How Systems Biology is Cracking the Cancer Code

The arrival and development of high-throughput technologies open a window to explore cellular activity, from their genome to gene, protein and metabolic expression profiles on a massive scale. The interpretation of these data allows us to explore how living organisms modulate their response at different biological scales and to discover the principles that govern them.

Osbaldo Resendis-Antonio, researcher of the Research Support Network (RAI-UNAM), of the National Institute of Genomic Medicine (INMEGEN), and member of the Center for Complexity Sciences (C3) of the UNAM, explained the above and acknowledged:

“Notwithstanding the interest in the proposal, this objective is not easy. Together with the advance of these technologies, there is a pressing need to develop new conceptual schemes capable of integrating data and generating hypotheses about the mechanisms that govern living systems. This is where the sciences of complexity, i.e., interdisciplinarity, come together. The implementation of its methods undoubtedly contributes to projects to understand problems in areas of biomedical sciences and with strategies for the study of complex diseases”.

At the Systems Biology Seminar, in the series of conferences Contemporary Approaches to Biology and Complexity Science, organized by C3, the expert addressed the topic Metabolism in cancer and the microbiome: a systems biology perspective.

Resendis-Antonio informed that this discipline (systems biology) seeks to merge databases in the area of biology with computational models, through classical areas (such as physics and bioinformatics) and new trends in information technologies, such as machine learning.

In general, the purpose of systems biology paradigms, he explained, is to build mathematical/computational models that allow three objectives: to unite, interpret and integrate the large amounts of biological data of an organism, such as its genome, transcriptome (measurement of the expression levels of all the genes in the genome of an organism), proteome and metabolome; to quantitatively model biological networks and generate hypotheses about their principles of organization, and finally to evaluate these hypotheses experimentally.

In other words, their strategies deal with how to incorporate biological information about an organism (e.g., a human tissue, a cancer cell line or a bacterium in the microbiota) and develop models with predictive capabilities of its phenotype (observable traits).

“In the laboratory we work on metabolic alterations in cancer, particularly the effect of the microbiome, whose activity has recently been linked to human ailments ranging from obesity to psychiatric disorders,” he said.

Currently, the data generated by high-performance technologies are the fundamental basis of the information that later becomes biological knowledge. How to get from the first to the last? This is one of the central questions of systems biology.

They allow us to build layers of biological information for any living organism. Thus, from the genome and transcriptome of an organism we can explore how genes are activated (“turned on”) or repressed (“turned off”) by different stimuli. In addition to the massive measurement of genes and proteins, the metabolome has been added, a technology that provides an overview of “the dance of creations, transformations and annihilations of metabolites — any molecule involved (as a reactant or product) in metabolic pathways or biochemical transformations — that give rise to the sustenance of a living organism. All this under regulatory mechanisms and rules of impressive complexity”.

In the lab, he said, we are interested in designing strategies to elucidate the changes of metabolic alterations in conditions such as cancer and type 2 diabetes. “We build mathematical and computational models capable of integrating transcriptome, proteome, and metabolome data. With these in silico models we generate hypotheses of the mechanisms of organization in the cell, and subsequently evaluate them experimentally with the aim of understanding the alterations of a disease and potentially having an impact on the clinic,” said Resendis-Antonio.

The scientist and his team developed a model that, using gene expression data, is capable of predicting the proliferation speed of different cancer cell lines. “We went to a database of thousands of patients with many types of the disease, from which we chose 33 different types, with a total of 12 thousand patients.” They found that some types of cancer proliferate less and others more. But even within the same type there is heterogeneity in metabolic activity and expression profiles.

The next thing, he added, was to ask ourselves if we could establish, metabolically, what each one depends on because skin cancer is not the same as brain cancer. We studied several metabolic pathways and concluded that, although it is the same tumor, the metabolism changes according to each tissue and patient. This is a type of heterogeneity that needs to be explored.

But there is also intratumoral heterogeneity; one has the idea that cancer grows disproportionately, but no, within a cancerous tissue there are different subpopulations that carry out different functions.

For example, in a spheroid model (cell culture structures generated in 3D) of a breast cancer cell line, we have found three subpopulations. The first is associated with malignant cells that are growing, the second refers to those that are preparing to metastasize and cope with the immune system. Finally, a third population, which requires further study because it lacks a clear biological function. This intratumoral functional variability shows the difficulty of curbing the disease and highlights the relevance of heterogeneity in the efficiency of treatments.