Subscriber access provided by BALL STATE UNIV
Article
Accurate quantification of Selenoprotein P (SEPP1) in plasma using isotopically enriched seleno-peptides and speciesspecific isotope dilution with HPLC coupled to ICP-MS/MS Christian Lutz Deitrich, Susana Cuello-Nuñez, Diana Kmiotek, Frank Attila Torma, Maria-Estela Del Castillo Busto, Paola Fisicaro, and Heidi Goenaga-Infante Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b00715 • Publication Date (Web): 25 Apr 2016 Downloaded from http://pubs.acs.org on April 26, 2016
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
Accurate quantification of Selenoprotein P (SEPP1) in plasma using isotopically enriched seleno-peptides and species-specific isotope dilution with HPLC coupled to ICP-MS/MS Christian L. Deitrich1, Susana Cuello-Nuñez1, Diana Kmiotek1, Frank Attila Torma1, Maria Estela del Castillo Busto2, Paola Fisicaro2 and Heidi Goenaga-Infante1* 1 2
LGC, Queens Road, Teddington, London, TW110LY, United Kingdom Laboratoire National de Métrologie et d’Essais (LNE), 1, Rue Gaston Boissier - 75724 Paris cedex 15, France
* email:
[email protected]; Fax: +44(0)20 8943 2767
ABSTRACT: A novel strategy for the absolute quantification of selenium (Se) included in Selenoprotein P (SEPP1), an important bio-marker for human nutrition and disease, including diabetes and cancer, is presented here for the first time. It is based on the use of species-specific double isotope dilution mass spectrometry (SSIDA) in combination with HPLC-ICP-MS/MS for the determination of protein bound Se down to the peptide level in a complex plasma matrix with a total content of Se of 105.5 µg kg-1. The method enabled the selective Se speciation analysis of human plasma samples without the need of extensive clean-up or preconcentration steps as required for traditional protein mass spectrometric approaches. To assess the method accuracy, two plasma reference materials, namely BCR-637 and SRM1950, for which literature data and a reference value for SEPP1 have been reported, were analyzed using complementary hyphenated methods and the species-specific approach developed in this work. The Se mass fractions obtained via the isotopic ratios 78Se/76Se and 82Se/76Se for each of the Se-peptides, namely ENLPSLCSUQGLR (ENL) and AEENITESCQUR (AEE) (where U is SeCys), were found to agree within 2.4%. A relative expanded combined uncertainty (k=2) of 5.4% was achieved for a Se (as SEPP1) mass fraction of approximately 60 µg kg-1. This work represents a systematic approach to the accurate quantitation of plasma SEPP1 at clinical levels using SSIDA quantification. Such methodology will be invaluable for the certification of reference materials and the provision of reference values to clinical measurements and clinical trials.
INTRODUCTION A significant amount of laboratory testing involves the in vivo diagnosis of protein biomarkers.1 Metalloproteins are especially important in medical diagnosis since they represent about 30% of the whole proteome. Most metalloproteins are important indicators for diseases, deficiencies and inflammation or markers in treatment control. Metalloproteins such as selenoproteins have a high relevance in human nutrition or cancer treatment.2 However, there is a lack of reference methods and procedures capable to deliver results traceable to the international systems of units (SI) for many of those protein biomarkers. These methods are however needed for the development of certified matrix reference materials or calibration standards for biological species, which are required for the harmonization and standardization of chemical measurements. Providing SI traceable results for the quantification of elemental species, especially metalloproteins, in complex biological systems is a remaining challenge. An important group of protein biomarkers are selenoproteins. It has been recognized that most biological functions of selenium (Se) are mediated by specific proteins such a as glutathione peroxidase (GPx) and selenoprotein P (SEPP1).3 In contrast to traditional metalloproteins such as transferrin, the heteroatom (Se) is specifically and covalently incorporated as selenocysteine (SeCys)
into their structures. The focus of this study is SEPP1, which is the major selenoprotein in plasma responsible for transport and distribution of Se. Furthermore, SEPP1 has also been associated with neurodegenerative diseases such as Alzheimer and Type 2 diabetes.4 Considering that SEPP1 is a biological active form of Se and most of the plasma Se is associated with this protein, it represents one of the most accurate selenium-status biomarkers for human nutrition.5,6 For clinical purposes SEPP1 has usually been characterized and quantified using antibodybased enzyme immunoassays such as ELISA. However, those assays often suffer from a lack in selectivity and results are usually associated with high standard deviations. Furthermore, the characterization and validation of most commercial assays for SEPP1 are missing so far.7 Hence, there are some inconsistencies and lack of data comparability in the results obtained for SEPP1 in humans. Thus, there is an urgent need for reference methodology to accurately determine SEPP1 for the production of new speciated reference materials and the provision of reference values to clinical measurements. The presence of a metal or heteroatom associated with a protein, such as Se in SEPP1, opens the path to perform protein absolute quantification using heteroatom-targeted proteomics.8 Heteroatom targeted proteomics involves the coupling of a sensitive element-specific detector, usually inductively cou-
1
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
pled plasma mass spectrometry (ICP-MS), with an appropriate separation technique such as high performance liquid chromatography (HPLC).8 A molecular mass spectrometry technique is usually required to provide structural information on the target species. Selenoproteins such as SEPP1 have been separated from the matrix and detected as intact proteins by a combination of size exclusion chromatography and/or a double-column affinity HPLC approach consisting of a heparinand blue-sepharose stationary phase followed by Se-specific detection with ICP-MS.9-13 The use of anion-exchange or solid phase extraction for sample pre-treatment as well as high resolution ICP-MS have all been exploited for SEPP1 quantification.9,10 Alternative approaches include the application of gel electrophoresis with laser ablation ICP-MS14,15 or the enzymatic proteolysis of selenoproteins followed by the detection of the released selenium-containing amino acids.16,17 External calibration approaches, often used for SEPP1 quantification present remaining challenges. These include the lack in specificity due to the use of species-unspecific calibrants and the high impact of possible loss or transformation of species, during sample treatment and separation, on the accuracy of data. Since the introduction of internal standardization techniques such as isotope dilution analysis (IDA) for elemental speciation, the approach has been increasingly used to improve measurement accuracy and minimize uncertainty.18 Speciesunspecific or post-column IDA is most frequently applied for quantitative elemental speciation of proteins due to the lack of isotopically enriched matching protein species and/or well characterized calibration standards. This approach in combination with affinity chromatography and ICP-MS has been successfully used for the quantification of selenoproteins.12,19-21 However, there is a clear disadvantage of post-column IDA in comparison with species-specific IDA (SSIDA) for which a matching isotopically enriched species is added to the sample before sample preparation; losses of species before isotopic equilibration (which may occur after species separation) are not corrected for.22 Therefore, this IDA approach is not suitable to stand alone in the certification of matrix reference materials. Alternatively, SSIDA determinations, which are directly traceable to the SI through the use of a certified or well characterized calibration standard have been reported so far for only a few clinically relevant metalloproteins.23,24 The production of like-for-like isotopically enriched proteins for speciesspecific IDA quantitation of entire proteins is not always feasible since it is challenging, cost and time consuming. Moreover, the difficulty to achieve selective separation and recovery of target proteins in complex protein mixtures and sample matrices for the purpose of their accurate quantification is a very difficult task. Therefore, SSIDA quantification of proteins down to the aminoacid16,17 or peptide level25,26 by HPLCICP-MS or HPLC and molecular mass spectrometry with protein hydrolysis or tryptic digestion has been proposed as an attractive alternative. To the author’s knowledge, no methodology based on SSIDA quantification of proteins (including SEPP1) via element-specific detection of specific peptides has been reported so far. This work reports the development and validation of double SSIDA methodology for the accurate quantification of selenoprotein P (SEPP1) via Se-specific detection at the peptide level. Isotopically enriched selenium-containing peptides specific to SEPP1, including ENLPSLCSUQGLR (ENL) and
Page 2 of 9
AEENITESCQUR (AEE) were synthesized and characterized for their Se content, Se distribution and isotopic composition. The addition of peptide spikes to the plasma sample was followed by tryptic digestion, alkylation and isotope ratio determination using HPLC-ICP-MS. The same peptides with natural Se isotopic composition were also synthesized, characterized and used for double IDA quantitation using the exact matching approach. This was achieved without the need for sample pre-concentration and/or further cleanup steps. To assess the method accuracy, two plasma reference materials namely BCR-637 and NIST SRM 1950, for which literature data and an indicative value for SEPP1 have been reported, respectively, by using complementary hyphenated methods were analyzed using the species-specific approach developed in this work. The measurement uncertainty associated with the mass fraction of SEPP1 in the investigated plasma matrix by the SSIDA method proposed here was estimated and the main contributing factors were identified. EXPERIMENTAL SECTION Reagents, Solutions and Materials The investigated reference materials were SRM 1950 (NIST, Gaithersburg, USA) with a reference mass fraction for SEPP1 (as Se) of 50.2 ± 4.3 µg kg-1 and CRM BCR–637 (IRMM, Geel, Belgium). The natural and isotopically enriched Sepeptides ENL and AEE, with U=L-SeCys or L-Se76Cys, used as calibration standards and spikes were a requested synthesis by IsoSciences (King of Prussia PA, USA). Peptides were produced as trifluoracetate salts with a molecular weight of 1463.55 and 1466.55 g mol-1 for the ENL and 1426.40 and 1429.40 g mol-1 for the AEE peptides, respectively. The isotopic incorporation of 76Se in the enriched peptides was > 98 atom %. Both peptides have a chemical purity of > 95%.. The enriched Se used to synthesize the peptides was purchased as metal with an isotopic composition for 76Se of 98.5 ±0.2% (CK Gas Products Ltd, Ibstock, UK). For the preparation of the IDA calibration blends a filtrate < 10 kDa was prepared from human serum type AB male (Sigma-Aldrich, Gillingham, UK) by using Amicon Ultra-0.5 10 kDa cut-off centrifugal filter devices (Millipore, Watford, UK). For total Se determination of the peptides a primary Se standard SRM 3149 (NIST, USA) with a certified mass fraction of 10042 mg kg-1 and a secondary Se standard at 1000 mg L-1 (ROMIL Ltd., Cambridge, UK) were purchased together with a 77Se enriched spike (CK Gas, UK). Other materials used were of analytical reagent grade or better if not stated otherwise. Trypsin from bovine pancreas, trizma base (Tris), trizma hydrochloride (Tris-HCl), dithiothreitol (DTT), calcium chloride (CaCl2), trifluoroacetic acid (TFA) and iodoacetamide (IAA), acetic acid (AA) (all Sigma-Aldrich), formic acid (FA) (Fisher Scientific, Loughborough, UK), methanol (MeOH) and acetonitrile (ACN) were of optigrade HPLC special grade and purchased from LGC Standards (Luckenwalde, Germany). Nitric acid (HNO3) and hydrogen peroxide (H2O2) were of Upa grade purchased from ROMIL. All solutions and standards were prepared with ultrapure water 18 MΩ.cm (Veolia Water Technologies, High Wycombe, UK). Instrumentation For the determination of the total Se in the natural and enriched peptides, aliquots were digested using a microwave
2
ACS Paragon Plus Environment
Page 3 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
formed by ICP-OES. Approximately 200 µg of each peptide was digested using 0.4 mL of concentrated HNO3 and 0.2 mL of H2O2. An adequate amount of the 77Se spike (0.11 mg g-1 in 5% HNO3) was added to the natural peptide solutions. A total of 4 replicates were prepared for the natural peptides and 2 for the enriched peptides. For IDA of total selenium, the natural peptide samples were measured against a calibration blend prepared from SRM 3149 (NIST). The enriched peptide solutions were submitted to ICP-OES analysis and determined by external calibration using standards prepared by appropriate dilution of a 1000 mg L-1Se stock standard solution (ROMIL) in 5 % nitric acid. For further characterization of the Se distribution and fractionation as well as assessing the alkylation efficiency, Se-peptide solutions were analyzed by HPLC-ICPMS as described in the instrumental section. To confirm the species identity (structure identification) peptide- solutions were also analyzed by HPLC-IonTrap-ESI-MS in combination with ICP-MS and HPLC-qTOF-ESi-MS. Tryptic Digestion and Peptide Extraction SSIDA blends were prepared as described in detail in the supplementation information (section S-1) and submitted to enzymatic digestion as described below. Digestion was performed with slight variations following a protocol reported by Arsene at al., 2010.28 An overview of the workflow used for the determination of Se-peptides can be found in Figure S-1. In brief, all blends were treated at 56 ºC by adding 100 µL of 200 mM DTT solution (in 0.1 M Tris-HCl) for 1 hour. Solutions were left to cool down and then 30 µL of a freshly prepared 40 mg ml-1 trypsin solution in 50 mM AA was added to each blend. Blends were incubated at 37 ºC followed by another addition of a 30 µL aliquot of trypsin to the reaction mixture after 10, 30, 90, 150, 210 and 270 minutes. In parallel 100 µL of ACN was added every 30 min. starting at t=30 min. to a total of 330 min. (11 additions). The reaction was allowed to proceed for a total of 24 hours. After 24 hours 75 µL of the 200 mM DTT solution was added and blends reduced at 56 ºC for 90 minutes. The insoluble fraction was removed by centrifugation for 5 min. at 12000 g and the filtrate collected. The insoluble fraction was extracted two times with 0.8 and 0.5 ml of ACN/H2O (1:1, v/v) and the extract was combined with the filtrate previously obtained as described above. The solutions were lyophilized and re-suspended in 500 µL of water. The insoluble residue was removed by centrifugation 5 min. at 12000 g, washed twice with 0.25 mL of ACN/H2O (1:1, v/v) and the extract added to the filtrate. Alkylation was performed by adding 200 µL of a 200 mM IAA solution in 0.1 M TrisHCl and incubating 90 min. at room temperature in the dark. This was followed by 60 min. incubation with 300 µL of a 200 mM DTT solution at room temperature and a final addition of 100 µL of 5% FA. Samples were then lyophilized and resuspended in 500 µL of water before analysis. SSIDA of the samples was performed by bracketing each sample blend between two calibration blends using an exact matching doublespike SSIDA approach as described elsewhere.29 Procedures and details for SSIDA measurements and for data processing are given in the supplementary information (section S-1). RESULTS AND DISCUSSION Selection of SEPP1-specific Se-peptides SEPP1 contains 10 SeCys residues (Figure 1). Tryptic digestion of this protein is known to result in a total of eight seleni-
system (Discovery™, CEM, Buckingham, UK). Mixtures were heated up to 180 °C and kept at this temperature for 10 min. and maintaining the pressure below 21 bar. Total IDA on the natural peptide standards was performed using an Agilent 7700 ICP-MS (Agilent Technologies, Palo Alto, USA) in either He or H2 mode. Total Se in the enriched peptides was measured using an Optima 3000 RL ICP-OES instrument (Perkin Elmer, Buckingham, UK). For instrumental conditions see supplementation information Table S-1. For peptide characterization and SSIDA of plasma samples HPLC-ICP-MS was utilized. A 1200 HPLC system (Agilent, Technologies) was coupled to an Agilent 8800 QQQ ICP-MS instrument operated in MS/MS mode using oxygen as reaction gas. In the first quadrupole only the selected isotope masses are allowed to pass through. The ions are then reacted with oxygen to form the target species XO+ and target isotopes are then filtered through the second quadrupole with an m/z shift of +16 (Table S-1). Separation of the Se-peptides was achieved using an Agilent Zorbax Rx C18 (250x2.1 i.d.; 5 µm) column. Mobile phase A was 5% (v/v) MeOH in water with 0.1% (v/v) FA and mobile phase B was 100% MeOH with 0.1% (v/v) FA. Elution was achieved using a gradient reaching 50% B after 35 minutes. The flow rate was 0.2 mL min-1 with an injection volume of 20 µL. The column was connected to the ICP-MS, operated and tuned for high organics-introduction, via PEEK tubing. All Se isotopes were monitored in O2 mode with a mass shift of m/z +16 and a dwell time of 0.3 s (Table S-1). HPLC-Ion-Trap ESI-MS experiments were performed by coupling the HPLC system to a 6330 Ion Trap LC/MS system (Agilent Technologies). A splitter was used to divide some of the eluent to the ICP-MS for parallel Se detection. Se-peptides were analyzed using the chromatographic conditions described above with typical IonTrap conditions shown in Table S-1. For absolute confirmation of the species identity, accurate mass determination of the natural and enriched alkylated peptides was performed using HPLC-ESI-qTOF MS/MS (see supporting information). A G6530A qTOF ESI-MS (Agilent Technologies) was coupled to an Xbridge BEH C18 column (150x2.1; 3.5 µm) (Waters, Hertfordshire, UK). The chromatographic and measurement conditions for the ESI qTOF experiments are shown in Table S-2. For the preparation of SSIDA blends, peptide solutions were lyophilized under vacuum, using a Christ rotational vacuum concentrator RVC 2-25 in combination with an Alpha 1-2 LDplus freeze dryer (SciQuip, Shropshire, UK). For tryptic digestion samples were placed in a Microtherm heated shaker (Camlab, Cambridge, UK). Samples were centrifuged using a Biofuge pico from Heraeus Instruments (Osterode, Germany). Procedures Characterization of Se-peptides SI traceability of double SSIDA is achieved through the use of a certified or well characterized species calibrant. For this purpose, the Se-peptides with natural isotopic composition were characterized for their mass fraction of Se (using IDA ICP-MS and Se distribution (by HPLC-ICP-MS).27 This strategy has been demonstrated to be fit for purpose for quantitative IDA of elemental species.27 The stock solutions of peptides were prepared as described in the supplementary information section S-1. Quantification of the total Se content of the isotopically-enriched Se-peptide stock solutions was per-
3
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
um-containing fragments. After a BLAST search (BLASTP database), peptides such as SUCCHCR (SUC), TGSAITUQCK (TGS), ENLPSLCSUQGLR (ENL) and AEENITESCQUR (AEE), where U is SeCys, were selected and then custom synthesized together with their 76Se-enriched analogues, for protein quantification. The reason for this is, the peptides yielded exact matches exclusively with SEPP1 amongst other plasma proteins containing Se. Preliminary results from the analysis of SUC and TGS peptides suggested that these peptides are not suitable to be used as calibrants for SSIDA due to their very low purity and low content of Se. In addition, the Swiss-Prot entry for SEPP1 (P49908) shows a possible natural variation in the region of the TGS peptide, hence making it unsuitable to be used as calibration standard for protein quantification. Apart from a possible glycosylation site in the AEE peptide no post-translational modifications were observed for SEPP1 in the regions of the ENL and AEE peptides (Figure 1). For all these reasons, ENL and AEE peptides were selected and further characterized as detailed below. Determination of the total Se content in natural and enriched Se-peptides For SSIDA, the mass fraction of the natural calibrant is known to be one of the major contributing factors to the accuracy of the overall measurement uncertainty,22,27,29 leaving no need for additional work in the accurate determination of the spike mass fraction considering that the same spike solution is added to both sample and calibration blends.29 Therefore, the total Se determination in the enriched peptides (spikes) was carried out using ICP-OES as described in the experimental section. Using ICP-OES and external calibration the enriched ENL-peptide and AEE spike stock solutions were found to contain 43.9 ± 1.4 mg kg-1 of Se and 2.71 ± 0.06 mg kg-1 of Se, respectively (see Table S-3). Total Se IDA determination of the natural Se-peptides was performed using the procedure described in the experimental section. Results showed that the natural ENL peptide stock solution contained 43.0 ± 1.2 mg kg-1 of Se and the AEE solution contained 0.72 ± 0.06 mg kg-1 of Se, respectively (see Table S-3). The analysis of Se-peptide calibrants and spikes for total Se was followed by a second step to verify whether all the Se present elutes at the retention time of the target peptide. This approach was demonstrated to be fit for purpose to provide accurate results for Se species in the CCQM-P86 study.27 Determination of the elemental distribution and isotopically enrichment in natural and enriched Se-peptides Efforts were undertaken to develop HPLC-ICP-MS methodology to monitor the Se distribution of the natural and spike peptide stock solutions. Separation of peptides is usually achieved by using chromatography with an organic gradient. However, ICP-MS does not tolerate the introduction of high amounts of organic solvents without compromising sensitivity. Moreover, the deposition of carbon in the cones interface generated from the introduction of high organics into the ICP-MS also causes deterioration in the short and long-term measurement precision. A method was developed to achieve a good separation of our target peptides (as needed for SSIDA quantitation) with conditions suitable to achieve a stable ICP-MS run over a period of 24 hours. MeOH and ACN as well as FA and TFA were tested. Furthermore, the use of a C8 and C18 column with identical dimensions was compared. The addition of an
Page 4 of 9
Figure 1. Amino acid sequence of full length human SEPP1 (Swiss-Prot database entry: P49908) showing the peptide fragments ENL and AEE chosen for protein quantification. ion-pairing reagent (TFA) to the mobile phase did not improve the separation and peak shape of the Se-peptides, it rather made the AEE peptide to co-elute with other possible digestion products of plasma proteins. Therefore, FA, which is also compatible with ESI-MS analysis, was selected for further use. A minimum of 30% ACN was needed to elute the most retained ENL peptide from the C18 column compared to 50% MeOH. In addition, the ICP tolerates higher concentrations of MeOH in comparison to ACN. By using the C18 column, target peptides eluted slightly faster without loss of selectivity under the same conditions as those used for the C8 column and was therefore, chosen. The gradient was optimized to obtain full elution of all peptides within 45 min. followed be a 15 min. equilibration prior to the next injection. The optimal HPLC conditions used are summarized in Table S-1. For ICP-MS detection we exploited the capabilities of triple quadrupole ICP-MS instrumentation by measuring the Se isotopes in MS/MS mode using oxygen as reaction gas.30,31 This approach has been demonstrated to be more efficient in removing interferences on the target ions than using traditional methods such as hydrogen or helium using single quadrupole technology.30,31 Performing analysis with the oxygen mode can help improve Se determination in complex sample matrixes such as plasma. The detection limits of 78Se for the ENL and AEE peptides were 1.38 and 2.74 µg kg-1, respectively. No correction of isotope ratios for instrumental mass bias and/or detector dead time was undertaken. Using approximate matching double SSIDA ICP-MS, previous reports27,29 have demonstrated that there is no need for these corrections, on the condition that a good isotope ratio matching between sample and calibration blends (generally < 5%) is achieved and that the isotopic signals of both blends are similar. The effects of mass bias and linearity are then eliminated since both calibration and sample blends are affected equally. Figure 2 shows the chromatograms obtained for the unprotected (or non-alkylated) natural and enriched ENL peptide obtained using HPLC-ICP-MS. The chromatograms of natural and enriched AEE peptide are shown in Figure S-2. The natural and enriched ENL peptide elute as 3 distinctive peaks between 37 min. and 42 min. with no other peaks observed in the chromatographic run. The AEE peptides elute as one major peak at 24 min. with several minor peaks observed close to the
4
ACS Paragon Plus Environment
Page 5 of 9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Analytical Chemistry
peptides is achieved and (c) complete separation of unspecific Se from the target Se-peptides is achieved. Peptide alkylation and structural confirmation In order to obtain a single signal or peak for each peptide, as required for IDA measurements Cys (SeCys) residues have to be protected to avoid oxidation and reactions of their thiol (selenol) groups. This is usually done by alkylation of the corresponding peptides.32 To investigate the reaction behavior of the investigated natural and enriched Se-peptides, aliquots were submitted for alkylation using iodoacetamide (IAA). Reaction products were analyzed by HPLC-ICP-MS. Figure 2 (inserts) shows the HPLC-ICP-MS chromatograms of the natural and isotopically enriched alkylated ENL peptide. Only one major peak can be observed for both the natural and isotopically enriched ENL peptides. Compared to the unprotected peptide the alkylated peptide elute earlier at 34 min., which is due to the structural changes caused by the addition of two carboxamidomethyl groups on the Cys residues of the peptides. The alkylation efficiency was found to be a 100% for both the ENL species. Similar observations were found for the AEE peptides (alkylated peptides eluted at 20.3 min. as single peaks, with an alkylation efficiency of 100%), as shown in Figure S-2. By adjusting the amount of reagents and the incubation period during the sample treatment, full alkylation of the Se-peptides in the serum matrix was achieved. Confirmation of the structural identity of the alkylated Sepeptides was performed using molecular MS. For this purpose, HPLC-IonTrap ESI-MS hyphenated in parallel to ICP-MS was used as described in the experimental section. Figures 3 and S3 show the HPLC-ICP-MS chromatograms and ESI-MS mass spectra obtained for the protected natural peptides. The measured m/z ratios of the natural protected ENL and AEE peptides of 1581.7 and 1544.6 (m/z +1) (Figures 3c and S-3c), do agree well with their theoretical values of 1580.7 and 1543.6, respectively. The mass difference between the enriched alkylated peptides and the untreated enriched peptides showed the successful modification of the 2 Cys residues with carboxamidomethyl groups (see Figures S-4 & S-5). A mass difference of 114 was confirmed by additional qTOF experiments (data not shown). In addition, the typical Se isotopic pattern was observed for the natural peptides (Figure 3d and S-3d). For the enriched peptides a mass shift of m/z -4 on the main signal can be observed compared to their natural species. This reflects the enrichment of the peptides with 76Se compared to the most intense signal obtained for 80Se in the natural peptides (see Figure S-5). Parallel coupling of IonTrap-ESI-MS and ICPMS additionally confirmed the eluting order of the peptides by assigning the Se trace to the ESI-MS spectra in the obtained chromatograms (Figures 3 and S-4). Further structural confirmation of the alkylated peptides using HPLC-qTOF-ESI-MS for high accurate mass measurements was also undertaken (see Figure S-6). The characterization data provided enough proof of evidence for the suitability natural and enriched Se-peptides for further development of methodology for the SSIDA quantification of SEPP1 in human plasma. Development and validation of SSIDA methodology for the accurate quantification of plasma SEPP1 The determination of SEPP1 via its Se-peptides requires the complete digestion and release of target peptides from the protein in the plasma matrix. In addition, a complete
Figure 2. RP-HPLC-ICP-MS chromatograms of natural and enriched ENL peptides. (a) Natural ENL peptide and (b) enriched ENL. Inserts, the same peptides after alkylation. column dead volume and the beginning of the run. Both peptides contain cysteine (Cys) and SeCys residues. If unprotected, these residues are prone to oxidation and oligomer formation via disulfide (selenide) bonds. This explains the elution profile of the ENL peptide showing multiple peaks for a single peptide standard, most likely due to oligomer formation (Figure 2). However, the use of SSIDA requires the addition of the isotopically enriched peptides to the serum sample in their native form prior to sample digestion and peptide alkylation. Therefore, efforts were undertaken to develop methodology for efficient and like for like alkylation of endogenous and added peptides in the plasma matrix and for confirmation of their structural composition (see next section). To verify whether the Se isotopic composition of synthesized isotopically enriched peptides was similar to that of the precursor inorganic 76Se spike, repeated injections of the enriched ENL and AEE peptide stock solutions into the HPLC were performed and the Se isotopic pattern was detected using ICP-MS. The isotopic enrichment of 76Se in these peptides was found to be 98.6 ± 0.1% for ENL and 98.7 ± 0.1% for AEE, respectively. Those values are in good agreement with the certificate for the enriched Se used as a starting material with an isotopic abundance of 98.5 ± 0.2% for Se-76. Table S3 summarizes the characterization data for each of the investigated Se-peptides. The ENL peptide solution was found to contain a higher concentration of Se in comparison to the AEE peptide. Moreover, most of the Se in the ENL peptide was found to elute as a single chromatographic peak (comprising 99.4% of the total Se) for both the spike and the natural species. Therefore, it is the preferred standard for SSIDA quantification. However, both peptides were found promising to be used as standards for SSIDA of SEPP1 if (a) the distribution of Se in the peptides is accounted for, (b) the Se-peptides are stable before complete equilibration between spikes and native
5
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 6 of 9
sample must be avoided. Several issues needed to be addressed in order to have a sample treatment procedure which would be found fit for purpose. The peptides were added in their non-protected form at the very beginning of the procedure. Reductive cleavage of Cys–Cys bridges in the peptides was then performed after enzymatic extraction in the plasma. When alkylated peptides were added to the plasma at the beginning or after enzymatic extraction, the ENL peptide did elute as two separate peaks and additional peaks for Se-76 at 11 and 37 min. were observed (see Figure S-7). This indicated that matrix effects play an important role and can alter the alkylation efficiency or process. Therefore, it is important to add the spike at the earliest possible time in its native form so it can equilibrate and interact with the matrix in the same way as the analyte. A second issue occurred when the unprotected spike was added at the very beginning of the procedure. A second peak enriched in 76Se was observed at a later retention time of 37 min. (see Figure S-8). The initial thought was the incomplete protection of the spike peptide, thus the alkylation procedure was adjusted by increasing the amounts of reduction and alkylation reagents and lengthen equilibration times. However, changing the alkylation procedure did not have an influence on the spike behavior. The possibility of our spike interacting with the plasma matrix in a different way than our natural sample peptides would render IDA unfeasible. To avoid possible spike transformation and/or interaction with matrix components an additional reduction step (addition of 100 µl of 200 mM DTT and incubation at 56 ºC for 1 hour) was introduced at the very beginning of the analytical procedure straight after spiking of the peptides into the plasma. The optimized procedure was applied to the quantification of SEPP1 in plasma reference materials. Quantification of SEPP1 in plasma reference materials The workflow of the optimized procedure for the determination of SEPP1 in plasma using SSIDA is shown in the supplementation information, Figure S-1. As described earlier, the peptide mass fraction and Se isotopic ratios in both sample and calibration blends were chosen to match as close as possible (within 5%). This is done to eliminate systematic errors that result from the uncertainty associated with the concentration and isotopically enrichment of the spike used.33 The amount of SEPP1 in the samples is calculated from the isotopic ratios measured for each peptide in the sample and in the calibration blend. Since each of the investigated peptides only contains one SeCys residue but intact SEPP1 contains a total of 10 SeCys residues the results obtained for each peptide need to be corrected by a factor of 10 to obtain the final mass fraction of Se in SEPP1. The developed SSIDA method was applied to the quantification of SEPP1 in BCR-637 and NIST SRM 1950. Figure 4 shows an example for a chromatogram obtained for a sample blend (NIST SRM 1950) using the newly developed method. The SSIDA method was first used to quantify SEPP1 in the BCR-637. This material has been extensively studied and values have been reported for SEPP1 by several authors, as summarized elsewhere.34 The mass fraction of SEPP1 (as Se) in BCR-637 was found to be 56.2 ± 1.7 µg kg-1 (for ENL) and 58.3 ± 2.7 µg kg-1 (for AEE), respectively. These data was obtained by using the 78Se/76Se isotope ratio. Both values agree well with each other and are in good agreement with the mean value of 60 ± 7 ng ml-1 reported in the literature.34 For
Figure 3. Chromatogram obtained by parallel coupling of HPLC to Ion-Trap ESI-MS and ICP-MS for the measurements of the natural alkylated ENL peptide used for the quantification of SEPP1. (a) HPLC-ICP-MS chromatogram of the natural alkylated ENL peptide, (b) ESI-MS total ion chromatogram, (c) ESI-MS mass spectrum of the signal at 34.5 min. shown above and (d) isotopic pattern of the m/z +2 signal shown above representing the typical natural Se isotopic abundance. separation of the specific peptides from other Se-species present in the matrix and the complete equilibration between the spike and sample peptides are mandatory. Aiming at best accuracy, previous successful inter-comparison studies for Se speciation have demonstrated that matrix matching of the calibration blend and its submission to the same preparation procedure than that used for the sample blend are very important considerations. However, since it is difficult to obtain a plasma free from SEPP1, calibration blends were prepared in a plasma filtrate ( 90%) are (a) the measured isotopic ratios of the calibration blend, (b) the measured isotopic ratios of the sample blend, (c) the uncertainty associated with the primary calibration standard and (d) the variability between the repeated samples. Most contributions are attributed to the low levels of
7
ACS Paragon Plus Environment
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
proximately 60 µg kg-1 Se) in micrograms of plasma with a relative expanded uncertainty (k=2) of 5%. This has been achieved due to the development and validation of a protocol for peptide stabilization in solution and complete equilibration between native and added peptides (spikes) during sample digestion and due to the selective chromatographic separation and detection of key Se-peptides (specific to SEPP1) using HPLC-QQQ-ICP-MS with limits of detection of 1.38 and 2.74 µg kg-1 Se for the ENL and AEE peptides, respectively. Elemental SSIDA is tedious and expensive and, therefore, is not likely, at its current stage, to replace classical proteomics or antibody-based assays in routine clinical medical research analysis of SEPP1. However it offers an invaluable potential to be used for quality control of measurements in the clinic. Furthermore it helps in the provision of reference values to clinical trials and the certification of reference materials. Future work will focus on the extension of the current approach to SEPP1 and other selenoproteins relevant to health present at even lower levels in microsections of tissue (e.g. brain tissue).
(7)
(8) (9)
(10) (11) (12) (13) (14) (15) (16)
AUTHOR INFORMATION
(17)
Corresponding Author
(18) (19)
* Phone: +44 (0)208 943 7555. Email:
[email protected] (20)
Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.
(21) (22)
Notes The authors declare not competing financial interests.
(23)
ACKNOWLEDGMENT
(24)
The work described in this manuscript was supported by EURAMET as part of the European Metrology Research Program (EMRP) -HLT05 “Metrology for Metalloproteins”. Thanks go to Caroline Pritchard, Milena Quaglia and Sarah Hill (LGC) for valuable discussions. We also like to thank Anna Konopka and Wolf D. Lehmann (DKFZ, Heidelberg, Germany), for their help with the peptide selection.
(25) (26) (27)
ASSOCIATED CONTENT Supporting Information Additional information as noted in the text. Experimental procedures and measurement conditions, determination of the elemental distribution and isotopically enrichment in natural and enriched Se-peptides, structural confirmation of Se-peptides, development of an SSIDA methodology for the accurate quantification of plasma SEPP1 and quantification of SEPP1 in plasma reference materials, uncertainty budget. This material is available free of charge via the internet http://pubs.acs.org.
(28) (29) (30) (31) (32) (33) (34)
REFERENCES (1) (2) (3) (4) (5) (6)
Anderson, N. L.; Clin. Chem. 2009, 56, 177−185. Rayman, M. P. Proc. Nutr. Soc. 2005, 64, 527-542. Reeves, M.A.; Hoffmann, P. R. Cell. Mol. Life. Sci. 2009, 66, 2457-2478. Rayman, M. Lancet 2012, 379, 1256-1268. Thomson, C. D. Analyst 1998, 123, 827-831. Pedrero, Z.; Madrid, Y. Anal. Chim. Acta 2009, 634, 135152.
(35)
(36)
Page 8 of 9
Hybsier, S.; Wu, Z.; Schulz, T.; Strasburger, C. J.; Köhrle, J.; Minich., W. B.; Schomburg, L. Perspect. Sci. 2015, 3, 23-24. Sanz-Medel, A.; Montes-Bayón, M.; Bettmer, J.; FernándezSanchez, M. L.; Ruiz Encinar, J. TRAC 2012, 40, 52-63. Jitaru, P.; Prete, M.; Cozzi, G.; Turetta, C.; Caitrns, W.; Seraglia, R.; Traldi, P.; Cescon. P.;Barbante, C. J. Anal. At. Spectrom. 2008, 23, 402-406. Jitaru, P.; Cozzi, G.; Seraglia, R.; Traldi, P.; Cescon, P.; Barbante, C. Anal. Methods 2010, 2, 1382-1387. Jitaru, P., Roman, M.; Cozzi, G.; Fisicaro, P.; Cescon, P.;Barbante, C. Microchim Acta 2009, 166, 319-327. García-Sevillano, M. A.; García-Barrera, T.; Gómez-Ariza, J. L. Anal. Bioanal. Chem. 2014, 406, 2719-2725. Shigeta, K.; Sato, K.; Furuta, N. J. Anal. At. Spectrom. 2007, 22, 911-916. Baillihaut, G.; Kilpartrick, L. E.; Kilpartick, E. L.; Davis, W. C. Metallomics 2012, 4, 533-538. Baillihaut, G.; Kilpartrick, L. E.; Kilpartick, E. L.; Davis, W. C. J. Anal;. At. Spectrom. 2011, 26, 383-394. Ruiz Encinar J.; Schaumlöffel, D.; Ogra, Y.; Lobinski, R. Anal. Chem. 2004, 76, 6635-6642. Jitaru, P.; Goenaga-Infante, H.; Vaslin-Reimann, S.; Fisicaro, P. Anal. Chim. Acta 2010, 657, 100-107. Heumann, K. G. Anal. Bioanal. Chem. 2004, 378, 318-329. Baillihaut, G.; Kilpartrick, L. E.; Davis, W. C. Anal. Chem. 2011, 83, 8667-8674. Hinojosa Reyes, J.; Marchante-Gaýon, J. M.; GarcíaAlsonso J. I.; Sanz-Medel, A. J. Anal. At. Spectrom. 2003, 18, 1210-1216. Xu, M.; Yang, L.; Wang, Q. J. Anal. At. Spectrom. 2008, 23, 1545-1549. García-Alonso, J. I.; Rodríques-González, P. Isotope Dilution Mass Spectrometry; Royal Society of Chemistry: Cambridge, 2013. Deitrich, C.L.; Raab, A.; Pioselli, B.; Thomas-Oates, J. E.; Feldmann, J. Anal. Chem. 2007, 79, 8381-8390. Castillo Busto, M. E.; Montes-Bayón, M.; Sanz-Medel, A. Anal. Chem. 2006, 76, 8218-8226. Ezan, E.; Dubois, M.; Becher, F. Analyst 2009, 134, 825834. Brönstrup, M. Expert Rev. Proteomics 2004, 1, 503−512. Goenaga-Infante, H.; Sturgeon, R.; Turner, J.; Hearn, R.; Sargent, M.; Maxwell, P.; Yang, L.; Barzev, A.; Pedrero, Z.; Ćamara, C.; Díaz Huerta, V.; Fernández Sánchez, M. L.; Sanz-Medel, A.; Emese, K.; Fodor, P.; Wolf, W.; Goldschmidt, R.; Vacchina, V.; Szpunar , J.; Valiente, L.; Huertas, R.; Labarraque, G.; Davis, C.; Zeisler, R.; Turk, G.; Rizzio, E.; Mackay, L. G.; Myors, R. B.; Saxby, D. L.; Askew, S.; Chao, W. Anal. Bional. Chem. 2008, 390, 629-642. Arsene, C. G.; Henrion, A.; Diekmann, N.; Manolopoulou, J.; Bidlingmaier, M. Anal. Biochem. 2010, 201,228-235. Hearn, R.; Evans, P.; Sargent, M. J. J. Anal. Atom. Spectrom. 2005, 20, 1019-1023. Jackson, B. P.; Liba, A.; Nelson, J. J. Anal At. Spectrom. 2015, 30, 1179-1183. Anan, Y.; Hatakeyama, Y.; Tokumoto, M., Ogra, Y. Anal. Sci. 2015, 29, 787-792. Sechi, S.; Chait, B. T. Anal. Chem. 1998, 70, 5150-5158. Henrion, A. Fresenius. J. Anal. Chem. 1994, 350, 657-658. Jitaru, P.; Roman, M.; Barbante, C.; Vaskin-Reimann, S.; Fisicaro, P. Accred. Qual. Assur. 2010 15, 343-350. Ellison, S. L. R.; Williams, A. Ed.; Eurachem/CITAC guide: Quantifying Uncertainty in Analytical Measurement, 3rd ed.; 2012, ISBN 978-0-948926-30-3. Available from www.eurachem.org. Bolea-Fernandez, E.; Balcaen, L.; Resano, M.; Vanhaecke, F. Anal. Bioanal. Chem. 2015, 407, 919-929.
8
ACS Paragon Plus Environment
Page 9 of 9
Analytical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
for TOC only
9
ACS Paragon Plus Environment