A Hypersweet Protein: Removal of The Specific Negative Charge at Asp21 Enhances Thaumatin Sweetness

Abstract

Thaumatin is an intensely sweet-tasting protein that elicits sweet taste at a concentration of 50 nM, a value 100,000 times larger than that of sucrose on a molar basis. Here we attempted to produce a protein with enhanced sweetness by removing negative charges on the interacting side of thaumatin with the taste receptor. We obtained a D21N mutant which, with a threshold value 31 nM is much sweeter than wild type thaumatin and, together with the Y65R mutant of single chain monellin, one of the two sweetest proteins known so far. The complex model between the T1R2-T1R3 sweet receptor and thaumatin, derived from tethered docking in the framework of the wedge model, confirmed that each of the positively charged residues critical for sweetness is close to a receptor residue of opposite charge to yield optimal electrostatic interaction. Furthermore, the distance between D21 and its possible counterpart D433 (located on the T1R2 protomer of the receptor) is safely large to avoid electrostatic repulsion but, at the same time, amenable to a closer approach if D21 is mutated into the corresponding asparagine. These findings clearly confirm the importance of electrostatic potentials in the interaction of thaumatin with the sweet receptor.

Sweeter and stronger: enhancing sweetness and stability of the single chain monellin MNEI through molecular design

“…Sweetness threshold was evaluated by triangle test as previously described 23 24 . MNEI solutions were used as positive control and water as negative control.…”

“…Since complexes with sweet proteins have never been experimentally resolved, the wedge model has been built using a homology model of the receptor, based on the structure of the metabotropic glutamate receptor mGluR1 22 . Nonetheless, building on this model, it has been possible to rationalise the effects of point mutations affecting the potency of monellin, brazzein and thaumatin 23 24 25 . The widely accepted idea is that both proper surface charge distribution and three-dimensional shape have to be maintained in order to trigger the sweet sensation 23 25 26 27 28 .…”

“…The other factor that most significantly correlates with sweetness is surface charge: in fact, the surface of the T1R2-T1R3 complex that is described to bind sweet proteins is characterised by the presence of a large amount of acidic amino acids 17 21 . Studies on single and double chain monellins 23 28 39 40 , thaumatin 24 41 42 43 , brazzein 44 45 46 and lysozyme 47 have shown that, in general, mutations increasing the acidic character would consistently decrease or even cancel sweetness, whereas the outcome of the introduction of a positive charge is not immediately predictable. For instance, among four surface mutations, namely M42R, Y63R, Y65R and D68R, only Y65R would increase sweetness, whereas the other mutations, despite introducing a positive charge, would abate the taste intensity 23 .…”

“…These models have been able to account for many of the experimental outcomes of mutations of charged surface residues, and proved the possibility to predict at the atomic detail the complexes of mutants of MNEI and brazzein with the sweet receptor. More recently, a similar approach has been used to design and validate a new super-sweet mutant of thaumatin 24 . In the present study, we have started from the sweeter variant Y65R-MNEI 23 and have introduced new, rationally predicted mutations to further potentiate sweetness.…”

Positive Charges on the Surface of Thaumatin Are Crucial for the Multi-Point Interaction with the Sweet Receptor

“…Since sweet-tasting proteins are useful as low-calorie sweeteners as well as a substitute for sucrose, a number of investigations have been performed to clarify why sweet-tasting proteins elicit a sweet taste, and how sweet-tasting proteins interact with sweet receptor (Kohmura et al, 1992 ; Kim and Weickmann, 1994 ; Slootstra et al, 1995 ; Somoza et al, 1995 ; Assadi-Porter et al, 2000 , 2010a , b ; Kaneko and Kitabatake, 2001 ; Temussi, 2002 , 2006 , 2011 ; Jin et al, 2003 ; Jiang et al, 2004 ; Tancredi et al, 2004 ; Masuda et al, 2005a , b , 2013 , 2016 ; Esposito et al, 2006 ; Walters and Hellekant, 2006 ; Koizumi et al, 2007 , 2015 ; Kurimoto et al, 2007 ; Li et al, 2008 ; Nakajima et al, 2008 , 2011 ; Ohta et al, 2008 , 2011a , b ; Ide et al, 2009 ; Walters et al, 2009 ; Xue et al, 2009 ; Dittli et al, 2011 ; Templeton et al, 2011 ; Liu et al, 2012 , 2016 ; Cornilescu et al, 2013 ; Lee et al, 2013 ; Rega et al, 2015 ; Ghanavatian et al, 2016 ; Leone et al, 2016 ; Lim et al, 2016 ; Singarapu et al, 2016 ).…”