ABSTRACT
Testudines possess a rigid shell that influences the mechanics of the respiratory system. We studied respiratory mechanics in the terrestrial red-footed tortoise Chelonoidis carbonarius (Cryptodira), comparing juvenile individuals with a less ossified and more flexible carapace with adults with a well-ossified rigid shell. Combined with these ontogenetic differences, we analyzed respiratory system mechanics with animals in a supine and a prone position, as well as in the isolated lungs, to evaluate the impact of the viscera on breathing mechanics. To do so, we used established protocols to measure pulmonary volume (i.e. resting, VLr; and maximum, VLm), static (Cstat) and dynamic (Cdyn) compliance, and the work of breathing (W). We observed that isolated lungs displayed increased VLr, VLm, Cstat and Cdyn and decreased W. Additionally, pulmonary volume, compliance and W were affected by evaluated position, such as a smaller VLr in a supine position. Cdyn and W showed a volume dependency while frequency had less influence on these variables. At similar levels of ventilation, juveniles showed a lower W than adults when standardized by body mass, but similar W when standardized by VLr. Clear ontogenetic changes could be observed in breathing mechanics between juvenile and adult C. carbonarius. While these differences might largely be explained by variation in shell ossification, other explanations such as differences in visceral proportions or developmental degree of the post-pulmonary septum should also be taken into account.
INTRODUCTION
Among reptiles, Testudines show unique anatomical and morphological features that make them interesting models for studying breathing mechanics. In addition to multichambered lungs (Perry, 1998), Testudines present ribs fused with the spine and loss of intercostal muscles, as well as the presence of a bony plastron (Lyson et al., 2014). Consequently, Testudines present a rigid shell that impedes lung ventilation by the mechanisms seen in other amniotes and, thus, fundamentally alters the mechanics of the respiratory system (Jackson, 1971; Vitalis and Milsom, 1986a). Turtles, tortoises and terrapins use abdominal muscles to promote changes in intrapulmonary pressure during ventilatory activities (Gans and Hughes, 1967; Gaunt and Gans, 1969; Jackson, 1971; Lambertz, et al., 2010; Lyson et al., 2014). Additionally, their multichambered lungs are connected to the dorsal shell through mesopneumonia or may be fused directly to the carapace, while ventral mesopneumonia, or an associated structure named the post-pulmonary septum (PPS; Duncker, 1978; Perry, 1998; Perry and Sander, 2004), connect the lungs to the ventrally located viscera (Broman, 1904, 1937; Lambertz et al., 2010). Similar to all the anatomical and morphological features, the presence of the PPS in the red-eared slider Trachemys scripta has recently been demonstrated to significantly influence lung mechanics (Souza and Klein, 2021).
In terms of lung ventilation, Testudines, like other reptiles, show an intermittent ventilatory pattern, by which single breaths or burst of breaths are separated by non-ventilatory periods of varied duration (Glass and Wood, 1983; Vitalis and Milsom, 1986a). Ultimately, understanding the breathing pattern implies observing the relationship between ventilation, gas exchange and ventilatory mechanics (Pages et al., 1990). Importantly, ventilatory mechanics plays a crucial role in lung ventilation, influencing breathing frequency and tidal volume, as well as the cost to ventilate the lungs (Perry and Duncker, 1980; Milsom and Vitalis, 1984; Vitalis and Milsom, 1986a,b; Pages et al., 1990; Perry, 1998; Reichert et al., 2019).
To our knowledge, the mechanical properties of ventilation in Testudines have mostly been investigated in aquatic or semi-aquatic species, including T. scripta (Jackson, 1971; Vitalis and Milsom, 1986a,b; Lee and Milsom, 2016; Souza and Klein, 2021), Caretta caretta (Lutcavage et al., 1989) and Mauremys caspica (Pages et al., 1990). These studies analyzed lung volume, static and dynamic compliance and/or the work of breathing. Jackson (1971) indicated that the unique pulmonary mechanics of T. scripta enables them to adjust pulmonary volume and maintain constant pressure with little or no muscular effort (low cost of breathing). Lutcavage et al. (1989) and Williams et al. (2021) studied static compliance and lung volume in prone and supine positions in non-terrestrial turtles and observed alterations in lung mechanics depending on the position of the animal. The present study is the first that aims to describe lung mechanics in the terrestrial red-footed tortoise Chelonoidis carbonarius, including both juvenile and adult individuals. A recent study compared mechanical properties of the respiratory system in juvenile and adult Caiman yacare and found differences in lung compliance, indicating that the morphological and anatomical changes during development lead to a reduction in body wall compliance (Reichert et al., 2019). Therefore, because of their less ossified shell (Oliveira et al., 2023), we hypothesize that juvenile C. carbonarius might also present greater respiratory system compliance compared with adult animals. To test our working hypothesis, we studied lung mechanics in both juvenile and adult C. carbonarius. Specifically, we describe and compare pulmonary volume, static and dynamic compliance and the work of breathing among age groups. Furthermore, we also investigated these variables in both supine and prone positions, as well as within the lungs alone, to understand the role of visceral pressure on lung mechanics. This latter point is based on the widespread practice to study respiratory system mechanics with animals in supine position. In Testudines, the lungs are placed dorsally within the body cavity, receiving the full visceral mass when supine, but not prone, a more natural position.
MATERIALS AND METHODS
Animals
Juveniles (n=4, mean±s.e.m. body mass 0.08±0.01 kg) and adults (n=10, 3.25±0.89 kg) Chelonoidis carbonarius (Spix 1824) were obtained from Jacarezário of the Universidade Estadual Paulista Júlio Mesquita Filho (UNESP) in Rio Claro (São Paulo, Brazil) during 2014 and 2015. The experimental animals were transferred to the local animal care facility at the Faculdade de Filosofia, Ciências e Letras de Ribeirão Preto (São Paulo, Brazil), where they were housed for 3 months until experiments were performed. All animals were housed under a 12 h:12 h light:dark cycle, in a controlled environment of 25°C. They were fed 3 times a week with fruit and vegetables and provided with water ad libitum. All experimental procedures were conducted under license SISBIO (35221-1 and 35221-8) and protocol CEUA-FFCLRP-Campus de Ribeirão Preto (12.1.1541.53.0 and 16.5.835.59.0).
Experimental setup and data analysis
The animals were euthanized through an intraperitoneal injection of thiopental (150 mg kg−1) and lidocaine (5 mg kg−1), and the trachea was then cannulated and connected using a three-way valve to a syringe and a pressure transducer (DELTRAN). The pressure transducer signal was amplified (AECAD 04P) and recorded using a data acquisition system (PowerLab 8/35), while data were analyzed using Labchart 7.0 software (AdInstruments).
To obtain the resting pulmonary volume (VLr) and maximum pulmonary volume (VLm), we followed the experimental protocol proposed by Perry and Duncker (1978) and Klein et al. (2003). Starting with the respiratory system equilibrated with atmospheric pressure, air was removed from the lungs in a stepwise manner using a syringe until reaching a pressure of −10 cmH2O, giving VLr. Afterwards, air was added in a stepwise manner until reaching a pressure of +20 cmH2O to obtain VLm. Lastly, air was removed again in steps to return to the initial state of inflation. This procedure enabled us to generate the volume–pressure curves both for inflation and deflation of the respiratory system. This experiment was performed in all the animals.
W was calculated as the area of the pressure–volume loop delimitated by the inflation curve, the y-axis, and by a horizontal line connecting the y-axis to the end-inflation point (Vitalis and Milsom, 1986b; York et al., 2017). The area was obtained by using the software ImageJ to determine the number of pixels in a given area and transformed into ml cmH2O by determining the number of pixels in an area given by 1 ml and 1 cmH2O. The relationship between the work and the frequency (min−1) was used to reach the minute work (in ml cmH2O min−1 kg−1). Following Souza and Klein (2021), linear regressions were employed, utilizing log10-transformed work values standardized by body mass (log10W/MB) against log10-transformed pump ventilation (log10VP), with data segregated based on varying pump frequencies (f) to identify isopleths. Vertical lines corresponding to pump ventilations of 1, 5, 10, 25, 50, 100, 200, 300, 400, 500, 750 and 1000 ml min−1 were depicted on the same graph. The anti-log of the work values for each of these lines intersecting the isopleths was used to calculate minute work (), which was then plotted against respiratory frequency, with data segregated according to different ventilation levels.
All described procedures were performed in three different conditions: animals in supine position, followed by prone position, and lastly with the viscera removed in order to expose the lungs, but with the PPS maintained intact. The exposed lungs were humidified with saline solution (0.9%) during the entire procedure. In supine and prone positions, care was taken that neither the limbs nor the head was retracted into the carapace.
Statistical analysis
Data analysis was performed with R 4.2.1 (http://www.R-project.org/), and for graphic construction we used the software GraphPad Prism 5.0 (https://www.graphpad.com/). The relationship between the variables in all tested conditions and between the groups was determined by applying a generalized model for repeated measures (Generalized Estimating Equations, GEE). Models with Gamma and Gaussian distribution were compared with quasi-information criterion (QIC) and correlation information criterion (CIC) tests, along with adjustment of the residuals of each model to a Gaussian Q–Q plot, to choose the best-fitting model. The relationship between Cstat and body size was plotted in a log–log relationship with MB to obtain an allometric equation. The pairwise comparison of slopes through linear regression was conducted in R to assess whether the slopes in different combinations exhibited significant differences. P-values ≤0.05 were considered significant.
RESULTS
Lung volume and static compliance
Pulmonary volume was significantly affected by the test conditions (supine, prone and isolated) as well as life stage (Fig. 1). VLr was significantly greater in the prone position in adults and juveniles. However, VLr in juveniles in isolated lungs was similar to that in the prone position, and VLr in adults in isolated lungs was similar to VLr in the supine position. Also, VLr in adults was significantly decreased in isolated lungs and in the supine position compared with that in juveniles. VLm was significantly augmented in isolated lungs when compared with that in supine and prone positions in adults and juveniles, and the lowest VLm was observed in adults in the prone position. In all three tested conditions, VLm was significantly smaller in adults when compared with juveniles. Analyzing the relationship VLr/VLm, we observed that VLr contributed significantly more to VLm in the prone position, being greatest in adults. Compared with juveniles, adults showed a significantly greater VLr/VLm in the prone position and a significantly lower VLr/VLm in the supine position. VLr/VLm was lowest in the isolated lungs.
Next, Cstat was standardized by MB, VLr and VLm (Fig. 2). We observed that Cstat/MB was not significantly different between supine and prone positions, but adults showed significantly lower Cstat in all tested conditions when compared with juveniles. Significant differences between prone and supine positions were found in Cstat/VLr for both juveniles and adults, while for Cstat/VLm, only adults presented a higher compliance in the prone position. For the intact system (prone and supine), juveniles, when compared with adults, showed significantly higher Cstat/VLm only in the supine condition. Cstat was greatest in the isolated lungs for all standardizations and significantly different between age groups when standardized by MB and VLm.
Plotting non-standardized Cstat against MB, VLr and VLm gave the intraspecific regressions for each tested condition (Fig. 3). Cstat showed strong and significant correlation with MB and pulmonary volume (Table 1).
Dynamic compliance
The position of the body for the intact respiratory system (supine and prone) showed a significant influence on Cdyn, with juveniles presenting more differences between supine and prone positions at greater volumes (3 and 4 ml). The greatest Cdyn for the intact respiratory system was observed when we ventilated juveniles at the highest volume (4 ml). The isolated lungs, however, showed the greatest Cdyn in both age groups.
Juveniles exhibited a more predictable pattern regarding Cdyn. We show in Fig. 4 that Cdyn decreased in all three conditions in a comparable manner with increasing ventilation frequency, while showing a tendency to increase with increasing pump volume at each given frequency. In adults, Cdyn was lower than in juveniles and showed a (volume-dependent) remarkable significant decrease between frequencies of 10 and 30 cycles min−1 (Fig. 4).
To better compare Cdyn between juveniles and adults, we decided to use pulmonary data standardized by MB. Thus, we chose pump volumes of 0.6 ml for juveniles and 25 ml for adults, as these volumes represent, respectively, ventilatory volumes of 7.5 and 7.7 ml kg−1. Plotting Cdyn in this way for the intact respiratory system, the isolated lungs and the values calculated for the body cavity against pump ventilation (Fig. 5) indicates the lungs are the most compliant component of these animals' respiratory system, showing a greater ventilation dependency in juveniles than in adults. Comparing juveniles with adults, lung compliance (CL) did not show statistically significant differences between the groups, but for the total system (CT) and the body cavity (CB), juveniles showed a higher compliance, especially when frequency increased.
Work of breathing
Fig. 6 shows that work per breath (W) was significantly impacted by lung volume, whereas frequency seemed to have less influence. However, it is important to note that an increase in frequency at the smallest lung volume (0.3 ml for juveniles, and 5 and 10 ml for adults) resulted in a slight decrease in W in both supine and prone positions for juveniles, while for adults it resulted in a slight increase in W. At the other volumes, W was virtually unaltered by increasing frequency. The isolated lungs showed significantly lower W when compared with supine and prone positions, for both juveniles and adults (Fig. 6).
We then analyzed minute work of breathing (Ẇ). In Fig. 7, we show that, in the intact system and for any given level of ventilation, an increase in frequency led to a decrease in Ẇ, while in the isolated lungs, Ẇ was more variable, especially in juveniles, decreasing with frequency at lower pump ventilation, but increasing with frequency at greater pump ventilations. Ẇ was lowest for the isolated lungs in both juveniles and adults, and in juveniles Ẇ was lower in the prone position when compared with the supine one. Position had no effect on Ẇ in adults.
DISCUSSION
While a chelonian in a supine position might not be a common sight, animals occasionally will end up in an inverted position and will need to self-right, which comes at an increased metabolic cost (Ewart et al., 2022; Golubović et al., 2015). Also, chelonians are frequently positioned in a supine position during veterinary practice, fundamentally altering breathing mechanics and consequently demanding adjustments in artificial ventilation to maintain an animal's acid–base and blood gas status (Williams et al., 2021). Overall, we observed that our tested positions (supine and prone) significantly influenced lung mechanics (i.e. VLr, VLm, Cstat, Cdyn) in both adults and juveniles of C. carbonarius. Consistently, lung mechanics in both the supine and prone position significantly differed from that in isolated lungs. As expected, isolated lungs displayed increased VLr, VLm, Cstat and Cdyn. Such results indicate an expressive change in respiratory mechanics caused by the limitation of the lungs within body cavity, similar to the observation by Jackson (1971) in the semi-aquatic turtle T. scripta.
Regarding VLr, our experiment showed that the prone position presents a significant increase in VLr when compared with the supine position. In T. scripta, Williams et al. (2021) found a tendency for increased VLr in the prone position with the respiratory system open to the atmosphere, but with no significant difference in the supine position. In C. carbonarius, these differences can be explained by the dorsally located lungs in Testudines being impacted by the mass of the viscera when in the supine position. Such an increased load affects breathing mechanics, but the effect may be reduced by the presence of a relatively complete PPS (Perry, 1978; Lambertz et al., 2010), mitigating lung compression (Souza and Klein, 2021; in T. scripta). In the prone position, in contrast, gravity will pull the visceral organs away from the lungs, and as the viscera are connected to the PPS (Broman, 1904, 1937), the PPS will be pulled downwards, expanding the space available for the lungs within the carapace, increasing VLr. However, a significant difference in VLm might not be expected, as the pressures applied during static lung inflation are great, displacing the viscera in similar manner in the two positions, as has also been shown by Williams et al. (2021) for T. scripta. Interestingly, VLr values reported by Williams et al. (2021) are about 3 times greater and the VLm values are 2.5 times greater in T. scripta than in C. carbonarius, suggesting a significant variation in lung volume among testudine species.
From a functional point of view, the experimentally obtained values for VLr in the supine position might not reflect the VLr in a living terrestrial testudine, as gravity would pull the viscera downwards, increasing VLr as described above. Data on mechanical properties of the respiratory system in terrestrial Testudines might, therefore, be better obtained with animals in the prone position to better reflect the normal expansion of the lungs within the carapace. When it comes to aquatic or semi-aquatic species, however, hydrostatic pressure will act upon the flexible parts of the body wall, pushing the viscera into the body cavity, thereby reducing lung volume (Gaunt and Gans, 1969) and consequently VLr, as we observed before (Trevizan-Baú, 2016). Additionally, buoyancy control can significantly alter the volume of the lungs in submerged turtles, as lung volume and bladder volume are intrinsically linked when turtles need to regulate their specific density to maintain a given position within the water column (Jacobs, 1939; 1941; Jackson, 1969, 1971; Peterson and Gomez, 2008). For species breathing while submerged, prone and supine positions might be analyzed to fully understand the mechanical properties of the respiratory system, or ideally might be measured while the animal is submerged (Trevizan-Baú, 2016; Souza and Klein, 2021).
Moreover, differences between juveniles and adults, especially in VLr/VLm, indicate that a relative difference in visceral mass in adults could exert more pressure on the lungs in a supine position and pull the lungs down more when in a prone position. If this is the case, visceral pressure would be at least partially responsible for the differences observed between juveniles and adults in the intact system. However, when the viscera were removed, juveniles exhibited greater lung compliance compared with adults. This suggests a potentially more distensible lung in the juvenile life stage. Consequently, other hypotheses could be related to these factors: (i) because of the lesser ossification of the carapace and by standardizing data by MB, juvenile compliance values were consistently greater when compared with those of adults; and (ii) ontogenetic differences in the lungs and/or PPS between juveniles and adults could also influence compliance. Either way, a more detailed investigation of the function of the PPS and/or changes during ontogenetic development in the present and in other testudine species seems warranted.
Following Pages et al. (1990), we also normalized our data by pulmonary volume (i.e. VLr and VLm), in addition to MB. These normalizations could avoid the effect of physiological or nutritional factors that may occur when normalizing by MB and furthermore explain whether differences in breathing mechanics might be related to differences in respiratory system volume or mechanical properties of the respiratory system (York et al., 2018). Our comparisons between Cstat standardized by pulmonary volume indicate fewer changes between juveniles and adults when compared with Cstat standardized by MB. In addition, a decreased slope in regressions when Cstat was standardized by pulmonary volume may indicate an influence of MB standardization on the results, possibly as a result of differences in shell ossification (Oliveira et al., 2023).
Differences in relative lung volume (VLr/VLm), with juveniles presenting a relatively larger lung volume than supine adults, could indicate a possible ontogenetic aspect that explains differences in lung static compliance. Similar to our observation, Reichert et al. (2019) found a higher Cstat in juveniles compared with adult C. yacare. Presumably, the stiffness of the body wall in adult caimans leads to a less compliant system. In C. carbonarius, juveniles possess a less ossified carapace compared with adults (Oliveira et al., 2023). Thus, this could influence the greater total compliance found in juveniles. However, our Cdyn results indicate a similarity in CL between the groups. We only conducted a one-volume comparison between juveniles and adults and only from 5 to 20 cycles min−1, so more Cdyn comparisons between proportional lung volumes must be done to fully understand a possible ontogenetic factor that could influence lung compliance.
The combinations of frequencies and volumes used by the animals can be a strategy they use to minimize the cost of breathing. In our experimental animals, we observed that volume influences both Cdyn and W more than frequency does. As depicted in Fig. 5, frequency increase exhibited greater stability in CT compared with a pronounced decrease in CL and with a moderated decrease in CB. For T. scripta, Vitalis and Milsom (1986b) found that, as frequency increases, the work required to inflate the lungs contributes more to the total work of the system. The authors indicated that both static and dynamic mechanics, especially at lower frequencies, are influenced more by the body wall, rather than the lungs, and the increase in lung work with increasing frequency results from an escalation in non-elastic forces required to overcome ventilation (Vitalis and Milsom, 1986b).
As reported in previous studies, the combination of frequency and volume plays a crucial role in meeting respiratory demands while incurring minimal breathing costs. Oliveira et al. (2023) observed, under normoxic conditions, an elevation in instantaneous breathing frequency (f′, breaths min−1) despite a constant tidal volume (VT, ml), when both juveniles and adults were exposed to an increase in temperature (15–35°C). Additionally, Trevizan-Baú et al. (2018) found that adults exhibited heightened ventilation characterized by increased volume and frequency when exposed to hypoxic and hypercarbic environments.
Work of breathing (W) seems to be greater in juveniles when compared with adults, but this is due to the greatly different MB of the two groups and the different pump volumes used. To better compare the applied pump volumes, we divided (as we did for Cdyn; Fig. 5) pump volume by juvenile and adult MB. According to this calculation, the 0.6 ml pump volume of juveniles is proportional to the 25 ml pump volume of adults, as these volumes represent MB-corrected pump volumes of ∼7.6 ml kg−1. Comparing these two pump volumes reveals that W in supine adults is almost double (43.0–50.9 ml cmH2O min−1 kg−1) that in juveniles in the supine position (26.8–33.5 ml cmH2O min−1 kg−1). This difference remains when animals are in the prone position (juveniles: 26.1–33.9 ml cmH2O min−1 kg−1; adults 48.6–56.0 ml cmH2O min−1 kg−1), but not in the isolated lungs (juveniles: 7.6–10.9 ml cmH2O min−1 kg−1; adults 8.0–13.7 ml cmH2O min−1 kg−1) (Fig. 6). This greater W at proportional pump volumes can be attributed to the greater Cdyn seen in supine and prone juveniles when compared with adults (Fig. 4). Surprisingly, the greater Cdyn of isolated juvenile lungs did not translate into lower W when compared with adult isolated lungs.
Juvenile C. carbonarius individuals show a minute ventilation of about 6 ml min−1 kg−1 at 25°C and about 10 ml min−1 kg−1 at 35°C, while adults display greater levels of ventilation at both temperatures (∼20 ml min−1 kg−1 at 25°C and ∼60 ml min−1 kg−1 at 35°C, respectively) (Oliveira et al., 2023). Exposing adult C. carbonarius to hypoxia or hypercarbia, Trevizan-Baú et al. (2018) found that ventilation increased up to 150 ml min−1 kg−1 at 25°C. Only the lowest level of ventilation (1 ml min−1; Fig. 7) chosen in the present study reflects minute ventilation seen in resting juvenile C. carbonarius as it represents a MB-corrected ventilation of 12.5 ml min−1 kg−1. Regarding adults, the ventilation volumes of 10, 25, 50, 100, 200, 300, 400 and 500 ml min−1 represent MB-corrected ventilation volumes of 3.1, 7.7, 15.4, 30.7, 61.5, 92.3, 123.1 and 153.8 ml min−1 kg−1, respectively, values that cover the physiological range of minute ventilation seen in adult animals well. While was lowest for the isolated lungs in both groups, position showed a more pronounced effect in juveniles when compared with adults. Juveniles showed a greater in the supine position, which could be expected from the viscera resting on the lungs while being in an inverted position. Once this impact was removed in the prone position, decreased for all levels of ventilation and frequency. In adults, however, was very similar in the two body positions, which seems surprising given that the absolute visceral mass resting on the adult lungs while in a supine position should be significant. An explanation for the similar could be the presence of a well-developed PPS in combination with a well-ossified shell. In adults, a firm PPS separating the lungs from the remaining viscera while being attached to an immobile carapace might avoid a significant impact of the viscera while in an inverted position. In juveniles, in contrast, the carapace is less ossified and still flexible (Oliveira et al., 2023) and the PPS might be less developed, thereby increasing the effect of the visceral mass on the lungs when in a supine position. A detailed examination of the development of the PPS in this chelonian combined with its effect on breathing mechanics seems necessary to better understand the differences observed by us between juvenile and adult C. carbonarius.
Conclusion
Chelonoidis carbonarius shows some differences in mechanical properties, when comparing prone with supine positions. These differences can be explained by a certain influence of the viscera on the lungs. However, the complete post-pulmonary septum present in these animals could reduce the pressure of the viscera when animals are in the supine position. The isolated lungs, as expected, showed a large compliance and, consequently, a decreased work of breathing when compared with the intact system. Combinations of volume and frequency indicate that mechanics are more volume dependent, corroborating data published on the lung ventilation of this species. Unlike semi-aquatic species, such as T. scripta, which are more frequency dependent, this observation may be associated with the terrestrial habitat of the living animals. The differences found between juveniles and adults may be affected by the normalization of the lung volume (e.g. MB versus VLr), especially given the less ossified carapace in juveniles. Nevertheless, juveniles showed a larger relative lung volume than adults, indicating a possible large lung volume related to size in the early stages of life in C. carbonarius.
Acknowledgements
We would like to express our special gratitude to Prof. Dr Augusto S. Abe for providing the animals used in this study.
Footnotes
Author contributions
Conceptualization: P.R.C.d.O., P.T.-B., W.K.; Methodology: P.R.C.d.O., P.T.-B., W.K.; Software: P.R.C.d.O., P.T.-B., R.B.B.d.S., W.K.; Validation: P.R.C.d.O., P.T.-B., R.B.B.d.S., W.K.; Formal analysis: P.R.C.d.O., P.T.-B., R.B.B.d.S., W.K.; Investigation: P.R.C.d.O., P.T.-B., W.K.; Resources: P.R.C.d.O., P.T.-B., W.K.; Data curation: P.R.C.d.O., P.T.-B., W.K.; Writing - original draft: P.R.C.d.O., P.T.-B., W.K.; Writing - review & editing: P.R.C.d.O., P.T.-B., W.K.; Visualization: P.R.C.d.O., P.T.-B., W.K.; Supervision: W.K.; Project administration: P.R.C.d.O., P.T.-B., W.K.; Funding acquisition: W.K.
Funding
W.K. received support from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; 2012/18652-1) and from the Conselho Nacional de Desenvolvimento Tecnológico e Científico (CNPq; 302999/2022-1). P.T.-B. was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES Finance Code 001) Master's scholarship within the Programa de Pós-Graduação em Biologia Comparada (FFCLRP, Universidade de São Paulo). P.T.-B. currently receives the University of Florida McKnight Brain Institute Gator NeuroScholar Fellowship.Open access funding provided by University of Florida. Deposited in PMC for immediate release.
Data availability
All relevant data can be found within the article and its supplementary information.
References
Competing interests
The authors declare no competing or financial interests.