点击显示 收起
Dr Bistrian: I'm going to applaud the afternoon speakers for their discussion of several topical and exciting functional foods or functional nutrients—lycopene, green tea, polyphenols, and genistein—and open the meeting for discussion. Before we begin, Dr Fürst would like to make a comment.
Dr Fürst: I would like to raise the question of bioavailability, which is a problem when dealing with phytochemicals. A novel way to study bioavailability is the use of an isolated rat small intestine [Plauth M, Kremer I, Raible A, Stehle P, Furst P, Hartmann F. Nitrogen absorption from isonitrogenous solutions of l-leucyl-l-leucine and l-leucine: a study in the isolated perfused rat small intestine. Clin Sci (Colch) 1992;82:283–90; Hummel M, Pogan K, Stehle P, Furst P. Intestinal taurine availability from synthetic amino acid-taurine conjugates—an in vitro perfusion study in rats. Clin Nutr 1997;16:137–9]. By using this model (Figure 1), we assessed the uptake of luminally administered genistein and rutin. In steady state, 16% of the administered rutin was absorbed as intact glycoside, less as quercetin sulfate and quercetin glucuronide. No quercetin sulfate or quercetin glucuronide could be detected at the luminal side, and only minute amounts of quercetin were discovered (Figure 1).
FIGURE 1. . Schematic illustration of the perfusion model. 1, Perfusate reservoir with magnetic stirrer; 2, roller pumps; 3, oxygenator; 4, perfusion chamber with organ preparation, humidified and temperature controlled; 5, precision pump for luminal perfusate; 6, pressure gauge.
After luminal administration of genistein, a recovery rate of 101% was recorded. The bioavailability in a single pass considering vascular uptake was 41%; when the intestinal tissue uptake of genistein was included, the bioavailability rate was close to 47% (Table 1). It is important to emphasize that part of the luminally administered genistein is glucuronidated; the conjugation takes place presumably in the mucosa. Less likely is the role of microorganisms. We found a higher share of vascular genistein glucuronide after luminal administration of genistein because the "glucuronide transporter" at the site of the basolateral membrane (vascular) is more efficient than at the site of the brush border membrane (luminal) [Sfakianos J, Coward L, Kirk M, Barnes S. Intestinal uptake and biliary excretion of the isoflavone genistein in rats. J Nutr 1997;127:1260–8]. The combined results also demonstrate that intestinal tissue has the capacity to glucuronidate and sulfatize phenolic phytochemicals. However, species variations in metabolism of phytochemicals must be considered [Hollman PC. Bioavailability of flavonoids. Eur J Clin Nutr 1997;51(suppl):S66–9].
View this table:
TABLE 1.. Assessment of the bioavailability of genistein1
We hope that this new method will facilitate direct ex vivo investigations of intestinal handling of phytochemicals, thereby allowing future research on absorption, transport, distribution, and metabolism of phytochemicals. This knowledge is a prerequisite to evaluating the particular role of phytochemicals in human health and disease.
Dr Bistrian: We will now begin a general discussion of topics covered in this session.
Dr Field: Dr Weisburger, do clinicians or clinical researchers use patients with high levels of DNA adducts as an approach to screening, or do they submit them to some kind of a treatment to see if they can blunt neoplastic occurrences?
Dr Weisburger: As far as I know, the only laboratory that uses humans for this kind of activity is Dr Lipkin's at the Strang Cancer Prevention Center in New York. They obtain a small sample of rectal mucosa and if they find a high level of aberrant crypts with a high rate of cell duplication, they identify them for preferential treatment.
Dr Mukhtar: Dr Arab, epidemiologic studies seem not to be helpful in explaining results of animal studies that show an association between tea consumption and cancer risk. Do we know why the epidemiologic observations are not consistent with laboratory findings? Are there known chemopreventive agents for which the epidemiologic results are consistent with laboratory findings?
Dr Arab: There are many agents where there is consistency, particularly in studies of the cardiovascular system with cholesterol models. When we don't see consistency between the laboratory findings and the epidemiologic studies, it means 1 of 2 things. You have an interesting model in the laboratory, but it is irrelevant to humans, perhaps because the dose is wrong or because the model itself is wrong. Or, there may be a problem with the population studies.
There are potential flaws in tea studies. If you are doing population studies in England, where almost everyone drinks tea, you may not have enough variance within the population. There may be other confounding factors as well.
There are many models in which the epidemiologic studies confirm the biochemical findings. This isn't how the partnership works best. I think it works best if we have epidemiologic findings that look promising, and then we do laboratory investigation to identify a mechanism that will explain those findings.
Where epidemiology has been most effective is in early detection of problems before we know anything about the biologic basis or even the health consequences. Examples of that range from HIV and AIDS to stopping an epidemic by changing the water supply.
Dr Weisburger: The only epidemiologic study involving black tea was by Goldbohm et al [Goldbohm RA, Hertog MG, Brants HA, van Poppel G, van den Brandt PA. Consumption of black tea and cancer risk: a prospective cohort study. J Natl Cancer Inst 1996;88:93–100]. They reported that with a high intake of vegetables, fruits, and tea, the relative risk is 0.6 for diet-linked diseases. When they corrected, in my view improperly, for vegetables and fruits, the effect of tea disappeared. The effect of black tea should be reexamined.
I was asked earlier about the effect of using milk with tea in Ireland and the United Kingdom. I think in the animal studies they used too much milk and did not find anything. In England, the standard is 1.85% milk (2% milk). We found that there was a reduction in the incidence of breast cancer that was potentiated by milk, but we need more research on this in the United Kingdom. In a joint collaborative study between the National Cancer Institute and the Chinese, there was good evidence that in some situations tea is beneficial (gastric cancer and cancer of the esophagus are lower in tea drinkers) [Gao YT, McLaughlin JK, Blot WJ, Ji BT, Dai Q, Fraumeni JF Jr. Reduced risk of esophageal cancer associated with green tea consumption. J Natl Cancer Inst 1994;86:855–8].
Dr Bistrian: What are the possible drawbacks from using improvement of a function rather than using prevention, elimination, or reduction of a risk factor as a means of assessing the value of providing a functional food? For example, it is one thing to improve immune function if someone has diminished immune function. It's quite another to increase immune function in someone who has an intact immune system in which you may well also increase inflammation, and that might have adverse effects. Are there any other situations in which the function improves and the risk factor doesn't or even may actually go in the reverse direction?
Dr Watson: Perhaps because I work with the immune system, I think your examples are appropriate. If you give vitamin E to young healthy mice, you can often improve autoimmune systems and enhance their resistance to some pathogens. I think one of the risks we take with nutrition is to assume that whatever is the normal immune function or diet is also optimal; I'm not sure that that's the case.
Just because a 25-y-old has what is considered to be optimal immune function does not mean it can't be improved, particularly if you have a specific response in mind (eg, cancer resistance, disease resistance). I am not sure a priori that just because it is normal for healthy young adults at their prime that it is optimum.
Dr Bistrian: Generally, aren't we going to be recommending these functional foods across the broad range of immune function, from mildly deficient to normal to optimal? If we treat everyone, is there some possibility that we might raise the level for everyone but at some cost to one quintile?
Dr Watson: Yes, with AIDS there is a cytokine imbalance, with relatively high levels of the T-helper 2 cell's proinflammatory cytokines, which then, at least in theory, suppress those produced by T-helper cells, which reduces cancer resistance. However, increasing TH1 cytokines in healthy people or elderly persons with low levels should decrease their risk of cancer or infectious disease. Thus, we may want to push it slightly that way toward more cytokines and thus improved cellular defenses; that may have some overall benefit to society. I agree with you on the basic premise, but I still think that normal doesn't necessarily equate with optimal.
Dr Milner: With regard to soybean, lycopene, tomatoes, and green tea, is it appropriate to recommend to the general public that we increase each of these food items in the diet? If so, by how much? Let's start with soybean.
Dr Lamartiniere: I think that there can be benefits by giving soy products early in life. We might exert a protective effect from short-term exposure that would be a lasting benefit. However, until all the toxicity studies are carried out, especially perinatal studies and second- and even third-generation studies in animals, and we look closely at the potential toxicity in human beings, I would not make any such recommendation. There is no real problem with trying to improve health beyond what we think is normal and good, but we have to be sure that this would be in the best interest of the population and that there would be no side effects or more extreme toxicities.
Dr Arab: Based on what we know now about the tomato, I would not want to endorse the use of lycopene supplements, but we do have populations that are regularly consuming high amounts of tomatoes in which we have the safety factor fairly well covered. We are telling people to eat fruits and vegetables several times a day, so we are recommending fairly high levels of intake already. I think it is less of an issue if recommended intake is within the norm and we are not seeing any adverse effects. I would emphasize, however, that what we are showing in the myocardial infarction arena should be confirmed. It's a single study, and the relation should not be emphasized based on this alone.
I also do research on cabbage, and I would say that based on what we know and think is going on, consuming it once every week or 2 wk is more likely to be within the range of optimal. Since we see protective effects with colon cancer, we would not increase the dose and expect more protection. We are in a fine balance between toxic effects and enzyme induction, and we know that we can maximize enzyme induction within a low dose range.
Dr Bistrian: How much evidence are you going to require before allowing a company to add a supplement to a product? Would you base it on epidemiologic evidence alone, or would you require a human study with a randomized trial?
Dr Weisburger: We don't want to reinvent the wheel. I'm quite sure that a Chinese child at weaning will be exposed to and fed some soy foods. I'm sure that a Greek child at weaning will be fed some tomato products. So we don't have to redo safety studies just because we are Americans. Let's look at the world and see what has been done. If a supplement has been tested in children or infants without adverse effects, we should be able to conclude that it is probably okay worldwide.
Dr Lamartiniere: I would love to believe that, but there is also the possibility that we should not generalize between populations. Asians have eaten soy for so long that their gut microbes may be different from those of Americans, and the Asians can tolerate phytoestrogens during the perinatal period without experiencing toxicity. We don't know this about Americans because they have not had a traditional soy diet.
Dr Arab: We won't have intervention studies for all of the outcomes that we want them for. We won't have them with the range of doses that would be relevant, because the population is such that as soon as the product is available, there will be people who are consuming it over a range of probably 10-, 20-, or 50-fold of one another. Also, who will pay for these studies?
I'm less concerned about this problem with functional foods than I am with what is actually going on right now with herbal supplements. People are taking them in amazing doses; women diagnosed with breast cancer are taking dozens of products. There is a tremendous potential for harm as well as good. We know nothing about them. A national experiment is going on without adequate control subjects.
Dr Milner: In Europe, a lot more of these products are consumed as if they were recognized drugs. Maybe we need to be a little more open-minded and not assume at the outset that they are not going to function. I think Dr Arab's point was that we have to be concerned about toxicity of these products because they are not typical items with which we have experience.
Dr Arab: We do know something about the functions and health effects, but the health effects are not being studied either here or in Europe. I was in the office that would have been studying them, the Bundesgesundheitsamt, for 7 y, and what we saw was a compilation of descriptions of products. No one is doing epidemiologic studies or investigating the long-term consequence of their use. A number of the German chancellors were taking injections of embryonic proteins for longevity, but there have been no epidemiologic studies of long-term health effects, even though there is a tremendous respect for some of these products in Europe and a long-term history of their use.
Dr Field: Australians are thinking about fortifying wheat with soy because they believe it has protective effects on bone density. There may be some lessons to glean from their public health experience. Dr Milner: In Sweden and Norway, they are putting selenium in their fertilizer, and intake has gone up substantially.
Dr Mukhtar, should everyone drink green tea and how much?
Dr Mukhtar: Tea has been used for generations without evidence of toxicity in any population. I get many phone calls from patients with a variety of diseases who are drinking tea. I tell them all that even if drinking does not help you, it won't do any harm.
Dr Watson: I disagree. Tea has a great deal of caffeine in it, and it is a well-known, substantial central nervous system modifier. In pregnant women, caffeine accumulates in the fetus at 7 times higher levels than in the mothers. Caffeine can be addictive to the extent of causing mild withdrawal symptoms.
Dr Mukhtar: But there is decaffeinated tea as well.
Dr Watson: That's fine. But to say that there is no toxicity to tea, green or black, is not true. There are many studies on various types of toxicities associated with caffeine. These have to do mainly with nervous system and birth defects.
Dr Roberfroid: The bioavailability of caffeine from tea is different from what it is in coffee. You have to be careful with comparing the data because they may refer to different conditions.
To return to the discussion of function, it should be relevant to health issues. We know, for example, that cholesterol is related to health, so we consider that restoring blood cholesterol levels to normal improves the function and thus is good for health. There are many such functions. I agree that the one way to start is with good epidemiologic data and then to identify the functions that are related to the effect you see. This then can be used to develop into an hypothesis to be tested in human studies.
Dr Bistrian: Let me tell you, from a clinical point of view, what happened when drugs were used to reduce the risk of ventricular tachyarrhythmias. Drugs that were very effective at stopping ventricular tachycardia actually increased mortality rate. It's still important that we check the relevance of the function to the disease outcome.
Dr Roberfroid: Yes, but again we are speaking from the point of view of nutrition, so I don't want my comments to be related too much to disease. I am inclined to take the opposite position and look at effects relevant to well-being and health. It's a completely different approach. Drugs are used to treat a disease. Food is eaten to improve the quality of life and well-being.
Dr Bistrian: I have a physiologic question for Dr Arab. When lycopene is added to the adipose tissue stores, I presume it goes in postprandially under the stimulation of insulin. How does it come out? What are the controls? Does it come out postabsorptively as insulin levels drop? It would seem important to know the nature of the physiologic response as to how and at what time of day you should measure these levels in epidemiologic studies and in what relation to meals.
Dr Arab: We know little about the mobilization of anything that is stored in the adipose tissue, including fatty acids themselves. For our purpose, we are using these methods as measures of long-term exposure; it then becomes less of an issue. If our hypothesis was that men with MI [myocardial infarction] are protected from sudden death because they have an available depot of lycopene that can be mobilized at the time of need, it would be a different question.