Scott Peterman (Presenter)
Thermo Fisher Scientific
Authorship: Scott Peterman (1), David Sarracino (1), Scott Kronwitter (1), Mazi Mohiuddin (1), Shen Luan (1), Amol Prakash (2)
(1) Thermo Fisher Scientific BRIMS, Cambridge, MA (2) Optys Technologies, Inc. Boston, MA
| Short Abstract Routine hemoglobin profiling is used to determine putative disease states based on glycation and/or sequence variation. Thus, analytical workflows must be sensitive to detect low-level variants and selective to characterize compounds as truncated chains, modified, altered sequence, or any combination. To achieve this, we have developed a workflow incorporating sample preparation routines using cation exchange resins to rapidly and robustly concentrate, remove matrix interference, and facilitate molecular weights cutoff extraction centered at hemoglobin chains. To test this workflow, we have evaluated neat whole blood (WB) as well as WB spiked with bovine hemoglobin at various levels. |
Long Abstract
Introduction:
Targeted profiling of hemoglobin remains a routine focus for two primary applications: glycohemoglobin analysis and hemoglobinopothies. Both studies require profiling all forms of hemoglobin to determine potential risk of disease. While hemoglobin is present in whole blood at high concentrations, the analytical method must be able to detect, characterize, and perform relative quantitation of minor components attributed to SNPs and PTMs on native and truncated hemoglobin chains. This requires a routine, high-throughput method to concentrate, remove the background matrix, and perform rapid qualitative and quantitative analysis using LC-MS.
Methods:
A set of samples were created to evaluate the sample preparation, LC-MS detection, and automated data processing across a complex and wide dynamic range of hemoglobin samples. All samples were created using whole blood samples from an in-house donor and a stock solution of bovine hemoglobin was created from a commercially available source (Sigma-Aldrich). WB samples were diluted 100-fold using 90:10 (water/MeOH) and the stock bovine hemoglobin solution was prepared using the same solution. Samples were created either neat or mixed separate wells of a 96-well SOLA 10mg Retain CX plate. The ratio of human/bovine hemoglobin amounts were based on the reported human hemoglobin concentration of 200 µg/mL WB and the ratios used were neat human and bovine hemoglobin as well as mixtures of human/bovine hemoglobin stock solutions consisting of 1:9, 3:7, 5:5, 7:3, 9:1. The plate was washed using water and organic by vacuum in less than a minute per step. The specific protein molecular weight (MW) range extracted off of the cation exchange resin was accomplished using 750 mM of ammonium formate. For sample analysis, a 20 µL extract aliquot was injected onto a 3 x 120 mm SEC column created in-house for a 6 minute isocratic analysis (82:18) of standard +ESI binary solvents. Data acquisition was performed using alternating high resolution/accurate mass (HR/AM) MS and all ion fragmentation (AIF) analysis on a Q Exactive High Field mass spectrometer. Both MS and AIF data was acquired using the same settings except the AIF scan event which set the stepped HCD collision energies of 25, 28, and 30% with the default charge state of 5. Both scan events were acquired with 10 microscans per data point. All data was processed using Pinnacle software designed iterative searching of intact and truncated protein sequences, SNPs, and PTMs based on charge state distribution profiles as well as specific fragmentation patterns.
Results:
A set of 175 (25 per sample) samples were created as described above to evaluate the workflow for profiling capabilities in a short sample analysis cycle time with few sample handling steps. To achieve this goal, the present workflow incorporates an abbreviated sample preparation step and a SEC-mass spectral detection scheme. All samples were loaded into two SOLA Retain CX plates to concentrate the sample, remove the matrix compounds, and facilitate MWCO step to displace the molecular weight range around hemoglobin. Each of the sample preparation steps took less than 1 minute. The plate of extracted hemoglobin samples were then loaded directly into the autosampler and 20 µL was introduced into the mass spectrometer for precursor and product ion analysis in 6 minutes. The neat analysis of human hemoglobin identified 3 different truncated forms of the alpha and beta chains as well as glycated alpha chain compared to over 6 different forms of bovine alpha and beta hemoglobin chains. Each chain was evaluated based on mass error analysis and charge state and isotopic distribution profiles. In addition, specific product ions were identified in the resulting AIF spectra for sequence confirmation. Ratio analysis of the spiked hemoglobin mixtures was measured to be less than 25% for all levels as well as the response ratios for the measured variants. The overall time required to prepare, analyze, and process the samples was ca. 18 hrs.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | no | |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | yes |