Fernando Benavente

Evaluation of on‐line solid phase extraction‐capillary electrophoresis‐electrospray‐mass spectrometry for the analysis of neuropeptides in human plasma

In this study, the suitability of SPE coupled on‐line to CE‐electrospray‐MS (SPE‐CE‐ESI‐MS) was evaluated for the analysis of neuropeptides in human plasma. First, CE‐ESI‐MS was investigated and a sample pretreatment based on precipitation with ACN was used for cleanup of plasma samples. The main quality parameters were determined and were consistent with those previously obtained for the analysis of standard mixtures, e.g., the LODs were of around 1 μg/mL. SPE‐CE‐ESI‐MS employing microcartridges containing a C18 stationary phase was explored in order to decrease the LODs. A double‐step sample cleanup pretreatment consisting of precipitation with ACN and ultrafiltration through 3000 Da MW cut‐off membranes was needed to prevent the microcartridge saturation. Repeatability, reproducibility, linearity, and LODs of the SPE‐CE‐ESI‐MS method and the durability of the microcartridges were studied. The LODs were markedly improved, ranging between 10 and 0.1 ng/mL depending on the peptide.

Analysis of intact erythropoietin and novel erythropoiesis-stimulating protein by capillary electrophoresis-electrospray-ion trap mass spectrometry

Recombinant human erythropoietin (rHuEPO) and novel erythropoiesis‐stimulating protein (NESP) were analyzed by CE‐ESI‐MS using an IT as analyzer. The IT parameters were optimized by direct infusion of solutions of different intact proteins (myoglobin, transferrin, α1‐acid glycoprotein and fetuin) with different degrees of glycosylation (from 0 to 35% w/w). Two physically adsorbed capillary coatings from UltraTol™ Pre‐Coats (low normal (LN) and high reverse (HR)) were evaluated for the separation of rHuEPO and NESP glycoforms by CE‐ESI‐IT‐MS. The results obtained with the neutral LN coating suggest that an IT mass spectrometer enables identification of the main glycoforms of a complex glycoprotein such as rHuEPO. Although LN provided acceptable glycoform resolution for rHuEPO, the separation obtained for NESP was less significant due to the higher microheterogeneity of this glycoprotein. Reproducibility studies confirmed the lack of stability and bleeding of the LN coating, which caused problems with MS detection, such as a dramatic loss of sensitivity and the presence of peaks in the mass spectra corresponding to molecular ions in the coating. In contrast, the cationic HR coating gave faster but poorer glycoform separations due to the presence of an anodal EOF. However, the positive charge of the coating provided enhanced hydrolytic stability, making it more suitable than the LN coating for the on‐line MS coupling.

Analysis of recombinant human erythropoietin glycopeptides by capillary electrophoresis electrospray-time of flight-mass spectrometry.

Capillary electrophoresis electrospray-mass spectrometry was used to detect and characterize the great variety of O- and N-glycopeptide glycoforms of recombinant human erythropoietin (rhEPO) using an orthogonal accelerating time-of-flight mass spectrometer to obtain their exact molecular masses (CE-TOF-MS). rhEPO was digested with trypsin and Glu-C and analyzed by CE-TOF-MS to detect O(126), N(83), N(24)-N(38) and N(24) and N(38) glycopeptide glycoforms, respectively. Neuraminidase was first used to enhance the detection of the glycopeptides and detect all possible glycoforms contained in each glycosylation site. O(126) and N(83) glycopeptides were extensively characterized. Twelve sialoforms corresponding to 5 different glycoforms were detected in N(83), and for the first time, a sulfated sialoform of this glycopeptide was also detected. In the case of O(126), different sialoforms with different types of sialic acids (Neu5Gc and Neu5Ac) were detected and an estimation of the relative percentage of Neu5Gc versus Neu5Ac was also carried out for this glycopeptide. N(24) and N(38) glycosylation sites were also characterized by CE-TOF-MS after Glu-C digestion and these results permitted to rule out some glycan combinations for N(24)-N(38) glycopeptide glycoforms. This study provided a reliable glycopeptide map of rhEPO and may be regarded as an excellent starting point to analyze rhEPO glycopeptides in biological fluids and detect the use of this hormone in sports.

Transient isotachophoresis in on-line solid phase extraction capillary electrophoresis time-of-flight-mass spectrometry for peptide analysis in human plasma†

In this study, we evaluated the combination of transient isotachophoresis with on‐line solid‐phase extraction capillary electrophoresis time‐of‐flight mass spectrometry (SPE‐tITP‐CE‐TOF‐MS) to improve sensitivity of peptide analysis, using several opioid peptides as model compounds. First, standard solutions were analyzed in order to establish the tITP‐CE methodology using UV and TOF‐MS detection. The volume and composition of the leading and terminating electrolytes (i.e. LE and TE) for an efficient tITP were investigated to obtain optimum detection sensitivity and electrophoretic separation. In the best cases, LODs in tITP‐CE‐TOF‐MS were tenfold better than those obtained in CE‐TOF‐MS (i.e. 5 versus 50 ng/mL). Afterwards, the tITP‐CE‐TOF‐MS methodology was adapted to perform SPE‐tITP‐CE‐TOF‐MS. Repeatability, linearity and LODs were investigated and compared to the values obtained by SPE‐CE‐TOF‐MS. Furthermore, human plasma samples fortified with the opioid peptides were analyzed in order to show the potential of SPE‐tITP‐CE‐TOF‐MS for peptide analysis in biological fluids. The LODs attained in standard solutions and plasma samples for some of the studied peptides (i.e. 0.01 and 0.1 ng/mL, respectively) were tenfold better than those obtained in SPE‐CE‐TOF‐MS, proving the enhanced sensitivity that could be achieved when both on‐line preconcentration approaches were combined together.

Analysis of recombinant human erythropoietin and novel erythropoiesis stimulating protein digests by immunoaffinity capillary electrophoresis-mass spectrometry

In this work, we demonstrate that detection of a specific peptide marker by immunoaffinity capillary electrophoresis-mass spectrometry (IA-CE-MS) could be used to confirm the presence of recombinant human erythropoietin (rhEPO) in solution. Besides the carbohydrate content, the amino acid sequence of novel erythropoiesis stimulating protein (NESP) differs from human erythropoietin (hEPO) at five positions (Ala30Asn, His32Thr, Pro87Val, Trp88Asn, and Pro90Thr). After digesting both glycoproteins in solution by trypsin and PNGase F, two specific proteotypic peptides, EPO (77-97) and NESP (77-97) which differ in three amino acids, were selected as rhEPO and NESP markers, respectively. Both digests and their mixtures were analyzed by IA-CE-MS. The IA stationary phase was prepared from a custom made polyclonal anti-EPO (81-95) antibody immobilized on a solid support of CNBr-Sepharose 4B and was packed in a microcartridge near the inlet of the separation capillary. As the antibody was directed to a synthetic peptide EPO (81-95), only the proteotypic peptide EPO (77-97) was retained. The retained peptide was eluted, separated by electrophoresis and detected by MS. The method was specific to confirm the presence of rhEPO in solution. Although the limits of detection for the peptide marker were similar to those obtained with CE-MS (a few mg/L), these results show the potential of this novel approach to detect in the future rhEPO and its analogues selectively and unambiguously at the levels expected in biological fluids.

Sensitivity enhancement for the analysis of naproxen in tap water by solid‐phase extraction coupled in‐line to capillary electrophoresis

SPE coupled in-line to CE, as the strategy to enhance the concentration sensitivity in CE, has been used to enrich naproxen in tap water samples. In this study, a microcartridge containing an octadecyl silica (C18) sorbent was placed near the inlet within the separation capillary column. The optimum conditions were obtained when naproxen in an acidic aqueous solution (pH 3.5) was loaded into the capillary at 930 mbar for 30 min, and 20 mM ammonium acetate in methanol/water (70:30 v/v) was used as both an elution solution and a separation BGE. Under these conditions, the sensitivity was enhanced 1820-fold with respect to normal hydrodynamic injection, and the LOD achieved was 0.2 microg/L. To show the capability of the in-line SPE-CE method, tap water samples were analysed after a pretreatment consisting in an off-line C18-SPE procedure. The recovery of this procedure was higher than 80%. Under these conditions, naproxen could be detected at a concentration of 10 ng/L; so the potential of the procedure for the sensitive analysis of this type of drugs in water samples was demonstrated. Afterwards, these results were compared with those previously obtained for naproxen in water samples using different sample stacking techniques.

Identification of bioactive peptides in hypoallergenic infant milk formulas by capillary electrophoresis-mass spectrometry

In this study, we use capillary electrophoresis-mass spectrometry (CE-MS) for the identification of bioactive peptides in hypoallergenic infant milk formulas (IF), which are complex bovine milk protein hydrolysates. A sample clean-up pretreatment with a citrate buffer containing dithiothreitol and urea followed by solid-phase extraction (SPE) with different reversed-phase commercial cartridges was investigated to achieve optimum detection sensitivity in CE-MS. SPE with C18, StrataX and Oasis HLB cartridges allowed detection of the largest number of low molecular mass components, but combination of C18 and StrataX results was enough to achieve an excellent coverage of the studied IF. The monoisotopic molecular mass values of the low molecular mass components obtained by capillary electrophoresis ion-trap mass spectrometry (CE-IT-MS) allowed the tentative identification of nine bioactive sequences. Only the identification of five of them could be confirmed when accurate mass measurements were performed by capillary electrophoresis time-of-flight mass spectrometry (CE-TOF-MS), namely LKP, IPY, ALPM, PGPIHN and VAGTWY, which were reported to present angiotensin-converting enzyme (ACE) inhibitory and antimicrobial activity (only VAGTWY).

Migration behavior of therapeutic peptide hormones: Prediction of optimal separation by capillary electrophoresis

A general equation that relates electrophoretic mobility of polyprotic peptide substances and pH of the running electrolytes is established, taking into account the species in solution and the activity coefficients. Modelling electrophoretic mobility as a function of pH can be simultaneously used for determination of ionization constants and selection of the optimum pH for separation of mixtures of the modelled compounds. The proposed relationships allow an important reduction of the experimental data needed for development of new separation methods. The accuracy of the proposed equations is verified by modelling the migration behavior of a heterogeneous series of polyprotic amphoteric peptide hormones. By calculating the values of predicted resolutions, selection of the optimum pH to perform separation of their mixtures becomes a rapid and simple process.

Identification of bioactive peptides in hypoallergenic infant milk formulas by CE‐TOF‐MS assisted by semiempirical model of electromigration behavior

Biologically active peptides derived from complex bovine milk protein hydrolysates are of particular interest in food science and nutrition because they have been shown to play different physiological roles, providing benefits in human health. In this study, we used CE‐TOF‐MS for separation and identification of bioactive peptides in three hypoallergenic infant milk formulas. An appropriate sample cleanup using a citrate buffer with DTT and urea followed by SPE with Sep‐Pack® C18 and StrataXTM cartridges allowed the detection of a large number of low molecular mass bioactive peptides. This preliminary identification was solely based on the measured experimental monoisotopic molecular mass values (Mexp). Later, we evaluated the classical semiempirical relationships between electrophoretic mobility and charge‐to‐mass ratio (me vs. q/Mα, α = 1/2 for the classical polymer model) to describe their migration behavior. The assistance of migration prediction proved to be useful to improve reliability of the identification, avoiding misinterpretations and solving some identity conflicts. After revision, the identity of 24, 30, and 38 bioactive peptides was confirmed in each of the three infant milk formulas. A significant number of these peptides were reported as inhibitors of angiotensin‐converting enzyme, however, the presence of sequences with other biological activities such as antihypertensive, antithrombotic, hypocholesterolemic, immunomodulation, cytotoxicity, antioxidant, antimicrobial, antigenic, or opioid was also confirmed.

Evaluation of fritless solid-phase extraction coupled on-line with capillary electrophoresis-mass spectrometry for the analysis of opioid peptides in cerebrospinal fluid

Fritless SPE on‐line coupled to CE with UV and MS detection (SPE‐CE‐UV and SPE‐CE‐MS) was evaluated for the analysis of opioid peptides. A microcartridge of 150 μm id was packed with a C18 sorbent (particle size > 50 μm), which was retained between a short inlet capillary and a separation capillary (50 μm id). Several experimental parameters were optimized by SPE‐CE‐UV using solutions of dynorphin A (DynA), endomorphin 1 (End1), and methionine‐enkephaline (Met). A microcartridge length of 4 mm was selected, sample was loaded for 10 min at 930 mbar and the retained peptides were eluted with 67 nL of an acidic hydro‐organic solution. Using SPE‐CE‐MS, peak area and migration time repeatabilities for the three opioid peptides were 12–27% and 4–5%, respectively. SPE recovery was lower for the less hydrophobic DynA (22%) than for End1 (66%) and Met (78%) and linearity was satisfactory in all cases between 5 and 60 ng/mL. The LODs varied between 0.5 and 1.0 ng/mL which represent an enhancement of two orders of magnitude when compared with CE‐MS. Cerebrospinal fluid (CSF) samples spiked with the opioid peptides were analyzed to demonstrate the applicability to biological samples. Peak area and migration time repeatabilities were similar to the standard solutions and the opioid peptides could be detected down to 1.0 ng/mL.