1 Introduction
Retinoblastoma is the most common intraocular childhood cancer and originates in the neural retina [
1]. The majority of retinoblastomas are caused by bi-allelic
RB1 inactivation [
2]. However, in a subset of cases (< 5%)
RB1 is proficient and instead other oncogenic mutations such as copy-number amplifying mutations of the
MYCN gene (
MYCNA) drive the carcinogenesis [
3,
4]. Such retinoblastomas are formed and diagnosed earlier, have faster and more aggressive growth, are less differentiated and more prone for metastasis than
RB1-deficient retinoblastomas without
MYCNA [
3,
5,
6]. The
MYCNA amplified
RB1-proficient retinoblastoma tumours also have distinct molecular signatures from
RB1-deficient tumours [
5]. Tumorigenic growth can be induced with high frequency when
MYCN is over-expressed in
RB1-proficient human retinal organoids as well as in embryonic chicken retina in vivo [
7]. In both systems, the tumorigenic cells are anaplastic and express markers for active proliferation and undifferentiated cone photoreceptors (cPR) [
7]. Chickens that were hatched with stable retinal
MYCN expression, formed retinal tumours with metastatic growth that infiltrated the sclera and optic nerve and formed extraocular tumours within 6–9 weeks. The tumours had round proliferating cells, lacked Flexner-Wintersteiner rosettes but contained necrotic regions. The potential for the neoplastic growth in chicken retina is embryonic stage-dependent and feature a cell-specific resistance to apoptosis. The tumours express genes associated to the cone/horizontal cell lineage, but not to ganglion or amacrine cells [
7]. Moreover, lentivirus-mediated over-expression of
MYCN in human
RB1-proficient foetal retina also induces tumorigenic growth resembling retinoblastoma [
8]. The cellular origin lies within the undifferentiated cPR lineage. The cell-of-origin for the
RB1-proficient tumours as shown in Blixt et al. 2022 and Singh et al. 2022 resides in the same cell type-lineage as the
RB1-deficient tumours in the cPR lineage as shown in several independent models. However, the
MYCN-induced
RB1-proficient phenotype was consistently more immature and less differentiated with faster tumour progression than that of the
RB1-deficient [
5,
7‐
11]. The
MYCN-transformed human foetal retina formed tumours that were not fully constrained to immature cPRs but also to a less extent developed expression of markers for e.g. ganglion cells [
8]. The similarities and differences in the phenotype between the modelled variants are consistent with that seen in patients with
MYCNA retinoblastomas regardless of
RB1 status [
3,
5].
In vivo MYCN-transformed chicken retinal cells grow
in vitro as aggregates in suspension, similar to the way of established retinoblastoma lines such as WERI-RB1 and Y79. In this work we have analysed the transcriptome of established chicken retinoblastoma cells with stable
MYCN over-expression (henceforth referred to as “DMC cells”) taken acutely (“young” DMC cells) and after culturing for more than 200 days (“old” DMC cells). Cells with over-expression of
MYCN with the stabilizing and potentiating oncogenic mutation T58A were also analysed [
12].
The retinal cell-of-origin of these transformed chick retinal cells, the cone/horizontal cell progenitor, do not exhibit developmental cell death [
13,
14]. The immediate horizontal progenitor withstands DNA-damage during the terminal cell cycle and escapes both cell cycle arrest and apoptosis to continue the final cycle into differentiation [
15‐
19]. Such natural death-resistance has been suggested to be associated with high
MDM2 expression that increases susceptibility to neoplastic transformation [
15,
20]. Over-expression of
MYCN in the progenitor cells is sufficient to drive neoplastic growth into retinoblastoma and the resistance to apoptosis in the cone/horizontal cell lineage contributes to the tumorigenic phenotype [
7].
In this work we sequenced the transcriptome of
in vivo MYCN-transformed chicken cells that have been cultured and the expression profiles contained both enriched signatures of the
RB1-
E2F-axis and p53 signalling pathways. The levels of
RB1 mRNA were not altered but the levels of
E2F1 and
E2F3 mRNAs were 10-fold increased.
E2F1 and
E2F3 are among the many cell-cycle regulatory genes directly or indirectly induced by Myc-proteins [
21‐
23]. Inhibitory hyperphosphorylation of the retinoblastoma protein (Rb) in
RB1-proficient retinoblastomas has been proposed to explain the
RB1-deficient-like phenotype of
RB1-proficient retinoblastomas [
24]. The expression profile with high levels of
E2F mRNA as shown in this work, opens up for a hypothesis that the increased levels of
E2F1 and
E2F3 expression may outcompete the inhibitory capacity of Rb and render a functionally
RB1 deficient phenotype. We tested this hypothesis by blocking E2fs using the small molecular inhibitor HLM006474, which normalized the cell cycle behaviour and induced cell death in the
MYCN over-expressing cells. Moreover, blocking Cdk4/6 with Palbociclib also contributed to a partial normalization by arresting the cell cycle, supporting a partially proficient
RB1 status in the cells. The cells were arrested but did not die, and when promoting p53 using Nutlin-3a the cells did not increase apoptosis, consistent with the intrinsic p53 insensitivity of the cell-of-origin lineage. Taken together, this paper contributes to the understanding of regulatory events in
RB1-proficient
MYCN-overexpressing retinoblastoma. Furthermore, the study provides an explanation to how
MYCN may promote proliferation and carcinogenesis in this retinoblastoma model of retinoblastoma tumorigenesis.
2 Materials and methods
2.1 Animals
Fertilised White Leghorn eggs (Gallus gallus; Håtunalab AB, Bro, Sweden), were incubated at 37 °C in a humidified incubator (8204/MP, Grumbach, Asslar, Germany). Embryonic age (E) was staged according to the Hamburger and Hamilton 1951 stages (st) [
25]. Animal experiments were carried out in compliance with the guidelines set by the Association for Research in Vision and Ophthalmology and were approved by the regional animal ethics committee in Uppsala, Sweden (Dnr C90/16, C159/15, 5.8.18–09718/2021).
2.2 In ovo electroporation of retina
Fertilized chicken eggs were windowed at st22/E3.5. Embryos were staged and electroporated with piggyBac integration expression vectors with human
MYCN, or
MYCN-T58A, together with green fluorescent protein (GFP) [
7]. The air sac-chorioallantois boarder was marked, and a thin forceps tip was inserted in the air sac 1 mm from the mark. The inside of the egg was gently scraped in order to shift the air sac over the embryo. A piece of surgical tape was placed on the shell over the embryo and a 1 × 1 cm window was opened, producing a re-sealable “hatch”, to expose the embryo. The vitelline and chorioallantois/amniotic membranes were gently peeled open to expose the right eye of the embryo. 0.2 µl solution was injected into the subretinal space of the central retina using a capillary and mouth pipetting. 1–2 µg/µl per vector in 1x PBS with Ca
2+ and Mg
2+ “PBS+/+” (18912-014, Gibco, Waltham, MA, USA) and 0.02% Fast Green (F7252, Sigma-Aldrich, St. Louis, MO, USA). Fast green was added for visualization during injection. Injection was done dorsally next to the temporal posterior ciliary artery on the border of the eye and the prospective brain. The negative electrode was placed behind the central region of the eye, next to the injection, and the positive electrode was placed in front of the prospective beak. Five 50 ms pulses of 15 V with 1 s. intervals were applied using an ECM 830 Square pulse electroporator (BTX Harvard Apparatus, Holliston, MA, USA). 100 µl of Ringer’s solution was applied to the electroporated eye. The window-hatch was closed and sealed with surgical tape, and the egg incubated for continued development [
7].
2.3 Cell cultures
GFP positive (+) cells express MYCN and GFP + regions of successfully MYCN-electroporated st40/E14 retina was dissected, cells were dissociated and cultured in RPMI1640 (21875-034, Gibco) and supplemented with 10% FBS (16000044, Gibco), 1% MEM-NEAA (11140035, Gibco) and 1% pen-strep antibiotic mix (15140122, Gibco). Cells were counted by using a Countess 3 Automated Cell Counter (Invitrogen, MA, USA).
2.4 RNA-sequencing
Cells from three acutely established primary MYCN-cell lines (denoted “young” DMC) from E14 retinas electroporated at E3.5 and cultured for 14 days, three “old” DMC cell lines cultured for > 200 days and three “old” T58A MYCN cell lines (> 200 days in culture) were triturated and used for RNA preparation. Cells from unelectroporated E14 retina were used as control and reference. RNA was extracted using the Qiagen RNeasy Micro Kit (74,004, Qiagen, MD, USA) according to the manufacturers protocol and RNA integrity was determined on a TapeStation (Agilent, CA, USA). RIN value > 8. Libraries were prepared with the TruSeq Stranded mRNA protocol using polyA-selection and sequenced on a SP1 flow cell on Illumina NoveSeq 6000 (Illumina, CA, USA).
2.5 Sequencing data analysis
Analysis was performed at the National Bioinformatics Infrastructure Sweden. In brief, samples were checked for rRNA contamination with “bbduk” from BBMap (version 38.61) [
26]. QC and alignment of the data was performed by nf-core/rnaseq pipeline (version 3.4) by adjusting the alignment parameter “alignEnsProtrude 100 ConcordantPair”. The first 10 base pairs of reads showed biased base composition and were removed. For QC, MultiQC [
27] and FastQC [
28] results reported by nf-core pipeline were assessed. To extract fragment counts, featurecounts (version 2.0.0) [
29] was used with minimum mapping quality of 20 and a requirement of both pairs to be properly aligned on the same chromosome. Pairwise comparison was performed using edgeR [
30] with pairwise comparison (exactTest). Significant differentially expressed genes (DEGs) were selected by two criteria: (1) corrected p-value for multiple testing ≤ 0.05 (FDR by the Benjamini-Hochberg procedure) and (2) ∣log2 fold-change∣
≥ 1.
For gene ontology (GO) terms and Kyoto Encyclopaedia of Genes and Genomes (KEGG) pathway enrichment analysis, ClusterProfiler (version 4.0.6) [
31] was used on all DEGs. GO over-representation analysis was performed by “enrichGO” and GO gene set enrichment analysis was performed by “gseGO”. KEGG over-representation analysis was done with “enrichKEGG” while KEGG gene set enrichment analysis was done by “gseKEGG”. Pathway figures were obtained by Pathview (version 1.32.0). Differential expression analysis and visualization was performed in R (version 4.1.1) (R Core Team 2021) [
32].
2.6 Data availability
RNA expression data are deposited as GSE226458 at the NCBI Gene Expression Omnibus.
2.7 Chemical treatments
1–2 × 106 cells were incubated with the reagent for 24 h before analysis. The reagents targeted and inhibited MDM2 (Nutlin-3a, SML0580, Sigma-Aldrich), p53 (Pifithrin-α, P4359, Sigma-Aldrich), Cdk4/6 (Palbociclib, PZ0383, Sigma-Aldrich), E2F (HLM006474, SML1260, Sigma-Aldrich) and caused DNA damage (Cisplatin, 2251/50, Tocris, Abingdon, UK). Reagents were dissolved in DMSO, except Cisplatin that was dissolved in water, and vehicle controls consisted of an equivalent concentration of DMSO.
2.8 Cell viability assay by Trypan Blue
1.5 × 106 cells were seeded into each well of 6-well ULA plates (3471, Corning). At each time point, 200 µL cells was dissociated by gentle pipetting, filtered through 20 μm cell strainer (43-10020-60, pluriSelect, Leipzig, Germany) and mixed 1:1 with Trypan Blue (15250061, Gibco). Viability was analyzed by using a Countess 3 Automated Cell Counter (Invitrogen).
2.9 Cell cycle analysis
Cells were centrifuged at 300 rcf for 5 min, the medium aspirated, and the cells incubated in 1X PBS-/- (14190094, Gibco). Single-cell suspensions were obtained by gentle trituration, one wash in PBS-/-, and filtration through a 20 μm cell strainer (43-10020-60, pluriSelect, Leipzig, Germany). 1 × 106 cells were used per sample. The samples were washed once with PBS-/- and resuspended in 500 µl PBS-/-. To fix GFP, 500 µl of 2% PFA was added to each sample, followed by incubation at 4 °C for 1 h. The samples were centrifuged at 300 rcf for 5 min, the supernatant aspirated, and the samples washed once in PBS-/-. One ml of ice cold 70% ethanol was added dropwise under agitation and the samples were incubated at 4 °C overnight. The ethanol was aspirated following centrifugation at 1000 rcf for 5 min. The samples were resuspended in propidium iodide working solution (0.1% TritonX-100, 10 µg/ml PI; P4864, Sigma-Aldrich), and 100 µg/ml DNase-free RNase A (11119915001, Sigma-Aldrich) in PBS-/- and incubated at RT for 30 min before analysed with an easyCyte 8 Flow cytometer (Luminex Corp, the Netherlands). Aggregates or doublets were excluded from analysis. Fixed, non-electroporated retinal cells were used as a negative control to set the threshold for a positive signal. PI + or GFP + cells were gated after compensation with single-color controls, which were GFP + cells in culture and non-electroporated retinal cells with PI staining.
2.10 Western blot analysis
Retinas and cells were homogenized in RIPA buffer (89900, Thermo Fisher Scientific) containing Halt Protease and Phosphatase Inhibitor Cocktail (78440, Thermo Fisher Scientific). The total protein concentration was measured by Dc Protein Assay kit (5000112, Bio-rad, Laboratories AB, Hercules, CA, USA). Each sample was normalized to actin level. Primary and secondary antibodies are listed in the Supplementary Table
S1. The Western blot analysis was performed as previously described [
33] and according to the manufacturer’s instructions.
2.11 Quantitative reverse transcriptase PCR analysis
RNA was extracted with TRIzol (15596026, Thermo Fischer Scientific). The samples were DNase-treated (M6101, Promega, Madison, WI, USA) and cDNA was synthesized with the High-capacity RNA-to-cDNA kit (4387406, Thermo Fisher Scientific). Tests were run in triplicates using IQ™ SYBR® green Supermix (1708882, Bio-Rad) and the Ct values were normalized to β-actin and TATA box-binding protein (TBP). Control reactions containing Supermix and primers but no cDNA were run in parallel. PCR program consisted of initial denaturation step at 95 °C for 3 min, followed by 39 cycles of denaturation at 95 °C for 15 s, and annealing and extension at 60 °C for 30 s. Melt curve analysis was performed to confirm the presence of a single product. Primers were designed with Primer Express v2.0 (Applied Biosystems, Darmstadt, Germany). For primers, see Supplementary Table
S2.
2.12 Alamar blue proliferation assay
5 × 104 cells were transferred to each well in 96-well ULA plates (3474, Corning, Arizona, USA). Chemical treatments were for 72 h. Alamar Blue (DAL1025, Thermo Fisher Scientific) was added 1:100 and incubated for 6 h before plate reading. The absorbance was read by ClarioStar Plate reader (BMG Labtech, Ortenberg, Germany) and the data were processed by Mars Software (BMG Labtech). Each experiment was repeated 3 times or more.
2.13 Histo- and immunocytochemistry
Dissected chicken eyes were fixed with 4% PFA, embedded and cryo-sectioned (10 μm sections) for immunohistochemistry (IHC). For immunocytochemistry, cells were in transferred to chambered cell culture slides (354104, Corning) coated with Poly-D-Lysine (A3890401, Thermo Fisher Scientific). Suspension cells were seeded in the slides and left for sedimentation for 30 min in the incubator at 37 °C. The slides were then centrifuged at 1000 rcf for 5 min to attach the cells. Cells were washed twice with pre-warmed PBS+/+ and fixed with 4% PFA at RT for 15 min and washed twice with 1X PBS. The permeabilization was performed by incubation with 0.2% TritonX-100 in PBS for 20 min at room temperature and washed twice with 1X PBS. After permeabilization, the blocking and antibody incubation were the same as immunohistochemistry. TUNEL staining was performed with Click-iT TUNEL assay according to manufacturer’s instruction (C10619, Thermo Fisher Scientific). For antibodies, see Supplementary Table
S1.
2.14 Microscopy and Image analysis
Fluorescence micrographs were captured using a Zeiss Imager Z2 microscope (Carl Zeiss Microscopy GmbH, Jena, Germany). Quantification of TUNEL stained cells was assisted by the software Fiji-ImageJ2. Contrast of fluorescence images was enhanced at the microscope using the Zeiss capture and image analysis software (ZEN).
2.15 Data analysis
Data were analysed with one-way ANOVA followed by Tukey’s multiple comparison post-hoc test or Student’s t test using GraphPad Prism (GraphPad Software Inc. San Diego, CA, USA) and statistical significance was set to p < 0.05. Statistical analysis and number of replicates are presented in the figure legends.
4 Discussion
In this work, we have studied in vivo-induced retinoblastoma tumour cell lines with over-expression of
MYCN. The tumours were generated in normal chicken embryonic retina and represent an early established type of
RB1 proficient
MYCNA retinoblastoma cancer [
7]. The expression profiles show that the tumour cells have a clear cPR signature together with a wide spectrum of
MYCN-induced genes. This profile is consistent with an origin of the tumour in the cPR lineage, which is similar to the cell-of-origin for the
RB1-deficient cancers [
15]. The profile revealed that the expression of several of the activating
E2F gene family members were upregulated. The activity of E2fs is directly regulated by the Rb protein and in spite of
RB1 proficiency our results suggest that the increased levels of E2f contribute to a dysfunctional cell cycle regulation and hence to the cancer phenotype. Our results also suggest that the cells are to some degree resistant to p53 activation. Such resistance is also consistent with the properties and cellular origin of a cell in the cPR lineage.
The cell lines were established directly after the initial signs of tumour formation in the E14 embryonic retina and thus represents an early pristine form of this type of retinoblastoma. Previous research has suggested the existence of two types of retinoblastomas based on their genetics and pathology. Type 1 tumours harbour few genetic alterations other than the initiating
RB1-inactivating mutations and corresponds to differentiated tumours expressing mature cone markers. By contrast, type 2 tumours harbour frequently recurrent genetic alterations including
MYCN-amplifications. They express markers of less differentiated cones together with other neuronal and ganglion cell markers, with marked tumour heterogeneity. The undifferentiated phenotype in type 2 is associated with stemness features including low immune and interferon response,
E2F and
MYCN activation and a high propensity for metastasis [
5]. The tumour cells established in this work represent a tumour with
MYCN over-expression but without
RB1 inactivation or other genetic lesions. The phenotype is anaplastic and aggressive as shown by optic nerve infiltration and extraocular growth [
7]. The clear cPR signature of phototransduction gene expression is accompanied with genes for cPR progenitors such as
RXRG and
THRB, indicative of progenitors or immature cPRs reflecting the corresponding early events during carcinogenesis. The majority of the top enriched, down regulated GO terms are related to neuronal development. None of the DMC cells display any overt ganglion cell markers as several of the type 2 cancers do [
5], or that is seen in tumours after MYCN expression in foetal human retina transduced in vitro and xenografted to mice [
8]. This may be due to the pristine state of the chicken “young” DMC cells taken only a day after tumour formation and few days in culture. The cells have then not been under influence of
MYCN for the extended period after transformation which is the case for patient-derived or xenografted experimental tumours. The “old” DMC cells that had been in culture for more than 200 days and the “young” DMC cells were very similar with relatively few DE genes and without any significant enriched GO-terms or pathways. However, when looking at individual DE genes, a trend with lower expression of cPR genes such as
ARR3, RXRG, THRB as well as several phototransduction genes, was seen in the “old” (MYCN and T58A) cells. This would suggest that prolonged actions of
MYCN in the tumour cells may facilitate dedifferentiation of the original cPR phenotype. Genes related to early retinal development, such as
IRX3, was also higher in the “old” cells. Species differences may also contribute to the differences and the stage of cPR development in relation to the effective time of transformation. The chicken retina develops fast and cPR are early formed which may render a relatively advanced stage of cPR at transformation.
The DMC cells can be established with high frequency (100% of successful
in ovo electroporations) from dissected embryonic retinal tumours and the
MYCN tumour cells grow in suspension with concomitant cell death similar to patient derived
MYCN-retinoblastoma lines [
36,
42]. The transcriptional profile of the chicken tumour cells has a clear cone photoreceptor signature in addition to broader signatures with increased biosynthesis and cell cycle regulation while that of neuronal development, differentiation and signalling is decreased, compared to E14 retina. Such effects on the expression profile can be expected from the impact of
MYCN over-expression.
MYC family members are known to drive proliferation and stem cell-ness at the expense of a differentiated cell phenotype with upregulation of cell cycle regulators such as CDKs and E2fs [
21]. Upregulation of these genes has also been found in different retinoblastoma cell lines [
53]. While the
E2F family members were upregulated, the
RB1 expression was not changed in the DMC-cells and we confirmed presence of Rb-protein using immunocytochemistry for Rb IR as well as for phospho-S608-Rb. Phosphorylation of Rb on S608 was blocked by the Cdk4/6 inhibitor Palbociclib (Fig.
1). Furthermore, Palbociclib had a significant but limited effect on proliferation and the cell cycle of DMC cells, leading to more cells in G1-phase, which is significative for increased number of cells in G1-arrest as a result of reduced phosphorylation of Rb. Such effects corroborate that the DMC cells have functional Rb protein, which is phosphorylated by Cdk4/6 and that the cells in fact are
RB1 proficient.
RB1 mutations and gene loss have been shown to induce resistance to Palbociclib in breast cancer cells [
54]. DMC cells do not display resistance to Palbociclib, which support that the DMC cells are
RB1 proficient.
Upon phosphorylation of Rb, E2f1 is disinhibited and promotes transcription of factors that transit cells into S-phase. As long as Rb is unphosphorylated and the pool of Rb is sufficient to inhibit the E2fs, the cell cycle will be regulated. With a large pool of E2fs, as seen in the DMC-tumour cells, Rb will not be able to fully inhibit E2f in the cell, leading to uncontrolled proliferation. A block downstream of E2f will, however, reduce proliferation and we showed that HLM006474 led to decreased proliferation and increased G1-phase arrest. Addition of increasing concentrations of Palbociclib with HLM006474 did not increase the effect while when increasing the concentration of HLM006474 the inhibitory effects were significantly higher. Interestingly, prolonged treatment with HLM006474 induced cell cycle arrest and cell death (Fig.
5L-M), which is consistent with that E2fs and not Rb constitutes the rate limiting step in the regulatory pathway of the DMC cells.
As already introduced, the tumour cells are derived from a retinal progenitor that do not exhibit the extent of developmental cell death as seen in other retinal cell types [
13,
14], it withstands DNA-damage and escapes cell cycle arrest and apoptosis [
16‐
19]. Such resistance against p53-induced death was suggested to be a result of insensitivity to modulation of the p53 activity by the MDM2 inhibitor Nutlin-3a, the p53 coactivator Zac1 and p53 inhibitor pifithrin-a [
17,
19]. Insensitivity to Nutlin-3a and pifithrin-a was replicated in the DMC cells implying that such p53 resistance is remaining after transformation (Fig.
4). Furthermore, an aberrant p53 pathway regulation was identified as top-ranked in the enriched activated pathways (Fig.
3I) and together this implies that p53 is dysregulated, which may contribute to the tumorigenic phenotype. An intrinsic p53 insensitivity, rather than effects of
TP53 mutations, is consistent with that silencing mutations in
TP53 are seldom found in retinoblastoma [
55]. Altered p53 functions in these cells is also consistent with the absence of upregulation of
CDKN1A mRNA after Nutlin-3a or pifithrin-a treatments (Fig.
4I). However, the intrinsic p53 insensitivity seams not to give cells resistance to apoptosis as shown by
CDKN1A upregulation, cell cycle arrest and extensive apoptosis after treatment with HLM006474 (Fig.
5M and N). It is not clear where in the regulatory pathways the insensititvity resides. We do not find upregulation of
MDM2 mRNA but robust expression of
CDKN2A. Human cPR precursors express high levels of
MDM2 and patient-derived retinoblastoma cells require functional
MDM2 for survival which also suppresses
CDKN2A/p14ARF-induced apoptosis in cultured human retinoblastoma cells [
15]. Furthermore, there is an extensive literature on blocking Mdm2 and its effects on retinoblastoma (reviewed in Laurie et al. 2007) [
56].
MDM2 has been shown to promote proliferation in both neuroblastoma and retinoblastoma cells through p53-independent regulation of
MYCN [
57,
58]. High transgene
hMYCN expression is likely making this pathway redundant. Noteworthy, we see
CDKN1A upregulation after blocking E2f, which also is consistent with a potential p53-independent regulatory pathway (Fig.
5F).
In conclusion, the results from this work implies that
MYCN-induced E2f-expression, contributes to a dysfunctional cell cycle regulation. In spite of
RB1-proficiency, the increased levels of E2f cause a cancer phenotype that resembles that of a
RB1-deficient retinoblastoma. The expression profiles show that these tumour cells have a clear cone photoreceptor signature together with a wide spectrum of MYCN-induced genes. Such profile is consistent with an origin of the tumour in the cone progenitor lineage and is similar to the cell-of-origin for
RB1-deficient retinoblastomas. Our results also suggest that the cells are to some degree insensitive to p53 activation, which is consistent with the properties and cellular origin of a cell in the cone progenitor lineage. The results, based on the chicken as a model show that the mechanism for retinoblastoma is among several evolutionary conserved pathogenetic mechanisms for human disease [
59]. The implications of these results are that targeting the Cdk4/6 - E2f signalling pathway may be a relevant and complementary regime for retinoblastoma with
MYCNAs. In
RB1 proficient tumours with elevated
E2F expression, it may not be sufficient to target Cdk4/6 with Palbociclib but also target the E2fs.
MYCN mutations is a known aggravating factor in paediatric cancers and defines severe subtypes of medulloblastoma and neuroblastoma.
MYCN is therefore an attractive drug target, however, direct targeting of MYCN has proven difficult [
60]. Targeting key
MYCN down-stream effectors such as elevated E2fs may be a feasible way to treat those cancers including retinoblastoma with
MYCNA.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.