Histology and ultrastructure of olfactory and nasal respiratory mucosae in suckling and adult African grasscutters (Thryonomys swinderianus- Temminck, 1827)

Grasscutters (GRCs) are hystricognath rodents that predominate West African countries where they are captured and bred in captivity as “microlivestock” and for research. Consequently, research priority has, of late, shifted to aspects of GRC biology particularly with regard to morphofunctional aspects of its body systems. The olfactory system plays critical roles in regulating social, sexual, maternal and feeding behaviors. This study examines, histologically and ultrastructurally, the pattern and magnitude of remodeling of the GRC olfactory mucosa (OM) and nasal respiratory mucosa (NRM) between suckling and adult ages and compares these with what is documented for other mammals. In adults, tubular-type Bowman’s glands, olfactory receptor neuron (ORN) axon bundles and blood vessels were uniformly distributed in the OM lamina propria contrary to sucklings where acinar-type Bowman’s glands lay superficially and the bundles relatively deeper. Apically in the adult NRM epithelium, ciliated and non-ciliated cells were uniformly distributed as opposed to the sucklings where linearly arranged ciliated cells separated large numbers of non-ciliated cells. Quantitatively between the suckling and adult ages, respective increment values (%) were 28.2, 23.0, 28.1 and 52.9 for OM epithelial thickness, axon bundle diameter, ORN packing density and cilia number/ORN dendritic knob. Age-related increment in volume density (%) was 53.9, 31.6, 19.4 and 46.3 for Bowman’s glands, axon bundles, OM vessels and NRM glands, respectively. We conclude that microstructural refinement of the OM and NRM varies in qualitative and quantitative detail depending on age and species and that phenotypic plasticity in these structures suggests environmentally driven morphology.


Introduction
The grasscutter (GRC) (Thryonomys swinderianus), also called great cane rat, is a tropical African cavy-like hystricomorph rodent that predominate the West African subregion including Senegal, Ghana, Nigeria, Togo, Benin and Côte d'Ivoire (Ogunsanmi et al. 2002;Woods and Kilpatrick 2005). In these countries, GRCs are aggressively hunted for their meat and also captured and raised in cages as "microlivestock" (home consumption and sale) and for use in research (Jori et al. 2001;Fa et al. 2002;Opara 2010;Ibe et al. 2017;Ingweye and Kalio 2020). Indeed, the economic potential of GRCs is the reason many organizations (including Non-Governmental Organizations) interested in reducing poverty are promoting its production (Adu et al. 2017). Farming of GRCs is associated with good quality meat, inexpensive feeds, short gestation interval, large litter size and fast growth rate (Adu and Yeboah 2003;Matthews 2008;Owen and Dike 2012). The meat of GRCs enjoys a higher premium price per kilogram weight than chicken, beef, pork and mutton among many West Africans and elsewhere (Adu et al. 2017). In the wild, GRCs inhabit dense and thick cane-like grasses growing in damp places where they live in small groups led by a single male (Woods and Kilpatrick 2005;Aluko et al. 2015;Adu et al. 2017). These animals exhibit nocturnal and precocial behaviors and have an uncommon phenotype and life history (Mustapha et al. 2020). Despite their size and short limbs, GRCs are quick runners and skilled swimmers and their sense of sight is relatively poor, making communication highly dependent on hearing and well developed sense of smell (Opara 2010). They are mainly fossorial, nesting and living much of the time in burrows and feed on roots, shoots, and stems of various grasses (Wood 1994;Williams et al. 2011).
In mammals, the nasal cavity has important and diverse functions including filtering, warming and humidification of inhaled air as well as olfaction (Cole 1993). Among the mammalian species, the nasal cavity varies, not only in gross structure, but also in the distribution of epithelial types as follows: (1) keratinized stratified squamous epithelium lining the nasal vestibule, (2) ciliated pseudostratified columnar epithelium (respiratory epithelium) in the main nasal chamber covering the nasal turbinates and parts of the nasal septum, (3) non-ciliated columnar epithelium (also called transitional epithelium) located between the squamous and respiratory epithelia, and (4) olfactory epithelium (OE) found in the caudal roof of the nasal cavity mainly lining the ethmoidal concha (Harkema 2006;Harkema et al. 2006). Evolutionary pressure exerted on the olfactory cue defines two categories of mammals: (a) those with complex noses with the olfactory mucosa (OM) covering a great part of the nasal cavity having olfaction as the primary function (macrosmatic) e.g., rodents, rabbits and dogs and (b) those with simple noses (OM covers a small portion of nasal passage) and have breathing as the main function (microsmatic) e.g., humans and monkeys (Gross et al. 1982;Sorokin 1988).
In the main nasal chamber, the nasal respiratory mucosa (NRM) consists of a luminal surface epithelium and an underlying lamina propria containing glands, blood vessels and lymphatics and nerves embedded in a connective tissue matrix. The NRM epithelium has the following cell types: mucous (goblet) cells, ciliated cells, non-ciliated columnar cells and basal cells (Monteiro-Riviere and Popp 1984). The mucous cells are predominantly located in NRM epithelium lining the proximal part of the nasal septum whereas nonciliated columnar cells (also called serous cells) are the primary secretory cells in the remainder of the NRM epithelium (Harkema et al. 1989). Goblet cells represent 5-15% of cells in the respiratory mucosa and produce secretions for the endonasal mucus together with the submucosal glands. Non-ciliated cells represent up to 70% of the epithelium and have 300-400 microvilli on their surface (Mygind et al. 1982). Another 20-50% of epithelial cells are ciliated cells possessing 200-300 cilia on their apical surfaces, which are the morphological substrate of the mucociliary clearance (Beule 2010).
The structure of the OM has comprehensively been described in our previous papers (Kavoi et al. 2010(Kavoi et al. , 2012. In brief, the atypical OM epithelium has the basal (progenitor) cells residing at the bottom of the OE overlying the basement membrane. Above these cells is a layer of bipolar olfactory receptor neurons (ORN) whose dendrites extend to the apical surface of the epithelium to form knob-like structures from which several immotile cilia emanate. Axons of the ORN pierce the basement membrane to enter the lamina where they fasciculate before passing via the crevices of the cribriform plate to reach the olfactory bulb. The most apical zone of the OE is occupied by the nuclei and cytoplasm of the supporting cells, whose apices possess microvilli. On the apical surface of the OE, the ORN cilia are enmeshed with each other and with the supporting cell microvilli in the surface fluid thus providing an extensive surface area for reception of odorants (Menco 1980;Menco et al. 1997). The lamina propria beneath the basement membrane accommodates the ORN axon bundles, Bowman's glands and vasculature. The Bowman's glands have acini and narrow secretory ducts going out through the OE. The secretions released by these glands on the epithelial surface provide an environment for dissolving odorants thus allowing their diffusion to the sensory receptor sites on the cilia (Getchell and Getchell 1992).
Structural remodeling of the OM takes place as animals transit from suckling to adulthood and this has previously been demonstrated in the rat (Meisami 1989), dog (Kavoi et al. 2010) and rabbit (Kavoi et al. 2012). Anatomical data on nervous system structures are essential in understanding the adaptive physiology of an animal species including its behavior in the natural habitat and in captivity. In the literature, substantial amount of information is available on the GRC brain (Nzalak et al. 2008;Byanet et al. 2009;Ajayi et al. 2011;Byanet and Dzenda 2014;Ibe 2016;Ibe et al. 2017Ibe et al. , 2019, spinal cord (Mustapha et al. 2015(Mustapha et al. , 2017 and the eye (Peter-Ajuzıe et al. 2019). The olfactory system plays a pivotal role in communication by regulating multiple and integrative functions including feeding, social behavior, reproductive events (e.g., mother-neonate interaction) and emotional responses (Lledo et al. 2005). In view of the importance attached to the olfactory system with respect to animal survival, adaptive behavior and performance during domestication, it is necessary that the anatomical design of GRC olfactory system is well understood. To the best of our knowledge, the only study available on the GRC olfactory system anatomy is that by Kavoi et al. (2021) in which nasal and brain olfactory components were examined grossly and histologically. To explore this further and to generate new information, we analyzed the OM and NRM in suckling and adult GRCs for qualitative and quantitative characteristics at light and electron microscopy to understand their morphofunctional dynamics. Additionally, data obtained in this study are compared with what is previously recorded for other mammalian species.

Experimental animals
Animals used for this study were healthy male captivebred GRCs of two different age groups: ten sucklings aged 4-6 weeks (345-460 g) and ten adults aged 13-15 months (2.8-3.2 kg). These animals were randomly selected from a colony raised in a conventional animal housing facility in the Department of Animal Science, University of Ghana. Animals selected for this study were transported in well ventilated cages to the School of Veterinary Medicine, University of Ghana, Anatomy Laboratory, where they were immediately euthanized with lethal doses of sodium pentobarbitone (140 mg/kg, intravenously). All procedures performed on the GRCs were approved by University's Animal Care and Use Committee and strictly conformed to the guidelines provided in the Animals (Scientific Procedures) Act 1986.

Tissue fixation and harvesting
Euthanized animals were immediately perfused transcardially through the left ventricle of the heart with 10% formaldehyde for histology (n = 5 animals/ age group) and 2.5% phosphate-buffered glutaraldehyde (pH 7.4) for scanning electron microscopy (SEM) and toluidine blue staining (n = 5 animals/age group). Respectively, olfactory and nasal respiratory mucosae were obtained from nasal and ethmoidal conchae following mid-sagittal sectioning of the animal skull and removal of the nasal septum. In each case, the conchae were transected perpendicular to their long axes into caudal, middle and anterior portions (Kavoi et al. 2010) from which tissue sub-segments for microscopy were selected by systematic random sampling.

Processing of tissues for light and scanning electron microscopy
Tissues for histology (from animals perfused with formaldehyde) were decalcified in 5% EDTA (Alers et al. 1999), washed in distilled water and dehydrated in increasing concentrations of ethanol (50%, 70%, 80%, 90% and twice in 100%). Dehydrated tissues were then transitioned through methyl benzoate and infiltrated and embedded in paraffin wax, sectioned in the transverse plane at 5 µm using a rotary microtome and stained in Masson's trichrome.
Glutaraldehyde-fixed tissues were post-fixed in 1% osmium tetroxide, dehydrated in graded dilutions of ethanol (70%, 80%, 95% and 100% × 2), cleared in propylene oxide and embedded in epoxy resin. The resin blocks were cut with glass knives using a Sorvall ® ultramicrotome to obtain semi-thin sections which were picked on glass slides and stained with 0.5% toluidine blue. After dehydration, some sections were selected for SEM, critical point dried in liquid carbon dioxide, mounted on brass stubs and sputter coated with gold-palladium complex. The sample surfaces were then examined using a Jeol 330 SEM operating at 15 kV and images captured on Orion Saturn frame grabber system.

Morphometric analysis
Quantitative parameters were analyzed at both light microscopy and SEM levels using a graticule fitted inside the eyepiece of the microscope. Tissue samples from each age group were obtained, randomly selected and processed from 10 animals (five for light microscopy and five for SEM) from which 10-15 histological and 8-10 SEM micrographs were prepared. In both cases, morphometric analysis was done on 30-35 test fields generated from randomly selected micrographs. The OM was analyzed for epithelial thickness, diameter of ORN axon bundles and volume densities for the axon bundles, Bowman's glands and vessels at light microscopy and densities of ORN and cilia numbers/ORN knob at SEM. In the lamina propria of the NRM, volume density values were worked out for both glands and vessels. All the aforementioned parameters were determined following the steps detailed in Kavoi et al (2012). In brief, the thickness of OE was measured from the basement membrane to the apical surface while diameters ORN axon bundle were worked out from mean linear intercept lengths. Volume densities were estimated by point counting using an overlay of a coherent test system of points. Packing densities of ORNs were estimated by counting the ORN knobs projecting from the OE in a millimeter square area while taking into account the forbidden line rule. In determining cilia counts/ ORN knob, the obscured ones, which are estimated at 25%, were incorporated in the final total value. Data on morphometry were analyzed using Student's t test. In all cases, data were expressed as means ± standard deviation (SD). For differences in means between groups, statistical significance captured at p < 0.05.

Light microscopy
The GRC OM is shown in Fig. 1. The OM epithelial zones (nuclear, non-nuclear and free zones) were well established in both suckling and adult animals, with basement membranes of almost the same thickness. In the lamina 1 3 propria of both age groups were ORN axon bundles, Bowman's glands and vessels. These three structures were distributed uniformly in the lamina propria of the suckling animals in contrast to the situation in the adults where the bundles were placed deeper than the glands. The glands in the suckling animals were of the acinar type while those in adults were tubular.
In Fig. 2A1, semi-thin sections showed the cellular components of the NRM epithelium in both suckling and adult GRCs to be: (1) ciliated cells whose apical surfaces possess numerous cilia, (2) non-ciliated columnar cells whose nuclei are positioned deeper than those of the ciliated cells and, (3) basal cells that are restricted to the lowermost region of the epithelium superjacent to the basement membrane. The NRM lamina accommodates glands and blood vessels ( Fig. 2A, B). The glands in both age groups were of the acinar-type and in the adult the glands appeared more closely packed and the vessels were of greater diameters.

Scanning electron microscopy
The apical surfaces of the OM epithelium in suckling and adult animals are shown in Fig. 3. In both age groups, the apical surfaces of the OE showed ORN dendrites with cilia-bearing knobs, apices of supporting cells and tunnellike openings of Bowman's gland ducts. On the OE surface, individual ORN dendrites and supporting cell apices were almost equally distributed in the suckling animals (Fig. 3A1). This differed from the situation in the adults where the ORN dendrites appeared more numerous per unit area and were packaged in a manner that appeared to conceal the supporting cell apices (Fig. 3A2). Additionally, olfactory cilia in the adults emerged from around the ORN knobs to project in a radial manner as opposed to the suckling animals where the cilia arose from the knobs to run in parallel in form of a tuft (Fig. 3A2, B2). Figure 4 shows the ultrastructural features of the NRM epithelium in suckling and adult GRCs. In both age groups, ciliated cells projected numerous cilia above the general NRM epithelial surface whereas apical surfaces of  A1 and B: the OM presents a distinctly formed nuclear (Nu) and non-nuclear (Nn) zones and an equally thick basement membrane (asterisks). In the lamina propria, ORN axon bundles (Ab), Bowman's glands (Bg) and blood vessels (arrows) present a uniform pattern of distribution in suckling animals in contrast to the adults where the bundles take a deeper position than the glands. The glands are acinar type in suckling and tubular in the adults A2: shows the conspicuous free zone (double arrow heads) and terminal bar (arrow) that characterizes the OE in both age groups. A1 and A2-Masson's trichrome stain, B-toluidine blue stain. Scale bar = 40 µm in A1 and B, 20 µm in A2   Fig. 2 Semi-thin sections of GRC nasal respiratory mucosa in suckling (A1, A2) and adult (B) age groups. A1: the epithelium in both groups present three cell types namely, ciliated cells (C) projecting tufts of cilia (arrows) on the epithelial surface, non-ciliated columnar cells (N) with nuclei positioned below those of the ciliated cells and basal cells (B) that are confined to the basal part of the epithelium where they overlie the basement membrane (asterisks). A2 and B: the lamina propria is observed to have a greater number (per unit area) of acinar-type glands (G) and larger diameter blood vessels (V) in the adult than in the suckling animals. Toluidine blue stain. Scale bar = 10 µm in A1, 20 µm in A2 and B 1 3 non-ciliated cells were covered with plenty of microvilli. Intercellular spaces and tunnel-like openings of subepithelial gland ducts were well discerned in both postnatal age groups. The NRM epithelium in the suckling animals was characterized by linearly arranged ciliated cells separating relatively large numbers of non-ciliated cells (Fig. 4A). Conversely in the adults, individual ciliated and non-ciliated cells appeared uniform in terms of number per field and distribution (Fig. 4B). Table 1 presents data on thicknesses of OM epithelia and diameters of ORN axon bundles in the suckling and adult GRCs (current study) and in rabbits, dog and sheep (previous studies). Between the suckling and adult stages of development, increment values were 28.2% (59.5 ± 4.1-76.3 ± 5.4 µm) for OE thickness and 23.0% (57.4 ± 6.0-70.6 ± 8.6 µm) for axon bundle diameters.

Morphometry
Regarding the OM lamina propria, volume density values for Bowman's glands, axonal bundles and blood vessels in the GRC are compared with those earlier recorded in rabbits (Table 2). In the GRC between suckling and adult ages, volume density increment was In both age groups, the epithelium is characterized by ciliated (C) cells that project hundreds of cilia (arrow heads) above the epithelial surface, and non-ciliated cells (N) with a numerous microvilli on their apical surfaces. Distinct intercellular boundaries (arrows) and tunnel-like openings of subepithelial gland ducts (asterisks) also characterize the epithelia of both age groups A: the epithelium contains a preponderance of non-ciliated cells forming a carpet punctuated by linearly arranged ciliated cells (see dotted arrow) B: ciliated and non-ciliated cells occur in nearly equal numbers and their distribution pattern is fairly uniform. Scale bar = 10 µm in A and B 53.9% (32.3 ± 6.0-49.7 ± 5.1) for the glands and 19.4% (7.2 ± 1.1-8.6 ± 3.3) for the vessels. For the axon bundles, GRCs recorded a volume density increment (between suckling and adult ages) of 31.6% (23.7 ± 5.2-31.2 ± 4.4) ( Table 2).

Fig. 5
Mean volume densities (%) for glands and blood vessels in the NRM of suckling and adult GRCs. Volume density increment is agerelated and quite significant in the case of the glands, p < 0.05

Discussion
In the current study, the anatomical design of the OM and NRM are analyzed using histological, ultrastructural and morphometric methods to unravel salient features of the mucosae that are specific to this species and those that accompany growth to adulthood. Additionally, a comparison is made to clarify how and to what extent, between suckling and adult ages, structural remodeling of the OM and NRM in the GRC (order Rodentia) varies from other taxonomically different mammals already studied including dogs (order Carnivora), sheep (order Artiodactyla) and rabbits (order Langomorpha). At histological level, the OM in both suckling and adult GRCs presents features that typify the mammalian OM. As the animals grew to reach adulthood, OE thickness and ORN axon bundle diameters are noted to increase. We attribute this to the age-related increment in ORN densities and the associated increase in axon numbers arising from the newly forming ORNs. Furthermore, optimization of olfactory function during juvenile development would require greater numbers of ORN axons to converge upon central relay neurons in the brain. This enhances the physiological capacities of the olfactory afferent pathway by increasing the opportunity for spatial summation and facilitation (Meisami 1989). Extensive work on mammalian OM structure (Yamagishi et al. 1989;Ferrari et al. 2000;Kumar et al. 2000;Kavoi et al. 2010Kavoi et al. , 2012Onyono et al. 2017;Kavoi 2018) has shown the distribution of ORN axon bundles and Bowman's glands and the structural form of the glands to vary among species and postnatal age groups. Similar differences are noted in the GRC where the Bowman's are of the tubular-type and acinar-type in the adults and suckling animals, respectively. In the suckling animals, the glands and the bundles are uniformly distributed as opposed to the adults where the glands are located superficially while the axon bundles are placed deeper in the lamina propria. Prenatally, Bowman's glands develop as buds of epithelial tissue in the lamina propria (Sangari et al. 1992). Arguably therefore, age-related differences in distribution pattern, structural forms and volume densities of the Bowman's glands may be linked to the anatomical transformation that happens as the glands develop from primitive to definitive forms.
In the NRM epithelium, ciliated cells play role in mucociliary clearance whereas non-ciliated columnar cells have microvilli that increase their apical surface area for retention of moisture and to prevent drying of the epithelial surface (Mygind et al. 1982;Beule 2010). Basal cells, on the other hand, have progenitor properties and can self-renew and differentiate into other epithelial cell types (Rock et al. 2010). In suckling GRCs, the ciliated cells are linearly arranged leaving wide spaces occupied by the non-ciliated cells whereas adult GRCs showed ciliated and non-ciliated cells that are fairly uniformly distributed. In a study by Maina et al. (1992) in which the ultrastructure of the trachea was compared in two species of East African mole rats, differences in the number and distribution pattern of ciliated and non-ciliated cells were observed, a feature that was believed to signify differences in structural modulation of the airway and to have a bearing on phylogeny, functional requirements and behavior. Notably also, agerelated increase in vascularization in the GRC NRM and OM may, according to Sangari et al. (2000), be associated with increasing metabolic demands of replicating mucosal cells toward completion of their functional maturation.
The ultrastructural appearance of the apical surface of the OE is influenced by both the animal's age and species. The radial pattern of cilia projection, which has also been demonstrated in humans (Lenz 1977), horses (Kumar et al. 2000), dogs (Kavoi et al. 2010) and Sengis (Kavoi 2018) is associated with higher cilia numbers when compared with the parallel pattern commonly seen in bovids (Menco 1978;Kavoi et al. 2010). In rabbits (Kavoi et al. 2012), the transition from suckling to adulthood showed OE surface changes similar to those currently noted in GRCs. Such refinement in OE structure is function-related and an important prerequisite to self-dependency (change in lifestyle including feeding from sucking to gnawing) at adulthood. Furthermore, the aforementioned differences in OE cytoarchitecture reflects the great variability in the number of ORN receptor genes among species that are reminiscent of the species' lifestyles and which is caused by frequent gene gains and losses during evolution (Niimura et al. 2014). Besides, Dubied et al. (2021) compared ontogenetic trajectories of mandible shape in rodents (suborders Hystricomorpha and Myomorpha) to reveal that epigenetic interactions between the skull bones and muscles influence spatialization of bone formation and remodeling in response to biomechanical strain, a mechanism that was ascribed to developmental plasticity emanating from evolutionary divergences between species and clades.
This study compares GRC data on packing densities of ORNs and cilia number/ per ORN knob between suckling and adult ages with what is documented for the rabbit (Kavoi et al. 2012) whose altricial young rely heavily on olfaction for nipple attachment (Coureaud et al. 2008), the dog (Kavoi et al. 2010), a relatively generalized carnivore with well-developed olfactory cue for tracking and catching prey (Walker 1975) and the sheep, a specialized grazer whose olfactory cue is comparatively less important (Levy et al. 2004). We note that the above mentioned parameters are greatest in rabbits, followed by GRCs, dogs and then sheep, a pattern that appears to match with olfactory function 1 3 demand levels for each species. ORN cilia are the principal sites for the initial events of olfactory transduction and signaling (Menco et al. 1997) and therefore the higher the ORN density and cilia number/ORN knob, the greater the area of the cilia plasmalemma is made available for odor binding.
Overall, results of this work show that the GRC OM and NRM undergo, during juvenile development, histological and ultrastructural modifications at levels that result to functionally better specialized mucosae than is the case (for the OM) in the dog and sheep (Kavoi et al. 2010) but generally less so when compared with the rabbit (Kavoi et al. 2012). Behaviorally, GRCs are nocturnal herbivores that travel at night through trails in reeds and grass and that exhibit fossorial, precocial and solitary lifestyles (NRC 1991;Mustapha et al. 2020;Williams et al. 2011;Adu et al. 2017). Adaptation to these behavioral characteristics requires a sophisticated olfactory system whose structural details are presented here and in our earlier paper (Kavoi et al. 2021). We assert that different selective pressures, which are undoubtedly liked to dynamics of evolution, have acted upon the olfactory systems of the above species to produce the observed outcomes. These data form an important basis for understanding GRC biology and physiological adaptation and may find application in the broader areas of comparative anatomy, wildlife behavior and mammalian evolution. Future studies on the olfactory system in hystricognaths should incorporate more representative species.