Elsevier

Analytica Chimica Acta

Volume 1069, 3 September 2019, Pages 89-97
Analytica Chimica Acta

Advanced liquid chromatography-mass spectrometry enables merging widely targeted metabolomics and proteomics

https://doi.org/10.1016/j.aca.2019.04.013Get rights and content

Highlights

  • RPLC-HILIC–tailored SRM was proposed and verified as a viable choice for large-scale targeted bi-omics.

  • Online parameter optimization strategy offered higher MS response than Skyline software.

  • Comparative evaluation of 101 metabolites and peptides between two hepatic cell lines was conducted.

  • Both metabolites and proteins governed the inter-cell line differentiation.

Abstract

Either widely targeted metabolomics or quantitative proteomics usually requires unique analytical platform. However, cross-platform omics studies entail higher levels of complexity and uncertainty, and result in a significant obstacle for high throughput assay as well. It is thereby urgent to pursue an integrative approach being capable of merging these two omics terms, namely widely targeted bi-omics. As an eligible analytical tool for large-scale targeted metabolomics, reversed phase liquid chromatography-hydrophilic interaction liquid chromatography–tailored selected reaction monitoring (RPLC-HILIC–tailored SRM) was deployed here to further receive the tryptic peptides as the analytes. Comparative evaluation of metabolites and tryptic peptides, 101 ones in total, between HepG2 and SK-Hep1 cells was conducted as a proof-of-concept. All analytes, regardless of metabolites or peptides, exhibited satisfactory chromatographic behaviors on RPLC-HILIC. Quantitative MS parameters, such as SRM transitions and collision energies (CEs), of either tryptic peptides or metabolites were online optimized in a standard compound-independent manner. It was worthwhile to mention that the signal responses of the peptides-of-choice generated by the optimized CEs were significantly superior to those values suggested by Skyline software. Calibration curves of both metabolites and peptides were constructed by serially diluting a so-called universal metabolome standard (UMS) sample. The quasi-content of each peptide or metabolite was gained according to applying those regressive calibration curves. After subjecting the quasi-content dataset into SIMCA-P software, significant differences took place between the two hepatic cell lines, and not only metabolites but tryptic peptides contributed to the discrimination. Above all, RPLC-HILIC–tailored SRM offered a promising choice towards widely targeted bi-omics attributing to the advantage of simultaneous monitoring metabolites and tryptic peptides.

Introduction

Both proteomics and metabolomics are the integral members of system biology science. Metabolome reflects all endogenous metabolites, while proteome reveals all proteins and peptides. These two terms are directly interconnected because protein levels regulate the metabolic profile of a cell system and metabolites’ concentration usually affects protein expression pattern. Therefore, the integrated approach namely bi-omics that combines the complementary information from both proteomics and metabolomics, should be a robust tool to achieve a more detailed evaluation of physiological and pathological status, especially from the global and overall perspective [[1], [2], [3], [4], [5], [6], [7], [8]]. Moreover, in comparison of non-targeted omics, such as non-targeted metabolomics and shot-gun proteomics, the widely targeted terms, in particular widely targeted bi-omics, are advantageous at offering definitive information [9]. Although the emerging interests from all over the world as well as the rapid development on diverse spectroscopic and spectrometric techniques in recent years, either targeted metabolomics or proteomics still requires unique analytical platform, and afterwards, statistical analysis is separately conducted for either targeted omics dataset. However, cross-platform studies usually enroll higher levels in terms of complexity and uncertainty, and lead to an annoying obstacle for high throughput assay as well. It is therefore urgent to pursue an eligible analytical tool allowing widely targeted bi-omics study.

Among diverse available analytical tools, liquid chromatography–mass spectrometry (LC–MS) has evolved as a state of the art platform for either large-scale targeted metabolomics or bottom-up proteomics because it integrates the separation capacity towards metabolites and/or tryptic peptides from LC and the mass spectrometric detection ability from MS [10,11]. In particular, selected reaction monitoring (SRM) on a triple-quadrupole mass spectrometer (QqQ-MS) has been recommended as the golden standard for quantitative analysis attributing to the merits in terms of sensitivity, selectivity, and dynamic range [12,13]. However, none such approach is available till now to permit simultaneously targeting tryptic peptides and metabolites. Because of the dramatically different scales between the biomolecules-of-interest, there are some technical barriers towards the accomplishment of widely targeted bi-omics. In LC domain, it is challenging for a single column to comprehensively retain tryptic peptides along with hydrophilic and hydrophobic metabolites, and the precision of two dimensional LC, even multi-dimensional LC cannot fully address the demands of omics study either owing to the occurrences of time-triggered valve switching. In MS domain, simultaneously monitoring tryptic peptides (mainly 500 Da < M.W. <2000 Da) and metabolites (M.W. <1000 Da) is a tough task attributing to the significant differences between their tandem mass spectrometric behaviors. It is also difficult to obtain the optimal MS parameters, such as ion transitions and collision energies (CEs) for the quantitative analysis of tryptic peptides when the standard compounds are unavailable. Although some software, e.g. Skyline, is able to recommend ion transitions as well as CEs for the peptides-of-choice, the parameters usually suffer from limited precision and/or sensitivity, because of, to some extent, the inter-platform variations.

Herein, improvements were conducted on both LC and MS domains to tackle with the aforementioned obstacles. We previously proposed a viable workflow [14] as well as a promising analytical platform namely serially coupled reversed phase LC-hydrophilic interaction LC–tailored SRM (RPLC-HILIC–tailored SRM) [15] for large-scale targeted metabolomics via breaking through the bottlenecks such as extensive polarity span and large content range. The term namely quasi-content was demonstrated as well to be feasible for quantitative metabolomics even though the standard compounds are unavailable [15]. Therefore, RPLC-HILIC–tailored SRM was deployed as the backbone platform to receive the fortification of tryptic peptides as the analytes. RPLC-HILIC should facilitate the retention and separation of the tryptic peptides, attributing to the hybrid chromatographic mechanism [15]. Moreover, in response to the detection of tryptic peptides, multiply charged precursor-to-product ion transitions were utilized to extend the molecular weight threshold of QqQ-MS (1250 Da for SCIEX 5500Qtrap-MS) and the optimal CEs along with ion transitions were gained with a well-defined online parameter optimization strategy [16]. Applicability illustration (Fig. 1) was conducted by comparatively assaying a large group of metabolites and tryptic peptides, 101 ones in total, between HepG2 and SK-Hep1 cells. Either cell lysate was divided into metabolite- and tryptic peptide-enriched fractions that were individually subjected for LC–MS measurements. After structural annotation and parameter optimization, RPLC-HILIC–tailored SRM was employed for the quantitative analysis of the combined sample. We envision that the integrative tool could completely address the analytical requirements of widely targeted bi-omics because it enlarged the scales in terms of molecular weight, polarity, and content in comparison of conventional LC–MS. Moreover, the analytical pipeline should also benefit specifically measuring the so-called biomarkers, usually metabolites and tryptic peptides that have been declared in those well-conducted non-targeted omics studies.

Section snippets

Chemicals and materials

The compound library built in our previous article [17] was deployed here for metabolite identity consolidation. Four standard peptides, such as EITALAPSTMK, DSYVGDEAQSK, IDIIPNPQER, and YIDQEELNK were synthesized commercially by Yuanye Biotech Co., Ltd. (Shanghai, China). Moreover, three internal standards, including talatisamine and liquiritin apioside for positive- and negative-ionization favored metabolites, respectively, and bivalirudin corresponding to peptides, were purchased from

Qualitative characterization of analytes-of-interest

Because the metabolome and proteome jointly affected the phenotypic state of a given cell, we sought to develop a method permitting the simultaneous measurement of both metabolites and tryptic peptides. Regarding qualitative characterization, the protein- and metabolite-enriched fractions were separately extracted from the mixed cell lysates; the proteins were then degraded to tryptic peptides; and the metabolome and peptidome were qualitatively profiled using RPLC-HILIC−predictive SRM and

Conclusions

In current study, efforts were devoted to pursue a new integrative approach enabling widely targeted bi-omics because cross-platform omics studies usually entailed higher levels of complexity and uncertainty. RP and HILIC columns were directly hyphenated in series to assign acceptable chromatographic behaviors for both peptides and hydrophilic as well as hydrophobic metabolites. Optimal ion transitions and CEs of all analytes, in particular those 40 tryptic peptides, were online optimized in a

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was financially supported by National Science Fund of China (Nos. 81773875, 81573572, and 81530097), and Young Elite Scientists Sponsorship Program by CAST (No. 2017QNRC001). We would like to thank the technical assistances from Ms. Kerong Zhang from SCIEX Application Support Center (Shanghai, China).

References (51)

  • C. Groer et al.

    Absolute protein quantification of clinically relevant cytochrome P450 enzymes and UDP-glucuronosyltransferases by mass spectrometry-based targeted proteomics

    J. Pharm. Biomed. Anal.

    (2014)
  • Y. Shao et al.

    An integrated strategy for the quantitative analysis of endogenous proteins: a case of gender-dependent expression of P450 enzymes in rat liver microsome

    Talanta

    (2017)
  • X. Luo et al.

    Development of a simple and efficient method of harvesting and lysing adherent mammalian cells for chemical isotope labeling LC-MS-based cellular metabolomics

    Anal. Chim. Acta

    (2018)
  • C.Z. Ulmer et al.

    Optimization of Folch, Bligh-Dyer, and Matyash sample-to-extraction solvent ratios for human plasma-based lipidomics studies

    Anal. Chim. Acta

    (2018)
  • J. Sostare et al.

    Comparison of modified Matyash method to conventional solvent systems for polar metabolite and lipid extractions

    Anal. Chim. Acta

    (2018)
  • R. Liu et al.

    Evaluation of two-step liquid-liquid extraction protocol for untargeted metabolic profiling of serum samples to achieve broader metabolome coverage by UPLC-Q-TOF-MS

    Anal. Chim. Acta

    (2018)
  • H.X. Cao et al.

    Metabolomics-proteomics profiles delineate metabolic changes in kidney fibrosis disease

    Proteomics

    (2015)
  • L. Blanchet et al.

    Data fusion in metabolomics and proteomics for biomarker discovery

    Methods Mol. Biol.

    (2016)
  • C. Hu et al.

    Proteomics and metabolomics analyses reveal the cucurbit sieve tube system as a complex metabolic space

    Plant J.

    (2016)
  • H.I. Wettersten et al.

    Grade-dependent metabolic reprogramming in kidney cancer revealed by combined proteomics and metabolomics analysis

    Cancer Res.

    (2015)
  • M.Q. Wu et al.

    Metabolomics-proteomics combined approach identifies differential metabolism-associated molecular events between senescence and apoptosis

    J. Proteome Res.

    (2017)
  • Y.K. Yau et al.

    Proteomics and metabolomics in inflammatory bowel disease

    J. Gastroenterol. Hepatol.

    (2013)
  • H.X. Zhao et al.

    Integrative proteomics-metabolomics strategy for pathological mechanism of vascular depression mouse model

    J. Proteome Res.

    (2018)
  • J. Zhou et al.

    Strategies for large-scale targeted metabolomics quantification by liquid chromatography-mass spectrometry

    Analyst

    (2016)
  • P. Picotti et al.

    Selected reaction monitoring-based proteomics: workflows, potential, pitfalls and future directions

    Nat. Methods

    (2012)
  • Cited by (28)

    • Widely quasi-quantitative analysis of both metabolites and tryptic peptides in animal-originated medicinal materials: Bufonis Venenum as a case

      2023, Journal of Pharmaceutical and Biomedical Analysis
      Citation Excerpt :

      All 100 analytes were involved for method validation. Assays in regards of linearity, limit of detection (LOD), lower limit of quantitation (LLOQ), intra- and inter-day variation, stability and repeatability, were conducted to justify the applicability by tightly following the instructions proposed in the literature [13]. Noteworthily, calibration samples bearing 8×, 32×, and 128× levels were defined as high, medium, and low concentration levels to complete intra- and inter-day variation evaluations.

    • Widely quasi-quantitative analysis enables temporal bile acids-targeted metabolomics in rat after oral administration of ursodeoxycholic acid

      2022, Analytica Chimica Acta
      Citation Excerpt :

      The qualitative information of all captured BAs, 120 ones in total, was carefully aligned in terms of MS1, MS2, retention time (tR), and plausible identity (Table S3, Supplemental information). Online ER-MS was programmed to optimize the compound-dependent parameters, including SRM ion pairs and CEs, for the other BAs, by following the descriptions in previous articles [36,37]. Details regarding online ER-MS for parameter optimization can be found in Supplemental information.

    View all citing articles on Scopus
    View full text