TECHNICAL FIELD

The present invention relates to the field of biotechnology, cancer cell biology and molecular medicine. In particular, the present invention is aimed at an in vitro method for predicting the result of the treatment with temozolomide (TMZ) in a subject that suffers from central nervous system tumours, essentially gliomas and within this group the most malignant subtype is known as glioblastoma, based on the identification of the SOX1 biomarker present in a sample from the subject to be treated. It could also be used in other types of cancer where it is identified that SOX1 has an oncogenic role.

As such, the present invention particularly relates to SOX1 as a prognostic and predictive biomarker in the treatment of central nervous system tumours, the use of the SOX1 transcription factor as a prognostic and predictive biomarker in the treatment of glioblastoma multiforme specifically.

BACKGROUND OF THE INVENTION

Glioblastoma multiforme is the most common tumour with the highest mortality rate in adults. In Spain, the incidence rate ranges between 3 and 4 cases per 100,000 inhabitants. Less than 3% of all patients diagnosed with glioblastoma survive more than four years and the average survival is 12 months.

Current therapy includes surgery followed by radiotherapy and treatment with temozolomide.

Temozolomide belongs to a group of anticancer drugs called alkylating agents. Temozolomide transforms in the organism into another compound called 3-methyl- (triazen-1-yl)imidazole-4-carboximide (MTIC), which binds to the DNA of the cells when they reproduce, which slows cell division. As a result, the cancerous cell cannot reproduce, thus slowing the progression of the tumours.

In this regard, it is well known that methylguanine-DNA methyltransferase (MGMT), a DNA repairing enzyme, is responsible for removing lesions in tumour DNA caused by alkylating agents and that tumours that have high levels of MGMT activity reduce the response to chemotherapy with temozolomide.

As a result, chemotherapy with temozolomide is essentially indicated in those tumours in which the promoter of the MGMT gene is methylated, that is, tumours in which the MGMT DNA repairing activity is reduced, as a result of which the sensitivity to agents that cause alkylation in DNA, as is the case of temozolomide, is increased. That is, MGMT is used as a biomarker to determine the response of the treatment with temozolomide in glioblastoma multiforme.

However, this is not always true clinically and tumours with methylation of the MGMT gene continue to be refractory to the treatment with temozolomide.

The patent application WO2015200823 describes a method for determining the resistance to glioblastoma treatment with a TGF-βΙ inhibitor, which comprises detecting at least one of the following proteins: TGFBI, ITGA7, TNC, DDR2, MRC2, MGST1 , CLCC1 , PTGFRN, CRTAP, CD109, SLC16A1 , CD47, MYOF, ABCA1 , S100A10, CA12, SLC16A3, etc.

Furthermore, the Korean patent application KR20160107745 describes the use of a RRAD inhibitor for treating glioblastomas in patients that have temozolomide resistance.

The US patent application US2015141439 proposes an azathioprine composition for treating glioblastomas in patients that have temozolomide resistance.

The US patent application US201 1091375 describes the combination of temozolomide with TNFa for treating glioblastoma, and lastly, the US patent application US201021051 1 describes the composition of temozolomide and a VEGFR2 inhibitor for treating glioblastoma.

Notwithstanding the previously proposed solutions, there is still the need to identify genetic characteristics responsible for tumour development in glioblastoma multiforme in order to improve the current treatments of the illness aimed at improving survival.

In recent years, different studies have provided a high resolution image of the genetic, epigenetic and transcriptomic landscape of glioblastoma, revealing a large number of genetic mutations and molecular alterations that lead to the pathogenesis of the illness and establish this tumour type as a heterogeneous collection of different illnesses that can be biologically and clinically stratified into relevant subgroups.

In particular, the expression of the identifying markers of the different subtypes varies among the individual cells inside a tumour and there is also significant intra- tumour heterogeneity in glioblastoma tumours. This heterogeneity is a major challenge for targeted therapy and personalised medicine.

There is evidence indicating that transcription factors responsible for decisions during embryonic development can also function as oncogenes by means of the promotion of the re-acquisition of active programs during embryonic development and which are necessary for tumorigenesis. Furthermore, certain malignant tumours depend on a cell hierarchy, with privileged subpopulations called cancer stem cells (CSCs), which promote the propagation and growth of the tumour.

As such, molecular and genetic programs that are active during embryonic development, regulating the population of stem and progenitor cells, re-emerge in CSCs in order to support the progressive development and growth of tumours.

Significant progress has been made in the identification of the molecular mechanisms underlying the pathobiology and intra-tumour heterogeneity of glioblastoma. However, further clarification is required of the development programs that govern glioma stem cells (GSC) and the progression of glioblastoma in order to accelerate the development of new therapeutic targets and urgently needed treatments.

SOX (the region determining the sex Y (SRY)-box) are a family of transcription factors characterised for containing a DNA-binding domain of the conserved high mobility group (HMG). There are 20 members in humans, which are divided into 8 groups based on their HMG sequence identity. The members within the same group can have superimposed expression patterns, share biochemical properties, being able to perform synergistic or opposite functions depending on the cell tissue and context. As such, there are several cases where members of the same subgroup develop well- differentiated activities within a specific tissue. The members of the SOX family play a crucial role both in embryonic and postnatal development. They are also important for regulating and maintaining stem cells, particularly in the Central Nervous System.

There is a growing amount of evidence that shows that several members of SOX are involved in the development of cancer. In general, they play a role in tumours that arise in tissues that are superimposed with the expression pattern thereof during embryonic development. In particular, some members of the family are oncogenes, while others act as tumour suppressors. For example, SOX2, SOX4, SOX9 or SOX10 show various oncogenic functions in different types of cancers, including glioblastoma. However, SOX17, SOX1 1 and SOX1 have been shown to act as tumour suppressors in certain types of cancer, such as gastrointestinal cancers, mantle cell lymphomas, and also glioblastoma.

The transcription factor SOX1 is a member of the subgroup SOXB1 , which also contains SOX2 and SOX3. It is a well-established tumour suppressor in ovarian, hepatocellular, cervical and nasopharyngeal cancers and it is commonly silenced by hypermethylation of the promoting region thereof. These findings support the idea that hypermethylation of the promoter of tumour suppressor genes is a major contributor to carcinogenesis.

Mechanically, SOX1 functions as a tumour suppressor through interaction with β-catenin, and the resulting inhibition of the Wnt signalling pathway. SOX1 is expressed since embryonic development and its expression in adults has been linked to stem cells, mainly in the brain. During development, the members of the subgroup SOXB1 show different and superimposed expression patterns. SOX1 is the first known specific marker of neuroectodermal lineage, which is activated during gastrulation. The other members, SOX2 and SOX3, show wider expression patterns that are activated in the preimplantation and epiblast stages, respectively.

In the brain, several reports have shown that SOX1 is a key regulator of the identity of neural progenitors and the determination of the purpose of neuronal cells, maintaining the capacity of these cells to proliferate or differentiate themselves since the early development of adult stages. Although the members of the group SOXB1 are co-expressed in a subpopulation of neural stem cells and show a certain level of functional redundancy, their activity does not overlap since, while SOX2 is mainly expressed in quiescent stem cells, SOX1 and SOX3 are more closely identified with active stem cells and transient progenitor cells.

The role of SOX1 in cancer has been described in several tissues, such as liver, lung, breast, ovarian, nasopharyngeal or cervical tissue, and it acts as a tumour suppressor in all of them. In a high percentage of these tumours, the expression of SOX1 is reduced due to the fact that the promoter thereof is methylated. Significantly, the silencing of SOX1 is correlated to lower patient survival in liver cancer, oesophageal cancer and ovarian cancer. Moreover, studies with different cell lines have shown that the overexpression thereof reduces tumour proliferation and activity, while the silencing thereof promotes tumour formation and progression.

With regards to the activity of the members of SOXB1 in glioblastoma, the oncogenic function and clinical relevance of SOX2 is well established, the majority of the functions thereof being linked to GSC. On the other hand, little is known about the expression or function of SOX1 and SOX3. Interestingly, the transcriptomic analysis of glioma cells with SOX2 silencing identified SOX1 and SOX18 as members of the SOX family among the almost 500 genes whose expression was altered. In the case of SOX1 , the effect was the reduction in its expression, suggesting that both SOX2 and SOX1 genes can in some way be related in glioblastoma. This fact would be an exception since it is well known that in tumours where SOX1 is silenced and acts as a tumour suppressor (liver, ovarian, cervical, breast, oesophageal or lung cancer), SOX2 is increased and its role is oncogenic.

Surprisingly and contrary to existing knowledge on the function of SOX1 in various tumours, the present invention provides new in vitro methods for the prognosis and prediction of the treatment of glioblastoma multiforme based on the use of SOX1 as a biomarker.

OBJECT OF THE INVENTION The main object of the present invention is an in vitro method for determining the response of a subject suffering from glioblastoma and other central nervous system tumours to the treatment with temozolomide that comprises detecting and quantifying the expression levels of SOX1 in a sample of the sick subject, and comparing said expression levels with respect to a reference value, wherein an expression level of SOX1 lower than the reference value indicates a good response to the treatment with temozolomide.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma multiforme or other central nervous system tumours to the treatment with temozolomide that comprises the above characteristics, wherein the reference value is that obtained in a control sample of a healthy individual, for example, of normal brain tissue.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide, as defined above, wherein detecting and quantifying the expression levels of SOX1 , and comparing them to a reference value, includes detecting and quantifying the levels of SOX1 protein in a sample of a sick subject and comparing said expression levels with respect to the reference value measured in a control sample of a healthy individual, for example, of normal brain tissue.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide that includes the above characteristics, wherein the detection of the expression levels of the SOX1 protein is carried out through the extraction of RNA from the sample of the sick subject, such as for example, of tumour brain tissue, the reverse transcription thereof and qRT-PCR.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide as defined above, wherein the quantification of the expression levels of the SOX1 protein is carried out through immunohistochemistry, immunocytochemistry, ELISA or Western blot.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide according to the foregoing, wherein detecting and quantifying the expression levels of SOX1 , and comparing them to a reference value includes detecting, quantifying and comparing the expression levels of the mRNA of the SOX1 gene. An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide according to the foregoing, wherein the detection, quantification and comparison of the expression levels of the mRNA of the SOX1 gene is carried out through qRT-PCR, after total RNA extraction and reverse transcription.

An object of the present invention is also an in vitro method for determining the response of a subject suffering from glioblastoma or other central nervous system tumours to the treatment with temozolomide according to the foregoing, wherein said sample is a sample of a sick subject, that is, a sample of tumour tissue, blood, urine or cerebrospinal fluid from patients affected by tumours.

An object of the present invention are also compositions comprising temozolomide and a SOX1 inhibitor, wherein the SOX1 inhibitor can be selected from the group comprising compounds that enable the expression of the SOX1 to be reduced, blocked or removed or the transport or activity of some of the expression products thereof, selected from the group consisting of siRNAs, mircoRNAs complementary to the mRNA of the SOX1 gene and peptides capable of binding to the SOX1 protein and inhibiting the activity thereof. These compositions can also contain carriers that are pharmaceutically acceptable for the use thereof in medicine and are indicated for the treatment of glioblastoma or other central nervous system tumours.

Lastly, an object of the present invention are kits to carry out the above methods comprising primers to determine the gene expression levels of SOX1 and the necessary antibodies to determine the levels of SOX1 at protein level, as well as the necessary reagents to carry out the immunofluorescence, immunohistochemistry, Western blot and/or ELISA procedures. Said primers can be selected from the group consisting of SEQ. ID. No. V. AG AC CTAG AT G C C AAC AATTG G and SEQ. I D. No. 2: GCACCACTACGACTTAGTCCG, SEQ. ID. No. 3: ATGGGGAAGGTGAAGGTCGG and SEQ. ID. No. 4: GACGGTGCCATGGAATTTGC, and SEQ. I D. No. 5: GAGTGGAAGGTCATGTCCGAGG and SEQ. ID. No. 6: CCTTCTTGAGCAGCGTCTTGGT.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 (A - C): High levels of SOX1 are linked to poor clinical evolution and low survival of patients with glioblastoma

Figure 2 (A - C): Enrichment of SOX1 expression in the glioma stem cell population

Figure 3 (A - E): Silencing of SOX1 reduces the tumorigenic capacity of glioma cells

Figure 4 (A - D): Silencing of SOX1 reduces self-renewal capacity, tumour initiation and increases sensitises to temozolomide

Figure 5 (A - G): SOX1 regulates the malignant activity of glioma stem cells Figure 6 (A - E): The increased SOX1 activity facilitates the acquisition of self- renewal activity and increases the proliferative capacity of glioma cells

Figure 7A: SOX1 expression is modulated by temozolomide

DETAILED DESCRIPTION OF THE INVENTION

The transcription factor SOX1 is a member of the SOX family that is expressed since embryonic development and the expression of which in adults has been linked to stem cells, mainly in the brain. The role thereof in cancer has been described in several tissues, such as liver, ovarian, nasopharyngeal or cervical tissue, and it acts as a tumour suppressor in all of them. In a high percentage of these tumours, the promoter is methylated and this silencing is correlated to a lower survival of patients. Moreover, studies with different cell lines have shown that the overexpression thereof reduces tumour proliferation and activity.

However, as shown in Figure 1 , the authors of the present invention have surprisingly identified that the expression of SOX1 in patients with glioblastoma, using two independent cohorts (Figures 1A and 1 B), is heterogeneous and that the cases with increased expression of SOX1 are linked to a lower survival of patients (Figure 1 C).

Additionally, it has been verified that expression of SOX1 inversely correlates with cell differentiation and that SOX1 plays an important role in maintaining cancer stem cells (CSCs) in glioma. As such, high levels of SOX1 are linked to the undifferentiated stage of glioma CSCs, while low levels promote differentiation. That is, as shown in Figure 2, the authors of the present invention have observed that expression of SOX1 is increased in glioma stem cell lines derived directly from patients with respect to conventional glioma lines (Figure 2A). Furthermore, enrichment of the expression of SOX1 was observed when the conventional glioma lines were cultivated in stem cell conditions (Figure 2B). Lastly, it has been noted that the levels thereof reduce when the glioma stem cells are cultivated in differentiation conditions (Figure 2C).

Moreover, the genetic silencing of SOX1 reduces the proliferative capacity and tumour activity of glioma cells, measured by cell viability assays, the measurement of positive cells for the mitosis marker P-H3 (Figures 3A and 3B), and by studies of tumour growth in immunosuppressed animals, as well as by analysis of these tumours by immunohistochemistry (Figures 3C,3D and 3E).

Lastly, it describes how a decrease in levels of SOX1 entails a lower tumour initiating capacity and a greater differentiation of the tumour cells and a greater response to temozolomide, as can be seen through studies on tumour initiation (Figure 4A), oncosphere formation (Figure 4B), analysis of differentiation and stem cell markers by qRT-PCR (Figure 4C) and MTT cell viability assays (Figure 4D). These data on sensitivity to TMZ in the absence of SOX1 at cell level are reinforced by the results clinically obtained in patients that have been treated with temozolomide.

Strengthening these results, expression of SOX1 was silenced directly in a glioma stem cell line (GNS166), observing a similar phenotype. In fact, the authors of the present invention, as shown in Figure 5, identified that SOX1 silencing (Figure 5A) promotes a drastic decrease in cell growth (Figure 5B), in cell viability (Figure 5C) and proliferative capacity thereof (Figure 5D). Moreover, these actions overlap with the decrease in self-renewal activity (Figure 5E) and an increase in the expression of differentiation markers (Figure 5F). Lastly, SOX1 silencing significantly delays tumour formation when the cerebral stereotactic procedure is carried out (Figure 5G).

Lastly, levels of SOX1 in the same glioma stem cell line (GNS166) were overexpressed, observing that the increase in the expression of SOX1 protein (Figure 6A) increases the levels of stem cell markers (Figures 6A, 6B), decreases the expression of differentiation markers (Figure 6C), in addition to promoting a slight, but significant, increase in cell growth (Figure 6D), and in the proliferative capacity (Figure 6E) of glioma stem cells.

In conclusion, the authors of the present invention have identified that a high expression of SOX1 maintains the quiescent state of the CSCs of the tumour and greater tumorigenicity and, in turn, a poorer response to temozolomide both at cell level and clinical level.

As such, the present invention proposes the SOX1 gene and the protein that codes this gene as a target for identifying patients with glioblastoma multiforme, responsive to treatment with temozolomide, regardless of the methylation state of the MGMT promoter.

Additionally, the present invention provides a new strategy in the treatment of glioblastoma multiforme with temozolomide, overcoming the resistance to temozolomide by inhibiting the expression of SOX1 with alternative compounds with inhibitory activity of the SOX1 protein and/or by means of SOX1 gene silencing.

In order to make the content and scope of the present invention more readily understandable, the following definitions of terms and expressions used in the present specification are included. The terms "subject" and "patient" refer to any individual subject for whom the methods and compositions of the present invention are useful. These mainly include mammals, among others, humans, livestock, dogs, guinea pigs, rabbits, chickens, etc. Preferably, the subject or patient is a human, either man or woman, without limitation in terms of their age or race.

The term "sample" refers to a "tissue or cell sample" obtained from the subject or patient. The source of the tissue or cell sample can be solid tissue, such as a fresh, frozen and/or preserved organ, or tissue sample or biopsy or aspirate; blood or any constituent of blood; body fluids, such as cerebrospinal fluid, amniotic fluid, peritoneal fluid or interstitial fluid; cells of any gestation or development period of the subject. The tissue sample can also be cells or lines of primary or cultured cells. Optionally, the tissue or cell sample is obtained from a primary or metastatic tumour. The tissue sample can contain compounds that are not naturally interspersed in the tissue in nature, such as preservatives, anticoagulants, buffers, fixing agents, nutrients, antibiotics or the like.

The expression "Gene-biomarker or a variant thereof" as used in the present invention also comprises truncated "Biomarker", "Biomarker" fragments, mutated "Biomarker" in the different forms that can be generated.

The term "expression of SOX1" refers to the gene expression levels of SOX1 measured by means of qRT-PCR, using in particular the following pairs of primers with the following sequences SEQ. ID. No. 1 : AGACCTAGATGCCAACAATTGG and SEQ. ID. No. 2: G CAC C ACTAC G ACTT AGT C C G , normalised with the gene expression levels of the GAPDH gene SEQ. ID. No. 3: ATGGGGAAGGTGAAGGTCGG and SEQ. ID. No. 4: GACGGTGCCATGGAATTTGC, as well as SEQ. ID. No. 5: GAGTGGAAGGTCATGTCCGAGG and SEQ. ID. No. 6: CCTTCTTGAGCAGCGTCTTGGT, and quantified with respect to a reference value.

The term "reference value" refers to the expression levels of SOX1 and normalised with GAPDH preferably from the control sample First Choice® Human Brain Reference RNA, reference No. AM6050, marketed by the company ThermoFisher.

Other control samples that can be used for the purposes of the present invention are: a) tissue from peritumoral regions of patients affected by glioblastoma, b) brain tissue of healthy individuals that do not suffer from any neurodegenerative disease.

The term "low expression level" refers to expression levels lower than 1 , preferably lower than 0.7, obtained when compared with a control sample, for example, the control sample First Choice® Human Brain Reference RNA, wherein the value of this healthy control would be equivalent to 1 .

The term "high expression level" refers to expression levels higher than 1.5 with respect to levels obtained with respect to a control sample, for example, the control sample First Choice® Human Brain Reference RNA, wherein the value of this healthy control would be equivalent to 1.

The term "good response to the treatment with temozolomide" refers to obtaining a value in the expression level of SOX1 lower than the reference value, that is, a cell viability equal (1 ) to or less (lower than 1 , preferably lower than 0.7) than that of the glioma cells in which the levels of SOX1 have not been modified (wherein the value assigned to these cells is 1 ), with a temozolomide treatment at a concentration of 100uM for 72 hours.

The term "poor response to the treatment with temozolomide" refers to a cell viability higher than 1 .5, than that of the tumour cells in which the levels of SOX1 have not been modified (wherein the value assigned to these cells is 1 ), with a temozolomide treatment at a concentration of 10OuM for 72 hours.

The term "inhibitor of the expression of SOX1 " refers to any compound (monoclonal antibodies, small inhibitors, pharmacologically diverse compounds) that reduces the expression levels of SOX1 in a statistically significant manner and below the value 0.7 relative to the control value.

As mentioned in the foregoing sections, the main object of the present invention is an in vitro method for determining the response of a subject suffering from a tumour in the central nervous system, for example, glioblastoma multiforme, to the treatment with temozolomide that comprises detecting and quantifying the expression levels of SOX1 in an isolated sample of the subject, and comparing said expression levels with respect to a reference value, wherein a low expression level of SOX1 indicates a good response to the treatment with temozolomide.

The different methodologies and procedures used in order to put the present invention into practice are described below: SOX1

All the nucleotide sequences that code for a human and animal SOX1 protein, as well as the protein sequences expressed by said nucleotide sequences, along with the homologous and variant sequences, are understood to be comprised within the scope of the present invention.

In particular, the following nucleotide and protein sequences are comprised within the scope of the present invention:

Gene that codes a human transcription factor identified in the NCBI database with access number: NP_005977 of 26/08/2016; Human nucleotide sequence of the mRNA of the SOX1 gene identified in the NCBI database with access number NM_005986 of 26/08/2016;

Gene identified in the NCBI database with access number 6656 (https://www.ncbi.nlm.nih.gov/qene/6656) with 4108 base pairs that gives rise to the transcript with access number NM_005986.2.1 and codes a 391 aa protein

Gene Ensembl:ENSG00000182968 from ensembl.org and the transcript ENST00000330949.2 MIM:602148: Veaa:OTTHUMG00000017362

NM_009233.3 and NP 033259.2 Mus musculus

NM_001002483.1 and NP 001002483.1 Danio rerio

NM_204333.1 and NP 989664.1 Gallus aallus

Several variants of these sequences at mRNA or protein level can comprise modifications resulting from deletion, substitution or insertion. Gene modification can be the result of natural gene variability, that is, variations that are not the result of genetic engineering. Gene modification can be the result of processing the gene or gene product in the body and/or degradation product. Modification at a protein level can be due to enzyme or chemical modification in the body. For example, modification can be glycosylation or phosphorylation or farnesylation.

Procedure for obtaining the sample from the patient

In order to carry out the method of the present invention, there must be a biological sample from the patient from which the expression level of SOX1 can be measured. To do so, before surgery, patients sign an informed consent form wherein they agree to provide the tumour biopsy and a 30 cc blood sample to Biobanco Vasco de la Investigacion. The study on the expression of SOX1 in solid biopsies is carried out by creating TMAs (Tumor Microarrays) that are stained with SOX1 antibodies, cited above for immunohistochemistry. The quantification of the number of positive SOX1 cells is carried out by means of a scanning electron microscope and computer image analysis system (Ariol SL-50; Genetix) for the automation there. To study the expression of SOX1 in a blood sample, a "digital droplet PCR" system from BioRad is provided, which enables a sensitivity lower than 0.1 %, improving the sensitivity of the PCR in real time (qPCR), which is 1 %.

As shall be discussed later, the expression of SOX1 in a sample can be analysed by several methodologies, many of which are known in the state of the art and are understood by a person skilled in the art, which include but are not limited to, immunohistochemical and/or Western blot analysis, blood-based quantitative assays, serum-based ELISA, in order to examine, for example, expression levels of proteins, enzyme assays of biochemistry activity, in situ hybridisation, Northern blot analysis and/or PCR analysis of mRNA and genomic Southern blot analysis (to examine, for example, the deletion or amplification of genes), in addition to any of a wide range of assays that can be carried out by gene and/or tissue matrix analysis.

The typical protocols for evaluating the expression, detection and quantification of genes and gene products used in scope of the present invention are described, for example, in the following articles:

Garros-Regulez L, Aldaz P, Arrizabalaga O, Moncho-Amor V, Carrasco-Garcia E, Manterola L, Moreno-Cugnon L, Barrena C, Villanua J, Ruiz I, Pollard S, Lovell- Badge R, Sampron N, Garcia I, Matheu A. rmTOR inhibition decreases SOX2-SOX9 mediated glioma stem cell activity and temozolomide resistance. Expert Opin Ther Targets. 2016;20(4):393-405.

This article describes the protocols used to create cell lines derived from patients and create oncosphere cultures. In addition, the lentiviral infection protocols that enable gene manipulation in cell cultures are described. Protocols for carrying out (i) Western Blot, (2) quantitative PCR, (3) immunofluorescence, and (iv) immunohistochemistry, are also defined. Lastly, the cerebral stereotactic protocol is described.

Carrasco-Garcia E, Lopez L, Aldaz P, Arevalo S, Aldaregia J, Egafia L, Bujanda L, Cheung M, Sampron N, Garcia I, Matheu A. SOX9-regulated cell plasticity in colorectal metastasis is attenuated by rapamycin. Sci Rep. 2016 Aug 30;6:32350.

This article describes the protocols followed in tumorigenesis assays in immunosuppressed animals. It also establishes the protocols for carrying out (i) Western Blot, (2) quantitative PCR, (3) immunofluorescence, and (iv) immunohistochemistry, as detailed in the aforementioned article.

Matheu A, Collado M, Wise C, Manterola L, Cekaite L, Tye AJ, Canamero M,

Bujanda L, Schedl A, Cheah KS, Skotheim Rl, Lothe RA, Lopez de Munain A, Briscoe J, Serrano M, Lovell-Badge R. Oncogenicity of the developmental transcription factor Sox9. Cancer Res. 2012 Mar 1 ;72(5): 1301 -15.

This article describes the protocols followed to generate TMAs (Tissue microarrays) of tumour biopsies, as well as the quantification protocol for tumour cells in a solid biopsy.

Guan Z, Zhang J, Wang J, Wang H, Zheng F, Peng J, et al. SOX1 down- regulates beta catenin and reverses malignant phenotype in nasopharyngeal carcinoma. Mol Cancer. 2014 Nov 478 26;13:257.

This article describes antibodies and sequences of primers for the study of the expression of SOX1 . Moreover, it describes the tumour suppressor function of SOX1 in nasopharyngeal cancer, which is the opposite result to that proposed in this patent. Tsao CM, Yan MD, Shih YL, Yu PN, Kuo CC, Lin WC, et al. SOX1 functions as a tumor suppressor by antagonizing the WNT/beta-catenin signaling pathway in hepatocellular carcinoma. Hepatology. 2012 Dec;56(6):2277-87.

This article describes antibodies and sequences of primers for the study of the expression of SOX1 . Moreover, it describes the tumour suppressor function of SOX1 in hepatocellular carcinoma, which is the opposite result to that proposed in this patent.

Lin YW, Tsao CM, Yu PN, Shih YL, Lin CH, Yan MD. SOX1 suppresses cell growth and invasion in cervical cancer. Gynecol Oncol. 2013 Oct;131 (1 ):174-81.

This article describes antibodies and sequences of primers used for the study of the expression of SOX1 . Moreover, it describes the tumour suppressor function of SOX1 in cervical cancer, which is the opposite result to that proposed in this patent.

For the case in which SOX1 protein is detected and quantified:

If the aim is to measure the expression of the SOX1 protein, the method comprises obtaining binding antibodies specific for the SOX1 protein present in the biological sample, which functions as an antigen, putting the antibodies in contact with the SOX1 protein obtained from the sample of the patient, and then detecting and quantifying the complex: antibody-SOX1 protein.

Specifically, the following method is followed: Western Blot, immunofluorescence, immunohistochemistry and/or ELISA, according to the practices known in the state of the art by a person skilled in the art.

For the case in which mRNA of the SOX1 gene is detected and quantified: If the aim is to measure the expression of SOX1 mRNA, the method comprises obtaining total RNA present in the biological sample, the conversion thereof into complementary DNA and the specific amplification of the SOX1 gene by means of the use of primers specific for the gene, such as SEQ. ID. No. 1 : AG AC CT AG ATG C C AAC AATT G G and SEQ. ID. No. 2: GCACCACTACGACTTAGTCCG.

In order to quantify the levels of SOX1 , the SOX1 values are normalised with the values of any reference gene (GAPDH, B-actin, TBP, etc.). Preferably, it is normalised with gene expression levels of the GAPDH gene SEQ. ID. No. 3: ATGGGGAAGGTGAAGGTCGG and SEQ. ID. No. 4: GACGGTGCCATGGAATTTGC,

SEQ. ID. No. 5: GAGTGGAAGGTCATGTCCGAGG, SEQ. ID. No. 6: CCTTCTTG AG CAG CGTCTTG GT and they are compared to normalised values obtained from the total RNA of the control sample, particularly, sample First Choice®Human Brain Reference RNA, reference No. AM6050. As a control between the analysis of different patient samples, the Ct value obtained in the sample First Choice® Human Brain Reference RNA, reference No. AM6050, is controlled, and a variation coefficient of 5% is accepted. The Ct value enables the quantification of the gene of interest in relative terms. To do so, the change in the expression levels of messenger RNA is expressed and the expression level of the gene to be studied is compared in relation to a control gene also referred to as reference or internal gene which is preferably GADPH.

Specifically, the following method is followed: extraction of total RNA, reverse transcription and qRT-PCR, and quantification of values using the 2 ^"DDCt method. The mircoarray or RNAseq analysis can also be carried out. All this is carried out in accordance with the known practices of the state of the art for a person skilled in the art.

Comparison of the expression level of SOX1 with respect to a "reference value"

For the case in which SOX1 protein is compared:

The reference value is the levels of SOX1 protein obtained in control samples of healthy patients, such as for example, of healthy brain tissue. Specifically, the use for these purposes of commercially available samples such as, for example, the sample First Choice® Human Brain Reference RNA from Fisher, reference No. AM6050, is envisaged.

For the case in which mRNA of the SOX1 gene is compared:

The reference value is the expression levels of SOX1 normalised with the reference gene (GAPDH, B-actin, etc.) in normal brain tissue, as discussed above.

SOX1 inhibitors:

The term "SOX1 inhibitor" as used herein refers to all compounds that enable the expression of the SOX1 gene to be reduced, blocked or eliminated or the transport or activity of some of the expression products thereof. These include new compounds or identification of agents, compounds, small molecules, drugs that have SOX1 affinity greater than 30% and functionalised to enable the passage of the blood-brain barrier in order for the maximum amount possible of said compound to reach the brain. Therefore, the following characteristics of the general chemical structure of the candidate molecules shall be taken into account: (i) having a low molecular weight, (ii) with inhibitory activity against SOX1 , and (iii) with physical and chemical properties that enable it to pass through the blood-brain barrier. These characteristics are determined through the fine-tuning and validation of chemical synthesis techniques and methodologies. Specifically, the reaction condition are fine-tuned, carrying out the corresponding modifications and optimisations in order to obtain the desired chemical compounds in the necessary amount and purity level. As such, it is necessary to (i) Design a retrosynthetic route, (ii) Carry out successive chemical reactions in order to optimise the chemical yields and purification methods of the products. In vitro enzyme assays are then carried out in order to verify the activity and selectivity of the compounds generated or tested in the above section compared to the SOX1 gene sequence. Lastly, the expression of SOX1 is determined by Western blot assays and quantitative PGR and cell cytotoxicity is determined by means of MTT viability analysis and oncosphere formation.

Examples of inhibitors are, although not limited to, siRNAs or microRNAs complementary to the mRNA product of the transcription of this gene or peptides capable of binding to the SOX1 protein and inhibiting the activity thereof.

Compositions of the present invention:

The compositions of the present invention comprise on the one hand compositions and kits of parts containing the active agents and reagents necessary to carry out the method of the present invention and, optionally, instructions of use. Said active agents can be formulated together or separately for their separate, sequential or joint administration.

Additionally, the present invention describes pharmaceutical compositions with application in the treatment of glioblastoma multiforme in subjects that are resistant to treatment with temozolomide comprising at least one inhibitor of the expression of SOX1 and temozolomide in therapeutically effective amounts.

Said pharmaceutical compositions additionally contain pharmaceutically acceptable excipients known by a person skilled in the art.

The type of excipient depends on the route of administration. As such, the compositions of the present invention can be formulated for topical, oral, nasal, intravenous, intracranial, intraperitoneal, subcutaneous or intramuscular administration with the suitable excipients known by a person skilled in the art in, for example, the publications: "Handbook of Pharmaceutical Excipients 8th Edition", "Martindale 36th Edition" or "Remington, The Science and practice of Pharmacy, 21st Edition".

The pharmaceutical compositions can also contain any and all other active ingredients that intensify or delay the therapeutic effect, such as abraxane, which in addition to being antineoplastic, makes the stroma less fibrous, thus facilitating the action of other compounds, or clavulanic acid that prevents amoxicillin from being eliminated by the kidneys.

Kits

The kit of the invention can comprise, without any type of limitation, marked or unmarked primers, marked or unmarked probes, buffers, agents for preventing contamination, as well as all the mediums for the implementation thereof, in addition to the instructions for the use thereof in the method of the present invention. In particular, the kit of the present invention contains the following components: oligo sequences to determine the gene expression levels of SOX1 and the necessary antibodies to determine the levels of SOX1 at a protein level by immunofluorescence, immunohistochemistry, Western blot and ELISA. Among them, the following have been tested and work well for measuring the expression at a protein and mRNA level:

a) Antibodies for Western Blot, immunofluorescence and immunohistochemistry

AF3369 by R&D systems

4194 by Cell Signaling

ab87775 by Abeam b) Primer (oligos) sequences to detect messenger RNA

Pair 1

SEQ. ID. No. 1 : AGACCTAGATGCCAACAATTGG

SEQ. ID. No. 2: G CAC CACTAC G ACTTAGT C C G

Pair 2

SEQ. ID. No. 3: ATGGGGAAGGTGAAGGTCGG

SEQ. ID. No. 4: GACGGTGCCATGGAATTTGC

Pair 3

SEQ. ID. No. 5: GAGTGGAAGGTCATGTCCGAGG

SEQ. ID. No. 6: C CTT CTT GAG C AG C GT CTT G GT

EXAMPLES

The materials and methods used for experimental purposes of the present invention are the following: Tumour biopsies and patient information

Part of the excess diagnostic tissue stored in the Biobanco Vasco de Investigacion (http://www.biobancovasco.org) was used for the analysis of the levels of mRNA in patient samples. The information on low-grade gliomas and glioblastomas of the TCGA cohort was obtained using the TCGA-Assembler program. The experimental methods and protocols in human samples were carried out according to the current regulations. All glioblastoma patients whose samples were used to carry out this work signed the corresponding informed consent form. The use of these samples was approved by the Clinical Research Ethics Committee of the Guipuzcoa health area. Cell culture

The glioma cell lines U373MG (U373), U251 MG (U251 ), U87MG (U87), A172 and T98G were obtained from the American Type Culture Collection (ATCC). These adherent cells were cultivated in DMEM medium {Eagle culture medium modified by Dulbecco, Gibco, Waltham, MA, U), supplemented with 10% (v/v) of bovine foetal serum (Gibco), 2mM of glutamine, penicillin (100 U/mL) and streptomycin (100 μg/mL), medium referred to as "complete or adherence medium".

The glioma stem cell lines Glioma Neural Stem (GNS) GNS166 and GNS179 were provided by Steve Pollard (Edinburgh Cancer Research Centre, University of Edinburgh). These stem cell lines were cultivated in adhesion, prior to treatment of the surface with 10 g/mL of laminin (Sigma Aldrich) for 3 hrs at 37 °C in DMEM/F12 medium (Eagle culture medium modified by Dulbecco, Ham F12, Sigma) in a ratio of 1 :1 and supplemented with B27 10x (Gibco) N2 100x (Gibco), glucose 45% and 20 ng/mL of epidermal growth factor (EGF, Sigma, St. Louis, MO, USA) and of fibroblasts (bFGF,R&D Systems), a medium which is referred to as "stem medium or medium selective for stem cells". The lines GB1 and 2 were generated in the group and these cells and oncospheres of the conventional lines were also grown in the medium selective for stem cells. The differentiation studies were carried out by excluding the bFGF and EGF factors from the complete medium DMEM-F12 and adding 1 % of FBS. For the oncosphere counting assay, 5x10 ³ cells per well were seeded in triplicate and maintained in culture for 10 days in stem cell medium.

Viral infections The viral infections were carried out following protocols established in the group using a multiplicity of infection of 10 for 6 hours. To do so, we use the pLM-mCitrine- SOX2 (SOX2) construct, kindly provided by Dr. Izeta, pWPXL-SOX1 (SOX1 ) construct, which was cloned by Dr. Stevanovic, and pLKO.1 shSOXI (sh 1 and sh5) construct, which were obtained from Sigma. The cells infected with the pLKO.1 and shRNA plasmids were selected by adding puromycin at a concentration of 2μg mL.

Immunofluorescence

In order to determine the expression of certain proteins in cells, 2.10 ⁴ cells were seeded in their corresponding cell culture in wells with immunocytochemistry chambers (Lab-Tek II Chambers Thermo Scientific) and after 24 hours they were fixed with 4% paraformaldehyde for 10 minutes. After fixing, they were blocked and permeabilised with PBS-0.3% Triton X-100 (Cat. No. T8787, Sigma) supplemented with 5% SBF for one hour at room temperature.

After blocking, the samples were incubated with the primary anti-phospho- histone H3 antibody (PH3, 1 :2000; Ab14955, Abeam) for 2 hours at room temperature and washed three times with PBS 1 x to be subsequently incubated with a Cy3- conjugated secondary antibody (Jackson ImmunoResearch Laboratories Inc.) for one hour at room temperature in darkness. After 3 washes with PBS 1 x, the nuclear DNA was stained and they were mounted on slides using Vectashield Mounting Medium with DAPI Hard set (Cat. No. H-1400, Vector Laboratories). The immunofluorescence was observed under an Eclipse 80i microscope and the photos were processed with NIS Elements Advances Research (Nikon) software. MTT cell viability assay

2x10 ³ cells were seeded in plates with 96 wells and after 24 hours they were treated with the drug temozolomide. The solvent indicated for the drug was used as a control for the treatments. The treatments were carried out six times and once finished, MTT reagent was added at a final concentration of 1 .9 mg/mL and it was incubated for 3 hours at 37°C and 5 % of C0 ₂ . Once the content of the wells is aspirated, 150 μί of DMSO was added to dissolve the formazan crystals and the plate was continuously shaken at room temperature for 15 minutes to subsequently measure the absorbency at 570nm in a Multiskan Ascent plate reader (Thermo Scientific). RNA extraction, reverse transcription and gene expression

The total RNA was extracted by means of Tri reagent solution (Life Technologies). The reverse transcription was carried out using random primers and the MultiScribe Reverse Transcriptase kit (Life Technologies). For the quantitative RT- PCR, 20 ng of cDNA was used to analyse the expression of each gene by means of the Absolute SYBR Green mix (Applied Biosystem), in a LightCycler 96 thermo-cycler machine (Roche). The transcription levels were normalised compared to the expression of GAPDH and were measured using the AACt relative quantification method.

Immunohistochemistry

Tumours generated subcutaneously in mice were extracted and fixed in 10% of formalin for 48 hours at room temperature to be subsequently soaked in paraffin and cut into sections that are 4 μηη thick with a microtome. The cuts were dewaxed, rehydrated by means of successive baths in decreasing alcohol and boiled in citrate buffer for 10 minutes for antigen retrieval. They were incubated in blocking solution (PBS-0.3% Triton X- 100-5% FBS) and were incubated with the anti-Ki67 primary antibodies (AB15580, Abeam), SOX1 (4194, Cell Signaling), SOX2 (AB5603, Millipore) and PML (A301 -167A, Bethyl Laboratories) at 37°C for 2 hours. After incubation with the primary antibody, they were washed and incubated with the Enzyme-conjugated secondary antibodies. Subsequently, the samples were incubated for 10 minutes at room temperature with the substrate 3,3'-Diaminobenzidine which means that the enzyme conjugated to the secondary antibody activates and generates the product that can be detected in the tissue. The stains were observed under an Eclipse 80i microscope and were processed with NIS Elements Advances Research (Nikon) software.

Western blot

Immunoblots were carried out following conventional protocols. The primary antibodies that have been used are anti-SOX1 (4194, Cell Signaling) and anti-SOX2 (AB5603, Millipore). Detection was carried out by chemiluminescence using the NOVEX ECL HRP Chemiluminescent Substrate Reagent kit (WP20005, Invitrogen). In vivo carcinogenesis assays

All the test animals used in this work were kept at the Animal Experimentation Service of the Biodonostia Research Institute following the guidelines of the competent authority and in accordance with the rules established by the European Community Council, directive 86/609/EEC.

Stereotaxy was carried out on 6-8 week old NOD-SCID mice. The cells were subjected to enzymatic disaggregation by means of accutase before being resuspended in PBS 1x, after prior centrifugation and washing with PBS1 x of the same. 100,000 cells of different conditions were injected in a final volume of 1 μΙ_ of PBS 1x by means of a 75RN, 26s ga 2" Hamilton syringe in the striatum of the right hemisphere of the mice at the following coordinates: Bregma: +1 .0 mm rear, -2 mm left side and - 2.5 mm deep, at a constant flow of 0.05uL/min. The generation and/or progression of the tumour generated was tracked by tracking weight loss and movement ability of the mice. The Kaplan-Meier survival analysis was carried out using GraphPad Prism 5 software.

The formation of subcutaneous tumours was carried out in 8 week old immunosuppressed FOXn1 ⁿ 7FOXn1 ^nu mice. Therefore, the cells were detached with trypsin and were resuspended in PBS 1 x after prior centrifugation and washing with PBS1x of the same. The mice were inoculated with 0.1 mL of the cell suspension and were observed twice a week, measuring the size of the tumours generated by means of a calliper until the end point, which is considered to be when the tumour volume reaches 15mm ³ . The tumour volume was calculated by means of the equation V = (I ^* w2 ^* 0.52), where I is the larger diameter of the tumour and w is the smaller diameter.

Data Analysis

The data shown in the present work represent the average ± the standard deviation (SD) of at least three independent experiments. The averages were compared by means of the Student's t test for normal distributions, the asterisks indicating the different levels of statistical significance ( ^* p<; 0.05; ^** p<;0.01 ; ^* p<;0.001 ).

To calculate the statistical significance of survival in the samples of the patients and in vivo stereotaxy assays, a Chi-square test (X ² ) was carried out using the Kaplan Meier program and a value of p less than 0.05 was considered statistically significant.

Example 1 :

High levels of SOX1 correlate to reduced survival in patients with GBM.

Firstly, the authors of the present invention researched whether the expression of SOX1 is regulated by SOX2 in glioma cells. To do so, they modulated the gene expression of SOX2 and analysed what happened with the expression of SOX1. Thus, silencing the expression of SOX2 by lentiviral infection with interfering RNA in cells with high levels of SOX2 causes a reduction in the expression of the SOX1 gene. On the other hand, when the expression of SOX2 is ectopically activated in cells with low levels of SOX2, it is observed how the expression levels of SOX1 increase. These results show that SOX2 modulates the expression levels of the SOX1 gene at cell level. Then the correlation between SOX1 and SOX2 was studied in clinical biopsies, analysing the expression of these transcription factors in a cohort of human glioblastoma samples. To do so, total RNA was extracted from the tumour biopsies and, after carrying out a reverse transcription using random primers, a qRT-PCR was carried out wherein the expression levels of SOX1 and SOX2, normalised with the expression values of GAPDH, were measured. These levels were quantified with respect to the levels obtained in healthy brain tissue.

The results of this assay indicated that the expression of SOX2 was significantly increased in a cohort of glioblastoma samples (from the Donostia University Hospital) compared to the healthy brain tissue. In the case of SOX1 , the expression levels were more variable, finding that only 60% of the tumour biopsies with high levels of SOX2 showed overexpression of SOX1 (more than 1.5 times the change, wherein the value corresponding to healthy brain tissue is 1 ). These results indicate that there is a positive correlation between SOX1 and SOX2, but the expression thereof does not go hand-in-hand.

Additionally, the expression levels of SOX1 were analysed by means of RT- QPCR in a cohort of more extensive glioblastoma samples derived from the same Hospital and they were compared to non-neoplastic neural tissue, finding a frequency close to 50% of glioblastomas having high expression levels of SOX1 (over 1.5 increase with respect to healthy tissue).

Subsequently, the correlation between the expression levels of SOX1 and the survival of patients with glioma of different grade was studied by means of computational biology in the well-established cohort with extensive patients, with more than 300 biopsies from the TCGA. The bioinformatics data were downloaded using the TCGA-Assembler program. The results show that the levels of SOX1 in samples of low-grade glioma (LGG) and normal brain are similar. In the case of glioblastomas, the expression levels are very heterogeneous with a greater dispersion than the expression levels of the normal samples and in this cohort there is also a subgroup of biopsies with overexpression of SOX1 . A Kaplan Meier survival analysis was carried out and the results indicated that high expression levels of SOX1 correlate to a decrease in the survival of patients with glioblastoma. This shows that the expression levels of SOX1 in glioblastomas are heterogeneous and that high expression levels correlate to poor overall survival in patients. Example 2:

High expression of SOX1 contributes to tumorigenicity in glioma cells.

Given that, clinically, high expression of SOX1 negatively correlated to survival of patients, SOX1 was silenced in the glioma cell line that expressed high endogenous levels of this gene.

The cell line U251 was used transduced with plasmids that bring 2 interference sequences of the SOX1 RNA in a stable and independent manner, TRCN0000015928 and TRCN0000015929 by Sigma. To do so, the HEK293T cells were transfected with the packaging plasmids pRSV-Rev, pCMV-VSV-G and pMDLgpRRE, as well as the modified virus genome that includes the DNA of interest (plko vector or plko-shSOX1 ). After waiting 72 hours, the medium with the virus particles was collected, filtered and used to infect U251 glioma cells. In this step, the cells were in contact with the virus for 6 hours, after which a change of medium is carried out and they were kept in culture between 48-72 hours until a selection treatment with 2μg ml of puromycin was initiated, which lasted 72 hours (when all the uninfected control cells died with the treatment).

Subsequently, the consequences of reducing the levels of SOX1 in different processes associated with cancer were assayed. Firstly, it was observed that low levels of SOX1 significantly reduce the growth of these cells. Therefore, it was analysed by means of staining with P-H3 whether this low growth correlated to lower proliferative capacity compared to control cells, as was the case. To do so, 2.5 10 ⁴ cells of the control condition and the SOX1 silencing condition were seeded in immunocytochemistry chambers (Lab-Tek II Chambers, Thermo Scientific). After 24 hours or when a confluence close to 80% is reached, the cells were fixed with 4% of paraformaldehyde for 20 minutes and blocked and permeabilised with PBS-0.3% Triton X-100 supplemented with 5% bovine foetal serum (FBS) at room temperature for one hour. They were incubated with the primary antibody for phospho-Histone H3 (P-H3, reference Ab14955, Abeam) for 16 hours at 4°C and, after carrying out 3 washes with PBS 1 X, the cells were incubated with the secondary antibody Alexa Fluor® 555 Rabbit Anti- Mouse IgG Antibody (reference A21427, Invitrogen) for 2 hours at room temperature and in darkness. Staining the nuclear DNA and mounting the preparations was carried out using Vectashield Mounting Medium with DAPI, Hard set (reference H- 1400, Vector Laboratories). The images were taken in an Eclipse 80i microscope and processed with NIS Elements Advances Research (Nikon) software. The total number of cells and positive P-H3 cells in both conditions was counted, and the results were presented comparing the values of the shSOXI condition with respect to the control.

The cells U87 and U373 and U251 present an overpopulation of cells with greater malignancy with stem cell characteristics. Therefore, cells expressing low and high levels of SOX1 are cultivated in a serum-free medium for neurospheres (NSC) and the sphere formation and self-renewal capacity was examined. The number of spheres clearly and significantly reduces in cells with low levels of SOX1. In accordance with these results, the expression of GSC markers such as p27 and PML is also reduced. These results indicate that the expression of SOX1 can modulate the activity of the GSCs.

The tumour initiating activity measured by the tumour formation efficiency through limiting dilutions in athymic mice is a way of functionally characterising self- renewal in the in vivo GSC population. Therefore, the authors of the present invention researched whether SOX1 could regulate tumour initiation. To do so, parental U251 cells transduced with an empty vector or with SOX1 interference were injected subcutaneously in athymic mice. To do so, 5x10 ⁵ and 5x10 ⁴ cells of both conditions were injected subcutaneously in the 4 flanks of immunosuppressed Foxn1 nu/Foxn1 nu nude mice (8 weeks old, of random sex). The mice were observed weekly for 41 days, measuring the size of the tumours generated at the indicated times. The condition of SOX1 silencing generated less tumours and these were smaller than those of the control condition.

These results indicate that SOX1 silencing suppresses the tumorigenicity of glioma cells, probably inhibiting the self-renewal capacity thereof.

Example 3:

Low expression levels of SOX1 correlate to GSC cell differentiation.

The cells derived from GNS and GB patients were cultivated in a medium selective for stem cells DMEM/F12 supplemented with 1 % N2, 2% B27, 2mM L- glutamine, 100 U/ml penicillin, 100 μg ml streptomycin, 1 .34% glucose 45%, 10 ng/ml of epidermal growth factor (EGF), and 10 ng/ml of basic fibroblast growth factor (bFGF- 2). In order to grow them adhered, the GNS were cultivated after treating the plates with 10μg/ml of laminin for 3 hours at 37°. For the differentiation assays, the EGF and bFGF2 growth factors were removed and 1 % FBS was added, and the cells were kept in this medium for 7 days.

In order to research the association of SOX1 with the "stem cell" properties, we started to work with stem cells derived from patients with GBM since it would have a greater impact clinically. Glioma stem cell lines derived from patients were used, two growing as tumourspheres and another two that are adherent that showed high expression levels of SOX1 , at the same time as the levels of other stem cell markers SOX2, CD133, OCT4 were also expressed in high levels. Then, the expression thereof in differentiation conditions was verified, adding 1 % serum and it was observed that the expression levels of SOX1 decreased significantly similarly to other GSC markers, postulating the hypothesis that there is a positive correlation between the molecular signature of the GSCs and SOX1 . These results, together with the previous studies in cell lines U87, U251 and U373, indicate that S0X1 could function as a marker for human GSC differentiation.

The exposure of GSC to BMP4 induces differentiation in astrocyte both in vitro and in vivo. Therefore, the role of SOX1 in this differentiation model was verified. A global study at the genome level using RNA sequencing has identified that SOX1 is between the genes silenced after differentiation induced by BMP4 in two independent glioma stem cell lines. Furthermore, methylation studies identify that the promoter thereof is demethylated in the CpG island during the treatment with BMP4 of the GSC cells. Therefore, this decrease in expression during differentiation depends on the demethylation of the SOX1 promoter. In order to characterised this process in greater depth, the glioma cells U87 and GSC were treated with 5'-aza-2'-deoxycytidine (5-aza), which is a DNA methylation inhibitor or with trichostatin (TSA), which is a histone deacetylase inhibitor. The results showed an inducement in the expression of SOX1 with higher concentrations of 5-aza and TSA.

Example 4:

SOX1 silencing suppresses the proliferation and self-renewal capacity of GSCs

In order to verify the impact of SOX1 activity on the regulation of glioma stem cells, SOX1 was silenced in a stable manner in the glioma stem cell line (GNS166) grown in adherence (Pollard SM. 2009). Following the protocol detailed in example 2, SOX1 was silenced in the cell line GNS166 that expressed high endogenous levels of this gene, with plasmids that bring 2 interference sequences of the SOX1 RNA in a stable and independent manner, TRCN0000015928 and TRCN0000015929 by Sigma. To do so, the HEK293T cells were transfected with the packaging plasmids pRSV-Rev, pCMV-VSV-G and pMDLgpRRE, as well as the modified virus genome that includes the DNA of interest (plko vector or plko-shSOX1 ). After waiting 72 hours, the medium with the virus particles was collected, filtered, concentrated and used to infect GNS166 glioma cells. In this step, the cells were in contact with the virus for 6 hours, after which a change of medium is carried out and they were kept in culture between 48-72 hours until a selection treatment with 2μg ml of puromycin was initiated, which lasted 72-96 hours (when all the uninfected control cells died with the treatment).

Quantitative PCR effectively confirms the silencing thereof and, therefore, the tumour characteristics of these cells were characterised in detail. Functionally, SOX1 silencing entails a reduction of the number of cells and the proliferation mediated by the number of positive p-Histone3 cells (P-H3). To do so, and as detailed in example 2, 2.5 ^■ 10 ⁴ cells of the control condition and the SOX1 silencing condition were seeded in immunocytochemistry chambers treated with laminin (Lab-Tek II Chambers, Thermo Scientific). After 24 hours or when a confluence close to 80% is reached, the cells were fixed with 4% of paraformaldehyde for 20 minutes and blocked and permeabilised with PBS-0.3% Triton X-100 supplemented with 5% FBS at room temperature for one hour. They were incubated with the primary antibody for phospho-Histone H3 (P-H3, reference Ab14955, Abeam) for 16 hours at 4°C and, after carrying out 3 washes with PBS 1 X, the cells were incubated with the secondary antibody Alexa Fluor® 555 Rabbit Anti- Mouse IgG Antibody (reference A21427, Invitrogen) for 2 hours at room temperature and in darkness. Staining the nuclear DNA and mounting the preparations was carried out using Vectashield Mounting Medium with DAPI, Hard set (reference H- 1400, Vector Laboratories). The images were taken in an Eclipse 80i microscope and processed with NIS Elements Advances Research (Nikon) software. The total number of cells and positive P-H3 cells in both conditions was counted, and the results were presented comparing the values of the shSOXI condition with respect to the control.

Then, self-renewal characteristics (characteristic of this population) were studied by measuring the expression of several GSC markers by means of qRT-PCR, finding a reduction of Nestin or PML in cells with low expression of SOX1 . On the other hand, the differentiation markers or cell cycle regulators such as GFAP or p27 ^Kip are expressed in higher levels at a transcription level. Lastly, it was verified whether the cells with high levels of SOX1 were tumorigenic. To do so, these cells were injected in the brain of immunosuppressed mice and the symptoms of tumour development were monitored. In a final volume of 1 μΙ, 1x10 ⁵ GNS166 control cells (infected with plko) or with SOX1 inhibition by means of stereotaxy is injected in the frontal cortex of 8 week old NOD-SCID immunosuppressed mice. The following coordinates were used: Anteroposterior: Bregma +1 ; Lateral: Bregma -2; Dorsoventral: Bregma -2.5. The injection rate of the cells is carried out by means of a infusion pump at a constant flow of 0.05μί/η"ΐίη. The animals were observed once a week and the conditions were compared by carrying out a Kaplan-Meier survival analysis with the GraphPad Prism 5 program. While 100% of the animals fell ill and developed tumours at 45 weeks after the injection of control cells, only 50% of the mice injected with low levels of SOX1 formed tumours. In summary, the data show that genetic silencing of SOX1 results in a reduction of the self-renewal capacity and tumour initiation. Therefore, these results indicate that inhibiting the expression of SOX1 can be used for glioblastoma therapy.

Example 5:

Overexpression of SOX1 promotes the proliferation and self-renewal capacity of GSCs

In order to verify the impact of the increase of SOX1 on the regulation of glioma stem cells, SOX1 was overexpressed in a stable manner in the glioma stem cell line (GNS166) grown in adherence (Pollard SM. 2009). To do so, the cells were infected with the construct pWXL-SOX1 generated in the Milena Stevanovic laboratory, a partner of the group. The method of viral infection was the same as in example 4 but with the difference that treatment with puromycin is not required.

The Western Blot and Q-RTPCR effectively confirm the overexpression thereof and, therefore, the tumour characteristics of these cells were characterised in detail. Functionally, high SOX1 activity entails an increase in the number of cells in culture at day 5, in addition to increasing the proliferation studied by the number of positive p- Histone3 cells (P-H3). To do so, the protocol detailed in example 2 was used.

For the study of the role of high levels of SOX1 on the regulation of self-renewal and differentiation, the expression of several specific markers of each process was measured by means of Western Blot and qRT-PCR. Significantly, an increase in both the levels of SOX2 protein and messenger RNA was found. The levels of the PML stem cell marker were also higher in the cells with high expression of SOX1. On the other hand, the differentiation markers or cell cycle regulators such as GFAP, CNPase and p27 ^Kip are significantly reduced at a transcription level.

Example 6:

The reduction in levels of SOX1 increase sensitivity to the treatment with temozolomide.

The first-line treatment of patients with glioblastoma is surgical resection followed by radiotherapy and concomitant and adjuvant temozolomide. Gliomas are resistant to this conventional treatment because the glioma stem cells survive chemotherapy and radiation. The evidence observed confirms an enrichment in the expression of SOX1 in the glioma stem cells and the clinical correlation of high levels of SOX1 to greater survival of the patients confirms that there is a correlation between high levels of SOX1 and resistance to temozolomide.

These data also suggest that the SOX1 activity could modulate the treatment with temozolomide (TMZ). The response of the control cells and cells with SOX1 silencing to the treatment with 100uM of temozolomide for 72 hours was studied, observing that the cell viability measured by means of MTT assays is significantly reduced in cells with SOX1 silencing compared to in control cells. Furthermore, it was observed that a short exposure of 24 hours at the same concentration of temozolomide (100uM) affects the levels of SOX1 in glioma cells (Figure 7A).

Moreover, clinically, the presence of the MGMT protein (gene responsible for

DNA repair) induces cell resistance against temozolomide, while methylation of the MGMT promoter promotes an increase in sensitivity to TMZ and increase in the survival of patients. Therefore, it was analysed whether there was any correlation between the levels of SOX1 and the status of the MGMT promoter in in vivo human biopsies. Although the number of samples is low, the results of this study do not confirm that there is a significant link between the expression of SOX1 and the methylated/demethylated state of the MGMT promoter. These results indicate that SOX1 mediates additional temozolomide response mechanisms that do not depend exclusively on MGMT (Table 1 ).

Table 1 : Correlation between the levels of SOX1 and the status of the MGMT promotor in in vivo human biopsies

That is, overall the present invention identifies that a high expression of SOX1 maintains the quiescent state of the CSCs of the tumour, induces greater tumorigenicity and, in turn, a poorer response to temozolomide.