Patent Application
20240325429
This application claims the benefit of the European Patent Application EP21191594.7 filed 17 Aug. 2021, which is incorporated herein by reference.
The present invention relates to a pharmaceutical nucleic acid agent capable of downregulating or inhibiting the activity of the minor spliceosome (MiS), for use in treatment, or prevention of recurrence, of cancer. The invention also relates to a method for the diagnosis of cancer.
Cancer evolves continuously by modifying its transcriptome as it evades therapeutic interventions. Splicing, which is routinely leveraged, results in altered isoforms of key cancer genes involved in disease progression. Currently, little is known about MiS in cancer progression. Here the inventors explore PCa as an exemplar cancer to understand the role of MiS in disease progression.
Androgen deprivation therapy (ADT) is used to treat advanced PCa. However, for PCa resistant to ADT, second-generation potent androgen receptor signalling inhibitors (ARSi) such as enzalutamide and abiraterone are used. While initially effective, intrinsic or acquired resistance to ARSi in form of castration resistant prostate cancer (CRPC) eventually develops in this ultimately lethal disease. In CRPC resistance to ARSi is conferred by intra-tumoral heterogeneity driven by re-activation of the AR axis including AR amplifications, AR mutations, AR co-activators and ligand independent AR activations.
Another increasingly recognized mechanism in the context of prolonged ARSi treatment (AR suppression) is the trans-differentiation form adenocarcinoma (CRPC-adeno) to neuroendocrine PCa (CRPC-NE), an extremely lethal and AR-indifferent variant of PCa. Generally, CRPC-adeno and CRPC-NE are defined by expression/absence of characteristic markers such as AR and KLK3 (CRPC-adeno) or SYP and CHGA and AR absence (CRPC-NE). The expression of those markers can be quantified in so called AR- or NEPC-scores.
While CRPC-adeno and CRPC-NE share similar genomic landscapes, they have dramatically distinct transcriptomes, suggesting non-coding RNA events and RNA splicing as a potential mechanism of PCa transdifferentiation and progression. Indeed, multiple studies propose that alternative splicing of the AR transcript plays a key role in therapy resistance of CRPC-adeno. While the splicing factor SRRM4 has been identified as a crucial driver of CRPC-NE transdifferentiation. In general, non-canonical splicing has been extensively linked to prostate tumorigenesis and many PCa relevant genes display isoform switching during cancer development and progression. Yet little is known about the pathways controlling it and a unified hypothesis explaining the molecular origin of those isoforms is still lacking. Thus, understanding the mechanisms underlying aberrant splicing in PCa is essential both for predicting tumor progression and for discovering key regulators. While the role of canonical splicing in cancer has been studied extensively, the present understanding of the interplay between minor intron splicing and cancer is lacking.
Minor introns (<0.5%), which require the minor spliceosome, are found in genes with mostly major introns that are spliced by the major spliceosome. These minor intron-containing genes (MIGs) execute diverse functions in disparate molecular pathways. Despite the diverse functions, MIGs are highly enriched in the essentialome, a list of genes that are essential for survival. The essentiality of MIGs is reflected in early embryonic lethality when MiS is inhibited in mice, zebrafish, and Drosophila. Moreover, loss of U11 snRNA in the developing pallium results in aberrant splicing of MIGs that resulted in cell cycle defect and loss of rapidly dividing neural stem cells. Regarding cancer, the dysregulation of minor intron splicing has been linked to the Peutz-Jegher's syndrome and myelodysplastic syndrome which frequently proceed to gastrointestinal cancer and acute myeloid leukemia, respectively. There is further evidence that MiS components show a reliable association with an increased risk of scleroderma (U11/U12-65K protein), AML (U11-59K protein) and familial PCa (U11 snRNA). In fact, microRNA (miRNA) profiling studies of high-risk Finnish PCa families identified altered U11 snRNA expression as a risk factor to develop PCa.
Based on the above-mentioned state of the art, the objective of the present invention is to provide means and methods to treat or prevent the recurrence of cancer. This objective is attained by the subject-matter of the independent claims of the present specification, with further advantageous embodiments described in the dependent claims, examples, figures and general description of this specification.
Here the inventors show that MiS function, which is regulated by the AR-axis, increases with (prostate) cancer disease stage and degree of differentiation. In fact, MiS component, U6atac snRNA, might serve as an additional marker for cancer diagnostics. The inventors show that siU6atac-mediated MiS inhibition is more effective at blocking PCa cell proliferation than the current state of the art combination therapy such as EZH2 inhibitor/enzalutamide. Finally, the inventors show that other MiS components can also be targeted, and that MiS inhibition also blocks proliferation of other cancer cell types. In all, this work brings to light a novel pathway, the minor spliceosome, as point of entry for therapeutics against lethal PCa and that this strategy extends to other cancer types.
A first aspect of the invention relates to a pharmaceutical nucleic acid agent capable of downregulating or inhibiting the activity of the minor spliceosome (MiS) for use in treatment or prevention of recurrence of cancer.
A second aspect of the invention relates to a method for assigning a likelihood of having or developing cancer to a patient. A high likelihood of having or developing cancer is assigned if an expression level of snRNA U6atac is 2-3.
A third aspect of the invention relates to a pharmaceutical nucleic acid agent for use according to the first aspect, wherein a high likelihood of having or developing cancer, or of having or developing a cancer of advanced, therapy resistant phenotype is assigned to the patient according the method of the second aspect.
For purposes of interpreting this specification, the following definitions will apply and whenever appropriate, terms used in the singular will also include the plural and vice versa. In the event that any definition set forth below conflicts with any document incorporated herein by reference, the definition set forth shall control.
The terms “comprising,” “having,” “containing,” and “including,” and other similar forms, and grammatical equivalents thereof, as used herein, are intended to be equivalent in meaning and to be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items. For example, an article “comprising” components A, B, and C can consist of (i.e., contain only) components A, B, and C, or can contain not only components A, B, and C but also one or more other components. As such, it is intended and understood that “comprises” and similar forms thereof, and grammatical equivalents thereof, include disclosure of embodiments of “consisting essentially of” or “consisting of.”
Where a range of values is provided, it is understood that each intervening value, to the tenth of the unit of the lower limit, unless the context clearly dictate otherwise, between the upper and lower limit of that range and any other stated or intervening value in that stated range, is encompassed within the disclosure, subject to any specifically excluded limit in the stated range. Where the stated range includes one or both of the limits, ranges excluding either or both of those included limits are also included in the disclosure.
Reference to “about” a value or parameter herein includes (and describes) variations that are directed to that value or parameter per se. For example, description referring to “about X” includes description of “X.”
As used herein, including in the appended claims, the singular forms “a,” “or,” and “the” include plural referents unless the context clearly dictates otherwise.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art (e.g., in cell culture, molecular genetics, nucleic acid chemistry, hybridization techniques and biochemistry). Standard techniques are used for molecular, genetic and biochemical methods (see generally, Sambrook et al., Molecular Cloning: A Laboratory Manual, 4th ed. (2012) Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. and Ausubel et al., Short Protocols in Molecular Biology (2002) 5th Ed, John Wiley & Sons, Inc.) and chemical methods.
The term snRNA U6atac in the context of the present specification relates to
The term gene refers to a polynucleotide containing at least one open reading frame (ORF) that is capable of encoding a particular polypeptide or protein after being transcribed and translated. A polynucleotide sequence can be used to identify larger fragments or full-length coding sequences of the gene with which they are associated. Methods of isolating larger fragment sequences are known to those of skill in the art.
The terms gene expression or expression, or alternatively the term gene product, may refer to either of, or both of, the processes—and products thereof—of generation of nucleic acids (RNA) or the generation of a peptide or polypeptide, also referred to transcription and translation, respectively, or any of the intermediate processes that regulate the processing of genetic information to yield polypeptide products. The term gene expression may also be applied to the transcription and processing of a RNA gene product, for example a regulatory RNA or a structural (e.g. ribosomal) RNA. If an expressed polynucleotide is derived from genomic DNA, expression may include splicing of the mRNA in a eukaryotic cell. Expression may be assayed both on the level of transcription and translation, in other words mRNA and/or protein product.
The term downregulating or inhibiting expression in the context of the present specification relates to the ability to reduce the number of RNA molecules inside a cell.
The term Nucleotides in the context of the present specification relates to nucleic acid or nucleic acid analogue building blocks, oligomers of which are capable of forming selective hybrids with RNA or DNA oligomers on the basis of base pairing. The term nucleotides in this context includes the classic ribonucleotide building blocks adenosine, guanosine, uridine (and ribosylthymine), cytidine, the classic deoxyribonucleotides deoxyadenosine, deoxyguanosine, thymidine, deoxyuridine and deoxycytidine. It further includes analogues of nucleic acids such as phosphotioates, 2′O-methylphosphothioates, peptide nucleic acids (PNA; N-(2-aminoethyl)-glycine units linked by peptide linkage, with the nucleobase attached to the alpha-carbon of the glycine) or locked nucleic acids (LNA; 2′O, 4′C methylene bridged RNA building blocks). Wherever reference is made herein to a hybridizing sequence, such hybridizing sequence may be composed of any of the above nucleotides, or mixtures thereof.
The terms capable of forming a hybrid or hybridizing sequence in the context of the present specification relate to sequences that under the conditions existing within the cytosol of a mammalian cell, are able to bind selectively to their target sequence. Such hybridizing sequences may be contiguously reverse-complimentary to the target sequence, or may comprise gaps, mismatches or additional non-matching nucleotides. The minimal length for a sequence to be capable of forming a hybrid depends on its composition, with C or G nucleotides contributing more to the energy of binding than A or T/U nucleotides, and on the backbone chemistry.
In the context of the present specification, the term hybridizing sequence encompasses a polynucleotide sequence comprising or essentially consisting of RNA (ribonucleotides), DNA (deoxyribonucleotides), phosphothioate deoxyribonucleotides, 2′-O-methyl-modified phosphothioate ribonucleotides, LNA and/or PNA nucleotide analogues. In certain embodiments, a hybridizing sequence according to the invention comprises 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29 or 30 nucleotides. In certain embodiments, the hybridizing sequence comprises deoxynucleotides, phosphothioate deoxynucleotides, LNA and/or PNA nucleotides or mixtures thereof.
The term antisense oligonucleotide in the context of the present specification relates to an oligonucleotide having a sequence substantially complimentary to, and capable of hybridizing to, an RNA. Antisense action on such RNA will lead to modulation, particular inhibition or suppression of the RNA's biological effect. If the RNA is an mRNA, expression of the resulting gene product is inhibited or suppressed. Antisense oligonucleotides can consist of DNA, RNA, nucleotide analogues and/or mixtures thereof. The skilled person is aware of a variety of commercial and non-commercial sources for computation of a theoretically optimal antisense sequence to a given target. Optimization can be performed both in terms of nucleobase sequence and in terms of backbone (ribo, deoxyribo, analogue) composition. Many sources exist for delivery of the actual physical oligonucleotide, which generally is synthesized by solid state synthesis.
The term siRNA (small/short interfering RNA) in the context of the present specification relates to an RNA molecule capable of interfering with the expression (in other words: inhibiting or preventing the expression) of a gene comprising a nucleic acid sequence complementary or hybridizing to the sequence of the siRNA in a process termed RNA interference. The term siRNA is meant to encompass both single stranded siRNA and double stranded siRNA. siRNA is usually characterized by a length of 17-24 nucleotides. Double stranded siRNA can be derived from longer double stranded RNA molecules (dsRNA). According to prevailing theory, the longer dsRNA is cleaved by an endo-ribonuclease (called Dicer) to form double stranded siRNA. In a nucleoprotein complex (called RISC), the double stranded siRNA is unwound to form single stranded siRNA. RNA interference often works via binding of an siRNA molecule to the mRNA molecule having a complementary sequence, resulting in degradation of the mRNA. RNA interference is also possible by binding of an siRNA molecule to an intronic sequence of a pre-mRNA (an immature, non-spliced mRNA) within the nucleus of a cell, resulting in degradation of the pre-mRNA.
The term shRNA (small hairpin RNA) in the context of the present specification relates to an artificial RNA molecule with a tight hairpin turn that can be used to silence target gene expression via RNA interference (RNAi).
The term sgRNA (single guide RNA) in the context of the present specification relates to an RNA molecule capable of sequence-specific repression of gene expression via the CRISPR (clustered regularly interspaced short palindromic repeats) mechanism.
The term miRNA (microRNA) in the context of the present specification relates to a small non-coding RNA molecule (containing about 22 nucleotides) that functions in RNA silencing and post-transcriptional regulation of gene expression.
The term specific binding in the context of the present invention refers to a property of ligands that bind to their target with a certain affinity and target specificity. The affinity of such a ligand is indicated by the dissociation constant of the ligand. A specifically reactive ligand has a dissociation constant of ≥10−7 mol/L when binding to its target, but a dissociation constant at least three orders of magnitude higher in its interaction with a molecule having a globally similar chemical composition as the target, but a different three-dimensional structure.
As used herein, the term pharmaceutical composition refers to a compound of the invention, or a pharmaceutically acceptable salt thereof, together with at least one pharmaceutically acceptable carrier. In certain embodiments, the pharmaceutical composition according to the invention is provided in a form suitable for topical, parenteral or injectable administration.
As used herein, the term pharmaceutically acceptable carrier includes any solvents, dispersion media, coatings, surfactants, antioxidants, preservatives (for example, antibacterial agents, antifungal agents), isotonic agents, absorption delaying agents, salts, preservatives, drugs, drug stabilizers, binders, excipients, disintegration agents, lubricants, sweetening agents, flavoring agents, dyes, and the like and combinations thereof, as would be known to those skilled in the art (see, for example, Remington: the Science and Practice of Pharmacy, ISBN 0857110624).
As used herein, the term treating or treatment of any disease or disorder (e.g. cancer) refers in one embodiment, to ameliorating the disease or disorder (e.g. slowing or arresting or reducing the development of the disease or at least one of the clinical symptoms thereof). In another embodiment “treating” or “treatment” refers to alleviating or ameliorating at least one physical parameter including those which may not be discernible by the patient. In yet another embodiment, “treating” or “treatment” refers to modulating the disease or disorder, either physically, (e.g., stabilization of a discernible symptom), physiologically, (e.g., stabilization of a physical parameter), or both. Methods for assessing treatment and/or prevention of disease are generally known in the art, unless specifically described hereinbelow.
A first aspect of the invention relates to a pharmaceutical nucleic acid agent capable of downregulating or inhibiting the activity of the minor spliceosome (MiS) for use in treatment or prevention of recurrence of cancer.
In certain embodiments, the agent is capable of downregulating or inhibiting expression of snRNA U6atac, particularly by hybridizing to, and leading to degradation or inhibition of, SEQ ID No 001.
In certain embodiments, the agent is or encodes an antisense oligonucleotide. In certain embodiments, the agent is or encodes an siRNA.
In certain embodiments, said cancer is a lethal cancer of advanced, therapy resistant phenotype (neuroendocrine and adenocarcinomas, particularly neuroendocrine carcinomas).
In certain embodiments, said cancer is selected from glioblastoma, breast cancer, chronic myeloid leukemia (CML), bladder cancer, colon cancer, kidney cancer and prostate cancer.
In certain embodiments, said cancer is prostate cancer.
In certain embodiments, said cancer is selected from castration-resistant prostate cancer (CRPC), neuroendocrine prostate cancer (NEPC), castration-resistant neuroendocrine prostate cancer (CRPC-NE) and small cell prostate cancer. In certain embodiments, said cancer is a lethal cancer of androgen receptor negative NE type and/or androgen receptor negative and NE negative type. Advanced resistant prostate cancers do not respond to ADT (androgene deprivation therapy) and ARSi (Androgene receptor inhibitors).
In certain embodiments, said agent is administered in combination with a platinum-containing complex.
In certain embodiments, said agent is administered in combination with a platinum-containing drug selected from carboplatin, satraplatin, cisplatin, dicycloplatin, nedaplatin, oxaliplatin, picoplatin, and/or triplatin.
A second aspect of the invention relates to a method for assigning a likelihood of having or developing cancer to a patient. A high likelihood of having or developing cancer is assigned if an expression level of snRNA U6atac is 2-3. For in situ testing there are 4 possible levels 0, 1, 2, 3. 2-3 would be equivalent to moderate/strong expression.
In certain embodiments, a high likelihood of having or developing cancer is assigned if the expression level of snRNA U6atac is 2 or 3.
In certain embodiments, a high likelihood of having or developing a cancer of advanced, therapy resistant phenotype is assigned if an expression level of snRNA U6atac is 2 or 3. In certain embodiments, a high likelihood of having or developing a cancer of advanced, therapy resistant phenotype is assigned if the expression level of snRNA U6atac is 2 or 3.
The inventors have observed a strong difference, regarding U6atac expression, between 1) benign tissue and cancer 2) primary and advanced or metastatic cancer. Their conclusion is that the relationship is valid for any cancer, wherein a higher U6atac score signals a more advanced stage of cancer.
In certain embodiments of the second aspect, the cancer is selected from glioblastoma, breast cancer, chronic myeloid leukemia (CML), bladder cancer, colon cancer, kidney cancer and prostate cancer. In certain embodiments of the second aspect, the cancer is prostate cancer.
A third aspect of the invention relates to a pharmaceutical nucleic acid agent for use according to the first aspect, wherein a high likelihood of having or developing cancer, or of having or developing a cancer of advanced, therapy resistant phenotype is assigned to the patient according the method of the second aspect.
A further aspect of the invention relates to a method for treatment or prevention of recurrence of cancer in a patient, said method comprising administering a pharmaceutical nucleic acid agent according to the first aspect to a patient.
A further aspect of the invention relates to a method for treatment or prevention of recurrence of cancer in a patient, said method comprising the steps:
In certain embodiments, the patient is treated via administering a pharmaceutical nucleic acid agent according to the first aspect.
A further aspect of the invention relates to a system for performing the method according to the second aspect.
A further aspect of the invention relates to a use of an agent being able to determine the expression level of snRNA U6atac in the manufacture of a kit for the detection of cancer.
Similarly, within the scope of the present invention is a method or treating cancer in a patient in need thereof, comprising administering to the patient a nucleic acid sequence vector according to the above description.
Similarly, a dosage form for the prevention or treatment of cancer is provided, comprising a non-agonist ligand or antisense molecule according to any of the above aspects or embodiments of the invention.
The skilled person is aware that any specifically mentioned drug compound mentioned herein may be present as a pharmaceutically acceptable salt of said drug. Pharmaceutically acceptable salts comprise the ionized drug and an oppositely charged counterion. Non-limiting examples of pharmaceutically acceptable anionic salt forms include acetate, benzoate, besylate, bitatrate, bromide, carbonate, chloride, citrate, edetate, edisylate, embonate, estolate, fumarate, gluceptate, gluconate, hydrobromide, hydrochloride, iodide, lactate, lactobionate, malate, maleate, mandelate, mesylate, methyl bromide, methyl sulfate, mucate, napsylate, nitrate, pamoate, phosphate, diphosphate, salicylate, disalicylate, stearate, succinate, sulfate, tartrate, tosylate, triethiodide and valerate. Non-limiting examples of pharmaceutically acceptable cationic salt forms include aluminium, benzathine, calcium, ethylene diamine, lysine, magnesium, meglumine, potassium, procaine, sodium, tromethamine and zinc.
Dosage forms may be for parenteral administration, such as subcutaneous, intravenous, intrahepatic or intramuscular injection forms. Optionally, a pharmaceutically acceptable carrier and/or excipient may be present.
Another aspect of the invention relates to a pharmaceutical composition comprising a compound of the present invention, or a pharmaceutically acceptable salt thereof, and a pharmaceutically acceptable carrier. In further embodiments, the composition comprises at least two pharmaceutically acceptable carriers, such as those described herein.
In certain embodiments of the invention, the compound of the present invention is typically formulated into pharmaceutical dosage forms to provide an easily controllable dosage of the drug and to give the patient an elegant and easily handleable product.
The pharmaceutical composition can be formulated for enteral administration, particularly oral administration or rectal administration. In addition, the pharmaceutical compositions of the present invention can be made up in a solid form (including without limitation capsules, tablets, pills, granules, powders or suppositories), or in a liquid form (including without limitation solutions, suspensions or emulsions).
The pharmaceutical composition can be formulated for parenteral administration, for example by i.v. infusion, intradermal, subcutaneous or intramuscular administration.
The dosage regimen for the compounds of the present invention will vary depending upon known factors, such as the pharmacodynamic characteristics of the particular agent and its mode and route of administration; the species, age, sex, health, medical condition, and weight of the recipient; the nature and extent of the symptoms; the kind of concurrent treatment; the frequency of treatment; the route of administration, the renal and hepatic function of the patient, and the effect desired. In certain embodiments, the compounds of the invention may be administered in a single daily dose, or the total daily dosage may be administered in divided doses of two, three, or four times daily.
In certain embodiments, the pharmaceutical composition or combination of the present invention can be in unit dosage of about 1-1000 mg of active ingredient(s) for a subject of about 50-70 kg. The therapeutically effective dosage of a compound, the pharmaceutical composition, or the combinations thereof, is dependent on the species of the subject, the body weight, age and individual condition, the disorder or disease or the severity thereof being treated. A physician, clinician or veterinarian of ordinary skill can readily determine the effective amount of each of the active ingredients necessary to prevent, treat or inhibit the progress of the disorder or disease.
The pharmaceutical compositions of the present invention can be subjected to conventional pharmaceutical operations such as sterilization and/or can contain conventional inert diluents, lubricating agents, or buffering agents, as well as adjuvants, such as preservatives, stabilizers, wetting agents, emulsifiers and buffers, etc. They may be produced by standard processes, for instance by conventional mixing, granulating, dissolving or lyophilizing processes. Many such procedures and methods for preparing pharmaceutical compositions are known in the art, see for example L. Lachman et al. The Theory and Practice of Industrial Pharmacy, 4th Ed, 2013 (ISBN 8123922892).
The invention further encompasses, as an additional aspect, the use of a nucleic acid agent as identified herein, or its pharmaceutically acceptable salt, as specified in detail above, for use in a method of manufacture of a medicament for the treatment or prevention of cancer.
Similarly, the invention encompasses methods of treatment of a patient having been diagnosed with a disease associated with cancer. This method entails administering to the patient an effective amount of a nucleic acid agent as identified herein, or its pharmaceutically acceptable salt, as specified in detail herein.
The invention further encompasses the use of a nucleic acid agent able to detect an snRNA U6atac expression level identified herein for use in the manufacture of a kit for the detection of cancer.
Wherever alternatives for single separable features such as, for example, a target gene, nucleic acid agent type or medical indication are laid out herein as “embodiments”, it is to be understood that such alternatives may be combined freely to form discrete embodiments of the invention disclosed herein.
The invention is further illustrated by the following examples and figures, from which further embodiments and advantages can be drawn. These examples are meant to illustrate the invention but not to limit its scope.
Boxplots display values of minimum, first quartile, median, third quartile, and maximum (two-sided unpaired t-test, ns p>0.05, ***p<0.001, ****p<0.0001). Each data point represents a single experiment, experiments were performed in triplicates. C) Normalized luminescence values of minor spliceosome luc-reporter plasmids in LNCaP (siU6atac n=5, Anisomycin n=8) and PM154 (siU6atac n=5, Anisomycin n=4) cells treated with siU6atac (96 h) or Anisomycin (1 ug/ml for 4 h) (mean±SEM, ordinary one-way Anova; ns p>0.05, ***p<0.001, ****p<0.0001). Each data point represents a single experiment, experiments were performed in triplicates. D) U6atac snRNA expression as x-fold of siScrambled (siScr.) normalized to the mean of GAPDH and ACTB gene transcription in LNCaP, C4-2, 22Rv1 and PM154 cells treated with siU6atac RNA for 48 and 96 hours (n=5) Data are represented using box and whisker plots. Boxplots display values of minimum, first quartile, median, third quartile, and maximum (one-way Anova; ****p<0.0001). Experiments were performed in triplicates.
Since MIGs are highly enriched in cell cycle regulation and survival, the inventors explored whether MIGs are enriched in biological pathways exploited by cancer-causing genes. For this, the inventors determined whether MIG-encoded proteins were enriched amongst proteins that interact with proteins encoded by cancer-causing genes in a network of 160,881 protein-protein interactions (PPI) between 15,366 human proteins as of the HINT database (Das, J et al., BMC Syst Biol 2012, 6, 92). The inventors hypothesized that MIGs are prime recipients of molecular information that cascade from cancer-causing genes through their protein-protein interactions. Considering 403 cancer-causing genes in the set of 15,366 interacting proteins as of the Cancer Genome Interpreter database (Bigas, N et al., Genome Medicine 2018, 10 (1), 25), the inventors computed the shortest distance d of all proteins encoded by cancer-causing genes, where d=1 indicates a direct interaction. In each distance bin, the inventors determined the enrichment of 542 MIG-encoded proteins (inset,
That MIGs interact directly with cancer-causing proteins led the inventors to explore the extent to which MIGs exhibit a greater degree of differential expression between distinct cancer types (relative to non-MIGs). To address this question, the inventors first used normalized MIG expression data to perform hierarchical sample clustering (HSC) of 9 different solid cancer types, and then evaluated the extent to which this MIG expression successfully clusters samples by cancer type. The inventors first evaluated the quality of the resultant clustering by visualizing the clustered data with a dendrogram and its associated heatmap, wherein structure can readily be seen in the case for which only MIGs were used in clustering the samples (
The inventors then took a more objective, quantitative approach to measure this deterioration in clustering (i.e., visual structure within these heatmaps). To quantitatively evaluate the extent to which a given expression dataset may be clustered, the inventors used the Silhouette coefficient, wherein higher coefficient values designate more clearly-defined and correct clustering. The inventors employed a simulation (data re-sampling-based scheme) to generate 1000 gene sets for each fraction of non-MIG genes, and then calculated the Silhouette coefficient associated with each resampled gene set. This quantitative approach recapitulated what the inventors had observed visually in the heatmaps: the Silhouette coefficients associated with clustering expression data in gene sets with higher fractions of non-MIGs resulted in less effective in clustering for these 9 different cancer types (
The inventors next evaluated whether MIG expression may likewise exhibit greater differential gene expression across different stages of PCa progression (relative to non-MIGs). This was carried out for the transdifferentiation analysis across the distinct transcriptomes, illustrated by principal component analysis. Specifically, to obtain data representing different stages of PCa progression, the inventors used prostate samples derived from GTEx (normal tissues), TCGA (primary PCa samples), and SU2C (CRPC-adeno and CRPC-NE) datasets. As with the inventors' clustering analysis on 9 distinct cancer types detailed above, the inventors observed that MIG expression resulted in better clustering performance on these distinct phases of prostate cancer progression, although the visual disparities between the heatmaps associated with 0% non-MIGs and 100% non-MIGs are not as pronounced as what the inventors had observed in the context of the inventors' pan-cancer analysis. The same simulation-based scheme in which Silhouette coefficients were measured in the inventors' pan-cancer analysis was similarly applied in the context of these broad stages of cancer progression. Again, the inventors observed that MIG expression results in greater clustering performance than non-MIG expression (
The upregulation of MIGs with PCa progression and transdifferentiation would necessitate increased MiS activity, which in turn is regulated by the levels of MiS components. Therefore, the inventors queried the expression kinetics of U11 snRNA (RNU11), a crucial MiS component in PCa (PRAD) TCGA data. The inventors observed a significant (Wilcoxon Test, p=0.023) association between high-grade (Gleason score 8 and 9) PCa patients and high RNU11 expression (
To test the prediction that U6atac expression is elevated with cancer onset and progression, the inventors developed an in-situ probe specifically against U6atac. The inventors used this probe to survey different cancer types, and found significantly higher expression of U6atac in cancer tissue compared to matching benign samples (Wilcox test, p=0.0038). This finding demonstrates a strong specificity of U6atac expression for highly proliferative tissues (pan-cancer Tissue Microarray (TMA)) (
Consistent with the increased expression of U6atac in more aggressive PCa cell lines (
AR signaling plays a critical role in PCa progression, and is often the apex of oncogenic pathways. Therefore, the inventors hypothesized that MiS activity across PCa progression might be linked to AR signaling. To simulate stress response and re-activation of AR signalling in PCa, the inventors mimicked therapy resistance mechanisms, and subjected PCa cells to long term androgen deprivation therapy (ADT) and ARSi using charcoal-stripped (C/S) media, abiraterone and enzalutamide. The inventors observed a significant increase in MiS activity in cells exposed to ADT/ARSi, whereas the treatment had only limited effects on major splicing (one-way ANOVA; *p=0.0117, ***p and ****p<0.001) (
Next, the inventors explored the relationship of AR signaling to minor intron splicing. Here, the inventors used luciferase reporter as the readout of minor intron splicing in therapy-sensitive LNCaP cells. The overexpression of AR led to a significant increase in MiS activity (
The association between AR activation and MiS activity suggests the existence of an AR-MiS regulatory axis. To address this, the inventors determined whether AR interacts with MiS proteins. Co-immunoprecipitations (Co-IPs) with an antibody directed against the AR revealed an interaction with PDCD7, one of the seven proteins unique to the MiS complex, and vice versa, implying a direct regulation of the MiS by AR through PDCD7 (
Based on the inventors' finding that the MiS plays a crucial role in PCa progression, the inventors next inhibited the MiS in PCa by targeting the U6atac snRNA, which normally exhibits higher rates of turnover but is detected at higher levels in advanced cancer stages. The inventors used siRNA against siU6atac in four PCa cell lines: LNCaP (primary PCa, therapy-sensitive), C4-2 (CRPC-adeno, therapy-resistant), and 22RV1 (CRPC-adeno, therapy-resistant) cell lines and a patient-derived organoid, PM154 (CRPC-NE, therapy-resistant). First, the inventors chose C4-2 cells to establish the kinetics of U6atac downregulation (
To capture a comprehensive effect of MiS inhibition, the inventors performed ribo-depleted total RNAseq on the four cell types treated with siU6atac and siScrambled for 96 hours. Equivalent levels of U6atac KD were observed in all 4 cell lines (
Next, the inventors explored whether there was a cell-type specific effect of MiS inhibition by performing an intersection analysis of the MIGs with significantly elevated minor intron retention or AS. Here, the inventors separated the three cell lines from the organoid due to their different origins, culture conditions, genomic architectures and PCa phenotypes (CRPC-NE), which was also reflected in principal component analysis. Of the 380, 419, and 390 MIGs with significantly elevated minor intron retention in LNCaP, C4-2, and 22RV1 cells, respectively, the inventors found that 337 MIGs were common to all (
The aberrant minor intron splicing in siU6atac-treated samples should impact the overall transcriptome of PCa, which the inventors captured by differential gene expression analysis. The inventors set a 1 transcript per million (TPM) threshold for gene expression and, using isoDE2, the inventors found 68 genes that were significantly upregulated (log 2FC≥1, p≤0.01) and 691 genes significantly downregulated (log 2FC≤−1, p:s 0.01) in the siU6atac-treated LNCaP cells compared to siScrambled (
Next, the inventors sought to untangle the downstream molecular defect of aberrant minor intron splicing in conjunction with the transcriptomic changes captured by RNAseq. For this, the inventors took the list of MIGs with elevated minor intron retention common to either the LNCaP, C4-2, and 22Rv1 cell lines (
To determine whether the transcriptional changes captured for each cell line extended to differential protein production, the inventors performed LC-MS/MS on the same cell lines treated with siRNA (
Comparing the proteomic data of the four analysed cell lines representing PCa disease progression revealed a cell type, and thus probably PCa subtype and context dependent MiS-dependent proteome, much like the RNAseq analysis. Each cell line expressed a unique set of up and downregulated proteins including MIGs and non-MIGs, (
The inventors next explored concordance between the transcriptome and proteome data, which were generally discordant in each cell line (
Next, the inventors wanted to identify cellular heterogeneity in the response to siU6atac mediated MiS inhibition. The inventors thus also performed single cell RNAseq (scRNAseq) on LNCaP cells and PM154 organoids post siU6atac. After standard data processing and quality control procedures (methods) the inventors obtained transcriptomic profiles for 8206 siScrambled and 6730 LNCaP cells and 11181 siScrambled and 11475 PM154 organoids. The inventors employed unsupervised clustering to identify heterogeneity in response to siU6atac. K-means clustering and UMAP projection of the combined data across both genotypes, which revealed 9 major cell clusters (
The comparison of distribution of cells from each condition in the 9 clusters and the distribution of cells in the different phases of cell cycle led the inventors to investigate cell cycle defects through scRNAseq analysis (
Given that siU6atac is successful at blocking cell cycle progression in PCa and that AR signaling is a crucial driver of PCa, the inventors next explored the AR score (a crucial metric of CRPC-adeno progression) in siU6atac-treated LNCaP cells and PM154 organoids (
Together, scRNAseq Revealed siU6atac Mediated Transcriptomic Remodeling that Contributes to PCa progression and lineage plasticity by decreasing cell cycle, AR signaling and EMT.
Application of siU6atac significantly decreased proliferation in hormone responsive LNCaP cells and in therapy-resistant L-AR, C4-2, 22RV1 and PM154 cell and organoids (
While the inventors targeted U6atac to inhibit the MiS, the minor spliceosome complex consists of other unique components, including snRNAs and snRNPs. Therefore, the inventors chose to inhibit the MiS by targeting RNPC3, another component of the MiS. Indeed, RNPC3 KD provoked similar reactions in PCa cells to siU6atac, indicating that the observed decrease in cell proliferation and viability can be attributed to a decrease in MiS activity in general and is not a U6atac-specific observation (
Finally, the inventors wanted to further explore the effects of MiS inhibition in a model system that captures PCa disease heterogeneity better than the previously used cell culture studies. Therefore, the inventors applied siRNA against U6atac in PCa patient-derived organoids that were cultured in 3D. Superimposing the RNAseq results of those organoids with clinical data from the SU2C study (
In summary, the inventors conducted the first-in-field evaluation of MiS in PCa using a multi-pronged approach that combined RNAseq, mass spectrophotometry, FACS, and scRNAseq. This demonstrates that MiS inhibition results in severe cell cycle defects through aberrant splicing of MIGs, which fundamentally impacts PCa progression, survival and lineage plasticity
The rational for the inventors' usage of reporter assays as splicing readout is that the assessment of splicing kinetics of endogenous genes is often complicated by multiple factors (chromatin architecture, rate of transcription, trans-acting factors and the activity of the nonsense mediated decay (NMD) pathway). In the field of splicing especially minor intron splicing there is therefore a long history of using splicing reporters like the one the inventors used in the current study. The rational especially when comparing splicing in different cell types or cancer stages is that the expression of the transfected splicing reporter is independent of the endogenous regulatory modalities. Thus, the same amount of expression of the reporter across different cell lines allows the inventors to truly interrogate the efficiency of MiS activity.
To assess the whether the changes seen by reporter assays reflect the changes of endogenous MIGs within each cell line. This question relates to the inventors' splicing kinetics analysis of CoA3 which the inventors performed to verify that siU6atac indeed impacts minor splicing of MIGs (
The inventors further analyzed the RNAseq datasets of the four cell lines representing different stages of PCa progression (LNCaP, C4-2, 22Rv1 and PM154). Specifically, the inventors analyzed the RNAseq data of the samples treated with scrambled siRNA which represents the de novo control transcriptome (
Interestingly the inventors observed that while therapy responsive LNCaPs expressed MIGs that play a role in anti-viral response, MIGs of C4-2 cells encompassed proteins involved in DNA damage and autophagy. MIGs of 22Rv1 included MAPK11 as well as tumor antigens (CTAG2) or splice factors (SRPK3) and the MIG repertoire of PM154 organoids included amongst others MAPK10, chromatin remodeling factors such as ACTL6B and several proteins involved in cytoskeletal signaling (CEP170, EML4, PDE6D) but also proteins involved in DNA damage repair (MSH3). One could thus argue that as PCa progresses, the cellular MIG pool favors cancer survival. Off note, the inventors also compare endogenous AS events in the RNAseq data of the samples treated with scrambled siRNA found only one AS event that is significantly different in all four cell lines, which is occurring more often in PM154. This event occurs in TSPYL2 which recently was shown to contribute to abiraterone, an ARSi, resistance in CRPC via CYP gene transcription regulation.
To answer the question why CRPC-NE cells/organoids display high MiS activity although they usually have low AR activity and if there are multiple mechanisms to regulate MiS activity, the inventors explored an AR independent mechanism of MiS regulation in NEPC. pP38MAPK was recently found to trigger neuroendocrine differentiation in LNCaP cells. Moreover p38MAPK has been implicated in therapy-resistant PCa cells to enhance invasion, metastasis, and immune evasion. Interestingly, p38MAPK is in part regulated by the rapid turnover of U6atac snRNA. The Dreyfuss lab showed that anisomycin mediated activation of the p38MAPK results in increased U6atac stability thereby activating the MiS. Based on these lines of evidence, the inventors hypothesized that in NE organoids p38MAPK regulates MiS activity.
To address this, the inventors tested this effect of anisomycin2 to activate p38MAPK in LNCaP and C4-2 PCa-adeno cells, and H660 PCa-NE cells as well as in 5 of the inventors' organoids (MSK8, MSK10, MSK16, PM154 and PM1262) transfected with a luc-splicing reporter assay constructs. According to a recently published study on patient derived organoids from MSKCC (designated MSK) and The inventorsill Cornell (designated PM), Tang et al. classify MSK8 as stem-cell like PCa-adeno, MSK16 and PM1262 organoids group into organoids which enrich for WNT signalling and MSK 10 and Pm154 genomically enrich for a NEPC signature. Surprisingly, the effect on minor intron splicing was only observed in PCa-NE cells and organoids (H660, PM154 and MSK10) thereby suggesting that MiS activity in neuroendocrine PCa might be regulated through the p38MAPK axis
To directly test this, the inventors explored the expression and minor intron splicing of all 4 p38MAPKs (MAPK11,12,13 and 14) in the inventors' omincs datasets. Integration of both siU6atac RNAseq and proteomics showed that MAPK14 and 13 might be the key components that could be leveraged to regulate MiS activity. The inventors found that all 4 p38MAPK family members display intron retention or AS in LNCaP, C4-2, 22Rv1 and PM154 cells and organoids upon a U6atac KD. Comparing RNAseq with proteome data, the inventors found that MAPK14 protein expression however is only decreased in 22Rv1 cells which are considered an intermediate between adeno and neuroendocrine stages and MAPK13, which promotes neurotoxicity and is required for prostate epithelial differentiation was only decreased in PM154 organoids.
Therefore, the inventors performed siRNA mediated KD of MAPK14 or 13 in LNCaP, C4-2, 22Rv1 and H660 cells as well as in MSK8,10,16, PM154 and PM1262 organoids along with the luc-splicing reporter assay (
To answer the question whether MIGs can influence mRNA levels within the transcriptome and whether U6atac is not impacting gene expression through its function in the MiS. Currently U6atac is only known to regulate minor intron splicing through its interaction with U4atac as part of the MiS. Knockdown of U6atac was predicted to result in elevated minor intron retention which was captured by the inventors' RNAseq analysis (
It is surprising that in this case the inventors do not observe significant downregulation in expression levels of MIGs with intron retention. However, there are multiple explanations for this outcome. First major NMD players including UPF1 and NCBP2 are themselves MIGs that show aberrant splicing. Thus, NMD pathway itself is compromised thereby allowing these aberrantly spliced transcripts to escape NMD, which is reflected in the lack of downregulation of MIGs. Despite this lack of downregulation, the aberrant splicing of MIGs will affect the function of the encoded protein because the majority of the transcripts fail to encode the full-length protein. In fact, the inventors have previously shown that aberrant splicing of MIGs through MiS inhibition results in transcripts that escape mRNA surveillance mechanisms8. Moreover, the inventors have previously also shown that aberrantly spliced MIG transcripts are bound to the ribosome as such production of novel potentially toxic proteins might contribute to the observed cell cycle defects. Finally, there is new literature support for intron retention as a distinct form of regulation that is not just restricted to NMD.
To analyze MIGs influencing transcriptome, MIGs execute varies functions that are disrupted upon aberrant splicing of these MIGs. For example, MIGs are splicing factors (SFs), subunits of polymerase II and III (Pol II, III), transcription factors (TFs), RNA binding proteins (RBPs), chromatin remodeling factors (CRFs) and ribonucleoprotein particles (RNPs). In all, U6atac KD results in aberrant splicing of MIGs, which partly escape NMD and are not observed to be downregulated, end up impacting transcription directly and/or indirectly the transcriptome of the cell.
As described above siU6atac mediated aberrant splicing of MIGs has a global transcriptomic effect. Since the inventors' objective was to understand how the primary defect of inhibiting MiS activity manifests in the cancer cell to ultimately result in cell cycle defect and cell death, the inventors endeavored to capture the overall transcriptome changes that underpin the cell cycle defect and cell death. Thus, initially the inventors emphasized genes either MIGs or non-MIGs that are affected by siU6atac.
Next, the inventors added additional PCa relevant MIGs to the listing such as JNK2 and JNK3, E2F1 and E2F2, RNPC3, Brd9, BRAF, ACTL6A, SMARCA1 and ACTL6B (also known as BAF53b), which the inventors recently described as a crucial NEPC biomarker. Below is stated a table that underlines the dynamic and context dependency of minor splicing in different PCa cell lines (Table 1). In that sense the inventors found that siU6atac triggered context dependent decrease in expression of many MIGs and non-MIGs with known functions in NEPC transdifferentiation such as the previously mentioned p38MAPK or BAF53B and BAF53A, EED, EZH2, AURKA which is likely based on intron retention and AS events (Table 1).
The primary defect in cell cycle is mediated by MIGs that are actively participating in the regulation of cell cycle. Importantly the inventors show that MIGs with elevated retention that are also downregulated at the protein level highly enrich for cell cycle regulation (GO-Term enrichment analysis). Thus, the primary molecular defect driving the observed cell cycle defect is aberrant splicing of MIGs. Additionally, it is now well established in the literature that disruption of minor splicing is associated with cell cycle defects, affecting S-phase, mitosis, and cytokinesis during brain development. Given that the inventors have demonstrated that siU6atac results in minor intron splicing defects similar to that observed in other model systems it is not a stretch to connect aberrant splicing to the observed cell cycle defects in PCa. Indeed, this is reflected in the inventors' molecular transcriptome/proteome intersection analysis (
Next, the inventors found that differential clustering noticeably improves when including more cancer types (and thus samples). In particular, the disparities in clustering based on MIGs and non-MIGs grew as a result of including more cancer types (i.e., the differences in Silhouette scores between MIGs and non-MIGs grew after including more cancer types in the inventors' analysis). The results from this revised analysis are given in
t-statistic and p-value associated with the slope:
Residual standard error: 0.0005889 on 9 degrees of freedom
Specifically, the inventors chose to include only those cancer types for which a sufficient number of samples with expression data are available. The inventors now only enforce that each cancer type be represented by a sufficiently large number of samples (we chose a threshold of 18). This enabled the inventors to retain 23 cancer types across a sample size of N=1224.
To answer the central question in the field of understanding PCa drug resistance to ARSi, whether MiS is indeed linked to the CRPC/NE transition in Pca, the authors previously reported that 10-15% of CRPC-adenos undergo lineage plasticity to a AR negative state, some express neuroendcrine surface markers and others are double negative. For neuroendocrine transdifferentiation, REST is a known critical factor.
RE1-silencing factor (REST) is a well-defined repressor of neural differentiation. Loss of REST expression has been associated with up-regulation of genes that are used to define CRPC-NE (e.g., SYP and CHGA). Moreover, loss of REST expression precedes neuroendocrine differentiation in PCa. As such loss of REST has been considered a potential NEPC driver PCa. Relevant to this study, REST is regulated through dynamic alternative splicing such that a miniexon in intron 3 when included results in a premature stop codon truncating the open reading frame. This truncated form of REST protein, referred to as REST4, cannot bind to the RE1 silencing element but can block REST FL/DNA contact. REST4 thus acts as a dominant negative and consequently increased levels of REST4 would result in inhibition of REST function enabling NE differentiation. The inventors observed that REST expression by qPCR is dynamic across the four different cell lines representing PCa progression (LNCaP>C4-2>22Rv1>PM154) (
When the inventors related this information to minor splicing, the inventors found that U6atac KD increases canonical REST after 96 h and decreases REST4 expression after 48-72 h in NE-like cells (22Rv1) and the NE-organoids (PM154) (
First, the inventors show that MIGs are crucial players in executing the oncogenic program downstream of PCa causing genes. This placement of MiGs positions the MiS as an important target for PCa therapeutics. Specifically, the inventors show that not only is the MiS directly downstream of AR-axis which blocks U6atac turnover (
Next, the inventors revisited the cell doubling (
In agreement with the inventors' assumption that MiS inhibition will disproportionally affect PCa cells over normal cells the inventors show that MS2514 (normal mouse prostate cells), that have a low doubling time (
To further address tumor selectivity, the inventors performed co-culture experiments with C4-2-GFP cancer cells and HS27-mCherry fibroblasts (
Taken together the inventors show that siU6atac mediated MiS inhibition is not just linked to proliferative rate, rather it is specifically affected in PCa as it uses a lot of MIGs to execute the oncogenic program. In that sense the inventors are currently building infrastructure to address the issue of therapeutic window to titrate MiS inhibition for translating the inventors' discovery into the clinics.
In PCa the inventors found that MiS downstream signaling encompasses critical MIGs such as BRAF; PARP1, EED, ACTL6A and B, MAPK8, 9, 13 and 14. The exact MIG repertoire is however cell type dependent. In that sense the inventors found for example that a U6atac KD triggers a decrease of ACTL6A transcript in LNCaP and C4-2 cells but not in 22Rv1 and PM154 cells (
Here the inventors identified PCa relevant MIGs via RNAseq, qRT-PCR (
In order to show that disruption of the PCa relevant MIGs primarily goes through U6atac the inventors have performed rescue experiments (
To test whether the inventors can rescue the cell death phenotype observed in siU6atac experiments, the inventors used CRISPR mediated KO of endogenous RNU6ATAC in cells overexpressing U6atac snRNA from a transfected plasmid in C4-2 cells (C4-2′U6atac cells) (
As expected overexpression of U6atac plasmid increased growth of C4-2 cells significantly (
The inventors performed the pulldown experiments with anti-PDCD7 ABs for 2 reasons: 1) anti-PDCD7 so far is the only reliable AB among all 7 unique MiS protein ABs. 2) Unlike major intron 5′ splice sites, minor intron 5′ splice sites are initially recognized and bound by a protein which is PDCD7. Since the inventors wanted to pull-down the pre-mRNAs, the inventors chose to perform the PD with this direct interactor protein.
For 1H and 11 the inventors have run the correlation analysis for all 7 proteins unique to the MiS as well as for ZRSR2 and CENPA which are proteins shared by minor and major spliceosome (
Of course, ZRSR protein loss of function in the testis has also reported major intron defect. Intriguingly the inventors observe while the majority of MiS proteins shows a trend towards a positive correlation with EZH2, ZMAT5 and PDCD7 show a trend towards a negative correlation with EZH2. Together these findings show that MiS and its components are dynamically regulated across PCa progression and the combinatorial changes in expression of the MiS components need further exploration. Moreover, these proteins have not been extensively studied as such the inventors do not understand other potential functions they might execute. For example, PDCD7, as the name suggests was identified as Programmed Cell Death Protein 7 gene and was reported as a potential transcription factor that activates the transcription e-cadherins. In all this analysis is one of the first to begin to place the entire MiS components in PCa progression. One caveat of course is that proteins of the MiS can have multiple functions, but snRNAs are currently thought to be exclusively involved in executing minor splicing.
Here the inventors identified the number of MIGs expressed in all three cell lines (LCR-515 MIGs) or the PM154 organoid (555 MIGs) and submitted these lists to DAVID, which yielded 81 and 99 significant GO terms, respectively. The inventors then intersected these GO terms, which reflect the functional enrichment of MIGs at baseline, with those generated by MIGs with elevated minor intron retention in the cell lines (39 GO terms) and the organoid (42 GO terms). For the cell lines, this intersectional approach identified 54 GO terms unique to MIGs expressed at baseline, 12 were unique to MIGs with retention, and 27 were shared. For the organoid, 70 GO terms were unique to MIGs expressed at baseline, 13 were unique to MIGs with retention, and 29 were shared. For both the cell lines and the organoid, the inventors found that: GO terms unique to MIGs expressed at baseline notably included cell cycle; spliceosomal complex assembly, RNA metabolic processes, DNA repair, small GTPase binding etc., shared GO terms included MAP kinase activity, nucleotide excision repair, snRNA transcription from RNA Polymerase II promoter, ATP binding, nuclear pore etc.; and GO terms unique to MIGs with retention included RAN GTPase binding, protein serin/threonine kinase activity, polA RNA binding and protein transport. Thus, the enrichment of pathways by mis-spliced MIGs reflects only some of the pathways that MIGs regulate at baseline, and also includes novel terms that do not emerge when only focusing on MIGs expressed at baseline.
In the PCa field there is active debate on classification on PCa stages. Therefore, the inventors have presented the classification scheme of the different PCa cell lines based on Bluemn et al. to clarify the inventors' analysis (
Here the inventors report for the first time that siU6atac and MiS inhibition is a potential therapeutic target. It will take considerably effort and time to translate this finding into the clinic to specifically treat therapy resistant lethal PCa. The inventors wanted to therefore show that this novel strategy is indeed a therapeutical viable approach by comparing it to the most exciting possible therapeutics in the field such as the combination of enzalutamide with EZH2 inhibition. This combinational approach is not FDA/EDA approved but there are several Phase 1 and 2 trials exploring this combination in CRPC. This combination approach is based on the demonstration by several groups (Mu, P. et al., Science 2017, 355 (6320), 84-88; Ku, S. Y et al., ibid) that EZH2 inhibitors re-sensitize therapy-resistant PCa towards enzalutamide treatment. Therefore, the significance of the finding of
MIG-expression discriminates cancer from benign tissue. MIG-expression and the pathways they regulate are intimately linked to oncogenes (
MiS activity increases with cancer progression. MIG expression is uniquely dependent on the splicing efficiency of minor introns. As such, it is not surprising that the inventors found differential efficiencies for the splicing of minor introns when the inventors compared different PCa subtypes. Indeed, studying U6atac levels during progression of prostate cancer showed that U6atac expression is closely correlated to the progression of Pca and is positively associated with Pca metastasis (
siU6atac blocks cell cycle of cancer cells. Given that U6atac snRNA has no other known function outside of the MiS, siU6atac specifically inhibited MiS function, which was reflected in elevated minor intron retention (
The cell type-specific response to MiS inhibition was reflected in the different numbers of MIGs with altered minor intron splicing for each cell line (
Perturbation of predicted biological pathways based on transcriptomic changes are ultimately implemented by the proteins produced. Mass spectrophotometry analysis showed downregulation of many MIGs that were aberrantly spliced. However, the inventors did not observe a one-to-one correlation (
ScRNAseq reveals reduction in AR- and EMT-score in siU6atac-treated cells and organoids. Cell cycle defects observed by FACS analysis (
The minor spliceosome is essential for PCa growth and viability. Whereas siU6atac-mediated MiS inhibition substantially impacted cancer cells, it did not strongly affect human fibroblasts or benign mouse prostate cells (
Interestingly, U6atac KD alone did not impact bladder cancer cells, but it was very effective when combined with cisplatin (
Taken together, the inventors show that MiS activity plays a crucial role in the progression of PCa and that MiS inhibition is a viable therapeutic target. The inventors show that inhibiting different MiS components can block cancer cell proliferation and viability, but in the inventors' study, U6atac was the most effective. The inventors show that MiS activity corresponds to AR signalling across stages of PCa progression, which is reflected by the increase in U6atac expression. Indeed, this discovery suggests that U6atac could also be employed as a diagnostic marker for lethal PCa and other cancers. Regardless of the stage of PCa, siU6atac can successfully inhibit proliferation and viability through disruption of pathways such as MAPK, cell cycle and DNA repair. While in the current study PCa is used as a model to investigate the efficacy of MiS inhibition as a therapeutic strategy, these observations could extend to other cancer types.
LNCaP (male, ATCC, RRID: CVCL_1379), C4-2 (male, ATCC, RRID: CRL-3314), 22Rv1 (male, ATCC, RRID: CRL-2505), PC3 (male, ATCC, RRID: CRL-3470), DLD-1 (2 (male, ATCC, RRID: CCL-221), L-rENZ and L-AR cells were maintained in RPMI medium (Gibco, A1049101), supplemented with 10% FBS (Gibco, 10270106), and 1% penicillin-streptomycin (Gibco, 11548876) on poly-L-lysine coated plates. RWPE cells (male, ATCC, RRID: CVCL_3791) were maintained in Keratinocyte Serum Free Medium (Gibco, 17005075) supplemented with bovine pituitary extract and human recombinant EGF (included), and 1% penicillin-streptomycin (Gibco, 11548876). HEK293T cells (female, ATCC, RRID: CVCL_0063), VCaP (male, ATCC, RRID: CRL-2876), MDA-MB-231 (female, ATCC, RRID: HTB-26), K-562 (female, ATCC, RRID: CCL-243), LN-18 (male, ATCC, RRID: CRL-2610) PC-3M-Pro4 and DU145 cells (male, ATCC, RRID: CVCL_0105) were maintained in DMEM (Gibco, 31966021), supplemented with 10% FBS, and 1% penicillin-streptomycin. NCI-H660 cells (male, ATCC, RRID: CRL-5813) were maintained in RPMI medium (Gibco, A1049101), supplemented with 5% FBS, 1% penicillin-streptomycin (Gibco, 11548876), 0.005 mg/ml Insulin (Sigma-Aldrich, 19278), 0.01 mg/ml Apo-Transferrin (Sigma-Aldrich, T1147), 30 nM Sodium selenite (Sigma-Aldrich, S9133), 10 nM Hydrocortisone (Sigma-Aldrich, H6909) 10 nM beta-estradiol (Sigma-Aldrich, E2257) and L-glutamine (for final conc. of 4 mM) (Sigma-Aldrich, G7513). PC-3M-Pro4 cells were a kind gift from Dr. Kruithof-De Julio. LNCaP-AR cells were a kind gift from Dr. Sawyers and Dr. Mu (Memorial Sloan Kettering Cancer Center). L-ENZ cells were established through constant enzalutamide exposure. Briefly low passaged LNCaP cells were treated over night with 20 uM enzalutamide in C/S media. The media was exchanged to normal RPMI (10% FBRS, 1% P/S) the next day and surviving LNCaP cells (˜10%) were maintained until they reached a confluency of ˜80%. This procedure was repeated twice. Subsequently the enzalutamide concentration was increased for three treatments to 40 uM and for 25 treatments to 80 uM. Cells are treated since them once a week with 80 uM enzalutamide.
All cell lines were grown at 37° C. with 5% CO2. All cell lines were authenticated by STR analysis and regularly (every 3 month) tested for mycoplasma.
MSKCC-PCa8,10,14 and 16 CRPC-Adeno patient derived organoids were a kind gift from Dr. Chen (Memorial Sloan Kettering Cancer Center). All organoids including WCM154 were maintained in three-dimension according to the previously described protocol (Puca, L. et al., Nat Commun 2018, 9 (1), 2404; Gao, D., et al., Cell 2014, 159 (1), 176-187). Briefly Advanced DMEM (Thermo Fisher Scientific, 31966047) with GlutaMAX 1× (Thermo Fisher Scientific, 35050061), HEPES 1 mM (Thermo Fisher Scientific, 15630056), AA 1× (Life Technologies, 15240-062), 1% penicillin-streptomycin, B27 (Thermo Fisher Scientific, 17504001), N-Acetylcysteine 1.25 mM (Sigma-Aldrich, A9165), Recombinant Murine EGF 50 ng/ml (PeproTech, 315-09), Human Recombinant FGF-10 20 ng/ml (Peprotech, 100-26), Recombinant Human FGF-basic 1 ng/ml (Peprotech, 100-18B), A-83-01 500 nM (Tocris, 29-391-0), SB202190 10 μM (Sigma-Aldrich, S7076), Nicotinaminde 10 mM (Sigma-Aldrich, N0636), (DiHydro) Testosterone 1 nM (Fluka, 10300), PGE2 1 μM (Tocris, 2296), Noggin conditioned media (5%) (PeproTech, 120-10C) and R-spondin conditioned media (5%) (PeproTech, 315-32). The final resuspended pellet was mixed with growth factor-reduced Matrigel (VWR, BDAA356239) in a 1:2 volume ratio. Droplets of 40 pl cell suspension/Matrigel mixture were pipetted onto each well of a six-well cell suspension culture plate (Huberlab, 7.657185) To solidify the droplets the plate was placed into a cell culture incubator at 37° C. and 5% CO2 for 30 min. Subsequently 3 ml of human organoid culture media was added to each well. 50% of the media was exchanged every 3-4 day during organoid growth. organoids were passaged as soon as they reached a size from 200 to 500 um. To this end, organoid droplets were mixed with TrypLE Express (Gibco) and placed in a water bath at 37° C. for a maximum of 5 min. The resulting cell clusters and single cells were washed and re-cultured, according to the protocol listed above.
Tissue micro-arrays were kindly provided by the Translational Research Unit (TRU) Platform, Bern (www.ngtma.com). For PCa the inventors used TMAs from the Bern PCBM cohort (Briganti, A. et al., Eur Urol 2013, 63 (4), 693-701) (28 patients) and a tissue microarray of 210 primary prostate tissues, part of the European Multicenter High Risk Prostate Cancer Clinical and Translational research group (EMPaCT) (Tosco, L. et al., Eur Urol Focus 2018, 4 (3), 369-375; Chys, B. et al., Front Oncol 2020, 10, 246; Dawson, H. et al., Histopathology 2020, 76 (4), 572-580).
LNCaP, C4-2, 22Rv1 and PM154 cells (400 000) were seeded in a 6 well and treated for 96 hours with siScrambled or siU6atac RNA (16 pmol). 96 hours post transfection cells were harvested and 50% of the cell pellet was used for U6atac KD confirmation by qRT-PCR. The remaining pellet was washed twice with PBS and subjected to mass spectrometry (MS) analysis:
Cells were lysed in 8M urea/100 mM Tris pH8/protease inhibitors with sonication for 1 minute on ice with 10 seconds intervals. The supernatant was reduced, alkylated and precipitated overnight. The pellet was re-suspended in 8M urea/50 mM Tris pH8 and protein concentration was determinate with Qubit Protein Assay (Invitrogen). 10 μg protein was digested with LysC 2 hours at 37C followed by Trypsin at room temperature overnight. 800 ng of digests were loaded in random order onto a pre-column (C18 PepMap 100, 5 μm, 100A, 300 μm i.d.×5 mm length) at a flow rate of 50 μL/min with solvent C (0.05% TFA in water/acetonitrile 98:2).
After loading, peptides were eluted in back flush mode onto a home packed analytical Nano-column (Reprosil Pur C18-AQ, 1.9 μm, 120A, 0.075 mm i.d.×500 mm length) using an acetonitrile gradient of 5% to 40% solvent B (0.1% Formic Acid in water/acetonitrile 4, 9:95) in 180 min at a flow rate of 250 nL/min. The column effluent was directly coupled to a Fusion LUMOS mass spectrometer (Thermo Fischer, Bremen; Germany) via a nano-spray ESI source.
Data acquisition was made in data dependent mode with precursor ion scans recorded in the orbitrap with resolution of 120′000 (at m/z=250) parallel to top speed fragment spectra of the most intense precursor ions in the Linear trap for a cycle time of 3 seconds maximum.
LV290591-RNU6ATAC Lentiviral Vector (Human) (CMV) (pLenti-GIII-CMV-GFP-2A-Puro) as well as the corresponding empty vector control were purchased from ABM. DNA was amplified via chemical transformation of One Shot Machi T1 Phage-Resistant Chemically Competent E. coli cells (Invitrogen, C862003). Lentivirus was produced in HEK293T cells by transfection with the constructs, and subsequent virus containing media was used to transduce C4-2 cells. Three days post transduction the cells were subjected to puromycin selection (1 pg/mL). After the selected cells reached a confluence of 80%, they were FACS sorted for GFP positivity. This was repeated 3 times.
For DHT stimulation experiments cells were starved of hormone for 48 hours in phenol red-free RPMI media (Gibco, 11-835-030) with 10% charcoal stripped FBS (Gibco, A3382101), then treated with 10 nM dihydrotestosterone (Fluka, 10300) for 24 hours.
For long-term ADT treatment cells were exposed weekly to 20 μM enzalutamide (Selleck Chemicals, S1250) or 10 uM Abiraterone.
For growth experiments cells were treated with siRNA and two hours later with 20 uM enzalutamide. Enzalutamide was refreshed 3 days later.
For splicing reporter assays cells were exposed to Anisomycin 1 ug/ml for 4 h (Sigma Aldrich, A9789).
L-AR cells (50 000) were seeded in p-Slide 8 The inventorsll (ibidi, 80826). The next day cells were washed once in ice-cold PBS and fixed in 4% PFA for 10 minutes. Subsequently cells were permeabilized with PBS+0.2% Triton for 10 minutes. Proximity ligation assay using the Duolink® In Situ Red Starter Kit Mouse/Rabbit (Sigma-Aldrich, DUO92101-1KT) was performed according to the manufacturer's instructions. Briefly primary monoclonal antibodies against mouse-AR (Thermo Fisher Scientific, MA5-13426), rabbit-HSP90 (Abcam, ab203085), rabbit-PDCD7 (Abcam, 121258) and rabbit IgG Isotype Control antibody (Thermo Fisher Scientific, 026102) were diluted in Duolink Antibody Diluent (1:50, 1:200, 1:100 and 1:1000). Cells were incubated in the AB solution over night at 4 C. The next day cells were washed twice and incubated for one hour at 37 C in a moisture chamber with PLUS and MINUS PLA probes. Subsequently cells were washed twice and incubated at 37 C (humidity chamber) for 30 min in the ligation mix and 100 minutes in the amplification solution. After two final washes for 10 minutes slides were mounted with DAPI containing media and monitored with a fluorescence microscope (LEICA, DMI4000 B).
Cell Transfection and siRNA-Mediated Knock-Down
ON-TARGET plus siRNA SMARTpool siRNAs against U6atac, AR, EZH2, PDCD7, mouse RNPC3 and the Non-targeting (siScrambled) siRNA were purchased from Dharmacon. siRNAs against RNU6atac and RNU12 and the Silencer Select Negative Control were purchased from Thermo Fisher Scientific and siRNA against mouse U6atac was purchased from Ambion. Transfection was performed for the respective timepoints on attached cells using the Lipofectamine™ RNAiMAX Transfection Reagent (Thermo Fisher Scientific, 13778150) to the proportions of 16 pmol of 20 μM siRNA per well.
Before transfection organoids were cultured for 2-3 weeks in human organoid growth medium. Media was removed and organoids were first mechanically dissociated. To obtain single cells organoids were trypsinized in 1 ml TripIE (Thermo Fisher Scientific, 12605036) for 15-18 minutes at 37 C. The reaction was stopped with 1 ml growth media and cells were spun for 5 minutes at 300 g. Subsequently the cells were strained and counted. Per condition one million cells were plated in a 6 well. Lipofectamine™ RNAiMAX complexes were prepared according to the standard Lipofectamine™ RNAiMAX protocol. In short, 5 ul of RNAiMAX reagent and 40 nM of siRNA plus 10% FBS were each diluted in 125 ul Opti-MEMH medium. Both mixes were pooled and incubated for 10 minutes before the siRNA-reagent complex was added to the cells. Cell/siRNA mix was centrifuged at 600 g at 32C for 60 min, and then incubated over night at 37 C. The next day cells were resuspended and collected by centrifugation (300 g, 5 min, RT). The pellet was resuspended in 280 ul Matrigel and the mix was separated into 7 drops that were added into a 6 well. Organoids were grown in human organoid media for 96 h (CTG assay) or seven days (cell counting assay).
RNA Extraction from Cells and qRT-PCR
Cells were harvested for RNA isolation using the ReliaPrep™ miRNA Cell and Tissue Miniprep System (Promega, Z6212). Synthesis of complementary DNAs (cDNAs) using FIREScript RT cDNA Synthesis Kit (Solis BioDyne, 06-15-00200) and real-time reverse transcription PCR (RT-PCR) assays using HOT FIREPol EvaGreen qPCR Mix Plus (Solis BioDyne, 08-24-00020) were performed using and applying the manufacturer protocols. Quantitative real-time PCR was performed on the ViiA 7 system (Applied Biosystems). All quantitative real-time PCR assays were carried out using three technical replicates. Relative quantification of quantitative real-time PCR data used GAPDH, ACTB as housekeeping genes.
Cell counting and viability assessments were conducted using a ViCell XR Cell counter and viability analyzer (Beckman Coulter, BA30273). Thereafter, GEM generation & barcoding, reverse transcription, cDNA amplification and 3′ gene expression library generation steps were all performed according to the Chromium Next GEM Single Cell 3′ Reagent Kits v3.1 User Guide (10× Genomics CG000204 Rev D) with all stipulated 10× Genomics reagents. Generally, 11.8.-27.5 μL of each cell suspension (600-1′400 cells/μL) and 15.7-31.4 μL of nuclease-free water were used for a targeted cell recovery of 10′000 cells. GEM generation was followed by a GEM-reverse transcription incubation, a clean-up step and 10-12 cycles of cDNA amplification. The resulting cDNA was evaluated for quantity and quality using a Thermo Fisher Scientific Qubit 4.0 fluorometer with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Q32854) and an Advanced Analytical Fragment Analyzer System using a Fragment Analyzer NGS Fragment Kit (Agilent, DNF-473), respectively. Thereafter, 3′ gene expression libraries were constructed using a sample index PCR step of 11-12 cycles. The generated cDNA libraries were tested for quantity and quality using fluorometry and capillary electrophoresis as described above. The cDNA libraries were pooled and sequenced with a loading concentration of 300 μM (150 μM in runs using XP workflow), paired end and single indexed, on an illumina NovaSeq 6000 sequencer using a NovaSeq 6000 S2 Reagent Kit v1.5 (100 cycles; illumina 20028316) and two NovaSeq 6000 S4 Reagent Kits v1.5 (200 cycles; illumina 20028313). The read set-up was as follows: read 1: 28 cycles, i7 index: 8 cycles, i5: 0 cycles and read 2: 91 cycles. The quality of the sequencing runs was assessed using illumina Sequencing Analysis Viewer (illumina version 2.4.7) and all base call files were demultiplexed and converted into FASTQ files using illumina bc12fastq conversion software v2.20. All steps were performed at the Next Generation Sequencing Platform, University of Bern.
Total RNA was extracted from LNCaP, C4-2, 22Rv1 and PM154 cells treated for 96 h with siU6atac or siScrambled. The recommended DNase treatment was included. The quantity and quality of the extracted RNA was assessed using a Thermo Fisher Scientific Qubit 4.0 fluorometer with the Qubit RNA BR Assay Kit (Thermo Fisher Scientific, Q10211) and an Advanced Analytical Fragment Analyzer System using a Fragment Analyzer RNA Kit (Agilent, DNF-471), respectively. Thereafter cDNA libraries were generated using an illumina Stranded Total RNA Prep, Ligation with Ribo-Zero Plus (illumina, 20040529) in combination with IDT for Illumina RNA UD Indexes Sets A and B (Illumina, 20040553 and 20040554, respectively). The illumina protocol was followed exactly with the recommended input of 100 ng total RNA. The quantity and quality of the generated NGS libraries were evaluated using a Thermo Fisher Scientific Qubit 4.0 fluorometer with the Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific, Q32854) and an Advanced Analytical Fragment Analyzer System using a Fragment Analyzer NGS Fragment Kit (Agilent, DNF-473), respectively. As a further quality control step, prior to NovaSeq 6000 sequencing, the pooled cDNA library pool underwent paired end sequencing using iSeq 100 i1 reagent v2, 300 cycles (illumina, 20040760) on an iSeq 100 sequencer. The library pool was re-pooled to ensure an equal number of reads/library and then paired end sequenced using a NovaSeq 6000 S4 reagent kits v1.5, 300 cycles (illumina, 20028312) on an Illumina NovaSeq 6000 instrument. The quality of the sequencing runs was assessed using illumina Sequencing Analysis Viewer (illumina version 2.4.7) and all base call files were demultiplexed and converted into FASTQ files using illumina bcl2fastq conversion software v2.20. The average number of reads/libraries was 82 million. The RNA quality-control assessments, generation of libraries and sequencing runs were performed at the Next Generation Sequencing Platform, University of Bern, Switzerland.
mRNA ISH was performed by automated staining using Bond RX (Leica Biosystems) and Basescope® technology (Advanced Cell Diagnostics, Hayward, CA, USA). All slides were dewaxed in Bond dewax solution (product code AR9222, Leica Biosystems) and heat-induced epitope retrieval at pH 9 in Tris buffer based (code AR9640, Leica Biosystems) for 15 min at 950 and Protease treatment for 5 min. The following probes from RNAscope 2.5 LS (Advanced Cell Diagnostics) were used: BaseScope™ LS Probe—BA-Hs-RNU6ATAC-1zz-st-C1ref 1039918, PPIB-1zz ref 710178 and DapB-1zz ref 701028, were used as positive and negative control respectively. Probe efficiency was tested using U6atac overexpressing C4-2 cells (5 million) of which 50% were treated with siU6atac RNA.
All probes were incubated at 370 for 120 min. Basescope™ 2.5 LS Assay (Ref 323600, Advanced Cell Diagnostics) was used as pre-amplification system. Subsequent the reaction was visualized using Fast red as red chromogen (Bond polymer Refine Red detection, Leica Biosystems, Ref DS9390) for 20 min. Finally, the samples were counterstained with Haematoxylin, air dried and mounted with Aquatex (Merck). Slides were scanned and photographed using Pannoramic 250 (3DHistech). U6atac intensity was scored manually by a pathologist (Mark Rubin) blinded to the clinical data, using the digital online TMA scoring tool Scorenado (University of Bern, Switzerland) especially developed for TMA scoring on de-arrayed spots.
For the analysis of TMA data, samples annotated as ‘center’ were used. U6ATAC score was calculated by multiplying the percentage of positive cells by the intensity. The sample with the highest score was used where more than one value was recorded for a block. Comparisons between groups were carried out using Wilcox test.
325 primary, 25 primary with metastatic potential and 32 metastatic samples, from 24 patients were used for the comparison of U6ATAC expression in PCa and PCBM.
C4-2 and PM154 cells were seeded in a 6 well (500 000/well) and transfected with siU6atac or siScrambled RNA for 72 and 96 hours as previously described. Flow Cytometry cell cycle analysis was performed using the Click-iT™ EdU Alexa Fluor™ 488 Flow Cytometry Assay Kit (Thermo Fisher Scientific, C10420). Briefly EdU (10 uM) was added into the media and cells were incubated for one hour at 37 C. cells were washed with 1% BSA in PBS and fixed in 100 ul Click-iT fixative for 15 minutes. After three additional washing step cells were permeabilized for 15 minutes in 100 ul 1×Click-iT saponin based reagent. Click-iT reaction cocktail was prepared according to manufacturer's instructions and 500 ul reaction mix/condition were incubated for 30 min with the cells at room temperature. Cells were washed and resuspended in 500 ul saponin-based permeabilization buffer. Hoechst (1 ug/ml) was added 20 minutes prior analysis to the reaction mix. Cells were analyzed using the FACSDiva Software on a BD LSR II Flow Cytometer (BD Biosciences) in the FACSIab Core facility of the University of Bern. Data was further quantified with FlowJo 10.7.1. Values were calculated as fold-change as compared to siScrambled treated controls.
Cells were lysed in GST-Fish buffer (10% (v/v) Glycerol, 50 mM Tris-HCl pH7.4, 100 mM NaCl, 1% (v/v) Nonidet P-40, 2 mM MgCl2, 1 mM PMSF) with freshly added protease and phosphatase inhibitors. Total protein concentration was measured using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). 50 ug protein samples were resolved 4-15% Mini-Protean TGX gels (BioRad, 456-1084) in SDS-PAGE and transferred to nitrocellulose membranes using the iBlot2 system (Thermo Fisher, 1B23001). Blots were blocked for 1 hour at room temperature in 5% milk/TBST or BSA/TBST and incubated overnight at 4° C. with primary antibodies which were dissolved in 5% BSA/TBST buffer. After 3 washes, the membrane was incubated with secondary antibody conjugated to horseradish peroxidase for 1 h at room temperature. After 3 washes, signal was visualized by chemiluminescence using the Luminata Forte substrate (Thermo Fisher Scientific, WBLUF0100) for strong antibodies and The inventorssternBright Sirius-HRP Substrate (Witec AG, K-12043-D10) for weak antibodies. Images were acquired with the FUSION FX7 EDGE Imaging System (Witec AG).
CMV-luc2CP/ARE (major intron splicing reporter), CMV-luc2CP (empty vector backbone control) and luc1CFH4 (minor splicing intron reporter) were a kind gift from Dr. Gideon Dreyfuss (University of Pennsylvania). DNA was amplified via chemical transformation of One Shot Mach1 T1 Phage-Resistant Chemically Competent E. coli cells (Invitrogen, C862003) and Sanger sequenced.
To determine the minor/major intron splicing rate cells were seeded in white 96 well plate (Huberlab, 7.655 098) (8000/well) and treated according to assay conditions. 24 hours prior analysis cells of each condition were co-transfected with each reporter plasmid and the empty vector backbone plasmid. In short 1.5 ul of P3000 reagent plus 0.5 ug of DNA and 1.5 ul Lipofectamine P300 were each diluted in 25 ul Opti-MEMH medium. Both mixes were pooled and incubated for 20 minutes before the solution was added to the cells. Luciferase expression was measured with the Dual-Glo® Luciferase Assay System (Promega, E2940): media was removed, 100 ul of PLB were added and cells were frozen for two hours at −20 C. After a one hour shaking step 100 ul of LAR substrate was added and Firefly luciferase expression was measured with a Tecan Infinite M200PRO reader. Values were calculated as x-fold of CMV-luc2CP expression and subsequently as the x-fold of the respective reference control.
pRSV2-p120-AmpR XL10 (p120), pCMV7.1 SCN4A frag (hSCN4A) and pLUX hSCN8A-minigene AmpR STbl3 (hSCN8a) were a kind gift from Dr. Mark-David Ruepp (King's College London). DNA was amplified via chemical transformation of One Shot Mach1 T1 Phage-Resistant Chemically Competent E. coli cells (Invitrogen, C862003).
To determine the minigene splice index cells were seeded in a 6-well (350 000/well) and treated according to assay conditions. SiRNA was added 96 hours prior measurement. 72 hours prior measurement cells were transfected with the respective minigene. Briefly 10 ul of P3000 reagent plus 1.5 ug of DNA and 7.5 ul Lipofectamine P3000 were each diluted in 125 ul Opti-MEMH medium. Both mixes were pooled and incubated for 20 minutes before the solution was added to the cells. 48 hours before the measurement media was exchanged for media with 10% charcoal stripped FBS (Gibco, A3382101) and 24 hours prior measurement 100 nM DHT was added to the respective condition. qRT-PCR was performed to verify the KD and to determine the splice index of each minigene. The minigene splice index was calculated by forming the ratio of normalized mRNA levels of cells transfected with the minigene versus mRNA levels of WT cells to consider the transfection efficiency. Subsequently the values corresponding to the spliced minigene were divided by the values corresponding to the unspliced minigene.
Cells were seeded in a 6 well (400 000) and treated according to assay conditions over night. Cells were then seeded in Poly-L-Lysine coated 96-well plates (8000 cells/well, n=3 per condition) and WCM154 cells were seeded in a collagen-coated 96-well plates (5000 cells/well, n=3 per condition). Remaining cells were used for U6atac KD control via qRT-PCR. Cell viability was determined after 24, 48, 72, and 96 h with a Tecan Infinite M200PRO reader using the CellTiter-Glo® Luminescent Cell Viability Assay according to manufacturer's directions (Promega, G9243). Viability values were calculated as x-fold of cells transfected with siRNA for 0 h. Cell confluence (n=4 per condition) was determined using the Incucyte S3 instrument and the IncuCyte S3 2018B software (Essen Bioscience, Germany). Values were calculated as x-fold of timepoint 0 and then as fold-change in confluency as compared to siScrambled treated controls.
Organoids were transfected with siRNA as described previously. The following day 160000 cells were resuspended in 320 ul Matrigel and drops of 40 ul (one drop/well, four timepoints, n=2) were plated in a suspension 48-well plate (Huberlab, 7.677 102). Remaining cells were plated in a 6-well for q RT-PCR U6atac KD control. The 24-well plate was incubated for three minutes in 37C and for 20 minutes upside down in 37 C. Subsequently 500 ul of organoid media were added and viability was measured using the CellTiter-Glo 3D Cell Viability Assay (Promega, G9683).
For the co-immunoprecipitation (co-IP), cytosolic fractions of LNCaP-AR cells were isolated using the Universal ColP Kit (Active Motif, 54002). Chromatin of the cytosolic fraction was mechanically sheared using a Dounce homogenizer Fisher Scientific, 11898502). Cytosolic membrane and debris were pelleted by centrifugation and protein concentration of the cleared lysate was determined with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, 23227). One microgram of the rabbit anti-PDCD7 (ab131258, Abcam) and rabbit anti-AR (ab133273, Abcam) antibodies and 1 μg of rabbit IgG Isotype Control antibody (Thermo Fisher Scientific, 026102) were incubated with 1 mg protein supernatant overnight at 4° C. with gentle rotation. The following morning, 30 μl of Protein G Magnetic Beads (Active Motif were washed twice with 500 μl ColP buffer and incubated with Antibody-containing lysate for 2 hours at 4° C. with gentle rotation. Bead-bound PDCD7 or AR complexes were washed twice with ColP buffer and subsequently twice with a buffer containing 150 mM NaCl, 50 mM Tris-HCL (pH 8) and Protease and Phosphatase inhibitors. Washing procedure was executed at 4° C. with gentle rotation. Bead-bound protein, supernatant and Input controls were reduced and denatured in 40 μl Laemmli buffer containing DTT through boiling for 5 min at 95° C. Magnetic beads were removed from solution using Magnetized Pipette Racks (Thermo Fisher Scientific, 11757325) and 20 μl of reduce protein was loaded on an SDS-PAGE gel with subsequent immunoblotting using iBlot (Life Technologies). Membranes were blocked in 5% BSA solution and then incubated over night with respective antibodies against targets of interest: AR, PDCD7, AR45 (Androgen Receptor Antibody (Carboxy-terminal Antigen), Cell Signaling Technologies, 54653S), ZRSR2 (Abcam, ab223062). Protein signal was detected using HRP-labeled native anti-rabbit IgG antibody (CST, #5127) and Luminata Forte substrate (Thermo Fisher Scientific, WBLUF0100) using the FUSION FX7 EDGE Imaging System (Witec AG).
Assuming that cancer genes perturb a large molecular network through their interactions with other genes, distance to MIGs in a protein-protein interaction network of 160,881 interactions between 15,366 human proteins as of the HINT database was determined (Das, J., et al., ibid). In particular, a MIG that directly interacts with a cancer-causing gene is a distance d=1 away, while a protein that is separated by a protein in between is a distance d=2 away from a cancer-genes in question. To find these interaction distances, a list of 403 cancer-causing genes from 186 different cancer types as of the Cancer Genome Interpreter database was considered (https://www.cancergenomeinterpreter.org) in the underlying protein-protein interaction network of 160,881 interactions between 15,366 human proteins as of the HINT database21. A profile of the numbers of MIGs with a given distance d away from each cancer gene, NdMIG (d=1, . . . , 5) was obtained. To assess if the presence of MIGs in the vicinity of cancer genes is significant, sets of 542 MIGs found in the underlying interaction networks were randomly sampled. In particular, the corresponding distances of cancer genes to these randomly sampled MIGs was measured and a profile of numbers of random MIGs with a given distance d away from cancer genes Ndr,MIG (d=1, . . . , 5) was obtained. The enrichment of MIGs is defined a distance d away from cancer genes as
with average E over 100,000 random samples of MIGs.
To find a tight-knit web of MIGs and cancer genes, such direct interactions of cancer genes and MIGs were distracted in the underlying human protein-protein interaction network and determined the size of the largest connected subnetwork S. Randomly sampling MIGs N=100,000 times. The sizes of the largest subnetworks was calculated using these random sets, Sr, and the significance of S by
was determined.
MIG expression data from 3 sources: prostate samples from GTEx (healthy prostate tissue), prostate cancer samples from TCGA (primary prostate cancer samples), and prostate cancer samples from SU2C (advanced prostate cancer) was merged. Gene expression values were normalized following the protocol adopted by GTEx (and detailed in the inventors' pan-cancer analysis). PCA analysis on this normalized gene expression matrix was carried out, thereby enabling visualization of the resultant data as the projections onto the space spanned by the first 2 PCs. Notably, this visualization appears to capture the progression from these 3 broad phases of prostate cancer progression, from healthy tissue in GTEx (at lower ends of the first PC) to advanced stages in SU2C (at higher ends of the first PC).
The Silhouette coefficient64 was used in order to characterize the relative performance of gene expression clustering for gene sets containing different relative abundances of MIGs and non-MIGs. The Silhouette coefficient provides an objective metric for measuring what is visually discerned to be structure (or any lack thereof) in a given heatmap. This coefficient constitutes an unsupervised approach to provide a score ranging −1 and 1, with scores closer to 1 indicative of well-defined and dense clustering (i.e., more meaningful structure in a given heatmap). This coefficient quantifies how similar a given data point (i.e., sample) is to its own cluster relative to different clusters. In the inventors' study, a data point consists of a N-length vector, where N is the number of distinct samples in an expression matrix, and the number of distinct clusters is pre-defined to be n_cluster=9 in the inventors' pan-cancer analysis, since this analysis was carried out on a dataset of 9 distinct cancer types. The Silhouette coefficient has also been adopted for similar purposes in previous studies (Belorkar, A. et al., BMC Bioinformatics 2016, 17 (Suppl 17), 540; Van Laere, S. J. et al., Clin Cancer Res 2013, 19 (17), 4685-96; Zhao, S. et al., Biol Proced Online 2018, 20, 5; Lovmar, L. et al., BMC Genomics 2005, 6, 35).
Prior to calculating the Silhouette coefficient, an agglomerative clustering on a normalized gene expression matrix was performed. Gene expression values from 9 cancer types (Biliary-, Breast-, ColoRect-, Lung-, Ovary-, Panc-, Prost-, Stomach-, Thy-adenocarcinoma) totaling N=444 samples were taken from PCAWG, and the gene expression normalization was performed using the same approach as that adopted by GTEx (PMID: 29022597). Briefly, the entire gene expression matrix was normalized using quantile normalization. Then, inverse quantile normalization was applied to this quantile-normalized matrix in order to map to a standard normal (this also enabled the inventors to remove outliers). For the PCa progression analysis, gene expression values from GTEx, TCGA, and SU2C were merged into one expression matrix. Briefly the pre-normalized expression matrix was normalized using quantile normalization, followed by inverse quantile normalization to map to a standard normal distribution; this step also removed outlier genes. This normalization scheme was the same as that adopted previously by GTEx (PMID: 29022597). Using this normalized gene expression matrix, clustering was performed by using hierarchical clustering by employing the Euclidean metric and Ward linkage.
The entire analysis was run using different expression matrices with varying fractions of MIG genes, and the results are plotted in
Primary tumor RNA-seq patient samples from The Cancer Genome Atlas (TCGA) were randomly queried using the GenomicDataCommons package in R (https://github.com/Bioconductor/GenomicDataCommons). The minor intron retention pipeline developed by Olthof et. al. (https://github.com/amolthof/minor-intron-retention) was run on the queried TCGA samples from the following cohorts: breast invasive carcinoma (BRCA; N=20), cholangiocarcinoma (CHOL; N=20), colon adenocarcinoma (COAD; N=20), lung adenocarcinoma (LUAD; N=20), ovarian serous cystadenocarcinoma (OV; N=20), pancreatic adenocarcinoma (PAAD; N=20), prostate adenocarcinoma (PRAD; N=16), and thyroid carcinoma (THCA; N=20). For the prostate cancer analyses, the following additional samples from the Stand Up to Cancer dataset were analyzed: androgen receptor (AR; N=20) and neuroendocrine (NE; N=22). NE samples were samples which had an NEPC score of greater than 0.4 while AR samples had an NEPC score of less than or equal to 0.4. Prostate cancer samples from the Genome-Tissue Expression Portal were analyzed as well (GTEX; N=20). A Kruskal-Wallis with post-hoc Dunn's Test was performed between the TCGA cohorts and the prostate cohorts. A heatmap was generated with the gplots package in R with the default clustering method for the pan-cancer TCGA cohorts and the prostate cohorts across the MIGs (https://CRAN_R-pr2ect.org/package=gplots). A GO Enrichment Analysis was performed on the genes that clustered for each cancer cohort, grouping genes with a mean MSI value in the ranges of 0, 0 to 0.04, 0.04 to 0.75, and 0.75 to 1.
Gene-expression data of primary prostate cancer specimen was retrieved from The Cancer Genome Atlas (TCGA) in form of raw-counts. Sequencing reads were aligned to the human reference genome (hg38) using STAR (Lovmar, L. et al., BMC Genomics 2005, 6, 35). Gene-expression was quantified at gene-level using Gencode annotations (v29) (Harrow, J. et al., Genome Res 2012, 22 (9), 1760-74). Subsequent analysis and library-size normalization were performed using edgeR pipeline (Chen, Y. et al., F1000Res 2016, 5, 1438). RNU11 mapping reads were identified in 23 out of 497 samples (5%) which reflects the difficulty of capturing this gene product using canonical PolyA+ sequencing techniques. Nonetheless, a clear association between Gleason score and RNU11 mRNA expression in these 23 samples was identified. Significance was assessed using non-parametric Wilcoxon Test.
Cell ranger analysis pipeline v6.0.1 was used to align reads to the human genome reference sequence (GRCh38) and generate a gene-cell matrix from these data. The gene expression matrix was analyzed using Seurat 4.0.3 (https://github.com/satijalab/seurat). The inventors removed low quality cells and multiplets by excluding genes detected in less than 5 cells and by discarding cells with more than 10000 fewer than 1000 detected genes. Cells containing mitochondrial gene counts greater than 25% were also removed.
UMI counts were normalized with the NormalizeData Seurat function using the LogNormalize normalization method with default parameters (10000 scale.factor).
Prediction of cell cycle phase for each cell was performed using Seurat CellCycleScoring function. A score was computed and a cell phase (G2/M, S and G1) was assigned to the cell as described previously (Tirosh, I. et al., Science 2016, 352 (6282), 189-196). Fisher's exact test was performed to check whether the slU6 cells have significantly a different number of cells than the Scr in G1 or S phase using the R fisher.test function.
The inventors used Seurat AddModuleScore function to evaluate the degree to which individual cells express a certain pre-defined gene set. The inventors defined scores to estimate the activities of prostate AR pathway, and EMT state, as described previously (Dong, B. et al., Communications Biology 2020, 3 (1), 778). The AR pathway gene set included AR, KLK3, KLK2, FKBP5, TMPRSS2, FOXA1, GATA2, SLC45A3 and EMT state CDH2, CDH11, FN1, VIM, TWIST1, SNAI1, ZEB1, ZEB2 and DCN. Violin plots were drawn using Seurat and p-values were calculated using Wilcoxon test (Hadley Wickham, ggplot2: Elegant Graphics for Data Analysis 2016).
For merging multiple datasets and minimizing the batch effect between them, the inventors integrated the inventors' 6 samples (3 replicates SCR and 3 replicates siU6) for each cell line following the procedure of Seurat v4.0.3 (Stuart, T. et al., Cell 2019, 177 (7), 1888-1902.e21).
Briefly, the inventors selected the most variable genes for each dataset using the FindVariableFeatures function (selection.method=“vst”) and ranked them according to the number of datasets in which they were independently identified as highly variable. The 2000 most variable genes were thus integrated by merging pairs of datasets according to a given distance.
Integration anchors, representing two cells that are predicted to originate from a common biological state in both datasets using a Canonical Correlation Analysis (CCA), were done using the FindIntegrationAnchors function. The expression of the target dataset was corrected using the difference in expression between the two expression vectors for each pair of anchor cells. This step was performed using the IntegrateData function. This process resulted in an expression matrix containing the batch-effect-corrected expression for the 2000 selected genes for all cells from the 6 samples for each cell line.
A PCA was performed on the scaled data using RunPCA Seurat function (npcs=30). Uniform Manifold Approximation and Projection (UMAP), a nonlinear dimension reduction method, was run using RunUMAP Seurat package function in order to embed cells in a 2-dimensional space. A K-nearest neighbor graph (KNN) based on the Euclidean distance in PCA space was constructed to cluster the cells with the Louvain algorithm (resolution=0.2) using the FindNeighbors and FindClusters Seurat functions. The inventors selected the optimal clustering resolution using the clustree R package (v0.4.3) (Zappia, L., et al., GigaScience 2018, 7 (7)). Barplots were performed using dittoSeq R package (Bunis, D. G. et al., Bioinformatics 2020, 36 (22-23), 5535-5536).
Differentially expressed genes (DEGs) were identified between the different clusters using the FindAllMarkers function from the Seurat package (one-tailed Wilcoxon rank-sum test, p values adjusted for multiple testing using the Bonferroni correction). To compute DEGs, all genes were tested provided they were expressed in at least 25% of cells in either of the two compared populations, and the expression difference on a natural log scale was at least 0.25. The heatmap was produced using the DoHeatmap Seurat function by selecting the top five genes for each cluster.
Paired-end, total RNA reads for each replicate (N=4) were mapped to the mm10 genome via Hisat2 (Kim et al., 2015). Reads mapping to multiple locations were removed. Gene expression values were calculated using IsoEM2 (Mandric et al., 2017). Differential gene expression was calculated by IsoDE2 (Mandric et al., 2017), which uses 200 rounds of iterative bootstrapping to produce a 95% confidence interval for the expression of each gene, then statistically compares these values between experimental conditions. A threshold of log 2FC≥1, P<0.01 for upregulation, and log 2FC <−1, P s 0.01 for downregulation was employed.
The inventors report here minor intron retention as a mis-splicing index through the methodology described in Olthof et al., 2019. Briefly, uniquely mapped reads from the region of interest around minor introns (from two exons upstream to two exons downstream) were extracted. The mis-splicing index was then calculated by summing reads that map to the 5′ splice site and 3′ splice site of a minor intron, divided by the sum of reads that map to the 5′ splice site, 3′ splice site, and 2× canonically spliced reads. The inventors only considered introns that pass the inventors' filtering criteria, which requires >4 exon-intron boundary reads, >1 read mapping to both the 5′ splice site and 3′ splice site, and >95% intron coverage, in all replicates of a condition as retained. Statistically significant global minor intron retention was determined using a Kruskal-Wallis test with post-hoc Dunn's test (P≤0.05). Determination of individual MIGs with significantly elevated minor intron retention was calculated using a two-tailed student's T-test (P s 0.05).
The inventors employed the methodology reported by Olthof et al., 2017 for the inventors' alternative splicing analysis. Briefly, the inventors used BEDTools to classify differential 5′ splice site and 3′ splice site usage around the region of interest for all minor introns and binned them into one of 8 categories (
Gene lists were submitted to DAVID for gene ontology (GO) enrichment. The inventors considered only GO Terms with Benjamini-Hochberg adjusted P-value s 0.05 as significant.
For LNCaP. C4-2, and 22Rv1, overlapping MIGs with significantly elevated minor intron retention that were also found to be associated with prostate cancer-causing genes were grouped with overlapping protein coding genes with significant downregulation. This list was submitted to STRING under the default parameters to obtain the gene interaction network. Subsequently, the same list was submitted to IPA as a core analysis using default parameters. All reported biological networks and pathways from IPA were significant using a Benjamini-Hochberg adjusted P-value s 0.05 cut-off. The same analysis was performed for PM154 alone.
Principal component analyses were performed using clustvis (Metsalu & Vilo, 2015) with default parameters. Ellipses show 95% confidence interval.
To perform gene set enrichment analysis RNAseq data was pre-ranked using the metric: log 10(Pvalue)/sign (log FC). GSEA was performed using the GSEA v.4.0.3 software. Hallmark gene sets, obtained from the GSEA website (www.broadinstitute.org/Qsea) was used for enrichment of siU6atac related pathway genes. Dotplot was used to visualize the most significant enriched terms. Normalised enrichment score (NES) and False discovery rate (FDR) were applied to sort siU6atac pathway enrichment after gene set permutations were performed 1000 times for the analysis.
MS data was interpreted with MaxQuant (version 1.6.1.0) against a SwissProt human database (release 2019_07) using the default MaxQuant settings, allowed mass deviation for precursor ions of 10 ppm for the first search and maximum peptide mass of 5500 Da; match between runs with a matching time window of 0.7 min was activated, but prevented between different groups of replicates by the use of non-consecutive fractions. Furthermore, the four cell lines were treated as different parameter groups and normalized independently. Settings that differed from the default also included: strict trypsin cleavage rule allowing for 3 missed cleavages, fixed carbamidomethylation of cysteines, variable oxidation of methionines and acetylation of protein N-termini.
Protein intensities are reported as MaxQuant's Label Free Quantification (LFQ) values, as well as Top3 values (sum of the intensities of the three most intense peptides); for the latter, variance stabilization was used for the peptide normalization. Missing peptide intensities were imputed in the following manner, provided there was at least one identification in the group: two missing values in a group of replicates would be replaced by draws from a Gaussian distribution of width 0.3×sample standard deviation centred at the sample distribution mean minus 1.8× the sample standard deviation, whereas a single missing value per group would be replaced following the Maximum Likelihood Estlimation (MLE) method. Imputation at protein level for both LFQ and Top3 values was performed if there were at least two measured intensities in at least one group of replicates; missing values in this case were drawn from a Gaussian distribution of width 0.3× sample standard deviation centered at the sample distribution mean minus 2.5× the sample standard deviation. Differential expression tests were performed using the moderated t-test empirical Bayes (R function EBayes from the limma package version 3.40.6) on imputed LFQ and Top3 protein intensities. The Benjamini and Hochberg method was further applied to correct for multiple testing. The criterion for statistically significant differential expression is that the maximum adjusted p-value for large fold changes is 0.05, and that this maximum decrease asymptotically to 0 as the log 2 fold change of 1 is approached (with a curve parameter of one time the overall standard deviation). The protein imputation step was repeated 20× so as to be able to flag those proteins that are persistently significantly differentially expressed throughout the cycles.
Imputed iTop3 was used to calculate relative protein abundances. Differential expression was calculated using the Empirical Bayes test. Protein upregulation and downregulation was determined by setting a threshold of Benjamini-Hochberg adjusted P-value <0.05, log 2FC >1 or <−1, respectively.
| Number | Date | Country | Kind |
|---|---|---|---|
| 21191594.7 | Aug 2021 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2022/072931 | 8/17/2022 | WO |
