to BioTechniques free email alert service to receive content updates.
Editing the Genome Here, There, and Everywhere

Casey McDonald

For over thirty years, researchers have dreamt of editing disease-associated mutations out of the genomes of human patients. Casey McDonald takes a look at how these dreams are slowly becoming a reality through a variety of targeted and genome-wide techniques.

Bookmark and Share

In June 2009, the pharmaceutical company Genzyme detected a virus in a bioreactor in its Allston Landing facility. The virus impaired the growth of CHO cells used to produce its products. As a result, the company stopped production for over a month to sanitize the plant, creating a temporary shortage of two of the company’s drugs. While the company promised improved screening of the cell media components, which were identified as the probable source of the contamination, researchers have a better solution: edit the genome.

ZFNs designed by companies like Sangamo Biosciences can cut at target sequences using restriction nucleases bound to DNA binding domains. Source: Sangamo Biosciences

“[A new genetic code] could genetically isolate organisms, making them safer, creating a genetic firewall, rendering organisms resistant to viruses,” says Yale University researcher Farren Isaacs.

The genome is often described as a recipe book, providing the instructions on how to create the proteins that work together to maintain life. By tweaking and editing these recipes, researchers could make cells resistant to viral infections so as to support pharmaceutical production or replace disease-causing genes with healthier ones to cure human patients.

Despite these prospects and decades of research, editing genomes remains a nontrivial task. But this past summer, several teams have reported new genome editing techniques for both targeted and genome-wide editing that may usher in a new era beyond simply reading and analyzing base-pairs.

Comparative Literature

At the University of California, Berkeley, genetics professor Barbara Meyer’s lab uses the roundworm Caenorhabditis elegans as a model to study various aspects of molecular biology, including sex determination during development and epigenetic control of X-chromosome gene expression.

But because of the lack of site-directed heritable mutagenesis techniques in C. elegans organism, working with this species is sometimes frustrating.

“It’s been a 30 year quest to be able to control how you mutagenize C. elegans,” says Meyer. “It’s such a great model organism, but yet there are limitations.”

For example, to study the functional aspects of genes, researchers use induce mutations in their model organisms by exposure to mutagenic chemicals or radiation. After this mutation, the researchers then screen for a phenotype and isolate the gene associated with that particular phenotype. However, these mutation-induction techniques are, for the most part, random.

In contrast, RNAi gene knockdown experiments are precisely targeted. In these experiments, researchers disrupt the function of a particular gene by introducing the corresponding siRNA for that gene’s mRNA. Once the gene’s expression is decreased, scientists can determine how that particular gene affects the cell’s phenotype. But the problem is that these methods are not feasible for the high-throughput screening of a multiple genes.

So, working with researchers from Sangamo BioSciences, Meyer’s team developed two genome-editing techniques for targeted genomic modifications in C. elegans (1). In their first approach, the team engineered zinc-finger nucleases (ZFN) to produce a double-stranded break at a specific site in C. elegans genome. These breaks are then repaired by the cell’s non-homologous end-joining DNA repair machinery, which is error-prone, thereby creating heritable mutations in the targeted gene.

“I’ve always been enamored with the ZFN technology because it gave us that possibility, that we could at will knock out anything we wanted,” says Meyer. Using the ZFNs, Meyer’s team targeted the gene ben-1 in C. elegans and the gene sdc-2 in the related species C. briggsae. In both experiments, the researchers found that the ZFNs mutated the targeted genes without significant off-target results. The ability to target a gene in both C. elegans and C. briggsae will enable a cross-species, evolutionary comparison of the gene’s function.

But engineering ZFNs with minimal toxicity has been a rather difficult and expensive process, restricting its widespread adoption. In contrast, the recently introduced transcription activator-like effector nucleases (TALENs) can produce similar targeted double-stranded breaks but without the need for complicated protein engineering. “TALENs are much easier to design. You can really predict exactly what sequence they’ll bind to. TALENs have now become the best way to approach things,” says Meyer.

Using the TALENs in the same ZFN protocols for targeted mutagenesis, Meyer’s team targeted the ben-1 region of the C. elegans. The group found similar results for both ZFNs and TALENs, showing that both were viable additions for a genome-editing toolbox.

Find and Replace Mutations

C. briggsae is separated from the more commonly studied model worm C. elegans by 15-30 million years of evolution. Source:

At the Children's Hospital of Philadelphia (CHoP), researcher Katherine High is looking to cure hemophilia and other congenital diseases. And she believes that Sangamo’s ZFN technology might be part of the solution.

“A long standing goal of the field has been the concept that if you could get correction of a mutant gene in some physiologically relevant target tissue or cell type, then you would reconstitute the mutant gene under the control of all the regulatory signals. And you would be assured of long lasting expression because you’ve actually corrected the genome and so that the correction would be past all the daughter cells,” says High.

But there are several concerns about using ZFNs to edit disease-causing mutations in human patients. For example, ZFNs must be designed with extreme specificity to a particular gene in order to minimize off-target affects that could cause disease or even cell death.

In a paper published in Nature in 2005 (2), Sangamo researchers reported that ZFNs efficiently corrected an X-linked severe combined immune deficiency (SCID) mutation in cultured cells. These cells carrying the healthy, corrected gene could then be inserted into the body where they could proliferate.

So High and her colleagues at CHoP’s Center for Cellular and Molecular Therapeutics were interested to see if ZFNs could do the same in vivo. “The tissue we were most interested in, the liver, which is where you produce clotting factors, does not lend itself well to being taken out, manipulated and put back in,” says High.

To develop an in vivo genome editing approach, High’s team decided to target a mouse model for hemophilia B. This model contains a mutation in the gene F9 that results in a deficiency of coagulating protein factor IX. The team injected two hepatotropic adeno-associated vectors into the liver cells of living mice; one contained ZFNs designed to cleave the gene F9, the other contained wildtype exons 2-8 to replace the disease-causing mutation within that gene during DNA repair. The treated mice began producing factor IX at levels close to normal, essentially curing the disease in these animals. The team reported their results in a paper published in Science in June (3).

New Genetic Language

After dividing the E. coli genome into 32 regions, large scale genome engineering is carried out by a hierarchical assemble message called CAGE. Source: Science

While researchers at George Church’s Harvard Medical School lab are interested in editing the genome, they have an entirely different concept in mind. Instead of just replacing base-pairs, Church’s group is interested in changing the entire genomic alphabet altogether.

“There’s a lot of concern regarding the use of genetically modified organisms and the fact that they can spread throughout the environment,” says Yale researcher and one of Church’s collaborators Farren Isaacs. “Creating organisms that are biologically contained, created [with a] genetic dependency on synthetic amino acids, this could insure that they can only grow in specific environments.”

In 2009, Isaacs, working with Church and other researchers from Harvard, introduced a technique to introduce produce multiple mutations across the genome in single cells or populations of cells (4). The technique, called multiplexed automated genome engineering (MAGE), introduces synthetic oligonucleotides containing sequence changes into the cells which are then incorporated into the genome by homologous recombination. This allows the production of billions of genomic variants within a day, significantly more than could be achieved by previous methods. By using this method, researchers could quickly optimize cells used in industrial and commercial applications such as pharmaceutical manufacturing and biofuel development.

More recently, the team developed a method to produce a bacterial cell line in which codons at 314 locations in a genome were replaced. The technique, called hierarchical conjugative assemble genome engineering (CAGE), uses MAGE and exploits the process of bacteria conjugation (5). Through conjugation, one bacterial strain exchanges a region of its genome with another strain.

“If you can control that process to transfer the part that you want and leave behind the part that you want, not overwriting the region in the recipient that you want to keep, that is an extremely powerful way of doing large-scale genome stitching,” says Massachusetts Institute of Technology researcher Peter Carr who collaborated with Isaacs and Church on the project.

In their paper, the researchers created 32 strains of Escherichia coli, each of which had ten different codons replaced (except for one, which only had four codons replaced). They then guided one strain to exchange genomic regions with another strain by conjugation to produce a strain with 20 modified sites. To guarantee that the desired portions of the genome were transferred during conjugation, they placed selectable markers at strategic points defining the boundaries of the incoming segments as well as the important part of the recipient genome. Then the researchers selected the strains that contained both modified regions. “This requires a bit of artistry using the right markers and placing them correctly, but once you figure out the details of how to get it to work, then it is highly repeatable,” says Carr.

The resulting strain with 20 modified sites was then combined with another strain with a different set of 20 modified sites to generate a strain with 40 modified sites, and so forth to 80 sites, 160 sites, and then a single strain containing 314 replaced codons. As a result, the group produced massive genome-wide changes in the cells. Now, they are looking beyond just changing base-pairs in the genome to incorporate unnatural amino acids into proteins, expanding the chemical repertoire of proteins.


  1. Wood A.J., Lo T.W., Zeitler B., Pickle C.S., Ralston E.J., Lee A.H., Amora R., Miller J.C., et al. 2011. Targeted genome editing across species using ZFNs and TALENs. Science. 333: 307.
  2. Urnov FD, Miller JC, Lee YL, Beausejour CM, Rock JM, Augustus S, Jamieson AC, Porteus MH, Gregory PD, Holmes MC. 2005. Highly efficient endogenous human gene correction using designed zinc-finger nucleases. Nature 435:646-51.
  3. Li H., Haurigot V., Doyon Y., Li T., Wong S.Y., Bhagwat A.S., Malani N., Anguela X.M., et al. 2011. In vivo genome editing restores haemostasis in a mouse model of haemophilia. Nature. 475: 217-221.
  4. Wang HH, Isaacs FJ, Carr PA, Sun ZZ, Xu G, Forest CR, Church GM. Programming cells by multiplex genome engineering and accelerated evolution. 2009 Nature 460:894-8.
  5. Isaacs F.J., Carr P.A., Wang H.H., Lajoie M.J., Sterling B., Kraal L., Tolonen A.C., Gianoulis T.A., et al. 2011. Precise manipulation of chromosomes in vivo enables genome-wide codon replacement. Science. 333:348-353.

Keywords:  genomics genome editing