The characterisation of Mozzarella cheese microstructure using high resolution synchrotron transmission and ATR-FTIR microspectroscopy

Highlights

• S-FTIR microspectroscopy successfully characterised Mozzarella cheese microstructure.

• Transmission and ATR showed similar spectral features for proteins and lipids.

• Minor structural differences were detected across the sampled area by PCA.

• Differences due to imaging mode or sample preparation were also identified through PCA.

• This method may be used for label free analysis of other dairy products.

Abstract

Synchrotron Fourier transform infrared (S-FTIR) microspectroscopy allows the label-free examination of material microstructure but has not been widely applied to dairy products. Here, S-FTIR microspectroscopy was applied to observe the microstructure of Mozzarella cheese and assess the protein and lipid distribution within individual cheese blocks. High lipid and high protein areas were identified in transmission and attenuated total reflectance (ATR) analysis modes and the secondary structures of cheese proteins determined. Hierarchical cluster analysis and principal component analysis identified variation in random coil, water content, lipid carbonyl and methylene stretching across the sampled area. Similar spectral features were obtained in both analysis modes; spatial resolution was higher with ATR and small differences were noted, potentially as a result of differences in sample preparation. S-FTIR is a useful microscopy tool that can detect structural alterations that may affect product properties and may assist reverse engineering of a range of dairy products.

Introduction

Fourier transform infrared (FTIR) spectroscopy is a well-established technique that is widely used in the dairy industry to measure gross composition including lipid, protein and moisture (Agnet, 1998) and has been used to differentiate dairy products, such as butter, from different geographic regions (Bassbasi, De Luca, Ioele, Oussama, & Ragno, 2014). When combined with microscopy, in a technique known as FTIR microspectroscopy, this provides insights into the distribution of major chemical components across the sample (Baker et al., 2014). This method has been used successfully to characterise and track chemical and structural changes within a number of different foods, such as meat (Kirschner, Ofstad, Skarpeid, Høst, & Kohler, 2004), fish (Bocker, Kohler, Aursand, & Ofstad, 2008) and films composed of gelatin and starch (De Giacomo, Cesàro, & Quaroni, 2008), highlighting the potential of this technique for the spatial mapping of components within a broader range of foods, including dairy products.

To date, FTIR microspectroscopy has only had limited application in cheese products, where it has been used to map the spatial distribution of protein, lipid and starch in imitation cheese (Noronha et al., 2008). A limitation of this technique is the lower lateral resolution that can be achieved when compared to other commonly applied advanced microscopy techniques, including confocal laser scanning microscopy (CLSM), with a diffraction-limited spatial resolution between 3 μm and 5 μm in the mid-infrared spectral region typically used for chemical analysis. Steps that can be taken to improve the signal to noise ratio of the data obtained at this resolution limit include the use of a synchrotron source, which provides a highly focussed bright infrared light source, preventing energy loss when small apertures are used for high resolution mapping (Wetzel, Sweat, & Panzer, 1998). This concept has been demonstrated to study food grains, where synchrotron FTIR (S-FTIR) was successfully applied to differentiate protein and starch structures within individual grains (Bonwell, Fisher, Fritz, & Wetzel, 2008). When data collected using a synchrotron IR source are combined with multivariate data analysis there is potential to improve the quality and quantitative interpretation of S-FTIR microspectroscopy data. Such multivariate data analysis has previously been applied to discriminate between varying concentrations of a mixture of porcine and bovine gelatin, where small differences in composition and structure could be identified (Cebi, Durak, Toker, Sagdic, & Arici, 2016). A synchrotron source has not previously been applied to dairy products nor combined with multivariate data analysis for such materials, despite the potential of this approach.

Some recent dairy studies have employed a related vibrational technique, confocal Raman microscopy, to examine the spatial distribution of key components in processed cheese using cryo-microtomed sections (Smith, Holroyd, Reid, & Gordon, 2017). Detailed information on protein structures could not be determined, however, as the Raman bands are typically stronger for non-polar bonds (Wellner, 2013), limiting the study of proteins. In contrast, FTIR can provide information on protein secondary structure (Barth, 2007). For this reason, Raman and FTIR microscopy techniques are often employed together, in order to take advantage of the complementary information provided by each technique. FTIR microspectroscopy can also be applied in different modes of analysis, including transmission, attenuated total reflectance (ATR) and reflectance, whereas the majority of analysis in Raman microscopy is conducted in reflectance mode, providing information on only the surface components of the food product.

The present study aimed to assess the suitability of S-FTIR microspectroscopy using a synchrotron based FTIR for the analysis of dairy products, with a specific focus on Mozzarella cheese. Mozzarella cheese and other pasta filata type cheeses account for the third largest volume of cheese produced in Australia, making the study of these cheeses relevant to industry. Two different sampling modes, transmission and ATR-FTIR, were applied to obtain spatially-resolved chemical images of the cheese microstructure.