Custom PEPscreen™ Peptide Libraries

Biowire Fall 2011 — Screening — microRNA Target Identification

An Economical Approach to Proteome-Level Protein Interaction Analysis

Characterizing protein-protein interactions across the entire proteome has been problematic because the conventional one protein, one experiment strategy is too time consuming and expensive. Rationally designed peptide libraries, however, afford an economical approach to screen protein interactions at an unbiased, proteome level.

Here, we review two high-throughput approaches that combine genomic data and peptide libraries to characterize large numbers of kinase-substrate interactions. Both methods represent natural enzyme substrates using Sigma’s PEPscreen peptide libraries.

PEPscreen Peptide Libraries

Sigma’s PEPscreen peptide libraries are widely used in drug discovery, vaccine development, and protein interaction research. Synthesized using a proprietary, state-of-the-art robotic platform and optimized Fmoc chemistry, Sigma’s affordable PEPscreen libraries are manufactured to the following specifications:

  • Peptide quantity: 0.5–2 mg or 2–5 mg
  • Peptide length: 6–20 amino acids
  • Chemical modifications: Phosphorylation, biotinylation, PEGylation, acylation, etc.
  • Dye labeling: FLC, FITC, Dansyl, Dabcyl, TAMRA, Lissamine, etc.
  • QC: MALDI–TOF mass spectrometry performed on all peptides
  • Format: Supplied lyophilized in 96-well tube rack
  • Delivery: Majority of orders are shipped within 7 business days

back to top Back to Top

Parallel, Label-Free Identification of Kinase Clients with Tandem Mass Spectrometry

Traditionally, protein kinase assays have been low throughput: pairing a single kinase with a single phosphoryl acceptor, followed by laborious analyses to identify the phosphorylated site(s). New instrumentation has enabled proteome-level phosphorylation studies, but limitations remain: surface-phase assays compromise native protein formation and solution-phase methods necessitate development of specific antibodies or radiolabeling. In addition, the presence of multiple phosphate acceptor residues in a peptide or protein can preclude localization of the phosphorylated site.

In 2010, the University of Missouri–Columbia Associate Professor Jay Thelen and colleagues created an approach for identifying client proteins in their natural conformation, without labels, and in a highthroughput manner using quantitative mass spectrometry1.

Thelen’s lab incubated the well-characterized pyruvate dehydrogenase kinase (PDK) with a 46-peptide PEPscreen library, representing 11–20 mer long overlapping sequences of the kinase target protein’s alpha subunit, all of its Ser, Thr, and Tyr residues, and then PDK itself. The peptide substrate and phosphorylated peptide product were separated, identified, and quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS) followed by a newer, more sensitive technique called multistage activation.

Demonstrating the assay’s specificity, tandem mass spectrometry correctly identified the only natural phosphorylation site, a sequence on pyruvate dehydrogenase’s alpha subunit. Subsequent spectral counting quantified the relative abundance of both the substrate and phosphorylated products, enabling the determination of kinase-substrate reaction kinetics. The data followed non-Michaelis-Menten kinetics, suggesting that the phosphorylated peptide acts as a PDK inhibitor. The utility of spectral counting was further validated by observing the negative effect of methionine oxidation on peptide phosphorylation.

A panel of peptides, with directed substitutions at certain residues flanking the phosphorylation site, was generated to characterize “mutagenesis” effects. (Those flanking sequences are highly conserved across species.) LC-MS/MS screening of a 15-peptide PEPscreen panel indicated that retaining phosphorylation activity required certain amino acids near the C-terminus, including serine at a defined position and hydrophobic residues at certain sites.

Thelen and colleagues note that this solution-based method and convenient access to affordable peptide libraries yields a promising approach for rapid identification and quantification of protein-kinase interactions. With libraries constructed from phosphoproteomic screens of cell lines, tissues, and organs, clients of the 500+ human kinases can be screened in a high-throughput manner and substrate interactions described in detail for the first time.

back to top Back to Top

High-Throughput Investigation of Prenylation Substrate Specificity

About a year earlier at the University of Michigan, parallel screening with PEPscreen peptide libraries enabled Professor Carol Fierke and colleagues to identify prenylation of approximately twice as many proteins as had been indentified in the past 20 years2.

Prenylation, the covalent attachment of an isoprenoid lipid near a protein’s C-terminus, is essential to the proper localization and function of many proteins, including members of the Ras and Rho superfamilies of small GTPases implicated in about 30% of cancers. Two enzymes, protein farnesyltransferase (FTase) and Geranylgeranyltransferase type I (GGTase-I), were proposed to prenylate proteins whose C-terminus has a certain four-amino acid “Ca1a2X” motif. “C” refers to a cysteine three residues removed from the C-terminus that is prenylated at the thiol group to form a thioether, “a” refers to any aliphatic amino acid, and “X” refers to a subset of amino acids that are proposed to determine specificity for FTase (Met, Ser, Gln, Ala) or GGTase-I (Leu, Phe).

However, evidence emerged that this model was too simple. Fierke and colleagues set out to improve the predictive model of prenylation substrates, as well as identify novel prenylated proteins potentially valuable as therapeutic targets. To do so, they scanned all human proteins in the human genome database for the “Ca1a2X” motif, identifying some 620 proteins of unknown prenylation status. For 213 of those proteins, Fierke requested that Sigma® Life Science prepare a library of peptides representing those proteins’ C-termini, capped by N-terminal Thr and Lys residues, and appended with a dansyl fluorophore (dansyl-TKCxxx). Control peptides were also prepared for proteins with known prenylation status.

The peptide library was then screened for FTase-catalyzed prenylation, called farnesylation, under multiple turnover ([E] << [S]) and single turnover ([E] > [S]) reaction conditions. Farnesylation was detected by either monitoring the change in the fluorescence of the dansyl group or by radioactivity-based assays using [3H]FPP and autoradiography. Consistent farnesylation between control peptides and parent proteins in vivo, as well as comparable specificity constants (kcat/kmpeptide), confirmed the peptide library’s utility for evaluating farnesylation status.

The screening revealed novel farnesylation of 236 proteins. Surprisingly, two classes of substrates for FTase with distinct reactivity profiles emerged. Screening identified 106 novel FTase substrates farnesylated under multiple turnover conditions. Another 130 peptides were farnesylated only under single turnover conditions. Fierke reasoned that modulation of single versus multiple turnover FTase activity might reflect an unanticipated mechanism for regulating farnesylation, with potential to affect the localization, trafficking, and activity of prenylated proteins within the cell.

In a second study, Fierke requested a library of 88 peptides, designed to apply structure-function analysis to define FTase’s selectivity criteria for the a2 residue of substrate peptides3. Called a positional scanning library, this library’s peptides had their a2 position substituted with each of the 20 amino acids. Correlation of reactivity data and each amino acid’s properties revealed that FTase recognizes both the size and hydrophobicity of the residue at the a2 position, in contrast to predominantly polarity-based recognition at the X residue in the Ca1a2X model. Further comparison indicated the identity of the adjacent X residue also affects a2 selectivity, as well as multiple versus single turnover farnesylation behavior. Such context-dependent substrate recognition suggested that the Ca1a2X model of FTase selectivity predicts only a subset of potential FTase substrates, raising the possibility of novel FTase substrates whose C-terminal sequences do not conform to the canonical Ca1a2X motif.

back to top Back to Top

Synthesizing Custom PEPscreen Peptide Libraries

High-throughput screening with custom peptide libraries has accelerated a range of experiments, from microarray-based antibody detection, antigenicity testing, T- and B-cell epitope mapping, to peptide therapeutic stability and efficacy optimization. You can review publications for these methods, and design your own peptide library for synthesis at

back to top Back to Top


  1. Huang YD, Houston NL, Tovar-Mendez A, et al. A quantitative mass spectrometry-based approach for identifying protein kinase clients and quantifying kinase activity. Anal Biochem. 2010;402:69–76.
  2. Houghland JL, Hicks KA, Hartman HL, et al. Identification of Novel Peptide Substrates for Protein Farnesyltransferase Reveals Two Substrate Classes with Distinct Sequence Selectivities. J Mol Bio. 2010;395(1):176–90.
  3. Houghland JL, Lamphear CL, Scott SA, et al. Context-Dependent Substrate Recognition by Protein Farnesyltransferase. Biochem. 2009;48(8):1691–701.

back to top Back to Top

Related Links