Researchers at Greehey Children’s Cancer Research Institute Greehey CCRI have contributed more novel therapies in national and international pediatric trials than most groups in the US.
Despite playing critical roles in the central dogma of molecular biology, the importance of RNA has been only recently appreciated. Spurred on by the seminal findings of Tom Gingeras in the early 2000s that entire genomes can be transcribed into RNA and that the majority of RNA transcripts are not destined to be translated into proteins, the subsequent discovery of numerous new classes of RNAs; ranging from the small (miRNAs) to the large (lncRNAs), and the characterisation of their functions across the spectrum of biological processes, the study of RNA has revolutionised modern biology. Key areas of research within the Greehey Children’s Cancer Research Institute include establishing the role of RNA metabolism in paediatric cancers, predicting and validating RNA-RNA interactions relevant to paediatric cancers, and developing and evaluating RNA-based and RNA-targeted therapies for treating children’s cancers. The goal is to capitalise on the solid foundation of the Greehey CCRI in preclinical models and drug development to understand the role of RNA biology in health and disease and apply that understanding to improve patient outcomes.
RNAs represent the next generation of therapeutic approaches
Non-coding RNAs, including microRNAs (miRNAs) and long non-coding RNAs (lncRNAs) are endogenously expressed RNAs that have been demonstrated to play critical roles in cancer growth, metastasis, and drug resistance. miRNAs regulate gene expression at the post-transcriptional level through degradation of their target messenger RNAs (mRNAs) or inhibition of their translation. lncRNAs interact with and regulate coding and non-coding transcripts, including miRNAs. RNA binding proteins (RBPs) bridge the RNA and protein worlds by allowing proteins to control RNA function and vice versa. Making sense of this complex set of interleaved interactions requires in vitro, in vivo, and in silico approaches.
The ability to either increase or decrease the intratumor levels of specific miRNAs through multiple methods has provided an opportunity to develop miRNA-based therapeutic strategies. Several miRNAs have progressed to phase I clinical trials for treating adult cancers. Our laboratories at Greehey CCRI have undertaken concerted efforts to understand the role of miRNAs, lncRNAs, and RBPs in paediatric cancers and to explore the possibility of using RNA-based therapeutics for treatment. Their strategies include chemical modification of RNAs and encapsulation in delivery vehicles to improve stability and protect against degradation and prediction of targets to identify the on-target and off-target cellular pathways likely to be disrupted.
Identification of interactions between RNA species can elucidate function
Identifying the interactions between miRNAs, lncRNAs and mRNAs is a critical part of understanding how their therapeutic effects are achieved and the likely consequences of manipulating intracellular levels of miRNAs, lncRNAs, or RBPs in both tumour and normal cells. RBP interactions are identified experimentally through cross-linking and immunoprecipitation, the targets of miRNAs that mediate their effects are identified both indirectly, through measurements of differential expression after cells are treated with one or more miRNAs and identification of motifs that contribute most to the observed changes in transcript expression, and directly, through a method that links miRNAs and their targets within the RISC complex and employs sequencing of the resulting chimeric sequences to identify interactions between miRNAs and their mRNA targets. Efforts in this area are advanced by our genome sequencing facility, which operates state of the art genomic platforms to generate high-quality data across multiple sequencing methodologies and our computational biology and bioinformatics Initiative, which provides computational infrastructure and bioinformatics expertise to manage, distribute and analyse large and complex datasets.
Chemical modification confers stability and resistance to degradation
Between miRNAs, lncRNAs, and RBPs, miRNAs are the most straightforward to manipulate, but also the most susceptible to degradation. Stability and binding affinity of the interaction between a miRNA and a target mRNA can be enhanced by several different chemical modifications, including phosphorothioate-containing oligonucleotides, 2′-O-methyl- (2’-O-Me) or 2′-O-methoxyethyl-
oligonucleotides (2’-O-MOE) and locked nucleic acid (LNA) oligonucleotides. Phosphorothioate modifications replace one of the non-bridging oxygens in the phosphate group by sulphur increasing nuclease resistance, uptake, absorption, and distribution. 2’-O-Me and 2’-O-MOE groups on the ribose moiety increase binding stability and oligoribonucleotides from nuclease degradation. LNAs lock the ribose moiety by a bridge connecting the 2′-oxygen and 4′-carbon in an N-type (C3′-endo) conformation, improving hybridisation affinity, mismatch discrimination, and solubility. Greehey CCRI investigators use these methods to enhance oligo stability and binding affinity in conjunction with conjugation with cholesterol or folate moieties to improve uptake by tumour cells. The pharmacokinetic behaviour can be further enhanced through the mechanism by which the oligos are delivered.
Encapsulation for delivery can improve pharmacokinetic behaviour
Efficacious delivery of candidate RNAs to the site of a tumour is the Achilles heel of RNA therapeutics. Liposomes, initially based on cationic lipids and now based on neutral emulsions and viral systems based on lentivirus or adeno associated virus have been employed, but their clinical application is limited by either technical hurdles or safety concerns. Our laboratories have adopted several different, innovative approaches to systemically deliver miRNAs. One approach involves using hydrogels, or crosslinked polymeric networks swollen in biological fluids. Hydrogels have the potential to overcome some limitations of other systemic delivery approaches, including liposomes and viral based systems through their high biocompatibility, tunable physicochemical properties, and ability to recognise and respond to stimuli. In addition, use of hydrogels can help maintain spatial and temporal control over the administration of biotherapeutics within the body and can therefore substantially reduce the drug dose needed to achieve anti-tumour effects, mitigating potential off-target effects. Another approach uses poly D, L-lactic-co-glycolic acid (PLGA) nanoparticles, which are widely used for systemic and targeted drug delivery with reduced toxicity. PLGA nanoparticles have several desirable properties, including biocompatibility, ability to entrap both water soluble and insoluble molecules and deliver them over sustained periods, enhanced endosomal escape and improved tumour homing.
RNA-binding proteins as therapeutic targets
Over 1,500 RBPs have been identified in the human genome to date – these regulate gene expression via numerous processes such as splicing, poly adenylation, mRNA decay, and translation. RBPs are differentially expressed in tumours at significantly higher levels compared to other classes of genes. As expected, a growing number of RBPs has been implicated in tumorigenesis, and several oncogene and tumour suppressor candidates have been identified. Inhibitors against RBPs can be developed by taking advantage of the unique characteristics of their RNA binding domains. Moreover, due to their broad regulatory impact, RBP targeting could be more effective in suppressing oncogenic pathways than other therapeutic approaches.
Researchers at Greehey CCRI have discovered that specific RNA binding proteins are highly expressed in and play causal roles in several paediatric cancers. One of us (Penalva) has shown that RBPs play significant roles in the progression of brain tumours through their support of cancer stem proliferation and self-renewal. Another (Rao) has identified an RBP that is highly amplified and supports the development and progression of osteosarcoma. Using unbiased high throughput screening approaches, we and other Greehey CCRI scientists are identifying small molecule inhibitors of those RBPs and evaluating them in preclinical in vitro and in vivo models. These investigations are facilitated by our high throughput/high-content screening facility, which conducts genome wide screening of siRNA, miRNA, and lncRNA libraries, and by the Center for Innovation in Drug Discovery, is a state of the art technological hub that performs high throughput screening of chemical and natural product libraries.
In Vivo validation and translation to the clinic
Mouse xenograft models based on human tumour cells are widely used to predict the overall efficacy of therapeutic agents in the clinic. Established at Greehey CCRI in 2016, the Texas Paediatric Patient Derived Xenograft Facility develops and validates preclinical models based on transplantation of tumour tissue from paediatric cancer patients into immunodeficient mice. The facility has already developed over 80 childhood cancer PDX models and plans to heterograft and propagate 200-250 new PDX preclinical models over the next few years, providing investigators with molecular characterisation, de-identified patient data, and cryopreserved xenograft tissues.
PDXs are selected to match the cell lines used with respect to oncogene mutational status, protein markers, and chromosomal copy number profile. Greehey CCRI investigators use established methods to edit cells derived from tumours to express luciferase and re-implant them in vivo for tracking primary tumour growth and detecting metastatic processes. This facility is key to successfully evaluating candidate miRNAs and RBPs in the most physiologically relevant preclinical models. After demonstrating therapeutic efficacy of miRNA replacement or inhibition with minimal toxicity in vivo, candidates are moved into a phase I clinical trial through the Mays Cancer Center Institute for Drug Development (IDD), which integrates basic, translational and clinical research to identify and develop novel cancer treatments.
One of our groups (Penalva) studies RBPs and miRNAs at the intersection of neurogenesis and brain tumour development. They identified the RNA-binding stem cell protein Musashi1 (Msi1) as a key contributor to medulloblastoma and glioblastoma, regulating multiple cancer relevant processes including apoptosis, cell cycle, proliferation, migration, invasion, and adhesion via a complex network of target genes. They showed that Musashi1 is highly expressed in high risk medulloblastoma sub-groups 3 and 4 and linked to poor prognosis through its critical role in the survival of tumour-initiating cells and influence on both radio and chemo resistance. They recently developed an inhibitor that blocks Msi1 RNA binding domains and is developing derivatives with improved properties for translation into the clinic.
Two of our groups (Penalva and Pertsemlidis) collaborate on the study of three miRNAs that function as important regulatory switches, influencing cell fate decisions, and tumour development. miR-124, miR-128, and miR-137 are among the most highly expressed miRNAs in the brain, required for neuronal production and displaying parallel increased expression as cells differentiate. These three miRNAs are often repressed in glioblastoma and neuroblastoma and are predicted to act as tumour suppressors. Recent results indicate that miR-124, -128 and -137 act synergistically and regulate overlapping target sets in both glioblastoma and neuroblastoma. Interestingly, miR-124, -128 and -137 share a significant number of targets with Musashi1, with potentially opposite impact on their expression, suggesting that the balance between miRNAs and MBPs can control neural stem cell fate.
Our third group (Rao) recently discovered that miR-584-5p is expressed at a very low levels or is completely absent in medulloblastoma, which is the most common childhood malignant brain cancer. By using physiologically relevant pre-clinical tumour models, they and their colleagues showed that increasing the levels of miR-584-5p can block the growth of medulloblastoma without affecting healthy brain cells. Notably, miR-584-5p significantly improved the efficacy of chemotherapeutic drug vincristine and radiation, a therapy combination that is routinely used to treat medulloblastoma patients. Since vincristine and radiation cause severe toxicities, miR-584-5p will make it possible to treat these patients with a dose that is markedly lower and therefore, less toxic.
The future of RNA-based approaches for the treatment of paediatric cancers
With recent progress in leveraging chemical modifications and conjugation of RNA-based therapeutics to improve stability and resistance to degradation and the availability of encapsulation methods for delivery that were not available a decade ago, RNA based therapies now hold significant promise. Although pharmacotherapy and drug development have seen decades of progress with protein-based therapeutics, the targets that they can reach remain limited – the majority of proteins remaining undruggable through conventional methods. Considering protein coding genes represent that less than 2% of the human genome, large numbers of non-coding RNAs remain unexplored as therapeutic, offering new opportunities for drug development in our search for effective, less toxic, curative therapies for paediatric cancers.