Advertisement

Domain-specific Quantification of Prion Protein in Cerebrospinal Fluid by Targeted Mass Spectrometry*

      Therapies currently in preclinical development for prion disease seek to lower prion protein (PrP) expression in the brain. Trials of such therapies are likely to rely on quantification of PrP in cerebrospinal fluid (CSF) as a pharmacodynamic biomarker and possibly as a trial endpoint. Studies using PrP ELISA kits have shown that CSF PrP is lowered in the symptomatic phase of disease, a potential confounder for reading out the effect of PrP-lowering drugs in symptomatic patients. Because misfolding or proteolytic cleavage could potentially render PrP invisible to ELISA even if its concentration were constant or increasing in disease, we sought to establish an orthogonal method for CSF PrP quantification. We developed a multi-species targeted mass spectrometry method based on multiple reaction monitoring (MRM) of nine PrP tryptic peptides quantified relative to an isotopically labeled recombinant protein standard for human samples, or isotopically labeled synthetic peptides for nonhuman species. Analytical validation experiments showed process replicate coefficients of variation below 15%, good dilution linearity and recovery, and suitable performance for both CSF and brain homogenate and across humans as well as preclinical species of interest. In n = 55 CSF samples from individuals referred to prion surveillance centers with rapidly progressive dementia, all six human PrP peptides, spanning the N- and C-terminal domains of PrP, were uniformly reduced in prion disease cases compared with individuals with nonprion diagnoses. Thus, lowered CSF PrP concentration in prion disease is a genuine result of the disease process and not an artifact of ELISA-based measurement. As a result, dose-finding studies for PrP lowering drugs may need to be conducted in presymptomatic at-risk individuals rather than in symptomatic patients. We provide a targeted mass spectrometry-based method suitable for preclinical quantification of CSF PrP as a tool for drug development.

      Graphical Abstract

      Prion disease is a fatal and incurable neurodegenerative disease caused by misfolding of the prion protein 1 (PrP), and may be sporadic, genetic, or acquired (
      • Prusiner S.B.
      Prions.
      ). Therapies currently in preclinical development for prion disease seek to lower PrP levels in the brain, a genetically well-validated strategy (
      • Raymond G.J.
      • Zhao H.T.
      • Race B.
      • Raymond L.D.
      • Williams K.
      • Swayze E.E.
      • Graffam S.
      • Le J.
      • Caron T.
      • Stathopoulos J.
      • O'Keefe R.
      • Lubke L.L.
      • Reidenbach A.G.
      • Kraus A.
      • Schreiber S.L.
      • Mazur C.
      • Cabin D.E.
      • Carroll J.B.
      • Minikel E.V.
      • Kordasiewicz H.
      • Caughey B.
      • Vallabh S.M.
      Antisense oligonucleotides extend survival of prion-infected mice.
      ). Clinical trials of PrP-lowering agents will rely on quantification of PrP in cerebrospinal fluid (CSF) as, at a minimum, a pharmacodynamic biomarker (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). This marker may, however, have even greater importance. Predictive testing of presymptomatic individuals harboring highly penetrant genetic mutations (
      • Minikel E.V.
      • Vallabh S.M.
      • Lek M.
      • Estrada K.
      • Samocha K.E.
      • Sathirapongsasuti J.F.
      • McLean C.Y.
      • Tung J.Y.
      • Yu L.P.C.
      • Gambetti P.
      • Blevins J.
      • Zhang S.
      • Cohen Y.
      • Chen W.
      • Yamada M.
      • Hamaguchi T.
      • Sanjo N.
      • Mizusawa H.
      • Nakamura Y.
      • Kitamoto T.
      • Collins S.J.
      • Boyd A.
      • Will R.G.
      • Knight R.
      • Ponto C.
      • Zerr I.
      • Kraus T.F.J.
      • Eigenbrod S.
      • Giese A.
      • Calero M.
      • de Pedro-Cuesta J.
      • Haïk S.
      • Laplanche J.-L.
      • Bouaziz-Amar E.
      • Brandel J.-P.
      • Capellari S.
      • Parchi P.
      • Poleggi A.
      • Ladogana A.
      • O'Donnell-Luria A.H.
      • Karczewski K.J.
      • Marshall J.L.
      • Boehnke M.
      • Laakso M.
      • Mohlke K.L.
      • Kähler A.
      • Chambert K.
      • McCarroll S.
      • Sullivan P.F.
      • Hultman C.M.
      • Purcell S.M.
      • Sklar P.
      • van der Lee S.J.
      • Rozemuller A.
      • Jansen C.
      • Hofman A.
      • Kraaij R.
      • van Rooij J.G.J.
      • Ikram M.A.
      • Uitterlinden A.G.
      • van Duijn C.M.
      • Daly M.J.
      • MacArthur D.G.
      Exome Aggregation Consortium (ExAC). Quantifying prion disease penetrance using large population control cohorts.
      ) that cause prion disease provides an opportunity for early therapeutic intervention to preserve healthy life, but randomization to a clinical endpoint in this population appears infeasible (
      • Minikel E.V.
      • Vallabh S.M.
      • Orseth M.C.
      • Brandel J.-P.
      • Haïk S.
      • Laplanche J.-L.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Safar J.
      • Kenny J.
      • Fong J.C.
      • Takada L.T.
      • Ponto C.
      • Hermann P.
      • Knipper T.
      • Stehmann C.
      • Kitamoto T.
      • Ae R.
      • Hamaguchi T.
      • Sanjo N.
      • Tsukamoto T.
      • Mizusawa H.
      • Collins S.J.
      • Chiesa R.
      • Roiter I.
      • de Pedro-Cuesta J.
      • Calero M.
      • Geschwind M.D.
      • Yamada M.
      • Nakamura Y.
      • Mead S.
      Age at onset in genetic prion disease and the design of preventive clinical trials.
      ). The U.S. Food and Drug Administration has indicated its willingness to consider lowered CSF PrP in this population as a potential surrogate endpoint for Accelerated Approval (
      • Vallabh S.M.
      Antisense oligonucleotides for the prevention of genetic prion disease.
      ,
      Food, U. S., Drug Administration. Critical Path Innovation Meeting: Genetic Prion Disease..
      ). Precise quantification of PrP in CSF will be essential to the development of prion disease therapeutics.
      PrP is an extracellular GPI-anchored protein that can be shed from the plasma membrane by ADAM10 and other peptidases (
      • Altmeppen H.C.
      • Puig B.
      • Dohler F.
      • Thurm D.K.
      • Falker C.
      • Krasemann S.
      • Glatzel M.
      Proteolytic processing of the prion protein in health and disease.
      ,
      • Linsenmeier L.
      • Mohammadi B.
      • Wetzel S.
      • Puig B.
      • Jackson W.S.
      • Hartmann A.
      • Uchiyama K.
      • Sakaguchi S.
      • Endres K.
      • Tatzelt J.
      • Saftig P.
      • Glatzel M.
      • Altmeppen H.C.
      Structural and mechanistic aspects influencing the ADAM10-mediated shedding of the prion protein.
      ). CSF PrP is predominantly soluble and full-length (
      • Tagliavini F.
      • Prelli F.
      • Porro M.
      • Salmona M.
      • Bugiani O.
      • Frangione B.
      A soluble form of prion protein in human cerebrospinal fluid: implications for prion-related encephalopathies.
      ), suggesting that it originates chiefly from this proteolytic shedding near the C terminus, although lower molecular weight fragments of PrP have also been identified in CSF (
      • Schmitz M.
      • Schlomm M.
      • Hasan B.
      • Beekes M.
      • Mitrova E.
      • Korth C.
      • Breil A.
      • Carimalo J.
      • Gawinecka J.
      • Varges D.
      • Zerr I.
      Codon 129 polymorphism and the E200K mutation do not affect the cellular prion protein isoform composition in the cerebrospinal fluid from patients with Creutzfeldt-Jakob disease.
      ), which may originate from other endoproteolytic events (
      • Altmeppen H.C.
      • Puig B.
      • Dohler F.
      • Thurm D.K.
      • Falker C.
      • Krasemann S.
      • Glatzel M.
      Proteolytic processing of the prion protein in health and disease.
      ,
      • Chen S.G.
      • Teplow D.B.
      • Parchi P.
      • Teller J.K.
      • Gambetti P.
      • Autilio-Gambetti L.
      Truncated forms of the human prion protein in normal brain and in prion diseases.
      ), and anchored PrP is also released from cells on exosomes (
      • Fevrier B.
      • Vilette D.
      • Archer F.
      • Loew D.
      • Faigle W.
      • Vidal M.
      • Laude H.
      • Raposo G.
      Cells release prions in association with exosomes.
      ). PrP is sufficiently abundant in CSF, at concentrations of tens or hundreds of nanograms per milliliter, to be readily quantified with enzyme-linked immunosorbent assay (ELISA). Studies using ELISA have reproducibly found that CSF PrP is decreased in the symptomatic phase of prion disease (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ,
      • Meyne F.
      • Gloeckner S.F.
      • Ciesielczyk B.
      • Heinemann U.
      • Krasnianski A.
      • Meissner B.
      • Zerr I.
      Total prion protein levels in the cerebrospinal fluid are reduced in patients with various neurological disorders.
      ,
      • Dorey A.
      • Tholance Y.
      • Vighetto A.
      • Perret-Liaudet A.
      • Lachman I.
      • Krolak-Salmon P.
      • Wagner U.
      • Struyfs H.
      • De Deyn P.P.
      • El-Moualij B.
      • Zorzi W.
      • Meyronet D.
      • Streichenberger N.
      • Engelborghs S.
      • Kovacs G.G.
      • Quadrio I.
      Association of cerebrospinal fluid prion protein levels and the distinction between Alzheimer disease and Creutzfeldt-Jakob disease.
      ,
      • Abu Rumeileh S.
      • Lattanzio F.
      • Stanzani Maserati M.
      • Rizzi R.
      • Capellari S.
      • Parchi P.
      Diagnostic Accuracy of a Combined Analysis of Cerebrospinal Fluid t-PrP, t-tau, p-tau, and Aβ42 in the Differential Diagnosis of Creutzfeldt-Jakob Disease from Alzheimer's Disease with Emphasis on Atypical Disease Variants.
      ,
      • Villar-Piqué A.
      • Schmitz M.
      • Lachmann I.
      • Karch A.
      • Calero O.
      • Stehmann C.
      • Sarros S.
      • Ladogana A.
      • Poleggi A.
      • Santana I.
      • Ferrer I.
      • Mitrova E.
      • Žáková
      • Pocchiari D.M.
      • Baldeiras I.
      • Calero M.
      • Collins S.J.
      • Geschwind M.D.
      • Sánchez-Valle R.
      • Zerr I.
      • Llorens F.
      Cerebrospinal fluid total prion protein in the spectrum of prion diseases.
      ). Therefore, even though CSF PrP is brain-derived and exhibits good within-subject test-retest reliability in individuals without prion disease (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ), it might be difficult to use this biomarker to read out the effect of a PrP-lowering drug in symptomatic individuals, because it is unclear whether to expect that such a drug should cause a further decrease in CSF PrP as a direct pharmacodynamic effect, or an increase in CSF PrP because of alleviation of the disease process. This confounder could potentially limit the use of ELISA-based CSF PrP quantification as a pharmacodynamic biomarker to presymptomatic individuals only.
      Prion disease is caused by a gain of function (
      • Prusiner S.B.
      Prions.
      ), and animal studies have shown that total PrP in the brain increases over the course of prion disease as misfolded PrP accumulates (
      • Schulz-Schaeffer W.J.
      • Tschöke S.
      • Kranefuss N.
      • Dröse W.
      • Hause-Reitner D.
      • Giese A.
      • Groschup M.H.
      • Kretzschmar H.A.
      The paraffin-embedded tissue blot detects PrP(Sc) early in the incubation time in prion diseases.
      ,
      • Safar J.G.
      • DeArmond S.J.
      • Kociuba K.
      • Deering C.
      • Didorenko S.
      • Bouzamondo-Bernstein E.
      • Prusiner S.B.
      • Tremblay P.
      Prion clearance in bigenic mice.
      ,
      • Moreno J.A.
      • Radford H.
      • Peretti D.
      • Steinert J.R.
      • Verity N.
      • Martin M.G.
      • Halliday M.
      • Morgan J.
      • Dinsdale D.
      • Ortori C.A.
      • Barrett D.A.
      • Tsaytler P.
      • Bertolotti A.
      • Willis A.E.
      • Bushell M.
      • Mallucci G.R.
      Sustained translational repression by eIF2α-P mediates prion neurodegeneration.
      ). The paradoxical decrease in PrP in CSF during prion disease might be because of its incorporation into plaques (
      • Parchi P.
      • Giese A.
      • Capellari S.
      • Brown P.
      • Schulz-Schaeffer W.
      • Windl O.
      • Zerr I.
      • Budka H.
      • Kopp N.
      • Piccardo P.
      • Poser S.
      • Rojiani A.
      • Streichemberger N.
      • Julien J.
      • Vital C.
      • Ghetti B.
      • Gambetti P.
      • Kretzschmar H.
      Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects.
      ), diversion into intracellular locations (
      • Caughey B.
      • Raymond G.J.
      • Ernst D.
      • Race R.E.
      N-terminal truncation of the scrapie-associated form of PrP by lysosomal protease(s): implications regarding the site of conversion of PrP to the protease-resistant state.
      ,
      • Goold R.
      • McKinnon C.
      • Rabbanian S.
      • Collinge J.
      • Schiavo G.
      • Tabrizi S.J.
      Alternative fates of newly formed PrPSc upon prion conversion on the plasma membrane.
      ), or down-regulation as a function of the disease process (
      • Mays C.E.
      • Kim C.
      • Haldiman T.
      • van der Merwe J.
      • Lau A.
      • Yang J.
      • Grams J.
      • Di Bari M.A.
      • Nonno R.
      • Telling G.C.
      • Kong Q.
      • Langeveld J.
      • McKenzie D.
      • Westaway D.
      • Safar J.G.
      Prion disease tempo determined by host-dependent substrate reduction.
      ). However, although the ELISA assay has been described as measuring “total PrP,” the assay's reactivity for different conformations or proteolytic fragments of PrP has not been evaluated, leaving doubt as to what the disease-dependent reduction in CSF PrP means. Occlusion of epitopes because of misfolding (
      • Safar J.
      • Wille H.
      • Itri V.
      • Groth D.
      • Serban H.
      • Torchia M.
      • Cohen F.E.
      • Prusiner S.B.
      Eight prion strains have PrP(Sc) molecules with different conformations.
      ) or up-regulation of proteolytic cleavage in disease (
      • Altmeppen H.C.
      • Puig B.
      • Dohler F.
      • Thurm D.K.
      • Falker C.
      • Krasemann S.
      • Glatzel M.
      Proteolytic processing of the prion protein in health and disease.
      ,
      • Mays C.E.
      • Kim C.
      • Haldiman T.
      • van der Merwe J.
      • Lau A.
      • Yang J.
      • Grams J.
      • Di Bari M.A.
      • Nonno R.
      • Telling G.C.
      • Kong Q.
      • Langeveld J.
      • McKenzie D.
      • Westaway D.
      • Safar J.G.
      Prion disease tempo determined by host-dependent substrate reduction.
      ,
      • Watts J.C.
      • Stöhr J.
      • Bhardwaj S.
      • Wille H.
      • Oehler A.
      • Dearmond S.J.
      • Giles K.
      • Prusiner S.B.
      Protease-resistant prions selectively decrease Shadoo protein.
      ) could render PrP invisible to ELISA even if its concentration were constant or increasing. We therefore sought to establish an orthogonal method for CSF PrP quantification. In addition, because the commercially available ELISA kit is specific to human PrP (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ), we required a multi-species assay applicable to the preclinical phases of drug development.
      Here, we describe quantification of CSF PrP relative to an isotopically labeled recombinant protein standard using multiple reaction monitoring mass spectrometry (MRM-MS) (
      • Carr S.A.
      • Abbatiello S.E.
      • Ackermann B.L.
      • Borchers C.
      • Domon B.
      • Deutsch E.W.
      • Grant R.P.
      • Hoofnagle A.N.
      • Hüttenhain R.
      • Koomen J.M.
      • Liebler D.C.
      • Liu T.
      • MacLean B.
      • Mani D.R.
      • Mansfield E.
      • Neubert H.
      • Paulovich A.G.
      • Reiter L.
      • Vitek O.
      • Aebersold R.
      • Anderson L.
      • Bethem R.
      • Blonder J.
      • Boja E.
      • Botelho J.
      • Boyne M.
      • Bradshaw R.A.
      • Burlingame A.L.
      • Chan D.
      • Keshishian H.
      • Kuhn E.
      • Kinsinger C.
      • Lee J.S.H.
      • Lee S.-W.
      • Moritz R.
      • Oses-Prieto J.
      • Rifai N.
      • Ritchie J.
      • Rodriguez H.
      • Srinivas P.R.
      • Townsend R.R.
      • Van Eyk J.
      • Whiteley G.
      • Wiita A.
      • Weintraub S.
      Targeted peptide measurements in biology and medicine: best practices for mass spectrometry-based assay development using a fit-for-purpose approach.
      ). We analyze n = 55 clinical samples from prion and nonprion disease patients, and find that six out of six PrP tryptic peptides, spanning N- and C-terminal domains of the protein, are uniformly decreased in prion disease. Thus, PrP concentration is genuinely lowered in prion disease CSF, meaning that dose-finding studies for PrP-lowering drugs may need to be conducted in presymptomatic individuals. To provide similar capability to measure drug-dependent changes in PrP concentration in tissues from preclinical species of interest, we also developed assays for mouse, rat, and cynomolgus macaque based on quantification relative to isotopically labeled synthetic peptide standards. Our findings supply an alternative method for validating the findings of ELISA-based studies of CSF PrP, and provide a potential assay for use as a pharmacodynamic biomarker in preclinical drug development and in human trials.

      EXPERIMENTAL PROCEDURES

       Experimental Design and Statistical Rationale

      We designed this study to compare the levels of PrP tryptic peptides in CSF samples from individuals referred for diagnostic testing for prion disease, later determined to have prion disease (cases, n = 34) or not (controls, n = 21), details provided in next section. We selected this set of CSF samples because we had previously analyzed them by ELISA and found that the difference in “total PrP” level between cases and controls was highly significant (p = 0.0001), suggesting our analysis should be well-powered to replicate or refute the ELISA findings. Operators were blinded to case/control status and samples were randomly assigned to different analysis days using an R script. Each sample was analyzed in duplicate and the mean value for each peptide from two replicates was used. Data were analyzed primarily by visual inspection and the use of confidence intervals. Because our study was limited to examining previously reported hypotheses, and not exploring new ones, p values are nominal where reported.

       Cerebrospinal Fluid and Brain Samples

      This study was approved by the Broad Institute's Office of Research Subjects Protection (ORSP-3587). Written consent for research use of samples was obtained from patients or next of kin as appropriate.
      All CSF samples in this study have been previously reported (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). CSF samples for assay development were large volume normal pressure hydrocephalus samples provided by MIND Tissue Bank at Massachusetts General Hospital. Clinical CSF samples (n = 55) were premortem lumbar punctures from rapidly progressive dementia patients referred to prion surveillance centers in Italy (Bologna) or Germany (Göttingen) with suspected prion disease and who were later either determined by autopsy or probable diagnostic criteria (
      • Hermann P.
      • Laux M.
      • Glatzel M.
      • Matschke J.
      • Knipper T.
      • Goebel S.
      • Treig J.
      • Schulz-Schaeffer W.
      • Cramm M.
      • Schmitz M.
      • Zerr I.
      Validation and utilization of amended diagnostic criteria in Creutzfeldt-Jakob disease surveillance.
      ) including real-time quaking-induced conversion (RT-QuIC (
      • Orrú C.D.
      • Groveman B.R.
      • Hughson A.G.
      • Zanusso G.
      • Coulthart M.B.
      • Caughey B.
      Rapid and sensitive RT-QuIC detection of human Creutzfeldt-Jakob disease using cerebrospinal fluid.
      )) as prion disease, or confirmed as nonprion cases on the basis of autopsy, patient recovery, or definitive other diagnostic test. Individuals with nonprion diagnoses (n = 21) included autoimmune disease (n = 8), nonprion neurodegenerative disease (n = 6), psychiatric illness (n = 3), stroke (n = 1), brain cancer (n = 1), and other (n = 2). Sporadic prion disease cases (n = 23) included probable cases (n = 10) and autopsy-confirmed definite cases (n = 13, of subtypes: 6 MM1, 3 VV2 and 4 other/unknown). Genetic prion disease cases (n = 11) included D178N (n = 2), E200K (n = 7), and V210I (n = 2). After receipt in our lab, samples were thawed, spiked with 0.03% CHAPS detergent (final concentration) and stored in 30 μl aliquots for mass spectrometry analysis. Sample handling histories before receipt in our lab are not well-documented and are likely variable because of the large number of referring physicians sending these samples to surveillance centers, but systematic differences between diagnostic groups are unlikely because all samples were similarly referred for diagnostic testing on suspicion of prion disease.
      Samples were de-identified and broken into five batches (to be run on different days) randomly using an R script. Assay operators were blinded to diagnosis. The methods and values obtained for PrP ELISA, hemoglobin, and total protein measurements on these CSF samples were previously reported (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ).
      Rat and cynomolgus monkey CSF were purchased from BioIVT. Human brain tissue was from a nonprion disease control individual provided by the National Prion Disease Pathology Surveillance Center (Cleveland, OH). Mouse brain tissue from Edinburgh PrP knockout mice (
      • Manson J.C.
      • Clarke A.R.
      • Hooper M.L.
      • Aitchison L.
      • McConnell I.
      • Hope J.
      129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal.
      ) backcrossed to a C57BL/10 background (
      • Striebel J.F.
      • Race B.
      • Pathmajeyan M.
      • Rangel A.
      • Chesebro B.
      Lack of influence of prion protein gene expression on kainate-induced seizures in mice: studies using congenic, coisogenic and transgenic strains.
      ), and matching tissue from wild-type C57BL/10 mice, were provided by Gregory J. Raymond (NIAID Rocky Mountain Labs, Hamilton, MT).

       Recombinant Prion Protein Preparation and Isotopic Labeling

      Untagged recombinant HuPrP23–230 (MW = 22,878) and MoPrP23–231 (MW = 23,151), corresponding to full-length post-translationally modified human and mouse PrP without the signal peptide or GPI signal but retaining an N-terminal methionine, were purified by denaturation and Ni-NTA affinity from E. coli inclusion bodies as previously described (
      • Wilham J.M.
      • Orrú C.D.
      • Bessen R.A.
      • Atarashi R.
      • Sano K.
      • Race B.
      • Meade-White K.D.
      • Taubner L.M.
      • Timmes A.
      • Caughey B.
      Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays.
      ,
      • Orrù C.D.
      • Groveman B.R.
      • Hughson A.G.
      • Manca M.
      • Raymond L.D.
      • Raymond G.J.
      • Campbell K.J.
      • Anson K.J.
      • Kraus A.
      • Caughey B.
      RT-QuIC assays for prion disease detection and diagnostics.
      ), using a vector generously provided by Byron Caughey (NIAID Rocky Mountain Labs, Hamilton, MT). 15N incorporation was achieved by growing the E. coli in 15N cell growth medium (Cambridge Isotope Laboratories CGM-1000-N) induced with 15N auto-induction medium (Millipore 71759–3). Purified protein was formulated in 0.03% CHAPS and aliquoted in single-use tubes to avoid repetitive freeze/thaw cycles. Protein concentration was determined by amino acid analysis (AAA, New England Peptide). Percent 15N isotopic incorporation was estimated using LC-MS/MS. 15N labeled human recombinant prion protein (10 μg) was digested and desalted following the procedure as described in PrP MRM assay and analyzed as described in Pilot LC-MS/MS analysis. Precursor masses for 15N were extracted from the chromatograms using XCalibur software Qualbrowser software (Thermo) 3.0.63 with a 6 m/z window of centered on the precursors and charge states listed in supplemental Table S1. Isotopic envelopes between protein expressed in 15N-containing media and standard media were compared visually. Summation of all observed m/z peak areas less than the 12C monoisotopic mass peak were compared with summation of all expected isotope peak to estimate the overall completeness of 15N incorporation (supplemental Fig. S1).

       LC-MS/MS Analyses of CSF and Recombinant PrP

      Samples of dried digested recombinant proteins or human cerebrospinal fluid (processed as described in PrP MRM assay) were reconstituted in 3% acetonitrile/5% acetic acid to a final concentration of ∼1 μg total protein per 1 μl and analyzed in a single injection using a standard 2 h reversed-phase gradient. LC-MS/MS was performed using a QExactive mass spectrometer (Thermo) equipped with a Proxeon Easy-nLC 1200 and a custom built nanospray source (James A. Hill Instrument Services). Samples were injected (1 to 2 μg) onto a 75 μm ID PicoFrit column (New Objective) packed to 20 cm with Reprosil-Pur C18 AQ 1.9 μm media (Dr. Maisch) and heated to 50 °C. MS source conditions were set as follows: spray voltage 2000, capillary temperature 250, S-lens RF level 50. A single Orbitrap MS scan from 300 to 1800 m/z at a resolution of 70,000 with AGC set at 3e6 was followed by up to 12 MS/MS scans at a resolution of 17,500 with AGC set at 5e4. MS/MS spectra were collected with normalized collision energy of 25 and isolation width of 2.5 amu. Dynamic exclusion was set to 20 s and peptide match was set to preferred. Mobile phases consisted of 3% acetonitrile/0.1% formic acid as solvent A, 90% acetonitrile/0.1% formic acid as solvent B. Flow rate was set to 200 nL/min throughout the gradient, 2–6% B in 1 min, 6–30% B in 84 min, 30–60% B in 9 min, 60–90% B in 1 min with a hold at 90% B for 5 min. MS data were analyzed using Spectrum Mill MS Proteomics Workbench software Rev B.06.01.202 (Agilent Technologies). Similar MS/MS spectra acquired on the same precursor m/z within ± 60 s were merged. MS/MS spectra were excluded from searching if they failed the quality filter by not having a sequence tag length > 0 (i.e. minimum of two masses separated by the in-chain mass of an amino acid) or did not have a precursor MH+ in the range of 600–6000. All extracted spectra were searched against a UniProt database containing human and mouse reference proteome sequences (UniProt.human.mouse.20141017.RNFISnr.150contams, n = 100,236 entries) downloaded from the UniProt web site on October 17, 2014 with redundant sequences removed. A set of common laboratory contaminant proteins (150 sequences) were appended to this database and verified to contain the sequences for human and mouse major prion protein. The database was searched with the following parameters. ESI-QEXACTIVE-HCD-v2 scoring, parent and fragment mass tolerance of 20 ppm, 40% minimum matched peak intensity and 'trypsin’ enzyme specificity up to 2 missed cleavages. Fixed modification was carbamidomethylation at cysteine and variable modifications were oxidized methionine, deamidation of asparagine and pyro-glutamic acid. Database matches were autovalidated at the peptide and protein level in a two-step process with optimized scores & R1-R2 score thresholds with maximum false discovery rate (FDR) of 1.2% across each LC run by target-decoy-based searches using reversed sequences. The list of identified proteins was further filtered to contain proteins and protein isoforms with at least 2 unique peptides and an aggregate protein score greater than 20. Protein-peptide comparison report comprised of all validated peptides was exported which included a ranked summary by intensity of all peptides unique to prion protein. Hits were ranked by total summed MS1 intensity of all identified peptides for each protein (totalIntensitySpecies column). Summary data are available as a supplementary Excel file or in the online GitHub repository for this study (see Data Availability).

       Selection of PrP Peptides for MRM Assay Development

      Nine peptides covering 4 species were selected from computational and empirical data (supplemental Table S2 and supplemental Figs. S2–S4). PrP amino acid sequences (supplemental Fig. S4) were obtained from UniProt (
      The UniProt Consortium. UniProt: the universal protein knowledgebase.
      ) and aligned using Clustal O 1.2.4 (
      • Sievers F.
      • Higgins D.G.
      Clustal Omega for making accurate alignments of many protein sequences.
      ). Peptides were prioritized based our criteria previously described (
      • Kuhn E.
      • Carr S.A.
      Multiplexed immunoaffinity enrichment of peptides with anti-peptide antibodies and quantification by stable isotope dilution multiple reaction monitoring mass spectrometry.
      ,
      • Hoofnagle A.N.
      • Whiteaker J.R.
      • Carr S.A.
      • Kuhn E.
      • Liu T.
      • Massoni S.A.
      • Thomas S.N.
      • Townsend R.R.
      • Zimmerman L.J.
      • Boja E.
      • Chen J.
      • Crimmins D.L.
      • Davies S.R.
      • Gao Y.
      • Hiltke T.R.
      • Ketchum K.A.
      • Kinsinger C.R.
      • Mesri M.
      • Meyer M.R.
      • Qian W.-J.
      • Schoenherr R.M.
      • Scott M.G.
      • Shi T.
      • Whiteley G.R.
      • Wrobel J.A.
      • Wu C.
      • Ackermann B.L.
      • Aebersold R.
      • Barnidge D.R.
      • Bunk D.M.
      • Clarke N.
      • Fishman J.B.
      • Grant R.P.
      • Kusebauch U.
      • Kushnir M.M.
      • Lowenthal M.S.
      • Moritz R.L.
      • Neubert H.
      • Patterson S.D.
      • Rockwood A.L.
      • Rogers J.
      • Singh R.J.
      • Van Eyk J.E.
      • Wong S.H.
      • Zhang S.
      • Chan D.W.
      • Chen X.
      • Ellis M.J.
      • Liebler D.C.
      • Rodland K.D.
      • Rodriguez H.
      • Smith R.D.
      • Zhang Z.
      • Zhang H.
      • Paulovich A.G.
      Recommendations for the generation, quantification, storage, and handling of peptides used for mass spectrometry-based assays.
      ) as outlined and described in detail in supplemental Fig. S2. Peptides were checked for uniqueness to human PrP using the Peptide String Match utility in Spectrum Mill (http://proteomics.broadinstitute.org). Peptides were selected based on PrP biology and desired assay applications described in Results (Fig. 1). One peptide, PIIHFGSDYEDR, was included after being detected in CSF despite an N-terminal proline.
      Figure thumbnail gr1
      Fig. 1.Design of the PrP MRM assay. A, Selection of PrP tryptic peptides for MRM. The full sequence of human PrP (residues 23–230) after post-translational modifications (removal of signal peptide residues 1–22 and GPI signal residues 231–253) is shown, GPI-anchored to the outer leaflet of the plasma membrane, with the position of selected peptides and their rodent or monkey orthologs shown relative to the positions of N-linked glycans, a disulfide bond, and endogenous proteolytic events (
      • Altmeppen H.C.
      • Puig B.
      • Dohler F.
      • Thurm D.K.
      • Falker C.
      • Krasemann S.
      • Glatzel M.
      Proteolytic processing of the prion protein in health and disease.
      ). B, PrP MRM workflow as described in Methods.
      All nine peptides were synthesized (New England Peptide) using stable isotope labeled [15N413C6]Arg or [15N213C6]Lys at the C terminus and purified peptide specifications previously outlined (>95% chemical purity, >99% isotopic purity, quantified by AAA) to qualify as standards for Tier 1 or 2 assays (
      • Carr S.A.
      • Abbatiello S.E.
      • Ackermann B.L.
      • Borchers C.
      • Domon B.
      • Deutsch E.W.
      • Grant R.P.
      • Hoofnagle A.N.
      • Hüttenhain R.
      • Koomen J.M.
      • Liebler D.C.
      • Liu T.
      • MacLean B.
      • Mani D.R.
      • Mansfield E.
      • Neubert H.
      • Paulovich A.G.
      • Reiter L.
      • Vitek O.
      • Aebersold R.
      • Anderson L.
      • Bethem R.
      • Blonder J.
      • Boja E.
      • Botelho J.
      • Boyne M.
      • Bradshaw R.A.
      • Burlingame A.L.
      • Chan D.
      • Keshishian H.
      • Kuhn E.
      • Kinsinger C.
      • Lee J.S.H.
      • Lee S.-W.
      • Moritz R.
      • Oses-Prieto J.
      • Rifai N.
      • Ritchie J.
      • Rodriguez H.
      • Srinivas P.R.
      • Townsend R.R.
      • Van Eyk J.
      • Whiteley G.
      • Wiita A.
      • Weintraub S.
      Targeted peptide measurements in biology and medicine: best practices for mass spectrometry-based assay development using a fit-for-purpose approach.
      ). Mixtures of all 9 heavy peptides were formulated in 30% acetonitrile/0.1% formic acid and aliquotted into single-use tubes to avoid freeze thaw throughout the study. Before each set of samples, 50 fmol was injected and analyzed using the same LC-MRM-MS method used for samples to confirm LC column and MS performance.

       System Suitability Standards

      An equimolar predigested “Bovine 6 Protein Mix” (PTD/00001/63) was purchased from Bruker-Michrom, Inc. Pierce™ Peptide Retention Time Calibration Mixture (88320) was purchased from Thermo. Both dried peptide standard mixtures were resuspended in 3% acetonitrile/5% acetic and 50 fmol were injected and analyzed by LC-MRM-MS at the beginning, the middle and at the end of each set of samples and visually inspected in a Skyline document to confirm LC column and MS performance. Transitions for Bovine protein mix and PRTC are provided in supplemental Tables S1B and S1C respectively.

       PrP MRM Assay

      In devising a CSF sample preparation protocol, we drew upon our experience with MRM analysis of plasma (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ) and published mass spectrometry protocols for prion studies (
      • Moore R.A.
      • Head M.W.
      • Ironside J.W.
      • Ritchie D.L.
      • Zanusso G.
      • Choi Y.P.
      • Pyo Choi Y.
      • Priola S.A.
      The distribution of prion protein allotypes differs between sporadic and iatrogenic Creutzfeldt-Jakob disease patients.
      ,
      • Moore R.A.
      • Ward A.
      • Race B.
      • Priola S.A.
      Processing of high-titer prions for mass spectrometry inactivates prion infectivity.
      ). Except where otherwise specified, all samples contained 0.03% CHAPS, a zwitterionic detergent, because this reduces preanalytical loss of PrP because of plastic adsorption (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). The elution conditions of the desalting step (see below) were designed to reduce the amount of CHAPS in the sample before LC-MS. Unlike many other detergents, CHAPS impact on chromatography and peptide ionization is minimized because it elutes off C18 later in the reversed-phase gradient (>60% acetonitrile).

       15N Protein Standard Addition For Human Samples

      For human CSF clinical samples, endogenous PrP was quantified relative to uniformly labeled 15N-labeled recombinant HuPrP23–230 (starting concentration 2.42 mg/ml determined by AAA) with an estimated isotopic incorporation >97.5% (see Recombinant Protein Preparation) diluted 1:5000 in phosphate-buffered saline containing 1 mg/ml bovine serum albumin and 0.03% CHAPS. This solution was then further diluted 1:20 (1.5 μl added into 30 μl) into CSF samples (final concentration 24.2 ng/ml) before the denaturation and digestion workflow described below. ELISA analysis indicated that this concentration of carrier protein and detergent was enough to keep recombinant PrP in solution and avoid loss to plastic, without appreciably affecting CSF total protein content.

       Sample Digestion

      All concentrations listed below are final concentrations. For each replicate, 30 μl of CSF with 0.03% CHAPS was incubated with 6 m urea (Sigma U0631) and 20 mm TCEP (Pierce 77720) at 37 °C while shaking at 800 rpm in an Eppendorf Thermomixer for 30 min to denature the protein and reduce disulfide bonds. 39 mm iodoacetamide was added for 30 min in the dark at room temperature to alkylate cysteine residues. Urea was diluted to 900 mm by the addition of 0.2 Trizma pH 8.1 (Sigma T8568) to permit trypsin activity. One microgram of trypsin (Promega V5113) was added (final concentration of ∼1.4 ng/μl), providing at least a 1:50 trypsin:substrate ratio for CSF samples with total protein content <1.6 mg/ml, which includes 97% of CSF samples we have analyzed (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). Trypsin digestion proceeded overnight shaking at 800 rpm at 37 °C. Digestion was stopped with 5% formic acid and samples were transferred to 4 °C until desalt. To permit quantification of PrP in preclinical species samples, a mix containing 100 fmol of each 15N/13C-labeled synthetic heavy peptide was then added to the CSF digests (3.33 nm peptide, equivalent to ∼76 ng/ml full-length PrP based on an approximate molecular weight of 22.8 kDa). Digestion was quenched by 5% formic acid and samples were kept at 4 °C until desalt.

       Synthetic Peptide Addition For Multi-species Assay

      To permit quantification of PrP in preclinical species samples, a mix containing 100 fmol of each 15N/13C-labeled synthetic heavy peptide (3.33 nm peptide, equivalent to ∼76 ng/ml full-length PrP based on an approximate molecular weight of 22.8 kDa) was added after digest and before sample desalt. Other aspects of the procedure described above and below were unchanged.

       Sample Desalt

      To desalt the samples, StageTips (
      • Rappsilber J.
      • Mann M.
      • Ishihama Y.
      Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips.
      ) comprised of two punches of C18 material (Empore 66883-U) fitted into a 200 μl pipette tip using a 16 gauge needle with 90° blunt ends (Cadence Science 7938) and a PEEK tubing puncher (Idex 1567) were placed onto microcentrifuge tubes using an adapter (Glycen CEN.24). Tubes were centrifuged at 2500 × g for 3 min after each step, as follows: conditioning with 50 μl 90% acetonitrile/0.1% trifluoroacetic acid; equilibration with 50 μl 0.1% trifluoroacetic acid and priming with 10 μl 0.1% trifluoroacetic acid (no spin after priming); addition of CSF digest in increments of 150 μl; two washes with 50 μl of 0.1% trifluoroacetic acid; and two elutions into a new microcentrifuge tube with 50 μl of 40% acetonitrile/0.1% trifluoroacetic acid. Eluates were frozen at −80 °C.

       LC-MRM-MS Analysis

      Frozen samples were dried under vacuum centrifugation and resuspended in 12 μl 3% acetonitrile/5% acetic acid and placed into a vortexer for 5 min at room temperature. Samples were then centrifuged at 12,000 × g for 5 min and 10 μl of the supernatant was transferred to an HPLC vial (Waters 186000273). HPLC vials were centrifuged briefly (30–60 s) at 1200 × g to remove air bubbles and transferred into the nanoLC autosampler compartment set to 7 °C. Samples were analyzed on a TSQ Quantiva triple quadrupole mass spectrometer installed with a Nanospray Flex source and Easy-nLC 1000 system (Thermo). Ion source was set to positive ion mode with capillary temperature of 300 °C, spray voltage of 2000 and sweep gas set to 0. The Easy-nLC 1000 system was primed with mobile phase A (3% acetonitrile/0.1% formic acid), mobile phase B (90% acetonitrile/0.1% formic acid). Samples were injected (2 μl, 20% of digested sample) onto a 0.075 mm ID PicoFrit (New Objective) column pulled to a 10 μm emitter and custom-packed to 20 cm with 1.9 μm 200Å C18-AQ Reprosil beads (Dr. Maisch). The LC gradient was 0% B to 30% B for 55 min, 30% B to 60% B in 5 min, 60% B to 90% B in 1 min using a flow rate of 200 nL/min. Collision energies were optimized over 4 steps, 2.5 V per step in batches of less than 500 transitions per batch. Three to four transitions were monitored per peptide using the MRM transitions listed in supplemental Table S1 using a 1.5 s cycle time. In addition, even though the corresponding heavy peptides were not synthesized, we monitored for the transitions that corresponded to the oxidized methionine version of the peptide VVEQMCITQYER.

       Data Analysis

      Extracted Ion chromatograms (XIC) of all transition ions were verified and integrated using a Skyline document as described (
      • Broudy D.
      • Killeen T.
      • Choi M.
      • Shulman N.
      • Mani D.R.
      • Abbatiello S.E.
      • Mani D.
      • Ahmad R.
      • Sahu A.K.
      • Schilling B.
      • Tamura K.
      • Boss Y.
      • Sharma V.
      • Gibson B.W.
      • Carr S.A.
      • Vitek O.
      • MacCoss M.J.
      • MacLean B.
      A framework for installable external tools in Skyline.
      ) (Skyline version 4.1.0.11796, https://brendanx-uw1.gs.washington.edu/labkey/project/home/software/Skyline/begin.view) that contained the sequences and spectral libraries derived from LC-MS/MS of the 15N/13C-labeled synthetic heavy peptides. After peak integration, the Skyline report file was exported as a text delimited file where the peak areas in the columns labeled as “Light,” “Heavy,” or “15N” for the single most intense, interference-free, reproducibly measured transition (supplemental Table S1) were used for quantification and subsequent statistical analysis. Interferences were identified by manual inspection of XICs between light and heavy peptides. Light or heavy transitions with different relative intensity ratios compared with standards or had asymmetric peaks were excluded from further analysis. Columns included for export were: Protein Name, Protein Gene, Protein Species, Peptide Sequence, Peptide Modified Sequence, File Name, Acquired Time, Replicate Name, SampleGroup, Peptide Retention Time, Precursor m/z, Fragment Ion, Area, Area Ratio, Total Area, Total Area Ratio.
      To determine the response of each peptide in terms of L:15N ratio as well as evaluate dilution linearity of the assay, we spiked 0, 2.4, 24, or 240 ng/ml of 15N-labeled recombinant human PrP into a single control CSF sample (from an individual with normal pressure hydrocephalus) in triplicate. For each peptide, we then fitted a linear model correlating the (nonzero) spiked concentrations to the observed 15N:light ratios with the intercept fixed at zero, yielding slopes ranging from 39 to 448 ng/ml. Each peptide was then assigned a response factor equal to the highest slope observed for any peptide (448 ng/ml) divided by its own slope. This response factor was multiplied by the L:15N ratio for each peptide in each sample to obtain a normalized estimate of protein concentration.
      In n = 12 individual replicates (out of 110) of the clinical samples, the oxidized methionine (met-ox) version of the VVEQMCITQYER peptide was more abundant than the reduced version, despite the inclusion of a reduction step in sample preparation. The VVEQMCITQYER peptide was omitted from analysis for these replicates.

      RESULTS

       Design of the PrP MRM Assay

      PrP ranked number 8 in intensity out of 322 confidently detected proteins in single-shot, LC-MS/MS analysis of human CSF digested with trypsin (see Methods). This indicated that PrP was a good candidate for direct analysis by LC-MRM-MS in CSF without additional fractionation (
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Kuhn E.
      • Carr S.A.
      Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
      ) or enrichment methods (
      • Kuhn E.
      • Addona T.
      • Keshishian H.
      • Burgess M.
      • Mani D.R.
      • Lee R.T.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Developing multiplexed assays for troponin I and interleukin-33 in plasma by peptide immunoaffinity enrichment and targeted mass spectrometry.
      ). PrP peptides with the highest MS intensities after digestion of recombinant human or mouse PrP as well as human CSF were preferentially ranked according to criteria described in Methods and supplemental Fig. S2. We selected six human peptides, as well as three orthologous peptides specific to mouse, rat, and/or cynomolgus macaque PrP, to support assay application to preclinical drug development (Fig. 1A, supplemental Fig. S4, and supplemental Table S2). Peptides were chosen to span the N- and C-terminal domains of PrP, up- and down-stream of alpha and beta cleavage sites, allowing us to quantify proteolytic fragments of cleaved PrP (Fig. 1A and supplemental Fig. S4).
      We further designed a workflow for the PrP MRM assay (Fig. 1B) incorporating an incubation in the presence of a strong chaotrope to denature both properly folded and misfolded forms of PrP. We then reduced and alkylated the protein mixture to break the disulfide bonds and prevent them from refolding, and thereby make the whole protein accessible to the enzymatic processing of r-trypsin. To permit quantification of endogenous unlabeled (hereafter “light”) PrP, we added uniformly 15N-labeled recombinant human PrP (hereafter “15N”) into clinical samples before analysis (Fig. 1B). Compared with synthetic heavy peptides, the use of an isotopically labeled full-length protein standard controls for analytical variability that can occur during enzymatic digestion and solid phase extraction (SPE) using StageTips (
      • Rappsilber J.
      • Mann M.
      • Ishihama Y.
      Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips.
      ), and can, in principle, provide a more accurate quantitative measure of the amount of each of the peptides derived from the target protein (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ,
      • Abbatiello S.E.
      • Schilling B.
      • Mani D.R.
      • Zimmerman L.J.
      • Hall S.C.
      • MacLean B.
      • Albertolle M.
      • Allen S.
      • Burgess M.
      • Cusack M.P.
      • Gosh M.
      • Hedrick V.
      • Held J.M.
      • Inerowicz H.D.
      • Jackson A.
      • Keshishian H.
      • Kinsinger C.R.
      • Lyssand J.
      • Makowski L.
      • Mesri M.
      • Rodriguez H.
      • Rudnick P.
      • Sadowski P.
      • Sedransk N.
      • Shaddox K.
      • Skates S.J.
      • Kuhn E.
      • Smith D.
      • Whiteaker J.R.
      • Whitwell C.
      • Zhang S.
      • Borchers C.H.
      • Fisher S.J.
      • Gibson B.W.
      • Liebler D.C.
      • MacCoss M.J.
      • Neubert T.A.
      • Paulovich A.G.
      • Regnier F.E.
      • Tempst P.
      • Carr S.A.
      Large-scale interlaboratory study to develop, analytically validate and apply highly multiplexed, quantitative peptide assays to measure cancer-relevant proteins in plasma.
      ).

       Assessment of Human PrP Assay Performance

      To support measurement of endogenous unlabeled PrP in n = 55 human CSF clinical samples (see next section), we performed quality control analysis using the 15N protein added into each sample. Clinical samples were divided into 5 batches run on separate days; each sample was processed in duplicate on the same day, stored at −80 °C, and later analyzed in duplicate, also on the same day. A common control sample was also measured in duplicate on each day.
      The mean absolute MS response, either from 15N recombinant or from endogenous light PrP, varied by over an order of magnitude between the six PrP peptides (Fig. 2A2B). The recovery of the six peptides from endogenous PrP relative to one another was preserved across CSF patient samples (Fig. 2A), but differed from the recovery of the corresponding peptides derived from 15N recombinant PrP (Fig. 2B), resulting in a ∼10-fold difference in mean light:15N ratio between different peptides (Fig. 2C and supplemental Table S3), which was consistent across days (supplemental Fig. S5). Such differences in response are expected and may reflect differences in both digestion efficiencies under the urea/trypsin conditions used here (
      • Scott K.B.
      • Turko I.V.
      • Phinney K.W.
      Quantitative performance of internal standard platforms for absolute protein quantification using multiple reaction monitoring-mass spectrometry.
      ) and electrospray ionization efficiencies (
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Kuhn E.
      • Carr S.A.
      Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Mani D.R.
      • Shi X.
      • Kuhn E.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Quantification of cardiovascular biomarkers in patient plasma by targeted mass spectrometry and stable isotope dilution.
      ,
      • Fusaro V.A.
      • Mani D.R.
      • Mesirov J.P.
      • Carr S.A.
      Prediction of high-responding peptides for targeted protein assays by mass spectrometry.
      ) of these peptides.
      Figure thumbnail gr2
      Fig. 2.Relative recovery of six human PrP peptides in CSF. For each of n = 55 clinical samples, panels show each peptide's A, light peak area, B, 15N peak area, and C, light:15N ratio. Gray lines connect the dots representing distinct peptides from the same individual. Peak area ratios shown here are not normalized by peptide response factors, and so reflect differences in recovery between peptides.
      Another factor driving the difference in response between peptides may be the post-translational modification (Fig. 1A) of endogenous PrP in CSF relative to the bacterially expressed recombinant 15N version used as reference. For example, a significant proportion of brain PrP is N-terminally truncated (
      • Chen S.G.
      • Teplow D.B.
      • Parchi P.
      • Teller J.K.
      • Gambetti P.
      • Autilio-Gambetti L.
      Truncated forms of the human prion protein in normal brain and in prion diseases.
      ), and PrP cleavage products have been observed in CSF as well (
      • Schmitz M.
      • Schlomm M.
      • Hasan B.
      • Beekes M.
      • Mitrova E.
      • Korth C.
      • Breil A.
      • Carimalo J.
      • Gawinecka J.
      • Varges D.
      • Zerr I.
      Codon 129 polymorphism and the E200K mutation do not affect the cellular prion protein isoform composition in the cerebrospinal fluid from patients with Creutzfeldt-Jakob disease.
      ). PrP is known to be variably glycosylated at residue N197, but our assay will only detect the nonglycosylated form of the GENFTETDVK peptide containing this site. This may account for the much lower response of this peptide in CSF versus the 15N standard (Fig. 2). For the C-terminal peptide ESQAYYQR, our assay might not detect proteolytically shed PrP if the cut site for ADAM10, the predominant PrP sheddase (
      • Altmeppen H.C.
      • Prox J.
      • Puig B.
      • Kluth M.A.
      • Bernreuther C.
      • Thurm D.
      • Jorissen E.
      • Petrowitz B.
      • Bartsch U.
      • De Strooper B.
      • Saftig P.
      • Glatzel M.
      Lack of a-disintegrin-and-metalloproteinase ADAM10 leads to intracellular accumulation and loss of shedding of the cellular prion protein in vivo.
      ), in human PrP is homologous to its reported cut site in rodent PrP (
      • Linsenmeier L.
      • Mohammadi B.
      • Wetzel S.
      • Puig B.
      • Jackson W.S.
      • Hartmann A.
      • Uchiyama K.
      • Sakaguchi S.
      • Endres K.
      • Tatzelt J.
      • Saftig P.
      • Glatzel M.
      • Altmeppen H.C.
      Structural and mechanistic aspects influencing the ADAM10-mediated shedding of the prion protein.
      ,
      • Taylor D.R.
      • Parkin E.T.
      • Cocklin S.L.
      • Ault J.R.
      • Ashcroft A.E.
      • Turner A.J.
      • Hooper N.M.
      Role of ADAMs in the ectodomain shedding and conformational conversion of the prion protein.
      ). For the most N-terminal peptide monitored, RPKPGGWNTGGSR, we detected a retained N-terminal methionine three residues upstream of this sequence in bacterially expressed PrP (supplemental Fig. S3), consistent with reported N-terminal methionine excision patterns in E. coli (
      • Frottin F.
      • Martinez A.
      • Peynot P.
      • Mitra S.
      • Holz R.C.
      • Giglione C.
      • Meinnel T.
      The proteomics of N-terminal methionine cleavage.
      ). This could alter this peptide's trypsin digest efficiency relative to brain and CSF PrP. Because we lacked access to purified full-length mammalian PrP to serve a reference standard, we cannot definitively dissect the reasons for the differences in recovery between peptides. Accordingly, we assigned each peptide a response factor based on the slope of the light:15N ratio observed in the 15N dose-response experiment (Methods, supplemental Fig. S6). Multiplying the raw light:15N ratios by these peptide-specific response factors brought each peptide's abundance into line with the highest-responding peptide, and yielded normalized estimates of CSF PrP concentration in CSF that averaged 421 ng/ml across samples and all peptides (supplemental Table S3, supplemental Fig. S6).
      All six peptides exhibited strong technical performance on par with previously published MRM assays (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Kuhn E.
      • Carr S.A.
      Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
      ,
      • Kuhn E.
      • Addona T.
      • Keshishian H.
      • Burgess M.
      • Mani D.R.
      • Lee R.T.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Developing multiplexed assays for troponin I and interleukin-33 in plasma by peptide immunoaffinity enrichment and targeted mass spectrometry.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Mani D.R.
      • Shi X.
      • Kuhn E.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Quantification of cardiovascular biomarkers in patient plasma by targeted mass spectrometry and stable isotope dilution.
      ). Mean same-day process replicate CVs were below 15% both overall (Table I) and within each quartile across the range of low- to high-PrP samples (supplemental Table S4). Inter-day process replicate CVs were below 25%. Although PrP MRM is currently a Tier 2 assay, this intra-day and inter-day reproducibility would be consistent with Tier 1 assay requirements as described for targeted MS-based peptide measurement fit-for-purpose and is also within the best practice guidelines for clinical MRM assays (
      • Carr S.A.
      • Abbatiello S.E.
      • Ackermann B.L.
      • Borchers C.
      • Domon B.
      • Deutsch E.W.
      • Grant R.P.
      • Hoofnagle A.N.
      • Hüttenhain R.
      • Koomen J.M.
      • Liebler D.C.
      • Liu T.
      • MacLean B.
      • Mani D.R.
      • Mansfield E.
      • Neubert H.
      • Paulovich A.G.
      • Reiter L.
      • Vitek O.
      • Aebersold R.
      • Anderson L.
      • Bethem R.
      • Blonder J.
      • Boja E.
      • Botelho J.
      • Boyne M.
      • Bradshaw R.A.
      • Burlingame A.L.
      • Chan D.
      • Keshishian H.
      • Kuhn E.
      • Kinsinger C.
      • Lee J.S.H.
      • Lee S.-W.
      • Moritz R.
      • Oses-Prieto J.
      • Rifai N.
      • Ritchie J.
      • Rodriguez H.
      • Srinivas P.R.
      • Townsend R.R.
      • Van Eyk J.
      • Whiteley G.
      • Wiita A.
      • Weintraub S.
      Targeted peptide measurements in biology and medicine: best practices for mass spectrometry-based assay development using a fit-for-purpose approach.
      ,
      • Lee J.W.
      • Devanarayan V.
      • Barrett Y.C.
      • Weiner R.
      • Allinson J.
      • Fountain S.
      • Keller S.
      • Weinryb I.
      • Green M.
      • Duan L.
      • Rogers J.A.
      • Millham R.
      • O'Brien P.J.
      • Sailstad J.
      • Khan M.
      • Ray C.
      • Wagner J.A.
      Fit-for-purpose method development and validation for successful biomarker measurement.
      ,
      • Grant R.P.
      • Hoofnagle A.N.
      From lost in translation to paradise found: enabling protein biomarker method transfer by mass spectrometry.
      ). We did not formally determine lower and upper limits of quantification, but we observed response linearity across a dilution series of high- and low-PrP human CSF samples (supplemental Fig. S5) and over two orders of magnitude with spiked 15N recombinant human PrP (supplemental Fig. S6) as surrogate analyte across the range of expected PrP concentrations in clinical samples.
      Table IPerformance reproducibility of six human peptides in human CSF samples. Mean intra-day CV (based on same-day process duplicates of n = 55 samples); mean inter-day CV (based on a single inter-day control CSF sample analyzed in duplicate on n = 5 separate days; and inter-individual CV among the 55 different samples
      CodonsPeptideMean intra-day assay CVMean inter-day assay CVInter-individual CV
      25–37RPKPGGWNTGGSR10%16%80%
      38–48YPGQGSPGGNR12%22%52%
      137–148PIIHFGSDYEDR10%12%56%
      195–204GENFTETDVK9%12%58%
      209–220VVEQMCITQYER9%12%54%
      221–228ESQAYYQR10%18%70%
      These data suggest that PrP MRM is suitable for estimating the amount of PrP in CSF and measuring changes in abundance within and between patients. In further support of the applicability of this multiplex assay to answering biological questions in clinical samples, we found that for every peptide, the variability in amount of PrP between patient samples was much larger than the analytical variability, with inter-individual CVs of 52–80% contrasting with the observed tight process replicate agreement of ∼10% CV (Table I). Given that analytical variability was much smaller than biological variability, all six peptides were deemed suitable for analysis in clinical samples, and, owing to their different positions within PrP's amino acid sequence (Fig. 1A), each peptide was deemed able to inform independently upon the presence of its particular protein domain in CSF.

       PrP Peptide Abundance Is Reduced in the CSF of Patients With Prion Disease

      We used PrP MRM to quantify CSF PrP peptides in n = 55 clinical samples from individuals with rapidly progressive dementia referred to prion surveillance centers for testing and who ultimately either received nonprion disease diagnoses, or in whom sporadic or genetic prion disease was confirmed by autopsy (see Methods). All six human PrP peptides quantified by PrP MRM showed a marked decrease in abundance in prion disease patients compared with nonprion diagnoses, and all six peptides showed the same general pattern, with nonprion disease patients' CSF samples giving the highest mean peptide level, followed by sporadic prion disease, followed by genetic prion disease (Fig. 3A). The results from MRM mirrored the previously reported PrP ELISA results for these same 55 individuals (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ) (Fig. 3B), but differed in the estimated absolute amounts of PrP by ∼3-fold.
      Figure thumbnail gr3
      Fig. 3.Relative abundance of PrP as measured by six independent peptides is decreased in the CSF of prion disease patients. CSF PrP concentrations in n = 55 clinical CSF samples determined by A, PrP MRM for each of six peptides, normalized by peptide response factors to obtain PrP concentrations in ng/ml and arranged with the most N-terminal peptide at left and the most C-terminal peptide at right, compared with B, previously reported PrP ELISA results for the same samples, reproduced from Vallabh et al. (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). Black squares and bars show the mean and 95% confidence interval of the mean for each group.

       Relationship Between PrP MRM and ELISA

      Across the clinical samples, each peptide's abundance was positively correlated to the full-length PrP concentration determined by ELISA (Fig. 4A). The coefficients of correlation, from 0.40 to 0.72, are within the ranges reported for other MRM assays compared with corresponding immunoassays (
      • Kuhn E.
      • Addona T.
      • Keshishian H.
      • Burgess M.
      • Mani D.R.
      • Lee R.T.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Developing multiplexed assays for troponin I and interleukin-33 in plasma by peptide immunoaffinity enrichment and targeted mass spectrometry.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Mani D.R.
      • Shi X.
      • Kuhn E.
      • Sabatine M.S.
      • Gerszten R.E.
      • Carr S.A.
      Quantification of cardiovascular biomarkers in patient plasma by targeted mass spectrometry and stable isotope dilution.
      ,
      • Kuhn E.
      • Wu J.
      • Karl J.
      • Liao H.
      • Zolg W.
      • Guild B.
      Quantification of C-reactive protein in the serum of patients with rheumatoid arthritis using multiple reaction monitoring mass spectrometry and 13C-labeled peptide standards.
      ). All peptides were strongly correlated to one another, with coefficients of correlation ranging from 0.67 to 0.96, and no obvious differences within versus between protein domains (N- and C-terminal; Fig. 4B). The linear relationships between peptides were preserved across the range of samples analyzed (supplemental Fig. S5). These results, together with the fact that the magnitude of decrease in abundance in prion disease cases was similar for all peptides (Fig. 4A), suggested that PrP MRM and ELISA may be measuring the same analyte – predominantly full-length PrP. We therefore asked whether PrP MRM could serve as an orthogonal method to validate findings recently reported for ELISA.
      Figure thumbnail gr4
      Fig. 4.Correlations among PrP MRM peptides and with ELISA. A, Spearman's correlation between each peptide measured in MRM versus total PrP by ELISA. B, Spearman's correlation between every combination of peptides measured in MRM. All p < 0.01.
      Because plastic adsorption is reported to cause substantial loss of PrP in preanalytical handling, and detergent is reported to largely mitigate this (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ), we analyzed replicates of one CSF sample by MRM with and without 0.03% CHAPS detergent. As with ELISA, we found that the addition of CHAPS increased PrP peptide recovery by an average of 51% (p = 2.3e-8, Type I ANOVA). (The clinical CSF samples and recombinant protein used in this study contained added 0.03% CHAPS, see Methods).
      To compare PrP MRM and ELISA results while introducing covariates, we calculated a final estimated PrP concentration from MRM for each CSF sample by averaging the normalized PrP concentration across the six peptides. The estimated PrP concentrations obtained by MRM and by ELISA were correlated across CSF samples (r = 0.61, Spearman's correlation, p = 1.3e-6). MRM PrP concentration was uncorrelated with CSF hemoglobin (p = 0.85, Spearman's correlation), supporting the conclusion that blood contamination is not a source of CSF PrP (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ).
      The concentration of PrP in CSF measured by ELISA is correlated with the total protein concentration in CSF (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). This might reflect a true biological correlation across individuals, or it could reflect pre-analytical factors, if other proteins in CSF serve a blocking function, mitigating PrP loss to plastic during sample handling (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ). A potential concern, however, is that such a correlation could also arise if other proteins found in the human CSF also nonspecifically bind in the ELISA and contribute to background signal. If true, this would call into question the ability of ELISA-based PrP measurement to accurately measure a specific drug-dependent or disease state-dependent decrease in PrP concentration. To distinguish between these possible explanations for the reported PrP-total protein correlation, we tested the relationships between ELISA PrP concentration, MRM PrP concentration, and total protein concentration among our clinical samples. The correlation between ELISA PrP concentration and total protein concentration was marginal but observable among the 55 samples analyzed here (+94 ng/ml PrP per 1 mg total protein, p = 0.043, linear regression: ELISA PrP ∼ total protein), but this relationship vanished completely when MRM PrP concentration was included as a covariate (p = 0.60 for total protein in linear regression: ELISA PrP ∼ MRM PrP + total protein). Likewise, MRM PrP concentration was itself correlated to total protein (+238 ng/ml PrP per 1 mg/ml total protein, p = 0.017, linear regression: MRM PrP ∼ total protein). Together, the observations that the relationship between PrP and total protein was replicated in MRM, and that total protein did not explain any residual variance in ELISA-measured PrP after controlling for MRM-measured PrP, suggest that the correlation between CSF PrP and total protein in CSF is a genuine property of the samples analyzed, and that ELISA is specifically measuring PrP in human CSF.

       Application of PrP MRM to Preclinical Species of Interest

      Because the existing ELISA assay is specific to human PrP (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ), we sought to apply PrP MRM to analysis of tissues from species of interest for preclinical drug development. Synthetic peptides can offer advantages over full-length recombinant proteins for the development of targeted MS assays: they are quicker and less expensive to generate and quantify, and can be multiplexed to measure a number of proteins simultaneously. To test this for PrP and construct species-specific MRM assays, we quantified endogenous “light” PrP relative to 15N/13C single residue labeled (“heavy”) synthetic peptides that were spiked after trypsin digestion to quantify mouse, rat, or monkey PrP (supplemental Fig. S7). To assess cross-species selectivity and sensitivity, we analyzed human, rat, and cynomolgus macaque CSF as well as mouse and human brain homogenate, and compared observed results to the expected presence or absence of each peptide for each species based on amino acid sequence (Fig. 5A). For the six PrP peptides harboring sequence differences between species (Fig. 1A, supplemental Table S2), we observed slight differences in the retention time, best transition ion, and/or response level (supplemental Table S1, supplemental Table S5). We also found the MRM assay to be highly selective when the MS response between species for these peptides was compared. Each peptide consistently detected in sequence-matched species above the background level observed in nonsequence-matched species (Fig. 5B). Process replicate mean coefficients of variation (CVs) were <15% for all peptides, in line with our previous experience (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ) suggesting that heavy peptide standards can provide precision on par with full-length recombinant protein, albeit with different recovery levels (supplemental Table S5). We found that the total protein and lipid content of brain tissue complicated analysis of ≥1% w/v brain homogenates, but 0.5% w/v brain homogenates proved technically tractable in PrP MRM. Using mixtures of wild-type mouse brain homogenates titrated into a background of PrP knockout mouse brain homogenate, we confirmed a linear response for three mouse sequence-matched peptides, demonstrating the specificity of the assay for lowered PrP within a complex brain-derived peptide mixture (Fig. 5C).
      Figure thumbnail gr5
      Fig. 5.Development of the PrP MRM assay for preclinical species of interest. A, Control chart of the species expected for the peptides selected for this study, n = 6 human, n = 5 monkey, n = 5 mouse, and n = 5 rat. B, Sensitivity and selectivity across species. Data from n = 19 samples (n = 4 cynomolgous macaque CSF, n = 10 human CSF, n = 1 human brain, n = 1 mouse brain, and n = 4 rat CSF) in a total of n = 35 replicates were analyzed. L:H peptide ratios are shown for peptides expected in the respective species' samples (sequence-matched, orange) versus not expected (nonmatched, gray). All species-specific peptides were observed in the sequence-matched species at least an order of magnitude above the noise observed in nonsequence-matched species, except for ESQAYYDGR (sequence-matched species: mouse, rat), for which the separation was only about half an order of magnitude. C, 10% brain homogenate from wild-type mice (WT) or Edinburgh PrP knockout mice (
      • Manson J.C.
      • Clarke A.R.
      • Hooper M.L.
      • Aitchison L.
      • McConnell I.
      • Hope J.
      129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal.
      ) (KO) were mixed in seven different proportions (all KO, 10/90, 25/75, 50/50, 75/25, 90/10, and all WT), further diluted to 0.5% brain homogenate in saline and 0.03% CHAPS, and assayed by PrP MRM. Of the five peptides sequence-matched to mouse PrP, the three with best performance in this experiment (mean process replicate CV <10%) are shown here, again with individual replicates jittered along the x axis so that separate points are visible. Each peptide's L:H ratio is normalized to the average value of the two “all WT” replicates, and best-fit lines are shown. All three peptides exhibit good linearity, with y-intercepts very close to zero, as expected for PrP knockout mice, and adjusted R2 values ranging 98.2% - 99.0% (linear regression).

      DISCUSSION

      Here we describe a targeted mass spectrometry assay for measuring CSF PrP. Six of six human PrP peptides we quantified, from the N to the C terminus, were lowered in prion disease patients compared with nonprion disease patients. CSF PrP may therefore be difficult to interpret as a pharmacodynamic biomarker in symptomatic prion disease patients, because the direct effect of a PrP-lowering drug and the effect of disease process alleviation would be expected to push CSF PrP in opposing directions. Instead, trials to demonstrate target engagement and perform dose-finding for a PrP-lowering drug will likely need to be conducted in presymptomatic individuals at risk for genetic prion disease (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ,
      • Vallabh S.M.
      Antisense oligonucleotides for the prevention of genetic prion disease.
      ). Preliminary evidence suggests that test-retest stability of CSF PrP in individuals without active prion disease is good (CV < 13%), and a clinical research study to validate this finding in at-risk PrP mutation carriers is underway (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ,
      • Vallabh S.
      Prion Alliance sponsors MGH research study.
      ). If CSF PrP, absent drug treatment, is indeed stable over time in presymptomatic mutation carriers, then a drug-dependent decrease of 40%, as recently observed for mutant huntingtin in an antisense oligonucleotide trial (
      • Tabrizi S.J.
      • Leavitt B.R.
      • Landwehrmeyer G.B.
      • Wild E.J.
      • Saft C.
      • Barker R.A.
      • Blair N.F.
      • Craufurd D.
      • Priller J.
      • Rickards H.
      • Rosser A.
      • Kordasiewicz H.B.
      • Czech C.
      • Swayze E.E.
      • Norris D.A.
      • Baumann T.
      • Gerlach I.
      • Schobel S.A.
      • Paz E.
      • Smith A.V.
      • Bennett C.F.
      • Lane R.M.
      Targeting huntingtin expression in patients with Huntington's disease.
      ), should be readily quantifiable.
      Our data support the interpretability of CSF PrP in such a trial context, for three reasons. First, we provide evidence that CSF PrP is a simple, well-behaved analyte: all PrP peptides behave similarly. This contrasts with the complex situation reported for tau isoforms in CSF (
      • Meredith J.E.
      • Sankaranarayanan S.
      • Guss V.
      • Lanzetti A.J.
      • Berisha F.
      • Neely R.J.
      • Slemmon J.R.
      • Portelius E.
      • Zetterberg H.
      • Blennow K.
      • Soares H.
      • Ahlijanian M.
      • Albright C.F.
      Characterization of novel CSF Tau and ptau biomarkers for Alzheimer's disease.
      ,
      • Sato C.
      • Barthélemy N.R.
      • Mawuenyega K.G.
      • Patterson B.W.
      • Gordon B.A.
      • Jockel-Balsarotti J.
      • Sullivan M.
      • Crisp M.J.
      • Kasten T.
      • Kirmess K.M.
      • Kanaan N.M.
      • Yarasheski K.E.
      • Baker-Nigh A.
      • Benzinger T.L.S.
      • Miller T.M.
      • Karch C.M.
      • Bateman R.J.
      Tau kinetics in neurons and the human central nervous system.
      ), and suggests that various approaches to measuring PrP in CSF can all be interpreted to reflect the level of the relevant, disease-causing protein. Second, we confirm that CSF PrP is not correlated with CSF hemoglobin. This supports the brain and not blood origin of CSF PrP and leads us to expect that pharmacologic lowering of brain PrP will be mirrored in CSF. Third, we replicate the reported (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ) correlation between CSF PrP and CSF total protein content. The correlation between PrP and total protein might reflect genuine biological variables such as age and CSF flow rate (
      • McCudden C.R.
      • Brooks J.
      • Figurado P.
      • Bourque P.R.
      Cerebrospinal fluid total protein reference intervals derived from 20 years of patient data.
      ), and/or it might arise from pre-analytical variables (
      • Vallabh S.M.
      • Nobuhara C.K.
      • Llorens F.
      • Zerr I.
      • Parchi P.
      • Capellari S.
      • Kuhn E.
      • Klickstein J.
      • Safar J.G.
      • Nery F.C.
      • Swoboda K.J.
      • Geschwind M.D.
      • Zetterberg H.
      • Arnold S.E.
      • Minikel E.V.
      • Schreiber S.L.
      Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
      ), but our observation of this correlation by both mass spectrometry and ELISA argues against the possibility that this correlation simply results from matrix interference in ELISA. This finding supports the overall interpretability of findings from prior, ELISA-based, studies of CSF PrP.
      Our study has several limitations. First, we have only compared samples between prion and nonprion disease patients to examine the effect of the disease state on CSF PrP. Determining the effect of PrP-lowering drug treatment on CSF PrP is a priority for future work. Second, we still cannot exclude the possibility that protein misfolding contributes somewhat to the decrease in CSF PrP that we observe, because the chaotrope used here–6 m urea–has not been proven to denature all misfolded PrP. This concentration of urea was shown to abolish 99.99% of hamster prion infectivity (
      • Prusiner S.B.
      • Groth D.
      • Serban A.
      • Stahl N.
      • Gabizon R.
      Attempts to restore scrapie prion infectivity after exposure to protein denaturants.
      ), but prion strains differ in their conformational stability (
      • Safar J.
      • Wille H.
      • Itri V.
      • Groth D.
      • Serban H.
      • Torchia M.
      • Cohen F.E.
      • Prusiner S.B.
      Eight prion strains have PrP(Sc) molecules with different conformations.
      ). Human prions unfold at ∼3 M guanidine hydrochloride (
      • Kim C.
      • Haldiman T.
      • Cohen Y.
      • Chen W.
      • Blevins J.
      • Sy M.-S.
      • Cohen M.
      • Safar J.G.
      Protease-sensitive conformers in broad spectrum of distinct PrPSc structures in sporadic Creutzfeldt-Jakob disease are indicator of progression rate.
      ,
      • Kim C.
      • Haldiman T.
      • Surewicz K.
      • Cohen Y.
      • Chen W.
      • Blevins J.
      • Sy M.-S.
      • Cohen M.
      • Kong Q.
      • Telling G.C.
      • Surewicz W.K.
      • Safar J.G.
      Small protease sensitive oligomers of PrPSc in distinct human prions determine conversion rate of PrP(C).
      ,
      • Cescatti M.
      • Saverioni D.
      • Capellari S.
      • Tagliavini F.
      • Kitamoto T.
      • Ironside J.
      • Giese A.
      • Parchi P.
      Analysis of conformational stability of abnormal prion protein aggregates across the spectrum of Creutzfeldt-Jakob disease prions.
      ), but urea is a less potent denaturant (
      • Greene R.F.
      • Pace C.N.
      Urea and guanidine hydrochloride denaturation of ribonuclease, lysozyme, alpha-chymotrypsin, and beta-lactoglobulin.
      ). Third, because bacterially expressed recombinant PrP is an imperfect standard by which to quantify mammalian PrP, our data do not support any firm conclusions about the baseline composition of PrP in terms of different cleavage products in human CSF generally. A 15N protein standard purified from a mammalian expression system would better account for the digestion efficiencies of PrP in CSF. Nevertheless, by comparing the abundance of each PrP peptide between diagnostic categories – individuals with and without prion disease – we do establish that any changes driven by the disease state apparently affect all domains of PrP equally. This finding is not inconsistent with existing literature: for example, the PrP C2 fragment resulting from beta cleavage is known to be increased in brain parenchyma during prion disease (
      • Watts J.C.
      • Stöhr J.
      • Bhardwaj S.
      • Wille H.
      • Oehler A.
      • Dearmond S.J.
      • Giles K.
      • Prusiner S.B.
      Protease-resistant prions selectively decrease Shadoo protein.
      ), but if C2 is then retained in intracellular aggregates rather than being shed, whereas its counterpart N2 is rapidly degraded, then increased beta cleavage might result in both N- and C-terminal PrP peptides being decreased in prion disease CSF, as observed here.
      The specificity for PrP peptides observed across species and in wild-type versus PrP knockout mouse brain suggests that PrP MRM should be useful in preclinical development of PrP-lowering drugs. PrP MRM is currently a Tier 2 targeted assay (
      • Carr S.A.
      • Abbatiello S.E.
      • Ackermann B.L.
      • Borchers C.
      • Domon B.
      • Deutsch E.W.
      • Grant R.P.
      • Hoofnagle A.N.
      • Hüttenhain R.
      • Koomen J.M.
      • Liebler D.C.
      • Liu T.
      • MacLean B.
      • Mani D.R.
      • Mansfield E.
      • Neubert H.
      • Paulovich A.G.
      • Reiter L.
      • Vitek O.
      • Aebersold R.
      • Anderson L.
      • Bethem R.
      • Blonder J.
      • Boja E.
      • Botelho J.
      • Boyne M.
      • Bradshaw R.A.
      • Burlingame A.L.
      • Chan D.
      • Keshishian H.
      • Kuhn E.
      • Kinsinger C.
      • Lee J.S.H.
      • Lee S.-W.
      • Moritz R.
      • Oses-Prieto J.
      • Rifai N.
      • Ritchie J.
      • Rodriguez H.
      • Srinivas P.R.
      • Townsend R.R.
      • Van Eyk J.
      • Whiteley G.
      • Wiita A.
      • Weintraub S.
      Targeted peptide measurements in biology and medicine: best practices for mass spectrometry-based assay development using a fit-for-purpose approach.
      ), with partial analytical validation suggesting a potential for further work to elevate it to Tier 1 and enable clinical use. Given the likely important role of CSF PrP as a biomarker in future prion disease therapeutic development, and the different pros and cons of ELISA and MRM methods, availability of both assays will be an asset. For clinical measurements of a single protein, wider instrument availability sometimes favors the use of ELISA, but there are counter-examples, such as thyroglobulin, where patients' anti-idiotypic antibodies can lead to false ELISA readings (
      • Hoofnagle A.N.
      • Wener M.H.
      Serum thyroglobulin: a model of immunoassay imperfection.
      ). Targeted MS measurements avoid such interference, because any autoantibodies are reduced to peptides before analysis. In addition, PrP MRM may offer other advantages over ELISA, including the use of a single assay across preclinical animals and clinical studies as well as wide dynamic range without sample dilution. Finally, because PrP MRM monitors well-defined peptide analytes, any potential for interference from post-translational modifications or patient missense mutations can be determined a priori based on sequence information. Adaptation of PrP MRM for clinical use will require modifications and improvements to the protocol described here. The LC/MS gradient of 45 min we used is longer than the 5–10 min expected for high-throughput clinical biomarker assays. To transition to clinical use, feasibility and performance should be assessed at a faster gradient under higher (e.g. microflow) conditions using commercially available C18 columns. An increase in assay throughput may come at the cost of some sensitivity, but because all PrP peptides in this study demonstrated comparable behavior across this set of clinical samples, a future implementation of PrP MRM might choose to monitor fewer or even a single peptide, facilitating the implementation of a significantly faster assay. In addition, we ported assay parameters that we use for plasma-based MRM (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Kuhn E.
      • Carr S.A.
      Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
      ) without extensive optimization. Thus, testing and improvement of digest and cleanup conditions could further improve recovery and performance. Finally, once such an assay is implemented, robust bioanalytical method validation will be expected if the assay is to be used in clinical decision-making (
      • Kuhn E.
      • Whiteaker J.R.
      • Mani D.R.
      • Jackson A.M.
      • Zhao L.
      • Pope M.E.
      • Smith D.
      • Rivera K.D.
      • Anderson N.L.
      • Skates S.J.
      • Pearson T.W.
      • Paulovich A.G.
      • Carr S.A.
      Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
      ,
      • Keshishian H.
      • Addona T.
      • Burgess M.
      • Kuhn E.
      • Carr S.A.
      Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
      ,
      • Whiteaker J.R.
      • Halusa G.N.
      • Hoofnagle A.N.
      • Sharma V.
      • MacLean B.
      • Yan P.
      • Wrobel J.A.
      • Kennedy J.
      • Mani D.R.
      • Zimmerman L.J.
      • Meyer M.R.
      • Mesri M.
      • Rodriguez H.
      CPTAC Assay Portal: a repository of targeted proteomic assays.
      ,
      Food and Drug Administration. Bioanalytical Method Validation.
      ).
      Our data and methods provide a proof of concept for MRM-based quantification of PrP in CSF and provide a roadmap for further development of the assay for eventual use in the clinic.

      DATA AVAILABILITY

      All processed data and source code for this study are provided in a public GitHub repository at https://github.com/ericminikel/prp_mrm and are sufficient to reproduce the analyses and figures herein. This repository also includes a summary table for download containing the MRM results (light and 15N peak areas, light:15N ratio and normalized PrP concentration in ng/mL) for all clinical samples and all peptides. Skyline files have been uploaded to Panorama (
      • Sharma V.
      • Eckels J.
      • Schilling B.
      • Ludwig C.
      • Jaffe J.D.
      • MacCoss M.J.
      • MacLean B.
      Panorama public: a public repository for quantitative data sets processed in Skyline.
      ) at https://panoramaweb.org/yVRdi1.url and data can be downloaded from ProteomeXchange (http://proteomecentral.proteomexchange.org/cgi/GetDataset?ID=PXD014781).

      Supplementary Material

      REFERENCES

        • Prusiner S.B.
        Prions.
        Proc. Natl. Acad. Sci. U.S.A. 1998; 95: 13363-13383
        • Raymond G.J.
        • Zhao H.T.
        • Race B.
        • Raymond L.D.
        • Williams K.
        • Swayze E.E.
        • Graffam S.
        • Le J.
        • Caron T.
        • Stathopoulos J.
        • O'Keefe R.
        • Lubke L.L.
        • Reidenbach A.G.
        • Kraus A.
        • Schreiber S.L.
        • Mazur C.
        • Cabin D.E.
        • Carroll J.B.
        • Minikel E.V.
        • Kordasiewicz H.
        • Caughey B.
        • Vallabh S.M.
        Antisense oligonucleotides extend survival of prion-infected mice.
        JCI Insight. 2019; 5: 131175
        • Vallabh S.M.
        • Nobuhara C.K.
        • Llorens F.
        • Zerr I.
        • Parchi P.
        • Capellari S.
        • Kuhn E.
        • Klickstein J.
        • Safar J.G.
        • Nery F.C.
        • Swoboda K.J.
        • Geschwind M.D.
        • Zetterberg H.
        • Arnold S.E.
        • Minikel E.V.
        • Schreiber S.L.
        Prion protein quantification in human cerebrospinal fluid as a tool for prion disease drug development.
        Proc. Natl. Acad. Sci. 2019; 116: 7793-7798
        • Minikel E.V.
        • Vallabh S.M.
        • Lek M.
        • Estrada K.
        • Samocha K.E.
        • Sathirapongsasuti J.F.
        • McLean C.Y.
        • Tung J.Y.
        • Yu L.P.C.
        • Gambetti P.
        • Blevins J.
        • Zhang S.
        • Cohen Y.
        • Chen W.
        • Yamada M.
        • Hamaguchi T.
        • Sanjo N.
        • Mizusawa H.
        • Nakamura Y.
        • Kitamoto T.
        • Collins S.J.
        • Boyd A.
        • Will R.G.
        • Knight R.
        • Ponto C.
        • Zerr I.
        • Kraus T.F.J.
        • Eigenbrod S.
        • Giese A.
        • Calero M.
        • de Pedro-Cuesta J.
        • Haïk S.
        • Laplanche J.-L.
        • Bouaziz-Amar E.
        • Brandel J.-P.
        • Capellari S.
        • Parchi P.
        • Poleggi A.
        • Ladogana A.
        • O'Donnell-Luria A.H.
        • Karczewski K.J.
        • Marshall J.L.
        • Boehnke M.
        • Laakso M.
        • Mohlke K.L.
        • Kähler A.
        • Chambert K.
        • McCarroll S.
        • Sullivan P.F.
        • Hultman C.M.
        • Purcell S.M.
        • Sklar P.
        • van der Lee S.J.
        • Rozemuller A.
        • Jansen C.
        • Hofman A.
        • Kraaij R.
        • van Rooij J.G.J.
        • Ikram M.A.
        • Uitterlinden A.G.
        • van Duijn C.M.
        • Daly M.J.
        • MacArthur D.G.
        Exome Aggregation Consortium (ExAC). Quantifying prion disease penetrance using large population control cohorts.
        Sci. Transl. Med. 2016; 8: 322ra9
        • Minikel E.V.
        • Vallabh S.M.
        • Orseth M.C.
        • Brandel J.-P.
        • Haïk S.
        • Laplanche J.-L.
        • Zerr I.
        • Parchi P.
        • Capellari S.
        • Safar J.
        • Kenny J.
        • Fong J.C.
        • Takada L.T.
        • Ponto C.
        • Hermann P.
        • Knipper T.
        • Stehmann C.
        • Kitamoto T.
        • Ae R.
        • Hamaguchi T.
        • Sanjo N.
        • Tsukamoto T.
        • Mizusawa H.
        • Collins S.J.
        • Chiesa R.
        • Roiter I.
        • de Pedro-Cuesta J.
        • Calero M.
        • Geschwind M.D.
        • Yamada M.
        • Nakamura Y.
        • Mead S.
        Age at onset in genetic prion disease and the design of preventive clinical trials.
        Neurology. 2019; 93: e125-e134
        • Vallabh S.M.
        Antisense oligonucleotides for the prevention of genetic prion disease.
        PhD dissertation, Harvard University, Cambridge, MA. 2019;
      1. Food, U. S., Drug Administration. Critical Path Innovation Meeting: Genetic Prion Disease..
        Requestor: Broad Institute. 2017;
        • Altmeppen H.C.
        • Puig B.
        • Dohler F.
        • Thurm D.K.
        • Falker C.
        • Krasemann S.
        • Glatzel M.
        Proteolytic processing of the prion protein in health and disease.
        Am. J. Neurodegener. Dis. 2012; 1: 15-31
        • Linsenmeier L.
        • Mohammadi B.
        • Wetzel S.
        • Puig B.
        • Jackson W.S.
        • Hartmann A.
        • Uchiyama K.
        • Sakaguchi S.
        • Endres K.
        • Tatzelt J.
        • Saftig P.
        • Glatzel M.
        • Altmeppen H.C.
        Structural and mechanistic aspects influencing the ADAM10-mediated shedding of the prion protein.
        Mol. Neurodegener. 2018; 13: 18
        • Tagliavini F.
        • Prelli F.
        • Porro M.
        • Salmona M.
        • Bugiani O.
        • Frangione B.
        A soluble form of prion protein in human cerebrospinal fluid: implications for prion-related encephalopathies.
        Biochem. Biophys. Res. Commun. 1992; 184: 1398-1404
        • Schmitz M.
        • Schlomm M.
        • Hasan B.
        • Beekes M.
        • Mitrova E.
        • Korth C.
        • Breil A.
        • Carimalo J.
        • Gawinecka J.
        • Varges D.
        • Zerr I.
        Codon 129 polymorphism and the E200K mutation do not affect the cellular prion protein isoform composition in the cerebrospinal fluid from patients with Creutzfeldt-Jakob disease.
        Eur. J. Neurosci. 2010; 31: 2024-2031
        • Chen S.G.
        • Teplow D.B.
        • Parchi P.
        • Teller J.K.
        • Gambetti P.
        • Autilio-Gambetti L.
        Truncated forms of the human prion protein in normal brain and in prion diseases.
        J. Biol. Chem. 1995; 270: 19173-19180
        • Fevrier B.
        • Vilette D.
        • Archer F.
        • Loew D.
        • Faigle W.
        • Vidal M.
        • Laude H.
        • Raposo G.
        Cells release prions in association with exosomes.
        Proc. Natl. Acad. Sci. U.S.A. 2004; 101: 9683-9688
        • Meyne F.
        • Gloeckner S.F.
        • Ciesielczyk B.
        • Heinemann U.
        • Krasnianski A.
        • Meissner B.
        • Zerr I.
        Total prion protein levels in the cerebrospinal fluid are reduced in patients with various neurological disorders.
        J. Alzheimers Dis. 2009; 17: 863-873
        • Dorey A.
        • Tholance Y.
        • Vighetto A.
        • Perret-Liaudet A.
        • Lachman I.
        • Krolak-Salmon P.
        • Wagner U.
        • Struyfs H.
        • De Deyn P.P.
        • El-Moualij B.
        • Zorzi W.
        • Meyronet D.
        • Streichenberger N.
        • Engelborghs S.
        • Kovacs G.G.
        • Quadrio I.
        Association of cerebrospinal fluid prion protein levels and the distinction between Alzheimer disease and Creutzfeldt-Jakob disease.
        JAMA Neurol. 2015; 72: 267-275
        • Abu Rumeileh S.
        • Lattanzio F.
        • Stanzani Maserati M.
        • Rizzi R.
        • Capellari S.
        • Parchi P.
        Diagnostic Accuracy of a Combined Analysis of Cerebrospinal Fluid t-PrP, t-tau, p-tau, and Aβ42 in the Differential Diagnosis of Creutzfeldt-Jakob Disease from Alzheimer's Disease with Emphasis on Atypical Disease Variants.
        J. Alzheimers Dis. 2017; 55: 1471-1480
        • Villar-Piqué A.
        • Schmitz M.
        • Lachmann I.
        • Karch A.
        • Calero O.
        • Stehmann C.
        • Sarros S.
        • Ladogana A.
        • Poleggi A.
        • Santana I.
        • Ferrer I.
        • Mitrova E.
        • Žáková
        • Pocchiari D.M.
        • Baldeiras I.
        • Calero M.
        • Collins S.J.
        • Geschwind M.D.
        • Sánchez-Valle R.
        • Zerr I.
        • Llorens F.
        Cerebrospinal fluid total prion protein in the spectrum of prion diseases.
        Mol. Neurobiol. 2018; 56: 2811-2821
        • Schulz-Schaeffer W.J.
        • Tschöke S.
        • Kranefuss N.
        • Dröse W.
        • Hause-Reitner D.
        • Giese A.
        • Groschup M.H.
        • Kretzschmar H.A.
        The paraffin-embedded tissue blot detects PrP(Sc) early in the incubation time in prion diseases.
        Am. J. Pathol. 2000; 156: 51-56
        • Safar J.G.
        • DeArmond S.J.
        • Kociuba K.
        • Deering C.
        • Didorenko S.
        • Bouzamondo-Bernstein E.
        • Prusiner S.B.
        • Tremblay P.
        Prion clearance in bigenic mice.
        J. Gen. Virol. 2005; 86: 2913-2923
        • Moreno J.A.
        • Radford H.
        • Peretti D.
        • Steinert J.R.
        • Verity N.
        • Martin M.G.
        • Halliday M.
        • Morgan J.
        • Dinsdale D.
        • Ortori C.A.
        • Barrett D.A.
        • Tsaytler P.
        • Bertolotti A.
        • Willis A.E.
        • Bushell M.
        • Mallucci G.R.
        Sustained translational repression by eIF2α-P mediates prion neurodegeneration.
        Nature. 2012; 485: 507-511
        • Parchi P.
        • Giese A.
        • Capellari S.
        • Brown P.
        • Schulz-Schaeffer W.
        • Windl O.
        • Zerr I.
        • Budka H.
        • Kopp N.
        • Piccardo P.
        • Poser S.
        • Rojiani A.
        • Streichemberger N.
        • Julien J.
        • Vital C.
        • Ghetti B.
        • Gambetti P.
        • Kretzschmar H.
        Classification of sporadic Creutzfeldt-Jakob disease based on molecular and phenotypic analysis of 300 subjects.
        Ann. Neurol. 1999; 46: 224-233
        • Caughey B.
        • Raymond G.J.
        • Ernst D.
        • Race R.E.
        N-terminal truncation of the scrapie-associated form of PrP by lysosomal protease(s): implications regarding the site of conversion of PrP to the protease-resistant state.
        J. Virol. 1991; 65: 6597-6603
        • Goold R.
        • McKinnon C.
        • Rabbanian S.
        • Collinge J.
        • Schiavo G.
        • Tabrizi S.J.
        Alternative fates of newly formed PrPSc upon prion conversion on the plasma membrane.
        J. Cell Sci. 2013; 126: 3552-3562
        • Mays C.E.
        • Kim C.
        • Haldiman T.
        • van der Merwe J.
        • Lau A.
        • Yang J.
        • Grams J.
        • Di Bari M.A.
        • Nonno R.
        • Telling G.C.
        • Kong Q.
        • Langeveld J.
        • McKenzie D.
        • Westaway D.
        • Safar J.G.
        Prion disease tempo determined by host-dependent substrate reduction.
        J. Clin. Invest. 2014; 124: 847-858
        • Safar J.
        • Wille H.
        • Itri V.
        • Groth D.
        • Serban H.
        • Torchia M.
        • Cohen F.E.
        • Prusiner S.B.
        Eight prion strains have PrP(Sc) molecules with different conformations.
        Nat. Med. 1998; 4: 1157-1165
        • Watts J.C.
        • Stöhr J.
        • Bhardwaj S.
        • Wille H.
        • Oehler A.
        • Dearmond S.J.
        • Giles K.
        • Prusiner S.B.
        Protease-resistant prions selectively decrease Shadoo protein.
        PLoS Pathog. 2011; 7: e1002382
        • Carr S.A.
        • Abbatiello S.E.
        • Ackermann B.L.
        • Borchers C.
        • Domon B.
        • Deutsch E.W.
        • Grant R.P.
        • Hoofnagle A.N.
        • Hüttenhain R.
        • Koomen J.M.
        • Liebler D.C.
        • Liu T.
        • MacLean B.
        • Mani D.R.
        • Mansfield E.
        • Neubert H.
        • Paulovich A.G.
        • Reiter L.
        • Vitek O.
        • Aebersold R.
        • Anderson L.
        • Bethem R.
        • Blonder J.
        • Boja E.
        • Botelho J.
        • Boyne M.
        • Bradshaw R.A.
        • Burlingame A.L.
        • Chan D.
        • Keshishian H.
        • Kuhn E.
        • Kinsinger C.
        • Lee J.S.H.
        • Lee S.-W.
        • Moritz R.
        • Oses-Prieto J.
        • Rifai N.
        • Ritchie J.
        • Rodriguez H.
        • Srinivas P.R.
        • Townsend R.R.
        • Van Eyk J.
        • Whiteley G.
        • Wiita A.
        • Weintraub S.
        Targeted peptide measurements in biology and medicine: best practices for mass spectrometry-based assay development using a fit-for-purpose approach.
        Mol. Cell. Proteomics. 2014; 13: 907-917
        • Hermann P.
        • Laux M.
        • Glatzel M.
        • Matschke J.
        • Knipper T.
        • Goebel S.
        • Treig J.
        • Schulz-Schaeffer W.
        • Cramm M.
        • Schmitz M.
        • Zerr I.
        Validation and utilization of amended diagnostic criteria in Creutzfeldt-Jakob disease surveillance.
        Neurology. 2018; 91: e331-e338
        • Orrú C.D.
        • Groveman B.R.
        • Hughson A.G.
        • Zanusso G.
        • Coulthart M.B.
        • Caughey B.
        Rapid and sensitive RT-QuIC detection of human Creutzfeldt-Jakob disease using cerebrospinal fluid.
        mBio. 2015; 6: e02451-14
        • Manson J.C.
        • Clarke A.R.
        • Hooper M.L.
        • Aitchison L.
        • McConnell I.
        • Hope J.
        129/Ola mice carrying a null mutation in PrP that abolishes mRNA production are developmentally normal.
        Mol. Neurobiol. 1994; 8: 121-127
        • Striebel J.F.
        • Race B.
        • Pathmajeyan M.
        • Rangel A.
        • Chesebro B.
        Lack of influence of prion protein gene expression on kainate-induced seizures in mice: studies using congenic, coisogenic and transgenic strains.
        Neuroscience. 2013; 238: 11-18
        • Wilham J.M.
        • Orrú C.D.
        • Bessen R.A.
        • Atarashi R.
        • Sano K.
        • Race B.
        • Meade-White K.D.
        • Taubner L.M.
        • Timmes A.
        • Caughey B.
        Rapid end-point quantitation of prion seeding activity with sensitivity comparable to bioassays.
        PLoS Pathog. 2010; 6: e1001217
        • Orrù C.D.
        • Groveman B.R.
        • Hughson A.G.
        • Manca M.
        • Raymond L.D.
        • Raymond G.J.
        • Campbell K.J.
        • Anson K.J.
        • Kraus A.
        • Caughey B.
        RT-QuIC assays for prion disease detection and diagnostics.
        Methods Mol. Biol. 2017; 1658: 185-203
      2. The UniProt Consortium. UniProt: the universal protein knowledgebase.
        Nucleic Acids Res. 2017; 45: D158-D169
        • Sievers F.
        • Higgins D.G.
        Clustal Omega for making accurate alignments of many protein sequences.
        Protein Sci. Publ. Protein Soc. 2018; 27: 135-145
        • Kuhn E.
        • Carr S.A.
        Multiplexed immunoaffinity enrichment of peptides with anti-peptide antibodies and quantification by stable isotope dilution multiple reaction monitoring mass spectrometry.
        Methods Mol. Biol. 2016; 1410: 135-167
        • Hoofnagle A.N.
        • Whiteaker J.R.
        • Carr S.A.
        • Kuhn E.
        • Liu T.
        • Massoni S.A.
        • Thomas S.N.
        • Townsend R.R.
        • Zimmerman L.J.
        • Boja E.
        • Chen J.
        • Crimmins D.L.
        • Davies S.R.
        • Gao Y.
        • Hiltke T.R.
        • Ketchum K.A.
        • Kinsinger C.R.
        • Mesri M.
        • Meyer M.R.
        • Qian W.-J.
        • Schoenherr R.M.
        • Scott M.G.
        • Shi T.
        • Whiteley G.R.
        • Wrobel J.A.
        • Wu C.
        • Ackermann B.L.
        • Aebersold R.
        • Barnidge D.R.
        • Bunk D.M.
        • Clarke N.
        • Fishman J.B.
        • Grant R.P.
        • Kusebauch U.
        • Kushnir M.M.
        • Lowenthal M.S.
        • Moritz R.L.
        • Neubert H.
        • Patterson S.D.
        • Rockwood A.L.
        • Rogers J.
        • Singh R.J.
        • Van Eyk J.E.
        • Wong S.H.
        • Zhang S.
        • Chan D.W.
        • Chen X.
        • Ellis M.J.
        • Liebler D.C.
        • Rodland K.D.
        • Rodriguez H.
        • Smith R.D.
        • Zhang Z.
        • Zhang H.
        • Paulovich A.G.
        Recommendations for the generation, quantification, storage, and handling of peptides used for mass spectrometry-based assays.
        Clin. Chem. 2016; 62: 48-69
        • Kuhn E.
        • Whiteaker J.R.
        • Mani D.R.
        • Jackson A.M.
        • Zhao L.
        • Pope M.E.
        • Smith D.
        • Rivera K.D.
        • Anderson N.L.
        • Skates S.J.
        • Pearson T.W.
        • Paulovich A.G.
        • Carr S.A.
        Interlaboratory evaluation of automated, multiplexed peptide immunoaffinity enrichment coupled to multiple reaction monitoring mass spectrometry for quantifying proteins in plasma.
        Mol. Cell. Proteomics. 2012; 11
        • Moore R.A.
        • Head M.W.
        • Ironside J.W.
        • Ritchie D.L.
        • Zanusso G.
        • Choi Y.P.
        • Pyo Choi Y.
        • Priola S.A.
        The distribution of prion protein allotypes differs between sporadic and iatrogenic Creutzfeldt-Jakob disease patients.
        PLoS Pathog. 2016; 12: e1005416
        • Moore R.A.
        • Ward A.
        • Race B.
        • Priola S.A.
        Processing of high-titer prions for mass spectrometry inactivates prion infectivity.
        Biochim. Biophys. Acta Proteins Proteomics. 2018; 1866: 1174-1180
        • Rappsilber J.
        • Mann M.
        • Ishihama Y.
        Protocol for micro-purification, enrichment, pre-fractionation and storage of peptides for proteomics using StageTips.
        Nat. Protoc. 2007; 2: 1896-1906
        • Broudy D.
        • Killeen T.
        • Choi M.
        • Shulman N.
        • Mani D.R.
        • Abbatiello S.E.
        • Mani D.
        • Ahmad R.
        • Sahu A.K.
        • Schilling B.
        • Tamura K.
        • Boss Y.
        • Sharma V.
        • Gibson B.W.
        • Carr S.A.
        • Vitek O.
        • MacCoss M.J.
        • MacLean B.
        A framework for installable external tools in Skyline.
        Bioinforma. Oxf. Engl. 2014; 30: 2521-2523
        • Keshishian H.
        • Addona T.
        • Burgess M.
        • Kuhn E.
        • Carr S.A.
        Quantitative, multiplexed assays for low abundance proteins in plasma by targeted mass spectrometry and stable isotope dilution.
        Mol. Cell. Proteomics. 2007; 6: 2212-2229
        • Kuhn E.
        • Addona T.
        • Keshishian H.
        • Burgess M.
        • Mani D.R.
        • Lee R.T.
        • Sabatine M.S.
        • Gerszten R.E.
        • Carr S.A.
        Developing multiplexed assays for troponin I and interleukin-33 in plasma by peptide immunoaffinity enrichment and targeted mass spectrometry.
        Clin. Chem. 2009; 55: 1108-1117
        • Abbatiello S.E.
        • Schilling B.
        • Mani D.R.
        • Zimmerman L.J.
        • Hall S.C.
        • MacLean B.
        • Albertolle M.
        • Allen S.
        • Burgess M.
        • Cusack M.P.
        • Gosh M.
        • Hedrick V.
        • Held J.M.
        • Inerowicz H.D.
        • Jackson A.
        • Keshishian H.
        • Kinsinger C.R.
        • Lyssand J.
        • Makowski L.
        • Mesri M.
        • Rodriguez H.
        • Rudnick P.
        • Sadowski P.
        • Sedransk N.
        • Shaddox K.
        • Skates S.J.
        • Kuhn E.
        • Smith D.
        • Whiteaker J.R.
        • Whitwell C.
        • Zhang S.
        • Borchers C.H.
        • Fisher S.J.
        • Gibson B.W.
        • Liebler D.C.
        • MacCoss M.J.
        • Neubert T.A.
        • Paulovich A.G.
        • Regnier F.E.
        • Tempst P.
        • Carr S.A.
        Large-scale interlaboratory study to develop, analytically validate and apply highly multiplexed, quantitative peptide assays to measure cancer-relevant proteins in plasma.
        Mol. Cell. Proteomics. 2015; 14: 2357-2374
        • Scott K.B.
        • Turko I.V.
        • Phinney K.W.
        Quantitative performance of internal standard platforms for absolute protein quantification using multiple reaction monitoring-mass spectrometry.
        Anal. Chem. 2015; 87: 4429-4435
        • Keshishian H.
        • Addona T.
        • Burgess M.
        • Mani D.R.
        • Shi X.
        • Kuhn E.
        • Sabatine M.S.
        • Gerszten R.E.
        • Carr S.A.
        Quantification of cardiovascular biomarkers in patient plasma by targeted mass spectrometry and stable isotope dilution.
        Mol. Cell. Proteomics. 2009; 8: 2339-2349
        • Fusaro V.A.
        • Mani D.R.
        • Mesirov J.P.
        • Carr S.A.
        Prediction of high-responding peptides for targeted protein assays by mass spectrometry.
        Nat. Biotechnol. 2009; 27: 190-198
        • Altmeppen H.C.
        • Prox J.
        • Puig B.
        • Kluth M.A.
        • Bernreuther C.
        • Thurm D.
        • Jorissen E.
        • Petrowitz B.
        • Bartsch U.
        • De Strooper B.
        • Saftig P.
        • Glatzel M.
        Lack of a-disintegrin-and-metalloproteinase ADAM10 leads to intracellular accumulation and loss of shedding of the cellular prion protein in vivo.
        Mol. Neurodegener. 2011; 6: 36
        • Taylor D.R.
        • Parkin E.T.
        • Cocklin S.L.
        • Ault J.R.
        • Ashcroft A.E.
        • Turner A.J.
        • Hooper N.M.
        Role of ADAMs in the ectodomain shedding and conformational conversion of the prion protein.
        J. Biol. Chem. 2009; 284: 22590-22600
        • Frottin F.
        • Martinez A.
        • Peynot P.
        • Mitra S.
        • Holz R.C.
        • Giglione C.
        • Meinnel T.
        The proteomics of N-terminal methionine cleavage.
        Mol. Cell. Proteomics. 2006; 5: 2336-2349
        • Lee J.W.
        • Devanarayan V.
        • Barrett Y.C.
        • Weiner R.
        • Allinson J.
        • Fountain S.
        • Keller S.
        • Weinryb I.
        • Green M.
        • Duan L.
        • Rogers J.A.
        • Millham R.
        • O'Brien P.J.
        • Sailstad J.
        • Khan M.
        • Ray C.
        • Wagner J.A.
        Fit-for-purpose method development and validation for successful biomarker measurement.
        Pharm. Res. 2006; 23: 312-328
        • Grant R.P.
        • Hoofnagle A.N.
        From lost in translation to paradise found: enabling protein biomarker method transfer by mass spectrometry.
        Clin. Chem. 2014; 60: 941-944
        • Kuhn E.
        • Wu J.
        • Karl J.
        • Liao H.
        • Zolg W.
        • Guild B.
        Quantification of C-reactive protein in the serum of patients with rheumatoid arthritis using multiple reaction monitoring mass spectrometry and 13C-labeled peptide standards.
        Proteomics. 2004; 4: 1175-1186
        • Vallabh S.
        Prion Alliance sponsors MGH research study.
        PrionAlliance.org. 2017;
        • Tabrizi S.J.
        • Leavitt B.R.
        • Landwehrmeyer G.B.
        • Wild E.J.
        • Saft C.
        • Barker R.A.
        • Blair N.F.
        • Craufurd D.
        • Priller J.
        • Rickards H.
        • Rosser A.
        • Kordasiewicz H.B.
        • Czech C.
        • Swayze E.E.
        • Norris D.A.
        • Baumann T.
        • Gerlach I.
        • Schobel S.A.
        • Paz E.
        • Smith A.V.
        • Bennett C.F.
        • Lane R.M.
        Targeting huntingtin expression in patients with Huntington's disease.
        N. Engl. J. Med. 2019; 380: 2307-2316
        • Meredith J.E.
        • Sankaranarayanan S.
        • Guss V.
        • Lanzetti A.J.
        • Berisha F.
        • Neely R.J.
        • Slemmon J.R.
        • Portelius E.
        • Zetterberg H.
        • Blennow K.
        • Soares H.
        • Ahlijanian M.
        • Albright C.F.
        Characterization of novel CSF Tau and ptau biomarkers for Alzheimer's disease.
        PloS One. 2013; 8: e76523
        • Sato C.
        • Barthélemy N.R.
        • Mawuenyega K.G.
        • Patterson B.W.
        • Gordon B.A.
        • Jockel-Balsarotti J.
        • Sullivan M.
        • Crisp M.J.
        • Kasten T.
        • Kirmess K.M.
        • Kanaan N.M.
        • Yarasheski K.E.
        • Baker-Nigh A.
        • Benzinger T.L.S.
        • Miller T.M.
        • Karch C.M.
        • Bateman R.J.
        Tau kinetics in neurons and the human central nervous system.
        Neuron. 2018; 97: 1284-1298.e7
        • McCudden C.R.
        • Brooks J.
        • Figurado P.
        • Bourque P.R.
        Cerebrospinal fluid total protein reference intervals derived from 20 years of patient data.
        Clin. Chem. 2017; 63: 1856-1865
        • Prusiner S.B.
        • Groth D.
        • Serban A.
        • Stahl N.
        • Gabizon R.
        Attempts to restore scrapie prion infectivity after exposure to protein denaturants.
        Proc. Natl. Acad. Sci. U.S.A. 1993; 90: 2793-2797
        • Kim C.
        • Haldiman T.
        • Cohen Y.
        • Chen W.
        • Blevins J.
        • Sy M.-S.
        • Cohen M.
        • Safar J.G.
        Protease-sensitive conformers in broad spectrum of distinct PrPSc structures in sporadic Creutzfeldt-Jakob disease are indicator of progression rate.
        PLoS Pathog. 2011; 7: e1002242
        • Kim C.
        • Haldiman T.
        • Surewicz K.
        • Cohen Y.
        • Chen W.
        • Blevins J.
        • Sy M.-S.
        • Cohen M.
        • Kong Q.
        • Telling G.C.
        • Surewicz W.K.
        • Safar J.G.
        Small protease sensitive oligomers of PrPSc in distinct human prions determine conversion rate of PrP(C).
        PLoS Pathog. 2012; 8: e1002835
        • Cescatti M.
        • Saverioni D.
        • Capellari S.
        • Tagliavini F.
        • Kitamoto T.
        • Ironside J.
        • Giese A.
        • Parchi P.
        Analysis of conformational stability of abnormal prion protein aggregates across the spectrum of Creutzfeldt-Jakob disease prions.
        J. Virol. 2016; 90: 6244-6254
        • Greene R.F.
        • Pace C.N.
        Urea and guanidine hydrochloride denaturation of ribonuclease, lysozyme, alpha-chymotrypsin, and beta-lactoglobulin.
        J. Biol. Chem. 1974; 249: 5388-5393
        • Hoofnagle A.N.
        • Wener M.H.
        Serum thyroglobulin: a model of immunoassay imperfection.
        Clin Lab Int. 2006; 8: 12-14
        • Whiteaker J.R.
        • Halusa G.N.
        • Hoofnagle A.N.
        • Sharma V.
        • MacLean B.
        • Yan P.
        • Wrobel J.A.
        • Kennedy J.
        • Mani D.R.
        • Zimmerman L.J.
        • Meyer M.R.
        • Mesri M.
        • Rodriguez H.
        CPTAC Assay Portal: a repository of targeted proteomic assays.
        Nat. Methods. 2014; 11: 703-704
      3. Food and Drug Administration. Bioanalytical Method Validation.
        Guidance for Industry. 2018; : 1239
        • Sharma V.
        • Eckels J.
        • Schilling B.
        • Ludwig C.
        • Jaffe J.D.
        • MacCoss M.J.
        • MacLean B.
        Panorama public: a public repository for quantitative data sets processed in Skyline.
        Mol. Cell. Proteomics MCP. 2018; 17: 1239-1244