Masking bitterness in food and beverages
Amongst the five tastes that we can perceive, bitterness is the least appreciated. Many toxic compounds have a bitter taste explaining why humans developed a high sensitivity to it. Yet, if many toxics are bitter, that doesn’t mean that all bitter foods are to be avoided. In a positive or negative way, bitter tasting substances are often bioactive. Nutraceuticals and healthy food often include bitter nutrients demonstrating health benefit effects.
There is a spread of sensitivity to this taste within the population. If you don’t like eating vegetables, part of the explanation may lie in the bitter-tasting flavonoids they contain. If food presents or develops a bitter taste, it will impact its acceptability. Creating and distributing a product poorly accepted undoubtedly results in economic losses. Repeating buyers rarely prioritize health benefits over taste. It seems wise to put in place ways to estimate a product's taste, and especially bitterness, using sensory panels or lab equipment. If we can measure it, we can better understand the potential market desirability. Using sensory panels is not always an option for reliability issues, health, and safety problems, or economic reasons. The electronic tongue is an alternative option. This analytical device can give an objective number on the taste of any food sample. It can also be used to test taste-masking strategies which is a major advantage. You can indeed use the e-tongue to mask, hide, and sometimes get rid of the undesired bitter taste early in the development of the product. It enables you to quantify the change in intensity of the different tastes and compare different recipes to get the best results. If well exploited, it can speed up formulation time, and save on research and development costs. This post will explain more in detail the principles of bitterness masking and why the electronic tongue can be a great supporting tool.
Grapefruit, coffee and celery are examples of bitter foods
Evaluating food bitterness
As bitterness is the most unwanted taste, many research studies exist to understand how it can be measured, avoided, or modified. Traditional analytical methods involve the detection in a food sample of all the molecules known as having a bitter taste. Those methods are chromatographic methods. They allow the identification and quantification of different molecules. Once all the information on all the molecules is available, we could analyze them all together to get a prediction of bitterness. Yet, it doesn’t always work. First, it implies to know the taste of all the potential existing bitter molecules that can be in the sample. Even then, they are all perceived at different thresholds (minimal concentrations needed to start tasting something) and can interact one with the other in many ways. Bitter compounds also interact with other constituents, some of which may enhance or suppress the taste perception. It is unlikely that for complex food with thousands of different molecules we can have access to the totality of the taste information for all of them. Also, bitterness is often a taste and an aftertaste. This dynamic information increases the level of complexity. The taste sensors on the electronic tongue, however, give an overview of the tastes and aftertastes generated by all the atoms and molecules in the sample. So yes, you won’t know what molecule is the reason for which taste. But if you are dealing with a specific taste, it may not be what matters. The electronic tongue, thanks to its highly selective taste sensors can analyze the bitterness resulting from all the compounds and their interactions. Also, it simulates a swallowing action and gives an evaluation of the aftertastes.
When it comes to taste analysis, this technology is still new. Yet, it has already been extensively studied to ensure its ability for evaluating bitterness. Most applications are in the pharmaceutical industry for measuring the bitter taste of drugs (antihistamines, quinine hydrochloride, ibuprofen, and different antibiotics), and to develop masking strategies for these. Given the importance of bitterness in food and beverages, application examples are also available. The electronic tongue has been used for research on the bitterness of beer, wine, olive oil, tonic water, fruit juices, coffee, tea, plant based foods, and meat products. Several studies on beers have shown that the electronic tongue is an efficient tool to predict the bitter taste of different beer samples. The following “taste map” is a great example of how bitterness data - here used alongside sourness data - can be represented [1]. It represents beers from around the world, evaluated with the e-tongue, and plotted based on the measured intensity of those two tastes.
For the case of beers, there is an internationally recognized way to estimate the bitterness. This unit is called IBU (international bitterness unit) and correlates well with data obtained with the e-tongue. In most cases, the electronic tongue is a promising method to get food or beverage bitterness without relying on actual tasting. This is especially important when it comes to tasting alcoholic beverages because of related ethical, legal, or religious restrictions. As well, it helps avoid problems linked with the diversity of people's preferences, physiological ability to taste bitterness (some people are not capable of perceiving it at all!), and the inconsistency in the score they can give [2].
Developing a bitterness masked formulation using the electronic tongue
The electronic tongue is constituted with two main parts. The autosampler with the taste sensors does the actual measurements and the computer part that translates this information into understandable taste data. If we look especially at the electronic tongue developed by Insent they offer a range of taste sensors depending on the application. Insent offers a set of taste sensors, originally designed for pharmaceuticals evaluation and optimized for bitterness perception. They realized series of studies to make sure that the results obtained from the e-tongue were comparable to the ones obtained with humans. For some pharmaceuticals evaluated and represented on the graph below, the correlation is very strong. This means that the results obtained with the electronic tongue are a good indicator of the bitterness that people taste.
Given the diversity of existing bitter compounds, it is unsure that all the taste sensors available for bitterness evaluation can give valuable information. When starting to look at a food product or a bitter nutraceutical, we need to make sure that we know how to best measure it. If there is some data already available then we just need to follow the protocol. Otherwise, there is a way to define what sensors are the best, what product concentration to use and under which protocol. When we know the best method to measure the taste of the bitter product with the e-tongue, it becomes easy to start evaluating taste-making strategies. For this, you need to carefully choose what masking material is best for your product. Once done, you also need to consider what are the best concentrations to assess and what are the other ingredients of the formula that can change the results. The following graph shows how the taste intensity of a solution can decrease when increasing the concentration of the masking agent. This masking effect corresponds to the loss in taste intensity obtained when adding the taste modifier. By measuring the concentration over a range, we can then optimize the concentration to use. This helps to get the best results while keeping in consideration the other ingredients as well as cost or supply related problematics.
There are some exceptions where using the e-tongue will, per design, not give satisfactory results. Indeed, the taste sensors have some limitations. Yet, as long as we use it in its operative range, the electronic tongue has some strong advantages:
A good prediction of sensory bitterness scores. The e-tongue delivers data you can trust and are representative of people’s perception
Rapid screening of a range of masking ingredients, allowing testing of many masking strategies
Easy evaluation of masking performances with taste intensity modification. It gives actual numbers which are great tools for formulation and communication
Short time and minimal cost to gather a large quantity of data
Opportunity to optimize formulation using mathematical methods and experimental design engineering rather than countless trials and errors
Bitterness masking in food and drinks and the ingredient used
Being able to measure the taste is the first step as a preventive or corrective action to ensure product quality. Preventive to predict if the taste profile of the product should correspond to what the customer is looking for. Corrective to understand and react to customer feedback or complaints. If a taste needs improvement, having data to start from is essential. Yet, correcting taste can’t be done by any possible means. The bitter compound may have a functional effect on a specific part of the body. In this case, we need to ensure that the rest of the food or beverage containing this compound doesn’t interfere. This ‘rest of food’ can be seen as the ‘delivery system’ for this bioactive compound. There are 4 principles this delivery system needs to meet besides its taste improvement properties. It should :
Be able to contain enough of the compound
Protect the compound from degradation (during production, storage, transport, etc)
Be able to release the compound in the body, where it should be bioactive
Be compatible with the compound, to avoid, change in texture, colored dots or poor mixing with the other ingredients
If we focus on bitterness masking, there is a range of molecules that are already used and have proven their effectiveness.
Cyclodextrins
Cyclodextrins are cyclic molecules used as a cage that can surround other smaller compounds in their center. Their chemical structure is derived from sugars and obtained from natural origins. They have a sweetness level of about ⅓ of the sugar sweetness. Widely accepted as food additives (also known under the number E459 in the ingredient list), they find many applications. For example, they can trap aromas so they last longer in the food, offer protection against the environment (rancidity, light, heat), increase food preservation, or even sequestrate cholesterol. Of course, they can also remove or mask bitterness or other disgusting tastes and odors. Some foods have a very peculiar smell that can be trapped using cyclodextrins, like soybean milk or some fishy odors. Aiming at taste modification, they can for example be used to mask bitterness from navel oranges, grapefruit juices, or caffeine.
Proteins and other polymers
A polymer is a large molecule that is constituted by a repetition of smaller molecules. They can be found in nature or synthetic. A common example of a polymer that can be found in natural sources is a protein. Proteins are an assembly of smaller units called an amino acid. Polysaccharides, like starch or cellulose, are other polymers, made of individual sugar linked to each other in a long chain. There are many other types of natural and synthetic polymers used, in particular in the pharmaceutical industry. For food applications, only natural ones, extracted from plants, animals or from microbial origins are allowed. There is a strong diversity in polymer size and chemical properties. Some carefully chosen have the potential to interact with smaller molecules we want to target. Depending on the strength of the binding, this can mask the taste of a molecule but also be a controlled delivery system.
Caseinate is a protein very present in milk. Drying caseinates and a bitter compound together can give a complexation between both. This complexation can be done in such a way that the bitterness is masked. This solution might not be the best for all foods as this complex can disassemble when putting in water. But in a dryer food, like a protein bar, for example, it is an interesting application. The complex would not have the time required to break apart in the mouth, and the protein bar can give both proteins and an active compound with hidden bitterness. Tannins present in great quantities in wine and teas are very bitter and astringent antioxidants but bind easily to certain proteins. Exploring the interaction between tannins and proteins is also an interesting strategy for bitterness making.
Emulsifiers
Emulsifiers are small molecules that have two parts. One side has more tendency to interact with oil while the other one has more tendency to interact with water. When added to a blend of water and oily elements (a salad dressing for example), they position themselves at the limit between the two surfaces. By doing so they also make the mixing of those two parts easier. Oil doesn’t mix with water. But adding an emulsifier makes it doable. For this unique property, they are used a lot in foods. Yet, the lipids contained in oil or fat are not only in the food but also make up a great part of the cell membrane. This includes the cells responsible for gustatory responses. This can explain why some surfactants can bind with those membranes. Lecithin, extracted from eggs and soy is one example. After binding the taste receptors in those membranes are less available to receive information. Thus, it decreases the taste perception. Through this action mechanism, bitterness it tasted with a lower intensity. This bitterness masking strategy may, however, also impact other tastes. The taste of the surfactant itself might also be unpleasant. An effective design for masking bitterness with emulsifiers isn't easy to produce.
Lipids and emulsions
Many bitter molecules have a higher affinity with lipids (fats) than water. When blending a bitter compound, water, and oil together, there will be a higher proportion of bitter molecules going in the lipid part than in the water part. If trapped in the lipid phase, it is more likely that it won’t be able to get to the tongue bitter receptors before being swallowed. So, if there is a lipid phase in the formulation, making sure that the bitter molecule is mainly present in the fat phase is a good approach. Because of the different lipid/water affinity of the molecules, part of the concentration will end up in the water. Some chemical properties of the matrix (pH especially), and diffusion effects condition this movement. Only the part in water can demonstrate its bitterness, which can be minimized. This technique is very dependent on the property of the bitter molecule but works well for quinine and some phenolics for example. Besides, this phenomenon works if the lipid phase is well separated from the water phase but also in emulsions, where both are mixed in a stabilized form. It can be under the form of a very small droplet of water in fat (like in butter) or a very small droplet of fat in water (like in a salad dressing). We can also play on the size of the droplets and the quantity of surfactant used to make those droplets similar to taste buds cells. Then, the bitter compound can interact with those micro-droplet and not the taste receptors, which is a complementary method to mask the bitterness.
Bitterness is aversive but many routes exist to mask this taste. The electronic tongue can be a universal predictor of bitterness and be used for the evaluation of bitter masked formulations. While it cannot replace sensory panels, it reduces its need to a minimum. Its use can speed up a formulation process and offer new options to improve an existing range.
Florian Woisel
Sources:
[1] Tahara, Y., & Toko, K. (2013). Electronic tongues–a review. IEEE Sensors Journal, 13(8), 3001-3011.
[2] Penn State University news, https://news.psu.edu/story/159262/2011/04/01/research-shows-taste-perception-bitter-foods-depends-genetics, accessed on the 10th of August 2020
And
Aliani, M., & Eskin, M. N. (2017). Bitterness: perception, chemistry and food processing. John Wiley & Sons.
Coupland, J. N., & Hayes, J. E. (2014). Physical approaches to masking bitter taste: lessons from food and pharmaceuticals. Pharmaceutical research, 31(11), 2921-2939.