Research Article Received: 14 April 2014
Revised: 28 May 2014
Accepted: 2 June 2014
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763 (wileyonlinelibrary.com) DOI: 10.1002/rcm.6957
Identification of site-specific heterogeneity in peptide drugs using intact mass spectrometry with electron transfer dissociation Ashley C. Gucinski and Michael T. Boyne II* U.S. Food and Drug Administration, CDER/OPS/OTR Division of Pharmaceutical Analysis, 10903 New Hampshire Ave, Silver Spring, MD 20993, USA RATIONALE: Protamine sulfate is a peptide drug product consisting of multiple basic peptides. As traditional high-
performance liquid chromatography (HPLC) separation methods may not resolve these peptides, as well as any possible peptide-related impurities, a method utilizing top-down mass spectrometry was developed for the characterization of complex peptide drug products, including any low-level impurities, which is described in this study. METHODS: Herring protamine sulfate was used as a model system to demonstrate the applicability of the method. Direct infusion mass spectrometry and tandem mass spectrometry (MS/MS) on a high-resolution, mass accurate instrument with electron transfer dissociation (ETD) were used to identify all the species present in the herring protamine sulfate sample. Identifications were made based on mass accuracy analysis as well as MS/MS fragmentation patterns. RESULTS: Complete sequence coverage of the three abundant herring protamine peptides was obtained using the top-down ETD-MS/MS method, which also identified a discrepancy with the published herring protamine peptide sequences. Additionally, three low-abundance related peptide species were also identified and fully characterized. These three peptides had not previously been reported as herring protamine peptides, but could be related to the published sequences through amino acid additions and/or substitutions. CONCLUSIONS: A method for the characterization of protamine, a complex peptide drug product, was developed that can be extended to other complex peptide or protein drug products. The selectivity and sensitivity of this method improves a regulator’s ability to identify peptide impurities not previously observed using the established methods and presents an opportunity to better understand the composition of complex peptide drug products. Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
The advancement of solid-phase peptide synthesis (SPPS) and recombinant technologies has resulted in an expanding class of peptide-based therapeutic molecules.[1] However, the regulatory review of these peptide products faces significant challenges as a result of the complexity of the drug substances paired with the lack of general guidelines for establishing method validation and performing comparative analytical characterization for these peptide drug products.[2–6] Depending on the methods used to produce peptide drugs, a variety of impurities may be present. Peptides generated using SPPS may feature impurities from side reactions or amino acid deletions from incomplete coupling,[7] while recombinantly produced or naturally derived products may feature impurities resulting from transcription, translation errors, or host cell products that are carried through processing. In contrast to most small molecules, peptides are susceptible to a variety of degradation pathways (e.g., oxidation or deamidation), as well as to changes in secondary structure and aggregation.[8,9] All of these different types of ’impurities’ may significantly affect the quality, safety and efficacy of the peptide drug products.[10]
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763 Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1757
* Correspondence to: M. T. Boyne II, U.S. Food and Drug Administration, CDER/OPS/OTR Division of Pharmaceutical Analysis, 10903 New Hampshire Ave, Silver Spring, MD 20993, USA. E-mail: [email protected]
Peptide drug products are generally excluded from general guidances because of the additional complexity associated with these products. Many of the current ICH guidelines regarding the testing and quality of pharmaceutical products do not include peptides, and established method validation guidances may not be directly applicable to peptides due to the general lack of available internal standards and/or reference materials.[3–5,11] In the absence of general guidelines, generic drug companies usually set their own standard for characterization and quality assessment, primarily relying on comparative testing and quality control assessments with Reference Listed Drugs (RLDs). Typically, firms develop analytical methods to detect peptide impurities in the early stages of product development, although it is quite challenging to develop methods that would be capable of identifying and resolving impurities when they are present at low levels. Unofficial guidelines have suggested that any peptide impurity present over 1% should be identified, fully characterized, and qualified based on ICH guidelines, despite peptide drugs being specifically excluded in some of these documents.[4,12] While high-performance liquid chromatography/ ultraviolet (HPLC-UV) analysis is commonly used, HPLCUV alone may not be capable of separating different impurities formed during the synthesis, production, or from degradation processes. These challenges have urged the development of advanced analytical approaches for identification and characterization of related impurities for peptide products.
A. C. Gucinski and M. T. Boyne II
1758
Techniques like liquid chromatography/mass spectrometry (LC/MS) are capable of detecting and identifying impurities at low levels, including those that co-elute, but are not widely used because LC/MS validation is challenging, often requires extensive technical expertise, and is difficult to implement in the quality control (QC) setting for complex products.[13] However, co-eluting peptides are of great regulatory concern. If such co-eluting impurities arise due to degradation of the peptide during storage, the peptide may continue to appear to meet the purity specifications from HPLC-UV analysis, but would actually have less pharmaceutical potency and may have changed the safety profile of the product.[14] Such impurities can be missed unless analytical methods specifically are designed to detect them. Protamine sulfate is a peptide-based pharmaceutical product used to counteract the effects of the anticoagulant behavior of heparin.[15] USP defines the active drug substance as a purified mixture of simple protein principles obtained from the sperm or testes of suitable species of fish, which has the property of neutralizing heparin. Each mg of protamine sulfate, calculated on a dried basis, neutralizes not less than 100 USP heparin units.[16] Pharmaceutical grade protamine is a natural product extracted from the sperm of chum salmon. Chum salmon protamine consists of four highly related major peptide components with 31 to 32 residues each. The peptides are highly basic, containing 21 to 22 arginine residues in each peptide. While many peptide species are generally amenable to reversed-phase (RP)-HPLC analysis, most reported methods are unable to resolve the four most abundant peptides in protamine sulfate.[17] While Hvass and coworkers have reported a RP-HPLC method capable of separating all four peptides, additional small peaks are also observed which to date have not been identified. Moreover, the mobile phases are not compatible with MS analysis. Thus, it has not been determined if any of the chromatographic peaks consist of co-eluting species. Despite the presence of additional low-level peaks, the current assay described in the protamine sulfate USP monograph features only a heparin titration.[16] The lack of a specific and sensitive identification test presents an opportunity for adulteration analogous to the heparin crisis.[18] Moreover, the lack of a well-defined drug substance would present a regulatory challenge in the event the source chum salmon was restricted. Ideally, all species observed in HPLC analyses would be fully characterized, as is possible with MS or nuclear magnetic resonance (NMR).[19] However, the limited relative sensitivity of NMR when compared to MSbased approaches, paired with the need for a priori knowledge of the protamine amino acid composition, suggests these orthogonal modern analytic approaches would need to be used in tandem in order to fully characterize the drug substance including the presence of impurities. In this work, we explored the suitability of intact mass spectrometry for the analysis of protamine sulfate drug products using herring protamine sulfate as a model system. The use of top-down mass spectrometry using electrontransfer dissociation for de novo sequencing of peptides has been reported previously.[20,21] Similar to chum salmon protamine, herring protamine also consists of highly basic peptides ca. 30 amino acids in length. While chum salmon consists of four products, herring protamine features only three peptides (YI, YII, and Z).[22–26] The sequences have been
wileyonlinelibrary.com/journal/rcm
well characterized using LC and thermolysis approaches. In contrast to other techniques, top-down and intact mass spectrometry do not require an a priori knowledge of the peptide sequence, nor optimization of a chromatographic separation for peptides of this size. Impurities or degradation products are separated in m/z space from the targeted protamine peptides, allowing for their identification. While the protamine peptides are not resolved from one another using standard reversed-phase chromatographic conditions, the sequence permutations allow for each of the peptides also to be separated by m/z. Because the peptide species are highly basic, electron-transfer dissociation (ETD) is well suited to also provide characterization of the peptide or impurity sequences.[27] Additionally, mass spectrometric detection features high sensitivity that allows for relatively low levels of impurity or degradation products to be identified.[28]
EXPERIMENTAL Herring protamine sulfate (grade III, histone free, Country of origin: Japan) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Solid protamine sulfate was reconstituted in Optima Grade water (Fisher, Fair Lawn, NJ, USA) to a concentration of ~10 μg/mL. Excess salt was removed from samples using Amicon Ultra 3 kDa molecular weight cut off (MWCO) filters (Millipore, Billerica, MA, USA). After MWCO filtration, samples were dried via speed-vacuum before being reconstituted in 49:49:2 water/acetonitrile/formic acid to a concentration of ~4 μg/mL for direct infusion. Samples were introduced into the mass spectrometer using a Triversa NanoMate (Advion, Ithaca, NY, USA) nanospray source. The spray voltage used was +1.8 kV with a source gas flow of 0.3 psi. MS and ETD-MS/MS experiments were performed using a LTQ Orbitrap XL mass spectrometer (Thermo Scientific, Bremen, Germany) with an operating resolution of 60,000 @ m/z 400. ETD reaction time was 100 ms with a ±3 m/z isolation window. Each MS or ETD MS/MS spectrum was the composite of 100 averaged scans. Mass spectral deconvolution was performed using the Xtract program in the Thermo QualBrowser software. Fragments ions were considered a match if within 10 ppm of the theoretical value.
RESULTS AND DISCUSSION The sequences, theoretical masses, and experimentally observed masses of herring (clupeine) protamine YI, YII, and Z are summarized in Table 1. All three major peptides are clearly observed in the intact mass spectrum shown in Fig. 1. Making identifications based on accurate mass values, several other species are also detected. Additional species include sulfate-adducted peptides as a result of the sulfate counter ion present in the drug product as well as peptides featuring the loss of arginine residues or side-chain fragments. Because of the high number of arginine residues and the multiple fragmentation pathways of the arginine side chain,[29] the prevalence of these populations is expected. While protamine peptides YI, YII, and Z were identified based on accurate mass assignments (within 2 ppm of theoretical values), detailed sequence confirmation requires tandem mass spectral(MS/MS) analysis. MS/MS spectra
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity Table 1. Summary of published herring (clupeine) protamine peptide sequences and mass values Experimental Theoretical Error (Da) MH+ (Da) MH+ (ppm)
Sequence ARRRR SSSRP PRRRT TRRRR PRRRT RRASR RPRRV SRRRR ARRRR SRRAS RRPRR VSRRR
IRRRR AGRRR R PVRRR ARRRR RPVRR RARRR R
4110.523
4110.517
1.46
4047.529
4047.521
1.98
4163.594
4163.591
0.72
Figure 1. Intact mass spectrum of herring protamine sulfate, m/z 500–550, highlighting the +8 charge state region of the spectrum. Inset: intact mass spectrum over full m/z range from m/z 450–1200.
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1759
were acquired for each of the three protamine peptides using ETD. Preliminary studies explored the possibility of using collision-induced dissociation (CID), but very low levels of fragmentation were observed as a result of the high number of basic residues in the peptides, consistent with the mobile proton model.[30] Labeled ETD MS/MS spectra along with fragment ion maps are shown in Fig. 2 with a summary of the published and observed peptide sequences listed in Table 1. (Note: all z-type ions detected were of the z• fragment ions, but are labeled z for simplicity.) The lack of fragmentation at the c9/z•22, c10/z•20 and c11/z•20 positions for clupeine protamine YI, YII and Z is consistent with established ETD fragmentation behavior, as fragmentation N-terminal to proline is disfavored as the backbone attachment of the proline side chain requires two bonds to be broken.[31]As such, the observation of the c15/z•16, c16/z•14 and c17/z•14 ions for YI, YII and Z clupine protamine N-terminal to proline is unexpected fragmentation behavior. This observation combined with the absence of fragmentation at the adjacent amino acid residue (c16/z•15, c17/z•13 and c18/z•13) suggests that the order of the arginine-proline residues at these positions are reversed and the published sequences are incorrect.[22–26,32] Further examination of the sequence determination studies for the clupeine protamine peptides supports the possibility of this sequence inversion. Suzuki and Ando used a combination of trypsin, thermolysin and carboxypeptidase digestion to deduce the sequence of YI.[26] While carboxypeptidase digestion
can identify the overall amino acid composition of the peptide, the exact sequence order cannot be deduced, even when interpreted along with the chromatographic analysis of the tryptic digest of the peptide. Instead, the sequence order of this arginine-proline pair is assigned based on expected homology with clupeine protamine Z and YII. In the report on the sequence determination of clupeine Z, Iwai and coworkers used a mixture of carboxypeptidase A and B to digest the peptide, reporting the release of nine or more arginine residues without additional explanation.[24] The release of nine arginines is consistent with our proposed sequence, while the release of ten arginines is necessary to be consistent with the sequence proposed by Iwai and coworkers. Furthermore, variability observed in the amino acid analysis of this specific peptide is consistent with either sequence. While a change in arginine-proline sequence order for these highly conserved peptides as a result of a mutation cannot be ruled out, the plausibility of the sequence inversion, as supported by the specificity afforded by MS/MS analysis, suggests that previously reported sequences were in error. Three distinct species were observed that could not be explained by adduct formation or fragmentation of the three abundant herring peptides, as denoted with (●) in Fig. 1. The +8 charge states were observed at m/z 508.322, 510.321, and 522.829, corresponding to MH+ values of 4059.522 Da, 4075.514 Da, and 4175.580 Da, respectively. These three species were present at 11%, 1.5%, and 2% intensity relative to the most abundant peak in the spectrum, which corresponds to the 8+ charge state of the herring protamine YII peptide. All other species present at >1.0% relative intensity could be assigned using accurate mass analysis. Each of the three species were isolated and subjected to fragmentation by ETD MS/MS. Fragmentation maps were constructed in order to determine the overall sequence of each species. Figure 3 shows the ETD MS/MS fragmentation spectra and maps for each of the three peptide sequence variants. A summary of the peptide sequences identified is given in Table 2, while the observed and theoretical masses along with associated mass error for each peptide are shown in Table 3. Full sequence coverage is observed for all three peptide variants, with at least one fragment ion if not both the c and z• ions present at each position, excluding positions Nterminal to proline residues. Lack of fragmentation between two residues is consistent with the position of proline residues in the sequences. Based on the observed fragment ions, the sequences of the three previously unidentified herring protamine peptides are proposed below. Sequence variants #1 and #2 display significant sequence homology with clupeine protamine YII, while variant #3 is similar to clupeine protamine Z; all three variants detected feature inversion of the most C-terminal proline residue with an arginine residue C-terminal to the proline, consistent with the more abundant peptide sequences determined above. While YII is a 30 residue peptide, variants #1 and #2 both feature 31 residues. The change in sequence occurs at the 2nd and 3rd amino acids where the variants feature an amino acid substitution for an arginine as well as an insertion; for YII, the first three residues are PRR while they are PAP in variant #1 and PSP in variant #2. The change in sequence can be explained by either the substitution of proline for arginine with an alanine/serine insertion or an alanine/ serine substitution for arginine with a proline insertion. Similarly, variant #3 increases in length to 32 amino acids
A. C. Gucinski and M. T. Boyne II
1760
Figure 2. ETD-MS/MS spectra for herring protamine YI (a), YII (b), and Z (c). Published and proposed sequences and fragment maps are inset in each spectrum with the proposed site of amino acid sequence inversion shown in bold.
wileyonlinelibrary.com/journal/rcm
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1761
Figure 3. ETD-MS/MS spectra and fragment maps for three herring protamine sequence variants.
A. C. Gucinski and M. T. Boyne II Table 2. Summary of protein herring amino acid sequences Sequence
Relation to published sequences
ARRRR SSSRP IRRRR RPRRT TRRRR AGRRR R PRRRT RRASR PVRRR RRPRV SRRRR ARRRR ARRRR SRRAS RPVRR RRRPR VSRRR RARRR R PAPRR TRRAS RPVRR RRRPR VSRRR RARRR R PSPRR TRRAS RPVRR RRRPR VSRRR RARRR R APARR RSRRA SRPVR RRRRP RVSRR RRARR RR
Protamine-YI,[26] P16↔R17 inversion Protamine-YII,[25] P17↔R18 inversion Protamine-Z,[23] P18↔R19 inversion R → P substitution at position 3, alanine insertion at position 2 relative to Protamine-YII with P18↔ R19 inversion R → P substitution at position 3, serine insertion at position 2 relative to protamine YII with P18↔R19 inversion R → P substitution at position 2, alanine insertion at position 3 relative to protamine Z with P19↔R20 inversion
Table 3. Comparison of observed and theoretical masses for observed herring protamine peptides
Sequence ARRRR PRRRT ARRRR PAPRR PSPRR APARR
SSSRP RRASR SRRAS TRRAS TRRAS RSRRA
IRRRR RPRRT PVRRR RRPRV RPVRR RRRPR RPVRR RRRPR RPVRR RRRPR SRPVR RRRRP
TRRRR SRRRR VSRRR VSRRR VSRRR RVSRR
AGRRR ARRRR RARRR RARRR RARRR RRARR
R R R R RR
relative to the 31 residues in clupeine protamine Z, with the first three residues changing from ARR to APA. This change can be explained by a proline substitution for arginine with an alanine insertion or an alanine insertion for arginine with a proline insertion.
CONCLUSIONS
1762
In this work, we have demonstrated the suitability of a combined direct infusion MS and ETD-MS/MS approach for identifying low-abundance peptides present in a peptide mixture as a case study for the characterization of peptide drugs. Previously unidentified and unreported peptides, which may result from amino acid insertions and/or substitutions, were observed and characterized. As such, this approach can be used to identify site-specific heterogeneity of peptide components even without a priori information of the peptide sequences. While MS data provided information on the presence and relative intensity of unanticipated species present, the interpretation of ETD MS/MS fragmentation patterns allowed for unequivocal identification of the herring protamine peptide sequences. Additionally, this approach identified a discrepancy among the published herring protamine sequences, as supported by ETD fragmentation patterns. This methodology presented in this work provides a straightforward framework for the characterization of a variety of protein and peptide therapeutics that may feature a mixture of peptide species, including species with sitespecific heterogeneity. Furthermore, modern analytical technologies such as the one described in this work provide information-rich assays for simultaneous structural and compositional analysis of
wileyonlinelibrary.com/journal/rcm
Experimental (Da) MH+
Theoretical (Da) MH+
Error (ppm)
4110.523 4047.529 4163.594 4059.522 4075.514 4175.580
4110.517 4047.521 4163.591 4059.510 4075.505 4175.580
1.46 1.98 0.72 2.96 1.98 0.00
complex drug products. Using methods like top-down or intact characterization of complex peptide drugs enables the drug product to be well defined analytically, which can translate to a greater assurance of drug product quality. The presence of additional peaks or intensity changes within a spectrum can be an indication of impurities, contaminants, or structure alterations that were previously unknown or not well described. The use of additional sensitive and specific fragmentation techniques allows for the unequivocal identification and characterization of these multiple components better defining the characteristics of a complex drug product.
Acknowledgements Funding and support for this work was provided by the CDER Critical Path Program (MTB) and is gratefully acknowledged. Helpful discussions with David Keire also are gratefully acknowledged. The findings and conclusions of this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any Agency determination or policy.
REFERENCES [1] F. Albericio, H. G. Kruger. Therapeutic peptides. Future Med. Chem. 2012, 4, 1527. [2] B. De Spiegeleer, V. Vergote, A. Pezeshki, K. Peremans, C. Burvenich. Impurity profiling quality control testing of synthetic peptides using liquid chromatography-photodiode array-fluorescence and liquid chromatography-electrospray ionization-mass spectrometry: the obestatin case. Anal. Biochem. 2008, 376, 229.
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity [3] U.S. Department of Health and Human Services, Food and Drug Administration Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research. Guidance for Industry: Q3A Impurities in New Drug Substances, 2008. [4] International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Impurities in New Drug Substances: Q3A (R2), 2006. [5] S. Van Dorpe, M. Verbeken, E. Wynendaele, B. De Spiegeleer. Purity profiling of peptide drugs. J. Bioanal. Biomed. 2011, S6, 003. [6] V. Vergote, C. Burvenich, C. Van de Wiele, B. De Spiegeleer. Quality specifications for peptide drugs: a regulatorypharmaceutical approach. J. Peptide Sci. 2009, 15, 697. [7] M. Schnolzer, A. Jones, P. F. Alewood, S. B. H. Kent. Ionspray tandem mass spectrometry in peptide synthesis: Structural characterization of minor by-products in the synthesis of ACP(65-74). Anal. Biochem. 1992, 204, 335. [8] D. Jain, P. K. Basniwal. Forced degradation and impurity profiling: recent trends in analytical perspectives. J. Pharmaceut. Biomed. Anal. 2013, 86, 11. [9] M. Verbeken, S. Suleman, B. Baert, E. Vangheluwe, S. Van Dorpe, C. Burvenich, L. Duchateau, F. H. Jansen, B. De Spiegeleer. Stability-indicating HPLC-DAD/UV-ESI/MS impurity profiling of the anti-malarial drug lumefantrine. Malar. J. 2011, 10, 51. [10] W. P. Kelley, S. Chen, P. D. Floyd, P. Hu, S. G. Kapsi, A. S. Kord, M. Sun, F. G. Vogt. Analytical characterization of an orally-delivered peptide pharmaceutical product. Anal. Chem. 2012, 84, 4357. [11] U.S. Department of Health and Human Services, Food and Drug Administration. Guidance for Industry: Analytical Procedures and Methods Validation, 2000. [12] International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products: Q6B, 1999. [13] C. Govaerts, E. Adams, A. Van Schepdael, J. Hoogmartens. Hyphenation of liquid chromatography to ion trap mass spectrometry to identify minor components in polypeptide antibiotics. Anal. Bioanal. Chem. 2003, 377, 909. [14] M. Verbeken, E. Wynendaele, R. A. Lefebvre, E. Goosens, B. De Spiegeleer. The influence of peptide impurity profiles on functional tissue-organ bath response : The 11-mer peptide INSL6[151–161] case. Anal. Biochem. 2012, 421, 547. [15] T. A. Gill, D. S. Singer, J. W. Thompson. Purification and analysis of protamine. Process Biochem. 2006, 41, 1875. [16] United States Pharmacopeia and National Formulary (USP 36–NF 31). Protamine Sulfate for Injection. United States Pharmacopeia, Rockville, MD, 2011. [17] A. Syncerski, J. Dudkiewics-Wilczynska, J. Tautt. Determination of protamine sulphate in drug formulations using high performance liquid chromatography. J. Pharmaceut. Biomed. Anal. 1998, 18, 907.
[18] D. A. Keire, M. L. Trehy, J. C. Reepmeyer, R. E. Kolinski, W. Ye, J. D. Dunn, B. J. Westenberger, L. F. Buhse. Analysis of crude heparin by 1H NMR, capillary electrophoresis, and stronganion-exchange-HPLC for contamination by over sulfated chondroitin sulfate. J. Pharmaceut. Biomed. Anal. 2010, 51, 921. [19] D. Awotwe-Otoo, C. Agarabi, D. Keire, S. Lee, A. Raw, L. Yu, M. J. Habib, M. A. Khan, R. B. Shah. Physicochemical characterization of complex drug substances: Evaluation of structural similarities and differences of protamine sulfate from various sources. AAPS J. 2012, 14, 619 [20] C. Jia, L. Hui, W. Cao, C. B. Lietz, X. Jiang, R. Chen, A. D. Catherman, P. M. Thomas, Y. Ge, N. L. Kelleher, L. Li. High-definition de novo sequencing of crustacean hyperglycemic hormone (CHH)-family neuropeptides. Mol. Cell. Proteomics 2012, 11, 1951. [21] C. Jia, Q. Yu, J. Wang, L. Li. Qualitative and quantitative topdown mass spectral analysis of crustacean hyperglycemic hormones in response to feeding. Proteomics 2014, 14, 1185. [22] T. Ando, K. Iwai, S. Ishii, M. Azegami, C. Nakahara. The chemical structure of one component of clupeine. Biochim. Biophys. Acta 1962, 56, 628. [23] M. Azegami, S. Ishii, T. Ando. Studies on protamines XIV. Partial chemical structure of one component of clupeine, clupeine Z. J. Biochem. 1970, 67, 523. [24] K. Iwai, C. Nakahara, T. Ando. Studies on protamines XV. The complete amino acid sequence of the Z component of clupeine. Application of N→O acyl rearrangement and selective hydrolysis in sequence determination. J. Biochem. 1971, 69, 493. [25] K. Suzuki, T. Ando. Studies on protamines XVI. The complete amino acid sequence of clupeine YII. J. Biochem. 1972, 72, 1419. [26] K. Suzuki, T. Ando. Studies on protamines XVII. The complete amino acid sequence of clupeine YI. J. Biochem. 1972, 72, 1433. [27] J. E. Syka, J. J. Coon, M. J. Schroeder, J. Shabanowitz, D. F. Hunt. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc. Natl. Acad. Sci. USA 2004, 101, 9528. [28] A. C. Gucinski, M. T. Boyne 2nd. Evaluation of intact mass spectrometry for the quantitative analysis of protein therapeutics. Anal. Chem. 2012, 84, 8045. [29] P. Y. I. Shek, J. Zhao, Y. Ke, K. W. M. Siu, A. C. Hopkinson. Fragmentations of protonated arginine, lysine, and their methylated derivatives: Concomitant losses of carbon monoxide or carbon dioxide and an amine. J. Phys. Chem. A 2006, 110, 8282. [30] A. R. Dongre, J. L. Jones, A. Somogyi, V. H. Wysocki. Influence of peptide composition, gas-phase basicity, and chemical modification on fragmentation efficiency: Evidence for the mobile proton model. J. Am. Chem. Soc. 1996, 118, 8365. [31] L. M. Mikesh, B. Ueberheide, A. Chi, J. J. Coon, J. E. P. Syka, J. Shabanowitz, D. F. Hunt. The utility of ETD mass spectrometry in proteomic analysis. Biochim. Biophys. Acta 2006, 1764, 1811. [32] T. Tobita, M. Yamasaki, T. Ando. XII. Determination of the carboxyl-terminal structure of clupeine and salmine using enzymatic procedures. J. Biochem. 1968, 63, 119.
1763
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
Revised: 28 May 2014
Accepted: 2 June 2014
Published online in Wiley Online Library
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763 (wileyonlinelibrary.com) DOI: 10.1002/rcm.6957
Identification of site-specific heterogeneity in peptide drugs using intact mass spectrometry with electron transfer dissociation Ashley C. Gucinski and Michael T. Boyne II* U.S. Food and Drug Administration, CDER/OPS/OTR Division of Pharmaceutical Analysis, 10903 New Hampshire Ave, Silver Spring, MD 20993, USA RATIONALE: Protamine sulfate is a peptide drug product consisting of multiple basic peptides. As traditional high-
performance liquid chromatography (HPLC) separation methods may not resolve these peptides, as well as any possible peptide-related impurities, a method utilizing top-down mass spectrometry was developed for the characterization of complex peptide drug products, including any low-level impurities, which is described in this study. METHODS: Herring protamine sulfate was used as a model system to demonstrate the applicability of the method. Direct infusion mass spectrometry and tandem mass spectrometry (MS/MS) on a high-resolution, mass accurate instrument with electron transfer dissociation (ETD) were used to identify all the species present in the herring protamine sulfate sample. Identifications were made based on mass accuracy analysis as well as MS/MS fragmentation patterns. RESULTS: Complete sequence coverage of the three abundant herring protamine peptides was obtained using the top-down ETD-MS/MS method, which also identified a discrepancy with the published herring protamine peptide sequences. Additionally, three low-abundance related peptide species were also identified and fully characterized. These three peptides had not previously been reported as herring protamine peptides, but could be related to the published sequences through amino acid additions and/or substitutions. CONCLUSIONS: A method for the characterization of protamine, a complex peptide drug product, was developed that can be extended to other complex peptide or protein drug products. The selectivity and sensitivity of this method improves a regulator’s ability to identify peptide impurities not previously observed using the established methods and presents an opportunity to better understand the composition of complex peptide drug products. Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
The advancement of solid-phase peptide synthesis (SPPS) and recombinant technologies has resulted in an expanding class of peptide-based therapeutic molecules.[1] However, the regulatory review of these peptide products faces significant challenges as a result of the complexity of the drug substances paired with the lack of general guidelines for establishing method validation and performing comparative analytical characterization for these peptide drug products.[2–6] Depending on the methods used to produce peptide drugs, a variety of impurities may be present. Peptides generated using SPPS may feature impurities from side reactions or amino acid deletions from incomplete coupling,[7] while recombinantly produced or naturally derived products may feature impurities resulting from transcription, translation errors, or host cell products that are carried through processing. In contrast to most small molecules, peptides are susceptible to a variety of degradation pathways (e.g., oxidation or deamidation), as well as to changes in secondary structure and aggregation.[8,9] All of these different types of ’impurities’ may significantly affect the quality, safety and efficacy of the peptide drug products.[10]
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763 Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1757
* Correspondence to: M. T. Boyne II, U.S. Food and Drug Administration, CDER/OPS/OTR Division of Pharmaceutical Analysis, 10903 New Hampshire Ave, Silver Spring, MD 20993, USA. E-mail: [email protected]
Peptide drug products are generally excluded from general guidances because of the additional complexity associated with these products. Many of the current ICH guidelines regarding the testing and quality of pharmaceutical products do not include peptides, and established method validation guidances may not be directly applicable to peptides due to the general lack of available internal standards and/or reference materials.[3–5,11] In the absence of general guidelines, generic drug companies usually set their own standard for characterization and quality assessment, primarily relying on comparative testing and quality control assessments with Reference Listed Drugs (RLDs). Typically, firms develop analytical methods to detect peptide impurities in the early stages of product development, although it is quite challenging to develop methods that would be capable of identifying and resolving impurities when they are present at low levels. Unofficial guidelines have suggested that any peptide impurity present over 1% should be identified, fully characterized, and qualified based on ICH guidelines, despite peptide drugs being specifically excluded in some of these documents.[4,12] While high-performance liquid chromatography/ ultraviolet (HPLC-UV) analysis is commonly used, HPLCUV alone may not be capable of separating different impurities formed during the synthesis, production, or from degradation processes. These challenges have urged the development of advanced analytical approaches for identification and characterization of related impurities for peptide products.
A. C. Gucinski and M. T. Boyne II
1758
Techniques like liquid chromatography/mass spectrometry (LC/MS) are capable of detecting and identifying impurities at low levels, including those that co-elute, but are not widely used because LC/MS validation is challenging, often requires extensive technical expertise, and is difficult to implement in the quality control (QC) setting for complex products.[13] However, co-eluting peptides are of great regulatory concern. If such co-eluting impurities arise due to degradation of the peptide during storage, the peptide may continue to appear to meet the purity specifications from HPLC-UV analysis, but would actually have less pharmaceutical potency and may have changed the safety profile of the product.[14] Such impurities can be missed unless analytical methods specifically are designed to detect them. Protamine sulfate is a peptide-based pharmaceutical product used to counteract the effects of the anticoagulant behavior of heparin.[15] USP defines the active drug substance as a purified mixture of simple protein principles obtained from the sperm or testes of suitable species of fish, which has the property of neutralizing heparin. Each mg of protamine sulfate, calculated on a dried basis, neutralizes not less than 100 USP heparin units.[16] Pharmaceutical grade protamine is a natural product extracted from the sperm of chum salmon. Chum salmon protamine consists of four highly related major peptide components with 31 to 32 residues each. The peptides are highly basic, containing 21 to 22 arginine residues in each peptide. While many peptide species are generally amenable to reversed-phase (RP)-HPLC analysis, most reported methods are unable to resolve the four most abundant peptides in protamine sulfate.[17] While Hvass and coworkers have reported a RP-HPLC method capable of separating all four peptides, additional small peaks are also observed which to date have not been identified. Moreover, the mobile phases are not compatible with MS analysis. Thus, it has not been determined if any of the chromatographic peaks consist of co-eluting species. Despite the presence of additional low-level peaks, the current assay described in the protamine sulfate USP monograph features only a heparin titration.[16] The lack of a specific and sensitive identification test presents an opportunity for adulteration analogous to the heparin crisis.[18] Moreover, the lack of a well-defined drug substance would present a regulatory challenge in the event the source chum salmon was restricted. Ideally, all species observed in HPLC analyses would be fully characterized, as is possible with MS or nuclear magnetic resonance (NMR).[19] However, the limited relative sensitivity of NMR when compared to MSbased approaches, paired with the need for a priori knowledge of the protamine amino acid composition, suggests these orthogonal modern analytic approaches would need to be used in tandem in order to fully characterize the drug substance including the presence of impurities. In this work, we explored the suitability of intact mass spectrometry for the analysis of protamine sulfate drug products using herring protamine sulfate as a model system. The use of top-down mass spectrometry using electrontransfer dissociation for de novo sequencing of peptides has been reported previously.[20,21] Similar to chum salmon protamine, herring protamine also consists of highly basic peptides ca. 30 amino acids in length. While chum salmon consists of four products, herring protamine features only three peptides (YI, YII, and Z).[22–26] The sequences have been
wileyonlinelibrary.com/journal/rcm
well characterized using LC and thermolysis approaches. In contrast to other techniques, top-down and intact mass spectrometry do not require an a priori knowledge of the peptide sequence, nor optimization of a chromatographic separation for peptides of this size. Impurities or degradation products are separated in m/z space from the targeted protamine peptides, allowing for their identification. While the protamine peptides are not resolved from one another using standard reversed-phase chromatographic conditions, the sequence permutations allow for each of the peptides also to be separated by m/z. Because the peptide species are highly basic, electron-transfer dissociation (ETD) is well suited to also provide characterization of the peptide or impurity sequences.[27] Additionally, mass spectrometric detection features high sensitivity that allows for relatively low levels of impurity or degradation products to be identified.[28]
EXPERIMENTAL Herring protamine sulfate (grade III, histone free, Country of origin: Japan) was obtained from Sigma-Aldrich (St. Louis, MO, USA). Solid protamine sulfate was reconstituted in Optima Grade water (Fisher, Fair Lawn, NJ, USA) to a concentration of ~10 μg/mL. Excess salt was removed from samples using Amicon Ultra 3 kDa molecular weight cut off (MWCO) filters (Millipore, Billerica, MA, USA). After MWCO filtration, samples were dried via speed-vacuum before being reconstituted in 49:49:2 water/acetonitrile/formic acid to a concentration of ~4 μg/mL for direct infusion. Samples were introduced into the mass spectrometer using a Triversa NanoMate (Advion, Ithaca, NY, USA) nanospray source. The spray voltage used was +1.8 kV with a source gas flow of 0.3 psi. MS and ETD-MS/MS experiments were performed using a LTQ Orbitrap XL mass spectrometer (Thermo Scientific, Bremen, Germany) with an operating resolution of 60,000 @ m/z 400. ETD reaction time was 100 ms with a ±3 m/z isolation window. Each MS or ETD MS/MS spectrum was the composite of 100 averaged scans. Mass spectral deconvolution was performed using the Xtract program in the Thermo QualBrowser software. Fragments ions were considered a match if within 10 ppm of the theoretical value.
RESULTS AND DISCUSSION The sequences, theoretical masses, and experimentally observed masses of herring (clupeine) protamine YI, YII, and Z are summarized in Table 1. All three major peptides are clearly observed in the intact mass spectrum shown in Fig. 1. Making identifications based on accurate mass values, several other species are also detected. Additional species include sulfate-adducted peptides as a result of the sulfate counter ion present in the drug product as well as peptides featuring the loss of arginine residues or side-chain fragments. Because of the high number of arginine residues and the multiple fragmentation pathways of the arginine side chain,[29] the prevalence of these populations is expected. While protamine peptides YI, YII, and Z were identified based on accurate mass assignments (within 2 ppm of theoretical values), detailed sequence confirmation requires tandem mass spectral(MS/MS) analysis. MS/MS spectra
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity Table 1. Summary of published herring (clupeine) protamine peptide sequences and mass values Experimental Theoretical Error (Da) MH+ (Da) MH+ (ppm)
Sequence ARRRR SSSRP PRRRT TRRRR PRRRT RRASR RPRRV SRRRR ARRRR SRRAS RRPRR VSRRR
IRRRR AGRRR R PVRRR ARRRR RPVRR RARRR R
4110.523
4110.517
1.46
4047.529
4047.521
1.98
4163.594
4163.591
0.72
Figure 1. Intact mass spectrum of herring protamine sulfate, m/z 500–550, highlighting the +8 charge state region of the spectrum. Inset: intact mass spectrum over full m/z range from m/z 450–1200.
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1759
were acquired for each of the three protamine peptides using ETD. Preliminary studies explored the possibility of using collision-induced dissociation (CID), but very low levels of fragmentation were observed as a result of the high number of basic residues in the peptides, consistent with the mobile proton model.[30] Labeled ETD MS/MS spectra along with fragment ion maps are shown in Fig. 2 with a summary of the published and observed peptide sequences listed in Table 1. (Note: all z-type ions detected were of the z• fragment ions, but are labeled z for simplicity.) The lack of fragmentation at the c9/z•22, c10/z•20 and c11/z•20 positions for clupeine protamine YI, YII and Z is consistent with established ETD fragmentation behavior, as fragmentation N-terminal to proline is disfavored as the backbone attachment of the proline side chain requires two bonds to be broken.[31]As such, the observation of the c15/z•16, c16/z•14 and c17/z•14 ions for YI, YII and Z clupine protamine N-terminal to proline is unexpected fragmentation behavior. This observation combined with the absence of fragmentation at the adjacent amino acid residue (c16/z•15, c17/z•13 and c18/z•13) suggests that the order of the arginine-proline residues at these positions are reversed and the published sequences are incorrect.[22–26,32] Further examination of the sequence determination studies for the clupeine protamine peptides supports the possibility of this sequence inversion. Suzuki and Ando used a combination of trypsin, thermolysin and carboxypeptidase digestion to deduce the sequence of YI.[26] While carboxypeptidase digestion
can identify the overall amino acid composition of the peptide, the exact sequence order cannot be deduced, even when interpreted along with the chromatographic analysis of the tryptic digest of the peptide. Instead, the sequence order of this arginine-proline pair is assigned based on expected homology with clupeine protamine Z and YII. In the report on the sequence determination of clupeine Z, Iwai and coworkers used a mixture of carboxypeptidase A and B to digest the peptide, reporting the release of nine or more arginine residues without additional explanation.[24] The release of nine arginines is consistent with our proposed sequence, while the release of ten arginines is necessary to be consistent with the sequence proposed by Iwai and coworkers. Furthermore, variability observed in the amino acid analysis of this specific peptide is consistent with either sequence. While a change in arginine-proline sequence order for these highly conserved peptides as a result of a mutation cannot be ruled out, the plausibility of the sequence inversion, as supported by the specificity afforded by MS/MS analysis, suggests that previously reported sequences were in error. Three distinct species were observed that could not be explained by adduct formation or fragmentation of the three abundant herring peptides, as denoted with (●) in Fig. 1. The +8 charge states were observed at m/z 508.322, 510.321, and 522.829, corresponding to MH+ values of 4059.522 Da, 4075.514 Da, and 4175.580 Da, respectively. These three species were present at 11%, 1.5%, and 2% intensity relative to the most abundant peak in the spectrum, which corresponds to the 8+ charge state of the herring protamine YII peptide. All other species present at >1.0% relative intensity could be assigned using accurate mass analysis. Each of the three species were isolated and subjected to fragmentation by ETD MS/MS. Fragmentation maps were constructed in order to determine the overall sequence of each species. Figure 3 shows the ETD MS/MS fragmentation spectra and maps for each of the three peptide sequence variants. A summary of the peptide sequences identified is given in Table 2, while the observed and theoretical masses along with associated mass error for each peptide are shown in Table 3. Full sequence coverage is observed for all three peptide variants, with at least one fragment ion if not both the c and z• ions present at each position, excluding positions Nterminal to proline residues. Lack of fragmentation between two residues is consistent with the position of proline residues in the sequences. Based on the observed fragment ions, the sequences of the three previously unidentified herring protamine peptides are proposed below. Sequence variants #1 and #2 display significant sequence homology with clupeine protamine YII, while variant #3 is similar to clupeine protamine Z; all three variants detected feature inversion of the most C-terminal proline residue with an arginine residue C-terminal to the proline, consistent with the more abundant peptide sequences determined above. While YII is a 30 residue peptide, variants #1 and #2 both feature 31 residues. The change in sequence occurs at the 2nd and 3rd amino acids where the variants feature an amino acid substitution for an arginine as well as an insertion; for YII, the first three residues are PRR while they are PAP in variant #1 and PSP in variant #2. The change in sequence can be explained by either the substitution of proline for arginine with an alanine/serine insertion or an alanine/ serine substitution for arginine with a proline insertion. Similarly, variant #3 increases in length to 32 amino acids
A. C. Gucinski and M. T. Boyne II
1760
Figure 2. ETD-MS/MS spectra for herring protamine YI (a), YII (b), and Z (c). Published and proposed sequences and fragment maps are inset in each spectrum with the proposed site of amino acid sequence inversion shown in bold.
wileyonlinelibrary.com/journal/rcm
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
1761
Figure 3. ETD-MS/MS spectra and fragment maps for three herring protamine sequence variants.
A. C. Gucinski and M. T. Boyne II Table 2. Summary of protein herring amino acid sequences Sequence
Relation to published sequences
ARRRR SSSRP IRRRR RPRRT TRRRR AGRRR R PRRRT RRASR PVRRR RRPRV SRRRR ARRRR ARRRR SRRAS RPVRR RRRPR VSRRR RARRR R PAPRR TRRAS RPVRR RRRPR VSRRR RARRR R PSPRR TRRAS RPVRR RRRPR VSRRR RARRR R APARR RSRRA SRPVR RRRRP RVSRR RRARR RR
Protamine-YI,[26] P16↔R17 inversion Protamine-YII,[25] P17↔R18 inversion Protamine-Z,[23] P18↔R19 inversion R → P substitution at position 3, alanine insertion at position 2 relative to Protamine-YII with P18↔ R19 inversion R → P substitution at position 3, serine insertion at position 2 relative to protamine YII with P18↔R19 inversion R → P substitution at position 2, alanine insertion at position 3 relative to protamine Z with P19↔R20 inversion
Table 3. Comparison of observed and theoretical masses for observed herring protamine peptides
Sequence ARRRR PRRRT ARRRR PAPRR PSPRR APARR
SSSRP RRASR SRRAS TRRAS TRRAS RSRRA
IRRRR RPRRT PVRRR RRPRV RPVRR RRRPR RPVRR RRRPR RPVRR RRRPR SRPVR RRRRP
TRRRR SRRRR VSRRR VSRRR VSRRR RVSRR
AGRRR ARRRR RARRR RARRR RARRR RRARR
R R R R RR
relative to the 31 residues in clupeine protamine Z, with the first three residues changing from ARR to APA. This change can be explained by a proline substitution for arginine with an alanine insertion or an alanine insertion for arginine with a proline insertion.
CONCLUSIONS
1762
In this work, we have demonstrated the suitability of a combined direct infusion MS and ETD-MS/MS approach for identifying low-abundance peptides present in a peptide mixture as a case study for the characterization of peptide drugs. Previously unidentified and unreported peptides, which may result from amino acid insertions and/or substitutions, were observed and characterized. As such, this approach can be used to identify site-specific heterogeneity of peptide components even without a priori information of the peptide sequences. While MS data provided information on the presence and relative intensity of unanticipated species present, the interpretation of ETD MS/MS fragmentation patterns allowed for unequivocal identification of the herring protamine peptide sequences. Additionally, this approach identified a discrepancy among the published herring protamine sequences, as supported by ETD fragmentation patterns. This methodology presented in this work provides a straightforward framework for the characterization of a variety of protein and peptide therapeutics that may feature a mixture of peptide species, including species with sitespecific heterogeneity. Furthermore, modern analytical technologies such as the one described in this work provide information-rich assays for simultaneous structural and compositional analysis of
wileyonlinelibrary.com/journal/rcm
Experimental (Da) MH+
Theoretical (Da) MH+
Error (ppm)
4110.523 4047.529 4163.594 4059.522 4075.514 4175.580
4110.517 4047.521 4163.591 4059.510 4075.505 4175.580
1.46 1.98 0.72 2.96 1.98 0.00
complex drug products. Using methods like top-down or intact characterization of complex peptide drugs enables the drug product to be well defined analytically, which can translate to a greater assurance of drug product quality. The presence of additional peaks or intensity changes within a spectrum can be an indication of impurities, contaminants, or structure alterations that were previously unknown or not well described. The use of additional sensitive and specific fragmentation techniques allows for the unequivocal identification and characterization of these multiple components better defining the characteristics of a complex drug product.
Acknowledgements Funding and support for this work was provided by the CDER Critical Path Program (MTB) and is gratefully acknowledged. Helpful discussions with David Keire also are gratefully acknowledged. The findings and conclusions of this article have not been formally disseminated by the Food and Drug Administration and should not be construed to represent any Agency determination or policy.
REFERENCES [1] F. Albericio, H. G. Kruger. Therapeutic peptides. Future Med. Chem. 2012, 4, 1527. [2] B. De Spiegeleer, V. Vergote, A. Pezeshki, K. Peremans, C. Burvenich. Impurity profiling quality control testing of synthetic peptides using liquid chromatography-photodiode array-fluorescence and liquid chromatography-electrospray ionization-mass spectrometry: the obestatin case. Anal. Biochem. 2008, 376, 229.
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
ETD of peptide drugs identifies site-specific heterogeneity [3] U.S. Department of Health and Human Services, Food and Drug Administration Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research. Guidance for Industry: Q3A Impurities in New Drug Substances, 2008. [4] International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Impurities in New Drug Substances: Q3A (R2), 2006. [5] S. Van Dorpe, M. Verbeken, E. Wynendaele, B. De Spiegeleer. Purity profiling of peptide drugs. J. Bioanal. Biomed. 2011, S6, 003. [6] V. Vergote, C. Burvenich, C. Van de Wiele, B. De Spiegeleer. Quality specifications for peptide drugs: a regulatorypharmaceutical approach. J. Peptide Sci. 2009, 15, 697. [7] M. Schnolzer, A. Jones, P. F. Alewood, S. B. H. Kent. Ionspray tandem mass spectrometry in peptide synthesis: Structural characterization of minor by-products in the synthesis of ACP(65-74). Anal. Biochem. 1992, 204, 335. [8] D. Jain, P. K. Basniwal. Forced degradation and impurity profiling: recent trends in analytical perspectives. J. Pharmaceut. Biomed. Anal. 2013, 86, 11. [9] M. Verbeken, S. Suleman, B. Baert, E. Vangheluwe, S. Van Dorpe, C. Burvenich, L. Duchateau, F. H. Jansen, B. De Spiegeleer. Stability-indicating HPLC-DAD/UV-ESI/MS impurity profiling of the anti-malarial drug lumefantrine. Malar. J. 2011, 10, 51. [10] W. P. Kelley, S. Chen, P. D. Floyd, P. Hu, S. G. Kapsi, A. S. Kord, M. Sun, F. G. Vogt. Analytical characterization of an orally-delivered peptide pharmaceutical product. Anal. Chem. 2012, 84, 4357. [11] U.S. Department of Health and Human Services, Food and Drug Administration. Guidance for Industry: Analytical Procedures and Methods Validation, 2000. [12] International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. Specifications: Test Procedures and Acceptance Criteria for Biotechnological/Biological Products: Q6B, 1999. [13] C. Govaerts, E. Adams, A. Van Schepdael, J. Hoogmartens. Hyphenation of liquid chromatography to ion trap mass spectrometry to identify minor components in polypeptide antibiotics. Anal. Bioanal. Chem. 2003, 377, 909. [14] M. Verbeken, E. Wynendaele, R. A. Lefebvre, E. Goosens, B. De Spiegeleer. The influence of peptide impurity profiles on functional tissue-organ bath response : The 11-mer peptide INSL6[151–161] case. Anal. Biochem. 2012, 421, 547. [15] T. A. Gill, D. S. Singer, J. W. Thompson. Purification and analysis of protamine. Process Biochem. 2006, 41, 1875. [16] United States Pharmacopeia and National Formulary (USP 36–NF 31). Protamine Sulfate for Injection. United States Pharmacopeia, Rockville, MD, 2011. [17] A. Syncerski, J. Dudkiewics-Wilczynska, J. Tautt. Determination of protamine sulphate in drug formulations using high performance liquid chromatography. J. Pharmaceut. Biomed. Anal. 1998, 18, 907.
[18] D. A. Keire, M. L. Trehy, J. C. Reepmeyer, R. E. Kolinski, W. Ye, J. D. Dunn, B. J. Westenberger, L. F. Buhse. Analysis of crude heparin by 1H NMR, capillary electrophoresis, and stronganion-exchange-HPLC for contamination by over sulfated chondroitin sulfate. J. Pharmaceut. Biomed. Anal. 2010, 51, 921. [19] D. Awotwe-Otoo, C. Agarabi, D. Keire, S. Lee, A. Raw, L. Yu, M. J. Habib, M. A. Khan, R. B. Shah. Physicochemical characterization of complex drug substances: Evaluation of structural similarities and differences of protamine sulfate from various sources. AAPS J. 2012, 14, 619 [20] C. Jia, L. Hui, W. Cao, C. B. Lietz, X. Jiang, R. Chen, A. D. Catherman, P. M. Thomas, Y. Ge, N. L. Kelleher, L. Li. High-definition de novo sequencing of crustacean hyperglycemic hormone (CHH)-family neuropeptides. Mol. Cell. Proteomics 2012, 11, 1951. [21] C. Jia, Q. Yu, J. Wang, L. Li. Qualitative and quantitative topdown mass spectral analysis of crustacean hyperglycemic hormones in response to feeding. Proteomics 2014, 14, 1185. [22] T. Ando, K. Iwai, S. Ishii, M. Azegami, C. Nakahara. The chemical structure of one component of clupeine. Biochim. Biophys. Acta 1962, 56, 628. [23] M. Azegami, S. Ishii, T. Ando. Studies on protamines XIV. Partial chemical structure of one component of clupeine, clupeine Z. J. Biochem. 1970, 67, 523. [24] K. Iwai, C. Nakahara, T. Ando. Studies on protamines XV. The complete amino acid sequence of the Z component of clupeine. Application of N→O acyl rearrangement and selective hydrolysis in sequence determination. J. Biochem. 1971, 69, 493. [25] K. Suzuki, T. Ando. Studies on protamines XVI. The complete amino acid sequence of clupeine YII. J. Biochem. 1972, 72, 1419. [26] K. Suzuki, T. Ando. Studies on protamines XVII. The complete amino acid sequence of clupeine YI. J. Biochem. 1972, 72, 1433. [27] J. E. Syka, J. J. Coon, M. J. Schroeder, J. Shabanowitz, D. F. Hunt. Peptide and protein sequence analysis by electron transfer dissociation mass spectrometry. Proc. Natl. Acad. Sci. USA 2004, 101, 9528. [28] A. C. Gucinski, M. T. Boyne 2nd. Evaluation of intact mass spectrometry for the quantitative analysis of protein therapeutics. Anal. Chem. 2012, 84, 8045. [29] P. Y. I. Shek, J. Zhao, Y. Ke, K. W. M. Siu, A. C. Hopkinson. Fragmentations of protonated arginine, lysine, and their methylated derivatives: Concomitant losses of carbon monoxide or carbon dioxide and an amine. J. Phys. Chem. A 2006, 110, 8282. [30] A. R. Dongre, J. L. Jones, A. Somogyi, V. H. Wysocki. Influence of peptide composition, gas-phase basicity, and chemical modification on fragmentation efficiency: Evidence for the mobile proton model. J. Am. Chem. Soc. 1996, 118, 8365. [31] L. M. Mikesh, B. Ueberheide, A. Chi, J. J. Coon, J. E. P. Syka, J. Shabanowitz, D. F. Hunt. The utility of ETD mass spectrometry in proteomic analysis. Biochim. Biophys. Acta 2006, 1764, 1811. [32] T. Tobita, M. Yamasaki, T. Ando. XII. Determination of the carboxyl-terminal structure of clupeine and salmine using enzymatic procedures. J. Biochem. 1968, 63, 119.
1763
Rapid Commun. Mass Spectrom. 2014, 28, 1757–1763
wileyonlinelibrary.com/journal/rcm
Published in 2014. This article is a U.S. Government work and is in the public domain in the USA.
