Sizing, stabilising, and cloning repeat-expansions for gene targeting constructs

Abstract

Aberrant microsatellite repeat-expansions at specific loci within the human genome cause several distinct, heritable, and predominantly neurological, disorders. Creating models for these diseases poses a challenge, due to the instability of such repeats in bacterial vectors, especially with large repeat expansions. Designing constructs for more precise genome engineering projects, such as engineering knock-in mice, proves a greater challenge still, since these unstable repeats require numerous cloning steps in order to introduce homology arms or selection cassettes. Here, we report our efforts to clone a large hexanucleotide repeat in the C9orf72 gene, originating from within a BAC construct, derived from a C9orf72-ALS patient. We provide detailed methods for efficient repeat sizing and growth conditions in bacteria to facilitate repeat retention during growth and sub-culturing. We report that sub-cloning into a linear vector dramatically improves stability, but is dependent on the relative orientation of DNA replication through the repeat, consistent with previous studies. We envisage the findings presented here provide a relatively straightforward route to maintaining large-range microsatellite repeat-expansions, for efficient cloning into vectors.

Graphical abstract

Fig. 1

CRISPR/Cas9 digestion of C9orf72-ALS patient BAC. (A) Schematic of the C9orf72-ALS patient BAC vector, harbouring the C9orf72 gene (light blue; exons numbered), a pathogenic GGGGCC repeat expansion in intron 1 (red), and large flanking genomic regions (dark blue; sizes indicated in light blue). sgRNA target sites A, B, and C are shown together with the expected size of the released fragments in red (and a fully retracted band size in brackets), following digestion with sgRNA A + B or A + C, assuming a (GGGGCC)₈₀₀ repeat. (B) Agarose gel electrophoresis images of CRISPR-Cas9 digested C9orf72-BAC DNA sub-clones. Repeat bands are marked with red asterisks, large retractions are marked with black asterisks. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 2

CRISPR-Cas9 cloning of C9orf72 repeat into pJazz. (A) Schematic of the pJazz cloning vector showing chloramphenicol resistance cassette (CAM^R) on the short arm, lacZ stuffer cassette (purple), and origin of replication (ori) on the long arm. sgRNA guides A + C were used to release the C9orf72 repeat (red) containing region, which was gel purified and blunt-cloned into pJazz (replacing the lacZ casette). The repeat containing fragment is illustrated in two possible orientations within pJazz (exons 1a and 1b added for reference), with expected BamHI and XbaI restriction fragment sizes shown below, assuming a 728x repeat. (B) Agarose gel electrophoresis gel images showing digestion of miniprepped clones with BamHI and XbaI, revealing 4 clones carrying a repeat in the forward orientation (#1, #3, #6, #7); asterisks indicate repeat containing band. The table displays the estimated repeat sizes for clones #1, #3, #6, #7; it is assumed clone #6 is unretracted^§. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

Sanger sequencing through 5′ and 3′ repeat junctions from pJazz-C9orf72-repeat clone #6. Sequencing chromatograms via sequencing primers F1 (top) and R1 (bottom) are shown, revealing the presence of human intron 1 (blue text) correctly juxtaposed to hexanucleotide GGGGCC repeats (red text). Green asterisks represent 2 hexanucleotide sites that read as GGGCC (missing G) in PacBio sequencing. Purple text highlights the insertion/deletion event (−GTGGTC + CGGGCCCG) downstream of the repeat. Arrowheads indicate erroneous lower peaks; independent primer sets (Supplementary Data 2) show these to be in random positions, indicating sequencing noise. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Stability assessments of the C9orf72 repeat inside pJazz. (A) 500 μl clone #6 was split into 8x 5 ml subcultures, denoted 6.1–6.8, and purified DNA was digested with XbaI. Clones #3 (~350 repeats), 7 (~500 repeats), 1 (~500 repeats), and 6 (728 repeats, assuming unretracted), plus pJazz vector were run as controls in size order (left). Red asterisks indicate repeat bands, with approximate repeat lengths indicated in red text. (B) 15 sub-clones from each original clone carrying repeats were subjected to growth at different temperatures. Sub-clones 3c, 6j, 7g, 7j displayed evidence of retraction (blue asterisk). DNA from the original clones were included as controls, indicated by C, with control repeat bands indicated by red asterisks. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5

Summary workflow to screen and subclone C9orf72 repeat region. (A) Frozen glycerol stock of repeat-BAC harbouring DH10b E. coli bacteria is streaked out onto LB-Agar plates. (B) Colonies are picked, and (C) grown in liquid culture for (D) BAC DNA extraction and purification. (E) BAC DNA is CRISPR/Cas9 digested to release the repeat region and screen for subclones that harbour large repeats. (F) Large repeat containing bands can be excised from the gel, purified and (G) blunt cloned into pJazz vector for stabilisation and further cloning steps if required.

Fig. 6

Schematic to demonstrate the potential directional impact of DNA replication on repeat stability. Lagging strand DNA synthesis is prone to slippage events when repetitive DNA is present. On the left, DNA replication is shown to run through the G-rich GGGGCC repeat strand (red) in a 3′ to 5′ direction, resulting in G-quadruplexes forming on the lagging synthesis strand, which can lead to expansion events. On the right, DNA replication is shown to run through the GGGGCC sequence in a 5′ to 3′ direction resulting in G-quadruplexes forming on the lagging template strand, which can lead to expansion events. Blue text represents the antisense CCCCGG repeat strand. Origin of replication is denoted by ori. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)