R+D CSIC

A platform to boost therapeutic modified oligonucleotides

They are one of the most promising pharmacological tools, custom-designed for diseases such as Alzheimer's, various types of cancer, or multiple sclerosis. A multicenter project, with the participation of the CSIC, aims to create the XNA-HUB platform for designing new therapeutic molecules based on modified nucleic acids. For doing this, the platform will utilize supercomputing techniques and predictive models of Artificial Intelligence.

Nucleic acids are large molecules that contain genetic information and are present in all types of cells, as well as in viruses. Their modification allows the creation of tailored molecules that can recognize, block, and degrade pathogenic genetic sequences from viruses or cancer cells. Additionally, they can interfere with basic biological processes for pathogens, hindering their proliferation.

Modified nucleic acids represent one of the most promising pharmacological strategies at this moment. This field is expanding. The market volume for these therapies is expected to grow by 13.35% annually until 2027, with estimated revenues of $3.9 million dollars, according to industry sources.

Among nucleic acid-based therapies, messenger RNA (mRNA) vaccines stand out. These vaccines contain an RNA sequence designed to produce an antigenic protein, which will be recognized by the immune system, triggering a defensive response against the virus.

Other therapies that use oligonucleotides are designed to recognize specific and exclusive fragments of certain cancer cells. This recognition allows for the selective elimination of these fragments, thus avoiding the side effects of current therapies. An example is a nanodrug developed by teams from CSIC, IIB Hospital de Sant Pau, UAB and CIBER-BBN, to block metastatic spread. That was possible because he designed molecule has an affinity for a receptor that is overexpressed in metastatic stem cells, which allows the drug to enter and activate only within the metastatic cell.

Furthermore, there are strategies based on oligonucleotides designed to recognize pathogenic RNA and prevent its action, such as antisense oligonucleotides (ASOs), which help block the production of a specific protein. Another example is small interfering RNAs (siRNAs), which suppress the expression of specific genes.

Additionally, there is the strategy based on aptamers, which are characterized by a special three-dimensional structure that allows them to recognize and bind with high affinity to target molecules, leading to the inhibition of their activity that causes the disease.

Supercomputing and AI Tools

However, the design of all these molecules is extremely complex because one needs to select and design the target sequence, determine which modifications are most suitable, and decide in which positions modifications should be introduced. These processes generate countless combinations for treating a pathology, making it challenging to select the most active molecule. In this regard, supercomputing and AI tools would facilitate speeding up this process.

A project involving the CSIC is the creation of a platform for biomolecular data, both experimental and modeling, supported by predictive AI models. This platform aims to bring new innovative molecules based on modified oligonucleotides to the market. It is known as the XNA-HUB platform (XNA, "Xeno Nucleic Acid"), with the goal of enhancing the properties of these molecules and developing drugs for various chronic and oncological diseases, such as Alzheimer's, diabetes, multiple sclerosis, and different types of cancer.

The consortium comprises Nostrum Biodiscovery, a joint spin-off of the Barcelona Supercomputing Center and the IRB Barcelona; the Nucleic Acids Chemistry group of the Institute of Advanced Chemistry of Catalonia (IQAC-CSIC), led by Prof. Ramón Eritja, an expert in the field of nucleic acids modifications; the Molecular Modeling and Bioinformatics group of the IRB Barcelona, led by Prof. Modesto Orozco, an expert in the study of nucleic acids in all their forms; and the Nuclear Magnetic Resonance of Nucleic Acids group of the IQF-CSIC, led by Prof. Carlos González, expert in the field of determining the three-dimensional structures of nucleic acids with chemical modifications.

Computational techniques emerge as the ideal methods to focus synthetic work on sequences of greatest interest, which would also allow for shortening research time and reducing costs

The project has recently been selected in the latest call for Strategic Projects for Economic Recovery and Transformation (PERTE) in Cutting-Edge Health, which will provide a significant boost for the creation of the Platform.

Predicting the best combinations among a virtually infinite number

As the researchers explain, for a normal antigen, "there are on average 3²⁰⁰ possible messenger RNA sequences to generate the same vaccine, and it is not known a priori which of these will be the most effective". Similarly, "to block a target gene there may be more than 100 ASO oligonucleotide sequences, and these are susceptible to being modified in more than 20 different ways at each position, leading to 20¹⁰⁰ different variants - a figure hard to imagine, which is about 131 digits long, well exceeding trillions, figures which are about 13 digits each.

In that sense, computational techniques emerge as the ideal methods to focus synthetic work on sequences of greatest interest, which would also allow for shortening research time and reducing costs.

XNA-HUB aims to develop a comprehensive and cross-cutting platform of biomolecular data, both experimental and modeling, supported by predictive AI models, enabling the introduction of innovative molecules based on oligonucleotides to the market.

Mercè Fernández / Communication CSIC Catalunya

• Next