Avian air sacs and neopulmo: their evolution, form and function

Abstract

The avian respiratory system is composed of an exchange structure (parabronchi) and a pump (air sacs) to perform gas exchange. While there are many studies dealing with the morphology and function of the palaeopulmonic parabronchi, the air sacs and the neopulmo have been somewhat neglected from a comparative and functional point of view, not always receiving a closer examination that they deserve. While a decent amount of data are available regarding air sac and neopulmo morphology on a family level or for domestic species, several orders of birds have yet to be investigated. Owing to the lack of detailed specific data, we did not perform a comparative phylogenetic analysis but compiled data regarding air sac and neopulmo morphology and analysed them from the viewpoint of current phylogenetic relations while also discussing aspects of these structures regarding avian physiology.

This article is part of the theme issue ‘The biology of the avian respiratory system’.

1. Introduction

Coiter [1] first described avian respiratory anatomy in written form in 1573. Since then, there has been continuous interest in the respiratory system of this group of animals, and looking up ‘bird’, ‘lung’, ‘air sac’ and ‘morphology’ in a scientific search engine yields several thousand hits. It is therefore beyond the scope of the present work to give an exhaustive revision of the subject; rather, we intend to discuss the knowledge regarding avian air sacs within a functional framework.

As mentioned, the morphology of the lung–air sac complex in birds has been intensely studied, and we refer the reader to King [2], Duncker [3,4], McLelland [5], Maina [6–9] and Hsia et al. [10] for detailed descriptions regarding the structure of the avian respiratory system. This morphological knowledge served as a basis for understanding the complex airflow patterns between the air sacs, with a solely ventilatory function, and the gas exchange surfaces (= parabronchi; figure 1a ). It has been established that airflow through the parallel parabronchi between dorsal and ventral secondary bronchi is unidirectional in a caudal–cranial direction during both inspiration and expiration and is maintained by aerodynamic valving of airflows [11–17]. The gas exchange structure receiving unidirectional airflow has been termed palaeopulmo as it is present in all birds, while a set of parabronchial tubes at the entrance of the (mainly posterior) air sacs, which are missing in penguins and emus but may account for up to 25% of a bird’s gas exchange tissue in other species, are referred to as neopulmo [3,4]. When present to its greatest extent, the neopulmo occupies the entire lateral part of the lungs, not only being present at the entrance of the posterior air sacs but also at the entrance of the cranial thoracic and interclavicular sacs [4,5] (figure 1b,c ). Airflow in the neopulmo is considered bidirectional as air moves over gas exchange surfaces once it enters an air sac and again when it leaves an air sac. While the detailed airflow patterns in respiratory systems with a well-developed neopulmo have not yet been investigated, a functional correlation between large anterior thoracic air sacs and cranially reaching neopulmonic bronchi has been suggested for plovers, curlews, sandpipers and fowl-like birds [3], but needs future investigations to be corroborated.

Figure 1.

Schematic representation of the basal avian respiratory system, (a) indicating relationships between airways (trachea, primary bronchus, dorso- and ventrobronchi), primary gas exchange structure (palaeopulmo; parabronchi indicated by straight lines between dorso- and ventrobronchi) and air sacs (cervical, clavicular, cranial thoracic, caudal thoracic and abdominal). While (a) represents a respiratory system lacking neopulmonic gas exchange structures, (b) represents neopulmonic tissue (indicted by red lines at the entrance of the posterios air sacs) at the entrance of the posterior air sacs and (c) depicts a well-developed neopulmo (indicated by a more extensive network of red lines) occupying the latero-inferior section of the lung, connecting both to the posterior air sacs as well as the clavicular and cranial thoracic air sacs. Extension of different air sacs and palaeopulmo (see (a) for corresponding colours) is depicted for the Rockhopper penguin Eudyptes cristatus (d, Spheniscidae), the Ring-necked pheasant Phasianus colchicus (e, Phasianidae), the Razor-billed auk Alca torda (f, Alcidae) and the Grey heron Ardea cinerea (g, Ardeidae). Images adapted from [3] and not to scale. (Online version in colour.)

The air sacs of the avian respiratory system have been studied in great detail in domestic and common species, but compared with the over 11 000 extant species of birds, our comparative knowledge is still sparse. Embryologically, six pairs of air sacs form [18], but air sac fusions subsequently occur, normally resulting in 7–9 air sacs in total. These are the cervical, (inter)clavicular, cranial thoracic, caudal thoracic and abdominal air sacs [3,4] (figure 1a ). On functional grounds, the first three are grouped together as anterior air sacs, while the latter two are grouped as posterior air sacs.

The organization of the air sacs within the avian body cavity is directly related to the presence of horizontal and oblique septa, both derivatives of the post-pulmonary septum [3,19,20]. Briefly, the parabronchial lungs are confined to the dorsal part of the thoracic cavity by the horizontal septum into a pleural cavity. Owing to the rigidity of the parabronchial gas exchange structures, this pleural cavity is maintained at a constant volume during ventilatory movements of the ribcage, as costoseptal muscles at the margin of the horizontal septum modulate its stiffness [21]. The oblique septum separates most air sacs into a subpulmonary cavity, while only the abdominal air sacs (except for in the kiwi and cassowary) [22,23] invade the abdominal body cavity.

2. Systematic analysis

To place the morphological variation seen in avian air sac morphology into a phylogenetic context, one is faced with the dilemma that a reasonable number of species have been studied over time, but not all species have had their morphological details presented in the literature. Air sac morphology has often been presented by grouping families together and describing their general pattern among groups. Figure 2 gives an approximate estimate of the number of species that have been studied, but a detailed examination resulted in only 32 species being available for comparative phylogenetic analysis. We therefore refrain from performing such an analysis because the number of species available is low compared with the absolute number of bird species, and furthermore, data are biased towards passeriforms, charadriiforms, galliforms and anseriforms, with several orders not yet having been studied. However, we have compiled the data available in the literature (mainly from [2–5]) into an overview to highlight their variation in air sac morphology, to present some suggested functional relationships, and to indicate directions for future studies (figure 2). Some trends, however, can be observed, such as a predominance of the cranial and caudal thoracic air sacs when compared with the other anterior or posterior air sacs, a notable degree of fusion among anterior air sacs in galliforms, nightbirds (Strisores) and passeriforms and a well-developed neopulmo in passeriforms, columbiforms and galliforms.

Figure 2.

Avian phylogeny according to Stiller et al. [24]. Respiratory system variables compiled from [2–5,25,26]. 1—Approximate number of species whose air sacs were studied, but whose morphological details were not necessarily reported; 2—cervical air sacs; 3—clavicular air sacs; 4—cranial thoracic air sacs; 5—caudal thoracic air sacs; 6—abdominal air sacs; 7—presence of neopulmonic parabronchi. In the groups for which species were investigated but where no symbol has been added, no mentions of any particular characteristic feature have been found in the literature regarding that structure. ++, very large; +, large; –, small; – –, very small; m, missing; f, air sac fusion present; * clavicular air sac with more than one chamber.

The cervical air sacs are usually paired structures, well developed in penguins, ducks, geese, swans, falcons and buzzards, but small in storks, herons, galliforms and passeriforms. Loons, grebes and the red-winged tinamou [25] have been reported to lack cervical air sacs. Junglefowl, turkeys, hummingbirds and house sparrows are described to have fused both air sacs, resulting in one unpaired cervical air sac, or in some cases they even have fused the cervical with the clavicular air sac.

The clavicular air sac generally is a relatively large, unpaired structure that develops embryologically by fusion of primordial paired lateral and medial parts, but that persists in gulls and several stork species as four separate clavicular sacs. In turkey, on the other hand, the primordial medial parts remain as separate sacs, while the lateral clavicular sacs fuse with an unpaired cervical sac, resulting in a single cervicoclavicular sac.

The cervical and clavicular sacs occupy the anterior space of the thoracic cavity and support the trachea and primary bronchi, oesophagus, blood vessels and nerves ventrally to the vertebral column, and they may contain varying numbers of internal septa subdividing their inner space [3]. Both the cervical and clavicular air sacs are also known for their diverticula invading the cervical column, sternum, ribs, pectoral girdle and shoulder joint, as well as intermuscular, subcutaneous or intrathoracic spaces. These variations in the diverticula of cervical and clavicular air sacs, as well as the varying degrees of fusion seen between those two groups, are responsible for the great variation regarding the total number of air sacs among birds given in the literature.

Much less variation is present in the other air sacs. Both the cranial and caudal thoracic air sacs are paired structures in the subpulmonary cavity. The cranial ones tend to be smaller than the caudal ones, reaching only one-quarter in most species of the caudal thoracic sac's size in most species, but can be larger (one-third of the size of the caudal ones) in swans and storks, or even the other way around in pigeons, fowl-like birds, birds of prey and the Eurasian coot. In passeriforms, the cranial thoracic sacs are fused with the clavicular sac, but possibly to varying degrees, as demonstrated by a recent study in the zebra finch [28], where two out of five individuals showed small, but separated right cranial thoracic air sacs. Only the turkey has been reported as missing the caudal thoracic sacs entirely. Only a few intrathoracic diverticula are reported as associated with these air sacs.

The paired abdominal air sacs are the only ones advancing into the peritoneal cavity, posterior to the post-pulmonary septum, occupying the dorsal portion of the abdominal cavity and reaching between the intestinal loops. The notable exceptions are the kiwi and the cassowary, where this air sac is located within the subpulmonary cavity [3,22,23]. These air sacs are very large in the Atlantic puffin, flamingos, common moorhen and cuckoo, but poorly developed in birds of prey, coots, frigatebirds, ratites, penguins, loons, large passeriforms and parrots, and have been found to be especially small in hummingbirds, the cassowary and the kiwi. The left abdominal sac may be smaller than the right one owing to visceral asymmetry, but there are exceptions (common loons, Atlantic puffins and gulls). The abdominal sacs can form diverticula that extend between the kidneys and pelvis, invading the synsacrum and pelvic girdle, as well as forming around the head of the femur, penetrating between thigh muscles, caudal vertebrae and femur.

Diving birds and those swimming underwater have been found to possess a simple air sac structure with only a few diverticula, whereas terrestrial birds and those swimming mainly on the water surface show a much greater degree of pneumatization to reduce specific gravity [3]. In the birds venturing frequently underwater, the body is generally elongated and the abdominal cavity is protected by a relatively long sternum, with ribs covering the abdominal wall, and longer uncinate processes (figure 1d,f ) when compared with non-diving birds (figure 1e,g ) [3,29,30].

Duncker [31] summarizes various morphological variables of the avian respiratory system. Total respiratory system volume is about 20% of body volume in good flyers, increasing up to 34% in large, good-flying birds like geese and swans, whereas the large galliform birds only show a relative respiratory system volume of 10%. Such variation can easily be observed in birds that swim on the water surface, with species like swans barely sinking below the surface, while other species with lower respiratory system volumes may sink deep below the water surface [31]. The gas-exchanging lungs occupy 2.5–3.0% of body volume in good flyers, but less than 2% in the larger galliforms. Within the respiratory system, air sacs account for about 80% of its volume, with one-third to one-half allocated within the anterior air sacs and two-thirds to one-half in the posterior air sacs.

Data are scarce regarding the volume of specific air sacs [4,32,33]; there is agreement that the anterior air sacs are at best equal in volume when compared with the posterior air sacs, but they are generally smaller in volume. However, the determination of air sac volume is a challenging task because the methods based on the injection of some material into the air sacs will usually result in maximum extension of the air sacs. This is especially relevant for the abdominal air sacs, which might occupy only a small volume within the abdominal cavity of a living bird but can be expanded to a large volume when force-filling the respiratory system. From a functional point of view, determining air sac volume using various levels of inflation [34] seems more informative in understanding avian physiology. Using this approach, Martinez et al. [28] found cranial air sacs (ca 48.8%) to be roughly equal volume to the caudal air sacs (ca 51.2%) in zebra finches. CT scans have been performed on anaesthetized geese to estimate anterior and posterior air sac volumes for both resting and maximally inflated animals, showing that proportional volumes did not change owing to the filling status of the respiratory system, with anterior air sacs representing ca 40% of respiratory system volume [33].

As detailed above, several air sacs give origin to air-filled diverticula that penetrate skeletal elements and/or soft tissues. The presence of air sac diverticula within fossil skeletal elements has been of fundamental importance to our understanding of the avian respiratory system's origin [35–40]. Among living birds, air sac diverticula represent only a very small volume of the entire respiratory system (ca 4%) [31,33] and they do not play any part in bird respiration [41,42]. A great expansion of air sac diverticula, especially those reaching subcutaneous tissues, has been suggested to increase the risk of respiratory system infection, which might easily spread to other tissues owing to the great contact area between air sacs and surrounding structures [10].

Air sac diverticula reduce the body mass of birds [4] and are commonly associated with locomotor specializations. A detailed comparative investigation regarding post-cranial pneumatization [43], however, failed to support a positive relationship between body mass and skeletal pneumatizations but instead found correlations with wing morphology, for example. Recently, Schachner et al. [42] have demonstrated that the subpectoral diverticulum contributes significantly to the biomechanics of the pectoral muscle in soaring birds when compared with non-soaring birds. These recent studies [42,43] highlight the importance of detailed morphological studies using a comparative phylogenetically informed framework to better understand the diversity and evolution of possible non-respiratory functions of the avian respiratory system, as well as the relation between respiratory system morphology and variables like body size, mode of locomotion (diving, terrestrial and flight) and metabolic demands.

3. Air sac function

The best-investigated function of the air sacs is their movement of air over the respiratory surfaces. The classical model [44,45] of rib motion as related to ventilation has recently been updated [30,46,47], showing that the motion of sternum and ribs is far more complicated than previously realized, possibly allowing birds a precise regulation of expansion and contraction of air sacs, as well as of overall intra-abdominal pressures. Such a control seems fundamental for a unidirectional flow of air through the palaeopulmonic parabronchi, as physiological experiments [11,48], as well as mathematical modelling [49,50], have shown that the aerodynamic valves do not operate at 100% efficiency and depend on gas density, flow velocity and air sac compliance [11,32]. Air sac compliance seems worthwhile investigating in particular, since mathematical models highlight this variable [49,50] while physiological experiments [17,32] have shown that posterior air sacs seem to be more compliant than the anterior air sacs in ducks. Additionally, a significant imbalance in ventilation rate for specific air sacs has been found, mainly for the cranial and caudal thoracic sacs, which receive three times more ventilation than the clavicular and abdominal sacs [32]. Considering (a) the different relative volumes seen in air sacs among birds (figures 1 and 2), (b) the presence of internal septations in the cervical and clavicular sacs [3] and (c) varying degrees of rib development within the abdominal wall, it seems necessary to assume a significant variation in compliance of individual air sacs, and this should be investigated in the future to better understand airflow within the avian respiratory system.

The avian respiratory system is capable of more than traditional caudal–cranial unidirectional airflow in the palaeopulmonic parabronchi, which can be seen in penguins lacking neopulmonic parabronchi. Measurements of the partial pressure of CO₂ in the caudal thoracic sacs yielded higher than expected values [51], suggesting to the authors a cranial–caudal flow direction of air through the parabronchi. Considering the presence of pressure fluctuation between air sacs in penguins during locomotion underwater [52], it therefore seems possible that these animals are able to move air back and forth between anterior and posterior air sacs, increasing their extraction of oxygen by better mixing the volumes of air residing within their anterior and posterior air sacs and thereby prolonging their dive time.

Air sacs are not being equally ventilated. Ducks have been shown to proportionally ventilate more of the cranial and caudal thoracic sacs when compared with the clavicular and abdominal sacs [32]—a pattern that changes towards better ventilated abdominal air sacs in anaesthetized ducks. The same study found washout to be greatest from the clavicular air sac (10.3%) but lower from the thoracic (approx. 3.3 %) and abdominal sacs (0.4 %), indicating that birds possess a large reserve capacity residing in their air sacs when it comes to increasing ventilation by increasing tidal volume [13]. One can, therefore, ask if there are other functions associated with the extension of air sacs that are relevant for avian function. An important behaviour among birds is communication using vocalizations. It has been shown that sound production is directly related to airflow and air sac pressures [53,54] and one must consider possible functional constraints on the anterior air sacs, given that the syrinx is positioned among air sacs, contributing to the production and transmission of sound. It is important to note that the clavicular air sac need to be pressurized for sound production, as experimental rupture of this air sac in chickens and Mallard ducks abolishes vocalizations [55,56], and air sac pressure may be up to 10 times greater than tracheal pressures during vocalizations [57].

Another function may well be the evolution of bipedalism in the theropod lineage that includes the modern birds [58], as increasing levels of body pneumatization influence the distribution of body mass and an animal’s centre of gravity [59,60]. While air sacs seem instrumental for the evolution of bipedalism in the ancestor of modern birds, they are not a fundamental pre-requisite for unidirectional airflow since such an airflow pattern has been demonstrated in a species of alligator and other non-avian reptiles lacking air sacs [61–63]. An evolutionary sequence can be postulated, with unidirectional airflow evolving within a multichambered lung of an ancestral archosaur while air sacs evolved later and concomitantly with bipedalism—a scenario backed up by variations in vertebral column morphology [64]. One has furthermore to consider that these adaptations may have occurred during an epoch of lower environmental oxygen concentrations than today, possibly being as low as 15% during the early Mesozoic, and at much higher temperatures than today [65]. Evolving an efficient respiratory system composed of a structure specialized for gas exchange (parabronchi) and another structure for ventilation (air sacs) under hypoxic conditions might explain why birds are the only vertebrates able to fly in a sustained manner at great altitudes. While mammalian lungs allowed bats also to evolve the ability to fly sustainably, only birds can do so at great altitudes in a strongly hypoxic and below-freezing atmosphere [10].

A large amount of air sac diverticula invading spaces outside of a bird’s coelomic cavity has naturally been associated with a reduction in body mass, facilitating flight. In a thought-provoking article, Duncker [31] presents a functional context for the morphological variation seen in the muscular, circulatory and respiratory systems that are mainly responsible for a bird’s ability to fly. Regarding respiratory system volume, he suggests that in good long-distance flyers, about 20% of body volume is occupied by lungs and air sacs, a value that can reach up to 34% in large birds that are good flyers. Such an enlarged body volume would allow for a greater surface area for the insertion of flight muscles, resulting in larger wings and consequently lower wing loading [31]. Compared with these large birds with their good flight capacity, the lineage of galliforms pursued another evolutionary path with increasing body size and reduced flight capacity [31]. Small galliform species like quails are good long-distance flyers, whereas the large galliform species show only limited aerobically sustained flight capacity, with the largest member, the turkey, only able to perform a short flight onto a nearby tree to escape a predator. Such a limited ability to fly over longer distances can be associated with a reduced amount of flight muscles (10–16% in domestic fowls when compared with 22–24% in other galliforms), an increased amount of anaerobic (white) muscle fibres in the pectoral muscle, a smaller heart, a smaller blood volume, a lower pulmonary diffusion capacity and a lower respiratory system volume (ca 10% of body volume) [31]. Duncker [31] concludes that large galliform birds, and especially the domesticated fowl, show significant morphological and physiological deviations from the general picture seen in regular flying birds and should only be used with caution when considering cardiorespiratory physiology. Nevertheless, despite their reduced flight capacities, galliforms, are known to possess a well-developed neopulmo [3–5]. None of these ideas, however, has yet been tested using comparative analytical tools, but the finding that the subpectoral diverticulum improves mechanical leverage of the pectoral muscle in soaring birds [42] provides some support for Duncker’s hypotheses.

4. The case of the neopulmo

Regarding the neopulmonic parabronchi, which are closely associated with air sacs, it has long been claimed that this gas exchange tissue is either lacking or only poorly developed in less derived groups such as emus, kiwis and penguins [3–5]. While this seems true for the Palaeognathae (tinamiforms, rheiforms, apterygiforms, casuariiforms, and struthioniforms; figure 2), the phylogenetic position of penguins either implies that this group secondarily lost neopulmonic structures during their evolution or that the evolution of neopulmonic gas exchange tissue is more complicated than previously anticipated. The phylogenetic distance between passeriforms, columbiforms and galliforms, which all developed a pronounced neopulmo as part of their respiratory system, recommends a more detailed investigation regarding the phylogenetic history of the neopulmo and to provide a better understanding of its functional importance. If penguins, for example, lost their neopulmo while adapting to become efficient divers, one could imagine that the bidirectional flow of air through the neopulmo, in addition to the gas exchange in the palaeopulmo, transfers too much oxygen from the respiratory system into the blood, resulting in a faster-decreasing oxygen supply and thereby shortening dive duration. Regarding the poorly developed neopulmo of the ostrich, it has furthermore been suggested [66] that these animals are able to pant for 8 h without altering their acid–base balance [67], as the posterior air sacs may function as storage spaces for CO₂. As thermoregulation is an important function of the avian respiratory system owing to the ease with which it transports heat from the body cavity through the air sacs to the outside, the relation between panting and neopulmonic gas exchange becomes relevant and should be investigated in a greater number of species than have been studied to date [31,61,68].

Another question is the claim [3] that birds with neopulmonic parabronchi should be able to supply their oxygen demand during rest solely through the gas exchange surface of the neopulmo. Palaeopulmo and neopulmo are structurally identical [69,70] and the neopulmo may represent up to 25% of a bird’s gas exchange parenchyma. As the neopulmo can be relatively well perfused [71] and shows a high ventilation/perfusion ratio [72], a significant amount of gas exchange seems possible. The neopulmonic gas exchange has been quantified for two birds, the domestic duck [73] and goose [72], both studies attributing less than 20% of resting O₂ consumption and CO₂ release to exchange occurring in the neopulmo. Morphological diffusion capacities have been estimated [74] to be relatively similar for duck and goose (3.85 ± 0.21 and 3.59 ml O₂ min⁻¹ mmHg⁻¹ kg⁻¹, respectively). Assuming the neopulmo to occupy roughly 10% of the lung’s gas exchange parenchyma, and assuming a diffusion gradient of 100 mmHg between inspired air and blood reaching the neopulmo, morphological diffusion capacity in the neopulmo would be between 35.9 and 38.5 ml O₂ min⁻¹ kg^-1, which is much larger than the O₂ consumption of a resting duck (14.8) [27] or goose (8.86 ml O₂ min⁻¹ kg⁻¹) [72]. While it seems morphologically possible for a bird to cover its resting oxygen consumption solely through the neopulmonic gas exchange surfaces, the physiologically measured values do not confirm Duncker’s hypothesis.

5. Suggestions for future directions

While the general workings of the avian respiratory system have been well investigated, there are still questions that need to be addressed. Both morphological and physiological variations are poorly represented in view of the phylogenetic diversity of birds, with some 13 orders not yet having been studied. Stereological data of parabronchial and neopulmonic gas exchange tissues, volumes, pressures and ventilation rates of individual air sacs, data on oxygen extraction and perfusion of the neopulmo when compared with the palaeopulmo seem important starting points to improve our knowledge of the avian respiratory system, especially when obtained under conditions of rest, vocalization and locomotion (terrestrial, aerial, swimming and diving), as well as allometric relationships. Such an enlarged database should also help in understanding the evolution of the respiratory system within specific lineages, as well as the evolution of the respiratory system among archosaurs, especially when applying current comparative tools.

Regarding air sacs, the mechanical properties of individual air sacs and their influence on airflow are not known among different species and might be correlated with the mode of locomotion. In the case of air sac diverticula invading muscle tissues and/or skeletal elements, it seems worthwhile to compare the possible biomechanical impacts among birds of different size or with different modes of locomotion. Also, the functional advantage, if there is any, of fusion between different air sacs merits further attention. In passeriforms, for example, one could ask if there is a correlation between the large degree of fusion seen among anterior air sacs, the mainly small abdominal air sacs and a well-developed neopulmo. Possibly, moving air mostly between large caudal thoracic air sacs and a large fused anterior air sac improves oxygen extraction not only by palaeopulmonic but also by the well-developed neopulmonic parabronchi.

The neopulmo, as part of the avian gas exchange structure, is currently underrepresented within the literature, both from a morphological as well as a physiological point of view. The characterization of neopulmo size among different lineages of birds, combined with a better understanding of airflow patterns and oxygen extraction, should greatly improve our knowledge regarding the evolution of this structure, especially when correlated with different modes of locomotion, body size, thermoregulatory demands or metabolic demands. Thereby, the evolution of a well-developed neopulmo in passeriforms, galliforms and columbiforms could be contrasted against the species with less-developed or lacking neopulmo. From an animal welfare perspective, one might investigate the putative relationship between the extension of air sac diverticula and susceptibility to respiratory diseases. In summary, despite several centuries of study, many open questions remain, and detailed analysis of morphological and physiological variation should provide renewed insight into the form and function relationships of the avian respiratory system.

Ethics

This work did not require ethical approval from a human subject or animal welfare committee.

Data accessibility

This article has no additional data.

Declaration of AI use

We have not used AI-assisted technologies in creating this article.

Authors’ contributions

W.K.: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, validation, writing—original draft; V.P.R.: data curation, investigation, methodology, writing—original draft; R.B.B.d.S.: conceptualization, data curation, formal analysis, methodology, validation, writing—original draft.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration

We declare we have no competing interests.

Funding

WK received support from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; no. 302999/2022-1). VPR was supported by a scholarship from the University of São Paulo. R.B.B.d.S. received support by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP; no. 2021/11389-2).