Diversity and complexity of mouse allergens in urine, house dust, and allergen extracts assessed with an immuno-allergomic approach.

BACKGROUND: Mouse allergy is an important cause of indoor asthma and allergic rhinoconjunctivitis. The major mouse allergen, Mus m 1, is a complex of homologous pheromone-binding lipocalins called major urinary proteins (MUPs). METHODS: We analyzed the proteome of MUPs in mouse urine, commercial mouse epithelial extracts, and environmental samples using several approaches. These include as follows: two-dimensional electrophoresis and immunoblotting; liquid chromatography-high-resolution mass spectrometry (LC/HRMS); multiple reaction monitoring (MRM) mass spectrometry; and LC/HRMS analysis of glycans at the N-66 residue of MUP3. RESULTS: Albumin is predominant in the extracts, while MUPs are predominant in urine. LC/HRMS of 4 mouse allergen extracts revealed surprising heterogeneity. Of 22 known mouse MUPs, only 6 (MUP3, MUP4, MUP5, MUP13, MUP20, and MUP21) could be identified with MRM using unique peptides. Assessment of MUP content in urine, extracts, and dust samples showed good correlation between MRM and other methods working with different detection principles. All 6 identifiable MUPs were found in electrophoretically separated urine bands, but only MUP3 and MUP20 were above LOQ in unseparated mouse urine, and only MUP3, MUP4, and MUP20 were found in mouse epithelial extracts. Glycan heterogeneity was noted among 4 individual inbred mice: of 13 glycan structures detected, 8 were unique to one mouse, and only 2 glycan modifications were present in all 4 mice. CONCLUSIONS: Using mass spectrometry and MRM, mouse allergen extracts and urine samples are shown to be complex and heterogeneous. The efficacy and safety of commercial mouse allergen extracts will be improved with better controls of allergen content.

Measurement of German cockroach allergens and their isoforms in allergen extracts with mass spectrometry.

BACKGROUND: Allergen extracts are the primary tool for diagnosis and treatment of allergic diseases. In the United States, most allergen extracts are non-standardized. More sophisticated analytical approaches are needed to characterize these products and enable manufacturers and regulators to better determine potency. OBJECTIVE: To expand the multiple reaction monitoring (MRM) assay for an in-depth characterization of German cockroach (GCr; Blattella germanica) allergen extracts. METHODS: We applied advanced liquid chromatography (LC) and mass spectrometry (MS) techniques including MRM. The expanded LC/MRM-MS method was optimized to measure known GCr allergens and their isoforms/variants in commercial extracts and environmental samples. We performed isoform-specific allergen measurements in multiple extracts from four commercial sources and extracts prepared using environmental samples from urban homes. To investigate causes of heterogeneity, we examined over 30 extraction process variables. RESULTS: Evaluation of the commercial extracts confirmed the variability of production lots and commercial sources. Commonly used defatting and extraction protocols yielded extracts with comparable allergen profiles and content. However, the identity and quality of source materials was a major contributor to variability. In comparing commercial GCr extracts to environmental samples, relative quantities of Bla g 1, Bla g 2, Bla g 3, Bla g 4 and Bla g 11 were similar, while Bla g 5, Bla g 6, Bla g 7 and Bla g 8 were present in the environmental samples but largely absent for the commercial extracts. CONCLUSIONS AND CLINICAL RELEVANCE: LC/MRM-MS can be used to measure all known GCr allergens in commercial allergen extracts and environmental samples. Significant differences exist between allergen profiles of commercial extracts and the profiles of environmental samples from dwellings. This analytical platform can serve as a template to achieve better product characterization of similarly complex products.

Strand-Specific Dual RNA Sequencing of Bronchial Epithelial Cells Infected with Influenza A/H3N2 Viruses Reveals Splicing of Gene Segment 6 and Novel Host-Virus Interactions.

Host-influenza virus interplay at the transcript level has been extensively characterized in epithelial cells. Yet, there are no studies that simultaneously characterize human host and influenza A virus (IAV) genomes. We infected human bronchial epithelial BEAS-2B cells with two seasonal IAV/H3N2 strains, Brisbane/10/07 and Perth/16/09 (reference strains for past vaccine seasons) and the well-characterized laboratory strain Udorn/307/72. Strand-specific RNA sequencing (RNA-seq) of the infected BEAS-2B cells allowed for simultaneous analysis of host and viral transcriptomes, in addition to pathogen genomes, to reveal changes in mRNA expression and alternative splicing (AS). In general, patterns of global and immune gene expression induced by the three IAVs were mostly shared. However, AS of host transcripts and small nuclear RNAs differed between the seasonal and laboratory strains. Analysis of viral transcriptomes showed deletions of the polymerase components (defective interfering-like RNAs) within the genome. Surprisingly, we found that the neuraminidase gene undergoes AS and that the splicing event differs between seasonal and laboratory strains. Our findings reveal novel elements of the host-virus interaction and highlight the importance of RNA-seq in identifying molecular changes at the genome level that may contribute to shaping RNA-based innate immunity.IMPORTANCE The use of massively parallel RNA sequencing (RNA-seq) has revealed insights into human and pathogen genomes and their evolution. Dual RNA-seq allows simultaneous dissection of host and pathogen genomes and strand-specific RNA-seq provides information about the polarity of the RNA. This is important in the case of negative-strand RNA viruses like influenza virus, which generate positive (complementary and mRNA) and negative-strand RNAs (genome) that differ in their potential to trigger innate immunity. Here, we characterize interactions between human bronchial epithelial cells and three influenza A/H3N2 strains using strand-specific dual RNA-seq. We focused on this subtype because of its epidemiological importance in causing significant morbidity and mortality during influenza epidemics. We report novel elements that differ between seasonal and laboratory strains highlighting the complexity of the host-virus interplay at the RNA level.

Impact of Influenza A Virus Infection on the Proteomes of Human Bronchoepithelial Cells from Different Donors.

Susceptibility to influenza A virus is determined by a balance of viral and host factors. The genetic background of the host contributes to the severity of disease, but the influenza-related proteomes of cells from different individuals have not been compared. We used high-resolution mass spectrometry to identify proteins in normal human bronchial epithelial (NHBE) cells isolated from three different donors. Infection of each NHBE cell culture with influenza A/California/07/2009 (H1N1) resulted in expression of viral proteins and a variety of host proteins, including interferons, interferon-stimulated genes, and secreted chemokines/cytokines. The expression level of viral proteins corresponded to the level of host proteins that support influenza infection (i.e., pro-viral proteins); however, production of infectious virus was inversely related to the levels of antiviral proteins, suggesting that a balance of pro-viral proteins and the antiviral response controls virus replication. In summary, our results demonstrate that expression levels of pro-viral as well as antiviral factors are different for each donor and suggest that relative quantitation of these factors may provide a way to identify individuals or population groups who are susceptible to severe influenza disease.

Assessment of influenza A neuraminidase (subtype N1) potency by ELISA.

Antibodies that inhibit neuraminidase (NA) activity of influenza virus provide resistance against disease and have been associated with milder epidemics. Although studies have demonstrated a correlation between NA inhibition antibody titers and vaccine efficacy, neither the quantity nor form of NA is measured in seasonal and pandemic influenza vaccines. In this report, we describe development of enzyme-linked immunosorbent assays (ELISAs) that are suitable for quantitation of the native form of NA of subtype N1. The assays use mouse monoclonal antibodies (mAbs) 1H5 and CD6 to capture NAs of viruses, and a different mAb 4E9 to detect bound antigen. The 1H5-capture ELISA detects NAs of seasonal and pandemic H1N1 viruses as well as H5N1 viruses and has a limit of quantitation (LOQ) of 5.5ng/mL for seasonal H1N1A/Brisbane/59/2007 NA. The CD6-capture ELISA is specific for NA of the 2009 pandemic viruses with a LOQ of 67ng/mL for A/California/07/2009 NA. The ELISA signals in both assays are proportional to NA enzymatic activity and correlate with NA immunogenicity. The ELISAs we describe may expedite the development of NA-based influenza vaccines by providing a practical assay to measure NA potency.

The proteomic dataset for bone marrow derived human mesenchymal stromal cells: Effect of in vitro passaging.

Bone-marrow derived mesenchymal stromal cells (BMSCs) have been in clinical trials for therapy. One major bottleneck in the advancement of BMSC-based products is the challenge associated with cell isolation, characterization, and ensuring cell fitness over the course of in vitro cell propagation steps. The data in this report is part of publications that explored the proteomic changes following in vitro passaging of BMSCs [4] and the molecular heterogeneity in cultures obtained from different human donors [5], [6].The methodological details involving cell manufacturing, proteome harvesting, protein identification and quantification as well as the bioinformatic analyses were described to ensure reproducibility of the results.

System-wide survey of proteomic responses of human bone marrow stromal cells (hBMSCs) to in vitro cultivation.

Human bone marrow stromal cells (hBMSCs, also loosely called bone marrow-derived mesenchymal stem cells) are the subject of increasing numbers of clinical trials and laboratory research. Our group recently reported on the optimization of a workflow for a sensitive proteomic study of hBMSCs. Here, we couple this workflow with a label-free protein quantitation method to investigate the molecular responses of hBMSCs to long-term in vitro passaging. We explored the proteomic responses of hBMSCs by assessing the expression levels of proteins at early passage (passage 3, P3) and late passage (P7). We used multiple biological as well as technical replicates to ensure that the detected proteomic changes are repeatable between cultures and thus likely to be biologically relevant. Over 1700 proteins were quantified at three passages and a list of differentially expressed proteins was compiled. Bioinformatics-based network analysis and term enrichment revealed that metabolic pathways are largely altered, where many proteins in the glycolytic, pentose phosphate, and TCA pathways were shown to be largely upregulated in late passages. We also observed significant proteomic alterations in functional categories including apoptosis, and ER-based protein processing and sorting following in vitro cell aging. We posit that the comprehensive map outlined in this report of affected phenotypes as well as the underpinning molecular factors tremendously benefit the effort to uncovering targets that are not just used only to monitor cell fitness but can be employed to slowdown the in vitro aging process in hBMSCs and hence ensure manufacturing of cells with known quality, efficacy and stability.

Global proteomic signature of undifferentiated human bone marrow stromal cells: evidence for donor-to-donor proteome heterogeneity.

The clinical application of human bone marrow stromal cells (hBMSCs) largely depends on their capacity to expand in vitro. We have conducted a comprehensive comparative proteomic analysis of culture-expanded hBMSCs obtained from different human donors. The data reveal extensive donor-to-donor proteomic heterogeneity. Processing and database-searching of the tandem MS data resulted in a most comprehensive to date proteomic dataset for hBMSC. A total of 7753 proteins including 712 transcription and translation regulators, 384 kinases, 248 receptor proteins, and 29 cytokines were confidently identified. The proteins identified are mainly nuclear (43.2%) and the share of proteins assigned to more than one subcellular location constitutes 10% of the identified proteome. Bioinformatics tools (IPA, DAVID, and PANTHER) were used to annotate proteins with respect to cellular locations, functions, and other physicochemical characteristics. We also compared the proteomic profile of hBMSCs to recently compiled datasets for human and mouse pluripotent stem cells. The result shows the extent of similarity between the three cell populations and also identified 253 proteins expressed uniformly by all lines of hBMSCs but not reported in the proteomic datasets of the two pluripotent stem cells. Overall, the proteomic database reported in this paper can serve as a reference map for extensive evaluation of hBMSC to explain their biology as well as identify possible marker candidates for further evaluation.

Mass spectrometry-based proteomic analysis of human liver cytochrome(s) P450.

The major objective of personalized medicine is to select optimized drug therapies and to a large degree such mission is determined by the expression profiles of cytochrome(s) P450 (CYP). Accordingly, a proteomic case study in personalized medicine is provided by the superfamily of cytochromes P450. Our knowledge about CYP isozyme expression on a protein level is very limited and based exclusively on DNA/mRNA derived data. Such information is not sufficient because transcription and translation events do not lead to correlated levels of expressed proteins. Here we report expression profiles of CYPs in human liver obtained by mass spectrometry (MS)-based proteomic approach. We analyzed 32 samples of human liver microsomes (HLM) of different sexes, ages and ethnicity along with samples of recombinant human CYPs. We have experimentally confirmed that each CYP isozyme can be effectively differentiated by their unique isozyme-specific tryptic peptide(s). Trypsin digestion patterns for almost 30 human CYP isozymes were established. Those findings should assist in selecting tryptic peptides suitable for MS-based quantitation. The data obtained demonstrate remarkable differences in CYP expression profiles. CYP2E1, CYP2C8 and CYP4A11 were the only isozymes found in all HLM samples. Female and pediatric HLM samples revealed much more diverse spectrum of expressed CYPs isozymes compared to male HLM. We have confirmed expression of a number of "rare" CYP (CYP2J2, CYP4B1, CYP4V2, CYP4F3, CYP4F11, CYP8B1, CYP19A1, CYP24A1 and CYP27A1) and obtained first direct experimental data showing expression of such CYPs as CYP2F1, CYP2S1, CYP2W1, CYP4A22, CYP4X1, and CYP26A1 on a protein level.

Improved proteomic profiling of the cell surface of culture-expanded human bone marrow multipotent stromal cells.

A comprehensive analysis of the membrane proteome is essential to explain the biology of multipotent stromal cells and identify reliable protein biomarkers for the isolation as well as tracking of cells during differentiation and maturation. However, proteomic analysis of membrane proteins is challenging and they are noticeably under-represented in numerous proteomic studies. Here we introduce new approach, which includes high pressure-assisted membrane protein extraction, protein fractionation by gel-eluted liquid fraction entrapment electrophoresis (GELFREE), and combined use of liquid chromatography MALDI and ESI tandem mass spectrometry. This report presents the first comprehensive proteomic analysis of membrane proteome of undifferentiated and culture-expanded human bone marrow multipotent stromal cells (hBM-MSC) obtained from different human donors. Gene ontology mapping using the Ingenuity Pathway Analysis and DAVID programs revealed the largest membrane proteomic dataset for hBM-MSC reported to date. Collectively, the new workflow enabled us to identify at least two-fold more membrane proteins compared to published results on hBM-MSC. A total of 84 CDs were identified including 14 CDs identified for the first time. This dataset can serve as a basis for further exploration of self-renewal, differentiation and characterization of hBM-MSC.