A Proteomic View on Genome-Based Signal Peptide Predictions

All listed proteins were identified by 2D PAGE and subsequent MALDI-TOF mass spectrometry. Putative signal peptides were predicted as described by Tjalsma et al. (2000). The hydrophobic H-domain is shaded grey. The residues at positions −3 to −1 relative to the predicted SPase I or SPase II cleavage sites are underlined and bold. For lipoproteins the conserved cysteine residue, which is part of the lipobox and resides at the +1 positions of the mature lipid-modified lipoprotein, is also shown. Proteins with a potential twin-arginine motif (RRX##; Cristóbal et al. 1999) in the N-domains of their signal peptides are marked ^RR, and the twin-arginine motif is shown in bold. Lipoproteins are marked^lipo. Proteins containing predicted transmembrane segments in the mature part are marked ^TM, and proteins containing wall-binding repeats are marked ^W. Proteins that are part of the extracellular proteome of phosphate-starved cells are labeled^pst. Proteins that were analyzed by amino-terminal sequencing are marked ^#. In the case of the two WapA-specific spots, the low molecular mass 58 kDa spot represents a carboxy-terminal fragment of WapA, starting with Ser₁₇₂₅. Proteins with a known dual localization (cellular and extracellular) are labeled ^dual, and proteins that lack a typical signal peptide but have a known extracytoplasmic localization are labeled ^ex. Proteins that are released into the medium of B. subtilis Δlgt are marked^lgt. Proteins identified by Hirose et al. (2000) are marked^H.

XynD and YrpD were predicted previously as lipoproteins (Tjalsma et al. 2000). However, it is uncertain whether these proteins are lipoproteins, because both proteins have predicted type I SPase cleavage sites. As shown by amino-terminal sequencing, YrpD is processed at its SPase I cleavage site. In addition XynD was previously shown to be processed at its SPase I cleavage site (Hirose et al. 2000). Furthermore, the paralogs of XynD and YrpD (XynA and YncM, respectively) have signal peptides with a cleavage site for SPase I.

WprA is known to be a major cell wall protein (Margot and Karamata 1996), but it lacks a typical cell wall binding motif. PbpA lacks a typical SPase cleavage site. For unknown reasons, the signal peptide of YweA was not identified previously with the SignalP algorithm. The signal peptide of YxkC was not identified previously, because only the first 60 deduced amino acid residues of each B. subtilis protein were used for signal peptide predictions. Ironically, the SPase I recognition sequence of the signal peptide of YxkC (residues −3 to −1, relative to the cleavage site) starts at position 61.