Organization

The Executive Management is responsible for the overall strategies of the consortium and for managing the overall program:

Prof Mathias Uhlen, Program Director
Kalle von Feilitzen, IT, web and data handling
Dr Cecilia Lindskog, Tissue section
Dr Jan Mulder, Brain section
Dr Linn Fagerberg, Bioinformatics and integrative omics
Prof Emma Lundberg, Subcellular section
Dr Fredrik Edfors, Targeted proteomics andr plasma profiling
Prof Fredrik Ponten, Pathology section
Prof Sophia Hober, Protein science research
Prof Jens Nielsen, Metabolic section
Prof Peter Nilsson, Protein arrays technologies
Prof Jochen Schwenk, Plasma profiling
Dr Hanna Tegel, Protein factory
Dr Åsa Sivertsson, Annotation
Assoc Prof Adil Mardinoglu, Systems medicine
Dr Cheng Zhang, Single Cell profiling
Inger Åhlen, Administrative Coordinator

AlbaNova site, Stockholm

Antigen and Antibody Factory

Head: Dr Hanna Tegel

Responsibility: (i) Production of recombinant PrEST expression clones including cDNA synthesis, cloning, and plasmid purification. (ii) Production and purification of PrEST-proteins used for preparation of antigens and affinity columns. (iii) Management of immunization outsourcing. (iv) Generation of purified antibodies through affinity purification of polyclonal antisera. (v) Western blot (WB) analysis of antibodies approved in protein array analysis. (vi) WB antibody validation using over-expression lysates. (vii) Storage and distribution of antigens and antibodies within the program. (viii) Production of the Human Secretome in mammalian cells

Description: PrEST regions are first amplified with RT-PCR from total RNA template pools with specific oligonucleotide primers for each PrEST. Amplicons are automatically processed with solid phase restriction, and ligated into the plasmid vector pAff8c (Larsson, M. et al, 2000) where the human gene fragment is fused to a histidine tag and albumin binding protein (His6ABP). After transformation into E. coli Rosetta(DE3), inserts are verified by DNA sequencing to omit clones with mutations and approved clones are single cell streaked. Plasmids are collected from all purified clones for deposition in the clone library and glycerol stocks are prepared and used as starting material for protein production.

All proteins are expressed as His6ABP fusions in E. coli shake flask cultures upon induction with IPTG. A fully automated protein purification system has been developed to allow for purifications of up to 60 cell lysates at a time. One-step purification is enabled by the hexahistidine affinity tag and metal affinity chromatography (IMAC) and performed under denaturing conditions. After evaluation of protein concentration and purity, the molecular weight of the PrEST proteins is determined by mass spectrometry as a final quality control. The purified proteins are then used to prepare antigens and affinity columns with PrEST-ligands. In addition, affinity resin with His6ABP-ligand is also produced.

After immunization of the antigens the polyclonal antisera, generated together with collaborative partners, are carefully purified in a three-step fashion consisting of: depletion of unwanted specificity, capture of wanted specificity and a final buffer exchange step. A manual process using gravity-flow columns carries out depletion of antibodies with unwanted specificity. The following steps are performed on the ÄKTAxpress chromatography system enabling a high-throughput semi-automated process where captured antibodies are eluted by a low pH glycine buffer and automatically loaded onto a desalting column for buffer exchange. Antibodies are supplemented with 50% glycerol and 0.02% sodium azide for long-term storage at -20°C. The binding specificity of all antibodies is determined on protein microarrays to certify that only antibodies with high specificity and low background binding are approved for immunohistochemistry analysis. All approved antibodies are further analyzed in a high-throughput WB platform using protein lysates from human cell lines (RT-4 and U-251 MG), human plasma depleted of IgG and HSA and whole tissue lysates from human liver and tonsil. A selection of the published antibodies, initially scored as uncertain in the standard WB panel, have been revalidated in a WB set-up comprising an over-expression lysate (VERIFY Tagged Antigen™, OriGene Technologies, Rockville, MD) as a positive control.

The group is also responsible for production of full-length proteins in mammalian cells within the Human Secretome project. The overall aim of the Human Secretome project is to produce, purify and characterize the majority of all human secreted proteins to build up a resource of reagents for drug discovery and development. A high-throughput protein production workflow, using Chinese Hamster Ovary (CHO) cells as host cells, has been set up and to purify the produced proteins a mild and effective one-step affinity purification based on affinity chromatography has been developed.

ABP - Albumin Binding Protein IPTG - Isopropyl-B-D-Thiogalactopyranoside IMAC - Immobilized Metal Affinity Chromatography

Protein Science

Head: Prof Sophia Hober

Responsibility: Protein science research.

Description: The module coordinates and conducts research projects aiming to extend the scientific outcome of data generated within the Human Protein Atlas project as well as improving current methodology. Both biological and technical research projects based on the vast amount of data generated are performed to further explore the function, localization and interactions of human proteins.

SciLifeLab site, Stockholm

Protein Array Technologies

Head: Prof Peter Nilsson

Responsibility: To validate the specificity and selectivity of all purified HPA antibodies. To develop and utilize peptide, antigen and antibody based microarray methodologies for large scale analysis of body fluids in the context of biomarker discovery and autoantibody profiling.

Description: Methodology for microarray based analysis of antibody specificity has been developed, where all purified antibodies are analyzed on protein arrays with immobilized PrESTs. Each microarray is divided into 21 replicated subarrays with 384 PrESTs, enabling the analysis of 21 antibodies simultaneously. The antibodies are detected through a fluorescently labeled secondary antibody. A specificity plot is generated for each antibody, where the signal from the binding to its antigen is compared to the unspecific binding to all the other PrESTs. A dual color system is used in order to verify the presence of the spotted PrESTs. Several complementary microarray formats for systematic analysis of body fluids are being utilized and under constant development. The PrEST-arrays have been implemented for systematic antigen-based plasma profiling for the screening of new autoimmunity components. The antibody microarrays with the possibility for simultaneous analysis of large amounts of analytes with high sensitivity and the reverse phase serum microarrays which enable serum from very large patient cohorts to be analyzed simultaneously are both utilizing in-house produced planar microarrays. The main platform for systematic antibody-based plasma profiling, is although the suspension bead array format with capacity for multiplexing in two dimensions, enabling the simultaneous profiling of 384 antibodies on 384 samples, see Plasma profiling.

Cell Profiling

Head: Prof Emma Lundberg

Team leader: Dr Ulrika Axelsson

Responsibility: Creation of the HPA Cell Atlas: (i) Determination of the subcellular distribution of proteins using high-resolution confocal microscopy, (ii) validation of antibody specificity using gene editing and silencing technologies (iii) annotation and knowledge-based curation of subcellular distribution profiles in the Cell Atlas.

Description: The Cell Profiling group is responsible for determination of the spatiotemporal subcellular distribution of proteins in human cells. The distribution is systematically assessed at single-cell level using the antibodies generated within the Human Protein Atlas program. Each protein is studied in up to three cell lines, which are selected based on RNA expression levels from a panel of cell lines of different origin. Subcellular distribution is determined by indirect immunofluorescence followed by confocal microscopy. The resulting high-resolution images show the protein of interest as well as markers for the nucleus, microtubule cytoskeleton, and endoplasmic reticulum. This enables manual annotation of protein localization to one or more subcellular structures, as well as detection of cell-to-cell variability. The high-resolution confocal images, annotations, and gene expression data can be explored in an interactive gene-centric manner on the Cell Atlas. The Cell Profiling group uses different strategies for enhanced antibody validation. Most common is validation by independent antibodies, but we also perform genetic validation by knocking down the gene of interest using siRNA, and validation by comparative immunostaining in cells co-expressing a GFP-tagged recombinant version of the protein of interest.

Plasma Profiling

Head: Prof Jochen M. Schwenk

Responsibility: Utilizing HPA antibodies in a variety of multiplexed assay systems to profile human proteins in blood plasma.

Description: The Plasma Profiling group is responsible for developing and applying antibody-based assays for protein biomarker analysis in human plasma or other body fluids. The group has established high-throughput single-binder assays, where antibodies are immobilized on color-coded beads to create antibody arrays in suspension. These discovery arrays are composed of up to 384 antibodies to analyze up to 384 biotinylated and heat-treated body fluids at a time.
For target validation and quantification of proteins in blood, the group built a multiplexed workflow to select pairs of antibodies for capture and detection of proteins. These dual binder sandwich assays enable a specific and sensitive protein detection.
The Plasma Profiling group uses different means for validating antibodiey selectivity in plasma: This includes paired antibodies raised against different epitopes, sequential affinity capture assays, mass spectrometry detection of proteins enriched by antibodies from plasma, as well as the development of dual binder sandwich assays. The group also uses orthogonal methods, such as targeted mass spectrometry or other immunoassay platforms, as well as genetic data to assess the performance of the antibodies.

Bioinformatics and integrative omics

Head: Dr Linn Fagerberg

Responsibility: Analysis and integration of large-scale biological data using advanced bioinformatic methods focusing on: (i) transcriptomics sequencing of cells and tissues; (ii) the quantitative transcriptomics-based classification of the human proteome; and (iii) advancing personalized medicine to allow for the profiling of human health and disease based on multi-level omics strategies.

Description: The bioinformatics and integrative omics group is specialized in utilizing various methods for mining “big data”. The main focus is analyzing the human proteome and transcriptome using integrative approaches to perform a functional and spatial classification. We are also leading the integrative data analysis work within the SCAPIS SciLifeLab Wellness Profiling (S3WP) program, based on the Swedish CArdioPulmonary bioImage Study (SCAPIS). In the S3WP program, the participants are profiled based on a combination of classical clinical chemistry, advanced medical imaging and extensive omics profiling and include both healthy as well as disease cohorts.

Tissue cell type profiling

Head: Dr Lynn Butler

Responsibility: The development and application of computational-based analysis methods to profile cell type specificity of gene expression, within different tissue types. This data is presented in the Tissue Cell Type section of the Human Protein Atlas.

Description: The group use correlation based integrative co-expression analysis to mine information on gene expression at the cell type level, using unfractionated bulk RNAseq data. We apply our method to existing large datasets, sourced though resources such as the GTEx consortium.

Targeted Proteomics and Plasma Profiling

Head: Dr Fredrik Edfors

Responsibility: The group is using mass spectrometry to validate reagents generated within the scope of the Human Protein Atlas. This includes orthogonal antibody validation by Capture MS, as well as bottom-up proteomics analysis of samples of human origin.

Description: The group is focusing on the use of Stable Isotope-labeled Standard (SIS) Protein Fragments based on the PrEST sequences to precisely quantify and accurately quantify proteins across human tissues, cell-lines and in blood plasma. A wide variety of bottom-up proteomics techniques are used to generate targeted assays, including Selective Reaction Monitoring (SRM) and Data Independent Acquisition (DIA) strategies.

IT, web and data handling

Head: Kalle von Feilitzen

Responsibility: (i) To deliver custom made software solutions for all operations in the Human Protein Atlas project, (ii) to provide the collected data to the public via the Human Protein Atlas, (iii) to map and quantify RNA-seq data, and (iv) to initiate the analysis of human proteins by in silico selection and design of Protein Epitope Signature Tags (PrESTs).

Description: With the LIMS (Laboratory Information Management System) as the backbone, data is collected from each module in the pipeline. The protein expression profiles, RNA-seq data and raw data from the project is published on the Human Protein Atlas public web site through annual releases. The group is also involved in research performed in the project, such as data collection, data interpretation and visualization as well as statistical analysis.

Biomedicum site, Stockholm

Brain Profiling

Head: Dr Jan Mulder

Advisor: Prof Tomas Hökfelt

Responsibility: Generation of transcriptomics and antibody-based protein data for the Brain Atlas: (i) Validation of antibodies against human targets on rodent tissues, (ii) profiling the distribution of proteins in the developing, adult, and diseased nervous system, (iii) quantification, annotation, and presentation of whole brain protein distribution profiles, (iv) dissection and collection of brain samples (various species), (v) RNAsequencing of human brain and rodent tissues using the MGI DNBSEQ-T7 platform, (vi) 3D volume imaging of human and animal samples (iDISCO protocol).

Description: The brain is a complex organ from a functional and anatomical perspective. To capture this complexity, we complemented our standard human tissue and organ analysis with a more in-depth analysis of the central nervous system by including more brain regions and by investigating regional protein expression in different mammalian species. By visualizing protein distribution in the ‘small’ mouse brain we are able to provide a more complete overview of protein distribution in the mammalian brain including most brain regions and cell types. For this, HPA antibodies against proteins expressed in the mouse nervous system are validated on mouse brain tissue using immunofluorescence and results are compared to available expression data. Antibodies that pass validation are used to generate detailed protein distribution profiles using 20-30 coronal sections of the mouse brain with a 400 μm section interval covering all major brain nuclei. Whole slide immunofluorescence captured at 10x primary objective is analyzed and regional, cellular and subcellular protein distributions are quantified. Data and images are optimized for online publication.

Pig brain project: The pig brain transcriptomics project is a collaborative project between human protein atlas and the Lars Bolund institute of regenerative Medicine (Dr Yonglun Luo), BGI-Qingdao, China.

Human prefrontal cortex project: The human prefrontal cortex project is a collaboration with Prof Miklós Palkovits, Human Brain Tissue Bank Microdissection Laboratory of the Semmelweis University, Budapest, Hungary

Uppsala site

Tissue Profiling

Head: Dr Cecilia Lindskog

Responsibility: Generation of antibody-based protein data for the Tissue Atlas and the Pathology Atlas: (i) handling and processing of tissues (biobank material), (ii) handling and storage of antisera, (iii) testing of antibodies, (iv) immunohistochemical staining of tissues slides to be scanned, (iv) scanning of tissue slides and image processing, (v) validation of antibody target specificity, (vi) annotation and final approval of immunohistochemically stained tissues, (vii) determination of antibody reliability based on enhanced antibody validation strategies, (viii) generation of knowledge-based protein expression profiles, and (viiii) prepare the data for release on the Tissue Atlas and Pathology Atlas.

Description: Formalin fixed, paraffin embedded tissue specimens are collected from the Department of Pathology, Uppsala University Hospital and subsequently processed for production of tissue microarrays (TMAs), and protein extraction for Western blot analysis. In addition to standard TMAs, additional tissue samples are used for extended profiling of certain proteins. Immunohistochemistry is performed using standardized procedures and includes testing antibodies on TMAs and whole sections of tissues in order to create protein expression profiles. Approved immunostained slides are scanned to generate high-resolution digital images, using a 20x objective. Images are imported to an in-house built software to facilitate subsequent manual scoring of images. All procedures follow strict guidelines and include quality control. Optimal antibody dilution and target specificity is assessed by microscopical examination and comparison of immunohistochemical staining with internal and external gene/protein characterization data. For each approved antibody, a final immunostaining protocol is defined and subsequently applied to the standard 8 full-scale TMAs. For extended profiling of certain proteins, also slides containing extended samples are used. Manual annotation of immunohistochemistry is performed using an in-house software, and subsequently curated by an independent second observer. The generated protein profiles then undergo further evaluation and antibodies that pass quality criteria are assigned a reliability score, a knowledge-based protein expression profile and are scheduled for publication in the next version of the Human Protein Atlas.

Clinical Pathology

Head: Prof Fredrik Ponten

Responsibility: (i) Develop strategies to identify potential biomarkers based on the HPA database and other efforts, (ii) validate proteins that can be used as clinical biomarkers for disease, (iii) participate in clinical studies, collect tumor material and clinical data to generate specific cancer TMAs coupled to clinical databases, (iv) perform statistical analysis and validate the clinical usefulness of identified biomarkers.

Description: The HPA database is actively mined for potential biomarkers with the aim to identify protein expression patterns that could be of medical or biological significance. Projects include various forms of human disease with an emphasis on cancer. Most projects are focused on identification and validation of biomarker candidates that can fulfill currently unmet clinical needs related to diagnostics, prognostics and treatment prediction. To address such questions, patient cohorts representing different cancers are defined and tumor material as well as clinical data is collected. These specifically designed cancer TMAs are produced and used for extended analysis of protein expression patterns to test and validate candidate proteins as useful biomarkers. The biomarker discovery and validation efforts include both internal projects and external collaborative projects.