A single-stage mass spectrometer provides one dimension of mass analysis. As shown in the schematic in Figure 2, a single mass spectrometer can be viewed simply as providing a survey of all ions generated in the ion source. For example, if a 10-component mixture is injected into a chromatograph and each component is separated with adequate resolution, then the mass spectrometer would provide discrete measurement of mass-to-charge of each of the 10 components. The resulting mass spectrum of each component is generated for subsequent interpretation and assignment of molecular mass.

For some sample types and applications, a direct insertion probe is used to introduce the sample into the mass spectrometer. The sample, usually a pure compound or relatively pure compound, is dissolved in an appropriate solvent. A small amount of the sample (typically 1 µL or less) is deposited onto the probe. Typically, the probe consists of a metal filament, such as platinum, located at its tip. The metal filament is heated, and the sample is desorbed into the ion source of the mass spectrometer. The probe then is removed from the mass spectrometer and is prepared for the next sample.

Many MS pharmaceutical analyses use the positive ion mode. For example, compounds that contain a basic functional group such as an amine (e.g., proteins and peptides) are excellent candidates for MS analyses in the positive ion mode because under acidic conditions these compounds readily accept a proton to form positively charged ions.

Electron ionization (EI) is used mostly with GC applications when the analytes of interest are nonpolar and are easily volatilized. Mass spectra produced by EI procedures are characterized by extensive fragmentation. Because EI typically produces extensive fragmentation, EI is considered a hard-ionization mode. Mass spectra are obtained when a 70 eV electron beam enters the source and impacts the analyte molecules present in the gas phase:

Nearly all EI applications are conducted in the positive ion mode. These spectra are highly reproducible. Thus, EI spectral libraries can be used to determine the structure and to confirm the identity of unknown compounds. Additionally, fragmentation patterns have been extensively studied for electron ionization and can help determine structure of unknown compounds.

Chemical ionization (CI) procedures rely on electron ionization of reagent ions such as methane, ammonia, or isobutene. The reagent ions react with the analyte molecules (ion–molecule reaction) in the source of the mass spectrometer. In the positive ion mode, analyte molecules are ionized via proton transfer and/or adduct formation, producing even electron molecular ion species or adduct ions. Depending on the analyte, both positive and negative ion spectra can be obtained with CI procedures. CI is considered a softer ionization (less fragmentation of the molecular ions) mode than EI although some approaches can induce fragmentation depending on the CI gas chosen and the structure of the analyte molecule. CI is very useful for reactive and unstable compounds where a molecular mass determination is desired. Representative CI ionization reactions for both GC–MS and LC–MS methods follow.

Protonated molecule formation (GC–MS example):

Protonated molecule formation (LC–MS example):

Adduct ion formation (LC–MS example):

Atmospheric pressure ionization (API) procedures allow the direct introduction of samples into the mass spectrometer from a liquid interface such as an HPLC. The two most common forms of API used in pharmaceutical analyses are electrospray ionization (ESI) and atmospheric pressure chemical ionization (APCI).

Matrix-assisted laser desorption ionization (MALDI) is a soft-ionization procedure used primarily for biomolecules that relies on the addition of a chemical matrix dried with the analyte of interest. This matrix compound can absorb laser energy at a particular wavelength during the laser ablation process. By a mechanism that has not been fully elucidated, some of the ions generated in the matrix can transfer protons to the analyte, and the resulting gas-phase ions are focused into the mass spectrometer. Specific matrices have been empirically developed for various classes of analytes, including 3,5-dimethoxy-4-hydroxycinnamic acid (sinapinic acid) and -cyano-4-hydroxycinnamic acid for proteins and peptides and picolinic acid for oligonucleotides. Unlike ESI, MALDI typically produces lower charge-state ions, with singly charged ions being the most favorable.

Ambient ionization generally refers to a collection of MS procedures that permit direct sampling and interrogation of analytes from sample matrices or surfaces under ambient conditions with little or no pretreatment. In recent years, the number and type of ambient ionization procedures have rapidly expanded. This is evident from the proliferation in recent literature of acronyms that represent various ambient ionization procedures. Ambient ionization has enjoyed broad application in fields that include forensics; food safety; monitoring of environmental contaminants; polymers; fuels; detection of explosives and drugs of abuse; molecular imaging of surfaces and tissues; profiling and characterization of metabolites, proteins, and biomolecules; as well as monitoring of chemical reactions and processes.

Ambient ionization procedures typically produce analyte ions directly from sample surfaces or require initial production of analytes that subsequently are ionized by any of several processes. Therefore, ambient ionization procedures can be broadly categorized as representing either direct or multistage ionization mechanisms. Direct ionization procedures generate analyte ions from a sample solution or droplet in an electric field, or desorb analyte ions by impinging a sample surface with charged droplets or solvent ions, photons, or metastable atoms. Multistage ionization procedures initially produce analyte particle or droplets with a liquid or gas stream, thermal desorption, irradiation or ablation with a laser, or nebulization. Analytes subsequently react with a charged species or with metastable atoms generated by ESI, APCI, or photoionization to produce analyte ions.

Table 1 is a representative, but not comprehensive, listing of ambient ionization procedures and illustrates the diversity and similarity among various approaches.

A time-of-flight (TOF) mass spectrometer uses differences in transit time through a field-free drift region to separate ions. Ions generated in the ion source are pulsed into the field-free drift region (flight tube) by an electric field. Lighter ions have a higher velocity and reach the detector sooner. The TOF mass spectrometer has benefited significantly from the use of fast electronics and fast computers to provide systems characterized by high speed, high sensitivity, and high resolution.

The triple quadrupole mass spectrometer is used for both qualitative and quantitative analysis. A triple quadrupole mass spectrometer features a first stage of mass analysis (Q1) for the selection of a precursor ion, followed by an RF-only collision quadrupole region (Q2). A second stage of mass analysis (Q3) is used for product ion analysis. Most quantitative LC–MS-based analyses feature the use of triple quadrupole mass spectrometers. Furthermore, the triple quadrupole mass spectrometer is capable of providing all the most common qualitative scan modes: full scan, product ion, precursor ion, and neutral loss (see Figure 4 and 4.2 MS/MS and MSn Spectrometry Operational Modes).

Collision-Induced Dissociation (CID) or Collisionally Activated Dissociation (CAD) is the process by which a selected precursor ion undergoes collisions with neutral gas molecules in a collision region to yield product ions. Fragmentation occurs at relatively low energies (1–100 eV) primarily at the site(s) of ionization to yield a fragment ion and a neutral molecule. Nitrogen, argon, or helium typically is used as the collision gas with collision energies in the 10–50 eV range.

Electron Capture Dissociation (ECD) and Electron Transfer Dissociation (ETD) are two distinct but related fragmentation methods used primarily to fragment proteins and peptides to generate c- and z-type ions that are useful for determining protein sequence and characterization of protein modifications. The fragmentation in each case relies on the introduction of either a free electron (ECD) or a radical anion (ETD) into the gas phase of positively charged molecules to induce chemical fragmentation. The advantage of these methods for proteins and peptides versus CID is that the chemical fragmentation occurs almost exclusively along the peptide backbone, thereby preserving the position and context of protein modification that often are lost during CID.

In MS, resolution is the extent to which two ions of adjacent m/z can be distinguished from each other, and has been expressed in different ways. Two adjacent peaks in a spectrum at values m and m Dm may be separated by a valley, which at its lowest point is a specified maximum amount of overlap between the peaks. The resolution would be expressed as m/Dm. For example, the term “10% valley” describes a 10% amount of overlap between two peaks and is appropriate when the peaks are of equal height and shape. In practice, however, adjacent peaks of equal height and shape are rarely observed. Therefore, resolution also can be calculated by using the apparent width of a single peak at a given point in mass range as Dm, and it often is expressed as the measured width of a peak at half-height (full-width half-height maximum, or FWHM). Resolution is an important factor in establishing mass accuracy, evaluating isotope patterns and abundance, determining charge state of multiply charged ions, and distinguishing nominally isobaric interferences.

Nominal mass spectrometers instruments typically provide mass resolution on the order of 1000, whereas for high-resolution mass spectrometry (HRMS) this can range from 5000 to 1,000,000. Improved resolution includes the ability to separate and distinguish isobaric interferences (analytes with the same nominal mass but with different empirical formulae) and to mine complex data sets more efficiently (using increased specificity).

Details of the process for characterizing proteins and peptides are provided in Biotechnology-Derived Articles—Peptide Mapping 1055 and are only summarized in this section to provide context.

The analysis of drug substances in this class is discussed in Glycoprotein and Glycan Analysis—General Considerations 1084. Thus this section addresses only the application of MS to the characterization of protein glycosylation.

Many biopharmaceuticals are glycoproteins for which glycosylation plays critical roles in maintaining function, stability, and solubility. The most common glycosylations are N-linked and O-linked glycosylation in addition to other glycosylation modifications such as GPI anchors. N-glycosylation refers to glycan modification on the asparagine side chain, and O-glycosylation occurs to the side chains of serine and threonine. MS analysis of protein glycosylation can be performed on three different levels: intact proteins, peptides, and glycans. This section provides a brief overview of glycoprotein analysis.

Vaccines present a unique challenge because often they include complex mixtures of proteins, polysaccharides, and lipids in the form of intact bacteria or viruses, outer membrane vesicle preparations, or combinations of recombinant proteins. General information regarding bacterial vaccines is available in Vaccines for Human Use—Bacterial Vaccines 1238. Quality control of such mixtures typically relies on characterization by SDS–PAGE, SEC, amino acid or monosaccharide compositional analyses, RP–HPLC, and other analytical methods. However, to date MS has not been used as a primary QC method even though it is frequently used for confirmation of identity and characterization of impurities. Because of the many advantages of MS described in this chapter and its widespread use in the characterization of biotherapeutics and vaccines that contain complex protein mixtures, MS and MS–MS offer the ability to identify product and impurity simultaneously, and likely will be adapted in routine QC environments in the future. Additionally, when a complication occurs (e.g., discrepancy between assays, activity changing after formulation, or posttranslational modification), the quantitation of the recombinant protein (within a complex proteome) or key antigen in a complex protein vaccine via a highly specific procedure such as MS may be desirable. The use of isotope dilution MS–MS in combination with multiple reaction monitoring scans can provide highly specific and accurate quantitation of protein in complex samples.

For optimal accuracy and precision, internal standards should be employed at a consistent concentration in both the test samples and reference standards. Internal standards compensate for small sample-to-sample differences in dilution volume or loss of material during sample processing. They also compensate for analysis-to-analysis fluctuations in MS response or sensitivity drift over the course of batch analyses, both of which occur in all MS-based methods. In practice, selective detection schemes are set up to monitor both the target compound (analyte) and its corresponding internal standard. The key instrumental measure for each standard and test sample is, therefore, a ratio of the analyte-to-internal standard chromatographic peak areas. Internal standards can be either structural or stable-isotope-labeled analogs of the target compound. The former may be adequate and often has the advantage of lower cost and availability. The isotopically labeled (e.g., 2H, 13C, or 15N) alternative offers superior method performance, because it is essentially identical to the analyte in chemical and physical properties, with the exception of greater molecular mass. The only requirements are that the isotopic labels are not subject to chemical exchange during storage or sample preparation, are retained in the MS ionization process, and provide a sufficient molecular mass shift (typically 3+ Da using multiple labels) versus the analyte to ensure negligible cross-talk (interference) between analyte and internal standard signals (because of natural isotopes and variable m/z resolution across different mass analyzers). In addition, care must be taken to ensure that the internal standard contains inconsequential levels of unlabeled target compound impurities that potentially could bias results. Suitability of the internal standard is confirmed after successful completion of the analytical procedure validation.

For accurate integration of analyte and internal standard chromatographic peaks, the data acquisition parameters (e.g., sampling rate, scan range, or masses monitored) must provide a sufficient number of intensity samples across the peak width. The number of samples may vary depending on method performance requirements (see Table 1 in general chapter 736) and practical considerations related to chromatography conditions, mass spectrometer type, and the number of analytes and internal standards monitored. For example, for single-analyte assays on quadrupole-based instruments, analyte and internal standard SIM or SRM detection schemes are alternately sampled throughout the analytical run. With modern quadrupole instruments, dwell times of 100 ms or less per sampling point are readily achieved and should result in at least eight samples across each peak (this is the suggested minimum for good quantitative representation of the peak).

A calibration curve is created by plotting analyte-to-internal standard peak area ratios versus concentration for the reference standards analyzed. For the somewhat narrow calibration range required for Category I and Category II measurements, a simple (nonweighted) linear fit is sufficient to define this response versus concentration function. However, 1/x or 1/x2 weighting also can be used, as well as nonlinear (e.g., quadratic) functions, as supported by appropriate and provided validation data.

The calibration concentration range is selected based on instrument sensitivity and the required method performance attributes. In general, the lowest concentration should yield sufficient signal for reproducibility, and the highest concentration ideally should be within the linear response range of the MS system. In extreme cases, the upper end of the linear response range may result from limitations of the mass analyzer or detector (e.g., as with TOF and ion trap MS instruments, but much less so with quadrupole mass analyzers). More often, loss of linearity results from diminishing ionization efficiency at higher concentrations. This is a common consideration with, for example, electrospray ionization where high analyte concentrations may increasingly deplete available ionizing adduct ions, leading to a less-than-proportional increase in detected signal with increasing concentration. This so-called self-suppression phenomenon can be somewhat compensated by the use of a stable-isotope-labeled internal standard that co-elutes with the analyte and, therefore, experiences similar proportional suppression. However, best-method performance is achieved when calibration falls within the linear range.

The scope of acceptable sample preparation procedures for quantitative MS analysis is broad and depends on many factors, including the analyte chemical class, the composition and complexity of the matrix, the target analyte concentration, and the type of analytical instrumentation available. If appropriate, the preferred approach is simple dissolution followed by internal standard addition and then dilution to a target concentration within the calibration range. This simplicity minimizes the potential for analyte recovery issues and contamination or other procedural errors that may compromise the robustness of more complex methods. Provided sufficient instrument sensitivity, optimal method performance typically is achieved with greater dilution factors (lower analyte concentrations in the prepared sample). This approach not only avoids self-suppression issues but also minimizes the potential for ionization suppression due to co-eluting matrix components.

More complex sample preparation procedures may be required to address issues related to the chemical class or concentration of the analyte, as well as the nature of the matrix. In such cases, the use of an internal standard (stable-isotope-labeled standards in particular) is almost essential to ensure adequate method performance. For best results, the addition of the internal standard should take place at the earliest practical stage of sample preparation (e.g., immediately following test sample dissolution).

For example, procedures such as liquid–liquid and solid-phase extraction may be required to enrich (concentrate) target compound(s) for greater assay sensitivity or for isolation from high levels of ionization-suppressing components within the matrix (e.g., salts or other excipients). Chemical derivatization of certain functional groups of a target compound may be desirable to stabilize the sample (e.g., for GC–MS analysis), improve method chromatographic performance, or enhance intrinsic sensitivity for detection (e.g., for ESI). Although such procedures add complexity, they are acceptable provided the overall analytical procedure can be properly validated.

A special case of complex sample preparation that may be required for quantification of macromolecular compounds such as vaccines and biotherapeutics involves the intentional degradation of the target molecule(s) into smaller entities that are more amenable to quantitative MS analysis. For example, partial degradation of a protein therapeutic may be achieved by hydrolytic (chemical) or enzyme-based (e.g., tryptic) digestion to yield peptide fragments that are characteristic of the protein. These smaller molecules can be analyzed by using principles described in this section (including the use of stable-isotope-labeled internal standards of each targeted peptide), and the results can be correlated to determine the concentration of the original target protein. Further, by measuring multiple peptides derived from the protein of interest, analysts can achieve qualitative (identity) verification of the protein by means of the same method. Of course, this indirect approach usually requires some degree of isolation and qualitative identification of the macromolecule to ensure that the subsequent quantitative measure of the representative peptides is relevant.

Counterfeit drug products may contain multiple or different active ingredients or no active ingredient at all. GC–MS, LC–MS, or direct-infusion MS procedures can provide a verification of the presence or absence of an active ingredient or the identity of other active ingredients that may be present. Counterfeit drug products containing multiple active ingredients or very high levels of a single active ingredient (superpotent samples) typically are of greater concern to human safety than those that contain little or none. In cases such as these, MS can provide positive, unequivocal proof of identity, as well as quantitative information sufficient to pursue those responsible for producing the counterfeit products.