«PLANT GENOME ENGINEERING WITH SEQUENCE-SPECIFIC NUCLEASES: METHODS FOR EDITING DNA IN WHOLE PLANTS A DISSERTATION SUBMITTED TO THE FACULTY OF THE ...»
PLANT GENOME ENGINEERING WITH SEQUENCE-SPECIFIC NUCLEASES: METHODS FOR
EDITING DNA IN WHOLE PLANTS
A DISSERTATION SUBMITTED TO THE FACULTY OF THE UNIVERSITY OF MINNESOTA
Nicholas J. Baltes
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF
Advisor: Daniel F. Voytas
August 2014 © Nicholas J. Baltes, 2014 Acknowledgements I would like to express my sincere gratitude to everyone in the Voytas lab, both past and present.
And I would like to specifically thank my advisor, Dan Voytas, for his support and encouragement.
i Dedication For my wife, Ashley, and my daughter, Adalyn.
ii ABSTRACT The development and function of all living organisms, from bacteria to humans, is encoded within a universal blueprint–deoxyribonucleic acid (DNA). The ability to re-write this code of life promises great benefits, ranging from a better understanding of gene function to correcting genetic diseases. Therefore, there is high value for tools and techniques that enable genome editing in living cells.
Two revolutionary discoveries have facilitated the development of current tools and methodologies for genome engineering. The first came from studies in yeast and mice demonstrating that synthetic donor DNA–fragments of DNA containing homologous sequences to a chromosomal target–can recombine with the target through homologous recombination, thereby incorporating the information carried by the synthetic donor into the genome. Whereas these results established a method for editing genomic DNA, the absolute frequency of recombination was quite low, ranging from one correctly-modified cell in 10 to 10. The second discovery came from studies demonstrating that damaging DNA greatly increases recombination frequencies between like sequences. Together, these discoveries have led to a powerful method for efficiently editing DNA: delivering a donor DNA while simultaneously breaking the target locus.
In the last 20 years, multiple classes of enzymes have been developed that can be ‘rewired’ to recognize and break a DNA sequence of interest. These enzymes (herein referred to as sequence-specific nucleases) have proven to be powerful reagents for editing DNA in highereukaryotic cells. However, the ability to modify plant DNA does not solely depend on the activity of the genome engineering reagents. Instead, it also depends on the efficiency with which the genome engineering reagents are delivered, the cells they are delivered to, and the effectiveness of selecting (or screening) for cells with the desired modification.
Studies within this dissertation seek to develop novel methods for delivering genome engineering reagents to whole plants. First, in chapter 2, we focused our attention on geminiviruses—a large family of plant DNA viruses. Prior to these studies, geminiviruses were primarily used as vectors for virus-induced gene silencing or for protein expression; however, their circular DNA genomes, and their ability to replicate extrachromosomally, makes them an attractive vector for delivering genome engineering reagents (sequence-specific nucleases and donor molecules). Here, we were the first demonstrate their utility for editing plant DNA. Proof-ofconcept experiments in Nicotiana tabacum established that replicons based on the bean yellow dwarf virus can indeed deliver genome engineering reagents to leaf cells, and that these modified cells could grow into calli and seedlings. Interestingly, we also observed an enhancement in homologous recombination in leaf cells, relative to our non-viral controls. This enhancement was due to a replicating donor molecule and by pleiotropic activity of the virus replication proteins.
iii In addition to DNA viruses, we explored the use of RNA viruses for the delivery of sequence-specific nucleases. Tobacco rattle virus—a plant RNA virus—is an attractive vector system for delivering foreign protein because it can systemically infect plants, it causes very mild symptoms, and it is strictly RNA which circumvents government regulation regarding the delivery of foreign DNA to plants. Previous studies demonstrated its utility by delivering sequence-specific nucleases to tobacco and petunia. Here, we show its use in the major model organism Arabidopsis thaliana (chapter 3). We first characterize virus movement throughout Arabidopsis plants showing its trafficking to rosettes, cauline leaves and floral tissue. We then demonstrate transient delivery of zinc-finger nucleases to leaf tissue by detecting mutations in cells where the virus has moved. Finally, we assessed the ability for tobacco rattle virus to facilitate mutagenesis of germline cells by screening next-generation seedlings for zinc finger nuclease-induced modifications. From ~10,000 seeds (from 16 different infected plants), we observed 5 seedlings with the modification event, suggesting that TRV can enter and facilitate mutagenesis of seedprogenitor cells.
The remaining of the dissertation focuses on expanding the utility of stable integration into plant genomes by applying this approach to additional plants, additional target genes, and additional genome modifications. We demonstrate targeted knockout of Arabidopsis genes that have not been previously knocked out by traditional mutagenesis methods, including T-DNA insertion or ethyl methyl sulfonate (appendix A). At the time, this was the first study to use this method to knockout a biologically-interesting and endogenous gene. This method was then applied in Glycine max to knockout several genes involved in RNA interference (chapter 5), and it was used in Arabidopsis to generate large chromosomal deletions, inversions and duplications (chapter 4). Taken together, studies conducted in this dissertation improve upon the methods and technologies for delivering reagents to whole plants.
APPENDIX A: TARGETED KNOCKOUT OF THE ARABIDOPSIS THALIANA 162
GENES TSPO, JMJC AND TZP BY STABLE INTEGRATION OF ZINC-FINGERNUCLEASES
APPENDIX B: DNA REPLICONS FOR PLANT GENOME ENGINEERING: 171
DETERMINING GENE TARGETING FREQUENCIES IN TOBACCO
CHAPTER 4 Table 4-1 Frequency of ZFN-induced chromosomal deletions 100 Table 4-2 Zinc-finger arrays, recognition sites and recognition helices 101 CHAPTER 5 Table 5-1 The gene target accessions, target sequence, and RHs of 125 CoDA ZFNs that generated mutations in the target genes Table 5-2 Supplemental Table; The gene target accessions, target 126 sequence and recognition helices (RH) of CoDA zinc-finger nucleases that did not generate mutations in the target genes.
Table 5-3 Supplemental Table; Whole plant transformation summary 127 of the ZFN transgene targeting DCL4.
CHAPTER 5 Figure 5-1 Detection of ZFN-induced mutations at a GFP transgene in 128 soybean.
Figure 5-2 Detection of ZFN-induced mutations in soybean hairy-root 129 tissue.
Figure 5-3 Sequences of induced ZFN mutations in soybean hairy- 130 root tissue.
Figure 5-4 The mutagenic specificity of the RDR6a and RDR6b ZFN 131 transgenes was assessed by performing PCR enrichment assays with gene-specific primers for each homeolog.
Figure 5-5 Supplemental Figure; Analysis of the ZFN-induced dcl4a 132 mutation recovered from whole plant soybean.
Figure 5-6 ZFN mutagenesis and heritability in whole-plant soybean. 134 Figure 5-7 Supplemental Figure; ZFN-induced mutations recovered 135 from DCL4b T0 and T1 whole plant soybean.
Figure 5-8 Supplemental Figure; The Context-dependent assembly 136 (CoDA) method for engineering multi-finger arrays.
1. Baltes NJ, Gil-Humanes J, Cermak T, Atkins PA, Voytas DF (2014). DNA replicons for plant genome engineering. The Plant Cell 26: 151-63
2. Qi Y, Li X, Zhang Y, Starker CS, Baltes NJ, Zhang F, Sander JD, Reyon D, Joung KJ, Voytas DF (2013). Targeted deletion and inversion of tandemly arrayed genes in
Arabidopsis thaliana using zinc finger nucleases. G3 doi:10.1534/g3.113.006270. PMID:
3. Curtin S, Zhang F, Sander J, Haun W, Starker C, Baltes NJ, Reyon D, Dahlborg E, Goodwin M, Coffman A, Dobbs D, Joung K, Voytas DF, Stupar RM (2011). Targeted mutagenesis of duplicated genes in soybean with zinc finger nucleases. Plant Physiology 156: 466-473 PMID: 21464476
1.1 Significance Severe societal challenges will be faced in the upcoming decades, including food shortages due to an increasing population. By the year 2050, world population is expected increase from the current 7 billion people to over 9 billion people. Counteracting an escalating food demand will require a 50% increase in agricultural production by 2030 (Ronald, 2011).
Unfortunately, the amount of remaining arable land is limited, necessitating an increase in food production on currently-used land. Compounding these challenges are the predicted crop losses due to extreme temperatures, pest attacks, and pathogen outbreaks.
A powerful approach that may help overcome these challenges is to modify DNA sequences within plant chromosomes. For example, herbicide tolerance can be introduced by altering a few DNA bases within native plant genes (e.g., modification of the acetolactate synthase genes results in resistance to imidazolinone and sulphonylurea). Furthermore, plants can be engineered to have increased tolerance to environmental stresses (e.g., drought) and increased resistance to pathogens (e.g., viruses, fungi, bacteria, insects and nematodes). These, and additional crop improvements, suggest plant genome engineering will play a vital role in meeting the agricultural demands caused by an expanding population.
In addition to improving the genetics of crops, genome engineering can also be used to produce valuable plants or products for non-agricultural purposes. For example, there is great potential for plants to be used as bioreactors for pharmaceutical proteins. This process (referred to as molecular pharming) has many advantages over current production strategies, including the low cost of production, facile scalability, and reduced risk of contaminating human pathogens.
However, to realize the potential benefits from these applications, it is critical that we generate effective tools and approaches for editing plant DNA.
1.2 Harnessing Cellular DNA Double Strand Break Repair Pathways for Introducing Desired Sequence Changes One method to efficiently modify plant genomes involves introducing targeted DNA double-strand breaks at a locus of interest. Double-strand breaks are highly toxic lesions: a single break can arrest the cell cycle and, if left unrepaired, can lead to cell death. To preserve the integrity of their genomes, all living organisms have evolved pathways to repair genetic lesions. In general, cells have are two main repair pathways: non-homologous end joining and homologous recombination. Repair mechanisms of these two pathways can be exploited to introduce sequence changes within plant genomes.
1.2.1 Non-Homologous End Joining and Targeted Mutagenesis Non-homologous end joining is the major pathway plant cells use to repair double-strand breaks and it is active in all phases of the cell cycle (Rothkamm et al., 2003). Currently, there are believed to be two major forms of non-homologous end joining: canonical and alternative. Repair mechanics of the canonical pathway involve the direct rejoining of the two exposed DNA ends.
Shortly after a double-strand break is generated, a complex of Ku70 and Ku80 proteins are recruited to the break site to protect and stabilize the exposed DNA ends. The Ku70/80 heterodimer then recruits DNA-PKcs (DNA-dependent protein kinase catalytic subunit). After possible processing by the MRN complex (MRE11-RAD50-NBS1) complex, DNA gaps are filled by DNA polymerases µ and λ, and the free DNA ends are ligated together by a complex of LIG4 (DNA ligase IV) and XRCC4 proteins (West et al., 2000) (Figure 1.1). Because non-homologous end joining is non-template directed, repair can result in insertions and deletions of nucleotides at the break site. On the other hand, alternative non-homologous end joining is a Ku-independent pathway, and frequently uses microhomology for repair. Here, exposed DNA ends are bound by PARP1 which then recruits MRN, CTIP and BRCA1 for subsequent end processing. The processed DNA ends are then ligated together by either LIG3 and XRCC1 or LIG1.
The error-prone nature of the non-homologous end joining pathway can be exploited to introduce sequence changes within genomes. If double-strand breaks are directed to a DNA sequence of interest within the host’s genome, targeted mutagenesis can be facilitated (Lloyd et al., 2005). Furthermore, if breaks are directed to coding sequences within genes, targeted gene knockout can be facilitated (Zhang et al., 2010). For example, gene knockout can occur from frameshift mutations or deletions that remove nucleotides coding for essential amino acids.
Lastly, if two or more breaks are directed to sequences within a single chromosome, one can facilitate multiplex targeted mutagenesis, targeted deletion or targeted inversion of intervening sequence (Qi et al., 2013a). And, on the other hand, if two or more breaks are made on different chromosomes, one can facilitate translocations (Piganeau et al., 2013). Taken as a whole, exploiting the repair mechanics of non-homologous end joining provides genome engineers with an approach to knockout genes, or delete or rearrange sequences within living cells (Figure 1.2).