Rainbow Of Risks: A look at the potential harmful impact of artificial food coloring agents.
Cover photo credit: Center For Science In The Public InterestThis is a summary of the research titled “Food Dyes: A rainbow of risks” by Sarah Kobylewski, Ph.D. Candidate of the Molecular Toxicology Program at the University of California, Los Angeles, & Michael F. Jacboson, Ph.D. Executive Director for the Center For Science In The Public Interest.
All tables and figures presented are taken directly from the publication listed above. Much of the content in the summary below was taken directly from the publication, with only the introduction including some of my own ideas.
Food Dyes: A Rainbow Of Risks
This research delves into the major food dyes currently approved by the FDA and used readily today. Although this research doesn’t discuss chocolate specifically, these colours and “lakes” (insoluble pigments) are the same ones used readily in the chocolate industry, including high end chocolate shops and even within the craft bean to bar chocolate sector to make beautiful looking bonbons.
DISCLAIMER: The purpose of this summary isn’t to demonize or belittle those who consume chocolate with artificial dyes or use them in the making of their chocolate confections. Rather, it’s intended to make all of us aware of potential risks you may not have known about regarding synthetic dyes, or to offer more tangible evidence to reinforce any vague ideas you may have on the matter. This information may help you consider whether using dyes on your chocolates or consuming chocolates with these dyes is worth the potential risks. The debate isn't settled, as you will see from reading this summary. As well, it may help nudge you to question, from a craftsman’s point of view, why use these dyes in the first place. If bean to bar chocolate is intended to focus on flavour and quality of ingredients above all else, does adding synthetic dyes complement these values or work contrary to them? It’s something I question myself very often working both as a bean to bar maker/educator and chocolatier. This paper analyzes 9 dyes approved by the FDA as safe to use, and the various studies associated with each dye, the outcome of the studies, as well as faults with the research or lack thereof. The researchers analyze the impacts on health in regards to:
Carcinogenicity (how long-term exposure may cause cancer)
Genotoxicity (how the dyes may cause mutations or damage to chromosomes/DNA, which indicates the dye may cause cancer)
Neurotoxicity (harmful impacts on the brain and central nervous system)
It’s interesting to note that most of the studies reviewed here, which are few to begin with, are over 30 years old. It may be helpful to repeat these studies or conduct new ones with the updated technology we have today. As well, these 30 year old studies don’t necessarily take into consideration the increased consumption levels of synthetic dyes in processed foods consumed worldwide today.
Some of the jargon here may require you to search what the terms are. I do my best to briefly give you an idea of what some of them mean. Two terms to keep in mind are in vitro vs In vivo.
In vitro (Latin for “in the glass”) are tests are conducted on single cells or bacteria within a culture.
In vivo (Latin for “within the living”) are tests conducted on fully functioning live animals, not just cells.
Many of the citations for the following information can be found in the full paper listed below! If you enjoy the summary here and want to delve deeper, I encourage you to download it and read over it yourself.
To read and download the full version of this research paper in .pdf form, please Click Here.
What are food dyes?
Artificial food dyes (AFDs) were originally synthesized from coal tar, but now produced from petroleum. Food dyes are not used simply to make our food look pretty, but are used by manufacturers to help compete in the market and sell their products. AFDs are more economical than natural dyes, and improve the appearance of generally bland looking foods (starches, brown chocolate) while grabbing the consumer’s attention. Essentially it makes foods that lack colour more appealing.
This is one reason why you see so many images online and in social media repeating the same foods (often bland on their own) in different colours, such as pasta with colourful sauces in various shades of red, green, and yellow, and why fine dining establishments add a twig of green and a petal of red to a dish. It’s why you see this push for natural food dyes made from algae to produce bright blue or green oatmeal bowls, perfect for that next instagram shot! The point is, our brains are wired to seek out these colourful looking foods. Since humans first walked the earth, colour gave us information on the nutritional content of the foods available, and our brains work in the same manner today. Colourful foods rule our world!
Why not use natural food dyes?
They’re expensive, they have shorter shelf lives (their colour fades more easily), and don’t have the same impact visually as artificial food dyes. When it comes to chocolate and confections, it’s very difficult to use natural food dyes. I’ve used natural dyes on chocolate before, and they always appear more dull and muted against the brown chocolate as opposed to AFDs. As well, natural colours are less stable over time and begin to fade and mute especially if exposed to sunlight. Since solid chocolate can last for months or years without spoiling, they would require the colours to last just as long without fading. For instance, using matcha powder to colour white chocolate, as a way to create chocolate leaves for a showpiece, is more of a challenge than if one were to mix the white chocolate with green AFDs. The matcha powder coloured white chocolate will fade dramatically within a few weeks while exposed to sunlight, while the chocolate with AFD will stay vibrant much longer.
So what are some natural food dyes? Some natural food colouring agents include:
beta-carotene
paprika
turmeric powder
freeze dried fruit (beet, raspberry, passion fruit)
algae powders (blue spirulina)
matcha tea powder
These are just a few ideas. You can mix them with white chocolate, or with cocoa butter to spray onto any chocolate. A downside is using too much will impart the flavour of that natural dye. It’s not impossible to use these with fine results, but if placed beside products with AFDs, its hard to compete as they would not be as vibrant. However, you can argue bright blue and neon orange foods may look more artificial and less appetizing to some as well. But I digress.
As well, chocolate requires fat-based food dyes, that can be mixed with cocoa butter. This is more of a challenge when it comes to natural food dyes. Mixing turmeric or beet powder with cocoa butter will be a bit grainy (versus AFDs) and make it more of a challenge to spray on if using a spray gun (the bits of powder clog up the passages in small spray guns). These freeze dried powders are also quite expensive. Using water based food dyes, such as extracting colour by boiling foods, is not possible when colouring solid chocolate because of the water content. Something to keep in mind.
To use or not to use artificial food dyes?
Artificial food dyes, unlike many other food additives, add absolutely no nutritional or health benefit. It’s purpose is to market the products. Some people may view artificial food dyes as harmless, and are indifferent to consuming them. However, it’s long been argued that most or all of the food dyes pose some health hazard (as you will read below), and are not worth the risk. There are many foods and drugs we consume (or over consume) that we know are not particularly healthy, but continue to do so. As well, what’s harmful for one individual may not be as harmful to another.
For hundreds of years, chocolate has been adulterated in many ways. Adding suet or brick dust back in the 16th & 17th Century, removing cocoa butter and adding palm kernel oil during the 20th Century, or adding copious amounts of sugar to mask poor quality cacao were a few of the ways chocolate has been adulterated. Food dyes can be viewed as a form of adulteration. According to the Food, Drug, & Cosmetic Act, Sections 402(b)(3) and (b)(4), they state “A food shall be deemed to be adulterated… (3) if damage or inferiority has been concealed in any manner; or (4) if any substance has been added thereto or mixed or packed therewith so as to… make it appear better or of greater value than it is.” It may be fair to apply this to the use of food dyes in our chocolate.
You may argue that it is large processed food companies forcing these artificial food dyes (AFDs) on the consumer, but some large corporations have stated they will use them until the consumer says otherwise. I have worked as a chocolatier for companies that use plenty of AFDs, and I know for a fact that if they got rid of them, many of their customers would rather go elsewhere. The added colour not only looks pretty, but can also look artistic and even “expensive”, which is an important consideration for consumers since most chocolate is purchased as a gift or to share with friends and family.
That said, it is up to us to be informed and begin voicing our desire for less or no AFDs in our food. I’m not going to hide that I am biased in that I do prefer to not use colour, I will also acknowledge that I have and continue to use it. That said, I would love continue my livelihood creating chocolate, but without the use of AFDs, and spreading this information is a great way to begin to shift the minds of my current and future customers.
Summary of the food dyes
So with all this in mind, we will return to the study here. This is a summary of a paper that summarized other research papers, so the information here will be brief. It’s not here to “answer” all your questions, but rather to make you aware and help you come up with better questions you may want to ask and further look into.
This table shows 15 million pounds or 6.5 million kilos of food dye was certified by the FDA in 2009.
The 9 food dyes discussed in the paper are listed in Table 1. For the purpose of this summary, the dyes summarized here will include Blue 1, Blue 2, Green 3, Orange B, Red 40, Yellow 5, and Yellow 6. I wont include summaries on Citrus Red 2, Orange B, and Red 3 mostly because they are rarely used in the lakes (pigments) of chocolate related dyes. You can see in Table 1 that the most used food dyes include Red 40, Yellow 5, and Yellow 6.
Below we will go over the 6 food dyes one by one, with a short summary for each one. Not all the information from the main paper will be included, but rather some points which are worth noting.
Please check out the full paper to read more in detail about the studies, and even follow up with some of the references you may be curious to know more about. Below will be my summary of this paper and a few tables which organize all this information very well.
This table offers an overview of all the dyes studied in of the research paper, highlighting potential risks of each dye including allergic reactions, carcinogenic contaminants, and cancer test findings.
Blue 1
Generally not found to be toxic in studies on rats and mice, although the only thorough study was sponsored by the dye industry. An unpublished study suggested that it may cause kidney tumors in mice, and a preliminary in vitro study raised questions about possible effects on nerve cell growth. Blue 1 may not be found to cause cancer, but more studies would be needed to confirm this.
Metabolism
96% is excreted in the feces within 36 hours of consumption, with only a few percentage found in the bile, which indicates Blue 1 is poorly absorbed by the gastrointestinal (GI) tract, and therefore isn’t broken down by the intestinal microorganisms, but about 5% is absorbed via the GI tract.
Genotoxicity
Blue 1 appears to test positive for chromosomal aberrations as seen in Table A1 below. Chromosomal aberrations are essentially errors during DNA replication, such as a missing or additional chromosomes. Testing for chromosomal aberrations is way to to predict for potential carcinogenicity.
Chronic Toxicity/Carcinogenicity
Two studies appear to be inadequate to confirm anything in favor of or against the use of Blue 1. The studies on both rats and dogs consisted of a very small sample size, and only lasted 1-2 years, which isn’t long enough to consider long term effects.
One study on mice showed that 7 out of 30 surviving male mice developed kidney tumors, compared with only 1 out of 45 in the control group (mice who didn’t consume Blue 1). However, there was no dose-response relationship, which means the rate of kidney tumors didn’t increase with greater amounts of Blue 1, or decrease with lesser amounts, so it suggest perhaps it’s not Blue 1 which is directly involved in creating the kidney tumors.
Another study by Borzelleca et al. for the Certified Color Manufacturers Association (CCMA) found that female rates in the group fed the highest amount of Blue 1 had a 15% body weight reduction vs the control group not fed any Blue 1, and also a decreased survival rate to the controls.
Neurotoxicity
As a nerve cell grows it produces long extensions in the form of dendrites or axons, known as neurite growth. These extensions interconnect nerve cells with other cells. Neurotoxicity measures the extent of neurite growth. If growth is inhibited, it may indicate a neurotoxic is involved.
Ingestion of Blue 1 showed 20% neurite growth inhibition, meaning it appeared to inhibit neurite growth overall. When Blue 1 was combined with L-glutamic acid (an amino acid that humans can create themselves to build proteins) the synergistic impact of both the Blue 1 and L-glutamic acid on the neurites was a 46.1% neurite growth inhibition, more than double Blue 1 on its own. This study raises concerns of Blue 1 inhibiting neuron growth.
The metabolism of dyes hasn’t been studied much if at all in children. The blood-brain barrier (which protects the brain from harmful substances ingested into the body and find their way into the bloodstream) is not fully developed in children until 6 months. Synaptogenesis (a brain growth-spurt) occurs in children from 3 months before birth to a few years old, during the time the blood-brain barrier is not fully formed. We know a small amount of Blue 1 is absorbed by the GI tract in rats (as stated in the “metabolism paragraph above), so there is a chance Blue 1 can cross the blood-brain barrier of children, and if so, we don’t know what impact this may have on their brain development during this early stage in their life.
Blue 2
Blue 2 was listed as safe by the FDA in 1983, and used in beverages, candies, pet foods, other foods, and drugs. It is shown to induce the formation of tumors such as brain gliomas in male rats. Gliomas are a type of tumor that starts in the glial cells of the brain or spine. The dye is broken down in the GI tract, but no metabolism studies in humans have been conducted. Two toxicity/carcinogenicity studies were too short and didn’t involve an in utero phase in order to be satisfied with their results. In utero studies are important because although certain substances may only mildly harmful or not harmful at all in adults (think alcohol or certain medications) may be extremely harmful during for a developing fetus, and have lifelong implications.
Metabolism
Studies in rats show that most Blue 2 dye is excreted in the feces (and not absorbed), with small amount of absorbed dye found in the urine.
Genotoxicity
All studies shown in Table A2 were negative except for a chromosomal aberration (DNA mutation) assay, which the author suggests an independant lab to repeat the study and perform an in vivo (in a living organism) chromosomal aberration test.
Chronic Toxicity/Carcinogenicity
A study by Borzelleca et al. (1984-86) suggests possible evidence of carcinogenicity. Male rats on Blue 2 displayed an increase in the incidence of transitional cell neoplasms (a mass of abnormal tissue, either cancerous or non cancerous on the urinary bladder). This hints that it may cause cancer of the bladder.
Male rats in the 2% group (the group with highest consumption level of Blue 2) showed a significant increase in:
cancerous mammary-gland tumors
and brain gliomas (tumors associated with the brain or spine).
The authors of this study concluded this increase was not related to Blue 2 for a few reasons: no dose-effect relationship, no decrease in survival time, and that the incidence of gliomas in the rats who consumed Blue 2 were consistent with “historical controls” (old data is used to compare to new data from a current study). The last point is more contentious. Essentially, the rate of tumor growth in this study was much higher than the control group in this study (rats who didn’t consume Blue 2), signaling perhaps Blue 2 did induce tumor growth. However, the authors of this study compared the rate of tumor growth in the test subjects with older data from other control groups (known as “historical controls”). If these historical controls from previous studies had a higher incidence of tumor growth, then by comparison, the rate of tumor growth on test rats in the current study wouldn’t appear as high.
In the end, the FDA approved Blue 2 as non cancerous based on the argument comparing it to historical data, lack of proof in regards to dose-related trends, lack of non-neoplastic cellular changes, no reduction in latency period, no varying progression of brain tumors, the inability of Blue 2 to cross the blood-brain barrier in adults, negative mutagenicity assays, and lack of evidence in structure-activity analysis (FDA 1983).
However, Blue 2 did show statistically significance of the male rates in the high dose group of the study above. As well, other consultants stated that these findings “cannot be dismissed as accidental” (Robert Squire) and Dr. William Lijinsky, a cancer specialist, stated that in his own laboratory “…this would be considered prima facie evidence of carcinogenicity of a treatment. This is especially so because the tumor is so rare, and my conclusion is that Blue 2 is a carcinogen, and should be regulated accordingly.” However, other consultants/scientists disagreed with this and agreed in favour with the FDA.
Reproductive Toxicity and Tetratogenicity
A reproductive study conducted by Borzelleca et al. on Blue 2 in rats saw that parents and pups were normal in terms of appearance and behavior, with no impact on fertility, length of gestation, viability, or lactation. A tetraogenicity (study of physical development) study also saw no adverse effects of Blue 2.
Green 3
Also known as Fast Green FCF, is a dye used in food (candies, beverages, dessert powders, ice cream, sorbet), drugs, personal care products, and cosmetics. Shown to cause increase in bladder and testes tumors in male rats. Because of this, further testing should be conducted.
Metabolism
In rats, 94% of the dye was excreted in the feces, and no traces were found in the urine. In dogs, no colour was found in the urine and only 2% found in the bile of 2 of the 3 dogs (Hess & Fitzhugh).
Genotoxicity
Overall, the tests didn’t raise any significant concerns.
Chronic Toxicity/Carcinogenicity
Various doses of Green 3 were offered to rats before mating, and then their offspring (F1 generation) were also fed doses of Green 3 for about 30 months. Pups born to parents of the mid- and high-dose groups had a higher mortality rate, and a decrease in survivorship was seen in all treated groups, but there was no dose-response trend. All treated groups appeared to be negatively impacted, but the higher dose groups weren’t necessarily impacted more than lower dose groups, so its difficult to interpret these results.
Histopathological examination (looking at body tissue under a microscope) showed the high-dose group of male rats had:
greater instances of urinary bladder neoplasms (abnormal tissue growth that may be cancerous)
testes Leydig’s cell tumors (rare in humans)
and liver neoplastic nodules (aka “tumors” that may or may not be cancerous).
A statistician, Mark Nicolich, stated in regards to this study in 1982: “Therefore, there is statistical evidence that the high dose of the test material increases the occurrence of certain types of tumors in rats.”
However, the FDA scientists concluded that the tumors in the testes were not compound-related, and the urinary bladder neoplasms were not statistically significant (although the “not statistically significant” was added as an addendum before the final submission of the paper, after already stating rats had a significantly increased incidence of benign tumors.)
Red 40
The most-widely used dye, also known as Allura Red. It is approved to use in beverages, bakery goods, dessert powders, candies, cereals, foods, drugs, and cosmetics.
It may accelerate the appearance of immune-system tumors in mice, hypersensitivity reactions in some consumers and potential hyperactivity in children.
The evidence for red 40 is controversial and inconclusive at best. For the most widely used dye in the world, there doesn’t appear sufficient data to consider it completly safe.
Metabolism
In an unpublished report (White, 1970) 29% of the intact dye was excreted in the feces, and 0.1% was excreted in the urine. This means the rest of the dye appears to be broken down in the gut into cresidine-4-sulfonic acid (a carcinogenic contaminant) and 1-amino-1-naphthol-6-sulfonic acid.
In another study, after 72 hours from ingestion, 92-95% was recovered in the feces of dogs (5.7-19.8% in urine), and 76-92% (2.7-3.6% in urine) was recovered from the feces of the rats.
They found a significant amount of Red 40 in the guts of sacrificed animals. However, the author states that there isn’t enough published metabolism data, and more need to be conducted.
Genotoxicity
Table A5 displays the positive in vivo comet assay (a test measuring DNA strand breaks) of glandular stomach, lungs, and colon cells of mice, indicating Red 40 can cause DNA damage in vivo (in living animals), not just in vitro.
Hypersensitivity
Patients who were suffering from urticaria (hives) and angiodema (swelling) for more than 4 weeks were put on a 3-week elimination diet. When Red 40 was ingested in doses of 1 or 10 mg, it induced hypersensitivity in 15% of the patients who were symptom free at the time.
Chronic Toxicity/Carcinogenicity
Hazleton Laboratories in the 1970’s found that in a Red 40 feeding study, the only statistically significance was a decrease in body weight of females in the high-dose Red 40 group, but no other consistent adverse effects.
They also performed chronic toxicity studies in mice by adding Red 40 to their food one week before breeding, during gestation, and during lactation. The pups were sampled at random, and at 42 weeks of age, 6 reticuloendothelial (RE) tumors (tumors related to the immune system) occurred in both males and females (0 tumors in the control group, 1 in each of the low and mid-dose, and 4 in the high dose groups). Some of the mice were then examined at 104 weeks, continuing to be fed Red 40, but they didn’t see any acceleration of the appearance of the RE tumors and it was concluded that Red 40 didn’t cause acceleration in RE tumors.
However, pathologist Dr. M. Adrian Gross stated there was clear evidence to support an acceleration effect of RE tumors, mainly because there was a decreased latency period even though there was no increase in overall tumor incidence. This means that there may not have been more tumors, the time between being exposed to red 40 and developing tumors was decreasing, so occurring more quickly. However, it should be noted that they also found a small number of RE tumors in all treatment groups before the 42 week point, with the highest incidence of tumors being in the high-dose group. As well, a problem with this study was also that after sacrificing 36% of the mice at 42 weeks, it left a small number of mice at the end of the study, reducing the ability to properly analyze tumor incidence rate.
A second mouse study, similar to the one above but different in some aspects, only found an increase in relative and absolute thyroid weight in the high-dose male and female mice. However, there were problems with this study. Two problems include caging and litter effects. Mice housed in the upper row of racks experienced higher incidences of RE tumors, than mice in the lower ages, likely due to not rotating the cages (rotating cages is a normal procedure in test labs). It’s also unsure to know if mice were housed with siblings, which might have had an influence on tumor incidence. According to the authors here, these two issues decrease the credibility of the study.
There was a great difference in the RE tumor rates between the two mouse studies, limiting the conclusiveness of the results. Lagakos and Mosteller concluded both studies on mice suggested that mice treated with Red 40 saw both a decrease in latency period and increase in incidence of RE tumors, so something worth looking into further.
Carcinogenic contaminats:
Red 40 might contain cancer-causing contaminates. Health Canada scientists identified amounts of aniline, p-cresidine, and 1-naphthylamine in the dye. The International Agency for Research on Cancer (IARC) notes p-cresidine to be possibly carcinogenic to humans. FDA considers aniline to be weakly carcinogenic to rats.
Reproductive toxicity/Tetragenicity:
A study on female rats saw no negative effects on maternal reproduction, embryolethality, or fetoxicity observed.
Yellow 5
Also known as Tartrazine, is used in baked goods, beverages, dessert powders, candies, cereals, gelatin desserts, pet food, pharmaceuticals, and cosmetics.
Not carcinogenic in rats, but wasn’t adequately tested in mice to confirm the same. It may be contaminated with cancer-causing chemicals such as Benzidine (linked to bladder and pancreatic cancer), & 4-amino-biphenyl. It may cause sometimes-severe hypersensitivity reactions in a small number of people, and trigger hyperactivity in children.
Metabolism
Yellow 5 is metabolized (broken down) by the gastrointestinal flora into its metabolite sulfanilic acid. Sulfanilic acid can be found in the urine of rats, rabbits, and humans after Yellow 5 was administered.
Genotoxicity
Table A6 lists 6 out of 11 studies were positive for genotoxic effects, including chromosomal aberrations (a change in the structure of the chromosome, such as extra or missing parts of chromosomes). Sasaki et al. demonstrated that Yellow 5 induces DNA damage in vivo.
Chronic Feeding/Carcinogenicity
In the earliest study, the sample size of rats were well below the FDA recommended minimum, and the rats were not exposed in utero.
A well-designed study by Borzelleca and Hallagan on rats found Yellow 5 did not show any carcinogenic or toxic effects. The same scientists also performed a chronic toxicity/carcinogenicity study on mice. However, in this study, the mice were not exposed in utero, and were 42 days old at the start of the study, which was seen as a serious drawback. In the end, they didn’t find any significant compound related effects in this study either.
Carcinogenic Contaminates
Yellow 5 may be contaminated with benzidine and 4-aminobiphenyl. The FDA limits free benzidine to 1 part per billion (ppb). FDA tests in an early 1990s study found some batches of Yellow 5 contained up to 83 ppb of free and bound benzidine (attached to other substances and therefore less active). The FDA doesn’t test for bound benzidine in Yellow 5, but this bound benzidine may become free after being metabolized in the gut, which we know it is.
In 1985, the FDA calculated a risk assessment of Yellow 5, made with projections of consumption levels of the 1990s. They found 4 cancers in 10 million people, smaller than the “concern” level of 1 cancer in 1 million people. However, the risk assessment failed to consider:
greater sensitivity of carcinogens to children
greater consumption of Yellow 5 by children (candies, processed snacks) vs by adults
substantial increase in per capita consumption of Yellow 5 since the 1990s
that some batches of Yellow 5 can contain high amounts of bound benzidine and other carcinogens
presence of similar contaminants in Yellow 6
It’s also important to note that foods and dyes imported by China, India, and other countries may not have been routinely tested by the FDA, and may have bound contaminants such as benzidine.
Hypersensitivity
Yellow 5 has had documented incidences of sensitivity which shows up as hives or asthma. Neuman et al. reported that 26% of patients with a variety of allergic disocerders had a positive reaction (heat-wave, general weakness, blurred vision, increased nasopharyngeal secretions, feeling of suffocation, palpitations, severe itching, swelling under the skin, and hives) 10-15 minutes after ingesting 50 mg of Yellow 5.
Tartrazine is also used in many medications. A study between aspirin-intolerance and Tartrazine-sensitivity by Stenius and Lemola demonstrated that about 50% of patients with positive reactions to aspirin also had positive reactions to Yellow 5.
A few examples of the severity of sensitivity to Yellow 5 listed in the full publication of this paper include individual cases of:
severe asthma and being hospitalized, lost off taste and smell
anaphylactic shock after receiving an enema with Yellow 5 and 6
swelling, hives, purple skin spots
purple skin spots, ulcerations, pain, and swelling of the legs
Yellow 6
Also known as Sunset Yellow, and is used in baked goods, cereals, beverages, dessert powders, candies, gelatin desserts, sausage, cosmetics, and drugs.
Shown to cause adrenal tumors in animals, but disputed by the industry and FDA. It may be contaminated by cancer-causing chemicals such as in Yellow 5, and occasionally causes severe hypersensitivity reactions (swelling, hives, stomach cramps) in well documented cases.
Metabolism and Metabolic Effects
Metabolites were found mostly in the urine of rabbits who ingested 0.5 mg/kg of Yellow 6. It was broken down by the gut flora into metabolites: sulfanilic acid, 1-amino-2-naphthol-6-sulfonic acid, and p-acetamidobenzene-sulfonic acid. Only 1-2% of the intact dye was found in the feces, which means most of it is broken down. In another rat study, 0.8% of intact dye was found in the feces, the rest being found as the metabolites.
Genotoxicity
Yellow 6 was negative in most cases except it did induce forward mutations and chromosomal aberrations in two tests.
Chronic Toxicity/Carcinogenicity
The National Toxicology Program (run by the US Department of Health) conducted carcinogesis studies in rats and mice in the early 1980’s. Unfortunately, neither test included in utero exposure. In the rat experiment, they found no statistically significant Yellow 6-related lesions (lesions are parts of an organ or tissue that were damaged in some way such as by tumors). In the mouse study, only the low-dose group of male mice saw a higher incidence of liver cancer and non-cancerous tumors. Because (amont other reasons) the mice in the high-dose didn’t experience equal or greater rates of this, therefore not dose-dependent, they concluded Yellow 6 was not carcinogenic for the mice as well.
In a 1982 study by Bio/dynamics Inc. on mice found that compared to controls:
females in the high-dose groups experienced an increase relative kidney weights
males and females in the high-dose groups experienced an increase in thyroid weights in
males and females in the high-dose groups experienced a statistically significant increase in incidences of non-cancerous tumors of the adrenal glands (although these tumors can still alter how the adrenal glands function)
males in the high-dose group had an increase in testicular tumors
In spite of all this, the investigators concluded no evidence of carcinogenicity.
Some of the reasons the FDA concluded tumors were not related to Yellow 6 included:
lack of dose-response in the high dosage groups
likelihood of false positives
lack of precancerous lesions
similar appearance of adrenal medullary lesions in the control and the Yellow 6 groups
lack of difference in the latency periods before tumors occurred
tumors seen are common spontaneous tumors in older rats
lack of other studies finding any association between Yellow 6 and tumors
In the 1960’s, the FDA conducted a 7-year Yellow 6 feeding study on a small number of beagles. The authors state this study wasn’t long enough to be considered a valid study. It’s important to take away that an FDA veterinarian, Kent J. Davis, found “tears, eye lid encrustations, pannus (corneal inflammation), and corneal opacity approaching blindness” to beagles who ingested Yellow 6. He concluded that because of all this “…immediate desertification of this color is necessary in order to protect the public health…”
carcinogenic Contaminants
Similar to Yellow 5, Yellow 6 can be contaminated with benzidine (linked to bladder and pancreatic cancer) and 4-aminobiphenyl (although its commercial production was banned in the US decades ago). Note that FDA tests for free benzidine in the dye batches, but there may be benzidine bound to other substances that is released when the dye is metabolized after ingestion. The authors here also note that, just as with Yellow 5, the risk assessment for these contaminants don’t consider:
sensitivity to children
greater consumption of Yellow 6 by children
increase in per capita consumption of Yellow 6 since these assessments were placed in the 1990’s
FDA scientists state that in 1992, one company eliminated benzidine contamination of Yellow 5. However, Health Canada found that Sunset Yellow (AKA Yellow 6) was still contaminated in a 1998 study. The authors point out that many dyes are being imported from other countries that manufacturer them, and these dyes may not be tested to the same extent.
Hypersensitivity
Human hypersensitivity to Yellow 6 has been reported since 1949. Some cases included:
anaphylactic shock
stomach cramps
skin lesions, indigestion, retching, belching, abdominal pain, and vomiting
A study in 1973 by Michaelsson and Juhlin on human patients offered some insight into sensitivity at different doses of Yellow 6. Symptoms observed after ingestion of various levels of Yellow 6 included:
hives
swelling of the lips, eyes, or face
reddening of the eyes
sweating
increased tear secretion
nasal congestion
sneezing
runny nose
hoarseness
wheezing
The control group had no history of recurrent hives, and only 2 out of 33 patients showed signs of rhinitis (swollen, stuffy, runny nose) when administered Yellow 6 and Yellow 5. Of the 27 patients with recurrent urticaria (hives) administered Yellow 6, 10 developed hives, 6 developed subjective symptoms, and 11 were negative for symptoms. 8 out of 9 patients with positive reactions to Yellow 6 also experienced positive reactions to aspirin (people sensitive to Yellow 5 are also often sensitive to aspirin).
in 1974 Michaelsson et al. tested patients with 5 mg of Yellow 6 who suffered from purplish spots (vascular purpura). One woman, who suffered for 12 years with these lesions, had a strong positive reaction to the dye. After being on a dye-free (and benzoates-free, a preservative) diet for 6 months, she became essentially free from the lesions.