Rapid target gene validation in complex cancer mouse models using re-derived embryonic stem cells

Abstract

Human cancers modeled in Genetically Engineered Mouse Models (GEMMs) can provide important mechanistic insights into the molecular basis of tumor development and enable testing of new intervention strategies. The inherent complexity of these models, with often multiple modified tumor suppressor genes and oncogenes, has hampered their use as preclinical models for validating cancer genes and drug targets. In our newly developed approach for the fast generation of tumor cohorts we have overcome this obstacle, as exemplified for three GEMMs; two lung cancer models and one mesothelioma model. Three elements are central for this system; (i) The efficient derivation of authentic Embryonic Stem Cells (ESCs) from established GEMMs, (ii) the routine introduction of transgenes of choice in these GEMM-ESCs by Flp recombinase-mediated integration and (iii) the direct use of the chimeric animals in tumor cohorts. By applying stringent quality controls, the GEMM-ESC approach proofs to be a reliable and effective method to speed up cancer gene assessment and target validation. As proof-of-principle, we demonstrate that MycL1 is a key driver gene in Small Cell Lung Cancer.

Figure 1

Optimization of ESC culture and injection procedures. A FACS profile of three core ESC transcription factors, Nanog, Oct4 and Sox2, in C57BL/6J ESC clone 1.4cultured in 2i medium (Blue). Red population represents the isotype control. B BFACS profile of three core ESC transcription factors, Nanog, Oct4 and Sox2, in FVB/n ESC clone 1.3 cultured in 2i medium (Blue). Red population represents the isotype control. C Three ESC injection procedures for C57BL/6J ESC clone 1.4 were evaluated on basis of chimeric contribution. Injecting 4–8 ESC per FVB/n morula followed by overnight culture in KSOM medium provided the best chimeras, with nine mice showing 100% coat color contribution (entirely black). male, female, n.d. D Two ESC injection procedures for FVB/n ESC clone 1.3 were evaluated on basis of chimeric contribution. Blastocyst injections resulted in reasonable chimeras whereas ESC injections into morulae in combination with overnight culture improved chimerism with three out of four live borns showing 100% chimerism (entirely white). The 80% chimera was a runt and died before weaning. male, female, n.d. E,F Efficiency of ESC injection procedures shown in (C) and (D) based on number of viable chimeras born compared to the total number of implanted embryos for C57BL/6J ESC clone 1.4 (E) and FVB/n ESC clone 1.3 (F). Note, for both ESC clones fewer chimeras were observed relative to the total number of implanted embryos when comparing ESC injected morulae to ESC injected blastocysts.

Figure 2

Validation of chimeras. A,B Three Rb1^F/F ;Trp53^F/F ESC clones (A) and two Nf2^F/F ;Trp53^F/F ;Cdkn2a^*/* ESC clones (B) were injected into C57BL/6N blastocysts and scored for their chimeric contribution. All ESC clones gave reasonable numbers of chimeric animals relative to the implanted embryos (supplementary Table S1) and, as expected, most of the chimeras were males as we exclusively used male ESC clones. male, female, n.d. C,D Comparison between chimeric contribution estimated on basis of coat-color versus genetic chimerism, tested in various tissues. Southern blot analysis was performed with a probe that distinguishes between a wild-type Trp53 allele or the floxed Trp53 allele reflecting the contribution by the host ESCs or cultured ESCs, respectively (example in supplementary Fig S3). Controls are wild-type spleen (0% chimerism expected) and F1 offspring of chimeras (50% chimerism expected). (C) Genetic chimerism of Rb1^F/F ;Trp53^F/F chimeras with coat color chimerism ranging from 70 to 100% (average 84%, n = 7). (D) Genetic chimerism of Nf2^F/F ;Trp53^F/F ;Cdkn2a^*/* chimeras with coat color chimerism ranging from 85 to 100% (average 95%, n = 4). E Survival curves of Rb1^F/F ;Trp53^F/F mice intratracheally injected with Ad5-Cre. Black line, conventional mice; red line, chimeras. F Survival curves of Nf2^F/F ;Trp53^F/F ;Cdkn2a^*/* mice intrathoracically injected with Ad5-Cre. Black line: conventional mice; Red line: chimeras. G Typical example of a neuroendocrine carcinoma (Small Cell Lung Cancer) in the lung (left panel) and a metastatic lesion in the liver (right panel). H Typical example of a mesotheliomatous lesion in the thoracic cavity. Tumor cells are either spindle sarcomatoid cells (left panel) or vacuolated epithelioid cells (right panel).

Figure 3

Genomic stability of targeted GEMM-ESC clones.

1. Comparison of chimeric contribution between the parental Rb1^F/F ;Trp53^F/F ESC clone 1.5 and three Col1a1-frt targeted derivatives. Correct targeting was confirmed by Southern blot analysis using a 3′ probe in the Col1a1 locus (supplementary Fig S4A and B). Two Col1a1-frt targeted clones, i.e. 1.5_1A10 and 1.5_1B1, provided good and germline-competent chimeras (supplementary Table S1). One chimera from the Rb1^F/F ;Trp53^F/F ESC clone 1.5_1B1 was backcrossed twice to the original strain and ESC were re-derived, i.e. clone 1.5_1B1 re-derived 4 (Table 1). This ESC clone resulted in improved chimeras compared to the parental clones. male, female, n.d.

2. Parts of whole representation of genetic aberrations observed in GEMM-ESCs cultured in 2i medium and subjected to either gene targeting, Flp-in integration and subcloning (supplementary Table S4). Last box represent the genetic aberrations observed in ESCs cultured under classic culture conditions as reported by Liang et al, .

3. Summary of CNVs observed in Rb1^F/F ;Trp53^F/F ESC clones as detected by aCGH. Two Col1a1-frt targeted clones acquired four independent CNVs. Some CNVs can be transmitted via the germ line as CNV-4.1 was maintained after backcrossing twice to the original strain, see ESC clone 1.5_1B1 re-derived 4. A detailed description of all CNVs is provided in supplementary Table S3.

Figure 4

Luciferase imaging of SCLC in chimeras.

1. In vivo imaging of a invCAG-Luc;Rb1^F/F ;Trp53^F/F chimeric mouse injected intrathoracically with Ad5-Cre. Tumor growth was monitored weekly by bioluminescence imaging.

2. Luciferase activity emitted from the thorax of 10 chimeric invCAG-Luc;Rb1^F/F ;Trp53^F/F mice. Each line represents measurements of an individual mouse. The chimeric mouse with the lowest coat-color chimerism (○, 20%) did not develop a tumor, while the second lowest chimera (□, 35%) did develop SCLC though with a long latency. One chimera (♦, 962975) failed to show any Luciferase activity but did develop SCLC. Analysis of the tumor revealed a lack of Cre-mediated switching of the invCag-Luc transgene (supplementary Fig S7).

3. Survival curves of chimeric Rb1^F/F ;Trp53^F/F mice containing either the invCag-Luc (black line) or the invCag-MycL1-Luc (red line) transgene, intratracheally injected with Ad5-Cre. Median survival indicated by the dotted line was 250 and 167 days, respectively.

4. Survival curves of F1 Rb1^F/F ;Trp53^F/F mice containing either the invCag-Luc (black line) or the invCag-MycL1-Luc (red line) transgene, intratracheally injected with Ad5-Cre. Median survival indicated by the dotted line was 235 and 140 days, respectively.

5. Luciferase activity emitted from the thorax of 11 F1 invCAG-MycL1-Luc;Rb1^F/F ;Trp53^F/F mice. Each line represents measurements of an individual mouse.

6. MycL1 copy number in SCLC tumors from three different genotypes determined by real-time PCR and aCGH. Each circle represents a primary SCLC tumor. All tumors with more than four copies (dotted line) were considered positive for MycL1 amplification. Note that overexpression of MycL1 by the transgene significantly reduces the frequency of genomic MycL1 amplifications in tumors as compared to the Rb1^F/F ;Trp53^F/F control ( P = 0.002 Fisher's Exact Test) and the invCAG-Luc;Rb1^F/F ;Trp53^F/F control ( P = 0.035 Fischer's Exact Test).

Figure 5

Efficiency of the GEMM-ESC approach.Schematic representation of the GEMM-ESC approach including the performance of the individual steps. The approach is divided in two phases: a resource phase and an experimental phase. The resource phase includes ESC's derivation and targeting with the Col1a1-frt vector, performed once per GEMM and takes ˜6 months, including the necessary quality controls. The experimental phase is mainly focused on introducing a transgene-coding plasmid in a validated GEMM-ESC clone using the Flp-in method that allows for consecutive manipulations and takes ˜4 months to obtain a chimeric cohort. Alternatively, GEMM-ESC clones are also suitable for direct targeting of a specific gene or the introduction of mutant alleles using gene editing (arrows with dotted lines). The experimental phase also includes the option to follow an F1 route as almost all GEMM-ESC clones showed germline transmission (GLT). In practice, we advise that for each model (i) multiple Col1a1-frt targeted GEMM-ESC clones are screened for their ability to efficiently generate high quality chimeras, (ii) two of the best-performing clones are selected for the Flp-in procedure, and (iii) at least two transgene-coding GEMM-ESC clones are used to generate cohorts. The final clones should originate from different Col1a1-frt targeted parental clones to minimize the chance of miss-interpreting phenotypes due to possible unwanted genetic alterations introduced by long-term culture. The selection of best-performing Col1a1-frt targeted GEMM-ESC clones is crucial for the efficiency to later generate experimental cohorts as the number of chimeras born per injected embryo is likely to decline after additional manipulations and propagation in culture.