The production and sale of counterfeit drugs has risen sharply in recent years. The World Health Organization (WHO) estimates that counterfeit medicines account for approximately 1% of sales in developed countries and well over 10% in developing countries. These substandard versions of medications not only represent a significant safety threat to patients, but also challenge the credibility of the pharmaceutical industry and its ability to provide patients with safe and effective products. These counterfeit products may contain either the incorrect dose or none of the intended active compound. In some cases, these counterfeit medicines contain different active drug components and, in the worst cases, may even contain toxic substances. Because of these concerns, the pharmaceutical industry vigilantly monitors the global market for counterfeit products. When a suspected counterfeit product is detected at Bristol-Myers Squibb, it is fully characterized to assess the potential risks to patient safety. Liquid chromatography coupled to mass spectrometry (LC–MS), with accurate-mass capability, is a powerful tool for investigating counterfeit pharmaceuticals because it allows the rapid assignment of the molecular weight and formula of each component, which can then be used to search the literature or internet for a potential match. This approach is frequently successful because counterfeiters often formulate their products from commonly available, relatively simple materials, rather than novel products. 

The World Health Organization (WHO) defines 

 as drugs that are "deliberately mislabeled with respect to identity and/or source. Counterfeiting can apply to both branded and generic products with counterfeit products including drugs with the correct ingredients or with the wrong ingredients; without active ingredients, with insufficient active ingredient or with fake packaging" (1). Counterfeit pharmaceutical products have been detected since approximately 1990 (2). Since then, the number of cases investigated by the US Food and Drug Administration (FDA) quadrupled to an average of about 20 per year in 2001 and 2002 (3). In an effort to ensure patient safety and brand integrity, the pharmaceutical industry vigilantly monitors the global market for counterfeit products. Counterfeit products at Bristol-Myers Squibb are fully characterized to assess the risks to patient health and safety.  

Figure 1: Decision-tree diagram of the general strategy for counterfeit product analysis.			

The first step in detecting a counterfeit medicine is to conduct a visual inspection of its physical characteristics as well as the appearance of the accompanying packaging materials. The effectiveness of a visual inspection may have limited success given the increased sophistication adopted by counterfeiters. An even greater challenge than identifying a pharmaceutical product as a counterfeit is the identification and quantitation of all components present in the material. This process is necessary to make an accurate assessment of toxicology and patient risk. Identification and quantitation of components in a counterfeit product requires chemical analysis. Throughout the industry, various modern analytical techniques have been applied for the characterization of counterfeit pharmaceuticals, including thin-layer chromatography (TLC) (4), gas chromatography (GC) (5), high performance liquid chromatography (HPLC) (6), Raman and near infrared (NIR) spectroscopy (7–11), mass spectrometry (MS) (12,13), and nuclear magnetic resonance (NMR) spectroscopy (14,15). Each of these techniques is capable of providing rich analytical data to assist in the characterization of counterfeit medicines. In practice, two or more of these techniques will be used orthogonally because the ideal goal of these investigations is to achieve absolute identification and quantitation of all components that are present (16).  

Figure 2: LCâUV chromatogram (210 nm) of product A tablet extract.			

To enable the complete characterization of detected counterfeit medicines, a general strategy has been developed that uses vibrational spectroscopy (Raman and NIR), liquid chromatography coupled to mass spectrometry (LC–UV-MS), and NMR as an orthogonal structural confirmational technique. An overview of this strategy is depicted in Figure 1. A visual inspection followed by analysis with Raman or NIR spectroscopy is used as an initial screen to identify product authenticity. If the product is deemed authentic based on a comparison of the collected spectra of the possible counterfeit product to library spectra acquired on the authentic product, then the investigation is complete. If the drug product is deemed a counterfeit based on this initial screen, then the material is subjected to low-resolution LC–UV-MS to assess the number of components present and assign molecular weights to all components observed in the LC–UV and MS total ion chromatograms (TIC). The initial screen by LC–UV-MS is performed on systems that are equipped and maintained for open-access usage. Use of the open-access systems improves the efficiency of the analysis process since these systems are preconfigured and available to run samples. If low-resolution MS analysis provides adequate ionization of the counterfeit components, then the sample is analyzed on a high-resolution instrument for accurate-mass measurement and assignment of molecular formulas. To achieve a narrow list of possible formulas, the elemental composition search is restricted with respect to included elements and the allowed deviation of calculated masses from the measured mass. Initially, the search is restricted to formulas that contain only C, H, N, and O. The allowed deviation from the measured mass is set to a maximum of 1.25 mmu. The isotopic distribution is also evaluated to check for the presence of elements other than C, H, N, and O that should be included in the element set.  

Figure 3: Negative-ion ESI mass spectrum of component 1 observed in product A.			

As an alternative, the sample can also be analyzed initially on an accurate-mass instrument, thereby bypassing the need for the low-resolution screen. If the low-resolution LC–MS analysis indicates a single component, or if the response in the mass spectrometer is found to be poor by electrospray ionization (ESI) then the sample is analyzed by NMR next. For samples that fail to ionize by ESI, and were observed to be relatively pure by NMR, desorption chemical ionization (DCI) is attempted to obtain a molecular weight and elemental composition for the unknown component. Quantitation of components is performed following identification of the components. For counterfeit products containing a single component, such as aspirin, LC–UV can be utilized to quantify the analyte by comparison with an authentic standard. Quantitative NMR (qNMR) is also used to determine the level of analytes in counterfeit products (17). For more-complex mixtures, a combination of techniques may be required to identify and quantify all of the individual components present in the counterfeit medicine. For example, techniques such as LC–UV-MS coupled with NMR spectroscopy (qualitative and quantitative), LC–UV-MS coupled with Raman or NIR spectroscopy, or these techniques coupled as needed with preparative LC isolation, X-ray diffractometry, or ICP-MS for metals analysis (18,19) can be used.  

Figure 4: Accurate mass, positive-ion ESI mass spectrum of component 2 observed in product A.			

			The strategy described above has been used to complete numerous counterfeit investigations at Bristol-Myers Squibb. In general, LC–MS has proven to be a powerful tool for the characterization of counterfeit products. The ability to rapidly assign a molecular weight and often a single formula for each counterfeit component affords the opportunity to expedite the putative identification using literature or internet searching. This process could be likened to electron ionization (EI) library searching in its ability to quickly filter potential candidates. Investigations using this approach are generally successful because counterfeiters often produce their products from commonly available, relatively simple materials, not novel products. A series of case studies are presented here that demonstrate the general strategy for conducting investigations of counterfeit materials.  

Figure 5: Positive- and negative-ion ESI total ion chromatograms of the product B extract.			

Initial LC–MS assessment is conducted using a system comprising a Shimadzu analytical HPLC system equipped with an SPD-10AV UV–vis detector and coupled to a Waters ZQ 2000 mass spectrometer. Chromatographic separation is performed using a 50 mm × 4.6 mm, 2.7-μm 

 Ascentis Express C18 column. The HPLC system is controlled using Shimadzu's Class VP software and the mass spectrometer is controlled by MassLynx v 4.0 SP4 software (Waters). These systems are configured for open-access use. The mass spectrometer is operated in both positive- and negative-ion modes (alternating). Data are acquired, in both modes, in the 

 75–1200 range. Typical flow rates are 1.0–1.5 mL/min. The flow is split to deliver approximately 300 μL/min of eluent to the mass spectrometer. Mobile-phase A is 95:5 0.01 M ammonium acetate in water–acetonitrile. Mobile-phase B is 5:95 0.01 M ammonium acetate in water–acetonitrile. A rapid gradient from 0% B to 100% B is used.   

Figure 6: Positive- and negative-ion mass spectra of the component at tR = 0.44 min from the 			

High-resolution mass spectrometry is conducted using a Waters Q-Tof Premier or a Thermo Fisher Scientific LTQ Orbitrap Discovery mass spectrometer. Both instruments are equipped with a Waters Acquity pump and photodiode-array detector. Chromatographic separations are run on a 100 mm × 2.1 mm, 1.7-μm 

 Waters BEH C18 column. The Q-Tof Premier system is controlled by MassLynx v 4.1 software (Waters) and is operated in V-mode with a resolution of 10,000 (full width at half maximum [FWHM]). The LTQ Orbitrap Discovery system is controlled by Xcalibur v 2.0.7 software (Thermo Fisher Scientific) and data are acquired at a resolution of 30,000 (FWHM). 

Figure 7: Accurate mass, positive-ion mass spectrum of the component at tR = 0.44 min from the 			

Positive-ion DCI with methane is used for the characterization of components that do not ionize adequately by LC–MS with ESI. The analysis is performed on a Waters GCT Premier system equipped with a DCI probe for sample introduction. The instrument is controlled using MassLynx v 4.1. The instrument is operated at a resolution of 7000 (FWHM) and the data is acquired over a range from 

-butylamine) is introduced as an internal reference for accurate-mass measurement.  

Figure 8: LCâUV chromatogram (220 nm) of product C tablet extract.			

Product A tablet was confirmed as counterfeit by Raman spectroscopy and was suspected to contain ascorbic acid based on comparison of the Raman spectrum to spectra in the library. A portion of the counterfeit tablet was extracted with 50:50 acetonitrile–water and sonicated for 20 min. The extracted liquid was filtered and diluted (1:15) for LC–UV-MS analysis. Figure 2 shows the LC–UV chromatogram obtained from the preliminary low-resolution LC–MS experiment. Component 1 was ionized using negative-ion ESI and was confirmed to be ascorbic acid based on the determined molecular weight of 176 Da as shown in Figure 3. Detection of ascorbic acid was consistent with the initial Raman analysis. The molecular weight of component 2 was assigned as 122 Da based on the low-resolution mass spectrum. To obtain the molecular formula of component 2, the sample was analyzed on a high-resolution instrument using positive-ion ESI, as shown in Figure 4. For component 2, evaluation of possible elemental compositions resulted in a single formula (C

O). Because only a single formula possibility was obtained, the formula was entered into an internet search engine to obtain a list of matching structures. The internet search on C

O returned nicotinamide as a possible structure. Next, 

H NMR data of the extract were collected to check for organic compounds that may not have been visible in the LC–UV-MS analysis and to further confirm the structure proposals for component 1 and component 2. Component 1 and component 2 were confirmed by 

H NMR to be ascorbic acid and nicotinamide. No additional components were detected by NMR. The 

H NMR data were also used to determine the mole ratio of ascorbic acid:nicotinamide as 4:1 (qNMR). An internet search confirmed the availability of commercial vitamin supplements with ascorbic acid and nicotinamide present in the same ratio.  

Figure 9: Positive-ion ESI mass spectrum of the tR = 3.51 min component in the product C extract.			

A tablet of suspected counterfeit product B was received following initial determination of counterfeit status by Raman spectroscopy. No additional structural information was obtained from the Raman analysis. A portion of the counterfeit tablet was extracted with 50:50 acetonitrile–water and sonicated for 20 min. The extracted liquid was filtered and diluted (1:10) for LC–UV-MS analysis. No significant components were detected in the LC–UV chromatogram; however, a poorly retained component was detected in both the positive- and negative-ion ESI total ion chromatograms as shown in Figure 5. The molecular weight of this component was determined to be 133 Da based on the positive- and negative-ion mass spectra (Figure 6). An accurate-mass analysis was conducted in positive-ion ESI mode and the data is shown in Figure 7. A single possibility for the molecular formula was obtained from the accurate-mass data (C

). An internet search conducted on this molecular formula suggested the identification was aspartic acid, which was confirmed by analyzing the extraction mixture by 

H NMR data additionally confirmed the absence of other organic components.  

Figure 10: Positive-ion ESIâMS-MS spectrum of the tR = 3.51 min component in product C (m/z 278.2).			

Product C was received and tested by LC–UV-MS after preliminary Raman screening confirmed its status as a counterfeit. No additional structural information could be gleaned from the Raman spectral analysis because the spectrum failed to yield a high-quality match with any of the compounds in the library. Similar to previous work, a portion of the counterfeit tablet was extracted with 50:50 acetonitrile–water and sonicated for 20 min. The extracted liquid was filtered and diluted (1:5) for LC–MS analysis. Figure 8 shows the LC–UV chromatogram at 220 nm that was obtained for the tablet extract, which indicates the presence of only one major component. The mass spectrum obtained for the major component observed in the HPLC analysis using positive-ion ESI is shown in Figure 9. The molecular weight of the compound was determined to be 277 Da. The elemental composition obtained from the accurate-mass measurement was C

N. This was the only possible formula returned by the software within the restricted accuracy window of 1.25 mmu. The formula was searched on the internet for possible structure matches and amitriptyline was the result. For additional confirmation of the identity of the counterfeit as amitriptyline, an accurate-mass product ion mass spectrum was collected and the product ion assignments are shown in Figure 10. All of the observed product ions are consistent with the identification of the active component of amitriptyline. As in the previous case studies, 

H NMR was used to confirm the structure of amitriptyline and ensure the absence of additional organic components. Amitriptyline is a tricyclic antidepressant that is used in the treatment of a variety of medical conditions including depressive and anxiety disorders, attention deficit disorder, hyperactivity disorder, migraines, eating disorders, insomnia, and bipolar disorder, among others (20). Though this product did contain an approved active pharmaceutical ingredient, amitriptyline, it did pose a significant risk to the patient because this compound is not an approved therapy or active against the target disease indicated for treatment with product C.  

Figure 11: Partial structure elucidated by NMR analysis for the active component in product D.			

Product D was initially tested by Raman and NIR spectroscopy, which indicated a counterfeit material composed of a possible mixture of lactose, hydroxypropyl cellulose (HPC), and starch. The compound did not ionize upon LC–UV-MS analysis with either positive- or negative-ion ESI. The sample was found to dissolve completely in H

O solution was used to obtain structural information using NMR. The NMR analysis indicated a single, pure component. Comprehensive one-dimensional (1D) and two-dimensional (2D) NMR analysis were successful in elucidating the majority of the unknown structure as shown in Figure 11. Mass spectrometry was needed to identify the para substituent labeled as "X" in Figure 11 because it was not represented in the NMR data set. DCI was used to obtain a mass spectrum of the compound because ESI had failed in the preliminary LC–UV-MS sample analysis. The mass spectrum obtained from the DCI analysis is shown in Figure 12. The molecular formula was elucidated as C

 based on the DCI accurate-mass analysis. Coupled with the structural information obtained from the NMR analysis, the para substituent was identified as a hydroxyl (-OH) group. An internet search using the molecular formula resulted in the further identification of this compound as arbutin (Figure 13). Arbutin is used as a skin-lightening agent, generally in topical applications, and is extracted from the bearberry plant (21). Its toxicology upon oral ingestion is unknown. This case study shows another example of a counterfeit product that contains an active component whose indication is misaligned with the intended disease area of the authentic product.  

Figure 12: Positive-ion DCI mass spectrum and elemental composition data for Product D analysis. 			

The general strategy that is used for testing counterfeit products has also been applied to the characterization of customer complaint samples. A complaint sample (product E) was received for characterization after preliminary testing by Raman and NIR spectroscopy indicated that the product contained within the bottle was not authentic. Visual examination of the packaging revealed the presence of pink residue on the threads of the dropper bottle, which indicates that at one point the bottle contained the authentic product (pink liquid), but was subsequently refilled with an unknown yellow liquid material. Two peaks, connected by a low horizontal plateau, were detected in the LC–UV chromatogram (see Figure 14). This type of peak shape is frequently indicative of an on-column interconversion or degradation (22). Molecular weights of 217 Da and 311 Da were assigned to these two observed components. The mass spectrum of the 311 Da component is shown in Figure 15. Accurate-mass data afforded molecular formula information. The compound was determined to contain sulfur, and a loss of H

 was observed from in-source fragmentation. NMR analysis performed in parallel indicated the presence of a monosubstituted phenyl ring, three methyl groups, and a methylene group. The MS data coupled with the information obtained from the NMR analysis was used to elucidate the structures of the two compounds as shown in Figure 16. Based on the structures, on-column degradation could easily be rationalized. The primary component (311 Da) was identified as metamizole, a powerful analgesic and antipyretic (23) that is available over the counter (OTC) as a yellow solution. The indication for product E does not align with that of metamizole.  

Product F was received for characterization after preliminary testing by Raman and NIR spectroscopy indicated that the product packed within the gel capsules was not the authentic drug substance. The investigation focused on characterization of the bulk yellow solid. The solid was observed to be insoluble in virtually all common solvents; however, it dissolved in hydrochloric acid to produce a yellow solution. Based on the solubility characteristics and the observation of effervescence upon dissolution it was suspected that the solid was primarily composed of an inorganic component. Reevaluation of the Raman data with a library containing inorganic compounds resulted in identification of the primary component as CaCO

 in the bulk solid was determined as 81.6 by titration. NMR analysis performed in DCl identified the presence of starch and acetaminophen in the solid. Quantitative NMR was used to assign a combined weight percent of 17.0 to starch and acetaminophen. Because none of the positively identified components could account for the color of the bulk solid, additional testing was conducted to understand the cause of the yellow color. A sample was analyzed by LC–UV-MS with UV detection at 430 nm and with only a single component observed in the UV chromatogram. The molecular weight and elemental composition of the component were determined by MS. An internet search on the molecular formula led to its identification as tartrazine (the United States Federal Food, Drug, and Cosmetic Act [FD&C] Yellow 5). Figures 17 and 18 show a comparison of the HPLC retention and UV spectra of the yellow component observed in product F with an authentic standard of tartrazine.  

Figure 14: LCâUV chromatogram of product E (260 nm).			

A series of case studies were presented to demonstrate a general strategy for conducting investigations of counterfeit pharmaceutical products. The case studies highlight the role of mass spectrometry in enabling the rapid characterization of suspected counterfeit products. The use of this process enables an efficient assessment of patient safety risks that are introduced by these products. The ability to readily assign molecular weights and often a single formula for each counterfeit component expedites the process of identification through literature and internet searching. The characterization is enhanced by coupling the information obtained from the LC–UV-MS experiments with NMR data. Analysis of the counterfeit samples by NMR provides full confidence in the structural proposals from the LC–UV-MS analysis and can be leveraged to enable the quantification of components by qNMR.   

Figure 15: Mass spectrum of the 311 Da component identified in product E.			

Figure 16: Structure of the primary component in product E and the resulting on-column degradation product.			

Figure 17: LCâUV chromatograms of product F and an authentic sample of tartrazine (430 nm).			

Figure 18: UV spectra of primary component in product F and an authentic sample of tartrazine (430 nm).			

Michael B. Peddicord, Charles Pathirana, Holly M. Shackman, Mark S. Bolgar, and Scott A. Miller 

are with Bristol-Myers Squibb, Analytical and Bioanalytical Development in New Brunswick, New Jersey. Please direct correspondence to: 

http://www.who.int/medicines/services/counterfeit/overview/en/

Articles

The production and sale of counterfeit drugs has risen sharply in recent years. The World Health Organization (WHO) estimates that counterfeit medicines account for approximately 1% of sales in developed countries and well over 10% in developing countries.

Application of Mass Spectrometry to Support Verification and Characterization of Counterfeit Pharmaceuticals

Here, a method to characterize edible oils and edible oil mixtures through fingerprinting and the isolation of individual analyte differences is reported. Aroma and flavor analytes in extra virgin olive, olive, peanut, grapeseed, and vegetable oils were sampled with headspace solid-phase microextraction (HS-SPME). The samples were analyzed with complementary gas chromatography (GC) systems; two-dimensional GC paired with time-of-flight mass spectro-metry (GC×GC–TOF-MS) and one-dimensional GC coupled to high resolution TOF-MS. These methods allowed for comparing edible oil varieties by their chromatographic fingerprints, characterizing the samples with the identification of individual analyte differences, differentiating oil mixtures from pure oil varieties with principal component analysis (PCA), and confirming analyte identities with accurate mass data. 

Characterization of food products, including distinguishing differences, is important in the food industry. Differences that are intentionally introduced are of particular interest. The global marketplace has seen an increase in cases of 

, which is loosely defined as the deliberate misrepresentation of a product for the purpose of monetary gain, with cost estimates of these types of fraud at $10–15 billion dollars per year (1). Olive oil adulteration ranks near the top of all reported food fraud cases with common adulterations including the substitution of olive-derived oils with other less expensive edible oils or the mislabeling of regular olive oils as extra virgin olive oil (2). 

These adulterations are often difficult for a consumer to detect, but can decrease health benefits and have the potential for unintended exposure to allergens — for example, by adulteration with nut oils. Detecting food fraud experimentally is also challenging because of the inherent variations of natural products and the wide range of potential methods of adulteration. Targeted analytical tools to screen for specific analytes or known adulterants have the risk of missing new or unanticipated mechanisms of food fraud. Nontargeted analytical methods, especially those that isolate individual analytes within a food matrix and characterize complex food products, are well-suited for addressing these challenges. 

Gas chromatography (GC) is a well-established analytical approach used in the food industry for isolating individual analytes within a complex matrix. For many samples, one-dimensional GC is sufficient to isolate the components of the mixture, but as sample complexity increases so does the occurrence of coelution. One approach to handle increased complexity and analyte coelution is the use of comprehensive two-dimensional GC (GC×GC), which couples a complementary second separation dimension to the primary column and leads to an increase in the peak capacity, theoretically as high as the multiplicative of the individual peak capacities of each column. This technique has been shown to be a powerful method for analyzing food products (3).  

Coupling GC separations with mass spectrometry (MS) provides a means for identifying the isolated analytes present in the food sample. When unanticipated analytes are present, identification can provide insight to the mode of adulteration. Time-of-flight (TOF) MS is especially beneficial for nontargeted analyses as it is capable of collecting full mass range data at fast acquisition rates and target analytes do not need to be specified before acquisition. Full mass range spectra can be library searched for identification and accurate mass information from high-resolution MS systems provides added confidence to identifications with formula computation capabilities. In this study, both one-dimensional (1D) and two-dimensional (2D) GC and both nominal and accurate mass MS systems were utilized, providing complementary and confirmatory information to characterize edible oils. These analytical tools have the potential to identify food fraud through differentiation of oil varieties and oil mixtures.

Sample Preparation and Solid-Phase Microextraction Parameters

Three extra virgin olive oil varieties as well as one of each olive, vegetable, peanut, and grapeseed oils were analyzed. A 2-g sample was pipetted into a 10-mL headspace vial and sealed. The headspace solid-phase microextraction (HS-SPME) sampling was performed by incubating the samples at 40 °C for 5 min immediately before 25 min of extraction at the same temperature. A divinylbenzene–carboxen–polydimethylsiloxane (50/30 µm DVB/CAR/PDMS, Supelco) fiber was used, and the fiber was exposed in the GC inlet at 250 °C for 2 min to desorb analytes for injection. Three replicates of each oil variety were analyzed. 

GC×GC analyses were performed with LECO's Pegasus 4D system, consisting of an Agilent 7890 GC system, modified with LECO's dual-stage quad jet thermal modulator and secondary oven and paired with LECO's Pegasus TOF-MS system. GC–high resolution (HR)-TOF-MS analyses were performed with LECO's Pegasus HRT system consisting of an Agilent 7890 GC system paired with LECO's high-resolution TOF-MS system. Method parameters are compiled in Table I. 

Table I: Instrument method parameters 			

Peak finding and library searching were performed with automated peak finding tools in LECO's ChromaTOF software. Principal component analyses (PCA) were implemented with the XLSTAT software package through Excel. PCA on the chromatographic fingerprint was performed on raw total ion current (TIC) GC×GC traces that were exported from ChromaTOF and compiled in Excel. The data were prepared for feature selection within the ChromaTOF software by compiling peak information across all samples. Various feature selection tools were utilized including ChromaTOF's reference feature, review of fingerprinting, and Student's 

-tests comparisons between edible oil varieties. The aligned analytes were exported, standardized for pre-processing, and submitted to PCA in XLSTAT.  

The analysis of adulterated olive oils was explored with complementary GC–MS instrument platforms. A nominal mass MS system was coupled with GC×GC and a high-resolution MS system was paired with GC. These analytical tools provided fingerprint characterization as well as the ability to isolate and identify individual analytes for further insight to the oil varieties.  

GC–TOF-MS Fingerprinting  Characterization of Edible Oils

Because food fraud can involve multiple edible oil varieties, a collection of representative edible oil samples were characterized and compared to investigate adulteration. Three extra virgin olive oil samples and one of each olive, peanut, grapeseed, and vegetable oils were prepared and analyzed with these methods. Example chromatograms for each sample are shown in Figure 1 and many similarities and differences were noted. The visually distinct two-dimensional chromatograms that are generated with GC×GC allowed for rapid preliminary comparisons of the samples. 

Figure 1: Representative TIC chromatograms for three extra virgin olive oil varieties and olive, peanut, grapeseed, and vegetable oils, analyzed by GCÃGCâTOF-MS. 			

The TIC chemical fingerprint traces were exported and submitted to PCA to investigate the ability to classify and group the oil varieties by type with the visual chromatogram. The corresponding scores plot is shown in Figure 2a with each sample represented as a circle. Each edible oil variety is clearly and distinctly grouped in the first three principal components, which capture a combined 86.1% of the variance. To simulate adulteration, 50% mixtures of each oil variety mixed into extra virgin olive oil were prepared and submitted to PCA along with the pure oil varieties with the scores plot shown in Figure 2b. The mixture samples, indicated with blue outer circles, have scores that fall between the pure oil variety and that of extra virgin olive oil, indicating their difference from either pure group and suggesting the potential to detect food fraud by GC×GC fingerprinting. 

Figure 2: The TIC traces were submitted to PCA. (a) The score plots show the edible oil varieties clearly distinguished and (b) the distinction of simulated fraud mixtures (with blue outer circles) as different from the pure oils. 			

Complementary Isolation of Individual  Analytes with GC

In addition to fingerprinting to compare the samples overall, specific analytes that distinguished the oil types were also determined to gain further insight. These complementary analytical tools increase resolving power, chromatographically and mass spectrally, to allow for more comprehensive sample characterization and better isolation of individual analyte components. ChromaTOF's automated peak finding tools were used to isolate individual analyte peaks within the data and compile peak areas and identification information. A peak from the extra virgin olive oil sample in the GC–HR-TOF-MS data is shown in Figure 3a. The TIC trace appeared to be a single analyte and the associated mass spectrum matched to octane with a similarity of 786. Upon comparison to the library spectra, however, several extra 

 values were noted — for example, 56, 67, 72, and 82. This suggested that another analyte was coeluted with octane and that the spectrum in Figure 3b was a combination of these analytes. Deconvolution can mathematically resolve coeluted analytes, but absolute coelution, as is suspected here, is typically beyond the capabilities of deconvolution, thus requiring chromatographic adjustments. 

Figure 3: (a) Octane chromatogram and (b) spectrum isolated from the GCâHR-TOF-MS data. The extra m/z values (56, 67, 72, and 82) in the Peak True spectrum suggested that a coeluted analyte was contributing to the spectrum. 			

One of the primary benefits of GC×GC is the ability to chromatographically isolate analytes in the second dimension that coelute in the first dimension. Figure 4a shows the corresponding GC×GC separation, where two analytes were clearly resolved in the second dimension. The chromatography data software's automated peak finding feature isolated mass spectra for each analyte, and the associated library matches are shown in Figures 4b and 4c. The similarity for octane improved to 947, and hexanal (with masses 56, 67, 72, and 82) was chromatographically resolved with a library similarity of 906. Hexanal and octane have retention indices of 806 and 816, respectively, so their similar retention on the primary nonpolar column was expected. Their structural differences, aldehyde versus alkane, resulted in different retention on the secondary polar stationary phase column. With this GC×GC column combination, the analytes were distinctly resolved in the second dimension in spite of their first dimension coelution. Hexanal is known to be one of the major volatile components of olive oil while octane has gasoline odor properties, so both are important for characterization of the olive oil sample. 

Figure 4: (a) Octane and hexanal are coeluted in the first dimension and the increased peak capacity of GCÃGC allowed for isolating these analytes in the second dimension. The isolated mass spectra were library matched with similarities of 906 and 947 for (b) hexanal and (c) octane, respectively. 			

Closer review of the GC–HR-TOF-MS spectrum in Figure 3 adds confidence to these identifications with accurate mass information. In Figure 5, overlaid extracted-ion chromatograms (XICs) of traces associated with each individual analyte, 85.101 and 82.077, apex together in agreement with the suspected absolute coelution. Based on library reference spectra and formula calculations of fragments, the 

 values were assigned to either hexanal or octane. Mass fragments associated with each analyte are compiled in Figure 5. The good mass accuracies of the fragments provided confidence to the identifications determined through nominal mass library searches of the chromatographically resolved analytes in the GC×GC data. Of particular note, hexanal and octane have a common nominal mass fragment of 

 85 that is indistinguishable with low-resolution MS. Octane has a C

O. Although the nominal masses are identical, the accurate masses differ by 0.0364 amu, 85.1012 and 85.0648. These individual fragments were both observed with good mass accuracies, also reported in Figure 5. 

Figure 5: XICs for masses specific to each analyte apex together indicating absolute coelution. The mass spectrum associated with the GC peak shown in Figure 3 is the combination of the octane and hexanal spectra. With accurate mass information, individual fragments were assigned to each analyte with good mass accuracies observed. Octane and hexanal have a shared nominal mass of 85 with accurate masses differing by 0.0364 amu that were distinguished with the high-resolution MS system.			

The spectrum shown in Figure 3 was separated with fragments assigned to two analytes through chemical reasoning, but the absolute coelution prevented mathematical deconvolution from isolating each individual spectrum. For other coeluted analytes with a larger degree of chromatographic separation, mathematical deconvolution provided an additional level of resolving power that isolated even more individual analytes. Even with the increased peak capacity of GC×GC, coelutions still occurred and the software's deconvolution feature mathematically resolved some of these overlapped analytes. Figure 6 shows that what appeared to be a single peak in the GC×GC contour plot was two analytes separated by 60 ms at their peak apexes in the second dimension. The software identified unique masses to define chromatographic shape, shown in Figure 6b, and isolated mass spectra for determining analyte identity (Figures 6c and 6d). These two analytes were identified as 1-penten-3-one and 1-penten-3-ol with similarity matches to the library spectra of 824 and 861, respectively. 

Figure 6: 1-Penten-3-one and 1-penten-3-ol appear as a single peak in the GCÃGC data. With deconvolution two distinct peak shapes are apparent and the isolated mass spectra were library matched with similarities of 824 and 861, respectively. 			

Feature Selection to Distinguish Edible Oil Varieties 

With these complementary analytical tools and powerful peak finding algorithms, sample characterization and differentiation were investigated and hundreds of analytes were measured and identified in each sample. Many similarities and differences were readily apparent in the fingerprinting comparison, shown in Figures 1 and 2. The software's reference feature was applied as a feature selection tool to further probe specific analyte differences in an automated way. Distinguishing analytes were located within the data by computing the relative concentration of analytes in a sample with respect to a specified reference sample. The reference feature tags each peak as a match, out of tolerance, not found, or unknown to indicate the similarity to the reference. Match and out of tolerance tags indicate that the analyte was present in the sample within or outside of the specified concentration tolerance relative to the reference, respectively. Not found and unknown tags indicate analytes that were only present in the reference or the sample, respectively. The extra virgin olive oil sample was set as the reference and each oil variety was the sample, such that peaks in the sample with not found or out of tolerance tags (with small concentration values) were indicative of extra virgin olive oil. Conversely, unknown and out of tolerance tags (with large concentration values) were indicative of the other oil varieties. The results were filtered by their type tag to locate analyte differences between the samples. 

Figure 7: Î±-Cubebene, an analyte indicative of extra virgin olive oil, is shown. XIC 119 in the GCÃGC chromatograms and XIC 119.085 in the chromatograms for the GCâHR-TOF-MS data show the analyte peak present in extra virgin olive oil (orange) and absent (or at drastically lower levels) in the light olive oil (green). The observed and library spectra are shown, as well as accurate mass information for the molecular and fragment ions. 			

Many analyte differences were present and a selection of representative analytes that differed between extra virgin olive oil and olive oil are demonstrated here. α-Cubebene is a sesquiterpene with herbal odor properties that was observed to be elevated in the extra virgin olive oil relative to the light olive oil, as shown in Figure 7. Phenylethyl alcohol, a compound with known floral odor properties, was observed to be present in both the extra virgin and light olive oils at different levels (Figure 8). And heptane, which has sweet and ethereal odor properties, was elevated in the light olive oil relative to the extra virgin olive oil (Figure 9). In each case, the accurate mass information for the molecular and fragment ions added confidence in the analyte identifications to the library search results, with mass accuracies less than 1 ppm. 

Figure 8: Phenylethyl alcohol, an "out of tolerance" analyte indicative of extra virgin olive oil, is shown. XIC 91 in the GCÃGC chromatograms and XIC 91.054 in the chromatograms for the GCâHR-TOF-MS data show the analyte peak is higher in extra virgin olive oil (orange) than in the light olive oil (green). The observed and library spectra are shown, as well as accurate mass information for the molecular and fragment ions. 			

The reference feature was used to find differences between extra virgin olive oil and each other oil variety and a collection of additional feature selection tools were also implemented to compile a representative list of specific analyte differences. In addition to the reference feature, paired 

-tests between each oil variety and the fingerprint comparisons highlighted analytes with clear distinctions between the oil varieties. A collection of these analytes is shown in Figure 10a. The analyte peak areas (mean centered per analyte) are shown with some analytes consistently elevated in each oil variety relative to the others. This subset of analytes was submitted to PCA, and the corresponding scores plot is shown in Figure 10b. By selecting a subset of analytes that were specifically intended to distinguish the samples, PCA resulted in tighter clusters and more distinct grouping of each edible oil variety. The 50% mixture sample scores are also clearly distinguished as different from either pure group. These analytes appear to distinguish the oil varieties and therefore have potential to be used as food fraud markers. 

Figure 9: Heptane, an "out of tolerance" analyte indicative of light olive oil, is shown. XIC 100 in the GCÃGC chromatograms and XIC 100.125 in the GCâHR-TOF-MS chromatograms show the analyte peak is higher in the light olive oil (green) than in the extra virgin olive oil (orange). The observed and library spectra are shown, as well as accurate mass information for the molecular and fragment ions.			

Complementary GC–TOF-MS systems were used to characterize edible oils and clearly distinguish representative varieties. Aroma and flavor analytes in extra virgin olive, olive, peanut, grapeseed, and vegetable oils were sampled with HS-SPME and analyzed with GC×GC–TOF-MS and GC–HR-TOF-MS. Distinct visual two-dimensional chromatograms were generated for each edible oil variety sample, and clear similarities and differences were observed. Fingerprinting with PCA resulted in clustered grouping of each oil variety. Individual analytes were isolated and identified through chromatographic separation, mass spectral resolution, and mathematical deconvolution, and those that distinguished the oil varieties were determined through a variety of feature selection tools. These sample-distinguishing analytes were submitted to PCA to also classify and group the samples by edible oil type. These nontargeted analytical methods allowed for comparing edible oil varieties by their chromatographic fingerprints, characterizing the samples with the identification of individual analyte differences, differentiating simulated fraudulent oil mixtures from pure oil varieties with PCA, and confirming analyte identities with accurate mass data where mass accuracies of less than 1 ppm were routinely achieved. These data suggest that this method has the potential to detect food fraud.  

Figure 10: (a) Analytes determined to differ between oil varieties are shown. Peak areas are mean centered for each analyte across the samples. These analytes distinctly differ between the oil varieties and help distinguish them from each other. (b) PCA scores plot with analytes selected via feature selection for pure varieties and simulated fraud mixtures, denoted with blue outer circles, added as samples. The simulated fraud mixtures have scores between those of the pure variety and extra virgin olive oil. 			

is an appplications chemist with LECO Corporation in St. Joseph, Michigan. Direct correspondence to: 

(1) Grocery Manufacturers Association (GMA), "Consumer Product Fraud: Deterrence and Detection"(Grocery Manufacturers Association, Washington, D.C., 2010). Available at: 

http://www.gmaonline.org/downloads/wygwam/consumerproductfraud.pdf

(3) P.A. Tranchida, P. Donato, F. Cacciola, M. Beccaria, P. Dugo, and L. Mondello, 

Here, a method to characterize edible oils and edible oil mixtures through fingerprinting and the isolation of individual analyte differences is reported. Aroma and flavor analytes in extra virgin olive, olive, peanut, grapeseed, and vegetable oils were sampled with headspace solid-phase microextraction (HS-SPME).

Comparison of Edible Oils by GCxGC–TOF-MS and GC–High Resolution TOF-MS for Determination of Food Fraud: A "Foodomics" Approach

Despite the advantages of soft ionization ion-source technologies for improving confidence in the identification of a range of challenging analytes, soft ionization remains a niche technique for gas chromatography–mass spectrometry (GC–MS). This article discusses the reasons for this limitation as well as new developments that make lower-energy electron ionization possible without the disadvantages that have historically been associated with it. The specific benefits of this technique are discussed by reference to several examples across the GC–MS field.

The term "soft ionization" encapsulates a range of techniques that ultimately result in analyte molecules that become ionized without imparting excess energy to them. The result is a limited degree of analyte fragmentation, meaning that the molecular ion passes intact through the mass spectrometer and to the detector. This ability to provide information about the unfragmented molecule makes soft ionization of great value to analysts.  

For macromolecules, the ready fragmentation of the molecular ion means that soft ionization is the only suitable technique and, in this field, matrix-assisted laser desorption–ionization (MALDI) is a popular method for ionizing DNA, proteins, peptides, sugars, polymers, and dendrimers. 

However, for smaller analytes amenable to gas chromatography (GC), high-energy electron ionization (EI), typically at 70 eV, is by far the most popular technique. Large libraries of spectra such as those curated by the National Institute of Standards and Technology (NIST) and Wiley are available for matching and identification, and have long been used by analysts across a wide range of gas chromatography–mass spectrometry (GC–MS) applications. In this field, methods for soft ionization have historically been viewed as more specialized, and more often used in cases where EI at 70 eV does not provide adequate results. 

There are multiple reasons for this. The major factor is that EI, which although by its nature is applicable to almost every vaporized substance, cannot be used successfully at lower energies. This is because of inefficient channeling of electrons from the filament ("e-gun") into the ion chamber, giving an extremely low signal and an unacceptable loss of sensitivity. 

Therefore, to achieve soft ionization, analysts have turned to other techniques, of which chemical ionization (CI) is the most common. CI also results in a drop in sensitivity, but to a far lesser extent than EI. Unfortunately, however, CI requires a different ion-source configuration, with additional source pressurization and the use of reagent gases. If a single instrument is being used, this can be a time-consuming transition and a considerable drain on laboratory resources if required on a regular basis. 

The challenge, therefore, remained of developing a convenient soft ionization technique that retained the performance and wide applicability of standard 70-eV electron ionization. Hence, we were excited by the development of an e-gun design that removes the link between the energy of the electrons and the ionization efficiency, by introducing an additional electrostatic element between the e-gun and the ion chamber. This allows the ionization energy of the electrons to be varied on a sliding scale from conventional 70 eV to lower energies, without loss of sensitivity. 

The Advantages of Variable-Energy Electron Ionization

A key feature of this variable-energy electron ionization is the avoidance of sensitivity losses; absolute ion intensities at low energies have been found to be equal to or greater than at 70 eV. On its own, this would not be particularly useful, but in conjunction with the improved signal-to-noise ratios at low energies (see below), the result is lower detection limits for target compounds. Furthermore, greater sensitivity allows analysts to use the same sample loading and analytical conditions for both hard and soft ionization, meaning less time is spent on sample preparation and method development.  

Like other soft ionization techniques, using EI at low energies inherently results in reduced fragmentation, and (as a consequence) an enhanced molecular ion. Importantly, however, we have found that a degree of fragmentation is usually retained even at the lowest energies. The ions produced are typically those that arise from the lower-energy fragmentation pathways, that is, larger ions that are more significant in terms of understanding the structure of the molecule. These larger ions are more often those required for quantitation, and so their increased intensities also benefit calibration.  

In addition to reduced fragmentation of the analyte, fragmentation of chromatographic background and carrier gases is also reduced. This greatly improves the signal-to-noise ratios for analytes, reduces demand on the dynamic range, and improves limits of detection, particularly in complex or more "dirty" samples. 

Another important point to note is that switching from regular 70-eV ionization to lower energies is inherent to the design, and can be carried out by changing a method parameter in the MS software. As well as avoiding the inconvenience of changing from EI to an alternative soft ionization technique, this allows sequences of samples to be repeated automatically at different energies, with subsequent savings of time. 

The inherent variability of the technique is also helpful, by allowing the analyst to adjust the ionization energy to suit the situation, and discover optimal parameters during method development. Relatively small differences in ionization energies at the low end of the scale can have significant differences. For example, note the difference in relative intensity of the 

 105 and 182 signals at 16 eV and 14 eV for benzophenone (Figure 1). 

Figure 1: Comparison of mass spectra for benzophenone at 70, 16, and 14 eV.			

The variable-energy ionization technology described above was developed by modifying the ion source for a time-of-flight (TOF) mass spectrometer (BenchTOF, Markes International), which is able to produce reliable 70 eV spectra that are a close match to those in standard libraries. The following three examples illustrate the use of the new system.  

Even when using a TOF-MS that produces spectra free from mass discrimination, uncertainties can arise when compounds have weak molecular ions or extensive fragmentation at regular 70-eV ionization energy. As illustrated in Figure 2 for the common pesticide precursor hexachlorocyclopentadiene, reducing the ionization energy can greatly enhance the molecular ion, thereby providing an additional degree of confidence in identification.  

Figure 2: Comparison of 70-eV and 11-eV EI spectra for hexachloro-cyclopentadiene, showing the greatly enhanced molecular ion at the lower energy.			

Example 2: Differentiating Between Isomeric Compounds

The ability to provide enhanced molecular ions while retaining structurally significant fragment ions improves discrimination between compounds that have similar spectra at 70 eV. Figure 3 shows an example of this increased orthogonality for four hydrocarbons in a crude oil. 

Figure 3: Top: GCÃGCâTOF-MS contour plot of a crude oil, showing overlaid EICs (m/z 268 + 282 + 296 + 310 + 324 + 338 + 352), analyzed with an ionization energy of 14 eV. The panels below (aâd) show the 70-eV and 14-eV EI spectra for  each of the four labeled peaks AâD, respectively, illustrating that it is much easier to discriminate between these compounds using the low-energy spectra.			

Example 3: Enhancing Signal-to-Noise Ratios

At low ionization energies, the reduction in the degree of fragmentation of both analytes and chromatographic artifacts helps improve signal-to-noise ratios, with subsequent lowering of the limits of detection. This is illustrated in Figure 4 for the odorous compounds safrole and coumarin, in which the increase in molecular ion intensity and reduction in fragmentation on moving from 70 eV to 15 eV result in an approximately twofold increase in the signal-to-noise values for the molecular ions.  

Figure 4: Comparison of 70-eV and 15-eV EI spectra for the allergens safrole and coumarin, showing the enhanced signal-to-noise ratios at the lower energy.			

The variable-energy ionization approach described in this article is a rugged technique that has the potential to deliver significant benefits to GC–MS laboratories across a wide range of applications. 

The ability to easily switch between hard and soft EI allows both regular 70-eV spectra to be generated for matching against existing libraries, alongside lower-energy spectra that provide complementary information on the molecular ion and structurally significant fragment ions. This complementary information enables more confident identification and discrimination between isomers that previously would not have been resolved. 

Furthermore, the reduced fragmentation of analytes, matrix interferences, and carrier gases significantly improve signal-to-noise ratios for target substances, and allow lower limits of detection to be achieved.  

For more information about variable-energy electron ionization, and detailed experimental information on the examples covered in this article, please contact the authors. 

Laura McGregor, Nick Bukowski, and David Barden

 are with Markes International, at the Gwaun Elai Medi-Science Campus in Llantrisant, UK. Direct correspondence to: 

Despite the advantages of soft ionization ion-source technologies for improving confidence in the identification of a range of challenging analytes, soft ionization remains a niche technique for gas chromatography–mass spectrometry (GC–MS).

A New Outlook on Soft Ionization for GC–MS

Translational proteomics has not been as successful as originally anticipated. Because mass spectrometry (MS) can separate proteins at the sequence level, it provides the selectivity needed for this application; however, traditional challenges still exist, including time-to-result, throughput, and sample-size requirements. For analytical validation and verification purposes, sample preparation times must be reduced from days to hours. Scientists recently coupled a previously developed immunoaffinity enrichment method to selected reaction monitoring (SRM) MS. The MS immunoassay–SRM method combines liquid chromatography–tandem mass spectrometry (LC–MS-MS) for target identification, a microscale immunoaffinity capture method for enrichment, and subsequent SRM analysis. Using the MS immunoassay–SRM workflow, a standard high-throughput method for developing targeted biomarker identification of proteins in human plasma and serum for clinical research was developed. 

Mass spectrometry (MS)-driven proteomics has made progress in the identification and quantification of disease biomarkers, including C-reactive protein as an indicator for myocardial infarction and prostate-specific antigen (PSA) for prostate cancer. Despite these and other successes, 

, which is defined as the translation of biomarker discovery to routine analysis, has not been nearly as successful as originally anticipated because comprehensive proteomic analysis of plasma, serum, and other biological fluids has proved exceedingly challenging. The "look alike" nature of molecular isoforms, the enormous dynamic range of protein concentrations of potential interest (>10 orders of magnitude in blood plasma), and the fact that molecules of interest are often in low abundance have all slowed the progress of biomarker hunters across the globe.  

One example of the challenge presented by protein analyte isoforms is demonstrated by the need to distinguish between full-length parathyroid hormone (PTH) 1-84 and multiple N-terminally truncated PTH variants. The differences between these isoforms are critical to accurate diagnosis of endocrine and osteological diseases (1). Similarly, in clinical testing, PSA typically presents in truncated and modified isoforms, making precise detection and quantification difficult which contributes to a high false-positive rate (2).  

				From a processing point of view, binding protein detachment and the high dynamic range of samples have hampered assays for insulin-like growth factor 1 (IGF-1), a marker for growth-related illness that is important in cell proliferation, differentiation, apoptosis, and tissue growth from a research point of view (3). 

To add to these challenges, verification and population-scale biomarker validation require the analysis of hundreds or even thousands of high-quality samples. Sample collection and storage must use standard protocols to reduce potential variations attributable to endogenous enzymes or sample contamination. Verification and validation studies require multiple control groups and subjects in disease subcategories — all gathered over the course of disease progression. The analysis of many samples is required to distinguish normal human genetic heterogeneity and heterogeneity attributable to disease. High-throughput detection methods are essential to achieving statistical confidence in the accurate identification of molecules of interest (4). 

The current gold standard for biomarker detection uses enzymes to bind antibody ligands to target molecules. The enzyme-linked immunosorbent assay (ELISA) has played a central role in diagnostic and clinical research applications for the last two decades. However, as molecular biology advances and as scientists are able to access new technologies with higher recognition accuracy, ELISA is emerging as a potential source of inaccuracy leading to imprecise biomarker detection and resulting in incorrect diagnosis and treatment. In certain cases, demonstrated poor concordance among assays from different manufacturers has raised questions about ELISA's accuracy (5). Poor concordance stems from the variety of proprietary antibodies used to detect different epitopes, as well as natural biological diversity manifested in post-translational modifications, single nucleotide polymorphisms, and cross-reactivities between antibodies to off-target proteins. Various interferences have also been observed, including antireagent antibodies and endogenous auto-antibodies, which can generate incorrect results. These indirect, nonspecific signals mask consequential variations in molecular structures (5). An October 2013 study investigated the specificity of the commercial CUZD1 ELISA assay (6). CUZD1 is a biomarker for ovarian and pancreatic cancer, as well as inflammatory bowel disease (7). Using a combination of western blot, high performance liquid chromatography (HPLC), and MS, the study confirmed that instead of CUZD1, the commercial assay recognizes a nonhomologous cancer antigen, CA125. The study concluded that the poor characterization of the commercial ELISA assays may lead to false biomarker discovery (6). 

MS has been used for quantification of small molecules for years and, more recently, has been applied to protein quantification and analysis. Because MS can separate proteins at the sequence level, it provides the selectivity needed to address the specificity issues outlined above (4). As noted, for validation and verification purposes, high throughput and high accuracy are vital; unfortunately, the traditional challenges surrounding MS-based techniques include time-to-result, throughput, and sample-size requirements. For MS to be used for validation and verification purposes in translational proteomics, sample preparation times must be reduced from days to hours (4). Despite their shortcomings, both ELISA and MS methods are widely used in routine clinical research and analysis. In recent years, researchers have used immunoaffinity enrichment coupled to MS to combine the selectivity of immunoaffinity extraction with the specificity of MS detection (8,9). 

One of the hurdles that scientists have worked to overcome to improve MS accuracy involves sample preparation before MS. The wide dynamic range of proteins in biological fluid samples compromises the ability of MS to achieve sufficient sensitivity to accurately quantify potential biomarkers. Scientists have explored numerous sample preparation approaches to eliminate proteins of little interest that increase dynamic range. These approaches include fractionation using multiple liquid chromatography (LC) columns, depletion of abundant proteins, enrichment using solid-phase extraction, nanoparticles, and immunoaffinity enrichment by various techniques that include magnetic beads and microcolumns (10).  

Scientists at Thermo Fisher Scientific, in partnership with Arizona State University, the Institut de Recherche Clinique de Montréal, the Sahlgrenska Academy at the University of Gothenberg, the University Health Network at the University of Toronto, and Kings College Hospital in London, have coupled a previously developed immunoaffinity enrichment method (MSIA, Thermo Fisher Scientific) (8) to selected reaction monitoring (SRM) MS (11). The MS immunoassay–SRM approach combines LC–MS-MS for target identification, a microscale immunoaffinity capture method for enrichment, and subsequent SRM analysis.  

To identify optimal peptides and transitions for SRM, the system combines algorithmic and spectral library prediction using Pinpoint software with results from high-resolution LC–MS-MS analysis resulting in a list of targets. After the first strategy is built and recombinant protein or peptide standards are obtained, scientists can narrow the list by iteratively optimizing collision energies, LC gradient, and SRM scheduling windows. Only the highest intensity transitions are retained and, after three or four repetitions, the system is ready for sample analysis. Figure 1 shows the workflow for the optimization of SRM analyses. 

Figure 1: Automated workflow for optimization of SRM analyses. 			

In the immunoaffinity capture step, antibodies are surface-immobilized in small, porous microcolumns fixed into pipette tips. Samples are repeatedly aspirated and dispensed through these affinity pipettes to expose the immobilized antibody to the targeted proteins (10). This iterative step has the equivalent effect of lengthening an LC column. After the proteins are captured, they are eluted for subsequent MS detection. In a single iterative separation step, target proteins and variants are captured using antibodies aimed at epitopes in the protein sequence. The approach is applicable to low-abundance proteins because of the robust immunoaffinity enrichment of the target analyte. Figure 2 summarizes the immunoaffinity enrichment workflow. 

Figure 2: Immunoaffinity enrichment workflow. 			

Most targeted protein detection methods add surrogate peptides and corresponding isotope-labeled counterparts to samples for standardization after digestion, just before the MS step. The MS immunoassay–SRM design, however, adds internal reference standards at the outset, before enrichment. This means the standard goes through all the same steps that the sample undergoes, acting as a normalizer for all processing and data collection steps. In this type of MS immunoassay–SRM experimental workflow, recombinant proteins rather than isotope-labeled peptides are used.  Recombinant proteins, if available, are less expensive and can be easier to obtain. After digestion, heavy isotope-labeled peptides are added to ensure quality control of the LC and MS detection steps in each sample to measure run-to-run variance.  

Using the MS immunoassay–SRM workflow, a standard high-throughput workflow for developing targeted biomarker identification of proteins in human plasma and serum was developed (11). The standardized protein identification workflow included targeted methods for 16 different protein analytes, including members of the apolipoprotein family: ApoE, ApoA1, ApoCI, ApoCIII, and ApoJ; some medium-high abundance proteins: ceruloplasmin, vitamin D binding protein, beta-2 microglobulin, and C-reactive protein; and several important low-abundance proteins, including procalcitonin, PTH, IGF-1, PSA, erythropoietin, proprotein convertase subtilisin/kexin type 9, and amyloid beta.  

Results for PTH and IGF-1 MS immunoassay–SRM were compared to clinical analyzer immunoassays. Correlation between the two methods yielded 

 values ranging from 0.67 to 0.87, depending on the surrogate target peptide. For PTH, the correlation of the analyzer data with MS immunoassay–SRM of the single, N-terminal peptide versus the average of all peptides suggests that the analyzer-based immunoassay likely includes uncharacterized, intact, and truncated protein species, leading to an overestimation of functional protein. In comparison, SRM characterizes all isoforms; therefore, all species of interest are individually quantified, resulting in improved accuracy. In addition, the method allowed discovery of previously undescribed isoform variants of PTH.  

Table I: Assay format comparison: MS immunoassay tip versus beads 			

Subsequently, Thermo Scientific scientist Eric Niederkofler presented data in a webinar (12) that compared IGF-1 results obtained using the MS immunoassay–SRM approach with results obtained using a magnetic bead immunoenrichment technique. This work demonstrated that the MS immunoassay technique was five times more effective than the bead method, lowering the limit of quantification by 10-fold (see Table I) and producing improved signal-to-noise ratio at the MS stage (Figure 3).  

Figure 3: High-resolution orbital trap MS-MS data of IGF-1 peptide obtained from MS immunoassay tips (top) versus magnetic beads (bottom). 			

In our published results (11), we presented linear regression curves that showed 

values from 0.089 to 0.098. In Figure 4 we present similar data for two apolipoprotein E peptides. 

Figure 4: Calibration curves for ApoE targeted peptides.  			

Figure 5 presents SRM chromatograms for ApoE surrogate peptides from digested human plasma. Peak shapes are symmetrical and relatively free of significant noise. SRM transition ion ratios were within ±15% of internal standard reference ratios. To ensure selectivity, we used several SRM transitions for each peptide. Detection limits and dynamic ranges for all targeted peptides were well within the useful range for the selected analytes, and lower limit of quantification values were all within the lower portion of existing ranges.  

Figure 5: Representative SRM chromatogram results. 			

While developing this approach to immunoaffinity enrichment over the last five years, we found that, in many cases and for several reasons, antiprotein antibodies are preferable binding agents to antipeptide antibodies. Antiprotein antibodies typically enrich multiple protein isoforms, making it easier for SRM analyses to distinguish active from inactive forms. Because antipeptide antibodies must be highly specific, each target requires a separate antibody. For example, for PTH protein detection using MS immunoassay–SRM and antipeptide antibodies, investigators would need to generate seven different antipeptide antibodies to reveal the same information available from one antiprotein antibody. Using an antiprotein antibody raised toward an epitope common to all isoforms or variants costs less, because only one protein is required to bind multiple forms. Using antipeptide antibodies also makes it more difficult to add new targets to a test, because multiple new antipeptide antibodies must be developed for each analyte.  

Antiprotein antibody capture provides distinct advantages in the trypsin digestion stage as well. Because digestion depends on the protein-to-trypsin ratio, complex samples such as raw plasma or serum require relatively large amounts of trypsin and long digestion times. Samples enriched at the protein level are digested after enrichment. Because they are less complex than raw samples, this results in reduced digestion times, typically 3 h or less.  

In comparison, samples enriched at the peptide level require crude digestion before enrichment and, as a result, take more time to process. Crude digestion also injects a potential source of error stemming from uncontrolled or unknown causes. By their nature, samples from different states of disease are expected to contain varying levels of proteins. Digestion variability because of illness may impact results and has not been adequately studied. Finally, if surrogate peptides are not unique, digestion and subsequent enrichment may result in incorrect measurements, because one peptide sequence may be present in multiple proteins.  

In our observation, all samples can benefit from immunoenrichment, not just those in which tests are searching for low-abundance molecules. Enrichment of high-abundance targets reduces processing times, lowers costs, and improves specificity. First, the lower the sample complexity, the faster it can go through trypsin digestion. Second, the process uses less trypsin than if the sample were unenriched; using less trypsin makes a test less expensive. Third, although protein targets may be in high abundance, important isoforms are probably rarer, and enrichment promotes isoform capture. The work with PTH and ApoAI showed that immunoenrichment and detection with the MS immunoassay–SRM method identified new isoforms, as well as new C- and N-terminally truncated variants, which may be present in much lower amounts. A practical advantage lies in reduced cleaning and maintenance requirements for mass spectrometers and LC columns because they are not injected with plasma or serum. Using purified samples also improves routine measurement consistency. 

Multiplexed ELISA or other immunoassay technologies present limitations because as the number of analytes increases, sensitivity typically decreases. The immunoenrichment technique allows multiplexing of analytes by serial extraction of the same sample. To test for different targets, samples can be enriched using different antibodies embedded in the microcolumns on the pipette tips. This approach is fast, conserves sample volumes, and avoids the problem of accommodating several antibodies with different optima for binding and elution conditions on a single substrate.  

Conclusion: Immunoenrichment for MS-Based Translational Proteomics

MS-based proteomics is a powerful tool for research and facilitates two important areas: identification, verification, and validation of new biomarkers, and detection and quantification of those markers in clinical research settings. Potential clinical research and biopharmaceutical applications might include screening, tracking progression, predicting course and outcome, monitoring for recurrence, and personalized assessment of drug response and toxicity (4). MS immunoassay–SRM–based proteomics, using the immunoenrichment method described, may offer a significantly shorter development time, reduced costs, and higher specificity for the identification of proteins and their isoforms in comparison to existing methods.  

 are with the Thermo Fisher Biomarkers Research Initiatives in Mass Spectrometry Center (BRIMS) in Cambridge, Massachusetts. Direct correspondence to: 

Translational proteomics has not been as successful as originally anticipated. Because mass spectrometry (MS) can separate proteins at the sequence level, it provides the selectivity needed for this application; however, traditional challenges still exist, including time-to-result, throughput, and sample-size requirements.

An Enhanced Immunoaffinity Enrichment Method for Mass Spectrometry–Based Translational Proteomics

Issue PDF

Click the title above to open the March 2014 issue of Current Trends in Mass Spectrometry, Volume 12, Number 1, in an interactive PDF format.

Special Issues-03-01-2014

Application of Mass Spectrometry to Support Verification and Characterization of Counterfeit Pharmaceuticals

Comparison of Edible Oils by GCxGC–TOF-MS and GC–High Resolution TOF-MS for Determination of Food Fraud: A "Foodomics" Approach

A New Outlook on Soft Ionization for GC–MS

An Enhanced Immunoaffinity Enrichment Method for Mass Spectrometry–Based Translational Proteomics

Vol 12 No 1 Current Trends in Mass Spectrometry March 2014 Issue PDF