CRISPR, a gene-editing technology borrowed from humble bacteria, has taken the biomedical research world by storm. Researchers can use the technique to inactivate genes, add point mutations, or insert entire genes into virtually any spot in the genome of any species. Neuroscientists are just starting to dabble with CRIPSR’s gene-editing recipe, and the first studies are cropping up (see part 1). CRISPR promises to speed up production of transgenic animals, reverse fatal diseases in animal models, and maybe one day even correct disease-causing mutations in people.
Transgenic Animals Made Easy
In addition to manipulating cell and animal genomes, another major advance CRISPR offers is the rapid generation of transgenic animals. Such models have been the cornerstone for much of the basic and translational research in the neurodegenerative disease field, but they require an extraordinary amount of time and resources to make. The tried-and-true way to generate transgenic mice involves transfecting embryonic stem cells with the chosen DNA, selecting transfected cells, injecting them into mouse blastocysts, and eventually selecting for offspring in which the transgene jumps into the germline. In contrast, simple direct injection of Cas9 plus multiple guide RNAs into mouse embryos has allowed researchers to skip several of these steps and produce multi-knockout mice in one fell swoop.
Mike Sasner of The Jackson Laboratory in Bar Harbor, Maine, expressed enthusiasm about CRISPR’s ability to accelerate development of transgenic mice. “It’s trimming the first part of the pipeline significantly, in terms of time and cost, and that means more mice can be made.”
In a recent study, researchers led by Rudolf Jaenisch at the Massachusetts Institute of Technology used the technique to disrupt up to five genes—Tet1, 2, and 3, and the Sry and Ury genes. The Tet genes are involved in the production of 5-hydroxymethylcytosine, and knocking down more than one of them at once allowed researchers to avoid the pitfall of redundancy that frequently undermines single-gene knock-out models (see Wang et al., 2014).
Benedikt Wefers of the German Center for Neurodegenerative Diseases (DZNE) in Munich is using CRISPR to make mice with AD-associated point mutations, as well as knock-ins of human disease-related genes. The homology-dependent process of adding a specific mutation into a mouse gene or a whole human gene is less efficient than using NHEJ to inactivate genes, Wefers said. Hence he found that most of the initial offspring harbored inactivated genes rather than added ones. “If you are interested in a knock-in or point mutations, you have to screen more mice than you do for knockouts,” he said. Wefers, who directs DZNE’s transgenic-mouse unit, estimated that the homology-directed repair works about 5 percent of the time, so the researchers typically screen 30 to 60 pups to find appropriate founders. Still, he said the process is far more streamlined than creating mice by traditional techniques based on embryonic stem cells.
Other transgenic animals could be generated in the same way. The creation of transgenic mammals has, for the most part, been limited to mice, JAX’s Sasner added, because embryonic stem cell technology does not readily work in other species. With this step removed, researchers could theoretically generate a menagerie of transgenics. “CRISPR can work in any species,” Sasner said. “It is a huge step forward.” Researchers have already used the technique to make transgenic monkeys with simultaneous alterations in the recombination activating gene 1 (Rag1) and peroxisome proliferator-activated receptor gamma (PPAR-γ) genes, which are crucial for adaptive immunity and metabolism, respectively (see Niu et al., 2014).
Zebrafish have proven to be easy targets for CRISPR, according to Bettina Schmid, also at DZNE. “We are absolutely fascinated by the CRISPR/Cas9 system. It allows us to do experiments we only were dreaming about years ago,” she wrote in an email to Alzforum. “It is easy, cheap, everybody can do it, and it is efficient.” Her lab started off using CRISPR to inactivate genes in zebrafish, and is now introducing point mutations associated with neurodegenerative disease.
Sasner said that he expects he will soon see CRISPR-built transgenic mouse models of neurodegenerative disease, although none have yet been submitted to the Jackson Lab. This not-for-profit organization maintains, and supplies researchers with more than 7,500 well-characterized mouse strains, including many Alzheimer’s models. “In 2015, I think we are going to see a flood of mice created with this technology,” he said. Sasner added that CRISPR will allow mouse modelers to keep closer pace with the discovery of new disease-associated genes and genetic variants found in massive genome-wide association studies. Rather than rolling the dice by targeting a single gene with a single mutation, researchers can more readily make multiple mice that model several different GWAS hits.
Todd Golde at the University of Florida in Gainesville agrees. “It’s going to be very hard to use traditional means to find out how all of these novel [GWAS] genes supposedly influence disease,” he said. “CRISPR gives us a faster, more facile tool to do that.”
Rather than generating transgenic animals from the embryo stage, Golde plans to use CRISPR to inactivate genes in mice after they are born. In collaboration with Edgardo Rodriguez, a new investigator at the University of Florida who works on CRISPR delivery systems, Golde plans to use adeno-associated virus (AAV) vectors to target CRISPR directly to the central nervous system (CNS). The scientists plan to inactivate genes while avoiding potential pitfalls associated with transgenic mice, such as developmental defects and uncertainty about the influence of genetic background, he said. While Golde hopes eventually to examine the effects of novel disease genes, his lab will begin their venture into CRISPR territory by looking at the effects of knocking down familiar culprits, such as BACE1 and APP.
Beyond mice, CRISPR will allow researchers to rapidly model a slew of genetic mutations in human cells. For example, researchers will be able to easily introduce, or correct, mutations in induced pluripotent stem cells (iPSCs) derived from patients, and then study the effects of those manipulations in a variety of differentiated cell types, including neurons.
“Anyone studying iPS cells would be crazy not to pursue this technology,” said Mathew Blurton-Jones of the University of California, Irvine. Blurton-Jones plans to use CRISPR to compare the effects of disease-associated mutations in the TREM2 gene, a known risk factor for AD. He will study the effects of either correcting the mutations in iPSCs derived from AD patients, or introducing them into iPSC lines from normal controls. Differentiating these cells into microglia, which express TREM2, will allow him to model disease and hunt down pathogenic mechanisms.
CRISPR for Therapy
Beyond basic and translational research, will CRISPR ever work as gene therapy? The answer to this question isn’t yet in, but some researchers are proposing that an early therapeutic application could be to correct mutations or add therapeutic genes to iPSCs before transplanting them back into the donor (see Li et al., 2014, and Kim et al., 2014). The hope is that such cells may treat diseases including amyotrophic lateral sclerosis, Parkinson’s, and perhaps even Alzheimer’s.
Along those lines, Blurton-Jones plans to use CRISPR to insert potentially therapeutic genes—such as the growth factor BDNF or the Aβ-degrading enzyme neprilysin—directly into iPSCs. In a recent study, he reduced pathology in AD mouse models by injecting them with iPSC-derived neural precursors overexpressing neprilysin (see Blurton-Jones et al., 2014). However, he inserted that transgene randomly into the cell genome. “That’s fine for a basic science experiment, but not for the clinic,” Blurton-Jones said, adding, “We would need to target a specific locus that we know is safe, and CRISPR dramatically increases our ability to do that.”
There are early hints that CRISPR could even be used to wipe out genetic diseases. A study published in Science Express on August 14 reported preventing Duchenne muscular dystrophy in mice through CRISPR repair of a single disease-causing genetic variation (see Long et al., 2014). Led by Eric Olson at Univeristy of Texas Southwestern in Dallas, first author Chengzu Long and colleagues corrected a point mutation that creates a stop codon in the dystrophin gene, which is crucial for muscle function. They injected mouse embryos with DNA encoding Cas9, guide RNAs for the mutated gene, and a wild-type replacement fragment. Dystrophin expression was restored in the resulting pups, and their muscle fibers never degenerated as the mice grew. Olson told Alzforum that his lab is now pursuing the next step toward a therapy, i.e., to reverse causative mutations in adult mice.
In another example, earlier this summer researchers led by Daniel Anderson at the Massachusetts Institute of Technology used CRISPR to rescue adult mice from hereditary tyrosinemia type I (HTI), a fatal liver disease caused by a point mutation in the fumarylacetoacetate hydrolase (FAH) gene (see Yin et al., 2014). The researchers delivered CRISPR/Cas9 into the mice’s liver via hydrodynamic injection, a method that uses large volumes of fluid to force nucleic acids into target cells. This delivery method causes some cellular damage and is unlikely to be applied to humans.
The major obstacle that stands in the way of CRISPR as a therapy for neurodegenerative disease is the same as for other potential gene therapies, that is, crossing the blood brain barrier (BBB) to deliver the genetic cargo to target cells. Scientists have tried to overcome it by using the adeno-associated virus AAV9, which crosses the BBB and infects neurons and astrocytes. Alas, AAV9 is too small to readily carry the Cas9 gene plus the other components that would be needed for CRISPR. Some researchers are trying to pare the Cas9 sequence back to its bare essentials to squeeze it into AAV9, while others propose using multiple viruses containing different parts of the gene that splice together when expressed in the same cell. A study in the September 4 Biotechnology Journal reported generating an “AAV toolbox” for CRISPR, which managed to pack the Cas9 and guide RNA sequences into one vector (see Senis et al., 2014). The researchers used this streamlined vector to target liver cells in mice.
The University of Florida’s Rodriguez is optimizing the AAV system to deliver CRISPR to the CNS. While he prefers to keep the details of his upcoming study under wraps, he hinted to Alzforum that the Cas9 protein derived from S. pyogenes represents but one form of the nuclease, and those from other bacteria could potentially work better with AAV or perform different functions. “There is an immense number of bacterial species that one can tap to develop these systems, so we have an enormous chest of tools,” he said.
Casey Maguire of Harvard University, who develops systems for delivering gene therapies into the brain, thinks AAV’s capacity may be the least of the problems. “When people want to fit something into AAV, they usually find a way to do it,” he said. But once the CRISPR system is fashioned to fit into AAV, other hurdles may lie ahead, he predicted. These include immune responses to the bacterial Cas9 protein and sustained expression after AAV infection. “Long-term expression is great if you want to overexpress a gene,” Maguire said. “But with a [gene] correction system, you want to correct it and get the heck out of there.” Maguire’s concern stems from the potential for off-target effects of the Cas9 nuclease, which has been shown to latch onto the wrong sequences in some cell culture studies. Theoretically this could lead to unpredictable disruptions in essential genes, although studies in mice and monkeys have not shown evidence of this effect so far. Some claim that off-target effects are less of a concern in vivo, where lower amounts of CRISPR components are transfected per cell. Researchers are also rapidly developing tricks to maximize CRISPR’s specificity (see Fu et al., 2014; Veres et al., 2014; and Tsai et al., 2014).
Maguire suggested non-viral delivery systems such as exosomes as one strategy to avoid the risks of long-term expression. Exosomes are membrane-bound vesicles that are shed from cells and can be engulfed by other cells. Because researchers could package them with therapeutic nucleic acids and proteins, exosomes derived from a patient’s own cells could become a promising delivery system, Maguire said.
Keith Joung of Massachusetts General Hospital in Charlestown has for the past decade worked on gene editing and tweaked CRISPR to boost its specificity. He told Alzforum that the sheer amount of enthusiasm and effort researchers are now bringing to this technology should propel it forward rapidly. “It’s a very exciting time. I think the technology really is transformative, both in terms of what it can do for the research community and in terms of therapeutics.”
Science. 2014 Sep 5;345(6201):1184-8. Epub 2014 Aug 14 PubMed.. Prevention of muscular dystrophy in mice by CRISPR/Cas9-mediated editing of germline DNA.
Biotechnol J. 2014 Sep 4; PubMed..CRISPR/Cas9-mediated genome engineering: An adeno-associated viral (AAV) vector toolbox.
Cell. 2013 May 9;153(4):910-8. Epub 2013 May 2 PubMed.. One-step generation of mice carrying mutations in multiple genes by CRISPR/Cas-mediated genome engineering.
Cell. 2014 Feb 13;156(4):836-43. Epub 2014 Jan 30 PubMed.. Generation of gene-modified cynomolgus monkey via Cas9/RNA-mediated gene targeting in one-cell embryos.
J Biol Chem. 2014 Feb 21;289(8):4594-9. Epub 2013 Dec 20PubMed.. A cut above the rest: targeted genome editing technologies in human pluripotent stem cells.
Stem Cells Dev. 2014 Jul 30; PubMed..Genomic editing tools to model human diseases with isogenic pluripotent stem cells.
Stem Cell Res Ther. 2014 Apr 16;5(2):46. PubMed.. Neural stem cells genetically-modified to express neprilysin reduce pathology in Alzheimer transgenic models.
Science. 2014 Sep 5;345(6201):1184-8. Epub 2014 Aug 14PubMed.. Prevention of muscular dystrophy in mice by CRISPR/Cas9-mediated editing of germline DNA.
Nat Biotechnol. 2014 Jun;32(6):551-3. Epub 2014 Mar 30 PubMed.. Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype.
Biotechnol J. 2014 Sep 4; PubMed.. CRISPR/Cas9-mediated genome engineering: An adeno-associated viral (AAV) vector toolbox.
Nat Biotechnol. 2014 Mar;32(3):279-84. Epub 2014 Jan 26 PubMed.. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs.
Cell Stem Cell. 2014 Jul 3;15(1):27-30. PubMed.. Low incidence of off-target mutations in individual CRISPR-Cas9 and TALEN targeted human stem cell clones detected by whole-genome sequencing.
Nat Biotechnol. 2014 Jun;32(6):569-76. Epub 2014 Apr 25 PubMed.. Dimeric CRISPR RNA-guided FokI nucleases for highly specific genome editing.
To view commentaries, primary articles and linked stories, go to the original posting on Alzforum.org here.
Copyright 1996–2014 Biomedical Research Forum, LLC. All Rights Reserved.