Genome Complexity Amplified (mentions CMT)

[Guest guest]

Genome Complexity Amplified

Copy number variation adds further complexity to the already

enormous task of assigning faulty genes to disease phenotypes.

Netterwald, PhD, MT (ASCP)

Senior Editor

http://www.genpromag.com/CurrentIssue.aspx

Variability is an inherent part of nature. It is omnipresent,

especially in the natural world, where diversity is king. Genomes of

the various life-forms on earth are, by their very nature, variable.

But diversity could not be present without genomic variability. And

so it goes that biologists want to understand variability at all of

its levels, from the gene to the whole organism. And therein lies

the reason why scientists study one form of genetic variability,

copy number variation (CNV).

" The reason that we study CNV is that it is one more factor that we

have to take into account in terms of human genetics and the

association between common diseases and human polymorphisms (and

variations in general), " says Jiannis Ragoussis, PhD, head of

genomics at The Wellcome Trust Center for Human Genetics, Oxford

University, Oxford, UK.

The first case to associate CNV and a specific human disease was

made by R. Lupski, MD, PhD, Cullen professor and vice chairman

in the Department of Molecular and Human Genetics and a professor of

pediatrics at The Baylor College of Medicine, Houston, Texas. It was

1991, the human genome sequence was a long way from being completed,

but somehow Lupski was able to find CNV in the myelin protein gene,

PMP22. The protein PMP22 makes up only part of the multi-subunit

myelin structure—which is necessary for conducting nerve cell

impulses. " Perturbing the ratio or stoichiometry [of one of the

subunits of myelin] may be enough to perturb the entire structure, "

says Lupski. And that's only one example of the havoc that CNV in

human genes can cause.

After these early discoveries, there have been a number of examples

of CNV-associated human diseases. " It is clear that there will

definitely be a component that comes through CNV and either

predisposes or is directly responsible for diseases that have a

genetic basis, " says Ragoussis. Also, there is already evidence for

CNV's impact. According to Ragoussis, about 17% of the differences

in gene expression between individuals can be linked or associated

with CNV. CNV is found in loci associated with genetic disease—for

example, spinal muscular atrophy on chromosome 5 and Charcot-Marie-

Tooth disease on chromosome 17. CNV is also directly implicated in

complex traits, such as CCL3L1 copy number and susceptibility to

AIDS.

But it was long road to get to that point, to find those examples.

And this was due in part to a lack of tools to make the discoveries.

The tools

" What made it challenging before was that we did not have the

genomics tools, " says Lupski. " But what we have now is a high-

resolution genome enabling detection of alterations of the genome

such as deletions that were too small to be detected by chromosome

analysis and too large to be seen by sequencing technologies, " he

says.

What came to the rescue and made the CNV field a reality by

providing that high resolution was the microarray. " There are many

different genome-wide array platforms available, but what they have

in common is that they each try to recapitulate the human genome at

a given resolution " , says Lee, PhD, FACMG, Director of

Cytogenetics for the Harvard Cancer Center, Harvard Medical School,

Boston.

The main tool used to study CNV is array comparative genomic

hybridization (aCGH)—microarrays typically composed of bacterial

artificial chromosome (BAC) clones, each containing a segment of the

genome of interest. aCGH can be used to analyze part of a chromosome

to look for chromosomal aberrations including copy number variation.

There are two basic types of these arrays. One type is a targeted

array in which a region containing a suspected chromosomal

rearrangement is targeted on the array. The number of clones on this

type of array ranges from about 1,000 to 1,500. This type of array

is used in the clinic to diagnose chromosomal disorders. The other

type of array, called the tiling array, has about 30,000 clones on

it, making it more expensive and less practical for clinical testing.

The way this is all done is similar to using a gene expression

array, involving differential labeling of genomes, mixing and co-

hybridization. And because the hybridization is competitive, an

equal number of copies of a particular locus in each sample will

yield a 1:1 ratio. However, any deviation from this 1:1 ratio

provides evidence for a gain or a loss in the number of copies of

particular locus, i.e., copy number variation.

" When we looked at the representation of Illumina's [infinium] 300K

to 650K chips and the Affymetrix 500K chip in terms of covering the

regions where CNVs have been detected, we found that the coverage

was between 50 and 60% for the Infinium and a bit higher for the

Affymetrix chip, but never 100%, " says Ragoussis. But the companies

have worked out these issues, he says, adding that both companies

have developed new generations of chips but also produce

supplementary assays to cover the regions not covered in the earlier

chips.

Challenges

Despite the impact and importance of CNV in human diseases and the

capabilities of the tools used to study it, there are still

challenges to studying CNV. One researcher who knows the challenges

click to enlarge of CNV research all too well is Manolis

Dermitzakis, PhD, investigator at The Wellcome Trust Sanger

Institute in Hinxton, Cambridge, UK. " The problem is

multidimensional, " Dermitzakis says of the challenges specific to

CNV research. And he explains the reason for this: CNV has a much

wider range of allelic options than just the usual two. In fact,

there can be anywhere from two to sixty copies of a particular

region of the genome...And it is not a trivial task to measure

copies of a gene due to limitations in the technologies, he says.

The biological effects of CNV are even more difficult to

interpret. " When you have 15 copies, you don't know if the copies

are nearby the original or somewhere else in the genome. And so, the

actual effect may not be due to the fact that there are extra copies

of a particular region, but could be due to where the extra copies

landed in the genome, " says Dermitzakis. And more importantly, if

the extra copy lands in the middle of an essential gene, it can

disrupt that gene's function, analogous to a stop codon.

As with the current state of any new field, there are challenges

that researchers hope to overcome in the future. " In the future,

there is going to be an attempt to try to capture all of the CNV in

healthy individuals, to determine their structures, and to catalog

the information in a meaningful manner . . . and that will happen

over the next three years, " says Lee. " We know that CNV is not just

in humans but also in other mammals and vertebrates, as far down as

zebrafish, " Lee says and explains that because CNV has been found in

many animal models of human disease, they are important to medical

research.

And as for the tools, Lupski says " The future of this whole thing is

oligo arrays and the future is now. "