) were was crystallized from 22% (w/v) PEG5K MME, 0.10 M obtained from robotic crystallization trials using the HEPES buffer, pH 6.75, 0.20 M KCl at a concentration of robotic CrystalMation system (Rigaku) at The Scripps 50 mg/ml. For cryo-protection, crystals were briefly Research Institute. For co-crystallization, CBTAU-27.1 immersed in a solution consisting of 75% reservoir and and CBTAU-28.1 Fabs were mixed with peptides 2833–1 25% glycerol. X-ray diffraction data were collected at and W1805, respectively, in a molar ratio of 1:10 before temperature of 100 K at the Swiss Light Source. Data were screening. Crystals of apo-form CBTAU-27.1 Fab were integrated, scaled and merged using XDS [25]. The struc- obtained at 20 °C from a reservoir solution containing 0. ture was solved with MOLREP [43] and refined with 1 M Hepes, pH 7.5, 20% PEG8000, while crystals of REFMAC5 [44]. Manual model completion was carried CBTAU-27.1 Fab with 2833–1 were grown at 20 °C from out using Coot [14]. The quality of the final model was 0.1 M Hepes, pH 7.0, 0.1 M KCl, 15% PEG5000 MME. verified PROCHECK [29] and the validation tools avail- Crystals of CBTAU-28.1 with W1805 were obtained at able through Coot [14]. The dmCBTAU-27.1 diffraction 20 °C from 0.085 M Tris-HCl, pH 8.5, 0.17 M sodium data were indexed in space group C2 and the dmCBTAU- acetate, 25.5% PEG4000. Before data collection, the crys- 28.1 data in space group P2 2 2 . Data were processed 1 1 1 tals of CBTAU-27.1 Fab and CBTAU-28.1 Fab were using the programs XDS and XSCALE (see Additional file soaked in the reservoir solution supplemented with 25% 1: Table S3). The phase information necessary to deter- (v/v) and 15% (v/v) glycerol, respectively, for a few seconds mine and analyze the structure was obtained by molecular and then flash-frozen in liquid nitrogen. X-ray diffraction replacement using previously solved structure of data of CBTAU-27.1 Fab apo-form were collected to 1.9 Å dmCBTAU-27.1 and dmCBTAU-28.1 Fabs were used resolution at beamline 23ID-D at the Advanced Photon search models. Subsequent model building and refinement Source (APS). X-ray diffraction data of CBTAU-27.1 Fab was performed according to standard protocols with the with 2833–1 and CBTAU-28.1 with W1805 were collected software packages CCP4 and COOT. For calculation of R- to 2.0 and 2.1 Å resolution, respectively, at beamline 12–2 free, 6.2% of the measured reflections were excluded from at the Stanford Synchrotron Radiation Lightsource (SSRL). refinement (see Additional file 1: Table S3). TLS refine- HKL2000 (HKL Research, Inc.) was used to integrate and ment (using REFMAC5, CCP4) resulted in lower R-factors scale the diffraction data (Additional file 1: Table S2). The and higher quality of the electron density map for structures were determined by molecular replacement dmCBTAU-27.1-A8119 using five TLS groups. Waters with Phaser [35] using search models of a human antibody were added using the “Find waters” algorithm of COOT 2-23b3 Fab (PDB ID 3QOS) for CBTAU-27.1 Fab and a into Fo-Fc maps contoured at 3.0 sigma followed by re- human antibody 1-69b3 (PDB ID 3QOT) for CBTAU-28.1 finement with REFMAC5 and checking all waters with the Fab. The models were iteratively rebuilt using Coot [14] validation tool of COOT. The occupancy of side chains, and refined in Phenix [1]. Refinement parameters included which were in negative peaks in the Fo-Fc map (contoured rigid body refinement and restrained refinement including at − 3.0 σ), were set to zero and subsequently to 0.5 if a TLS refinement. Electron density for both peptides 2833– positive peak occurred after the next refinement cycle. 1 and W1805 were clear and the peptides were built in The Ramachandran plot of the final model shows 88.7% the later stages of the refinement. Final refinement statis- (dmCBTAU-27.1) and 88.1% (dmCBTAU-28.1) of all resi- tics are summarized in Additional file 1: Table S2. dues in the most favored region, 10.8% (dmCBTAU-27.1) and 11.6% (dmCBTAU-28.1) in the additionally allowed Crystallization, data collection, and structure determination region, and 0.3% (dmCBTAU-27.1) and 0.0% (dmCBTAU- of dmCBTAU-27.1 Fab and dmCBTAU-28.1 Fab 28.1) in the generously allowed region. Residue Ala51(L) For crystallization, the dmCBTAU-27.1 and dmCBTAU- is found in the disallowed region of the Ramachandran 28.1 Fab fragments (in 20 mM HEPES buffer, pH 7.5, 7. plot (Additional file 1: Table S3) for both Fabs, as is 55 mM NaCl) were incubated with 4 mM of the respective frequently found in other Fabs [3] . This was either peptide on ice overnight and concentrated to a final confirmed from the electron density map or could not be Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 6 of 17 modelled in another more favorable conformation. Statis- and ThT. Kinetic measurements were monitored at 37 °C tics of the final structure and the refinement process are in a Biotek Synergy Neo2 Multi-Mode Microplate Reader listed in Additional file 1: Table S3. (Biotek, VT, USA) by measuring ThT fluorescence at 485 nm (20 nm bandwidth) upon excitation at Affinity maturation by phage display 440 nm (20 nm bandwidth) upon continuous shaking The coding sequence for scFv directed against CBTAU- (425 cpm, 3 mm). 28.1 epitope was cloned into an inducible prokaryotic expression vector containing the phage M13 pIII gene. Atomic force microscopy Random mutations were introduced in the scFv by error For each sample, 20 μl of rtau solution was deposited onto prone PCR (Genemorph II EZClone Domain Mutagen- freshly cleaved mica surface. After 3 min incubation, the esis kit) after which the DNA was transformed into surface was washed with double-distilled water and dried TG1bacteria. The transformants were grown to mid-log with air. Samples were imaged using the Scanasyst-air phase and infected with CT helper phages that have a protocol using a MultiMode 8-HR and Scanasyst-air genome lacking the infectivity domains N1 and N2 of silicon cantilevers (Bruker Corporation, Santa Barbara, protein pIII, rendering phage particles which are only in- USA). Height images of 1024 × 1024 pixels in size and fective if they display the scFv linked to the full-length surface areas of 10 × 10 μm were acquired under pIII [28]. Phage libraries were screened using magnetic ambient environmental conditions with peak force beads coated with rtau in immunotubes. To deselect frequency of 2 KHz. nonspecific binders, the tubes were coated with a tau peptide lacking the CBTAU-28.1 epitope. To ensure Assessment of CBTAU-27.1 binding to rtau and PHFs by maturation against the correct epitope, selection was SEC-MALS continued using beads coated with the cognate A6940 15 μM monomer or aggregated rtau were incubated with peptide. Eluted phages were used to infect XL1-blue F′ dmCBTAU-27.1 in a rtau:IgG1 = 1:0.6 ratio for 15 min E. coli XL1-blue F′ which were cultured and infected and samples were subsequently centrifuged for 15 min with CT-helper (or VCSM13) phages to rescue phages at 20,000 g. Same procedure was also applied for used for subsequent selection rounds. After three rounds controls containing only monomer rtau, aggregated rtau of panning, individual phage clones were isolated and or IgG. All samples were analyzed by SEC-MALS screened in phage ELISA for binding to rtau and cognate upon fractionation on a TSKgel G3000SWxl (Tosoh CBTAU-28.1 peptide A6940. Bioscience) gel filtration column equilibrated with 150 mM sodium phosphate, 50 mM sodium chloride In vitro tau aggregation assay at pH 7.0. at a flow rate of 1 ml/min. For molar mass Stock solutions of 500 μMthioflavin T (ThT)(Sigma- determination, in-line UV (Agilent 1260 Infinity MWD, Aldrich, St Louis, MO, USA) and 55 μM heparin Agilent Technologies), refractive index (Optilab T-rEX, (Mw = 17–19 kDa; Sigma-Aldrich, St Louis, MO, USA) Wyatt Technology) and 8-angle static light scattering were prepared by dissolving the dry powders in reaction (Dawn HELEOS, Wyatt Technology) detectors were used. buffer (0.5 mM TCEP in PBS, pH 6.7), and filtered through a sterile 0.22 μm pore size PES membrane filter Immunohistochemistry (Corning, NY, USA) or a sterile 0.22 μm pore size PVDF Brain samples were obtained from The Netherlands Brain membrane filter (Merck Millipore, Tullagreen, Cork, IRL), Bank (NBB), Netherlands Institute for Neuroscience, respectively. The concentration of the ThT solution was Amsterdam. All donors had given written informed consent determined by absorption measurements at 411 nm using for brain autopsy and the use of material and clinical infor- − 1 − 1 an extinction coefficient of 22,000 M cm .The mation for research purposes. Neuropathological diagnosis huTau441 concentration was determined by absorption was assessed using histochemical stains (haematoxylin and measurements at 280 nm using an extinction coefficient eosin, Bodian and/or Gallyas silver stains [41] and immuno- − 1 − 1 of 0.31 ml mg cm . For spontaneous conversions, histochemistry for Amyloid beta, p-tau (AT8), α-synuclein, mixtures of 15 μM huTau441 in 200 μl reaction buffer TDP-43 and P62, on formalin-fixed and paraffin-embedded containing 8 μM heparin and 50 μM ThT were dispensed tissue from different parts of the brain, including the frontal in 96-well plates (Thermo Scientific, Vantaa, Finland) that cortex (F2), temporal pole cortex, parietal cortex (superior were subsequently sealed with plate sealers (R&D Systems, and inferior lobule), occipital pole cortex, amygdala and the Minneapolis, MN). For seeding experiments, preformed hippocampus, essentially CA1 and entorhinal area of the seeds were added to the wells before sealing the plate. To parahippocampal gyrus. Staging of pathology was assessed assess the effect of IgG or Fab on the conversion, according to Braak and Braak for tau pathology and Thal huTau441 and IgG or Fab were mixed and incubated for for Amyloid beta pathology [5, 40]. Formalin-fixed and par- 20 min in reaction buffer before the addition of heparin affin embedded tissue sections (5 μm thick) were mounted Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 7 of 17 on Superfrost Plus tissue slides (Menzel-Gläser, Germany) were added to the beads together with 90 μl of the 1:1 and dried overnight at 37 °C. Sections were deparaffinised antibody-brain extract mixture. Samples were incubated and subsequently immersed in 0.3% H O in phosphate- over night at 4 °C, rotating at 5 rpm. The following day, 2 2 buffered saline (PBS) for 30 min to quench endogenous the immunodepleted fractions were separated from the peroxidase activity. Sections were either treated in sodium beads by pulling down the beads with the magnet, trans- citrate buffer (10 mM sodium citrate, pH 6.0) heated by ferred to a new 96-well PCR plate and stored at − 80 °C autoclave (20 min at 130 °C) for antigen retrieval or proc- until tested. Each condition was tested in duplicate. essed without heat pretreatment. Between the subsequent Immunodepleted fractions were incubated for 10 min with incubation steps, sections were washed extensively with Lipofectamine 2000 (Invitrogen) in Opti-MEM (Gibco) in PBS. Primary antibodies were diluted in antibody diluent a 96-well cell culture plate (Greiner Bio-one) before 5.5 × (Immunologic) and incubated overnight at 4 °C. Secondary 10 HEK biosensor cells (provided by M. Diamond, EnVison™ HRP goat anti-rabbit/mouse antibody Washington University School of Medicine) were added HRP (EV-GαM , DAKO) was incubated for 30 min at room to each well. After a 2-day incubation at 37 °C, cells were temperature (RT). 3,3-Diaminobenzidine (DAB; DAKO) washed twice with PBS, detached using Trypsin/EDTA was used as chromogen. Sections were counterstained with (Gibco) and transferred to a polypropylene round bottom haematoxylin to visualize the nuclei of the cells, dehydrated plate (Costar) containing FACS buffer (Hank’s Balanced and mounted using Quick-D mounting medium (BDH Salt Solution (Sigma), 1 mM EDTA (Invitrogen), 1% FBS Laboratories Supplies, Poole, England). For the Gallyas (Biowest)). Cells were then analyzed for FRET positivity silver staining, 30 μm thick sections were rinsed in distilled by flow cytometry using a FACS Canto II (BD Bioscience). water and incubated in 5% periodic acid for 30 min at RT, Each plate contained a brain extract only condition (to followed by an incubation in silver iodide solution assess baseline FRET response) and an antibody isotype (4% sodium hydroxide, 10% potassium iodide and 0.35% control. Results are reported as normalized values, relative silver nitrate in distilled water) for 30 min at RT. Subse- to condition without antibody. quently, sections were washed in 0.5% acetic acid and de- veloped with developer working solution (10 volumes 5% Microglia assay sodium carbonate solution, 3 volumes solution 0.2% am- Aggregated recombinant 2N4R tau (rtau) was generated monium nitrate, 0.2% silver nitrate and 1% Tungstosilicic in the absence of ThT under the conditions described for acid solution, and 7 volumes 0.2% ammonium nitrate, 0. the in vitro tau aggregation assay and covalently labelled 2% silver nitrate, 1% Tungstosilicic acid and 0.3% formal- with pHrodo® Green STP Ester (Invitrogen) following dehyde solution. After color development, sections were manufacturer’s instructions. Briefly, rtau aggregates were rinsed in 0.5% acetic acid, after which sections were incu- spun down by centrifugation at 20800 rcf for 30 min and bated in 5% sodium thiosulphate and rinsed in distilled then resuspended in 0.1 M sodium bicarbonate buffer, water. Stained sections were mounted on coated glass pH 8.5, at a final concentration of 2 mg/ml. Efficiency of slides (Menzel-Gläser) and dried for at least 2 h at 37 °C. aggregation was assessed by detecting presence of tau in Subsequently sections were fixed in ethanol 70% for the supernatant using SEC-MALS. Prior to labeling, tau 10 min, counterstained with hematoxylin, dehydrated and aggregates were briefly sonicated. Ten moles of dye were mounted with Quick D mounting medium. added per mole of protein and the mixture was incubated for 45 min at room temperature, protected from light. FRET based cellular immunodepletion assay Unconjugated dye was removed using a PD10 column Cryopreserved brain tissue was acquired from the (GE Healthcare) equilibrated with 0.1 M sodium bicar- Newcastle Brain Tissue Resource biobank. Frozen brain bonate buffer pH 8.5 and eluting the protein with the tissue samples from 17 AD patients were homogenized same buffer. Eluted fractions were evaluated for their in homogenization buffer (10 mM Tris (Gibco), 150 mM protein content by BCA assay (Thermo Fisher Scientific) NaCl (Gibco) containing protease inhibitors (cOmplete™ following the manufacturer’s instructions. Protein contain- ULTRA tablets EDTA free, Roche) to obtain a 10% (w/v) ing fractions were pooled, aliquoted and stored at − 20 °C. pooled brain homogenate. Individual antibody dilutions BV-2 cells were cultured in DMEM supplemented with were prepared in PBS pH 7.4 (Sigma), mixed with brain 10% FBS, 100 U/ml penicillin, 100 μg/ml streptomycin extract in a 1:1 ratio in a 96 well PCR plate (Thermo and 2 mM L-Glutamine. Cultures were maintained in Scientific), and incubated until the beads were washed. humidified atmosphere with 5% CO2 at 37 °C. In order to Protein-G DynaBeads (Life Technologies) were added in generate immunocomplexes, 250 nM aggregated rtau, a 96-well PCR plate (Thermo Scientific) and washed covalently labelled with pHrodo Green dye, was incubated twice with PBS, 0.01% Tween20 (Sigma) by pulling down with a serial dilution (12.5–150 nM) of a chimeric version the beads with a magnet (Life Technologies). Wash buffer (mouse Fc region) of CBTAU-28.1 (parental and high was removed completely and 10 μlofPBS, 0.1% Tween20 affinity mutant) or CBTAU-27.1 (parental and high affinity Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 8 of 17 mutant) in serum-free medium. Tau immunocomplexes 52–71 for CBTAU-28.1 (Additional file 1:FigureS1).The were also generated with 300 nM Fab fragments of both specificity of these antibodies for tau was confirmed by CBTAU-28.1 and CBTAU-27.1, in the parental and high Western blot (Additional file 1:Figure S2). affinity mutant format. In each experiment, a mouse IgG1 isotype control was included together with cells incubated Recognition of a structurally identical, germline encoded with only aggregated rtau. Immunocomplexes were incu- hotspot motif bated over night at 4 °C and the day after applied to BV2 Crystal structures of the Fab fragments of CBTAU-27.1 and cells for 2 h at 37 °C with 5% CO . During the incubation, CBTAU-28.1 in complex with tau peptides spanning resi- antibody-independent tau uptake was prevented by block- dues 299–318 and 52–71 were determined at 2.0 and 2.1 Å ing the Heparan Sulfate Proteoglycan Receptor with resolution, respectively (Fig. 1b and c, Additional file 1: 200 μg/ml Heparin. After incubation, cells were harvested Table S2). The structures reveal that an intriguing similarity with 0.25% trypsin-EDTA for 20 min thus simultaneously exists in the way they bind despite recognition of very removing tau bound to the extracellular membrane, cen- distinct regions on the tau protein (Fig. 1d). Both light trifuged at 400 rcf to remove medium, washed twice with chains harbor a pocket made of aromatic tyrosine or PBS, and resuspended in flow cytometry buffer (PBS 1× phenylalanine side chains that form a binding site for a plus 0.5% BSA and 2 mM EDTA). Cells were analyzed proline residue in the N-terminal region of the different with a Canto II flow cytometer (BD) gating for live single peptides. This interaction is further stabilized by a peptide cell population, as identified by forward and side scatter backbone hydrogen bond to LCDR3 Phe (CBTAU-27.1) profiles. Results are reported as geometric mean fluores- and LCDR3 Tyr (CBTAU-28.1). Similarly, the two heavy cent intensities. Each experiment was performed twice. chains interact with a lysine in the peptide C-terminal For the microscopy experiments, cells were seeded in 96- region that involves identical hydrogen bonding networks well μClear® plate (Greiner Bio-one). After incubation with with two HCDR2 aspartates and the backbone of HCDR3 97 103 the immunocomplexes, nuclei were stained with Hoechst Ala (CBTAU-27.1) and Ser (CBTAU-28.1), respectively. (Sigma) and the acidic cellular compartment with Lyso- Both lysines are flanked by HCDR1 Trp and HCDR2 Tracker Red dye (Thermo Fisher). Live-cell imaging was Tyr that align and stabilize the aliphatic part of the Lys performed using the Opera Phenix™ High Content Screen- side chains. However, the central parts of the tau epitopes ing System (PerkinElmer) with temperature set to 37 °C differ significantly (see Fig. 1d, column 3). The four-residue and in presence of 5% CO . For high quality images, a 63× central region adopts an extended structure in CBTAU-27.1 water immersion objective was used and 0.5 μmplanes and inserts Leu into a pocket formed by LCDR3, (20 per Z-stack) were acquired per imaged field. HCDR3, HCDR1 and HCDR2. In the same spatial location, the seven-residue central region of the CBTAU-28.1 epitope Results spans the same distance between the conserved proline and Identification of naturally occurring anti-tau antibodies in lysine residues by adopting a more compact helical struc- healthy donors ture (Fig. 1e). CBTAU-28.1 Asp makes a salt bridge with 6 59 102 In total, 2.6 × 10 memory B cells from nine healthy blood Arg (HCDR2) and a hydrogen bond with Tyr (LCDR3). donors aged 18–65 years were interrogated against a pool These two antibodies thus recognize a Pro – X – Lys of 10 overlapping peptides spanning the length of tau441 motif in different tau peptides, where n is from 4 up to at (Additional file 1: Table S1). Ninety-two tau-reactive B least 7 amino acids. The proline and lysine binding pockets cells were sorted and 30 heavy and light variable chain are germline-encoded and specificity towards one or other sequences were recovered and full-length IgGs were epitope arises from the CDR3 loops, which interact with cloned and expressed. Two unique tau binding antibodies, the X region (Fig. 1d). CBTAU-27.1 and CBTAU-28.1, which are both derived The CBTAU-27.1 epitope encompasses residues 310 317 from the V 5–51 and V 4–1 germline families, were iden- YKPVDLSK in the R3 domain (Fig. 1f). The R3 H L tified. Both antibodies carry high numbers of somatic mu- domain is part of the core of PHFs [12, 15], where a 306 311 tations with 38 and 28 nucleotide substitutions for the hexapeptide VQIVYK is crucial for PHF assembly heavy and light chains of CBTAU-27.1, and 19 and 16 for [12, 15, 32, 45]. Since the CBTAU-27.1 epitope overlaps the heavy and light chains of CBTAU-28.1, respectively. this hexapeptide, in particular the key Lys [32], we Since memory B cell selections were performed using a hypothesize that CBTAU-27.1 binding to tau could peptide pool, an ELISA-based binding assay with the 10 block the nucleation interface and thus prevent aggre- individual tau peptides was performed and established gation. The CBTAU-28.1 epitope encompasses residues 58 67 that CBTAU-27.1 and CBTAU-28.1 bound to peptide EPGSETSDAK in the first N-terminal insert (Fig. 1f). A6897 (residues 299–369) and peptide A6940 (residues Since the N- and C-terminal tau regions that surround 42–103), respectively. Further mapping narrowed the epi- the repeat domains have been shown to be disordered and tope regions to residues 299–318 for CBTAU-27.1 and project away from the PHF core to form a flexible fuzzy Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 9 of 17 coat [15], binding of CBTAU-28.1 is unlikely to interfere control and AD brain tissue (Fig. 3). CBTAU-27.1 did not with PHF formation, but may—like previously described show immunoreactivity in either control or AD cases, antibodies [6, 42, 48, 49]—hamper the spreading of aggre- whereas dmCBTAU-27.1 showed immunoreactivity in the gates after they are formed. However, the affinities of both cytosol of neurons of the control cases and clear recogni- antibodies, at least to their cognate tau peptides, are in the tion of aggregated tau in neuropil threads and NFTs in high nanomolar range (Additional file 1: Figure S3A and AD brains, but only after heat pretreatment which is a B), which may limit their functional activity. Therefore, we routine ‘antigen retrieval’ procedure to recover reactivity set out to generate affinity-improved mutants of CBTAU- in formalin-fixed paraffin-embedded tissue sections. No 27.1 and CBTAU-28.1 by employing a combination of immunoreactivity of CBTAU-28.1 was detected in the rational design and random mutagenesis approaches. control cases, whereas dmCBTAU-28.1 showed diffuse immunoreactivity of neurons after heat pretreatment. In Affinity-improved antibodies retain specificity AD brains, both CBTAU-28.1 and dmCBTAU-28.1 recog- For CBTAU-27.1, we used a rational structure-based nized neuropil threads and NFTs regardless of the sample approach through analysis of the co-crystal structure treatment. CBTAU-28.1 thus recognizes PHFs without (Fig. 2a). LCDR3 Thr was identified as one location heat pretreatment whereas CBTAU-27.1, even in its high where additional hydrophobic interactions could be formed affinity variant, requires heat pretreatment to recognize without affecting the structure of the tau peptide. Isoleucine pathologic tau. This is in line with the epitope of CBTAU- introduced at this position better filled the gap between 27.1 being buried within the PHFs and becoming exposed 313 315 58 Val ,Leu and the aliphatic portion of V Arg .In upon heat pretreatment. The diffuse neuronal immunore- 27D LCDR1, Ser was mutated to tyrosine to remove the activity of dmCBTAU-27.1 and dmCBTAU-28.1 observed unfavorable contact between the serine hydroxyl and the in control brain tissue after heat pretreatment shows that proline pyrrolidine sidechain and create additional these antibodies bind to physiological tau. The observed hydrophobic interactions. These two mutations improved immunoreactivity under identical conditions shows a clear theaffinitybymorethan50-fold to thelow nanomolar improvement in the detection of tau by affinity-improved range (Fig. 2c and Additional file 1:FigureS3).For CBTAU- antibodies relative to parental antibodies without affecting 28.1, analysis of the structure did not reveal potential specificity (Fig. 3). Similar results were obtained with affinity-improving mutations and, therefore, a random mu- immunohistochemical staining on post-mortem brain tagenesis strategy was employed (Fig. 2b). This approach tissue of other tauopathies like frontotemporal lobar 32 35 led to the identification of Ser ➔Arg and Glu ➔ Lys degeneration (FTLD), Pick’s disease, progressive supra- mutations in the light chain that combined led to an ~ nuclear palsy (PSP) and primary age-related tauopathy 4-fold improvement in affinity compared to the parental (PART) cases (Additional file 1: Figure S5). Both CBTAU- antibody (Fig. 2d and Additional file 1:FigureS3). 27.1 and CBTAU-28.1 recognize pathological tau struc- Co-crystal structures of the Fab fragments of the CBTAU- tures in all these diseases, but detection by CBTAU-27.1 is 27D 94 27.1 double mutant Ser ➔ Tyr / Thr ➔ Ile (from here dependent on heat pretreatment to make its epitope on referred to as dmCBTAU-27.1) and the CBTAU-28.1 accessible. Furthermore, the detection of tau is improved 32 35 double mutant Ser ➔ Arg / Glu ➔ Lys (from here on for the affinity-improved antibodies. referred to as dmCBTAU-28.1) in complex with peptides A8119 and A7731, respectively, were determined at 3.0 and 2.85 Å resolution (Additional file 1: Table S3). Alignment of Binding domain-dependent functional activities the structures of the double mutants in complex with their The observation that the epitope of CBTAU-27.1 forms tau epitopes to the corresponding parental antibody co- part of the core of the PHFs led us to consider that crystal structures showed that both dmCBTAU-27.1 and CBTAU-27.1 might be able to prevent aggregation of tau dmCBTAU-28.1 retained the binding mode of the parental by inhibiting the initial nucleation step. While the antibody (Fig. 2e and f), with RMSD values for the peptide molecular mechanism of tau aggregation is not fully Cα atoms of 0.44 Å and 0.24 Å, respectively. The similarity understood, the current paradigm is that it follows a between the double mutants and their parental antibodies nucleation-dependent polymerization (NDP) process regarding the nature of their interactions with tau was con- [2, 11, 23, 39]. An NDP mechanism is characterized by an firmed by biolayer interferometry using buffers of different initial nucleation step (nuclei formation) followed by an ionic strengths (Additional file 1:FigureS4).Furthermore, exponential growth step (fibril elongation). Nucleation, the binding of the different antibodiestosetsoftau peptides rate-limiting step of the aggregation process, is a stochastic was assessed to confirm conservation of the specificity phenomenon and refers to the formation of high energy (Additional file 1:FigureS4). nuclei. Once nuclei are formed, they rapidly recruit tau The tau specificity of the antibodies was further assessed monomer (growth step), and convert into thermodynamic- by immunohistochemical staining on post-mortem ally stable aggregates. These aggregates can undergo Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 10 of 17 312 313 Fig. 2 Generation of affinity-improved mutants of CBTAU-27.1 and CBTAU-28.1a Structure-based design of mutants around Pro (left panel) and Val 312 94 (right panel). The tau epitope is illustrated as in Fig. 1b. Antibody loops and the key residues interacting with Pro and Thr are plotted in white. Proposed 27D mutations are shown as orange sticks on top of the corresponding wild-type side chains. Ser is mutated to tyrosine (left panel) to 312 94 313 enlarge the hydrophobic pocket of Pro ,and Thr is mutated to isoleucine (right panel) to fill the empty cavity surrounding Val and Leu . By introducing both mutations, additional hydrophobic contacts between tau and the antibody loops could be formed, potentially resulting in a lower desolvation penalty and increased affinity. b Schematic representation of the CBTAU-28.1 affinity maturation process by random mutagenesis. Mutations were introduced randomly by error prone PCR in the coding sequence for the single-chain variable fragment (scFv) directed against the CBTAU-28.1 epitope. M13 phage libraries displaying the scFv were screened against rtau and peptide A6940. Affinity-matured variants were identified by phage ELISA and converted into an IgG1 format to assess binding in solution. c and d Association and dissociation profiles for parental and affinity improved CBTAU-27.1 (c) and CBTAU-28.1 (d) variants to their corresponding cognate peptides as determined by Octet biolayer interferometry. Affinities as determined by ITC (K ) are shown on the individual graphs. (e and f) Co-crystal structures of the Fabs of dmCBTAU-27.1 (e) and dmCBTAU-28.1 (f) with tau peptides A8119 and A7731, respectively. Antibodies are illustrated as molecular surfaces (colored as in panel A), together with tau epitopes as sticks with yellow carbons. The corresponding parental co-crystal structures have been aligned using their variable regions, and their tau epitopes are shown as blue mesh on top of the mutant epitopes. fragmentation generating more fibril ends that are capable to as “seeding”. To assess whether CBTAU-27.1 can inter- of recruiting tau monomers and converting them into de fere with the aggregation of tau, we have developed a robust novo fibrils. This process is in most general terms referred and highly reproducible in vitro assay that monitors the Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 11 of 17 Fig. 3 Detection of immunoreactivity in human brain tissue by CBTAU-27.1 and CBTAU-28.1 and affinity-improved variants. Immunohistochemistry was performed on 5 μm thick formalin-fixed paraffin embedded sections of the hippocampal region using a 0.1 μg/ml antibody concentration. Immunodetection using CBTAU-27.1 (a-d), dmCBTAU-27.1 (e-h), CBTAU-28.1 (i-l), dmCBTAU-28.1 (m-p) and PHF-tau-specific mouse antibody AT8 (q-t) in control and AD brain tissue without or with heat pretreatment using sodium citrate buffer. Gallyas staining for detection of NFTs and neuropil threads is shown from the same control and AD case of corresponding areas for comparison (u, v). Immunoreactivity was visualized using DAB (brown) and nuclei were counterstained with haematoxylin (blue). Representative areas of the CA1/subiculum of the hippocampus are shown. Scale bars represent 50 μm heparin-induced aggregation of full-length recombinant microscopy (AFM) (Additional file 1:FigureS6C) and were Tau441 (rtau) by thioflavin T (ThT) fluorescence. The extremely efficient in seeding de novo aggregation of tau aggregation behavior of tau in our assay fulfills the expected (Additional file 1:FigureS6D andE). features of an NDP: sigmoidal kinetic curves with a well- In agreement with our hypothesis, CBTAU-27.1 inhib- defined lag phase followed by exponential growth ending in ited tau aggregation, as reflected by longer lag phases a stationary phase (Additional file 1:FigureS6).The aggre- and lower final ThT fluorescence signal in the kinetic gation kinetics of tau were highlyreproducibleand dis- curves. This inhibitory effect was strongly enhanced played the expected concentration dependence (Additional after affinity improvement (Fig. 4a, and b, and file 1: Figure S6B). Furthermore, the obtained tau aggregates Additional file 1: Figure S7). To shed further light on the displayed PHF-like morphology as assessed by atomic force mechanism, we assessed the ability of dmCBTAU-27.1 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 12 of 17 Fig. 4 CBTAU-27.1, but not CBTAU-28.1, inhibits the aggregation of recombinant tau in vitro. Aggregation of rtau in the absence (black) or presence of CBTAU-27.1 (a), dmCBTAU-27.1 (b), Fab CBTAU-27.1 (c), Fab dmCBTAU-27.1 (d), CBTAU-28.1 (e), dmCBTAU-28.1 (f), Fab CBTAU-28.1 (g)orFab dmCBTAU-28.1 (h), as monitored continuously for 120 h by ThT fluorescence. Three different rtau-IgG (1:0.2 – red, 1:0.4 – blue, and 1: 0.6 – purple) and rtau-Fab (1:0.4 – red, 1: 0.8 – blue, and 1:1.2 – purple) stoichiometries were tested. Each condition was tested in quadruplicate and one representative curve is shown for each condition. For complete datasets, see Additional file 1: Figures S7-S10 to alter the tau conversion when added at later times sequestering monomeric tau (Additional file 1: Figure after initiating the aggregation. In all cases, our results S11). The hypothesis that CBTAU-27.1 targets mono- show that stoichiometric amounts of dmCBTAU-27.1 meric tau and does not interact with tau aggregates was can not only prevent tau aggregation, but also arrest it further confirmed by sedimentation experiments even in the exponential phase where significant amounts followed by SEC-MALS size determination, which indi- of seeds are already present, presumably by binding and cates that the antibody binds to two tau monomers Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 13 of 17 using its two Fab arms while not being able to co-sedi- ment with preformed tau aggregates (Additional file 1: Figure S12). In sum, these results confirm the initial hy- pothesis that an antibody that targets the key PHF inter- face of the monomeric tau can block its misfolding and aggregation. Both double mutant and parental CBTAU-28.1 showed comparable alterations in the aggregation kinetics (Fig. 4e,and f, and Additional file 1:Figure S8). While somewhat longer lag phases and lower end-point fluorescent signals were observed in the presence of anti- body, the effect did not seem to be dose dependent and the kinetics were strikingly irreproducible. Furthermore, visual inspection of reaction mixtures after 120 h incuba- tion revealed that CBTAU-28.1 induced formation of large polymeric structures (Additional file 1: Figure S13), sug- gesting it can crosslink tau aggregates. This notion is sup- ported by the fact that the Fabs of both parental and dmCBTAU-28.1 did not affect tau aggregation (Fig. 4g, and h, and Additional file 1: Figure S10). In contrast, CBTAU-27.1 and dmCBTAU-27.1 Fabs showed similar in- hibitory effects as their corresponding antibodies (Fig. 4c, Fig. 5 CBTAU-28.1, but not CBTAU-27.1, is capable of immunodepleting and d, and Additional file 1: Figure S9), emphasizing the seeds from AD brains. Residual seeding activity of human AD brain different mechanisms by which CBTAU-27.1 and homogenates following immunodepletion with different concentrations CBTAU-28.1 interfere with tau aggregation. of CBTAU-27.1 and dmCBTAU-27.1 (a)orCBTAU-28.1and dmCBTAU-28.1 (b) as measured by FRET signal in biosensor cells expressing the We next assessed the ability of the antibodies to bind microtubule repeat domains of tau (aa 243–375) fused either to yellow or tau aggregates and thus potentially block the propagation cyan fluorescent protein. Uptake of exogenous tau aggregates into the and spreading of tau pathology. We therefore incubated cells results in aggregation of the tau fusion proteins, which is detected human AD brain homogenate containing PHFs with the by FRET. As positive and negative controls, a human IgG1 chimeric antibodies and depleted the antibody-antigen complexes. version of murine anti-PHF antibody AT8 and anti-RSV-G antibody RSV-4.1 were taken along, respectively. For the controls, the same data The residual seeding capacity was assessed using a cell- are shown in plots A and B for visualization purposes. Error bars indicate based biosensor assay [24, 48]. In line with its inability to the SD of two independent experiments bind PHFs, CBTAU-27.1 did not reduce the seeding activ- ity of the AD brain homogenate, while dmCBTAU-27.1 showed only minor reduction and only at the highest con- CBTAU-27.1 and dmCBTAU-27.1, was confirmed by centration tested (Fig. 5a). In contrast, CBTAU-28.1, like confocal microscopy (Fig. 6b,and d, and Additional file 1: mouse anti-PHF antibody AT8, depletes seeding activ- Figure S14). ity from AD brain homogenate and this in vitro activ- ity is enhanced for dmCBTAU-28.1 (Fig. 5b). The observation that CBTAU-28.1 and dmCBTAU-28 Discussion can bind PHFs led us to explore the possibility that these Implications for therapy and vaccines antibodies may furthermore enhance the uptake of tau ag- By binding to the region that is critical for the aggrega- gregates by microglia, the resident macrophage cells of tion of tau and which forms the core of PHFs, CBTAU- central nervous system [17]. Indeed, CBTAU-28.1 and 27.1 prevents aggregation of tau in vitro. This functional dmCBTAU-28.1 promoted the uptake of aggregated rtau activity identifies its epitope as a potential target for im- into mouse microglial BV2 cells and the affinity-improved munotherapy and could perhaps allow earlier interven- antibody appeared to mediate tau uptake to a greater ex- tion than antibodies that inhibit the spreading of already tent (Fig. 6a). ThefactthatFabs of bothparentaland formed tau seeds [6, 42, 48, 49]. Evidence that interfering dmCBTAU-28.1 did not increase basal tau uptake indi- with tau aggregation through immunotherapy may be cates that the uptake is Fc mediated. As expected, possible is provided by murine antibody DC8E8 which CBTAU-27.1 and dmCBTAU-27.1 did not show activity in also targets monomeric tau, inhibits tau aggregation in this assay (Fig. 6c). Antibody-mediated tau uptake and vitro, and reduces tau pathology in a murine AD model localization of rtau aggregates in the endolysosomal com- [27]. An alternative approach could be the development of partment by CBTAU-28.1 and dmCBTAU-28.1, but not a small molecule drug targeting monomeric tau that Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 14 of 17 Fig. 6 CBTAU-28.1, but not CBTAU-27.1, enhances uptake of tau aggregates into microglial BV2 cells. a and c Aggregated recombinant tau was covalently labelled with pHrodo Green dye and incubated with chimeric versions (containing mouse instead of human Fc region) of CBTAU-28.1, dmCBTAU-28.1, CBTAU27.1, dmCBTAU-27.1, their Fab fragments, a mouse IgG1 isotype control antibody (IC), or no antibody (rtau). Immunocomplexes were subsequently incubated with BV2 cells and their uptake was assessed by flow cytometry as expressed by the geometric mean fluorescent intensity. Error bars indicatethe SD of two independent experiments. b and d Preformed pHrodo-Green labeled immunocomplexes of rtau with chimeric dmCBTAU-28.1 or dmCBTAU-27.1 (at a concentration of 150 nM) were incubated with BV-2 cells. After incubation, nuclei were stained with Hoechst (blue) and the acidic cellular compartment with LysoTracker Red dye and uptake of immunocomplexes was assessed by live-cell imaging. Images represent maximum intensity projections of a 20 planes Z-stack (0.5 μm planes) acquired with a 63× water immersion objective interacts with the CBTAU-27.1 epitope. This epitope does functional activities) to be used for different purposes at dif- not overlap with the microtubule binding motifs (Fig. 1f), ferent stages of disease. The fact that both antibodies inter- suggesting that such a drug may not interfere with the act with tau aggregates from different tauopathies normal function of physiological tau while preventing its (Additional file 1: Figure S5) suggests that they may hold aggregation. Furthermore, the epitope of CBTAU-27.1 is therapeutic potential for various related neurodegenerative specific for tau, and is not present on MAP2, a micro- diseases. Clearly, effective therapy would require binding to, tubule stabilizing protein closely related to tau [13]. Like and clearance of different tau aggregate species (e.g. previously described antibodies, CBTAU-28.1 may be able low and high molecular weight species) and assess- to inhibit the spread of tau pathology. In addition, it dem- ment of the ability of these antibodies to do so will onstrated an Fc-dependent enhancement of the uptake of be the subject of future studies. tau aggregates by microglial BV2 cells. It is probably able Both antibodies are derived from the V 5–51 and to enhance the uptake of aggregates because its epitope is V 4–1 germline families and bind their respective epi- distant from the core of the PHFs and remains accessible. topes through hotspot interactions that are remarkably In summary, CBTAU-27.1 binds an epitope crucial for alike, pointing towards a conserved structural motif in tau aggregation that becomes buried inside PHFs and there- tau that could not have been predicted from sequence fore inhibits aggregation by binding and sequestering tau. analysis alone. The motif containing proline and lysine However, it does not decrease seeding activity of previously separated by 4 to 7 amino acids, is commonly found in formed aggregates and is thus functional at earlier stages of tau, and appears nine times on 2N4R. The lack of som- tau aggregation. In contrast, CBTAU-28.1 binds PHFs, atic mutations around the hotspot proline and lysine cross-links tau aggregates and depletes seeding activity, but suggests these two key tau residues could be responsible does not affect initial tau aggregation. CBTAU-27.1 and for the initial recognition of the V 5–51 and V 4–1 H L CBTAU-28.1 (and their respective affinity-improved vari- germline combination. The same V 5–51 or V 4–1 H L ants) thus have complementary activities that may allow hotspot interactions can be separately found in other these antibodies (or drug modalities mimicking these crystallized antibody complexes [7, 19, 33, 36], but the Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 15 of 17 combination of the two germlines could be more preva- Ethics approval and consent to participate Brain samples were obtained from The Netherlands Brain Bank (NBB), lent against tau, as they seem to get triggered by the Pro Netherlands Institute for Neuroscience, Amsterdam. All donors had given – X – Lys motif abundantly present on the protein. written informed consent for brain autopsy and the use of material and clinical Identification and characterization of these antibodies information for research purposes. Whole blood from healthy male and female donors was obtained from the San Diego Blood Bank (ages 18–65 years) after may thus be exploited to develop antibody or drug regi- informed consent was obtained from the donors. mens for distinct phases of progression of tau pathology and pave the way towards a peptide-based tau vaccine, Competing interests by taking advantage of the apparent immunogenicity of The authors declare that they have no competing interests. the identified motif and presenting both hotspot residues in the right spacing and orientation. Publisher’sNote Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Additional file Author details Janssen Prevention Center, Janssen Pharmaceutical Companies of Johnson Additional file 1: Figure S1. Peptide epitope mapping. Figure S2. & Johnson, Archimedesweg 6, 2333, CN, Leiden, the Netherlands. Janssen Reactivity of CBTAU-27.1 and CBTAU-28.1 to PHF-tau. Figure S3. Affinity Prevention Center, Janssen Pharmaceutical Companies of Johnson & of CBTAU-27.1, CBTAU-28.1 and their affinity-matured mutants for their Johnson, 3210 Merryfield Row, San Diego, CA 92121, USA. Janssen cognate tau peptides. Figure S4. Affinity-improved antibodies Neuroscience Discovery, Janssen Pharmaceutical Companies of Johnson & dmCBTAU-27.1 and dmCBTAU-28.1 retain both the nature and specificity Johnson, Turnhoutseweg 30, 2340 Beerse, Belgium. Molecular and Cellular of the interactions of the parental antibodies with tau. Figure S5. Detection Pharmacology, Discovery Sciences, Janssen Pharmaceutical Companies of of immunoreactivity in various tauopathies by CBTAU-27.1 and CBTAU-28.1 Johnson & Johnson, Turnhoutseweg 30, 2340 Beerse, Belgium. Department and affinity-improved variants. Figure S6. Set-up of an in vitro rtau of Integrative Structural and Computational Biology, The Scripps Research aggregation assay. Figure S7. Complete dataset for in vitro tau aggregation Institute, La Jolla, CA 92037, USA. Proteros Biostructures GmbH, in the absence or presence of CBTAU-27.1 and dmCBTAU-27.1. Figure S8. Bunsenstraße 7a, 82152 Planegg, Germany. Department of Pathology, Complete dataset for in vitro tau aggregation in the absence or Amsterdam Neuroscience, VU University Medical Center, De Boelelaan 1117, presence of CBTAU-28.1 and dmCBTAU-28.1. Figure S9. Complete dataset 1081, HV, Amsterdam, the Netherlands. Skaggs Institute for Chemical for in vitro tau aggregation in the absence or presence of Fab-CBTAU-27.1 Biology, The Scripps Research Institute, La Jolla, CA 92037, USA. Department and Fab-dmCBTAU-27.1. Figure S10. Complete dataset for in vitro of Epidemiology, Harvard T.H. Chan School of Public Health, 677 Huntington tau aggregation in the absence or presence of Fab-CBTAU-28.1 and Avenue, Boston, MA 02115, USA. Department of Neurology, Amsterdam Fab-dmCBTAU-28.1. Figure S11. In vitro tau aggregation in the presence of Neuroscience, Academic Medical Center, Meidreefberg 9, 1105, AZ, dmCBTAU-27.1 added at different time points. Figure S12. Assessment of Amsterdam, the Netherlands. Present address: Janssen R&D US, 3210 CBTAU-27.1 binding to rtau and PHFs by SEC-MALS. Figure S13. Merryfield Row, San Diego, CA 92121, USA. Janssen Vaccines and Macroscopic image of rtau fibrils generated in the absence and presence of Prevention, Janssen Pharmaceutical Companies of Johnson and Johnson, CBTAU-28.1. Figure S14. Tau aggregates are internalized by BV-2 cells and Archimedesweg 6, Leiden, CN 2333, the Netherlands. localize in cellular acidic organelles. Table S1. Names and sequences of tau peptides used in this study. The first 10 peptides listed were used as Received: 6 April 2018 Accepted: 7 May 2018 baits in the BSelex method. Table S2. Data collection and refinement statistics for CBTAU-27.1 Fab and CBTAU-28.1 Fab. Table S3. Data collection and refinement statistics for dmCBTAU-27.1 - A8119 and References dmCBTAU-28.1 - A7731 complexes. (DOCX 30952 kb) 1. Afonine PV, Grosse-Kunstleve RW, EcholsN,Headd JJ,MoriartyNW, Mustyakimov M, Terwilliger TC, Urzhumtsev A, Zwart PH, Adams PD (2012) Towards automated crystallographic structure refinement with phenix.Refine. Acta Crystallogr Sect D Biol Crystallogr 68:352–367. https://doi.org/10.1107/S0907444912001308 Acknowledgements 2. Apetri AC, Vanik DL, Surewicz WK (2005) Polymorphism at residue 129 modulates We would like to thank Mohammed Drissi Saidi, Hector Quirante, Başak Kükrer, the conformational conversion of the D178N variant of human prion protein 90- Otto Diefenbach, Tariq Nahar and Imke Sprengers, for protein generation and 231. Biochemistry 44:15880–15888. https://doi.org/10.1021/bi051455+ analysis, Alberto Carpinteiro Soares and Tjado Morrema for technical assistance, 3. Arevalo JH, Stura EA, Taussig MJ, Wilson IA (1993) Three-dimensional Frederique Bard and Louis de Muynck for valuable comments and advice. structure of an anti-steroid Fab' and progesterone-Fab' complex. J Mol Biol Human brain tissue for the immunodepletion experiments performed in this 231:103–118. https://doi.org/10.1006/jmbi.1993.1260 study was provided by the Newcastle Brain Tissue Resource which is funded in 4. Billingsley ML, Kincaid RL (1997) Regulated phosphorylation and dephosphorylation part by a grant from the UK Medical Research Council (G0400074), by NIHR of tau protein: effects on microtubule interaction, intracellular trafficking Newcastle Biomedical Research Centre and Unit awarded to the Newcastle and neurodegeneration. Biochem J 323(Pt 3):577–591 upon Tyne NHS Foundation Trust and Newcastle University, and as part of the 5. Braak H, Braak E (1991) Neuropathological stageing of Alzheimer-related Brains for Dementia Research Programme jointly funded by Alzheimer’s changes. Acta Neuropathol 82:239–259 Research UK and Alzheimer’sSociety. 6. Bright J, Hussain S, Dang V, Wright S, Cooper B, Byun T, Ramos C, Singh A, Parry G, Stagliano N, Griswold-Prenner I (2015) Human secreted tau increases amyloid-beta production. Neurobiol Aging 36:693–709. https://doi.org/10. Authors’ contributions 1016/j.neurobiolaging.2014.09.007 Project design by AA, GP, JW, and JG; aggregation assay and in vitro 7. Chen X, Zhao, Y., Harlos, K., Snir, O., Sollid, L.M. (2017) Crystal structure of antibody screening by AA and RC; antibody discovery, optimization and anti-gliadin 1002-1E03 Fab fragment in complex of peptide PLQPEQPFP. protein expression and purification by JJ, GP, RJ, EK, TH, JW, HV, BS, EB, DZ, DOI102210/pdb5ijk/pdb. doi:https://doi.org/10.2210/pdb5ijk/pdb DM and AA; FRET based cellular immunodepletion assay by MB, KD, KK, and 8. Cleveland DW, Hwo SY, Kirschner MW (1977) Physical and chemical MM, aggregation, labeling and microglia assay RT, DM, RC, JA, and AA; properties of purified tau factor and the role of tau in microtubule biophysical characterization by AA, RC, BS, JA, HI, MW; immunohistochemical assembly. J Mol Biol 116:227–247 analysis by JJH and KU, X-ray work and analysis by MM, SS, HZ, XZ, WY and 9. Cleveland DW, Hwo SY, Kirschner MW (1977) Purification of tau, a microtubule- IAW; statistical analysis by MK, and manuscript written by WK, EJMS, and AA. associated protein that induces assembly of microtubules from purified tubulin. All authors read and approved the final manuscript. J Mol Biol 116:207–225 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 16 of 17 10. Concepcion J, Witte K, Wartchow C, Choo S, Yao D, Persson H, Wei J, Li P, 30. Lee VM, Balin BJ, Otvos L Jr, Trojanowski JQ (1991) A68: a major subunit of paired Heidecker B, Ma W, Varma R, Zhao LS, Perillat D, Carricato G, Recknor M, Du K, helical filaments and derivatized forms of normal tau. Science 251:675–678 Ho H, Ellis T, Gamez J, Howes M, Phi-Wilson J, Lockard S, Zuk R, Tan H (2009) 31. Lee VM, Goedert M, Trojanowski JQ (2001) Neurodegenerative tauopathies. Annu Label-free detection of biomolecular interactions using BioLayer interferometry Rev Neurosci 24:1121–1159. https://doi.org/10.1146/annurev.neuro.24.1.1121 for kinetic characterization. Comb Chem High Throughput Screen 12:791–800 32. Li W, Lee VM (2006) Characterization of two VQIXXK motifs for tau fibrillization 11. Crespo R, Rocha FA, Damas AM, Martins PM (2012) A generic crystallization- in vitro. Biochemistry 45:15692–15701. https://doi.org/10.1021/bi061422+ like model that describes the kinetics of amyloid fibril formation. J Biol 33. Liao HX, Bonsignori M, Alam SM, McLellan JS, Tomaras GD, Moody MA, Chem 287:30585–30594. https://doi.org/10.1074/jbc.M112.375345 Kozink DM, Hwang KK, Chen X, Tsao CY, Liu P, Lu X, Parks RJ, Montefiori DC, 12. Daebel V, Chinnathambi S, Biernat J, Schwalbe M, Habenstein B, Loquet A, Ferrari G, Pollara J, Rao M, Peachman KK, Santra S, Letvin NL, Karasavvas N, Akoury E, Tepper K, Muller H, Baldus M, Griesinger C, Zweckstetter M, Yang ZY, Dai K, Pancera M, Gorman J, Wiehe K, Nicely NI, Rerks-Ngarm S, Mandelkow E, Vijayan V, Lange A (2012) Beta-sheet core of tau paired Nitayaphan S, Kaewkungwal J, Pitisuttithum P, Tartaglia J, Sinangil F, Kim JH, helical filaments revealed by solid-state NMR. J Am Chem Soc 134:13982–13989. Michael NL, Kepler TB, Kwong PD, Mascola JR, Nabel GJ, Pinter A, Zolla- https://doi.org/10.1021/ja305470p Pazner S, Haynes BF (2013) Vaccine induction of antibodies against a 13. Dehmelt L, Halpain S (2005) The MAP2/tau family of microtubule-associated structurally heterogeneous site of immune pressure within HIV-1 envelope proteins. Genome Biol 6:204. https://doi.org/10.1186/gb-2004-6-1-204 protein variable regions 1 and 2. Immunity 38:176–186. https://doi.org/10. 14. Emsley P, Lohkamp B, Scott WG, Cowtan K (2010) Features and 1016/j.immuni.2012.11.011 development of Coot. Acta Crystallogr Sect D Biol Crystallogr 66:486–501. 34. Mandelkow E, von Bergen M, Biernat J, Mandelkow EM (2007) Structural https://doi.org/10.1107/S0907444910007493 principles of tau and the paired helical filaments of Alzheimer's disease. 15. Fitzpatrick AWP, Falcon B, He S, Murzin AG, Murshudov G, Garringer HJ, Brain Pathol 17:83–90. https://doi.org/10.1111/j.1750-3639.2007.00053.x Crowther RA, Ghetti B, Goedert M, Scheres SHW (2017) Cryo-EM structures 35. McCoy AJ, Grosse-Kunstleve RW, Storoni LC, Read RJ (2005) Likelihood-enhanced of tau filaments from Alzheimer's disease. Nature 547:185–190. https://doi. fast translation functions. Acta Crystallogr Sect D Biol Crystallogr 61:458–464. org/10.1038/nature23002 https://doi.org/10.1107/S0907444905001617 16. Frost B, Jacks RL, Diamond MI (2009) Propagation of tau misfolding from 36. Ofek G, Zirkle B, Yang Y, Zhu Z, McKee K, Zhang B, Chuang GY, Georgiev IS, the outside to the inside of a cell. J Biol Chem 284:12845–12852. https://doi. O'Dell S, Doria-Rose N, Mascola JR, Dimitrov DS, Kwong PD (2014) Structural org/10.1074/jbc.M808759200 basis for HIV-1 neutralization by 2F5-like antibodies m66 and m66.6. J Virol 17. Ginhoux F, Lim S, Hoeffel G, Low D, Huber T (2013) Origin and 88:2426–2441. https://doi.org/10.1128/JVI.02837-13 differentiation of microglia. Front Cell Neurosci 7:45. https://doi.org/10.3389/ 37. Pascual G, Wadia JS, Zhu X, Keogh E, Kukrer B, van Ameijde J, Inganas fncel.2013.00045 H, Siregar B,PerdokG,Diefenbach O,Nahar T, SprengersI, Koldijk MH, 18. Goedert M, Spillantini MG, Jakes R, Rutherford D, Crowther RA (1989) Multiple der Linden EC, Peferoen LA, Zhang H, Yu W, Li X, Wagner M, Moreno isoforms of human microtubule-associated protein tau: sequences and V, Kim J, Costa M, West K, Fulton Z, Chammas L, Luckashenak N, localization in neurofibrillary tangles of Alzheimer's disease. Neuron 3:519–526 Fletcher L, Holland T, Arnold C, Anthony Williamson R, Hoozemans JJ, 19. Gorny MK, Sampson J, Li H, Jiang X, Totrov M, Wang XH, Williams C, O'Neal Apetri A, Bard F, Wilson IA, Koudstaal W, Goudsmit J (2017) T, Volsky B, Li L, Cardozo T, Nyambi P, Zolla-Pazner S, Kong XP (2011) Immunological memory to hyperphosphorylated tau in asymptomatic Human anti-V3 HIV-1 monoclonal antibodies encoded by the VH5-51/VL individuals. Acta Neuropathol 133:767–783. https://doi.org/10.1007/ lambda genes define a conserved antigenic structure. PLoS One 6:e27780. s00401-017-1705-y https://doi.org/10.1371/journal.pone.0027780 38. Sanchez C, Diaz-Nido J, Avila J (2000) Phosphorylation of microtubule-associated protein 2 (MAP2) and its relevance for the regulation of the neuronal 20. Greenberg SG, Davies P (1990) A preparation of Alzheimer paired helical cytoskeleton function. Prog Neurobiol 61:133–168 filaments that displays distinct tau proteins by polyacrylamide gel electrophoresis. Proc Natl Acad Sci U S A 87:5827–5831 39. Surewicz WK, Jones EM, Apetri AC (2006) The emerging principles of mammalian prion propagation and transmissibility barriers: insight from 21. Guo JL, Lee VM (2011) Seeding of normal tau by pathological tau studies in vitro. Acc Chem Res 39:654–662. https://doi.org/10.1021/ar050226c conformers drives pathogenesis of Alzheimer-like tangles. J Biol Chem 286: 40. Thal DR, Rub U, Orantes M, Braak H (2002) Phases of a beta-deposition in 15317–15331. https://doi.org/10.1074/jbc.M110.209296 the human brain and its relevance for the development of AD. Neurology 22. Guo JL, Narasimhan S, Changolkar L, He Z, Stieber A, Zhang B, Gathagan RJ, Iba M, 58:1791–1800 McBride JD, Trojanowski JQ, Lee VM (2016) Unique pathological tau conformers 41. Uchihara T (2007) Silver diagnosis in neuropathology: principles, practice from Alzheimer's brains transmit tau pathology in nontransgenic mice. J Exp Med and revised interpretation. Acta Neuropathol 113:483–499. https://doi.org/ 213:2635–2654. https://doi.org/10.1084/jem.20160833 10.1007/s00401-007-0200-2 23. Harper JD, Lansbury PT Jr (1997) Models of amyloid seeding in Alzheimer's disease and scrapie: mechanistic truths and physiological consequences of 42. Umeda T, Eguchi H, Kunori Y, Matsumoto Y, Taniguchi T, Mori H, Tomiyama T the time-dependent solubility of amyloid proteins. Annu Rev Biochem 66: (2015) Passive immunotherapy of tauopathy targeting pSer413-tau: a pilot study 385–407. https://doi.org/10.1146/annurev.biochem.66.1.385 in mice. Ann Clin Transl Neurol 2:241–255. https://doi.org/10.1002/acn3.171 43. Vagin A, .Teplyakov, A. (1997) MOLREP: an Automated Program for 24. Holmes BB, Furman JL, Mahan TE, Yamasaki TR, Mirbaha H, Eades WC, Belaygorod Molecular Replacement J Appl Cryst 30:1022–1025 L, Cairns NJ, Holtzman DM, Diamond MI (2014) Proteopathic tau seeding predicts 44. Vagin AA, Steiner RA, Lebedev AA, Potterton L, McNicholas S, Long F, tauopathy in vivo. Proc Natl Acad Sci U S A 111:E4376–E4385. https://doi.org/10. Murshudov GN (2004) REFMAC5 dictionary: organization of prior chemical 1073/pnas.1411649111 knowledge and guidelines for its use. Acta Crystallogr Sect D Biol 25. Kabsch W (2010) Xds. Acta Crystallogr Sect D Biol Crystallogr 66:125–132. Crystallogr 60:2184–2195. https://doi.org/10.1107/S0907444904023510 https://doi.org/10.1107/S0907444909047337 45. von Bergen M, Friedhoff P, Biernat J, Heberle J, Mandelkow EM, Mandelkow 26. Kfoury N, Holmes BB, Jiang H, Holtzman DM, Diamond MI (2012) Trans-cellular E (2000) Assembly of tau protein into Alzheimer paired helical filaments propagation of tau aggregation by fibrillar species. J Biol Chem 287:19440–19451. depends on a local sequence motif ((306)VQIVYK(311)) forming beta https://doi.org/10.1074/jbc.M112.346072 structure. Proc Natl Acad Sci U S A 97:5129–5134 27. Kontsekova E, Zilka N, Kovacech B, Skrabana R, Novak M (2014) Identification of structural determinants on tau protein essential for its 46. Weingarten MD, Lockwood AH, Hwo SY, Kirschner MW (1975) A protein factor pathological function: novel therapeutic target for tau immunotherapy essential for microtubule assembly. Proc Natl Acad Sci U S A 72:1858–1862 in Alzheimer's disease. Alzheimer's Res & Ther 6:45. https://doi.org/10. 47. Wu JW, Herman M, Liu L, Simoes S, Acker CM, Figueroa H, Steinberg JI, 1186/alzrt277 Margittai M, Kayed R, Zurzolo C, Di Paolo G, Duff KE (2013) Small misfolded tau species are internalized via bulk endocytosis and anterogradely and 28. Kramer RA, Cox F, van der Horst M, van der Oudenrijn S, Res PC, Bia J, retrogradely transported in neurons. J Biol Chem 288:1856–1870. https://doi. Logtenberg T, de Kruif J (2003) A novel helper phage that improves phage org/10.1074/jbc.M112.394528 display selection efficiency by preventing the amplification of phages 48. Yanamandra K, Kfoury N, Jiang H, Mahan TE, Ma S, Maloney SE, Wozniak DF, without recombinant protein. Nucleic Acids Res 31:e59 Diamond MI, Holtzman DM (2013) Anti-tau antibodies that block tau aggregate 29. Laskowski RA, MacArthur, M.W., Moss, D. S. & Thornton, J.M (1993) PROCHECK: seeding in vitro markedly decrease pathology and improve cognition in vivo. a program to check the stereochemicai quality of protein structures. J Appl Neuron 80:402–414. https://doi.org/10.1016/j.neuron.2013.07.046 Crystallogr 26:283–291 Apetri et al. Acta Neuropathologica Communications (2018) 6:43 Page 17 of 17 49. Yanamandra K, Patel TK, Jiang H, Schindler S, Ulrich JD, Boxer AL, Miller BL, Kerwin DR, Gallardo G, Stewart F, Finn MB, Cairns NJ, Verghese PB, Fogelman I, West T, Braunstein J, Robinson G, Keyser J, Roh J, Knapik SS, Hu Y, Holtzman DM (2017) Anti-tau antibody administration increases plasma tau in transgenic mice and patients with tauopathy. Sci Transl Med 9. https://doi.org/10.1126/scitranslmed.aal2029

Journal

Acta Neuropathologica CommunicationsSpringer Journals

Published: May 31, 2018

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off