Researchers have identified a genetic variant that increases risk of follicular lymphoma (FL), a type of blood cancer.

FL is a type of Non-Hodgkin lymphoma (NHL), the fifth most common cancer in the U.S. Nearly 20,000 patients die and 66,000 more are diagnosed with NHL each year. The incidence of FL has doubled within the past 30 years.

In a study by Skibola et al., scientists analyzed the genomes of 4,805 people of European descent from the United States, Germany and Canada to search for genetic associations with four different subtypes of lymphoma. The results, published online this week in the journal Nature Genetics, newly linked one SNP to an elevated risk for developing FL in European populations.

FL is a slow-progressing cancer of the white blood cells that affects middle-aged and elderly people. It is usually diagnosed in advanced stages, but patients live approximately 10 to 12 years after diagnosis.

Skibola et al. revealed that having one copy of the C version of rs6457327, a SNP located on chromosome 6, is associated with 1.69 times increased odds of AF. Carrying two copies of C at this SNP is associated with 2.22 times increased odds of FL. A majority of the population carries the C version, so those who have the A version of this SNP seem to be protected from FL.

(23andMe customers can check their data for rs6457327 using the Browse Raw Data feature.)

This SNP is located near a region that is associated with susceptibility to psoriasis. The researchers said that further studies are required to evaluate whether rs6457327 is also associated with psoriasis due to this genetic overlap. Future research may also determine whether FL is associated with exposure to viruses such as HIV.

ShareThis

Researchers have identified a genetic variant that increases risk of atrial fibrillation (AF), an electrical disorder of the heart, which in turn increases risk of stroke.

AF affects one in four people over 40 years old, and stroke is the third leading cause of death in the United States.

In one of two studies, scientists scanned the genomes of 5,806 people from the United States, Iceland and Norway to look for genetic variations associated with AF. The other study combined data from several large studies comprising almost 47,000 people of European ancestry.

Their results, published online this month in Nature Genetics, newly linked one SNP to slightly elevated risk for developing AF in European populations.

Most strokes are caused by a blood clot that blocks circulation to the brain. Because AF causes ineffective beating of the heart, blood can pool and clot in the upper chambers, resulting in stroke if a clot leaves the heart and lodges in the brain. AF increases the risk of stroke four to five times across all age groups and is the cause behind 10 to 15 percent of all blood clot-related strokes.

In the first study, Gudbjartsson et al. studied people of European descent from the United States, Iceland and Norway to reveal that the T version of rs7193343, a SNP on chromosome 16, is associated with 1.21 times increased odds of AF. This SNP was also associated with 1.11 times increased odds of ischemic stroke and 1.22 increased odds for cardioembolic stroke.

In the second study, Benjamin et al. associated AF with another SNP near rs7193343. Both SNPs are likely to be marking a third genetic variation that actually has a functional effect on AF. Further research will be necessary to identify both the variation and its effect.

(23andMe customers can check their data for rs7193343 using the Browse Raw Data feature.)

Gudbjartsson et al. also assessed the association of AF with the T version of rs7193343 in more than 3,000 Han Chinese, but the association was not significant. They noted that the T allele of this SNP is much more frequent in Han Chinese than in European populations.

ShareThis

Much to the surprise of many scientists, a lot of the SNPs identified in genomewide association studies have not been in the parts of genes that encode the molecular machinery of a cell.

Instead, many SNPs have been found on the edges of genes, in regions of DNA that control when the genes get turned on or off, in parts of genes that get cut out before the final proteins are made, or even in so-called “gene deserts,” areas of DNA that don’t seem to contain any genes at all.

Rs6983267 is one of these gene desert SNPs.  People with two copies of the G version of this SNP have about 1.4 times the odds of developing colorectal cancer compared to people who have two Ts, but so far no one has been able to figure out why.  Two new reports show that, even though this SNP seems to be out in the middle of nowhere in the genome, it can interact with components of a signaling pathway known to be overactive in more than 90% of all colorectal cancers.

(23andMe customers can see their data for rs6983267, as well as two other SNPs associated with increased colorectal cancer risk, in Health and Traits.)

Results from Tuupanen et al. and Pomerantz et al., both published online this week in the journal Nature Genetics, show that the region of DNA containing rs6983267 is an “enhancer” that can turn up the amount of protein made from the MYC gene. The riskier G version of the SNP appears to make the enhancer stronger than the T version.

In colorectal cancer, increased MYC expression can often be traced to overactivity of a molecular signaling pathway known as Wnt.  Both groups of researchers found that the region of DNA containing rs6983267 was responsive to Wnt signaling, thus connecting this SNP to a well-established cancer mechanism.

Drugs that attack the Wnt pathway are attractive candidates for cancer therapies.  According to Tuupanen et al., the new results suggest that these same types of drugs might be useful for personalized cancer prevention treatments in people who carry the riskier version of rs6983267.

ShareThis

It’s no secret that obesity rates are rising — quickly.  Between 2000 and 2005 the prevalence of obesity rose by 24%.  Extreme obesity increased by more than 50%.  If current trends continue, more than half of all Americans will be clinically obese by the year 2030.

Rapid changes in the prevalence of a disorder suggest that environmental rather than genetic factors are to blame. But scientists know from twin and adoption studies that body mass index (BMI) is highly heritable.  So which is it: nature or nurture?

As with many aspects of human health, it’s both.  Some people have the bad luck to have inherited genetic variations that together with the modern environment – sedentary jobs and hobbies, easy access to calorie-dense foods – create the perfect storm for the onset of obesity.

Changing the environmental factors that lead to obesity in some people seems simple enough – more exercise, less food.  But public health campaigns touting these commonsense fixes have had little effect against the obesity epidemic. By understanding the genetic aspects of obesity, and how they interact with the environment, scientists may be able to develop more effective treatments and prevention strategies.

In the July issue of Nature Reviews Genetics Andrew J. Walley of Imperial College London and colleagues review the current state of obesity genetics research.  While much progress has been made, the authors make it clear that there is still a long way to go, as the genes identified thus far explain only a tiny fraction of the total genetic component of obesity.

One key to future advancements in obesity genetics research, say Walley et al., lies in improvements to current genomewide association study (GWAS) methods.

First, the authors recommend that researchers focus on recruiting extremely obese people for their studies to increase the likelihood of finding genes with large effects. They also suggest that scientists should stop using BMI as their primary measurement of obesity.  While simple and cheap, this method does not take fat distribution or the ratio of fat to muscle into account.  There are more sophisticated methods, such as CT scans, MRI scans and ultrasound imaging, as well as machines that use air displacement to measure body volume and weight and can calculate fat and fat-free body mass.  Technological advances that will reduce the costs associated with genotyping, and ultimately genetic sequencing, are also needed so that larger numbers of subjects can be studied.

Beyond these improvements to current GWAS methods, Walley et al. say studies of more than just common variations are needed.  They suggest that investigations of rare SNPs, copy number variations (duplications, insertions and deletions of DNA) and inherited changes that don’t affect the actual DNA sequence will be needed to fully understand the genetics of obesity.

There are several competing theories about the overall biological basis of obesity.  Some suggest that obesity is a disorder of energy balance in the body, while others think regulation of the growth of fat cells is the key.  Still others think obesity may be due to defects in the neurological control of appetite and food intake.  Continued advancements in understanding the genetics of obesity will help scientists tease these theories apart, and hopefully lead to a healthier future.

(23andMe customers can check their data for a SNP in the FTO gene in the Obesity Research Report.  So far, this is the genetic variant most strongly associated with the risk of obesity.  There is also an Obesity Preliminary Research Report.)

Related posts in the Spittoon:

SNPwatch: New Obesity SNPs Point To The Brain’s Role In Regulating Appetite

SNPwatch: Gene Variant Linked To Obesity Affects Food Choices In Children

Genetics May Dull Brain’s Pleasure Response To Food, Causing Weight Gain

SNPwatch: Genetic Variants Affect Weight Loss Drug Effectiveness

SNPwatch: Genetic Link Between Obesity and Colorectal Cancer

It’s Not Genes Or Environment, It’s Genes AND Environment

SNPwatch: Researchers Find Genetic Variants That May Influence Risk For Obesity

SNPwatch: MC4R Gene Associated With Body Mass

ShareThis

In the current issue of Nature Genetics, two teams of researchers identify nine new genetic locations associated with obesity. Almost all of them are within or close to genes that are expressed in the brain, especially the part that regulates appetite.

“This underscores the importance of genes that regulate food intake over those involved in metabolism,” wrote a team of researchers led by Gudmar Thorleiffsson and G. Bragi Walters of deCODE Genetics.

Thorleiffsson et al. performed a two-step study that initially examined more than 30,000 people of European and African American descent from Iceland, Holland and the United States. It then narrowed down the list of promising genetic regions to 11 using a sample of 5,586 Danish individuals. Four of those regions — in or close to the genes FTO, MC4R, BDNF and SH2B1 — were already known to be associated with obesity.

The other Nature Genetics paper, authored by the members of the Genetic Investigation of Anthropometric Traits (GIANT) consortium, initially looked for genetic links to obesity in more than 32,000 people of European descent, and then pursued their most promising leads in a second group of more than 59,000. The GIANT researchers confirmed two well-known obesity associations — the genes FTO and MC4R — and discovered an additional six.

Altogether, the two teams found nine new obesity-associated genetic variants; but none of them had a very substantial effect. For example, even someone who had the riskier version at all six locations found by the GIANT consortium would be only about four pounds heavier compared to average.

There are two possible explanations for the small effect of these new associations. Either there are dozens more associations with similarly small effects, or relatively rare genetic variations are much more important that more common ones in determining a person’s weight. Either way, finding those additional sources of genetic obesity risk is going to require studies even larger than the ones published this week in Nature Genetics.

23andMe customers who want to know their own genotypes at the new obesity-associated variants can use the following chart, which identifies the weight-increasing genotype for each SNP.

ShareThis

The online publication of three papers by Nature Genetics this week has roughly doubled the number of genetic locations associated with levels of cholesterol and triglycerides in the blood.

With the addition of so many new SNPs, researchers have a wealth of opportunities to better understand how the body regulates cholesterol and triglyceride levels, find new targets for drugs to control them and identify people who are at increased risk of cardiovascular disease.

Two of the Nature Genetics papers combined data from a number of previous studies to increase their statistical power to detect SNPs associated with cholesterol and triglyceride levels. One looked at between 17,798 and 22,562 European subjects (depending on the specific measurement being examined), and found six genes or locations on five chromosomes that were significantly associated with total cholesterol, HDL, LDL or triglycerides. The other studied 20,623 people from Europe and the United States, and identified 11 genetic regions that were not previously known to be associated with cholesterol or triglyceride levels.

The third paper found five genetic regions associated with cholesterol or triglycerides in a cohort of 4,763 people born in northern Finland in 1966.

A number of the new associations were located near places in the genome where rare mutations are already known to completely disable cholesterol-regulating genes, causing serious disruptions in cholesterol metabolism. The newly discovered SNPs apparently cause much subtler effects, however; in most cases they appear to modulate the activity of the genes they affect.

The authors of one study (Kathiresan et al.) also provided evidence that seven of their new SNPs modulate the expression of genes in the liver, where cholesterol is produced.

In spite of their success in both discovering new genetic locations associated with cholesterol and triglyceride levels — the papers also replicated virtually all previously known associations as well — the value of all this information for personal genomics is somewhat limited. After all, blood levels of cholesterol and trigylcerides are themselves indicators of cardiovascular disease risk. Knowing a person’s genetic risk on top of their actual cholesterol levels provides only an “incremental” amount of additional information, Kathiresan et al. wrote.

Even so, Kathiresan et al. developed an “allelic dosage score” that consisted of a person’s number of elevated-risk genotypes out of 32 previously known and newly discovered SNPs. Study subjects with scores in the top tenth of the distribution were more than twice as likely to have LDL cholesterol about 160 mg/dl, HDL cholesterol below 40 mg/dl and trigylcerides above 200 mg/dl.

Researchers still have captured only a small fraction of the gene variation that explains why one person’s cholesterol level is higher or lower than the next person’s. The authors of the Finnish study, who found one cholesterol-increasing version of the AR gene that was present in less than 2% of subjects, suggest that much of that unknown risk may lie in similarly rare variants.

23andMe customers can use our Browse Raw Data feature to check their genotypes at most of the newly reported SNPs. The following table indicates which version of each SNP is less common among Europeans, and whether that version causes an increase or decrease in LDL cholesterol, triglycerides or HDL cholesterol (lower HDL raises a person’s risk of cardiovascular disease).

ShareThis

Researchers have added chronic lymphocytic leukemia (CLL), the most common type of leukemia in western countries, to the long list of diseases that have been associated with common genetic variants.

CLL is caused by abnormal, uncontrolled growth of immune cells called B-cells. Usually there are no obvious symptoms and the disease is detected only during a routine blood test. In the United States, about 15,000 new cases of the disease are diagnosed each year.

Because there is a familial basis to the disease it tends to run in families, a role for genetics in CLL has long been suspected. First-degree relatives of patients have a seven-fold increased risk for the disease. But so far, no clear contenders for the “CLL gene” have been found. This suggested to researchers that instead of there being one or two genes with large effects to be found, there might instead be multiple genetic risk factors for CLL, each contributing a small amount to the risk.

By studying about 1,500 British people with CLL and about 3,000 controls, Di Bernardo et al found six distinct SNPs associated with CLL. These results, published online today in Nature Genetics, not only begin to shed light on the genetics of the disease, but they also give researchers clues about its biological basis.

(All six SNPs are available from 23andMe. See the table at the end of this post for more information.)

For the about 54% of people of European ancestry who have three or fewer “risky” copies out of a possible 12 (two copies each for six SNPs), there is no significant difference in the risk of developing CLL. Having six risky copies ups a person’s odds about 3.8 times over those with none, and the 2.1% of that population who are unlucky enough to have seven risky copies have approximately eight-fold increased odds of developing CLL.

Although it’s far from certain that these six SNPs are actually causative factors for CLL, the authors note in their report that it will be intriguing to see if the SNPs are associated with CLL in non-Europeans. Both Asian and African populations have lower frequencies of the risky versions of each of the six SNPs — and lower rates of CLL.

“Effect” is the increase in odds of developing CLL conferred by each copy of the riskier version of a SNP.

Image: Mary Ann Thompson/GNU Free Documentation License

ShareThis

Tobacco use is the leading cause of preventable deaths in the United States — every year it kills nearly 440,000 Americans.

While about 70 percent of the 45 million Americans who currently smoke have said they want to quit, the Centers for Disease Control and Prevention estimates that only about 10 percent of them will succeed, due in large part to the addictive qualities of the active ingredient in tobacco, nicotine.

Researchers have been looking at the genetic basis of nicotine dependence, characterized by tolerance to the drug and the presence of withdrawal symptoms in its absence, with the long-term goals of better assessing a person’s risk for addiction and potentially developing targeted treatments.

Last year, Laura Bierut and her colleagues at the Washington University School of Medicine identified a single variant in a nicotine receptor that increases a person’s risk of developing nicotine dependence. A later study indicated the same SNP can actually reduce the risk of developing cocaine dependence. And now a study published this month by Bierut and her colleagues shows that the variant also influences how much pleasure a first-time smoker will experience from the first puff of a cigarette.

“The type of variant you have at this particular receptor — the alpha-5 nicotinic receptor — may actually predict whether or not you will do well on nicotine replacement therapy,” Bierut said in a statement at the time.

The version of SNP rs16969968 that can increase the risk for nicotine addiction is found at appreciable levels only in European populations. People with two copies of the less common A version of the SNP are twice as likely to develop nicotine dependence when exposed to smoking than people with two copies of the more common G version.

(This SNP isn’t in 23andMe’s database, but we do have data on an equivalent SNP: rs1051730, which has also been linked to nicotine dependence in European populations, as The Spittoon previously mentioned here. As with rs16969968, the rarer A version of rs1051730 is associated with increased nicotine dependence compared to the more common G version. Though both SNPs have the same effect on nicotine dependence, the size of the effect caused by rs1051730 may be slightly larger or smaller compared to the SNP Bierut and her colleagues study.)

A new study from Bierut’s group published this month in the journal Addiction shows that the same version of rs16969968 previously associated with nicotine dependence is also linked to the pleasurable rush some first-time smokers get. Each A at rs16969968 increased a person’s odds of having found that first puff pleasurable by 1.6 times compared to people with no copies of an A. This “buzz” was in turn associated with an 8-fold increase in odds of being a current smoker.

“The finding of a genetic association with pleasurable early smoking experiences may help explain how people get addicted — and, of course, once addicted, many will keep smoking for the rest of their lives,” said the study’s senior author Ovide Pomerleau in a statement.

“It appears that for people who have a certain genetic makeup, the initial physical reaction to smoking can play a significant role in determining what happens next. If cigarette smoking is sustained, nicotine addiction can occur in a few days to a few months.”

Bierut’s research on SNP rs16969968 also extends to addictive substances other than nicotine. Back in June, her group released a study that examined cocaine addicts and non-drug abusing controls to show that having the nicotine dependence-predisposing A version of this SNP actually decreases the risk for cocaine dependence. Having one A decreases the odds of cocaine dependence risk by about 33 percent; two copies decrease the odds by about 45 percent.

The involvement of this SNP in both nicotine and cocaine addiction, and its opposing effects, may seem perplexing. But Bierut and her colleagues suggest that the answer lies in how each addictive substance affects dopamine in the brain’s reward pathway – a key player in the pleasurable feelings produced by drugs that can lead to addiction.

Nicotine triggers the release of dopamine, a neurotransmitter that activates pleasure pathways in the brain. In contrast, cocaine produces pleasurable feelings by preventing the dopamine that is naturally released in the brain from being reabsorbed when its job is done. Excess dopamine hangs around in the brain and produces the “high” associated with the drug.

The nicotine receptor that rs16969968 affects is involved in normal dopamine release, even in the absence of nicotine. So when the A version of this SNP dials down the receptor’s activity, natural dopamine release is reduced. This means there is less dopamine available for cocaine to keep around in the brain, and the amount of pleasure the drug can induce is diminished, possibly enough to keep a person from getting addicted.

Image from FreeDigitalPhotos.net

ShareThis

Even as the genetic studies on schizophrenia released this week illustrate our progress toward the ultimate goal of personalized medicine, they also bring to mind the challenges that still lie ahead.

All three studies focus on identifying the genetic bases of schizophrenia, a mental disorder characterized by hallucinations, delusions and the decreased ability to plan and organize. These symptoms typically manifest during a patient’s late teens or early 20s.

One study led by Cardiff University researcher Michael O’Donovan identified single nucleotide gene variants (SNPs) associated with schizophrenia. As detailed in the journal Nature Genetics, the researchers started with 12 variants, narrowed the list to six after replicating the tests on several thousand study participants and finally identified three SNPs associated with the disease.

Unfortunately, the variant ultimately found to have the strongest association with schizophrenia — rs1344706 — is not included in 23andMe’s database. Yet the SNP’s effects are so small — it increases the chances of schizophrenia by only about 12 percent among people who have the higher risk version — that it offers limited information.

The other studies — one from the International Schizophrenic Consortium and one from deCODE and the SGENE Consortium — which appear in the journal Nature opted to look for genetic associations by checking copy number variants, DNA segments with deletions or duplications that can be inherited.

One group found three deletions — one on chromosome 1 and two on chromosome 15 — that showed up in a few dozen people out of thousands sampled and were more common among people with schizophrenia. The other group identified two of those three deletions. They also found that people with schizophrenia tend to have more rare copy number variants — both duplications and deletions — in their genomes compared to people without the disease.

“This work opens up an entirely new way to think about schizophrenia and eventually will suggest new avenues for researching effective therapies for the sake of patients and families suffering from this terrible disorder,” said International Schizophrenic Consortium member and study co-author Pamela Sklar in a statement.

One of the frustrations schizophrenia researchers have encountered is that while the disease is fairly common — affecting one percent of Americans — the genetic basis of schizophrenia has been difficult to pinpoint. One as-yet unsolved mystery is how the disease, which has a high degree of heritability, is as prevalent as it is when correlated with the fact that people who have schizophrenia and other similar mental disorders have fewer offspring.

These studies suggest that very rare gene mutations might play more of a role in the disease than had previously been supposed. That’s valuable information for scientists, and may explain why the genetics of schizophrenia have been so difficult to figure out. But it also suggests that it will take much more research to understand the genetic risk factors underlying schizophrenia than many other diseases.

ShareThis

No one understands how or why millions of Americans are afflicted with restless legs syndrome, a “creepy-crawly” sensation that compels them to move their legs while lying in bed at night. The phenomenon is so strange and poorly understood, in fact, that some physicians even question whether the seriousness and extent of restless legs syndrome has been over-hyped.

A report published today in Nature Genetics could change some minds about the mysterious condition. New research shows that two SNPs, both located in a gene that encodes a brain signaling protein, may influence who is affected by restless legs syndrome (RLS).

A study of 2,458 sufferers and 4,749 people without the condition from Germany, Austria, The Czech Republic and Canada has shown that versions of both rs1975197 and rs4626664 increase the risk of having RLS.

23andMe customers can check their data for rs1975197 using our Browse Raw Data feature. According to the study, compared to the much more common GG genotype, each A at rs1975197 increases the odds of RLS by 1.31. The riskier version of rs4626664 had a similar effect, with each copy increasing the odds of having RLS by 1.44 times. (23andMe does not report data for this SNP at this time)

RLS has been recognized in the medical literature since 1683, but didn’t gain its official name until 1945. The involvement of genes in the disorder has been suspected since the 1920s. Studies of affected families have identified broad regions of the genome that could contain genes influencing RLS, and recently scientists have been narrowing in on various candidates using genome-wide association studies of SNPs.

In 2007 the same research group behind the new study by Schormair et al published results describing three other RLS-associated SNPs. One of these SNPs (rs3923809 in BTBD9) was replicated by a separate group and is described in detail in the 23andMe My Gene Journal (now called Health and Traits) article on Restless Legs Syndrome.

The newly identified SNPs, rs1975197 and rs4626664, are located in PTPRD, a gene that encodes a signaling protein expressed in the brain. Mouse studies have shown that this protein is important for motor neuron development. These nerve cells connect the central nervous system to the musculature, making their involvement in RLS an attractive theory.

RLS cannot be definitively diagnosed with a blood test or other measure. Instead, patients are classified as having the disorder if the symptoms they describe fit with several criteria. Some have argued that “disease mongering” on the part of drug companies and the media has convinced healthy people that they have RLS and should seek treatment. As more genetic associations like those reported today are found for RLS, the underlying biology of the disorder may become more clear, allowing more definitive diagnoses.

ShareThis