In this project, combining high-throughput molecular biology and translational
approaches, we aim to identify the genetic and transcriptomic signatures that influence
primary sensitivity and acquired resistance to CDK4/6 inhibitors in MB. Several key
genetic events have been identified as determinants in MB. For example, MYC is among the
most frequently amplified and best characterized oncogenes in G3-MB, a highly aggressive
form of MB. PTCH1 or SUFU (negative regulators of the SHH signaling pathway) are
recurrently mutated in SHH-MB, both in the germline and somatically, supporting a causal
role for SHH signaling in these tumors. On the other hand, somatic mutations in CTNNB1
have been observed in patients with WNT-MB, leading to the activation of the WNT
signaling pathway. Mutations in the CDK4/6 pathway also play an important role in the
development of MB and are associated with poor outcomes for patients. It is important to
emphasize that both the MYC pathway and the cyclin CDK4/6 RB pathway, which are
frequently altered in MB, control the key cell fate decision to replicate the genome. For
this reason, cell cycle regulators, such as CDK4/6 inhibitors, are currently considered
attractive therapeutic targets. We have previously shown that the E3-ligase substrate
receptor and autophagy regulator AMBRA1 promotes unrestricted CDK4/6 activity and confers
resistance to palbociclib treatment in a wide range of tumors. Furthermore, our
preliminary results suggest that the genetic alterations that drive MB onset influence
the efficacy of CDK4/6 inhibitors. In this project, we aim to identify the genes and
biological processes involved in the response to CDK4/6 inhibitors and to determine their
relevance to MB.
Specific Aim 1: Genome-wide CRISPR/cas9 knockout screening of sensitivity to CDK4/6
inhibitors in MB. Recent studies have shown that cell cycle disruption is a key factor in
MB onset and relapse. However, a unifying molecular mechanism involved in this process is
still lacking. Furthermore, many of the therapeutic approaches that are under study in
clinical trials lack solid molecular bases for systematic patient stratification. To fill
this knowledge gap, we propose to develop an unbiased approach to identify genes that
influence sensitivity to CDK4/6 inhibitors (palbociclib and abemaciclib). This approach
will also help us better understand the regulatory networks that control progression and
aggressiveness. We will use CRISPR/Cas9 technology to perform a genome-wide knockout (KO)
screen. To this end, we will use the Human GeCKO v2 Pooled Knockout CRISPR library in MB
cell lines expressing Cas9 ONS76 (SHH-MB), D283-MED (G3-MB) and HD-MB03 (G3-MB). These
cell lines are representative of two distinct MB subgroups and also share p53 competence.
This feature is relevant in the context of CDK4/6 inhibitors, because their efficacy
strictly depends on the ability of p53 to induce senescence and cell cycle arrest. Cells
will be treated or not with palbociclib and abemaciclib and maintained in log-growth
phase at a 1000x coverage for sgRNA throughout the screen. The doses of palbociclib (0.07
μM) and abemaciclib (0.19 μM) were chosen as those that cause a 50% reduction in
proliferation rate by cell count assays. After 14 days of treatment, genomic DNA will be
extracted from each population and sgRNA sequencing libraries will be prepared.
Identified genes will be ranked based on the protective effect that their KO exerts in
both treated conditions. Subsequently, we will further characterize the top 5
non-redundant genes of interest (GOIs) that confer a growth advantage by biochemical and
molecular biology approaches. In detail, we will assess the ability of GOIs to modulate
the cell cycle by quantitative image-based cytometry (QIBC) and flow cytometry, in basal
condition or upon CDK4/6 inhibition. To extend our results, we will validate the selected
GOIs in other MB cell lines that we routinely use for cell biology assays (DAOY,
D341-MED, CHLA-MED-01). Furthermore, we will implement our analysis by exploiting two
cell lines (DAOY and ONS76) that we have already engineered to express two fluorescent
reporters: FUCCI CA and DHB mVenere-p2a-mCherry-CDK4KTR. The former is a sophisticated
technology that discriminates between cell cycle phases in a spatio-temporal manner using
a double color combination. The latter is a reporter system that allows simultaneous
monitoring of CDK4/6 and CDK2 activities in single cells. GOIs will be deleted in these
cell lines by CRISPR/Cas9 technology and the effect of their ablation on cell growth and
CDK4/6 inhibition will be studied. To gain mechanistic insights into how GOIs confer
resistance to CDK4/6 inhibitors and to evaluate whether they enable acquired resistance,
we will use an orthogonal approach to the CRISPR/Cas9 screen. We will develop in vitro
models of acquired resistance by treating our panel of MB cells with abemaciclib or
palbociclib. We will then perform RNA-seq to investigate whether the GOIs, or the
pathways to which they belong, are modulated in resistant cells. Furthermore, since
altering pre-mRNA processing regulation by treatment with an arginine methyltransferase
PRMT5 inhibitor (GSK3326595) has recently been shown to restore sensitivity to
palbociclib in tumor cells that have acquired resistance to this drug, we will test the
combination of CDK4/6 inhibitors and PRMT5 in resistant MB cells.
Specific Aim 2: To establish ex-vivo translational models of MB, we will develop
patient-derived organoids (PDOs) from primary surgical biopsies of patients. To this end,
the Organoid Research Core Facility available at the P.I.'s Research Entity will be
instrumental. Surgical biopsies will be enzymatically and mechanically digested and the
tumor cell suspension (1x106 cells) will be mixed with Matrigel, allowed to assemble for
15 minutes at 37 °C and then seeded in tissue culture plates. Tumor organoids will be
allowed to form for 4 days in culture in a humidified incubator at 37 °C. At the end of
this incubation, tumor organoids will be transferred to an orbital shaker and will be
cultured under agitation conditions (70 rpm) until passage for experimental assays. Once
established, SHH-MB and G3-MB PDOs will be tested for their ability to capture the
oncogenic characteristics of the original tumor. First, we will perform parallel
immunohistochemical characterization of MB markers currently used for subtype
specification in clinical practice. Subsequently, MB PDOs and original tissues will be
sequenced to determine the mutational status of >500 genes frequently altered in human
tumors using the TSO-500 (Illumina), as recently published at our Organoids Facility.
Only PDOs that recapitulate the histological and genetic characteristics of the original
tumor will be further propagated and used for functional studies. We will then evaluate
the effects of CDK4/6 inhibitors on selected MB PDOs. Since these cell models grow in 3D
and not all cells are directly accessible to the compounds delivered in the medium, we
will first assess the cellular uptake and localization of palbociclib and abemaciclib.
Since these compounds can be accumulated in lysosomes and other acidic cellular
compartments, we will analyze the cellular uptake of CDK4/6 inhibitors in PDOs using the
fluorescent properties of these drugs (~500 nm when excited at 405 nm light) in
combination with fluorescent dyes used to label organelles in vivo (e.g., lysotracker,
mitotracker). Subsequently, the impact of drug localization will be assessed by
determining the proliferation, survival and senescence of PDOs in response to selected
doses of CDK4/6 inhibitors using methods already in use at our facility. Furthermore,
since the establishment and maintenance of PDOs depend on the presence of stem cells and
stemness contributes to chemoresistance in MB, we will evaluate the effects of treatments
on PDO stemness/differentiation characteristics by PCR and western blot analysis of
previously established markers (CD133, Nestin, SOX1/2 for stemness; GFAP, CD44 for
astrocytes; CD24, ßIII-tubulin for neuronal progenitors and neurons, respectively). PDOs
sensitive to these drugs will then be cultured with suboptimal doses of CDK4/6 inhibitors
for 2 months in order to select clones that acquire resistance to the treatments. RNA
sequencing analyses of parental and CDK4/6-resistant PDOs will be performed and compared
with the results obtained with MB cell lines (Specific Aim 1). Resistant PDOs will then
be treated with CDK4/6 inhibitors with or without the PRMT5 inhibitor. Furthermore, the
GOIs selected in Specific Aim 1 will be silenced in resistant PDOs by electroporation of
sequence-specific siRNAs or CRISPR/Cas9 sgRNAs and the effects on sensitivity to CDK4/6
inhibitors will be evaluated. These studies will validate the functional relevance of
GOIs and PRMT5 inhibition in a clinically relevant disease model.
Specific Aim 3: Development of innovative technologies for the evaluation of new clinical
biomarkers for diagnosis and prognosis. To investigate the correlation between the GOIs
selected in Specific Aims 1 and 2 and clinical parameters, we will analyze their
expression in histological samples from MB patients. To this end, we will use the tissue
slides from MB patients (n=60) that are stored at the UO1 Hospital and for which clinical
follow-up is available. To date, immunohistochemistry (IHC) is the only clinically
validated and commercially available approach to detect the distribution and amount of
proteins in tissues by specific antigen-antibody reactions. However, IHC suffers from
some drawbacks when new genes are under investigation. First, antibody production
requires time-consuming processes that can lead to high analysis costs. Second, IHC
involves several steps that affect the turnaround time of the technique. To overcome
these limitations, we propose an alternative approach based on the use of synthetic
DNA-based nanoswitches. More specifically, we will develop and characterize DNA
nanoswitches capable of responding to specific proteins through a conformational change
mechanism. Nanoswitches are composed of single-stranded DNA molecules (20-40 nucleotides)
that serve as molecular scaffolds for the conjugation of different recognitions (protein
epitopes, antibodies, aptamers) and signaling labels (fluorophore, quencher tag). The
recognition will confer to the nanoswitch the ability to specifically bind to the protein
of interest. The fluorophore/quencher pair will instead allow the signaling of the
binding event. Nanoswitches are rationally designed to undergo a change upon binding to
the protein of interest. This conformational change will force the fluorophore to move
away from the quencher, leading to an increase in signal. This approach offers several
advantages. Nanoswitches provide a signal only upon binding to the target protein, no
washing and reaction steps are required. Nanoswitches are also versatile and can be
labeled with different fluorophores, allowing for the orthogonal detection of several
proteins simultaneously in the same sample.
In summary, this project aims to thoroughly investigate the role of CDK4/6 inhibitors in
medulloblastoma (MB) treatment. By combining high-throughput molecular biology techniques
with translational approaches, we will identify genetic and transcriptomic signatures
that influence sensitivity and resistance to these inhibitors. We will also develop and
utilize patient-derived organoids (PDOs) to validate our findings in a clinically
relevant context. Furthermore, we will create innovative diagnostic tools using DNA
nanoswitches to improve the detection of key biomarkers in MB. These studies have the
potential to significantly advance our understanding of MB biology and pave the way for
more effective and personalized treatment strategies.
