Article Preview
TopIntroduction
Sweetness is one of the most important tastes of mankind and has long been sought after as a dietary constituent to produce a pleasant sensation. Since sucrose is the most commonly used sweetener, it is the standard substance employed for measuring and comparing the Relative Sweetness (RS) of other sweet substances (Singh, Khan, & Singh, 2014). RS is defined as the ratio of a standard sucrose concentration to the iso-sweet concentration of another sweetener (Bassoli et al., 2001). In other words, a standard solution of sucrose has a sweetness perception rating of 1 (or 100), and the sweetness of whatever sweetener is being evaluated is rated relative to sucrose.
The development and search for new sweeteners is complicated. On one hand, there are multiple factors that affect sweetness; e.g., solubility, stability at wide pH and temperature ranges, clean sweet taste without post-flavor effects, sweetening effect as compared to low-cost sucrose, and finally, the most important factor is the safety of human health (Belitz, Grosch, & Schieberle, 2009). On the other hand, the measurement of RS involves a high cost due to the use of trained panels for the use of the “sip and spit” taste method for comparing the RS between a sweetener and the standard. Because of all these factors, there is a clear advantage to develop quantitative-structure models in order to understand the sweetness mechanism and to use these models to develop and synthesize new potent sweeteners (Yang, Chong, Yan, & Chen, 2011; Zhong, Chong, Nie, Yan, & Yuan, 2013).
The QSAR/QSPR theory suggests that the biological activities and properties of molecules can be determined in terms of specific molecular attributes, i.e., these chemical structures can be represented theoretically although they do not offer specific details of the usually complex mechanism/path of action involved. However, the molecular structure is quantified by using a set of suitable molecular descriptors, which are numbers carrying information on the constitutional, topological, geometrical, hydrophobic, and/or electronic aspects of the chemical structure (Diudea, 2001; A. R. Katritzky, Lobanov, V. S., Karelson, M., 1995; Todeschini & Consonni, 2009; Trinajstic, 1992). Hence, the goal is to build a useful model selecting only the most representative descriptors among thousands of them. If successful, such a model can be useful to investigate and understand specific characteristics of the relationship between the molecular structure and its activity/property that is under investigation.
The first Quantitative Structure-Activity Relationship (QSAR) study of sweeteners was carried out by Edna W. Deutsch and Corwin Hansch (Deutsch & Hansch, 1966) in which they used some derivatives of 2-amino-4-nitrobenzene to model RS. They concluded that this property is well-correlated with the hydrophobicity and the Hammett constant descriptors given by the substituents. Since this pioneering study, there were several applications of the QSAR/QSPR theory to model and predict the RS for diverse sweeteners(Arnoldi, Bassoli, Merlini, & Ragg, 1991; Barker, Hattotuwagama, & Drew, 2002; Bassoli et al., 2001; Bassoli, Drew, Merlini, & Morini, 2002; Drew et al., 1998; Hansch, 1970; Iwamura, 1980, 1981; Katritzky et al., 2002; Pietrzycki, 2001; Rao & Kumar, 1986; Spillane, 1983; Spillane & McGlinchey, 1981; Spillane, McGlinchey, Muircheartaigh, & Benson, 1983; Spillane et al., 2000; William J. Spillane et al., 1996; Spillane & Sheahan, 1989; van der Heijden, Brussel, & Peer, 1979;Walters, 2002; Walters, 2006; Walters & Hinds, 1994).