VCOM Carolinas Research Day 2023

Biomedical Studies

In vitro genome-wide identification and mapping of single-stranded DNA sequences that bind to the metastasis-associated protein PCBP1

Krishna Patel, OMS I, Chirag Lodha, OMS II, Bidyut Mohanty, PhD VCOM Carolinas, Cell Biology and Physiology, Spartanburg, SC

Abstract # BIOM-10

Abstract

Results

Discussion, Conclusions, and Future Directions

Polycytosine binding protein PCBP1 suppresses cancer metastasis by inhibiting expression of several metastasis-associated proteins and it does so by binding site- and structure (hairpin)- specifically to 3’ untranslated regions (3’ UTRs) of their respective mRNAs. Recently, PCBP1 has been shown to bind polycytosine-rich single-stranded DNA (ssDNA) sequences present at promoters of several oncogenes and at telomeres. These ssDNA sequences are known to form a secondary structure of DNA called the intercalating-motif (i-motif) in which C-C+ and C-C interaction occurs. However, the scope of PCBP1 binding to these sequences and their i-motifs across the genome, along with the mechanisms of action of PCBP1 at these sites is largely unknown. Although in vivo studies have identified a few sites, these studies do not reveal all the potential binding sites in the genome or their structures. A robust in vitro study is needed to identify all the potential PCBP1-binding sites and their i-motifs. To identify these PCBP1 binding sites genome-wide, we have developed a new technique by which we generate a library of single-stranded DNA from genomic DNA in vitro and purify ssDNA molecules that bind to PCBP1 protein. In this technique, total genomic DNA prepared from a mouse cell line (NMuMG) is sonicated to generate fragments of ~200 nucleotide pairs, which are then treated with E. coli exonuclease III or T7 exonuclease to generate ~100 base-long ssDNA fragments. The DNA fragments are then incubated with hnRNP E1 protein immobilized on glutathione-agarose beads. The interacting protein-ssDNA complexes on agarose beads are washed multiple times and then, the protein-bound ssDNA fragments are purified for DNA sequencing.

References In order to determine the potential binding sites for hnRNP E1 genome wide, a new technique to obtain single-stranded DNA was used. Please refer to Chirag Lodha’s presentation (at the current Research day) for an in-depth exploration of this technique. First, the genomic dsDNA was broken down into about 200 bp using a sonicator, with the sonication pattern shown in Figure 2. This was successful after 21 cycles of sonication of 30 seconds each. Next, the DNA was chewed using E. coli Exo III Exonuclease and T7 Exonuclease in order to yield ssDNA (see Figure 1) with a length of about 100 bases. Figure 3 shows the DNAs before and after treatment with the exonucleases. Although the fractionation pattern of ssDNA does not show much change from dsDNA, which is expected, the intensity of the ssDNA is less than that of dsDNA suggesting reduction of the amount of DNA. New experiments are underway to confirm ssDNA generation and its difference from dsDNA. A glutathione S-transferase-tagged E1 (GST-E1) fusion protein was then immobilized on glutathione-agarose beads. The ssDNA pool was mixed with the immobilized GST-E1 protein and incubated at room temperature to generate protein-ssDNA complexes. The beads were washed with buffer of hnRNP E1 (25 mM tris-HCl, pH 7.5, 50 mM NaCl, 1 mM EDTA, 10 mM β mercaptoethanol and 2% glycerol), and the hnRNP E1 bound DNA molecules were eluted from the agarose beads by treatment with Proteinase K (that degrades hnRNP E1 protein). Experiments are underway to determine which sequences were bound by hnRNP E1 by two methods: (1) PCR amplification of a region around KRAS oncogene promoter that contains G-4/i-motif forming poly-G/poly-C-rich sequences. (2) Genome wide sequencing of hnRNP E1-bound DNA sequences will be conducted commercially by outsourcing the ssDNA pool. In addition, a similar assay will be carried out for the binding of i-motif-forming ssDNA sequences in the ssDNA pools to E1. A comparison between the two pools of PCBP1-bound DNA sequences will reveal a part of the mechanism of PCBP1-binding to ssDNA and its i-motif and will lead the way to future structural and mechanistic studies. This is especially important for studies on DNA damage and cancer, as an absence of PCBP1 causes DNA damage and is known to cause cancer metastasis. This study can lead to additional studies such as analyzing i motifs in Burkitt lymphoma and metastatic cancer cells that have defects in PCBP1 (hnRNP E1), and other i-motif binding proteins. After identification of the sequences, i-motifs in cancer cells can be stabilized using chemical stabilizers and can be targeted with DNA damaging drugs. 1. Mohanty, B. K. et al (2021) Heterogeneous nuclear ribonucleoprotein E1 binds polycytosine DNA and monitors genome integrity. Life Sci Alliance. 4(9):e202000995. doi: 10.26508/lsa.202000995. 2. Peña et al., (2022) Human genomic DNA is widely interspersed with i-motif structures. doi: https://doi.org/10.1101/2022.04.14.488274. I thank Dr. Bidyut Mohanty for his mentorship, knowledge, and support, and Chirag Lodha for his contributions to this work. Acknowledgements

Figure 2 . 1.5% agarose gel showing sonication pattern of NMuMG (mouse cells) chromosomal DNA.

Figure 3 . 1.5 % agarose gel showing Exo III Exonuclease- and T7 Exonuclease- treatment of sonicated NMuMG mouse chromosomal DNA.

Methods

Figure 1. Protocol to Isolate E1- bound ssDNA

Figure 4 . Technique showing purification of ssDNA bound to PCBP1 protein.

14

2 0 2 3 R e s e a r c h R e c o g n i t i o n D a y