33. Analysis of Food Contaminants, Residues, and Chemical Constituents of Concern

33.2.1 Choice of Analytical Method

33.2.2 Sample Preparation

33.2.2.1 Introduction

Food samples are usually too dilute (e.g., beverages) or too complex (e.g., meat) for direct analysis of trace contaminants and residues. Therefore, sample preparation, including homogenization, extraction, fractionation/cleanup, concentration, and/or derivatization, normally precedes the analysis of food contaminants and residues (Fig. 33.2). Analysts are faced with continuous decrease of legislative limits for food contaminants and residues, resulting in the need for more sensitive, precise, and accurate measurements. Sample preparation techniques for the analysis of food contaminants and residues are continuously improved to guarantee high recovery and reproducibility (see Ref. [8] as an example). Fortunately, there also have been significant advances in analytical methods and equipment. The advanced technology of mass spectrometers (see Chap. 11), specifically, their enhanced “recognition features” when used in tandem, allows them to take over the “selectivity” of classical sample preparation methods. However, for quantification purposes, cleaner extracts are preferred. In both target and multiresidue methods, isotope dilution and addition of internal standard provide the most accurate results, compensating for sample matrix effects and ion suppression in MS. Also, faster and more efficient extraction methods are now available for the analysis of food contaminants and residues, as will be discussed in subsequent sections.

Analytical chemists often focus on perfecting the analytical technique (e.g., chromatographic analysis), overlooking the importance of sampling, sample storage, and sample preparation. Sampling and sample preparation are labor intensive and time consuming but are essential prerequisites for acquiring meaningful analytical data. There is a high probability of error and contamination during sampling and sample preparation, and these cannot be corrected at any point during the analysis. Therefore, an adequate plan for sampling, storing, and preparing samples should be implemented and validated by statistical analysis (see Chap. 5 for more details). Obtaining a sample that is representative of the level of trace substances in a heterogeneous mixture is not an easy task. Sampling for the analysis of a particular contaminant or residue will be briefly discussed in relevant sections of this chapter.

33.2.2.3 Extraction and Cleanup

33.2.2.3.1 Introduction

Almost all food samples, with the exception of samples directly soluble in an organic solvent (e.g., vegetable oil), require a solvent extraction step (Chap. 12, Sect. 12.​2; Chap. 14, Sect. 14.​2.​2.​4) to isolate the target analytes from the matrix. Defatting of lipid-rich food matrices is often required, using hexane or isooctane, prior to the extraction of target contaminants. Extraction of contaminants and residues is traditionally done by solubilizing them in a suitable organic solvent, generally acetonitrile or acetone. Anhydrous salt (NaCl or Na₂SO₄) can be added to absorb water. In some cases, water is added so that the crude extract can be purified using a subsequent partitioning step with a second, water-immiscible solvent. When the extraction process is complete, the solvent is separated from insoluble solids by filtration.

Often, the crude sample extract is purified and concentrated before the separation/determination steps. The degree of cleanup required depends on the method of analysis such as the chromatography mode and detector type. The objective of cleanup is to separate the target analytes from coextractives that can interfere with their detection. Often, the preliminary cleanup step is followed by a preparative chromatography step (see Chap. 12 for basic information on chromatography). Cleanup of crude water-acetone or water-acetonitrile extract, for example, can be done by partitioning with a relatively nonpolar organic solvent. The residues then can be further purified by column chromatography (either adsorption or size-exclusion chromatography). Separate fractions of column eluate can be analyzed. The choice of packing material is both analyte and matrix dependent. In some cases, such as determining aflatoxin in milk [9], monoclonal antibody affinity chromatography is used as a one-step column cleanup.

Great efforts have been made to develop extraction techniques that are faster and more efficient and require less solvent. Extraction and purification techniques have been developed based on the characteristics of the target analyte(s), which can be categorized into three classes: (1) volatile compounds (VC) that can be analyzed via headspace techniques, after derivatization if needed, (2) semi-volatile compounds (SVC) that are GC amendable (thermally stable), and (3) the nonvolatile (thermally labile) compounds (NVC), which are mostly analyzed by HPLC after extraction. Some of the widely used extraction techniques will be covered briefly in subsequent sections. References [10–14] contain more details on the various extraction techniques.

33.2.2.3.2 Solid-Phase Microextraction

All three classes of compounds listed above can be extracted using solid-phase microextraction (SPME), which is a type of solid-phase extraction (SPE) (Chap. 14, Sect. 14.​2.​2.​5). VCs are best analyzed by headspace techniques (see Sect. 14.​2.​2.​2). Headspace sampling applications include the analysis of pesticides, furan, and residues from packaging materials. Headspace sampling can be done by immersing the SPME polymer-coated fibers into the headspace. SPME also can be used for SVCs and NVCs, with the polymer-coated fiber directly immersed in the aqueous samples, or placed inside a hollow cellulose membrane.

33.2.2.3.3 The QuEChERS

QuEChERS method [15, 16], which stands for quick, easy, cheap, effective, rugged, and safe, is a dispersive SPE (DSPE) technique that is by far the best method to extract multiple pesticide residues (Sect. 33.3.2) (see Chap. 14, Sect. 14.​2.​2.​5, for SPE). A single sample preparation method and one chromatography method linked to MS (preferably in tandem) to determine as many pesticides as possible are very much desired. The advantages of the QuEChERS method (outlined in Fig. 33.3) include speed, ease, minimal solvent use, and lower cost as compared to conventional SPE. The types of adsorbents and solvents, as well as the pH and polarity of the solvents, can be adjusted based on the sample matrix and types of analytes. QuEChERS kits are now commercially available (e.g., Sigma Aldrich/Supelco, Restek, and United Chemical Technologies).

33.2.2.3.4 Microwave-Assisted Solvent Extraction

Microwave-assisted solvent extraction (MASE), which utilizes electromagnetic radiation to desorb organics from their solid matrix, can decrease significantly the extraction time and the quantity of solvent required to efficiently extract target analytes (see Ref. [17] for example application). The efficiency of MASE is attributed to the elevated temperatures that exceed the boiling point of the solvent(s) and the rapid transfer of the analyte(s) to the solvent phase. Commercial systems are available that incorporate the capacity for simultaneously extracting multiple samples within closed, lined (perfluoroalkoxy), pressurized vessels (up to 8–12), using microwave absorbing solvents. Disadvantages of MASE include lack of selectivity and loss of thermolabile analytes.

33.2.2.3.5 Accelerated-Solvent Extraction

Accelerated-solvent extraction (ASE) (Chap. 17, Sect. 17.​4.​4) utilizes limited quantities of organic solvents at elevated temperature (up to 200 °C) and pressure (1,500–2,000 psi) to statically or dynamically extract solid samples for short periods of time (often <10 min) [18]. See Ref. [19] for example application of ASE for pesticides and antibiotic residues. Disadvantages of ASE include diluting effect and lack of selectivity, requiring further cleanup and concentration.

33.3.1 Introduction

A pesticide is any substance or a mixture of substances formulated to destroy or control “pests” including weeds, microorganisms (e.g., fungi or bacteria), insects, and even mammals. Currently, there are over 1,300 registered pesticides [21], main classes of which include: (1) herbicides [for weed control, e.g., triazine (e.g., atrazine)], (2) insecticides [e.g., organochloride (OC) (e.g., dichlorodiphenyltrichloroethane (DDT)), organophosphates (OP) (e.g., malathion, dimethoate, omethoate), and methyl carbamates (e.g., aldicarb)], and (3) fungicides [e.g., phthalimide (e.g., captan)]. Other types of pesticides may include acaricides, molluscicides, nematicides, pheromones, plant growth regulators, repellents, and rodenticides. Reference [5] lists active substances used in various pesticides products that belong to different chemical groups, and FDA provides a glossary of pesticide chemicals [22].

Pesticide residues may occur in food as a result of the direct application to a crop or a farm animal or as a postharvest treatment of food commodities. Pesticide residues also can occur in meat, milk, and eggs as a result of the farm animal consumption of feed from treated crops. Additionally, pesticide residues can occur in food from environmental contamination and spray drift.

If pesticides were not applied, an estimated one-third of the crop production would be lost. However, pesticides may have adverse effects on human health, including, but not limited to, cancer, acute neurologic toxicity, chronic neurodevelopment impairment, and dysfunction of the immune, reproductive, and endocrine systems. Accordingly, there are strict regulations for pesticide registration and use all over the world.

Risk assessment studies, required by the EPA, are done to determine the nature and extent of toxic effects and to establish the level at which no adverse effects are observed [no observed adverse effect level (NOAEL)]. Based on risk assessment studies, tolerance levels [or maximum residue level (MRL)] of pesticides (the active ingredients as well as toxic metabolites, and transformation products) in food are set and enforced by government agencies, as mentioned in Sect. 33.1, to protect stakeholders and regulate international trade. In fact, tolerance levels must be established prior to registration. In general, the tolerance levels in foods range between 0.01 and 10 mg/kg, depending on the commodity and the pesticide used. Low tolerance levels have necessitated the development of more accurate and sensitive analytical methods to meet the requirements in food.

33.3.2 Types of Analytical Methods

Several factors make the analysis of pesticide residues in food a complex task. These factors include the complexity of the food matrix, the possibility of having pesticide levels as low as pictogram or femtogram amounts, and the considerable differences in the physical and chemical properties of pesticides.

Analytical methods employed for the analysis of pesticides can be categorized into either single-residue methods (SRM) or multiple-residue methods (MRM). SRMs are designed to measure a single analyte and, often, its toxic metabolites. The majority of SRMs have been developed for purposes of registration and setting tolerance levels, or for investigating the metabolism and environmental fate of a specific pesticide. Sampling, extraction, purification, and determination in SRMs are optimized for the target pesticide. The majority of the SRMs currently in use, which have undergone EPA review or have been published in peer-reviewed scientific journals, are found in Volume II of the Pesticide Analytical Manual (PAM II) [23]. For purposes of monitoring quality and safety, given the large number of pesticides and the considerable differences in their physical and chemical characteristics (e.g., acidic, basic, or neutral, polar or nonpolar, and volatiles or nonvolatiles), MRMs can determine various pesticides in a single run and are much preferred. Many of the MRMs are found in Volume I of the Pesticide Analytical Manual (PAM I) [24]. AOAC International [25] also has developed an MRM for pesticide residues, the AOAC Pesticide Screen (AOAC Method 970.52). The identification and quantification in MRMs currently used by the FDA and USDA are based on GC and HPLC analysis.

Prior to chromatographic analysis, sampling, extraction, and fractionation/cleanup in MRMs are optimized to ensure an efficient transfer of most if not all of the pesticide residues present in the sample matrix to the organic phase. To achieve efficient transfer of as much of the pesticide residues as possible from the sample matrix into the organic phase, partitioning using water-miscible solvent is performed. This is followed by partitioning with nonpolar solvent that is miscible with the polar solvent, yet immiscible with water. After the extraction step, a cleanup step is performed using adsorption columns, in particular Florisil, alumina, and silica gels, with solvent mixtures of low polarity for elution. Commercial kits based on solid-phase extraction chromatography are available for pesticide cleanup. Reference [26] includes detailed descriptions of different MRM extraction and cleanup protocols specific for pesticide analysis.

33.3.3 Analytical Techniques Used for the Detection, Identification, and/or Quantification

A wide range of analytical techniques can be used in pesticide analysis. This section highlights some of the current analytical techniques used for the detection and determination of pesticide residues.

33.3.3.3 Mass Spectrometry Detection

Details on MS instrumentation, modes of ionization, and mass analyzers are provided in Chap. 11. The subsequent sections will provide sample applications of GC-MS and LC-MS used in pesticide residue analysis. References [6, 7, 16, 26, 28–31] describe more application examples and details on method development.

33.3.3.3.1 Gas Chromatography-Mass Spectrometry

The most common GC-MS technique for the analysis of pesticide residue involves single quadrupole instruments with electron impact (EI) ionization. The selected ion monitoring (SIM) option enhances selectivity, increases sensitivity, and minimizes interferences from co-extracted compounds. Ion-trap detectors (ITD) are also used in the analysis of pesticide residues. Using ITD, the analyzing in full-scan mode provides higher sensitivity as compared to single quadrupole analysis and allows for confirmation by library searches (NIST library spectrum search). Additionally, the ITD enables tandem MS analysis (MS/MS) by means of collision-induced dissociation (CID). The use of tandem MS improves selectivity and significantly reduces background without loss of identification capability, thus enabling the analysis of pesticides at trace levels in the presence of many interfering compounds. Figure 33.4 provides an illustration of enhanced compound identity confirmation upon the use of MS/MS vs. only MS. For the determination of multiclass residues, triple quadrupole (QqQ) MS is becoming a powerful and fast analytical tool, requiring minimum sample preparation. The chromatogram separation becomes less important when the analysis is carried out by GC/QqQ-MS/MS, since the analyzer is able to monitor simultaneously a large number of co-eluting compounds. Reliable quantitation and confirmation can be easily achieved, even at trace concentration levels. Time-of-flight (TOF) MS instruments also are gaining popularity for the simultaneous analysis of multiclass pesticides [16, 30].

33.3.3.3.2 High-Performance Liquid Chromatography-Mass Spectrometry

Similar to GC-MS, the use of LC-MS in the analysis of pesticide has undergone great development. Tandem MS methods, using atmospheric-pressure ionization (API), atmospheric-pressure chemical ionization (APCI), and electrospray ionization (ESI), have been developed for many pesticides such as OP, carbamate, and sulfonylurea pesticides [29]. Specifically, the ESI coupled with MS/MS was found to have high sensitivity and selectivity for a wide range of pesticides in foods. Because of HPLC solvent interferences, single quadrupole instruments are not as widely used in LC-MS as compared to GC-MS for pesticide analysis. When using LC-MS for pesticide residue analysis, QqQ are the most widely used mass analyzers. LC/TOF-MS has also gained popularity due to its high speed, sensitivity, and selectivity [31].

33.4.1 Introduction

Molds, which are filamentous fungi, can develop on food commodities and produce various types of chemical toxins, collectively known as mycotoxins. The main producers of mycotoxins are fungal species belonging to the genera Aspergillus, Fusarium, and Penicillium. Crops can be directly infected with fungal growth and subsequent mycotoxin contamination as a result of environmental factors such as temperature, humidity, weather fluctuations, mechanical damage of crops, and pest attack. In addition, plant stress due to extreme soil dryness or lack of a balanced nutrient absorption can induce fungal growth. Fungal infection and mycotoxin contamination can occur at any stage of the food chain (Sect. 33.1). The United Nation’s Food and Agriculture Organization estimated that up to 25 % of the world’s total crops per year are affected with “unacceptable” levels of mycotoxins [32].

Mycotoxin contamination may occur in food as a result of the direct mold infection of plant-origin commodities, such as cereals, dried fruit, spices, grape, coffee, cocoa, and fruit juices (especially apple based). Mycotoxins also can occur in milk, eggs, and, to a minor extent, meat, as a result of the farm animal consumption of feed from contaminated crops. Additionally, humans and animals can be exposed to mycotoxins through heavily contaminated dust, as in the case of harbors and warehouses.

While molds can be destroyed by natural causes or processing of food, mycotoxins can survive. To limit human exposure to mycotoxins, a key critical control point is avoiding the processing of raw materials with unacceptable mycotoxin levels. Therefore, implementing periodic testing is a necessity, including the adoption of reliable sampling procedures and validated methods of analysis. Organizations such as the AOAC International, American Oil Chemists’ Society (AOCS), AACC International, and the International Union of Pure and Applied Chemistry (IUPAC) have method validation programs for mycotoxin analysis. References [33–37] provide detailed information on mycotoxin occurrence, health effects, control, sampling and sample preparation, and analysis.

33.4.2 Sampling

Compared to the analysis of other types of residues, the sampling step in the analysis of mycotoxins is by far the largest contributor to the total error. The possible variability is associated with the level and distribution of mycotoxins in the food commodity. An unevenly distributed 0.1 % of the lot is usually highly contaminated, resulting in an overall level above the tolerance limit. Due to this heterogeneous distribution, an appropriate sampling plan is needed to ensure that the concentration in a sample is the same as that in the whole lot. An incorrect sampling protocol can easily lead to false conclusions, often false negative, leading to undesirable health, economic, and trade impacts.

The Commission Regulation 401/2006 [36] provided specifications for the sampling of regulated mycotoxins in various food commodities including cereals, dried fruits, nuts, spices, milk and derived products, fruit juice, and solid apple products. Also, AOAC International, AOCS, and the FDA in collaboration with the USDA have developed detailed sampling plans for separate commodities. In general, an acceptable plan involves obtaining a large number of samples from multiple locations throughout the lot, creating a composite sample, grinding or slurring the composite sample (to reduce particle size and/or increase homogeneity), and subsampling for laboratory analysis. The number and size of collected samples and laboratory subsamples are dependent on the matrix and the size of the lot.

An example of sampling for mycotoxin analysis is the sequential sampling of raw shelled peanuts (with a tolerance level of less than 15 μg/kg for aflatoxins); a bulk sample of approximately 70 kg is randomly accumulated (at a rate of one incremental portion per 225 kg of lot weight). This bulk sample is divided randomly into three 21.8-kg samples using a Dickens mechanical rotating divider and then ground separately. A subsample (1,100 g) is taken from one of the samples, formed into a slurry, and analyzed for aflatoxins, in duplicate. If the average of the two determinations is ≤8 μg/kg, the lot is passed and no further testing is performed. If the average is ≥ 45 μg/kg, the shipment is rejected. For averages >8 μg/kg and <45 μg/kg, a second 21.8-kg sample is analyzed in duplicate, and the average of the four results is used to decide whether to accept (≤12 μg/kg) or reject (≥23 μg/kg) the lot. If the average falls between the second set of determining values, the third 21.8-kg sample is analyzed. If the average of the six determinations is <15 μg/kg, the lot is accepted.

33.4.3 Detection and Determination

Post sampling, sample preparation commonly includes extraction, cleanup, and concentration. Sample preparation steps (discussed in Sect. 33.2.2) that are specific for mycotoxin analysis have been developed. Shaking or blending often is used for mycotoxin extraction, while SPE often is used for cleanup, especially when multi-mycotoxin analysis is needed. For a specific mycotoxin, immunoaffinity LC (IAC) is used for cleanup.

Tremendous effort has been made to develop and optimize qualitative and quantitative methods for mycotoxin analysis. A list of official methods of mycotoxin analysis is included in Ref. [32] and others in Refs. [33–35, 37]. Mycotoxin analysis kits, both for cleanup (for GC, HPLC, or TLC analysis) and for detection (e.g., ELISA, LFS, immunoaffinity with HPLC), are commercially available. The sections below briefly discuss some of the current and newly developed analytical techniques used for the detection and determination of mycotoxins in food.

33.5.1 Introduction

Animals intended for human consumption may not only be given drugs (e.g., antibiotics, antifungals, tranquilizers, and anti-inflammatory drugs) at therapeutic levels to combat diseases, but also some countries allow use of drugs (mainly antibiotics) at subtherapeutic levels to reduce the incidence of infectious diseases and for weight gain advantages. The Center for Veterinary Medicine (CVM), a branch of the FDA, regulates the manufacture, distribution, administration, and withdrawal period (time between the last drug treatment to the animal and the slaughter or use of milk or eggs by humans) of drugs given to animals, including animals from which human foods are derived. Most drug residues of concern for human food are antibiotic residues and thus are the focus of this section. Some of the major families/types of antibiotics, referred to in later sections, include β-lactams, sulfa drugs, cephalosporins, tetracyclines, and chloramphenicol.

Residual levels of any antibiotics given to animals used for human consumption are of concern for a variety of reasons, including the fact that some consumers are allergic to certain antibiotics, and the possibility that some microbes may develop antibiotic resistance due to constant exposure. Some antibiotics can be carcinogenic as well, such as nitrofuran compounds [47]. Also, on a practical level for dairy products such as cheese made with starter cultures, antibiotic residues would reduce the intended microbial growth and therefore reduce acid production. For all these reasons and more, the FDA has strict regulations on antibiotic residues in human food. Due to these regulations, antibiotic-contaminated meat and milk (including milk products) are considered to be adulterated. The FSIS monitors antibiotic residues in meat products from cattle, swine, lamb, goat, and poultry. Number of violations vary from year to year, for example, 38 antibiotics violations were reported in 2004, while 8 were reported in 2011 [48]. FDA’s Center for Food Safety and Nutrition monitors antibiotic residues in milk and its products, including milk from pick-up tanks, pasteurized fluid milk, cheeses, etc. In 2014, out of four million samples tested, 703 were reported to have antibiotic residues including β-lactam and sulfonamide [49].

Samples are tested for antibiotic residues using rapid screening methods. A positive result from a screening test suggests that one or more types of antibiotic are present, so further testing is required to identify and quantify the specific antibiotics present.

33.5.2 Detection and Determination

Often, procedures such as defatting, protein hydrolysis (in case of meat or egg samples), protein precipitation (in case of dairy samples), and aqueous wash (in case of honey to remove excess sugar) precede the extraction of antibiotic residues. For the extraction of many antibiotics, liquid-liquid extraction and SPE are commonly used, followed by a partial purification step often using ion-exchange cleanup systems that take advantage of the acid/base character of the antibiotics. Sample preparation steps, including extraction and isolation (as discussed in Sect. 33.2.2) specific for drug residues, are reviewed in Ref. [50].

Tolerance levels are established for some antibiotics, while others have a zero tolerance (as is the case for nitrofurans and chloramphenicol). Chloramphenicol, for instance, is an antibiotic of considerable current concern in the USA, European Union, and other countries. It is used in some parts of the world in producing shrimp and has been found in imported seafood products (e.g., shrimp, crayfish, and crab). Because of the adverse health effects on humans, the FDA has banned the use of chloramphenicol in animals raised for food production and set a zero tolerance in human food [21 CFR 522.390 (4)]. Therefore, the analytical methods need to be as sensitive and selective as possible.

A wide variety of analytical methods have been developed and optimized for the analysis of antibiotics, categorized as screening or determinative and confirmatory. Reference [4] lists the methods commonly used in the analysis of several antibiotics, and Ref. [51] gives the process used by the AOAC to validate test kits. One excellent resource for antibiotic test methods, specific for milk and dairy products, is Ref. [52]. The sections below briefly discuss some of the analytical techniques used for the detection and determination of antibiotics in susceptible food commodities.

33.6.1 Introduction

Agriculturally important plants may be genetically modified by the insertion of DNA from a different organism (transgene) into the plant’s genome, conferring novel traits that it would not have otherwise, such as herbicide tolerance or insect resistance. The modified plant is termed a genetically modified organism, or GMO. GMO production relates to a number of food crops, including the following: corn, soybeans, cotton, canola (a cultivar of rapeseed), rice, sugar beets, and papaya. Although protection from pests and herbicide tolerance are the most common GMO traits, other GMO crops include plants that have been modified to improve postharvest quality or enhance the nutritional makeup of the food. Examples include vegetables with an extended shelf life, apples that have reduced browning rate (Granny Smith apples), and “golden rice,” which produces the precursor to vitamin A.

Despite the prevalence of GMO crops, there are concerns about limiting crop variation (i.e., monocropping); unintended impacts on other plants, insects, wildlife, and nearby communities; and possible allergies. The debate about the use of GMO crops has resulted in government regulation for application and labeling. There are specific guidelines put forth by various governments regarding the analyses used. Therefore, even producers must be able to accurately test ingredients and other products.

Many companies produce high-quality, easy-to-use test kits, many of which are specific for a certain GMO protein or gene. These kits generally fall into two categories, based on the methodology: polymerase chain reaction (PCR) kits that specifically amplify the DNA of the GMO gene (specific or shared by many GMOs) so that it can be detected and immunoassays (ELISA and LFS) that are specific for the proteins. Further reading on the GMO and GMO detection topics is listed in the Refs. [61–63].

33.6.2 DNA Methods

Detection of the transgene DNA is an effective method of testing for GMO material in a sample. The analysis involves three distinct steps: (1) extraction of the DNA from the sample, (2) amplification of the DNA by PCR, and (3) identification and quantification of the amplified DNA (note: for real-time quantitative PCR analyses, amplification and detection/quantification occur at the same time). While all three steps are important, the PCR amplification is critical to the specificity and success of the analysis.

33.6.3 Protein Methods

The protein methods for GMOs are immunoassays, primarily ELISA (Chap. 27, Sect. 27.​3.​2) and LFS (Chap. 27, Sect. 27.​3.​4), generally used for testing raw agricultural products and not processed products. Both qualitative and quantitative ELISA test kits for GMOs are commercially available, with the latter having typical detection limits of 0.01–0.1 %. The LFSs are generally less sensitive (0.1–1.0 %), but their speed and ease of use make them ideal for testing in the growing field, at storage areas, and at points of transport.

33.7.1 Introduction

Food allergens are food proteins that trigger an allergic response. Symptoms of an allergic response include hives, face and tongue swelling, and difficulty breathing and can include the severe, life-threatening allergic reaction called anaphylactic shock. It is important to note that food allergy, which triggers an immune system reaction, is distinct from other adverse responses to food, such as food intolerance (e.g., lactose intolerance), pharmacologic reactions (due mainly to food additives such as sulfites and benzoate), and toxin-mediated reactions (due to residues such as pesticides and mycotoxins).

A significant percentage of the population has food allergies, and the prevalence is rising. More than 160 foods can cause allergic reactions in people with food allergies, but over 90 % of the food allergic reactions in the USA are caused by the eight most common allergenic foods, referred to as ‘the big eight’: milk, eggs, fish, crustacean shellfish, tree nuts, peanuts, wheat, and soybeans [64]. In the USA, the Food Allergen Labeling and Consumer Protection Act of 2004 targets these eight food allergens [65]. It applies to all foods regulated by the FDA (both domestic and imported), requiring that labels list all ingredients by their common names and identify the source of all ingredients derived from the eight most common food allergens. Food allergies are an issue around the world; numerous other countries have or are considering labeling regulations. Since there is no cure for food allergies, strict avoidance of food allergens is the only effective action. All of these facts point to the importance of both screening and quantitative methods for analysis of allergenic foods.

Methods available for the detection of food allergens are mainly based on protein or DNA detection, as discussed in subsequent sections. A commercially available rapid screening method that is not based on protein or DNA detection is the swab and adenosine triphosphate (ATP)-sensitive detection method. This method is based on the detection of ATP present on the surface of multiple allergenic foods, e.g., peanut butter, whole egg, soybeans, and milk. It is used mainly to prevent food allergen cross contact during cleaning of processing equipment. Test kits for food allergen analysis (protein or DNA based and others) are commercially available. A review of food allergen analytical methods is given in Ref. [66].

33.7.2 Protein Methods

33.7.2.1 General Considerations

Similar to the analysis of many other food constituents of concern present in small amounts, sampling adequacy and the detection limits are of concern with analysis for allergens. Another concern is the adequate extraction of the different allergens. Unlike the trace analytes of concern mentioned thus far, since the analytes are proteins, the extraction solution is normally a buffer at various pHs and salt concentration. Extraction buffers differ in their ability to extract food allergens, with some solutions extracting the same allergens in different concentrations, while others cannot extract all the allergens. For example, phosphate buffer fails to extract the major peanut allergen (Ara h 3); however, the efficiency of extracting this allergen is greatly enhanced upon the addition of salt (Fig. 33.5). Additionally, it is crucial that the extraction solution is compatible with the assay used (e.g., immunoassay) and does not alter the chemical structure of the analyte. The choice of extraction procedure should also take into consideration the food processing conditions employed. Upon processing, the solubility of the proteins can be reduced due to denaturation and aggregation, resulting in reduced protein recovery [67]. Therefore, to obtain reliable and accurate results, it is crucial to select the right extraction solution for the target analyte(s).

33.7.3 DNA Methods

There are several advantages and disadvantages for the use of DNA-based methods in the analysis of food allergens. DNA-based methods do not target the allergen in the sample; therefore, the detection of the allergen-encoding DNA does not always correlate with the presence of the allergen, especially when the food has been fortified with purified protein. Upon processing, for example, production of protein isolates, protein, and DNA could be separated resulting in false conclusions regarding the presence of the allergen in the sample. Regardless of these disadvantages, DNA-based methods are very specific and sensitive techniques with the advantage that targeted DNA is less affected by several processing and extraction conditions as compared to proteins. The choice to use DNA-based methods is dependent on the type of sample being analyzed.

DNA-based methods involve the extraction of the DNA (see Sect. 33.6.2.1) followed by amplification by PCR using a thermostable polymerase (see Sect. 33.6.2.2). The amplified sample is then visualized by fluorescence staining or by Southern blotting following agarose gel electrophoresis. This procedure normally provides qualitative data, or semiquantitative data if internal standards were used. Quantification can be achieved if real-time PCR (see Sect. 33.6.2.1 and Ref. [66]) or PCR-ELISA was used. PCR-ELISA method involves linking the amplified DNA fragment of an allergenic food to a specific protein-labeled DNA probe, which then is coupled with a specific enzyme-labeled antibody. Quantification of the DNA is based on the enzyme-substrate color producing reaction. Reference [60] includes a list of commercially available real-time PCR and PCR-ELISA test kits; for the analysis of allergens, readers are referred to Ref. [66].

33.8.1 Introduction

Pesticide residues, mycotoxins, antibiotic residues, and allergens in foods have been of concern for many years. However, in any given time period, there are numerous additional chemical hazards that must be addressed. Many of these substances fall into the following categories: (1) banned (e.g., coumarin) or allowed in some countries but not others, (2) legally limited (e.g., sulfites, benzene), (3) intentional contaminants (e.g., melamine), (4) approved for use but of concern (e.g., monosodium glutamate), or (5) natural constituents of concern (e.g., acrylamide, furan). This section would be very lengthy to cover in detail the screening and quantitative methods for many of these chemical hazards. Instead, only a summary (Table 33.3) is given of select current and emerging hazards, with their major method(s) of analysis [69]. To reach the limit of detection (LOD) desired and to positively identify the compounds, many of the major analyses rely on gas or liquid chromatography, with mass spectrometry. Described in more detail below is the analysis for two compounds of concern, sulfites and nitrates/nitrites, which are not as commonly analyzed by chromatography as by other methods. There are legal limits in foods for both of these compounds, so they are monitored regularly in select foods for quality control purposes.

33.8.2 Sulfites

Although sulfites are classified as allergens, they are covered in this separate section because the nature of the chemical compound, symptoms, and methods of analysis are quite different than for other allergens. Sulfites and sulfiting agents are a group of chemical compounds that include sulfur dioxide (SO₂), sulfurous acid (HSO₃), and the following inorganic sulfite salts that can liberate SO₂: sodium (Na) and potassium (K) sulfite, Na and K bisulfite, and Na and K metabisulfite [70]. In some foods they occur naturally, but in other foods they are added for a variety of reasons, including preventing microbial growth and browning. Sulfites naturally occur to some extent in all wines but are commonly added to stop fermentation at the appropriate time and to prevent spoilage and oxidation. Dried fruits and vegetable products are sometimes treated with sulfites to reduce browning. Shrimp, lobster, and related crustaceans can be treated with sulfite to prevent “black spot.” Some consumers are highly intolerant to sulfite residues in food, most commonly resulting in asthma attacks [71]. Therefore, the FDA forbids sulfites from being added to foods intended for raw consumption (e.g., salad bar foods) and requires the phrase “contains sulfites” on the label of foods that contain greater than 10 ppm sulfites, whether naturally occurring or added during manufacturing [21 CFR 101.100 (a)(4)] [72]. Many countries besides the USA have set strict limits on the residual levels of sulfites in various foods.

Reactions of sulfites with other food components make analysis challenging, often resulting in decreased levels during storage. Most methods of analysis detect free forms of sulfite plus some bound forms. However, none of the available methods, alone, measures all forms of sulfite in foods, which includes free inorganic sulfite plus the many sulfite bound forms. It is not known which forms of sulfite cause the adverse responses in sulfite-sensitive consumers, so the focus has been on measuring as much as possible of the residual sulfite, both free and bound forms [63].

One of the long-time, quantitative methods for sulfite analysis of foods is the Monier-Williams procedure (AOAC Method 990.28) [25], which measures “total” SO₂ (actually, free sulfite plus reproducible portion of bound sulfites, such as carbonyl addition products). The FDA refers to this method in regulations for labeling sulfite-containing foods. In this method, the test sample is heated with HCl, converting sulfite to SO₂. Nitrogen gas bubbled through the sample sweeps SO₂ through a condenser and a hydrogen peroxide solution, oxidizing SO₂ to H₂SO₄. The sulfite content of the sample is directly related to the amount of H₂SO₄ generated, measured by either a gravimetric or turbidimetric procedure.

33.8.3 Nitrates/Nitrites

Sodium nitrate and sodium nitrite are both food preservatives commonly used to cure processed meat products, but there are legal limits on the level of sodium nitrite in the finished product. Nitrates (NO₃) and nitrites (NO₂) are very similar chemically, and sodium nitrate is readily converted to sodium nitrite. Sodium nitrate is naturally occurring in some vegetables, and it converts to sodium nitrite when it comes in contact with your saliva. The naturally occurring high levels of nitrates in celery and Swiss chard make powders of them common ingredients in some cured meat products. Both sodium nitrate and sodium nitrite give cured meat an appealing pinkish-reddish color and prevents it from turning brown. More importantly, sodium nitrite works with sodium chloride to prevent the growth of Clostridium botulinum. However, when products with high levels of residual nitrites are cooked at high temperatures (e.g., frying of bacon), the nitrites react with amines naturally present in the meat to form nitrosamines, which are suspected carcinogens. Also, high levels of nitrates or nitrites in the diet may induce the disorder methemoglobinemia, which can be fatal due to the reduced ability for oxygen transport in the blood. In addition to the residual nitrates in processed meat products, there is concern about the levels of nitrates and nitrites in drinking water, and in the water and soil used to grown vegetables, especially when consumed by infants and young children [76, 77].

During the curing of processed meat products, processors ensure the legal limits of sodium nitrite in the finished product largely by carefully monitoring the amount of sodium nitrate used in the formulation of the product. The levels of nitrates/nitrites may be monitored throughout the production system with relatively rapid tests for nitrates/nitrites. There are both official methods and more rapid rapids of analysis used for testing both processed meat products and water, as described below.

Rapid methods of analysis include ion-selective electrodes (ISEs) and test strips. Regarding ISE, both nitrate and nitrite ISEs are commercially available for testing food products (liquid samples) (see Chap. 21, Sect. 21.​3.​4, for details of ISEs). Regarding the test strips, various companies make test strips intended for checking the nitrate content of water. For example, AquaChek nitrite/nitrate test strips (Hach, Loveland, CO) are able to test for either nitrates or nitrites, or both, using two pad areas on the test strip, with one area measuring nitrates and the other pad area measuring nitrites. The nitrate test area contains a combination of chemicals to reduce nitrates to nitrites. Nitrites, at an acid pH, react with sulfanilic acid to form a diazonium compound, which couples with an indicator to produce a pink color. The color intensity is proportional to the concentration of nitrate. Color blocks for interpretation of results are provided for both nitrate N (0–50 ppm) and nitrite N (0–3 ppm).