Anal Bioanal Chem (2015) 407:749–759 DOI 10.1007/s00216-014-8172-2
PAPER IN FOREFRONT
Modern analytics for naturally derived complex drug substances: NMR and MS tests for protamine sulfate from chum salmon Ashley C. Gucinski & Michael T. Boyne II & David A. Keire
Received: 1 July 2014 / Revised: 1 August 2014 / Accepted: 7 August 2014 / Published online: 27 September 2014 # Springer-Verlag Berlin Heidelberg (outside the USA) 2014
Abstract This work describes orthogonal NMR and MS tests for the structure and composition of the drug protamine sulfate derived from chum salmon. The spectral response pattern obtained by 1D-1H-NMR and MS methods from salmon protamine, a mixture of four predominant peptide chains, is dependent on the amino acid sequence and abundance of each peptide. Thus, an assay was developed based on the ratios of alanine, glycine and arginine amino acid residue NMR peaks (relative to the arginine CδH proton signal) in this mixture that are unique to the salmon source. In addition, MS analysis provided sensitive sequence determination and impurity analysis based on shifts from exact masses. Spectra from protamine sulfate active pharmaceutical ingredient (API) suppliers and from a formulated drug product purchased from the US market were examined. Based on these marketplace survey data, NMR acceptance criteria for chum salmon derived protamine sulfate could be based on the absence of aromatic amino acid signals and on ratios of Ala βH/Arg δH, Gly αH/Arg δH and Arg αH/Arg δH integrated areas of 2.4± 1 %, 9.4±3 % and 50±5 %, respectively. For MS, acceptance criteria based on the presence of specific mass to charge (m/z) ratio peaks (m/z=+8 of 530.455, 540.841, 532.208 and 508.950) could be used for the four major peptides present in the mixture with relative abundances of 17±1 %, 31±2 %, 27±1 % and 25±3 %, respectively. The specificity of the combined NMR and MS assay was tested by comparison to data obtained from herring protamine which contains a different mixture of peptides with related amino acid sequences.
Published in the topical collection celebrating ABCs 13th Anniversary. A. C. Gucinski : M. T. Boyne II : D. A. Keire (*) Division of Pharmaceutical Research, Office of Testing and Research, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, 645 S. Newstead Ave., St. Louis, MO 63110, USA e-mail: [email protected]
Both assays were able to clearly distinguish protamine derived from these different natural sources. Keywords NMR . LC-MS . Peptide drug quality
Introduction Protamines are a class of highly positively charged peptides whose natural function is to package DNA in sperm nuclei at high densities [1]. Protamine is also used as a drug product (i.e. protamine sulfate) which is isolated from chum salmon testes and used as an antagonist to heparin anticoagulation activity post-surgery or as a stabilizing agent in the formulation of long acting insulin products. Protamine sulfate is one of the few reversing agents for heparin and is also partially effective in reversing low molecular weight heparin (LMWH) activity [2,3]. Protamine sulfate is widely used in large amounts and is on the World Health Organization list of essential medicines. For example, a one year drug use evaluation of protamine and heparin from a single 263 bed acute care US-based general hospital across 239 procedures estimated the use of heparin and protamine at 6.3 million IU of heparin (or 35 g using a conversion factor of 180 IU/mg) and 52 g protamine, respectively [4]. This usage can be further extrapolated using the American Hospital Association 2012 Annual Survey which found 5,723 registered hospitals with a total of 920,829 beds [5]. Thus, at 52 g of protamine per 263 bed-years an estimated ~183,000 g of protamine is used in these hospitals per year. The use of ~122,000 g of heparin sodium can be estimated in the same manner. The primary source of chum salmon used for protamine sulfate drug products in the US is the Japanese fishery. After the earthquake and subsequent tsunami in Japan in March of
750
2011, the Japanese salmon fishing grounds were moved from the Honshu fishing area to a new area north of Hokkaido island (see [6]). Thus, because there could be variation in protamines derived from different fishing grounds, methods sensitive to the structure and composition of the drug are needed to assure its quality. The current United States Pharmacopeia (USP) monograph for protamine sulfate drug substance has no test for the structure or composition of this naturally derived complex drug substance. Tests listed in the USP National Formulary (USP 37 NF37) include a heparin neutralization assay, nitrogen content determination , UV absorbance , loss on drying and sulfate content. Therefore, the FDA initiated studies to assess methods for surveillance of this drug [7,8]. For example, Awotwe-Otoo et al. tested 1D 1H-NMR in 90 % water, thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), Fourier transform infrared (FTIR) spectrometry and circular dichroism spectropolarimetry (CD) for protamine sulfate API quality assessment. These authors determined that the protamine sulfate peptides were largely unstructured in solution and that there were differences in thermal properties between APIs from different sources. Because these peptides are unstructured in solution, the FTIR and CD methods which respond to secondary structure elements would not be sensitive to sequence differences unless they caused such structures to form. Furthermore, the thermal methods that assess bulk properties of the powdered drug are not amenable to direct testing of the formulated drug product. The 1D 1H-NMR test implemented by AwotweOtoo et al. used 95 % water and 5 % D2O and was sensitive to the composition and sequence of the drug. However, the need to use NMR water suppression for this assay limits its applicability as a regulatory test because of inter-laboratory variability. In a separate report, Awotwe-Otoo et al. applied quality by design (QbD) concepts to optimize HPLC conditions used to separate and quantify protamine peptides [8]. However, the traditional HPLC-UV chromatography approach to assess protamine peptide composition has well-known limitations. First, protamine consists of highly charged peptides, which are not well retained or separated on common reverse phase chromatography columns using organic/aqueous gradient methods. Second, chum salmon protamine sequences lack aromatic amino acids so that short wavelength UV (~
PAPER IN FOREFRONT
Modern analytics for naturally derived complex drug substances: NMR and MS tests for protamine sulfate from chum salmon Ashley C. Gucinski & Michael T. Boyne II & David A. Keire
Received: 1 July 2014 / Revised: 1 August 2014 / Accepted: 7 August 2014 / Published online: 27 September 2014 # Springer-Verlag Berlin Heidelberg (outside the USA) 2014
Abstract This work describes orthogonal NMR and MS tests for the structure and composition of the drug protamine sulfate derived from chum salmon. The spectral response pattern obtained by 1D-1H-NMR and MS methods from salmon protamine, a mixture of four predominant peptide chains, is dependent on the amino acid sequence and abundance of each peptide. Thus, an assay was developed based on the ratios of alanine, glycine and arginine amino acid residue NMR peaks (relative to the arginine CδH proton signal) in this mixture that are unique to the salmon source. In addition, MS analysis provided sensitive sequence determination and impurity analysis based on shifts from exact masses. Spectra from protamine sulfate active pharmaceutical ingredient (API) suppliers and from a formulated drug product purchased from the US market were examined. Based on these marketplace survey data, NMR acceptance criteria for chum salmon derived protamine sulfate could be based on the absence of aromatic amino acid signals and on ratios of Ala βH/Arg δH, Gly αH/Arg δH and Arg αH/Arg δH integrated areas of 2.4± 1 %, 9.4±3 % and 50±5 %, respectively. For MS, acceptance criteria based on the presence of specific mass to charge (m/z) ratio peaks (m/z=+8 of 530.455, 540.841, 532.208 and 508.950) could be used for the four major peptides present in the mixture with relative abundances of 17±1 %, 31±2 %, 27±1 % and 25±3 %, respectively. The specificity of the combined NMR and MS assay was tested by comparison to data obtained from herring protamine which contains a different mixture of peptides with related amino acid sequences.
Published in the topical collection celebrating ABCs 13th Anniversary. A. C. Gucinski : M. T. Boyne II : D. A. Keire (*) Division of Pharmaceutical Research, Office of Testing and Research, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, 645 S. Newstead Ave., St. Louis, MO 63110, USA e-mail: [email protected]
Both assays were able to clearly distinguish protamine derived from these different natural sources. Keywords NMR . LC-MS . Peptide drug quality
Introduction Protamines are a class of highly positively charged peptides whose natural function is to package DNA in sperm nuclei at high densities [1]. Protamine is also used as a drug product (i.e. protamine sulfate) which is isolated from chum salmon testes and used as an antagonist to heparin anticoagulation activity post-surgery or as a stabilizing agent in the formulation of long acting insulin products. Protamine sulfate is one of the few reversing agents for heparin and is also partially effective in reversing low molecular weight heparin (LMWH) activity [2,3]. Protamine sulfate is widely used in large amounts and is on the World Health Organization list of essential medicines. For example, a one year drug use evaluation of protamine and heparin from a single 263 bed acute care US-based general hospital across 239 procedures estimated the use of heparin and protamine at 6.3 million IU of heparin (or 35 g using a conversion factor of 180 IU/mg) and 52 g protamine, respectively [4]. This usage can be further extrapolated using the American Hospital Association 2012 Annual Survey which found 5,723 registered hospitals with a total of 920,829 beds [5]. Thus, at 52 g of protamine per 263 bed-years an estimated ~183,000 g of protamine is used in these hospitals per year. The use of ~122,000 g of heparin sodium can be estimated in the same manner. The primary source of chum salmon used for protamine sulfate drug products in the US is the Japanese fishery. After the earthquake and subsequent tsunami in Japan in March of
750
2011, the Japanese salmon fishing grounds were moved from the Honshu fishing area to a new area north of Hokkaido island (see [6]). Thus, because there could be variation in protamines derived from different fishing grounds, methods sensitive to the structure and composition of the drug are needed to assure its quality. The current United States Pharmacopeia (USP) monograph for protamine sulfate drug substance has no test for the structure or composition of this naturally derived complex drug substance. Tests listed in the USP National Formulary (USP 37 NF37) include a heparin neutralization assay, nitrogen content determination , UV absorbance , loss on drying and sulfate content. Therefore, the FDA initiated studies to assess methods for surveillance of this drug [7,8]. For example, Awotwe-Otoo et al. tested 1D 1H-NMR in 90 % water, thermogravimetric analysis (TGA), differential scanning calorimetry (DSC), Fourier transform infrared (FTIR) spectrometry and circular dichroism spectropolarimetry (CD) for protamine sulfate API quality assessment. These authors determined that the protamine sulfate peptides were largely unstructured in solution and that there were differences in thermal properties between APIs from different sources. Because these peptides are unstructured in solution, the FTIR and CD methods which respond to secondary structure elements would not be sensitive to sequence differences unless they caused such structures to form. Furthermore, the thermal methods that assess bulk properties of the powdered drug are not amenable to direct testing of the formulated drug product. The 1D 1H-NMR test implemented by AwotweOtoo et al. used 95 % water and 5 % D2O and was sensitive to the composition and sequence of the drug. However, the need to use NMR water suppression for this assay limits its applicability as a regulatory test because of inter-laboratory variability. In a separate report, Awotwe-Otoo et al. applied quality by design (QbD) concepts to optimize HPLC conditions used to separate and quantify protamine peptides [8]. However, the traditional HPLC-UV chromatography approach to assess protamine peptide composition has well-known limitations. First, protamine consists of highly charged peptides, which are not well retained or separated on common reverse phase chromatography columns using organic/aqueous gradient methods. Second, chum salmon protamine sequences lack aromatic amino acids so that short wavelength UV (~
