Mass measurement error, protein

Fig. 25.5. (a) Example of direct tissue MS and MS/MS data (American cockroach neurohemal glands (cc and ca-complex) sprayed with the DHB matrix). MS profiling of ca spectrum averaged over the ca area of the tissue sample (see Note 5). Molecular mass measurement errors in ppm are calculated (Protein Calculator) and indicated for obsen/ed peptides (see Table 25.1 for peptide identifications), (b) Orbitrap (FTMS) MS/MS spectrum of m/z 996.6467, yielding high accuracy fragment ions confirming peptide sequence [LVPFRPRLamide Pea-PK-lll]. 

Pioneering laser desorption (LD) capabilities (Posthumus et al. 1978) have been greatly extended (Karas and Hillenkamp 1988) by dissolving the sample in a solid matrix that is a highly efficient absorber of the laser radiation, so as to ionize molecules as large as 300 kDa (Hillenkamp et al. 1991). The absence of matrix effects and mass measuring errors of 0.01%-0.1% make LD a particularly promising method for protein mixtures (Chait et al. 1990). 

Protein was measured using the crude protein method (25). Analysis of the percentage of protein may be low, leading to mass balance error. 

The problem is slightly different in the case of mutants obtained by directed mutagenesis. In that case, the expected modification is known and it is thus not necessary to verify the whole sequence. Comparing the molecular mass calculated from the mutant known sequence with that precisely measured using MALDI or ESI is sufficient. However, in the case when the mass difference between the substituted amino acids is smaller than the error in the measured molecular mass of the protein, a cleavage is necessary to obtain peptides with smaller masses. A cleavage is also necessary when several mutations were introduced. 

Today, most mass spectrometers used for protein analysis can achieve mass accuracy errors of +0.01% = +100 ppm. A mass accuracy of 10 ppm means that one could measure a 1000-amu peptide to 0.01 amu. For a 10,000-amu protein, one could measure the m/z to +0.1 amu. This mass accuracy error is considered excellent. 

Mass mapping is not without possible errors in accuracy. Mass loss accounts for inaccuracies, especially in proteins. Wall and Hainfield (17) report that a dose of 10e /0.1nm2 is a reasonable compromise between structural preservation and accuracy of mass measurements. Use of a cold specimen stage is requisite. Since proteins, nucleic acids, and lipids suffer mass loss at different rates, a dose-... 

A sample of the protein, horse heart myoglobin, was dissolved in acidified aqueous acetonitrile (1% formic acid in HjO/CHjCN, 1 1 v/v) at a concentration of 20 pmol/1. This sample was injected into a flow of the same solvent passing at 5 pl/min into the electrospray source to give the mass spectrum of protonated molecular ions [M + nH] shown in (a). The measured ra/z values are given in the table (b), along with the number of protons (charges n) associated with each. The mean relative molecular mass (RMM) is 16,951,09 0.3 Da. Finally, the transformed spectrum, corresponding to the true relative molecular mass, is shown in (c) the observed value is close to that calculated (16,951.4), an error of only 0.002%. 

In the example shown in Figure 1.23 using the peaks at m/z 939.2 and 1372.5 (j = 6), we obtain z = 6(1372.5 - 1.0073)/(1372.5 - 939.2) = 19 and we can number all the peaks measured according to the number of charges. M can be calculated from their mass. The average value obtained from all of the measured peaks is 17 827.9 Da with a mean error of 2.0 Da. This technique allowed the determination of the molecular masses of proteins above 130 kDa with a detection limit of about 1 pmol using a quadrupole analyser. 

Figure 9.11 Release of bovine serum albumin from EVAc matrices. Controlled release of BSA from an EVAc matrix. Solid particles of BSA either (a) 45 to 75 /rm or (b) 150 to 250 /rm in diameter were dispersed within EVAc by solvent evaporation to achieve final protein loadings from 10, 20, 30, 40, or 50%. In each case five identical slabs, each 70 mg and 1 mm thick, were incubated in cacodylate-buffered water containing 0.02% gentamicin. Periodically, the buffered water was replaced, and the amount of protein released from the matrix was determined by measuring the concentration of protein in the solution that was removed. Each symbol represents the average cumulative fraction of protein released (cumulative mass of protein released/initial mass of protein within the matrix) for the five samples error bars indicate the standard deviation, which in some cases are smaller than the symbols. Data from [16].

---