science news logo

science service logo


Ketones to the Rescue
Fashioning therapies from an adaptation to starvation

Ben Harder

In times of plenty, both the mind and the body thrive. But deprived of basic sustenance, the mind perishes before the body does. That's not New Age philosophy; it's basic metabolic chemistry. While most of the body manages food shortages with relative ease, the tissues of the brain are vulnerable during periods of scarcity. So when blood sugar dips, the brain must fall back on special biochemistry to meet its energy needs. From studying that metabolic back-up system, a coterie of scientists has drawn inspiration that could lead to a new treatment for conditions as diverse as epilepsy, diabetes, Alzheimer's disease, and heart failure.

Most of the time, the body makes its fundamental fuel, glucose, from ingested carbohydrates. With each meal, the bloodstream gets replenished with glucose to replace the blood sugar that hungry cells have consumed to satisfy their metabolic needs. The body can't store glucose well, yet cells must be fed continually. So the body puts away extra energy in the form of fat, which it can break down into energy-supplying fatty acids when needed. A starving animal or a person with normal fat stores can thus sustain most of the body's cells for weeks or months without eating.

But brain cells, even hungry ones, can't avail themselves of these emergency stores. A physiological barrier that blocks toxins in the bloodstream so they can't enter the delicate brain also keeps out fat and fatty acids. As a consequence, when glucose in the blood runs low, brain cells can run into trouble.

People are uniquely vulnerable to such glucose starvation because of their disproportionate braininess. Although the brain makes up about 2 percent of a normal adult's weight, it commands roughly 20 percent of the body's resting metabolic budget.

A condition found only in people and a few ruminants can protect against this Achilles' heel. The state, known to followers of the popular Atkins diet, is called ketosis. When blood-glucose concentrations get low, the liver converts a portion of fatty acids into acids called ketone bodies or ketones. These substances can substitute for glucose and fatty acids as cellular fuel. However, unlike fatty acids, ketones can penetrate the blood-brain barrier.

While ketosis may guard the brain in times of starvation, Richard L. Veech has additional applications in mind. Veech, who works at the National Institutes of Health in Rockville, Md., argues that ketones might be therapeutic any time cells are threatened by energy deprivation. Such threats could arise both from a lack of fuels and from cells' failure to properly metabolize the fuels at their disposal. The latter category covers a broad array of diseases.

Veech and others have been suggesting for several years that ketosis could help treat, among other conditions, Alzheimer's and Parkinson's diseases, certain insulin disorders such as type 1 diabetes, and several metabolic disorders caused by rare mutations.

"These diseases appear wildly different," Veech says. Treating "all these different things with some magic substance sounds improbable," he adds. Yet across a wide range of specialties, doctors who've dabbled with ketone-based therapies are warming to that seemingly outlandish idea, and a vanguard of research on ketone therapies is appearing in scientific journals. At NIH earlier this fall, Veech hosted a gathering of researchers who have studied ketones.

Ketone cuisine

There is one medical condition in which ketones find proven, if limited, application. Since the 1920s, a ketosis-inducing diet has been used to treat some cases of severe childhood epilepsy. This high-fat, low-protein, low-carb regimen shifts the body's main fuel supply from glucose to ketones and fatty acids. This ketogenic diet is more extreme than the high-protein Atkins diet, which produces ketones in urine but not necessarily in the blood, says Veech.

Whereas most people consume less than a third of their calories in the form of fat and the rest as carbohydrates or protein, people on the medical ketogenic diet obtain at least two-thirds of their calories from fat.

"It's a hideous diet," says Kieran Clarke of the University of Oxford in England, who attended Veech's summit. "Think of eating pounds of butter at a time, and eating cream on top of that," she says. Not surprisingly, many children find the diet unpalatable. In studies, refusal to eat has been a primary cause for the treatment's failure. Implementing the diet, furthermore, usually requires a hospital stay and the involvement of numerous dietitians and pediatricians. Enforcing it requires parents to weigh foods and calculate ratios of calories from different sources.

With the widespread introduction in the 1960s of more effective drugs for epilepsy, the unwieldy diet's use declined. Its reputation has recently enjoyed a resurgence, however, because some seizures that had been resistant to the drugs were observed to stop during ketosis. The well-publicized case of one boy, whom doctors at Johns Hopkins Medical Institutions in Baltimore successfully treated with the ketogenic diet, inspired the 1997 movie First Do No Harm.

Even so, no more than a few hundred people in the United States are on the medical ketogenic diet at any one time, estimates Eileen P.G. Vining, a pediatric neurologist at Johns Hopkins. For one thing, she says, it's prescribed almost exclusively for children because doctors are concerned about the heart attack risk that adults might face from chowing down on so much fat.

To get a feel for how serious the side effects of the ketogenic diet might be, Vining, Peter O. Kwiterovich, and their colleagues studied 141 epileptic children they'd treated for at least 6 months at Johns Hopkins since 1994. The children's blood concentrations of total cholesterol, triglycerides, and other markers associated with cardiovascular disease had jumped by as much as 60 percent during the ketogenic treatment. Meanwhile, blood concentrations of high-density lipoproteins, or good cholesterol, fell by an average of 13 percent, the researchers reported in the Aug. 20 Journal of the American Medical Association.

Even if those numbers translate into a health risk for children on a ketogenic diet, which is far from certain, "it's a price worth paying" when children are wracked by drug-resistant seizures, Vining contends.

Nevertheless, there could be a better way. Studies suggest that certain ketones are directly involved in inhibiting seizures. That raises the possibility of supplanting the ketogenic diet with pure ketones as drugs.

Rescue mission

Limited quantities of ketones are produced for research purposes, but they are expensive and difficult to test because the body breaks them down quickly. Nevertheless, neurologist Serge Przedborski of Columbia University is experimenting with ketones. Przedborski's main research interest is Parkinson's disease (SN: 5/3/03, p. 285: Available to subscribers at http://www.sciencenews.org/articles/20030503/note10.asp), the symptoms of which include tremors, muscle stiffness, and loss of balance and coordination. The physiological hallmark of Parkinson's is the loss of certain neurons that respond to the brain chemical dopamine.

The degeneration of those neurons and of similar brain cells in Alzheimer's disease has been linked to defects in the cells' energy-producing machinery, or mitochondria. In both diseases, mitochondria in some neurons are inefficient at metabolizing glucose. But the process by which mitochondria metabolize ketones isn't necessarily impaired in the two diseases.

Veech, Clarke, and four of their colleagues from Japan demonstrated 3 years ago that, in test tubes, the ketone D-beta-hydroxybutyrate protects neurons that have the mitochondrial defects associated with Parkinson's and Alzheimer's.

To test the effectiveness of the approach in animals, Przedborski and his colleagues implanted into some laboratory mice a pump that gradually released D-beta-hydroxybutyrate, and into other mice a dummy pump. A day later, the researchers gave the animals a neurotoxin that inhibits glucose metabolism in the critical neurons. That procedure is commonly used in the lab to induce a condition similar to Parkinson's disease.

After a week, the researchers counted surviving neurons. While mice given dummy pumps had lost about two-thirds of a certain neuron type associated with Parkinson's, mice treated with 160 milligrams of D-beta-hydroxybutyrate per kilogram of body weight per day appeared to have lost only one-third of those cells. Mice receiving lower doses of the ketone didn't fare noticeably better than the animals that had gotten the dummy pumps.

"It's not a dramatic effect," Przedborski acknowledges. "But, sure enough, we were able to recover some of the function of the mitochondria." Most important, the scientists report in the September Journal of Clinical Investigation, the high dose of the ketone prevented the mice from developing Parkinson's-like movement problems.

To confirm that defective mitochondria use ketones to detour around their obstructed metabolic pathway, Przedborski's team administered a second toxin, which interferes with ketone metabolism. In mice with both metabolic pathways blocked, the ketone therapy didn't rescue any neurons.

The Columbia researchers' findings support the idea that ketones could help people with Parkinson's disease, says Theodore B. VanItallie of St. Luke's-Roosevelt Hospital Center in New York City. "There's enough evidence available now to encourage people to test the hypothesis," he says. "There's at least a reasonable possibility that these things [ketones] will work."

VanItallie and his colleagues recently put several people with Parkinson's disease on a ketogenic diet, but the researchers haven't yet gathered enough data to draw conclusions. VanItallie is looking for funding to mount a full-size trial.

Diabetes, too, can affect the brain. Children with type-1 diabetes lose some mental acuity when their glucose metabolism slows, says Jullie W. Pan of the Albert Einstein College of Medicine in New York. That can eventually affect their academic performance.

In type 1 diabetes, the body doesn't have enough insulin to do its normal job of transporting glucose into cells that would metabolize it. In fact, ketosis is a symptom of diabetic shock because it arises when glucose metabolism is suppressed. Insulin injections can boost glucose metabolism, but blood insulin can vary considerably between injections.

Pan is now studying the effect of ketone infusions in diabetic children to see whether the therapy might compensate for the effects of glucose-metabolism fluctuations on the brain.

Putting heart into it

Brain effects of low glucose availability aren't the only problems that ketones could conceivably fight. An international team of doctors recently reported successes in using ketones to treat three children with the rare genetic disease known as multiple acyl-CoA dehydrogenase deficiency, or MADD. The metabolic defect renders the body unable to process certain fatty acids. A low-fat diet and other interventions sometimes help affected children, but weakened muscles, particularly heart muscles, and damage to other tissues can lead to early death.

The first child the researchers treated with ketones was a 2-year-old boy receiving standard treatment for MADD who suddenly developed quadriplegia and could no longer speak. To supply energy to poorly functioning cells, doctors gave the boy oral doses of ketones every 4 hours. The boy recovered gradually until, after 19 months of the ketone therapy, he could again walk unassisted, a development that NIH's Veech hails as remarkable.

The researchers subsequently gave the same treatment to two other children with MADD, both of whom had suffered heart failure. These patients showed substantial recovery, pediatrician Johan L.K. Van Hove, now at the University of Colorado Health Sciences Center in Denver, and his colleagues reported in the April 26 Lancet.

Oxford's Clarke suggests that ketones could treat heart failure from other muscle-weakening causes, as well. One hypothesis of heart disease suggests that a heart can gradually fail after a nonfatal heart attack because the organ's muscle cells become inefficient at both taking up glucose and metabolizing fatty acids, says Clarke. Ketones could provide heart muscle with an alternative energy source. Clarke is experimenting on failure-prone rat hearts to test this idea.

If this treatment is ever to be practical, an abundant and reasonably inexpensive source of purified ketones will be needed, she says. "The only way now we can produce ketones in the body is a high-fat diet," she says. "You couldn't feed a high-fat diet to a heart-failure patient. That would be a disaster."

Even in the small amounts needed by the children that Van Hove and his colleagues treated, purified ketones could cost $20,000 per patient per year, says Veech. The higher quantity of ketones that an adult would require would lead to even more expense.

That's only a temporary obstacle, according to Veech. In the lab, scientists can already use bacteria to manufacture a compound that can be processed into ketones. If researchers can improve on the current method for refining the precursor, then ketones could be inexpensively produced, he says, and his theories about their broad medical effectiveness could be put to the test.

********

Letters:

This article describes attempts to use ketones as a substitute for glucose when it's unavailable to the brain because converted fats can't penetrate the brain. Wouldn't it be simpler to feed or inject glucose directly?

Oliver H. Winn
Corona del Mar, CA

The researchers who conducted the study say that giving glucose might help in some cases, but the conditions discussed in the article involve a defect in glucose metabolism.—B. Harder

References:

Kashiwaya, Y. … K. Clarke, and R.L. Veech. 2000. D-β-hydroxybutyrate protects neurons in models of Alzheimer's and Parkinson's disease. Proceedings of the National Academy of Sciences 97(May 9):5440–5444. Available at http://dx.doi.org/10.1073/pnas.97.10.5440.

Kwiterovich, P.O., E.P.G. Vining, et al. 2003. Effect of a high-fat ketogenic diet on plasma levels of lipids, lipoproteins, and apolipoproteins in children. Journal of the American Medical Association 290(Aug. 20):912–920. Abstract available at http://dx.doi.org/10.1001/jama.290.7.912.

Tieu, K. … and S. Przedborski. 2003. D-β-hydroxybutyrate rescues mitochondrial respiration and mitigates features of Parkinson disease. Journal of Clinical Investigation 112(Sep. 15):892–901. Available at http://dx.doi.org/10.1172/JCI200318797.

Van Hove, J.L.K., et al. 2003. D,L-3-hydroxybutyrate treatment of multiple acyl-CoA dehydrogenase deficiency (MADD). Lancet 361(April 26):1433–1435. Summary available at http://dx.doi.org/10.1016/S0140-6736(03)13105-4.

Vanitallie, T.B., and T.H. Nufert. 2003. Ketones: Metabolism's ugly duckling. Nutrition Reviews 61(October):327–341. Abstract available at http://dx.doi.org/10.1301/nr.2003.oct.327-341.

Veech, R.L., et al. 2001. Ketone bodies, potential therapeutic uses. IUBMB Life 51:241–247. Abstract available at http://dx.doi.org/10.1080/152165401753311780.

Further Readings:

Seppa, N. 2003. Protein implicated in Parkinson's disease. Science News 163(May 3):285. Available to subscribers at http://www.sciencenews.org/articles/20030503/note10.asp.

Sources:

Kieran Clarke
Department of Biochemistry
University of Oxford
Oxford OX1 3QU
United Kingdom

Philippe Demaerel
Department of Radiology
University Hospital Gasthuisberg
Katholieke Universiteit Leuven
B-3000 Leuven
Belgium

Peter O. Kwiterovich
Lipid Research Atherosclerosis Division
Johns Hopkins University
Lipid Clinic
550 North Broadway
Suite 308
Baltimore, MD 21205

Jullie W. Pan
Department of Neurology and Neuroscience
The Gruss Magnetic Resonance Research Center
Room 211
Albert Einstein College of Medicine of Yeshiva University
1300 Morris Park Avenue
Bronx, NY 10461

Serge Przedborski
BB-307
Columbia University
650 West 168th Street
New York, NY 10032

Kim Tieu
Department of Neurology
Columbia University
650 West 168th Street
New York, NY 10032

Johan L.K. Van Hove
Department of Pediatrics
Box C233
University of Colorado Health Sciences Center
4200 East Ninth Avenue
Denver, CO 80262

Theodore B. Vanitallie
1678 Jose Gaspar Drive
Boca Grande, FL 33921

Richard L. Veech
Unit on Metabolic Control
Laboratory of Membrane Biochemistry and Biophysics
National Institute of Alcohol Abuse and Alcoholism
12501 Washington Avenue
Rockville, MD 20852

Eileen P.G. Vining
Pediatric Epilepsy Center
Department of Pediatrics
Johns Hopkins Medical Institutions
Baltimore, MD 21205


From Science News, Volume 164, No. 24, December 13, 2003, p. 376.