How sweet! Right, call me miraculin

Miracle berries from Richadella dulcifica plant. Credit: SITS Girls (Flickr)
Miracle berries from Richadella dulcifica. Credit: SITS Girls (Flickr)

We perceive the world surrounding us through our senses. Thanks to them we can have a pleasant time listening to Claude Debussy or get delighted by the scent of rosemary. We blindly trust them, but are they always truthful? A good answer would be they are accurate most of the time, although it seems nature has found funny ways to cheat us by misleading them.

An interesting example of the former comes from a protein with a quite unique feature, known as miraculin. Found within the red berries from the Richadella dulcifica (a shrub native to Africa also known as miracle fruit), it has a taste-modifying activity which transforms sour taste to sweet one. In essence, chewing one of these berries makes lemon taste like orange and the effect lasts for about an hour after consumption. There are other proteins, like monellin and thaumatin known to be intensely sweet, but none of these have such a transformative effect in other tastes1.

African natives have known about the berries for centuries, but it was in 1968 when it was formally described by Kenzo Kurihara and Lloyd M. Beidler in the Science journal2. Miraculin is a glycosylated protein consisting of 191 amino acids which must form a dimer (two molecules covalently bound) to show its taste-modifying activity at acidic pH. Furthermore, it is produced six weeks after pollination and its activity is lost two hours after harvesting the fruit, thus imposing obvious limitations to its use (1).

Although miraculin was first described more than 40 years ago, little work has been done in order to elucidate its mode of action at the molecular level. It is known that its effect is mediated through the G-protein coupled receptors hT1R2+hT1R3 (which form a heterodimer), expressed in sensory organs known as taste buds. This set of specialized cells are mainly found in our tongue, allowing us to discriminate a range of basic flavors which are fine-tuned by combining taste-related information processed in other sensory cells (including olfactory cells and trigeminal primary afferents). Those heterodimers are responsible for the sweet taste modality as opposed, for instance, to detection of umami taste, mediated by hT1R1+hT1R3. Different combination of receptors leads to the perception of different taste modalities.

To characterize the molecular mechanism by which miraculin transforms sour to sweet taste, a group of Japanese scientist under the leadership of Keiko Abe published in 2011 a cell-based assay in which a widely used cell line (HEK293T) expressing the receptors was treated with miraculin 3. Once the protein was washed out, the scientist reduced the pH of the extracellular solution bathing the cells, thus mimicking the acidic environment the taste buds encounter after the berries are eaten by the individual. In addition, they used a Ca2+-sensitive molecule which allow scientist to monitor the activation of the heterodimers. Briefly, when the receptors are activated, extracellular Ca2+ flows into the cells and this entry leads to the appearance of a cytosolic fluorescent signal that can be measured with a microscope and an appropriate software. When they pre-incubated the cells with as less as 100nM miraculin and thereafter applied an acidic buffer to the solution, they could see a consistent and reproducible fluorescent signal which was absent in cells not expressing the sweet-detecting heterodimers.

The authors suggest that hT1R2+hT1R3 activations occurs through a conformational change of the extracellularly bound miraculin which is promoted by the protonation of two key Histidines (His) present in the protein. Mutation of these two His into Alanine amino acids (much smaller and unable to be protonated) abolishes the acidic-mediated activation. In summary, miraculin is an agonist of the sweet receptors when the pH is reduced. Perhaps more remarkable is the observation that pre-incubation of the same cells with miraculin at neutral pH antagonizes the effect of additional sweet substances in a dose-dependent fashion, presumably because the affinity of the protein for the receptor is higher than that of the other molecules. Thus, miraculin can be both an agonist and an antagonist of hT1R2+hT1R3 depending on the pH (3).

Despite the exciting properties of miraculin, it is somehow surprising that scientist’s curiosity has only grown in recent years (55 out of 90 peer reviewed articles published after 2005). More importantly, sweet proteins as a whole are currently the focus of intense research. We could arguably link this growing interest to the ever increasing incidence of diabetes seen in many countries. Only in the UK, the NHS spends an estimated 14 billion per year on treating diabetes (mainly type II) and its associated complications. Quality of life in people with diabetes is heavily compromised, so excessive carbohydrate intake should be avoided (with emphasis on highly processed ones). Still, this is difficult to achieve especially in a society always short of time (no time for cooking) with predilection for sugary meals. Therefore, alternative ways to satisfy people needs without compromising their health should be found.

Replacement of sucrose and other processed carbohydrates with sweet proteins could alleviate the problem as they do not trigger insulin secretion, making more difficult to develop resistance and therefore, diabetes. A study from 2006 in rats even showed that miraculin led to improved insulin sensitivity 4. In addition, the calories contained on the miraculin needed to elicit sweetness is negligible, thus not contributing to obesity. Finally, another interesting application would be as a flavor enhancer for patients under chemotherapy. They often suffer from reduced appetite as well as severe taste alterations which could be improved with the magic berry. Few clinical studies have shown promising results, although with small sample sizes 5.

An additional factor contributing to the lack of use seems to be the difficulties in handling the protein. The bush only grows under specific conditions and does not produce many berries. Moreover, the protein is prone to activity loss although the berries can be stored for three months at -20ºC maintaining miraculin intact. To overcome these limitations, genetic engineered alternatives are being developed in order to produce miraculin, specially in Japan. The idea is to produce massively the protein, while improving the stability profile too. The best candidate seems to be the tomato for a number of reasons. Lettuce and strawberries have also been explored as candidates, but gene silencing problems in the first, and insufficient production in the second makes them unlikely options 6. Future will tell, but it looks like the story of miraculin has just begun.


  1. Misaka T. Molecular mechanisms of the action of miraculin, a taste-modifying protein. Semin Cell Dev Biol. 2013 Mar;24(3):222-5. doi: 10.1016/j.semcdb.2013.02.008.
  2. Kurihara K, Beidler LM. Taste-modifying protein from miracle fruit. Science. 1968 Sep 20;161(3847):1241-3. doi: 10.1126/science.161.3847.1241
  3. Koizumi A, Tsuchiya A, Nakajima K, Ito K, Terada T, Shimizu-Ibuka A, Briand L, Asakura T, Misaka T, Abe K. Human sweet taste receptor mediates acid-induced sweetness of miraculin. Proc Natl Acad Sci U S A. 2011 Oct 4;108(40):16819-24. doi: 10.1073/pnas.1016644108.
  4. Chen CC, Liu IM, Cheng JT. Improvement of insulin resistance by miracle fruit (Synsepalum dulcificum) in fructose-rich chow-fed rats. Phytother Res. 2006 Nov;20(11):987-92. doi: 10.1002/ptr.1919
  5. Thorne T, Olson K, Wismer W. A state-of-the-art review of the management and treatment of taste and smell alterations in adult oncology patients. Support Care Cancer. 2015 Sep;23(9):2843-51. doi: 10.1007/s00520-015-2827-1.
  6. Hiwasa-Tanase K, Hirai T, Kato K, Duhita N, Ezura H. From miracle fruit to transgenic tomato: mass production of the taste-modifying protein miraculin in transgenic plants. Plant Cell Rep. 2012 Mar;31(3):513-25. doi: 10.1007/s00299-011-1197-5.

Written by

Leave a Reply

Your email address will not be published.Required fields are marked *