Page 1
Draft
Olfactory epithelium ontogenesis and function in postembryonic North American bullfrog tadpoles
(Lithobates catesbeiana)
Journal: Canadian Journal of Zoology
Manuscript ID cjz-2019-0213.R1
Manuscript Type: Article
Date Submitted by the Author: 21-Nov-2019
Complete List of Authors: Heerema, Jody; University of Lethbridge, Department of Biological SciencesBogart, Sarah; University of Lethbridge Department of Biological Sciences, Helbing, Caren; University of Victoria, Department of Biochemistry & MicrobiologyPyle, Greg; University of Lethbridge, Department of Biological Sciences
Is your manuscript invited for consideration in a Special
Issue?:Not applicable (regular submission)
Keyword:Lithobates catesbeiana, North American bullfrog tadpole, OLFACTION < Organ System, BEHAVIOUR < Discipline, METAMORPHOSIS < Discipline, EOG
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 2
Draft
1
Title: Olfactory epithelium ontogenesis and function in postembryonic North American
bullfrog tadpoles (Lithobates catesbeiana)
Authors: J. L. Heerema1, S. J. Bogart1*, C. C. Helbing2, G. G. Pyle1
Author affiliations:
1Department of Biological Sciences, University of Lethbridge, Lethbridge, AB, Canada
2Department of Biochemistry and Microbiology, University of Victoria, BC, Canada
Author emails: JLH: [email protected] ; SJB: [email protected] ;
CCH: [email protected] ; GGP: [email protected]
Author ORCID iDs: JLH: 0000-0002-2202-8819; SJB: 0000-0002-2313-8649;
CCH: 0000-0002-8861-1070; GGP: 0000-0003-4338-7674
*Corresponding author: Sarah J. Bogart, Dept. of Biological Sciences, University of Lethbridge,
4401 University Drive, Lethbridge, AB, Canada, T1K 3M4. Phone: 403-332-4048. E-mail:
[email protected]
Page 1 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 3
Draft
2
Title: Olfactory epithelium ontogenesis and function in postembryonic North American
bullfrog tadpoles (Lithobates catesbeiana)
Authors: J. L. Heerema1, S. J. Bogart1*, C. C. Helbing2, G. G. Pyle1
Abstract
During metamorphosis, the olfactory system remodelling in anuran tadpoles—to transition from
detecting waterborne odorants to volatile odorants as frogs—is extensive. How the olfactory
system transitions from the larval to frog form is poorly understood, particularly in species that
become (semi-)terrestrial. We investigated the ontogeny and function of the olfactory epithelium
of Lithobates (Rana) catesbeiana Shaw 1802 tadpoles at various stages of postembryonic
development. Changes in sensory components observable at the epithelial surface were examined
by scanning electron microscopy (SEM). Functionality of the developing epithelium was tested
using a neurophysiological technique (electro-olfactography; EOG), and behaviourally, using a
choice maze to assess tadpole response to olfactory stimuli (algae extract, amino acids). The
youngest (premetamorphic) tadpoles responded behaviourally to an amino acid mixture despite
having underdeveloped olfactory structures (cilia, olfactory knobs) and no EOG response. The
consistent appearance of olfactory structures in older (prometamorphic) tadpoles coincided with
reliably obtaining EOG responses to olfactory stimuli. However, as tadpoles aged further, and
despite indistinguishable differences in sensory components, behavioural- and EOG-based
olfactory responses were drastically reduced, most strongly near metamorphic climax. This work
demonstrates a more complex relationship between structure and function of the olfactory system
during tadpole life history than originally thought.
Key words: Lithobates catesbeiana, North American bullfrog tadpole, olfaction, behaviour,
metamorphosis, EOG
Page 2 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 4
Draft
3
Introduction
The postembryonic development of anurans includes the metamorphosis of the free-
living aquatic, herbivorous tadpole to a (semi-)terrestrial, carnivorous frog (Gilbert et al. 1996).
During metamorphosis, there are major developmental changes in the anuran olfactory system as
it shifts from functioning exclusively in water environments in tadpole larvae to functioning in
both water and air environments as frogs (reviewed in Gascuel and Amano 2013). These
metamorphic changes may affect olfactory detection and response to food and predators by
larval anurans.
In the broadest sense, olfactory detection occurs at the surface of the olfactory
epithelium. The olfactory epithelium is composed of three cell types, which we will describe
briefly: sustentacular cells (SCs), which provide physical and metabolic support for the
epithelium, phagocytose dead neurons, and can help modulate odorant detection (reviewed in
Getchell and Getchell 1992, and Schwob 2002; Lucero 2013); basal (progenitor) cells, which are
involved in the maintenance and regeneration of the epithelium; and olfactory sensory neurons
(OSNs), which function to detect odours (reviewed in Manzini and Schild 2010). The OSNs are
bipolar neurons that project either cilia or microvilli—both richly supplied with olfactory
receptors (ORs)—to the apical surface of the epithelium (reviewed in Schwob 2002, and Manzini
and Schild 2010). Logically, if the surface structures of OSNs are underdeveloped, the olfactory
sense may be impaired.
In frogs, the development of the olfactory system throughout metamorphosis has been
well studied in Xenopus laevis Daudin 1802 (African clawed frog), but other species remain less
well documented (Dittrich et al. 2016; Syed et al. 2017). Although there are interspecies
differences in the olfactory system on a cellular level, the overall structural development is
Page 3 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 5
Draft
4
conserved (Gascuel and Amano 2013). Generally, the larval olfactory epithelium originates
from a placode, which gives rise to an olfactory pit (Klein and Graziadei 1983). Each olfactory
pit differentiates into a principal cavity (PC) and vomeronasal organ (VNO) (Hansen et al. 1998).
In the tadpole, the olfactory epithelium lies within the PC and contains both ciliated OSNs and
microvillous OSNs, which are specialized to detect only waterborne odorants (Hansen et al.
1998). The VNO contains microvillous OSNs and is presumed to be involved in the detection of
pheromones (conspecific signalling odorants; reviewed in Døving and Trotier 1998), but may
have other functions. As development proceeds towards metamorphosis, the middle cavity (MC)
is formed de novo between the PC and VNO (Hansen et al. 1998; Dittrich et al. 2016) and the PC
olfactory epithelium is remodelled extensively (Hansen et al. 1998). After metamorphosis, the
PC contains only ciliated OSNs (Mezler et al. 1999) and is specialized to detect volatile odours,
which are typical of a terrestrial environment (Higgs and Burd 2001; Gascuel and Amano 2013).
In most species, the MC contains only non-sensory epithelium. However, in the case of X.
laevis, the adult MC resembles the larval PC, and contains both microvillous OSNs and ciliated
OSNs, and functions to detect odours in aquatic environments (Hansen et al. 1998; Wang et al.
2008; Gascuel and Amano 2013; Dittrich et al. 2016). The frog VNO does not change from the
tadpole form and continues to contain only microvillous OSNs (Taniguchi et al. 1996; Hansen et
al. 1998). It should be noted that in contrast to the majority of anurans, X. laevis remains aquatic
in adult form (despite it also becoming air-breathing). Thus, due to interspecies differences, X.
laevis may not be a true model of olfactory development for other anurans undergoing
metamorphosis, especially regarding their olfactory-mediated function and behaviours.
The development of the olfactory system and many other tissues during metamorphosis is
driven by thyroid hormones (THs) thyroxine (T4) and 3,3’,5-triiodothyronine (T3) (Brown and
Page 4 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 6
Draft
5
Cai 2007). Metamorphic changes include cell apoptosis resulting in tail resorption and loss of
gills; organogenesis of limbs, stomach, and bone marrow; and the complete remodelling of the
intestine, brain, and spinal cord (Tata 2006; Brown and Cai 2007). Metamorphosis is divided
into three stages with respect to endogenous TH concentrations. During premetamorphosis, the
thyroid gland is inactive and no TH circulates through the anuran body. During
prometamorphosis, the thyroid gland becomes active and begins secreting TH and initiating
metamorphic changes. At metamorphic climax, endogenous TH is at its highest concentration,
triggering extensive changes and remodelling throughout the anuran body (reviewed in Tata,
2006; Brown and Cai 2007). Given the substantial changes that occur in the tadpole olfactory
system during development, changes in chemosensory function, including olfaction, are
expected. However, very little is known about the nature of these structural and functional
changes in frogs that more completely transition to a terrestrial existence, such as in the North
American bullfrog, Lithobates (Rana) catesbeiana Shaw 1802.
At a functional level, previous studies have utilized chemosensory cues to evoke tadpole
behavioural responses in a variety of species using many different assays (Mirza et al. 2006;
Smith et al. 2008; Takahara et al. 2012). Two common approaches to studying chemosensation,
particularly olfaction, are to either use calcium imaging (Sachse and Galizia 2002) or to measure
the odour-evoked extracellular field potentials using electro-olfactography (EOG) (reviewed by
Scott and Scott-Johnson 2002). Early studies on Lithobates pipiens Schreber 1782 frogs
demonstrated the utility of EOG measurements for defining electrical responses to a series of
stimuli (Getchell 1974). Odour molecules interact with ORs on the cilia or microvilli of OSNs,
triggering a molecular signal transduction cascade that leads to bulk depolarization of the OSNs.
The EOG measures the loss of cations from the extracellular environment (i.e. at the surface of
Page 5 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 7
Draft
6
the olfactory epithelium) resulting from this bulk depolarization. The amplitude of the resulting
electro-olfactogram is related to the magnitude of the bulk depolarization of the OSNs after
odorant stimulation, thus indicating the magnitude of the olfactory response. Previous studies
have used EOG to measure the olfactory response of fish to odorants (Green et al. 2010; Dew et
al. 2014), and the technique has recently been modified for use with L. catesbeiana tadpoles
(Heerema et al. 2018a).
The purpose of the present study was to investigate ontogenetic changes in the L.
catesbeiana tadpole olfactory system during metamorphosis, as measured by changes in the
surface ultrastructure of the olfactory epithelium and in the olfactory-based, neurophysiological
and behavioural responses to odorants. Overall, we hypothesize that 1) changes in olfactory
structures during tadpole development will lead to measurable differences in neurophysiological
responses and ultimately alter tadpole behaviour. We used scanning electron microscopy (SEM)
to compare the surface ultrastructure of the olfactory epithelium, EOG to examine for
measurable olfactory responses at the neurophysiological level, and behavioural choice maze
tests to elucidate any behavioural changes in tadpoles, at various stages of post-embryonic
development. For this study, tadpoles were grouped into previously published, developmental
groups that are classified according to differences in the concentration of endogenous TH and
outwardly recognizable markers of development (i.e. premetamorphosis, prometamorphosis, and
nearing metamorphic climax are marked by limb development). Additional investigations were
conducted within these developmental groups, as needed, to further understand variations in
structure, function, and behaviour.
Page 6 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 8
Draft
7
Materials and Methods
Animals
Lithobates catesbeiana (Shaw, 1802) tadpoles of mixed sex were wild caught in Victoria,
BC, Canada, by Westwind Sealab Supplies under permit from the Capital Regional District
Parks and Recreation. Tadpoles were housed for two weeks in dechlorinated city water in a re-
circulatory system in the Aquatic Research Facility (ARF) at the University of Lethbridge, AB,
Canada, prior to experimentation. Housing conditions included 15°C water temperature on a
light: dark 16:8 h photoperiod and tadpoles were fed ad libitum daily with Spirulina spp. Turpin
Ex Gomont, 1893 algae flakes. Prior to experiments tadpoles were fasted and temperature-
acclimated to 24 °C for 72 hours in aerated, dechlorinated water in 15 L polypropylene buckets.
During the temperature acclimation water quality including temperature (mean, range; 24.2, 21.4
– 25.0 °C, n = 16), dissolved oxygen (mean, range; 78.9, 46.7 – 94.7 %; 6.7, 4.1 – 7.9 mg/L, n =
15) was measured daily and pH (median, range; 8.27, 7.79 – 8.39, n = 9) every other day.
Hardness (mean ± SE; 178 ± 1 mg/L as CaCO3) and alkalinity (mean ± SE; 136 ± 2 mg/L as
CaCO3, n = 30) are routinely measured in the ARF and are stable through time. Tadpoles were
staged according to Taylor and Kollros (1946) and will be referred to by developmental group
for ease of comparison according to Table 1. All procedures involving tadpoles were approved
by the Animal Welfare Committee at the University of Lethbridge (protocol #1401) and the
University of Victoria Animal Care Committee (protocol #2011-030), for compliance under the
Canadian Council on Animal Care guidelines.
<Table 1 placement>
Page 7 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 9
Draft
8
Experimental design
The initial development of the olfactory sense was examined using the youngest tadpoles.
Specifically, the EOG response of premetamorphic and early prometamorphic tadpoles (mean ±
SE body mass: 7.0 ± 0.4 g; median, range: TK VI, I – XIII, n = 7) was measured to examine the
ontogenetic progression of a measurable olfactory response to a food cue (Spirulina extract;
algae). To determine whether olfaction was used to behaviourally respond to chemical stimuli at
early stages, premetamorphic tadpoles (mean ± SE body mass 4.6 ± 0.28 g; median, range: TK
IV, I – VIII, n = 10 - 12) were rendered anosmic and their response to an olfactory stimulus was
tested in a choice maze relative to unaffected animals. In this test, an amino acid mixture was
used as the stimulus because premetamorphic tadpoles did not respond to Spirulina in
preliminary behavioural tests (see section on preparation of olfactory stimuli). To link the ability
to obtain EOG responses to the ontogenetic appearance of key sensory structures (cilia and
olfactory knobs) on the surface of the olfactory epithelium, olfactory pits from premetamorphic
tadpoles used in EOG measurements were excised and examined by SEM.
Later stage tadpoles were primarily examined for ontogenetic changes in olfactory
function and behaviour as metamorphosis progressed. Behavioural attraction responses to
Spirulina extract were measured in early prometamorphic tadpoles (mean ± SE mass 10.1 ± 0.9
g; median, range TK: XII, XI –XII, n = 9) using a linear trough-style choice-maze. Later
developmental stage tadpoles (mean ± SE mass 22.5 ± 1.1 g, n = 23) were divided into two
experimental groups: late prometamorphic and tadpoles approaching metamorphic climax (n =
10 – 11). For these groups, behavioural responses to Spirulina extract were recorded in the
linear trough-style choice maze. Then, EOG responses to Spirulina extract and an amino acid (L-
alanine) were measured on the same tadpoles. Lastly, olfactory pits were excised from these
Page 8 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 10
Draft
9
individuals and observed by SEM. Details of all preparations and procedures are as described in
the sections below.
Preparation of olfactory stimuli
All olfactory stimuli used in this study were previously established as appropriate cues to
study tadpole behaviour and neurophysiological responses (Manzini et al. 2002; Hassenklöver et
al. 2012; Heerema et al. 2018a, 2018b). The type and concentration of cues used in EOG and
behaviour experiments was dependent upon the ability to elicit consistent responses to the
stimuli, at the lowest concentration possible, as previously described by Heerema et al. (2018a,
2018b). Preliminary testing of lower concentrations of stimuli described below did not elicit
consistent behavioural responses in the tadpoles. Stimuli included Spirulina (algae) extract and
an amino acid (L-alanine), representing food cues, or an equimolar amino acid mixture (L-
alanine, L-serine, glycine), representing a predator cue. To prepare the Spirulina extract for
behavioural and EOG experiments, Spirulina flakes (2 g/L; Nutrafin Max, Hagen, Montreal,
Canada) were added to dechlorinated water, stirred for 30 minutes, and then filtered through
aquarium filter floss (Aqua-Fit, Hagen, Montreal, Canada). For EOG experiments, 10-2 M L-
alanine (USP grade, Sigma-Aldrich, Oakville, Canada) was dissolved in dechlorinated water.
Due to a lack of consistent behavioural responses to Spirulina and to single amino acids in
preliminary testing of premetamorphic tadpoles, an amino acid mixture was used instead
(Heerema et al. 2018a, 2018b). To prepare the amino acid mixture, equimolar concentrations
(0.022 M each) of L-alanine (USP grade, Sigma-Aldrich, Oakville, ON, Canada), L-serine (USP
grade, VWR, Radnor, USA), and glycine (proteomics grade, AMRESCO, Cleveland, USA) were
dissolved into dechlorinated water, producing a mixture concentration of 0.066 M. It should be
Page 9 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 11
Draft
10
noted that the stimuli concentrations reported here for use in behavioural tests are stocks and do
not reflect the final concentrations in the testing maze (see the section Behavioural choice maze
assays for details). All stimuli were prepared fresh daily.
Electro-olfactography (EOG)
Methodology of Heerema et al. (2018a) was used to conduct EOG on L. catesbeiana
tadpoles wherein tadpole olfactory responses to olfactory stimuli (cues; 2 g/L Spirulina extract
and 10-2 M L-alanine) were measured. In short, tadpoles were anaesthetized in buffered tricaine
methanesulfonate (500 mg/L; TMS, Aqua Life, Syndel Laboratories, Nanaimo, Canada) until the
heartbeat slowed to approximately one beat every two seconds. Preparation and placement of
the EOG electrodes were as previously described by Heerema et al. (2018a). Olfactory stimuli,
without dilution, were individually delivered to the exposed olfactory pit in pulses of five
seconds and a minimum of two minutes were allowed to pass between each stimulus delivery to
prevent habituation. The delivery order of the cues was randomized and the response to each cue
was measured three times throughout the EOG test. Responses to each stimulus were averaged
for each individual. Mean EOG responses to olfactory stimuli were blank corrected by
subtracting any response elicited by the blank across each individual.
Inducing anosmia
To induce anosmia, tadpoles were anesthetised in buffered TMS (500 mg/L) and the
olfactory pits were exposed in the same manner used for tadpole EOG (Heerema et al. 2018a).
Surgical glue (Vetbond, 3M Canada, London, Canada) was applied to the olfactory pits,
completely capping them, and allowed to dry for approximately 30 seconds, hence rendering the
Page 10 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 12
Draft
11
tadpole anosmic by blocking external access to the neurons but without removing them (i.e. the
least invasive option). The tadpoles were placed in dechlorinated water and allowed to recover
for 48 h before their behavioural responses to an olfactory stimulus was challenged in a choice
maze. The lack of response to an olfactory stimulus confirmed anosmic condition (see results
section).
Behavioural choice maze assays
To measure tadpole behavioural responses, the choice maze test protocol of Dew et al.
(2014) was used with a few species-specific modifications made for animal size. Linear troughs
(75 cm x 20 cm x 15 cm; l x w x h), representing the choice maze, were divided lengthwise into
three zones (see Dew et al., 2014 for maze illustration). The two distal zones served as areas to
receive olfactory stimuli (stimulus-delivery zones) while the middle section served as the
acclimation zone. Placed within the acclimation zone was a clear plastic, bottomless container
(21.5 cm x 15 cm x 11.5 cm; l x w x h), serving as an acclimation chamber. This chamber
allowed tadpoles to adjust to maze conditions while it prevented their access to either of the
stimulus-delivery zones during the acclimation period, which immediately preceded each
behavioural trial.
For each trial, 8 L of dechlorinated water was added to the maze. A single tadpole was
placed in the acclimation chamber and allowed to acclimate to maze conditions for 20 min prior
to the start of the choice assay. Then, to each stimulus-delivery zone, 50 mL of blank
(dechlorinated water) or stimulus (2 g/L of Spirulina extract or an equimolar mixture of L-
alanine, L-serine, and glycine; total molarity 0.066 M) were remotely administered via silicone
tubing and syringes. Once administered to the maze, the cue and blank were allowed to diffuse
Page 11 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 13
Draft
12
for one minute (as determined in odorant diffusion tests) before the acclimation chamber was
remotely lifted via an overhead pulley system. The lifted acclimation chambers allowed the
tadpole to swim freely throughout the maze. Four tadpoles in separate mazes were tested
simultaneously, and the position of each tadpole in the maze was recorded every 10 sec for five
minutes. Time spent in each arm (stimulus zone) was calculated as in Dew et al. (2014). Note
that, the stimulus concentrations after complete cue dispersal in the stimulus delivery zone
(~3.14 L based on maze dimensions) were less than the administered concentration. The final
(dispersed) stimulus concentration in the maze would be approximately 0.001 M for the amino
acid mixture and approximately 0.03 g/L for Spirulina extract. The Spirulina extract did not
change the colour of the behavioural testing water, meaning that the tadpole visual system could
not be responsible for any resultant behaviour.
Several quality control measures were used in behavioural experiments. Without
animals, odorant diffusion tests were used to determine the diffusion time required for the cue
(50 mL of dechlorinated water mixed with food colouring) to reach the acclimation zone post
delivery. To ensure tadpoles did not inherently favour one arm of the maze, 50 mL of
dechlorinated water was administered to each arm of the maze and behavioural trials were run as
described above, prior to running behaviour experiments. The assignment of the blank or
olfactory stimulus to one end of the maze or other was randomized for each successive trial and
the researcher was blinded as to which end each stimulus was administered. To not influence
trials, all trials were recorded with a webcam (HD 720p, Logitech, Romanel-sur-Morges,
Switzerland) and observed on a laptop computer (MacBook Air, Apple, Cupertino, USA).
Page 12 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 14
Draft
13
Scanning electron microscopy (SEM)
Immediately following EOG experiments, tadpoles were euthanized in pH-buffered
tricaine methane sulfonate (MS-222; 1000 mg/L) solution. Olfactory pits were excised under a
dissecting microscope and fixed by immersion in Karnovsky’s fixative (Karnovsky 1965) for 24
h at room temperature. Fixed tissues were successively rinsed twice in cacodylate buffer (0.2 M;
pH 7.2; Ted Pella Inc., Redding, USA) for 15 min each, and dehydrated in a series of graded
ethanol dilutions (75% to 100% anhydrous; Commercial Alcohols, Brampton, Canada). Tissues
were dried and mounted on 15 mm aluminum stubs, sputter coated with platinum, and observed
with an SEM (S-3400N SEM, Hitachi, Tokyo, Japan) at the Lethbridge Research and
Development Centre, Lethbridge, AB, Canada. Entire olfactory pits were examined and imaged
via SEM in the youngest tadpoles. However, only the PC olfactory epithelium was imaged in
older tadpoles, which was differentiated from the VNO based on sensory cell types present (re
Døving and Trotier 1998; Hansen et al. 1998). SEM images were adjusted for brightness and
contrast using Adobe Photoshop CC v2015.5.0 and assembled into an image plate in Adobe
Illustrator CC v2015.3.0 (both Adobe Systems Inc., San Jose, USA).
Statistical Analyses
All statistical analyses were completed using R version 3.2.2 (R Core Team 2015), and R
Studio version 0.99.484 (R Studio Team 2015). For behavioural trials tadpoles that failed to
leave the acclimation chamber for the duration of the test were removed from the dataset, an a
priori decision. The average time spent in the stimulus arm and the control arm was compared
for each experimental group. Parametric assumptions were tested using the Shapiro-Wilk
normality test and Bartlett’s test of homoscedasticity on the paired differences of time spent in
Page 13 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 15
Draft
14
the stimulus and blank arms of the maze. Data meeting parametric assumptions were analyzed
with a paired t-test. Data not meeting parametric assumptions even after transformation were
analyzed with a paired Wilcoxon Rank Sum test. Power analyses and sample sizes were
calculated using the “effsize” package (Torchiano 2016). For behavioural experiments, the test
power of the behavioural test of the response of early prometamorphic tadpoles to odorants was
calculated as 0.74.
For EOG experiments, parametric assumptions were tested using the Shapiro-Wilk’s
normality test and the Bartlett’s test for homogeneity of variances. When parametric
assumptions were not met, data were transformed with a log10 (1 + x) transformation to reclaim
assumptions. Therefore, the transformed corrected mean EOG responses were compared with an
independent-samples t-test. Where parametric assumptions could not be reclaimed after
transformation, the Wilcoxon Rank Sum test was used to compare mean EOG responses. Mean
differences were considered to be significant when p ≤ 0.05.
Data availability
Data are available as part of the Supplementary Materials submitted with the manuscript.1
Results
Early stage development of olfactory structures vs EOG response
The surface morphology of the PC olfactory epithelium in premetamorphic tadpoles
changed considerably as metamorphosis progressed, and the appearance of sensory structures
(cilia and olfactory knobs) generally coincided with the ability to obtain EOG responses to
Spirulina extract and L-alanine (Fig. 1). At TK V (n = 7), the olfactory epithelial cells had a
1 Supplementary Materials, Tables S1, S2, S3, and S4.
Page 14 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 16
Draft
15
polygonal shape and short microvillous protrusions were evident (Fig. 1A, B), however, EOG
responses were not detectable at this stage (Fig. 1C). By TK VI, the polygonal cells were domed
and continued to bear short microvilli (Fig. 1D). A few specimens at TK VI were further
developed, with cilia from non-sensory cells (SCs) and mucus covering the epithelium (Fig. 1E).
These more developed TK VI specimens also exhibited the appearance of common sensory
structures, olfactory knobs with ciliated projections, which were commonly observed in later
stages (insets in Fig. 1G). Note that olfactory knobs, which are ~2 µm in diameter (Fig. 1G
insets), should not be confused with mucus appearing as larger globules, which obscured
underlying structures (Fig. 1E inset). Overall, EOG responses were undetectable at TK VI (Fig.
1F), although one individual did exhibit variable EOG responses to Spirulina extract. At TK IX,
olfactory knobs with cilia (ciliated OSNs) were consistently observed on the olfactory epithelium
(Fig. 1G). At TK IX, EOG responses to Spirulina extract and L-alanine were detectable (Fig. 1I)
but were still inconsistent across individuals. By early prometamorphosis (TK XIII), the
olfactory epithelium was densely covered by long cilia and mucus (Fig. 1H), and consistent
responses to Spirulina extract and L-alanine were measurable by EOG (Fig. 1I). Overall, ciliated
projections became longer and more prominent with increasing developmental stage, until they
completely covered the entire epithelial surface.
<Fig. 1 placement>
Anosmic vs control behavioural response of premetamorphic tadpoles
Tadpole response to the 0.001 M amino acid mixture (equimolar glycine, L-alanine, and
L-serine) was measured using tadpoles with intact olfactory pits and tadpoles that were rendered
Page 15 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 17
Draft
16
anosmic. When olfactory pits were left intact, tadpoles avoided the amino acid mixture,
spending significantly more time in the blank arm than the stimulus arm of the linear trough-
style choice maze (V = 51.5; p = 0.017; Fig. 2A). Even when the data were constrained to
tadpoles of age less than TK VI, i.e. before the first OSNs appeared (Fig. 1), the tadpoles
(median TK IV, range TK I - V; mean ± SE of 4.44 ± 0.40 g) still significantly avoided the cue
relative to the blank (V = 33.5, p = 0.035; Fig. S12). Conversely, there was no significant
difference between time spent in the blank and stimulus arms when premetamorphic tadpoles
were rendered anosmic (t11 = 0.81, p = 0.43; Fig. 2B).
<Fig. 2 placement>
Olfactory ontogeny in prometamorphic and older tadpoles
Attraction responses to Spirulina extract from early prometamorphic tadpoles were
measured with a linear trough-style choice maze (Fig. 3). Early prometamorphic tadpoles spent
significantly more time in the stimulus arm than the blank arm of the maze (t8 = -2.9, p = 0.01;
Fig. 3A).
<Fig. 3 placement>
Attraction responses from late prometamorphic (TK XVII) and tadpoles approaching
metamorphic climax (TK XVIII-XX) were also measured with the linear trough-style choice
maze (Fig. 3). Then, the same individuals were tested with EOG (Fig. 4) and their olfactory
epithelium surface anatomy was observed via SEM (Fig. 5). TK XVII tadpoles spent 3-fold
2 See Supplementary Materials
Page 16 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 18
Draft
17
more time in the stimulus arm than in the blank arm of the linear trough-style choice maze,
however this difference was not statistically significant (W = 6, p = 0.19; Fig. 3B). Tadpoles
approaching metamorphic climax (TK XVIII-XX tadpoles) were also not attracted to Spirulina
and spent the same amount of time in the stimulus and blank arms of the maze (t8 = -0.37, p =
0.71; Fig. 3C). Conversely, late prometamorphic tadpoles exhibited significantly higher EOG
responses to Spirulina extract than metamorphic climax tadpoles. Spirulina extract EOG
responses were over two-fold higher from late prometamorphic tadpoles when compared to
tadpoles approaching metamorphic climax (t21 = -2.4, p = 0.02, Power = 0.45; Fig. 4A). Late
prometamorphic tadpole EOG responses to L-alanine were also two-fold higher when compared
to tadpoles approaching metamorphic climax; however, this difference was not significant (W =
43, p = 0.16; Fig. 4B). SEM observation of the olfactory epithelium showed no qualitative
change in the sensory structures between late prometamorphic and tadpoles approaching
metamorphic climax. In all specimens observed, and similar to tadpoles at the TK XIII stage
(Fig. 1H), long ciliated projections and abundant mucus densely covered the olfactory epithelium
(Fig. 5).
<Fig. 4 placement>
<Fig. 5 placement>
Discussion
An ontogenetic shift in the olfactory system of L. catesbeiana tadpoles is evident on a
cellular and functional level throughout larval development. The results of the present study
demonstrate that discrete ontogenetic changes occur at specific stages of the larval period, as
Page 17 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 19
Draft
18
observed with EOG, behaviour, and scanning electron micrographs. To our knowledge, this is
the first study to compare olfactory responses between developmental stages in L. catesbeiana
with behaviour and EOG. Additionally, the surface anatomy of L. catesbeiana olfactory
epithelium has not previously been studied with SEM.
Early stage structural and functional development of the olfactory epithelium
The types of sensory structures observed here in bullfrog tadpoles are similar to those
observed in previous studies, which have mainly focused on X. laevis (Klein and Graziadei 1983;
Hansen et al. 1998). Polygonal-shaped cells observed in TK V – VI (Fig. 1A, B, D) in the
present study are similar to those Hansen et al. (1998) observed in the early developing PC of X.
laevis. We observed cilia and microvilli in tadpoles at all TK stages (Figs 1, 5), which were also
previously described by Hansen et al. (1998). In general, the cilia and microvilli observed in our
study could be projections from OSNs or SCs, since both of these cell types have been reported
in the larval PC in other species including X. laevis, Rana japonica Boulenger 1879, and
Ascaphus truei Stejneger 1899 (Taniguchi et al. 1996; Hansen et al. 1998; Benzekri and Reiss
2012). Although we did specifically identify OSNs (Fig. 1) that are required for an olfactory
response, additional histological investigation is warranted due to the presence of mucus that
obscured epithelial surface structures.
Previous studies on X. laevis have shown that not all cell types of the mature olfactory
epithelium are present at early developmental stages, but rather cells differentiate gradually
throughout development. For example, Klein and Graziadei (1983) and Hansen et al. (1998)
observed that microvilli develop before cilia in the PC, but both are present by Nieuwkoop and
Faber (NF) 39 (Nieuwkoop and Faber 1956), which precedes TK I. A similar progression of
Page 18 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 20
Draft
19
tissue development was also observed in R. japonica tadpoles (Taniguchi et al. 1996). Here, the
lack of detectable EOG olfactory responses to olfactory stimuli in early stage premetamorphic
tadpoles was seemingly due to the underdeveloped sensory structures in the olfactory epithelium.
The results (Fig. 1) clearly show that there is a close relationship between EOG responses to
olfactory stimuli and the cellular structures present on the olfactory epithelium, including cilia
and olfactory knobs. Although a few specimens at TK VI and all specimens at TK IX had
developed olfactory knobs and sparse cilia, EOG responses were variable (Fig. 1D, E, F, G, I).
Comparatively, at TK XIII, the dense cover of cilia over the olfactory epithelium resulted in
consistent EOG responses (Fig. 1H, I). However, the lack of EOG responses to olfactory stimuli
at developmental stages earlier than TK VI in our study brings into question when exactly the
olfactory epithelium begins to function.
For X. laevis, Hansen et al. (1998) suggests the olfactory system begins functioning at
approximately NF 45 (which precedes TK I), when OSNs first become functional and the
tadpole begins feeding. Given the various staging regimes used, it can be difficult to compare
both the developmental stage (McDiarmid and Altig 1999) and the characteristics that might be
unique to specific stages across species. Even considering these discrepancies, the tadpoles used
in the present study were more developed than those in Hansen et al. (1998). Therefore, the
olfactory epithelium of tadpoles in the present study should have been functioning.
However, the overall lack of EOG responses at early stages (≤ TK VI) in L. catesbeiana
tadpoles in the present study suggested that the olfactory epithelium was not yet functional.
Nevertheless, when tested in a linear trough-style choice maze, TK I – VIII tadpoles with their
olfactory senses intact avoided the olfactory stimulus (Fig. 2A and Fig. S13), but those rendered
anosmic (by occluding their entire nares with glue) did not exhibit the same avoidance response
3 Ibid
Page 19 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 21
Draft
20
to the olfactory stimulus (Fig. 2B and Fig. S14). Therefore, intact premetamorphic L.
catesbeiana tadpoles of age less than TK VI must have had an alternate, functional
chemosensory system in their nasal cavity and were using it to respond to cues since there was
no visual evidence of developed OSNs until tadpoles were at least of stage TK VI (Fig. 1). One
possibility is that there are solitary chemosensory cells (SCCs) present in the epithelial lining of
the nasal cavity of premetamorphic L. catesbeiana tadpoles. SCCs are sensory cells embedded
in epithelial tissues of vertebrates, including in the skin of tadpoles of other ranid species (e.g.
Rana temporaria L.), and in the nasal epithelium of American alligators (Alligator
mississippiensis Daudin, 1802), that are not associated with the olfactory nerve but are instead
linked to fibers of the trigeminal nerve (Whitear 1976; Kotrschal 1991; Hansen 2007; and
references therein). Like OSNs, SCCs can respond to chemosensory cues. When present, SCCs
sparsely populate epithelial tissue relative to other cells and are difficult to positively identify
using light microscopy or SEM without validation by techniques such as transmission electron
microscopy (TEM) and immunocytochemistry (Hansen 2007).
Moreover, the aforementioned olfactory-driven behavioural response in premetamorphic
tadpoles also indicates that early stage olfactory sensory structures are not developed enough to
generate a detectable EOG response and second, EOG analysis is not sensitive enough to record
olfactory responses at early developmental stages. It is plausible that a low density of olfactory
sensory structures at these early developmental stages, as observed in X. laevis larvae (Hansen et
al. 1998), resulted in the absence of EOG response in our study. Finally, the broad range of
anatomical development and EOG responses observed in TK VI individuals is likely due to
natural biological variation. This variation may be a sign of precocious metamorphic
development and suggests that there is limited fidelity between external markers of tadpole age
4 Ibid
Page 20 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 22
Draft
21
(i.e. small changes in limb development; Taylor and Kollros 1946) and timing of olfactory
remodelling, particularly around TK VI.
Shift in olfactory function in older tadpoles
Further ontogenetic changes in olfactory function were observed in L. catesbeiana
tadpoles in later developmental stages. An ontogenetic decrease in olfactory response, as
measured by EOG, was observed as tadpoles progress through prometamorphosis to
metamorphic climax (Fig. 4). Similarly, we observed a decrease in behavioural preference for
Spirulina between early prometamorphic tadpoles and older tadpoles closer to metamorphic
climax (Fig. 3). Previous studies have demonstrated that during the transition of the PC
olfactory epithelium, from the larval to the adult form, all OSNs suffer apoptosis and are
replaced (Higgs and Burd 2001; Dittrich et al. 2016). In X. laevis, the highest rate of OSN
apoptosis coincides with the maximum concentrations of plasma THs during metamorphic
climax (Dittrich et al. 2016). In L. catesbeiana, THs increase to their maximal concentrations
after TK XVII (reviewed in White and Nicoll 1981). Therefore, the reduced olfactory responses
to Spirulina extract in the older, TK XVIII-XX tadpoles when compared to younger,
prometamorphic tadpoles may be attributed to the widespread apoptosis of OSNs in the larval PC
olfactory epithelium. Alternately, metamorphic climax is also the period where the PC prepares
to detect airborne odorants (Taniguchi et al. 1996; Hansen et al. 1998). Therefore, it is also
possible that OSNs in the tadpole PC are not tuned to detect aqueous Spirulina extract and L-
alanine stimuli at TK stage XVIII-XX. However, qualitative ontogenetic changes in the cellular
structure of the olfactory epithelium were not observed in the PC between prometamorphic L.
catesbeiana tadpoles and those approaching metamorphic climax. Mucus sometimes limited our
Page 21 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 23
Draft
22
surface investigation of the ontogeny of the olfactory epithelium (Fig. 1). Future use of TEM
and histology would allow for a more detailed picture of the ontogenetic changes in the olfactory
epithelium, including cell turnover.
Interestingly, there was a disparity between stage-specific responses to Spirulina extract
when considering behavioural and EOG measurements in late prometamorphic (TK XVII)
tadpoles and those approaching metamorphic climax (TK XVIII-XX). Measured EOG responses
represent the bulk depolarization of OSNs to which odour molecules have bound. Action
potentials are generated when OSNs depolarize to the threshold potential, then the signal is
propagated to the brain where it is processed and an appropriate behavioural response is returned
(Baldwin and Scholz 2005). However, EOG does not indicate if the depolarization leads to
propagation of an action potential to the brain. Instead, it only measures a bulk influx of cations
from the extracellular environment associated with neural depolarization. Regardless, early
prometamorphic tadpoles exhibited a significant attraction response to Spirulina extract (Fig.
3A), but this behavioural response was considerably more variable in late prometamorphic
tadpoles and absent in tadpoles approaching metamorphic climax (Fig. 3B, C). Our results
highlight an interesting correlation between the known cessation of feeding at metamorphic
climax (Hourdry et al. 1996), and an ontogenetic change in olfactory response.
In summary, the results of the present study demonstrate the ontogenetic changes in the
olfactory system in L. catesbeiana tadpoles as they progress through to metamorphic climax. It
is evident that the olfactory structures on the PC olfactory epithelium develop gradually. At
early stages of premetamorphosis, sensory structures are underdeveloped, and therefore EOG is
unsuccessful. In addition, tadpoles approaching metamorphic climax also exhibit olfactory-
based ontogenetic changes, with a decreased olfactory response when compared to early and late
Page 22 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 24
Draft
23
prometamorphic tadpoles. This decrease in olfactory response may be related to other changes
co-occurring during metamorphosis and should be further investigated. Our results also showed
differences in the timing of olfactory development and initiation of function in L. catesbeiana as
compared to previous work on X. laevis, highlighting the importance of assessing olfactory
ontogeny and function in a diversity of anuran species.
Acknowledgements
Thank you to Doug Bray for aiding with preparation of specimens for SEM and to Grant
Duke for the use of the SEM at the Lethbridge Research and Development Centre, Lethbridge,
AB. This work was financially supported by the Natural Sciences and Engineering Research
Council [grant number STPGP 447250] and a Campus Alberta Innovation Program (CAIP)
Chair in Aquatic Health to G.G.P.
References
Altig, R., Whiles, M.R., and Taylor, C.L. 2007. What do tadpoles really eat? Assessing the
trophic status of an understudied and imperiled group of consumers in freshwater habitats.
Freshw. Biol. 52: 386-395.
Baldwin, D.H., and Scholz, N.L. 2005. The electro-olfactogram: An in vivo measure of
peripheral olfactory function and sublethal neurotoxicity in fish. In Techniques in Aquatic
Toxicology. Edited by G.K. Ostrander. CRC Press/Taylor & Francis Group, New York. pp.
257-276.
Benzekri, N.A., and Reiss, J.O. 2012. Olfactory metamorphosis in the coastal tailed frog
Ascaphus truei (Amphibia, Anura, Leiopelmatidae). J. Morphol. 273: 68-87.
Page 23 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 25
Draft
24
Brown, D.D., and Cai, L. 2007. Amphibian metamorphosis. Dev. Biol. 306: 20-33.
Dew, W.A., Azizishirazi, A., and Pyle, G.G. 2014. Contaminant-specific targeting of olfactory
sensory neuron classes: Connecting neuron class impairment with behavioural deficits.
Chemosphere 112: 519-25.
Dittrich, K., Kuttler, J., Hassenklöver, T., and Manzini, I. 2016. Metamorphic remodeling of the
olfactory organ of the African clawed frog, Xenopus laevis. J. Comp. Neurol. 524: 886-98.
Døving, K.B., and Trotier, D. 1998. Structure and function of the vomeronasal organ. J. Exp.
Biol. 201: 2913-2925.
Gascuel, J., and Amano, T. 2013. Exotic models may offer unique opportunities to decipher
specific scientific question: the case of Xenopus olfactory system. Anat. Rec. 296: 1453-
1461.
Getchell, T.V. 1974. Unitary responses in frog olfactory epithelium to sterically related
molecules at low concentrations. J. Gen. Physiol. 64: 241-261.
Getchell, M.L., and Getchell, T.V. 1992. Fine structural aspects of secretion and extrinsic
innervation in the olfactory mucosa. Microsc. Res. Techniq. 23: 111-127.
Gilbert, L.I., Tata, J.R., and Atkinson, B.G. (Editors) 1996. Metamorphosis: Postembryonic
reprogramming of gene expression in amphibian and insect cells. Academic Press, San
Diego.
Green, W.W., Mirza, R.S., Wood, C.M., and Pyle, G.G. 2010. Copper binding dynamics and
olfactory impairment in fathead minnows (Pimephales promelas). Environ. Sci. Technol. 44:
1431-7.
Page 24 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 26
Draft
25
Hansen, A. 2007. Olfactory and solitary chemosensory cells: Two different chemosensory
systems in the nasal cavity of the American alligator, Alligator mississippiensis. BMC
Neurosci. 8: 64.
Hansen, A., Reiss, J.O., Gentry, C.L., and Burd, G.D. 1998. Ultrastructure of the olfactory organ
in the clawed frog, Xenopus laevis, during larval development and metamorphosis. J. Comp.
Neurol. 398: 273-88.
Hassenklöver, T., Pallesen, L.P., Schild, D., and Manzini, I. 2012. Amino acid- vs. peptide-
odorants: responses of individual olfactory receptor neurons in an aquatic species. PLoS One.
7(12), e53097. doi:10.1371/journal.pone.0053097
Heerema, J.L., Helbing, C.C., and Pyle, G.G. 2018a. Use of electro-olfactography to measure
olfactory acuity in the North American bullfrog (Lithobates (Rana) catesbeiana) tadpole.
Ecotoxicol. Environ. Saf. 147: 643-647.
Heerema, J.L., Jackman, K.W., Miliano, R.C., Li, L., Zaborniak, T.S.M., Veldhoen, N., et al.
2018b. Behavioral and molecular analyses of olfaction-mediated avoidance responses of
Rana (Lithobates) catesbeiana tadpoles: Sensitivity to thyroid hormones, estrogen, and
treated municipal wastewater effluent. Horm. Behav. 101: 85-93.
doi:10.1016/j.yhbeh.2017.09.016
Higgs, D.M., and Burd, G.D. 2001. Neuronal turnover in the Xenopus laevis olfactory epithelium
during metamorphosis. J. Comp. Neurol. 433: 124-130.
Hourdry, J., L'hermite, A., and Ferrand, R. 1996. Changes in the digestive tract and feeding
behavior of anuran amphibians during metamorphosis. Physiol. Zool. 69: 219-251.
Ishizuya-Oka, A., Hasebe, T., and Shi, Y.B. 2010. Apoptosis in amphibian organs during
metamorphosis. Apoptosis. 15: 350-64.
Page 25 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 27
Draft
26
Karnovsky, M.J. 1965. A formaldehyde-glutaraldehyde fixative of high osmolarity for use in
electron microscopy. J. Cell Biol. 27: 1A-149A.
Klein, S.L., and Graziadei, P.P.C. 1983. The differentiation of the olfactory placode in Xenopus
laevis - A light and electron-microscope study. J. Comp. Neurol. 217: 17-30.
Kotrschal, K. 1991. Solitary chemosensory cells – taste, common chemical sense or what? Rev.
Fish Biol. Fish. 1: 3-22.
Leivas, P.T., Leivas, F.W.T., and Moura, M.O. 2012. Diet and trophic niche of Lithobates
catesbeiana (Amphibia: Anura). Zoologia. 29: 405-412.
Liu, X., Luo, Y., Chen, J., Guo, Y., Bai, C., and Li, Y. 2015. Diet and prey selection of the
invasive American bullfrog (Lithobates catesbeiana) in Southwest China. Asian Herpetol.
Res. 6: 34-44.
Lucero, M.T. 2013. Peripheral modulation of smell: Fact or fiction? Semin. Cell Dev. Biol.
24(1): 58–70. doi:10.1016/j.semcdb.2012.09.001
Manzini, I., Peters, F., and Schild, D. 2002. Odorant responses of Xenopus laevis tadpole
olfactory neurons: A comparison between preparations. J. Neurosci. Methods. 121: 159-167.
Manzini, I., and Schild, D. 2010. Olfactory coding in larvae of the African clawed frog Xenopus
laevis. In The Neurobiology of Olfaction. Edited by A. Menini. CRC Press/Taylor & Francis,
Boca Raton. pp. 113-129.
Mcdiarmid, R.W., and Altig, R. 1999. Introduction - The tadpole arena. In Tadpoles the biology
of anuran larvae. Edited by R.W. McDiarmid and R. Altig.University of Chicago Press,
Chicago. pp. 1-6.
Mezler, M., Konzelmann, S., Freitag, J., Rossler, P., and Breer, H. 1999. Expression of olfactory
receptors during development in Xenopus laevis. J. Exp. Biol. 202: 365-76.
Page 26 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 28
Draft
27
Mirza, R.S., Ferrari, M.C.O., Kiesecker, J.M., and Chivers, D.P. 2007. Responses of American
toad tadpoles to predation cues: Behavioural response thresholds, threat-sensitivity and
acquired predation recognition. Behaviour. 143: 877-889.
Nieuwkoop, P.D., and Faber, J. 1956. Normal table of Xenopus laevis (Daudin): A systematical
and chronological survey of the development from the fertilized egg till the end of
metamorphosis. North Holland Publishing Company, Amsterdam.
R Core Team. 2015. R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. https://www.R-project.org/ [accessed 23 July 2016].
RStudio Team. 2015. RStudio: Integrated development for R. RStudio Inc., , Boston, MA, USA.
http://www.rstudio.com/ [accessed 23 July 2016].
Sachse, S., and Galizia, C.G.. 2002. Role of inhibition for temporal and spatial odor
representation in olfactory output neurons: A calcium imaging study. J. Neurophysiol. 87:
1106-1117.
Schwob, J.E. 2002. Neural regeneration and the peripheral olfactory system. Anat. Rec. 269: 33-
49.
Scott, J.W., and Scott-Johnson, P.E. 2002. The electroolfactogram: A review of its history and
uses. Micros. Res. Techniq. 58: 152-160.
Smith, G.R., Boyd, A., Dayer, C.B., and Winter, K.E. 2008. Behavioral responses of American
toad and bullfrog tadpoles to the presence of cues from the invasive fish, Gambusia affinis.
Biol. Invasions. 10: 743-748.
Syed, A.S., Sansone, A., Hassenklöver, T., Manzini, I., and Korsching, S.I. 2017. Coordinated
shift of olfactory amino acid responses and V2R expression to an amphibian water nose
during metamorphosis. Cell. Mol. Life Sci. 74(9): 1711–1719.
Page 27 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 29
Draft
28
Takahara, T., Kohmatsu, Y., Maruyama, A., Doi, H., Yamanaka, H., and Yamaoka, R. 2012.
Inducible defense behavior of an anuran tadpole: Cue-detection range and cue types used
against predator. Behav. Ecol. 23: 863-868.
Taniguchi, K., Toshima, Y., Saito, T.R., and Taniguchi, K. 1996. Development of the olfactory
epithelium and vomeronasal organ in the Japanese reddish frog, Rana japonica. J. Vet. Med.
Sci. 58: 7-15.
Tata, J.R. 2006. Amphibian metamorphosis as a model for the developmental actions of thyroid
hormone. Mol. Cell. Endocrinol. 246: 10-20.
Taylor, A.C., and Kollros, J.J. 1946. Stages in the normal development of Rana pipens larvae.
Anat. Rec. 94: 7-24.
Torchiaano, M. 2016. Effsize: Efficient effect size computation. R package version 0.6.4.
https://CRAN.R-project.org/package=effsize [accessed 23 July 2016].
Trakimas, G., Jardine, T., Barisevičiūtė, R., Garbaras, A., Skipitytė, R., and Remeikis, V. 2011.
Ontogenetic dietary shifts in European common frog (Rana temporaria) revealed by stable
isotopes. Hydrobiologia. 675: 87-95.
Wang, H., Zhao, H., Tai, F., and Zhang, Y. 2008. Postembryonic development of the olfactory
and vomeronasal organs in the frog Rana chensinensis. Zoolog. Sci. 25: 503-508.
White, B.A., and Nicoll, C.S. 1981. Hormonal control of amphibian metamorphosis. In
Metamorphosis: A problem in developmental biology. Edited by L.I. Gilbert and E. Frieden.
Plenum Publishing, New York. pp. 636-696.
Whitear, M. 1976. Identification of the epidermal “Stiftchenzellen” of frog tadpoles by electron
microscopy. Cell Tissue Res. 175: 391–402.
Page 28 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 30
Draft
29
Tables
Table 1. Classification of Taylor and Kollros (TK; 1946) stages used to assign North American
bullfrog tadpoles (Lithobates catesbeiana Shaw 1802) into the developmental groups used in the
present study.
TK Stages Developmental Group
I - X Premetamorphic
XI - XV Early prometamorphic
XVII Late prometamorphic
XVIII - XX Approaching metamorphic climax
Page 29 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 31
Draft
30
Figure captions
Fig. 1. Representative scanning electron micrographs, with insets of higher magnification
showing key features, of the principle cavity olfactory epithelium of North American bullfrog
(Lithobates catesbeiana Shaw 1802) tadpoles and corresponding EOG responses to Spirulina
spp. Turpin Ex Gomont 1893 (algae) extract. A – B: Developmental stage TK V. Polygonal
cells (asterisks) are evident in the olfactory epithelium, and some appear domed. Short
microvilli (arrows) are also present. Insets show a higher magnification view of microvilli (A)
and polygonal cells (B). C: TK V. EOG responses to Spirulina extract are undetectable. D: TK
VI. Polygonal cells (asterisks) are evident with microvillous projections (inset). E: TK VI. In a
few individuals, extensive cilia originating from non-sensory cells (arrows and left inset bracket)
and mucus (asterisks and right inset) cover the olfactory epithelium. Olfactory knobs with
ciliated projections (see insets in G for higher magnification exemplars) are present in only a few
individuals. F: Lack of EOG response to Spirulina extract at TK VI. One TK VI individual did,
however, exhibit a variable EOG response to Spirulina extract. G: TK IX. Cilia (arrows) and
mucus (black asterisks) cover the olfactory epithelium. Olfactory knobs with ciliated projections
(insets; white asterisks) are present. H: TK XIII. Relative to earlier stages, the entire surface of
the olfactory epithelium is more densely covered with cilia (arrows and inset) and mucus
(asterisks) obscuring olfactory knobs. I: Beginning at TK IX, EOG measurements of responses
to Spirulina extract are successful. Scale bars: A, D, E, G = 30 μm, B = 40 μm, H = 50 μm. Inset
scale bars: A, D, and H = 7.5 μm, B = 10 μm, E = 5 µm (both), G = 4 μm (both). Dashed line in
C, F, and I indicate stimulus delivery. Faint black horizontal line in G is an artefact of the SEM
scan.
Page 30 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 32
Draft
31
Fig. 2. Time premetamorphic North American bullfrog (Lithobates catesbeiana Shaw 1802)
tadpoles (TK range I – VIII) spent in the blank (dechlorinated water) and stimulus (amino acid
mixture) arms of the linear trough-style choice maze with or without anosmia induced (means ±
SE). A: Response when tadpole olfactory pits were left intact (n = 10; paired Wilcoxon rank
sum t-test, p = 0.017. B: Response when tadpoles were rendered anosmic (n = 12; paired t-test, p
= 0.43). An asterisk above a bar denotes a significant difference (p ≤ 0.05).
Fig. 3. Time spent by North American bullfrog (Lithobates catesbeiana Shaw 1802) tadpoles
undergoing prometamorphosis and those approaching metamorphic climax in the blank
(dechlorinated water) and stimulus (Spirulina spp. Turpin Ex Gomont 1893 extract) arms of the
choice maze (means ± SE). A: Early prometamorphic tadpole response (TK XI – XII, n = 9;
paired t-test, p = 0.01). B: Late prometamorphic tadpole response (TK XVII, n = 7; paired
Wilcoxon Rank Sum test, p = 0.19). C: Response of tadpoles approaching metamorphic climax
(TK XVIII – XX, n = 9; paired t-test, p = 0.71). An asterisk above a bar denotes a significant
difference (p ≤ 0.05).
Fig. 4. Raw EOG responses (EOG traces) and blank-corrected EOG responses of North
American bullfrog (Lithobates catesbeiana Shaw 1802) prometamorphic tadpoles
(prometamorphic, n = 11) and tadpoles approaching metamorphic climax (metamorphic, n = 12)
(means ± SE). A: Response to Spirulina spp. Turpin Ex Gomont 1893 extract (independent
samples t-test, p = 0.02). B: Response to L-alanine (Wilcoxon Rank Sum test, p = 0.16). An
asterisk denotes a significant difference (p ≤ 0.05).
Page 31 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 33
Draft
32
Fig. 5. Scanning electron micrographs of the principle cavity olfactory epithelium of North
American bullfrog (Lithobates catesbeiana Shaw 1802) tadpoles in (A) late prometamorphosis
(TK XVII) and (B) approaching metamorphic climax (TK XVIII – XX). At both stages, dense
cilia (arrows) and mucus (asterisks) are present in abundance, forming a dense matrix over the
olfactory knobs as in TK XIII tadpoles (see Fig. 1H). Scale bars: A = 50 μm, B = 40 μm.
Page 32 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 34
Draft
Tables
Table 1. Classification of Taylor and Kollros (TK; 1946) stages used to assign North American
bullfrog tadpoles (Lithobates catesbeiana Shaw 1802) into the developmental groups used in the
present study.
TK Stages Developmental Group
I - X Premetamorphic
XI - XV Early prometamorphic
XVII Late prometamorphic
XVIII - XX Approaching metamorphic climax
Page 33 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 35
Draft
Page 34 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 36
Draft
Fig. 1. Representative scanning electron micrographs, with insets of higher magnification showing key features, of the principle cavity olfactory epithelium of North American bullfrog (Lithobates catesbeiana
Shaw 1802) tadpoles and corresponding EOG responses to Spirulina spp. Turpin Ex Gomont 1893 (algae) extract. A – B: Developmental stage TK V. Polygonal cells (asterisks) are evident in the olfactory
epithelium, and some appear domed. Short microvilli (arrows) are also present. Insets show a higher magnification view of microvilli (A) and polygonal cells (B). C: TK V. EOG responses to Spirulina extract are undetectable. D: TK VI. Polygonal cells (asterisks) are evident with microvillous projections (inset). E: TK VI. In a few individuals, extensive cilia originating from non-sensory cells (arrows and left inset bracket)
and mucus (asterisks and right inset) cover the olfactory epithelium. Olfactory knobs with ciliated projections (see insets in G for higher magnification exemplars) are present in only a few individuals. F: Lack of EOG response to Spirulina extract at TK VI. One TK VI individual did, however, exhibit a variable
EOG response to Spirulina extract. G: TK IX. Cilia (arrows) and mucus (black asterisks) cover the olfactory epithelium. Olfactory knobs with ciliated projections (insets; white asterisks) are present. H: TK XIII.
Relative to earlier stages, the entire surface of the olfactory epithelium is more densely covered with cilia (arrows and inset) and mucus (asterisks) obscuring olfactory knobs. I: Beginning at TK IX, EOG
measurements of responses to Spirulina extract are successful. Scale bars: A, D, E, G = 30 μm, B = 40 μm, H = 50 μm. Inset scale bars: A, D, and H = 7.5 μm, B = 10 μm, E = 5 µm (both), G = 4 μm (both). Dashed line in C, F, and I indicate stimulus delivery. Faint black horizontal line in G is an artefact of the SEM scan.
Page 35 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 37
Draft
Fig. 2. Time premetamorphic North American bullfrog (Lithobates catesbeiana Shaw 1802) tadpoles (TK range I – VIII) spent in the blank (dechlorinated water) and stimulus (amino acid mixture) arms of the
linear trough-style choice maze with or without anosmia induced (means ± SE). A: Response when tadpole olfactory pits were left intact (n = 10; paired Wilcoxon rank sum t-test, p = 0.017. B: Response when tadpoles were rendered anosmic (n = 12; paired t-test, p = 0.43). An asterisk above a bar denotes a
significant difference (p ≤ 0.05).
Page 36 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 38
Draft
Fig. 3. Time spent by North American bullfrog (Lithobates catesbeiana Shaw 1802) tadpoles undergoing prometamorphosis and those approaching metamorphic climax in the blank (dechlorinated water) and
stimulus (Spirulina spp. Turpin Ex Gomont 1893 extract) arms of the choice maze (means ± SE). A: Early prometamorphic tadpole response (TK XI – XII, n = 9; paired t-test, p = 0.01). B: Late prometamorphic tadpole response (TK XVII, n = 7; paired Wilcoxon Rank Sum test, p = 0.19). C: Response of tadpoles
approaching metamorphic climax (TK XVIII – XX, n = 9; paired t-test, p = 0.71). An asterisk above a bar denotes a significant difference (p ≤ 0.05).
Page 37 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 39
Draft
Fig. 4. Raw EOG responses (EOG traces) and blank-corrected EOG responses of North American bullfrog (Lithobates catesbeiana Shaw 1802) prometamorphic tadpoles (prometamorphic, n = 11) and tadpoles approaching metamorphic climax (metamorphic, n = 12) (means ± SE). A: Response to Spirulina spp.
Turpin Ex Gomont 1893 extract (independent samples t-test, p = 0.02). B: Response to L-alanine (Wilcoxon Rank Sum test, p = 0.16). An asterisk denotes a significant difference (p ≤ 0.05).
Page 38 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology
Page 40
Draft
Fig. 5. Scanning electron micrographs of the principle cavity olfactory epithelium of North American bullfrog (Lithobates catesbeiana Shaw 1802) tadpoles in (A) late prometamorphosis (TK XVII) and (B) approaching
metamorphic climax (TK XVIII – XX). At both stages, dense cilia (arrows) and mucus (asterisks) are present in abundance, forming a dense matrix over the olfactory knobs as in TK XIII tadpoles (see Fig. 1H).
Scale bars: A = 50 μm, B = 40 μm.
Page 39 of 39
https://mc06.manuscriptcentral.com/cjz-pubs
Canadian Journal of Zoology