Whole-Proteome Peptide Microarrays for Profiling Autoantibody

Subscriber access provided by Kansas State University Libraries

Article

Whole-proteome peptide microarrays for profiling autoantibody repertoires within multiple sclerosis and narcolepsy. Arash Zandian, Björn Forsström, Anna Häggmark-Manberg, Jochen M Schwenk, Mathias Uhlén, Peter Nilsson, and Burcu Ayoglu J. Proteome Res., Just Accepted Manuscript • DOI: 10.1021/acs.jproteome.6b00916 • Publication Date (Web): 25 Jan 2017 Downloaded from http://pubs.acs.org on January 26, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Proteome Research is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of Proteome Research

Running title: Whole-proteome peptide microarrays

Whole-proteome peptide microarrays for profiling autoantibody repertoires within multiple sclerosis and narcolepsy

Arash Zandian1, Björn Forsström1, Anna Häggmark-Månberg1, Jochen M. Schwenk1, Mathias Uhlén1, Peter Nilsson1 and Burcu Ayoglu1,§,* 1

Address: Affinity Proteomics, SciLifeLab, School of Biotechnology, KTH - Royal Institute of Technology, Stockholm, Sweden § Current address: Department of Medicine, Division of Immunology and Rheumatology, Stanford University School of Medicine, Stanford, California, USA *Corresponding author: [email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT The underlying molecular mechanisms of autoimmune diseases are poorly understood. In order to unravel the autoimmune processes across diseases, comprehensive and unbiased analyses of proteins targets recognized by the adaptive immune system are needed. Here, we present an approach starting from high-density peptide arrays to characterize autoantibody repertoires and to identify new autoantigens. A set of ten plasma and serum samples from subjects with multiple sclerosis, narcolepsy and without any disease diagnosis were profiled on a peptide array representing the whole proteome, hosting 2.2 million 12-mer peptides with a six amino acid lateral shift. Based on the IgG reactivities found on these whole-proteome peptide microarrays, a set of 23 samples was then studied on a targeted array with 174,000 12-mer peptides of single amino acid lateral shift. Finally, verification of IgG reactivities were conducted with a larger sample set (n=448) using the bead-based peptide microarrays. The presented workflow employed three different peptide microarray formats to discover and resolve the epitopes of human autoantibodies, and revealed two potentially new autoantigens: MAP3K7 in MS and NRXN1 in narcolepsy. The presented strategy provides insights into antibody repertoire reactivity at a peptide level and may accelerate the discovery and validation of autoantigens in human diseases.

KEYWORDS Peptide microarrays, autoantibody profiling, epitope mapping, narcolepsy, multiple sclerosis

1


Page 2 of 42

Page 3 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Peptide microarrays are highly multiplexed assay platforms for high-resolution mapping of antibody binding sites against thousands of peptides in parallel. Following the early demonstrations of the possibility of in situ synthesis of peptides (1), the peptide microarray technology has been advanced by implementation of combinatorial synthesis based on photolithography (2), followed by replacement of photomasks with digital micromirrors (3, 4) and by development of mask-free photolithographic methods (5, 6). High-density peptide microarrays are usually designed based on a tiling strategy in order to represent each target protein sequence by a set of overlapping short peptides. Such peptide microarrays have so far been primarily applied for characterization of linear epitopes of antibodies based on available protein sequence information (7-10). In recent years, the potential of high-density peptide microarrays for a detailed analysis of antibody profiles in human serology has also been recognized. A handful of studies have utilized tiled peptide microarrays for the analysis of antibody repertoire in various autoimmune conditions (11-14), as well as in infectious diseases (15-18). In addition to tiled arrays, the utility of high-density arrays of random sequence peptides (19) has been shown for the analysis of cancer-associated autoantibodies in sera (20). Multiple sclerosis (MS) is a chronic and inflammatory neurological disease, resulting in demyelination and axonal loss in the central nervous system. MS is generally considered to be a disease with an autoimmune character, where an immune response towards self-proteins is involved. Even though T-cells are assumed to be the central mediators of the autoimmune response in MS, B-cells and autoantibodies produced by terminally differentiated B-cells are considered to be relevant team players (21, 22), as indicated by studies demonstrating the decline of disease activity following the depletion of peripheral B-cells in MS patients (23). Despite the growing evidence that autoantibodies play a significant role in the pathogenesis of MS, our understanding of autoimmune targets in MS is still constrained by our inability to completely characterize the molecular targets of the autoimmune mechanisms. In line with this, high-density peptide microarrays representing the entire, or a significant portion of the human proteome, are powerful tools for an unbiased exploration of the autoantibody repertoire in body fluids on the proteome-scale. There are a handful of studies on the use of peptide microarrays for serum antibody profiling within adult-onset and pediatric-onset MS (24-26). In all of these studies,

2



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 42

arrays of peptides and protein were generated, where a few proteins of potential relevance for inflammatory

demyelinating

diseases,

such

as

myelin

basic

protein

(MBP),

myelin/oligodendrocyte glycoprotein peptide (MOG) or myelin-associated oligodendrocyte basic protein (MOBP), were represented as peptides. Autoantibody reactivities towards this limited set of selected peptides were analyzed also in CSF (27) or matched CSF-serum samples of MS patients (28). Very recently, two studies reported the use of arrays consisting of almost 4,000 (12) and 9,000 (29) peptides representing human and viral antigens selected from literature for the analysis of serum and CSF antibody repertoire within MS. These two recent studies are the largest studies so far utilizing high-density peptide microarrays for profiling the autoantibody repertoire in MS. Even so, these efforts so far have been built on a targeted analysis strategy in which arrays of peptides representing a very limited portion of the human proteome has been represented on the arrays. Narcolepsy is another chronic, neurological disease manifesting itself as excessive daytime sleepiness, muscle paralysis (cataplexy) and dream-like hallucinations. It is caused by a selective loss of specific hypothalamic neurons producing the peptide neurotransmitters, orexins, also called hypocretins, which regulate the sleep and arousal states. Based on its strong genetic association with certain immune system gene polymorphisms, namely with human leukocyte antigen (HLA) and T cell receptor (TCR) polymorphisms (30), narcolepsy is considered as an immune-mediated disease with a possible autoimmune component due to molecular mimicry (31, 32). In recent years, epidemiological observations from studies in China (33), Scandinavia (34-36), and other European countries (37) revealed that susceptibility to narcolepsy might be associated with influenza A (H1N1) viral infections and certain H1N1 vaccine preparations. These observations strengthened the hypothesis that immune cells which are activated by H1N1 epitopes could contribute to an autoimmune reaction due to mimicry between H1N1 epitopes and self-proteins (38). However, studies dedicated to identify targets of immune cells have so far been scarce and the target autoantigens in vaccine-associated narcolepsy are yet to be identified. In a recent study, we utilized arrays of human protein fragments for an unbiased profiling of the serum autoantibody repertoire within vaccine-associated narcolepsy (39). Yet, high-density peptide microarrays have been utilized so far only in a single study, where arrays representing the influenza A (H1N1) viral proteome were used for profiling H1N1-specific serum antibody reactivities before and after H1N1 vaccination or H1N1 infection (40). Efforts dedicated for

3


Page 5 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


an unbiased exploration of the serum autoantibody reactivity on proteome-scale peptide microarrays have so far been missing within narcolepsy. In this work, we used high-density peptide microarrays of 2.2 million overlapping peptides representing the protein products of all human protein-coding genes. We analyzed serum and plasma samples from multiple sclerosis, as well as flu vaccine-associated narcoleptic patients and matched control individuals using an unbiased and untargeted strategy. We aimed to identify the targets of the circulating autoreactive antibodies in these sample sets and to fine-map the epitopes of the autoantibodies, as well as to map these reactivities back to their corresponding proteins. We utilized a high-throughput bead-based array format to create peptide bead microarrays for verification of the identified reactivities in larger sample collections. To the best of our knowledge, this work is the first study, which explores the circulating autoantibody repertoire of patients with multiple sclerosis and narcolepsy on peptide microarrays representing the entire human proteome. While giving a unique possibility for a global analysis of the autoantibody repertoire in blood-derived samples and its diversity, the presented work also reports a handful of interesting and new potential autoimmune targets within the context of MS and narcolepsy, which should be assessed further in larger sample collections for their pathogenic relevance in these two neurological diseases.

4



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

MATERIALS AND METHODS

Design of the planar peptide microarrays The whole-proteome peptide microarrays, covering all human proteins, were designed based on the protein set (version 37.1) available by the human Consensus Coding Sequence (CCDS) project (41). These arrays contained 2,198,610 12-mer peptides with a lateral shift of six amino acids. These 2.2 million peptides represented a total of 17,958 human protein-coding genes and they were randomly distributed on the array in a checkerboard pattern. Peptides that revealed reactivity in serum samples profiled on the wholeproteome microarrays were selected for the design of the targeted arrays. These targeted arrays contained a total of 174,235 peptides with a single amino acid lateral shift and consisted of 12 identical subarrays The arrays contained 12-mer peptides that represented 4,513 different human protein-coding genes. In addition to the peptides selected from the whole-proteome arrays, the targeted arrays contained peptides representing known autoantigens from the literature and in-house unpublished data. (Figure 1).

Synthesis of the planar peptide microarrays The peptide arrays were generated based on the principle of in situ photolithographic synthesis (4) in a Roche-Nimblegen Maskless Array Synthesizer (MAS) as described previously (8, 10). As solid support, microscope slides (1 x 3 inch) were utilized which were amino functionalized with 6-amino hexanoic acid as a spacer and amino acid derivatives carrying a photosensitive 2-(2-nitrophenyl)propyl-oxycarbonyl-group (NPPOC) at the α-amino function (42). Coupling of the amino acids to the array was done by a preactivation of 30 mM amino acid, 30 mM 1-hydroxy-benzotriazole / 2-(1H-Benzotriazole-1yl)-1,1,3,3-tetramethylaminium hexafluorophosphate and 60 mM ethyldiisopropylamine, which were in dimethylformamide for a total of 5-7 minutes before flushing the substrate for 5 min. The substrates were then washed with N-methyl-2-pyrrolidone (NMP) and a sitespecific cleavage of the NPPOC group was made by irradiation of an image created by a Digital Micromirror Device (Texas Instruments, SXGA+ graphics format), projecting light at 365 nm wavelength to a selection of approximately 1.4 million features of 13 x 13 µm2 size at

5


Page 6 of 42

Page 7 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


a total dose of approximately 10 J/cm2 in NMP. Final treatment of the slides with trifluoroacetic acid/water/triisopropylsilane for 30 min removed the side-chain protection of the amino acids.

Samples The biological specimen used in this study were blood-derived serum/plasma samples and originated from two different sample collections of individuals with multiple sclerosis and narcolepsy (Table 1). The narcolepsy sample set, collected both in Sweden and Finland, included sera from control individuals who have been vaccinated with Pandemrix®, but with no onset of narcolepsy, and from vaccinated patients who developed narcolepsy. The MS-related sample collection consisted of EDTA plasma samples and was part of a cohort collected during routine neurological diagnostic work-up at the neurology clinic of Karolinska University Hospital Stockholm, Sweden. In addition to patients with various MS subtypes, samples from patients with a single demyelinating event, referred to as clinically isolated syndrome (CIS), as well as individuals with other neurological diseases (OND) and ONDs with signs of inflammation (iOND) were included. Original study enrolment for both cohorts followed the recommendations of the Declaration of Helsinki and the studies were approved by the Ethics Committee of the Karolinska Institute. Further information on both of the sample collections is provided elsewhere (39, 43).

Assays on planar peptide microarrays As described previously by Forsström et al.(10), de-protected peptide microarray slides were first washed twice with TBSTT (20 mM Tris, 0.9% NaCl (w/v), pH 7.4, 0.1% Tween-20 (v/v), 0.4% Triton X-100 (v/v)) in a LockMailer slide jar (Sigma) by inverting the jar for 2 min. Then, the slides were transferred to new slide jars and washed twice in TBS (20 mM Tris, 0.9% NaCl (w/v), pH 7.4) for 2 min by inverting the jar, then rinsed three times with de-ionized water and dried. Peptide microarray reaction wells (subarrays) were created by attaching mixer masks (Roche NimbleGen Inc.) to the slide. Serum/plasma samples, which were diluted 1:200 in assay buffer (10 mM Tris, 0.45% NaCl (w/v), pH 7.4, 0.5 % alkali soluble casein (w/v) (Novagen, EMD chemicals)), were injected into the mixing compartments. The slides were incubated overnight in a NimbleGen Hybridization Station (Roche NimbleGen Inc.) according to the manufacturer’s instructions.

6



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

After the primary incubation, the slides were submerged in TBSTT and the mixer masks were removed. The slides were washed twice with TBSTT, followed by two washes with TBS, as described above. For the assay read-out, DyLight 649-conjugated anti-human IgG (Jackson ImmunoResearch) was diluted to 0.15 µg/ml in binding buffer in LockMailer jars and the slides were incubated 3 hr on a shaking table. Then, as described above, the slides were washed twice with TBSTT, twice with TBS, and rinsed three times in de-ionized water and dried. Finally, the peptide microarray slides were scanned using a NimbleGen MS200 scanner (Roche NimbleGen Inc.) at a resolution of 2 µm.

Image and data analysis of the planar peptide microarrays The images generated by the NimbleGen MS200 scanner were first aligned using the NimbleScan 2 software (Roche NimbleGen Inc.) and a feature report of all the array spots and corresponding fluorescence intensities was exported. This feature report was imported to MATLAB 7.14 (The Mathworks Inc.) for extraction of fluorescence intensities from all pixels on the slides, resulting in median intensities for all spots, median intensities for the local background, as well as the standard deviation of the spot intensities. The resulting spot intensities were imported into R, a program for statistical computing and graphics (44). To distinguish the spots from any background noise, spots were filtered by applying cut-off values based on the local background of each spot, which was set as 2 times the local background intensity. The local background was based on the median intensity in the non-synthesized blank spots around the spot of interest, which could be utilized due to the checkerboard synthesis pattern of the array. In addition, possible artifacts due to the spot morphology were addressed by calculating the variation of the pixel intensities within each spot and by setting a coefficient of variation threshold of maximum 50%, assuming intensity ranges with a large variation could indicate the presence of morphological artifacts. This tailored pre-filtering strategy reduced the complexity of the datasets, as well as minimized the effect of possible false positives, e.g. due to dust particles or zone variations within array or subarray (Figures S-1 & S-2). Data was dichotomized where spots with signal intensities passing the above-mentioned filtration criteria were assigned a binary value of 1 and considered as reactive. Spots not passing the filtration were assigned a binary value of 0 and considered as not reactive. For the analysis of the whole-proteome peptide microarray data, the dichotomized data was used to calculate reactivity frequencies and thereby find groupenriched peptide sequences for further analysis. For the analysis of data generated on the 7


Page 8 of 42

Page 9 of 42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


targeted peptide microarrays, the binary values were used to perform Fisher’s exact test per peptide, with the summed antibody reactivities for the sample groups. Peptides with significant differences (p

Whole-Proteome Peptide Microarrays for Profiling Autoantibody

Recommend Documents