Development of Novel Isobaric Tags Enables Accurate and Sensitive Multiplexed Proteomics Using Complementary Ions

Abstract

High-throughput quantitative analysis of the cells’ proteomes across multiple conditions such as various perturbations and different time points is essential for gaining insights into treatment-induced biological responses or disease pathological states. The advancements in mass spectrometry instrumentation and isobaric labeling methods provided useful tools to help address such demands. However, the current widely adopted isobaric labeling methods such as tandem mass tag (TMT) and isobaric tags for relative and absolute quantitation (iTRAQ) are based on low-mass reporter ions, which are indistinguishable among different peptide analytes, to achieve relative quantification. Therefore, these methods intrinsically suffer from severe ratio distortion when analyzing complex samples due to peptide coelution and cofragmentation. Here, we developed a novel set of isobaric tags named dimethylated leucine complementary ion (DiLeuC) and relied on complementary ions for relative quantification, in which the complementary ions are the remanent peptide segments after fragmentation in the high-mass range. Since those residual peptide fragments are precursor-specific, they retain the relative abundance information in an interference-free manner even in a complex matrix environment. The quantification accuracy of our method was validated in a two-proteome model where the yeast proteome was spiked with a strong background human proteome as interference. In addition, we also applied this strategy to single-cell proteome analysis, demonstrating its potential utility for sensitive high-throughput quantitative proteomics.

Graphical Abstract

INTRODUCTION

The quantitative analysis of the proteomes across multiple conditions such as different time points and various perturbations is essential for gaining insights into biological processes or disease states. The combinatorial advancement of mass spectrometry (MS) instrumentation and isobaric labeling techniques have provided powerful tools for such purposes. In quantitative proteomics enabled by isobaric labeling strategy, the isobaric tags such as TMT and iTRAQ produce differentially labeled peptides that are indistinguishable in full-MS spectra but after isolating and fragmenting in tandem MS (MS/MS), quantitative information can be acquired by comparing intensities of the corresponding reporter ions in the low mass range (1). However, the accuracy and precision of this method can be compromised when contaminating peptide ions are co-isolated with the target peptide ions, as the reporter ions from contaminating precursors are indistinguishable from the target analytes. This so-called interference effect is an inherent limitation of all MS2-based quantification methods (2). Ting, L. et al were the first to systemically investigate this effect using a two-proteome system by mixing yeast and human peptides together, and they found a severe distortion of reporter ion intensities for the yeast peptide ions when spiking with abundant human proteome as interference (3). They then proposed a strategy that applies triple-stage tandem mass (MS3) to alleviate this problem. In their design, the MS2 spectrum was only used for peptide identification, and the most intense fragment ion in MS2 was selected and subjected to another round of fragmentation, i.e., MS3. The quantification was then achieved by reporter ions in the MS3 spectrum. This approach was proven to be successful in eliminating the interference effect and providing accurate quantification.

Although the standard MS3 method successfully countered the detrimental effects of the interfering ions, this method reduces sensitivity markedly. By fragmenting the initial precursor ions into all the possible product ions and selecting a single one of them for subsequent interrogation, only a small percentage of MS1 precursor signals are converted into the MS3 reporter ion signal, resulting in reduced quantitative sensitivity. To address this limitation, Gygi and coworkers described a solution using isolation waveforms with multiple frequency notches for synchronous precursor selection of multiple MS2 fragments, namely SPS-MS3 (4). In this manner, a 10-fold increase in TMT reporter ion intensities was observed relative to the original MS3 approach. This method soon became the gold standard in multiplexed proteomics until very recently the real-time search SPS-MS3 gains popularity with the advanced generation of mass spectrometer – the Orbitrap Eclipse – become commercially available (5). With real-time search feature, the duty cycle is reduced and the depth of quantification can be further improved.

Besides the SPS-MS3 method, another way to solve the interference effect is through a complementary-ion-based strategy initially proposed by Wühr and colleagues (6). This strategy does not require a higher-order MS/MS scan so it would be beneficial to those labs who do not have the most advanced instrumentations such as Fusion Lumos or Eclipse. The rationale behind this strategy is that when TMT-labeled peptides fragment at the MS2 stage to produce the low-mass reporter ions, corresponding complementary ions are formed as the remanent parts which could be also informative. These complementary ions encode information from different channels in the same way the low-mass reporter ions do. Furthermore, these ions are different for each peptides, which means they are precursor specific. Therefore, complementary ions can provide accurate quantification even in condition that peptides are coeluted and coisolated. They named their method TMTc, and in their first work, the whole envelope of the precursors was isolated, and it resulted in complex data deconvolution and reduced accuracy (6). This limitation was addressed by narrowing the isolation window and optimizing instrument parameters in the following work TMTc+, in which higher accuracy and precision were achieved (7). Similar idea was extended to 8-plex analysis using TMTpro tags, representing the highest sample processing capacity using the complementary-ion-based strategies to date (8). In addition, these studies stimulated the development of specialized tags for employing this strategy, as inefficient complementary ion formation was reported for TMT and TMTpro tags (9-11).

The structures of classic isobaric tags and tags that employ complementary-ion-based strategy are similar (Figure S1). However, to qualify for adopting complementary-ion-based strategy, tags need to leave over a signature equalizer on labeled peptides after fragmentation. Consequently, some classic isobaric tags are not inherently applicable for this strategy including iTRAQ (12) and the N, N-dimethyl leucine (DiLeu) tags developed in our lab previously (13-17), because these tags utilize the carbonyl group as the balancer group, and it will lose as a neutral loss during the MS/MS to produce the reporter ions as a-type fragments, which does not leave much room for additional labeling on the peptide part, making the remnant peptide segments indistinguishable. Here, by introducing a beta-alanine linker, a novel set of isobaric tags, DiLeuC, was designed and made to achieve multiplexed quantitative proteomics by adopting the complementary-ion-based strategy. Moreover, partial channels of DiLeuC tag could also be used as classic isobaric labeling reagents. Therefore, we evaluated and compared the quantification performance among different methods using these channels and reconfirmed that the complementary-ion-based strategy can avoid ratio distortion without sacrificing sensitivity. We also devised a compatible sample preparation workflow, and carefully tuned the instrument parameters, to establish a streamlined complementary-ion-based method using DiLeuC isobaric tags. Eventually, taking the advantages of DiLeuC tag’s structure for facile labeling and efficient fragmentation, we applied complementary-ion-based method in the analysis of MCF7 proteome at the single-cell level for the first time, demonstrating potential utility of this novel method for sensitive multiplexed proteomics.

MATERIALS AND METHODS

Chemicals and reagents

Acetic acid (AA), acetonitrile (ACN), acetone, dimethyl sulfoxide (DMSO), formic acid (FA), methanol (MeOH), and water were purchased from Fisher Scientific (Pittsburgh, PA). Beta-alanine, ¹⁵N-beta-alanine, 3-¹³C-¹⁵N-beta-alanine, 1,2-¹³C₂-¹⁵N-beta-alanine, ¹³C3-beta-alanine, L-leucine, 1-¹³C-leucine, 1-¹³C, ¹⁵N-leucine, formaldehyde, ¹³C-formaldehyde, sodium cyanoborohydride, ¹⁸O water, 1-Ethyl-3-(3-dimethylaminopropyl)-carbodiimide (EDCI), Dicyclohexylcarbodiimide (DCC), hydroxybenzotriazole (HOBt), N-hydroxysuccinimide (NHS), were purchased from Sigma-Aldrich (St. Louis, MO). Trypsin, LysC, LysN and yeast digests were purchased from Promega (Madison, WI). Peptide standards were customer synthesized by GenScript Biotech (Piscataway, NJ). Ethylene bridged hybrid C18 was purchased from PolyLC Inc. (Columbia, MD). Fused silica capillary tubing (i.d., 75 μm, o.d., 375 μm) was purchased from Polymicro Technologies (Phoenix, AZ). All reagents were used without additional purification.

Cell culture and sample preparation

MCF7 cells were grown in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin. All cells were cultured in a 37 °C incubator with 5% CO₂. After harvest, cells were washed with cold PBS (pH 7.4) there times, then homogenized in lysis buffer (8 M Urea, 50 mM tris) using a probe sonicator, and protein concentration was determined using a BCA Protein Assay Kit (Thermo Pierce, Rockford, IL) per manufacturer instructions. Then, 2-5 mg of proteins were reduced with 100 mM DTT at 56°C for 1 hr and alkylated with 200 mM IAA for 30 min in dark under room temperature before quenching with 100 mM DTT. Proteins were digested by LysN at 37°C for 16 hr in a 50:1 (protein: enzyme) ratio. Digests were quenched by lowering the pH below 3 with 10% TFA. 100-200 μg peptides (determined by Thermo Pierce Peptide Assay) were desalted with SepPak C18 solid-phase extraction (SPE) cartridges 1 cc Vac, 50 mg Sorbent per Cartridge, 55 - 105 μm, (Waters, Milford, MA). For yeast protein extract purchased from Promega Corporation, the sample preparation follows the same procedure as stated above. All samples were lyophilized and stored at −80°C until LC-MS/MS analysis.

Single-cell proteomics sample preparation

SCP sample preparation followed by the cellenONE user manual as in the previous report(18). Briefly, MCF7 cells were suspended and washed using PBS three times before sorting. 100 nL lysis buffer (0.2% DDM, 100mM TEAB, 20ng/ul LysN) was dispensed into each well at high humidity. After cell deposition, the proteoCHIP was incubated at 37 °C for 4 hrs. For the DiLeuC labeling step, 100 nL DiLeuC tag in anhydrous ACN was added and incubated for 1 hour at room temperature. Each channel of tags was used for labeling of each row of wells in an array. The labeling reaction was subsequently quenched with 50 nL 0.5% hydroxylamine and 3% HCl. Then, samples were pooled via centrifugation using the proteoCHIP funnel part and transferred into the sample vial before LC-MS/MS analysis.

LC-MS/MS analysis

Two-proteome samples were analyzed on the Orbitrap Fusion Lumos Mass Spectrometer (Thermo Fisher Scientific, Bremen, Germany). The following MS parameters were utilized: MS1 spectra were acquired in the Orbitrap from mass range m/z 400 −200 at 120K resolution and the AGC target was set to 4e5; maximum injection time was set to 50 ms while RF Lens was set to 30%; Dynamic exclusion was employed for 30 s excluding all charge states for a given precursor. Tandem MS was operated at Top 10 data dependent acquisiction mode with first mass at m/z 100 and reolution at 20K with the normalized collision energy of 30. The chromatographic separation was carried out via mobile phase A which consists of 0.1% FA in water and mobile phase B consisting of 0.1% FA in ACN. Peptides were loaded onto a 75 μm × 15 cm homemade column packed with 1.7 μm, 130 Å, BEH C18 material obtained from a Waters UPLC column (part no. 186004661). Emitter tips were pulled from capillary tubing 75 μm i.d. using a model P-2000 laser puller (Sutter Instrument Co., Novato, CA). The LC gradient was set as follows: 3%-10% B (18-31 min), 10%-24% B (31-95 min), and 24%-35% B (95-127 min) with a flow rate of 300 nL/min. SCP samples were analyzed on the trapped ion mobility spectrometry coupled to time-of-flight mass spectrometer (timsTOF MS, Bruker) combined with a high-performance applied chromatographic system ACQUITY UPLC M-class (Waters). The chromatographic separation conditions and MS parameters were kept the same/close as stated above. Peptides were loaded onto an in-house packed column (75 μm × 15 cm homemade column packed with 1.7 μm, 130 Å, BEH C18 material). For the analysis of isobaric labeling peptides with DiLeuC, the mass spectrometer was operated in the parallel accumulation serial fragmentation (PASEF) data combined with dependent acquisition (DDA) mode.

Data analysis

The mass spectra data were searched against either the yeast or human UniProt database (version updated December 2022). A reverse database for the decoy search was generated automatically in MaxQuant (version 1.6.7.0) (19). Enzyme specificity was set to ‘LysN’, and a minimum number of seven amino acids were required for peptide identification. Static modifications included carbamidomethylation of cysteine residues (+57.02146 Da). Dynamic modifications included the oxidation of methionine residues (+15.99492 Da), acetylation of the protein N-terminal. Andromeda search engine was used with FDR<1% on both peptide and protein levels. The quantification of DiLeuC-tag-coupled reporter ions was set in MaxQuant. For MS2 and SPS MS3 based methods, m/z of repoter ions were set as: 117.1318, 116.1345, 115.1312 and 114.1278 for channel 7-1 to 7-4, respectively. All peptide-spectrum matches (PSMs) and peptides found by MaxQuant were exported in the msms.txt and the evidence.txt files. For complementary-ion-based quantification, peak assignments are extracted using a Python script as reported in the previous study (20). Bioinformatic analysis was performed using Perseus and the R language statistical computing environment (21).

RESULTS AND DISCUSSION

Design and Synthesis of 7-plex DiLeuC Tag

While the traditional isobaric labeling methods rely on reporter ions in low mass range for relative quantification, the complementary-ion-based approach counts on larger remanent peptide parts. This requires the isobaric tags designed for using this strategy to have characteristic equalizer groups among different channels. However, our previously developed DiLeu isobaric tags encode the balancer/equalizer in the carboxylic acid and turn into amide group after labeling. During MS/MS, the amide bond breaks followed by the loss of one molecule of carbon monoxide to produce the a-type reporter ions (Figure S2), therefore, DiLeu tags cannot be used as complementary-ion-based isobaric tags directly. The simple way to overcome this obstacle is to introduce a bigger equalizer in the structure which is stable and retained on peptides after fragmentation. Here, we used beta-alanine as it is structurally simple and has various isotopologues available (Figure 1A). By distributing the isotopes ¹³C, ¹⁵N, and ¹⁸O between mass loss region (used to be DiLeu reporter ion) and equalizer group, a set of 7-plex DiLeuC isobaric tag is designed with a moderate overall mass at 333 Dalton (Da) and small incremental mass ranging from 72 Da to 78 Da when used for complementary-ion-based quantification.

Figure 1.

(A) The structure design of 7-plex DiLeuC isobaric tags (for detailed isotopic structures of each channel, please refer to Figure S3); (B) The synthetic route of DiLeuC in five steps.

The synthesis of DiLeuC is straightforward and only requires limited synthetic expertise. It takes five steps in total and the isotopologues are introduced in the former two steps with the remaining three steps streamlined (Figure 1B). N-hydroxysuccinimide (NHS) ester form of DiLeuC was adopted for fast amine coupling, similar to the design of TMT and iTRAQ, instead of using tedious in-situ carboxylic acid activating reagent 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methyl-morpholinium chloride (DMTMM) reported in the original research of DiLeu¹³. The overall yield of this route is around 40% and it achieved a delicate balance between economic cost and analysis throughput comparing with the current methods (Table S1). However, although above 99% isotopic purity has been achieved for channels 7-1 to 7-4, impurity seems to be higher for channels 7-6 and 7-7 (Figure S3), and this is probably due to the oxygen swap during ester coupling in the last step (22). The reduced isotopic purity can be detrimental to quantitative analysis and needs to be further improved. It is also worth noting that the first four channels of DiLeuC can be used as classic isobaric tags as well since the a-type mass loss fragments have distinct m/z ranging from m/z 117 to m/z 114, and for a proof-of-concept method development, we only used the four channels of DiLeuC 7-1 to 7-4 for the following experiments and made side-by-side comparisons among different methodologies including reporter-ion-based, SPS-MS3 and complementary-ion-based strategies.

Evaluation of Tag Performance and Establishment of Sample Preparation Workflow

We tested the labeling efficiency of our DiLeuC first as the complete reaction yield is the prerequisite for achieving accurate quantification. The tag was designed to label all free amine groups including peptide N-terminus and lysine side chains. A peptide standard was incubated with non-isotopic DiLeuC at a 1:5 (w/w) ratio and no unlabeled peptide peak was observable in just 10 min, demonstrating complete labeling with high efficiency (Figure S4). Similar results were also demonstrated on a more complex sample, the tryptic digests of MCF7 cells (Table S2). After database search, the percentage of coverage on the N-terminus and lysine side chain when setting the DiLeuC as either variable or fixed modification closely matched with each other, indicating the complete labeling efficiency. One major challenge for TMT and TMTpro to be used as the complementary-ion-based method is the low efficiency of forming peptide-coupled ions, and it largely drives the development of EASI tag⁹, SOT tag¹⁰, and Ac-Ag tag²⁴. We investigated the fragmentation behavior of DiLeuC-labeled peptides under various normalized collision energies (NCEs). This evaluation confirmed that our DiLeuC-labeled peptides fragmented efficiently at lower collision energies and with lower median NCEs than those required for peptide backbone fragmentation compared with TMT tags (Figure S5). This is in accordance with a previous report concluding that the chemical structure of our DiLeuC breaks much readily than commercial isobaric tags (23).

As stated above, DiLeuC conjugates to peptides N-terminus and lysine side chains, therefore, when conducting classic proteomics sample preparation using trypsin/LysC as the enzyme, both peptide N-terminus and C-terminus will be labeled when the lysine residues are present in peptides. This does not cause a big trouble in classic reporter-ion-based isobaric labeling but would become intractable when adopting the complementary-ion-based approach because mass loss could occur at either N- or C-terminus and it complicates the MS/MS spectrum and precludes peptide identification and complementary ion recognition. We replaced trypsin/LysC with LysN in the new workflow (Figure 2A), as it not only generates similar-length digested peptides compared to trypsin but also creates peptides with consensus N-terminus lysine. In this case, the labeling only happens on the peptide lysine N-terminus at two labels per peptide ratio. The benefit of this workflow is that during the fragmentation, the y ions of the same peptide are identical among different channels and thus facilitating the database searching. We compared the identification performance among different enzymes (Figure S6A) and found that LysN was only slightly underperformed than trypsin but evidently better than LysC. Furthermore, an asymmetric isolation window around the monoisotopic peak at the width of 0.4 Thomson (Th) with an offset of −0.15 Th was set (Figure 2B) to suppress the isolation of the isotopic peak for precursor ions. We screened different isolation window settings and found that the asymmetric isolation window with a narrow width did not markedly reduce the transmission and identification of peptides (Figure S6B).

Figure 2.

(A) Optimized sample preparation workflow using LysN for enzymatic digestion to produce peptides with N-terminus lysine; then through labeling, the DiLeuC tags were only introduced into the peptide N-teminus at two labels per one peptide ratio, in which during tandem MS, these labeled peptides would produce the same y-ions as unlabeled ones as well as the signature b-ions for complementary-ion based quantification; (B) Narrow asymmetric isolation window helps to select monoisotopic peak and facilitates complementary-ion-based quantification.

Quantification Accuracy Assessment on Peptide Standards and a Two-Proteome Model

We validate our design first on two peptide standards using the first four DiLeuC channels, as differentiable low-mass reporter ions can be generated along with the formation of complementary ions and therefore provide a good opportunity to compare the ratio distortion between reporter-ion-based and complementary-ion-based strategies. Two coeluting peptides were selected from previous DiLeuC labeled MCF7 LysN digests experiment. After DiLeuC labeling, the +2 ions of peptide 1 (KEAAAAAL) have close m/z to the +3 ions of peptide 2 (KLQFIFTNIDP), which could be coisolated and cofragmentated together. Peptide 1 was labeled with four channels of DiLeuC at a 3:3:1:1 ratio while peptide 2 was labeled with the reverse ratio at 1:1:3:3. Then both labeled peptides were equally mixed before analysis. As shown in Figure 3, the precursor ions of the two peptides were coisolated and cofragmented together, and the abundant fragments of both peptides were present in one MS/MS spectrum. Since the low-mass reporter ions can be cleaved off from both peptides, the ratios among them were greatly distorted and cannot reflect the underlying real abundances for either of the two peptides. In contrast, the complementary ions of both peptides in the high-mass range preserved the true ratios and did not interfere with each other, exhibiting the interference-resistant nature of this method, resulting in accurate quantitation.

Figure 3.

(A) Representative MS/MS of four channels of DiLeuC labeled two co-fragmented peptides; Pep1: KEAAAAAL; Pep2: KLQFIFTNIDP; (B) Low-mass region showing four reporter ions ranging from m/z 114 to 117; (C) Precursor ions profiles of the two co-fragmentated peptides in full MS before isolation; the doubly charged precursor ion is Pep1 while triply charged ion denoting Pep2; isolation widow is 1 Th; (D) Complementary ions of Pep1; (E) Complementary ions of Pep2.

Next, a larger-scale investigation was conducted on a more complex two-proteome system. In this model, human MCF7 digested peptides were labeled with DiLeuC in ratios of 1:1:1:1 while yeast digested peptides were labeled in ratios of 1:2:5:10. Then, the two labeled samples were combined with five parts of human peptides for every one part of yeast peptides (Figure 4A). The combined mixture is highly complex and simulates the environment where the low abundance peptides have different concentrations among different conditions (yeast peptides) in a background of highly abundant peptides that do not change concentration (MCF7 peptides). When isolating yeast peptides, coisolation, and cofragmentation of MCF7 peptides will tend to bias the measured ratios toward 1, making quantification less accurate. We evaluated and compared three quantification methods: reporter-ion-based, SPS-MS3 and complementary-ion-based strategies using the same sample (Figures 4B and C). As shown in the box plot in Figure 4D, the classic reporter-ion-based method distorts theoretical ratios severely, while the SPS-MS3 strategy achieves accuracy in the sacrifice of sensitivity, resulting in the loss of almost half of the quantified proteins/peptides (Table S3). However, the complementary-ion-based method seems to inherit the advantages of aformentaioned two methods, preserving accurate and precise quantitative information while producing comparable coverage depth compared to the classical reporter-ion-based strategy.

Figure 4.

(A) Two-proteome model consists of MCF7 human peptides as background and yeast peptides as analytes; (B-D) Quantification results comparison using DiLeuC on two-proteome model among three different methods: standard MS2 reporter-ion-based, SPS-MS3, and complementary-ion-based method; (B) Comparison of quantified peptides; (C) Comparison of quantified proteins; (D) Comparison of the ratio compression among three methods.

Application to Sensitive and Accurate Multiplexed Quantitative Single-cell Proteomics

Single-cell proteomics (SCP) can provide unique insights into biological processes by revealing cellular heterogeneity (24, 25). In spite of fast development driven by MS-based method evolvement and advanced instrumentation, SCP is still in a more nascent state compared to genomic and transcriptomic technologies (26). There are several major obstacles awaiting to be addressed such as delicate sample preparation and sensitive peptide/protein quantification. For example, Zhu, Y. et al have constructed a microfluidic sample-preparation platform using nanodroplet processing in one-pot for trace samples (nanoPOTS) to reduce the sample loss (27); Mann lab developed a highly sensitive technique using the Bruker timsTOF SCP platform combined with 384-well plastic plates sample preparation (28); Nikolai and colleagues uilized TMT isobaric labeling to develop a high-throughput method named single cell protEomics (SCoPE-MS) (29). Here, we integrated the commercially available automated minute sample handling platform, the cellenOne, and complementary-ion-based quantification method, to demonstrate the potential application of our DiLeuC isobaric tags in the field of SCP study.

CellenOne is a powerful automated platform for nanoliter sample dispensing and single cell isolation based on piezo driven pulses on an inert glass capillary (30). Combined with their 12 × 16 proteoCHIP, all procedures of sample preparation including single-cell isolation, on-chip enzymatic digestion and isobaric tag labeling can be streamlined (31). We used the first four channels of DiLeuC 7-1 to 7-4 to label MCF7 cells. Briefly, the cells were suspended before being sorted by piezo dispense capillaries. Each well in an arrary was spotted with one single cell, generating an arrary that contains 16 cells in total. The lysis buffer was then transferred uniformally followed by adding the LysN enzyme. Then, each row in an array was dispensed with different channels of DiLeuC tags from 7-1 to 7-4 respectively and the labeled single-cell samples were pooled by centrifugation before LC-MS/MS analysis (Figure 5A). Overall, 252 proteins from 431 peptides based on 1140 PSMs can be identified throughout four replicate injections (Figure 5B). Peptides without complementary-ion intensities in all channels were then filtered. With this procedure, 78 proteins can be strictly quantified. The 4-plex DiLeuC ratios for all quantified peptides were plotted against each other (Figure 5C). Across all channels, the median ratios measure within 10% of the 1:1 ratios, indicating the homogeneuous expression of these proteins at the single-cell level. In addition, GO analysis showed that these quantified proteins were enriched in several biological processes, including DNA replication, protein folding, and chromatin remodeling (Figure S7).

Figure 5.

(A) Streamlined sample preparation for SCP using cellenOne and proteoCHIP; (B) Identification results from sixteen DiLeuC labeled MCF7 cells; (C) Box plot showing quantified peptides across four channels using complementary-ion-based method.

CONCLUSIONS

In summary, we developed a set of novel isobaric tags DiLeuC capable of achieving complementary-ion-based quantification. The outstanding fragmentation performance of DiLeuC was inherited from original DiLeu chemical structure so the ability of generating complementary ions outperformed commercially available products TMT or iTRAQ. We streamlined the sample preparation and optimized the instrument parameters to support complementary-ion-based quantification. LysN was employed for enzymatic digestion to produce N-terminus lysine so that the y-ions from different channels were identical and therefore it facilitated the database searching. The proof-of-concept experiments were performed using peptide standards and a yeast-human two proteome model system, demonstrateing the interference-free characteristic of the complementary-ion based quantitation method. We also integrated automated sample processing platform and our DiLeuC based isobaric labeling method together to perform a simple SCP analysis, indicating the potential of using complementary-ion-based quantification strategy for sensitive proteome profiling at the single-cell level. We anticipate that the concise synthesis, efficient fragmentation, accurate and precise quantification, and capability of achieving classic MS2-reporter-ion-based, SPS-MS3 and complementary-ion-based quantification strategies offered by the DiLeuC tags will benefit a wide range of multiplex quantitative proteomics studies.