Background

At a time when increasing amounts of omics data are being generated on a daily basis, it is becoming even more important to functionally characterise genes in a context-dependent manner. In this regard, the application of genome-wide CRISPR-Cas9 screenings in different cell culture models emerge as powerful large-scale approaches to identify promising candidates whose disruption elicits a specific phenotype or physiological effect.

The rationale behind this technique starts by choosing the specific readout that cells will manifest after the screening (i.e drug resistance, cell reprogramming) and knocking out every gene that could be implicated in this readout. Importantly, only one gene per cell will be knocked-out, and after incubation, some cells will start to show an altered phenotype, providing a list of candidate genes that can be further studied.

Genome-wide CRISPR-Cas9 screen can be designed for a wide variety of applications. And here is where the fun part begins! In order to identify key regulators of developmental transitions, different reporter systems have been engineered as optimal readouts of fate transition [1,2,4]. This preprint by Nikhil Gupta and colleagues is an example of the great utility of these reporter systems in conjunction with whole genome screening, and provides an interesting list of novel regulators of early pluripotency states and germline genes.

Summary

Epigenetic mechanisms represent an important regulatory layer essential both for the maintenance of cellular identities and for the optimal transition between different developmental states. The main actors behind these regulatory processes are the commonly named epigenetic “writers” and “erasers”: different factors that establish or remove epigenetic marks and whose modifications will be interpreted by several “readers”, translating them into a transcriptional response. Thanks to massive screening techniques, the catalogue of epigenetic effectors is rapidly increasing, providing a progressively more complete picture of the complexity of epigenetic regulatory mechanisms.

In this preprint, Nikhil Gupta and colleagues establish a mESC-based reporter system responsive to epigenetic modifications. In a search for repressors, they take advantage of the tunability of the epigenetic state of mESCs depending on culture conditions. On the one hand, cells cultured in 2i+LIF are grown in the presence of two inhibitors of MEK/ERK and GSK3b signalling pathways plus leukemia inhibitory factor. As a consequence, they have lower DNA methylation levels  and show features of naïve pluripotency. On the other hand, cells in S+L are cultured in the presence of fetal bovine serum (undefined chemical composition) and leukemia inhibitory factor, and show a repressive and methylated chromatin landscape and display features of metastable pluripotency.

In this specific transition context, it becomes crucial to choose an optimal endogenous reporter gene. In the search for the perfect candidate, the authors considered three main criteria: epigenetic mechanisms of repression, biological function of the gene and its expression level. Taking all three into consideration, they chose Dazl  (a germ cell and 2C developmental stage marker) as  a reporter gene given its expression (2i=ON, S/L= OFF) and methylation pattern (2i=unmethylated, S/ L=methylated). Next, they engineered a cell line with a monoallelic knock-in of two selection markers (mScarlett and hygromycin resistance) driven by the Dazl promoter (DASH cell line). Interestingly, Dazl is informative as a reporter for genome wide CRISPR-Cas9 screens, as its repression depends on the synergistic action of ncPRC1.6 and DNA methylation, broadening the search for putative repressors. This DASH reporter cell line is the key tool of the study, and is subjected to a CRISPR-Cas9 genome-wide screening with the Brie library, a lentiviral library of vectors co-expressing sgRNAs and Cas9, targeting approximately 20000 genes [3].Following this experimental pipeline, the authors reach the goal of the study: the identification of repressive factors whose knock out triggers gene reactivation in serum condition (Figure 1).

Figure 1: Graphical representation of the reporter system used in the study. In 2i conditions, Dazl  is expressed while in S/L it is repressed trough DNA methylation and ncPRC1.6 complex. Knocking out factors essential for repression either at local or global scale triggers the reporter reactivation in serum condition. CpG DNA methylation status is represented by white (unmethylated) or black (methylated) lollipops. Figure modified from Gupta et al. 2021 (preprint).

The screening yielded a list of expected hits (DNMT1, UHRF1, MGA) involved in DNA methylation maintenance. These results served as an initial validation of the reporter system After narrowing down the list of candidates based on its statistical significance and biological function, the authors focused on six genes that were not known to regulate Dazl: Zbtb14, Kdm5c, Spop, Mcm3ap, Bend3 and Kmt2d. Next, they performed independent knock outs of these candidates in the DASH cell line in order to validate the results initially obtained in heterogeneous KO populations. Interestingly, the knock-outs of Kmt2d and Kdm5c resulted in the highest Dazl inductions. They confirmed that the observed effects were strictly due to the knock outs by rescue experiments, which reduced the mScarlet signal and sensitized cells to hygromycin. Then, they proceeded to further characterize the knock-outs by transcriptomics and methylation analysis. PCA analysis based on the transcriptome of the knock-outs grouped together Kdm5c, Mcm3ap, Spop and Zbtb14, while Kmt2d and Bend3 remained isolated, close to serum grown cells. Regarding methylation levels, only Spop KO showed global hypomethylation in comparison to parental cells.

In addition, knock out of Zbtb14, Kdm5c, Spop and Mcm3ap reactivates the poorly characterized 2-cell like cell (2CLC) state, an in-vitro model of totipotency, whose transcriptome highly overlaps with that of two-cell stage embryos and models the transition from maternal to zygotic expression program in early mouse development

Altogether, this work describes an endogenous DNA methylation-sensitive reporter system that can be useful in the search for transcriptional repressors in different cell state transition models and using different representative loci as reporters. Last but not least, the work by Defoussez´s Lab opens the door for further characterization of the novel epigenetic repressors discovered in this study.

Key results

• Development of DASH reporter cell line as a suitable reporter system for the study of epigenetic repressive pathways.

 

• Discovery of novel epigenetic repressors of germline genes, repeat elements and 2-cell like cell state, prioritizing ZBTB14, KDM5C, SPOP, MCM3AP, BEND3 and KMT2D.

 

Why I chose this preprint

As part of my daily twitter scroll early in May I found this preprint from Defossez´s Lab about genome-wide CRISPR KO screening. Turns out I had been recently hanging out with a friend who is applying it to cancer research and we had been discussing the technique´s potential as a massive search for oncogenes. Therefore, I decided to proceed to further reading, finding a very interesting experimental approach that not only helped me gain some theoretical background, but also gave me new ideas for my project. The perfect Prelight, isn´t it?

Epigenetic mechanisms represent a complex and context-specific set of regulatory processes whose coordination rules are still far from being fully understood. The regulatory and signalling cascades dictating cell identity and cell fate transitions still miss molecular effectors in order to get global pictures of the mechanisms involved. These new discoveries are of crucial relevance for the prevention, diagnosis and treatment of several developmental syndromes and diseases such as cancer.

This preprint adds to recent works searching for epigenetic regulators using endogenous reporters whose expression is dependent on effectors involved in the regulatory mechanisms explored [1,4]. These publications are providing the missing pieces of the epigenetics puzzle, and further study will provide more accurate predictions on the mechanisms ruling cell identity.

Questions for the authors

• The authors mention that one limitation of this reporter approach is the loss of information from the knock out of essential or near-essential genes. Another possible restraint of the study could be the presence of false negatives caused by mechanisms of redundancy between transcription factors and/or transcriptional compensation due to the absence of full gene knock outs. How could the authors overcome such limitations in further studies?

 

• What is the rationale behind using two selection markers (mScarlet and Hygromycin resistance) rather than using only mScarlett and filtering positive cells by FACS?

 

• One of the top hits is KMT2D (MLL4), which is considered as an activator mostly in the context of enhancers. How can KMT2D contribute to Dazl repression? Could the repression of Dazl be an indirect effect due to silencing of other genes in the absence of KMT2D? How can the authors distinguish between direct and indirect regulators of Dazl?

References

1. Dixon, et al. QSER1 protects DNA methylation valleys from de novo methylation. Science 372, (2021).
2. Alda-Catalinas, C. et al. A Single-Cell Transcriptomics CRISPR-Activation Screen Identifies Epigeneti- Regulators of the Zygotic Genome Activation Cell Systems 11, 25-41.e9 (2020).
3. Doench, J. G. et al. Optimized sgRNA design to maximize activity and minimize off-target effects of CRISPRCas9. Nature Biotechnology 34, 184–191 (2016).
4. Douillet, D., Sze, C., Ryan, C. et al. Uncoupling histone H3K4 trimethylation from developmental gene expression via an equilibrium of COMPASS, Polycomb and DNA methylation. Nature Genetics 52, 615–625 (2020).

 

 

Tags: crispr-cas9, epigenetic, knock out, pluripotency, repression, stem cells, transcription factor

Posted on: 15 June 2021

doi: https://doi.org/10.1242/prelights.29562

Read preprint (No Ratings Yet)