N-Terminal Peptide Detection with Optimized Peptide-Spectrum Matching and Streamlined Sequence Libraries

Abstract

We identified tryptic peptides in yeast cell lysates that map to translation initiation sites downstream of the annotated start sites using the peptide-spectrum matching algorithms OMSSA and Mascot. To increase the accuracy of peptide-spectrum matching, both algorithms were run using several standardized parameter sets, and Mascot was run utilizing a, b, and y ions from collision-induced dissociation. A large fraction (22%) of the detected N-terminal peptides mapped to translation initiation downstream of the annotated initiation sites. Expression of several truncated proteins from downstream initiation in the same reading frame as the full-length protein (frame 1) was verified by western analysis. To facilitate analysis of the larger proteome of Drosophila, we created a streamlined sequence library from which all duplicated trypsin fragments had been removed. OMSSA assessment using this "stripped" library revealed 171 peptides that map to downstream translation initiation sites, 76% of which are in the same reading frame as the full-length annotated proteins, although some are in different reading frames creating new protein sequences not in the annotated proteome. Sequences surrounding implicated downstream AUG start codons are associated with nucleotide preferences with a pronounced three-base periodicity N1^G2^A3.

Keywords: Mascot; OMSSA; peptide mass spectrometry; streamlined sequence libraries.

Figure 1.

Mascot is significantly more accurate when screening a, b, and y ions. In assessment of conformance to parent proteins before trypsin digestion, all 4560 peptides detected by Mascot screening for a, b, and y CID ions were also detected when screening for only b and y ions. However, the 264 additional peptides detected only in the b/y screen had very poor conformance to parent protein MWs (62.9%; arrow). Bootstrap analysis with 2000 samples of 264 peptides (with replacement) from the a/b/y screen revealed 99% confidence limits between 79.5 and 90.9% (broken arrows). This confirms that the poor b/y conformance of 62.9% is significantly lower than the a/b/y conformance.

Figure 2.

Western analysis of frame-1 truncated proteins. DownPeptide genes with predicted frame-1 truncated proteins were tested in TAP-epitope-tag expression lines (DOT1, SKN7, GCS1). Cells were grown under various conditions: normal (n), stationary (s), synthetic minimal medium with glucose (SMD), starved in synthetic minimal medium lacking nitrogen (SD-N), or heat-treated (37 °C, 15 min). Five experiments are illustrated where truncated, and full-length proteins (arrows) were detected under different conditions. A truncated protein of DOT1 was also previously reported. Note that both the annPeptide and downPeptide were detected by MS/MS for DOT1, but only the downPeptide was detected by MS/MS for SKN7. The downPeptide for GCS1 was only detected with one OMSSA parameter set. The detected truncated and full-length proteins ran appropriately according to MW size markers (not shown).

Figure 3.

Stripped sequence libraries facilitate MS/MS analysis of large proteomes. Sequence libraries stripped of duplicated trypsin fragments give much smaller search spaces for PSM algorithms. The mRNAs from alternative splicing produce proteins with high sequence redundancy, most of which is removed when duplicated trypsin fragments are removed from the proteins of the sequence library. Alternative transcription start sites are shown (*).

Figure 4.

Lengths of detected frame-2 and -3 downORFs compared with all frame-2 and −3 ORFs initiated within 100 nucleotides downstream of the annAUG (and ≥15 nucleotides long). The ORFs for downPeptides detected in yeast (A) are longer than expected by random selection (chi-square goodness of fit p < 0.01). Although generally longer, the Drosophila downORFs (B) are not significantly longer by chi-square test. Illustrated are ORF lengths for downPeptides detected by OMSSA or Mascot with the standard yeast library (A; 5% FDR) and OMSSA with the stripped Drosophila library (B; 2% FDR).

Figure 5.

(A) Sequences flanking the implicated frame-1 downAUGs of Drosophila downPeptide genes have a pronounced 3-nucleotide periodicity with depression of G at position 2 and A at position 3 of the codons. Average nucleotide frequencies of aligned sequences are shown relative to the downAUG at positions 1, 2, and 3. (B) Frequencies of G are depressed at the second nucleotide of codons downstream of the AUG of frame-1 and frame-2 downPeptide ORFs (*) despite the frame-2 ORF being out of frame with the overlapping frame-1 ORF. (C) Average deviations from background, measured as log₂(freq_obs/freq_background), are illustrated for codon positions 1, 2, and 3 for windows upstream (positions −20 to −1) and downstream (positions +4 to +20) of the start codons of frame-1 and frame-2 downPeptide ORFs (based on background nucleotide frequencies in ORFs; fA: 25.6%, fC: 27.1%, fG: 26.8%, fU: 20.5%). Compared with random samples of 1000 downAUGs, frame-1 downPeptides show G₂ and A₃ depression at positions 2 and 3 of codons upstream and downstream of the start codon (*). Frame-2 downPeptides show G₂ depression (*) downstream of the start codon at positions corresponding to the wobble position of frame-1. The G₂ and A3 depressions (*) are significant by bootstrap analysis (p < 0.01). Only downPeptides (2% FDR) with downAUGs > 20 nucleotides downstream of the annAUG were used in this analysis.