(466f) CRISPR/Cas9-Mediated Knock-in of an Optimized TetO Repeat for Live Cell Imaging of Endogenous Loci

Zhao, H., University of Illinois at Urbana-Champaign
Tasan, I., University of Illinois at Urbana-Champaign
Zhang, L., University of Illinois at Urbana-Champaign
Sustackova, G., University of Illinois at Urbana-Champaign
Sivaguru, M., University of Illinois at Urbana-Champaign
HamediRad, M., University of Illinois at Urbana-Champaign
Kim, J., University of Illinois at Urbana-Champaign
Wang, Y., Carnegie Mellon University
Genova, J., University of Illinois at Urbana-Champaign
Ma, J., Carnegie Mellon University
Belmont, A., University of Illinois at Urbana-Champaign
CRISPR-mediated gene targeting has been used for various applications including epitope or GFP tagging of endogenous proteins. One application that has not been explored yet is the tagging of the DNA itself for visualizing the position and dynamics of specific chromosomal regions. Nuclear organization has an important role in determining genome function; however, it is not clear how spatiotemporal organization of the genome relate to the functionality. Thus, in order to relate genome function to the 4D nuclear organization, a direct, microscopy-based method for live cell tracking of the dynamics of any specific endogenous locus of interest is necessary. A commonly used method for imaging a specific region of interest within the chromosome is DNA fluorescence in situ hybridization (DNA-FISH); however, the procedure is technically challenging and expensive, making it a low throughput method. Moreover, DNA-FISH is performed in fixed cells, which cannot be used for tracking dynamics of DNA. Live cell imaging of DNA in mammalian cells was previously done by using fluorescent repressor-operator system (FROS). The idea was based on enrichment of a fluorescent protein (FP) at a specific site on the DNA to obtain enough fluorescent signal that is above the background noise. For this purpose, fluorescent proteins such as GFP were fused to DNA binding proteins that can recognize a repeating DNA sequence. If the repetitive DNA is positioned close enough, it can serve as a tag for the region of interest. In the two commonly used systems, repeating sequences of lac operators (LacO) or tet operators (TetO) were used as a DNA tag and FP fused Lac repressors (LacI) or Tet repressors (TetR) were used for visualization of the tag. In previous examples, tagging was mostly done either within a plasmid or BAC containing the gene of interest, which might not fully represent the endogenous locus. An attempt for tagging an endogenous locus yielded very low efficiency. Discovery of novel modular proteins such as transcription activator-like effectors (TALEs) and the CRISPR/Cas9 system, whose DNA recognition specificity can be easily tailored, lead to the development of alternative approaches for visualizing endogenous loci. Coupled with fluorescent tags, both TALEs and a catalytically inactive version of Cas9 (dCas9) were used for visualization of naturally occurring repetitive sequences within mammalian genomes. However, repetitive sequences are not equally distributed throughout the genome, necessitating other strategies for visualizing such non-repetitive regions. Visualization of non-repetitive sites by TALEs or dCas9 requires introduction of a set of many TALEs or sgRNAs recognizing multiple different sites, which could be challenging and inefficient. Thus, an easy and efficient method for visualization of non-repetitive endogenous loci is highly desirable.

In this work we developed a new strategy for live cell imaging of specific chromosomal regions. Our approach was based on knock-in of a TetO repeat into endogenous loci via HR. HR is normally inefficient in mammalian cells. Thus, in our strategy we took advantage of the CRISPR/Cas9 system to increase the efficiency of knock-in of a TetO repeat via HR in HCT116 colorectal carcinoma cell line. One bottleneck in the traditional HR method is the use of long homology arms, which requires the arms to be PCR-amplified from the genomic DNA first, followed by their cloning into a donor vector. Construction of such a donor vector could be challenging, especially for higher throughput applications requiring creation of many such donor vectors. Based on a previous work using limited homology arms of down to 50 base pairs, we utilized a one-step, PCR-based method to create a linear donor DNA containing a TetO repeat. In this strategy, the TetO repeat and a selection marker are PCR-amplified using primers with 50 nt 5’ extension sequences, which are homologous to the target site sequence. The resulting linear donor DNA contains 50 bp homology arms flanking the insert, and the donor can then be gel or PCR-purified to be used directly for transfection.

Direct repeats of TetO could be unstable during plasmid preparation from bacteria, making the protocol tedious. Moreover, it is also difficult to PCR-amplify direct repeats. Thus, to improve the TetO repeats and add more beneficial features to it, we created a multimer of TetO (19 nt) repeats separated by random 10 nt sequences that do not contain CpG sites. Optimized repetitive sequences enabled easy PCR of both 48mer and 96mer TetO donors with a clear band. With the help of the improved donor DNA design and an optimized TetO repeat, it was possible to achieve knock-in efficiency of up to 20% using the 48mer TetO repeat. 48 repeats of TetO were also efficient at visualizing endogenous loci when TetR-EGFP was expressed. The observed number of spots was consistent with the genotyping results; thus, our suggested technique for visualization of endogenous sites was highly specific. Moreover, a comparison with wild-type cells showed that our method could resemble natural compartmentalization of the loci.

We also demonstrated applicability of our method in various different regions, representing both gene-rich, transcriptionally active and nuclear speckle-associated sites as well as gene-poor, transcriptionally inactive and nuclear lamina-associated sites. Knock-in efficiencies were 20% and 18% in two different gene-rich loci and the efficiency was 2% and 13% in two different gene-poor loci. As an interesting finding, we observed that gene-poor and heterochromatin regions exclusively preferred non-homologous end joining (NHEJ) for integration of the knock-in cassette, instead of HR. A direct ligation of the donor DNA after the expected cut site was observed in all the knock-in clones analyzed. Our results show that an NHEJ-based approach can be more successful at knock-in into heterochromatin sites. An encouraging finding was the fact that we could also visualize heterochromatin/gene-poor regions by expressing TetR-EGFP. Our method was also successful at live cell tracking of the dynamics of different loci.

The ease and the potential of scalability of the suggested method can help analysis of various loci in parallel to increase our understanding of relationships between gene compartmentalization and function. Given the general applicability of our method, we anticipate that it can be utilized to understand nuclear localization of various gene-rich and poor regions and enhance future causative analysis between gene function and compartmentalization.