INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

FOR BIRDS, panting and gular fluttering are usually the dominant routes for heat dissipation (9). However, some birds, such as the white-necked raven (when flying), Corvus cryptoleucus (11), a few species of sandgrouse, pteroclidae (14, 15, 26), a few columbidae species (14-16, 20, 21, 27, 28), and the emu, Dromaius novaeholandia (13), are also known to use substantial cutaneous evaporative cooling.

The epidermis is the final barrier for the passage of water to the skin surface (19). It is well known that in mammals the permeability of the cutaneous barrier is formed by keratin filaments and by the deposition of lipid contents of lamellar bodies in the stratum corneum (19, 22). In the avian epidermis, nonpolar lipids may be one of the main components contributing to the permeability of the epidermal barrier, and the absence of keratin filaments in the corneocytes makes them less substantial compared with those of mammals (19, 22). The "loose" structure of the stratum corneum suggests that bird skin might have a lower resistance to vapor diffusion than the mammalian integument (19). In a hot, dry environment a large vapor-pressure gradient is established (16), and the combined effect of this large gradient and the loose structure of the skin enables the transfer of water to the skin surface from which it is evaporated. Over the last decade, it was shown (14-17) that well-developed cutaneous water evaporation (CWE) enables heat-acclimated (HA) rock pigeons to maintain normal body temperature (T_b) without panting or gular fluttering when exposed to temperatures up to 60°C for 4-5 h. Under these conditions CWE was 19.1 ml H₂O · cm² · h¹ at a normothermic T_b of 41.5-42.5°C (14-17). It is likely that the major water source for persistent cutaneous evaporation is the circulating blood. A number of factors may affect the passage of water from the capillary lumen to the extracellular space. First, extravasation of plasma albumin and resultant changes in the colloid osmotic pressure and in the net ionic charge across the capillary wall may provide a major driving force for water flux, mainly via the endothelial intercellular clefts (1, 2, 4, 8, 12). Second, the pericytes that encircle the microvessel wall can promote endothelial gap formation by modifying the shape of endothelial cells and thus increase capillary permeability (3, 7, 25). Third, activation of adrenergic receptors on the endothelial wall can affect capillary permeability (3, 5, 8, 18, 24, 30), and recent studies (17, 21) provide evidence that adrenergic transduction pathways are involved in enhanced cutaneous cooling. The administration of propranolol [a -adrenoreceptor (AR) antagonist] as well as clonidine (₂-AR agonists) activated intensive CWE in these birds. In contrast, the administration of isoproterenol and salbutamol (- and ₂-AR agonists, respectively) prevented enhanced CWE. Isoproterenol also eliminated heat-induced CWE, suggesting adrenergic control of the cutaneous evaporative cooling mechanism. None of the evaporation inducers (adrenergic drugs and heat stress) had any effect on cold-acclimated (CA) pigeons (17, 21).

The purpose of the present investigation was twofold: to study ultrastructural modifications in the vascular bed of the HA pigeon's skin and the resultant increase in microvessel permeability that is associated with cutaneous evaporation and to investigate the mechanism of the adrenergic transduction pathway to this cooling mechanism.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Two different experimental series were carried out on HA (n = 46) and CA (n = 41) adult rock pigeons (Columba livia) with mean body weights of 214 ± 7 (± SE) and 227 ± 9 g, respectively. In the first series we investigated the ultrastructure and permeability of the cutaneous blood vessels after pharmacological treatments and heat exposure. In the second series, the permeability of the skin capillaries was investigated under the effect of propranolol, using a spectrophotometric detection of extravasation of the Evans blue (EB) dye. The T_b and CWE were monitored at several stages during both series of experiments.

Animal housing. The HA pigeons were housed in an environmental room (Hotpack 883-13; ±0.1°C), each pair in a cage measuring 71 × 53 × 48 cm. The CA birds were housed in identical cages in the animal unit. Food and water were freely available.

Acclimation protocol. Heat acclimation was achieved by means of daily 4-5 h exposure to ambient temperatures of 50-60°C, relative humidity 10-15%, from the first day of hatching. The CA pigeons were exposed to T_a of 5-20°C (seasonally) controlled by the air-conditioning system of the animal unit.

T_b measurements. T_b was measured for each bird before the treatment and during the experiment by inserting a telethermistor needle (YSI-513) into the bird 1-2 cm lateral to the pubis.

CWE. Measurements of CWE were carried out on the sparsely feathered abdominal skin concomitantly with a variety of treatments. We used a porometer (Delta-T Devices, AP4), which enabled rapid and accurate measurements of the skin's diffusive resistance within a range of 0.5-20 s/cm, by gently pressing the sensor head on the skin area under investigation. The porometer was calibrated using a calibration kit (supplied by the manufacturer) twice per week or one day before each day of measurements. The CWE measurements were taken concomitantly (but on the opposite side) with skin biopsies used for ultrastructural or capillary permeability studies before the treatment (heat exposure or administration of drugs) and under the effect of the treatment.

Pharmacological treatments. The drugs propranolol (0.3 mg/0.1 ml saline) or clonidine (20 µg/0.1 ml saline; Sigma) were administered intramuscularly to five birds in each group, and CWE measurements were taken under normothermic conditions (23-26°C). Isoproterenol (0.2 mg/0.1 ml saline; Sigma) was administered to five HA pigeons 15 min before the onset of heat exposure, and CWE measurements were taken in T_a of 50°C (17, 21). Drugs were administered before the ultrastructural or microvessel permeability studies.

Electron microscopy. The enzyme horseradish peroxidase (HP, 5000 U; Sigma) in 0.1 ml saline was slowly injected (0.1 ml/min) in a small subcutaneous artery leading to the skin of the abdomen of restrained (by hand) HA and CA birds (6 birds for each treatment and for the control). HP administration was followed either by heat exposure (50°C for 4 h) or by adrenergic drug administration. For controls we used untreated HA and CA birds. Under local anesthesia (hostacain 2%), which did not affect CWE, small skin biopsies (~20 mm²) were taken from the abdomen of each bird at four of the following times: 5, 10, 20, 60, and 240 min (one sample at each of the times; total area of ~80 mm² for each bird) after the treatment (except one bird from which we took 5 samples). Immediately thereafter the biopsies were prepared for electron microscopy (EM) (19). HP was traced using diaminobenzidine (Sigma). Each skin specimen was cut into 50-60 cross sections, 50-70 nm in thickness (Ultratome, Ultratome type 8801A), for examination using EM (Jeol 100 XC).

EB extravasation. We used this method (5, 10) to reinforce the linkage between the onset of CWE, ultrastructural changes, and protein permeability of the capillary wall. In ambient temperatures of 23-26°C, EB (in a dosage of 1.26 mg/0.1 ml saline) was slowly injected (0.1 ml/min) in the pectoral vein of 14 HA and 11 CA pigeons. This dose of EB was found to bind completely to plasma proteins (23), thus allowing the utilization of the dye as a marker for plasma albumin. Immediately thereafter, propranolol was administered intramuscularly to the pigeons. For control, 0.1 ml of saline was injected intramuscularly to nine HA and seven CA pigeons before EB administration. Under local anesthesia (hostacain 2%), small skin biopsies were taken from the abdomen of each bird at 10, 20, 40, and 60 min after drug administration. The skin biopsies were weighed (Mettler H 10 T, 0.1 mg), incubated in formamide at 60°C for 48 h, and EB was extracted by centrifugation (20,000 g, 10 min). The extracted EB in each tube was measured, using a spectrophotometer at 740 nm (Hewlett-Packard 8452 A) (10).

Calculations. CWE values were calculating using the following equation where s (gH₂O/m³) is vapor density on the surface of the skin (100% relative humidity, skin temperature), a (gH₂O/m³) is vapor density in the air (determined from a table in Ref. 12a) according to the temperature and pressure measured that day, and r is the overall resistance of the feathers and skin to water diffusion (as measured and corrected for the skin temperature) measured by the sensor of the porometer (27).

Density of endothelial gaps and fenestrae. To calculate the density of endothelial fenestrae and gaps (gaps/mm² skin), we first calculated the mean skin area for the microscopic sections that were taken from each pigeon. The mean density was then calculated for each pigeon and for the entire experimental group.

Statistics. All the results are presented as means ± SE. Student's unpaired t-test was used to assess statistical significance. P  0.05 was accepted as a significant difference.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

T_b measurements. The T_b measurements of untreated HA (n = 46) and CA pigeons (n = 41) in normothermic conditions were 42.5 ± 0.4 and 42.2 ± 0.6°C, respectively. The T_b values during the treatments are presented in Table 1.

CWE measurement. In HA birds, we measured intensive CWE in response to either heat exposure (T_a 50-60°C) or adrenergic drug (clonidine and propranolol at room temperature) administration. Isoproterenol administration before heat exposure to this group resulted in intensive panting and almost abolished CWE. In the control CA pigeons (administered with saline) as well as in propranolol-administered CA pigeons we measured very low values of CWE (Table 1). CWE ceased 3 h after propranolol administration or 1.5 h after the administration of clonidine. Significant differences in CWE were found between drug-treated HA and CA birds (P < 0.01) and between the control (saline administered) HA and drug-treated HA pigeons (P < 0.01). There were no significant differences between the CWE values of the treated CA pigeons and those of the control groups.

EM. The subcutaneous continuous capillary of the pigeon is similar to that of mammals. This capillary was observed in the skin of untreated or isoproterenol-treated HA birds and in CA pigeons. As shown in Fig. 1, A and B, the common microvessel wall consists of one or more endothelial cells separated from each other by intercellular clefts and one or two pericytes encircling the capillary wall. Five minutes after the onset of heat exposure or after clonidine or propranolol administration, some capillaries exhibited intracellular fenestrae-like formations, as shown in Fig. 2. This micrograph was taken 5 min after propranolol administration, but the intracellular fenestrations were observed for ~1 h after the initiation of treatment. Ten minutes after the onset of heat exposure, we also found intracellular gaps in the endothelial wall, as clearly shown in Fig. 3. It is likely that the intercellular gap is an extended endothelial cleft, as shown in Fig. 4. This micrograph was taken 10 min after exposure to a T_a of 50°C. Gap formation, however, was observed during the entire duration of heat exposure (4 h). In both cases (heat exposure or drug administration), the gaps were 0.1-1.0 µm in diameter. The fenestration and gap densities with exposure time are summarized in Table 2. Both endothelial fenestrations and gaps appeared within 5 min of the onset of propranolol administration and could not be detected after ~1 h. The effect of clonidine on fenestration and gap appearance was similar, although the fenestrae could no longer be detected 60 min after the administration of the drug (Table 2). The differences with time, within each group, and between the experimental groups were usually insignificant due to high standard error values.

View larger version (127K): [in this window] [in a new window]   Fig. 1.   Cross section of capillary of continuous type [taken from untreated cold-acclimated (CA, A; scale bar = 2 µm) and heat-acclimated (HA) pigeons (B; scale bar = 1 µm)], which is dominant type in skin. One pericyte (P) with its nucleus (N) or more encircle endothelial wall (ET). Continuity of capillary wall is interrupted by narrow clefts (C). Capillary lumen (L) contains an erythrocyte (EC) and dark traces of horseradish peroxidase (HP). Capillaries in HA pigeon were usually larger in diameter and more flattened compared with capillaries in CA pigeon.

View larger version (147K): [in this window] [in a new window]   Fig. 2.   Micrograph taken 5 min after propranolol administration shows thin ET of a capillary with fenestrae (F). Fenestra is an intracellular aperture ~50-80 nm in diameter closed off by a diaphragm (area enlarged, a).

View larger version (132K): [in this window] [in a new window]   Fig. 3.   Intercellular gaps (G), ~0.1 µm wide, were formed in ET 10 min after heat exposure. Entire left half of micrograph is occupied by an EC and its N. Note small but obvious swelling in EC membrane that has invaded gap zone (area enlarged, a).

View larger version (123K): [in this window] [in a new window]   Fig. 4.   Formation (or disappearance) of G in HA pigeon, ~0.5 µm wide, 10 min after onset of heat stress (50°C) is presented in this micrograph. This gap is clearly an augmentation of C. M, mitochondrion.

We also observed rearrangement of the pericytes encircling the endothelial wall. It is evident from Fig. 5 that after heat stress or adrenergic drug administration, the arrangement of the pericytes allows the free passage of plasma proteins and water via the fenestrations and the gaps to the interstitial space. Pericytes in this particular rearrangement were observed in 78% of the opened capillaries.

View larger version (173K): [in this window] [in a new window]   Fig. 5.   Capillary seen 10 min after propranolol administration. Position of P opens up a free passage from capillary L via intercellular G (dashed line; area enlarged, a) to interstitial space. For comparison see Fig. 1, A and B.

To follow the fate of plasma proteins during intensive CWE, the enzyme HP was used as a marker. The data obtained are presented in Figs. 6 and 7. Twenty minutes after the administration of HP, we found dark enzyme traces outside the vessels between nerve cells in the dermis. Within 60 min the protein molecules had extended into the upper layers of the epidermis. Neither of the structural changes described were observed in the saline-treated (control) or ISP-treated HA birds, nor in the control or treated CA pigeons, and no traces of HP were found in the interstitial compartment.

View larger version (189K): [in this window] [in a new window]   Fig. 6.   One hour after onset of heat exposure, dark traces of HP originating in capillary bed of dermis (D) can be seen in between epidermal cells. SC, stratum corneum; SI, stratum intermedium; SB, stratum basal; ET, endothelial cell.

View larger version (142K): [in this window] [in a new window]   Fig. 7.   Twenty minutes after initiation of heat exposure, HP molecules were found between dermal neurons. A, axon; M, myelin.

Extravasation of EB. Figure 8 shows the changes in EB extravasation in response to propranolol administration. Under the effect of propranolol, the rate (expressed by the slope of the line) of EB extravasation was faster in the HA pigeons than in the CA pigeons (8 and 16 min, respectively, to attain 50% of maximal concentration). Within 40 min EB concentration ([EB]) had reached its maximal value in both groups, and it was clearly demonstrated that 40 min after propranolol administration, the [EB] in the skin of the HA pigeon was two times higher than in the CA pigeons (P < 0.01). In contrast, propranolol treatment of CA birds did not cause a significant increase in [EB] compared with the control CA group (treated with saline). A significant difference (P < 0.01) was found between the control HA and CA after saline administration.

View larger version (19K): [in this window] [in a new window]   Fig. 8.   Evans blue (EB) extravasation in skin of CA (triangles) and HA (squares) pigeons treated with either saline (open shapes) or propranolol (closed shapes) in room temperature was observed over a period of 60 min. Significant differences (P < 0.01) were found between propranolol- and saline-administered HA pigeons. Differences between 2 CA groups were also significant (P < 0.04). Also, significant differences (P < 0.01) were found between HA and CA birds treated with propranolol. ***P < 0.01, significant difference within each experimental group. Numbers in parentheses are number of birds in each group. Results are presented as means ± SE, and absolute EB concentration (%) from each skin sample was corrected for sample weight (in mg). Results are in agreement with results presented in Fig. 7.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The innovation of the current study is our finding that there are significant ultrastructural changes in the microvessel wall of HA rock pigeons. These changes are manifested by increased leakage of plasma proteins from the capillary wall and coincide with intensive CWE. Moreover, an adrenergic transduction pathway controls these phenomena.

In birds as well as mammals, cutaneous capillaries are predominantly of the continuous type (6, 22), as we also found in the skin of all pigeons of all experimental groups. Interestingly, heat exposure and adrenergic drugs both provoked the appearance of fenestrated capillaries in the skin of the HA pigeons. A novel finding was the formation of intercellular gaps in response to both of the imposed CWE inducers (heat exposure or drug). It is probable that the intercellular endothelial gaps are extended intercellular clefts, as demonstrated in Fig. 4. These gaps were 0.1-1 µm in diameter, wide enough to allow extravasation of plasma proteins from the capillary lumen to the interstitial compartments (the albumin molecule is ~40 Å). We suggest that the transiently formed fenestrae and gaps observed in heat-stressed HA pigeons play an important role in the passage of proteins and water from the capillary bed to the interstitium, providing water for CWE. This hypothesis is supported by the well-established role of fenestrated capillaries in the kidneys and gut. In these organs there is an intensive turnover of water and solutes between blood vessel lumen and tissue, and the endothelial fenestrae serve as a sievelike organelle, allowing the rapid transfer of water and solutes (6, 7). A similar role is attributed to the intercellular gaps, although in the context of pathological events, such as edema caused by inflammation or asthma (7, 18, 25, 29). This investigation provides evidence for a physiological function for these gaps. We also demonstrated that endothelial -ARs are involved in the CWE mechanism. Baluk and McDonald (3) and Zink et al. (30) showed that stimulation of endothelial -AR (by isoproterenol or formoterol) caused relaxation of the endothelial cells, resulted in constriction of the intercellular gaps, and decreased permeability of the endothelial wall. The experiments performed with isoproterenol (Table 2) demonstrated that isoproterenol stopped the heat-induced CWE (see also Ref. 21) and the formation of structural changes in the HA endothelial cells, supporting the involvement of an adrenergic transduction pathway in CWE. It was shown in previous studies (17, 21) that the adrenergic drugs used in the present study induced or stopped CWE in HA pigeons when administered locally (propranolol) or systemically (propranolol, clonidine, and isoproterenol). This was associated with vasodilation (propranolol, subcutaneously or intramuscularly) (17, 21) and alterations in heart rate (propranolol intramuscularly) (21). Also, in preliminary investigation we found that both propranolol and clonidine induce CWE, coincidentally with augmented arterial blood flow and a small reduction in venous blood flow, in the cutaneous abdominal vasculature of HA pigeons. This implies an augmented arterial-venous resistance difference, leading in turn to increased hydrostatic pressure in the microvasculature, thus accelerating water passage via the endothelial wall. Altogether, these findings further confirm the involvement of adrenergic signaling in the CWE cooling mechanism through a local effect on the endothelial wall and via the myocardium. We observed changes in the pattern of the pericytes encircling the endothelial wall. It was reported that in mammal capillaries, histamine-induced gap formation leads to blockage of the area adjacent to the gap by pericytes (3, 7). As shown in Fig. 5, heat stress or drug administration led to a rearrangement of the pericytes, allowing the free passage of plasma proteins and water into the interstitial space via the gaps and fenestrae. In contrast, no similar structural changes were found in the vasculature of the integument in isoproterenol-treated birds and in untreated HA birds. It is evident from Figs. 6 and 7 that the response to both treatments is indeed a leakage of proteins, such as the enzyme HP, from the microvessel lumen to the interstitial space of the adjacent tissue. Twenty minutes after injection, HP was found between neurons (Fig. 7) and fat cells (not shown) in the dermis. It took 60 min for the HP to diffuse from the dermis across the basal membrane and into the epidermis, where there is a complete absence of blood vessels.

The EB results support these findings. Propranolol administration caused a marked increase in the leakage of EB from the capillary lumen in HA pigeons. This increase in EB extravasation was stabilized 40 min after propranolol administration, presumably by EB-labeled albumin drainage via the lymphatic system (as the dose of EB we injected fully binds to plasma albumin) (23). This intensified plasma extravasation may establish adequate colloid osmotic pressure gradient and altered net ionic charge of the molecules on both sides of the capillary wall, to provide a major driving force for water efflux, until equilibrium is achieved. The ultrastructural alterations observed in the present study are very well correlated with the EB extravasation and with the CWE data in Table 1 [and previous studies (14-17, 21)]. The onset of CWE occurs 10-15 min after the initiation of heat exposure or propranolol administration and reaches its maximum rate within 30-40 min. The structural changes observed exhibit similar temporal dynamics. We therefore suggest that on heat stress, endothelial gap formation and the increased intracellular fenestration are specific acclimatory responses, essential for CWE.

Perspectives