| Trace/single-cell analysis reveals cellular heterogeneity,thereby facilitating the investigation of physiological and pathological processes of biological organisms,which cannot be achieved using the averaged results yielded by bulk sample analysis.As the executor of cell function,protein more directly reflects the state of cells.However,due to the inability to amplify proteins like nucleic acids,as well as the current sensitivity limitation of proteomics analysis instrument,trace/single-cell proteomics still faces many challenges.To address the issue of the limited sensitivity of trace/single-cell proteomics analysis,new methods in two aspects of analysis process were proposed in this study.The first one is to solve the problem of poor identification of MS/MS spectra of trace/single-cell proteomics.A strategy named same-precursor-produced-unidentified-spectra merging（SPPUSM）is proposed.Due to the low amount of peptides in trace/single-cells,the precursor ions accumulation in the mass spectrometer is insufficient to produce adequate fragment ions with high-signal-to-noise ratio for peptide identification.The low-quality MS/MS spectra result in an obviously lower spectra identification rate compared with the obtained in bulk analysis.Therefore,new methods to improve the interpretation of these low-quality unidentified spectra is of particular importance for trace/single-cell studies.Reducing the adsorption and transfer losses in the sample processing procedure has always been a key point in trace/single-cell analysis,as minute sample loss results in huge identification decreasing in trace analysis.Therefore,we developed an angle-tip based trace sample processing method to decrease the surface adsorption resulted peptide loss and efficiently improved proteome identification of 1000,100 and 10 cells.Additionally,a preliminary study on a new labeling method for de novo peptide sequencing was also explored in this study.We developed a new database searching method by merging unidentified MS/MS spectra produced by the same precursor ions to improve the spectra identification ratio.By merging the unidentified MS/MS spectra produced by the same precursor ions from multiple technical replicates,the number of fragment ions in the merged spectra is increased,the signal-to-noise ratio of the fragments is enhanced,both of which facilitate spectra identification.The feasibility of this approach was first tested on standard peptide.No false identification was discovered even merging the spectra from isoform peptides with the same amino acid composition,indicating the reliability of this method.After applying it to trace and single-cell analysis,this technique increased proteome identification scale with an 18.2%increase in protein identification for 1 ng samples and a maximum 60.7%enhancement for single He La cells.The newly identified peptides/proteins were mostly low-abundant ones in the sample and can effectively detect cellular heterogeneity between He La and HEK293T cell at the single-cell level.The error rate of this method was evaluated by the“entrapment database”strategy using two mixed species and approximately 3%FDRs were obtained when 4-5 files were merged,demonstrating the reliability of our SPPUSM.For trace/single-cell sample processing,we developed a new micro-cell processing system prepared by folding a 10μL pipette tip into a"V"shape,named as V-tip.Due to its decreased internal surface area to reduced non-specific adsorption,the sample loss was efficiently reduced during the processing process.The V-tip can be inserted into a96-well plate and directly applied to a flow cytometry for parallel cell collection and sample processing.Moreover,when compared to the commonly used micro-cell processing container,such as PCR tubes,V-tip also displayed obvious advantages.In three experimental groups of 10,100,and 1000 cells,the highest improvement was achieved in the 100 cells group,with more than twice as many protein groups as obtained by the PCR tubes.The V-tip method was further applied to the analysis of mouse tissue cells.Using 100 cells obtained from mouse liver and brain the identified proteins by V-tip clearly distinguished the two cell types.All the above results highlight the superiority of the V-tip method for processing trace/single-cell sample.Finally,we developed an isotopic labeling-based approach for peptide de novo sequencing.In this method C-terminal 18O exchange and N-terminal amino acetylation of the peptides were employed to introduce two types isotopic modification with minimal mass difference of 0.0049 Da to facilitate the production of b/y ion pairs and to enable more efficient de novo sequencing.The labeling efficiency was verified using standard peptides and complex samples,in which more than 99%isotopic labeling was reached.Using this labeling strategy,abundant b/y ion pairs were found in the MS/MS spectra.We further optimized the mass spectrometry parameters by altering the precursor ion isolation window for fragmentation,and discovered that 2.5 m/z was an appropriate isolation window size,enabling the retrieval of the maximum number of ion-pairs-containing spectra.By retaining the light and heavy labeled b/y ions in the ion pair spectra,only the theoretical peptide fragment ions were remained in the spectra,which could be used for further de novo sequencing. |
