Evaluation of endogenous peptide stereochemistry using liquid chromatography-mass spectrometry-based spiking experiments

Abstract

Functional d-amino acid-containing peptides are endogenously found throughout nature, generated through non-ribosomal peptide synthesis or through post-translational modification of ribosome-derived peptides. Despite the high functional importance of peptide stereochemistry in biomolecular recognition, identification of amino acid residue isomerization can be challenging using most common peptide characterization workflows. Here, we describe a relatively simple approach to test hypotheses regarding the stereochemistry of endogenous peptides via liquid chromatography-mass spectrometry (LC-MS) spiking experiments with synthetic isotope-labeled peptide standards. Our protocol details the synthesis of ¹³C-labeled synthetic peptide diastereomers via solid-phase peptide synthesis, their purification and characterization, and LC-MS experiments to evaluate putative stereochemical assignments. This approach does not require demanding purification of the endogenous peptide and is compatible with small quantities of endogenous extracts. Overall, this procedure complements current endogenous peptide characterization methods, granting the ability to relatively quickly verify the sequence and stereochemistry in endogenous peptide diastereomers directly in complex biological extracts.

1. Introduction

Although the ribosome generates proteins and peptides using exclusively l-amino acids, polypeptides containing d-amino acid residues are found throughout nature. Perhaps the most well-known d-amino acid-containing peptides are those in the bacterial peptidoglycan (Lovering et al., 2012). However, endogenous antimicrobial peptides, peptide toxins, neuropeptides, and peptide hormones can also contain d-amino acid residues (Checco et al., 2018; Cotter et al., 2005; Erspamer et al., 1989; Heck et al., 1996; Heck et al., 1994; Kessler et al., 2004; Knerr et al., 2013; Kreil et al., 1989; Ling et al., 2015; Livnat et al., 2016; Mast et al., 2019; Süssmuth et al., 2017). In these cases, the d-residue is incorporated either through non-ribosomal peptide synthesis or as an enzymatic post-translational modification following synthesis by the ribosome. Additionally, peptide isomerization has also been found in long-lived proteins, usually generated through spontaneous isomerization of aspartic acid or asparagine residues (Lambeth et al., 2019; Maeda et al., 2015; Tao et al., 2014; Tao et al., 2012). In most known d-amino acid-containing peptides, the d-residue has critical functional consequences, oftentimes greatly increased biological function relative to their all-l-residue analogues (Mast et al., 2021; Ollivaux et al., 2014).

Despite the functional importance of d-residues in most d-amino acid-containing peptides, peptide stereochemistry is often difficult to detect. This is because most modern peptide discovery approaches rely primarily on mass spectrometry (MS)-based peptide identifications, and diastereomeric peptides have identical masses. This makes isomerization “invisible” to most common peptide characterization workflows. Significant effort has been devoted to developing methods to reliably identify peptide diastereomers, including chiral amino acid analysis methods (Bhushan et al., 2004; Fujii et al., 1997a; Fujii et al., 1997b), novel mass spectrometry methods (Adams et al., 2005; Bai et al., 2013; Jia et al., 2014; Koehbach et al., 2016; Tao et al., 2014; Tao et al., 2012), and enzymatic screening approaches (Ewing et al., 2008; Livnat et al., 2016; Maeda et al., 2015). However, many methods of stereochemical confirmation require arduous purification of the endogenous peptide or sophisticated MS/MS methods not readily available to most laboratories. Here, we outline a relatively simple method of confirming peptide stereochemistry through spiking experiments with synthetic isotope-labeled peptide standards (Checco et al., 2018; Mast et al., 2019). This method does not require rigorous purification of the endogenous peptide, making this protocol ideal for relatively quickly assessing putative stereochemical assignments for endogenous extracts with limited sample quantity. This protocol has been written to aid laboratories with little or no prior experience in peptide synthesis, and outlines relatively simple MS and MS/MS analysis that can be performed on most LC-MS instruments readily available in core instrumentation facilities.

2. Overview of the protocol

The method described in this chapter is designed to test hypotheses generated about endogenous peptide stereochemistry (Figure 1). To accomplish these goals, Section 4.1 of this protocol outlines detailed methods for the synthesis, purification, and analysis of ¹³C-labeled synthetic peptide analogues. Following synthesis, Section 4.2 outlines spiking these isotope-labeled peptides into endogenous peptide extracts, followed by analysis using reversed-phase high-performance liquid chromatography (HPLC) coupled to MS. This method relies on the fact that most peptide diastereomers elute at different retention times during reversed-phase HPLC. (Bai et al., 2013; Checco et al., 2018; Fujisawa et al., 1992; Heck et al., 1994; Livnat et al., 2016; Mast et al., 2019; Mast et al., 2021; Morishita et al., 1997) As a result, co-elution of a synthetic standard with the endogenous peptide of interest when spiked into the endogenous peptide extracts provides strong evidence that the correct stereochemical assignment was made (provided the chromatographic method used is able to separate peptide diastereomers under consideration). The use of isotope-labeled peptides is important to unambiguously differentiate synthetic peptides from endogenous peptides, preventing any possibility of unintentional cross-contamination which would lead to false-positive results. For the purpose of this chapter, we show data for naturally occurring diastereomers of the Aplysia californica allatotropin-related peptide (ATRP).(Checco et al., 2018) Two endogenous forms of ATRP exist in the Aplysia central nervous system: the all-l-residue form (L-ATRP), and its diastereomer bearing a d-Phe residue at position 2 (D2-ATRP).

Figure 1.

Overview of the method. In Step 1, peptides suspected to contain d-residues are identified from animal tissues through a variety of methods. The cartoon depicts an extracted ion chromatogram (EIC) displaying two distinct peaks with identical MS1 spectra, a common characteristic of endogenous diastereomers. (Diastereomeric peaks will often also have very similar MS2 spectra.) In Step 2, ¹³C-labeled synthetic peptides corresponding to the putative diastereomers are synthesized by solid-phase peptide synthesis and purified by reverse-phase HPLC. In Step 3, ¹³C-labeled synthetic peptide is spiked into endogenous peptide extracts. Co-elution of endogenous and synthetic peptide adds high confidence to the stereochemical assignment of the endogenous peptide.

3. Before you begin – determining putative positions of isomerization

Before carrying out spiking experiments, the researcher must have already identified candidate DAACPs and putative positions of isomerization. Several methods are available for untargeted identification of candidate DAACPs, which are covered in detail elsewhere.(Jansson, 2018; Livnat et al., 2016; Mast et al., 2019; Mast et al., 2021) Putative positions of isomerization must also be posited because the researcher must synthesize the hypothesized diastereomers for the spiking experiments outlined here. Although it is possible to synthesize and evaluate all possible diastereomers of short peptides (e.g., peptides of 3 or 4 residues), this strategy dramatically loses feasibility as peptide length increases. As a result, preliminary studies must be performed to narrow down the possible isomerized position(s).

There are several ways that one may make putative stereochemical assignments for evaluation without needing to purify large amounts of material for chiral amino acid analysis.(Mast et al., 2021) At the simplest level, homology to a previously discovered DAACP can often point the researcher in productive directions. Developing and testing hypotheses based on sequence homology has proven successful for many discovered DAACPs, including Aplysia ATRP (Checco et al., 2018) and other bioactive neuropeptides from mollusks (Bai et al., 2013; Ohta et al., 1991; Yasuda-Kamatani et al., 1997), DAACPs from frog skin secretions (Kreil et al., 1989), peptide toxins (Buczek et al., 2005b), and diastereomers of crustacean hyperglycemic hormone and vitellogenesis inhibiting hormone (Aguilar et al., 1995; Ollivaux et al., 2009).

In the absence of sequence homology, controlled protease digestion of the endogenous peptide of interest can be an effective technique for narrowing down the position of isomerization. In practice, there are at least two fundamentally different types of enzymatic digestion experiments that have been successful to localize isomerization. The first strategy is to utilize exopeptidases to identify regions of the peptide that are stable to degradation. Because proteases hydrolyze amide bonds near d-residues significantly slower than those near l-residues, exopeptidases dramatically slow their digestion as they approach an isomerized residue. In the presence of an exopeptidase such as aminopeptidase M, endogenous peptides stable to degradation likely indicate the presence of a d-residue near the N-terminus (Checco et al., 2018; Livnat et al., 2016; Mast et al., 2019; Torres et al., 2002). Similarly, putative internal d-residues may be identified by carefully examining the peptide fragments stable to degradation after terminal l-residues are removed through LC-MS analysis (Mast et al., 2019). A second distinct approach is to digest both the endogenous peptide and synthetic all-l-residue diastereomer with an endopeptidase and compare the chromatographic properties of resulting peptide fragments (Buczek et al., 2005a; Soyez et al., 1994). In this approach, peptide fragments that contain the isomerized residue(s) will differ in retention time between the synthetic all-l-version and the endogenous peptide, while peptide fragments that contain only l-residues will be identical between the two. From careful analysis of these digests, regions of the peptide likely to differ between the diastereomers can be identified.

In the ATRP example highlighted in this chapter, the putative location of the d-residue (Phe2) was determined using a combination of homology and exopeptidase digestion (Checco et al., 2018). It is also important to note that endogenous ATRP also displayed LC-MS and LC-MS/MS behavior characteristic of peptides endogenously present as multiple diastereomers: two chromatographically resolved peaks with identical MS parent mass and nearly identical MS/MS fragment spectra (Figure 2). Peptides showing this behavior could be considered as possible DAACP candidates.

Figure 2.

LC-MS and LC-MS/MS analysis of putative peptide diastereomers. Shown here are experimental results examining two distinct peaks from Aplysia californica neuropeptide extracts. (A) Sequence of Aplysia ATRP and relevant calculated monoisotopic masses. Suspected isomerized residue is highlighted in red. (B) Extracted ion chromatogram (EIC) for the ATRP parent mass (m/z = 511.9 ± 0.1, z = +3). (C) Representative MS spectra for Peak 1 and Peak 2. (D) Representative MS/MS fragmentation spectra (CID) for Peak 1 and Peak 2. Figure adapted from (Checco et al., 2018).

4. Step-by-step methods

4.1. Chemical synthesis of isotope-labeled peptide analogues

Once putative stereochemical assignments of peptide diastereomers have been generated, one should synthesize the all-l-residue isomer in addition to any suspected d-amino acid-containing peptides, all containing one or more ¹³C-isotope labels. (More ¹³C-isotope labels provide a greater mass difference between endogenous and synthetic peptides, and reduce overlap of isotopic envelopes. For the example highlighted in this chapter, we utilize synthetic peptides containing two ¹³C atoms.) Note that the protocol described in this chapter differs slightly from the protocol previously reported to generate ¹³C-labeled L- and D2-ATRP (Checco et al., 2018). Both protocols are likely to be successful for most peptides, but we include the protocol described in this chapter because it uses fewer reagents and thus may be more amenable for labs aiming to begin peptide synthesis with little experience. Figure 3 shows an overview of the chemical reactions necessary to assemble peptides via solid-phase peptide synthesis.

Figure 3.

Schematic outline of solid-phase peptide synthesis for (A) C-terminal carboxylic acid using pre-loaded Wang resin, or (B) C-terminal amide using Rink Amide resin. Grey circle indicates resin beads. “P” indicates a side-chain protecting group. DMF washing steps, which take place between each reaction, are omitted in this schematic.

4.1.1. Materials and reagents

• Protected amino acid monomers suitable for Fmoc solid-phase peptide synthesis. Monomers should be protected on their N-terminus with an Fmoc group, and sidechain protected with appropriate base-stable, acid-labile protecting group. We obtain these monomer building blocks from Novabiochem, Chem Impex, or Anaspec. The appropriate monomers for standard synthesis are the following:

  • Alanine: Fmoc-Ala-OH
  • Cysteine: Fmoc-Cys(Trt)-OH
  • Aspartic acid: Fmoc-Asp(OtBu)-OH
  • Glutamic acid: Fmoc-Glu(OtBu)-OH
  • Phenylalanine: Fmoc-Phe-OH
  • Glycine: Fmoc-Gly-OH
  • Histidine: Fmoc-His(Trt)-OH
  • Isoleucine: Fmoc-Ile-OH
  • Lysine: Fmoc-Lys(Boc)-OH
  • Leucine: Fmoc-Leu-OH
  • Methionine: Fmoc-Met-OH
  • Asparagine: Fmoc-Asn(Trt)-OH
  • Proline: Fmoc-Pro-OH
  • Glutamine: Fmoc-Gln(Trt)-OH
  • Arginine: Fmoc-Arg(Pbf)-OH
  • Serine: Fmoc-Ser(tBu)-OH
  • Threonine: Fmoc-Thr(tBu)-OH
  • Valine: Fmoc-Val-OH
  • Tryptophan: Fmoc-Trp(Boc)-OH
  • Tyrosine: Fmoc-Tyr(tBu)-OH
  • Desired Fmoc-protected d-amino acids with appropriate side chain protecting groups.
• Resin for solid phase peptide synthesis. The choice of resin primarily depends on the functionality on the C-terminus of the desired peptide.

  • For C-terminal carboxylic acid, we recommend buying Wang resin preloaded with the C-terminal amino acid. For example, for the sequence GFFD-OH, we would purchase Fmoc-Asp(OtBu)-Wang resin (Novabiochem, 856005).
  • For C-terminal amide, we recommend Rink Amide resin, such as NovaPEG Rink Amide resin (Novabiochem, 855047).

• Appropriate ¹³C-labeled Fmoc-protected amino acid monomers. For the example outlined in this chapter, we used Fmoc-¹³C-Gly-OH (Cambridge Isotope Laboratories, CLM-3639-1 or Anaspec, AS-61572-1), which contains a single ¹³C atom at the carbonyl carbon. However, if the peptide(s) of interest do not contain a glycine residue for replacement, another labeled amino acid will have to be chosen (e.g., Fmoc-¹³C-Ala-OH, Cambridge Isotope Laboratories, CLM-818-PK).

• Benzotriazole-1-yl-oxytris-pyrrolidino-phosphonium hexafluorophosphate (PyBOP, Novabiochem, 8510090100).

• Dimethylformamide (DMF), peptide synthesis or sequencing grade.

• Dichloromethane (DCM).

• Piperidine.

• N,N-diisopropylethylamine (DIEA).

• Trifluoroacetic acid (TFA).

• Triisopropylsilane (TIS).

• 1,2-ethanedithiol (EDT) (required only if peptide contains one or more cysteine residues).

• Ultrapure water.

• Diethyl ether.

• 5 mL peptide synthesis reaction vessels (Torviq, SF-0500).

• Pressure caps for reaction vessels (Torviq, PC-SF).

• System for vacuum aspiration of solvent from peptide synthesis reaction vessels. There are many possible options for this. Our setup uses a vacuum manifold designed for DNA and RNA purification, with modification of chemically-resistant stopcocks. Set-up of apparatus is shown in Figure 4.

  • Vac-Man Vacuum Manifold (Promega, A7231).
  • Chemically resistant stopcocks for vacuum manifold (e.g, polypropylene). (Note that the polycarbonate stopcocks sold with the Vac-Man Vacuum Manifold are not resistant to the solvents used in solid-phase peptide synthesis.)

• Micro stir bars, ~7 mm length (Fisher Scientific, 1451363SIX).

• HPLC-grade H₂O with 0.1% TFA.

• HPLC-grade acetonitrile with 0.1% TFA.

• “Preparatory”-scale HPLC instrument, C18 column, vials, and collection tubes capable of reverse-phase separations and fraction collection.

  • We use C18 columns with internal diameter range of ~9–21.2 mm for “preparatory”-scale purifications, usually running flow rates between 5–15 mL/min.
• “Analytical”-scale HPLC instrument, or alternative method for evaluating purity of peptide after large-scale HPLC Purification.

  • We use C18 column with internal diameter of 4.2 mm running a flow rate of 1 mL/min.

• 10 mL syringe.

• 0.22 μm syringe filters (we use 33 mm diameter units with PVDF membrane).

• Method for removing solvents after HPLC purification. We prefer lyophilization, but other methods such as centrifuge-based vacuum concentrators and rotary evaporation can also work.

• Micropipettes and tips capable of accurately dispensing 1–1000 μL.

• Microcentrifuge tubes (~1.5 mL).

• Conical centrifuge tubes (50 mL).

• Temperature-controlled centrifuge capable of centrifuging 50 mL conical tubes at 3000 × g, 4 °C.

• Mass spectrometer capable of measuring the mass of synthesized peptide. This instrument must have sufficient resolving power and mass accuracy to confirm incorporation of the isotope label. We prefer to use matrix-assisted laser desorption ionization (MALDI)-MS for this step, although methods using electrospray ionization (ESI) will also produce comparable results.

• UV-Visible spectrophotometer with cuvette capable of measuring absorbance at 280 nm (if peptide contains tyrosine or tryptophan residues) or 214 nm (if the peptide does not contain tyrosine or tryptophan residues). (“QS”-grade quartz cuvettes work well for wavelengths examined here. Plastic cuvettes are not usable because they absorb strongly in the wavelength range of interest.)

Figure 4.

Vacuum manifold system for manual solid-phase peptide synthesis. (A) Set-up for manifold system, including manifold, two trap flasks, and a small vacuum pump. (Pump may be replaced with house vacuum line or water aspirator.) (B) Two reaction vessels on stir plate during coupling or deprotection reaction. (C) Reaction vessel being drained on vacuum manifold during washing steps.

4.1.2. Before you begin

• Choose a location for incorporation of ¹³C-labeled amino acid residue. If peptide sequence has glycine or alanine residues, we recommend replacing these residues using Fmoc-¹³C-Gly-OH or Fmoc-¹³C-Ala-OH as the building blocks, respectively, due to relatively low cost. However, other isotope-labeled amino acids can be obtained and used for the same effect.

• Although peptide sequences are always written from N-terminus to C-terminus, chemical synthesis proceeds from C-terminus to N-terminus. Researchers new to peptide synthesis should take special care to ensure that they are synthesizing the peptide in the correct order.

• Choose a scale for your synthesis. We recommend 10-25 μmol scale for most peptides.

• Based on the peptide’s sequence, choose a resin as follows:

  • For C-terminal acid, Wang resin preloaded with the C-terminal amino acid.
  • For C-terminal amide, Rink Amide resin.

• Prepare a solution of 20% piperidine in DMF for deprotection of the Fmoc protecting group. This solution is generally usable for up to 3 days.

• For all washing steps, fill reaction vessel with ≥3 mL of appropriate solvent and then filter using aspiration apparatus such as a vacuum manifold (See Figure 4).

4.1.3. Protocol: Solid-phase peptide synthesis

1. Based on the chosen scale and the resin’s loading capacity, weigh the appropriate amount of resin into reaction vessel with micro stir bar. (For example, for a 25 μmol scale synthesis and resin with 0.70 mmol/g loading capacity, weigh 35.7 mg of resin into the reaction vessel.)

2. Wash resin 3-5 times with DCM, then let resin swell in DCM for at least 30 minutes.

3. Wash resin 3-5 times with DMF.

4. If resin is protected with an Fmoc group, deprotect Fmoc group by adding 1-3 mL of 20% piperidine in DMF and stirring for 15-25 minutes at room temperature. (If resin is not Fmoc protected, move directly to coupling the first amino acid, as in Step 6.)

Note: Extended deprotection reaction times (≥40 minutes) should be avoided as they may lead to unwanted side reactions.

• 5.Wash resin 3-5 times with DMF.
• 6.For coupling amino acid monomer onto the chain, dissolve 4 equivalents (equiv.) Fmoc-protected amino acid, 4 equiv. PyBOP, and 8 equiv. DIEA in 1-3 mL DMF. Vortex to ensure completely dissolved, then add to resin in reaction vessel. Stir reaction for ≥40 min at room temperature.

Note: If using resin that is not preloaded with the C-terminal residue, the first coupling will be the C-terminal residue in the sequence. If using a resin preloaded with the C-terminal residue, the first coupling will be the penultimate residue in the sequence.

Note: Equivalents (equiv.) are based on scale of synthesis, as determined by resin loading. As an example, for a 25 μmol synthesis, 4 equiv. is 100 μmol.

Note: Duration of coupling reactions can vary significantly, and are flexible. We typically allow coupling reactions to proceed for 40-60 minutes, although longer reactions times (up to 12-18 hours) are permissible without major loss in product quality. Note that depending on the specific stir bar used, overnight reactions may cause grinding of the resin into small particles, making resin filtration difficult. If performing overnight coupling reactions, we recommend replacing stirring with agitation/rocking to avoid this.

Note: ¹³C-labeled amino acids and d-amino acids are incorporated in solid-phase peptide synthesis using the same conditions as standard l-amino acids.

• 7.Wash resin 3-5 times with DMF.
• 8.Deprotect Fmoc group by adding 1-3 mL of 20% piperidine in DMF. Stir for 15-25 minutes.

Note: Extended deprotection reaction times (≥40 minutes) should be avoided as they may lead to unwanted side reactions.

• 9.Wash resin 3-5 times with DMF.
• 10.Repeat Steps 6-9 for each amino acid in the sequence, moving from C-terminus to N-terminus.
• 11.After final Fmoc deprotection and wash with DMF, wash resin 3-5 times with DCM. Allow resin to air dry.

Note: Remember to remove the Fmoc group on the N-terminal residue with 20% piperidine in DMF solution at the end of the synthesis. This protecting group is not fully removed during TFA cleavage reaction.

Pause point: Assembly of the peptide can be paused after any step by washing the resin with DMF, followed by DCM and air drying. Peptide on dried resin is relatively stable, and can be stored at 4 °C for weeks–months without significant issue. When resuming synthesis, first swell resin in DCM, wash resin with DMF, then continue with the next step.

• 12.
  Cleave the peptide from the resin and remove sidechain protecting groups using 2-4 mL of cleavage cocktail:

  1. For peptides without cysteine residues, use: 95% TFA, 2.5% ultrapure water, 2.5% TIS.
  2. For peptides with cysteine residues, use: 94% TFA, 2.5% ultrapure water, 2.5% EDT, 1% TIS.

Note: TFA is extremely corrosive, and is a significant hazard. Be sure to only handle only if properly trained and while using the appropriate PPE in a chemical fume hood.

• 13.Allow cleavage reaction to proceed for 3-7 h with gentle agitation. Agitation can be provided via stirring, or plunger and cap can be placed on reaction vessel and allowed to mix on rocker.
• 14.When cleavage reaction is completed, use plunger of reaction vessel to transfer the solution to clean 50 mL conical centrifuge tube (leaving the resin in the reaction vessel). The cleaved peptide is now in this solution.
• 15.Using a gentle stream of air or nitrogen in a fume hood, carefully evaporate the cleavage solution until ~1 mL remains in the 50 mL conical tube.
• 16.Add ~40 mL cold diethyl ether (pre-cooled on dry ice) to precipitate peptide.
• 17.Centrifuge ≥3000 × g, 4°C for 10 min. A visible white/off-white pellet should collect at the bottom of the tube.

Note: Some peptides, particularly those that are very short (<6 residues), may not effectively precipitate during this step. If no visible pellet is seen after centrifugation, we recommend evaporating the ether/TFA mixture via rotary evaporation and continuing directly to HPLC purification.

• 18.Decant supernatant. Dry the remaining pellet with a gentle stream of air or nitrogen.

Pause point: Dried, crude peptide can be stored at −20 °C for extended periods of time (e.g., months) prior to purification.

4.1.4. Protocol: HPLC purification

Synthetic peptides should be purified to remove undesired side products and other unwanted material via reverse-phase HPLC equipped with a UV detector and fraction collector. There are many acceptable ways to purify peptides via HPLC other than the approach outlined in this chapter. Specific mobile phase solvents, columns, gradients, injection volumes, and others can be altered to meet the individual researcher’s needs or laboratory preferences. Here, we walk through our lab’s standard protocol, which we find works well to obtain peptides of high purity (>95%) for analysis in spiking experiments. We generally use columns with C18 stationary phase and internal diameter from 9–21.2 mm. An example of HPLC purification for ¹³C-D2-ATRP is shown in Figure 5.

Figure 5.

HPLC purification of ¹³C- D2-ATRP (Sequence = [G*][dF]RLNSASRVAH[G*]Y-NH₂, where [dF] = d-Phe, and [G*] = ¹³C-Gly containing one ¹³C atom). Purification gradient was 18-28% B over 20 min for purification of major peptide peak (21.2 mm ID column, 12 mL/min flow rate). Peptide was purified over multiple injections similar to the one shown. Black solid line indicates absorbance at 220 nm, while the blue dashed line indicates solvent gradient. Red bar indicates region collected via fraction collector for purification.

One important consideration for HPLC purification and drying is to avoid any possibilities of cross contamination of biological samples with the synthetic peptide. Thus, it is recommended to purify the synthetic peptide using instruments and columns that will not come in contact with biological extracts.

1. Dissolve crude peptide in 2-5 mL of 50% water/acetonitrile + 0.1% TFA.

Note: It is critical that the product is fully dissolved in solution at this step. If product is found to be insoluble in the above solution, the researcher may wish to try other ratios of water:acetonitrile, or alternative HPLC-compatible solvents such as dimethyl sulfoxide (DMSO).

• 2.Filter solution through 0.22 μm syringe filter to remove insoluble particulates.
• 3.
  To determine the optimal gradient for peptide purification, inject a small portion of sample (e.g., 100 μL) onto HPLC instrument using the following method:

  1. Solvent A = water + 0.1% TFA; Solvent B = acetonitrile + 0.1% TFA
  2. Flow rate = 12-15 mL/min on C18 column (21.2 mm internal diameter)
  3. Starting composition = 5% B. Gradient = 0 – 50 min, 5–55% B; 50–50.1 min, 55–90% B; 50.1–55 min, 90% B; 55–55.1 min, 90–5% B; 55.1–60 min, 5% B.

Note: If peptide looks to be of relatively high purity at this stage (i.e., one major peak without closely eluting products), one may choose to purify using exclusively the above gradient (1% B/min), performing several larger volume injections and collecting the major peak via fraction collector. However, if chromatogram shows multiple peaks or side products closely eluting to major peak, it is recommended to purify on a shallower gradient (0.5% B/min), as described below.

Note: It is often helpful to collect all major peaks via fraction collector at this stage and analyze by mass spectrometry to determine which peak corresponds to the desired product. However, if easy access to a mass spectrometer is not available, one can proceed by collecting all major peaks at this stage and determining which peak is the desired product during subsequent analysis.

• 4.Based on the results from injection on the 5-55% B gradient described above, determine the approximate % B in which the product peptide elutes. Purify peptide using a 0.5% B/min gradient centered around the % B where the peptide elutes.

Note: In the example in Figure 5, ¹³C-D2-ATRP eluted at ~23% B during the initial 5-55% B gradient injection, so it was purified using an 18-28% B gradient with the following method: Solvent A = water + 0.1% TFA, Solvent B = acetonitrile + 0.1% TFA. Flow rate = 12 mL/min on C18 column (21.2 mm internal diameter). Starting composition = 18% B. Gradient = 0 – 20 min, 18–28% B; 20–20.1 min, 28–90% B; 20.1–25 min, 90% B; 25–25.1 min, 90–18% B; 25.1–30 min, 18% B.

Note: It is recommended to purify over several injections of 300-1000 μL, depending on the yield and current volume of the crude peptide solution. One wants to be sure to inject a small enough volume to ensure adequate peak separation and avoid column overloading, but inject enough each run to ensure purification of crude peptide in a reasonable amount of time.

Note: When calculating the % B at which the peptide elutes, take into account the system’s dwell/dead times, which delay elution from the column relative to when pumps run at a given composition.

• 5.Combine collected fractions corresponding to the peak of interest and remove solvent via lyophilization, vacuum concentrator, or alternative method.

Pause point: Most dried purified peptides can be stored at −20 °C for extended periods of time (years) without significant degradation. Peptides containing cysteine or methionine residues may oxidize over time, so it is recommended to confirm purity of these compounds immediately before use via HPLC and repurify if necessary.

4.1.5. Analysis of synthetic peptide

After synthesis and purification of a synthetic, isotope-labeled peptide, it is recommended to perform three analytical tests: 1) UV absorbance analysis to quantify peptide yield 2) mass spectrometry analysis to confirm peptide identify, and 3) analytical HPLC analysis to assess peptide purity. These steps can be performed in any order. Following these tests, the synthetic peptide may be aliquoted and dried for long-term storage, or diluted and used immediately for spiking experiments.

1. Dissolve purified peptide in 1-3 mL of HPLC-grade water and determine concentration by UV absorbance on a UV-Visible spectrophotometer.

   1. If the sequence contains a tyrosine or tryptophan residue, it is recommended to calculate concentration using absorbance at 280 nm (Conibear et al., 2012; Gill et al., 1989). At 280 nm, the extinction coefficient of the purified peptide can be calculated as:

      $${\epsilon }_{280nm}=(no.ofTrp\times 5500)+(no.ofTyr\times 1490)+(no.ofdisulfidebridges\times 125)$$
   2. If the peptide sequence contains no tyrosine or tryptophan residues, you may also calculate the concentration based on the absorbance at 214 nm (Buck et al., 1989; Conibear et al., 2012). At 214 nm, the extinction coefficient of the purified peptide can then be calculated as:

      $$\begin{gathered} {\epsilon }_{214nm}=([no.ofAminoacids-1+no.ofAsn+no.ofGln]\times 2846)+(no.of \\ Phe\times 7200)+(no.ofHis\times 6309)+(no.ofTrp\times 22735)+(no.ofTyr\times 5755) \end{gathered}$$

Note: The first term in the above equation ([no. of Amino acids – 1 + no. of Asn + no. of Gln]) is a calculation of the number of amide bonds in the peptide, and is applicable for peptides bearing C-terminal carboxylic acids. For peptides bearing C-terminal amides, the first term should be modified to [no. of Amino acids + no. of Asn + no. of Gln].

Note: We recommend that researchers calculate concentration based on A₂₁₄ only when the peptide contains no Trp or Tyr residues, as we have found peptide concentrations based on A₂₈₀ to be more accurate than those based on A₂₁₄. If peptide sequence does not have any Trp or Tyr residues, measuring peptide mass using accurate analytical balance is also acceptable.

• c.Calculate the concentration of peptide stock using a rearranged form of the Beer-Lambert law: c = A/(ε×l), where c is concentration (in M), A is measured absorbance at the predetermined wavelength, ε is calculated extinction coefficient at the predetermined wavelength (in M⁻¹ cm⁻¹), and l is path length (in cm).

Note: If purified peptide is found to be insoluble in pure water, it may be helpful to try dissolving in water + 0.1% TFA (or formic acid), or in a solution of 25-50% acetonitrile/water.

Note: It is recommended to prepare several dilutions of peptide (e.g., 1:5 – 1:100) and consider only those that give absorbance values between 0.1–1 at the wavelength being monitored. Absorbance values significantly outside of this range may not lead to accurate estimations of peptide concentration.

• d.This peptide solution with determined concentration can then be used directly for follow-up experiments, or peptide can be aliquoted into known amounts and dried via lyophilization or vacuum concentrator.

• 2.
  Analyze purified peptide by analytical-scale HPLC to evaluate relative purity (Figure 6A). Use integrated peak areas to determine relative purity. If below 90% purity, we recommend repurifying the peptide via HPLC. We generally evaluate purity using the method outlined below:

  1. Solvent A = water + 0.1% TFA, Solvent B = acetonitrile + 0.1% TFA
  2. Flow rate = 1 mL/min on C18 column (4.6 mm internal diameter)
  3. Starting composition = 5% B. Gradient = 0 – 50 min, 5–55% B; 50–50.1 min, 55–90% B; 50.1–55 min, 90% B; 55–55.1 min, 90–5% B; 55.1–60 min, 5% B.
• 3.Analyze purified peptide by mass spectrometry to determine exact mass of peptide. Be sure to pay careful attention to isotopic distribution to confirm the desired incorporation of ¹³C. Analyze synthetic peptide by LC-MS and LC-MS/MS using collision-induced dissociation (CID) in data-dependent acquisition (DDA) mode to determine MS2 fragmentation pattern on LC-MS system. See Figure 6 and Figure 7 for examples.

Figure 6.

Analytical characterization of synthesized and purified ¹³C- D2-ATRP (Sequence = [G*][dF]RLNSASRVAH[G*]Y-NH₂, where [dF] = d-Phe, and [G*] = ¹³C-Gly containing one ¹³C atom). (A) Analytical-scale purity check (4.6 mm ID column, 1 mL/min flow rate). Black solid line indicates absorbance at 220 nm, while the blue dashed line indicates solvent gradient. (B) MALDI-MS analysis of purified peptide. Only the monoisotopic peak of the product is labeled. (C) Zoom of Panel B to examine isotopic distribution of synthetic peptide. Calculated monoisotopic [M+H]⁺ = 1535.81 (z = +1) for peptide with two ¹³C-labeled glycine residues. Calculated monoisotopic [M+H]⁺ = 1533.80 (z = +1) for peptide with no ¹³C label.

Figure 7.

Sequences and MS/MS (CID) spectra for endogenous D2-ATRP (top spectrum, black) and synthetic ¹³C-D2-ATRP (bottom spectrum, red). Red [G*] indicates ¹³C-labeled Gly, and [dF] indicates d-Phe residue. The spectrum for endogenous D2-ATRP is the same as that shown in Figure 2D, replotted here for comparison. Figure adapted from (Checco et al., 2018).

Note: Our workflow includes a mass check by MALDI-MS as part of the synthetic process to verify correct product produced (Figure 6), followed by more detailed MS and MS/MS analysis on LC-MS instrument (Figure 7 and Figure 8). Our choice for a first check by MALDI-MS is due to the relative speed and ease of MALDI-MS analysis relative to LC-ESI-MS, which requires longer sample preparation and run times. However, analysis by MALDI-MS is not necessary if thorough LC-MS and LC-MS/MS analysis is performed on synthetic peptide.

Figure 8.

Example spiking experiment to determine endogenous stereochemistry of allatotropin-related peptide (ATRP) diastereomers in Aplysia cerebral ganglia peptide extracts. (A) “No spike” control showing endogenous L- and D2-ATRP diastereomers separable by LC-MS. Data is the same as that displayed in Figure 2, replotted here for comparison. (B) Endogenous extracts are spiked with ¹³C-labeled D2-ATRP. (C) Endogenous extracts are spiked with ¹³C-labeled L-ATRP. The left plot in each panel shows the Total Ion Current (TIC) chromatogram for each sample. The center plot in each panel shows two overlaid extracted ion chromatograms (EICs) for endogenous ATRP (black, m/z = 511.9 ± 0.1, z = +3) and ¹³C-labeled synthetic ATRP (red, m/z = 512.6 ± 0.1, z = +3). Endogenous ATRP is expected to have low signal in the m/z = 512.6 channel due to endogenous ¹³C isotope abundance. The right plot in each panel shows a representative MS1 spectrum for the peak at ~12 min (indicated by arrow in the center panel). Bars indicate windows used for generation of EICs in center panel for endogenous (grey) and ¹³C-labeled synthetic (red) peptides. Figure adapted from (Checco et al., 2018).

4.2. LC-MS-based spiking experiments

With the isotope-labeled peptides in hand, LC-MS-based spiking experiments can provide strong evidence to support stereochemical assignments of endogenous peptides extracted from biological tissues. Individually spiking each suspected isomer into tissue extracts allows for rigorous comparisons of LC retention times, which generally differ for peptide diastereomers. Isotope labeling of synthetic standards ensure that these peptides will not be misidentified as endogenous peptides, avoiding the possibility of unintentional cross contamination leading to false positive results.

4.2.1. Materials and reagents

• Peptide extracts from tissue of interest. There are a number of different methods to extract peptides from animal tissues (or cells). The researcher should explore several extraction methods to find what works best for peptide(s) of interest. Two possible methods are highlighted below:

  • Three-stage extraction consisting of consecutive homogenizations in boiling water, acidified acetone (40:6:1 acetone:water:concentrated HCl, v/v/v), and acidified water (0.25% acetic acid in water).(Lee et al., 2010; Yang et al., 2017). At each stage, homogenized tissue is centrifuged and supernatant removed and dried by vacuum concentrator. Extracts from each stage can be combined or analyzed separately.
  • Homogenization in acidified acetone (40:6:1 acetone:water:concentrated HCl, v/v/v) (Bai et al., 2013; Checco et al., 2018; Livnat et al., 2016), followed by high-speed centrifugation, removal of supernatant, and drying of supernatant by vacuum concentrator.

• Isotope-labeled peptides (from Section 4.1)

• Micro BCA kit (ThermoFisher, 23235), along with UV-Visible spectrophotometer or plate reader capable of measuring absorbance at 562 nm.

• Method for performing C18 solid-phase extraction (SPE) on peptide extracts. We recommend one of the following:

  • Pierce 8 mg C18 spin columns (ThermoFisher, 89870) for <30 μg total peptide extract.
  • HyperSep C18 Cartridges for ≥30 μg total peptide extract: 50 mg (ThermoFisher, 60108-390), 100 mg (ThermoFisher, 60108-302), or 500 mg (ThermoFisher, 60108-304).

• HPLC-grade H₂O

• HPLC-grade acetonitrile

• HPLC-grade methanol

• HPLC-grade formic acid

• Micropipettes and tips capable of accurately dispensing 1–1000 μL.

• Microcentrifuge tubes (~1.5 mL).

• Temperature-controlled centrifuge capable of centrifuging microcentrifuge tubes at >10,000 × g, 4 °C.

• Nano-LC instrument equipped with reversed-phase column/solvents and interfaced with a high-resolution mass spectrometer (e.g., Q-TOF, Orbitrap, etc.). For the example shown in this chapter, a Thermo Dionex Ultimate 3000 RSLC with a nanoflow selector coupled to a Bruker Impact HD UHR-QqTOF mass spectrometer with a CaptiveSpray Nanosource was used. However, many alternative instruments and configurations will also work for this purpose. The LC instrument must be able to reliably separate peptide diastereomers of interest, and the MS instrument must have sufficient resolving power and mass accuracy to definitively determine if a peptide signal is arising from endogenous peptide or isotope-labeled synthetic peptide.

4.2.2. Protocol

1. Dissolve dried peptide extracts in 20-500 μL of 5% acetonitrile/H₂O + 0.1% formic acid.

Note: A solution of 5% methanol/H₂O + 0.1% formic acid may also be used if using a methanol/water solvent system for SPE below.

• 2.Determine total peptide concentration using MicroBCA assay kit, according to manufacturer’s instructions. If total peptide content < 30 μg, proceed to SPE with 8 mg C18 spin columns. If total peptide content ≥30 μg, proceed to SPE with HyperSep C18 cartridges (≥50 mg).
• 3.Desalt peptide extracts by SPE, using column/cartridge choice identified in previous step, and following manufacturer’s instructions. Either H₂O/acetonitrile or H₂O/methanol solvent systems generally work well for desalting of endogenous peptides. These two different solvent systems may retain somewhat different subsets of peptides, and we recommend researchers explore both solvent systems when working with particular peptide extracts for the first time to determine what is most appropriate for their system.

Note: Solvent composition for washing and elution steps during SPE may be modified to fit needs of specific sample. For the ATRP example highlighted in this chapter, equilibrations, sample loading, and washing steps were performed in 100% H₂O + 0.1% formic acid (no organic solvent) to minimize analyte loss due to the high polarity of the target peptides.

• 4.Dry eluted, desalted peptide extract (e.g., using vacuum concentrator).

Pause point: Dried, desalted peptide extracts can be stored at −20 °C for extended periods of time (usually at least several months).

• 5.Resuspend dried peptide extracts in 10-50 μL of LC-MS solvents matching the starting composition of your LC-MS run (e.g., 3% acetonitrile/water + 0.1% formic acid).
• 6.Determine concentration of resuspended peptide extracts using MicroBCA assay kit, according to manufacturer’s instructions. Determining the total concentration of resuspended peptide extracts is important to control the amount of total peptide injected onto the nano-LC column.
• 7.Dilute peptide extracts to the appropriate concentration for sample injection and split into several microcentrifuge tubes for spiking experiment. One tube of sample is needed for each peptide spike condition, along with one tube for “No Spike” control sample.
• 8.For each peptide spike condition, spike in purified ¹³C-labeled peptide at final concentration which gives signal of roughly similar intensity to endogenous peptide being examined. This concentration should be predetermined for each system using pilot experiments to evaluate the endogenous peptide signal for an injection of a given sample and separate injections of varied concentrations of ¹³C-labeled synthetic peptides. In the examples shown in Figure 8, ¹³C-labeled peptides are spiked into peptide extracts at a final concentration of ~75 nM, and 5 μL of this spiked solution is injected onto the LC-MS instrument.

Note: There are many possible LC gradients that will suffice for this analysis. The most important aspect of choosing an LC gradient for this experiment is the ability to reliably separate the diastereomeric peptides of interest. The experiment shown in Figure 8 uses the following gradient:

1. Solvent A = 95% water, 5% acetonitrile, 0.1% formic acid; Solvent B = 95% acetonitrile, 5% water, 0.1% formic acid; loading solvent (for pre-column trap) = 98% water, 2% acetonitrile, 0.1% formic acid, 0.01% TFA.

2. Flow rate = 300 nL/min

3. Starting composition = 4% B. Gradient = 0 – 3 min, 4% B; 3–90 min, 4–50% B; 90–93 min, 50–90% B; 93–100 min, 90% B; 100–110 min, 90–4% B; 110 – 120 min, 4% B.

Note: When injecting endogenous extracts and synthetic peptides separately, researchers should carefully examine the MS2 spectra for each peptide. Although the masses of the ¹³C-labeled synthetic peptides will differ from the endogenous peptide, the general fragmentation pattern should be nearly identical for the endogenous and synthetic peptides (see Figure 7). If MS/MS fragments differ significantly between endogenous peptide and the ¹³C-labeled peptide, this likely indicates the endogenous peptide sequence was incorrectly determined.

• 9.Prepare and inject a “No Spike” control sample by diluting peptide extracts with the same volume of solvent (no peptide) used for each spike condition.
• 10.
  Data analysis is performed by creating extracted ion chromatograms (EICs) for both the endogenous peptide and the ¹³C-labeled synthetic peptide. Following this, the MS1 and MS2 spectra for each putative diastereomer should be closely examined to determine if putative stereochemical assignments match those of the endogenous peptides. It is important to look for the following:

  1. The “No Spike” control sample should show putative peptides with predicted endogenous isotope ratio (Figure 8A).
  2. In each spike condition, EIC of endogenous peptide and spiked ¹³C-labeled synthetic peptide should co-elute (Figure 8B and 8C). Failure of synthetic ¹³C-labeled peptide to co-elute with endogenous peptide indicates an incorrect assignment of endogenous peptide sequence or stereochemistry.
  3. Inspection of MS1 spectra for spiked peptide peak should show a mixture of isotope patterns for endogenous peptide and ¹³C-labeled synthetic peptide. The non-spiked diastereomer should show only the endogenous isotope pattern (Figure 8B and 8C).

Note: For peptides with masses higher than ~1000 Da, the most intense m/z peak will often be an ion with z > +1 using electrospray ionization (ESI). We suggest generating EICs for all m/z values associated with the peptides of interest. Further analysis can then be performed on the most intense m/z ion for the peptide of interest.

4.2.3. Pre-spiking control experiment

Performing the spiking experiment described above helps to determine the stereochemistry of endogenous peptides present in an extract after sample processing. However, it is possible that isomerization may result as an artifact of sample preparation. In order to evaluate this possibility, it is recommended to also perform a “pre-spiking” experiment (Figure 9). In this control experiment, ¹³C-labeled synthetic all-l-residue peptide is spiked into extraction solution before endogenous peptide extraction from tissue. The tissue is then processed and analyzed by LC-MS. If spontaneous isomerization occurs as part of the extraction process, the presence of isomerized ¹³C-labeled peptide should be observable by inspecting the isotopic distributions of each peptide peak in LC-MS. (Note that the amount of synthetic peptide that is spiked for this experiment will need to be optimized based on instrument sensitivity and endogenous peptide abundance. The experiment shown in Figure 9, 10 pmol ¹³C- L-ATRP was spiked into 500 μL acidified acetone prior to tissue extraction.)

Figure 9.

Pre-spiking control experiment to evaluate possible isomerization as a result of peptide extraction and sample preparation protocols. (A) Overview of experiment. ¹³C-labeled all-l-residue diastereomer is spiked into extraction solution before tissue is added. Tissue is then processed as usual and peptide extracts analyzed by LC-MS. Isotope label should be found in all-l-residue peptide, but absent from d-amino acid-containing diastereomer. Evidence of the isotope label in d-amino acid containing peptide indicates spontaneous isomerization during extraction and sample preparation. (B) Representative overlaid EICs for endogenous ATRP (black, m/z = 511.9 ± 0.1, z = +3) and ¹³C-labeled synthetic ATRP (red, m/z = 512.6 ± 0.1, z = +3) following pre-spike with ¹³C-L-ATRP. (C) Representative MS1 spectra for each peak shown in panel B. The isotope label is present in L-ATRP peak, but no evidence of ¹³C-labeled synthetic peptide is found in the D2-ATRP peak. Figure adapted from (Checco et al., 2018).

5. Conclusions

In order to validate the stereochemical assignment of peptide diastereomers using this protocol, three conditions must be met. First, putative diastereomeric peptides must have identical m/z values in MS1. Second, putative diastereomeric peptides should have nearly identical MS2 fragmentation patterns. Although it is known that changes in stereochemistry can alter the fragmentation of peptide diastereomers (especially fragment intensity) (Adams et al., 2005; Bai et al., 2013; Koehbach et al., 2016; Tao et al., 2014; Tao et al., 2012), the majority of fragments are usually identical between diastereomers. Observed MS2 fragmentation pattern for synthetic peptide should match that of the endogenous peptide being assigned. Significant difference in MS2 fragment patterns indicates two peptides have different sequences, even if they share the same parent mass. Third, possible peptide diastereomers must be separable during the LC-stage of the LC-MS run, and spiked synthetic standards must co-elute with peptides of interest. As demonstrated in Figure 8 and in many other cases (Bai et al., 2013; Checco et al., 2018; Fujisawa et al., 1992; Heck et al., 1994; Livnat et al., 2016; Mast et al., 2019; Mast et al., 2021; Morishita et al., 1997) peptide diastereomers often elute at different retention times during reverse-phase chromatography. However, it is important to note that this may not be the case for all possible peptide isomers, and optimization of LC conditions may be required to reliably separate diastereomers. Furthermore, for most peptides it is generally not feasible to synthesize and spike every possible diastereomer. Thus, in most cases additional experiments such as homology studies, protease digestions, or others will be required prior to spiking experiments to narrow down possible positions of isomerization, as discussed in Section 2.