What is a knockout mouse?

Short answer: A knockout mouse is a science-bred mouse that has a specific gene or regulatory region of the genome deleted.

More detail:

Genetics is the study of how traits are passed from generation to generation, and specifically focuses on how changes in the DNA encoding those traits can lead to individual trait differences and disease. Simply put, a phenotype is the observable outcome of a particular gene. For example, you could have a gene leading to brown or blue pigment in your eyes. If one variant of this gene (or allele) codes for blue eyes and the other codes for brown, you could end up with a brown-eyed “phenotype.” (Click here for a better explanation of alleles and dominant/recessive inheritance; for a better explanation of the genotype-phenotype distinction, check this out).

Scientists are able to figure out what different genes do by introducing mutations into the genome and then asking “what broke?” and “how did it break?” by screening for particular phenotypes. For example, you can introduce mutations with radiation and then look for cells that divide more slowly. Then, isolate these cells and ask “which genes were destroyed or changed when I introduced the mutations that caused slow cell division?”

Through screens or through genome wide association studies (which use SNPs in the human genome to track disease-associated regions), we’ve discovered many genes involved in disease processes.

When we want to study those genes further, though, we want to know exactly what the gene does and how it works with other genes to drive disease or other cellular processes. To study these, scientists often turn to model organisms and genetic knockouts.

A model organism is an organism with a fast reproductive cycle and conserved processes (as in: development in frogs approximates the same steps of development that happens in many other animals and thus represents a good system in which to study those steps). These organisms are well defined (we know a lot about them), and they are carefully bred. Choice of a model organism is guided by what process a scientist is studying; some common examples include worms, fruit flies, frogs, rats, and mice.

Model organisms allow researchers to do thoroughly controlled scientific studies that would be highly unethical in humans (where we cant manipulate the genome or define breeding pairs). These experiments allow us to understand basic processes in biology as well as learn about new drugs and therapies.

Model organisms can be used to study the impact of a gene on development and physiology through a genetic knockout. Scientists can engineer the genomes of some model organisms such as flies or mice, and can selectively delete a gene from specific tissues or the entire organism. These mutants are called “conditional knockouts” and “knockouts,” respectively, because the gene has been knocked out of the organism’s genome.

Many knockout mice are already available to help researchers study processes from cancer to neurodegeneration.

23andMe: my foray into personalized genomics

Just over a year ago the personalized genomics company 23andMe dropped the price of their genetics testing service to $99. My family and I decided to try their product after hearing a few stories of people who gained valuable health information as a result of the service. As an added bonus for my dad, a genealogy buff, the service includes ancestry tracing and an option to ‘share genomes’ with potential distant relatives.

23andMe takes a sample of your DNA (from spit) and analyzes a set of markers at thousands of locations across your genome called single nucleotide polymorphisms (SNPs). Since this analysis uses a small plastic chip coated with probes to query these sites, this genome-wide genotyping analysis is called a “SNP-chip.” Basically it asks which of the letters that make up the genetic code, A, T, C, or G, occur at a particular spot in your DNA, and then repeats this question 32,000 times. SNPs are essentially little genetic flag posts. The presence of a certain SNP (an A at a site that usually carries a G) can indicate a predisposition for heart disease, or may be associated with curly hair.

We gathered around the table and spit into our little sample collection tubes, registered online, sent the samples off in the mail, and waited.

Two of us opted in to their ‘contribution to research’ and began filling in prescription histories, general health background information, and answering questions such as whether or not our mother smoked while pregnant with us. Two of us opted to not share any further information at this time.

These answers, along with our self-reported medical history and drug responsiveness are entered into the 23andMe data crunchers to be processed in studies called Genome Wide Association Studies (GWAS). GWAS studies look for correlations between SNPs and a particular trait such as responsiveness to a drug, ability to curl your tongue or wiggle your ears, or the likelihood of developing breast cancer.

Within one to two months of sending our samples in, our data was processed and uploaded, and we could read through a carefully curated list of genetic risks, carrier status (diseases we could potentially pass on to our children), fun traits, and ancestry information.

The test is well worth the $99 just for the ancestry and fun trait information. For those who know what they’re doing with the raw SNP data, these files are also made available. The selling point, though, was the genetic risk and carrier status information. I have a fairly complete medical history from almost all of my immediate relatives going back at least two generations, so I had a pretty solid idea for what to expect for the most part, but having a gene-based confirmation of those assumptions was massively relieving. Further, I identified a trait or two which I deemed actionable, and I feel confident that the measures I’m taking post 23andMe revelation are making me healthier and happier.

It’s important to keep three things in mind, though, when going through this genetic data.
1. There are always error rates associated with these tests. If I were to send in my spit four more times I may get slightly different genotypes for any given SNP in each of those samples.
2. This is a SNP-Chip and associated GWAS studies, not sequencing and molecular mechanism. SNPs are often simply associated with functional regions of genetic code, and may not actually indicate an error in a functional gene. And GWAS studies are correlations, and don’t always indicate causation. Take these annotations with a grain of salt. An exact DNA sequence for your entire genome would be far more informative, and is probably on the way. Even so, for most of these disease correlations, vague association/correlation is as good as it gets at this point in the game.
3. These are DNA samples from your spit. No amount of habit changing or exercise will change these results, but they’re also not written in stone. You have the same DNA in every one of your cells, but a skin cell is different from a liver cell or a brain cell. There are many layers of regulation governing which genes get turned on when and where. Just because you have a gene associated with a predisposition to high blood pressure doesn’t mean you’re doomed to develop high blood pressure. This just means that perhaps you should consider a diet lower in sodium and a more active lifestyle.

23andMe makes a significant effort at educating their clients on exactly what they’re purchasing with the service, and offers loads of mini biology lessons. However, with direct-to-consumer genotyping making such intimate health details available without a physician intermediary the FDA was bound to get involved sooner or later. Since people may start making medical choices based on their data, the FDA ordered a cease and desist while it investigates regulating the genetics test as a medical device. For the time being, the health and disease risk results are shut down to newcomers on the site, but the genetic trait and ancestry information is still available.

Ribozymes are catalytic RNAs!

I’m studying for my second qualifying exams (the exam which, should I pass, will officially qualify me as a PhD candidate). In an effort to bolster my general knowledge of all things RNA-related, I’m reading a textbook called the “Molecular Biology of RNA.”

This book as an entire chapter devoted to catalytic RNAs. Since my days as an AP Biology student I’ve known that some RNA molecules can behave as enzymes and catalyze reactions, but this chapter opened up the world of ribozymes as I’ve never understood them before.

A Ribozyme is an RNA molecule that can catalyze a reaction. The ribosome is probably the most famous ribozyme, which catalyzes peptide bond formation, but other RNAs exist that also behave as true catalysts. A true catalyst is not destroyed or changed by the reaction it speeds up.

Some RNAs with enzymatic activity ARE exhausted by the reactions they catalyze, and the chemistry here is equally fascinating. These include self-splicing RNA introns and self-cleaving viral genomes.

Most catalytic RNAs use acid-base chemistry and leverage the heightened reactivity of the ribose sugar via the 2′-OH functional group.

Let’s take a step back. To understand how RNA can act as an RNA-cutting enzyme, we need to know two things. 1) how structure and function are intimately related in biology, and 2) how generally unstable RNA is as a polymer.

1) Structure and Function:

This is something I also learned in high school biology that I took for granted until late in college. Primary sequences of proteins and RNAs are closely tied to secondary or tertiary structures, and these are intimately related to how a molecule functions. In the case of RNAs, primary sequence influences base-pairing or secondary structure, and RNA molecules can fold up into tertiary structures that bring reactive groups into close proximity. All you need is an environment that encourages this folding and can stabilize the reactive groups in this close proximity.

2) RNA instability

RNA is less stable than DNA. The namesake ribose sugar has an extra functional group (-OH) attached to carbon #2. This functional group is reactive, and particularly so when de-protonated. The partially-negatively charged oxygen can attack the phosphate in RNA’s phosphate backbone, triggering cleavage of the backbone.

(I understand pictures would be greatly useful here…I’ll get on that)

So for enzymatic RNAs to function, all we need is creation of an active site by RNA-folding, and then stable acid-base chemistry to trigger cleavage. This can occur within the RNA sequence itself (self-cleavage) or by sequence-recognition in a separate RNA molecule (site-specific cleavage).