>>Effects of protein stabilizers and storage time on the mass spectrometry dynamics of human serum proteins
Presentation Start/End Time:
Monday, Apr 16, 2007, 1:00 PM - 5:00 PM
Location:
Exhibit Hall, Los Angeles Convention Center
Poster Section:
25
Poster Board Number:
28
Author Block:
Amit S. Dhamoon, Rick Saul, Chenwei Liu, Elise C. Kohn, Gordon Whiteley. National Institutes of Health, Bethesda, MD, SAIC-Frederick, Clinical Proteomics Reference Laboratory, Gaithersburg, MD
Optimization of pre-analytical variables in serum proteomics research is necessary to further develop this technology for biomarker discovery. We hypothesized that the presence of protein stabilizers (PS) in serum collection tubes at different storage times would affect the quality of mass-spectrometry (MS) data. We analyzed the effect of these variables on putative biomarkers for ovarian cancer (transthyretin variants and ApoA1) and on possible phosphorylation events measured by a novel bioinformatics method. Serum from 20 apparently healthy volunteers was collected with and without PS (BD P200 tube, Becton, Dickinson & Co; PS components proprietary). Serum was then frozen (-76 oC) immediately after preparation (t=0), or after 24, 48, or 72 hours of storage at 4 oC. Serum was fractionated using strong anion exchanger Ciphergen Q10 chips, and was studied using the Ciphergen SELDI platform. Separate aliquots of serum were fractionated using PerkinElmer Proexpression kits, which concentrate albumin-bound peptides, and studied using a PerkinElmer PrOTOF MS. Paired data were compared at each time point and repeated measure analysis was utilized to determine data trends over time. Total ionic current (TIC), as a surrogate of total protein load, was lower in serum tubes with PS. TIC measured on the Ciphergen MS was stable over time, while TIC on the PrOTOF MS decreased (p<0.05) over time in both tube types. Intensities of ApoA1 (m/z=28000), and unmodified (m/z=13736) and cysteinylated (m/z=13858) transthyretin peaks were statistically lower in tubes with PS at each time point (p<0.02, p<0.0002, p<0.015, respectively). Conversely, intensity of glutathionylated transthyretin (m/z=14050) was consistently higher (p<0.0003) in tubes with PS. With increasing storage time, intensity of unmodified transthyretin decreased (p<0.0001) in both tube types while intensities of ApoA1 and cysteinylated transthyretin did not show a significant change. Truncated transthyretin (m/z=12795) decreased (p=0.0003) without PS and glutationylated transthyretin increased (p=0.0025) with PS over greater storage time. PrOTOF datastreams were evaluated and 6 un- and putatively monophosphorylated peptides, or peak pairs, were identified by MS mass shifts of approximately 80 Da. Statistically significant differences in these peptide phosphorylation pairs between the two tube types were identified in 5 of 6 peak pairs. The presence of PS significantly alters the serum proteome over time as examined by SELDI and MALDI platforms. It is unclear how the presence of PS in serum collection tubes will impact the search for clinically useful biomarkers of disease. Further investigation is required to determine whether PS should be universally implemented in serum proteomics research.<<
Both of these underscore the difficulty SELDI still faces as a diagnostic tool. Various things have been tried with mixed results. Here protein stabilizers. There the filtering out of low abundance proteins. The above study is particularly interesting because it discusses a couple of the biomarkers CIPH is using in its soon to launched (yeah, right) ovarian cancer test. And as one can see, protein stabilizers mess up the biomarkers, and not uniformly. This might explain CIPH's 25% slide in the last couple of weeks.
Or it might be the patent re-examination mentioned in the recent 10-K, which isn't going well. CIPH could lose $2 million from Bio-Rad, and it could lose exclusivity in detecting single biomarkers with SELDI. My guess is Qiagen, who bought LumiCyte, is behind this.
Cheers, Tuck |
