The consequences of aneuploidy and chromosome instability
Schukken, Klaske Marijke
DOI:
10.33612/diss.135392967
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2020
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Schukken, K. M. (2020). The consequences of aneuploidy and chromosome instability: Survival, cell death and cancer. University of Groningen. https://doi.org/10.33612/diss.135392967
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
5
Klaske M. Schukken1, Colin Pritchard2, Petra Bakker1, Alicia Borneman,
Ivo Huijbers2, and Floris Foijer1,
1European Research Institute for the Biology of Ageing (ERIBA), University
of Groningen, University Medical Center Groningen, 9713 AV, Groningen, the Netherlands
2 Mouse Clinic, National Cancer Institute (NKI), Antoni Van Leeuwenhoek
hospital, 1066 CX, Amsterdam, the Netherlands
Aneuploidy tracker- a novel mouse
model to track aneuploidy in live cells
Chapter 5
5
5
Aneuploidy tracker- a novel mouse model to track aneuploidy in live
cells
Klaske M. Schukken1, Colin Pritchard2, Petra Bakker1, Alicia Borneman, Ivo
Huijbers2, and Floris Foijer1,
1European Research Institute for the Biology of Ageing (ERIBA), University of
Groningen, University Medical Center Groningen, 9713 AV, Groningen, the Netherlands
2 Mouse Clinic, National Cancer Institute (NKI), Antoni Van Leeuwenhoek
hospital, 1066 CX, Amsterdam, the Netherlands
Abstract Three out of four cancers are aneuploid, and Chromosomal Instability (CIN) frequently enhances tumor progression. Many cancers select for specific chromosomes that are beneficial to growth, but which chromosomes are selected for depends greatly on the specific cancer type. Most of the aneuploidy measuring techniques currently available to the field (Immunofluorescence staining, karyotype spreads, CGH sequencing and single cell sequencing) are endpoint experiments. Additionally, there are no mouse models available that can be used to observe aneuploidy within living tissue. Here, we test two techniques to monitor chromosome copy number in living cells, fluorescent tetR with TRE and fluorescent dCas9 with sgRNAs. We find that the tetR TRE approach is the more reliable of the two strategies evaluated. We therefore engineered a beyond state of the art mouse model in which the TRE repetitive binding site is integrated into the mouse genome, flanked by PiggyBac transposon sites, allowing mobilization of the element to other chromosomes in mouse ES cells. In the future, this mouse model can be used to determine copy numbers of specific chromosomes in vivo, and monitor copy number alterations during tumor progression and aging, or in unperturbed tissues following the induction of a CIN phenotype. Keywords: Aneuploidy, mouse model, cancer, intravital imaging Introduction Cells can mis‐segregate their DNA, a process called Chromosome Instability (CIN), which leads to daughter cells with extra or missing chromosomes, a state called aneuploidy. While aneuploidy is generally detrimental to cell growth and survival2,42,75,77,86, cancers have found a way to select for specific chromosome gains and losses that are beneficial for cancer cell
proliferation7,9,17,27,187. Indeed, aneuploidy is a hallmark of cancer cells9, with
about 3 out of 4 cancers being aneuploid7,8. Importantly, patients with aneuploid cancers have a lower expected survival11–13,105–110. While there is no singular cancer karyotype, cancers of the same types tend to gain and loose the same chromosomes191, and cancers induced with the same mutation have been shown to independently select for similar chromosome gains and losses27. Per cancer type, certain chromosome gains or losses are known to strongly influence patient outcome 187. However, despite the frequency of aneuploidy in cancer, aneuploidy’s role in cancer development is not yet fully understood. While there is very little aneuploidy reported in healthy human tissues91, aneuploidy is a hallmark of aging103,182,192, and the frequency of aneuploid cells has been shown to increase within a few of our tissues as we age117,118. Mosaic variegated aneuploidy (MVA), a condition in which patients develop random aneuploidies due to defects in their chromosome segregation machinery121,122leads to heterogeneous aneuploidy in the affected individuals, and has also been shown to lead to enhanced aging phenotypes119,120,124,125,127. Despite the increase of aneuploid cells as we age, senescent and highly aneuploid cells have been shown to be eliminated by natural killer cells in co‐culture experiments97. However, how our bodies deal with aneuploid cells and tissues in vivo, how many aneuploidies have accumulated within our tissues, and how this changes over time, is not yet known. While there are many methods available to quantify aneuploidy44, these methods are all endpoint measurements, and therefore do not allow one to monitor the fate of aneuploid cells over time. It is therefore also not possible to evaluate the development of specific aneuploidies during tumor development. Therefore, tools to determine aneuploidy over time in vivo are a key step to better understand the effects of aneuploidy on cell survival, tissue aging, and tumor development. In this chapter we describe our attempts to create cell lines to monitor copy numbers of individual chromosomes using time‐lapse imaging, and
1
5
Abstract Three out of four cancers are aneuploid, and Chromosomal Instability (CIN) frequently enhances tumor progression. Many cancers select for specific chromosomes that are beneficial to growth, but which chromosomes are selected for depends greatly on the specific cancer type. Most of the aneuploidy measuring techniques currently available to the field (Immunofluorescence staining, karyotype spreads, CGH sequencing and single cell sequencing) are endpoint experiments. Additionally, there are no mouse models available that can be used to observe aneuploidy within living tissue. Here, we test two techniques to monitor chromosome copy number in living cells, fluorescent tetR with TRE and fluorescent dCas9 with sgRNAs. We find that the tetR TRE approach is the more reliable of the two strategies evaluated. We therefore engineered a beyond state of the art mouse model in which the TRE repetitive binding site is integrated into the mouse genome, flanked by PiggyBac transposon sites, allowing mobilization of the element to other chromosomes in mouse ES cells. In the future, this mouse model can be used to determine copy numbers of specific chromosomes in vivo, and monitor copy number alterations during tumor progression and aging, or in unperturbed tissues following the induction of a CIN phenotype. Keywords: Aneuploidy, mouse model, cancer, intravital imaging Introduction Cells can mis‐segregate their DNA, a process called Chromosome Instability (CIN), which leads to daughter cells with extra or missing chromosomes, a state called aneuploidy. While aneuploidy is generally detrimental to cell growth and survival2,42,75,77,86, cancers have found a way to select for specific chromosome gains and losses that are beneficial for cancer cellproliferation7,9,17,27,187. Indeed, aneuploidy is a hallmark of cancer cells9, with
about 3 out of 4 cancers being aneuploid7,8. Importantly, patients with aneuploid cancers have a lower expected survival11–13,105–110. While there is no singular cancer karyotype, cancers of the same types tend to gain and loose the same chromosomes191, and cancers induced with the same mutation have been shown to independently select for similar chromosome gains and losses27. Per cancer type, certain chromosome gains or losses are known to strongly influence patient outcome 187. However, despite the frequency of aneuploidy in cancer, aneuploidy’s role in cancer development is not yet fully understood. While there is very little aneuploidy reported in healthy human tissues91, aneuploidy is a hallmark of aging103,182,192, and the frequency of aneuploid cells has been shown to increase within a few of our tissues as we age117,118. Mosaic variegated aneuploidy (MVA), a condition in which patients develop random aneuploidies due to defects in their chromosome segregation machinery121,122leads to heterogeneous aneuploidy in the affected individuals, and has also been shown to lead to enhanced aging phenotypes119,120,124,125,127. Despite the increase of aneuploid cells as we age, senescent and highly aneuploid cells have been shown to be eliminated by natural killer cells in co‐culture experiments97. However, how our bodies deal with aneuploid cells and tissues in vivo, how many aneuploidies have accumulated within our tissues, and how this changes over time, is not yet known. While there are many methods available to quantify aneuploidy44, these methods are all endpoint measurements, and therefore do not allow one to monitor the fate of aneuploid cells over time. It is therefore also not possible to evaluate the development of specific aneuploidies during tumor development. Therefore, tools to determine aneuploidy over time in vivo are a key step to better understand the effects of aneuploidy on cell survival, tissue aging, and tumor development. In this chapter we describe our attempts to create cell lines to monitor copy numbers of individual chromosomes using time‐lapse imaging, and
engineer an “aneuploidy tracker” mouse model that allows for monitoring aneuploidy in live cells, and track cell/tissue fate over time. To quantify aneuploidy, we fluorescently label a specific chromosome, a technique used previously to observe DNA damage and chromosome movement35–37,193, allowing us to view chromosome copy number per cell in living cells. Results Testing tagging of individual chromosomes in living cells In order to monitor aneuploidy in live cells, we wanted to fluorescently label a specific chromosome in living cells (Figure 1A). While this method would not detect all forms of aneuploidy, it would display chromosome copy number variations for the labeled chromosome. Two previously published methods for fluorescently labelling chromosomes were tested, the tet‐TRE method35,193 and the dCas9‐sgRNA method36,37(see Materials
and Methods). For the first approach, we transduced previously‐described 3T3 NIH cells containing a TRE tandem193 with a construct fluorescently labeling tetR, H2B and alpha‐tubulin that localize to the TRE repeat, chromatin and tubulin, respectively (vector 487, Supplementary Table 2). After adding 1 μg/ml doxycycline to induce tetR binding to the TRE, we found that these cells indeed had clear fluorescent foci (Figure 1B). Since strong dCas9 or tetR protein binding to DNA during mitosis could cause problems for chromosome segregation, we next studied the localization of these proteins during mitosis using time‐lapse imaging. We found that the tetR fluorescent protein is removed from the chromatin during mitosis (Figure 1B), but returns several minutes after mitotic exit. Next, we tested the dCas9 approach. For this, we transduced 3T3 NIH cells with a fluorescently tagged non‐cleaving variant of Cas9 (Vector 552, Supplementary Table 2). To test functionality, we first tested this approach with previously‐published small guide RNAs (sgRNAs) targeting telomeres, a known long repetitive region present on all chromosomes. Indeed, 3T3 cells expressing telomere‐targeting sgRNAs and fluorescently‐tagged cCas9 showed multiple foci, corresponding to telomeres within the nucleus during interphase (Figure 1C). Similar to the tetR model described above, dCas9 was removed from the chromatin during mitosis (Figure 1C); the Figure 1: Overview of methods used to fluorescently label chromosomes. A) Schematic of method used to fluorescently label chromosomes. B) NIH 3T3 MEFs with a TRE repetitive binding site, and a fluorescent marker expressing tetR‐YFP, H2B‐GFP and mTurquoise2‐ Tubulin. TetR loci is visible in interphase but nor in mitosis. C) 3T3 cells with dCas9‐mCherry and telomere sgRNA in interphase where upon it forms foci in the nucleus, and in mitosis, where Cas9 does not form foci and is evenly distributed throughout the dividing cell. D) RPE1 A) B) C) Fluorophore DNA binding protein Repetative DNA binding site 50+ repeats D) C hr .9 s gR N A dCas9 H2B Merge Mitosis 3T 3 + te lo m er e Interphase E) 0 30 60 90 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X Y Chromosome R epeat C op y N umber Database TRD_Human BLAST Primary BLAST alternative 0 50 100 150 200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 X Chromosome R epeat C op y N umbe r Database TRD_Mouse BLAST BLCK6 BLAST Mixed Mouse Human
tetR Tubulin H2B Merge
M ito si s In te rp ha se
1
5
engineer an “aneuploidy tracker” mouse model that allows for monitoring aneuploidy in live cells, and track cell/tissue fate over time. To quantify aneuploidy, we fluorescently label a specific chromosome, a technique used previously to observe DNA damage and chromosome movement35–37,193, allowing us to view chromosome copy number per cell in living cells. Results Testing tagging of individual chromosomes in living cells In order to monitor aneuploidy in live cells, we wanted to fluorescently label a specific chromosome in living cells (Figure 1A). While this method would not detect all forms of aneuploidy, it would display chromosome copy number variations for the labeled chromosome. Two previously published methods for fluorescently labelling chromosomes were tested, the tet‐TRE method35,193 and the dCas9‐sgRNA method36,37(see Materialsand Methods). For the first approach, we transduced previously‐described 3T3 NIH cells containing a TRE tandem193 with a construct fluorescently labeling tetR, H2B and alpha‐tubulin that localize to the TRE repeat, chromatin and tubulin, respectively (vector 487, Supplementary Table 2). After adding 1 μg/ml doxycycline to induce tetR binding to the TRE, we found that these cells indeed had clear fluorescent foci (Figure 1B). Since strong dCas9 or tetR protein binding to DNA during mitosis could cause problems for chromosome segregation, we next studied the localization of these proteins during mitosis using time‐lapse imaging. We found that the tetR fluorescent protein is removed from the chromatin during mitosis (Figure 1B), but returns several minutes after mitotic exit. Next, we tested the dCas9 approach. For this, we transduced 3T3 NIH cells with a fluorescently tagged non‐cleaving variant of Cas9 (Vector 552, Supplementary Table 2). To test functionality, we first tested this approach with previously‐published small guide RNAs (sgRNAs) targeting telomeres, a known long repetitive region present on all chromosomes. Indeed, 3T3 cells expressing telomere‐targeting sgRNAs and fluorescently‐tagged cCas9 showed multiple foci, corresponding to telomeres within the nucleus during interphase (Figure 1C). Similar to the tetR model described above, dCas9 was removed from the chromatin during mitosis (Figure 1C); the Figure 1: Overview of methods used to fluorescently label chromosomes. A) Schematic of method used to fluorescently label chromosomes. B) NIH 3T3 MEFs with a TRE repetitive binding site, and a fluorescent marker expressing tetR‐YFP, H2B‐GFP and mTurquoise2‐ Tubulin. TetR loci is visible in interphase but nor in mitosis. C) 3T3 cells with dCas9‐mCherry and telomere sgRNA in interphase where upon it forms foci in the nucleus, and in mitosis, where Cas9 does not form foci and is evenly distributed throughout the dividing cell. D) RPE1 A) B) C) Fluorophore DNA binding protein Repetative DNA binding site 50+ repeats D) C hr .9 s gR N A dCas9 H2B Merge Mitosis 3T 3 + te lo m er e Interphase E) 0 30 60 90 120 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X Y Chromosome R epeat C op y N umber Database TRD_Human BLAST Primary BLAST alternative 0 50 100 150 200 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 X Chromosome R epeat C op y N umbe r Database TRD_Mouse BLAST BLCK6 BLAST Mixed Mouse Human
tetR Tubulin H2B Merge
M ito si s In te rp ha se
cells with dCas9‐mCherry and sgRNA for chromosome 9 with 2 visible loci. E) Bar graph of largest number of repetitive elements Figure 1 continued: (pattern >22bp, %match>95, containing sp dCas9 PAM site when possible) per chromosome in humans and mice. Number of correct repeats in the Tandem Repeat Database, and number of repeats found in 2 separate genomes in BLAST. fluorescent Cas9 does not form foci, and is even distributed throughout the dividing cell. This suggests that dCas9 is also removed from chromatin during mitosis. Next, we tested the same approach, but now in human RPE1 cells with previously published sgRNAs for chromosome 9 and chromosome 13 36. Both published sgRNAs tagging chromosome 9 and chromosome 13 showed clear fluorescent localization with 2 clear dots per cell, suggesting that the respective chromosomes were indeed tagged by dCas9 (chrm.9: Figure 1D, chrm.13: data not shown). However, in both RPE1 and 3T3 MEF cells dCas9 expression in combination with sgRNA presence was very toxic to the cells within a few weeks, suggesting that dCas9 localization to chromatin is poorly tolerated. Therefore, while cells may be able to deal with sgRNA and dCas9 co‐ expression for short periods of time, long term co‐expression was severely toxic, making the dCas9 chromosome labeling model too toxic for mouse models, at least for the published repeats on chromosome 9 and 13. Testing new sgRNAs for dCas9 chromosome loci Despite the toxicity of sgRNA and dCas9 co‐expression for chrs. 9 and 13, we next investigated whether other loci might be more suitable to be tagged with dCas9. To identify possible repeat regions to tag (repeats are required as otherwise the dCas9 signal might be too dim to be observed in vivo), we made use of the Tandem Repeat Database that contains long repetitive sequences present in the human and mouse genetic code. Repeats were selected based on multiple minimum requirements, such as presence of the PAM sequence required for dCas9 localization, a minimum of a 23 base pair repeat size and a base pair match rate larger than or equal to 93% between repeats. Repeats were manually curated to find the regions with the most identical 23bp repeats ending with a PAM sequence. Using these conditions, we identified the longest repeats per chromosome. These sequences were compared to two separate genome sequences, to check that they are unique to a single chromosome, and that these repeats are present in multiple genome variants for the species (Figure 1E, Figure 2: Making a TRE mouse model for chromosome labeling. A) Schematic of i) PiggyBac enclosed TRE vector and ii) how the transgenic mice were made from the TRE vector. B) Number of TRE integrations in genomic DNA per mouse, based on qPCR results set relative to mouse 10. Ratio of MEF offspring with TRE integration listed below. C) Primary MEF cell with a single PiggyBac‐TRE integration transduced with tet‐YFP in interphase and mitosis. D) Primary MEF cell from a mouse containing two TRE integrations, transduced with tet‐YFP. mCherry channel used for background fluorescence. | White arrows point to TRE‐tet loci. Dotted line represents nuclear outline. A) B) C) i) Pr im ar y M EF In te rp ha se Pr im ar y M EF M ito si s
5’ PiggyBac TRE 96x Luciferase fragment 3’ PiggyBac
Inject into zygote Inject embryo into
foster mother Offspring
TRE vector i)
ii)
tetR tetR
tetR Background fluorescence Merge
D ou bl e TR E in te gr at io n D) ii) TRE+ MEF 6/8 7/8 4/11 --0 1 2 3 4 5 6 7 8 9 20 13 6 14 10 17 21 8 Mouse ID R el ati ve TR E int egr ati on s qPCR 1 2 Piggybac trans-posase mRNA
1
5
cells with dCas9‐mCherry and sgRNA for chromosome 9 with 2 visible loci. E) Bar graph of largest number of repetitive elements Figure 1 continued: (pattern >22bp, %match>95, containing sp dCas9 PAM site when possible) per chromosome in humans and mice. Number of correct repeats in the Tandem Repeat Database, and number of repeats found in 2 separate genomes in BLAST. fluorescent Cas9 does not form foci, and is even distributed throughout the dividing cell. This suggests that dCas9 is also removed from chromatin during mitosis. Next, we tested the same approach, but now in human RPE1 cells with previously published sgRNAs for chromosome 9 and chromosome 13 36. Both published sgRNAs tagging chromosome 9 and chromosome 13 showed clear fluorescent localization with 2 clear dots per cell, suggesting that the respective chromosomes were indeed tagged by dCas9 (chrm.9: Figure 1D, chrm.13: data not shown). However, in both RPE1 and 3T3 MEF cells dCas9 expression in combination with sgRNA presence was very toxic to the cells within a few weeks, suggesting that dCas9 localization to chromatin is poorly tolerated. Therefore, while cells may be able to deal with sgRNA and dCas9 co‐ expression for short periods of time, long term co‐expression was severely toxic, making the dCas9 chromosome labeling model too toxic for mouse models, at least for the published repeats on chromosome 9 and 13. Testing new sgRNAs for dCas9 chromosome loci Despite the toxicity of sgRNA and dCas9 co‐expression for chrs. 9 and 13, we next investigated whether other loci might be more suitable to be tagged with dCas9. To identify possible repeat regions to tag (repeats are required as otherwise the dCas9 signal might be too dim to be observed in vivo), we made use of the Tandem Repeat Database that contains long repetitive sequences present in the human and mouse genetic code. Repeats were selected based on multiple minimum requirements, such as presence of the PAM sequence required for dCas9 localization, a minimum of a 23 base pair repeat size and a base pair match rate larger than or equal to 93% between repeats. Repeats were manually curated to find the regions with the most identical 23bp repeats ending with a PAM sequence. Using these conditions, we identified the longest repeats per chromosome. These sequences were compared to two separate genome sequences, to check that they are unique to a single chromosome, and that these repeats are present in multiple genome variants for the species (Figure 1E, Figure 2: Making a TRE mouse model for chromosome labeling. A) Schematic of i) PiggyBac enclosed TRE vector and ii) how the transgenic mice were made from the TRE vector. B) Number of TRE integrations in genomic DNA per mouse, based on qPCR results set relative to mouse 10. Ratio of MEF offspring with TRE integration listed below. C) Primary MEF cell with a single PiggyBac‐TRE integration transduced with tet‐YFP in interphase and mitosis. D) Primary MEF cell from a mouse containing two TRE integrations, transduced with tet‐YFP. mCherry channel used for background fluorescence. | White arrows point to TRE‐tet loci. Dotted line represents nuclear outline. A) B) C) i) Pr im ar y M EF In te rp ha se Pr im ar y M EF M ito sis5’ PiggyBac TRE 96x Luciferase fragment 3’ PiggyBac
Inject into zygote Inject embryo into
foster mother Offspring
TRE vector i)
ii)
tetR tetR
tetR Background fluorescence Merge
D ou bl e TR E in te gr at io n D) ii) TRE+ MEF 6/8 7/8 4/11 --0 1 2 3 4 5 6 7 8 9 20 13 6 14 10 17 21 8 Mouse ID R el ati ve TR E int egr ati on s qPCR 1 2 Piggybac trans-posase mRNA
Supplementary Table 1). While human repeat copy numbers were fairly consistent between genome sequence origin, mouse repeat copy numbers dropped off dramatically in the mixed background genome (Figure 1E, Supplementary Table 1A, B). Next, we designed six sgRNAs targeting these sequences (Supplementary Table 1 A), and tested in 3T3 cells expressing fluorescently‐labeled dCas9 and the respective sgRNAs (Supplementary Table 2). Unfortunately, none of the cell lines showed fluorescent foci. We therefore selected another set of repeats from the tandem repeat database using less stringent filters. We designed another 15 repeat‐based sgRNAs (Supplementary Table 1 C, and Supplementary Table 3). Again, none of additional 15 sgRNA resulted in visible dCas9 foci (Supplementary Table 3). While our tests of sgRNAs for repetitive regions targeting the mouse genome is far from exhaustive, the high failure rate and excessive toxicity along with the variability of repeat lengths between mouse genome sequences, make the dCas9 approach in our opinion a poor choice for a mouse model for now. Engineering Mouse models with a PiggyBac‐flanked TRE tandem As described above, co‐expression of sgRNA and dCas9 was severely toxic to cells within a few weeks, while tet‐TRE loci did not display this toxicity, possibly because TetR localization relies on the presence of doxycycline and is thus inducible or maybe because the Tet repeat is an exogenous DNA element. We therefore decided to create a mouse model with the TRE tandem integrated into the genome in order to visualize aneuploidy in vivo. Additionally, in order to make a model which can be used to make several mouse lines labeling different individual chromosomes, we engineered a vector with PiggyBac transposon flanking the TRE repetitive element (Figure 2Ai, see Materials and Methods). This PiggyBac‐TRE vector was used to engineer a mouse model (Figure 2Aii), which will henceforth be referred to as the “AneuTracker” mouse. For this purpose, the targeting vector was injected into mouse zygotes together with a vector expressing PiggyBac transposase. This leads to random integration of the PiggyBac flanked TRE cassette into the zygotes’ genomes. Because of this approach, the number of TRE integrations per mouse was unknown. To determine if our founder mice indeed harbored the PiggyBac‐TRE insert, genomic DNA was extracted followed by genotyping, which revealed that seven of the twenty‐one founder mice tested harbored the PiggyBac‐TRE cassette. To quantify the number of integrations of the AneuTracker cassette, we performed quantitative genomic PCRs (qgPCR) for each mouse (Figure 2B). In parallel, we crossed three male TRE+ mice with C57/B6 mice to produce MEFs containing the TRE tandem. MEFs were also genotyped for presence of the transgene and the ratio of TRE+ MEFs produced from a heterozygous crossing (i.e. with a wildtype female), in combination with qgPCR data was used to estimate how many TRE integrations their parent contained (Table 1). To test functionality of the transgene, the primary TRE+ MEFs were transduced with tetR‐YFP and treated with 1 ug/ml doxycycline to localize tetR to the Tet‐tandem. Time‐lapse imaging revealed that the TRE+ MEFs displayed chromosome foci in interphase, but not mitosis (Figure 2C), similar to what was observed in TRE+ 3T3 MEFs (Figure 1B). Some of the MEFs tested contained two TRE integrations, resulting in two TRE foci within the cell (Figure 2D). Together, these analyses revealed that our founder mice harbor between 1 and 9 TRE integrations, and the TRE repeats are long enough to be visualized with tetR‐YFP binding. While our AneuTracker mice do not yet have a transgenic cassette to label the Tre tandem in vivo, engineering a Tet reporter mouse will be the next step for this project as well as the generation of a collection of ES cell lines with the AneuTracker cassette present on specific chromosomes. Table 1: Estimated TRE integrations per mouse. Mouse number, PCR of gDNA for TRE integration, delta Cp for TRE vector integrations, ratio of TRE positive offspring and estimated number of TRE inserts.
Mouse # gDNA TRE + ΔCp 1* qPCR ΔCp 2* qPCR offspring TRE + inserts # TRE
9 + 1,54 1,31 NA 9 20 + 1,11 0,93 NA 6 13 + 0,60 0,45 NA 3 6 + 0,41 0,33 6 of 8 2 14 + 0,39 0,34 7 of 8 2 10 + 0,18 0,15 NA 1 17 + 0,14 0,14 4 of 11 1 21 ‐ 0,00 0,00 NA 0 *qPCR data was used to get ΔCp for two luciferase primer pairs, and relative copy number of TRE was estimated based qPCR and TRE + offspring ratio.
1
5
Supplementary Table 1). While human repeat copy numbers were fairly consistent between genome sequence origin, mouse repeat copy numbers dropped off dramatically in the mixed background genome (Figure 1E, Supplementary Table 1A, B). Next, we designed six sgRNAs targeting these sequences (Supplementary Table 1 A), and tested in 3T3 cells expressing fluorescently‐labeled dCas9 and the respective sgRNAs (Supplementary Table 2). Unfortunately, none of the cell lines showed fluorescent foci. We therefore selected another set of repeats from the tandem repeat database using less stringent filters. We designed another 15 repeat‐based sgRNAs (Supplementary Table 1 C, and Supplementary Table 3). Again, none of additional 15 sgRNA resulted in visible dCas9 foci (Supplementary Table 3). While our tests of sgRNAs for repetitive regions targeting the mouse genome is far from exhaustive, the high failure rate and excessive toxicity along with the variability of repeat lengths between mouse genome sequences, make the dCas9 approach in our opinion a poor choice for a mouse model for now. Engineering Mouse models with a PiggyBac‐flanked TRE tandem As described above, co‐expression of sgRNA and dCas9 was severely toxic to cells within a few weeks, while tet‐TRE loci did not display this toxicity, possibly because TetR localization relies on the presence of doxycycline and is thus inducible or maybe because the Tet repeat is an exogenous DNA element. We therefore decided to create a mouse model with the TRE tandem integrated into the genome in order to visualize aneuploidy in vivo. Additionally, in order to make a model which can be used to make several mouse lines labeling different individual chromosomes, we engineered a vector with PiggyBac transposon flanking the TRE repetitive element (Figure 2Ai, see Materials and Methods). This PiggyBac‐TRE vector was used to engineer a mouse model (Figure 2Aii), which will henceforth be referred to as the “AneuTracker” mouse. For this purpose, the targeting vector was injected into mouse zygotes together with a vector expressing PiggyBac transposase. This leads to random integration of the PiggyBac flanked TRE cassette into the zygotes’ genomes. Because of this approach, the number of TRE integrations per mouse was unknown. To determine if our founder mice indeed harbored the PiggyBac‐TRE insert, genomic DNA was extracted followed by genotyping, which revealed that seven of the twenty‐one founder mice tested harbored the PiggyBac‐TRE cassette. To quantify the number of integrations of the AneuTracker cassette, we performed quantitative genomic PCRs (qgPCR) for each mouse (Figure 2B). In parallel, we crossed three male TRE+ mice with C57/B6 mice to produce MEFs containing the TRE tandem. MEFs were also genotyped for presence of the transgene and the ratio of TRE+ MEFs produced from a heterozygous crossing (i.e. with a wildtype female), in combination with qgPCR data was used to estimate how many TRE integrations their parent contained (Table 1). To test functionality of the transgene, the primary TRE+ MEFs were transduced with tetR‐YFP and treated with 1 ug/ml doxycycline to localize tetR to the Tet‐tandem. Time‐lapse imaging revealed that the TRE+ MEFs displayed chromosome foci in interphase, but not mitosis (Figure 2C), similar to what was observed in TRE+ 3T3 MEFs (Figure 1B). Some of the MEFs tested contained two TRE integrations, resulting in two TRE foci within the cell (Figure 2D). Together, these analyses revealed that our founder mice harbor between 1 and 9 TRE integrations, and the TRE repeats are long enough to be visualized with tetR‐YFP binding. While our AneuTracker mice do not yet have a transgenic cassette to label the Tre tandem in vivo, engineering a Tet reporter mouse will be the next step for this project as well as the generation of a collection of ES cell lines with the AneuTracker cassette present on specific chromosomes. Table 1: Estimated TRE integrations per mouse. Mouse number, PCR of gDNA for TRE integration, delta Cp for TRE vector integrations, ratio of TRE positive offspring and estimated number of TRE inserts.Mouse # gDNA TRE + ΔCp 1* qPCR ΔCp 2* qPCR offspring TRE + inserts # TRE
9 + 1,54 1,31 NA 9 20 + 1,11 0,93 NA 6 13 + 0,60 0,45 NA 3 6 + 0,41 0,33 6 of 8 2 14 + 0,39 0,34 7 of 8 2 10 + 0,18 0,15 NA 1 17 + 0,14 0,14 4 of 11 1 21 ‐ 0,00 0,00 NA 0 *qPCR data was used to get ΔCp for two luciferase primer pairs, and relative copy number of TRE was estimated based qPCR and TRE + offspring ratio.
Future directions and discussion The aim of this study was to create a mouse model we call the AneuTracker, which can be used to visualize aneuploidy in live cells and tissues. Such a model would allow us to better quantify the extent of aneuploidy within different tissues or tumors over time, and follow the fate of aneuploid cells in vivo. Currently, in vivo aneuploidy measurements are largely endpoint measurements44. Thus, creating the AneuTracker model would give us a tool to better understand the extent and effects of aneuploidy within living cells, tissues and tumors. Fluorescent dCas9 localization We were able to replicate previously published data showing fluorescent dCas9 foci at telomeres in both human and mouse cells, and dCas9 targeting to chromosome 9 and 13 in human cells. However, we were not able to create sgRNAs that led to dCas9 foci in mouse cells. This may be due to the mouse repetitive sequence lengths being less consistent than human repeats, indicating that a mouse’s strain may play a large role in whether or not a repeat is long enough to be visualized via fluorescent dCas9, or due to the limited number of sgRNAs tested. A third explanation is that repeats would typically be part of the heterochromatin and therefore poorly accessible to dCas9 binding. This could also explain the toxicity when targeting did work: maybe in this setting heterochromatin that was supposed to be closed is opened up leading to toxicity. Another explanation for the toxicity might be that while dCas9 did not bind to DNA during mitosis, it may be interfering with genome replication during interphase. Since more than 50 dCas9 proteins bind to the same genomic region, cCas9 binding may be causing repeated replication fork stalling and subsequent replication errors194. These replications errors may lead to DNA damage or chromosome mis‐segregation over time. To prevent this from occurring in our TRE‐based model, we used a binding protein (tet repressor rtTA) that only binds to the tandem repeat when doxycycline is present. This allows for much more precise control of how long the fluorescent binding proteins are bound to the chromatin, and may help prevent the genetic combination of rtTA & TRE from being toxic to cells. Goal of the aneuploid tracker mouse While the AneuTracker mice cannot yet be used for in vivo aneuploidy imaging, the mice do contain TRE integrations, and can be used to harvest tissues, MEFs, or derive organoids. The resulting cells can be transduced in vitro with tetR fluorescent constructs to visualize TRE copy number per cell and monitor the aneuploidy of these chromosome(s) in live cells. In future experiments, new genetically modified mice with tetR fluorescence expression will be engineered. As only fluorescent dots are not sufficient to count copy numbers in individual cells (as the cell boundaries are not visible), we will combine the AneuTracker detection cassette with H2B to label the chromatin and nucleus, and Tubulin to label the mitotic spindle and cytoplasm. The key advantage of our new mouse model is that chromosome numbers can be tracked over time, albeit form one individual chromosome. For this purpose, it is essential to derive mouse strains with the AneuTracker cassette integrated in a single chromosome. These mice will be crossed to homozygosity so that two dots will represent a euploid cell for the AneuTracker chromosome. In these mice aneuploidy can be tracked over time in living mice in tissues of choice, both in healthy tissue as well as tumors or other conditions. By making a collection of ES cell lines with the AneuTracker integrated on individual chromosomes available to the research community, one can chose his/her favorite chromosome to track in vivo in the near future. TRE integration into specific chromosomes via the PiggyBac transposon Zygote targeting of the AneuTracker construct yielded seven TRE+ mice, some with multiple TRE integrations (Figure 2B). These TRE integrations may all be located in different locations in the genome, up to 24 separate locations. To detect the exact integration site, splinkerette PCRs are currently being performed to map the chromosome location of all the integrated AneuTracker cassettes. In addition to having several mice with different TRE integrations, the TRE binding site is flanked by 5’and 3’ PiggyBac transposon sequences. Using PiggyBac transposase the PiggyBac flanked sequence can be mobilized throughout the genome in ES cells to generate new mice which have the TRE integrated at additional loci. This PiggyBac transposase method can be used to generate a host of mouse ES cell lines, with the TRE integrated into
1
5
Future directions and discussion The aim of this study was to create a mouse model we call the AneuTracker, which can be used to visualize aneuploidy in live cells and tissues. Such a model would allow us to better quantify the extent of aneuploidy within different tissues or tumors over time, and follow the fate of aneuploid cells in vivo. Currently, in vivo aneuploidy measurements are largely endpoint measurements44. Thus, creating the AneuTracker model would give us a tool to better understand the extent and effects of aneuploidy within living cells, tissues and tumors. Fluorescent dCas9 localization We were able to replicate previously published data showing fluorescent dCas9 foci at telomeres in both human and mouse cells, and dCas9 targeting to chromosome 9 and 13 in human cells. However, we were not able to create sgRNAs that led to dCas9 foci in mouse cells. This may be due to the mouse repetitive sequence lengths being less consistent than human repeats, indicating that a mouse’s strain may play a large role in whether or not a repeat is long enough to be visualized via fluorescent dCas9, or due to the limited number of sgRNAs tested. A third explanation is that repeats would typically be part of the heterochromatin and therefore poorly accessible to dCas9 binding. This could also explain the toxicity when targeting did work: maybe in this setting heterochromatin that was supposed to be closed is opened up leading to toxicity. Another explanation for the toxicity might be that while dCas9 did not bind to DNA during mitosis, it may be interfering with genome replication during interphase. Since more than 50 dCas9 proteins bind to the same genomic region, cCas9 binding may be causing repeated replication fork stalling and subsequent replication errors194. These replications errors may lead to DNA damage or chromosome mis‐segregation over time. To prevent this from occurring in our TRE‐based model, we used a binding protein (tet repressor rtTA) that only binds to the tandem repeat when doxycycline is present. This allows for much more precise control of how long the fluorescent binding proteins are bound to the chromatin, and may help prevent the genetic combination of rtTA & TRE from being toxic to cells. Goal of the aneuploid tracker mouse While the AneuTracker mice cannot yet be used for in vivo aneuploidy imaging, the mice do contain TRE integrations, and can be used to harvest tissues, MEFs, or derive organoids. The resulting cells can be transduced in vitro with tetR fluorescent constructs to visualize TRE copy number per cell and monitor the aneuploidy of these chromosome(s) in live cells. In future experiments, new genetically modified mice with tetR fluorescence expression will be engineered. As only fluorescent dots are not sufficient to count copy numbers in individual cells (as the cell boundaries are not visible), we will combine the AneuTracker detection cassette with H2B to label the chromatin and nucleus, and Tubulin to label the mitotic spindle and cytoplasm. The key advantage of our new mouse model is that chromosome numbers can be tracked over time, albeit form one individual chromosome. For this purpose, it is essential to derive mouse strains with the AneuTracker cassette integrated in a single chromosome. These mice will be crossed to homozygosity so that two dots will represent a euploid cell for the AneuTracker chromosome. In these mice aneuploidy can be tracked over time in living mice in tissues of choice, both in healthy tissue as well as tumors or other conditions. By making a collection of ES cell lines with the AneuTracker integrated on individual chromosomes available to the research community, one can chose his/her favorite chromosome to track in vivo in the near future. TRE integration into specific chromosomes via the PiggyBac transposon Zygote targeting of the AneuTracker construct yielded seven TRE+ mice, some with multiple TRE integrations (Figure 2B). These TRE integrations may all be located in different locations in the genome, up to 24 separate locations. To detect the exact integration site, splinkerette PCRs are currently being performed to map the chromosome location of all the integrated AneuTracker cassettes. In addition to having several mice with different TRE integrations, the TRE binding site is flanked by 5’and 3’ PiggyBac transposon sequences. Using PiggyBac transposase the PiggyBac flanked sequence can be mobilized throughout the genome in ES cells to generate new mice which have the TRE integrated at additional loci. This PiggyBac transposase method can be used to generate a host of mouse ES cell lines, with the TRE integrated intoany specific chromosome of interest, from which new mouse strains can be derived, all part of currently ongoing work. In conclusion, while the aneuploid tracker mouse model is not complete, it will ultimately allow researchers to monitor copy numbers of a specific chromosome, and track the fate of the aneuploid cells and tissues over time to better understand the consequences of aneuploidy in vivo. Material and Methods Cell culture RPE1 (ATCC), 3T3 NIH MEFs (ATCC), and NIH 3T3 MEFs with LacZ & TRE (a kind gift from Tom Misteli) were grown in DMEM (Gibco, ThermoFisher Scientific) with 10% FBS, and 1% Pen/Strep (Ref 15140122, Gibco, Thermo Scientific), at 37°C with 5% CO2 and 18% O2. Primary MEFs were grown in MEF media at 37°C, in 5% CO2 and 5%O2. MEF media was made from DMEM, with 10% FBS, 1% Pen/Strep, 1% MEM NEAA (Cat. Num. 11140050) (Gibco, ThermoFisher Scientific), and 0.1% 5.5x10^‐2 β‐Mercapto‐Ethanol (Cat. Num. 21985023, Gibco, Thermo Scientific). Imaging Cells were imaged on a Delta Vision microscope (Applied Precision Ltd.) at 37°C, with oxygen and carbon dioxide conditions identical to culture conditions. Cells were plated in glass bottom imaging dishes, and imaged with a 60x Olympus (Olympus co.) objectives, with an oil with a refractive index of 1.518. 10 to 15 z‐stacks at 0.5μM spacing were taken per picture, and a maximum projection was made. Methods for fluorescently labeling chromosomes Two genetic elements can be used to fluorescently label a chromosome in a live cell: a fluorescent DNA binding element, and a repetitive unique binding site for the protein. We looked into two methods of chromosome labeling, the first being a fluorescently labeled, endonuclear dead Cas9 (dCas9) with sgRNAs targeting the fluorescent dCas9 to repetitive sequences within the genome36,37. The second technique use fluorescently labeled rtTA (tetR) which localizes to the Tetracycline binding site, 96 of which are successively located on the Tetracycline Repetitive Element (TRE). We also briefly looked into using TALENs, however since they would have the same genomic binding sites as dCas9 while being more labor intensive to clone, we decided not to pursue this technique. In order to visualize fluorescent foci, multiple fluorescent proteins need to be localized to the same region, thus a long repetitive piece of DNA serving as multiple protein binding sites needs to be present. The TRE contains 96 copies of the tetracycline binding site, making it a robust sequence which has previously already been used to visualize chromosome loci when combined with a fluorescent tetR protein193. Alternatively, genomic DNA itself naturally contains repetitive regions, many of which are unique to the chromosome region it’s located on195. These gDNA repetitive regions can be targeted by a fluorescent dCas9 to fluorescently label these regions36,37. sgRNA primer design Small guide RNA (sgRNA) primers were designed based on repetitive regions from the tandem repeat database. Initially primers were designed based on tandem repeats with more than or equal to 23 base pair repeats, with more than 50 repeats, more than 98% match rate between repeats, and containing a “GG” sequence. Since NGG is the PAM site sequence for sp dCas9, the 20bp 5’of the NGG sequence was used. The number of times the exact 20bp sequence 5’ of the NGG site occurred in the repeat region was recorded. This sequence was run through BLAST196 to check that it only occurred as a repetitive region on only a single chromosome, and that it occurred in two different genome sequences. Several more sgRNAs were made using a similar protocol, but with less stringent initial screening: 14bp sequences with a minimum of 30 repeats, and 85% match rate between repeats. Repeat numbers and sequences used to design sgRNAs are listed in Supplementary Table 1A and 1C for mice, and 1B for humans. sgRNA primers were designed based on the 11 to 20bp regions 5’ of the PAM site. ACCG was added 5’of Fw primers, and AAAC was added 5’ of reverse primers. These primers were annealed and cloned into the pLH s. pyogenes shRNA2 vector backbone digested with BbsI. sgRNA primer sequences are listed in Supplementary Table 3. PiggyBac Transposon
1
5
any specific chromosome of interest, from which new mouse strains can be derived, all part of currently ongoing work. In conclusion, while the aneuploid tracker mouse model is not complete, it will ultimately allow researchers to monitor copy numbers of a specific chromosome, and track the fate of the aneuploid cells and tissues over time to better understand the consequences of aneuploidy in vivo. Material and Methods Cell culture RPE1 (ATCC), 3T3 NIH MEFs (ATCC), and NIH 3T3 MEFs with LacZ & TRE (a kind gift from Tom Misteli) were grown in DMEM (Gibco, ThermoFisher Scientific) with 10% FBS, and 1% Pen/Strep (Ref 15140122, Gibco, Thermo Scientific), at 37°C with 5% CO2 and 18% O2. Primary MEFs were grown in MEF media at 37°C, in 5% CO2 and 5%O2. MEF media was made from DMEM, with 10% FBS, 1% Pen/Strep, 1% MEM NEAA (Cat. Num. 11140050) (Gibco, ThermoFisher Scientific), and 0.1% 5.5x10^‐2 β‐Mercapto‐Ethanol (Cat. Num. 21985023, Gibco, Thermo Scientific). Imaging Cells were imaged on a Delta Vision microscope (Applied Precision Ltd.) at 37°C, with oxygen and carbon dioxide conditions identical to culture conditions. Cells were plated in glass bottom imaging dishes, and imaged with a 60x Olympus (Olympus co.) objectives, with an oil with a refractive index of 1.518. 10 to 15 z‐stacks at 0.5μM spacing were taken per picture, and a maximum projection was made. Methods for fluorescently labeling chromosomes Two genetic elements can be used to fluorescently label a chromosome in a live cell: a fluorescent DNA binding element, and a repetitive unique binding site for the protein. We looked into two methods of chromosome labeling, the first being a fluorescently labeled, endonuclear dead Cas9 (dCas9) with sgRNAs targeting the fluorescent dCas9 to repetitive sequences within the genome36,37. The second technique use fluorescently labeled rtTA (tetR) which localizes to the Tetracycline binding site, 96 of which are successively located on the Tetracycline Repetitive Element (TRE). We also briefly looked into using TALENs, however since they would have the same genomic binding sites as dCas9 while being more labor intensive to clone, we decided not to pursue this technique. In order to visualize fluorescent foci, multiple fluorescent proteins need to be localized to the same region, thus a long repetitive piece of DNA serving as multiple protein binding sites needs to be present. The TRE contains 96 copies of the tetracycline binding site, making it a robust sequence which has previously already been used to visualize chromosome loci when combined with a fluorescent tetR protein193. Alternatively, genomic DNA itself naturally contains repetitive regions, many of which are unique to the chromosome region it’s located on195. These gDNA repetitive regions can be targeted by a fluorescent dCas9 to fluorescently label these regions36,37. sgRNA primer design Small guide RNA (sgRNA) primers were designed based on repetitive regions from the tandem repeat database. Initially primers were designed based on tandem repeats with more than or equal to 23 base pair repeats, with more than 50 repeats, more than 98% match rate between repeats, and containing a “GG” sequence. Since NGG is the PAM site sequence for sp dCas9, the 20bp 5’of the NGG sequence was used. The number of times the exact 20bp sequence 5’ of the NGG site occurred in the repeat region was recorded. This sequence was run through BLAST196 to check that it only occurred as a repetitive region on only a single chromosome, and that it occurred in two different genome sequences. Several more sgRNAs were made using a similar protocol, but with less stringent initial screening: 14bp sequences with a minimum of 30 repeats, and 85% match rate between repeats. Repeat numbers and sequences used to design sgRNAs are listed in Supplementary Table 1A and 1C for mice, and 1B for humans. sgRNA primers were designed based on the 11 to 20bp regions 5’ of the PAM site. ACCG was added 5’of Fw primers, and AAAC was added 5’ of reverse primers. These primers were annealed and cloned into the pLH s. pyogenes shRNA2 vector backbone digested with BbsI. sgRNA primer sequences are listed in Supplementary Table 3. PiggyBac TransposonThe TRE genetic element integrated into the mouse genome was flanked by 5’ and 3’ PiggyBac transposon sequences (vector 617, see Supplementary Table 2). Thus, the TRE is integrated in a PiggyBac transposon, and can be activated via PiggyBac Transposase, which would flip the element out of the genome, and integrate it randomly into another region of the genome without leaving genetic scars197. Isolating single cell clones from transposase‐treated cells, or genotyping offspring from transposase treated mice would generate a library of cells or mouse lines with the TRE integrated into different chromosomes. A splinkerette PCR will be used to sequence the area of the genome that the TRE has integrated into. DNA cloning The list of vectors cloned for this chapter are listed in Supplementary Table 2, and the primers for the sgRNA vectors clones are listed in Supplementary Table 3. Tet vectors, dCas9 vectors and sgRNA vectors were cloned using standard cloning techniques: PCR with Phusion polymerase (M0530S, NEB), digest with NEB digestion enzymes, ligation (T4 DNA ligase, NEB), transformation, miniprep (GeneJET plasmid miniprep kit, K0502, Thermo Scientific) and transduction (Turbofect transfection reagent, R0531, Thermo Scientific). New vectors were sequenced with sanger sequencing (GATC, Eurofins Genomics) before being used. The vector with 96x repeats of TRE was a kind gift from David Spector. Vectors containing the TRE repeat were cloning using the Gibson assembly kit. The TRE repeat was never digested out or PCRed due to its highly repetitive sequence. Therefore, 5’ genetic inserts and 3’ inserts were cloned into the TRE vector sequentially. Since the TRE repeat cannot be PCRed or sequenced, TRE repeat retention was checked by digesting DNA and checking TRE DNA size. Vectors containing the TRE insert were never successfully transduced into cells due to the length and repetitive nature of the vector. The p3526 TRE tandem vector was a kind gift from Spector lab. All other vectors are available upon request. Engineering a new mouse model The genetically modified mice were created using the method described in Ivics et al. (2014)198. A few modifications were made to the protocol. Instead of our genetic insert containing the Sleeping Beauty transposon, it contained the PiggyBac transposon, therefore PiggyBac transposase (not Sleepying Beauty transposase) mRNA was injected along with the vector into the pronucleus of mouse zygotes. These zygotes were cultured to the 2‐cell stage and implanted into pseudo‐pregnant foster mice. Mouse lines and harvesting MEFs All animal experiments were performed in accordance with Dutch law and approved by the University of Groningen Medical Center Committees on Animal Care. Male mice with the aneuploidy tracker genotype were setup in a timed mating with female C57/B6 mice. Pregnant females were terminated on day 13.5 of their pregnancies. E13.5 embryos were harvested, and embryonic sac, internal organs and the head were removed in a laminar flow cabinet. Tissue was homogenized with a scalpel, then digested with Trypsin for 15 minutes. Cells were resuspended in MEF media. Resulting MEF cell lines were genotypes for inheritance of the desired genotype and negative MEFs were discarded. Genotyping Aneuploidy tracker mice were genotyped with three primer pairs targeting the 5’PiggyBac sequence, the 3’ Luciferase and PiggyBac sequence, and the TRE sequence, with expected sizes of 332bp, 1160bp, and 400bp with a smear, respectively: 5’PiggyBac primers: Fw 5’ TATATAGACGTCTTAACCCTAGAAAGATAGTCTGC Rv 5’ TATATACTCGAGTGATATCTATAACAAGAAAATATATATATAATAAG 3’Luciferse PiggyBac primers: Fw 5’ AAAGTGAAAGTCGTCGAGAATTCGTCCACGAACACAACACCAC Rv 5’ CCTGAAAATCTCGCCAAGCTTAACCCTAGAAAGATAATCATATTG TRE primers: Fw 5’ CCTGCCGCTTAACCCTAGAAAGATAGTCTGC Rv 5’ TATATACACTTTTCTCTATCACTGATAGGGAGTGG qPCR Two primer pairs corresponding to the luciferase fragment in the PiggyBac‐ TRE vector were used to quantify the number of PiggyBac‐TRE integrations in the genomic DNA of the aneuploidy tracker mice. Two Mad2 primer pairs
1
5
The TRE genetic element integrated into the mouse genome was flanked by 5’ and 3’ PiggyBac transposon sequences (vector 617, see Supplementary Table 2). Thus, the TRE is integrated in a PiggyBac transposon, and can be activated via PiggyBac Transposase, which would flip the element out of the genome, and integrate it randomly into another region of the genome without leaving genetic scars197. Isolating single cell clones from transposase‐treated cells, or genotyping offspring from transposase treated mice would generate a library of cells or mouse lines with the TRE integrated into different chromosomes. A splinkerette PCR will be used to sequence the area of the genome that the TRE has integrated into. DNA cloning The list of vectors cloned for this chapter are listed in Supplementary Table 2, and the primers for the sgRNA vectors clones are listed in Supplementary Table 3. Tet vectors, dCas9 vectors and sgRNA vectors were cloned using standard cloning techniques: PCR with Phusion polymerase (M0530S, NEB), digest with NEB digestion enzymes, ligation (T4 DNA ligase, NEB), transformation, miniprep (GeneJET plasmid miniprep kit, K0502, Thermo Scientific) and transduction (Turbofect transfection reagent, R0531, Thermo Scientific). New vectors were sequenced with sanger sequencing (GATC, Eurofins Genomics) before being used. The vector with 96x repeats of TRE was a kind gift from David Spector. Vectors containing the TRE repeat were cloning using the Gibson assembly kit. The TRE repeat was never digested out or PCRed due to its highly repetitive sequence. Therefore, 5’ genetic inserts and 3’ inserts were cloned into the TRE vector sequentially. Since the TRE repeat cannot be PCRed or sequenced, TRE repeat retention was checked by digesting DNA and checking TRE DNA size. Vectors containing the TRE insert were never successfully transduced into cells due to the length and repetitive nature of the vector. The p3526 TRE tandem vector was a kind gift from Spector lab. All other vectors are available upon request. Engineering a new mouse model The genetically modified mice were created using the method described in Ivics et al. (2014)198. A few modifications were made to the protocol. Instead of our genetic insert containing the Sleeping Beauty transposon, it contained the PiggyBac transposon, therefore PiggyBac transposase (not Sleepying Beauty transposase) mRNA was injected along with the vector into the pronucleus of mouse zygotes. These zygotes were cultured to the 2‐cell stage and implanted into pseudo‐pregnant foster mice. Mouse lines and harvesting MEFs All animal experiments were performed in accordance with Dutch law and approved by the University of Groningen Medical Center Committees on Animal Care. Male mice with the aneuploidy tracker genotype were setup in a timed mating with female C57/B6 mice. Pregnant females were terminated on day 13.5 of their pregnancies. E13.5 embryos were harvested, and embryonic sac, internal organs and the head were removed in a laminar flow cabinet. Tissue was homogenized with a scalpel, then digested with Trypsin for 15 minutes. Cells were resuspended in MEF media. Resulting MEF cell lines were genotypes for inheritance of the desired genotype and negative MEFs were discarded. Genotyping Aneuploidy tracker mice were genotyped with three primer pairs targeting the 5’PiggyBac sequence, the 3’ Luciferase and PiggyBac sequence, and the TRE sequence, with expected sizes of 332bp, 1160bp, and 400bp with a smear, respectively: 5’PiggyBac primers: Fw 5’ TATATAGACGTCTTAACCCTAGAAAGATAGTCTGC Rv 5’ TATATACTCGAGTGATATCTATAACAAGAAAATATATATATAATAAG 3’Luciferse PiggyBac primers: Fw 5’ AAAGTGAAAGTCGTCGAGAATTCGTCCACGAACACAACACCAC Rv 5’ CCTGAAAATCTCGCCAAGCTTAACCCTAGAAAGATAATCATATTG TRE primers: Fw 5’ CCTGCCGCTTAACCCTAGAAAGATAGTCTGC Rv 5’ TATATACACTTTTCTCTATCACTGATAGGGAGTGG qPCR Two primer pairs corresponding to the luciferase fragment in the PiggyBac‐ TRE vector were used to quantify the number of PiggyBac‐TRE integrations in the genomic DNA of the aneuploidy tracker mice. Two Mad2 primer pairswere used at the genomic control for each mouse gDNA. Mice that were TRE positive and mice that were negative were tested. The average Cp for three replicates for genomic DNA control qPCR was subtracted from each luciferase fragment qPCR Cp to get the delta‐Cp per primer pair. Two to the power of the delta Cp per mouse per primer pair were listed and set relative to the mouse with the lowest expression level (mouse 10) to get the copy number of TRE. Luciferase primer pair 1: FW 5’ CAGCTTCTTGGCGGTTGTA Rv 5’ AAAACCATGACCGAGAAGGA Luciferase primer pair 2: FW 5’ CAGTGTCTTACCGGTGTCCA Rv 5’ CGCAGTAGGCAAGGTGGT Mad2 primer pair 1: FW 5’ TTAGGGAGGGATTCGGAGTT Rv 5’ CAGGCGTAATGAGCCCTAAG Mad2 primer pair 2: FW 5’ TAGTGATGGCACAGCAGCTC Rv 5’ CGCTCAGCACTCACAGAAAA Supple mentar y Ta bl e 1A : Lo ngest re peats per chromosomes found in T RD B pe r m ou se chromosome, once filtered for several specific re quir ements. Listed are the ch romoso me n um be r, the pat tern si ze in basepair s, the co py nu mber o f re peats, a nd per cen t match re peats have to the m ea n, sg RN A nu mber made fr om the se quen ce, n um be r o f i de nti cal repeats foun d, a nd n um be r o f repeats fo und in two alternat ive genome se quen ces, a nd th e repeat pattern . Ch ro m oso m e Patter n Size Copy Nu m be r Match % sgRNA Go od repeat s BLCK6 BLAS T mix bl as t Pattern chr1 42 66 95 58 37 18 TCACCAGCTCT AAC ATCCATG GATTCCCCCAT CTG TCTGTAG chr2 37 59 97 32 37 7 AGGGGACAG AGCTGGGGTGG GGTATAGACAGA AGTGT chr3 31 87 99 3.2 66 66 12 GGTGATGAGAGAAGGGGTTGGTACAGTGGAT chr4 37 326 96 4.1 227 148 10 CAGGCAGGCAAAAGTGC ATTACATC ATAC AG GA CA GC chr5 42 117 94 61 46 27 CT CA GG GC AG CC TG TG CA CT GA CT GC AG AT AG T GAGTGAGGG chr6 50 51 99 6.1 49 104 0 CTCATCCACTG ATCCTGCAGGTAGACTCC ATGGC TA TC CT GC TC TC TG TA chr7 37 82 99 7.3 81 80 7 AA GG TG CA TG CT GG AA GC TC TG TA TA AG GA GCC TCGG chr8 51 108 98 108 110 14 AGCAGCTTTCTGTAGGTGTAGACGCATGCGCAC CTCCTGCCAAT CAATC AA chr9 34 154 97 110 122 19 CAGTCAGGAG CATCCTGTGCACAGTGGGAGGAG A
