Grand Challenge: Finding similarities and differences in mammalian brain organization

Ruth Benavides-Piccione^*

• Cajal Institute, Spanish National Research Council (CSIC), Madrid, 28002, Spain

There are over 6000 currently recognized species in the class Mammalia (Burgin et al., 2018). This biodiversity is the enduring result of past events and also presents environmental and ecological conditions (Jones and Safi, 2011). The ability to interact with the environment and display a range of behaviors depends on the coordinated interaction of different structures of the nervous system with specialized functions, from sensory receptors through to higher sensory and motor processing centers. Thus, it is crucial to comprehend how these circuits operate in order to understand mammalian behavior.Over the years, neuroscientists have been searching for the organizational principles underlying mammalian brain function. From classical studies to advanced modern techniques, significant efforts have gone into understanding brain circuit organization. The first diagrams of brain circuits were created in the late 19th century, primarily by Cajal, using the Golgi method (reviewed in DeFelipe, 2002). This technique allowed the detailed study of the morphology of neurons and their connections, marking a significant milestone in understanding brain architecture. Since then, the introduction of new methods and techniques have enabled researchers to progress in the study of brain organization to better understand brain function and its role in cognition and behavior. Nowadays, big interdisciplinary international projects (e.g., Human Brain Project, Blue Brain Project, Brain Initiative, Human Connectome Project, Allen Institute Human Program, The China Brain Project) are making use of advances in imaging, artificial intelligence, and computational neuroscience, with the aim being to fully map brain connections. However, despite the outstanding progress made by these projects, the vast majority of neuroscientific studies in mammals have traditionally focused on the investigation of only a few species (mainly rodents and non-human primates). These species have been primarily chosen due to their suitability for standardized laboratory studies or their genomic similarities to humans (as with monkeys). Nevertheless, many mammals exhibit unique capabilities that are not characterized in these species. Thus, one of the major challenges in mammal neuroscience is to support the study of a broad range of species, in order to reveal both conserved and species-specific features, ultimately leading to a better understanding of mammalian brains and their role in inducing behavior.Since the very early studies of brain organization, neuroscientists have been trying to understand how features such as brain size, the number of brain regions, cell lamination patterns, and interconnections between areas are organized in the different brain regions and how they relate to cognitive abilities. There has been a long-standing debate regarding the uniformity versus non-uniformity of the brain organization, with some researchers emphasizing the similarities, while others highlight the differences (reviewed in DeFelipe et al., 2002). For example, the cerebral cortex has been traditionally divided into a number of cytoarchitectonic fields that can be distinguished from their neighbors based on differences in the overall density, size and shape of the cells and their arrangement in cortical layers, supporting the idea that differences in cortical organization would give rise to a distinct and specialized neural architecture (e.g., Brodmann, 1909;von Economo, 1909; for a review, see Amunts and Zilles, 2015). Other researchers proposed that functional differences between areas are mostly due to connections (e.g., Szentágothai, 1978;Creutzfeldt, 1977;Rockel, et al., 1980, Douglas andMartin, 2004). Supporters of this view affirm that during evolution, the complexity of the neocortex increased in larger brains due to the addition of microcircuits with the same basic structure. However, when a range of species other than those commonly used (mouse, rat, cat, monkey, and human) are considered, particular new arrangements are found (e.g., Haug, 1987;Glezer et al., 1988;Reep et al., 1989;Stolzenburg et al., 1989;Hof et al., 2000;DeFelipe et al., 2002;DeFelipe, 2011, Chegentanai et al., 2020;Manger et al., 2021). That is, new insights can be revealed by examining the diversity of species. For example, analyzing brains that are larger than the human brain can be of great interest, such as the case of the brains of African elephants, in which it has been revealed that there are three times more neurons than in the average human brain; however, the great majority of these neurons are found in the cerebellum, showing that it is the larger absolute number of neurons in the human cerebral cortex (but not in the whole brain) that correlates with the superior cognitive abilities of humans (Herculano et al., 2014).Alternative developed methodologies have also allowed the study of new aspects of circuitry, and comparative studies are becoming more common. Indeed, there is increasing evidence that each species presents certain unique molecular, anatomical and physiological features, humans included (e.g., Preuss and Coleman, 2002;Oberheim et al., 2009;DeFelipe, 2011;Sherwood et al., 2012;Geschwind and Rakic, 2013;Hawrylycz et al., 2012;Kaas, 2013;Eyal et al., 2018;Sousa et al., 2017;Verendeev and Sherwood, 2017;Molnar et al., 2019;Elston et al., 2001;Marchetto et al., 2019;Hodge et al., 2019;Kalmbach et al., 2021;Lee et al., 2023;Galakhova et al., 2022;Luria et al., 2023;Benavides-Piccione et al., 2024;Kanari et al., 2024). In this regard, the concept of species-specific types of neurons is in fact a matter of debate since the definition of a cell type depends on the morphological, physiological, molecular and genetic composition (e.g., Ecker et al., 2017). As a consequence, it is key to examine the diversity of species from different perspectives and encourage anatomical, physiological and molecular researchers to reach consensus on current controversial terms and issues (preferably by meeting in person), in order to clarify brain organization in the different species (e.g., PING, 2008;Nelson, 2002;DeFelipe et al., 2013;Molnar et al., 2019;Yuste et al., 2020).The study of different brains allows comparisons of features across brain regions/species. If a particular brain microcircuit shows specific patterns responsible for the information processing of a particular brain region, it can be investigated whether such patterns can serve as a foundation for comparing organizational principles across various brain systems, aiming to uncover both shared principles and region/speciesspecific adaptations (reviewed in Shepherd and Grillner, 2010). Taking the pyramidal cells (the basic building block of the cerebral cortex) as an example, it is possible to analyze to what extent these cells have parallel morphologies in the different cortical regions and species, by comparing distinct anatomical features, which have important functional implications. Briefly, pyramidal cells are basically composed of distinct dendritic apical and basal compartments which receive and integrate information from functionally diverse areas (DeFelipe and Fariñas, 1992;Spruston, 2008;Aru et al., 2020). These cells have been shown to be characterizedamong different areas and species-by markedly different dendritic structures, which are directly related to function (reviewed in Elston et al., 2011). For example, certain areas of the prefrontal cortex of various primate species, including humans, have larger pyramidal cells, which are more branched and more spinous than their counterparts in the occipital, parietal and temporal lobes (e.g., Lund et al., 1993;Elston et al., 2001;Jacobs et al., 2001;Luebke, 2017;Benavides-Piccione et al.,, 2024). Also, there is a trend towards increasing pyramidal cell complexity with anterior progression in the occipito-temporal cortex (reviewed in Elston, 2003). Regional variation in pyramidal cell structure has also been shown in mice, albeit to a lesser degree (Benavides-Piccione et al., 2006;Ballesteros-Yañez et al., 2010). Briefly, the size of dendritic arbors influences their sampling geometry and the mixing of inputs; the patterns of dendritic branching may determine the degree to which the integration of inputs is compartmentalized within their arbors; and the density of dendritic spines influences various aspects related to the integration and co-operativity of inputs (e.g., Koch et al., 1982;Shepherd et al., 1985;Malach, 1994;Elston, 2003;London and Häusser, 2005;Spruston, 2008). Specifically, human pyramidal cells show greater, and not scalable, dendritic computation complexity in certain regions compared with pyramidal cells in other species, which accounts for the demonstrated singularity of biophysics of these neurons (e.g., Jacobs et al., 1997Jacobs et al., , 2001;;Jacobs and Sheibel, 2002;Elston et al., 2001;Zeba et al., 2008;Anderson et al., 2009;Hustler and Zang, 2010;Beaulieu-Laroche et al., 2018;Eyal et al., 2016Eyal et al., , 2018;;Gidon et al., 2020;Benavides-Piccione et al., 2020, 2021, 2024;Mertens et al., 2024;Kanari et al., 2024;Masoli et al., 2024). Nevertheless, there are also relatively small and simple human pyramidal neurons, such as those in the visual cortex (Elston et al., 2001;Benavides-Piccione et al., 2024). Interestingly, in brains that are larger than human brains -such as that of the African elephant-longer dendritic segments are found, but there is less intricate branching than that observed in human pyramidal cells. Also, African elephants show regional variation, as do other rodent and primate species (Jacobs et al., 2011;Bianchi et al., 2001). Thus, through detailed analyses of particular features of pyramidal neurons across regions and species, it is possible to find some common dendritic organizational patterns. Examples of such patterns that have so far been determined as conserved are as follows: pyramidal cells' dendritic diameter values decrease as the branch order increases; the length of dendritic segments increases with higher branch orders; intermediate segments are thicker and shorter than terminal segments; terminal segments of a pyramidal neuron exhibit similar widths; and main apical dendritic diameter correlates with axonal diameter and soma sizewhereas, there are other features whose variation is found to contribute to the region/species-specificity, such as the dendritic diameter, number of primary dendrites, branching complexity and spine density (Benavides-Piccione, 2024; see also Elston, 2002;Benavides-Piccione et al., 2020, 2021;de Kock and Feldmeyer, 2023). Thus, there are some features that reflect a general trend in the structural organization and design of pyramidal neurons, whereas other features represent specific morphological parameters that contribute to the existing diversity within pyramidal cell structures across different areas and species. In addition, by identifying the distinct and conserved features between regions and species, it is possible to hypothesize how pyramidal cell complexity may have possibly increased during cortical expansion: (1) an increase in dendritic diameter, followed by the further dendritic width enhancement of apical main and basal dendrites, along with an increase in axonal diameter; and/or (2) An enlargement in neuron size, involving: a) extension of distal dendritic segment lengths; b) increase in dendritic complexity (e.g., number of nodes and dendrites); and c) increase in the number of dendritic spines (Benavides-Piccione 2024).In addition, since morphological features highlight significant variations in the processing of information, it is possible to build models that demonstrate the biophysical and computational distinctiveness of neurons in each of the different regions and species (e.g., Eyal et al., 2016Eyal et al., , 2018)). Furthermore, since a relationship between microscale cytoarchitecture and macroscale connectome organization has been established in several species, including humans (e.g., Scholtens et al., 2014;Barbas, 2015;van den Heuvel et al., 2015van den Heuvel et al., , 2016;;Beul et al., 2017;Garcia-Cabezas et al., 2019;Wei et al., 2019), the more elaborate the identification and extraction of features of the pyramidal neuron, the more comprehensive the characterization of the macroscale organization. Therefore, it is essential to further identify and extract the features that capture the functional properties of the pyramidal neuron in the different cortical regions and species. Moreover, the study of the human brain in health and disease will not only help to better understand the mechanisms underlying human brain function, but will also provide new insights into the underlying disease mechanisms of neurodegenerative and neurodevelopmental brain disorders.A main concern regarding the comparison between species is the extrapolation of data between regions and species. The case of the prefrontal cortex (PFC) is particularly relevant in this regard since its function is still poorly understood, and potential interspecies differences remain the subject of much debate, as demonstrated by a recent workshop that brought together experimental and computational scientists to discuss this matter (https://www.humanbrainproject.eu/en/education-trainingcareer/workshops/pfc/). Here, we will focus only on a few of the most pertinent points that were debated at this workshop. The granular prefrontal cortex (gPFC) is involved in a variety of high-level cognitive processes, particularly those involving executive control, attention, memory, and social behavior. It has undergone dramatic expansion in primates and it is composed of diverse regions that vary in terms of the size, density and distribution of their components, displaying a complex set of connections and a diverse gene expression repertoire (reviewed in Povinelly and Preuss, 1995;Goldman-Rakic, 1996;Kaas, 2013;Fuster, 2001;Kolk and Rakic, 2022;Preuss and Wise, 2022). Nevertheless, the long-standing question alluded to above remains, that is, it has not yet been defined the extent to which it is possible to extrapolate from the whole PFC to specific regions of PFC, or species, to make comparisons (e.g., Preuss, 1995). For example, rodents have homologs of the agranular areas found in primates but lack homologs of the granular cortex, which constitutes the largest part of the PFC in most primate species. Likewise, connectivity observed in primates, as a result or consequence of the new areas generated in primates, cannot be studied in mice. Thus, it could be agreed that the delimitation of the PFC across species would be based on the presence of a gPFC. Similarly, the overall homology of the areas between species should be revised in order to define a more appropriate extrapolation of data. On a similar note, it ought to be better defined to what extent the same behavioral task can be applied to different species, highlighting potential limitations when comparing tasks across species. In particular, inferring from animal models to humans requires even more careful evaluationnot only due to species-specificity, but also because there are technical and ethical constraints that limit the methods that can be used to study of the human brain. Consequently, understanding the human brain requires its direct analysis whenever possible and there is a clear need for more strategic tools to achieve this efficiently. Likewise, it is important to outline the types of experiments or strategies that should be employed to examine each brain species. Finally, interindividual variability should also be considered, particularly in humans and the PFC region, which exhibit greater variability than that reported in other species (e.g., Jacobs and Scheibel, 1993;Peng et al., 2019;Benavides-Piccione et al., 2021).In summary, it is it is essential to support the study of a broad range of species. By actively studying a wider variety of species -rather than focusing solely on mice, rats, and other primates-the animal kingdom diversity becomes evident. Identifying both conserved and species-specific features will help uncover the neural mechanisms underlying diverse mammalian behaviors. The human brain has some many unique features, as each species has its own particular traits. Encouraging studies on the human brain is crucial to ensure it is better understood. Multidisciplinary approaches and collaboration between experimental and computational scientists are necessary to establish consensus on key issues related to brain organization across species.