A genome-wide MAGIC kit for recombinase-independent mosaic analysis in Drosophila

Reviewer #1 (Public review):

Summary:

In this manuscript, Shen et al. have improved upon the mitotic clone analysis tool MAGIC that their lab previously developed. MAGIC uses CRISPR/Cas9-mediated double-stranded breaks to induce mitotic recombination. The authors have replaced the sgRNA scaffold with a more effective scaffold to increase clone frequency. They also introduced modifications to positive and negative clonal markers to improve signal-to-noise and mark the cytoplasm of the cells instead of the nuclei. The changes result in increase in clonal frequencies and marker brightness. The authors also generated the MAGIC transgenics to target all chromosome arms and tested the clone induction efficacy.

Strengths:

MAGIC is a mitotic clone generation tool that works without prior recombination to special chromosomes (e.g., FRT). It can also generate mutant clones for genes for which the existing FRT lines could not be used (e.g., the genes that are between the FRT transgene and the centromere).

This manuscript does a thorough job in describing the method and provides compelling data that support improvement over the existing method.

Weaknesses:

It would be beneficial to have a greater variety of clonal markers for nMAGIC. Currently, the only marker is BFP, which may clash with other genetic tools (e.g., some FRET probes) depending on the application. It would be nice to have far-red clonal markers.

Reviewer #2 (Public review):

Summary:

In this study, the authors present the latest improvement of their previously published methods, pMAGIC and nMAGIC, which can be used to engineer mosaic gene expression in wild-type animals and in a tissue-specific manner. They address the main limitation of MAGIC, the lack of gRNA-marker transgenes, which has hampered the broader adoption of MAGIC in the fly community. To do so, they create an entire toolkit of gRNA markers for every Drosophila chromosome and test them across a range of different tissues and in the context of making Drosophila species hybrid mosaic animals. The study provides a significant and broadly useful improvement compared to earlier versions, as it broadens the use-cases for transgenic manipulation with MAGIC to virtually any subfield of Drosophila cell biology.

Strengths:

Major improvements to MAGIC were made in terms of clone induction efficiency and usability across the Drosophila model system, including wild-type genotypes and the use in non-melanogaster species.

Notably, mosaic mutants can now be created for genes residing on the 4th chromosome, which is exciting and possibly long-awaited by 4th chromosome gene enthusiasts.

Selection of the standard set of gRNA markers was done thoughtfully, using non-repetitive conserved and unique sequences.

The authors demonstrate that MAGIC can be used easily in the context of interspecific hybrids. I believe this is a great advancement for the Drosophila community, especially for evolutionary biologists, because this may allow for easy access to mechanistic, tissue-specific insight into the process of a range of hybrid incompatibilities, an important speciation process that is normally difficult to study at the level of molecular and cell biology.

In the same way, because it is not limited to usage in any particular genetic background, genome-wide MAGIC can be potentially used in wild-type genotypes relatively easily. This is exciting, especially because natural genetic diversity is rarely investigated more mechanistically and at the scale/resolution of cells or specific tissues. Now, one can ask how a particular naturally occurring allele influences cell physiology compared to another (control) while keeping the global physiological context of the particular genetic background largely intact.

Weaknesses:

It is not entirely clear how functionally non-critical regions were evaluated, besides that they are selected based on conservation of sequence between species. It may be useful to directly test the difference in viability or other functionally relevant phenotype for flies carrying different markers. Similarly, the frequency of off-targets could be investigated or documented in a bit more detail, especially if one of the major use-cases is meant for naturally derived, diverse genetic backgrounds. It is, at the moment, unclear how consistently the clones are induced for each new gRNA marker across different WT genetic backgrounds, for example, a set of DGRP genotypes, which could be highly useful information for future users.

Reviewer #3 (Public review):

Summary:

In the manuscript by Shen, Yeung, and colleagues, the authors generate an improved and expanded Mosaic analysis by gRNA-induced crossing-over (MAGIC) toolkit for use in making mosaic clones in Drosophila. This is a clever method by which mitotic clones can be induced in dividing cells by using CRISPR/Cas9 to generate double-strand breaks at specific locations that induce crossing over at those locations. This is conceptually similar to previous mosaic methods in flies that utilized FRT sites that had been inserted near centromeres along with heat-shock inducible FLPase. The advantage of the MAGIC system is that it can be used along with chromosomes lacking FRT sites already introduced, such as those found in many deficiency collections or in EMS mutant lines. It may also be simpler to implement than FRT-based mosaic systems. There are two flavors of the MAGIC system: nMAGIC and pMAGIC. In nMAGIC, the main constituents are a transgene insertion that contains gRNAs that target DNA near the centromere, along with a fluorescent marker. In pMAGIC, the main constituents are a transgenic insertion that contains gRNAs that target DNA near the centromere, along with ubiquitous expression of GAL80. As such, nMAGIC can be used to generate clones that are not labelled, whereas pMAGIC (along with a GAL4 line and UAS-marker) can be used much like MARCM to positively label a clone of cells. This manuscript introduces MAGIC transgenic reagents that allow all 4 chromosomes to be targeted. They demonstrate its use in a variety of tissues, including with mutants not compatible with current FLP/FRT methods, and also show it works well in tissues that prove challenging for FLP/FRT mosaic analyses (such as motor neurons). They further demonstrate that it can be used to generate mosaic clones in non-melanogaster hybrid tissues. Overall, this work represents a valuable improvement to the MAGIC method that should promote even more widespread adoption of this powerful genetic technique.

Strengths:

(1) Improves the design of the gRNA-marker by updating the gRNA backbone and also the markers used. GAL80 now includes a DE region that reduces the perdurance of the protein and thus better labeling of pMAGIC clones. The data presented to demonstrate these improvements is rigorous and of high quality.

(2) Introduces a toolkit that now covers all chromosome arms in Drosophila. In addition, the efficiency of 3 target different sites is characterized for each chromosome arm (e.g., 3 different gRNA-Marker combinations), which demonstrate differences in efficiency. This could be useful to titrate how many clones an experimenter might want (e.g., lower efficiency combinations might prove advantageous).

(3) The manuscript is well written and easy to follow. The authors achieved their aims of creating and demonstrating MAGIC reagents suitable for mosaic analysis of any Drosophila chromosome arm.

(4) The MAGIC method is a valuable addition to the Drosophila genetics toolkit, and the new reagents described in this manuscript should allow it to become more widely adopted.

Weaknesses:

(1) The MAGIC method might not be well known to most readers, and the manuscript could have benefited from schematics introducing the technique.

(2) Traditional mosaic analyses using the FLP/FRT system have strongly utilized heat-shock FLPase for inducible temporal control over mitotic clones, as well as a way to titrate how many clones are induced (e.g., shorter heat shocks will induce fewer clones). This has proven highly valuable, especially for developmental studies. A heat-shock Cas9 is available, and it would have been beneficial to determine the efficiency of inducing MAGIC clones using this Cas9 source.

Author response:

Reviewing Editor Comments:

The following are some consolidated review remarks after discussions amongst all three reviewers:

The reviewers feel the evidence level could be raised from 'convincing' to 'compelling' if the following key (and partially shared) suggestions by the reviewers are followed adequately:

(1) Expand labeling options for nMAGIC, which is currently just a BFP marker. This would increase the utility of the method. A far-red marker would be very helpful. Could the authors just do this for one chromosome arm and make the reagent available for others to generate other chromosome arms?

This is a great suggestion. We will make an nMAGIC vector containing a far-red fluorescent marker and generate a 40D2 version of this nMAGIC gRNA-maker to demonstrate its utility. This vector will be available for others to make additional nMAGIC gRNA-markers.

(2) Verify that destabilized GAL80 is potent enough to suppress GAL4. Repeat Figure 1C-E with tub-GAL80-DE-SV40.

We will use a tub-GAL80-DE-SV40 gRNA-marker to test suppression of pxn-Gal4.

(3) Concern about the health of the induced mitotic clones. This is an important consideration, but the reviewers were not sure what the necessary experiments would be. To gauge twin-spot clone sizes? Please address.

We will assess the health of induced mitotic clones in wing imaginal discs. We will do this by generating twin spots with a nMAGIC gRNA-marker in wing discs and compare the sizes of the two cell populations (BFP^+/+ and BFP^-/-) in twin spots.

(4) Include a schematic of the MAGIC method as Figure 1 or add it to Figure 1. Many may not be familiar with the method, so to promote its adoption, the authors should clearly introduce the MAGIC method in this paper (and not rely on readers to go to previous publications). For this paper to become a MAGIC reference paper, it should be self-contained.

We will add a diagram of the MAGIC method in the revised manuscript.

(5) Determine the utility of using a hs-Cas9 line for temporal induction of MAGIC clones. This is a traditional method for mitotic clone induction (with hsFLP/FRTs), and its use with the MAGIC system (especially pMAGIC) could also make it more attractive, especially to label small populations of neurons born at known times. To this point, the authors could generate pMAGIC clones using hs-Cas9 for commonly used adult target neurons, such as projection neurons, central complex neurons, or mushroom body neurons. The method to label small numbers of these adult neurons is well worked out with known GAL4 lines, and demonstrating that pMAGIC could have similar results would capture the attention of many not familiar with the pMAGIC method.

We thank the reviewers for this suggestion. We will test hs-Cas9 in inducing pMAGIC clones in one of the neuronal populations in the adult brain, as suggested by the reviewers.

In addition, we will address all other minor concerns of the reviewers.