Michael T. Davis

Towards defining the urinary proteome using liquid chromatography‐tandem mass spectrometry I.Profiling an unfractionated tryptic digest

The proteome of normal male urine from a commercial pooled source has been examined using direct liquid chromatography‐tandem mass spectrometry (LC‐MS/MS). The entire urinary protein mixture was denatured, reduced and enzymatically digested prior to LC‐MS/MS analysis using a hybrid‐quadrupole time‐of‐flight mass spectrometer (Q‐TOF) to perform data‐dependent ion selection and fragmentation. To fragment as many peptides as possible, the mixture was analyzed four separate times, with the mass spectrometer selecting ions for fragmentation from a subset of the entire mass range for each run. This approach requires only an autosampler on the HPLC for automation ( i.e, unattended operation). Across these four analyses, 1.450 peptide MS/MS spectra were matched to 751 sequences to identify 124 gene products (proteins and translations of expressed sequence tags). Interestingly, the experimental time for these analyses was less than that required to run a single two‐dimensional gel.

Automated LC-LC-MS-MS platform using binary ion-exchange and gradient reversed-phase chromatography for improved proteomic analyses.

A simple multidimensional liquid chromatography system utilizing an isocratic pump and a HPLC system is described for the comprehensive proteomic analysis of complex peptide digest mixtures by coupled LC-LC-MS-MS techniques. A binary ion-exchange separation was achieved through the use of a strong cation-exchange column followed by a reversed-phase column for data-dependent LC-MS-MS analysis of the unbound analytes, and following salt elution (and concomitant column reequilibration), the bound analytes. Off-line validation of the platform showed near quantitative recovery of fractionated peptides and essentially complete ion-exchange partitioning. In comparative analyses of a highly complex peptide digest mixture a >40% increase in the number of peptide and protein identifications was achieved using this multidimensional platform compared to an unfractionated control.

Towards defining the urinary proteome using liquid chromatography-tandem mass spectrometry. II. Limitations of complex mixture analyses.

With an emphasis on obtaining a multitude of high quality tandem mass spectrometry spectra for protein identification, instrumental parameters are described for the liquid chromatography‐tandem mass spectrometry analysis of trypsin digested unfractionated urine using a hybrid quadrupole‐time‐of‐flight (Q‐TOF) mass spectrometer. Precursor acquisition rates of up to 20 distinct precursors/minute in a single analysis were obtained through the use of parallel precursor selection (four precursors/survey period) and variable collision induced dissociation integration time (1 to 6 periods summed). Maximal exploitation of the gas phase fractionated ions was obtained through the use of narrow survey scans and iterative data‐dependent analyses incorporating dynamic exclusion. The impact on data fidelity as a product of data‐dependent selection of precursor ions from a dynamically excluded field is discussed with regards to sample complexity, precursor selection rates, survey scan range and facile chemical modifications. Operational and post‐analysis strategies are presented to restore data confidence and reconcile the greatest number of matched spectra.

Biomarkers as Predictors of Response to Treatment with Motesanib in Patients with Progressive Advanced Thyroid Cancer

CONTEXT Antiangiogenic therapies have shown potential in the treatment of advanced thyroid cancer, but it is uncertain which patients are most likely to benefit from therapy. OBJECTIVE This prespecified exploratory analysis investigated whether baseline levels and/or changes in circulating biomarkers could predict tumor response and/or progression-free survival (PFS) among patients enrolled in a phase 2 study of motesanib in advanced thyroid cancer. DESIGN/SETTING/PATIENTS Patients with progressive locally advanced or metastatic medullary or differentiated thyroid cancer received motesanib 125 mg once daily for up to 48 wk in a phase 2 interventional study. Samples for assessment of circulating biomarkers of angiogenesis or apoptosis were collected at study wk 1 (baseline), 2, 4, 8, 16, 24, 32, 40, 48, and 4 wk after cessation of motesanib treatment. Tumor response was assessed per Response Evaluation Criteria in Solid Tumors by independent review. RESULTS Change from baseline in serum placental growth factor (PlGF) after 1 wk of treatment correlated with best tumor response (Kendall rank correlation, 0.28; P < 0.0001). Using a Fisher exact test, the most significant separation between patients who had an objective response and those who did not was at a 4.7-fold increase in PlGF. The response rate among patients with a greater than 4.7-fold increase in PlGF was 30% compared with 3% below this threshold. There was also a significant separation between responders and nonresponders at a 1.6-fold decrease in soluble vascular endothelial growth factor (VEGF) receptor 2 after 3 wk of treatment. Patients with baseline serum VEGF less than 671 pg/ml had significantly longer PFS times than the remainder of patients. CONCLUSIONS Changes in PlGF and soluble VEGF receptor 2 levels after initiation of therapy predicted response to motesanib in patients with advanced differentiated thyroid cancer or metastatic medullary thyroid cancer. Lower baseline VEGF levels were associated with longer PFS.

Simplification of complex peptide mixtures for proteomic analysis: Reversible biotinylation of cysteinyl peptides

A rapid means of identifying many components in an enriched mixture of proteins is enzymatic digestion of the entire protein fraction. This complex peptide mixture is then subjected to reversed‐phase high performance liquid chromatography (HPLC) coupled on‐line with a mass spectrometer capable of data‐dependent ion selection for fragmentation (LC‐tandem mass spectrometry; MS/MS). Thus, as many peptides as possible in the sample are fragmented to produce MS/MS spectra, which can then be searched against sequence databases. Ideally, one peptide from each protein in the mixture would be fragmented and identified. To this end, we employed an affinity selection method to capture cysteinyl peptides and thereby simplify the mixture. Both the captured cysteinyl and the noncysteinyl peptides are analyzed by LC‐MS/MS, to increase the numer of proteins identified. The method was tested on a limited set of standard proteins and applied to the analysis of a protein fraction obtained from isolated mitochondria treated with atractyloside. To further increase the number of different precursor ions selected for fragmentation, dynamic exclusion and ion selection from multiple narrow mass ranges of consecutive runs were employed.

Maximizing the sensitivity and reliability of peptide identification in large-scale proteomic experiments by harnessing multiple search engines

Despite recent advances in qualitative proteomics, the automatic identification of peptides with optimal sensitivity and accuracy remains a difficult goal. To address this deficiency, a novel algorithm, Multiple Search Engines, Normalization and Consensus is described. The method employs six search engines and a re‐scoring engine to search MS/MS spectra against protein and decoy sequences. After the peptide hits from each engine are normalized to error rates estimated from the decoy hits, peptide assignments are then deduced using a minimum consensus model. These assignments are produced in a series of progressively relaxed false‐discovery rates, thus enabling a comprehensive interpretation of the data set. Additionally, the estimated false‐discovery rate was found to have good concordance with the observed false‐positive rate calculated from known identities. Benchmarking against standard proteins data sets (ISBv1, sPRG2006) and their published analysis, demonstrated that the Multiple Search Engines, Normalization and Consensus algorithm consistently achieved significantly higher sensitivity in peptide identifications, which led to increased or more robust protein identifications in all data sets compared with prior methods. The sensitivity and the false‐positive rate of peptide identification exhibit an inverse‐proportional and linear relationship with the number of participating search engines.

Development of a method for the determination of glycine in human cerebrospinal fluid using pre-column derivatization and LC-MS/MS.

An LC-MS/MS method using pre-column derivatization with phenylisothiocyanate (PITC) was developed to quantify glycine in human cerebrospinal fluid (CSF) and applied to the determination of glycine in human samples collected during clinical testing. The calibration curve range for the assay was 50-10,000 ng/mL and ¹³C₂¹⁵N-glycine was used as an internal standard. Artificial CSF was used as a surrogate matrix for standards due to the presence of endogenous glycine in human CSF and this approach was validated with additional experiments involving either standard addition, or stable labeled glycine as an alternate calibration standard for endogenous glycine. Interday bias (% RE) and precision (% CV) were -4.2 and 12.3% at the LLOQ, and less than ±0.9 and 8.3% for higher concentrations, respectively. Glycine was stable in artificial CSF for at least 5h at room temperature, 55 days at -70 °C (-60 to -80 °C range), and through three freeze-thaw cycles.

Identification of incompletely processed potential Carboxypeptidase E substrates from CpEfat/CpEfat mice

In an attempt to identify peptides that may be involved in the obese phenotype observed in CpEfat/CpEfat mice (deficient in Carboxypeptidase E, CpE) samples from fourteen neuroendocrine tissues in wild‐type and CpEfat/CpEfat mice were obtained. Peptides were purified from these tissues and potential CpE substrate peptides were enriched using an anhydrotrypsin column that captures peptides with basic C‐termini. Bound peptides were subjected to tryptic digestion and followed by liquid chromatography‐mass spectrometry analysis. The relative levels of CpEfat/CpEfat versus wild‐type peptides were determined by comparison of the ion intensities. Peptide ions elevated in the CpEfat/CpEfat samples were identified by targeted liquid chromatography‐tandem mass spectrometry. From those ions, 27 peptides derived from known neuropeptides (including CpE substrates) were identified, together with another 25 peptides from proteins not known to be components of the neuropeptide processing pathway. The known CpE substrates identified included the recently discovered proSAAS, granin‐like neuroendocrine peptide precursor that inhibits prohormone processing. The approach demonstrated the feasibility of using an affinity‐based method for identifying differences in specific classes of peptides between normal and mutant mice.

Cancer biomarker discovery via low molecular weight serum proteome profiling - Where is the tumor?

Time‐course analyses of rapidly processed serum performed in parallel by SELDI and nanoscale LC‐MS/MS have revealed the temporal correlation of several literature‐based disease markers with ex vivo driven events such that their in vivo existence in healthy subjects is questionable. Identification by MS/MS reveals these putative biomarkers to be byproducts of the coagulation cascade and platelet activation and suggests plasmatic analysis may be preferred. In a pilot plasmatic study, a cohort of naïve prostate cancer (PCa) samples were uniformly distinguished from their age‐matched controls (n = 20) on the basis of multiple peptidic components; most notably by a derivative of complement C4 at 1863 m/z (GLEEELQFSLGSKINVK, C41353–1369). The fully tryptic nature of this and other putative PCa discriminants is consistent with the cleavage specificity of common blood proteases and questions the need for tumor‐derived proteolytic activities as has been proposed. In light of the known correlation of disregulated hemostasis with malignant disease, we suggest the underlying differentiating phenomena in these types of analyses may lie in the temporal disparity of sample activation such that the case (patient) samples are preactivated while the control samples are not.

Cancer Biomarker Discovery via Low Molecular Weight Serum Profiling—Are We Following Circular Paths?

The rigors of attaining reproducible protein identifications from complex biological matrices have recently been described as the ascent of a “mountainous road” that must surmount a series of methodological, technical, and analytical barriers to attain reliable results (1). From the perspective of biomarker discovery, the difficulty of this trail, and the attendant requisite level of expertise, is greatest when broad discovery work flows are used but is lessened substantially if targeted strategies can be used. Taking broad poetic license to portray this visual image within the context of the American westward expansion, one can envision a process by which teams must leave the comfort of the gentle plains (the proof-of-concept phase) to scale the foothills and peaks that lie on the trail to clinical utility. There will be much debate on the choosing of the best path forward. With this simile in mind (and to our point of view), we still see, as we suggested a few years ago, that many advocates of the use of mass spectrometry (MS)1 for profiling the so-called low molecular weight fragmentome (LMWF) remain circling on the plains, retracing the paths of evidence laid down decades before that had revealed the prevalence of dysregulated hemostasis in malignant disease (2). The genesis of what has been referred to as the “SELDI fiasco” (3) traces back to the early success and attendant hyperbole associated with the apparent differentiation of ovarian cancer patients from their unaffected controls by a pattern of uncharacterized peaks presented in the low-mass region of native serum SELDI analyses (4). Although the discriminatory power of these results was ultimately attributed to methodological bias (5), the concept of the use of biomolecule patterns as disease-specific identifiers had been proposed, and the need for component identification had been disputed. Qualitative data obtained by liquid …