Anatomy, ontogeny, and evolution of the archosaurian respiratory system: A case study on Alligator mississippiensis and Struthio camelus

Abstract

The avian lung is highly specialized and is both functionally and morphologically distinct from that of their closest extant relatives, the crocodilians. It is highly partitioned, with a unidirectionally ventilated and immobilized gas-exchanging lung, and functionally decoupled, compliant, poorly vascularized ventilatory air-sacs. To understand the evolutionary history of the archosaurian respiratory system, it is essential to determine which anatomical characteristics are shared between birds and crocodilians and the role these shared traits play in their respective respiratory biology. To begin to address this larger question, we examined the anatomy of the lung and bronchial tree of 10 American alligators (Alligator mississippiensis) and 11 ostriches (Struthio camelus) across an ontogenetic series using traditional and micro-computed tomography (µCT), three-dimensional (3D) digital models, and morphometry. Intraspecific variation and left to right asymmetry were present in certain aspects of the bronchial tree of both taxa but was particularly evident in the cardiac (medial) region of the lungs of alligators and the caudal aspect of the bronchial tree in both species. The cross-sectional area of the primary bronchus at the level of the major secondary airways and cross-sectional area of ostia scaled either isometrically or negatively allometrically in alligators and isometrically or positively allometrically in ostriches with respect to body mass. Of 15 lung metrics, five were significantly different between the alligator and ostrich, suggesting that these aspects of the lung are more interspecifically plastic in archosaurs. One metric, the distances between the carina and each of the major secondary airways, had minimal intraspecific or ontogenetic variation in both alligators and ostriches, and thus may be a conserved trait in both taxa. In contrast to previous descriptions, the 3D digital models and CT scan data demonstrate that the pulmonary diverticula pneumatize the axial skeleton of the ostrich directly from the gas-exchanging pulmonary tissues instead of the air sacs. Global and specific comparisons between the bronchial topography of the alligator and ostrich reveal multiple possible homologies, suggesting that certain structural aspects of the bronchial tree are likely conserved across Archosauria, and may have been present in the ancestral archosaurian lung.

Keywords: 3D modeling; Aves; Crocodylia; computed tomography; lungs; pulmonary.

FIGURE 1

Phylogeny for Tetrapoda demonstrating the structural diversity of the tetrapod lung. (a) Amniota; (b) Sauropsida; (c) Archosauria. Volume rendered skeleton and surface model of the bronchial tree of: (d) Xenopus sp.; (e) Saguinus sp., (with a diagrammatic illustration of a standardized primate bronchial tree); (f) Iguana iguana (modified from Cieri et al., 2014); (g) Varanus exanthematicus (modified from Schachner et al., 2014); (h) Chelydra serpentina (modified from Schachner et al., 2017); (i) Alligator mississipiensis; and, (j) Struthio camelus. Images not to scale

FIGURE 2

Segmented 3D surface model of the thorax, lung, and bronchial tree of Alligator mississippiensis. Alligator mississippiensis 11 (live) during a natural apnea in dorsal (a, c, e), and left lateral (b, d, f) views. Parabronchi (i.e., connections between the primary, secondary and tertiary bronchi) are not shown

FIGURE 3

Diagrammatic models demonstrating the quantitative metrics. Simplified and reduced digital model of the bronchial tree of the right lung of Alligator mississippiensis in medial (a), craniomedial (b), and ventral (c, d) views, with the tertiary, medial, caudal and lateral bronchi all removed. Simplified and reduced digital model of the bronchial tree of the right lung of Struthio camelus in (e, f), dorsomedial (g), and ventral (h, i) views. These models are 3D representations and indications of the 2D quantitative metrics obtained from the DICOM images in OsiriX. (a) Gold rings represent the metrics obtained at the trachea (metric 1) and primary bronchus (metric 2) in the alligator. Pink oblique rings demonstrate the sites where metrics were taken for the area of the primary bronchus, perpendicular to the origin of the secondary bronchus (metric 3). (b) Red oblique rings represent the site where the metrics were taken for the area of the ostium for each secondary bronchus as it branched from the primary bronchus (metric 4). (c, d) Diagram of where metrics were taken for the distances from the carina to each of the large secondary bronchial ostia (metric 6) (c) with the bronchi labeled (d) (metrics 1 and 2 are also labeled in this view). (e) Pink oblique circles demonstrating the position of metric 3 in the ostrich. Blue oblique circle demonstrates metric 5 as measured around the large laterobronchus on S. camelus. (f,g) Red rings demonstrate metric 4 in S. camelus for the ventrobronchi and dorsobronchi. (h,i) Diagram of where metrics were taken for the distances from the carina to each of the large secondary bronchial ostia (metric 6) in S. camelus with the measurement diagram (h) and the labeled ostia (i). Abbreviations: CA, carina; CVB, cervical ventral bronchus; D2‐D5, dorsobronchi 2‐5; L, laterobronchus; PB, primary bronchus; V1‐4, ventrobronchi 1‐4. Numbers indicate specific metrics described in methods. Images not to scale

FIGURE 4

Measured and imputed body masses of Alligator mississippiensis scaled to pulmonary measures across a growth series. (a) Measured/imputed body mass scaled to the maximum diameter of the right primary bronchus just distal to the bifurcation from the trachea; (b) measured/imputed body mass scaled to the area of the right primary bronchus at the same location as (a)

FIGURE 5

Segmented 3D surface model of the primary, secondary, and large tertiary airways of Alligator mississippiensis 64. Specimen is deceased, artificially inflated, and all images in left craniolateral view (except for D). (a) The primary, secondary, and large tertiary bronchi; (b) the primary and secondary bronchi (the tertiary bronchi have been removed); (c) the primary bronchi, the cervical ventral bronchus, and the dorsobronchi; (d) image (c) in left lateral view; (e) the primary and medial (M) bronchi; (f) the primary bronchi, the laterobronchi, the cardiac lobes, and the caudal group bronchi. Parabronchi (i.e., connections between the primary, secondary and tertiary bronchi) are not shown. Abbreviations: C, cardiac lobes; CGB, caudal group bronchi; CVB, cervical ventral bronchus; D2‐5, dorsobronchi 2‐5; L, laterobronchi; M1‐5, medial bronchi; Pb, primary bronchus

FIGURE 6

Segmented 3D surface model of the bronchial tree of Alligator mississippiensis 64. Bronchial tree in dorsal (a) and ventral (b) views, with the ostia of the major secondary and a few tertiary branches represented as stumps to visually demonstrate the clear branching pattern. Abbreviations: C1‐4, cardiac lobes 1‐4; CVB, cervical ventral bronchus; CVB2, secondary branches off of the cervical ventral bronchus; D2‐5, dorsobronchi 2‐5; LB, laterobronchi; M1‐5, medial bronchi 1‐5

FIGURE 7

Segmented 3D surface model of the dorsal vertebrae and ribs, lung surface, and bronchial tree of a hatchling Alligator mississippiensis (AM041315‐1) and CT images of a live adult (“Stumpy”). Hatchling thorax model in left craniolateral (a–c) and ventral (d–f) views generated from µCT data. Lung surface and axial skeleton are shown in (a), (d), and (e). Surface of the lungs and the tertiary bronchi of the left lung are made semi‐translucent in (b,c) and (e,f) to demonstrate the position of the major primary and secondary airways within the lung and relative to the smaller interconnecting branches (=parabronchi). Lung surface and tertiary bronchi are removed from the left lung, and the tertiary bronchi are made semi‐translucent in (c) and (f) to further demonstrate these relationships. Axial (g) and parasagittal (h) CT images of a live adult A. mississippiensis (scanned in a supine position) demonstrating the pulmonary heterogeneity and regional distribution of the parenchyma within the lung. Abbreviations: CVB, cervical ventral bronchus; P, parenchyma

FIGURE 8

Intraspecific and methodological variation in the bronchial tree of Alligator mississippiensis. Segmented 3D surface models of the bronchial tree of four different individuals. Top row: The lungs of alligator 81 (deceased), shown in dorsal (a), ventral (b), and left lateral (c) views, were completely dissected out of the thorax, and inflated via a syringe prior to scanning. Second row: alligator 64 (deceased), shown in dorsal (d), ventral (e), and left lateral (f) views, was inflated via a syringe, and scanned in situ in the torso. Third row: “Stumpy,” shown in dorsal (g), ventral (h), and left lateral (i) views, was scanned live, unsedated, and in an upside down (supine) position. Bottom row: alligator 15, shown in dorsal (j), ventral (k), and left lateral (l) views, was scanned live, unsedated, and prone position. Images not to scale

FIGURE 9

Segmented 3D surface model of the post‐cranial skeleton and respiratory system of Struthio camelus 6 (deceased, artificially inflated). Struthio camelus model in left lateral (a), dorsal (b), and left craniolateral (c, d) views. No secondary pulmonary diverticula are shown. (c) Lung surface and air sacs. (d) Lung surface has been removed showing a solid representation of the bronchial tree and the direct connections to the extrapulmonary air sacs. Note that the parabronchi are not shown (connections between the dorsobronchi and ventrobronchi) as they are too small to be segmented from the CT data due to the resolution of a medical grade scanner. Additionally, the interclavicular and cervical air sacs have been segmented as a single unit due to the inability to differentiate between the boundaries because of the resolution of the scan. Abbreviations: AAS, abdominal air sac; CRTS, cranial thoracic air sac; CS, cervical air sac; CTS, cranial thoracic air sac; GL, gas exchanging lung; IAS, interclavicular air sac

FIGURE 10

Segmented 3D surface model of the gas exchanging lung and bronchial tree of Struthio camelus 6. Model is shown in left craniolateral (a–c) and left lateral (d) views. The surface of the gas exchanging lung is represented as semi‐translucent blue and the negative space within the bronchial tree is shown as solid. Note that the parabronchi are not shown (connections between the dorsobronchi and ventrobronchi) as they are too small to be segmented from the CT data due to the resolution of a medical grade scanner. Abbreviations: CRTS, cranial thoracic air sac; D1‐8, dorsobronchi 1‐8; LB, laterobronchi; LS, lung surface; PB, primary bronchus; TR, trachea; V1‐4, ventrobronchi 1‐4

FIGURE 11

Segmented 3D surface model of the bronchial tree of Struthio camelus 6. Model is shown in dorsal (a), ventral (b), and left dorsolateral (c) views, with the ostia of the major secondary branches represented as stumps to visually demonstrate clear branching patterns. Abbreviations: CR, carina; D1‐7, dorsobronchi 1‐7; LB, laterobronchi; PB, primary bronchus; T, trachea; V1‐4, ventrobronchi 1‐4

FIGURE 12

Segmented 3D surface model of the entire respiratory system of Struthio camelus 7. Model is shown in left craniolateral (a, c, d) and lateral views (b, e). The pulmonary diverticula are visible in (a) and (b) and can be clearly seen extending cranially and caudally to the gas‐exchanging lung, as well as positioned dorsally to the ventilatory air sacs. The pulmonary diverticula are removed in (c–e) and demonstrate the lack of continuity with the air sacs. Abbreviations: AAS, abdominal air sac; CS, cervical air sac; CRTS, cranial thoracic air sac; CTS, caudal thoracic air sac; D1‐2, dorsobronchi 1‐2; IAS, interclavicular air sac; L, laterobronchus; PB, primary bronchus; TR, trachea; V1, ventrobronchus 1. Images not to scale

FIGURE 13

Segmented 3D surface model of the skeleton and respiratory system of S. camelus 7. Model is shown in left craniolateral (a), left lateral (b), right lateral (c), and left dorsolateral views with the left ilium removed (d). The pulmonary diverticula are visualized as a solid in (D) to clarify the relationships between these structures and the adjacent skeletal tissues

FIGURE 14

Segmented 3D surface model of the gas‐exchanging lung, bronchial tree, and pulmonary diverticula of S. camelus 7. Model is shown in left lateral view, demonstrating the origin of the majority of the diverticula from the secondary airways, and directly from the surface of the lung, but distinct from the caudal extent of the primary bronchus as it extends beyond the gas‐exchanging lung to balloon into the abdominal air sac. Abbreviations: GL, gas exchanging lung; L, laterobronchus; PB, primary bronchus; TR, trachea; V1‐2, ventrobronchi 1‐2

FIGURE 15

Volume rendered model and coronal CT slices of S. camelus 10 demonstrating extensive axial and appendicular post‐cranial pneumaticity. (a) Volume rendered 3D model of a juvenile S. camelus in left lateral view with lines demonstrating the location of the two coronal DICOM slices shown at positions (b) and (c), and two axial slices shown at positions (d) and (e). Abbreviations: D, diverticula; DP, diverticula pelvica; F(p), femur (pneumatized); GL, gas‐exchanging lung; R(p), rib (pneumatized); V(p), vertebra (pneumatized)

FIGURE 16

Intraspecific variation in the bronchial tree of Struthio camelus. Segmented 3D surface models of S. camelus 7 in dorsal (a) and left lateral (b) views, S. camelus 8 in dorsal (c) and left lateral (d) views, S. camelus 10 in dorsal (e) and left lateral (f) views, and S. camelus 11 in dorsal (g) and left lateral (h) views. Images not to scale

FIGURE 17

Standard major axis regressions of the log of body mass and the log of the following anatomical measurements: (a) Cross‐sectional area of intrapulmonary primary bronchus at the level of CVB (alligator) and V1 (ostrich), (b) cross‐sectional area of the ostium of CVB and V1, (c) distance from the carina to the ostium of the CVB and V1. (d) Diagrammatic illustration of the A. mississippiensis bronchial tree with the cervical ventral bronchus highlighted in green. (e) Diagrammatic illustration of the bronchial tree of S. camelus with the cervical ventral bronchus highlighted in green. Abbreviations: CVB, cervical ventral bronchus; PB, primary bronchus; V1, ventrobronchus. Ostriches =magenta circles. Alligators =blue triangles

FIGURE 18

Standard major axis regressions of the log of body mass and the log of the following anatomical measurements: (a) Cross‐sectional area of intrapulmonary primary bronchus at the level of D2 (Alligator) and D1 (Struthio) and body mass,(b) cross‐sectional area of the ostium of the D2 and D1, (c) distance from the carina to the ostium of D2 and D1 (d) Diagrammatic illustration of the A. mississippiensis bronchial tree with dorsobronchus 2 (the first dorsobronchus) highlighted in lime. (e) Diagrammatic illustration of the bronchial tree of S. camelus with dorsobronchus 1 highlighted in lime. Ostriches =magenta circles. Alligators =blue triangles

FIGURE 19

Standard major axis regressions of the log of body mass and the log of the following anatomical measurements: (a) Cross‐sectional area of the intrapulmonary primary bronchus at the level of D3 (Alligator) and D2 (Struthio) and body mass, (b) cross‐sectional area of the ostium of the D3 and D2 (c) distance from the carina to the ostium of D3 and D2, (d) Diagrammatic illustration of the A. mississippiensis bronchial tree with dorsobronchus 3 (the second dorsobronchus) highlighted in neon green. (e) Diagrammatic illustration of the bronchial tree of S. camelus with dorsobronchus 2 highlighted in neon green. Ostriches =magenta circles. Alligators =blue triangles

FIGURE 20

Standard major axis regressions of the log of body mass and the log of the following anatomical measurements: (a) Cross‐sectional area of intrapulmonary primary bronchus at the level of D4 (Alligator) and D3 (Struthio) and body mass,(b) cross‐sectional area of the ostium of the D4 and D3 (c) distance from the carina to the ostium of D4 and D3, (d) Diagrammatic illustration of the A. mississippiensis bronchial tree with dorsobronchus 4 (the third dorsobronchus) highlighted in aqua. (e) Diagrammatic illustration of the bronchial tree of S. camelus with dorsobronchus 3 highlighted in aqua. Ostriches =magenta circles. Alligators =blue triangles

FIGURE 21

Standard major axis regressions of the log of body mass and the log of the following anatomical measurements: (a) Cross‐sectional area of intrapulmonary primary bronchus at the level of D5 (Alligator) and D4 (Struthio) and body mass,(b) cross‐sectional area of the ostium of laterobronchus 1, (c) cross‐sectional area of the ostium of laterobronchus 2. (d) Diagrammatic illustration of the A. mississippiensis bronchial tree with dorsobronchus 5 (the fourth dorsobronchus) highlighted in blue, and the laterobronchi highlighted in magenta. (e) Diagrammatic illustration of the bronchial tree of S. camelus with dorsobronchus 4 highlighted in blue, and the laterobronchi highlighted in magenta. Ostriches =magenta circles. Alligators =blue triangles

FIGURE 22

Ratio of the distances from the carina to the major secondary bronchi and total distance from the carina to D5 in A. mississippiensis and D4 in S. camelus. Top: The relative distances from the carina to the cervical ventral bronchus and then to each consecutive dorsobronchus (2‐5) in A. mississippiensis. Bottom: The relative distances from the carina to the first ventrobronchus and then each consecutive dorsobronchus (1‐4) in S. camelus. The colors follow the hypotheses of homology. There is limited intraspecific variation in all measures suggesting that the relative distances of secondary bronchi from the carina are strongly ontogenetically conserved. Furthermore, the only substantial difference between the two taxa is the distance from the carina to D2/D1 suggesting the other distances may be conserved within Archosauria

FIGURE 23

Schematic of hypotheses of pulmonary homology shared between the developing chick lung (a) and the adult alligator lung (b). (a) Diagrammatic image of the embryonic chick respiratory track at day 8 of development showing the initial emergence of the air sacs from the bronchial tree, prior to their massive expansion beyond the boundary of the gas exchanging lung; image redrawn and modified from Sakiyama et al. (2000). (b) Diagrammatic simplified illustration of the bronchial tree and lung of an adult alligator lung in left lateral view. Colors denote hypothesized homologous regions. Abbreviations: AAS, abdominal air sac; CGB, caudal group bronchi; CLS, clavicular air sac; CRTS, cranial thoracic air sac; CS, cervical air sac; CVB, cervical ventral bronchus; D2, dorsobronchus 2; LB, laterobronchi. Images not to scale

FIGURE 24

Homology hypotheses for the archosaurian bronchial trees. Segmented solid surface models of the bronchial tree of Alligator mississippiensis (a, d, g, i), and Struthio camelus (b, c, e, f, h, k), all in dorsal view. Colors represent hypothesized homologous primary and secondary bronchi for the two taxa with the “bronchial homology hypothesis 1” (b, e, h): the ostrich ventrobronchi are homologous to the alligator cervical ventral bronchus. “Bronchial homology hypothesis 2” (c, f, k): the avian ventrobronchi are homologous to the alligator medial bronchi, and ventrobronchi 2–4 are homologous to the alligator medial bronchi; (k) demonstrates the angle of orientation of the secondary bronchi in both taxa on the dorsal surface of the primary bronchus. Abbreviations: CVB, cervical ventral bronchus; D, dorsobronchi; M, medial bronchi; V, ventrobronchi; H1, hypothesis 1; H2, hypothesis 2