Functional morphology of the lungs of the green iguana , Iguana iguana , in relation of body mass ( Squamata : Reptilia )

Body mass is one of the most influencing factors of metabolic rate and gas exchange of animals, and also related to activity pattern and occupancy of ecological niches. This study aimed to understand the relationships between body mass (MB) and morpho-functional features of the lungs of Iguana iguana, through morphological and morphometric characterization of the structural elements of the respiratory system. Iguana iguana has lungs of the transitional type, the heterogeneously distributed parenchyma being faveolar in cranial and medial regions and trabecular in the caudal region. Within the parenchyma, 43.6 ± 25.5% corresponds to faveoli, 18.0 ± 5.9% to interfaveolar septa, and 38.7 ± 31.6% to trabeculae. Within the interfaveolar septa, 9.4 ± 4.0% corresponds to blood capillaries, 4.4 ± 1.0% to type I pneumocytes and 3.9 ± 1.1% to type II pneumocytes. Allometric analyses showed that lung (MB) and parenchymal volume (MB) scale with MB in I. iguana just as in other lizards with unicameral or transitional lungs, which was unexpected for lung volume, since reptilian lung volume is generally considered to scale as MB. The functional morphology of the lungs in I. iguana seems to play an important role to meet the metabolic demands through ontogenetic growth.

Among squamates, significant variations regarding the structural pulmonary type and distribution and kind of lung parenchyma are also well known (Perry, 1998).A recently proposed scenario for the evolution of the amniote respiratory system suggests that the last common ancestor of amniotes possessed multi-chambered lungs that evolved into single-chambered lungs among the Lepidosauria due to the small body size of stem lepidosauromorphs (laMbertz et al., 2015).The multichambered lungs found in the Gila monster Heloderma suspectum and monitor lizards of the genus Varanus are being interpreted as having secondarily evolved in this lineage from a simpler lung structure, since a multichambered stage can be identified during the ontogeny of a gekkotan single-chambered lung (laMbertz et al., 2015).Superimposed onto these evolutionary traits are the various metabolic demands that the respiratory system has to serve (Powell & HoPKins, 2004).Typically, reptiles show low metabolic rates compared to endothermic vertebrates with similar body mass (Powell & HoPKins, 2004), but the great diversity of lifestyles and habits resulted in a large variation in metabolic demands and distinct morpho-functional respiratory adaptations.
To maintain increased metabolic rates, a larger respiratory surface area is required, achieved through increased pulmonary compartmentalization. Monitor lizards, for example, show elevated metabolic demands due to their pattern of activity and larger body mass and have lungs more compartmentalized than less active species and with lower body mass (Maina et al. 1989).
The effects of body mass on reptilian metabolism have been widely studied (e.g.van MarKen-licHtenbelt et al. 1993, dawson & bartHoloMew 1956, wood et al. 1978, andrews & PougH 1985, Mcnab 2002, Maxwell et al. 2003) and it has been verified that intraspecifically an increase in body mass results in an increase in the total metabolic rate and in the density of respiratory surfaces.However, among lizards, bennett & dawson (1976( ), beucHat & vlecK (1990)), Karasov & anderson (1998) reported no significant metabolic changes between hatchling and adult iguanids.So, if intraspecific metabolic changes do not occur along with variations in body mass, it is expected that, proportionately, the density of respiratory surfaces is kept mass independent.In order to verify this relationship we choose the species Iguana iguana as experimental model, belonging to the family Iguanidae.This species shows a more than 350 fold increase in body mass from hatchling to adult (12 g -4.5 kg), allowing to study pulmonary morphology in animals of various body sizes.Thus, this study aimed to investigate the relationships between morpho-functional aspects of the lungs in Iguana iguana of different body masses through morphological and morphometric characterization of the structural elements of the respiratory system, which potentially are determinant for metabolic rate and thereby activity of the species.

Animals
Five iguanas of different body masses (M B ; 0.020, 0.026, 0.319, 0.546, and 2.4 kg) were obtained from the same population of animals being held and reproduced at the Jacarezário, São Paulo State University "Julio de Mesquita Filho" Rio Claro, SP, Brazil, the two smaller animals having hatched recently, the two intermediate animals being 1 year old and the largest animal being 2 years old.The animals were killed by an overdose of Sodium Thiopental (100 mg.kg -1 ), administered intraperitoneally (018/2014 Protocol of the Committee on Animal Use and Ethics, IBIO/UFBA).Subsequently, the lungs were fixed by means of intratracheal instillation of glutaraldehyde 5% in 0.2 M phosphate buffer (pH 7.4, 380 mOsm) at a pressure of 20 cmH 2 O for 12 hours.Following fixation the lungs were removed, stored for up to 10 days in glutaraldehyde 0.5% stock solution in 0.2 M phosphate buffer (pH 7.4; 380 mOsm) for light and electron microscopical processing.

Respiratory system measurements
The following variables regarding the respiratory system have been measured (adopted from Klein et al. (2005): length of the left lung (L le [mm]) and length of the right lung (L ri [mm]).To remove the effect of body size, the ratio of L le /L ri was calculated.

Total volume pulmonary determination
The total lung volume was determined by the Cavalieri method that estimates the total volume and area of cross sections multiplied by the distance between the sections (cruz-orive & weibel 1990).For this purpose, the size of each lung was measured in cranio-caudal direction and ten cross sections were performed from a random starting point and in fixed distances, based on the total length of each lung.A test system was superimposed on the cross sections and the numbers of points covering lung parenchymal and lung lumen were counted (Fig. 1A).The total volume (V L ) of the lungs was given by: V = T.a/p.ΣPi, where "T" is the distance between sequential sections, "a/p" is the area represents by each point and "Pi" is the number of points over the parenchyma and lumen.To estimate the parenchymal and lumen volume separately, the same methodology was applied by counting the points over each structure.
Relative lung volume (V L /M B ), relative parenchymal volume (V P /V L ) and the ratio of parenchymal volume and body mass were obtained by multiplication of the first two parameters ([

Morphometric parameters and sampling
The cranial, medial and caudal regions of the lungs of each specimen were separated for analysis and using a stereoscopic microscope samples of 5 mm 2 of these areas were removed (N = 5) and used for the histological processing for light microscopy and transmission electron microscopy, whereas subsamples of 0.5 mm 2 (N = 5) were used for scanning electron microscopy.

Morphological analysis
Stereomicroscopy.Samples of the cranial, medial and caudal regions were analyzed and photomicrographed using a Leica M165C stereomicroscope coupled to digi-tal camera DFC 450 and the images were used to characterize the internal pulmonary anatomy.
Light microscopy.The samples for light microscopy were dehydrated in a graded series of ethanol with their pleural surface facing down and were systematically rotated and sectioned in order to obtain vertical uniform sections (VUR), according to Howard & reed (2005).Subsequently, they were embedded in methacrylate (His-toresina® Leica).Histological sections of 5 μm were cut using a manual Zeiss Hyrax M15 microtome, mounted on slides, stained with toluidine blue (1%) and photomicrographed using a Zeiss Axio Lab.A1 and capture software Vision Rel.4.8.The images were used for structural characterization and quantitative analysis.
Transmission electron microscopy.The samples for transmission electron microscopy were post-fixed in osmium tetroxide 1%, 0.8% potassium ferrocyanide and 5 mM calcium chloride in 0.1 M sodium cacodylate buffer, pH 7.4 for one hour.They were washed three times for ten minutes each in the same buffer, dehydrated in a graded series of acetone, followed by three successive baths of 100% acetone and subsequent replacement of the acetone and Polybed® resin (1:1) for 6 hours.Afterwards the samples were transferred to pure resin, where they remained until polymerization.
Semithin (0.5 μm) and ultrathin (80 nm) sections were performed using an ultramicrotome Leica EM UC7.The ultrathin sections were stained using uranyl acetate and lead citrate, and analyzed with a transmission electron microscope JEOL 1230.The images were used for ultrastructural characterization and measurement of the distance of the air-blood diffusion barrier.
Scanning Electron Microscopy.The subsamples designated for scanning electron microscopy (SEM) were dehydrated through a crescent ethanol series and passed through drying by Critical Point in a Leica CPD 030.Subsequently, the samples were plated in gold and analyzed by scanning electron microscope JEOL JSM 6390LV.

Quantitative analysis
The quantitative analyzes were performed based on the method of cruz-orive & weibel (1990) and Howard & reed (2005).

Volume density (Vv).
The volume densities of the parenchymal components (faveoli, septa, trabeculae, type I and type II pneumocytes) were estimated by the ratio between the points over the different components and the reference area (Fig. 1B).This way: where, ΣP(cp) gives the sum of the points over the component to be estimated (faveoli, septa, trabeculae, type I or type II pneumocyte, respectively) and ΣP(ref ) gives the sum of the points over the reference space (faveoli + septa + trabeculae + type I + type II pneumocytes).A total of 1.125 micrographs were quantified at a final magnification of 200×.The estimate of the volume density of the trabeculae was given by the ratio between the points on the component to be estimated (muscle or capillary) and the points on the volume of reference (muscle + ca pillary) (Fig. 1C).A total of 20 micrographs were quan ti fied at a final magnification of 400×.

Thickness of air-blood barrier.
To determine the thickness of the blood-air diffusion barrier between the surface of the epithelium and the endothelial cell surface, ten electron micrographs of cranial, medial and caudal regions from three specimens (0.020; 0.319; 2.4 kg) were used, totaling 30 micrographs for each specimen, at a final magnification of 25.000×.On each micrograph the distance of the air-blood diffusion barrier at an angle defined by random from among five different angles (0º, 45º, 90º, 135º, 180º) was measured (Fig. 1D).These were used to obtain the arithmetic mean and standard deviation.The measurements were performed using Image J ® software.

Statistical analysis
The descriptive statistics (mean and standard deviation) and Pearson correlation coefficient and simple linear regression were calculated using the Biostat ® 5.3 software.Linear regression analyses were performed using Graph-Pad Prism 6.A statistical difference was considered significant at P < 0.05.

Pulmonary anatomy
Macroscopically lungs of all Iguana iguana were very similar, showing a pair of saccular lungs of different sizes, with the right lung being longer than the left one (L le /L ri : 0.84 ± 0.03) (Fig. 2A; Table 1), and with short extrapulmonary bronchi fused medially near the region of entrance into the lungs (Fig. 2B).Both lungs were situated in the pleuroperitoneal cavity, ventral to the digestive tract, laterally limited by dorsal ribs, intercostal muscles and the dorsal spine.
In cross section, the lungs were oval-shaped and composed of two chambers, separated by a longitudinal septum, characterizing it as a transitional lung (Fig. 2C).The smaller cranial chamber originated in the medial region and extended to the cranial region and the larger caudal chamber originated in the cranial region and extended through the medial to the caudal region.In longitudinal section two distinct epithelial types were present, the cranial and medial regions with pronounced faveoli and a relatively small lung lumen, and the caudal region com- Cranial chamber (crc), caudal chamber (cac), trachea (tr), right lung (L ri ) and left lung (L le ), faveolar region (fvr), trabecular region (trr), hilus (hi).
Table 1.Body mass (M B ), length of the left lung (L le ), length of the right lung (L ri ), ratio of L le /L ri , total volume of the lungs (V L ), volume parenchyma (V P ), and lumen volume (V Lum ) of Iguana iguana lungs.posed of trabecular epithelium with a relatively large lung lumen when compared to the previous regions (Fig. 2D).The hilus could be found in the cranial region at the termination of the short intrapulmonary bronchi (Fig. 2D).
A clear transitional region could be identified between medial and caudal lung regions with a significant reduc-tion of faveolar epithelia to give way for trabecular parenchyma (Fig. 2D).
The cranial and medial regions (Figs.3B and 4B) were composed of first, second and third order trabeculae, the latter delimiting the respiratory chambers, the faveoli (Figs.3A and 4A).The first order trabeculae were  located closest to the lumen and from the branch of these were formed second order trabeculae, which branched out forming third order trabeculae and were continuous with the interfaveolar septa (Figs.3C and 4C).The trabeculae contained bundles of smooth muscle and were covered by blood capillaries (Fig. 5).
In the caudal region, the faveoli were absent (Figs.6A and 6B), this region being composed of first order trabeculae fused to the pulmonary wall (Fig. 6C), featuring a trabecular parenchyma.

Pulmonary epithelium
Within the respiratory surface ciliated cells, type I and type II pneumocytes could be distinguished, the first one forming the respiratory gas exchange surface while the latter mainly secretes pulmonary surfactant.Ciliated cells were also present in parts of the surface of the first order trabeculae (Figs.7A -D).
The epithelium lining the septa was formed by two cell types characteristic of the gas exchange regions: squamous cells located on the blood capillaries, characterized as type I pneumocytes (Fig. 8A) and cuboidal cells, containing cytoplasmic granules, located in intercapillary spaces and characterized as type II pneumocytes (Fig. 8A).Type I pneumocytes exhibited squamous appearance and consisted of a thicker region containing the nucleus, located in the central region, and a thin re-gion formed by cytoplasmic extensions covering the surface of the capillary and forming the respiratory epithelium (Fig. 8C).
Blood capillaries could be seen on both sides of the interfaveolar septa, exposed to the lumen, forming a double capillary system.The capillaries were separated through the interfaveolar septa and only one side of each capillary was available for gas exchange (Fig. 8B).The air-blood barrier consisted of type I pneumocytes, a basement membrane, and endothelial cells forming the blood capillaries (Fig 8D).
The type II pneumocytes had a cuboid shape and were located between intercapillary spaces.Within the cytoplasm it was possible to notice numerous corpuscles inclusion present near the nucleus or in distal regions (Figs.9A  and 9B).These were formed by concentric lamellae, intensely osmiophilic and resembling lamellar bodies.At the junction regions between the type II pneumocytes it was possible to notice the presence of a junctional complex, consisting of an apical junctional zone and spot desmosomes along the lateral membrane (Fig. 9B).

Lung morphometry
The ratios between the V L /M B , the V P /V L and V P /M B are given in Table 2.While relative parenchymal volume (V P / M B ) remained unchanged by increasing M B , relative lung volume (V L /M B ) in the specimen of 0.020 kg was equal to 335 cm 3 .kg - and in the specimen of 2.4 kg this value was 199.2 cm 3 .kg - , averaging at 277.6 ± 58.83 cm 3 .kg - .The V P /V L in the lower specimen was 0.12 and in the largest specimen, this value was equal to 0.26.On average, the V P /V L was equal to 0.18 ± 0.05.
The volume density of the different components of pulmonary parenchyma of I. iguana is presented in Table 3.The density of faveoli and interfaveolar septa Table 2. Body mass (M B ) and relative total volume of the lungs/ body mass (V L /M B ), parenchymal volume/total volume of the lungs (V P /V L ) and parenchymal volume/body mass (V P /M B ) of the lungs in Iguana iguana.

Animal M B (kg)
V L /M B (cm 3 .kg - ) V P /V L (ml.ml -1 )  was greater in the larger specimen, showing 64.18% and 22.31% of faveolar and septal parenchymal volume, respectively, and least in the smallest specimen, representing 15.12% and 11.03%, respectively.The inverse relationship was found for the trabecular density, which presented the greatest value (74.01%) in the smallest specimen and the lowest (13.6%) in the largest specimen.
The volume densities of capillaries, type I and type II pneumocytes were greater in the largest specimen, representing 13%, 4.56% and 4.63%, respectively, and lowest in the smallest specimen, representing 5.22%, 3.76% and 2.12%, respectively.On average, 43.6 ± 25.5% of the parenchyma corresponded to faveoli, 18.0 ± 5.9% to septa and 38.7 ± 31.6% to trabeculae.In the interfaveolar septa, an average of 9.4 ± 4% (52% of the volume of the septa) corresponded to capillaries, 4.4 ± 1% (24% of the volume of the septa) to type I pneumocytes and 3.9 ± 1% (21% of the volume of the septa) to type II pneumocytes.
The arithmetic barrier thickness of the air-blood diffusion barrier (arithmetic mean and standard deviation) of I. iguana was equal to 0.58 ± 0.10 μm (Table 4).

Discussion
The present study confirms that Iguana iguana possesses lungs of the transitional type, consisting of two chambers, a cranio-medial chamber composed of faveolar parenchyma, and a caudal chamber showing trabecular parenchyma, resulting in an overall heterogeneously dis-

µm 2 µm
2 µm 2 µm tributed respiratory epithelium.This same structural type was evidenced in chamaeleonids (Klaver, 1973(Klaver, , 1977)), agamids (daniels et al., 1994) and other iguanids (wood & Moberly, 1970;crawford & KaMPe, 1971;bennett, 1973).In the latter group, the lungs are characterized by having two separate chambers, a short intrapulmonary bronchus and a caudal region of sac-like appearance (Perry, 1998).According to Perry (1983), although the lungs of the transitional type are formed by two separate chambers, these are considered as a single-chamber type arranged in series, due to the presence of a single communication between the chambers.Regarding the size differences found between the right and left lung in all specimens of I. iguana, the same was verified for other lizards.In Diploglossus warreni, D. fasciatus and D. te nuifasciatus (becKer, 1993), the right lung is smaller than the left, the inverse occurring in Ophiosaurus apo dus and O. harti (Milani, 1894;becKer, 1993).Klein et al. (2005) found for the familiy Teidae a tendency for the right lung to be shorter than the left lung, but no such asymmetry in other lizard families (Cordylidae, Gerrhosauridae, Gymnophthalmidae, Lacertidae, Scincidae, and Xantusiidae).In general, there is a tendency in the partial or total reduction of the lungs in species with elongated body and reduced legs (Perry, 1998), generating larger differences between the size of the lungs.
Regarding the distribution of the parenchyma, all specimens analyzed showed a faveolar parenchyma in the cranial and medial region, with a hierarchical arrangement of first, second and third order trabeculae and trabecular parenchyma in the caudal region, consisting only of first order trabeculae.This same type of parenchymal distribution was described for teiid lizards (KleMM et al., 1979), geckonids (Perry et al., 1989(Perry et al., , 1994)), lacertids (nielsen, 1961) and snakes (lucHtel & Kardong, 1981, Maina, 1989).Perry (1978Perry ( , 1983)), in studies on turtles and varanid lizards, found that more than half of the parenchyma is located in the cranial and medial regions of the lungs.Functionally, the parenchymal distribution in cranial and medial region is related to a greater specialization of these regions to carry out gas exchange, while the more sac-like caudal region does seem to provide a more ventilatory function.However, other roles have been attributed to the caudal (saccular) region of the reptilian lungs (varde, 1951;brattstorM, 1959;Mcdon ald, 1959), such as gas storage or decreasing the elastic work of breathing (Perry, 1983).Intrapulmonary flow measurements undertaken recently in I. iguana lungs suggest the presence of a consistent pattern of airflow inside both the cranio-medial and the caudal chambers, moving air preferentially in one direction over the gas exchange epithelia (cieri et al., 2014).The muscle bundles present in trabeculae act most likely to promote the change in faveoli diameter during phases of pulmonary inflation and deflation, as suggested for other lizards and snakes (Perry & duncKer, 1978;lucHtel & Kardong, 1981). daniels et al. (1994) propose a functional model to explain the role of muscle bundles in trabeculae by contraction and relaxation cycles, promoting an intrapulmonary ventilation mechanism and allowing entry and exit of air in the faveoli during ventilation.
The structure and ultrastructure of the lung epithelium of I. iguana did not vary with body mass and also did not differ with respect to results found in other lizards (KleMM et al. 1979;Meban, 1978;Perry, 1983;Perry et al. 1989), crocodilians (Perry, 1988) and snakes (Maina, 1989(Maina, , starcK et al. 2012)).Lung epithelium was composed of ciliated and mucous cells located on the surface of the trabeculae, type I pneumocytes located on capillaries to promote gas exchange by reducing the thickness of the air-blood barrier (nagaisHi et al., 1964;Maina, 2002), and type II pneumocytes located in intercapillary spaces related to the synthesis, storage and secretion of pulmonary surfactant (daniels et al., 1995; sullivan et al., 2001).
In I. iguana, surfactants may prevent atelectasis and act as an anti-adhesive compound (daniels et al. 1995), preventing the adhesion of epithelial interfaveolar septa during expiration and possibly also during voluntary dives performed by this species (Moberly, 1968).
The anatomical arrangement of a double pattern of capillaries in the interfaveolar septa of I. iguana is found in other air breathers, such as lungfish (Maina & Maloy, 1985), amphibians (goniaKowsKa-witalinsKa, 1978;Me ban, 1977;Maina, 1989), varanid and teiid lizards (Perry, 1983), as well as embryonic stages of the mammalian lung (KliKa & anataliKova, 1978).The arrangement of blood capillaries on both sides of the septa allows only one side of the capillary to exchange gases, i.e., only half the potential surface area is available to participate in gas exchange.On the other hand, a simple capillarity pattern provides a greater surface area for gas exchange, since the whole surface of the capillary takes part in this process (Maina, 1989) being functionally more efficient than standard double capillary nets.Thus, a simple pattern of capillarity is functionally advantageous, being found in reptiles whose parenchyma is more specialized to perform gas exchange (Perry, 1983).
The nature of the diffusion barrier of the lungs of I. iguana showed no morphologic variations compared with those found in other reptiles and other vertebrates (Meban, 1980), consisting of type I pneumocytes, the basal membrane and capillary endothelial cells.
Regarding lung morphometrics in I. iguana, the total lung volume showed a strong correlation with body mass (Table 5).An increase in lung volume as a function of body mass has also been demonstrated for other iguanids (Crotaphytus collaris and Sceloporus poinsettii), teiids (Tupinambis nigropunctatus), varanids (Varanus exan thematicus), crocodilians (Caiman sclerops) and snakes (Thamnophis ordinatus, Boa constrictor and Python re gius) (tenney & tenney, 1970;Perry, 1983;starcK et al., 2012).tenney & tenney (1970) report V L to show an allometric relationship with body mass as M B 0.75 for reptiles, including in their analysis species of lizards, snakes, testudines and crocodilians.However, alexander (1985) already suggested that determining lung volume in isolated dry-fixed lungs, as has been done by tenney & tenney (1970), may lead to erroneous estimates of lung volumes.Lung volume among other vertebrates has been shown to increase nearly isometrical with body mass in mammals as M B 1.05 (staHl, 1967), in birds as M B 1.048 (Maina et al., 1989), and in amphibians as M B 1.05 (ten ney & tenney, 1970).Despite the fact that mammals, , and in the present study I. iguana showed lung volume to scale as log 10 V L = 0.8949.log 10 M B -log 10 0.3271 (Fig. 10A).Taking these discrepancies between intra-and interspecific allometric lung volume relationships among reptiles into account, we gathered data from the literature on lung volume as well as parenchymal volume of lizards with unicameral and transitional lungs and compared these data with our own results (Fig. 10).Despite only few studies having measured lung volume in lizards with sack-like lungs (Table 5) a general trend can be identified.With the exception of Gekko gecko, lung volume increases more isometrically with body mass, since all intraspecific slopes are close to 1.0, and the combination of all data gives an interspecific regression of log 10 V L = 1.027.log-MB -log 10 0.8303, suggesting that lung volume in lizards with unicameral and transitional lungs increases isometrically with body mass and not as predicted by tenney & tenney (1970) as M B 0.75 .The regression obtained for Gecko gecko, may indicate a different scaling relationship in this species, but too few data are available and the regression also shows a very poor fit, needing data over a wider range of body masses to verify the results.
Parenchymal lung volume also increases isometrically with body mass in I. iguana, Lacerta spp.and Tupi nambis nigropunctatus (Table 5, Fig. 10B), but not in Gekko gecko, in the latter species probably due to the rea-sons mentioned before.Interestingly, combing the data for the three species with unicameral lungs (Lacerta spp., T. nigropunctatus, and G. gecko results in a linear regres sion with its slope not being significantly different from the regression of the transitional type I. iguana lung, but the intercepts are significantly different (log-10 V P = 1.083.logM B -log 10 1.835 versus log 10 V P = 1.030.log M B -log 10 1.389), just as could be expected from the general morphology of these lungs, where respiratory epithelia in unicameral lungs are exclusively located on the inner wall of the lung chambers, but in transitional lungs few septa dissect the central lung lumen (Perry, 1998).
The present study suggests that lung volume and parenchymal volume increase in a greater proportion than the metabolic rate of I. iguana, which showed an allometric exponent value equal to 0.73 (Maxwell et al., 2003).The body mass standardized V L showed an inverse relationship, being greater in the smaller animals when compared to the largest iguana.Similar results were observed in Gekko gecko and Crocodylus niloticus (Perry, 1994(Perry, , 1988)), indicating that animals of lower body mass possess greater lung volumes per unit body mass.The strong correlation found between parenchymal volume and body mass in I. iguana, increasing parenchymal density with increasing body size, was caused by a greater internal compartmentalization of the lungs, as indicated by the increasing densities of the faveoli and the interfaveolar septa (Table 3).
The volume density of trabeculae indicates a reduction throughout the body mass increase (Table 3).However, it was observed that the smaller specimens (0.020 and 0.026 kg, respectively) showed a lower density in interfaveolar septa, and may have contributed to the greater values of trabecular density found when compared to animals with larger body mass.The trabeculae were composed of 70% smooth muscle and 18.2% capillaries.The percentages of smooth muscle and capillaries are 75.9% a Among the data given by Tenney and Tenney (1970) only the lizards Crotaphytus collaris, Iguana iguana and Scleroporus poinsettia could be clearly associated to a given V L and M B .
and 5.1% in Tupinambis nigropunctatus, respectively, and in Varanus exanthematicus on average 54% and 16%, respectively (Perry, 1983).The higher bulk density of capillaries, type I and type II pneumocytes in specimens of greater body mass is related to the greater amount of parenchyma, since these components are within interfaveolar septa and the volume density of septa is shown ascending from smaller individuals (0.020 kg to 0.026 kg) to the largest ones (0.319 kg, 0.546 kg and 2.4 kg) (Table 3).This is the first study to determine the volume density of type I and type II pneumocytes within respiratory epithelia of reptiles.On average, the capillaries accounted for 52% of the total volume of septa.Similar values (70%) were found for Gekko gecko, a species with unicameral lungs and double standard capillarity (Perry, 1994).Since the thickness of the blood-air diffusion barrier in I. iguana and other reptiles was found to be similar (Table 4), it is reasonable to assume the occurrence of structural conservation of the pulmonary epithelium allowing gas exchange.Thus, in I. iguana, like in other reptiles, the diffusion barrier seems to be as thin as possible without prejudice to its mechanical stability, and in order to favor gas exchange.However, in a recent study gillooly et al. (2016) found no correlation between barrier thickness and body mass in ectothermic vertebrates.
I. iguana, as well as other vertebrates, meet their metabolic demands for locomotion mainly through aerobic metabolism (Moberly, 1968), which comprises a series of coordinated steps constituting the oxygen cascade (farMer & HicKs, 2000).According to these authors, restrictions on certain points of the cascade can significantly affect gas exchange.wang et al. (1997) found a decrease in pulmonary ventilation and oxygen extraction in I. iguana with the progressive increase in locomotor activity, attributing this to a mechanical constraint caused by the simultaneous occurrence of costal ventilation and mobility (car rier, 1987).The lateral undulations of the body reduce the effectiveness of the costal breathing movements because the intercostal muscles are also used in body movement during locomotion, contracting alternately and not simultaneously, as required during ventilation.farMer & HicKs (2000) found that both ventilatory air flow and pulmonary blood flow are affected in I. iguana by increasing locomotor activity.Ventilation is reduced due to muscular restrictions and blood flow due to pressure increases in the trunk, causing constriction of blood vessels.Thus, I. iguana is unable to expand its locomotor activity associated with adequate ventilation (fraPPell et al., 2002).
With increasing body mass there is a reduction in lung volume per unit of body mass, whereas the parenchymal volume remains constant.This indicates that neonate and juvenile animals have relatively larger lungs per unit body mass than adults, although they show the same proportion of parenchyma, potentiating a greater metabolic rate.These inferences are supported by activity pattern found in this species.swanson (1950) and Henderson (1974) found that juvenile I. iguana are seen predominantly foraging on the ground and have greater metabolic rates than adults (troyer, 1984) and, thus could possess a greater metabolic scope when compared to adults.

Conclusions
In conclusion, I. iguana possess lungs that bring together characteristics considered simple for gas exchange, but are functionally well adapted to the metabolic demands of the species.The functional morphology of the lungs  5. in I. iguana seems to play an important role to meet the metabolic demands, influencing the lifestyle dependent on different body masses.The simple pulmonary structure and potential restrictions between the joint occurrence of locomotion and respiration can be decisive in the adoption of high or low activity pattern, associated with different body masses and their respective niches.Our data, as well as other literature data, suggest that lung volume scales proportionally to body mass, and thereby significantly different from the pattern described previously for reptiles, recommending further analysis of scaling relationships of respiratory system variables among reptiles.

Fig. 1 .
Fig. 1. (A) Cross section of Iguana iguana lungs with point test system used to quantify total lung volume, parenchymal volume and lumen volume.(B) Longitudinal section of the lung parenchyma of the Iguana iguana with test point system used to quantify the volume density (Vv) of the main components of the parenchyma.(C) Longitudinal section of trabeculae with test point system used to quantify the major components of trabeculae.(D) Measurement of the distance of the air-blood diffusion barrier representing the possible angles used to determine diffusion distance.Erythrocyte (er).

→
Fig. 2. (A) Ventral view of Iguana iguana (M B = 2.4 kg) lungs.(B) Dorsal view of the lungs showing the medial fusion of the bronchi (arrow).(C) Cross-section of a lung.(D) Lateral view of the medial half showing the transitional region of parenchyma (arrow).

Fig. 7 .
Fig. 7. (A) Histological section of the trabeculae in the lung parenchyma of Iguana iguana (M B = 0.319 kg), showing cilia on the surface (arrows).Staining with toluidine blue 1%.(B) Scanning electron micrograph of the trabecular surface showing the presence of non-ciliated (arrows) and ciliated regions (asterisks) (M B = 2.4 kg).(C) Scanning electron micrograph of ciliated cells (M B = 2.4 kg).(D) Transmission electron micrograph of cilia in the trabecular surface (M B = 0.319 kg).

Fig. 8 .
Fig. 8. Structural characterization of Iguana iguana lung parenchyma.(A) Histological section of the lung parenchyma in luminal view.Staining with toluidine blue 1% (M B = 0.200 kg).(B) Histological section of the interfaveolar septa indicating the presence of capillaries on both sides (arrows).Staining with toluidine blue 1% (M B = 2.4 kg).(C) Transmission electron micrograph of a type I pneumocyte showing erythrocytes and endothelial cell (M B = 0.026 kg).(D) Transmission electron micrograph of the lung air-blood barrier (M B = 0.319 kg).For abbreviations see Fig. 3.

Fig. 9 .
Fig. 9. Structural characterization of pneumocytes of Iguana igua na.(A) Transmission electron micrograph of a type II pneumocyte with lamellar bodies (*) and small microvilli-like structures, covering a cell with prominent nucleus (n) and endoplasmic reticulum (er) (M B = 0.026 kg).(B) Transmission electron micrograph showing the apical junctional zone (arrow) and spot desmosomes (arrowhead) along the lateral membrane of two neighboring type II pneumocytes (M B = 0.546 kg).For abbreviations see Fig. 3.

Table 3 .
Volume densities (Vv) of major components of Iguana iguana lungs.

Table 4 .
Comparison of the mean thickness of the diffusion barrier in Iguana iguana lungs (mean ± standard deviation) and other reptiles (mean, mean ± standard deviation, or range).

Table 5 .
Linear OLS regressions of relationships of log 10 -lung volume (V L ) and log 10 -parenchymal volume (V P ) with log 10 -body mass (M B ) of different lizard species.N gives the number of individuals used to calculate the regression.