Computational protocol: Prioritization of Biomarker Targets in Human Umbilical Cord Blood: Identification of Proteins in Infant Blood Serving as Validated Biomarkers in Adults

Similar protocols

Protocol publication

[…] Chemicals. All chemicals were obtained from Sigma-Aldrich (Sigma Aldrich, St. Louis, MO, USA) with the following exceptions: Sequencing-grade modified porcine trypsin was obtained from Promega (Madison, WI, USA), and Bradford reagent was purchased from Bio-Rad (Hercules, CA, USA).UCB serum samples. The UCB serum samples were acquired from the cord blood cohort collection of the Baltimore Tracking Health-Related Environmental Exposures (THREE) study. This study was approved by the Johns Hopkins Medicine Institutional Review Board (IRB approval 04-04-22-02) and received a waiver from the Health Insurance Portability and Accountability Act (2002). The study showed U.S.-representative exposure levels (mean and maximum concentrations) of a) perfluorooctanoate (PFOA): 1.6 ng/mL, 7.1 ng/mL; b) perfluorooctane sulfonate (PFOS): 5 ng/mL, 34.8 ng/mL; c) organochlorine pesticides (e.g., trans-nonachlor): 94 pg/mL, 185.5 pg/mL; d) trans-permethrin: 36.3 pg/mL, 34.8 ng/mL; e) polychlorinated biphenyls (PCBs) (e.g., di-ortho): 17.4 ng/g lipid, 176.5 ng/g lipid; and f) heavy metals (e.g., lead: 0.66 μg/dL, 15.5 μg/dL; copper: 38.6 μg/dL, 265 μg/dL). Details on this cohort are described elsewhere (, ; ; ; ). Of this cohort study, 12 samples (eight male and four female) () were randomly chosen for proteomic profiling. Using the Witter cord cradle (), hospital-trained personnel collected the UCB by direct venipuncture of the umbilical vein, which assured that no maternal blood was present in the samples. In addition, analysis of the X:Y chromosome ratios in male newborns as described by further confirmed the absence of maternal blood. Up to five 10-mL UCB samples were collected per newborn and immediately stored at 4°C. Within < 3 hr, the refrigerated blood specimens were centrifuged at 1,000 × g for 15 min to collect the serum. Serum samples were then fractionated and stored in 2-mL polypropylene cryovials at –80°C. For proteomic analysis, frozen sample splits were shipped on dry ice to the Biodesign Institute at Arizona State University; the individual samples were thawed for the first time just before sample processing.Preparation of reference pool samples and immunodepletion. Aliquots of 100 μL of each of the 12 individual UCB samples were pooled to obtain a composite sample with a protein concentration of 79 mg/mL protein as determined by Bradford assay. Of this pool, 240 μL were taken and human serum albumin (HSA) depleted using a Vivapure anti-HSA kit (VivaScience, Hannover, Germany) according to the manufacturer’s description. We chose not to further deplete the samples because several of the other highly abundant blood proteins often routinely depleted are either U.S. Food and Drug Administration (FDA)–approved adult blood biomarkers () or proposed biomarker candidates for diverse diseases or exposures (; ; ). The HSA-depleted proteome fraction was then concentrated and desalted by ultrafiltration using the Vivaspin 500 concentrators (MWCO 3kDa; Sartorius, Goettingen, Germany). The total volume of the resultant composite sample was 150 μL, with a protein concentration of 33 mg/mL.Protein digest and sample fractionation. Proteins were denatured and reduced in 10 mM ammonium bicarbonate and 0.05% sodium dodecyl sulfate with 10 mM dithiothreitol at room temperature for 1 hr. Alkylation of proteins was accomplished by incubation in 40 mM iodoacetamide for 1 hr at room temperature in the dark. Ten microliters sequencing-grade modified porcine trypsine (1 mg/mL stock solution; Promega, Madison, WI, USA) was added, and the mixture was incubated at 37°C overnight. Tryptic digests (100–500 μg) were then fractionated with the Agilent 3100 OFFGEL Fractionator using the 3100 OFFGEL Low Res Kit, pH 3-10 (Agilent Technologies, Santa Clara, CA, USA). The isoelectric focusing (IEF) was performed without ampholytes and glycerol according to the manufacturer’s instructions for peptide focusing. In short, the peptides were separated in a linear gradient of up to 8,000 V. The potential was kept at 8,000 V until 56,000 Vh was reached. The 12 IEF fractions were then extracted with 0.1% trifluoroacetic acid in 50% methanol, vacuum-concentrated, and dissolved in 2% acetonitrile and 0.1% trifluoroacetic acid.Reverse-phase liquid chromatography (RP-LC) separation and MS-analysis. For the RP nano-LC separation, a Tempo LC MALDI Spotting system (Applied Biosystems/MDS SCIEX, Foster City, CA, USA) was used with a 2-μL injector loop and a Chromolith CapRod column (150 × 0.1 mm; Merck, Darmstadt, Germany). Separation was obtained by running a gradient at a 2-μL/min flow rate. Solution A contained 2% LC-grade acetonitrile and 0.1% trifluoroacetic acid; solution B contained 98% acetonitrile and 0.1% trifluoroacetic acid. A 30-min gradient elution with the following parameters was used: 2% B (0.5 min), 2%→40% B (0.5–15 min), 40%→65% B (15–22 min), 65%→80% B (22–24 min), 80% B (24–26 min), 80%→2% B (26–28 min), 2% B (28–30 min). The matrix-assisted laser desorption/ionization (MALDI) matrix solution (7 mg/mL recrystallized α-cyano-hydroxycinnamic acid, 0.1% trifluoroacetic acid, 70% acetonitrile) was added postcolumn with a flow rate of 2 μL/min. Every 7 sec the combined eluate was automatically spotted onto a matrix prespotted stainless steel MALDI target plate (Applied Biosystems/MDS SCIEX). For calibration, 13 calibrant spots (ABI 4700 Mix) were added to each plate manually. All spotted samples were analyzed with a 4800 MALDI-tandem time-of-flight (TOF/TOF) mass spectrometer (Applied Biosystems/MDS SCIEX). First, MALDI-MS spectra were acquired over a mass range of m/z 800–4,000 in positive-ion reflector mode using 70–500 laser shots/spectrum with a fixed relative laser power of 3,300 and a central biased spot search pattern. In each MS spectrum, up to 25 peaks were selected for MS/MS using an acquisition method that excluded ions with signal-to-noise (S/N) ratios of < 50. The precursor ion with the weakest S/N ratio was acquired first to achieve the maximum signal intensity for low-abundance peptides. Tandem MS mode was operated using air as the collision-induced dissociation gas and enabled metastable ion suppressor settings. The relative precursor mass window was set to 200 (full width half mass). The MS/MS acquisition of selected precursors was set to a maximum of 2,500 shots/spectrum with a fixed relative laser power of 4,200.Protein identification and pathway mapping. The combined MS/MS spectra were searched using ProteinPilot™ Software v3.0 (version 3.01 prior to July 2009; Applied Biosystems/MDS SCIEX) with the implemented Paragon and the Pro Group processing algorithm against the human subset of the National Center for Biotechnology Information (Bethesda, MD, USA) nonredundant protein database (downloaded on 7 January 2008) and the UniProt human proteome database (version from 20 March 2009). Peptide and protein identification was carried out with ProteinPilotTM. Most search parameters are not user-adjustable but confer to the molecular and cellular proteomics guidelines (). Adjustable search parameters included cysteine modification by iodoacetamide, methionine oxidation, tryptic digestion, and thorough search with biological modifications ID focus. Additional information on the ProteinPilotTM algorithm is found in . Because the samples were obtained according to normal clinical procedures and settings, they may include proteins that were at least partially digested, proteolysed, or degraded during sample handling (). To address this issue and also to identify partially digested proteins, additional database searches were performed allowing nonspecific digestion as search parameter. This search strategy was successfully employed for the generation of a plasma proteome reference map (; ; ). Protein identification was based on ProtScore unused score criteria (Pro Group Algorithm, ProteinPilotTM software; Applied Biosystems/MDS SCIEX). Only proteins identified with Protscore ≥ 1.3 and at least one unique peptide with ≥ 95% confidence were used for further analysis. False discovery rate (FDR) was estimated by a search against a randomized decoy database using the same parameters as the original search () as well as using the MAYU approach ().Blast2GO software was used to extract gene ontology (GO) information for each protein () for functional characterization. The resulting list was curated manually. Pathway information for the identified proteins was obtained by Basic Local Alignment Search Tool (BLAST searches of the proteins against the curated Kyoto Encyclopedia of Genes and Genomes (KEGG) database ().Supporting information. Beyond the results presented here and in the (http://dx.doi.org/10.1289/ehp.1104190), the additional information (EHP_Appendix 1–4 and the spectral data folder) is available on our homepage http://labs.biodesign.asu.edu/halden/publications/ () and at http://proteomecommons.org () project UCB proteome. Appendix 1 contains the extended version of the list of all identified UCB proteins and their descriptions. Appendix 2 lists those proteins that are shared between the UCB proteome and the proteome reported for adult blood. Appendix 3 lists UCB proteins assigned to GO category “multicellular organismal development” which are unique for UCB or shared with adult blood. Appendix 4 shows a detailed overview of UCB proteins and their respective KEGG pathways involved. […]

Pipeline specifications

Software tools ProteinPilot, Mayu, Blast2GO, BLASTN
Databases KEGG
Application MS-based untargeted proteomics
Organisms Homo sapiens