In a review of relevant literature for phosphopeptide enrichment using various titanium dioxide formats (1–
27) and immobilized metal affinity chromatography (IMAC) formats (
7,
28–
54), there was no clear consensus on sample and reagent formulations that yielded the optimum phosphopeptide enrichment ratio, the highest absolute phosphopeptide recovery, and the greatest reproducibility. In the majority of these studies, reagent formulations were optimized for the experiment at hand.
Similarly, phosphopeptide enrichment using AssayMAP Fe(III)-NTA and TiO2 cartridges requires some degree of chemistry optimization and method development depending on the desired analytical result. No one set of chemistry conditions will be universally optimal. However, there are some chemistry trends that can serve as reasonable starting points for method development.
Typically, samples for phosphopeptide enrichment using Fe(III)-NTA and TiO2 have first been subjected to a reversed-phase cleanup (desalting) step. Samples are generally in a phosphate-free, TFA-containing, low-pH buffer (pH 2.4 to 2.8) with some amount of organic solvent. Low-pH conditions reduce non-specific binding of non-phosphopeptides by facilitating protonation of acidic side-chains and C-termini of peptides, whereas the phosphoryl groups of phosphopeptides remain negatively charged at low pH to promote binding in a combination of ion exchange and metal coordination to the stationary phase (
1,
55). The presence of organic solvent (typically 25% to 80% acetonitrile) in the sample and wash solutions helps to reduce nonspecific binding of non-phosphopeptides. Organic solvent serves to modulate the pKa values of ionizable species and the effect can be very different depending on the chemical species (
56,
57). Achieving high phosphopeptide enrichment often means finding conditions that maximize the difference in the pKa values of phosphoryl and carboxyl groups such that under a given set of solvent conditions, the phosphoryl groups are negatively charged while carboxyl groups are neutral (
44,
48). This is accomplished by adjusting the type and concentration of both the acid modifier and organic solvent present in the samples and wash buffers. Specificity can also be improved for phosphopeptide enrichment by TiO
2 with the addition of small-molecule organic acids (such as, glycolic acid, lactic acid, and dihydroxybenzoic acid), which may further reduce nonspecific binding of non-phosphopeptides by competitively displacing weakly bound acidic, non-phosphopeptides.
For each AssayMAP Fe(III)-NTA and TiO2 cartridge, the quantitative binding capacity (
≥ 90% recovery) depends on the chemical characteristics of the phosphopeptide targets and the sample matrix. Matrix effects include the overall sample complexity and the presence and abundance of certain of solvents, buffers components, additives, and the pH of the sample.
For a tryptic digest of bovine a-casein, ~10 mole phosphate per mole protein (58), quantitative recovery of phosphopeptides is achieved with up to 1200-µg digest (~500 nmol phosphate) loaded onto TiO
2 cartridges and up to 150-µg digest (~ 65 nmol phosphate) loaded onto Fe(III)-NTA cartridges.
The default protocol settings assume that samples will be arranged in multiples of 8 in a column-based configuration. Also, the Bravo Platform applies differential pressure to seat cartridges based upon the number of full columns of cartridges. To achieve proper cartridge seating, entire columns must be used.