How Food Affects Drug Delivery
Grapefruit juice and St. John’s Wort, a popular herbal supplement, have something in common. Both interfere with the activity of some drugs.
It’s easy to find such examples of interactions between the foods and supplements people eat and the drugs they take. But it’s much harder to find out about the chemical interactions that drive those clashes. So Shana Sturla of the Swiss Federal Institute of Technology, Zurich (ETH); John Manthey of the U.S. Department of Agriculture; and Kyung Myung of Dow AgroSciences decided to make chemical fundamentals the focus when they organized a half-day symposium on food-drug interactions for the American Chemical Society national meeting in Boston.
The trio’s session, hosted by the Division of Chemical Toxicology, played out on a rainy Tuesday in a crowded room at the Boston Convention Center. Four speakers described different biomolecules involved in diet-drug interactions, such as liver enzymes, transporter proteins, and lipid-protein aggregates in the bloodstream. And one speaker focused on the impact foods can have on systems for delivering drugs.
The liver enzyme avenue to drug interactions is by far the most familiar to food-drug interaction aficionados. Liver enzymes known as cytochrome P450s help the body break down drugs, and many foods and herbal remedies contain compounds that alter these enzymes’ activities and thus change the drugs’ potencies.
In Boston, Paul F. Hollenberg of the University of Michigan, Ann Arbor, talked about his team’s search for a chemical explanation of what’s called the grapefruit juice effect—drinking just a glassful a day can lead to dangerous increases in levels of some drugs, such as atorvastatin (Lipitor), in the bloodstream. Researchers know that bergamottin, a molecule in grapefruit juice, and its metabolites inactivate a particular cytochrome P450 enzyme known as CYP3A4. Bergamottin molecules work by a process called suicide inactivation: The enzyme transforms them into highly reactive intermediates, which then react with the enzyme.
Mass spectrometry has clued Hollenberg’s team into bergamottin’s basic effect on the P450 enzyme: It irreversibly modifies a glutamine residue on the protein’s surface. Still, at a detailed mechanistic level, “we are not absolutely sure how bergamottin inactivates the enzyme,” Hollenberg said. The structure of the reactive intermediate hasn’t been determined yet, and it’s not clear why modifying a glutamine so far from the active site would have such a profound effect, he explained.
Despite its infamous role in the grapefruit juice effect, bergamottin may not be all bad, Hollenberg said. The problem is that it’s impossible to get a consistent dose of bergamottin in its natural context. People may not consume the same amount of grapefruit or grapefruit juice each day, and different grapefruit crops or juice batches will have different levels of bergamottin. If bergamottin could be administered in a reliable fashion, it might be used to boost the bioavailability of drugs or drug candidates that undergo cytochrome P450 metabolism, Hollenberg said. “It could even lower the doses needed of expensive drugs,” he added.
As for grapefruit lovers whose medication precludes them from consuming a favorite fruit, Hollenberg says solutions to their problem are in the offing. Researchers are trying several strategies to remove bergamottin from grapefruit juice, such as genetic engineering of grapefruit trees, as well as chemical or photochemical approaches for ridding the juice of bergamottin. “It’s like the idea behind decaffeinated coffee,” he said.
Botanicals can also affect cytochrome P450 enzymes and thus drug metabolism (C&EN, July 19, page 38). One example is milk thistle, a flowering plant in the daisy family that people have used to treat and prevent liver diseases for thousands of years. “We know nothing, almost, about the drug interaction potential of supplements,” said David J. Kroll, a pharmacologist at North Carolina Central University. His collaborators purified compounds from milk thistle and used them to reveal more about how the plant interacts with drugs. A small grant from the 2009 American Recovery & Reinvestment Act stimulus funds enabled Kroll’s collaboration, which included researchers in the labs of pharmacognosy specialist Nicholas Oberlies at the University of North Carolina, Greensboro, and drug metabolism expert Mary Paine of the University of North Carolina, Chapel Hill.
The team’s results suggest that some milk thistle compounds block cytochrome P450 enzymes that break down the blood thinner coumadin. Coumadin’s efficacy is known to be affected by foods, most notably leafy greens such as kale (C&EN, Aug. 16, page 15). Patients don’t always tell their doctors what supplements they take, so it’s critical to convey information on drug-supplement interactions, Kroll said.
The Paine lab also found that two milk thistle compounds reversibly inhibit certain P450 enzymes. This finding differs from the results of a 2004 study from Hollenberg’s group, which found that a mixture containing the two compounds inhibits the enzymes irreversibly.
It’s important to distinguish between the two possible effects, Kroll said in a phone interview after the meeting. If the P450 inhibition is irreversible, “the supplement’s effect will be much longer-lasting,” he said. He plans to send samples of compounds to Hollenberg so the Michigan team can test them in their system.
During a question-and-answer period, Hollenberg addressed the discrepancy, saying “one is always disappointed when one’s work cannot be replicated.” He discussed differences between the teams’ experimental protocols that could contribute to the inconsistencies and then expressed his desire to receive samples of the purified compounds and work with Paine’s team to understand the differences.
Cytochrome P450 enzymes may be familiar names in food-drug interactions, but membrane-embedded transporter proteins are up-and-coming players, according to Marilyn E. Morris of the State University of New York, Buffalo. “Transporters are important in tissue distribution, absorption in the intestine, and elimination via the kidney or through bile,” she explained in Boston. When compounds in fruits, vegetables, and botanicals inhibit or activate transporter proteins, they can affect the bodily distribution and clearance of drugs that rely on transporters to reach their destinations. That list of drugs is diverse, including the cholesterol-lowering medication pravastatin (Pravachol) and the allergy medication fexofenadine (Allegra).
In Boston, Morris used St. John’s Wort to demonstrate the importance of transporters in herb-drug interactions. In a 2000 study by another team, healthy volunteers taking the botanical saw a significant drop in levels of the HIV medication indinavir (Crixivan) in their blood. Indinavir is a transporter substrate, and its response to St. John’s Wort is dangerous because in addition to preventing HIV patients from getting enough drug to keep the virus at bay, it could encourage the development of resistance to the medication. Additional work has suggested that St. John’s Wort stimulates the intestine’s arsenal of P-glycoprotein, a transporter that flushes substances out of cells. A glut of P-glycoprotein is likely what lowers the drug’s levels in patients, Morris said.
Morris’ own research involves evaluating compounds such as isothiocyanates, which are found in vegetables such as broccoli as well as in botanicals, for their transporter-mediated effects on toxicity and on the efficacy of drugs such as breast cancer chemotherapeutics. She closed her lecture by noting that many effects of supplements and foods on transporters have not been investigated in animals or people. “We really need more in vivo studies to determine the clinical relevance of these interactions,” she said.
Diet-drug interactions can involve lipids as well as proteins. Kishor M. Wasan of the University of British Columbia told attendees about drug interactions with lipoproteins—the lipid-protein aggregates best known for shipping cholesterol around the bloodstream. “We all look at lipoproteins from a cardiovascular perspective,” he said. But lipoproteins can also have a profound influence on a drug’s behavior and toxicity in the body.
When a drug reaches the bloodstream, it can bind to lipoproteins or plasma proteins such as albumin, or it can remain free. Each of those situations confers different abilities on the drug. For instance, binding to some lipoproteins can facilitate drug uptake into cells. So a change in a person’s lipid profile, whether a temporary one from a high-fat meal or a long-term one due to a disease such as cancer or AIDS, can lead to changes in the amount of a drug reaching its site of action.
But there is no hard-and-fast rule dictating what a certain lipoprotein-drug interaction will do. So it’s critical to test drug candidates in a relevant lipoprotein environment, Wasan said. Different animal species have different lipoprotein distributions, and people can also display dramatic differences, he said. “Now that drugs being developed are becoming more hydrophobic, lipoprotein interactions are becoming a bigger factor,” he said.
Wasan is collaborating with pharmaceutical company Eisai to unravel how lipoprotein binding deactivates eritoran, Eisai’s drug for sepsis, which is in late-stage clinical trials. In Boston, Wasan presented his team’s unpublished work using synthetic lipoproteins to help narrow down which lipoprotein interactions deactivate eritoran. Now that they know the culprits, they hope to work on a formulation that will direct eritoran away from them.
As with Wasan’s work, lipids figure prominently in Rebecca L. Carrier’sresearch. At the Boston meeting, Carrier, a former formulations researcher at Pfizer who now heads an academic research team at Northeastern University, discussed how foods affect drug delivery systems. Many of these systems are lipid-based and enhance the solubility and absorption of hydrophobic drugs. But it’s not always clear how they work, so it’s not easy to predict how well they’ll perform with a new drug, she said.
Carrier’s team develops computational models to try to describe what happens when foods interact with delivery systems, and how that might affect oral drug absorption. Eating changes many conditions inside the body that must be considered in the models, such as a decrease in pH in the gut or the secretion of bile, Carrier said. And depending on the chemical structure of the drug delivery system, more specific chemical interactions may come into play. Carrier gave the example of cyclodextrin, a frequently studied carbohydrate-based drug delivery system. The components of lipid-based micelles that exist in the intestine after a person has eaten “can associate with hydrophobic cyclodextrin cavities,” she said. “This can then displace a drug that’s inside the cavity,” therefore changing the cyclodextrin drug delivery system’s effectiveness, she said.
Carrier’s team also conducts experimental work to test the theoretical models. In advice that paralleled Wasan’s, Carrier recommended that researchers test drug delivery systems in experimental environments that closely mimic environments in the body.
Researchers are still far from making all the connections about how diet-drug interactions work. If they could amass that knowledge, it might help drugmakers predict how their drug candidates will hold up in real life long before they get inside a patient. “We hope this session establishes a framework to better understand the molecular mechanisms of food-drug interactions,” Myung said.
Attendee Stephen A. Smith, a former medicinal chemist at GlaxoSmithKline who now is an independent consultant with Stort MedChem Consulting, in England, thought the session delivered what it promised. “Everyone talks about the potential effects of foods on drugs,” he said. “But few presentations offer systematic scientific content as opposed to gut feelings.”
Learn more here from Chemical & Engineering News.