CRISPRCINIMM · Revealing genes that mediate the CIN-induced interferon-inflammatory response using a Genome-wide CRISPR screen coupled to a Suicide Gene Switch.

Revealing genes that mediate the CIN-induced interferon-inflammatory response using a Genome-wide CRISPR screen coupled to a Suicide Gene Switch.

Chromosomal instability (CIN) is defined as the process by which cells experience an increased frequency of chromosome mis-segregation during mitosis over several successful cell divisions. Leading to abnormalities as gain, loss and/or translocation of (parts of) chromosomes. A state known as aneuploidy. CIN and the resulting aneuploidy are found in 60-80% of all human cancers, and is therefore a cancer hallmark. Cancers are well known to be heterogeneous, both intertumoral (between tumours) and intratumoral (within one tumour). The latter is known to be driven by CIN, and the resulting heterogenous tumour cell populations seem to provide the ability for cells to undergo selective evolution. Tumours are also found with low CIN and little heterogeneity but still highly aneuploid, potentially reducing their evolutionary capacity. CIN and aneuploidy in human cancers are tolerated, as both are highly prevalent and associated with tumour evolution, metastasis, multi-drug resistance and therefore poor prognosis, pointing towards a supportive role of CIN in tumorigenesis. Untransformed cells (non-cancerous) are found to be negatively affected by CIN and aneuploidy, where their presence induces multiple stress signalling pathways including proteotoxicity, metabolic stress and inflammatory responses. Therefore negatively impairing cell fitness. The differences between the responses to aneuploidy by transformed versus untransformed cells is referred to as the aneuploidy paradox. What suggests that tumour cells have developed mechanisms cope with stresses induced by CIN and the resulting aneuploidy. As earlier noted, mis segregation events during mitosis can lead to extranuclear packaging of chromosome (fragments), forming micronuclei found in the cytoplasm. The membrane of the micronucleus is easily damaged, and therefore prone to rupture, exposing genomic double-stranded DNA (dsDNA) to the cytoplasm. The presence of cytosolic DNA is detected by nucleic acid receptors like cyclic GMP-AMP synthase (cGAS) and retinoid acid inducible gene I (RIG-I), both pattern recognition receptors (PRRs). PRRs recognize pathogen- and danger-associated patterns (PAMPs) (DAMPs), and are responsible for the initiation of innate immunity against microbial pathogens and therefore, inflammation. Besides microbial DNA, it is now also known that other sources of cytoplasmatic DNA, like dead tumour cells, free telomeric DNA and cytoplasmic chromatin fragments are recognized by cGAS. When dsDNA is bound to cGAS it results in the synthesis of cGAMP, a stimulator of interferon genes (STING) leading to type I interferon (IFN) activation and secretion of chemokines like CCL5 and CXCL10, and interleukins as (IL)-6 and -8. Besides the resulting inflammation, cGAS-STING is also associated with cell senescence, cell death by apoptosis and necroptosis. And in cancers, cGAS-STING contributes in metastasis and is believed to mediate escaping immune surveillance. The role of cGAS-STING is important in not only cancer, but also autoimmune disorders and other diseases. Targeting cGAS-STING can help reduce inflammation and further development of autoimmune diseases, and can therefore lead to new therapeutic strategies. However, recent publications show that inducing CIN by genotoxic stress through micronuclei-inducing agents failed to activate cGAS in the HeLa cancer cell line. This goes against the claims that CIN-related micronuclei are a source of cGAS-STING-mediated inflammatory response. We aimed to help further with the investigation of the occurrence of CIN and the resulting phenotypes, and how this relates to inflammatory activation. Similar to microbially induced inflammation, sterile inflammation, following ligand recognition, can start the downstream signaling pathways, such as interferons. Interferons comprise a family of molecules divided into three main subfamilies: IFN-I (a and b), IFN-II (y), and IFN-III, and are the typical inflammatory mediators and have essential roles in potentiating acquired antitumor immune activities, and are transcription regulators. However, chronic inflammation that occurs when the offending agent is not removed or contained can be detrimental. The cytoplasmic DNA sensing pathway, subsequently triggers downstream JAK-STAT1/2-interferon signalling, later binding DNA elements termed interferon-sensitive response element (ISREs) (with the consensus sequence TTTCNNTTTC) for IFNI and a distinct DNA element termed a gamma-activated site (GAS; consensus sequence TTCNNNGGA) for IFNII. Patterns of genes induced by type I and II IFNs overlap, partly because target genes can contain both ISRE and GAS elements, and overlap may be secondary to the induction of transcription factors with shared target genes. This cascade of transcription factors, particularly IRF family members, which can interact with STATs and redirect their binding activity, can mediate the evolution of IFN signatures over time. Type I and II IFNs also activate noncanonical transcriptional complexes and additional STATs, and induce the expression of unphosphorylated STATs, thus contributing to the IFN signature. The nature of the IFN response is context dependent, because IFN-induced gene expression is modulated by distinct environmental stimuli via signal-transduction cross-talk. Therefore studying the role of inflammation activation, the associated signalling pathways, and how CIN cancers develop immune-evasive mechanisms might provide new insights that can help with new better-targeted strategies to treat CIN cancers. An improved understanding of the aneuploidy-coping mechanism could provide a powerful strategy to eradicate cancer cells exhibiting CIN and have a fundamentally defective IFN signalling system, could explain their inability to recruit immune cells into the tumor microenvironment and contribute to disease progression. With this project, we aimed to develop and use inflammation reporter cell lines to investigate inflammatory signalling pathways due to CIN in cancerous and non-transformed cells. The use of reporter cell lines allows visualization of the expression of a gene of interest by fluorescent protein (FP) expression. Because of the role of activation of IFN, CXCL10, and IL6 in both cGAS-STING signalling and the association with evasion and activation of immune clearance in cancers, we chose to use the expression of these cytokines as the reporter by tagging expression with a fluorescent protein, together with a suicide gene switch (HSV-TK or iCasp9). This inducible suicide gene can selectively destroy cells with an intact inflammatory response in the presence of activating drugs, and therefore, enhance the population that will not have an intact inflammatory response. An Unbiased genome-wide CRISPR screen, coupled to this reporter cell lines can help identifying genes that when deleted, can abolish or attenuate the inflammatory response induced by CIN. IL6/CXCL10, alongside other genes (HERC5, IFIT1, IFIT3, MX1, and OAS3) CRISPR mediated knock-in of the reporter sequence was made at the C-terminus of the genes, creating an Gene-reporter fusion protein. We selected the breast cancer cell line BT549, and the non-transformed human fibroblast cell line BJ. Despite successful integration in BT549, IL6/CXCL10-mWASABI expression levels remained similar in treated and untreated cells, when analyzing samples by flow cytometry (FC) complicating the analysis. Moreover, BT549-IL6 cells exhibited decreased viability over time, suggesting that the IL6-mWASABI fusion protein might impair cell function. However, since the BT549 cell line is a cancer cell line, and supporting literature has shown that tumor cells can evade immune surveillance, no IFN reporter activity can be explained by the believed impaired IFN signalling of cancer cells. Also when treating the BT549-IL6 cell line with IFNγ, higher growth arrest and cell death happened after 24 hours, compared to control cells. It can, therefore, be discussed that the loss of IL6 function allows IFNy to induce cancer cell death. On the contrary, BJ cells attempts to be targeted by CRIPSR failed. The reporter had to be changed, and an ISRE-type reporter construct was chosen and added to the BT549 and BJ cell lines by lentivirus. In this reporter, IFNI expression is translated to GFP expression through the control of a 4xISRE repeat and an ISG54 promoter. IFNa and Lipopolysaccharide (LPS, from the membrane of gram-negative bacteria) treatment induced GFP expression in BJ-ISRE cells, validating the reporter’s functionality. However, BT549-ISRE cells showed minimal GFP expression, indicating further defects in their IFN pathway activation. The lack of IFN reporter activation in BT549 suggests that the cells can ignore IFN signalling. Ideally, IFN stimulation should increase the secretion of inflammatory cytokines, along with growth arrest and apoptosis through IFN-STAT1 signalling. To verify that this construct worked in a normal breast cell line, MCF10a cell line was used and the construct was introduced, and in this cell line all controls used activated the reporter. Interstingly, IFNy was also able to activate the reporter in the 2 untransformed cell lines, which is in contrast with the current notion that ISRE recognition sequences can only be activated through IFNI. Next, after verifying activation of the reporter, to observe activation in response to CIN, reversine, an MPS1 inhibitor, was used. Reversine treatment has been shown to induce chromosome missegregation leading to, for example, the formation of micronuclei through lagging chromosomes in anaphase. BJ cells were followed by live cell imaging and showed varying GFP expression in response to the treatments peaking between 2-4 days, which was determined by a 7-day imaging trial. Reversine treatment presented a unique pattern of segregation errors, that could be explained by the working mechanism of reversine, which targets MPS1, leading to various types of chromosomal missegregations. BJ-ISRE cells presented varying GFP expression in response to the CIN-inducing treatment, suggesting that the type and severity of CIN can differently influence IFN expression. It was also noted that the GFP expression did not occur on cells that had noticeable missegregations, chromatin bridges or micronuclei. We were able to quantify the amount of mitotic errors after reversine treatment, and with the concentration used for the majority of assays, more than 90% of cells had mitotic errors after 24h. However, GFP signalling only started to peak after 72h in the treatment. While following the live cell imaging images, we noticed that after activation of the reporter in one cell, around 2-4 hours after, cells in the vicinity were also activated, with the number and intensity of cells activating the reporter increasing over time. However, after 5-6 days, the highest percentage of cells activating the reporter stopped at around 50-60%. This finding indicates that paracrine signalling is important and activates the IFN cascade in bystander and adjacent cells that did not recently divide. This finding prompted us to do an RNAseq of the cells that did not activate the reporter and the ones that activated the reporter after 6 days of reversine treatment. While untreated cells vs treated cells show a difference between the samples, with known inflammation, senescent, and DNA damage genes being upregulated or downregulated, treated samples that were GFP positive and GFP negative samples, did not show an important difference. This indicates that the activation of the signalling in bystander cells can mask the difference in genes that will start the IFN cascade or the ones that will stop it. Therefore, a CRISPR screen with reversine treatment would not lead to finding the genes that will be significant in the start of immune evasion in chromosomally unstable cells, as not all the cells will activate the reporter, and not all the cells activating the reporter did it because they had starting error, but by being a bystander cell. The finding that IFNy can activate a reporter known to be only activated by IFNI, indicates that there might be an unknown mechanism that can link IFNII to IFNI signalling. Therefore, the reporter coupled to a suicide gene, can be used for the CRISPR screen to uncover genes that regulate the activation of an ISRE-IFN response signalling by IFNy, in the BT549 and the BJ cell lines. These efforts will improve our understanding of the molecular mechanisms underlying the loss of a normal inflammation signalling in response to INFy by identifying genes that when deleted or activated, can rescue the dampened inflammatory response. As we wanted to keep analyzing how to keep paracrine signalling from creating noise in our experiments, we tried to knock out (KO) transmembrane receptors known to be activated by IFN, like IFNAR2. Unfortunately, all of the attempts to KO this receptor in the BJ cell line were unsuccessful. Next, as is known that IFN signalling in cells due to viral or bacterial infections can be transferred between cells in a variety of ways, including connexin‐dependent intercellular transfer via gap junctions, by entry into bystander cells following cell fusion, and by inclusion in extracellular vesicles (EV) for extracellular remote transport, some inhibitors of endocytosis, which involves two important proteins, clathrin and caveolin, were used. EVs are lipid bilayer-bound vesicles in biological fluids and are released and captured by cells, and evidence suggests that EVs are usually taken up into endosomal compartments via endocytosis. EVs vesicles, has been found to mediate intercellular communication in scenarios including infection, inflammation, and malignancy. The use of the two inhibitors of endocytosis, at different doses, decreased the amount of GFP positive cells after reversine treatment compared to the reversine only treatment. While the reversine treated cells showed decreased GFP signal, IFNy treatment plus endocytosis inhibitors did not decrease it, showing that the receptors and the rest of the signalling was still working. Also, using a cell proliferation tracker, it was noticeable that in cells treated with both reversine and the endocytosis inhibitors, proliferation diminished each day. However, with the combination of a high dose of the inhibitors and reversine treatment, more cells were able to continue proliferation than in the single treatments, showing that CIN and endocytosis decrease will affect cells, but in combination, will allow some cells to continue dividing despite rounds of mitotic errors. Further research regarding the supposed impaired IFN signalling in BT549 cells, should be more focussed on the underlying causes. Investigating the genetic and epigenetic factors that contribute to this phenotype, for example, through the performance of a CRISPR screen. Through selective death induced by iCasp9 activation, only the cells with defective inflammatory signalling can survive. Performing the CRISPR screen in these surviving cells is expected to identify important inflammatory mediators. Further research is needed to understand the impaired IFN signalling in BT549 cells, possibly through genetic and epigenetic analysis. Also, as part of our efforts to collaborate with other groups to co-operate and increase interdisciplinary work, we worked with another group that used a joint computational and experimental approach to identify some genes located in commonly amplified regions of the genome. These genes are detrimental to the cell when overexpressed upon gain, which triggers mechanisms of gene compensation. One of those genes, RBM14, perturbs the DNA damage response and cGAS/STING signalling, and its amplification is associated with increased vulnerability of tumors to radiation treatment in a clinical colorectal cancer cohort. RBM14 overexpression pro motes STING activation in the context of DNA damage, and how this effect impairs immunosuppressive STAT3 signalling and may enhance NK cell-mediated tumor recognition and killing. Overall, the project further provides evidence of the importance of Intercellular communication for immune signalling in sterile inflammation in non-transformed cell lines, and how it can be connected to CIN and the beginning of transformation. It also discusses insights into the possibility of a cross-regulation between IFNI and IFNII at the moment unknown, which also shows the need for further research.

Data: CORDIS, © European Union