Toxicology 101

Raw for Beauty.com

When I hear people talk about things being 'toxic', I cringe a bit. Merriam dictionary defines toxic as "containing or being poisonous material especially when capable of causing death or serious debilitation" . It's colloquial usage has deviated far from the true definition, with every food and substance being subject to the term.

Recent concerns regarding the toxicity of food products, like ones made by Food-Babe minded folks discussing chemicals in pumpkin spice lattes, often fail to mention the dose of the questionable compound, and when this is pointed out, they retort that there is a lack of long-term data on the topic in humans and invoke the precautionary principle.

mercola.com

In determining the toxicity of chemicals (be they vitamins, hormones, or environmental chemicals), there are always gaps in the data. Nothing in science is so certain that more data couldn't refine our understanding- indeed, we have no human lifelong data on any of these compounds, be they natural or synthetic - that's why we need a discipline of toxicology. I'm reminded of this 'need' for more data every time I read the Institute of Medicine's Dietary Reference Intakes (DRI). For many vitamins and minerals, we don't have excessive amounts of data across diverse populations for many years, and when a DRI is made, it's always a "well, we think this is what's best" recommendation. Despite this lack of data, we don't see food movements making a big stink about the lack of long-term intake data on 'XYZ' nutrient. Actually, we see the opposite - practitioners like Dr. Joe Mercola tell you to take way more vitamin D than is recommended, and to aim for blood levels where there is evidence for potential harm - see here.

Sense About Science

Obviously, most are not concerned with toxicological outcomes of vitamins and mineral; they're more focused on the potentially undesirable functional additives that have been added to foods. Individuals who call food additives toxic fail to point out that all foods contain components that can be toxic - for both added and endogenous compounds, toxicity is dependent upon frequency of consumption and the dose. Indeed, there are more naturally occurring rodent carcinogens in a single cup of coffee than 'potentially' carcinogenic pesticide residues in the average American diet in a year, and there are still a thousand chemicals left to test in roasted coffee (1). The FoodBabe and CSPI's movement to remove caramel coloring from coffee fail to acknowledge that both coffee and caramel coloring are classified as group 2b carcinogens (2). Where is the foodie movement to ban coffee? It contains untested chemicals! We don't know the lifelong effects! The irony comes from what the #FoodBabeArmy certainly won't tell you - caramel coloring show anti-carcinogenic properties at very low doses (3).

Environmental Working Group.

There are also food movements against the spraying of pesticides onto foods, often failing to take dose into the equation - work by Carl Winter at UC-Davis clearly shows that the levels in food don't pose a threat to humans (4,5,17). Essentially, humans consume pesticide residues at 1/10,000th the levels that show no health effects on animal models. Read below and you'll see that these levels *might* even have positive effects in certain cases. Many will still doubt this work (in some cases, like organophosphates, this may be warranted) and say we don't have long-term enough data and they don't want pesticides in their food. Unfortunately (or maybe fortunately..), if you don't want pesticides in your food, you'll need to only eat animal products - 99.99% of the pesticides you eat are made by plants naturally (6). It is estimated that Americans eat about 1.5 g of natural pesticides per person per day, which is about 10,000 times more than they eat of synthetic pesticide residues. And if you perform toxicology tests on these, they show just as scary effects as the synthetic versions do (some, like glucosinolates found in cabbage, can even break chromosomes at doses relevant to the amount in the plant). Oddly enough, these natural pesticides get to be called "phytonutrients" and their benefits are touted. Toxicity is all about perspective, and double standards abound.

Carcinogenicity data for compounds found naturally in foods.
Many occur at levels that are relevant to their contents in food.

My general point is that everything is toxic at a specific dose. How do we determine that dose though?

 When people think of toxicity, they generally think of linear or threshold models.

This is the most common model and how people that I regularly encounter think of toxicity - either a chemical is always toxic and keeps getting more toxic as you increase the dose, or low amounts pose no risk but at a certain threshold, the chemical becomes toxic.

While this is certainly a valid model, it does not apply universally to 'chemicals'. As a nutritional scientist, I can't rely on this model. When you look at the recommendations for nutrients, they have a U-shaped curve:

This should make intuitive sense - ingesting too little of a nutrient doesn't provide the necessary amounts for health, but ingesting too much leads to negative outcomes. Too little vitamin A during pregnancy can leave your offspring at increased risk of morbidity/mortality from infections due to hypovitaminosis A, but too much vitamin A (retinol) can be teratogenic. Nutrients don'ts act in the linear or threshold manners.

Another response trend also exists- the J-shape. This is known as the Hormetic Dose-Response Model. It is an emerging model that has been demonstrated to be even more common than the simpler threshold dose-response model (7,8).

The hormetic model essentially shows that low doses are beneficial, and as the dose increases, the compound becomes toxic. Examples of this can be found all over the toxicology literature. Low amounts of methylmercury have been shown to improve hatching success of eggs from mallards (9). In the free radical biology field, a concept known as mitohormesis is emerging (10), and forcing scientists to re-think the supposed detrimental effects attributed to oxidative stress.  This thinking is being re-applied to a number of concepts in nutrition (11,12),  including vitamin and mineral intake, energy restriction, acrylamide, pesticides (both natural and synthetic), and alcohol. Low levels of both man-made and naturally occurring chemicals may indeed have beneficial effects for humans.

I should note that hormesis, as it has been applied to BPA exposure, has caused some controversy, as some have suggested that the hormetic effect can be inverted, with low doses causing negative effects not seen at high doses (13,14); These studies have lacked repeatability and might be due to the strains of rats used, as better designed studies have not shown this trend (15,16).

What I hope to demonstrate here is not that we shouldn't be concerned about any chemicals in our environment - I would hope you don't go drinking DDT or super-heating your plastics to get a bit more BPA in - but rather that when someone tells you something is toxic, your next question should be: at what dose and in what model? Do low levels have a negative effect? Might they have a positive effect? And even more importantly, does the person telling me these things are toxic have a knowledge of basic toxicological principles?

Also please note that I'm in no way defending or advising regular consumption of Pumpkin Spice lattes. At 380kcals/16oz, they're a nice seasonal treat to enjoy on occasion, but lack many essential nutrients and fiber.

1. http://www.ncbi.nlm.nih.gov/pubmed/9677052
2. http://en.wikipedia.org/wiki/List_of_IARC_Group_2B_carcinogens
3. http://www.ncbi.nlm.nih.gov/pubmed/21075160
4.http://www.ncbi.nlm.nih.gov/pubmed/19059451
5. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3135239
6. http://www.pnas.org/content/87/19/7777.full.pdf
7. http://toxsci.oxfordjournals.org/content/71/2/246.full
8. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1299203/
9. http://www.ncbi.nlm.nih.gov/pubmed/20821490
10. http://www.sciencedirect.com/science/article/pii/S0531556510001282
11. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2836150/
12. http://www.nature.com/ejcn/journal/v61/n2/abs/1602507a.html
13. http://cancerres.aacrjournals.org/content/66/11/5624.abstract
14. http://www.ncbi.nlm.nih.gov/pubmed/16235731
15. http://toxsci.oxfordjournals.org/content/114/1/1.full
16. http://www.ncbi.nlm.nih.gov/pubmed/19864446
17. http://www.cof.orst.edu/cof/teach/agbio2010/Other%20Readings/Organic%20Food%20Rev%20Inst%20Food%20Technol%20Dec%202006.pdf