The LCGC Blog: Choosing the Right Sample Prep and Separation Tools for Food Analysis

What was it like many thousands of years ago, when our prehistoric ancestors realized that converting their meat from the tartare variety to a juicy flame-grilled variety held significant benefits in terms of preservation and taste? At what point were essential oils isolated from vegetables, fruits, and other plants brought into the mix to add more flavors, fragrances, and nutrients? And when did they begin to top that meal off with a nice fermented beverage? At that point I am sure it all must have started to taste like civilization. Today, while we certainly take for granted the variety of foods and beverages that are available to us, we are ever more conscious of the effects of that intake on our bodies. We think about caloric intake, but this is really a composite of the various carbon-containing chemicals that compose the food-a vast array of carbohydrates, fats, fibers, proteins, vitamins, and other species. We think about the type of each class: saturated and trans fats, gluten-full or gluten-free, sucrose or stevia glycosides, and so forth. Most of us take either a quick glance or long look at the nutritional facts provided on the labels of the foods we buy, but who beyond the nutritionalist and analytical chemist (perhaps the avid bodybuilder) really understands the complexity of the situation? Somehow foods must be characterized both qualitatively and quantitatively for their content. And, when you think about what it takes to optimally analyze a complex sample for its fat-, fiber-, protein-, or vitamin-contributing species, you invariably come up with different methods for each. From sample preparation to trace speciation by chemical analysis, speciation of food, through its variety of formats, processing, and content, is an exceedingly complex problem.

In preparation for my stay here in Messina, Italy, I have reviewed a number of articles on food analysis. Here, I propose to give a very quick overview of two reviews that I found quite informative. The first is one by Marazuela and Bogialli (1), which covers novel strategies for sample preparation and focuses on those that have been used primarily to isolate and target antibacterial determinations. I like this article not necessarily because of its focus on pervasive food contaminants, but rather because of the breadth of different sample preparation techniques it covers in this context. The applications discussed, however, are primarily limited to those that involve liquid chromatography (LC) chemical analysis, since it is the best choice for separation and detection of antibacterial compounds. In a greater context, this review does not necessarily consider determination of endogenous food matrix components, although the preparation techniques must certainly isolate the target analytes from interferences. The determination of endogenous food matrix components in their various forms requires considerations of different formats of chromatographic analysis. For this reason, I also liked a recent review by Tranchida  and colleagues (2) on the application of various types of comprehensive chromatography, including gas and liquid and combinations of the two, among others, for food analysis. I like this review, because it takes a practical stance. Based on the complexity of the sample, and the type of analysis desired (for example, targeted versus nontargeted), this review emphasizes the need for a proper choice when deciding whether one-dimensional (1-D) chromatography will be sufficient for a task, or when two-dimensional (2-D) techniques are warranted. More emphasis is placed on the latter in terms of applications, but the review is presented in a very balanced tone and provides critical insight into the topic.

Food safety is a topic of ever-increasing interest as the demand for production grows with the world’s population. Organic contaminants, whether natural or anthropogenic, can include pesticides, toxins, persistent organic pollutants, and drug compounds. As alluded to above, the latter category, which includes antibacterial compounds administered liberally to animals to aid growth and avoid disease, are of significant concern because they can be still present in food intended for human consumption. Although relatively simple and fast screening methods can be devised, if a product tests positive, it must then be passed on for more-detailed confirmatory analysis. Regulatory bodies such as the U.S. Food and Drug Administration and the World Health Organization provide guidance for method development and set tolerances in various forms for different drug contaminants. To create a sensitive and reliable method, one must combine efficient sample preparation with specific chemical analysis. For targeted chemical analysis of drugs from complex matrices, LC with tandem mass spectrometry (MS) detection is the gold standard (although care must be taken depending on the number of analytes targeted). In the case of antibacterials, there are enough physicochemical differences between classes that, to target multiple classes, a fairly nonselective sample preparation must be considered. This makes the LC–tandem MS determination more subject to matrix interferences, because they may also be transferred through preparation to the instrumental analysis.

Sample preparation can be performed both on- and off-line. The authors review a number of techniques, including pressurized liquid extraction, microwave-assisted extraction, solid-phase extraction and microextraction (and various formats for each), matrix solid phase dispersion (MSPD), restricted access materials, and turbulent flow chromatography. Besides the use of more-generalized extraction procedures to accommodate the determination of multiclass compounds, as mentioned above, the authors also emphasize other current trends in sample preparation, such as the use of smaller sample sizes, reduction or elimination of organic solvents, and the potential for automation and high-throughput determination. Each of the techniques mentioned above has advantages and disadvantages relative to these trends. The review does well to survey some novel applications of different formats, and it also highlights the current state of different technologies. Some techniques, such as MSPD, are less well known, and readers are given some nice insight into new approaches and how they can be used for various benefit.

From the chemical analysis standpoint, foods contain a variety of volatile and nonvolatile constituents with a variety of physicochemical character. Tranchida and colleagues present a nice concept in terms of defining the complexity of an analysis based on the number of constituents of interest desired to be visualized. It is in this framework that a discussion regarding the proper choice of 1-D and 2-D chromatographic separations makes the most sense. The decision point appears to reside in the neighborhood of 100–150 compounds of interest, above which, 2-D would certainly be desired based on the availability of increased separation space to resolve analytes, and below which, 1-D is sufficient and avoids the need for the more complex hardware and development inherent to 2-D methods. Key considerations include whether the analysis is targeted or untargeted, and in such a choice, one must choose the appropriate detector. While MS can often provide that additional level of qualitative information, the speed and resolution of the chosen (or available) MS system must be weighed and carefully considered.

Approximately half of the article is dedicated to comprehensive gas chromatography (GCxGC) applications, and the other half reviews comprehensive liquid chromatography (LCxLC) applications, as well as some emerging areas, such as the incorporation of supercritical fluid chromatography (SFC). GCxGC is the far more developed area. Since the introduction of GCxGC for food analysis in 2000 (3), the field has steadily increased to include applications ranging from what are deemed low-complexity food samples (for example, amino acids or fatty acid methyl esters from various food oils) to medium-complexity food samples (for example, aroma volatiles and marine salt volatiles) to high and very high complexity food samples (for example, wine volatiles and metabolites). For comparison, aroma volatiles might contain 100 or so compounds of interest, wine volatiles might approach 300–400 compounds, and metabolites can easily exceed 1000. Both targeted quantitation and nontargeted identification, using headspace and direct injection techniques, have been demonstrated in a number of GCxGC food applications. Further, the choice and development of different modulators used to transfer sample fractions from the first dimension to the second (termed modulation) is also a very active area of research in GCxGC.

Interestingly, the majority of LCxLC food analysis applications to date have been nontargeted in nature. This focus speaks to a less mature field relative to GCxGC, and the added complexity in choosing, optimizing, and coupling the different modes of LC to achieve independent selectivities (or, orthogonality). There are a couple of common ways of combining LC dimensions in a comprehensive fashion -through storage loops or parallel 2-D columns- and these choices dictate to some extent the modes of chromatography that can be combined. Where miscibility issues, precipitation problems, or extreme solvent strength variations exist in transferring the first dimension to the second, sufficient dilution of the modulated sample (essentially, the 2-D injection) must be accommodated, and this is best performed with a parallel 2-D column set-up. Applications to triacylglycerols, phospholipids, carotenoids, polyphenols, and glycosides are reviewed in the article. In many cases, various combinations of 1-D and 2-D modes are compared for a given application to determine which gives the best coverage and resolution. Nevertheless, the authors conclude that it is virtually impossible to achieve a truly orthogonal LCxLC system for real-world sample analysis.

They further conclude that comprehensive 2-D chromatography will likely be “repositioned” over the coming years. This is not to say that it will be replaced any time soon, but the authors note that the rate of MS instrument development and its associated capabilities are increasing faster than those offered by chromatography. The relative role of each technology in complex sample analysis has long been debated, but from my view, both will be equally needed for some time if reliable speciation and quantitation is to remain key in food characterization.

Overall, there are innumerable combinations of sample preparation and chemical analysis techniques that can be used to survey food components, whether endogenous or exogenous. Certainly, many more continue to be developed. What is striking to me is that we generally still operate from the standpoint of one preparation and injection to monitor or study a given class of compounds. Of course, we make compromises the wider we cast our net, as was discussed in the relatively simple terms of monitoring different classes of antibiotics. But what if we want to monitor different types of molecules from a single injection? I think that with the growing ability of analytical chemists to handle on-line formats that it is time to consider more actively how we can get more information from a single injection. In our lab, we are working on what I call a multipath liquid chromatograph that can segregate multiple classes (that is, small and large molecules) from a complex mixture on-line and allow their parallel chromatographic development. There is no reason why such a system could not involve parallel comprehensive chromatography in each path. That might not make complete sense to you the reader right now, but it is something that I will write more about in the future. For now, as with biofluids, food analysis provides a rich sample matrix with many compounds of interest for analysis to contemplate-but one must always take care that the right tool is chosen for the desired task.

References

1. M.D. Marazuela and S. Bogialli, Anal. Chim. Acta645, 5–17 (2009).

2. P.Q. Tranchida, P. Donato, F. Cacciola, M. Beccaria, P. Dugo, and L. Mondello, Trends Anal. Chem.52, 186–205 (2013).

3. J.-M.D. Dimandja, S.B. Stanfill, J. Grainger, D.G. Patterson Jr., J. High Res. Chromatogr.23, 208–214 (2000).

Kevin A. Schug is a Full Professor and Shimadzu Distinguished Professor of Analytical Chemistry in the Department of Chemistry & Biochemistry at The University of Texas (UT) at Arlington. He joined the faculty at UT Arlington in 2005 after completing a Ph.D. in Chemistry at Virginia Tech under the direction of Prof. Harold M. McNair and a post-doctoral fellowship at the University of Vienna under Prof. Wolfgang Lindner. Research in the Schug group spans fundamental and applied areas of separation science and mass spectrometry. Schug was named the LCGCEmerging Leader in Chromatography in 2009 and the 2012 American Chemical Society Division of Analytical Chemistry Young Investigator in Separation Science. He is a fellow of both the U.T. Arlington and U.T. System-Wide Academies of Distinguished Teachers.