New insight into the evolution of the vertebrate respiratory system and the discovery of unidirectional airflow in iguana lungs

Abstract

The generally accepted framework for the evolution of a key feature of the avian respiratory system, unidirectional airflow, is that it is an adaptation for efficiency of gas exchange and expanded aerobic capacities, and therefore it has historically been viewed as important to the ability of birds to fly and to maintain an endothermic metabolism. This pattern of flow has been presumed to arise from specific features of the respiratory system, such as an enclosed intrapulmonary bronchus and parabronchi. Here we show unidirectional airflow in the green iguana, a lizard with a strikingly different natural history from that of birds and lacking these anatomical features. This discovery indicates a paradigm shift is needed. The selective drivers of the trait, its date of origin, and the fundamental aerodynamic mechanisms by which unidirectional flow arises must be reassessed to be congruent with the natural history of this lineage. Unidirectional flow may serve functions other than expanded aerobic capacity; it may have been present in the ancestral diapsid; and it can occur in structurally simple lungs.

Keywords: diapsid; evolution; lizard; lung; respiratory system.

Fig. 1.

Anatomy of the respiratory system of the green iguana. (A) Dried left lung transected in the parasagittal plane (cutaway section inverted). White lines (b and c) mark approximate levels of transection shown in volume-rendered µCT data illustrated in B and C. (B) Dorsal view of both lungs transected in the coronal plane. Arrow points to the location where the primary bronchus enters the cranial chamber of the right lung. (C) Ventral view of the three chambered heart and lungs transected in the coronal plane. Arrow marks the ostium between the chambers of the left lung. Craniad, left.

Fig. 2.

Flow simulation and in vivo visualization of smoke. (A) Dorsomedial view of the model with planes of section (b,c and d,e) projected in B–E. (B) Inspiration: Simulated flow in plane b,c is primarily craniad (warm colors). Eye marks approximate location where smoke (circles) was visualized in vivo (Movie S1). Smoke flows craniad at this location. (C) Exhalation: simulated flow and visualized smoke move craniad. (D) Inspiration: simulated flow in plane d,e shows high-speed fluid streams emanating caudad (cool colors) from the primary bronchus and ostium. Smoke flows caudad (Movie S2). (E) Exhalation: Simulated and visualized flow is craniad. Flow velocity (meters per second). Note the change in direction as denoted by the arrows. Asterisk, a clump of smoke.

Fig. 3.

Flow simulation and ex vivo visualization of flow. (A) Ventromedial view of model with coronal plane of section projected in B and C. (B) Inspiration: Simulation shows a high-velocity fluid stream emanating caudad (cool colors) and laterad with flow along the medial wall craniad (warm colors). White square marks the approximate location on the model where flow was visualized in excised lungs using fluorescent microspheres in water. Green square panels show three consecutive frames of video (Video S3) of microspheres moving in the lungs while fluid was injected (top three panels) and while it was withdrawn (bottom three panels). (C) Expiration: simulated flow is largely craniad. Flow magnitude (meters per second). Red and blue arrows track microspheres moving craniad and caudad respectively.

Fig. 4.

Pulmonary and tracheal airflow in an excised lung. Direction of airflow measured along the walls of the lung with heated thermistor flow meters in the cranial chamber (A) and caudal chamber (B). Positive trace is cranial flow. (C) Direction of airflow in the trachea measured with a pneumotachograph. Positive trace is exhalation. Note air moved craniad in both chambers during both phases of ventilation.

Fig. 5.

Computational fluid dynamics simulation of airflow at four consecutive seconds in the respiratory cycle in coronal (Left) and transverse (Right) planes. (A) Medial views of the computational model with lines illustrating planes of view projected in B–E. (B) During peak flow of inspiration high-speed fluid streams emanate into both chambers laterocaudad. Most of the remainder of the flow is moving craniad. (C) One second later as inspiration ends and no bulk flow enters or exits the lungs, the internal flow recirculates in a laterocaudal to mediocranial direction. (D) During peak flow of exhalation most of the air flows craniad. (E) End of exhalation and no bulk flow enters or exits the lungs, but both chambers contain low-speed recirculating flow. Axial direction corresponds to the body axis.

Fig. 6.

Probe placement. (A) Right lateral view showing the caudal point of entry of the endoscope (white arrow). (B) Ventral view of lungs showing caudal point of entry of the endoscope (white arrow). (C) Lateral view of excised right lung showing location of thermistor flow meters. (Scale bar, 1 cm.)

Fig. 7.

Numerical verification of CFD solutions of airflow in the iguana lung. (A) Flow patterns in the iguana lung during inspiration computed using the coarser mesh (1.8 million elements) and time step (0.005 s). (B) Flow patterns at the same time as in A computed using the finer mesh (3.9 million elements) and time step (0.0005 s).