Model Stomach Takes Digestion Outside the Body

Twenty years ago, after Bob Quinn began growing Khorasan wheat, an ancient and vanishing Egyptian grain with an oversized, banana-shaped kernel, he noticed something unusual. Not only could people who have trouble eating regular wheat digest Khorasan, but it actually made them feel better. Today, food from the grain, which the Montana farmer has branded Kamut, is sold in health stores around the world and is prescribed by some doctors as a treatment for wheat allergies. Yet, says Quinn, ”we don’t know how it’s really acting in the body to create these differences.”

Now, though, two scientists at the Institute for Food Research (IFR) in Norwich, England, might be able to offer some answers. Later this year, Martin Wickham and Richard Faulks plan to feed Quinn’s pasta to the world’s first and most sophisticated artificial stomach and compare the output to that from a meal of conventional pasta. Not only does the IFR’s ”model gut,” as it’s called, break down food with the proper quantities of enzymes and acids, it also mimics the physical motion—the mixing and shearing—that occurs inside the stomach. Besides clarifying our understanding of digestion, the invention may revolutionize the way processed foods are designed and how drugs are delivered. Since the machine began operating, in November 2006, some 10 to 15 companies have used it to test their products.

The model gut originated with a study that aimed to uncover how carotenoids—pigments that color foods such as tomatoes and carrots—are released from food and absorbed by the body. Faulks, a food chemist and nutritionist, and Wickham, a chemist specializing in colloidal mixtures, were unable to measure exactly what transpires in the stomach. Clinical trials are difficult and expensive. ”You can’t obtain samples of solid dinner because you can’t easily aspirate them—you can’t suck them up through a pipe,” says Faulks. ”You have to physically get in there. This, of course, then generates huge ethical problems.”

Advances in echo-planar imaging, an ultrafast type of magnetic resonance imaging, provided a breakthrough. This technique relies on just a single excitation of the molecules under study rather than the sequence of energy bursts in a traditional MRI, allowing it to capture 10 images or more each second. Echo-planar imaging ”allows us to collect data from people’s stomach and small intestine while they’re digesting foods,” says Wickham, without invasive probes.

Given the mystery that has shrouded the workings of the stomach, it’s fitting that the bulk of Faulks and Wickham’s model is behind smoky plastic: a literal black box. One day in October 2007, the scientists agree to demonstrate the device for IEEE Spectrum. They place two cans of soup on the countertop and prepare to feed the machine a late-morning meal.

The machine is the result of nearly a decade’s worth of effort and more than US $2 million, financed by the British government. After capturing data from hundreds of volunteers and drawing up a provisional design, Wickham and Faulks went to engineers to actually build the device. ”We had to help specify what materials were to be used, what size things were, how fast things moved,” says Wickham. ”Unless you can describe it to an engineer, an engineer can’t create it,” adds Faulks.

With Wickham at the computer manning the controls, Faulks pours a can of chunky chicken and vegetable soup into the blue funnel encased in a clear plastic cylinder at the top of the machine. This serves as the fundus, the curved, upper portion of a human stomach where newly arrived, chewed food gets gently massaged.

To measure this movement, Faulks and Wickham prepared beads of various densities made from agar, a seaweed gel. ”We got volunteers to swallow these, and then we MRI imaged the stomach to see when the stomach could break them up,” says Faulks. Still, he adds, designing the fundus was the most challenging aspect of the project. In the model, the massaging is replicated by a pressurized warm-water bath that surrounds the blue envelope, squeezing it to a rhythm that Wickham now sets.

In response to the acidity and amount of food, the computer sets the stomach’s acid level, which in turn regulates the quantity of enzymes. ”It’s an average tum, but we can change the characteristics,” says Faulks. ”You can take this model and you can simulate a child’s stomach. Or you can simulate one or two of the gastric disease states, like hyperacidity.”

Once Wickham establishes the parameters, the soup descends into the machinery behind the darkened doors. This is the simulacrum of the antrum, the lower part of the stomach. The model simulates the mixing that takes place here with a barrel that moves up and down. Inside, a piston fitted with an interchangeable plate provides the shear. By changing the thickness and size of the plate, the scientists can adjust the shear force exerted on the food.

A human stomach eventually pushes food into the small intestine, where the nutrients are delivered to the rest of the body. Faulks sets a beaker before a spout in the front of the machine, and soon the first measure of partially digested soup comes spurting out. ”What you saw going in was soup, and what you get coming out of the stomach is gastrically processed, plus enzymes, plus acid,” says Faulks. ”But because it’s the first cycle out, it won’t have a lot of acid and enzyme in it. As the process continues, more and more enzymes are going to be added.” A few minutes later and, splat, more soup drops into the beaker. And then more. ”We can collect a sample every time the stomach processes it,” Faulks says.

By studying the samples, the scientists hope to understand how digestion makes nutrients available to the body. ”Let’s say, for example, you wanted to measure how much of the iron in a food became soluble or went into an absorbable form during its acid processing in the stomach,” says Faulks. ”You can measure that from the output from the stomach.”

For the moment, the model gut’s time is mostly bought by companies hoping to engineer ”foods that deliver all those necessary sensory qualities, but which also help to control appetite and satiety,” says Faulks. Scientists have already figured out how to control hunger with carbohydrates—for example, the candy bar that really satisfies—but now, Faulks adds, ”there’s an opportunity here for the industry to understand what happens to proteins or fats.” Once researchers determine which attributes are most conducive to alleviating hunger, they can take the results to clinical trial.

This focus reflects a couple of priorities set firmly by the British government. One is the growing alarm over obesity—at the time of the visit, a new report had come out placing the UK on a trajectory of gluttony over the next 50 years. Another is a desire to have science pay more of its own way. The model gut is what the IFR calls an ”exploitation platform.”

”It supports the research program from which it originated,” says Reg Wilson, head of IFR Innovation. The IFR, Wilson adds, would also like to broaden the potential customer base by moving into pharmaceuticals. Oral drug delivery is a logical extension of the work, says Faulks. ”It’s a chemical, the way a meal is a chemical. So the process is exactly the same.”

But these priorities could squeeze out other strains of research. Quinn, for instance, theorizes that industrial agriculture has created foods that are harder to digest than ancient, unmodified foods like his Kamut, but he can only afford to partially test his idea in the model gut. Quinn recalls that Faulks and Wickham were polite, though not especially enthusiastic, about his proposal until they teased out a link to obesity.

”We’re happy to help Bob with his research,” responds Hadyn Parry, business development manager for Plant Bioscience Limited, the government-owned company that funded the device. ”The model gut allows him to do research at a much lower cost than he’d otherwise have to spend, because he’d have to do a full clinical trial.”

The model gut is so popular that the IFR is planning to manufacture five more models as a prelude to a larger commercial endeavor. ”We might hire them out, get our feet in the market,” says Faulks. ”Or we might sell them.”

In the lab, the artificial gut is finishing its meal. Soup plops out of the spout two more times, and the beaker is full. Wickham points to visible bits of chicken and carrot. ”It’s great, isn’t it?” he says. ”The carrots are breaking up quite nicely.”