Biologist William K. Purves of Harvey Mudd College offers this explanation:

To understand how something can taste sweet and yet add no calories to the diet, we should address two questions. First, what are calories, nutritionally speaking? Second, what constitutes a sweet taste?

Calories are a measure of the energy made available when we digest and metabolize food. The energy drives the replacement of molecules we have lost, enables us to move, and so forth; we store excess energy as fat. A substance that we do not metabolize releases no energy it "has no calories" and is not a food.

A sweet taste results from the binding of molecules to specific receptor proteins in our taste buds. Sweet-taste-sensory cells in the taste buds have these receptor protein molecules embedded in their plasma membranes. Binding of a molecule to a receptor protein initiates a cascade of events within the taste-sensory cell that eventually releases a signaling molecule to an adjoining sensory neuron, causing the neuron to send impulses to the brain. Within the brain, these signals derived from the taste bud cause the actual sensation of sweetness. Other sensory cells, with different receptor proteins, report on other taste modalities: salty, sour, bitter, and "umami" (also referred to as glutamate, or "meat").

The events that occur between binding by the "sweet receptor" and the sensation in the brain have nothing to do with whether a molecule can be metabolized to yield energy and thus "has calories." The only factor in taste is whether the molecule can bind to the receptor.

So, what determines this binding ability? In April 2001, two research teams published independent contributions to answering this question. Both papers announced and described a protein, dubbed T1r3, which appears to be the primary receptor for sweet substances. The molecular structure of T1r3 can be seen here. Like all receptor proteins, T1r3 has a well-defined "pocket" where smaller molecules may enter and perhaps bind. Binding depends on a good fit of molecular shape and the presence of groups that interact chemically to stabilize binding.

Sucrose, the sugar in the sugar bowl, binds fairly well to T1r3 and hence leads to a sweet sensation in the brain. Enzymes readily metabolize sucrose, releasing energy and, if our diet contains excess calories, causing fat deposition.

Saccharin, once the most popular artificial sweetener, binds to T1r3 much more strongly than does sucrose, owing to the differing structures of the two molecules. Therefore, we sense saccharin as being approximately 300 times as sweet as the same amount of sucrose. Moreover, saccharin passes through the body without being metabolized and thus has no caloric content.

Aspartame (NutraSweetTM), currently the most-used artificial sweetener, also binds to T1r3 more strongly than sucrose does and tastes nearly 200 times as sweet. Unlike saccharin, however, aspartame is metabolized, yielding methanol and the amino acids phenylalanine and aspartic acid. Further metabolism of these products does yield calories, but far fewer than those obtained from the amount of sucrose required to produce the same sweetening effect.

Arno F. Spatola is a professor of chemistry and the director of the Institute for Molecular Diversity and Drug Design at the University of Louisville, where his current research focuses on peptides, including artificial sweeteners. He provides the following answer:

Is there any caloric value to artificial sweeteners? How am I able to have my cake (the sweetness of my food) and eat it too (avoid gaining weight from excess calories)? The answer to these questions, as in most areas of science, is that it depends.

Sweetness is a taste sensation that only requires interaction with receptors on our tongues. Many sugar substitutes, such as saccharin and acesulfame K (also known as SunetteTM), do not provide any calories. This means that they are not metabolized as part of the normal biochemical pathways that yield energy in the form of adenosine triphosphate, or ATP. In some cases, small quantities of additives such as lactose are added in order to improve the flow characteristics or to add bulk to the products. But the quantities of these added ingredients are so small that they do not represent a significant amount of energy-producing foodstuffs.