Alzheimer’s (AD) and Parkinson’s disease (PD) are neurodegenerative disorders characterized by an accumulation of protein aggregates in the brain. AD is the most common cause of dementia and presents with impairment of memory and cognition. PD results from the loss of dopaminergic neurons which results in motor symptoms such as bradykinesia, rest tremor and/or rigidity. In AD and PD neuropathological investigation and genetic association indicate dysfunctional proteostasis as a pathological feature. Proteostasis is maintained by autophagy and the endosomal-lysosomal system and by the ubiquitin-proteasome system. We aimed to investigate new potential cerebrospinal fluid (CSF) biomarkers by targeting proteins involved in proteostasis. Measurable changes in protein concentrations in CSF might reflect altered proteostasis in neurodegenerative diseases. We identified and targeted 50 peptides from 18 proteins by combining solid-phase extraction (SPE) and parallel reaction monitoring mass spectrometry (PRM MS). The CSF concentration of these proteins were measured in four cross-sectional studies including subjects with AD (N = 61), PD (N = 21), prodromal AD (N = 10), stable mild cognitive impairment (stable MCI; N = 15), as well as controls (N = 68). Two pilot studies (Study 1 and 2) showed increased concentrations of several proteins in subjects with biochemical evidence of AD pathology compared to biochemically characterized normal subjects. A study (Study 3) including clinically characterized subjects instead showed decreased CSF concentration of several proteins in PD compared to prodromal AD. A follow up cohort (Study 4) was analyzed, also including clinically characterized subjects. Again decreased CSF concentrations were identified in PD compared to controls and AD. Proteins identified with significant different protein concentrations in the studies were AP2B1, C9, CTSB, CTSF, GM2A, LAMP1, LAMP2, TCN2, and ubiquitin. Proteins with repeatedly identified significant differences in concentration, in more than one study, were AP2B1, CTSB, CTSF, GM2A, and ubiquitin. No differences in CSF concentration of the investigated proteins were identified in clinically characterized cases with AD compared to controls suggesting if such differences exist, compared to PD, are minor. Further studies are need to investigate whether the promising protein candidates identified herein might serve as potential CSF biomarkers in PD.

Protein and Peptide Standards: Crude stable isotope-labeled peptides, labeled with C-terminal 13C/15N-Lys or 13C/15N-Arg and modified by Cys carbamidomethylation, were purchased from JPT Peptide Technologies GmbH (SpikeTides L; Berlin, Germany) and Thermo Fisher Scientific Inc. (FasTrack 1; Waltham, MA, USA). Stock solutions of the peptides were prepared by dissolving SpikeTides L peptides in 10% acetonitrile and FasTrack 1 peptides in H2O with the exception of C9_497-508 which was dissolved in 23% acetonitrile and 0.76% formic acid. All peptides were frozen and stored at –20° C. Bovine serum albumin (BSA; full length, average mass 66430 Da; 100% purity by agarose electrophoresis; Sigma-Aldrich Co. Saint Louis, MO, USA) and uniformly labeled 13C-ubiquitin (average mass 8940 Da; >90% protein purity by SDS electrophoresis and >98% isotope enrichment purity; Silantes, GmbH, München, Germany) were dissolved in H2O and stored at –20° C and –80° C, respectively. An internal standard mixture of stable isotope-labeled peptides, BSA and 13C-ubiquitin was prepared in 50 mM NH4HCO3 for the addition to CSF samples. The standard solution was aliquoted, frozen and stored at −80 °C. A reverse calibration curve was prepared by serially diluting a mixture of stable isotope-labeled peptides and 13C-ubiquitin. Following dilution an equal concentration of BSA was added to each dilution. The reverse calibration curve dilutions were frozen and stored at −80 °C. Sample Digestion and SPE: Digestion and SPE were performed as described previously [1]. CSF, 100 µL, was mixed with 25 µL of a stable isotope-labeled mixture containing the corresponding heavy labeled peptides (JPT Peptide Technologies GmbH and Thermo Fisher Scientific Inc.) as well as uniformly labeled 13C-ubiquitin (Silantes, GmbH), full length bovine serum albumin (Sigma-Aldrich Co.), and a heavy labeled bovine serum albumin peptide (Thermo Fisher Scientific Inc.). Samples were reduced and alkylated by the addition of 25 µL 30 mM 1,4-dithiothreitol (Sigma-Aldrich Co.) in 50 mM NH4HCO3 and shaking incubation at +60° C for 30 min, incubation 30 min at room temperature, and addition of 25 µL 70 mM iodoacetamide (Sigma-Aldrich Co.) in 50 mM NH4HCO3 and incubation shaking at room temperature in dark for 30 min. Digestion was performed for approximately 18 h at +37° C following the addition of 25 µL sequencing grade modified trypsin (0.08 g/L, Promega Co., Madison, WI, USA) dissolved in 50 mM NH4HCO3. Digestion was ended by the addition of 10% trifluoroacetic acid. SPE was performed using Oasis HLB 96-well µElution Plates (2 mg sorbent and 30 µm particle size; Waters Co., Milford, MA, USA). The wells of the plate were conditioned by 2×300 µL methanol and equilibrated with 2×300 µL H2O. The digested samples were loaded followed by a wash of 2×300 µL H2O. Finally the samples were eluted in 2×100 µL methanol and dried by vacuum centrifugation. The samples were frozen and stored at −80° C. PRM MS: Samples were dissolved in 50 µL 50 mM NH4HCO3 shaking at room temperature for 1 h. Dissolved samples were injected and subjected to HPLC on a Dionex UltiMate 3000 standard-LC system (Thermo Fisher Scientific Inc.). 20 µL of the samples was injected in studies 1 and 3, and 10 µL in studies 2 and 4. The samples were separated on a Hypersil GOLD HPLC C18 column (length 200 mm; inner diameter 2.1 mm; particle size 1.9 μm; Thermo Fisher Scientific Inc.) with mobile phases being A: 0.1% formic acid in H2O (v/v); and B: 0.1% formic acid and 84% acetonitrile in H2O (v/v). Separation was performed at a flow rate of 100 µL/min at +40° C with a gradient going from 2 to 6% B over two minutes, 6 to 20% B over four minutes and finally 20 to 40% over 42 min. A PRM method was developed on a quadrupole-orbitrap hybrid mass spectrometer (Q Exactive, Thermo Fisher Scientific Inc.). Source settings were as follow: HESI-II ionization probe (Thermo Fisher Scientific Inc.), a spray voltage of +4.1 kV; a heater temperature of +300° C; a capillary transfer tube temperature of +320° C; a sheath gas flow rate of 25; and an auxiliary gas flow rate of 10. Single microscans were acquired in PRM mode using an isolation window of m/z 3, a resolution setting of 70 k, an AGC target of 1×106, a maximum injection time of 300 ms and fragmentation with beam type collision induced dissociation (or “higher energy collision induced dissociation” (HCD) [2]). The normalized collision energy, used for fragmentation with HCD, was optimized for each peptide by direct infusion of stable isotope-labeled peptide in solution and varying the energy while monitoring the intensity of the product ion in relation to the precursor ion. Peptide acquisition was scheduled within a two minutes retention time window. Precision and Accuracy: A quality control (QC) sample was prepared by pooling human CSF. The QC was analyzed in replicates of eight on five different occasions. The repeatability and intermediate precision were calculated according to ISO 5725-2 [3]. Using the QC reverse calibration curves were generated. Duplicate curves on three different occasions were analyzed. The mean of the duplicates followed by the mean of the three occasions were used to calculate the LOQ. The limit of quantification (LOQ) was determined by calculating the concentrations with a relative error of ≤20% of calculated to nominal concentration. The LOQ was calculated using least squares linear regression in GraphPad Prism v.7.02 (GraphPad Software, Inc., La Jolla, CA, USA). Data Analysis: Skyline v3.6 [4] was used for peak detection and peak area integration. Skyline was set to pick and integrate [M+H]+ y-ions from a data independent acquisition method with a fixed isolation window of m/z 3. Evaluation of which product ions to include in the integration was performed by manual inspection of MS/MS spectra, inspecting individually extracted product ion chromatograms in Skyline and by an in-house software comparing product ion ratios. The evaluation enabled exclusion of product ions not reproducibly detected and product ions potentially afflicted by interferences. The ratio of the sum of the product ion areas of tryptic to isotope-labeled peptide was used for quantification. References: 1. Brinkmalm G, Sjödin S, Simonsen AH, Hasselbalch SG, Zetterberg H, Brinkmalm A, Blennow K: A Parallel Reaction Monitoring Mass Spectrometric Method for Analysis of Potential CSF Biomarkers for Alzheimer's Disease. Proteomics Clinical Applications 2018, 12:1700131-n/a. 2. de Graaf EL, Altelaar AF, van Breukelen B, Mohammed S, Heck AJ: Improving SRM assay development: a global comparison between triple quadrupole, ion trap, and higher energy CID peptide fragmentation spectra. J Proteome Res 2011, 10:4334-4341. 3. International Organization for Standardization: ISO 5725-2:1994 Accuracy (trueness and precision) of measurement methods and results — Part 2: Basic method for the determination of repeatability and reproducibility of a standard measurement method. International Organization for Standardization; 1994. 4. MacLean B, Tomazela DM, Shulman N, Chambers M, Finney GL, Frewen B, Kern R, Tabb DL, Liebler DC, MacCoss MJ: Skyline: An open source document editor for creating and analyzing targeted proteomics experiments. Bioinformatics 2010, 26:966-968.

Participants: The participants included subjects biochemically characterized as having an AD CSF biomarker profile indicating AD pathology or having a normal profile. The biomarker profile was determined by measuring the CSF concentrations of amyloid β 1-42, total tau protein and tau phosphorylated at Thr181. These samples were collected after being analyzed in clinical routine at the Clinical Neurochemistry Laboratory, Mölndal, Sweden. Study 1 (AD, N = 7 and controls, N = 10) and 2 (AD, N = 12 and controls, N = 14) consisted of participants solely selected based on their biomarker profile. Study 3 consisted of subjects consecutively enrolled at the Center of Memory Disturbances of the University of Perugia. The cohort included participants diagnosed with AD dementia (N = 6) according to the National Institute on Aging-Alzheimer’s Association criteria [5, 6], patients diagnosed as PD (N = 10) according to the National Institute of Neurological Disorders and Stroke (NINDS) diagnostic criteria [7] and subjects diagnosed as mild cognitive impairment (MCI; N = 25) according to Petersen’s criteria [8]. Ten out of 25 developed AD (prodromal AD), while 15 remained stable for a median of two years follow-up (stable MCI). Study 4 consisted of a subset of participants from the Swedish BioFINDER study (www.biofinder.se) recruited at Skåne University Hospital, Sweden. The study included participants with AD dementia (N = 36), according to the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association [9] and participants diagnosed with PD (N = 11), according to the NINDS diagnostic criteria [7], as well as cognitively healthy volunteers (controls, N = 44). In studies 3 and 4 AD subjects and controls also had a CSF biomarker profile indicating AD pathology or being normal, respectively. All of the sample collection and procedures were performed following the Helsinki Declaration, after approval by local ethical committees. QC Sample: A QC sample was prepared by pooling human CSF. The QC sample was used to determine and monitor methodological precision and used for creating reverse calibration curves to determine the LOQ. Sample Collection: Human CSF was collected by lumbar puncture in polypropylene tubes, centrifuged at room temperature [10] or +4° C [11] for 10 min at 2000×g and aliquoted. Tubes were frozen at –80° C pending analysis. CSF amyloid β 1-42, total tau protein and tau phosphorylated at Thr181 concentrations were assessed by using commercially available enzyme-linked immunosorbent assays for diagnostic purpose (INNOTEST βAmyloid1–42, hTAU-Ag and p-TAU181Ag; Fujirebio Europe, Ghent, Belgium), according to the manufacturer’s instructions. References: 5. Albert MS, DeKosky ST, Dickson D, Dubois B, Feldman HH, Fox NC, Gamst A, Holtzman DM, Jagust WJ, Petersen RC, et al: The diagnosis of mild cognitive impairment due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011, 7:270-279. 6. McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Jr., Kawas CH, Klunk WE, Koroshetz WJ, Manly JJ, Mayeux R, et al: The diagnosis of dementia due to Alzheimer's disease: recommendations from the National Institute on Aging-Alzheimer's Association workgroups on diagnostic guidelines for Alzheimer's disease. Alzheimers Dement 2011, 7:263-269. 7. Gelb DJ, Oliver E, Gilman S: Diagnostic criteria for Parkinson disease. Arch Neurol 1999, 56:33-39. 8. Petersen RC: Mild cognitive impairment as a diagnostic entity. Journal of Internal Medicine 2004, 256:183-194. 9. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM: Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer's Disease. Neurology 1984, 34:939-944. 10. Chiasserini D, Biscetti L, Farotti L, Eusebi P, Salvadori N, Lisetti V, Baschieri F, Chipi E, Frattini G, Stoops E, et al: Performance evaluation of an automated ELISA system for Alzheimer's disease detection in clinical routine. Journal of Alzheimer's Disease 2016, 54:55-67. 11. Hall S, Surova Y, Öhrfelt A, Blennow K, Zetterberg H, Hansson O: Longitudinal Measurements of Cerebrospinal Fluid Biomarkers in Parkinson's Disease. Movement Disorders 2016, 31:898-905.