Protein, or colloid, osmotic pressure (COP) is of sufficient magnitude to exert a powerful effect on Starling forces across the capillary wall. Although avian skin lacks active mechanisms for fluid transmission, such as sweating, it is now hypothesized that passive oncotic forces may regulate fluid flow and distribution in the skin and related phenomena. In this study, serum protein profiles, as well as COP in serum (COPs) and in suction blister fluid (COPsbf), were determined in juvenile, young and adult female and male chickens. For assessing COP, the Ott-Ahlqvist equation was applied. This equation can formulate the effects of multiple serum proteins on COP, according to their molecular mass. The COP values determined in chickens were lower than those previously found in mammals. COPs increased with age in males, and was higher in adult males than in adult females. In contrast, COPsbf decreased with age in females, although it was better preserved in adult males. The age-dependent decrease in COP was associated with a parallel decrease in the concentration of α1-globulin and albumin, and a positive correlation between packed cell volume (PCV) and COP was found. It is concluded that ageing affects Ott's COP around the vessel wall, and that an oncotic mechanism preserves plasma volume. The preservation of COPsbf in males suggests better maintenance of the interstitial ground substance. Females seem to lose more of a tissue's COPs counteracting force than males, probably as a result of gender-specific changes in the composition of the interstitial matrix. Further studies are required to elucidate the structure and function of avian lymphatics and their role in skin rheology.
The skin has fundamental influence on the control of whole body homeostasis by being the largest organ of the animal body, and by holding a substantial volume of the total extracellular fluid (ECF) (Aukland and Nicolaysen, 1981; Levick, 1991; Aukland and Reed, 1993; Holliday, 1999). Studies in various animal species have revealed that about half of the total ECF volume in adult animals is located in the skin and the connective tissue interstitium; in juvenile animals, because of the higher surface-to-volume ratio, it is about 5% more (Holliday, 1999). In birds, the skin and its unique functional plasticity have shown to be crucial in adaptational responses to various stressors, such as heat, cold and dehydration, which all involve modifications in water transmission between the blood, tissues and the environment (Arad et al., 1983; Marder, 1983; Marder and Ben-Asher, 1983; Menon et al., 1986; Arad et al., 1989; Menon et al., 1989; Peltonen et al., 2000; Muñoz-Garcia and Williams, 2008). Characteristically, thermal fluid transmission between a bird and its environment does not involve exocrine glands, and could be characterized as a passive mechanism, driven mainly by physical factors (Ophir et al., 2003). However, fluid and electrolytes may be stored in the skin, probably as a part of various transmission processes. In mammals, the skin of humans and rats has been shown to store Na+ (Heer et al., 2000; Titze et al., 2003; Titze et al., 2004). In avians, studies in chickens using suction blister fluid (SBF) have revealed an accumulation of monovalent cations (Peltonen et al., 2006). In both cases, the accumulation of electrolytes took place without a significant influence on osmotic pressure. In rats, accumulation has been associated with the storage capacity of the interstitium, which is, in turn, regulated by processes involving the inhibition and disinhibition of glycosaminoglycan (GAG) chain polymerization (Titze et al., 2004). GAGs, together with proteoglycans, glycoproteins and the structural proteins collagen and elastin, form the highly organized extracellular matrix, which endows skin with its water-binding properties (Stern and Maibach, 2008). A similar mechanism for controlling skin storage capacity may be feasible also in chickens, owing to high concentrations of GAGs in their skin (Nakano and Sim, 1989; Nakano and Sim, 1991).
The osmotic pressure of a solution reflects its colligative properties and its behaviour when it is separated from another solution by a semipermeable membrane. About 95% of total osmotic pressure is built by inorganic electrolytes, 4-5% by glucose, urea and large-molecular-weight organic compounds, such as GAGs, and only about 0.5% by proteins (Guyton and Hall, 2006). Even though most of the total osmotic pressure is built by inorganic electrolytes, they do not exert osmotic pressure across the capillary wall in tissues; they are dissolved in water and move freely with it. In contrast, proteins exert a powerful effect on the convective fluid flow. This is because of the finite permeability of capillaries and the exclusion of proteins by the endothelium (Levick, 1991). Holding to the classical Starling hypothesis (Starling, 1896), fluid shifts across the capillary endothelium respond only to changes in the balance between protein osmotic pressure, i.e. difference in COP, and hydrostatic pressure (Boulpaep, 2009). However, discrepancies between the experimental data and the classical Starling predictions have resulted in a revised model in which the gradient of the Starling forces exists across the glycocalyx, not the endothelium. By this revision, the fluid flow is smaller than that predicted by Starling for bulk driving forces and, for a given driving force, the fluxes are greater for filtration than for absorption (Levick, 2004; Michel, 2004; Curry, 2005). The Donnan forces, tending towards electroneutrality, further emphasize the role of plasma proteins in fluid flow: owing to the net negative charge carried by proteins in normal pH, they affect COP by retaining cations and repelling anions (Hitchcock, 1924). Recently, protein charge has been shown to account for one-third of the total COP in rats (Rügheimer et al., 2008). Once filtrated, water will occupy the interstitial ground substance by binding to collagen and GAGs, especially to hyaluronan (reviewed by Stern and Maibach, 2008). These molecules maintain the space and make the extracellular fluid for a gel, in which water diffuses rather than flows (Guyton and Hall, 2006). Simply for geometrical reasons, collagen and GAGs exclude albumin and other plasma proteins from the densely packed gel, creating an osmotically, as well as rheologically, heterogeneous interstitium (Ogston and Phelps, 1961).
Now, a hypothesis is put forward that the variation of Starling forces across the capillary wall may function as a regulatory factor in the response to various homeostatic imbalances in birds of different age and gender. Homeostatic response often includes the transmission and holding of fluid and electrolytes, which obviously does not involve strictly regulated secretory mechanisms in the practically glandless avian skin. However, there are some previous findings to support the special role of COP as a dynamic regulatory variable. Firstly, plasma albumin and globulin profiles change considerably during development, especially in females. Early electrophoretic studies in Galliform species show that plasma protein compositions differ between females and males of the same species more than those between the males of different species (Sibley and Johnsgard, 1959). Secondly, the relative distribution of total body fluid is influenced by gender and age. In chickens, as the relative body fluid content (percent of body mass) decreases by age, the relative intracellular fluid increases and the ECF decreases, both in blood and interstitium (Sturkie, 1986). In some species, female blood tends to be also more diluted than that of the males, as indicated by lower values of packed cell volume (PCV) (Hebert et al., 1989). PCV seems to increase with age, even though this effect may be dependent on gender (Mitchell and Johns, 2008). Thirdly, the avian lymphatic system differs structurally, and probably functionally, from its mammalian counterpart. In mammals, lymph circulation is the major unidirectional pathway for returning proteins and excess fluid back to plasma, and for restoring the COP of the interstitial fluid (Aukland and Nicolaysen, 1981; Levick, 1991; Aukland and Reed, 1993; Levick, 2004; Michel, 2004). In birds, the lower numbers of lymphatic vessels and the less frequent valves in them (King and McLelland, 1975; Rose, 1985) imply a reduced capacity to drain fluid from tissues.
In order to elucidate the possible influence of the Starling COP difference on fluid transmission and accumulation in avian skin, we assessed, for the first time, the COP of serum and interstitial fluid in female and males chickens, and followed these values during the development of a number of individuals. The COP values were calculated by using the equation of Ahlqvist (Ahlqvist 2004), originally based on the experimental results of Ott (Ott, 1956), which take into account the whole spectrum of plasma protein fractions and, thus, the variability of plasma protein profiles during ageing.
MATERIALS AND METHODS
Serum protein profiles and Ott's COP in blood and in interstitial fluid were determined in female and male White Leghorn chickens (Gallus gallus domesticus Linnaeus 1758) at three successive ages: 10 weeks (juvenile), 20 weeks (transition to sexual maturity) and 30-33 weeks (sexual maturity). Birds were sexed and housed in 20 m2 rooms in ambient temperature of 20-23°C. Birds were fed with food manufactured for juvenile, growing and adult chickens. The onset of laying took place at the age of 20 weeks. At the age of approximately 33 weeks, all females had regular laying cycles of one egg per day. During the study, body mass increased almost linearly by 52% in females and by 53% in males. Birds were kept constantly under a steady 12 h:12 h light:dark photoperiod. All the experiments conformed to the guidelines for proper animal care and use, and were authorized by the local ethical committee.
The SBF and blood samples were collected under injectional anaesthesia by intramuscular ketamine in combination with xylazine in doses of 20 mg kg-1 (Ketalar, Pfizer Animal Health, Helsinki, Finland) and 5 mg kg-1 (Rompun, Bayer, Leverkusen, Germany). Blood samples were taken at the same time of the light:dark cycle and SBF samples within a two-hour time range. SBF was collected in vivo by the method first developed for humans by Kiistala and Mustakallio (Kiistala and Mustakallio, 1964). Individual suction cups were attached on the area of thoracic apterial skin devoid of large cutaneous blood vessels. Blood was removed from the brachial vein to 500 μl vials through a 21-22G needle. Serum was separated and collected after 1-2 h of incubation at room temperature and centrifugation. All samples were frozen for storage. PCV was determined immediately from fresh venous blood using 10 μl microhematocrit vials. Separation of proteins was carried out by agarose gel electrophoresis (SPE P/N 655900, Beckman Coulter, Fullerton, USA), according to the manufacturer's instructions. Quantitative results were obtained by software for quantitative gel analysis (ImageJ 1.36b, Wayne Rasband, National Institutes of Health, USA). Total protein (TPR) was determined by BCA Protein Assay (Pierce, Rockford, IL, USA), based on the biuret reaction by proteins, sensitized with bicinchoninic acid.
Equations and calculations
To determine the protein osmotic pressure (Π) of serum, we used an equation by Ahlqvist (Ahlqvist, 2004). Eqn 1 formulates the effects of each protein fraction on the Π of serum. These effects were first presented by Ott (Ott, 1956) and were later modified by Guyton (Guyton, 1981). Eqn 1 states that the effect of globulin concentration on Π is dependent on the molecular mass (Mr). It was revealed that the same concentrations of albumin or of various globulin fractions alone contribute to Π in the order: α1-globulin, albumin, α2-globulin, β-globulin and γ-globulin (Ahlqvist, 2004).
The first-order terms represent the van't Hoff law for osmotic pressure, and the square terms represent the deviations from that law caused by Donnan effects and interactions between proteins.
One-way analyses of variance (ANOVA) were used for statistical analyses. If the overall P-value for statistical significance was <0.05, various post hoc tests were used. These were the test for increasing and decreasing trends, based on one-way analyses of variance, and Tukey's or Bonferroni's tests for multiple comparisons. Selected data were analyzed also by Pearson's correlation.
Electrophoretic analysis of serum and suction blister fluid proteins
The electrophoretic pattern of serum and SBF consisted of seven protein fractions: transthyretin, albumin, α1-, α2-, β1-, β2- and γ-globulin. Transthyretin was present in both serum and SBF, and there was gender-specific variation by age in serum in male birds (one-way ANOVA, P<0.01): the percentage of total protein in serum peaked in 20-week-old male birds, the value being 1.3% and the mean concentration being 0.60±0.044 g l-1. No such variation was observed in females. Densitometrical profiles of serum proteins during development to sexual maturity are shown in Fig. 1. The most relevant proteins for COP, albumin and α1-globulin, varied differently with age. Although albumin concentration (g l-1) did not change by age and its concentration was independent of gender (Fig. 2C,D), α1-globulin concentration decreased in females, but not in males (Fig. 2A,B). As expected, the total amount of globulin in serum increased rather steeply in females (Fig. 1C,E; one-way ANOVA, P<0.001, post hoc test for increasing trend, slope 5.89, R2=0.545, P<0.001). Globulin concentration increased also in males, but levelled after the transitional phase (Fig. 1B,D; post hoc test for increasing trend, slope 1.81, R2=0.193, P<0.05). To summarize, the various serum globulins revealed the following differences in their profiles: (1) α1-globulin concentration was higher in young adult and adult males than in females of the same age (one-way ANOVA, Tukey-Kramer post hoc test, P<0.001 in both age groups); (2) β1-globulin concentration was higher in laying young adult and adult females than in males of the same age (one-way ANOVA, Tukey-Kramer post hoc test, P<0.05 for 20- and P<0.001 for 30-week-old birds); and (3) γ-globulin concentration was significantly higher in adult females, but not in juvenile and young adult females, than in males of the same age (one-way ANOVA, Tukey-Kramer post hoc test, P<0.01 for 30-week-old birds). Serum β1-, β2- and γ-globulin concentrations increased with age in females, whereas α1- and α2-globulin concentrations decreased or were stable. Only β2-globulin concentration increased with age in males; other serum proteins remained unchanged.
Neither albumin nor the total globulin concentrations in the SBF differed between genders with age (Fig. 1). However, in females the SBF versus serum ratios ([SBF]/[S]) for both protein fractions declined at the onset of laying and levelled after that: for albumin the decrease was about 20% (one-way ANOVA, post hoc test for decreasing trend, slope –9.24, R2=0.26, P<0.05) and for globulin was 28-30% (one-way ANOVA, post hoc test for decreasing trend, slope –15.45, R2=0.44, P<0.01). In males, albumin and total globulin concentrations in the SBF remained stable during development. A gender-specific difference was observed in α1-globulin and albumin concentrations (Fig. 2), which were higher in adult males than in adult females (one-way ANOVA, Tukey-Kramer post hoc test, P<0.01). In female SBF, both α1-globulin and albumin concentration showed an age-dependent decreasing trend (Fig. 2A,C).
Ott's protein osmotic pressures of serum and suction blister fluid
The calculated mean Ott's COPs values increased with age in males (one-way ANOVA, P=0.022, post hoc test for increasing trend, slope 1.3, R2=0.24 and P<0.05). In females, no such trend was observed (Fig. 3). By Bonferroni's post hoc test for selected pairs, mean COPs values were shown to be different in transitional and adult, but not in juvenile, birds (P-values <0.05). As expected, COPsbf valueswere lower than corresponding COPs values, varying between 35 and 55% of that of the serum. The mean COPsbf values of female and male chickens of varying ages were significantly different (one-way ANOVA, P=0.004). However, although a significant declining trend in females was shown (one-way ANOVA, post hoc test for linear trend, slope –1.20, R2=0.49, P<0.001), COPsbf values in males remained stabile (Fig. 3).
Derived from the COP data, the Starling oncotic gradient tended to increase with age in females, as shown by the variation of mean ΔCOP values with age (one-way ANOVA, P<0.001, post hoc test for linear trend, slope 2.1, R2=0.51, P<0.001). Generally, the oncotic gradient was better preserved in adult males (Fig. 3, insert).
PCV and COPOtt
The mean PCV value averaged 33.7% (±0.88%) in young adult, and 29.7% (±0.47%) in adult female chickens. The corresponding values in males were significantly higher and showed less variation, being 50.1% (±0.13%) and 47.7% (±0.13%), respectively (one-way ANOVA, P<0.001). Derived from the PCV and the corresponding Ott's COP data, intravascular and extravascular oncotic pressures decreased by hemodilution (Fig. 4). However, the decrease was not parallel, resulting in a steeper oncotic gradient in more hemoconcentrated blood.
Plasma protein composition has been shown to be both an age- and a gender-dependent variable in chickens, and it was hypothesized that this variability would reflect on Starling forces affecting fluid transmissions between the blood and skin tissue. The Starling principle states that transendothelial filtration is driven by capillary hydrostatic pressure and interstitial COP, and that the counteracting absorptive force is exerted by intravascular COP and interstitial hydrostatic pressure. In the present study, the Ott's protein osmotic pressures of serum and SBF, calculated from multiple protein fractions, changed reciprocally with age and showed gender-related differences at sexual maturity. This oncotic component alone implies an increased capacity of the blood to hold fluid and protect plasma volume with age. At the same time, the counteracting interstitial forces appear to weaken. However, males seem to be better able to preserve the oncotic gradient across the vessel wall by preserving COPsbf. Functionally, this would mean that the skin of adult male birds is able to compensate for a higher increase in net transcapillary filtration than the skin of laying females. In the skin, compensation would depend on the fluid-binding capacity of the tissue, on the composition and constancy of the ground substance, and on the capacity of the lymph system to accumulate or drain fluid. There is some evidence to support the better compensation observed in males. Visual inspection of the male skin reveals ‘a young or hydrated skin’, as it is less transparent and its stereoscopic effect (‘relief’) is finer and less visible than that of coeval females, similar to humans (Zahouani and Vargiolu, 2004). Male skin is also thicker and more resilient (Peltonen et al., 2006; Peltonen and Mänttäri, 2008), implying a greater mass of collagen and elastic fibres in the ground substance. Furthermore, GAGs are present in high concentrations in males, especially in the combs and wattles, hyaluronan being the main species (Nakano and Sim, 1989; Nakano and Sim, 1991).
The present results show a negative correlation between age and albumin and α1-globulin concentrations in SBF in females, parallel to the COP decrease, whereas no change was found in males. In rodents, it has been suggested that the decrease in general perimicrovascular protein content with age is associated with the loss of maintenance of the interstitial space (Barber et al., 1995). Conversely, the preservation of the interstitial albumin concentration during ageing seems to be connected with the tissue space becoming increasingly inaccessible to albumin (Parameswaran et al., 1995). Thus, in females the weakening of protein osmotic activity suggests a susceptibility to age-dependent modulation of the interstitial space available for plasma proteins. In humans, the ground substance is modulated as collagen becomes more sparse, less soluble and thinner, and GAGs become aberrantly localized, interfering with normal water binding irrespective of their amount (Waller and Maibach, 2006). The volume of the interstitial fluid increases, as does the amount of ‘free’ water that binds to itself, rather than to other components of the interstitium (Wright et al., 1998). So far, the protein exclusion effect per se has not been studied in birds.
Estrogen is an important modulator of the interstitial ground substance, as it stimulates collagen production and prevents its loss, and increases skin thickness and elasticity via its effects on water-binding capacity and GAG content (reviewed by Verdier-Sévrain et al., 2006; Farage et al., 2009). Hence, in egg-laying chickens estrogen may be involved in the short-term regulation of skin composition and function owing to its diurnally oscillating secretion pattern (Johnson and van Tienhoven, 1980). The cellular and subcellular mechanisms for the ‘collagen (or GAG) effect’ of estrogen are complex and, besides estrogen, other sex steroid hormones and their receptors may also be involved. Estrogen concentration in chickens increases markedly just before sexual maturity, peaks during the laying period and drops dramatically only at 70 weeks of age (reviewed by Beck and Hansen, 2004). Therefore, it could be speculated that the age-related decrease in COPsbf found in the present study may not be related to circulating estrogen, but to a decrease in the density of estrogen receptors (Hansen et al., 2003). The well-established downregulating effect of progesterone on estrogen receptors may diminish the estrogen effect in laying females, which display increased progesterone concentrations (Johnson and van Tienhoven, 1980; Herremans et al., 1993; Gahr, 2001). The lower levels of circulating progesterone in younger females and males may enhance the effects of estrogen and result in the higher COPsbf found in these birds.
The present serum COP values in chickens are considerably lower than those reported in mammals. The TPR of birds is generally lower (Carpenter et al., 2001), and thus is the most evident cause for the lower COP values. Depending on age and gender, the values in the present study varied between 33 and 55 g l-1, being maximally 35% below the mammalian level. However, it was found that the causal relationship between TPR and COP differed between genders. In juvenile birds, having uniform fractional protein profiles, males had higher COP values than females in association with their higher TPR, whereas, in the transitional and adult phase, male serum COP peaked above that of females despite having a lower mean TPR. Based on the work of Ott (Ott, 1956), the greater protein osmotic force of serum is probably due to the higher fraction of α1-globulin in the male blood. According to the computed model of COPOtt at body temperature in humans, at constant albumin concentration the influence of α1-globulin on COP is twofold compared with the influence of γ-globulins (Ahlqvist, 2004). In concert with TPR and α1-globulin, the relative amount of plasma may partly explain the observed differences in COP between genders. Even though the total volume of plasma could be estimated to be greater in male chickens in all age groups (Sturkie, 1986), the relative plasma volume in the present study is roughly 20% lower. Because the plasma (the loose intercellular ground substance of blood) provides less available space for proteins in males, it may function more efficiently in maintaining adequate blood volume than does the more diluted plasma in females.
To our knowledge, interstitial COP has not been measured in chickens. However, direct measurements of plasma carried out in 5- to 8-week-old male chicks (McDonagh and Gore, 1982) have shown approximately 1 mmHg lower values than our present calculations in juvenile males. As the male birds used in this earlier study were two to five weeks younger, the age-dependent increase in the TPR may partly explain the higher values in the present study. Otherwise, the relationship of plasma TPR and measured COP in young male chicks was found to be similar to that found by Ott (Ott, 1956) in humans.
Owing to the insufficient knowledge of COP in birds, the comparison between calculated COPOtt and measured COP cannot be made. In other species, such as in horses, the calculated COP values have been shown to differ from the measured values when the calculations have been based on total protein or albumin and total globulin concentrations, ignoring the effect of low-molecular-weight globulin fractions (Boscan et al., 2007). In humans, the difference between the measured COP and Ott's COP has been estimated to be ±4.5% when rare conditions, such as analbuminemia and nephrotic syndrome, have been excluded (J. Ahlqvist, personal communication). The calculated interstitial Ott's pressure is probably underestimated because of the lack of GAG influence. As yet, the impact of GAGs on COPsbf is not known.
Mobilization of extravascular fluid
The early argument of a greater capacity of birds to mobilize extravascular fluid (Djojosugito et al., 1968; Wyse and Nickerson, 1971) was based on findings of a better tolerance of avian species compared with mammals to severe blood loss (Kovách et al., 1969). Since that, evidence has been accumulated on the ability of birds to tolerate also other severe homeostatic fluid imbalances. Birds seem to stand extremely well osmotic stress during severe thermal dehydration (Arad et al., 1989), as well as dehydration stress at high altitudes during long-distance migration (Landys et al., 2000) or in the desert habitat (Marder, 1983; Muñoz-Garcia and Williams, 2008). Alternatively, in the condition of hydrothorax, the pathological effect of local overhydration on respiration may be avoided by shifting the excess fluid locally from the abdominal air sacs to the microvilli of the adjacent mesothelial cells (Weidner, 2000). A good capacity to mobilize fluid would mean either effective lymph flow in order to return the ultrafiltrated, proteinaceous fluid back to circulation, or effective absorption of fluid back to the venous capillaries. The present results suggest that the effectiveness of the absorption mechanism at the venous end of the capillary network may be related to age. However, little is known about the effects of the various unique characteristics of the avian circulatory and lymph system on the fluid flow between capillaries, tissues and lymphatics. These characteristics include a more efficient heart and a higher systemic blood pressure than mammals, a higher blood pressure of males than of females (Sturkie, 1986; Ruiz-Feria et al., 2004), relative thinness of the capillary endothelial cells compared with mammalian species (Watson et al., 2007), uniformity of the capillary wall ‘thinness’ and, yet, a large variation of the luminal cross-sectional area along very short distances in some capillary networks (West et al., 2010), and the existence of a thick and curious hyaline coating around the capillaries of specialized organs, such as the intraocular pecten (Goodman and Bercovich, 2008). In theory, these factors challenge the way the Starling hypothesis can be applied in birds, thus leaving the functional ‘dimensions’ of the protein osmotic gradient still unclear.
The present findings of intravascular and interstitial COP open up new vistas for explaining the phenomenon of thermoregulatory cooling via glandless skin. Cutaneous water evaporation (CWE) is a cooling mechanism, which is primarily driven by the vapour pressure gradient between the bird and its surroundings (Marder and Arieli, 1988; Wolf and Walsberg, 1996; Ophir et al., 2003). In hot and dry environments, CWE may be enhanced by making the pressure gradient steeper by increasing the hydration of the outermost layers of the skin (Peltonen et al., 1998). Thus, changes in the COP around the cutaneous capillary wall will influence fluid shifts in the skin and, eventually, the process of CWE. Evidence of a specific epidermal fluid build-up has been found in rock pigeons (Columba livia Gmelin 1789) acclimated to high ambient temperatures, but not in those acclimated to mesic or cold environments (Peltonen et al., 1998; Peltonen et al., 2000). The associated dermal changes have been found to be an increased number of skin capillaries in the superficial dermis (Peltonen et al., 1998), and increased capillary fenestrations and endothelial gaps, the latter implying an increased extravasation of plasma proteins (Arieli et al., 1999). CWE has also been shown to involve both central and local adrenergic control, the local control involving a β2-adrenergic decrease in the arterial resistance to blood flow and an increase in the venous resistance (Marder and Raber, 1989; Ophir et al., 2000). Obviously, these changes tend to increase filtration, creating a driving force for CWE (Ophir et al., 2003). However, the relative importance of CWE to thermoregulatory cooling varies among species. In columbids, CWE becomes the major cooling mechanism in adaptation to a hot and dry environment (Marder et al., 1986; Marder and Arieli, 1988; Withers and Williams, 1990; Marder et al., 2003; Dawson and Whittow, 2000), whereas in other diurnally active species its relative importance decreases with increasing ambient temperature (Wolf and Walsberg, 1996). Together, fluid shifts driven by Starling forces within the body, or out of the body, may function as crucial local and systemic regulatory factors in fighting homeostatic imbalances.
As for the lymph flow, more research is needed to elucidate its physiological role in the homeostatic responses of birds. The lack of flow-driving lymph hearts, the lower number of lymphatic vessels and the less frequent valves compared with mammals (King and McLelland, 1975; Rose, 1985) suggest either a reduced capacity for fluid drainage from the tissues, or a different strategy for recirculating fluid between the circulatory system and the tissues. During the course of evolution, the loss of a permanent external water environment has been compensated for by an internal water reservoir and a lymphatic drainage system. Many amphibian species have a large-volume lymph space, drained externally from the skin and internally from skin capillaries, and pumped by multiple lymph hearts that beat 60-70 times per minute (Priestley, 1878). Many avian species, however, have lost their lymph hearts, but have probably preserved the capacity for fluid storage and a multidirectional flow pattern because of the more primitive structure of the lymph system (von Rautenfeld and Schacht, 2006). The age-dependent decrease in COPsbf in the present study suggests that ageing may increase the volume of the cutaneous interstitial fluid or, at least, the volume that is available for plasma proteins. Besides the possible changes in the composition of the interstitial ground substance, ageing may impair the function of the lymph transport system. It has been shown in rodents that the negative impact of ageing is associated with the loss of the contractile capability of the lymph vessel as a result of atrophy of the smooth muscle cells and elastic matrix (Gashev and Zawieja, 2010). Further studies are warranted in order to elucidate how ageing and natural changes in the physiology of birds modulate the osmotic activity and volume of the various fluid spaces of the body.
To summarize, Ott's protein osmotic pressure, which takes into account the graded osmotic effect of various serum proteins, showed age- and gender-dependent variation in chickens (Gallus gallus). Serum COPOtt increased with age in males, in association with increasing α1-globulin concentration. In contrast, COP of the bulk interstitial fluid collected by suction, showed an age-dependent decrease in females, associated with decreasing concentrations of α1-globulin and albumin. The affect of other molecules on COP, such as lactose, glucose and GAGs, was not assessed in the present study. However, the specific changes in the serum protein concentrations suggest that ageing modulates the composition of the interstitial ground substance so that the volume of the space occupied by serum proteins changes, affecting water binding and fluid diffusion/flow within the interstitium. The positive correlation between serum COPOtt and PCV suggests that the geometrical space available for proteins may affect the Starling forces also on the luminal side of the vessel wall. We did not measure the volume of the interstitial fluid in the present study. However, the age- and gender-dependent decrease in the COPsbf values is indicative of increased fluid retention/storage in the interstitium. Further studies are required in order to elucidate the role of lymphatics in, and the effects of ageing on, the recycling of serum proteins and fluid back to the circulation in birds.
We are most grateful to a true scholar, the late Professor Johan Ahlqvist, for his advice and for his great insight into hemodynamics and pathophysiology.
LIST OF ABBREVIATIONS
This research was funded by the University of Helsinki and the Finnish Physiological Society.