biochemistry at Virginia Tech, an upper-level course in nutritional biochemistry at Cornell, and two new graduate-level courses in biochemical toxicology and molecular toxicology for a new graduate field of toxicology, also at Cornell. Like other faculty in these fields, I followed the typical textbook model of lecturing, mostly focusing on individual nutrients, individual toxic chemicals, individual mechanisms of action (i.e., biochemical explanations), and individual effects, as if there were, for each nutrient or chemical, one main mechanism that explains and perhaps controls the relationship between cause and effect.
When I taught nutrition in this traditional, reductionist way, here’s how it went. We began by considering the chemical structure of the nutrient. Then we discussed how it functions in the body: its absorption across the intestinal wall into the blood; its transport through the body; its storage; its excretion; and the amounts needed for good health. We talked about each nutrient on its own, as if it acted alone in a totally mechanical fashion. In other words, teaching nutrition meant getting students to memorize facts and figures and chemical pathways to pass tests without asking them to think about the context for these discrete bits of information.
We do the same thing in research as we do in education. The gold standard of nutritional research—the type that receives preference for funding and gets published in top-line journals—focuses on one nutrient and one explanation of its effect. My experimental research program focused on the effects of discrete causes, reactions, enzymes, and effects, oftentimes outside of the context of the body as a whole—in part because, as I mentioned, I, too, was taught to think this way, 1 but also because, in order to get research funding, we scientists are forced to focus our hypotheses and experimental objectives on outcomes that can be measured.
Let me give you a specific example from the initial stages of my own research on cancer formation initiated by aflatoxin (AF), a chemical known to cause liver cancer. (As you may recall from the introduction, AF was the carcinogen produced by the peanut fungus I was looking at in the Philippines.) Figure 5-1 summarizes the process we were studying (using a diet of 20 percent casein, or milk protein).
My lab research at this early stage was completely acceptable according to the reductionist rules. We focused on one kind of carcinogen (AF) that caused one kind of cancer (hepatocellular liver cancer) that depended on one kind of enzyme (mixed-function oxidase) that metabolized AF to produce one kind of highly reactive product (AF epoxide) that produced one biochemical effect (the very tight chemical bonding of the epoxide to DNA that causes genetic damage), each stage of which seemed internally consistent and biologically plausible. And we discovered that the more the carcinogen bound itself to the DNA, the greater the amount of cancer occurred. 2 Aha! This was the mechanism that “explained” the effect of protein on cancer!
FIGURE 5-1. A linear model of cancer causation from aflatoxin
A couple of thoughts about the previous paragraph: first, I don’t expect you to understand everything I wrote. I’m describing complex biological and chemical reactions in the kind of specialized language used by scientists everywhere to communicate with precision. All you need to know is that, according to this model, A causes B, which causes C, which leads to D. So the more A (cancer-causing chemical) you start with, the more D (cancer) you end up with.
Second, it probably sounds pretty convincing, even if you don’t really understand it. Research like this seems airtight because it deals with objective facts—reactions, genetic mutations, and carcinogenesis—as opposed to messy things like human behavior and lifestyle. Only by excluding messy and complex reality can we make linear, causal statements about biological chain
When I taught nutrition in this traditional, reductionist way, here’s how it went. We began by considering the chemical structure of the nutrient. Then we discussed how it functions in the body: its absorption across the intestinal wall into the blood; its transport through the body; its storage; its excretion; and the amounts needed for good health. We talked about each nutrient on its own, as if it acted alone in a totally mechanical fashion. In other words, teaching nutrition meant getting students to memorize facts and figures and chemical pathways to pass tests without asking them to think about the context for these discrete bits of information.
We do the same thing in research as we do in education. The gold standard of nutritional research—the type that receives preference for funding and gets published in top-line journals—focuses on one nutrient and one explanation of its effect. My experimental research program focused on the effects of discrete causes, reactions, enzymes, and effects, oftentimes outside of the context of the body as a whole—in part because, as I mentioned, I, too, was taught to think this way, 1 but also because, in order to get research funding, we scientists are forced to focus our hypotheses and experimental objectives on outcomes that can be measured.
Let me give you a specific example from the initial stages of my own research on cancer formation initiated by aflatoxin (AF), a chemical known to cause liver cancer. (As you may recall from the introduction, AF was the carcinogen produced by the peanut fungus I was looking at in the Philippines.) Figure 5-1 summarizes the process we were studying (using a diet of 20 percent casein, or milk protein).
My lab research at this early stage was completely acceptable according to the reductionist rules. We focused on one kind of carcinogen (AF) that caused one kind of cancer (hepatocellular liver cancer) that depended on one kind of enzyme (mixed-function oxidase) that metabolized AF to produce one kind of highly reactive product (AF epoxide) that produced one biochemical effect (the very tight chemical bonding of the epoxide to DNA that causes genetic damage), each stage of which seemed internally consistent and biologically plausible. And we discovered that the more the carcinogen bound itself to the DNA, the greater the amount of cancer occurred. 2 Aha! This was the mechanism that “explained” the effect of protein on cancer!
FIGURE 5-1. A linear model of cancer causation from aflatoxin
A couple of thoughts about the previous paragraph: first, I don’t expect you to understand everything I wrote. I’m describing complex biological and chemical reactions in the kind of specialized language used by scientists everywhere to communicate with precision. All you need to know is that, according to this model, A causes B, which causes C, which leads to D. So the more A (cancer-causing chemical) you start with, the more D (cancer) you end up with.
Second, it probably sounds pretty convincing, even if you don’t really understand it. Research like this seems airtight because it deals with objective facts—reactions, genetic mutations, and carcinogenesis—as opposed to messy things like human behavior and lifestyle. Only by excluding messy and complex reality can we make linear, causal statements about biological chain
Similar Books
