Engineered in vivo systems are emerging to characterize the lineage history and transcriptomic states of tissues in different organisms. However, existing CRISPR-based systems have typically required embryonic injection of editing reagents. A new study in Cell describes the CRISPR array repair lineage tracing (CARLIN) engineered mouse line that genomically encodes all the editing components, which can be activated in an inducible manner.

In their study, Bowling, Sritharan et al. engineered a two-part locus into mice: an expression cassette for ten guide RNAs (gRNAs) was followed by a GFP gene in which a barcode array targeted by the gRNAs was inserted upstream of its polyadenylation site. As for several existing CRISPR-based lineage approaches, the rationale is that cutting and repairing the target barcodes leaves genetic ‘scar’ edits at these sites that are inherited by cellular descendants, and these scars can be diversified further by subsequent editing in derivative subclones. The placement of the target barcode array was chosen so that poly(A)-based RNA sequencing (RNA-seq) of edited single cells could reveal both their transcriptional state (based on the overall transcriptome) and their lineage history (based on sharing of edits in the expressed lineage barcode array).

Credit: Ace_Create/DigitalVision/Getty

To complete the set of required genomically encoded components, the authors crossed these mice to a line encoding the M2rtTA reverse tetracycline transactivator and TetO-regulated Cas9. This combination of elements in the resultant CARLIN mice allows Cas9 expression (and hence editing activity) to be induced in the presence of doxycycline.

The team carried out various proof-of-concept studies in the CARLIN mice and embryonic stem cells, such as optimizing doxycycline scheduling, adjusting single-cell RNA-seq protocols to capture both the lineage barcode and wider transcriptome with sufficient sensitivity, and demonstrating that CARLIN could recapitulate known lineage relationships in vivo.

The team particularly focused on the barcode diversity generated after doxycycline exposure, estimating that CARLIN can generate ~44,000 distinct edited barcode arrays. They generated a database of CARLIN-edited barcode frequencies and incorporated it into their analysis pipeline to help to distinguish informative lineage barcodes that are shared as a result of descent from a common cellular ancestor, versus less-informative barcodes that commonly arise independently in different clones.

As an example biological application, the authors showed that after haematopoietic depletion by 5-fluorouracil, only a minor proportion of haematopoietic stem cells (HSCs) generated progeny that repopulate the haematopoietic system, and they identified transcriptional signatures that are associated with HSC repopulation potential.

address biological phenomena in diverse tissues

The CARLIN mouse and its accompanying bioinformatic pipeline represent an important step towards opening up in vivo lineage tracing systems for wider adoption by the community. Its germline, inducible nature should allow it to address biological phenomena in diverse tissues in both young and adult mice.