Rapid measurement of phytosterols in fortified food using gas chromatography with flame ionization detection

Highlights

• A new rapid method to quantify phytosterols in fortified food was developed.

• Analysis is based on gas chromatography with flame ionization detection.

• The method is more economical, time and labour efficient than current methods.

• Critical aspects of method validation and uncertainty are reported.

• This method can ensure accurate plant sterol fortification for the food industry.

Abstract

A novel method for the measurement of total phytosterols in fortified food was developed and tested using gas chromatography with flame ionization detection. Unlike existing methods, this technique is capable of simultaneously extracting sterols during saponification thus significantly reducing extraction time and cost. The rapid method is suitable for sterol determination in a range of complex fortified foods including milk, cheese, fat spreads, oils and meat. The main enhancements of this new method include accuracy and precision, robustness, cost effectiveness and labour/time efficiencies. To achieve these advantages, quantification and the critical aspects of saponification were investigated and optimised. The final method demonstrated spiked recoveries in multiple matrices at 85–110% with a relative standard deviation of 1.9% and measurement uncertainty value of 10%.

Introduction

In the plant world, phytosterols are the equivalent to cholesterol in animals and humans. Plant sterols belong to the triterpene family and can be found as free, steryl glycosides (SG), steryl esters (SE), hydroxycinnamic acid ester (HSE) and acylated steryl glycosides (ASG), with the latter four commonly known as phytosterol conjugates (Dutta, 2004, Moreau et al., 2002). Phytosterols are believed to be an integral component of plant structural membranes (Dutta, 2004, Moreau et al., 2002) with most phytosterols comprised of a 28–29 carbon ring based structure with a hydroxyl group at the Δ-3 position and a double bond at the Δ-5 position (Dutta, 2004, Moreau et al., 2002). The structure of a generic sterol molecule is shown in Fig. 1 and structures of some of the common phytosterol structures are presented in the Supplementary Material.

Food sources naturally rich in plant sterols include a wide range of cereals, fruits, vegetables and plant derived oils (Han et al., 2008, Moreau et al., 2002). In addition to natural foods, some processed foods are fortified with phytosterols, usually with the steryl esters as they are easily incorporated into the fat component of the product. Phytosterols found in these fortified food matrices include β-sitosterol, campesterol, stigmasterol, brassicasterol and stigmastanol (see structures in Supplementary Material). A range of processed food products are commonly fortified with sterols and these include dairy products, fat spreads, chocolates, snack bars and salad dressings.

In the last decade there has been a dramatic increase in public awareness and, consequently, in the consumption of phytosterols due to their demonstrated health benefits. Several reports have shown a direct correlation between phytosterol ingestion and the reduction of low density lipoprotein (LDL) cholesterol (Anon, 2005, Katan et al., 2003, Ostlund, 2002). The optimal and recommended steryl ester dosage to provide approximately a 10% reduction in LDL cholesterol is 2 g/day (Kritchevsky & Chen 2005). Higher dosages have been shown to offer minimal additional reduction (Katan et al., 2003; Ostlund 2002). As of 2002, the United States Food and Drug Administration (USFDA) has permitted health claims to be published on any food products containing plant steryl or stanyl esters (Anon 2005; Moreau et al., 2002).

To support the food industry and ensure fortification claims on nutritional labelling are correct, robust analytical techniques for the routine determination of fortified phytosterols in food are required (Chen et al., 2015, Mo et al., 2013, Revathi et al., 2013, Saha et al., 2014, Srigley and Haile, 2015). Common analytical procedures for phytosterol determination usually consist of an alkaline saponification mixture and conditions utilising potassium hydroxide or sodium hydroxide at concentrations ranging from 1 to 6 M (Lagarda et al., 2006, Liu, 2007, Moreau et al., 2002). This is typically followed by an organic solvent extraction with many reported studies showing success using hexane, heptane, toluene, and petroleum ether (Lagarda et al., 2006, Liu, 2007, Moreau et al., 2002). The main benefits of using these organic solvents is their opposing polarity to the aqueous saponification mixture, facilitating the extraction of the sterols which are more soluble in organic solvents than water (Dutta 2004). The aqueous phase of the saponification mixture will solubilise the cleaved fatty acid ligands (in salt form) allowing free extraction of the sterols into the organic solvent. This process should eliminate non-targeted compounds that are insoluble in the organic solvent from entering the extract solution (Du, 2002, Toivo et al., 2000). The selection of the organic solvent will be influenced by several factors including its affinity to the target compounds, low hydrophilicity, availability and safety (Du, 2002, Toivo et al., 2000).

In theory, organic compounds of similar structure and molecular weight extractable by the saponification/solvent extraction technique may interfere with the sterol quantitation. Compounds such as tocopherols, tocotrienols, retinol and β-carotene may be expected to interfere, however, in practice the levels of these fat-soluble vitamins are very low when compared to the sterol levels thus rendering any effect to the quantitation less than the statistical uncertainty. In addition, the chromatographic method will typically provide sufficient separation from the target sterols so as to negate this anticipated interference (Du, 2002, Dutta, 2004).

Organic solvent extraction is then followed by derivatisation and analysis by gas chromatography with a flame ionization detector (GC-FID) (Anon, 2005, Clement et al., 2010, Lagarda et al., 2006, Moreau et al., 2002). This has been found to be an effective technique for the analysis of most processed foods, in particular those fortified with high levels of steryl esters in the range of 300–8000 mg/100 g. However, the main drawback of these methods is that the extraction procedure is specific for steryl esters and free sterols only. Furthermore, this method is labour intensive and time consuming and therefore ill-suited as a routine procedure. A typical procedure for a batch of 10 samples including a quality assurance (QA) step can take up to 5 h to complete. In addition, other reported phytosterol analysis techniques include the use of GC-mass spectroscopy (MS) and liquid chromatography coupled with MS, photo diode array or evaporative light scattering detectors (Ahmida et al., 2006, Joseph, 2012, Raith et al., 2005, Soupas et al., 2004). A common challenge for sterol analyses is the co-elution of target compounds that have similar column affinity. The FID is non-discriminatory such that compound identification is limited to a referenced retention time. Confirmation can be achieved by either analysing the same extract by a different column stationary phase or by using other techniques such as GC-MS. This method, although less efficient for quantitation, employs the compound retention time in conjunction with the mass-spectrum to characterise the compound (Skoog et al., 1998).

Previous work on the analysis of phytosterols using GC-FID procedures has predominately used 3 main surrogates for quantification, namely betulin, 5α-cholestane and 5β-cholestan-3α-ol. Despite their structural similarities, some reports have proposed the preferential use of the latter compound. It is suggested that 5β-cholestan-3α-ol which contains a hydroxyl group, is structurally more similar to the target sterols and would therefore provide a better emulation of the process during extraction (Dutta, 2004, Katan et al., 2003, Moreau et al., 2002). In this paper, a rapid, accurate and robust method for the measurement of phytosterol esters in a range of fortified food is presented. Emphasis was given to high throughput efficiency while minimising labour and reagent costs to ensure effective implementation in a commercial laboratory.

The term surrogate standard refers to the use of a similar compound to the target analyte that is added at the beginning of the extraction process. A known amount of surrogate is added to the sample at the beginning of the anlaysis to enable the evaluation of the analyte during extraction (Crosby, Prichard, & Newman 1995).

An internal standard is a compound (not necessary similar to the target compound) that is added before the instrument analysis for the purpose or instrumentation monitoring (Crosby et al., 1995).