2019 Volume 25 Issue 6 Pages 879-884
Fatty acid polyglycerol esters are commonly used as emulsifiers in flavor emulsions. In recent years, transparent beverages based on clear flavor emulsions have been developed and are becoming increasingly popular in the market. Here, the quartz-crystal microbalance (QCM) method was applied to investigate the interaction between clear flavor emulsions containing lemon essential oil and lipid bilayers supported on electrodes. Comparative QCM resonance frequency data between clear- and cloudy-type emulsions demonstrated that clear-type emulsions showed smaller frequency changes, suggesting a size-dependent interaction with 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC). Furthermore, as indicated by the frequency changes, the affinity of clear-type emulsions increased with the increasing 1,2-dimyristoyl-sn-glycero-3-phosphatidyl-glycerol (DMPG) to whole lipid composition ratio. This indicated that surface regions of polyglycerol fatty acid ester emulsions may specifically interact with DMPG in mixed DMPC-DMPG systems. Our results may be helpful in understanding flavor expression and retention in cell membranes.
Flavor emulsions are commonly used in the beverage industry (Viela, 2018). Fatty acid polyglycerol esters (PGEs) are used as stable oil-in-water emulsifiers in food and beverages. A PGE is an amphiphilic molecule containing both a hydrophilic and a lipophilic component (Fig. 1). The hydrophilic-lipophilic balance (HLB) in PGEs depends on the length of the polyglycerol and lipophilic chains (Kunieda et al., 1985, 2000). The main advantages of using PGEs in flavor emulsions are their high storage stability and nano-dispersion ability. Further, depending on the HLB, PGE can increase the solubility of a flavor. A specific lamellar phase of highly purified decaglycerol-monolaurate (10G-1L) can emulsify d-limonene over its critical flavor solubility and form transparent flavor emulsions due to nano-dispersion (Nakamura et al., 2003). The emulsions enhance the physical stability for an extended 2-month period (Nakamura et al., 2004). In another example, nano-dispersed emulsions with decaglycerol-monolaurate containing beta-carotene increase the allowable storage period (Tan and Nakajima, 2005). Furthermore, an exquisite combination of arabic gum and PGEs produced a stable flavor emulsion (Paramita et al., 2010). Recently, nano-optical transparent emulsions with ultrafine droplets can be prepared (Piorkowski and McClements, 2014). Transparent beverages based on nanosized flavor emulsions have been a focus of the current market (Valoppi et al., 2017, Dasgupta and Ranjan, 2018).
Structure of flavor emulsions. (a) Structure of PGEs. (R; alkyl chain) (b) Schematic diagram of PGEs (c) The model of an emulsified particle and O/W emulsions. (d) Left: clear flavor emulsions (CLEAR), right: cloudy flavor emulsions (CLOUDY).
Fragrance and flavor molecules bind specifically to mammalian odorant receptors belonging to the G-protein coupled receptor (GPCR) family (Perricone et al., 2013). Winners of the 2004 Nobel Prize for Physiology or Medicine, Linda Buck and Richard Axel, discovered transmembrane odorant receptors (Buck and Axel, 1991). Odorant molecules reach odorant receptors via orthonasal transport through the nose, or retronasal transport through the oral cavity (Ruijschop et al., 2009, Sefton et al., 2011). Retronasal perception, involving the transport of volatile fragrance to receptors on nasal olfactory cells, plays a vital role in taste perception (Goldberg et al., 2018). In the case of flavor emulsions, interaction between cell membranes and emulsions in the oral cavity is triggered by retronasal transport. Hence, advanced emulsions have been designed to enhance flavor richness using properties including body, sharpness, impact, and taste persistence. Although interaction between stable flavor emulsions and cell membranes may be related to flavor richness, molecular level behavior associated with such interaction remains unclear.
Quartz crystal microbalance (QCM) is an in-situ mass sensing tool used to probe various biomolecular interactions (Sota, et al., 2002, Cooper and Singleton, 2007, Kira et al., 2009). QCM is used to investigate the interaction of biomolecules such as opioid peptides, lactoferrin-derived peptides, and tea catechins with membranes immobilized on an electrode (Kira et al., 2014, Tsutsumi et al., 2012, Umeyama et al., 2006, Kamihira et al., 2008). The QCM approach is also used as an odorant and taste-sensing technique. The odorant-sensing system was originally developed as an electronic noise system using QCM with an automatic-sampling stage (Ide et al., 1993 and 1996, Nakamura et al., 2000). Furthermore, a gas-sensing QCM was developed using a calixarene film that was able to capture volatile low molecular weight compounds (Koshets et al., 2005, Özbek et al., 2015). Previous studies have shown that the bitterness intensities of beer and coffee are coupled with changes in frequency due to absorption of beer and coffee components, respectively (Kaneda et al., 2002, 2003 and 2005).
Although studies using the QCM of low molecular compounds contained in foods and beverages have been reported, QCM studies on cell membrane interactions targeting flavor emulsions are scarce. Thus, as expected, this study was challenging. The objective of this study was to clarify the interactions between flavor emulsions and cell membranes; we expected to clarify the effect of the flavor release, as well as the flavor strength by investigating the interaction between CLEAR (nano-scale) flavor emulsions and membranes coated on an electrode via QCM methods.
Flavor emulsions Clear- and cloudy-type flavor emulsions (CLEAR and CLOUDY) contain uniformly nm and sub-nm sized particles, respectively (Fig. 1d). Both emulsions containing lemon essential oils were carefully prepared according to previously described methods (Kokubo and Sakurai, 1999, Nakamura et al., 2004). The CLEAR showed stable dispersion due to a negatively charged electric double layer formed by a soluble hydrophilic moiety of PGEs and a typical surface zeta potential of -23 mV. PGE contents were 5.61% and 3.74% in the CLEAR and CLOUDY systems, respectively. Both emulsions were diluted 250 times using artificial saliva for QCM measurements.
Phospholipids DMPC (1,2-dimyristoyl-sn-glycero-3-phosphocholine) and DMPG (1,2-dimyristoyl-sn-glycero-3-phosphatidyl-glycerol) phospholipids were purchased form Avanti Polar Lipids (Fig. 2).
Structural formulae of (a) DMPC and (b) DMPG.
QCM experiments Changes in the frequency (ΔF, Hz) of QCM measurements were monitored on an Affinix QN (Initium, Japan) with a 27-MHz resonator. The relationship between the changes in mass (Δm, g) and F is based on Sauerbrey's Equ. 1, (Sauerbrey, 1959), as follows:
 |
Experimental QCM sensing image during flavor emulsion-membrane interaction; Flavor emulsions (yellow particles) and membrane of self-assembled phospholipid molecules on Au electrode (deep blue).
The difference in CLOUDY and CLEAR interaction with the DMPC membrane was evaluated by QCM as shown in Fig. 4. Although the CLOUDY PGE content was smaller than that of CLEAR, the frequency change was obviously larger in CLOUDY; the ΔF of CLEAR and CLOUDY at 500 seconds is around -35 and -220 Hz, respectively. (Fig. 4. (a) and (b)). This result suggested that the large sized flavor emulsions interact with the DMPC membrane, and further, that this experimental system recognized size in the emulsion-membrane interaction.
Frequency curves after the injections of (a) CLEAR (red line) and (b) CLOUDY (black line) emulsions to the DMPC lipid bilayers immobilized on the QCM sensor.
Frequency curves on three types of membrane systems obtained immediately following CLEAR injection are shown in Fig. 5. After injection, the frequency systematically decreased for DMPC, DMPC/DMPG = 1:1, and DMPC/DMPG = 1:2 for approximately 30 min. Based on equation (a), the frequency changes were directly proportional to the weight of adsorbed CLEAR emulsions. Furthermore, all the three systems plateaued gradually. We estimated the average ΔF of all three systems (Fig. 6). The average ΔF was negatively larger as a following order; DMPC/DMPG = 1:2 > DMPC/DMPG = 1:1 > DMPC. It is indicated that DMPC may be required for CLEAR emulsion interactions, and that DMPG plays a supportive role in that interaction. Furthermore, our results suggest that the lipid components affect the interaction between the emulsions and cell membrane in the oral cavity.
Typical frequency curves of (a) DMPC (red line), (b) DMPC:DMPG = 1:1 (green line), and (c) DMPC:DMPG = 1:2 (blue line). The immobilized sensor is induced by CLEAR injection (lemon flavor emulsions). The injection point is at 0 sec.
Average frequency (ΔF) graph at 2000 s of CLEAR emulsion interaction with DMPC (red), DMPC:DMPG = 1:1 (green), and DMPC:DMPG = 1:2 (blue). Number of measurements, n = 4. Error bars represent the standard deviation.
QCM detects frequency changes in biomolecular interactions based on the DMPC/DMPG ratio (Kamihira, et al., 2008, Umeyama, et al., 2006). In this study, the frequency curves were based on DMPC and DMPC/DMPG (1:1 and 1:2 ratios), and estimated following electrode injection with flavor emulsions. Using 250 times-diluted emulsions, the change in frequency can be considered an interaction between the emulsion, and the membrane supported on the electrode. CLOUDY-type emulsions showed a larger frequency change, ΔF, suggesting a stronger affinity with the membrane (Fig. 4). When drinking and comparing beverages using emulsified flavors of the CLOUDY- and CLEAR-type, we often feel a difference in the subtle impact of the two flavors. It may be dependent on the rate of retro-nasal aroma. This may be related to the subtle slower retro-nasal delivery of flavor aftertaste by the CLOUDY emulsions in the beverage. Conversely, in beverages containing the CLEAR type flavor emulsions, the frequency change may be responsible for the faster retro-nasal aroma of flavor and refreshing aftertaste. Our result shows the possibility of semi-quantitative evaluation of such taste differences in two types of beverages by changing the frequency of QCM.
The average ΔF results indicated that the CLEAR affinities were DMPC/DMPG = 1:2 > DMPC/DMPG = 1:1 > DMPC (Fig. 5 and 6), but the mode of interaction between nanoparticles and cell membrane surfaces could not be clarified. This was because CLEAR are dispersed emulsions that are not of a low total weight, such as the compounds present in beer (Kaneda et al., 2005). Thus, based on the QCM results, we surmised that the following may apply to the interaction on the interface between emulsions and lipid bilayers: The polar head group of DMPC is neutrally charged, while that of DMPG is negatively charged (Fig. 2). Based on the zeta potential of -23 mV, the surface regions of CFEs are considered to be negatively charged. When the membrane and the particle surface are either neutrally or oppositely charged, common interactions that may lead to adhesion are electrostatic interactions and van der Waals forces. Electrostatic interactions are likely to be formed between the negatively charged particles on the surface of CLEAR and positively charged choline head groups (-CH2-CH2-N+(CH3)3). Since the choline group of DMPC does not function as a hydrogen bond donor or acceptor, the glycerol group of DMPG may act as a hydrogen bond acceptor and donor. The positive charge of DMPC weakens substantially due to the positive charge on the nitrogen atom in the choline group being shielded by three methyl groups (-CH3) (Sum et al., 2003). However, DMPC interacts easily with anions such as Cl− and SO42− according to the Hofmeister series (Hu et al., 2002). Thus, CLEAR may interact with DMPC via electrostatic interaction, and with DMPG via hydrogen bonding.
In addition, membrane viscosity and elasticity, which are related to a diffusion of individual phospholipid molecules and a large-scale deformation of the membrane, may also affect the interaction of the flavor emulsions. The DMPC membrane viscosity and membrane thickness decrease with increasing temperature beyond its gel-to-liquid-crystalline phase transition temperature (23 °C) (Dufrêne and Lee, 2000, Nagao et al. (2017)). When nanoparticles strongly adsorb to the membrane, the membrane deforms to sufficiently increase the contact area to cover the particles (Bahrami et al., 2014, Hamada, 2013). Reportedly, interaction with nanoparticles may also be initiated via elastic deformation of the membrane by simulation technique (Li et al., 2008). Therefore, even in the case of nano-sized CLEAR, adsorbed CLEAR may be stabilized by elastic membrane deformation. As the proportion of DMPG in the DMPC/DMPG membrane increases, the elastic interaction of the membrane may increase. As a result, it appears that the system with DMPG showed a larger frequency change. This interpretation of elastin interaction may be the reason for the reaction slowing to reach a plateau (Fig. 5).
These results indicate one of the most fundamental interactions between PGE flavor emulsions and cell membranes in the oral cavity. Furthermore, changes in affinity due to differences in membrane lipid components may be related to flavor strength and retention in the oral cavity when considering the release of retronasal-like flavors, because the emulsion-membrane interaction may be deeply associated with emulsion retention and flavor release.
With regards to the unclear membrane–flavor emulsion interaction, we showed that one of the fundamental interactions related to flavor expression and release using QCM measurements. Namely, a combination of electrostatic interaction, van der Waals forces, and elastic membrane interactions may be related to the affinity strength between clear-type emulsions and membranes. Our results provide insight into the expression of flavor, and be helpful in designing the next generation of emulsified beverages. Our results suggest the need for further study of flavor emulsions in the view of molecular and macroscopic levels to better understand the flavor expression mechanism.
The interaction of flavor emulsions with the membrane was investigated via QCM. Frequency changes differed between CLEAR and CLOUDY-type emulsions, and the difference was considered to be size-dependent. CLEAR had a certain affinity with DMPC, and the affinity increased when the proportion of DMPG in the DMPC/DMPG membrane increased. The fundamental interactions that occur between such cell membranes and nano-sized emulsified perfumes will improve our understanding of flavor expression within the oral cavity.
Acknowledgements This work was partly supported by Industry-University Collaboration Research Start-Up Grants from Yokohama National University to I. K.
