De novo Sequencing of Proteins With Mass Spectrometry
67
methylated peptide - allow to read out systematically correct amino acid
sequences. The disadvantages of the technique are that the peptides have to be
analyzed twice and that the chemical modification tends to increase the chemical
noise level in the electrospray spectrum.
The more elegant approach is to label the C-terminus by an ISO isotope during
the digestion. A mixture of 16 0 and ISO isotopes of every cleaved peptide is produced. When they are selected for fragmentation simultaneously all y ions are
distinguishable by their unusuall:1 16 0/ IS O isotopic pattern (Shevchenko A et al.
1997). Since the two isotopes differ only by 2 Da, isotopically resolved tandem
mass spectra should be acquired. This is achievable without any loss in sensitivity using a quadrupole - time of flight machine (Morris HR et al. 1996, Shevchenko A et al. 1997). Long amino acid sequences can be read out identifying y
ions by their specific isotopic distribution and their precise amino acid spacings.
The identification of y ions is limited by the fidelity with which the 1: 1 ratio of
the 16 0 to the ISO isotope is reflected in the spectrum. Some y ions may have been
generated with a low efficiency so that the isotope peaks contain only 5-10 ions
with a limited statistical representation of the real isotopic distribution. Or,
chemical noise ions on the same m/z value disturb the isotopic distribution when
fragmenting a peptide of low abundance. Here we present the differential scanning technique which addresses this issue. It facilitates peptide de novo sequencing on a triple quadrupole machine by allowing to use the ISO labeling and it
improves the sequence read out from fragment spectra generated from labelled
peptides with a quadrupole - time of flight mass spectrometer.
3
Methods
3.1
Sample Preparation
Chemicals used were of HPLC grade.
The peptides analyzed were derived from proteins purified in the course of
different biological projects. The examples had been selected to demonstrate
general principles of the de novo sequencing technique.
The proteins were generally separated on one dimensional SDS gels and Coomassie Blue stained with quantities in the range between 2 pmol-lO pmol. For
digestion they were de stained in 1:1 water/methanol,S % acetic acid, washed in
1:1 water/acetonitrile and acetonitrile, the proteins were reduced with DTT and
alkylated with iodoacetamide (Shevchenko A et al. 1996). Proteins were in-gel
digested with trypsin either in 50 % ISO water or 33 % ISO water. Isotopically
labelled water was purchased from Cambridge Isotope Laboratories (Andover,
MA, USA) and purified by distillation before use (Shevchenko A et al. 1997).
After extraction of the peptide mixture in a 1: 1 solution of 25 mM ammoniumbicarbonate and acetonitrile and a 1:1 solution of 10 % formic acid and acetonitrile, the peptide mixture was dried down for storage at 4°C. For the mass spectrometric analysis, the pep tides were taken up in 1 [tl 80 % formic acid, rapidly
diluted to 8 [tl with water and desalted on a 100 nL Poros™ R2 column (PerSept-
67
methylated peptide - allow to read out systematically correct amino acid
sequences. The disadvantages of the technique are that the peptides have to be
analyzed twice and that the chemical modification tends to increase the chemical
noise level in the electrospray spectrum.
The more elegant approach is to label the C-terminus by an ISO isotope during
the digestion. A mixture of 16 0 and ISO isotopes of every cleaved peptide is produced. When they are selected for fragmentation simultaneously all y ions are
distinguishable by their unusuall:1 16 0/ IS O isotopic pattern (Shevchenko A et al.
1997). Since the two isotopes differ only by 2 Da, isotopically resolved tandem
mass spectra should be acquired. This is achievable without any loss in sensitivity using a quadrupole - time of flight machine (Morris HR et al. 1996, Shevchenko A et al. 1997). Long amino acid sequences can be read out identifying y
ions by their specific isotopic distribution and their precise amino acid spacings.
The identification of y ions is limited by the fidelity with which the 1: 1 ratio of
the 16 0 to the ISO isotope is reflected in the spectrum. Some y ions may have been
generated with a low efficiency so that the isotope peaks contain only 5-10 ions
with a limited statistical representation of the real isotopic distribution. Or,
chemical noise ions on the same m/z value disturb the isotopic distribution when
fragmenting a peptide of low abundance. Here we present the differential scanning technique which addresses this issue. It facilitates peptide de novo sequencing on a triple quadrupole machine by allowing to use the ISO labeling and it
improves the sequence read out from fragment spectra generated from labelled
peptides with a quadrupole - time of flight mass spectrometer.
3
Methods
3.1
Sample Preparation
Chemicals used were of HPLC grade.
The peptides analyzed were derived from proteins purified in the course of
different biological projects. The examples had been selected to demonstrate
general principles of the de novo sequencing technique.
The proteins were generally separated on one dimensional SDS gels and Coomassie Blue stained with quantities in the range between 2 pmol-lO pmol. For
digestion they were de stained in 1:1 water/methanol,S % acetic acid, washed in
1:1 water/acetonitrile and acetonitrile, the proteins were reduced with DTT and
alkylated with iodoacetamide (Shevchenko A et al. 1996). Proteins were in-gel
digested with trypsin either in 50 % ISO water or 33 % ISO water. Isotopically
labelled water was purchased from Cambridge Isotope Laboratories (Andover,
MA, USA) and purified by distillation before use (Shevchenko A et al. 1997).
After extraction of the peptide mixture in a 1: 1 solution of 25 mM ammoniumbicarbonate and acetonitrile and a 1:1 solution of 10 % formic acid and acetonitrile, the peptide mixture was dried down for storage at 4°C. For the mass spectrometric analysis, the pep tides were taken up in 1 [tl 80 % formic acid, rapidly
diluted to 8 [tl with water and desalted on a 100 nL Poros™ R2 column (PerSept-