Engineering

Design

The goal of our project was to create a system capable of efficiently binding DNA molecules via Cas proteins. As shown in a number of studies, the use of a multi-agent CRISPR approach made it possible to increase the efficiency of genetic therapy, including approaches based on epigenetic editing. To reduce the size of the construct and facilitate its delivery, we decided to target two regions within the HBV cccDNA molecule simultaneously. Increasing cooperativity between Cas complexes could significantly enhance the specificity and efficiency of editing.

Double complexes of Cas nucleases and dCas proteins were often used as parts of dimeric assemblies. Dimerization was typically achieved either through antibody-derived protein dimers or through guide RNA modifications. However, the overall size of genetic constructs was critical for cell-therapy applications. For this reason, the use of large dimerization protein domains or fused Cas proteins was not suitable for the purposes of our project.

Therefore, the most optimal solution to our problem was to employ modified guide RNAs that facilitated dimerization of dCas proteins via RNA–RNA interactions. To achieve this, we planned to use known hairpin RNA motifs capable of forming kissing-loop structures, which could stabilize paired RNA molecules and promote complex formation between two Cas proteins.

Figure 1. General design logic of the dual-CRISPR dCas system. Modified sgRNAs form kissing-loop interactions that bring together two dCas proteins bound to nearby cccDNA sites, promoting cooperative epigenetic repression.

At the first stage of our work, we selected the RNA dimerization domain described in Mutsuro-Aoki H., Acquisition of Dual Ribozyme-Functions in Nonfunctional Short Hairpin RNAs through Kissing-Loop Interactions, Life, 2022. Using this motif, we modified a standard sgRNA by adding the domain to its 3′ end via a short linker.

We tested the efficiency of this RNA modification and compared it with the standard sgRNA using several analytical methods — electrophoretic mobility shift assay (EMSA), biolayer interferometry (BLI), and cell-based assays evaluating the biological activity of the modified RNA in model systems. The data obtained indicated partial improvement in dimer formation, but the results were not sufficiently strong or reproducible to confirm that this modification reliably increased complex stability. For this reason, we proceeded to design and construct our own RNA dimerization domain to achieve higher stabilization of Cas protein dimers.

Build

The second stage of our work focused on creating an improved RNA dimerization domain to enhance the stability of Cas–Cas complexes and to evaluate variants with different dimerization parameters. To achieve this, we developed a genetic selection system that allowed direct comparison of dimerization domains within a controlled regulatory network.

The main idea of this stage was to enable selection of the most stable kissing-loop structures that could stabilize two Cas proteins through interactions between their guide RNAs. The use of chimeric guide RNAs containing two scaffolds was excluded at this stage, since it would substantially increase RNA size, reduce activity, and potentially elevate off-target effects relative to genomic DNA. To select the most effective RNA–RNA pairs, we employed a Ptet/TetR-based genetic selection system designed to identify stable interactions between guide RNAs in living cells.

Figure 2. Schematic of the Ptet/TetR selection circuit. Three plasmids operated together: (1) a reporter plasmid (pTet–reporter), (2) a TetR-expressing sensor plasmid under a constitutive promoter with Cas-binding sites, and (3) an expression plasmid encoding dCas and the gRNA library.

The selection system consisted of three plasmids:

1. Reporter plasmid — contained a fluorescent or antibiotic-resistance gene under the control of the pTet promoter.
2. Expression plasmid — encoded dCas proteins and a library of guide RNAs carrying various 3′ RNA motifs.
3. Sensor plasmid — included a promoter with multiple Cas-binding targets and expressed the TetR repressor protein.

The system functioned as follows. When two Cas complexes formed a stable pair through RNA–RNA interaction, a DNA loop was generated. This structural change suppressed the activity of the constitutive promoter controlling TetR expression, resulting in a decrease in TetR concentration. As TetR levels fell, repression of the pTet promoter was lifted, leading to increased expression of the reporter gene. Such a configuration allowed rapid selection of guide-RNA sequences that provided the strongest stabilization of ribonucleoprotein (RNP) complexes.

In addition to known kissing-loop variants described in the literature, we performed one round of randomization at the 3′ end of the guide RNA. This approach allowed identification of novel sequences capable of effective RNA–RNA dimerization.

The resulting guide RNAs were subsequently tested in vitro using minicircle-derived recombined cccDNA (mcccDNA) — an experimental model that closely reproduced the structural and functional properties of native HBV cccDNA.

Test

We evaluated the functionality of the constructs both in vitro and in vivo.
– In vitro: The selected gRNA constructs were introduced into bacterial systems containing the three-plasmid circuit. Fluorescence intensity and antibiotic-resistance profiles were used as indicators of successful RNA–RNA dimerization and Cas complex stabilization.
– In vivo: Validation experiments were conducted in HepG2 cell models with integrated cccDNA to assess whether RNA-mediated dCas cooperation enhances transcriptional repression without double-strand breaks.

Additionally, several previously characterized RNA dimerization motifs were employed as positive controls to benchmark the efficiency of our newly designed sequences.

Learn

Analysis of the obtained data showed that standard RNA modifications resulted in only partial stabilization of Cas complexes, confirming the necessity of designing a custom dimerization RNA motif.

This first iteration of the Design–Build–Test–Learn cycle validated the feasibility of our concept and yielded quantitative metrics for the next optimization round. Based on the feedback from this iteration, we refined both the guide-RNA architecture and the sensitivity of the Ptet/TetR selection system, improving its dynamic range and reducing background activation.

Through this iterative process, we achieved our first engineering success — a complete cycle from conceptual design to experimental validation and analysis, demonstrating the applicability of engineering principles to RNA-mediated CRISPR complex formation.

Furthermore, within the same framework, we developed a new inducible chloramphenicol-resistance plasmid regulated by the tetracycline promoter. This construct served as a reversible selection module and could be adapted for clone selection or gene-repression studies in other synthetic-biology applications. Learn more in Contribution.