In Alzheimer’s disease (AD) brain, one of the histopathological hallmarks is the neurofibrillary tangles consisting of aggregated and hyperphosphorylated tau. Currently many tau binding antibodies are under development to target the extracellular species responsible for the spreading of the disease in the brain.
Journal of Chromatography A 1651 (2021) 462299 Contents lists available at ScienceDirect Journal of Chromatography A journal homepage: www.elsevier.com/locate/chroma Development of immunoprecipitation – two-dimensional liquid chromatography – mass spectrometry methodology as biomarker read-out to quantify phosphorylated tau in cerebrospinal fluid from Alzheimer disease patients Sebastiaan Bijttebier a,1,∗, Clara Theunis b,1, Farid Jahouh a, Dina Rodrigues Martins b, Marc Verhemeldonck a, Karolien Grauwen b, Lieve Dillen a, Marc Mercken b a b DMPK, Janssen Pharmaceutica, Turnhoutseweg 30, Beerse, Belgium R&D Neurosciences, Janssen Pharmaceutica, Turnhoutseweg 30, Beerse, Belgium a r t i c l e i n f o Article history: Received 19 March 2021 Revised 17 May 2021 Accepted 24 May 2021 Available online 28 May 2021 Keywords: Immunoprecipitation Two-dimensional liquid chromatography Metal oxide chromatography Alzheimer’s disease Phosphorylated tau Human cerebrospinal fluid a b s t r a c t In Alzheimer’s disease (AD) brain, one of the histopathological hallmarks is the neurofibrillary tangles consisting of aggregated and hyperphosphorylated tau Currently many tau binding antibodies are under development to target the extracellular species responsible for the spreading of the disease in the brain As such, an in-house developed antibody JNJ-63733657 with picomolar affinity towards tau phosphorylated at both T212 and T217 (further named p217+tau) was recently tested in phase I clinical trial NCT03375697 Following multiple dose administration in healthy subjects and subjects with AD, there were dose dependant reductions in free p217+tau fragments in cerebrospinal fluid (CSF) following antibody administration, as measured with a novel single molecule ELISA assay (Simoa PT3 x PT82 assay), demonstrating epitope engagement of the therapeutic antibody [Galpern, Haeverans, Janssens, TrianaBaltzer, Kolb, Li, Nandy, Mercken, Van Kolen, Sun, Van Nueten, 2020] Total p217+tau levels also were reduced in CSF as measured with the Simoa PT3 x PT82 assay In this study we developed an orthogonal immunoprecipitation – liquid chromatography – triple quadrupole mass spectrometry (IP-LC-TQMS) assay to verify the observed reductions in total p217+ tau levels In this assay, an excess of JNJ-63733657 is added to the clinical CSF to ensure all p217+tau is bound by the antibody instead of having a pool of bound and unbound antigen and to immunoprecipitate all p217+tau, which is followed by on-bead digestion with trypsin to release surrogate peptides Tryptic peptides with missed cleavages were monitored when phosphorylation occurred close to the cleavage site as this induced miscleavages Compared with acidified mobile phases typically used for peptide analysis, reversed phase LC with mobile phase at basic pH resulted in sharper peaks and improved selectivity and sensitivity for the target peptides With this setup a diphospho-tau tryptic peptide SRTPSLPTPPTREPK∗ could be measured with pT217 accounting for at least one of the phospho-sites This is the first time that the presence of a diphopsho-tau peptide is reported to be present in human CSF A two-dimensional LCTQMS method was developed to remove matrix interferences Selective trapping of diphospho-peptides via a metal oxide chromatography mechanism was achieved in a first dimension with a conventional reversed phase stationary phase and acidified mobile phase Subsequent elution at basic pH enabled detection of low picomolar p217+tau levels in human CSF (lower limit of quantification: pM), resulting in an approximate 5-fold increase in sensitivity This enabled the quantification of total p217+tau in CSF leading to the confirmation that in addition to reductions in free p217+tau levels total p217+tau levels were also reduced following administration of the tau mAb JNJ-63733657, correlating with the previous measurement with the PT3 x PT82 Simoa assay An orthogonal sample clean-up using offline TiO2 /ZrO2 ∗ Corresponding author E-mail addresses: sbijtteb@its.jnj.com (S Bijttebier), ctheuni3@its.jnj.com (C Theunis), fjahouh@ITS.JNJ.com (F Jahouh), DRodri39@its.jnj.com (D.R Martins), MVERHEME@its.jnj.com (M Verhemeldonck), kgrauwe@ITS.JNJ.com (K Grauwen), LDILLEN@its.jnj.com (L Dillen), MMERCKEN@its.jnj.com (M Mercken) Authors contributed equally to this work https://doi.org/10.1016/j.chroma.2021.462299 0021-9673/© 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 combined with 1DLC-TQMS was developed to confirm the presence of mono-ptau (pT217) tryptic peptides in CSF © 2021 The Authors Published by Elsevier B.V This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) phorylated at T217 in CSF is superior to traditional tau assays in differentiating amyloid status, was at the same time also demonstrated using a single molecule ELISA assay with the PT3 antibody [20,21] This single molecule ELISA assay (PT3 x PT82 simoa assay) could additionally be used as a target engagement assay for JNJ’3657 when combining it with a denaturing technique [21] The PT3 x PT82 Simoa assay was used as a target engagement assay to analyse CSF from the phase I clinical trial with JNJ’3657 (NCT03375697) Both free p217+tau and total p217+tau levels were measured, with total p217+tau being the free p217+tau fraction plus the JNJ’3657 – p217+tau complexes which can only be measured after denaturation of the complex Free p217+tau levels were found to be reduced, which shows epitope engagement of the large molecule [14] Additionally, total p217+tau levels were found to be reduced, which suggests clearance of JNJ’3657 and the targeted p217+tau molecules To verify clearance of total p217+tau levels after dosing of JNJ’3657, an orthogonal immunoprecipitation – liquid chromatography – triple quadrupole mass spectrometry (IP-LC-TQMS) assay was developed in our lab To achieve the selectivity and sensitivity, necessary for the detection of low pM p217+tau levels in CSF, a 2DLC-TQMS approach was developed This setup was compared with an offline TiO2 /ZrO2 clean-up selective for phospho-peptides The p217+tau levels in human CSF clinical samples were measured with the optimized IP-LC-TQMS methodology and correlation of results with the orthogonal PT3 x PT82 Simoa assay was examined Introduction By 2050, it is expected that almost 19 million people will be suffering from dementia in Europe only, representing 3% of the European population [1] With Alzheimer’s disease (AD) as the leading cause, there is an urgent need for disease modifying therapies that can prevent or halt the progression of the disease Besides neurodegeneration, neuropathological hallmarks for AD are the extracellular amyloid plaques and the intracellular aggregates of protein tau Due to the dominantly inherited mutations in amyloid precursor protein or the presenilin genes PSEN1 and PSEN2 that were found to cause AD [2,3], amyloid protein has been the main target for disease modifying therapies for many years Currently, the strong link between tau pathology and cognitive decline is recognized and targeting tau pathology has been included in research strategies [4–6] The intracellular neurofibrillary tangles and neuropil threads consist of misfolded and abnormally phosphorylated tau species that can be released in the extracellular environment and seed pathology intracellularly in neighbouring cells, contributing to the spatio-temporal progression of tau pathology [7–11] This extracellular tau seed is targeted by active and passive immunotherapy approaches that are currently under preclinical and clinical investigation A novel phosphorylated tau selective monoclonal Ab PT3 was generated with picomolar affinity towards tau phosphorylated at both T212 and T217 Reduced binding affinity is observed for tau monophosphorylation at either T212 or T217, while the effect of phosphorylation at S210 and S214 seemed limited [12,13] Its humanized variant, named JNJ-63733657 (JNJ’3657) is currently in clinical development [14] The phosphorylated tau species targeted by PT3 and JNJ’3657 will be further referred to as p217+tau The combined measurements of a decrease in Aβ 1–42 and an increase in total tau and tau phosphorylated at T181 in CSF with immuno-assays, have been commonly used as diagnostic and prognostic biomarkers for Alzheimer’s disease [16] However, considering currently developed therapeutic strategies and ongoing clinical trials, additional biomarkers are needed for a more precise and even earlier prediction of disease onset and a further increase in the accuracy of the diagnosis Cerebrospinal fluid (CSF) is in continuous exchange with brain interstitial fluid in direct contact with neurons [16] Tau species found in CSF however differ from tau in brain: Sato et al [17] discovered that the predominant forms of tau (99.9%) in CSF are Cterminally truncated containing the mid-domain but lacking the microtubule binding region and C-terminus, with cleavage between amino acid (AA) residues 222 and 225 As a consequence, only part of the human brain tau information is present in CSF Moreover, the concentration of tau in CSF is three orders of magnitude lower than in the brain [17] Barthélemy et al recently demonstrated by using liquid chromatography – mass spectrometry (LCMS) methodology that T217 phosphorylation (pT217) is considerably increased in CSF of AD patients [15,18] Moreover, T217 hyperphosphorylation occurred systematically in CSF of amyloid positive participants even at pre-symptomatic stage, thus making it a potential important target for (immuno)therapeutic development [15,18] In another study, Barthélemy et al showed that CSF pT217 outperforms pT181 as a means of AD diagnosis [19] That tau phos- Materials and methods 2.1 Materials 2.1.1 Chemicals Ammonium bicarbonate, formic acid (FA) 98–100%, Tween20, ammonia solution 25% (Suprapur), acetic acid (glacial), lactic acid, glycerol, 3-hydroxypropanoic acid, citric acid, glutamic acid, pyrrolidine and dithiothreitol were supplied by Merck MilliQ water with a resistivity of 18.2 M cm at 25 °C was generated with a MilliporeTM -purification system Acetonitrile (ULC-MS) was bought from Actu-All Chemicals Recombinant tau-441 was acquired from Promise Proteomics Synthetic peptides H-TPSLPTPPTR-OH, H(pT)PSLPTPPTR-OH, H-TPSLP(pT)PPTR-OH, H-(pT)PSLP(pT)PPTR-OH, H- (pT)PSLP(pT)PPTREPK-OH, H-SR(pT)PSLP(pT)PPTREPK-OH and H-SR∗ (pT)PSLP(pT)PPTR∗ EPK-OH (∗ : Arginine labelled with 13 C and 15 N) were purchased from Pepscan Phosphate buffered saline (PBS) was purchased from Roche Antibodies PT9 [12], and PT3 and its humanized variant JNJ-63733657 (JNJ’3657) [13] were produced in-house SCBiot-(dPEG4)-GTPGSR-S210-R-T212-P-S214LP-T217-PPTREPKK-amide with different phosphorylation patterns at amino acids S210, T212, S214 and T217 were obtained from New England Peptides 2.1.2 Biological samples All animal procedures were done strictly according to the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care International (AAALAC), with the European communities Council Directive of 24th November 1986 (86/609/EEC) and with protocols approved by the local Institutional Animal and Use Ethical committee S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 Three-month-old P301L tau transgenic mice [22], bred in-house, were anesthetized and forebrain was collected, weighed and immediately put on dry ice A volume of homogenization buffer (buffer H - consisting of 10 mM Tris/HCl, 0.8 M NaCl, 10% sucrose and mM EGTA in MilliQ at pH 7.6 – and tablet/10 mL phosphatase and protease inhibitors (Roche)) corresponding to six times the weight of the brain tissue was added and incubated on ice for 15 The samples were homogenized with a FastPrep®−24 instrument (MP Biomedicals) at 6.0 m/s for 20 s The samples were subsequently centrifuged at 20,0 0 × g for at 4°C (Heraeus Megafuge 8R centrifuge, Thermo Fisher Scientific) The supernatant was centrifuged at 34,0 0 × g for 20 at 4°C (OPTIMATM MAX-XP, Beckman Coulter) The supernatant was collected as total brain homogenate (BH) A pool of homogenates of forebrain was prepared by combining equal amounts of homogenate from each animal and vortex mixing Human CSF samples with low, medium and high tau concentrations were prepared by pooling de-identified CSF samples from healthy subjects and subjects with AD based on their total tau concentration as measured with ELISA (Innotest, Fujirebio) For analysis on the effect of JNJ’3657 on p217+tau in CSF, human CSF samples from the multiple ascending dose trial in subjects with prodromal or mild AD (NCT03375697) were used All CSF samples were collected with informed consent from the subjects In this study, subjects received either placebo or three doses of one of three dose levels of JNJ’3657 intravenously A total of 26 baseline and post-dose samples were analysed and were blinded to dose level [14] (Eppendorf) with 200 μL 0.01% Tween20 in PBS per IP reaction The tubes were placed in a DynaMagTM −2 Magnet (Thermo Fisher Scientific) and the supernatant was discarded 56 μL PBS + 0.1% Tween20, 100 nM JNJ’3657 and 500 μL (artificial) CSF were added to the beads The samples were mixed and incubated overnight at °C while rotating on a hula mixer (Thermo Fisher Scientific) The supernatant was collected as immunodepleted fraction 300 μL (50 mM ammonium bicarbonate (pH 8) + 10% of 0.1% Tween in PBS) was added to the beads and subsequently vortex mixed and spun (Minispin Plus centrifuge, Eppendorf) The supernatant was collected as wash and the beads were resuspended in 150 μL of 50 mM ammonium bicarbonate, pH The samples were stored on melting ice while preparing reagents for trypsinization: trypsinization was started on the day IP was finished 2.2.3 Trypsinization 15 μL of acetonitrile was added to the beads + 150 μL of 50 mM ammonium bicarbonate after IP, vortex mixed for s and spun down for 15 s with a Minispin Plus centrifuge 15 μL of 0.1 μg trypsin mL−1 in 50 mM acetic acid was added (Trypsin Gold, Promega) and the samples were subsequently incubated at 37 °C for 20 h while shaking at 1200 rpm (ThermoMixer, Thermo Fisher Scientific) for on-bead digestion Afterwards, the digestion was quenched by adding 15 μL of formic acid and briefly vortex mixing Subsequently, 20 μL of SIL working solution or 20 μL blank solvent (for ‘blank’ (supplemental information 1): 90:10:0.1 MilliQ water:acetonitrile:formic acid) was added Thereafter, 20 μL of non-labelled working solution (for calibration points and QCs for batch acceptance (supplemental information 1)) or blank solvent (90:10:0.1 MilliQ water:acetonitrile:formic acid, for CSF samples) was spiked The samples were vortex mixed and centrifuged at 20,0 0 × g for 10 (Heraeus Megafuge 8R centrifuge, Thermo Fisher Scientific) and the supernatant was transferred to micronic tubes (Micronic) The micronics were sealed in a 96 well plate with sealing mat and stored at °C until analysis 2.2 Sample preparation 2.2.1 Preparation of standard dilutions All standard dilutions were prepared in low bind Eppendorf tubes Calibration standard spike solutions: a stock solution of 200 μM was prepared in water:acetonitrile:acetic acid (89:10:1) The stock solution was aliquoted and stored at −80 °C Standard working dilutions were prepared in 90:10:0.1 water:acetonitrile:formic acid (stored at −20 °C) and used for the production of calibration points ranging from 0.5 pM to 10 pM as described in supplemental information Standard working dilutions were freshly prepared for each batch of samples Spike solutions for batch acceptance QCs (supplemental information 1) were prepared from the same stock solutions used for production of calibration standard spike solutions Preparation of stable isotopically labelled (SIL) stock solution and working solutions: a stock solution of 200 μM was prepared in water:acetonitrile:acetic acid (89:10:1), aliquoted and stored at −80 °C SIL working dilutions were prepared in 90:10:0.1 water:acetonitrile:formic acid (stored at −20 °C and freshly prepared for each batch of samples) 2.2.4 TiO2 /ZrO2 clean-up Clean-up of immunoprecipitated and trypsinized CSF samples was performed with TiO2 /ZrO2 solid phase extraction (SPE) followed by Oasis HLB SPE TopTips containing 10 mg TiO2 /ZrO2 (GlySci) were conditioned with × 50 μL 7.7% FA in 90:10 water:acetonitrile saturated with glutamic acid (conditioning buffer) by centrifugation at 20 0 rpm for with a Minispin Plus centrifuge 55 μL of a MilliQ solution saturated with glutamic acid and containing 7.7% formic acid was added to 120 μL of digest The sample was loaded on a conditioned TopTip by centrifugation at 10 0 rpm for The sample was reloaded twice The TopTip was washed with 50 μL conditioning buffer, × 50 μL 50:50 water:acetonitrile + 2% formic acid and × 50 μL water (at 20 0 rpm for min) Phospho-peptides were eluted with 50 μL 10% NH4OH and × 50 μL 5% pyrrolidine (each step at 20 0 rpm for min), eluants were pooled and 15 μL formic acid and 10 μL acetonitrile was added Ammonia and pyrrolidine were removed from the eluant with Oasis HLB (96-well Plates, 30 mg, 30 μm, Waters) The stationary phase was conditioned with 200 μL acetonitrile (centrifuged at 600 rpm for min), equilibrated with 100:2 water:formic acid (600 rpm for min), the TiO2 /ZrO2 eluant was loaded (5 at 300 rpm followed by at 600 rpm), followed by washing with × 80 μL 10 0:2 water:formic acid and 400 μL water (each step: at 10 0 rpm followed by at 20 0 rpm) Peptides were eluted with 150 μL 50:50 water:acetonitrile (5 at 10 0 rpm and at 20 0 rpm) The eluant was evaporated to dryness with nitrogen gas at 45 °C and redissolved in 150 μL of 90:10:0.1 water:acetonitrile:formic acid and stored at °C until analysis 2.2.2 Immunoprecipitation For IP-LC-TQMS assay development, either a 1/20 dilution of transgenic mouse brain homogenate in artificial CSF (12.4 mM NaCl, 0.125 mM NaH2 PO4 H2 O, 0.13 mM MgSO4 •7H2 O, 0.27 mM KCl, 2.6 mM NaHCO3 , 18 mM D-glucose.H2 O, mM ascorbic acid, mM CaCl2 in H2 O, pH 7.25) or pools of human CSF with either low (< 350 pg mL−1 tau), medium (350 >< 750 pg mL−1 tau) or high level (> 750 pg mL−1 tau) of tau were used and all spiked with JNJ’3657 antibody to mimic the presence of the dosed antibody in the clinical samples 2.2.2.1 Optimized immunoprecipitation procedure 93.5 μL of Dynabeads protein G (Thermo Fisher Scientific) corresponding to 2.8 mg beads was washed two times in LoBind Eppendorf tubes S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 Fig Schematic depiction of the column connections to the 6-way valve in the column oven Dotted line: valve position 0, solid line: valve position Table LC-gradient and valve switching time program of the 2DLC-TQMS method Mobile phase solvents A and B are connected, and mobile phase solvents C and D are connected 2.3 LC-TQMS analysis Both 1DLC- and 2DLC-TQMS analyses were performed on an ultra-high performance liquid chromatograph from Shimadzu consisting of Nexera LC30AD liquid chromatographs set up to provide dual binary solvent gradients, a SIL-AC30 autosampler, a CTO20AC column oven with an integrated 6-way valve, a communications bus module (CBM-20A) and a sample Rack Changer II, hyphenated via a Turbo-IonsprayTM Interface (Sciex) to a 6500 triple quadrupole mass spectrometer (Sciex) A separate Nexera LC20AD liquid chromatograph (Shimadzu) was used for post-column addition of 100 μL mL−1 acetonitrile via a T-piece Analyst 1.6.3 (Sciex) was used as instrument control and data processing software 2.3.1 1DLC-TQMS analysis For 1DLC-TQMS analysis, 50 μL of digest was injected on ˚ 1.7 μm, an ACQUITY UPLC Peptide BEH C18 Column, 300 A, mm × 100 mm (Waters) and thermostatically (60 °C) eluted The mobile phase (MP) solvents consisted of 100:1 water + 0.05% ammonia:acetonitrile (v:v) (A) and acetonitrile + 0.05% ammonia (v:v) (B), and the gradient was set as follows (min/A%): 0.0/100, 0.5/100, 5.0/70, 5.1/2, 6.6/2, 6.7/10 0, 10/10 The flow rate was set at 0.2 mL/min The probe vertical millimetre setting was adjusted to mm to improve sensitivity The peptides were ionized with electrospray ionisation (ESI) in positive ion mode The ionspray voltage was set to 4500 V, temperature to 400 °C, declustering potential to 60 V and entrance potential to 10 V Ion source gas 1, gas and curtain gas were set to 50, 40 and 30, respectively CAD gas was set to The selected MS transitions used for multiple reaction monitoring of the target peptides are provided in supplemental information Time (min) Module Events 0.0 0.0 0.0 2.1 2.6 2.7 3.7 9.7 14.7 14.8 14.8 15.4 15.5 20 Oven Pumps Pumps Pumps Pumps Pumps Pumps Oven Pumps Pumps Oven Pumps Pumps Pumps System Controller Valve Pump Pump Pump Pump Pump Pump Valve Pump Pump Valve Pump Pump Pump Stop Parameter B Conc D Conc B Conc B Conc B Conc B Conc D Conc D Conc D Conc D Conc D Conc 1 98 98 1 30 98 98 2.3.3 Analysis of clinical samples Clinical CSF samples (n = 26) from the multiple ascending dose trial in subjects with prodromal or mild AD (NCT03375697) were processed with the optimized IP-2DLC-TQMS protocol CSF samples were collected at different timepoints after dosing with either placebo or one of three dose levels of JNJ’3657 [14] Quality control criteria for the optimized IP-2DLC-TQMS protocol were used during analysis of human CSF samples as described in supplemental information Correlation between results obtained for surrogate peptide/p217+tau levels in the clinical CSF samples with IP-2DLCTQMS and the PT3 x PT82 Simoa assay was calculated by using a linear regression after transforming the data (X=Log(X); Y=Log(Y)) Results and discussion 2.3.2 2DLC-TQMS analysis For 2DLC-TQMS analysis, 50 μL of digest was injected ˚ 1.7 μm, An ACQUITY UPLC Protein BEH C4 Column, 300 A, 2.1 mm × 50 mm (Waters) was used in a first dimension with water + 0.1% formic acid (A) and acetonitrile (B) as mobile phase solvents and a flow rate of 0.4 mL mL−1 In a second di˚ 3.5 μm, mension an XBridge Peptide BEH C18 Column, 300 A, mm × 100 mm (Waters) was used with mobile phase solvents consisting of 100:1 water + 0.05% ammonia:acetonitrile (v:v) (C) and acetonitrile + 0.05% ammonia (v:v) (D) and a flow rate of 0.2 mL mL−1 Column oven temperature was set at 60 °C and the connections of the 6-way valve were as depicted in Fig The LC-gradient and valve switching time program are described in Table The same MS settings were used as during 1DLC-TQMS analysis 3.1 Digestion – formation of tryptic peptides with missed cleavages During the current study, an IP-LC-TQMS assay was developed to determine low pM quantities of p217+tau in human CSF and brain tissue homogenates of mice Detection of doubleor triple-phosphorylated tau tryptic peptides have not been reported up to now as their stoichiometry is assumed to be lower than mono-phosphorylated peptides, unless there is biological coordination (e.g priming effect) of site phosphorylation [23] The Ab PT3 and its humanized variant JNJ’3657 however exhibit picomolar affinity for tau phosphorylated at both T212 and T217, while its affinity for tau monophosphorylated at T212 or T217 is respectively 16-fold and 6-fold lower Phosphorylation at S210 and/or S214 has only a limited effect on the binding affinity S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 Table Tryptic digests of elongated peptide SCBiot-(dPEG4)-GTPG-S210-R-T212-P-S214-LP-T217-PPTREPKK-amide with different phosphorylations show different cleavage patterns Results are expressed as peak areas of the different tryptic peptides formed, relative to the peak area of the tryptic peptide with the highest peak area Data are recorded on an UHPLC–HRMS system in TOF MS mode (660 0, Sciex), mass range: m/z 30 0–180 The 1DLC-method described in section Materials and Methods was used ∗ loss of phospho-moiety at S210 by tryptic cleavage at R211 AAs in bold represent the elongation due to missed cleavage Phosphorylation – T217 S214/T217 T212/T217 T212/S214 T212/S214/T217 S210/S214/T217 TPSLPTPPTR SRTPSLPTPPTR TPSLPTPPTREPK SRTPSLPTPPTREPK 100 – 12 – 100 – 18 – 100 53 32 19 2.5 100 – 50 2.3 100 – 15 – 100 – 64 75∗ 100 26∗ 52 [13] Potential capture of tau with double and triple phosphorylation in the Ab epitope region was therefore also considered in this project A set of elongated peptides containing the tau epitope region of the Ab (SCBiot-(dPEG4)-GTPG-S210-R-T212-PS214-LP-T217-PPTREPKK-amide, AA numbering based on full length 2N4R tau) with different phosphorylation patterns at amino acids S210, T212, S214 and T217 was available from epitope mapping experiments, all biotinylated at N-terminal and amidated at Cterminal These elongated peptides were used to investigate the influence of phosphorylation on formation of miscleavages during trypsinization Separate dilutions of the elongated peptides were trypsinized overnight Table shows per elongated peptide the peak areas of the different tryptic peptides formed, relative to the peak area of the tryptic peptide with the highest peak area These data indicate that formation of miscleavages during trypsinization is dependant on the site of phosphorylation, as described previously by others [24] The negative charge of phosphorylated serine or threonine positioned next to the basic arginine or lysine forms salt bridges and competes with the complementary aspartic acid at the trypsin active site [25] For example, when a peptide is phosphorylated at T212, tryptic peptides miscleaved at R211 (based on full length 2N4R tau) are abundant (Table 2), for peptides phosphorylated at T212/T217 and T212/S214 predominantly tryptic peptides with miscleavage at R211 are detected, while for phosphorylation at S214/T217 the tryptic peptide with miscleavages is most abundant Additionally, when comparing the results for peptides phosphorylated at pT212/pS214/pT217 and pS210/pS214/pT217, phosphorylation at the N-terminal side of R211 (S210) seems to have less inhibitory activity on tryptic cleavage at R211 than phosphorylation at its C-terminal side (T212) (peptide [Btn]-GTPGSR(pS)RTP(pS)LP(pT)PPTR was not detected) Replacement of trypsin with another enzyme with cleavage sites not influenced by the phosphorylation could potentially result in the formation of peptides without missed cleavages, thereby simplifying data analysis This was not explored in the current work It is known that when glutamic acid or aspartic acid is located next to a trypsin cleavage site a similar inhibitory mechanism is exhibited [25] In the AA sequence of the elongated peptides under study, glutamic acid is present next to the R221 trypsin cleavage site (based on full length 2N4R tau), enhancing the probability of a missed cleavage being formed Missed cleavages at this location of the tau sequence have been reported previously [15–17] Formation of miscleavages should therefore be considered in the search for p217+tau tryptic peptides in BH and CSF samples for the peptides of interest when using 0.1% FA as mobile phase additive Moreover, this mobile phase composition rendered low chromatographic resolution of the target peptides, which consist of multiple isomeric compounds differing only in phosphorylation site In case of chromatographic overlap, diagnostic product ions are needed for differentiation of these isomers, thereby limiting product ion selection options (e.g selection based on absence of matrix interferences, sensitivity) Many studies report that phosphate groups are easily lost during collision-induced dissociation, as phospho-peptides are often preferentially fragmented at the phospho-sites thereby making localization of the phospho-site challenging [26] During the current work, predominantly proline (Pro) y-ions were observed: the Pro-effect is a well-known fragmentation in MSMS spectra of peptides, in which selective cleavage commonly occurs at the N-terminal side of Pro-with mobile protons to form abundant y-ions [27] Next to broad peak shapes and low chromatographic resolution obtained with acidified MP, high carry-over levels were observed for the target phospho-peptides, most probably due to interactions of the phospho-moieties with stainless steel and free silica at low pH [28] It has been stated that the use of higher pH mobile phases can limit these interactions [28,29] Others have suggested that these interactions have no relation to pH as indicated by the ionic condition of phosphate compounds, and because the interactions can be suppressed by making use of metal chelators such as EDTA and medronic acid or ion-pairing reagents [29,30] As the peaks observed for the non-phosphorylated target peptides (e.g TPSLPTPPTR) were also broad when using acidic mobile phase and as this was not the case for other tryptic tau peptides (investigated with recombinant tau digests on HRMS, data not shown), this indicates that peak broadening of the target peptides is also related to its amino acid sequence It has been reported that this can be caused by slow cis-trans peptide bond isomerisation of Pro-Promoieties [31] Most peptide bonds overwhelmingly adopt the trans isomeric form under unstrained conditions mainly because of the weaker steric repulsion effects, but peptide bonds to N-substituted amino acids such as Pro can populate both isomers [32] Because of the partial double bond character of the amide bond and reasonable high barrier of conformational transformation, the cis–trans isomerization of peptide bonds is a relatively slow process [32] Increase in column temperature however increases isomerization rates, resulting in symmetric peaks for peptides containing a ProPro-moiety [31] During this current study the influence of column temperature and pH on chromatography of the target peptides was investigated The results in Table show that peak widths at pH are similar for TPSLPTPPTR and TPSLP(pT)PPTR at 25 °C When the column temperature is increased, peak widths of the non-phospho-peptide decrease to s, which is in agreement with the increasing isomerization rates described by Griffits and Cooney [31] The peak widths of TPSLP(pT)PPTR however remain constant with increasing column temperature: this is most probably due to interactions of the phospho-moiety with free silanols and stainless steel Increas- 3.2 1DLC-TQMS analysis Tryptic digests of the elongated peptides were used as reference during 1DLC-TQMS method optimization, next to a set of (non)phosphorylated synthetic peptide standards representing tryptic tau peptides with miscleavages (TPSLPTPPTR) Most often, acidic mobile phases (e.g 0.1% FA) are used for the separation of peptide mixtures with LC However, broad peak shapes were observed 7.53 7.30 7.05 20 19 20 7.61 7.40 7.12 9.8 7.8 5.7 3.68 3.64 3.58 8.2 7.2 7.0 2.86 2.78 2.72 22 15 9.6 7.56 7.41 7.23 16 14 12 5.24 5.04 4.83 ing the retention by lowering the gradient speed (from 1% to 30% B in 15 instead of min) results in doubling of the peak widths at 25 °C This can be explained by the longer residence time of the peptides in the column (retention times approximately double) and thereby more time for peak broadening because of isomerization Peak fronting also increases with longer residence time in the column Similar to when a fast gradient is applied, increasing column temperature results in sharper peak widths for the non-phosphopeptide (from 21 s to 9.6 s) while the peak width of TPSLP(pT)PPTR remains constant In contrast with chromatography at pH 3, peaks of the TPSLP(pT)PPTR peptide eluted at pH 11 become narrower when the column temperature is increased, indicating less interactions of the phospho-moiety with the stationary phase and stainless steel Next to that, change of the mobile phase pH also results in a shift in selectivity The peptides TPSLPTPPTR and TPSLP(pT)PPTR co-elute at pH while at pH 11 they are easily separated: TPSLP(pT)PPTR elutes earlier, hence its narrower peak widths obtained at pH 11 (Table 3) Chromatography was compared with a larger set of reference tau peptides to investigate the influence of the mobile phase pH on selectivity (Fig 2) Best results were obtained with 0.05% ammonium hydroxide as mobile phase additive (pH 11), rendering sharp chromatographic peaks and separating target peptides based on degree of phosphorylation (following the trend for capacity factor: non-phospho- > mono-phospho- > di-phospho-peptides) Changes in the charge-states of phosphate groups and (e.g basic) amino acids by changes in pH contribute to the overall net charge and therefore hydrophilicity of peptides thereby altering chromatographic selectivity and retention [33] Next to increased selectivity and smaller peak widths, less carryover was observed at pH 11 Moreover, an average 5-fold increase in peak area and height was observed for both TPSLPTPPTR and TPSLP(pT)PPTR Therefore, ammonium hydroxide was retained as mobile phase additive for further experiments To optimize sensi˚ 1.7 μm, tivity, an ACQUITY UPLC Peptide BEH C18 Column, 300 A, 1.0 mm × 100 mm was used with mobile phase flow rate of 0.2 mL min−1 and post-column addition of 0.1 mL min−1 acetonitrile to enhance mass spec ionization efficiency Notwithstanding the small inner diameter of the column, injections of 50 μL of standards and samples rendered sharp peaks for the peptides of interest 3.3 Optimization of IP and digestion of diluted mouse BH and human CSF 10 8.0 5.8 25 °C 45 °C 60 °C 10 8.6 10 3.67 3.59 3.51 21 15 9.6 To be able to quantify low pM total p217+tau levels in human CSF derived from the multiple ascending dose trial in subjects with prodromal or mild AD (NCT03375697) with LC-TQMS, an IP protocol with JNJ’3657 was developed These clinical CSF samples already contain JNJ’3657 derived from the treatment As described in [14], the dosed antibody is bound to p217+tau in CSF resulting in a fraction of p217+tau in complex with JNJ’3657 and a fraction of free p217+tau The fraction of free p217+tau in CSF decreases after antibody dosing in a dose dependant manner, as more and more antibody will form a complex with p217+tau Total p217+tau (i.e the sum of p217+tau in complex with JNJ’3657 and free p217+tau) also was observed to be decreased In order to measure the total p217+tau levels with immunoassays, JNJ’3657 cannot be covalently bound to the beads prior to sample addition, as then only free p217+tau would be captured Instead, a protocol was designed in which JNJ’3657 is first added in excess to the sample to bind all p217+tau present after which magnetic beads with protein G were used to capture all JNJ’3657 either bound or not to p217+tau A scheme of the sample preparation procedure is provided in supplemental information As human CSF samples are scarce, a pool of total homogenates of forebrain of 3-month-old P301L tau transgenic mice [22] was prepared These mice overexpress human full 3.64 3.57 3.49 TPSLP(pT)PPTR TPSLPTPPTR peak width retention (s) time (min) peak width retention (s) time (min) TPSLP(pT)PPTR peak width retention (s) time (min) peak width retention (s) time (min) TPSLPTPPTR TPSLP(pT)PPTR TPSLPTPPTR peak width retention (s) time (min) peak width retention (s) time (min) column temperature TPSLP(pT)PPTR TPSLPTPPTR peak width retention (s) time (min) pH 11 long gradient pH long gradient pH 11 short gradient peak width retention (s) time (min) Journal of Chromatography A 1651 (2021) 462299 pH short gradient Table Peak width in seconds, at 5% peak height of peptides TPSLPTPPTR and TPSLP(pT)PPTR (average of replicate injections) under different chromatographic conditions Analytical column: ACQUITY UPLC Peptide BEH C18 Column, ˚ 1.7 μm, 2.1 mm × 50 mm Column temperature: 25 °C, 45 °C and 60 °C Mobile phase at pH 3: A: water + 0.1% FA, B: acetonitrile Mobile phase at pH 11: A: water + 0.05% ammonia, B: acetonitrile + 0.05% ammonia 300 A, Short linear gradient: from – min: 1% B to 30% B Long linear gradient: from – 15 min: 1% B to 30% B Injection volume: 10 μL Maximum relative standard deviation (RSD) of peak widths per condition: 10.7% Maximum RSD of retention time per condition: 0.4% S Bijttebier, C Theunis, F Jahouh et al S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 Fig Chromatography of reference peptides representing the target tryptic peptides of p217+tau and tau, analysed with the 1DLC-MSMS method described in Materials ˚ 3.5 μm, 1.0 mm × 100 mm Lower chromatogram: 0.1% FA in mobile phase Upper chromatogram: 0.05% NH4 OH in and methods, using an XBridge peptide BEH 300 A, mobile phase length 2N4R tau under the thy1 promotor and not develop aggregated forms of tau at this age As a human CSF surrogate for method optimization, the BH pool was diluted 20-fold in artificial CSF Experiments were conducted to make sure all p217+tau was captured and to increase method sensitivity, by optimising the amount of JNJ’3657, magnetic protein G beads, sample intake and volume of ammonium bicarbonate buffer added after IP A comparison was made between on-bead digestion after IP and digestion following tau elution after IP The elution protocol consisted of of boiling at 98 °C in the presence of dithiothreitol (DTT), immediately followed by 10 cooling of the sample in ice water and a centrifugation step of 30 at 20,0 0 × g As tau is a heat stable protein [34] it will remain in the supernatant while most other proteins, including the capture antibody JNJ’3657, will denature and be centrifuged down to the pellet When performing on-bead digestion, more substrates (both protein G and Ab together with ptau) for trypsin are available, potentially resulting in less efficient digestion of p217+tau and more matrix effects during LC-TQMS analysis On the other hand, tau could be partly lost during elution after IP, e.g by non-complete elution or by inclusion between the denatured antibody aggregates Moreover, it was observed that DTT partly inhibited trypsin, potentially leading to variability, and that matrix effects were still present during LC-TQMS analysis of sample extracts obtained with digestion following tau elution It was decided to use on-bead digestion in future experiments and minimize matrix effects via chromatography An experiment was conducted to compare p217+tau tryptic peptides detected in the 20-fold diluted BH pool and pools of human CSF samples with low, medium or high total tau levels (2, and 11 pM ptau, respectively, as measured with the PT3 x PT82 simoa assay [21]) The samples were IP-ed with JNJ’3657 and two human IgG1 antibodies (isotype control and 2) as negative controls to assess non-specific binding (Table 4) When p217+tau from diluted BH samples were IP-ed with JNJ’3657, both monophosphorylated (pT217) and double phosphorylated tryptic peptides of tau were detected For pT217, both TPSLP(pT)PPTR (0 miscleavages) and TPSLPTPPTREPK∗ p (one miscleavage at C-terminal side, phosphorylation site undetermined) were detected The double phos- phorylated tryptic peptide was only detected with two missed cleavages (SRTPSLPTPPTREPK∗ 2p) This formation of missed cleavages agrees with what was observed with the set of elongated p217+tau peptides: inhibition of trypsinization due to phosphorylation In the negative control samples and when using JNJ’3657, non-phosphorylated tau tryptic peptides are also detected Morris et al also detected a tryptic peptide diphosphorylated most likely at T212 and T217 in mouse BH [35] When an additional wash step was conducted after IP, the peak areas of non-modified tau tryptic peptides diminished, suggesting mainly antibody independent binding of non-phosphorylated tau to the beads Some low affinity binding of non-modified tau with JNJ’3657 during IP cannot be completely excluded as non-phosphorylated tau is present in excess over p217+tau in CSF and brain homogenates Also in human CSF pools, double phosphorylated tryptic peptides were detected (SRTPSLPTPPTREPK∗ 2p) Moreover, an increase in peak areas was observed for SRTPSLPTPPTREPK∗ 2p in the order low748.2 with product ion corresponding to ‘y6’ fragmentation TPSLP(pT)PPTR is chromatographically separated from its isomers phosphorylated at other sites allowing usage of the product ion ‘y6 - phospho moiety’ for unambiguous identification This increased sensitivity at least 5-fold (Fig 4) No improvement in sensitivity was observed for SRTPSLPTPPTREPK∗ 2p during 1DLC-TQMS analysis, indicating that not all matrix interferences were removed with the clean-up For this peptide, 2DLCTQMS proved to be the most sensitive option 3.5 TiO2 /ZrO2 clean-up to enable 1DLC-TQMS of target phospho-peptides 1DLC-TQMS sensitivity of monophospho-peptides was limited by matrix compounds originating from sample preparation The optimized 2DLC methodology did not allow to confirm the presence of tau phosphorylated at T217 in human CSF as the monophospho-peptides are not retained on the C4-column Notwithstanding these mono ptau peptides were previously reported by Barthélemy et al [15,23] A sample clean-up protocol was developed in our lab with TiO2 /ZrO2 SPE in order to confirm the presence of tau phosphorylation at T217 in human CSF In contrast to online metal oxide chromatography, its offline application is not constrained by the chemicals used, that could otherwise decrease sensitivity of the LC-MS system or limit lifetime of instrument consumables Metal oxides such as TiO2 and ZrO2 show HILIC properties as well as anion exchange properties under acidic conditions: below pH 2.7, the carboxy group at Aspand Glu-residues and the C-terminus of tryptic digests are largely undissociated and the amino group at Lys-, His-, and Arg-residues and the N-terminus are positively charged [39] Therefore, nonphosphorylated peptides including ordinary acidic ones are largely unretained by a WAX column because of electrostatic repulsion between the solute and the solid phase, while phospho-peptides are slightly retained since phosphate groups are dissociated under these conditions [39] Nonetheless, Asp- or Glu-rich peptides were proven to exhibit non-specific binding to TiO2 beads [26] Various acidic additives such as lactic acid and glutamic acid have been reported to decrease non-specific binding of acidic peptides because of competition for interaction with metal oxide binding sites [40] It was shown that these additives not compete with phospho-peptide binding, probably because of a different geometry of phospho-peptide binding compared with non-specific peptide binding [40,41] During this study several additives (lactic acid, glycerol, hydroxypropanoic acid, citric acid and glutamic acid) were tested in the loading solvent to minimize non-specific binding: based on the recovery of target phospho-peptides and removal of matrix interferences (matrix peptides and detergents), glutamic acid provided an overall best performance The release of bound peptides during phospho-peptide enrichment is usually done with alkaline solutions containing for example ammonium bicarbonate, ammonium hydroxide or pyrrolidine: different eluents provide different phospho-peptide spectra and successive elution with various elution buffers can significantly improve phospho-peptide recovery [42] Fukuda et al observed that there is little correlation be- 3.6 Analysis of clinical samples The project goal was to confirm the decline of total p217+tau levels over time in CSF from patients dosed with JNJ’3657 in the multiple ascending dose trial in subjects with prodromal or mild AD (NCT03375697), as measured with the PT3 x PT82 Simoa assay [21] We used the optimized 2DLC-TQMS methodology as orthogonal assay to monitor SRTPSLPTPPTREPK∗ 2p as surrogate peptide for p217+tau in human CSF clinical study samples The immunoprecipitation procedures were done as described in the Materials and Methods section In short, optimized levels of JNJ’3657 and protein G beads - to have an excess of both to connect all p217+tau to the beads but as little as possible to minimize matrix interference - were added to the clinical study CSF samples and incubated overnight at °C After incubation, the unbound fraction was removed, beads were washed and trypsinization buffer was added Method validation could not be executed in a traditional way because of the limited availability of human CSF and as (Cterminally truncated) ptau is naturally present in CSF Nonetheless, multiple test batches were analysed to optimize sample preparation and LC-TQMS methods and to confirm method reliability During analysis of clinical samples, all quality control parameters for batch acceptance, agreed prior to sample preparation and analysis, were met (e.g acceptance criteria for calibration points and quality control samples, etc., as described in supplemental information 1) Carry over (appr 5%) was observed during 2DLC-TQMS analysis: a duplicated gradient test as described by Vu et al [44] revealed that carry over is not caused by the injection system but originates from retention of diphospho-peptides in the LC-TQMS S Bijttebier, C Theunis, F Jahouh et al Journal of Chromatography A 1651 (2021) 462299 Fig 1DLC-TQMS analysis of TPSLP(pT)PPTR in tryptic digest of IP-ed human CSF before (left) and after (right) TiO2 /ZrO2 clean-up Results were obtained with the 1DLCMSMS setup described in Materials and Methods with MP gradient as follows (min/A%): 0.0/100, 5.0/70, 5.1/2, 6.6/2, 6.7/100, 10/100 Fig IP-2DLC-TQMS analysis of SRTPSLPTPPTREPK∗ 2p in human CSF clinical samples Graph A:% Reduction of total p217+tau (LC-TQMS) after normalization for total tau levels (Simoa) compared to baseline samples Graph B: correlation of total p217+tau measured with IP LC-TQMS and p217+tau simoa measurement (PT3-PT82: +boiling) Only values within calibration range are included Two samples were below quantification limit and one sample was above calibration range system and column As p217+tau tryptic peptides are very low in abundance (pM quantities) in human CSF, endogenous concentrations are close to the method LLOQ Limiting the dynamic range of calibration standards reduced carry over Fig shows the quantitative IP-2DLC-TQMS measurement of the SRTPSLPTPPTREPK∗ 2p tryptic peptide in the human CSF clinical samples, depicted as the normalized p217+tau level to the total tau level (left), and the correlation between the IP-LC-TQMS data and the PT3 x PT82 Simoa data (right) An average maximum reduction of 50% of total p217+tau (LC-TQMS) after normalization for total tau (HT-7 x PT82 simoa assay [21]) compared to baseline samples was observed, thereby confirming the reduction of total p217+tau after dosing as observed with the PT3 x PT82 Simoa Absolute levels measured with Simoa are slightly lower than with IP-2DLC-TQMS, which is reflected in the slope (