Plasticity of skin water permeability and skin thickness in the amphibious mangrove rivulus Kryptolebias marmoratus
The skin of amphibious fishes is a multipurpose organ, important for gas and ion exchange and nitrogen excretion when fish are out of water (emersed). We tested the hypothesis that skin permeability is altered to maintain water balance through changes in water permeability and skin thickness during salinity acclimation and/or when fish emerse, using the euryhaline, amphibious fish Kryptolebias marmoratus as a model. We first recorded the behaviour of fish out of water to determine which part of the cutaneous surface was in contact with the substrate. Fish spent about 70% of their time on their ventral surface when out of water. Osmotic permeability of the skin was assessed in fish acclimated to 0.3 or 45‰ using 3H2O fluxes in an in vitro micro-Ussing chamber setup. In freshwater-acclimated fish, 3H2O influx across the skin was significantly higher compared to hypersaline-acclimated fish, with no significant changes in efflux. Prolonged emersion (7 days) resulted in an increase in skin 3H2O influx, but not efflux in fish acclimated to a moist 45‰ substrate. In a separate experiment, dorsal epidermal skin thickness increased while the ventral dermis thickness decreased in fish emersed for over a week. However, there was no link between regional skin thickness and water flux in our experiments. Taken together, these findings suggest that K. marmoratus alter skin permeability to maximize water uptake while emersed in hypersaline conditions, adjustments that probably help them survive months of emersion during the dry season when drinking to replace water loss is not possible.
KeywordsAmphibious fish Water flux Desiccation stress Skin thickness Euryhaline Hypersaline
The skin of fishes primarily serves as a barrier, separating the internal and external environments. However, some fishes also use the skin as an exchange surface (e.g., Kirsch and Nonnotte 1977; Cooper et al. 2013; reviewed by; Glover et al. 2013). For example, in many amphibious fishes nitrogen excretion, ion regulation and respiratory gas exchange may occur across the skin when fish are out of water (e.g., Tamura et al. 1976; Feder and Burggren 1985; Fenwick and Lam 1988; Zhang et al. 2000; Frick and Wright 2002; Litwiller et al. 2006; LeBlanc et al. 2010; reviewed by; Graham 1997; Turko and Wright 2015; Wright and Turko 2016). The skin may also regulate the exchange of water. In aestivating freshwater lungfish (Protopterus dolloi), 3H2O efflux rates across the skin were significantly lower relative to rates in active aquatic lungfish (Wilkie et al. 2007). However, aestivating lungfish undergo a profound metabolic depression (Smith 1930) and changes in water flux may result from the general decrease in metabolism rather than emersion in particular.
Water fluxes across the skin of fishes depend, in part, on water salinity. Freshwater amphibious fishes may gain water passively from the substrate via osmosis. However, euryhaline or marine amphibious fishes face the risk of dehydration because of osmotic loss of water to the substrate or evaporative loss to the air when emersing onto salty moist surfaces (e.g., intertidal rock pools, mangrove swamps). In aquatic marine teleosts, passive water loss across body surfaces is compensated by drinking to replace lost water (Marshall and Grosell 2006), but drinking is not an option for amphibious fishes out of water. Do active amphibious fishes reduce cutaneous water flux to prevent desiccation when they are air-exposed?
The objective of this study was to determine the role of the skin in regulating water homeostasis in a euryhaline (0–114‰, King et al. 1989) amphibious fish—the mangrove rivulus Kryptolebias marmoratus. K. marmoratus frequently emerse (supplementary video, Turko and Wright 2015) and can survive prolonged terrestrial episodes during the dry season (> 66 days, Taylor 1990), hidden in leaf litter or packed into termite-excavated tunnels within moist mangrove logs (Taylor et al. 2008). Although they are known to forage on land (Taylor 1992; Pronko et al. 2013), K. marmoratus must return to water to consume food items because like most teleost fishes, they are suction feeders (McNeil Alexander 1970). Interestingly, body water content significantly increased by ~ 1% in K. marmoratus after 11 days emersed in the lab indicating that water balance is maintained in this species (Litwiller et al. 2006). In a previous study, we reported that whole body water efflux decreased in K. marmoratus following 9 days of air exposure on a hypersaline (45‰) substrate (LeBlanc et al. 2010), but whether these changes reflect renal or cutaneous modifications is unknown.
We tested the hypothesis that skin permeability is altered during acclimation to terrestrial conditions to maintain water balance through changes in water permeability and skin thickness. This hypothesis predicts a decrease in water efflux and an increase in influx across isolated skins in fish acclimated to hypersaline conditions. We examined osmotic permeability to understand how skin permeability in a euryhaline amphibious fish may be changing under conditions that mimic natural osmotic gradients. Osmotic skin permeability was assessed by measuring the unidirectional flux of 3H2O across isolated skins of K. marmoratus acclimated to different salinities (0.3 or 45‰) in water or air for 1 or 7 days. Regional variation in skin permeability was assessed using unidirectional 3H2O fluxes across paired dorsal and ventral skins. We also examined skin thickness in a separate group of fish. A decrease in skin thickness would equate to a decrease in diffusion distance and according to the Fick equation would, thus, enhance the flux of water, as well as solutes and gases (Toledo and Jared 1993; Moss et al. 2002; Takeuchi et al. 2012; Haslam et al. 2014). Changes in skin thickness may vary regionally. Therefore, variation in dorsal, ventral and lateral skin thickness in fish acclimated to 1 and 45‰ in water or air for 9 days was measured to assess morphological changes in response to emersion or salinity exposure. Fish were also filmed to determine which skin surface was most often in contact with the substrate.
Adult mangrove rivulus (0.18 g ± 0.01), Kryptolebias marmoratus (Belize strain DAN06, Tatarenkov et al. 2010) were acquired from a breeding colony housed under constant conditions (25 °C, 15‰, pH 8, 12L:12D cycle; Frick and Wright 2002) in the Hagen Aqualab at the University of Guelph. Fish were fed live Artemia nauplii three times per week. All experiments were approved by the University of Guelph animal care committee (Animal Utilization Protocol 2239).
Two series of experiments were conducted. First, we examined emersion behaviour to determine which part of the body was in contact with the moist substrate. The second series of experiments were designed to determine if differences in salinity or exposure to air affected skin permeability.
Series 1 Body orientation during emersion. Body position of fish out of water (1 day, 15‰) was quantified by mounting a GoPro camera (Hero HD, CA, USA) above emersed fish in plastic containers (15 cm3) lined with moistened filter paper (Ong et al. 2007). We chose a 1-day period because K. marmoratus move very little when out of water for 7 days and appear to maintain body orientation (Turko et al. 2014). Photos were captured every 5 s for 2 h and the body region in contact with the substrate (dorsal, ventral, lateral) was recorded. Care was taken not to disturb the fish over this time period.
Series 2 Influence of salinity and air exposure on skin permeability. To determine if regional differences in skin permeability existed, we performed in vitro experiments using isolated pairs of dorsal and ventral skins from the same individual in an Ussing chamber, following a previously validated protocol (Cooper et al. 2013). Fish were acclimated for 1 week to either control water (15‰) or air over a hypersaline moist surface (45‰) as previously described (Ong et al. 2007). These salinities (as well as the hyposaline treatment described below) are ecologically relevant, as K. marmoratus populations have been found in the wild at a wide range of salinities (0–70‰; Taylor 2012). During air exposure, the chambers remained at 99% relative humidity (RH). In the field, RH can be more variable (79–99%) (Gibson et al. 2015) but under mangrove leaf litter, it was ~ 99% (P. Wright, unpublished data). Fish were not fed during air acclimation because they are suction feeders and are unable to swallow out of water (Pronko et al. 2013). Following acclimation, fish were euthanized by spinal cord transection and the skin was carefully dissected. The skin was scraped free of muscle with a blunt instrument while being bathed in serosal saline solution [(mmol/L) 125 NaCl, 2 KCl, 1 MgSO4, 5 NaHCO3, 2 CaCl2, 1.25 KH2PO4, 5.55 glucose, pH adjusted to 7.5; Cooper et al. 2013]. The skin was then sandwiched between the two halves of the Ussing chamber (~ 0.6 mL each) and left for 1 h to stabilize (Cooper et al. 2013) in saline (serosal side) and an environmental bath (mucosal side; water of the appropriate salinity). The serosal solution was aerated with a humidified 0.5% CO2/O2 balance gas mix to mimic the gaseous environment of the blood plasma, while the mucosal solution was aerated with humidified air. During the 1-h stabilization period, a dye test was performed using food colouring (Club House, London ON, Canada; water, propylene glycol, tartrazine, citric acid, and sodium benzoate) to ensure that there was no leakage. At the end of this stabilization period, the chamber was rinsed and replenished with the appropriate solutions. 3H2O (Perkin Elmer Ltd., Ontario, Canada) was added to either the serosal or mucosal side (1 µCi/mL), mixed by gently pipetting, and triplicate samples (25 µL) were taken at t = 0 min from both the mucosal and serosal chambers. Flux was measured by the appearance of isotope on the cold side (Wood and Grosell 2012). Preliminary experiments showed that 3H2O flux across the skin was linear up to 30 min, so a 10-min flux period was chosen. Following the 10-min flux period, triplicate samples (25 µL) were taken from both chambers (t = 10 min). Immediately following this flux, the chamber was rinsed and replenished with unlabeled serosal/mucosal solutions. 3H2O was then added to the opposite side and samples were taken as before. The direction of flux was randomized, providing both influx and efflux data on each skin in random order. As a comparison, we also determined in vitro water influx and efflux rates across the skin of freshwater zebrafish (Daniorerio; n = 4). All procedures were identical to those described above for K. marmoratus, except fish were held in freshwater and the mucosal solution was 0.3‰ water. Following the unidirectional fluxes, a final dye test was performed to ensure the skin had not been compromised during the experimental period. The serosal/mucosal saline samples were mixed with aqueous counting scintillant (5 mL; 667 ml toluene, 333 ml Triton X-100, 4 g PPO, and 0.2 g POPOP; Sigma) and subsequently counted in a scintillation counter (Beckman Coulter LS6500 Multi-Purpose Scintillation Counter, California, USA).
To determine the influence of salinity and air exposure on 3H2O fluxes, fish were transferred from control water (15‰) to either 0.3‰ (hyposaline), or 45‰ (hypersaline) water and were acclimated for 1 week. Following acclimation, fish were placed in new water of the appropriate salinity or air-exposed over filter paper moistened with water of the appropriate salinity (0.3 or 45‰) as previously described (Ong et al. 2007). Fish were held for 1 day and were not fed during this time. An additional hypersaline emersion group with a 7-day air exposure period was tested to determine the effects of a more prolonged emersion as seen during tropical dry seasons. Longer periods of air exposure in the hyposaline group were not possible due to low survival. Following their respective acclimations, fish were euthanized and the skins were mounted in Ussing chambers as described above. Unidirectional water flux was measured through the skin in either the mucosal-to-serosal (influx) or serosal-to-mucosal (efflux) direction. 3H2O fluxes were sampled after 10 min as described above. Fluxes were conducted on individual dorsal skins, as preliminary experiments did not reveal significant differences in water permeability due to body region (P > 0.05; Table 1). Influx and efflux measurements were made on separate skin preparations from different fish. Transepithelial potential was also measured during flux experiments to verify the integrity of the skin. Electrodes were constructed using 4% agar bridges, 0.5 M KCl and ferric chloride-treated silver wires (A-M Systems, Washington, USA; Cooper et al. 2013). The electrode bridges were placed on either side of the skin while a modified BNC cable was submerged in the KCl solution, completing the electrical circuit. The signal was recorded in mV using a pH metre (Orion 520, Thermo Fisher Scientific Inc., Massachusetts, USA).
3H2O flux rates (µL/cm2/min) across paired dorsal and ventral skins of K. marmoratus
0.32 ± 0.11
0.29 ± 0.06
0.28 ± 0.03
0.24 ± 0.005
0.04 ± 0.13
0.05 ± 0.07
0.27 ± 0.07
0.23 ± 0.08
0.26 ± 0.10
0.19 ± 0.06
0.01 ± 0.13
0.03 ± 0.13
To measure skin thickness, we capitalized on tissues collected from a previous histological study focused on skin ionocytes in our lab (Belize strain 50.91, 0.06–0.14 g; LeBlanc et al. 2010). In that study, fish were acclimated to either hypo- (1‰) or hypersaline (45‰) water for at least 1 week prior to experiments. Following this, fish were either transferred to new water of the same salinity, or emersed over filter paper moistened with the appropriate salinity (1 or 45‰) for 9 days. Note, extended air exposure at 1‰ did not affect survival in this strain. K. marmoratus were then euthanized, embedded in paraffin, sectioned at 4 µm, and stained with haematoxylin and eosin. Sections were then photographed using a Nikon Eclipse 90i epifluorescent microscope equipped with a monochrome digital camera (Q-Imaging Retiga 1300). Measurements were taken using NIS Elements AR software from the basal dermis to the outermost epidermis on dorsal, ventral, and lateral regions of body.
Calculations and statistical analysis
For Series I, behavioural data were analyzed using a one-factor repeated measured ANOVA, followed by a post hoc Holm-Sidak test. For Series II, two-factor repeated measures ANOVAs were used to test whether differences existed between the dorsal and ventral skin permeability, and whether influx and efflux differed across a single skin preparation. A natural logarithm (ln) transformation was used to normally distribute efflux data. Two-factor ANOVAs with post hoc Holm–Sidak tests were used to test whether salinity or acute emersion induced changes in 3H2O flux across fish skins. Independent sample t tests were used to test whether emersion time (1 or 7 days) influenced 3H2O flux across hypersaline-acclimated fish skins. Two-factor repeated measures ANOVAs with post hoc Holm-Sidak tests were used to test whether salinity and emersion acclimations induce changes in epidermal and dermal skin thickness. SigmaPlot 11 (Systat Software, San Jose, CA, USA) was used for all analyses (critical α = 0.05). Throughout the text, values are given as means ± SEM.
Air-exposed fish spent significantly more time with their ventral skin surfaces in contact with the moist substrate (~ 71%), than dorsal (~ 22%) or lateral (~ 6%) regions (P < 0.05; Fig. 1).
Regional variability in skin water flux
There was no regional variation in 3H2O flux across the skin of K. marmoratus3H2O flux rates (Table 1). There were no significant differences in 3H2O efflux or in influx rates in the dorsal vs ventral skins in the control fish (15‰, P = 0.59) or in the hypersaline-acclimated and emersed (7 days) group (45‰, P = 0.41). There was no significant effect of flux direction, as influx and efflux were comparable in both control (P = 0.74) and hypersaline-acclimated (P = 0.23) fish. Net flux was approximately 0 in all groups and did not significantly vary by skin region (P = 0.86) or salinity/emersion acclimation (P = 0.94; Table 1). We also compared in vitro flux rates in K. marmoratus to those of fully aquatic freshwater zebrafish (Danio rerio, n = 4). There were no significant differences between species in 3H2O influx [in µL/cm2/min K. marmoratus 0.26 ± 0.05 (n = 10) vs D. rerio 0.35 ± 0.06 (P = 0.28)] or efflux [in µL/cm2/min K. marmoratus 0.37 ± 0.03 (n = 6) vs D. rerio 0.50 ± 0.24 (P = 0.62)] across isolated skins.
Freshwater-acclimated fish had higher rates of cutaneous 3H2O influx compared to hypersaline-acclimated fish (P = 0.026, Fig. 2a). One day of emersion in fish acclimated to a freshwater or hypersaline substrate had no effect on 3H2O influx (P = 0.20, Fig. 2a), however, 7 days of air exposure in the hypersaline group significantly increased 3H2O influx compared to the acutely emersed (1 day) group (P = 0.016, Fig. 2a). 3H2O efflux was not significantly affected by salinity (P = 0.11) or 1 day of emersion (P = 0.37, Fig. 2b). 3H2O efflux remained stable in emersed fish over time, as efflux rates were not statistically different between acute (1 day) or prolonged (7 days) emersion treatments (P = 0.45, Fig. 2b).
The TEP across the skin of freshwater-acclimated fish was negative (− 5.26 ± 0.65 mV, n = 9) and unchanged in fish exposed to air for 1 day (− 5.79 ± 0.50 mV, n = 8). In hypersaline-acclimated fish, the TEP across the skin was positive (+ 1.68 ± 0.06 mV, n = 12) and unaffected by air exposure (1 day, +3.32 ± 0.28 mV, n = 6).
Skin thickness was altered in response to both water salinity and air exposure. The dorsal epidermis of K. marmoratus was unaffected by salinity acclimation (P = 0.34), but thickened significantly after air exposure compared to control fish in water at both salinities (P = 0.024, Fig. 3I). The dorsal dermis was unaffected by emersion (P = 0.43), but thickened in hypersaline compared to hyposaline-acclimated fish (P = 0.032, Fig. 3I). The ventral epidermis was thicker following acclimation to hypersaline water (P = 0.006, Fig. 3J), but was unaffected by emersion (P = 0.81). The ventral dermis was thinner (18% thinner in 1‰ and 28% in 45‰) in response to prolonged emersion (P = 0.002, Fig. 3J), but was unaffected by salinity acclimation (P = 0.10). The lateral epidermis was unaffected by emersion and salinity (P = 0.37, 0.68 respectively, Fig. 3K), but the lateral dermis was thicker with hypersaline acclimation (P = 0.021, Fig. 3K), and was unaffected by emersion (P = 0.86). No salinity by medium interaction terms were significant (all P > 0.05).
Our data supports the hypothesis that terrestrial acclimation causes plastic changes to skin permeability in the amphibious K. marmoratus that may contribute to the ability of these fish to maintain water balance out of water. As predicted, increased 3H2O influx across isolated skins was observed following prolonged emersion over a hypersaline substrate. Hypersaline conditions promote osmotic water loss in fishes and when on land, amphibious fishes would be unable to replenish lost water by drinking. As well, the thickness of the ventral dermis was significantly reduced in air-exposed fish and fish spent most of the time on their ventral surface when out of water. However, we found no link between regional skin thickness and water flux in our experiments. Overall, the skin of K. marmoratus is highly plastic and changes in water permeability and morphology are part of a suite of cutaneous responses to extended air exposure (ammonia transporters, Hung et al. 2007; skin ionocytes, LeBlanc et al. 2010; angiogenesis, Cooper et al. 2012; Turko et al. 2014).
Unidirectional water flux
Overall rates of water flux across the isolated skin of K. marmoratus were comparable to studies on isolated amphibian skin. A study of four amphibian species reported that rates of inward water transport across ventral skins were between 0.3 and 1.3 µL/cm2/min, whereas across dorsal skins the values were between 0.0016 and 0.1 µL/cm2/min (Yorio and Bentley 1977). Our values (0.07–0.72 µL/cm2/min) were in the same range as the amphibian data, but there were no significant differences between the dorsal and ventral rates. Wilkie et al. (2007) reported an appreciable flux of 3H2O across the ventral surface in intact air-exposed African lungfish P. dolloi that approximated whole body water flux in immersed animals. The authors suggested that the ventral surface of lungfish, therefore, was the main site of water transport, similar to amphibians (McClanahan and Baldwin 1969; Bentley and Main 1972; Yorio and Bentley 1977) but dorsal permeability/flux was not measured. In wild K. marmoratus, water flux across all body surfaces may be beneficial when they seek moist cramped crevices out of water during the dry season (Taylor et al. 2008) and all parts of the body may be in contact with the substrate.
The TEP values in the current study were consistent with previous measurements. Cooper et al. (2013) reported TEP values of − 3.93 ± 0.24 mV across isolated skin of K. marmoratus acclimated to freshwater (1‰) and + 1.75 ± 0.13 mV in fish acclimated to brackish water (15‰), in a similar range to our values − 5.79 ± 0.5 mV (0.3‰) to + 3.32 ± 0.28 mV (45‰), albeit at more extreme salinities. The consistency of the TEP values indicates that the isolated skin preparations were intact, also verified by routine dye tests.
Effect of emersion
Prolonged emersion over a hypersaline substrate (7 days) resulted in a significant increase in cutaneous 3H2O influx rate. These findings suggest that K. marmoratus have mechanisms to improve water uptake in an environment with compounded stresses of osmotic and evaporative water loss, which would be advantageous during the dry season. It is possible that aquaporin density and/or tight junction complexes are altered to enhance water permeability (Cerdà and Finn 2010; Kolosov et al. 2013). Preliminary evidence showed an upregulation in the protein expression of claudin-10e, a pore-forming claudin, in the dorsal skin of K. marmoratus after only 1-day of emersion (F. Galvez, P. Wright and S. Kelly, unpublished data). Further research is required to determine the influence of aquaporins and tight junction proteins on water permeability in the skin of K. marmoratus.
Increased 3H2O influx in hypersaline-acclimated emersed fish is suggestive of water moving “uphill” against the natural osmotic gradient which would indicate active water transport. Active water transport through amphibian skin was first suggested by Reid (1892) and reported by others (Koefoed-Johnsen and Ussing 1953; House 1964). As well, uphill water movement has been reported in the intestines of marine teleosts and is thought to allow these fishes to absorb pure water from imbibed seawater (Curran 1960; Skadhauge 1969; Genz et al. 2011; Whittamore 2012). Ion uniporters and active ion transporters, such as Na+/K+-ATPase are thought to be the vectors of uphill water movement (Zeuthen and Stein 1994; Zeuthen 2010) and there is a strong interdependency between NaCl and water movement across bullfrog skin (Huf et al. 1951; Koefoed-Johnsen and Ussing 1953). LeBlanc et al. (2010) observed significantly larger Na+/K+-ATPase rich ionocytes in the epidermis of K. marmoratus after 9 days out of water on a hypersaline substrate. These cutaneous ionocytes may allow K. marmoratus to move water against the gradient to maintain water balance.
Unlike 3H2O influx, no changes in water efflux rates were observed in response to emersion. In the freshwater African lungfish P. dolloi and P. annectens, prolonged air exposure for 6–8 months resulted in a lower whole body water efflux (Wilkie et al. 2007; Patel et al. 2009). Similarly, whole body efflux in K. marmoratus acclimated for 9 days out of water on a moist hypersaline surface decreased markedly (LeBlanc et al. 2010). Our data suggest that changes in whole body efflux reported in our earlier study (LeBlanc et al. 2010) are more likely to result from decreased urine production rather than modifications to cutaneous water efflux in this species.
In hyposaline-acclimated fish, there were no significant changes in water fluxes after 1 day in air. We were unable to determine if longer periods out of water altered skin water permeability because survival declined. In their natural habitat, exposure to low salinity water would be associated with heavy rains and abundant water, and therefore, the signal to conserve body water may not be present. As well, passive osmotic uptake of water from the freshwater substrate may be sufficient to counterbalance evaporative water loss under these conditions.
Effect of salinity
When K. marmoratus were acclimated to hypersaline conditions, water influx across the skin was significantly reduced relative to fish acclimated to hyposaline conditions. Previous studies in teleosts have shown that whole animal water influx rates are consistently higher under freshwater relative to seawater conditions (Potts et al. 1967; Evans 1969; Motais et al. 1969). Thus, our findings in an isolated skin preparation were consistent with whole animal studies. 3H2O effluxes in isolated skins, however, were not affected by salinity acclimation.
Epidermal skin thickness measurements varied across body region, ranging from 7.7 to 39.2 µm. These values are consistent with previous measurements of K. marmoratus (~ 10–50 µm, Grizzle and Thiyagarajah 1987). What was surprising, however, was that skin thickness was plastic. To our knowledge, although skin structure has been studied (e.g., Park 2002), plasticity in skin thickness has not been reported in other amphibious species. We previously showed that skin thickness in K. marmoratus was not associated with individual variation in emersion preference (Turko et al. 2011). Changes in other aspects of the skin, such as the size of ionocytes (LeBlanc et al. 2010), angiogenesis (Cooper et al. 2012; Turko et al. 2014), and blood flow (Daxboeck and Heming 1982; Graham et al. 1985; Cooper et al. 2012) are known to occur in amphibious fish out of water. Prolonged air exposure resulted in a thickening of the dorsal epidermis and a thinning of the ventral dermis, whereas increased salinity alone increased the dorsal and lateral dermal thickness and ventral epidermal thickness. In other words, a combination of hypersaline and air acclimation resulted in an overall thickening of the skin with only the ventral dermis becoming thinner. These findings suggest that as the dry season approaches and water salinity rises, K. marmoratus increase the diffusive barrier to water loss. Once water is unavailable and the fish emerse, they modify the ventral dermis by reducing the thickness. A thinning of the ventral skin would reduce the diffusive distance between the ventral body and the moist substrate and thickening of the dorsal epidermis would do the opposite. Changes in skin thickness were not correlated, however, with changes in dorsal and ventral water flux (Table 1). There are several possible explanations. First, the genetic differences between fish (Belize strains: 50.91 and DAN06) used in the two experiments (H2O flux, skin thickness) may have resulted in different responses. Second, the skin is a complex organ with multiple layers and a blood supply. It is necessary to take into consideration the functional surface area, capillaries, and blood flow (Lillywhite and Maderson 1988; Lillywhite 2006). Epidermal capillaries are very close to the surface in K. marmoratus (1 µm; Grizzle and Thiyagarajah 1987), but in our in vitro experiments, there was no blood perfusion. In amphibians, increased circulation to the skin increases diffusive water exchange (Boutilier et al. 1992). Thus, the location of blood vessels within the skin and the volume of blood perfusing them would be valuable information for a more complete understanding of water flux. Third, the skin is a multifunctional organ (Glover et al. 2013) and water flux is only one of several important functions that may result in tradeoffs. For example, a thickening of the dorsal epidermis may reduce abrasive damage from locomotion (jumping or ‘tail flip’ locomotion; Gibb et al. 2011, 2013) in a terrestrial environment while altering the diffusion distance. Thus, there are many factors involved in understanding the relationship between the morphology of the skin and diffusive water flux.
Overall water balance
How important is the skin to overall water balance in intact fish out of water? Water exchange is controlled by skin, gills, and kidney in immersed fish, but likely only the skin and kidneys in K. marmoratus out of water. But skin is the only surface for water uptake, whereas the kidneys regulate the excretion of water. Water can also be gained metabolically from the oxidation of organic molecules (Schmidt-Nielsen 1997). Hence, the 1% improvement in body water content over long term emersion (Litwiller et al. 2006) may be achieved through a combination of metabolic water gain, reduced urine production and regulation of skin water fluxes. We know that cutaneous angiogenesis occurs in K. marmoratus acclimated to air (Cooper et al. 2012; Turko et al. 2014) suggesting that the skin exchange of gases, ions and/or water is important. Although we found a significant increase in cutaneous water influx after 7 days out of water (45‰), the mean efflux rate was similar (albeit values from different fish). It appears that cutaneous water influx may not result in substantial net water gain at this time point, but may prevent water loss under dehydrating conditions.
Water flux across the skin will depend on the microenvironment surrounding the fish. In the field, emersed K. marmoratus have been found in excavated tunnels in rotting mangrove logs (Taylor et al. 2008), under leaf litter (Taylor 2012), and above crab burrows on mud (Turko and Wright 2015) or on branches over the crab burrow (Wright 2012). The RH of these habitats will affect the rate of cutaneous water loss. The RH of air above the crab burrows can vary daily (Gibson et al. 2015) and no doubt seasonally, but under moist leaf litter it is near saturation (P. Wright, unpublished data) and is probably less variable. The RH in our lab experiments (99%) was matched to conditions under leaf litter, protected terrestrial habitat that may be preferred by K. marmoratus for prolonged episodes out of water. Besides RH, the salt content and water potential of the substrate will impact water uptake. The salt content of terrestrial substrates will vary seasonally in mangrove forests depending on rainfall and tidal cycles. We acclimated fish to water salinities (0.3 and 45%) that reflect aquatic conditions in the field (see Materials and methods), but to our knowledge the salt content of terrestrial substrates have not been measured in typical K. marmoratus terrestrial habitat. Substrate water potential will affect the rate of water uptake, as water should move from a higher to a lower water potential (e.g. the gradient between soil and the animal; Shoemaker et al. 1992). Although soil water potential has been determined for some types of soils and has been related to cutaneous water uptake in amphibians (reviewed by Shoemaker et al. 1992) there are no comparable data for mangrove forest ecosystems. In our lab experiments, K. marmoratus were placed on moist filter paper over top of water-soaked cotton balls, providing an ample supply of water, possibly reflecting conditions early in the dry season, but possibly not later. Future experiments should aim to characterize the microenvironment of terrestrial K. marmoratus habitat to gain a better understanding of the osmoregulatory challenges.
In conclusion, our data provide evidence that salinity and emersion alter the cutaneous water permeability and thickness in K. marmoratus. In hypersaline-acclimated fish out of water, skin 3H2O influx increased after 7 days in air suggesting a mechanism for obtaining water against an osmotic gradient. Although fish spent more time on their ventral surface and ventral skin thickness decreased in air-acclimated fish, we found no link between water permeability and skin thickness in this study, possibly because of the lack of blood flow in the isolated skin preparation. Regardless, dynamic remodelling of the cutaneous surface is probably an important physiological mechanism that helps to explain, in part, how K. marmoratus are able to maintain water balance (LeBlanc et al. 2010) and survive for months out of water during the dry season (Davis et al. 1990; Taylor et al. 2008; Taylor 2012; Wright 2012).
We thank Jim Ballantyne, Scott Kelly and Chris Wood for helpful comments on experimental design and calculations. Lori Ferguson and Hayley Ferguson are thanked for typographical assistance. Funding for this project was provided by an NSERC Discovery grant to P.A.W. and NSERC and OGS scholarships to A.J.T.
- Boutilier RG, Stiffler DF, Toews DP (1992) Exchange of respiratory gases, ions, and water in amphibious and aquatic amphibians. In: Environmental physiology of the amphibians. University of Chicago Press, Chicago, pp 81–124Google Scholar
- Davis WP, Taylor DS, Turner BJ (1990) Field observations of the ecology and habits of mangrove rivulus (Rivulus marmoratus) in Belize and Florida (Teleostei: Cyprinodontiformes: Rivulidae). Ichthyol Explor Freshw 1:123–134Google Scholar
- Graham JB (1997) Air-breathing fishes. Academic, San Diego USAGoogle Scholar
- Grizzle JM, Thiyagarajah A (1987) Skin histology of Rivulus ocellatus marmoratus: apparent adaptation for aerial respiration. Copeia 1987:237–240Google Scholar
- Kirsch R, Nonnotte G (1977) Cutaneous respiration in three freshwater teleosts. Respir Physiol 29:339–354Google Scholar
- Marshall WS, Grosell M (2006) Ion transport, osmoregulation, and acid-base balance. In: The physiology of fishes, 3rd edn. CRC, Florida, pp 177–230Google Scholar
- Schmidt-Nielsen K (1997) Animal physiology. Adaptation and environment, 5th edn. Cambridge Univeristy, Cambridge UKGoogle Scholar
- Shoemaker VH, Hillman SS, Hillyard SD, Jackson DC, McClanahan LL, Withers PC, Wygoda ML (1992) Exchange of water, ions, and respiratory gases in terrestrial amphibians. In: Environmental physiology of the amphibians. University of Chicago, Chicago, pp 125–150Google Scholar
- Smith H (1930) Metabolism of the lung-fish, Protopterus aethiopicus. J Biol Chem 88:97–130Google Scholar
- Taylor DS (1990) Adaptive specializations of the Crypinodont fish Rivulus marmoratus. Fla Sci 53:239–248Google Scholar
- Zeuthen T (2010) Water-transporting proteins. J Membr Biol 234:57–67Google Scholar