Genetic testing can give us tremendous insight into our own health and, more importantly, our optimal diet and lifestyle choices given our genetic predisposition. I’ve long held that the Paleo diet is far from a one-size-fits-all approach and try to steer away from Paleo dogma in all of my writing. In this article series, I discuss specific genes that can tell us a lot about disease risk and how we can modify our implementation of Paleo to mitigate that risk. You may be able to request genetic testing through your doctor (my functional medicine doctor ran genetic screening on me as part of my first evaluation) or you can order a genetic test to do at home. I recommend 23andMe which you can purchase here.
When it comes to Alzheimer’s and heart disease, one of the most important genes to know about is ApoE. ApoE codes the protein apolipoprotein E (or apoE), which plays a major role in how we metabolize and transport fat, cholesterol, and fat-soluble vitamins. Three main apoE variants exist (E2, E3, and E4), and all of us carry a combination of any two of those variants (so, our apoE phenotype can be apoE2/2, apoE2/3, apoE2/4, apoE3/3, apoE3/4, or apoE4/4). (There are also two extremely uncommon variants, E1 and E5, but they’re found in less than 0.1% of the population and aren’t commonly studied!)
Although each apoE variant differs only by one or two amino acids in its molecular structure, those tiny differences are enough to change how apoE functions in our body in a big way! As abundant research shows, our specific apoE phenotype can have a major impact on our disease risk, our response to saturated fat, our response to alcohol, our response to omega-3s, our ability to detox heavy metals, our ability to heal from traumatic brain injury… and more!
While apoE3 is considered neutral (it neither raises nor protects against apoE-related conditions) and is the most common variant (about 55% of the population carries two copies of apoE3 and another 40% carries one copy!), apoE4 and apoE2 come with some important quirks. In particular, apoE4 has grabbed attention as a major risk factor for Alzheimer’s disease (along with other neurological and cardiovascular conditions), while apoE2 appears to protect against Alzheimer’s disease (although it can raise the risk of other conditions at the same time). Let’s take a closer look at these two important variants and what they mean for our health!
ApoE4 is believed to be the oldest or “original” allele (it’s closest to the apoE variant that nonhuman primates have, suggesting it existed even before the human and chimpanzee lineages split), with apoE2 and apoE3 developing as mutations more recently in human history. Approximately 1-2% of the population is homozygous for apoE4 (meaning they carry two copies, AKA apoE4/4), and an additional 28-29% of the population carries one apoE4 copy (apoE3/4 or apoE2/4). But, this variant can be much more common among specific groups—especially hunter-gatherer societies like the Pygmies, Khoi San, Malaysian aborigines, Australian aborigines, Sami, Papuans, the Greenland Inuit, and some Native Americans. (One reason for apoE4’s greater frequency in these populations is that it enhances cholesterol absorption in the intestine and slightly raises blood cholesterol levels, which can be beneficial during times of food scarcity, undernutrition, and other conditions that foraging societies frequently face.) Not surprisingly, areas with longer-standing agricultural histories, where food supply is stable and apoE4 is less useful, have a lower prevalence of apoE4 (for example: southern Europe, the Middle East, and Southeast Asia).
Although the increase in certain disease risks is more dramatic for apoE4 homozygotes, anyone with at least one apoE4 copy faces a higher likelihood of developing certain conditions due to the way apoE4 functions in the body. Those risks include:
Alzheimer’s disease. Being an apoE4 carrier is the biggest known genetic risk factor for developing late-onset (after the age of 60) Alzheimer’s disease. In fact, carrying one apoE4 copy doubles or triples your risk of developing Alzheimer’s, and people with two apoE4 copies have at least a 10-15 times higher risk (compared to non-apoE4 carriers)! Researchers are still trying to pinpoint exactly how apoE4 contributes to this disease, but the link appears related to the role of apolipoprotein E in beta-amyloid clearance. Alzheimer’s disease involves the buildup of beta-amyloid aggregates in the brain, and apoE4 (compared to other apoE variants) has a reduced ability to promote the proteolytic reactions that break this peptide down. As a result, apoE4 carriers may be at higher risk of developing the amyloid plaques featured in Alzheimer’s disease. Recent research has also suggested that this link is much stronger in women than in men!
Cardiovascular disease. Several large cohort studies have shown that apoE4 carriers have up to a 2-times greater risk of coronary heart disease compared to non-apoE4 carriers, although not all studies confirm this link (most likely due to too-small sample sizes making it hard to establish statistical significance, as well as various factors that modify the effects of apoE4, such as age, diet, drinking, and smoking). Even at young ages, apoE4 carriers tend to have higher LDL cholesterol and triglycerides than the rest of the population (typically 5-10% higher than apoE3/3 individuals), and some research shows that apoE4 carriers have higher levels of the inflammatory cytokine interleukin (IL)-1β. And, although the evidence is limited to just a handful of studies that have been conducted so far, apoE4 carriers of Asian descent (especially homozygotes) appear to have a greater risk of high blood pressure and stroke than non-apoE4 carriers.
Sleep apnea. Being an apoE4 carrier significantly increases the risk of obstructive sleep apnea (which causes breathing to abruptly stop and start during sleep), especially in people under the age of 65. The risk is particularly high for people with two apoE4 copies! See also Sleep and Disease Risk: Scarier than Zombies!
HIV progression. For people infected with the human immunodeficiency virus (HIV), being an apoE4 carrier is associated with faster disease progression and earlier mortality (mechanistic studies show that apoE4 enhances the entry of HIV-1 into cells, which may help explain this). However, being an apoE4 carrier doesn’t necessarily increase the risk of acquiring HIV in the first place!
Traumatic brain injury. ApoE4 carriers who suffer traumatic brain injuries (typically from a violent blow, jolt, or penetration to the head) have a slower and less complete recovery than non-apoE4 carriers, due to the effect of apoE4 on the neuronal repair process. Research shows that apoE4 carriers have poorer memory performance following a traumatic brain injury, and also score worse on behavioral abnormalities, dysarthria (slurred or slow speech), cognitive abnormalities, dysphasia (communication impairment), and other functional outcomes. Whereas apoE2 appears neuroprotective, apoE4 isn’t as effective at constructing and regenerating the microtubules within a neuron’s axon.
Multiple sclerosis progression. Several studies have found that apoE4 carriers diagnosed with multiple sclerosis (MS) experience a faster progression of the disease and a greater severity of symptoms (in contrast to apoE2 carriers, who tend to have milder cases of MS). This is potentially due to apoE4 carriers with MS having lower levels of N-acetylaspartate in the brain, which is an amino acid that’s considered a marker for neuronal health. See also An Update from Dr. Wahls’ Research Lab
On top of increases in disease risk and progression, apoE4 carriers respond uniquely to certain dietary and lifestyle factors when compared to the rest of the population. For example:
Omega-3 fats. The protective effects of omega-3 fats may not apply to apoE4 carriers, and at high intakes, omega-3s may actually have adverse health effects! The omega-3 fat DHA has been shown to significantly increase LDL particle count in apoE4 carriers, while also reducing LDL particle size (both of which are risk factors for heart disease). Likewise, EPA appears to lower HDL particle count and concentration in apoE4 carriers. So, while the rest of the population experiences cardioprotective effects from omega-3 fats, apoE4 carriers may actually experience the opposite! Likewise, apoE4 carriers seem unable to benefit from the neuroprotective effects of omega-3 fats. Whereas omega-3 fats are generally believed to reduce the risk of Alzheimer’s and boost cognition, studies that divide the results based on people’s apoE status show that only apoE4 non-carriers get these benefits. See also Why Grains Are Bad–Part 2, Omega 3 vs. 6 Fats
Saturated fat. In studies, a high saturated fat intake among apoE4 carriers has been linked to a rise in LDL particle number and concentration, a rise in C-reactive protein , and a higher risk of Alzheimer’s disease. Likewise, saturated fat intake significantly raises the risk of dementia in apoE4 carriers compared to non-carriers (in one study, the risk increased sevenfold for apoE4 carriers with high saturated intakes!). In other words, saturated fat seems to be a major modifier for disease risk among apoE4 carriers in a way that isn’t seen with other apoE phenotypes. See also Saturated Fat: Healthful, Harmful, or Somewhere In Between?
Physical inactivity. The effects of sedentary living seem particularly detrimental for apoE4 carriers. Compared to physically active non-carriers, apoE4 carriers who are physically inactive during midlife have a significantly higher risk (77% greater!) of developing dementia in older age. See also The Benefits of Gentle Movement
Dietary cholesterol. The idea that cholesterol in our diet significantly raises cholesterol in our blood has been challenged by plenty of recent research, but for apoE4 carriers (especially homozygotes), the link may be real! ApoE4 enhances cholesterol absorption in the intestine relative to other apoE variants. One study found that adding 750mg of cholesterol per day to participants’ daily diets (in the form of egg yolks) caused total blood cholesterol, LDL, and apoB levels to increase more than twice as much in people with apoE4/4 compared to people with other apoE phenotypes.
Alcohol. For apoE4 carriers, even small amounts of alcohol fail to deliver the heart-protective perks seen for the rest of the population. Although more studies are needed to really help us understand the relationship, it appears that apoE4 carriers who completely abstain from alcohol have greater protection from cardiovascular disease and Alzheimer’s. See also The WHYs behind the Autoimmune Protocol: Alcohol
Heavy metal detox. ApoE4 carriers have a reduced ability to detox heavy metals, especially lead and mercury. This is because the structure of apoE4 gives it a more limited ability to bind to and remove metals from tissues within the body. See also The Best Foods and Nutrients to Support Liver Detox
As we can see, apoE4 has an important influence on our disease risk and our response to diet and lifestyle variables. But, the news isn’t all bad! ApoE4 is also associated with better episodic memory (the memory of specific events and experiences in our lives), higher levels of vitamin D (which might be another reason why apoE4 is higher in indigenous populations living at extreme latitudes), protection against non-alcoholic fatty liver disease (apoE4 carriers have only half the risk compared to apoE3/3 carriers!), and better verbal fluency. It can also protect against (or reduce the severity of) certain infectious diseases like Giardia and hepatitis C.
ApoE2 is the least common apoE variant, with only 0.5% of the population carrying two copies (apoE2/2) and another 18-19% carrying one copy (apoE2/3 or apoE2/4). It doesn’t have as consistent a distribution pattern as apoE4, but this variant is almost completely absent among Native Americans. And. although it gets less airtime in scientific studies than apoE4 does, it can still have an important influence on our health, offering a mixed bag of risk increases and disease protection!
First, the bad:
Type III hyperlipoproteinemia. A small portion of apoE2/2 carriers (less than 5%) develop a disorder called type III hyperlipoproteinemia, which results from abnormal function of the apoE receptor and delayed clearance of chylomicron remnants and IDL (intermediate density lipoprotein) from circulation. This leads to elevated cholesterol and triglycerides, an early onset of vascular diseases (including heart disease and peripheral vascular disease), and visible symptoms such as orange discoloration within skin creases (called palmar xanthomata). Although not everyone with apoE2 will develop this condition (the main modulator is high insulin levels), the people who do have it are typically apoE2 homozygotes. (The defect is recessive, so people with only one apoE2 copy aren’t at heightened risk. In fact, for people without type III hyperlipoproteinemia, apoE2 is associated with lower cholesterol levels!)
Parkinson’s disease. Some studies have uncovered a link between apoE2 and a higher risk of Parkinson’s disease. However, this finding hasn’t always been replicated in larger cohort studies, and might apply more to certain ethnic groups than others. So, more research is needed!
Psoriasis. ApoE2 has been linked to a higher risk of psoriasis (47% higher risk compared to apoE3 and apoE4) among European and Asian populations.
Multiple sclerosis. A meta-analysis of almost 7,000 people found that apoE2 carriers are more likely to develop multiple sclerosis (although, as we saw earlier, the cases tend to be more mild than for apoE4 carriers).
Vertebral fracture. Some research suggests that people with vertebral fractures are about twice as likely to be apoE2 homozygotes than healthy controls, due to the role of apoE as a regulator of bone metabolism and apoE2 resulting in different metabolic properties than the other variants. However, this link doesn’t seem to carry over to other forms of fractures.
But, apoE2 also has a major benefit: a reduced risk of Alzheimer’s disease! People who carry two copies (apoE2/2) or who carry one apoE2 copy and one apoE3 copy have a 40% reduced risk of developing Alzheimer’s disease, compared to apoE3 homozygotes. For people with both apoE2 and apoE4, the risk is higher than for apoE3 homozygotes (a 2.6-times higher risk), but still lower than for all other apoE4 combinations. The reason for these protective effects is because apoE2 results in significantly less amyloid-beta deposits in the brain (the opposite of what we see with apoE4).
For people who have apoE4 (especially homozygotes) and even apoE2, the news can seem grim in terms of disease risk. But, it’s important to remember that while we can’t change the genetic lot we were handed, the effects of apoE are heavily mediated by diet and lifestyle, which are in our control! No matter what apoE variants we have, we can take measures to reduce our disease risk as much as possible.
For apoE4 carriers:
- Eat (and live!) Paleo! Although we haven’t conducted any studies (yet!) on the impact of a Paleo diet and lifestyle on apoE4 carriers, we can make an educated guess that following diet and activity patterns that mimic those of hunter-gatherers could diminish apoE4’s impact on chronic disease risk (since hunter-gatherers don’t seem to experience the damaging effects of apoE4 that agricultural populations do, and instead, get its benefits). For example, among groups that haven’t adopted Western ways of life (such as Sub-Saharan Africans), there’s no association between apoE4 and Alzheimer’s or heart disease—suggesting that apoE4 only becomes problematic when our diet and lifestyles are out of sync with our genes. So, for us Westerners, the best plan of action is to eat a whole-foods Paleo diet with abundant vegetables, leaner animal foods, plenty of phytochemicals, and no oxidized oils, and possibly implement intermittent fasting, carb cycling, or fat cycling (to mimic periods of scarcity and abundance experienced by hunter-gatherers). Physical activity is also extremely important, and apoE4 carriers might benefit the most from weight lifting, climbing, interval training, and other forms of exercise that resemble the activity patterns of early humans.
- Limit omega-3 intake. Seafood (including both fish and shellfish) is an incredible source of nutrition, but for apoE4 carriers, it might be smart to choose leaner fish (like cod or tuna) instead of fatty fish, and to avoid fish oil (or other omega-3) supplements. Likewise, keeping omega-6 intake low is important to maintain a good omega 6/3 ratio and to reduce the need for more omega-3s from the diet.
- Limit heavy metal exposure. This means avoiding (or safely removing) amalgam fillings, testing for lead or copper in the water we drink at home, avoiding high-mercury seafood, and making sure the bones we use to make bone broth don’t come from animals raised close to industrial areas (since this can cause heavy metal accumulation in their bones, which then transfers to us in the broth!).
- Avoid traumatic head injuries! Okay, getting a serious head injury isn’t a good idea for anyone, but apoE4 carriers have the most difficult time recovering. For that reason, apoE4 carriers should generally avoid high-risk activities like wrestling, boxing, and football.
- Consider going alcohol-free. I know, this is a tough one to swallow! Current research suggests that avoiding alcohol could be very beneficial for apoE4 carriers, especially homozygotes. But, for people who aren’t willing to say goodbye to alcohol forever, drinking small amounts more frequently appears to be safer than drinking large amounts (or binge-drinking) on a sporadic basis.
- Regulate insulin levels. As we saw earlier, high insulin is one of the biggest factors separating apoE2/2 carriers who get type III hyperlipoproteinemia from those who don’t. Avoiding refined carbohydrates, having an adequate (not excessive) energy intake, eating our last meal early and our breakfast late (to increase the time spent in a fasted state), losing weight if we carry excess body fat, and engaging in frequent gentle activity as well as more strenuous exercise can all help keep insulin levels in check.
- Prioritize bone health. Because of the link between apoE2 and vertebral fractures (and a less consistent association with lower bone density), consuming plenty of vitamin K2, vitamin A, calcium, vitamin D (from the sun or from supplements), protein, vitamin C, and magnesium (all of which play important roles in bone metabolism) is crucial! Likewise, activities that improve bone mass, such as weight-bearing exercise, could be helpful.
So, who should get their apoE status tested? Although it can be useful information for virtually anybody, people with a family history of Alzheimer’s disease or type III hyperlipoproteinemia, and to a lesser extent cardiovascular disease, will particularly benefit from knowing their apoE phenotype. Tests can be ordered through your doctor and via independent testing companies, and genetic testing kits like 23andMe will also reveal apoE status.
Bottom line? ApoE variants can have a significant impact on our health, but even being an apoE4/4 carrier isn’t a death sentence. As with most things, knowledge is power when it comes to apoE! Knowing our phenotype can help us modify our diet and lifestyle to maximize our reduction in disease risk and live a long, healthy life.
Alexander DM, et al. “The contribution of apolipoprotein E alleles on cognitive performance and dynamic neural activity over six decades.” Biol Psychol. 2007 Jul;75(3):229-38. Epub 2007 Mar 16.
Bergeron N & Havel RJ. “Prolonged postprandial responses of lipids and apolipoproteins in triglyceride-rich lipoproteins of individuals expressing an apolipoprotein epsilon 4 allele.” J Clin Invest. 1996 Jan 1;97(1):65–72.
Burt TD, et al. “Apolipoprotein (apo) E4 enhances HIV-1 cell entry in vitro, and the APOE epsilon4/epsilon4 genotype accelerates HIV disease progression.” Proc Natl Acad Sci U S A. 2008 Jun 24;105(25):8718-23. doi: 10.1073/pnas.0803526105. Epub 2008 Jun 18.
Carter DB. “The interaction of amyloid-beta with ApoE.” Subcell Biochem. 2005;38:255-72.
Chapman J, et al. “APOE genotype is a major predictor of long-term progression of disability in MS.” Neurology. 2001 Feb 13;56(3):312-6.
Corbo RM & Scacchi R. “Apolipoprotein E (APOE) allele distribution in the world. Is APOE*4 a ‘thrifty’ allele?” Ann Hum Genet. 1999 Jul;63(Pt 4):301-10.
Corder EH, et al. “Protective effect of apolipoprotein E type 2 allele for late onset Alzheimer disease.” Nat Genet. 1994 Jun;7(2):180-4.
Crawford FC, et al. “APOE genotype influences acquisition and recall following traumatic brain injury.” Neurology. 2002 Apr 9;58(7):1115-8.
de Beer F, et al. “Expression of type III hyperlipoproteinemia in apolipoprotein E2 (Arg158 –> Cys) homozygotes is associated with hyperinsulinemia.” Arterioscler Thromb Vasc Biol. 2002 Feb 1;22(2):294-9.
De Feo E, et al. “A case-control study on the effects of the apolipoprotein E genotypes in nonalcoholic fatty liver disease.” Mol Biol Rep. 2012 Jul;39(7):7381-8. doi: 10.1007/s11033-012-1570-7. Epub 2012 Feb 15.
Eichner JE, et al. “Relation of apolipoprotein E phenotype to myocardial infarction and mortality from coronary artery disease.” Am J Cardiol. 1993;71:160–5.
Enzinger C, et al. “Lower levels of N-acetylaspartate in multiple sclerosis patients with the apolipoprotein E epsilon4 allele.” Arch Neurol. 2003 Jan;60(1):65-70.
Friedman G, et al. “Apolipoprotein E-epsilon4 genotype predicts a poor outcome in survivors of traumatic brain injury.” Neurology. 1999 Jan 15;52(2):244-8.
Gao J, et al. “Apolipoprotein E genotypes and the risk of Parkinson disease.” Neurobiol Aging. 2011 Nov;32(11):2106.e1-6. doi: 10.1016/j.neurobiolaging.2011.05.016. Epub 2011 Jul 8.
Godfrey ME, et al. “Apolipoprotein E genotyping as a potential biomarker for mercury neurotoxicity.” J Alzheimers Dis. 2003 Jun;5(3):189-95.
Gottlieb DJ, et al. “APOE epsilon4 is associated with obstructive sleep apnea/hypopnea: the Sleep Heart Health Study.” Neurology. 2004 Aug 24;63(4):664-8.
Haan MN & Mayeda ER. “Apolipoprotein E Genotype and Cardiovascular Diseases in the Elderly.” Curr Cardiovasc Risk Rep. 2010 Sep; 4(5): 361–368.
Han Y, et al. “Apolipoprotein E gene polymorphism in psoriasis: a meta-analysis.” Arch Med Res. 2013 Jan;44(1):46-53. doi: 10.1016/j.arcmed.2012.10.009. Epub 2012 Nov 6.
Hauser PS & Ryan RO. “Impact of Apolipoprotein E on Alzheimer’s Disease.” Curr Alzheimer Res. 2013 Oct; 10(8):809-817.
Huang X, et al. “APOE-[epsilon]2 allele associated with higher prevalence of sporadic Parkinson disease.” Neurology. 2004 Jun 22;62(12):2198-202.
Huang Y. “Mechanisms linking apolipoprotein E isoforms with cardiovascular and neurological diseases.” Curr Opin Lipidol. 2010 Aug;21(4):337-45. doi: 10.1097/MOL.0b013e32833af368.
Huebbe P, et al. “APOE ε4 is associated with higher vitamin D levels in targeted replacement mice and humans.” FASEB J. 2011 Sep;25(9):3262-70. doi: 10.1096/fj.11-180935. Epub 2011 Jun 9.
Jiang Q, et al. “ApoE promotes the proteolytic degradation of Abeta.” Neuron. 2008 Jun 12;58(5):681-93. doi: 10.1016/j.neuron.2008.04.010.
Kadotani H, et al. “Association between apolipoprotein E epsilon4 and sleep-disordered breathing in adults.” JAMA. 2001 Jun 13;285(22):2888-90.
Kehlmann I, et al. “Apolipoprotein E genotype and hepatitis C, HIV and herpes simplex disease risk: a literature review.” Lipids Health Dis. 2010; 9: 8.
Kerr ME & Kraus M. “Genetics and the central nervous system: apolipoprotein E and brain injury.” AACN Clin Issues. 1998 Nov;9(4):524-30.
Kivipelto M, et al. “Apolipoprotein E ε4 magnifies lifestyle risks for dementia: a population-based study.” J Cell Mol Med. 2008 Dec; 12(6b): 2762–2771.
Lehtimäki T, et al. “Cholesterol-rich diet induced changes in plasma lipids in relation to apolipoprotein E phenotype in healthy students.” Ann Med. 1992 Feb;24(1):61-6.
Marley RW.”Apolipoprotein E: from cardiovascular disease to neurodegenerative disorders.” J Mol Med (Berl). 2016 Jul;94(7):739-46. doi: 10.1007/s00109-016-1427-y. Epub 2016 Jun 9.
Mondadori CR, et al. “Better memory and neural efficiency in young apolipoprotein E epsilon4 carriers.” Cereb Cortex. 2007 Aug;17(8):1934-47. Epub 2006 Oct 31.
Moreno JA, et al. “The effect of dietary fat on LDL size is influenced by apolipoprotein E genotype in healthy subjects.” J Nutr. 2004 Oct;134(10):2517-22.
Pinholt M, et al. “The association between apolipoprotein E and multiple sclerosis.” Eur J Neurol. 2006 Jun;13(6):573-80.
Pizzorno L. “Omega-3s, ApoE Genotype and Cognitive Decline.” Longevity Medicine Review. Accessed April 19, 2017.
Potapov AA. “[Clinical and prognostic significance of genetic markers of ApoE gene in traumatic brain injury].” Zh Vopr Neirokhir Im N N Burdenko. 2010 Jul-Sep;(3):54-62; discussion 62.
Samatovicz RA. “Genetics and brain injury: apolipoprotein E.” J Head Trauma Rehabil. 2000 Jun;15(3):869-74.
Schmidt S, et al. “Association of polymorphisms in the apolipoprotein E region with susceptibility to and progression of multiple sclerosis.” Am J Hum Genet. 2002 Mar;70(3):708-17. Epub 2002 Feb 11.
Tikkanen MJ, et al. “Apolipoprotein E4 homozygosity predisposes to serum cholesterol elevation during high fat diet.” Arteriosclerosis. 1990 Mar-Apr;10(2):285-8.
Valcour V, et al. “Frequency of apolipoprotein E4 among older compared with younger HIV patients: support for detrimental effect of E4 on survival.” Proc Natl Acad Sci U S A. 2008 Oct 14;105(41):E66; author reply E67-8. doi: 10.1073/pnas.0806919105. Epub 2008 Oct 8.
Yin YW, et al. “Association between apolipoprotein E gene polymorphism and the risk of multiple sclerosis: a meta-analysis of 6977 subjects.” Gene. 2012 Dec 10;511(1):12-7. doi: 10.1016/j.gene.2012.09.010. Epub 2012 Sep 13.
Zhang SQ, et al. “Apolipoprotein E gene E2/E2 genotype is a genetic risk factor for vertebral fractures in humans: a large-scale study.” Int Orthop. 2014 Aug;38(8):1665-9. doi: 10.1007/s00264-014-2380-4. Epub 2014 Jun 1.