The smell of mystery

A study carried out at Unicamp's Biology Institute (IB) reveals for the first time the repertoire of non-coding RNAs (not carrying genetic codes) from one of the least understood regions of the nervous system, the neurons of the olfactory system. RNAs are molecules produced in the nucleus of cells and many of them transport information from genes located in DNA to the place where proteins are manufactured, therefore being called coding RNAs. The types of RNAs described in the research perform a different function: they work to regulate the expression and activity of other genes in cells. As they do not carry messages for the production of proteins, they are called non-coding RNAs.

This is a multidisciplinary work that used the modern computational approach known as machine learning (machine learning algorithm ) to discover and catalog non-coding RNAs in neurons involved in olfaction. Subsequently, laboratory experiments were carried out to confirm the identity and place of synthesis of these molecules, some of which can regulate the perception of odors, one of the least understood processes in the functioning of the nervous system and which has gained increasing prominence in the scientific community.

Seeking to unravel these mysterious RNAs, groups led by biologist Fábio Papes, professor at the Department of Genetics, Evolution, Microbiology and Immunology at the Institute of Biology (IB) at Unicamp, and biochemist Bettina Malnic, professor at the Institute of Chemistry at USP, demonstrated for the first time the expression of a large set of non-coding RNAs, a component that is still very little known in the cells of all living organisms. The work, which also included the collaboration of researcher Pedro Galante, from Hospital Sírio-Libanês de SP and received funding from FAPESP and CNPq, has just been published in the journal DNA Research, from Oxford Press (https://academic.oup.com/dnaresearch/article/26/4/365/5535672). The article resulted from the master's thesis by biologist Antonio Pedro Camargo, supervised by professor Fábio and co-supervised by researcher Marcelo Falsarella Carazzolle, who coordinates the bioinformaticians at the Genomics and Bioenergy Laboratory, where the work was developed. The study also included the significant participation of the professor's then doctoral student, Thiago Seike Nakahara, now a postdoctoral fellow of Professor Bettina, and other postgraduate students from the group coordinated by Professor Fábio Papes.

The professor has been dedicating himself for more than 15 years to studying the olfactory system in animals. In an article in Jornal da Unicamp published in May 2010, motivated by an article he co-authored and which received a cover illustration in the magazine Cell, he had already predicted that understanding this system could provide important support for the future understanding of behavioral diseases in humans , of great importance for the medical field. The article highlighted that the detection of chemical compounds present in the environment is essential for animals to be able to escape dangers and predators, hence its title The smell of fear (https://www.unicamp.br/unicamp/unicamp_hoje/ju/maio2010/ju462_pag12.php#). In April 2016, reflecting another publication by the professor, now in the journal BMC Biology, Jornal da Unicamp published the article entitled The smell of hate (https://www.unicamp.br/unicamp/ju/652/animais-dao-pistas-sobre-comportamentos-humanos), publicizing the discovery of the population of neurons in the noses of mice associated with infanticide, that is, the massacre of puppies by adult males who had not yet started their sexual life. The discovery could also provide clues to explain human behavior. With this new publication, The smell of mystery, allusion to the still mysterious function of non-coding RNAs, Jornal da Unicamp completes a trilogy on the work led by professor Fábio.

The author of the master's thesis that led to the current publication, Antonio Pedro Camargo, explains that the compounds present in the environment are essential for animals to be able to interact with the external environment. They allow the location of food, prevent the ingestion of toxic substances, warn of the presence of predators and facilitate interaction with individuals of the same species. For this reason, the olfactory system is the main sensory system and, probably, the most important for terrestrial animals, as it allows them to recognize a huge range of environmental stimuli and trigger behavioral, physiological and endocrine responses.

As a result, the central objective of the work was to identify non-coding RNAs possibly involved in biological processes in the olfactory organs of mice. This basic research provides elements for subsequent investigations of the molecular mechanisms and functional roles of these substances in the system in question, which could have repercussions in the medical field in the future. As will be seen below, the researchers' exhibition constitutes a beautiful example of multidisciplinary work developed using bioinformatics, artificial intelligence, mathematics, statistics, laboratory benches and microscopy in the field of modern molecular biology.

A necessary explanation

DNA, deoxyribonucleic acid, is known as genetic material even by laypeople who have heard of paternity testing. The general public knows that it is located in human cells and is responsible for the construction and functioning of the body. But perhaps few people are aware that DNA does not work alone. There is another category of genetic material present in the cells of organisms called RNA, called ribonucleic acids, whose molecules work together with DNA, without which they would not be able to perform their functions.

Several types of RNAs are found in cells, each of which performs its function. They were discovered more than fifty years ago, when people were trying to understand how cells worked from a molecular point of view. At this time, messenger RNAs were identified which, as the name suggests, carry genetic information from DNA to manufacture proteins and, therefore, are known as protein coding; transporters, responsible for transporting amino acids within cells to the site of protein synthesis based on information contained in DNA; ribosomes, which control the protein synthesis machinery, the ribosomes. All of these types are involved in the production of proteins.

The study developed by Antônio Pedro focused on another category of RNAs, discovered more recently and which, unlike those mentioned, does not specifically perform functions that lead to protein synthesis: they are called non-coding RNAs. They remain very mysterious and little is known about them, although they perform functions that the scientific community is now trying to unravel. "Discovering the function of these molecules constitutes the great challenge for the coming years", says Professor Fábio.

All cells in an organism, with very few exceptions, have the same DNA, whether from the skin, liver or brain, with very few differences. However, the RNAs of cells in different organs are distinct, which contributes to differences in function between cells. The research carried out did not just focus on investigating a single type of non-coding RNA, but sought to discover all mouse RNAs belonging to this class, and this is perhaps the biggest difference of the work.

Antonio Pedro explains why: “Studying only the olfactory system would not guarantee that we would discover all the non-coding RNAs existing in the cells of mouse organisms. That's why we use data from experiments available on the Internet and from research with various tissues of these animals, such as lung, brain, heart, liver, and then we apply the process of discovering non-coding RNAs to all of them. Then we focus on the olfactory system. Why? Because it has a series of very interesting characteristics. For terrestrial animals it is the most important sensory system, used to discover food, to prevent the ingestion of unsuitable foods, to avoid dangerous situations and to influence behavior in relation to the presence of the female and awaken in them instinct for maternal care. It is, therefore, a system involved in a very wide range of behaviors. Despite this importance, little is known about how it works at the cellular level.”

Furthermore, the advisor adds that "every researcher is always looking for themes that have not yet been studied, which offer the possibility of new inquiries, instead of restricting themselves to subjects that have already been exhaustively worked on. Hence the search for what is least known. And the sensory system olfactory is perhaps the least understood in nature".

Bioinformatics and artificial intelligence

When proposing new discoveries within cells, researchers can adopt several paths, including, as in the case under study, extract non-coding RNAs from olfactory cells and from them obtain the sequence of nucleotides, which are the bases that constitute them. With this approach, the identity of the RNAs is revealed. Another strategy employs bioinformatics. To carry it out, says Marcelo, “we collected data from the literature relating to all known RNA sequences from all mouse tissues, which resulted from studies by many research groups around the world over decades. This prospected biological database was transported to the computer. Bioinformatics tools, together with artificial intelligence resources, allow for transcriptomic analysis which, in turn, makes it possible to find answers to specific questions, such as, for example, which transcripts are present in the neurons of the olfactory system. Questions like these can only be formulated based on knowledge of the whole. Although we had the transcripts generated by all mouse tissues, we focused only on those specific to the olfactory system. Using bioinformatics resources, it is possible to process the millions of RNAs that make up organisms, a possibility that did not exist until the emergence of computational resources.”

Although the sequential patterns of coding RNAs are well known, the same is not true for non-coding RNAs. To establish these unknown patterns, in addition to bioinformatics, artificial intelligence tools were used, especially from the machine learning class (machine learning algorithm ). "There are algorithms that allow recognizing and establishing unknown sequences for the various groups of nucleotides. In this way, in the gigantic list of RNAs, non-coding RNAs can be distinguished, for which an identification pattern had not yet been determined", explains Marcelo.

Laboratory validation

But how do we know if the molecules described by computational tools correspond to those that actually exist in cells? In natural sciences in general, everything predicted by models needs to be validated on the laboratory bench, based on real materials, so that confirmation is given by the system or organism under study. The use of mice at this stage was due to several reasons. It is a living organism that has a series of characteristics that make the study viable: they are animals created in the laboratory and reproduce quickly; they can be observed from behavioral points of view, in this case, related to smell; offer the availability of a large amount of biological material, with the extraction of the olfactory organ, which would not be possible with humans, allowing the study of their non-coding RNAs on-site visit, through microscopy.

Professor Fábio summarizes the validation process: “With the resources of bioinformatics we discovered around ten thousand RNA sequences in the olfactory system of mice. It is an enormous amount of material that requires years of research. Therefore, from this list we selected only a few RNAs that, according to computational predictions, would be present in large quantities in the tissue studied, and therefore considered highly expressed. Furthermore, we were particularly interested in non-coding RNAs from olfactory neurons, specific to cells that detect odors and that were not present in cells from other organs. As we intend to study the function of these olfactory neurons, determine how they act in detecting smells and controlling behavior, it makes perfect sense to search for molecules that are present in a special way in this tissue. Therefore, in the list of ten thousand genes mentioned by Antônio Pedro, we focused on twenty of them, which are more specific to the olfactory system and which appear in greater quantities there, if the computational predictions are correct”.

It turns out that in the nose there are thousands of different types of neurons, each of which performs the detection of specific sets of odors, with the functions of most of them not being known. Therefore, there is still a need to determine in which of the immense varieties of neurons each non-coding RNA is present. There is the possibility that a certain RNA is present in all olfactory neurons, with no location specificity; but there is also the more favorable possibility that it is located in just a few types of neurons, performing specific functions. In addition to these, another possibility must be considered. Inside the nose, in addition to neurons, there are stem cells, which are those capable of proliferating, dividing and deriving other types of cells from the organism and which, in this case, are those that manufacture new olfactory neurons. Therefore, it is possible that the experiments localize a certain type of non-coding RNA in these stem cells, with the function of promoting the replacement of olfactory neurons within the nasal cavity. All these possibilities can be investigated in situ, directly at the location where they are expressed, using slides and microscopy.

In relation to the twenty non-coding RNAs, the most important part of the work, the professor makes a point of highlighting two. One of them is found specifically in the stem cells of the nose and is not present in other stem cells in the body. The finding suggests that this non-coding RNA is responsible for controlling how stem cells produce new neurons as the organism needs them. The answers that explain this possible functionality will require around five to ten years of research. Regarding the action of other RNA, he explains that olfactory cells have a series of points that are poorly understood. One of its functions is the detection of odors through proteins located on the surface of its neurons. These are the so-called olfactory receptor proteins that, in this case, identify odors. Although they are found in olfactory neurons, very little is known about how the organism determines and controls their production in the various types of neurons in the nose. There are more than a thousand types of these proteins and they are important because they are the largest family of proteins existing in nature, even when considering the numbers found in microorganisms, bacteria, fungi, yeasts, animals and plants. Despite this, it is known very little about how the body's cells decide how to express, how to synthesize these odor receptor proteins. The professor adds: “In his work, Antonio Pedro identified, among the twenty RNAs selected for further investigation, one present in mature olfactory neurons, which are those that manufacture odor receptors. Again, it can be suggested that these non-coding RNAs participate in the control process in the manufacture of these receptors, which is fundamental in the detection of smell through the nose.”

Marcelo informs that the knowledge produced by the group may be used by other researchers. To make this possibility possible, he says, “we produced another article, specifically about the bioinformatics methodology used in our work. This is software that will allow other researchers, using other types of data and asking specific biological questions, to be able to use the same methodology. This software has already been submitted to a magazine and we are waiting for feedback to make it available so that other researchers can learn about its architecture and functionality.”

Mention in The New York Times

Article recently published in The New York Times about research carried out by the group led by Brazilian Alysson Muotri at the University of California, San Diego, USA, (https://www.nytimes.com/2019/08/29/science/organoids-brain-alysson-muotri.html?fbclid=IwAR3TtXBWvxZhno-OJoAzgBl_wch23kd99h-IPRcg_J2Ge_ZykksOI84jPXo ) mentions the collaboration maintained with professor Fábio Papes and Unicamp. The collaborative project seeks to understand an autism spectrum disease known as Pitt-Hopkins Syndrome, which is being investigated in both groups using so-called mini-brains, which are clusters of cells vitro manufactured to simulate the development of children's brains, allowing in-depth studies of the disease.