Mammalian Genome (2023) 34:389–407 https://doi.org/10.1007/s00335-023-10005-4 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for disease model phenotyping Jesús Ruberte1,2 · Paul N. Schofield3,4 · John P. Sundberg3,5 · Alfonso Rodriguez‑Baeza6 · Ana Carretero1,2 · Colin McKerlie7,8 Received: 16 April 2023 / Accepted: 13 June 2023 / Published online: 8 July 2023 © The Author(s) 2023 Abstract The laboratory mouse is the foremost mammalian model used for studying human diseases and is closely anatomically related to humans. Whilst knowledge about human anatomy has been collected throughout the history of mankind, the first comprehensive study of the mouse anatomy was published less than 60 years ago. This has been followed by the more recent publication of several books and resources on mouse anatomy. Nevertheless, to date, our understanding and knowledge of mouse anatomy is far from being at the same level as that of humans. In addition, the alignment between current mouse and human anatomy nomenclatures is far from being as developed as those existing between other species, such as domestic animals and humans. To close this gap, more in depth mouse anatomical research is needed and it will be necessary to extent and refine the current vocabulary of mouse anatomical terms. Introduction comprehensive study of mouse anatomy was published by Margaret J. Cook in 1965 (Cook 1965). Since the release of The mouse remains the key animal model for exploring Cook’s atlas, several books and resources of mouse anat- human disease and, despite its small comparative size, the omy and have been published (Table 1). Nevertheless, the laboratory mouse is anatomically similar to humans, pro- understanding and published descriptions of mouse anatomy viding even unexpected anatomical analogies in structures (Mus musculus) is far from the body of knowledge available with high interspecies variation such as the presence of the for humans (Homo sapiens). The most recently discovered clavicle. Knowledge of human anatomy has been collected anatomical structure in humans, the cisterna chyli, was first and recorded throughout human history, however, the first described four hundred years ago (Natale et al. 2017). In contrast, there still remain examples of anatomical structures of the mouse such as the bulbourethral glands with only * Jesús Ruberte poorly described structural detail. Other examples of the gap jesus.ruberte@uab.es in complete description between human and mouse anatomy 1 Center for Animal Biotechnology and Gene Therapy, are: (1) the extant knowledge of normal human anatomical Universitat Autònoma de Barcelona, Barcelona, Spain variation is well documented and quantified, but is poorly 2 Department of Animal Health and Anatomy, Universitat characterized in the mouse. As an example, a previously Autònoma de Barcelona, Barcelona, Spain accepted anatomical difference between mouse and human, 3 The Jackson Laboratory, Bar Harbor, ME, USA the presence of a supratrochlear foramen in the humerus, is 4 now known to be incorrect because a percentage of humans Department of Physiology, Development and Neuroscience, University of Cambridge, Cambridge, UK present with this foramen; (2) the disparity between the 5 number of anatomical entities in “gold-standard” ontolo- Department of Dermatology, Vanderbilt University Medical Center, Nashville, TN, USA gies: 3257 in the Mouse Anatomy (MA) ontology (version 6 2017-02-07) (Hayamizu et al. 2005b) versus 7800 in the Ter- Department of Morphological Sciences, Universitat Autònoma de Barcelona, Barcelona, Spain minologia Anatomica (TA) (version 2021-16-08) (F.I.P.A.T 7 2019); and (3) existing mouse and human anatomic ontolo- The Hospital for Sick Children, Toronto, Canada gies (Rosse and Mejino 2008) have only ~ 50% of terms have 8 Department of Lab Medicine and Pathobiology, Faculty a conceptual or lexical match. of Medicine, University of Toronto, Toronto, Canada 13 Vol.:(0123456789) Table 1  Publications on mouse anatomy 390 Title Year Authors Publisher Characteristics 13 Biology of the Laboratory Mouse 1941 Snell GD Dover Publications One chapter on mouse anatomy by Fekete E. Draw- ings and photographs from histological sections The Anatomy of the Laboratory Mouse 1965 Cook MJ Academic Press Atlas of drawings. Freely available at: http://​www.​ infor​matics.​jax.​org/​cookb​ook Atlas of the Mouse Brain and Spinal Cord 1971 Sidman RL, Angevine JB, Pierce ET Harvard University Press Atlas with coronal, sagittal, and horizontal brain and spinal sections and diagrams. Nomenclature in Latin The Mouse in Biomedical Research 1983 Foster HL, Small JD, Fox JG Academic Press One chapter on mouse anatomy by MJ Cook Necropsy Guide: Rodents and the Rabbit 1988 Feldman DB, Seely JC CRC Press Anatomical basis for the mouse necropsy. Photo- graphs and text A Guided Tour of Veterinary Anatomy. Domestic 1992 Smallwood JE WB Saunders Company One chapter dedicated to laboratory mammal Ungulates and Laboratory Mammals anatomy. Text and diagrams Colour Atlas of Anatomy of Small Laboratory 1992 Popesko P, Rajtová V, Horák J Saunders Company Atlas of colour drawings. Uses Nomina Anatomica Animals. Volume two: Rat, Mouse, and Hamster Veterinaria in Latin and English Pathology of the Mouse 1999 Maronpot RR Cache River Press Anatomical and histological introduction in each of the chapters Systematic Evaluation of the Mouse Eye. 2000 Smith RS, John SWM, Nishina PM, Sundberg JP CRC Press One section (3 chapters) dedicated to the eye Anatomy, Pathology, and Biomethods anatomy and development. Text and photographs Atlas of Mouse Hematopathology 2000 Fredrickson TN and Harris AW Harwood Academic Three chapters about the anatomy, histology, and cytology of lymphatic and hematopoietic systems Pathology of Genetically Engineered Mice 2000 Ward JM, Mahler JF, Maronpot RR, Sundberg J, Iowa State Press Several chapters with anatomical introductions Frederickson RM A color Atlas of Sectional Anatomy of the Mouse 2001 Iwaki T, Yamashita H, Hayakawa T Braintree Scientific Inc Color atlas with photographs of skeleton, dissec- tions, and transverse, sagittal, and dorsal sections Laboratory Animal Medicine. Second Edition 2002 Fox JG, Anderson LC, Loew FM, Quimby, FW Academic Press Short introduction to the mouse anatomy taken from Cook’s book Atlas of Laboratory Mouse Histology 2004 Conti CJ, Gimenez-Conti IB, Benavides F, Fri- Texas Histopages Inc Atlas with photographs and diagrams. CD format jhoff AFW, Conti MA Anatomia degli Animali da Laboratorio 2006 Cozzi B, Ballarin C, Peruffo A, Carù F Casa Editrice Ambrosiana Atlas with photographs from dissections and histo- logical sections The Mouse Brain in Stereotaxic Coordinates. 2008 Paxinos G and Franklin KBJ Academic Press Atlas with 100 coronal histological brain sections Compact Third Edition (photographs and diagrams) Allen Reference Atlas. A Digital Color Brain 2008 Dong HW Wiley Atlas with 151 histological brain sections (coronal Atlas of the C57BL/6J Male Mouse and sagittal) accompanied by diagrams and a sys- tematic hierarchically organized taxonomy. https://​ mouse.​brain-​map.​org Histologic Basis of Mouse Endocrine System 2010 Kaufman M, Nikitin AY, Sundberg JP CRC Press Anatomy, histology and development of ovary, Development: A Comparative Analysis testis, pancreas, hypophysis, thyroid, parathyroid, and pineal glands Comparative Anatomy of the Mouse and the Rat 2011 Constantinescu GM CRC Press Color atlas with illustrations drawn from dissec- tions. Uses the Nomina Anatomica Veterinaria J. Ruberte et al. Table 1  (continued) Title Year Authors Publisher Characteristics The Mouse Nervous System 2011 Watson C, Paxinos G, Puelles L Academic Press Text with photographs providing a comprehensive anatomy of the central nervous system of the mouse Comparative Anatomy and Histology. A Mouse 2012 Treuting PM, Dintzis SM, Frevert CW, Liggitt D Academic Press Text and Atlas. For comparison with human and Human Atlas anatomy uses Netter images and MA ontology The Laboratory Mouse. Second edition 2012 Hedrich HJ Academic Press One chapter on mouse anatomy by Komárek V. Drawings taken from Popesko et al. (2002) A Practical Guide to the Histology of the Mouse 2014 Scudamore CL Willey Blackwell Histological text and atlas with the basis of mouse anatomy Morphological Mouse Phenotyping. Anatomy, 2016 Ruberte J, Carretero A, Navarro M Editorial Medica Pan- Text and atlas. 2200 original photographs from dis- Histology and Imaging 2017 america Academic Press sections, histological sections, and imaging. Uses the Nomina Anatomica Veterinaria Comparative Anatomy and Histology. A Mouse, 2018 Treuting PM, Dintzis SM, Montine KS Academic Press Text and Atlas. This second edition adds the rat as Rat, and Human Atlas comparator species The Laboratory Mouse. A guide to the Location of 2019 Johnson J, DelGuidice B, Bangari DS, Peterson E, CRC Press Anatomical basis for the location of principal ana- Tissues for Optimal Histological Evaluation Ulinski G, Ryan S, Thurberg BL, Callis G tomical structures in mouse X-Ray Annotation Mouse Atlas 2021 Ruberte J, Carretero A, Cater H, Gracia G, Lally Doctor Herriot SL Atlas. 152 X-rays and 590 anatomical references. C Uses the Nomina Anatomica Veterinaria Essentials of Laboratory Animal Science: Princi- 2021 Nagaran P, Ramachandra G, Srinivasan R Springer One chapter devoted to mouse anatomy and physiol- ples and Practices ogy Pathology of Genetically Engineered and other 2022 Sundberg JP, Vogel P, Ward JM Wiley Blackwell Several chapters with anatomical introductions Mutant Mice Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 13 391 392 J. Ruberte et al. Advances in imaging and high resolution phenotyping century, on generating genetically inbred strains of mice, in recent years have generated a need for a granular and has established the mouse as the most powerful organism complete anatomical terminology for the mouse which is for discovery of the genetic basis of disease. Mapping of cognate to that available for humans. To answer this need we spontaneous genetic variation between mouse strains, the conclude this commentary by discussing the development of experimental modification of the genome, and the ability to a standard anatomical nomenclature for the mouse following manipulate processes of development has greatly contrib- the principles of the TA terminology. Because the mouse uted to our understanding of the molecular mechanisms of and human should have similar complexity, this mouse Ter- human disease, and allowed the creation of models of human minologia Anatomica (mTA) should have a number of nor- diseases that can be used for the development of therapeutics malised anatomical concepts similar to that of the TA, and and further our understanding of pathobiology (Vandamme would greatly assist in the further development of the adult 2014). Furthermore, mice are a cost-effective research tool. mouse anatomy ontology and related resources. They have a short inter-generational lifespan, relatively short longevity, and they are easy to house and transport at scale. Mice are biologically very comparable to humans, have Mice are a small animal model ~ 15,000 genes with human orthologues, and are therefore but anatomically similar to humans susceptible to the majority of mono- and polygenic inher- ited disorders affecting the human population. Mice can be Recognition of the anatomical and physiological similarities genetically manipulated to mimic the causative or contribut- and commonalities between different species can be traced ing genetic mutation or mutations underlying many human in the literature to Aristotle who probably reflected estab- disease conditions (Brown 2021). Furthermore, mice are lished opinions when he wrote in the 4th C BCE: a cost-efficient research tool (Leader and Padgett 1980) “Ought we, for instance, to begin by discussing each with an accelerated lifespan (one mouse year equaling ~ 30 separate species-man, lion, ox, and the like—taking each human years), small footprint animal holding, and are rela- kind in hand independently of the rest, or ought we rather to tively easy to handle and transport; particularly as frozen deal first with the attributes which they have in common in germ plasm. virtue of some common element of their nature, and proceed Although in comparison with humans, mice are small from this as a basis for the consideration of them separately” (Fig. 1A), they display a relatively similar anatomy. As an (Περὶ ζῴων μορίων; On the parts of animals. Aristotle. example of this, Fig. 1B shows a comparison between the Trans. William Ogle, 1882). human and mouse femur; the latter enlarged using scanning The common anatomy, physiology and ontogeny of spe- electron microscopy. The same gross anatomical details: cies closely related to humans has made mammals the obvi- head of the femur, the lesser and great trochanter, as well ous model organisms for increasing our understanding of as, the trochanteric fossa, can be observed both in the proxi- human biology and disease since the Greco-Roman period, mal epiphysis of human and mouse femur. These anatomical and arguably earlier. Galen (129–216 CE) compared the sur- similarities not only represent a morphological homology gical anatomy of wounded gladiators from the arena of Per- between these two mammalian species, but they also reflect gamum and of his patients in Rome, with that of primates, similarities in the biomechanics and function of the bone. goats and pigs. Using the opportunities presented by dead The comparatively larger size of the head of the femur in and living individuals of all species, including occasion- humans reflects the different body positions of both species ally living humans, he attempted to establish the function (Fig. 1B). Humans are bipedal and the entire body weight of anatomical structures and their involvement in disease, is transmitted through the femoral heads to the ground. By and initiated a long tradition of developing surgical practice contrast, in the mouse, a quadrupedal species, the body using animal models (Matter 2013). weight is transmitted to the ground through all four limbs. Since Galen and his Greek precedents, animal models Sometimes, anatomical similarities between humans have historically played a critical role in the exploration and and mice are unexpected (Fig. 2). In bipedal primates and characterization of disease and in the development of novel humans, the clavicle, a bone belonging to the pectoral girdle, therapeutic agents and treatments, now being called the One is well-developed because there is an adaptation to tasks medicine, One pathology, One health concept (Sundberg that depend on using hands distant from the central trunk and Schofield 2009). The mouse has now become the most axis, such as climbing and reaching distant objects (Rock- widely used, and arguably the most important, mammalian wood et al. 2009). By contrast, in many quadrupedal mam- model for studying human disease. Mice are biologically mals, such as ungulates, carnivores and several rodents, the very similar to humans and manifest many of the same or clavicle has disappeared or has been dramatically reduced closely-related diseases (McGonigle and Ruggeri 2014). (De Souza et al. 2020). This reduction or loss is common to Most significantly, work since the beginning of the twentieth mammals that use the thoracic limb for cursorial locomotion 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 393 anatomical terms is lissencephalic, since its cerebral cortex does not have folds (gyri), and its telencephalon does not completely cover the mesencephalon and the cerebellum, as happens in humans (Fig. 3). Despite these overt ana- tomic differences, the mouse has nevertheless contributed enormously to our understanding of human lissencephaly and its underlying genetics, reflecting fundamental simi- larities in the development of the CNS between the species (Del-Valle-Anton and Borrell 2022; Mota and Herculano- Houzel 2015). There are other very well-established ana- tomical differences, such as the presence of a characteristic supratrochlear foramen in the mouse humerus (Fig. 4A), however, on close consideration this turns out not to be a real difference to humans. Proximal to the trochlea of the human humerus there is a thin region of bone that is sometimes perforated forming a supratrochlear foramen like the mouse (Fig. 4B). The prevalence of this foramen varies from 6.5% in Europeans (Mays 2008) to 26.7% of people from India (Shivaleela et al. 2016). Functionally, a supratrochlear foramen in human allows hyperextension of the elbow (de Wilde et al. 2004). Women who are more flexible than men and people that generally practice dis- ciplines that need improved flexibility, such as yoga, have higher prevalence of a supratrochlear foramen (Akabori 1934; Shivaleela et al. 2016). Similarly, it is reasonable to suppose that the supratrochlear foramen in mice pro- vides the elbow with the capacity to hyperextend. It is just as likely that careful examination of this structure in the Fig. 1  HYPERLINK "sps:id::fig1||locator::gr1||mediaobject::0" A Skeleton of a mouse relative to the human hand. B Comparison of the new, highly genetically variable, collaborative cross strains human and mouse proximal epiphyses of the femur. Similarity of the (Collaborative Cross 2012) and Diversity Outbred mice anatomical structures of the head of the femur (1), great trochanter (Churchill et al. 2012) will identify a wide variety features (2), lesser trochanter (3), and trochanteric fossa (4) using scanning that do or do not resemble human anatomy. microscopy and image size correction for the mouse. Mouse images modified from Ruberte et al. (2016) with permission As discussed below, the rate of discovery of new ana- tomical structures and entities in humans is now quite slow, but several recent discoveries were paradoxically (Rockwood et al. 2009). However, the quadrupedal mouse made first in the mouse. For example, the recent discov- has a well-developed clavicle that, as happens in man, joins ery of functional lymphatic vessels lining the dural sinuses the sternum and the acromion of the scapula (Fig. 2). A pos- originated in detailed functional and molecular studies in sible reason for this unexpected anatomical similarity may the mouse CNS (Aspelund et al. 2015; Eide et al. 2018; be the consequence of the thoracic limb freedom that mice Louveau et al. 2017, 2015). Understanding the functions have. Mice are often seen standing on their hind limbs trying and relationships of the glymphatic system and meningeal to climb their cage walls and this rearing action is associ- lymphatics (Eide et al. 2018). Similarly, with the discovery ated with exploration and manipulation of the environment. of the network of transcortical capillaries in long bones Recordings using high-speed close-up video of mice eating that now helps to explain previous puzzles in the migration seeds (Barrett et al. 2020) show well developed manual dex- of neutrophils and the closed circulation system of bone terity for manipulating small objects. (Grüneboom et al. 2019). It is noteworthy that in both of Despite these similarities, mice and humans do pre- these cases the development of imaging techniques and the sent clear anatomical differences. These manifest at the unique ability to use molecular markers, tissue prepara- macroscopic, microscopic and system levels. For exam- tions and experimental procedures in the mouse facilitated ple, the sexual dimorphism of the mouse kidney is well these discoveries. Are we moving into an era of discovery known (Butterfield 1972). One of the most notable macro- of new human anatomical features from the mouse or as a scopic differences lies in the mouse brain, which in human result of better analytical tools? 13 394 J. Ruberte et al. Fig. 2  The quadrupedal mouse has unexpected anatomical similarities to the bipedal human, such as the presence of the clavicle (1). The clavicle in both attaches the sternum (2) and the acromion of the scapula (3). Acromial end (4), body of clavicle (5), esternal end (6). Mouse images modified from Ruberte et al. (2016) with permission Mice and humans: a similar anatomical be different for genetically different populations. For exam- complexity but different level of knowledge ple, the supratrochlear foramen in human humerus was first and resources described two centuries ago (Meckel 1825). By contrast, the first comprehensive study of mouse anatomy was published Knowledge about human anatomy has been collected by Margaret J. Cook in 1965. Cook worked at the MRC throughout history. Since Galen's anatomical descriptions, Laboratory Animal Centre in Carshalton (UK) and carried and even before, up to the present, the level of detail found in out intracardiac latex injections describing for the first time human anatomical books, and other resources has continu- the mouse arterial and venous system. In Cook’s atlas fore- ously increased. This is especially true for anatomic varia- word (Cook 1965), W. Lane-Petter stated that at that time no tions. The classic, Bergman’s Comprehensive Encyclopedia complete study of the anatomy of the mouse had so far been of Human Anatomic Variation (Tubs et al. 2016), gives some made and that gap in the knowledge was even more surpris- an idea of the current level of knowledge; 1456 pages and ing as the mouse was at that time the most commonly used more than 12,500 references. vertebrate in the laboratory. An anatomic variation can be defined as difference in Subsequent to the publication of Cook’s atlas, several morphology that is outside of the canonical. It is impor- comprehensive textbooks and other resources in mouse tant to remember, however, that the canonical is often an anatomy and histology, singly or as part of larger com- accepted abstract standard, as the dominant morph may well pendia, have been published. Table 1 lists the title, year of 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 395 as a comparable and informative model of human biology and disease. Of originals and copies: the case of bulbourethral glands Anatomy is perceived as an objective scientific discipline, because it involves the observation and description of defin- able material entities. However, dissection, the most preemi- nent technique to physically dissociate and isolate individual structures for examination is, like other analytical scientific methods, susceptible to artifact. The discovery of bulboure- thral glands in the mouse and the historical way to under- Fig. 3  Mouse and human have anatomical differences in the brain. A stand its morphology and topography is a good example of The human cerebral cortex is gyrencephalic with characteristic gyri how dissection can be misleading, and that anatomy is not (ridges) and sulci (depressions or furrows). B The mouse cerebral always as obvious as the “nose on your face”. cortex is lissencephalic with a smooth surface and the telencephalon The bulbourethral glands are part of the male accessory (1) does not cover the mesencephalon (2) and cerebellum (3). Mouse image modified from Ruberte et al. (2016) with permission genital glands, which are located along the pelvic portion of the urethra producing secretions for the nutrition, transporta- tion and protection of spermatozoa. Not all mammals have bulbourethral glands. The dog lacks bulbourethral glands, which might explain why they were missed in Cook's book. In her atlas, six drawings were devoted to the male genital system, however, bulbourethral glands were not noted. In fact, it would be necessary to wait until 1983, when Cook, in a book about the use of mouse in biomedical research (Cook 1983) showed and drew the location of two paired bulboure- thral glands emptying into the urethra and associated with a urethral diverticulum (Fig. 5A). From that time, several clas- sic and modern mouse books used the same figure to locate and describe these glands (Maronpot et al. 1999; Treuting et al. 2018). However, this anatomical description for the mouse bulbourethral glands is not the only one that appears in textbooks. Several years before the publication of Cook´s atlas, The Jackson Laboratory published its celebrated book: Biology of Laboratory Mouse (Snell 1941) in which, E. Fekete, who at that time was responsible of the histology laboratory, carried out a different anatomical description for the mouse bulbourethral glands. Fekete’s representation Fig. 4  A A previously purported anatomical difference between showed the bulbourethral glands related with two urethral mouse and human humeri is the presence of a supratrochlear foramen (1) in the humerus of the mouse but not in the human humerus. B diverticula, each of them covered by the bulbospongiosus Variable presence of the supratrochlear foramen in humans. Trochlea muscle. As happened with Cook's description, Fekete's ana- of humerus (2). Mouse image modified from Ruberte et al. (2016) tomical interpretation was later followed by Popesko et al. with permission (1992) in its beautiful anatomy atlas of laboratory animals, and subsequently Popesko’s drawings were copied by V. publication, authors, publisher, and characteristics of the Komàrek in the first and second editions of the Laboratory extant resource. However, despite the steady release of text- Mouse book (Hedrich 2012; Hedrich and Bullock 2004). book or on-line resources since Cook’s seminal publication, Furthermore in 2011, G.M. Constantinescu, Professor of the depth of detailed description and understanding of vari- Veterinary Anatomy and Medical Illustrator at the Univer- ation in mouse anatomy is far from comparable to human sity of Missouri-Columbia, carrying out new original ana- anatomy thereby presenting a challenge for using the mouse tomical preparations for its book “Comparative Anatomy of 13 396 J. Ruberte et al. C.L. Scudamore (Scudamore 2014), which comments that in standard necropsy sampling protocols, the bulbourethral glands are usually missing. The difficulty of dissecting the bulbourethral glands is also obvious, when in a recent pub- lished guide for the location and orientation of mouse tissues for optimal histological evaluation (Johnson et al. 2019), no specific instruction is provided to dissect and sample the bul- bourethral glands. However, the abnormal swelling of these glands, that presents with male mice seeming to have “4 testicles”, enables easy identification of these structures such that they can be dissected and studied (Kiupel et al. 2000). To understand the anatomical relations of bulbourethral glands in the mouse and to clarify which of the two mor- phological representations is more accurate, injection of a polymerizable resin ­(Mercox®) in the urethra of C57/BL6J male mice was done via the urinary bladder. The four casts obtained were immersed in 60 °C soapy water for 24 h, cor- roded in 3% KOH and washed in distilled water. Casts were mounted on stubs, sputtered with gold and observed in a Hitachi S-570 scanning electron microscope at an acceler- ating voltage of 10–15 kV. Analysis of the urethral casts (Figs. 6B and 7B) and correlative dissections and histologi- cal sections (Figs. 6A, C and 7A) showed that the pelvic part of the urethra in mice, as happens in humans, presented a large membranous portion. There is a narrowing of the urethra when it turns around the ischial arch, the urethral isthmus, which is well documented in domestic mammals (Fig. 6B). At the penile urethra, inside the bulb of the penis, only a single urethral recess was noticed (Figs. 6B and 7B). Fig. 5  Comparison between Cook’s and Fekete’s representations of The histological sections showed that this diverticulum is mouse bulbourethral glands. A Cook described paired glands (1) with covered by the bulbospongiosus muscle and surrounded by only one urethral recess (2). B Fekete described two paired glands (1) and two paired urethral diverticula (2) (adapted from Cook 1965 and a narrow band of erectile tissue corresponding to the cor- Snell 1941) pus spongiosum (Fig. 6C). The two corpora cavernosa, that complete the penile erectile system, are inside the crura penis covered by the ischiocavernosus muscles (Fig. 6C). the Mouse and the Rat” (Constantinescu 2011) described The ducts of the two bulbourethral glands join the penile mouse bulbourethral glands as did E. Fekete in 1941. urethra distal to the urethral recess (Fig. 7). The urethral The last macroscopic organ discovered in humans was the recess of the mouse has a homonymous dilation of the ure- cisterna chyli, that was first described in 1651 by the French thra in male humans. Our study suggests that the anatomical physician Jean Pecquet, as a lymph reservoir at the origin description of Cook (1983), although incomplete, is more of the thoracic duct (Natale et al. 2017). In other words, accurate (Fig. 5). Only one diverticulum is present in the for almost four hundred years we have had a very complete mouse urethra and not two as Fekete (1941) and her follow- description of human macroscopic anatomy. By contrast, ers point out. in the mouse we have severe discrepancies in anatomical descriptions for important organs, such as the bulbourethral gland (Fig. 5). One reason for these discordances could be that the classic technique used to collect the accessory geni- Mouse vs human anatomical terminology tal glands in male mouse, a ventral approach after opening and nomenclature the abdominal cavity, is not optimal to identify and dissect the bulbourethral glands, which are located laterally at the Terminology is understood as a set of terms used in a base of the tail between the crus and the bulb of the penis specific scientific field, whereas nomenclature is a nor- (Fig. 6A). This difficulty was brought to light in several malized system of exactly defined terms arranged accord- mouse histology books, such as the one published in 2014 by ing to certain classification principles. A nomenclature 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 397 Fig. 6  Analysis of the urethra in C57BL/6J male mice identified only one urethral recess within the bulbospongiosus muscle. A Gross dissection of the pelvis (lateral aspect). B Scanning microscopy of corrosion cast of the urethra (lateral aspect). Hematoxylin and eosin stained histological section of the root of the penis. Urethral recess (1); bulbospongiosus muscle (2); bulbourethral gland (3); corpus cavernosum (4); ischiocaverno- sus muscle (5); urethra mem- branosa (6); urethral isthmus (7), penile urethra (8); penis (9); rectum (10). Mouse images modified from Ruberte et al. (2016) with permission is approved and refined by a scientific commission and is There are 7635 terms in the TA and 6500 in the NAV, widely accepted by the professional community (Kachlik that are equivalent and have the same principles: (1) each et al. 2008). structure should be designated by a single term; (2) each The origins of anatomical terminology date back to the term in the official list is in Latin, but with liberty to trans- ancient period, more than 2500 years ago, however, the late; (3) each term is, so far as possible, short and simple; first effort to compile a unified anatomical terminology, the (4) the terms are primarily memory signs, but do have some Basle Nomina Anatomica, was made in 1885 (His 1885). At informative or descriptive value, and (5) terms derived from that time, the number of human anatomical terms reached proper names (eponyms) should not be used. 50,000, many of them synonyms. Structures and organs were Largely driven by the advent of high throughput gene given eponyms from the person first credited with the first expression analysis it became important to capture knowl- description, and in many cases the same organ was asso- edge about anatomy that might be used computationally ciated with the names of different anatomists in different across very large datasets. This need was further reinforced countries (O’Rahilly 1989). This first Nomina Anatomica by the development of large phenotype databases for the was followed by seven revisions. The last revision was pub- mouse which needed a formally computable structure in lished with the title of Terminologia Anatomica (TA) and order to search it. Very rapidly it became apparent that consists of a list of terms in Latin including their corre- these needs would be answered by the development of for- sponding translation in English (F.I.P.A.T 2019). The TA mal ontologies to describe anatomy, at all levels, and pheno- is not applicable to domestic animals because the terms type—the latter necessarily needing to incorporate anatomy of direction were based on the erect position of the human into its concepts, and pathology where the location of a body and not in the quadrupedal position of animals, like the lesion needed to be captured. mouse. Therefore, a Nomina Anatomica Veterinaria (NAV) For these needs to be met, the “anatome”, the complete was established and today it is in its sixth revised edition set of anatomical structures associated with an organism, (I.C.V.G.A.N. 2017). needed to be organized in a computer-comprehensible way 13 398 J. Ruberte et al. navigating hierarchies of concepts—a knowledge structure that goes back to the Etymologiae of Isidore of Seville in the sixth century CE (Barney et al. 2006). A formal ontol- ogy allows for modelling of the knowledge within it and it is interesting that here several anatomical ontologies, nota- ble the Foundational Model of Anatomy (FMA) ontology (Rosse and Mejino 2008), the time-dependent embryological anatomy ontology of the mouse (EMAPA) (Hayamizu et al. 2013) and the Mouse Anatomy (MA) ontology, differ. As we will see below these differences in modelling decisions make it complicated to map and align these ontologies, spe- cifically the FMA, which is important for integrating human and mouse data. The development of the Uberon metazoan ontology (Mungall et al. 2012) (https://o​ bofou​ ndry.o​ rg/o​ ntol​ ogy/​uberon) was designed to bridge the anatomical ‘silos” particularly between vertebrates. The first attempt to develop a mouse anatomy ontology was an internet-accessible database of names and synonyms of the tissues in the first 22 Theiler stages of development (Bard et al. 1998). A few years later, an adult Mouse Anat- omy (MA) ontology was developed in The Jackson Labora- tory (Hayamizu et al. 2005a). MA ontology contains 3300 unique terms to provide standardized nomenclature for ana- tomical structures in the mouse. The MA ontology can be accessed at the Mouse Genome Informatics website: Fig. 7  A Gross dissection of the root of the penis (caudal aspect). (MGI, https://​www.​infor​matics.​jax.​org/​vocab/​gxd/​ma_​ B Scanning microscopy of corrosion cast of the urethra (caudal ontol​ogy). aspect). Urethral recess (1); bulbourethral gland (2); penile urethra (3); bulbospongiosus muscle (4); ischiocavernosus muscles (5); ure- MA is a polyhierarchy with several different axes of thra membranosa (6); bulbourethral gland ducts to the penile urethra organization. These include anatomic region organ system, (arrows). Mouse image modified from Ruberte et al. (2016) with per- organ, substance, and tissue. The grouping of individual mission organs and structures by function or physiology, for example the duodenal glands (formally called Brunner’s glands) is_a in order to support inference, reasoning, and estimates of exocrine gland and is part_of the small intestine. semantic similarity. Lists of anatomical terms alone, like TA The MA Browser enables one to navigate through the and the NAV, are inadequate for this formalization because ontology in two ways. Progressively, scrolling through the anatomical nomenclatures cannot easily put in the format various hierarchies and in the Term tab obtaining informa- appropriate for the structure of relational databases (Bard tion about individual terms, including ID and relationship 2005). to other terms. Alternatively, an anatomy search could be The solution adopted was to organize the anatomy of each done introducing any text sequence in the query field. Then, organism in a hierarchy of anatomical structures and linking all terms in the MA vocabulary, including any synonyms, relationships (e.g., the bulbourethral gland is_a male repro- contained in text will be displayed (Hayamizu et al. 2005a, ductive gland and is part_of male urethra). The adoption of 2015). The MA ontology can also be downloaded from the a formal ontology to describe the domain of anatomy has Open Biomedical Ontologies website (OBO; https://​sourc​ very important advantages over other approaches and allows eforge.​net/​proje​cts/​obo). The MA ontology is currently the use for formal semantic methods to make inference, sub- used to annotate: (1) developmental expression data in the sume related concepts under higher concepts (parents) and Gene Expression Database (GXD; http://​www.​infor​matics.​ to use automated reasoning. The advent of the Ontology jax.​org/​expre​ssion.​html) (Smith et al. 2015); (2) phenotype Web Language (OWL) for most bio-ontologies allows for data obtained in the International Mouse Phenotyping Con- much more precise and expressive modelling of the concept sortium (IMPC; https://w ​ ww.m ​ ousep​ henot​ ype.o​ rg) (Elmore domain, with restrictions, properties, instances and complex et al. 2018); and (3) pathology data in Pathbase (http://w ​ ww.​ relationships that allow the capture of a great deal of knowl- pathb​ase.​net) (Schofield et al. 2004). Post-natal mouse anat- edge. While an ontology is not intrinsically a graph, it may omy terms are also available as Theiler stages 27 and 28 be expressed as such, especially as humans are familiar with in the EMAPA ontology where developmental anatomy is 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 399 available in the same hierarchy as adult/postnatal (Hayamizu Carpal bones in mice are arranged in two rows (Fig. 8). et al. 2013). There are several aspects that could be considered regarding No scientific terminology can be considered complete and this hierarchy and terminology. First, the falciform carpal permanent as long as research in the field continues, and the bone is not a true carpal bone, in fact is a sesamoid bone MA ontology like others in the OBO Foundry continues to embedded in the flexor retinaculum (Wirtschafter and Tsu- be developed by its curators in response to new applications jimura 1961), therefore, it should not be considered to be a and the discovery of new knowledge. MA ontology devel- carpal bone. Furthermore, carpal bones in mouse present opers have continued to refine terminology, adding syno- many anatomic variations (Ruberte et al. 2021). In 15% of nyms, and revising anatomical concepts. However, due to the 28 C57BL/6J carpals analyzed a supernumerary bone, the increasing number of publications and books dealing with central carpal bone, was observed between the proximal and mouse anatomy published in recent years (Table 1) there is distal row. Additionally, this bone was fused to the scapholu- a need to continuously develop the ontology and to align nate (in 45% of cases) and to the trapezoid (40%), thus the mouse and human anatomy. central carpal bone should be considered, as happens in the We propose the development of a mouse version of the TA, as an actual carpal bone in mouse. It is a common prac- Terminologica Anatomica with high resolution detail, struc- tice to number the carpal bones from medial to lateral, first tured and consistent with the principles established in the TA in the proximal row and then in the distal row. This proce- and the NAV, the mTA. One example of how such a termi- dure was first adopted by Vesalius (Vesalius 1543) and after nology might deal with anatomical complexity and existing that all anatomy resources have followed this rule, including standards is for the carpal bones. TA and NAV. Fig. 8  Dorsal ventral radio- graphs of human hand (A) and mouse forepaw (B). Diagram showing the topography and organization of mouse carpal bones (C). Alizarin stained mouse carpus (palmar aspect) (D). Carpal bones (1); metacar- pal bones (2); phalanges (3); scapholunate (4); triquetral (5); pisiform (6); trapezium (7); trapezoid (8); capitate (9); hamate (10); falciform carpal bone (11), which is a sesamoid bone embedded in the flexor retinaculum and not a true car- pal bone; ulnar sesamoid bone (12). Roman numerals indicate the medial to lateral order of digits 13 400 J. Ruberte et al. As a corollary to the previous comments, it seems clear manual curation approaches approximately 830 matching that it would be useful to integrate or map terms in an mTA pairs were identified (Hayamizu et al. 2012). Afterwards, to mouse anatomical terms in the MA and EMAPA ontolo- the harmonization of both ontologies by addition of terms gies, using community effort. Table 2 shows the comparison represented only in one ontology, changes in the hierarchy, of the current terminologies with the proposed mTA. The and augmented synonymy, reached 1634 matching terms, great advantage of the Open Biomedical Ontology principles meaning that approximately half of the terms do not match is that they encourage community input to the development and are specific to one of the two ontologies (Hayamizu and continuous refinement of ontologies by the user com- et al. 2012). An overall comparison of automatically map- munity. Because the mouse, domestic animals, and humans pable terms (Bioportal; accessed 29.5.23) shows that while have a similar anatomical complexity, mTA probably should 74,000 of 105,000 FMA classes can be mapped lexically have a number of terms similar to those of the TA and NAV, and organisationally to MA, this includes a very large num- around 6000, considerably increasing granularity, although ber of one-to-many mappings (FMA to MA), contrasting attention needs to be paid to structuring classes in both ter- with the 1634 manually matched terms above. This is due minologies when considering these absolute numbers. More to the inclusion of concepts such as clustered anatomical importantly formal mapping or axiomatization of MA to structures, relations and subregions and the high degree of FMA could also be considered along with cross references granularity of the FMA. For example there are 15 terms in to classes in the human Terminologica Anatomica. We are FMA mapped onto one term, soleus, in the mouse. Only a confident that this development would greatly support mor- very limited number of the species-specific terms identified phological phenotyping of mouse models as modern imag- represented real anatomical differences between the two spe- ing methods increasingly need a higher degree of anatomical cies, and there were consequences of decisions made regard- granularity for annotation than is currently available. ing the views and resources used to build each ontology. The MA ontology was based on major sources and expertise, including mouse atlases as well as anatomy and histology Alignment of mouse and human anatomy text resources. Once the list of terms was generated, each ontologies and its application term was confirmed to represent a real mouse anatomical structure. However, MA ontology developers found that Alignment of mouse and human anatomy ontologies facili- this confirmation was sometimes ambiguous, since numer- tates the integration of mouse and human phenotype/gen- ous structures described in anatomy and histology textbooks otype data and promotes the translation of basic research do not have unambiguous evidence of their existence in the discoveries into clinical settings (Bodenreider et al. 2005). mouse, and then consequently were not included in the Such alignment is critical when integrating phenotype data ontology (Hayamizu et al. 2005a). from humans and mice in search of disease-causing genes There are additional complications in mappings from (Boudellioua et al. 2017; Hoehndorf et al. 2011, 2013, 2015, human to mouse, sometimes in the “views” of the topology, 2011; Mungall et al. 2010; Robinson and Webber 2014; such as where concepts have been created to reflect clusters Schofield et al. 2010). However, the modelling decisions for or features of more specific structures. In other cases, there the FMA were radically different from those in MA (Hay- are real debates about the way to describe structures and amizu et al. 2005a), reflecting that it was designed for differ- others where the topology is quite distinct across species. ent purposes, and the FMA contains a particularly rich set of For example, in the mouse the prostate has identifiable ante- relations, multiple axes of classification and concepts such rior, dorsal, lateral, and ventral lobes (Ruberte et al. 2017); as boundaries, landmarks, lines, voids and spatial relations, the dorsal and lateral being viewed in combination as the particularly relevant for human clinical imaging and surgical dorsolateral prostate. The human gland is much less easily applications, for example. Both are excellent at what they do, differentiated into lobes but in the NCI thesaurus anatomy but they were designed for different purposes. Some work the lateral, medial, and posterior lobes are defined, and an has been done with anatomy ontology matching with terms additional term used; “overlapping zones”. Originally the from UMLS (https://​www.​nlm.​nih.​gov/​resea​rch/​umls), a human prostate was believed to contain five lobes, but these widely used terminology that is not strictly an ontology, with are only recognizable during development where it origi- a degree of success allowing for cross-species searching for nates from five pairs of epithelial buds. The human prostate example but not complex computation. is unilobular but contains three identifiable zones: central, Under the Mouse–Human Anatomy Project (MHAP), the transitional and peripheral. The central zone surrounds the MA ontology and the human anatomical terms included in ejaculatory duct, the transitional zone surrounds the urethra, the National Cancer Institute (NCI) Thesaurus (Coronado and the peripheral zone makes up most of the prostate, lying et al. 2004) were compared. In the preliminary identification against the rectal wall on the dorsal surface of the gland of equivalent terms using a combination of automated and (Ittmann 2018). Given the profound structural differences 13 Table 2  Comparison between current anatomical terminologies/ontologies an the proposed mTA Terminol- Scope and Focus Structure and Annotation and Applications No of classes Ontology depth Axes of clas- Formal ontol- Knowledge Asserted refer- Authority ogy/Ontol- organization detail sification ogy (OWL) base/terminol- ences/imported ogy ogy classes UBERON Multi-species Built on the Aims to provide Widely used in 25,930 34 Complex hierar- Yes No Refers to or Uberon curation anatomy ontol- principles of a broad bioinformatics chy containing imports team ogy that aims the Open Bio- coverage of and compara- mixed biologi- more than 30 to provide medical Ontol- anatomical tive anatomy cal concepts anatomy, spe- a common ogies (OBO) entities across studies to and semantic cies specific, framework for Foundry, using species but integrate and types; not molecular and representing a hierarchical may not have analyze ana- all anatomi- functional anatomical structure of extensive tomical data cal Spatial, ontologies, structures, classes and detail at the across species taxonomic, e.g. GO, both adult and relationships level of indi- and to support developmental ChEBI, MA, developmen- to represent vidual species cross-species and systematic FMA etc. tal, across anatomical or develop- research relations Specifies diverse organ- entities, cells mental stages presence of isms, includ- tissues and classes in ing humans, features across taxon mice, and different spe- other model cies. Includes organisms many relation- ship types such as 'inner- vated_by", "part_of" etc MA Specifically Uses a hierarchi- Provides detailed Primarily used 3257 9 Anatomically Yes No MA local curation focuses on cal structure annotations in mouse systematic. team, The Jack- the anatomi- to organize specific to the research, such Region, organ son Laboratory cal structures anatomical anatomy of the as studies system, body of the adult structures adult mouse, involving fluid, sub- mouse, provid- within the including mouse models stance, organ ing detailed adult mouse, information on of human informa- including organ systems, diseases or Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… tion about regions, structures, and investigations the mouse organs, and their spatial into the mouse anatomy at the tissues. It pro- relationships anatomy for macroscopic vides specific experimental and micro- information design and scopic levels relevant to interpretation the mouse anatomy. "is_a" and "part_of" 13 401 Table 2  (continued) 402 Terminol- Scope and Focus Structure and Annotation and Applications No of classes Ontology depth Axes of clas- Formal ontol- Knowledge Asserted refer- Authority ogy/Ontol- organization detail sification ogy (OWL) base/terminol- ences/imported 13 ogy ogy classes FMA Focus on human Employs a com- Provide com- Extensively used 104,721 23 Anatomically Yes No FMA local cura- anatomy, plex structure prehensive in various systematic. tion team providing a based on a and detailed human clinical Physical and comprehensive foundational information fields, includ- non physical representa- model, utiliz- about human ing medical anatomi- tion of human ing a mix of anatomical education, cal entities, anatomical hierarchical structures, clinical including structures and relationships, includ- informatics, relations and their relation- partonomic ing spatial imaging and transformation ships relationships, relationships, anatomical entities and other hierarchical research, to axioms to rep- organization, provide a resent human and functional detailed and anatomical attributes standardized entities and representa- their attributes tion of human anatomy NAV NAV aims to NAV is NAV includes It provides a > 6500 N/A Hierarchical No Yes Maintained and establish a organized species- common lan- structure. updated by the standardized systematically, specific terms guage and set General divi- International nomenclature providing to address of terms for sions on major Committee on for anatomi- hierarchical anatomical accurately and anatomical Veterinary Gross cal structures relationships variations and consistently systems or Anatomical in veterinary between ana- differences describing regions of the Nomenclature medicine cov- tomical struc- among dif- anatomical body, specifc (ICVGAN) ering the main tures. Terms ferent animal structures regions or veterinary spe- are classified species. It in animals. structures cies. Covers based on recognizes that Used by within each a wide range their position, animals may veterinarians, general of animal spe- function, and have unique researchers, system. cies, including relationship anatomical and anatomists Hierarchical cat, dog, pig, with other features, and mainly in relationships cattle, sheep, structures. the terminol- the clinical captured goat, horse, This hierarchi- ogy provides domain. and species and rabbit cal organiza- species- Global reach specific termi- tion allows for specific terms and available nology clear identi- to accurately in Latin and fication and describe those English classification structures of anatomical entities J. Ruberte et al. Table 2  (continued) Terminol- Scope and Focus Structure and Annotation and Applications No of classes Ontology depth Axes of clas- Formal ontol- Knowledge Asserted refer- Authority ogy/Ontol- organization detail sification ogy (OWL) base/terminol- ences/imported ogy ogy classes mTA mTA aims to mTa will be mTA will It will provide ~ 6500 N/A It will be a No Yes It will be devel- establish a organised provide defini- a set of hierarchical oped and curated standardized systematically tions of mouse confirmed structure. by a expert nomenclature in several sec- anatomical anatomical General divi- panel of mouse for confirmed tions: general terms and structures for sions on major morphologists anatomical terms, regions its relation accurately and anatomical structures in of the body, to human consistently systems or laboratory osteology, anatomical describ- regions of the mice (Mus arthrology, concepts and ing mouse body, specific musculus) myology, structures morphology. regions or digestive appa- Definitions structures ratus, respira- of anatomical within each tory apparatus, terms will be organ thoracic cav- accompanied ity, urogenital by annotated apparatus, uri- drawing nary organs, images. It will male and be available female in Latin and genital organs, English and perineum, Latin peritoneum, endocrine glands, angiol- ogy, nervous system, and common integument. Anatomical terms are clas- Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… sified based on their posi- tion, shape, function and relationship with other structures 13 403 404 J. Ruberte et al. between the human and mouse prostates the question must (Alghamdi et al. 2019). This approach has been applied for also be raised as to whether a lobe-specific mapping between the high throughput pathology studies by the International the species, as is often attempted, is biologically justified. It IMPC (Elmore et al. 2018) and a large-scale study on the is, however, possible to map mouse and human at a broader pathology of aging laboratory mice (Sundberg et al. 2011) concept level and simply map the human to mouse prostate, using a dedicated data capture system, MODIS (Sundberg and that complete collection of lobes/zones from each spe- et al. 2008). cies. This is a useful mapping approach but only reflects a ground truth when made at this higher organizational level. The MA ontology and NCI thesaurus both contain broader Conclusions concepts prostate gland and prostate gland lobe, and these allow accurate mappings at a biologically meaningful level The ability to relate and map anatomical structures between of granularity. mice and humans is a critical part of using laboratory mice Breaking down the anatomical barriers between human for the discovery of disease genes and pathological mecha- and mouse anatomy, indeed between these and other meta- nisms. An accurate and granular formal terminology or zoan anatomies, is an important part of data integration, par- ontology is essential for these activities, and permits not only ticularly for functional genomics and, as mentioned above, accurate data capture, but also computational approaches discovery of human disease genes using model organism to cross-species phenotyping. Such terminologies need to phenotypes. Given the problems of anatomy ontology map- be grounded in biological truth and mappings made using ping a cross-species anatomy ontology was developed— knowledge-driven criteria. The problems mapping high Uberon. The problems with data integration across species resolution anatomical data from humans to mice and vice become more difficult with evolutionary distance and where versa are, however, impacted by a major difference in the the matching of human to model organism anatomies is completeness of the two species’ anatomical descriptions. desired to “de-silo” data, particularly phenotype data, map- It is currently not possible to represent mouse anatomy with pings can only be done at a low level of granularity. Such a the precision available to FMA using the MA, and indeed mapping of evolutionarily homologous structures and ana- the MA was not designed for the detailed representation tomical concepts was achieved with the Uberon ontology required to annotate and interpret high resolution imaging (Mungall et al. 2012; Haendel et al. 2014). While not as and modern anatomical techniques, or the small variations granular as MA, Uberon has become a de facto standard seen between species and genetic backgrounds. for expressing anatomical location in gene expression stud- While it is not the aim of this commentary to propose a ies, and also through equivalence axioms facilitating cross- new anatomy ontology for the adult mouse, we have dis- species phenotype similarity assessments using the major cussed how advances in imaging particularly have generated metazoan phenotype ontologies, for example MP (https://​ a requirement for more granular terminologies, the model www.​ebi.​ac.​uk/​ols/​ontol​ogies/​mp) and HPO (https://​hpo.​ for which is the TA or the NAV. Such knowledgebase level jax.​org) (Robinson and Webber 2014; Gkoutos et al. 2017). terminologies themselves have limited use computationally As well as the incompleteness issues, Uberon has complex and we would strongly support establishment of a close rela- axes of classification and contains mixed semantic types. tionship with MA curation to include new recommendations Detailed anatomical annotation of images, and high reso- in the existing, gold-standard ontology as appropriate. lution phenotype capture were not applications for which The study of human anatomy has a long history, mainly Uberon was designed. motivated by surgery and therapeutics. As we discuss above, The use of anatomy ontologies, and specifically the MA it is currently more detailed and complete than mouse ontology goes further than axiomatization and they have anatomy, so work to increase the accuracy and coverage proved extremely useful in collecting and coding data for of mouse anatomy will greatly assist in making accurate subsequent computational analysis. The collection of his- and meaningful mappings between the two species. The topathology data at scale is complicated by the inability to development of new imaging technologies, such as MRI, compute on existing terminologies and the limited scope of CT, SPECT, PET, two photon tomography (STP) (Ragan existing terminologies when attempting to cover the entire et al. 2012), light sheet microscopy (Stelzer et al. 2021) domain of pathology in the mouse. Such terminologies and experimental procedures such as CLARITY, 3DISCO would been many hundreds of thousands of concepts were and CUBIC (Ueda et al. 2020) have not only improved our they to pre-compose all the possible locations and types of knowledge of mouse anatomy but, in a sense, have turned lesions. This problem is solved by using an anatomy ontol- anatomical investigations inside out. Traditionally human ogy, in this case MA, in combination with a foundational anatomy proceeded from the outside-in. Contemporary pathology ontology (MPATH) (Schofield et al. 2013) to investigations can increasingly proceed from the inside-out post-compositionally create specific terms as required (Standring 2016) and we are entering a new phase in the 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 405 description and understanding of anatomy, where mouse References anatomical studies may inform those in humans rather than vice versa. Akabori E (1934) Septal apertures in the humerus on Japanese, Ainu Added to this is the increasing awareness of anatomi- and Koreans. Am J Phys Anthropol 18:395–400 Alghamdi SM, Sundberg BA, Sundberg JP, Schofield PN, Hoehndorf cally definable functional units which do not have morpho- R (2019) Quantitative evaluation of ontology design patterns for logically explicit physical boundaries and structures at a combining pathology and anatomy ontologies. Sci Rep 9:4025 macro-scale, yet can be demonstrated through high reso- Aspelund A, Antila S, Proulx ST (2015) A dural lymphatic vascular lution molecular, imaging and functional techniques, has system that drains brain interstitial fluid and macromolecules. J Exp Med 212:991–999 changed some of our concepts of what anatomy actually is, Bard JB (2005) Anatomics: the intersection of anatomy and bioinfor- for example the brain connectome and the identification of matics. J Anat 206:1–16 the angiome (D'Angelo and Jirsa 2022; Taylor and Palmer Bard JBL, Kaufman MH, Dubreuil C, Brune RM, Burger A, Baldock 1987). RA, Davidson DR (1998) An internet-accessible database of mouse developmental anatomy based on a systematic nomencla- Integration of cellular and tissue anatomy is a topic we ture. Mech Dev 74:111–120 have not addressed in this commentary, but multiscale inte- Barney S, Lewis W, Beach J, Berghof O (eds) (2006) The etymologies gration of anatomy is crucial to the generation of the virtual of Isidore of Seville. Cambridge University Press, Cambridge physiome (Hoekstra et al. 2018; Kokash and de Bono 2021) Barrett JM, RaineriTapies MG, Shepherd GMG (2020) Manual dexter- ity of mice during food-handling involves the thumb and a set of and much work is proceeding to make this computationally fast basic movements. PLoS One 15:e0226774 coherent in support of modelling physiological processes, Bodenreider O, Hayamizu TF, Ringwald M, De Coronado S, Zhang S which in turn allows for comparison of abnormal processes (2005) Of mice and men: aligning mouse and human anatomies. between species. AMIA Annu Symp Proc, pp 61–65 Boudellioua I, Mahamad Razali RB, Kulmanov M, Hashish Y, Bajic Finally, the need for training in these rapidly develop- VB, Goncalves-Serra E, Schoenmakers N, Gkoutos GV, Schofield ing fields will be important in implementing the advantages PN, Hoehndorf R (2017) Semantic prioritization of novel causa- gained from improved comparative anatomy between mice tive genomic variants. PLoS Comput Biol 13:e1005500 and humans, and training courses aimed at a new generation Brown SDM (2021) Advances in mouse genetics for the study of human disease. Hum Mol Genet 30:R274–R284 of investigators are under development in the community Butterfield LB (1972) The fine structure of Bowman’s capsule in male (Ruberte et al. 2020). C3H/HeJ mice. Lab Anim Sci 5:652–657 Churchill GA, Gatti DM, Munger SC, Svenson KL (2012) The Diver- Acknowledgements This work was supported by a grant from the sity Outbred mouse population. Mamm Genome 23:713–718 Erasmus Mundus of the Education, Audiovisual and Culture Execu- Collaborative Cross C (2012) The genome architecture of the col- tive Agency (EACEA) of the European Union (Reference Number: laborative cross mouse genetic reference population. Genetics 101082155). The name of the project is the Erasmus Mundus Joint 190:389–401 Master degree in “Human Diseases Models Morphological Phenotyp- Constantinescu GM (2011) Comparative anatomy of the mouse and the ing” (MorphoPHEN). The authors thank Prof. Martin Ringwald and Dr. rat: a color atlas and text. American Association for Laboratory Jerrold M. Ward for helpful comments on the manuscript. Animal Science, Memphis Cook MJ (1965) The anatomy of the laboratory mouse. Academic Author contributions J Ruberte, P Schofield, J Sundberg, and C MacK- Press, New York erlie wrote the main manuscript text. A Carretero and A Rodriguez Cook MJ (1983) Anatomy, chapter 7. In: Foster HL, Small JD, Fox JG prepared anatomical preparations and figures. All authors reviewed (eds) The mouse in biomedical research. Academic Press, New the manuscript. York, pp 101–120 Coronado S, Haber MW, Sioutos N, Tuttle M, Wright LW (2004) NCI Funding Open Access Funding provided by Universitat Autonoma de Thesaurus: using science based terminology to integrate cancer Barcelona. research results. Medinfo 2004:33–37 D’Angelo E, Jirsa V (2022) The quest for multiscale brain modeling. Declarations Trends Neurosci 45:777–790 De Souza P, Viotto-Souza W, Pereira V, Coelho F, Leite B, Abidu- Competing interests The authors declare no competing interests. Figueiredo M, Quagliatto A (2020) Clavicle in carnivorans: a forgotten bone. Anat Rec 303:1831–1841 de Wilde V, Maeseneer M, Lenchik L, Roy P, Beeckman P, Osteaux Open Access This article is licensed under a Creative Commons Attri- M (2004) Normal osseous variants presenting as cystic or lucent bution 4.0 International License, which permits use, sharing, adapta- areas on radiography and CT imaging: a pictorial overview. Eur tion, distribution and reproduction in any medium or format, as long J Radiol 51:77–84 as you give appropriate credit to the original author(s) and the source, Del-Valle-Anton L, Borrell V (2022) Folding brains: from development provide a link to the Creative Commons licence, and indicate if changes to disease modelling. Physiol Rev 2:511–550 were made. The images or other third party material in this article are Eide PK, Vatnehol SAS, Ke E (2018) Magnetic resonance imaging included in the article's Creative Commons licence, unless indicated provides evidence of glymphatic drainage from human brain to otherwise in a credit line to the material. If material is not included in cervical lymph nodes. Sci Rep 8:7194 the article's Creative Commons licence and your intended use is not Elmore SA, Cardiff R, Cesta MF, Gkoutos GV, Hoehndorf R, Keenan permitted by statutory regulation or exceeds the permitted use, you will CM, McKerlie C, Schofield PN, Sundberg JP, Ward JM (2018) A need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. 13 406 J. Ruberte et al. review of current standards and the evolution of histopathology Kokash N, de Bono B (2021) Knowledge representation for multi-scale nomenclature for laboratory animals. ILAR J 59:29–39 physiology route modeling. Front Neuroinform 15:560050 F.I.P.A.T (2019) Federative International Programme for Anatomi- Leader RW, Padgett GA (1980) The genesis and validation of animal cal Nomenclature (2019) Terminologia Anatomica, 2nd edn; models. Am J Pathol 101:S11-16 approved by the Federation of Association of Anatomist in 2020 Louveau A, Smirnov I, Keyes T (2015) Structural and functional Gkoutos GV, Schofield PN, Hoehndorf R (2017) The anatomy of phe- features of central nervous system lymphatic vessels. Nature notype ontologies: principles properties and applications. Brief 523:337–341 Bioinform 19(5):1008–1021. https://​doi.​org/​10.​1093/​bib/​bbx035 Louveau A, Plog BA, Antila S, Alitalo K, Nedergaard M, Kipnis J Grüneboom A, Hawwari I, Weidner D (2019) A network of trans-cor- (2017) Understanding the functions and relationships of the tical capillaries as mainstay for blood circulation in long bones. glymphatic system and meningeal lymphatics. J Clin Investig Nat Metab 1:236–250 127:3210–3219 Haendel MA, Balhoff JP, Bastian FB, Blackburn DC, Blake JA, Brad- Maronpot RR, Boorman GA, Gaul BW (1999) Pathology of the mouse. ford Y, Comte A, Dahdul WM, Dececchi TA, Druzinsky RE, Reference and atlas. Cache River Press, Vienna Hayamizu TF, Ibrahim N, Lewis SE, Mabee PM, Niknejad A, Matter SP (2013) The prince of medicine; Galen in the Roman Empire. Robinson-Rechavi M, Sereno PC, Mungall CJ (2014) Unification Oxford University Press, Oxford of multi-species vertebrate anatomy ontologies for comparative Mays S (2008) Septal aperture of the humerus in a mediaeval human biology in Uberon. J Biomed Semantics 5:21 skeletal population. Am J Phys Anthropol 136:432–440 Hayamizu TF, Mangan M, Corradi JP, Kadin JA, Ringwald M (2005a) McGonigle P, Ruggeri B (2014) Animal models of human dis- The Adult Mouse Anatomical Dictionary: a tool for annotating ease: challenges in enabling translation. Biochem Pharmacol and integrating data. Genome Biol 6:R29 87:162–171 Hayamizu TF, Mangan M, Jp C (2005b) The adult mouse anatomical Meckel JH (1825) A note on the septal apertures in the humerus of dictionary: a tool for annotating and integrate data. Genom Biol Central Indians. Eastern Anthropol 33:270–284 6:R29 Mota B, Herculano-Houzel S (2015) Cortical folding scales universally Hayamizu TF, Coronado S, Fragoso G (2012) The mouse–human with surface area and thickness, not number of neurons. Science anatomy ontology mapping project. Database article ID bar066 349:74–77 Hayamizu TF, Wicks MN, Davidson DR, Burger A, Ringwald M, Bal- Mungall CJ, Gkoutos GV, Smith CL, Haendel MA, Lewis SE, Ash- dock RA (2013) EMAP/EMAPA ontology of mouse developmen- burner M (2010) Integrating phenotype ontologies across multiple tal anatomy: 2013 update. J Biomed Semant 4:15 species. Genome Biol 11:R2 Hayamizu TF, Baldock RA, Ringwald M (2015) Mouse anatomy Mungall CJ, Torniai C, Gkoutos GV, Lewis SE, Haendel MA (2012) ontologies: enhancements and tools for exploring and integrating Uberon, an integrative multi-species anatomy ontology. Genome biomedical data. Mamm Genome 26:422–430 Biol 13:R5 Hedrich HJ (ed) (2012) The Laboratory Mouse. Academic Press, Natale G, Bocci G, Ribatti D (2017) Scholars and scientists in the his- Boston tory of the lymphatic system. J Anat 231:417–429 Hedrich HJ, Bullock G (2004) The laboratory mouse. Elsevier, O’Rahilly R (1989) Anatomical terminology, then and now. Acta Anat Amsterdam 134:291–300 His W (1885) Die anatomische Nomenclatur. Nomina anatomica, Popesko P, Rajtova V, Horak J (1992) A Color Atlas of Anatomy of Verzeichniss der von der anatomischen Gesellschaft auf ihrer IX. Small Laboratory Animals. Wolfe Publishing Ltd, London Versammlung in Basel angenommenen Namen. Metzger, Leipzig Ragan T, Kadiri LR, Venkataraju KU, Bahlmann K, Sutin J, Taranda Hoehndorf R, Schofield PN, Gkoutos GV (2011) PhenomeNET: a J, Arganda-Carreras I, Kim Y, Seung HS, Osten P (2012) Serial whole-phenome approach to disease gene discovery. Nucleic two-photon tomography for automated ex vivo mouse brain imag- Acids Res 39:e119 ing. Nat Methods 9:255–258 Hoehndorf R, Schofield PN, Gkoutos GV (2013) An integrative, trans- Robinson PN, Webber C (2014) Phenotype ontologies and cross-spe- lational approach to understanding rare and orphan genetically cies analysis for translational research. PLoS Genet 10:e1004268 based diseases. Interface Focus 3:20120055 Rockwood C, Matsen F, Wirth M, Lippit S (2009) The shoulder. Saun- Hoehndorf R, Schofield PN, Gkoutos GV (2015) The role of ontologies ders/Elsevier, Philadelphia in biological and biomedical research: a functional perspective. Rosse C, Mejino JLV (2008) The foundational model of anatomy ontol- Brief Bioinform 16:1069–1080 ogy. Comput Biol 6 Hoekstra AG, van Bavel ET, Siebes M, Gijsen FJH, Geris L (2018) Ruberte J, Carretero A, Navarro M (2016 and 2017) Morphological Virtual physiological human 2016: translating the virtual physi- mouse phenotyping. Editorial Medica Panamericana, Buenos ological human to the clinic. Interface Focus 8 Aires and Academic Press, Cambridge I.C.V.G.A.N. (2017) Nomina Anatomica Veterinaria (2017) World Ruberte J, Schofield PN, Brakebusch C, Vogel P, Herault Y, Gracia G, Association of Veterinary Anatomists (WAVA. I.C.o.V.G.A. McKerlie C, Hrabe de Angelis M, Hagn M, Sundberg JP (2020) Nomenclature) PATHBIO: an international training program for precision mouse Ittmann M (2018) Anatomy and histology of the human and murine phenotyping. Mamm Genome 31:49–53 prostate. Cold Spring Harb Perspect Med 8:a030346 Ruberte J, Carretero A, Cater H, Gracia G, Lally C (2021) X ray anno- Johnson J, DelGiudice B, Bangari D, Peterson E, Ulinski G, Ryan S, tation mouse atlas. Doctor Herriot S.L., Zaragoza Thurberg B (2019) The laboratory mouse: a guide to the location Schofield PN, Bard JB, Booth C, Boniver J, Covelli V, Delvenne P, and orientation of tissues for optimal histological evaluation. CRC Ellender M, Engstrom W, Goessner W, Gruenberger M, Hoefler Press, Boca Raton H, Hopewell J, Mancuso M, Mothersill C, Potten CS, Quintanilla- Kachlik D, Baca V, Bozdechova I (2008) Anatomical terminology Fend L, Rozell B, Sariola H, Sundberg JP, Ward A (2004) Path- and nomenclature: past, present and highlights. Surg Radiol Anat base: a database of mutant mouse pathology. Nucleic Acids Res 30:459–466 32:D512–D515 Kiupel M, Brown KS, Sundberg JP (2000) Bulbourethral (Cowper’s) Schofield PN, Gkoutos GV, Gruenberger M, Sundberg JP, Hancock gland abnormalities in inbred laboratory mice. J Exp Anim Sci JM (2010) Phenotype ontologies for mouse and man: bridging 40:178–188 the semantic gap. Dis Model Mech 3:281–289 13 Bridging mouse and human anatomies; a knowledge‑based approach to comparative anatomy for… 407 Schofield PN, Sundberg JP, Hoehndorf R, Gkoutos GV (2011) New Sundberg J, Berndt A, Sundberg B, Silva KA, Kennedy V, Bronson R, approaches to the representation and analysis of phenotype knowl- Yuan R, Paigen B, Harrison D, Schofield PN (2011) The mouse edge in human diseases and their animal models. Brief Funct as a model for understanding chronic diseases of aging: the his- Genom 10:258–265 topathologic basis of aging in inbred mice. Pathobiol Aging Age- Schofield PN, Sundberg JP, Sundberg BA, McKerlie C, Gkoutos GV Relat Dis 1:71719 (2013) The mouse pathology ontology, MPATH; structure and Taylor GI, Palmer JH (1987) The vascular territories (angiosomes) of applications. J Biomed Semant 4:18 the body: experimental study and clinical applications. Br J Plast Scudamore CL (2014) A practical guide to the histology of the mouse. Surg 40:113–141 Wiley, London Treuting P, Dintzis S, Montine KS (2018) Comparative anatomy and Shivaleela C, Afroze KH, Lakshmiprabha S (2016) An osteological histology. A mouse, rat, and human atlas. Academic Press, San study of supratrochlear foramen of humerus of south Indian popu- Diego lation with reference to anatomical and clinical implications. Anat Tubs R, Shoja MM, M L, (2016) Bergman’s comprehensive encyclo- Cell Biol 49:249–253 pedia of human anatomic variation. Wiley Blackwell, Hoboken Smith CM, Finger JH, Tf H (2015) GXD—a community resource of Ueda HR, Ertürk A, Chung K, Gradinaru V, Chédotal A, Tomancak P, mouse Gene Expression Data. Mamm Genome 26:314–324 Keller PJ (2020) Tissue clearing and its applications in neurosci- Snell GD (1941) Biology of the laboratory mouse, 1st edn. McGraw- ence. Nat Rev Neurosci 21:61–79 Hill, New York Vandamme TF (2014) Use of rodents as models of human diseases. J Standring S (2016) A brief history of topographical anatomy. J Anat Pharm Bioallied Sci 1:2–9 229:32–62 Vesalius A (1543) De humani corporis fabrica libri septem. Joannis Stelzer EHK, Strobl F, Chang B-J, Preusser F, Preibisch S, McDole Oporini, Basle K, Fiolka R (2021) Light sheet fluorescence microscopy. Nat Rev Wirtschafter ZT, Tsujimura JK (1961) The sesamoid bones in the C3H Methods Primers 1:73 mouse. Anat Rec 139:399–408 Sundberg JP, Schofield PN (2009) One medicine, one pathology, and the one health concept. J Am Vet Med Assoc 234:1530–1531 Publisher's Note Springer Nature remains neutral with regard to Sundberg JP, Sundberg BA, Schofield PN (2008) Integrating mouse jurisdictional claims in published maps and institutional affiliations. anatomy and pathology ontologies into a diagnostic/phenotyping database: tools for record keeping and teaching. Mamm Genome 19:413–419 13