Vespers and vampires: A lifelong microscopic search for the smallest of things

Histology, the study of tissues of the body, was greatly hindered prior to the middle of the 17th century. The communication of venous and arterial blood flow for instance, remained the subject of speculation, because vessels other than veins, arteries, venules, and arterioles are typically too small to see (Hwa & Aird, 2007). Did blood pool into organs and reside as in a sponge, arriving and departing by the blood vessels one could see with the naked eye, dissect, and hold with a forceps? Some biologists suspected that smaller connecting networks existed. The Italian anatomist Marcello Malpighi (1628–1694) was able to glimpse capillary networks in skin of a frog, an animal that he referred to as a “microscope of nature,” with the aid of only a magnifying lens (West, 2013). His definitive descriptions were accomplished later, with a compound microscope (reviewed by Hwa & Aird, 2007). The microscope resolved microanatomical mysteries, and centuries later, we continue to glimpse ever more details at the tissue and cellular levels. Big thinkers and those with a steadfast focus have been required to make strides in our understanding of histology. This commentary is a review of advances in histology and cytology, or microanatomy, during the last century, and a celebration of Professor Kunwar Bhatnagar (1934–2021, Prof. Emeritus, University of Louisville), who explored many of these methods in his anatomical studies, often to report findings in the pages of The Anatomical Record (Figure 1). Kunwar Prasad Bhatnagar was born on March 21, 1934, in Gwalior, Madhya Pradesh, India. He recalled living in a “castle-like” home with his parents and siblings (Bhatnagar, unpublished autobiography). He was named in honor of the Kunwar Baba Temple (Figure 1a), which his parents attended (Prasad translates as “offering”). His father, Narain Swaroop Bhatnagar, was a statistician. He fondly recalled day trips with his father, who was an avid walker. He credits his mother, Bhagawati Devi Bhatnagar, for helping him chart his career path, by asking an uncle (a professor of physics) to arrange his entry into biological studies. His early education was in India (B.S. at Agra University, 1956; M.S. in Zoology at Vikram University, 1958). He frequently mentioned his teacher Dr. Shyama Charan Srivastava at Agra University. Science was of lifelong importance to Prof. Bhatnagar, who wrote that in marriage he hoped to find a “girl of scientific parentage” (Bhatnagar, unpublished). A year after his father's passing, Kunwar came to Buffalo, NY, to begin his graduate education in the State University at Buffalo, under the mentorship of Prof. Frank Kallen. This career-building path was blessed and supported by his father-in-law, Brahma Shankar Varma, a chemist. A year after his arrival, Kunwar's wife Indu and his daughters, Divya and Jyoti, joined him in Buffalo (Bhatnagar, unpublished). During his lifetime, microscopy charted a path of ever-increasing resolution of more minute detail, and during his long career, he avidly used these instruments to solve more mysteries. Light (LM) and electron microscopy (EM) are twin tools, with differing orders of resolution, allowing us to discern tissue and cellular level detail of the body. LM offers far less cellular detail, but more context than EM. Epithelial tissues were a frequent subject of Prof. Bhatnagar's great interest, and offer a clear illustration of the differing magnitudes of observation possible with LM and EM (Figure 2). LM provides ample information for characterizing tissue organization but allows relatively little cellular detail. Both LM and EM harness light and electrons that are transmitted through tissue sections to assess cross-sectional microanatomy, or reflected off of whole or dissected specimens to assess 3D structure at a fine scale. LM is a centuries-old technique, whereas EM emerged less than a century ago (Table 1; Mayall, 1885; Hawkes, 1985; Von Arden, 1985). Prof. Bhatnagar was born near the time of the birth of EM (Table 1), and came to often employ LM and EM in his studies. He recalls using the compound microscope of his uncle, a chemist, in his earliest studies. He even borrowed this microscope in an attempt to build his own, but unable to replicate the lens, had to wait (Bhatnagar, unpublished). His initial research in the US bore on detailed anatomical, histological, and behavioral traits of two bats as relates to olfaction (Bhatnagar, 1972; Bhatnagar & Kallen, 1974a) and subsequently he surveyed the olfactory bulb size in 40 species of bats (Bhatnagar, 1972; Bhatnagar & Kallen, 1974b). Broadly, he saw evidence that dietary preference influenced olfactory anatomy, a finding borne out by many subsequent anatomical and genetic studies (e.g., Barton et al., 1995; Yohe et al., 2021). In his landmark paper “Cribriform plate of ethmoid, olfactory bulb and olfactory acuity in forty species of bats,” he was the first to quantify osteological features of ethmoid bone as a proxy for olfactory abilities (Bhatnagar & Kallen, 1974a). Even though associations of osteology with special senses had long been known descriptively, he was one of the pioneers in a new age of ecological morphology. As pertains to olfaction, this inspired a long term trend in efforts to understand the ontogeny, ecology, and evolution of sensory systems in mammals (e.g., Bird et al., 2014, 2018, 2020; Pang et al., 2016; Pihlström et al., 2005; Smith et al., 2007; Van Valkenburgh et al., 2014; Yee et al., 2016). Subsequent to his dissertation research, his use of microscopy accelerated. Most of his histological specimens were prepared for study in waves during the 1970s, often with the aid of Indu Bhatnagar. In a 1980 publication on the chiropteran vomeronasal organ, Prof. Bhatnagar wrote: “It is necessary to emphasize here the importance of examining complete and uninterrupted series of sections” (Bhatnagar, 1980, p. 291). This is particularly true on minute structures such as the vomeronasal organ, which is 1.6–3.9 mm long in bats that possess a functional organ, and even smaller in bats that possess non-functional vestiges of the organ (Bhatnagar, 1980; Cooper & Bhatnagar, 1976). This carefully prepared collection, sectioned serially at 10 μm, is well suited for identifying the smallest patches of tissues. The vomeronasal organ is an enigmatic structure, exactly the sort of topic that captivated his interest throughout his career (another is the pineal gland—Bhatnagar, 1992). Function of the vomeronasal organ is understood through experimental work demonstrating that sociosexual behavioral deficits arise if its innervation is ablated (Aujard, 1997; Powers & Winans, 1975). However, not all mammals possess this sensory organ; in some it is absent or only a vestige remains (Cooper & Bhatnagar, 1976; Smith et al., 2001; Wysocki & Preti, 2004). Using LM and EM, Prof. Bhatnagar sought to identify functional traits of the organ in chiropterans (Figure 2), a great challenge since most bats appear to lack a functioning vomeronasal system (see below). Macroscopic observations provide scant information on this organ, and he sought finer and finer details using microscopy. While gross observation allows many insights, the ever-improving means of microscopic investigation enabled new physiological insights. Professor Bhatnagar, in collaboration with colleagues at the Institut fur Anatomie, Universitätsklinikum (Essen, Germany) used the freeze-etching technique (via transmission EM, or TEM) to describe the membrane features of receptor cell microvilli of the vomeronasal organ. Their studies (Breipohl et al., 1979; Miragall et al., 1979) were the first to identify intramembranous particles on vomeronasal sensory neurons, mostly reflecting transmembrane proteins (such as receptors and ion channels). The presence of such proteins was first described only a few years before on mammalian olfactory cilia (Menco, 1997). The field of microscopy became ever more precise during Prof. Bhatnagar's lifetime, zeroing in toward functional attributes at the levels of cellular membranes; in this case seeking the site at which odorants, airborne chemical signals, contact receptor cell mucous membranes where they interact with receptors. Subsequent to the completion of his dissertation, Prof. Bhatnagar continued to enlarge his collection of preserved bats, and even kept live bats captive for study, as he began his faculty position in 1972 in the Department of Anatomy and Neurobiology, University of Louisville, Louisville, Kentucky. He traveled to Mexico to collect vampire bats (Desmodus rotundus; Figure 3) and numerous other species of New World bats. At multiple sites in India, including Gwalior, Guna, and Mandu, Madhya Pradesh he collected Old World bat species (Cooper & Bhatnagar, 1976). In these locations, he described cavernous forts and castles where the bats lived, in some cases over the course of centuries (Bhatnagar, unpublished). His fascination with South American species grew as he realized that one speciose family, Phyllostomidae, is unique as the only bat family in which all species possess a vomeronasal organ (Bhatnagar & Meisami, 1998). This chemosensory system connects to the accessory olfactory bulb, distinct from the synaptic connections for the main olfactory system (Meisami & Bhatnagar, 1998). His collection of phyllostomids grew from many collaborative relationships with other bat biologists, and especially after the expedition to Mexico, after a spontaneous invitation from Professor William Lopez-Forment. With then graduate students Greg Cooper, Leon Kundrotas, and Barry Spoonamore, they collected bats from attics, from caves, and from rock quarries. The ceilings of a huge cave in Vera Cruz reached 30 or 40 ft (Cooper, personal communication). They had to screw aluminum extensions to the nets to reach concavities from which the bats were exiting their roosts. Numerous phyllostomids were collected on this expedition, such as Desmodus rotundus (common vampire bat); Daiemus youngi (white-winged vampire bat), Carrolia perspicillata (Seba's short-tailed bat), and Artibeus jamaicensis (Jamaican fruit bat). Prof. Bhatnagar's studies on chiropterans have established that few bats aside from phyllostomids possess a vomeronasal organ with functional traits (e.g., neuroepithelium is present) (Figure 2). Miniopterines and pteronotids (even though they are insectivorous) also possess such an organ. The latter findings are curious and as yet unexplained since Miniopterus is but the only genus in a very large family of bats (Vespertilionidae; a.k.a., evening or vesper bats), most of which lack the organ. The genus Pteronotus possesses a functional vomeronasal organ, but bats of the closely related genus Mormoops possess only a vestige of the organ (Bhatnagar et al., 2006). Prof. Bhatnagar emphasized the frugi-nectari-sangui-pollenivorous nature of the phyllostomid diet, and left open possible a dietary role for their vomeronasal system. Most other workers explored only sociosexual roles for this system in mammals, although some dietary roles have been explored (e.g., Halpern et al., 2005), and a role for the vomeronasal system in prey detection in some reptiles and amphibians is well known (e.g., Halpern et al., 1997; Placyk Jr & Graves, 2002). Over the course of five decades, the variation in the chiropteran vomeronasal organ that was uncovered by Prof. Bhatnagar and colleagues exposed a parallel to primates, which likewise exhibit extremes variation of the vomeronasal system (Bhatnagar & Meisami, 1998; Bhatnagar & Smith, 2006). These orders of mammals in particular offer a unique opportunity to examine presence or absence of the vomeronasal organ in the context of reproductive and sensory biology (e.g., Garrett, 2015; Yohe et al., 2018). Subsequent to his initial surveys of bats, Prof. Bhatnagar conducted several in-depth studies on individual genera (Bhatnagar et al., 2001, 2006; Bhatnagar & Smith, 2007). Professor Bhatnagar's collection is currently curated by the author, housed at his institution. The collection is relatively well-studied regarding the nasal region, but also includes vast microanatomical information on other cranial structures that are available to future generations of researchers. Histologists have long sought to reconstruct tissues, organs, or whole structures into three-dimensional form, often using “stacked” physical models which are then rendered (e.g., Maier, 1980). Histologists also learn to “see through” distortions, or artifacts, created by histological processing, such as shrinkage introduced by dehydration that precedes embedding, or folding of tissues on the glass slides. New advanced in contrast-enhanced computed tomography have made the process of reconstruction faster with relatively less artefactual distortions than traditional histological sectioning (Table 1). In particular, diffusible iodine-based contrast-enhanced computed tomography (diceCT) allows visualization of soft-tissue structures of whole cadavers of small mammals (Gignac & Kley, 2014). In a recent study on a broad spectrum of bats (Smith, DeLeon, et al., 2021), the distribution of vessels along the nasal airways was mapped out using diceCT. The technique also allows visualization of whole heads, and within the space of the head, visualization of internal structures (Figure 4). Our team, which included Prof. Bhatnagar, has even been able to infer mucosa type using diceCT using his invaluable bat collection (Smith, Corbin et al., 2021). Microscopy is similarly finding ever-increasing means of gleaning fine details and placing them in broader context. Similar to the wide field of view (at least, from the perspective of microscopy), a new system of laser scanning confocal microscopy offers a similar means to analyze microscopic aspects of whole embryos (McConnell et al., 2016). These advances in radiographic and microscopic methods promise to provide ever broader context for microscopic anatomy. In his search for tiny chemosensory organs of the nose, Prof. Bhatnagar advocated for an exceedingly careful approach, involving the scrutiny of “uninterrupted” sectional series. Although the length of the vestigial vomeronasal organs varies extensively, they are as small as 270 or even 120 μm (Tadarida mexicana and Balantiopterix io, respectively) (Bhatnagar, 1980); thus, these rudiments of the organ could hide within “very few sections” in bats (Bhatnagar, 1980, p. 291). I was struck by these words while, in parallel, I searched for the same organ in primates, which have a similar degree of variation (Bhatnagar & Smith, 2006; Smith et al., 2001). It was fascination with this parallel that led me to invite Professor Bhatnagar to contribute a review for a special is of Microscopy Research and Technique, bearing on the vertebrate vomeronasal organ (VNO). The electronic correspondence is lost to me now, but I recall he initially expressed trepidation about the timeline. Yet he clearly saw this review as timely, and I was greatly relieved he was able to contribute a manuscript detailing the extreme variations in the VNO of bats and primates (Bhatnagar & Meisami, 1998). The human vomeronasal organ is another enigma that attracted his attention. The structure has been the subject of endless fascination and speculation (for appropriately thoughtful appraisals of the likelihood of functionality in humans, see Meredith, 2001, and Wysocki & Preti, 2004). Anatomical studies have hardly resolved confusion on the topic, since collectively they report the incidence of a vomeronasal organ in adult humans is anywhere between 6 and 100% (Bhatnagar & Smith, 2001; Garcia-Velasco & Mondragon, 1991; Moran et al., 1991; Stensaas et al., 1991; Trotier et al., 2000; Won et al., 2000; Zbar et al., 2000). Bhatnagar and Smith (2001) asserted that the basis for these varied estimates is rooted in methodological deficiencies. Prior studies that used macroscopic observation (e.g., endoscopy) provided much more varied percentages for presence of the organ and higher incidence of absence. There is an irony to the use of macroscopic methods, considering that the vomeronasal organ is but an epithelial tube. Macroscopic studies had been blindly asserting an identity to a lumen, which may be a duct for any number of septal glands (Bhatnagar & Smith, 2001). On the other hand, studies that employed histological methods in which large blocks of septal tissue were sectioned had better success locating the organ. Trotier et al. (2000) found the organ in 9 of 15 sectioned samples. However, most samples were extracted from the “anterior part” of the nasal septum; it is unclear whether a sufficient amount of the septum, anteroposteriorly or superoinferiorly, was sectioned to locate a small epithelial tube in all cases. Earlier, Johnson et al. (1985) found the VNO in 70% of adults examine histologically, but again, the authors might have missed pertinent tissue. Bhatnagar and Smith (2001) studied 20 adult human septa via serial sectioning. In addition, they endeavored to include all relevant tissue, even a small portion of the hard palate, in the tissue blocks. In doing so, they preserved the complete context for septal microanatomy. Using positional criteria, they found epithelial tubes are located superior to bilateral paraseptal cartilages in all specimens. This position precisely matches the location of the structure in late fetuses (Kölliker, 1877). Bhatnagar and Smith also found vomeronasal nerves and sensory neurons to be lacking, as observed previously in adults (Trotier et al., 2000), and also documented in late human fetuses (Ortmann, 1989). The human septum, sectioned serially, spanned approximately 1,500 sections using the methods of Bhatnagar and Smith (2001; 20–25 μm thick sections). The smallest vomeronasal organ measured 1.3 mm in length, occupying perhaps 65 sections. A more randomized search, therefore, may easily miss it, exactly as cautioned by Bhatnagar (1980) based on bats. The topic of the human vomeronasal organ also fascinated Prof. Bhatnagar in a historical sense. The organ was discovered prior to the widespread availability of microscopy. The first person to describe it grossly in mammals was Ludwig Jacobson (hence the sometimes used eponym, Jacobson's organ) in 1811 (Bhatnagar & Reid, 1996). The person most commonly credited with the discovery of the human vomeronasal organ was Frederick Ruysch, in 1703. And yet, Ruysch only made gross observations of an opening, or “pit” in the septum, into which he inserted a probe. Retrospectively, the position of the opening is consistent with that of the human vomeronasal organ, but histological confirmation was not possible. Thus, Bhatnagar and Smith (2003) credited Rudolf Albert Von Kölliker (1877) with identifying the human VNO, since he provided the first microscopic identification. A friend once commented to me that a histology laboratory is often quite like a cooking show, where boxes of ingredients, a bowl of batter, a cake in the oven, and one in the refrigerator, are all available at the same time. So it was in Prof. Bhatnagar's lab. His histology collection of bats spans 15 families, including over 40 species. His collection contains serially sectioned specimens, some of which remain unstained; some remain decalcified in paraffin blocks, and some remain in jars, awaiting study. Among the specimens remaining unsectioned, one fetal Artibeus jaimaciensis remained in paraffin until about the time of this writing. Consulting notes provided to me by Prof. Bhatnagar, I learned that this specimen was “still born on Jan. 28, 1971. [He received the bats live, while working on his dissertation.] The whole fetus was dropped in Bouin's. The block contains the whole head. In order to get the correct orientation, you may have to re-embed thereby noting the positioning of the head.” Holding this paraffin block to a bright light, I could see the orientation perfectly; it was completely fine for frontal sections. The entire snout was serially sectioned and stained with his favored procedure, Gomori Trichrome. One adult Artibeus was mostly unstained, and selected sections were stained for comparison to the term fetus (Figure 5). From birth to adult, one may infer proliferation of glandular tissue and enlargement of venous sinuses in support of the vomeronasal organ (Figure 5a,b). Already at birth, the vomeronasal neuroepithelium comprises the majority of cross-sectional area of the organ (Figure 5c). As in the adult, the non-sensory part (or receptor-free epithelium—Breipohl et al., 1979) is restricted to a small lateral patch (Figure 5c,d). Thus, this phyllostomid might be interpreted to have precocious development of the vomeronasal sense, although this comparison also suggests much postnatal growth awaits. In the stillborn specimen, the right vomeronasal neuropithelium measures in 0.7 mm in length and 0.01 mm3 in volume; in the adult, right vomeronasal neuropithelium measures in 1.2 mm in length and 0.037 mm3 in volume. Kunwar surely would have enjoyed seeing the excellent preservation of these specimens revealed after staining. Over the course of five decades, this collection was studied many times by Prof. Bhatnagar and visiting scholars (Bhatnagar, 1980; Bhatnagar et al., 1996, 2006; Bhatnagar & Kallen, 1974a, 1975; Bhatnagar & Meisami, 1998; Bhatnagar & Smith, 2007; Cooper & Bhatnagar, 1976; Eiting et al., 2014; Smith et al., 2012; Smith, DeLeon et al., 2021; Wible & Bhatnagar, 1996), and it remains one of his many great gifts to science. Some days we live, and sometimes we write, with a heavy heart. Writing this remembrance of Kunwar was bittersweet; throughout this writing, fond stories were shared with Kunwar's many colleagues, and new friendships were made. There were some surprises along the way. I had known for some time that Kunwar finished his dissertation in my hometown, at a time when I was too young to understand microscopy, and might or might not have seen a bat up close. But I now know that his daughters attended an elementary school a mere 11- minute drive away from the school my sister and I were attending (ourselves, the same ages as his daughters). Kunwar and my father may have passed one another in downtown Buffalo, and exchanged hellos. Many years later, it was a great privilege to know Kunwar and Indu and meet their family. My special thanks to Divya Cantor (Kunwar's daughter), Deborah Bird, Greg Cooper, and Bert Menco for sharing memories. The images in Figure 1 were sent to me by Indu Bhatnagar, Divya Cantor, and Jyoti Burruss. Divya Cantor and Greg Cooper provided comments that greatly helped the development of this essay. Thanks also to Bert Menco for helpful technical discussion of EM techniques and suggested changes to text on the topic. An anonymous reviewer provided many helpful suggestions and even provided an excellent transition sentence which was added verbatim. I am grateful to Sharlene Santana and Brock Fenton for sharing their wonderful photographs of Desmodus, and Valerie DeLeon for CT scans of Anoura.