Rare History, Common Disease

John Raelson never thought he would read so many French Canadian history books. It's not exactly the typical fare of a geneticist helping to build a biotechnology company. But for a while, where French Canadians came from was the main question on his mind; books like The First French Canadians: Pioneers in the St. Lawrence Valley (Hubert Charbonneau et al.) were the main research references on his desk.

Currently, on the outskirts of Montreal, in a warehouse-style laboratory encased by hallways of research offices, these history lessons are busily being transformed into genotypes at Genizon BioSciences, where Raelson is chief geneticist. Past a key-coded room securing Genizon's biobank freezers, in a seemingly abandoned and quiet room, eight Illumina BeadArray readers are churning out 576 million genotypes per week. Each sample comes from an individual with four French Canadian grandparents.

On a frozen day, sitting in a warm Genizon office with coworker Majid Belouchi, Raelson smiles and nods as Belouchi speaks emphatically about the history books that have helped them hone in on the genes involved in common diseases. The dry-erase marker board on the wall shows the etched signs of many meetings on the topic. It's work that can be epitomized by Genizon's 2007 study uncovering four novel loci associated with Crohn disease using genome-wide association studies (GWAS) on the Quebec population.¹ The novel loci include regions that contain novel candidate Crohn disease genes such as JAKMIP1, which is involved in interleukin-23 signaling, and LOC285484, which is similar to a secreted cytokine potentially involved in intestinal inflammation.

Raelson says the Quebec population is ideal for genetic research. Here, 2,600 individuals migrated from France between 1608 and 1760, then experienced a genetic bottleneck, creating a population with less "genetic noise" or variation. This homogeneity makes it easier to spot disease-causing mutations (see Figure). "Founder populations can help in identifying smaller [genetic] effects with more power," says Ariel Darvasi of the Hebrew University of Jerusalem, who works with the Ashkenazi Jewish population.

Raelson says that the genetic bottleneck - in which the Quebec founder population has remained isolated and reproduced wildly - makes it, in some respects, superior to other founder populations frequently mined for genetic information, such as those of Finland, Sardinia, Iceland, and groups such as Ashkenazi Jews, (see Criteria of various founder populations). Genizon has collected DNA samples from 47,000 participants in Quebec, counting on many of its efforts to identify disease genes via genome-wide association studies on what those individuals can show them.

Genizon scientists are not alone in their interest in the people of Quebec, who are affected by at least 22 monogenic diseases. Other scientists who study the Quebec population have found information about genes involved in coronary heart disease, hypertension, bipolar affective disorder, and asthma. Researchers have also tracked historical mutations back to France and analyzed the geographic distribution of mutations in Quebec, allowing limited medical resources to be targeted to the regions that need them most.

However, the diminishing isolation of the Quebec founder population means its genetic uniqueness may be lost in the next few generations. "Quebec is becoming highly multiethnic, and intermarriage between different [groups] is becoming more and more common," says Raelson. For studies on Alzheimer's and other diseases in an aging population, this may mean the time for research is now or never. (Indeed, Genizon is in the process of a GWAS involving Alzheimer's.) However, there is also debate over whether founder populations are any better than heterogeneous populations for studying complex traits. Focusing on a unique group for gene discovery may temporarily give a new biotech a competitive edge, but it also means the company must establish why its technique is better than others for enhancing drug development - and quickly.

Quebec is Canada's largest province, about three times the size of France. In 2007, 6 million of the 7.7 million people living there were the descendants of the original French Settlers. It's a population structure that has captured the attention of many scholars over the years.²

The tale of Quebec's genetic uniqueness began with Nouvelle France, the area of North America that the French explorers colonized as they were charting the St. Lawrence River. After the failed settlement attempt of Jacques Cartier in 1541, Samuel de Champlain and 28 French farmers arrived in 1608 on the St. Lawrence River, at the future site of Quebec City. Not too long after, Trois-Rivières had a post, and in 1634, Montreal became a permanent settlement.

"France didn't do a great job of populating the area," says Bertrand Desjardins of the University of Montréal. He is reconstructing the population of Quebec from its beginnings during French colonization. Estimates vary, but some say that only about 8,500 people permanently settled the area, with 2,600 highly reproductive settlers providing two-thirds of the gene pool, before French immigration essentially ceased when the British conquered Nouvelle France in 1760.²

Cultural barriers then set in. Even though English-speaking settlers and Americans surrounded the French population, intermarriage was minimal due to linguistic, religious, and historic differences. Nevertheless, a high birth rate among French Canadians sent the population soaring, with founding couples averaging 36 children and grandchildren. For example, the marriage of Pierre and Anne Tremblay in 1657 resulted in an estimated 280,000 descendants.

Now, 12 to 16 generations later, Quebec's French population has expanded by 80-fold in 300 years, while France's population has grown only six-fold during the same period. Because of Quebec's small gene pool, today's population carries only a small fraction of the total genetic variation that was present in France during the colonization. The resulting "founder effect" means that this population, originating from a small group of individuals, has high frequencies of otherwise rare alleles, which translates into genetic effects such as a higher incidence of specific genetic diseases.^2,3

To Desjardins' pleasure, from the very beginning the French Canadian population was well documented with birth, marriage, and death certificates in parish registries. In the early 1970s, Desjardins was hired as a student to read out loud the parish registries, handwritten in old French, so his partner could enter the information into the Historical Research Program in Demography database (PRDH). Now, 36 years later, "a French Canadian, in less than half a day, can easily dig up a family tree of about a thousand ancestors," he says. As he sits in his office at the University of Montréal, filled with bookshelves of transcribed parish registries, Desjardins points to a large plaque displaying half his family's genealogy. This and other databases have helped genealogists work on family trees, and geneticists study Quebec's founder effect, by identifying surviving relatives of this small group of founders.

At least 22 monogenic diseases have been found to be prevalent in regions of Quebec such as the Saguenay-Lac-St-Jean region, in the Northeast part of the province.² According to Claude Laberge, who has been studying the founder population at the University of Laval, 16% of people living in the Charlevoix region are carriers of one of four rare diseases: tyrosinemia (type 1), spastic ataxia, sensitivomotor ataxia, and lactic acidosis (cytochrome C oxidase [COX] deficiency). Contrast this, for example, with a worldwide carrier rate for tyrosinemia of 1 in 160-175.

A disease like hereditary tyrosinemia takes a toll. An autosomal recessive disorder, it can cause cirrhosis and hepatocellular carcinoma. Its estimated live birth incidence worldwide is 8-10 cases per million births; the Saguenay-Lac-St-Jean region is home to nearly 500 cases,with an estimated carrier rate of one in 22. The disease results from a mutation in the fumarylacetoacetate hydrolase (FAH) gene, and in French Canadians just one mutation accounts for 95.4% of the mutant alleles.The disease is effectively treated with nitisinone, and a long-term therapeutic trial of the compound in a Quebec cohort is providing data on its effectiveness.⁴

The carrier rates of certain diseases are "very humbling; it's pure chance," adds Charles Scriver, professor emeritus at McGill University Health Center. "Since we all carry mutations, there are certain conditions where those mutations will become manifest," he says, "and Quebec is an exciting example of this manifestation for geneticists."

Scriver has built a long career working with the Quebec population. In the 1960s, he and his colleagues set up a screening program for phenylketonuria (PKU) in Quebec. Years later, by using the PKU mutations discovered via the screening program, they traced historical PKU mutations (for example, a rare PKU allele, M1V) back to ancestors who emigrated from a small region in northwestern France. They also found that this emigration history resulted in different PKU mutations on the eastern side of Quebec versus the western side, showcasing another founder effect.⁵ This and other work has allowed the targeting of medical resources to the regions of Quebec that most need them.

Criteria of various founder populations Courtesy of Genizon
Founder Population	Estimated No. of Generations	Estimated No. of Founders	Current Population
Quebec	12–16	˜2,600	>6 million
Costa Rica	˜12	˜4,000	2.5 million
Finland (North-east)	15–20	200–400	1 million
Ashkenazi Jews	˜30	˜1,500	10 million
Sardinia	˜400	˜500	1.6 million
Newfoundland	˜16	˜25,000	500,000
Iceland	15–40	˜20,000	280,000
Umea (Sweden)	˜15	˜16,000	200,000
Mormons	7	˜43,000	1.6 million

People in Quebec have always been keen to participate in such work, Scriver says, and their history, reproduction, and isolation means that "there are opportunities here [in Quebec] for geneticists that aren't available in many other places."

In 2005, Robert Koenekoop's group at Montreal Children's Hospital Research Institute discovered a new gene (RP31) involved with retinitis pigmentosa (RP), an inherited eye disease, by working with a Quebec family that had an early-onset autosomal dominant form of retinitis pigmentosa.⁶ Koenekoop's group has also used the Quebec founder population to discover mutations in genes involved in Usher syndrome type 1, a leading cause of deafblindness.⁷

"Founder populations have been used to identify hundreds of disease genes," says Thomas Hudson, geneticist and president of Ontario Institute for Cancer Research. Of these, "dozens of genes causing rare diseases [have been] discovered in the Quebec population," he says, including mutations for autosomal recessive spastic ataxia of Charlevoix-Saguenay (ARSACS), lactic acidosis, agenesis of the callosum, and many others.

But what about the more common diseases that stem from something decidedly more complex than a handful of genes? In 1999, researchers at Myriad Genetics, a biopharmaceutical company based in Salt Lake City, began to wonder whether Quebec's founder population could reveal the genetics of common diseases such as obesity and asthma. At the same time, John Raelson, Majid Belouchi, and their colleague Nathalie Laplante (now all at Genizon) were working at Algene Biotechnologies, trying to use the Quebec founder population to study schizophrenia. Some of their findings included the discovery of a link between D-amino acid oxidase, a new gene (G72), and schizophrenia.⁸ Algene entered into negotiations with Myriad to collect samples from the Quebec founder population for an asthma study.

Algene folded, however, so Belouchi and Raelson decided to start their own company, Galileo Genomics, which became Genizon in 2005. As Galileo Genomics, they initially helped Myriad recruit from the Quebec founder population for studies on obesity, dislipidemia, and coronary heart disease. (Myriad subsequently provided $1.5 million to Galileo Genomics.) Although Myriad didn't find anything it is pursuing in the samples that Galileo Genomics had collected, Genizon is currently using those samples in a study aimed at identifying genes associated with metabolic syndrome.

Early on, John Hooper, an organic chemist with experience in drug development, joined the team. He had previously built up Montreal's Phoenix International Life Sciences before it was sold to MDS Inc. for $500 million. Upon hearing the Belouchi-Raelson business plan to hunt for specific disease genes in the Quebec founder population, Hooper was "intrigued," but he says it needed to be redirected to attract investors.

"We wrote a new business plan to address the probability of using the Quebec founder population to find genes in common diseases," says Hooper, "which is not an easy thing since common diseases appear to be caused by multiple genes, 20 or 30 maybe, plus the environment, plus lifestyle." The new business plan was to discover genes using the Quebec founder population, create unique "gene maps" (a combination of interacting genes, genetic markers, biochemical pathways, and drug targets involved in common diseases), patent these discoveries, and then license them to pharmaceutical companies.

It was a risky mandate, but Hooper liked it. Sitting in the Genizon boardroom in front of his laptop, which displays the stages of their various genome-wide association studies (24 in total), Hooper says the plan felt cutting edge back then, and it was also likely to be more attractive to investors.

Since 1999, the company has raised more than $130 million in funding from venture capital firms such as Quebec's Investissement Desjardins, the Dutch firm BTF, and European Bio Fund Management Oy Ltd. Originally using Perlegen's high-density SNP genotyping technology, Genizon has gone on to collect DNA samples from 47,000 participants in Quebec, in the search for genes involved with psoriasis, schizophrenia, Alzheimer's disease, coronary heart disease, acne, and baldness. Results aren't in on Genizon's overall success, but in 2007, the company published on how the technique facilitated the identification of new genes linked to Crohn disease, without the Quebec founder population having a higher incidence of the disease.

"Genizon has thrown a wide net across French Canadians in Quebec," says Serge Gauthier, director of the Alzheimer's Disease Research Unit at the McGill Center for Studies in Aging. Gauthier recently completed an Alzheimer's study with Genizon, in which he helped recruit patients with French Canadian parents and grandparents. In 2007, Pfizer licensed the company's gene maps created using the Quebec founder population for Alzheimer's disease, attention deficit hyperactivity disorder, and endometriosis. More recently, London Genetics Ltd began to work with Genizon to replicate the company's discoveries on a genome-wide association study of schizophrenia.

The findings are nice, but you don't need a founder population to get them, Hudson argues. To him, the main usefulness of a founder population in gene discovery research is related to the size of the population bottleneck. "[This] restricts the number of mutated alleles, such that descendents having the same allele share a large segment of DNA in common surrounding the mutated allele," says Hudson. By looking for these large tracks of DNA (haplotypes), geneticists can localize disease genes. The trick is that more founders produce more alleles coming through the population bottleneck - hence the difficulty in detecting long stretches of DNA in common among the contemporary population.

Hudson says that when he and others started collections in Quebec, Finland, and Iceland, everyone thought that the founder effect would be useful. It turns out, however, that too many founders carried a variety of common mutations, he says, leading to low statistical power when trying to detect genes for common diseases. "We were successful in finding many disease genes, but this is because we applied rigorous approaches and a lot of sweat - and not because of the founder effect." Nevertheless, Hudson notes that this doesn't diminish important discoveries in the Quebec population on complex diseases, including a notable linking of the G protein-coupled receptor for asthma susceptibility (GPRA) to asthma,⁹ work on the vitamin D receptor and asthma, and other findings related to diabetes and myocardial infarction.

Darvasi agrees that the usefulness of founder populations shouldn't be overestimated, but he argues for some small advantages such as sharing a homogeneous environment (e.g., a similar diet). This, combined with homogeneous genetics, makes a "better experiment." His team, for instance, recently completed a GWAS in Ashkenazi Jews to discover that reelin, a serine protease involved in corticogenesis, increases the risk of schizophrenia only in women.¹⁰

Talking instead about the study of the Quebec founder population as a whole, Scriver says you end up with something important: "Useful information for individuals, communities, and populations where we can say" a higher rate of disease is not the result of "a divine force and this is the cost of it. Instead, it's pure chance."

But how much longer will the Quebec founder population even exist for genetic sampling? "This may be the last generation of people over 60 where we can do founder studies," says Gauthier. Recently, it's become harder and harder to find people for his GWAS work. In Canada, interracial marriages have increased more than 30% from 2001 to 2006, according to Statistics Canada. Moreover, in Quebec, visible minorities (many of whom are new immigrants) also rose more than 30% during the same period, to 8.8% of the population. The changes definitely add to Gauthier's desire to complete the search for Alzheimer's genes with a Quebec population before another generation passes, he says. But Hooper is less worried: "I hope to have all the important gene mapping done well before that."

When I click on the little Page Tools box at the top to get a printable version, doesn't seem to work. Thanks.

This paper is very interesting, but it should be said first that most French people who came in Nouvelle-France settled mainly in the Maritimes (Acadia, now part of New-Brunswick and Nova-Scotia) and not only in Quebec. When the British conquered Nouvelle-France, a lot went in Louisiana, some came back to France, and a few headed to Quebec. A lot of these people came from Poitou-Charentes, and this can be traced back to this French region till today. This variety of origins between French people who came to Nouvelle-France is not detailed here.
Second, in this paper is mixed the effect of a small number of founders (1600-1760), with the subsequent long lasting in-breeding in some regions of Quebec like Saguenay-Lac St Jean.
Most of these mutations did occur randomly after the French settlement in Quebec, later amplified and concentrated by intensive in-breeding. Indeed, only a few can be traced back to France.
All in all, the real founder effect might be far less important than the genetic isolation and intensive in-breeding which lastled for centuries especially in the Saguenay Lac St Jean region.
Dr Hudson is perfectly right, the main thing here is the tight bottleneck and subsequent in-breeding. The "founder" effect is here (and not primarily in the small number of first settlers).

It makes no difference whatsoever regarding the genetic interest of this population.

Thanks for letting us know. The link was temporarily broken, but has been fixed.

Alison McCook
Deputy Editor

Why is it that no matter the link I click I get redirected to "A Planck Wall" instead???

Please change it!