vascular aspects of water uptake mechanisms in the toad skin: perfusion, diffusion, confusion

logy, Part A 148 (2007) 55–63www.elsevier.com/locate/cbpa

Comparative Biochemistry and Physio

Vascular aspects of water uptake mechanisms in thetoad skin: Perfusion, diffusion, confusion☆

Niels J. Willumsen a,⁎, Arne L. Viborg a, Stanley D. Hillyard b

a Institute of Molecular Biology and Physiology, August Krogh Building, University of Copenhagen, Denmarkb School of Dental Medicine, University of Nevada, Las Vegas, NV, USA

Received 17 July 2006; received in revised form 29 December 2006; accepted 31 December 2006Available online 12 January 2007

Abstract

Blood cell flow (BCF) in the water absorbing “seat patch” region of toad skin was measured with laser Doppler flow cytometry. BCF of dehydratedtoads increased by a factor of 6–8 when water contact was made and declined gradually as toads rehydrated. Water absorption was initially stimulatedand declined in parallel with BCF. Water absorption measured during the initial rehydration period did not correlate with BCF and hydrated toadsinjected with AVT increased water absorption without an increase in BCF indicating the lack of an obligate relation between blood flow and waterabsorption. Aquaporins 1–3 were characterized by RT-PCR analysis of seat patch skin. AQP 1 was localized in the endothelium of subepidermalcapillaries and serves as a pathway for water absorption in series with the apical and basolateral membranes of the epithelium. Dehydrated toadsrehydrated more rapidly from dilute NaCl solutions than from deionized water despite the reduced osmotic gradient. BCF of toads rehydrating on50mMNaCl was not different than on deionized water and blocking Na+ transport with 100 μMamiloride did not reduce water absorption from 50mMNaCl. Thus, neither circulation nor solute coupling explains the greater absorption from dilute salt solutions. Rehydration from 10 mM CaCl2 wasstimulated above that of DI water by a similar degree as with 50 mM NaCl suggesting the anion might control water permeability of the skin.© 2007 Elsevier Inc. All rights reserved.

Keywords: Amphibian skin; Toad; Water uptake; Dehydration; Aquaporins; Seat patch; Blood flow

1. Introduction

Except for the breeding period, toads (Bufonidae) spendthe greater fraction of the year terrestrially. At this time theamphibian antidiuretic hormone, arginine vasotocin (AVT), reducesglomerular filtration, stimulates reabsorption of stored bladderwater and stimulates the water permeability of the skin to facilitaterapid rehydration (Bentley, 1966). Water absorption occursprimarily through a specialized, richly vascularized region of thepelvic skin termed the ‘seat patch’ and may amount to as much as20–30% of the body mass per hour. Word and Hillman (2005)

☆ This paper was presented in the session “Water transport” at the society ofexperimental biology's annual meeting at the university of Kent, Canterbury,UK April 2nd–7th 2006.⁎ Corresponding author. Institute of Molecular Biology and Physiology,

August Krogh Building, University of Copenhagen, Universitetsparken 13,DK-2100 Copenhagen O, Denmark. Tel.: +45 3532 1635; fax: +45 3532 1567.

E-mail address: [email protected] (N.J. Willumsen).

1095-6433/$ - see front matter © 2007 Elsevier Inc. All rights reserved.doi:10.1016/j.cbpa.2006.12.032

recently showed that water absorbed across the seat patch of Bufomarinus is taken up entirely into the circulation and lead us to re-examine the question: Is water absorption limited by blood flow tothe skin (perfusion), osmotic flow across the skin (diffusion) orsome combination of these mechanisms? Christensen (1975)observed the high rate of AVT-stimulated water absorption acrossisolated seat patch skin from Bufo bufo only when the vasculaturewas perfused with a Ringer's solution. In contrast, Baldwin (1974)found significant stimulation of water flow across isolated, non-perfused, seat patch skin from B. punctatus following treatmentwith the mammalian antidiuretic hormone arginine vasopressin(AVP). However, values for isolated seat patch skin were about halfthat of water absorption by dehydrated B. punctatus, in vivo, heldwith modeling clay to apply the seat patch to a calibrated pipette(McClanahan and Baldwin, 1969). The first section of this reviewwill describe experiments designed to test the hypothesis that seatpatch blood flow is a limiting factor for water absorption byconscious, unrestrained toads andwill present newdata that identifyaquaporin 1 as awater conducting pathway in cutaneous capillaries.

mailto:[email protected]

http://dx.doi.org/10.1016/j.cbpa.2006.12.032

56 N.J. Willumsen et al. / Comparative Biochemistry and Physiology, Part A 148 (2007) 55–63

AVT also stimulates Na+ transport across isolated amphibianskin. This has been measured as short circuit current (Isc) inUssing chamber preparations and shown to be the result of anincreased number of epithelial Na+ channels (ENaCs) in theapical membrane of the epithelium (Helman et al., 1983). AVTstimulation of Isc across toad skin (Bufo woodhouseii) is muchgreater in the seat patch than in pectoral skin (Baker andHillyard, 1992) and NaCl transport has been shown to becoupled to water absorption, in vitro (Nielsen et al., 2007). Inlive toads, B. bufo were observed to rehydrate 64% morerapidly from 20 mM NaCl than from deionized water (Ferreiraand Jesus, 1972) and it was suggested water uptake was coupledto Na+ transport. In the second section of this review we presentour own studies that have shown enhancement of waterabsorption by dilute salt solutions (Hillyard and Larsen, 2001)and describe experiments aimed at elucidating the still un-resolved mechanism for this phenomenon (confusion).

2. Methods

2.1. Animal sources and maintenance

Bufo bufo and B. marinus were provided by commercialsuppliers. Toads were maintained in a 3×5 m room with waterand dry surface available ad lib. B. punctatus were captured inthe Spring Mountains, Clark County Nevada, under permit fromthe Nevada Department ofWildlife and maintained in 30 terrariawith rocks, sand and water arranged to approximate their naturalenvironment. All species were maintained and experimentsconducted at ambient (20–23 C) temperature. All weremaintained on a regular photoperiod, usually 12:12 light:dark,and fed on a regular basis to maintain or increase their bodymass. Under these conditions toads were considered to maintaina hydrated state (Jørgensen, 1994) and normally stored diluteurine in their urinary bladders to offset dehydration. The weightof a hydrated toad with an empty bladder, the standard weight(Ruibal, 1962), was used as a reference for evaluating levels ofdehydration.

2.2. Seat patch blood flow and water uptake rates

Viborg and Rosenkilde (2004) utilized laser Doppler flowcytometry to measure seat patch blood flow in B. bufo. Anexternal probe measured the velocity and density of blood cellsin the cutaneous circulation to a depth of approximately 1 mm.The output of the probe is expressed in units of mVand gives arelative measure of Blood Cell Flow (BCF) rather than anabsolute value for volume flow. The probe was positioned in thefloor of a Lucite chamber and toads were placed individually inthe chamber with the seat patch over the probe. BCF wasinitially measured on hydrated toads before and after placingthem in a water filled beaker to determine whether watercontact, per se, stimulated BCF. Toads were then dehydratedbetween 10–20% of their standard weight. BCF was againmeasured before and after water exposure to determine if waterexposure in addition to dehydration is necessary for stimulatingBCF. After the initial measurement of BCF, toads were allowed

to remain in water for 30 min periods at which time BCF andwater uptake, evaluated as weight gain, were recorded. Viborgand Hillyard (2005) and Viborg et al. (2006) further developedthis method with the laser-Doppler probe incorporated into acalibrated water reservoir so water absorption could bemeasured simultaneously with BCF. This allowed blood flowand water absorption to be compared during the initial periodfollowing water contact when both parameters are maximallystimulated. Because dehydration increases the plasma AVTconcentration of B. marinus (Konno et al., 2005) one mightexpect a parallel increase in BCF to accompany the increase inwater absorption (Parsons and Mobin, 1991). For this reason,Viborg and Rosenkilde (2004) also measured BCF and waterabsorption in hydrated B. Bufo injected with exogenous AVT(100 ng/100 g BW).

2.3. Water uptake routes in the toad skin

A polyclonal antibody was raised in rabbits against theterminal twenty two amino acid sequence of rat AQP 1;GQVEEYDLDADDINSRVEMKPK. The antibody was affin-ity purified and used in a 1:100 or 1:300 dilution. Paraffinsections (10 μM) were dewaxed and rehydrated. Endogenousperoxidase was blocked by 0.5% H2O2 in absolute methanol.To reveal antigens, the sections were boiled in 1 mM Tris, pH9 supplemented with 0.5 mM EGTA. Non-specific bindingwas quenched by incubating the sections in 50 mM NH4Cl andblocked in phosphate-buffered salt solution (PBS) supplemen-ted with 1 % bovine serum albumen (BSA), 0.05% saponinand 0.2% gelatin. The sections were incubated overnight at 4 Cwith the antibodies diluted in PBS supplemented with 0.1%BSA and 0.3% Triton X-100. After washing, the sections wereincubated with horseradish peroxidase conjugated goat antirabbit IgG (P0448 1:200 Dako) diluted in PBS supplementedwith BSA and Triton X-100. The peroxidase stain wasvisualized by 0.05% 3,3′ diaminobenzidine tetrahydrochloridedissolved in PBS with 0.1% H2O2. Mayer′s haematoxylin wasused for counterstaining and the sections were dehydrated andmounted in hydrophobic Eukitt mounting medium (O. Kindler,Freiburg, Germany). For control experiments, comparisonswere made between sections labeled with a 1:100 dilution ofthe primary antibody alone and preabsorbed with 30 μg of thepeptide.

We have also begun to examine the expression of aquaporinsthe pelvic and pectoral skin of B. bufo that were taken directlyfrom the maintenance terraria, dehydrated or kept in standingwater to insure complete hydration. Reverese transcriptase(RT)-PCR was conducted on homogenates of epithelial cellsusing sets of forward and reverse primers designed from basesequences of B. marinus AQP 1-3 obtained from GenBank:

Aquaporin 1: Accession Number AF020620AQP-t1Fw: GGCGGTGATAGCCGAGTTCTAQP-t1Rw: CGGTCCGTTAAGTCGCTGGT

Aquaporin 2: Accession Number AF020621AQP-t2Fw: TGAATTGGCAGTCGGCACTTAQP-t2Rw: CCTGGGCTGTTCTTGTCTTTCC


57N.J. Willumsen et al. / Comparative Biochemistry and Physiology, Part A 148 (2007) 55–63

Aquaporin 3: Accession Number AF020622AQP-t3Fw: GGTCGGGAGCTATGCCAACTAQP-t3Rw: AGGGAATTGAGGCCCACAGA

In one set of experiments epithelial cells were isolated by10 min′ immersion in 1 M potassium iodide (KI) solution(Kawada et al., 1969) and scraping from the dermis with ascalpel blade. In a second set of experiments epithelial cellswere isolated by 120 min incubation with collagenase (2 mg/ml,Worthington Type II) to minimize subepidermal tissue in theRT-PCR analysis. Homogenates were processed with the One-step RT-PCR kit from Quiagen: 30 min at 50 C for reversetranscription, 15 min at 95 C to inactivate reverse transcriptase.Twenty nine cycles of PCR amplification were used: 40 sec at94 C for denaturation, 20 sec at 55 C for annealing and 4 min at72 C for elongation.

Fig. 1. A. dehydrated toad (Bufo alvarius) increases its blood flow more than10-fold upon presentation for a hydration source. The blood flow (BCF, Y axis)is given in arbitrary units (volts as read from the laser-Doppler probe).Oscillations reflect flow changes associated with heartbeat. (From Viborg et al.,2006). B. Dehydrated (18–26% of standard weight), but not hydrated toads(Bufo bufo) increased their blood flow upon water contact (⁎:Pb0.001). BCF ofdehydrated toads prior to water contact was not different from hydrated toads.Figure summarizes 18 trials with 6 toads. (From Viborg and Rosenkilde, 2004).

Fig. 2. Dehydrated, but not hydrated Bufo bufo increased blood flow (A) as wellas cutaneous water uptake (B) upon water contact. Both parameters returnedtowards normal values over 2 h. ⁎: BCF and water absorption greater indehydrated toads (Pb0.001). (From Viborg and Rosenkilde, 2004).

2.4. Effect of salinity and tonicity of the hydration source

Hillyard and Larsen (2001) studied rehydration by Bufomarinus that were placed in beakers containing either deionizedwater (DI) or dilute (10–50 mM) NaCl solutions. As above,dehydration (10–15%) was evaluated as weight loss relative tothe standard weight. By the van't Hoff relationship:

P ¼ CdRdT

where π is the osmotic pressure, C is the total concentration ofsolutes, R the gas constant and T the absolute temperature. Thus, ifwater absorption is entirely passive, the osmotic gradient for waterabsorption should be directly proportional to the concentrationgradient. Hillyard and Larsen (2001) found the osmotic concen-tration of dehydrated B. marinus to average 260 mOsm/kg whichwill be the predicted gradient for absorption from deionized water.For 50 mM NaCl the gradient will be reduced by approximately93mOsm, assuming an osmotic coefficient of 0.93 forNaCl. Thus,the gradient will be reduced by 167/260 or 64%. Absorption atgreater rates than predicted requires additionalmechanisms such asincreased blood flow or coupling to active Na+ transport. Thesepossibilities were evaluated by monitoring BCF, as describedabove, ofB.marinus rehydrating from 50mMNaCl vsDIwater or

Fig. 3. Blood flow (BCF) in rehydrating B. punctatus was consistentlystimulated above the level of hydrated toads (indicated by dotted line) but wasnot correlated with water uptake when measured in individual animals (Bufopunctatus). Data from 18 trials in 3 animals. (From Viborg and Hillyard, 2005).


comparing rehydration rates from 50 mM NaCl solutions, in thepresence and absence of 100 mM amiloride, a blocker of epithelialNa+ channels. Grosso et al. (1994) also have suggested the

Fig. 4. A. Injection of AVT (100 ng/100 g BW) had no effect on BCF butincreased water uptake 2-fold in B. bufo (B). P: prior to AVT injection; I: rightafter AVT injection; H: water immersion. ⁎: Water absorption greater in AVT-injected toads (Pb0.001). (From Viborg and Rosenkilde, 2004).

insertion of aquaporins in skin and urinary bladder ofB. marinus issensitive to the concentration of Cl− in the intracellular fluid andthe regulation of cell volume. For this reason rates of waterabsorptionwere compared between toads rehydrating from 50mMNaCl and 50 mMNagluconate. As a control, rehydration was alsomeasured in toads placed in a 100mMsolution of a non-electrolye,sucrose. More recently we have compared rehydration rates ofB. punctatus placed in 50 mM NaCl vs 10 mM CaCl2 solutions.Because Ca2+ transport across the skin is negligible (Baldwin andBentley, 1981), enhanced water uptake from CaCl2 would imply arole for Cl− in the elevation of water uptake. In these experiments,toads were dehydrated by 10% of their standard weight and placedin beakers of water containing deionized water (DI), 50 mMNaCl,10 CaCl2 or 35 mM NaCl+10 mM CaCl2.

2.5. Statistical analysis

Throughout the paper values are reported as mean±std.Two-way analysis of variance (ANOVA) applied to the random

Fig. 5. Immunolocalization ofAQP 1 in the skin ofBufo bufo. AQP 1was confinedto the capillaries in the subepidermal connective tissue (A). Antibodies preabsorbedwith AQP 1 did not bind to tissue structures (B). Magnification ×640.

Table 1

Distribution of mRNA for aquaporins 1–3 in epidermal cells from pelvic (seatpatch) and pectoral skin of B. bufo that were isolated by treatment with eitherpotassium iodide (1 M, 10 min) or collagenase (2 mg/ml, 120 min)


block design was used to compare treatment groups where toadsserved as a blocking factor with 3 replications for each toad(Viborg et al., 2006).

3. Results


When dehydrated toads came in contact with water there was aprogressive increase in BCF (Fig. 1A). The increase was 6–8 timesthat of dehydrated toads prior to water contact. BCF of dehydratedtoads prior to water contact was not different than BCF of hydratedtoads (Fig. 1B), i.e. dehydration alone did not stimulate BCF.When dehydrated toads were allowed to rehydrate over a 2 hperiod, BCF andwater absorptionwere initially high and decreased

Fig. 6. Identification of aquaporins 1–3 in seat patch and pectoral skin oftoads that were either dehydrated (deprived of water for 24 h), hydrated (kept intap water for 24 h) or controls (kept dry with access to tap water ad lib.).Epithelial cells were obtained by 10 min incubation with 1 M KI (A) or 120 minwith 2 mg/ml collagenase (B).

RNA for AQP 1 was only seen in KI treated tissues while AQP 3 the onlyaquaporin seen consistently in the collagenase treated tissue. ‘+’ indicates astrong signal. ‘(+)’ indicates a weaker signal. (c: control, h: hydrated, dh:dehydrated).

to values near that of fully hydrated toads (Fig. 2). Viborg andHillyard (2005) further showed that BCF of rehydrating toads waselevated relative to that of hydrated animals but was not correlatedwith water absorption during the initial 20 min period, beforeappreciable rehydration had occurred (Fig. 3). However, whenhydrated toads were given exogenous AVT, water absorptionincreased without an increase in BCF (Fig. 4).


AQP 1 was consistently localized in the subepidermalcapillaries and not in the epithelial cells (Fig. 5A). Pre-absorption

Fig. 7. Water uptake in salt solutions relative to that in deionized water. In NaClsolutions thewater uptake is invariably higher thanwhat is expected from the van′tHoff equation (dotted line). Asterisks indicate significantly different frompredicted value. When Cl−was replaced by gluconate the uptake was as predictedby the equation. Also,whenNaCl was replaced by sucrose the water uptake was inaccordance with the van't Hoff equation. (From Hillyard and Larsen, 2001).

Fig. 8. There was no difference in blood flow between hydrated and dehydratedB. punctatus prior to water exposure (hatched bars). When dehydrated toadswere given access to deionized water BCF increased in the seat patch but not inthe pectoral skin (open bars). If the hydration source was 50 mMNaCl BCF waslikewise increased in the seat patch only (double hatched bars). Asterisksindicate BCF significantly higher than that measured in the dry chamber. Plussigns indicate BCF of pectoral skin significantly less than in the seat patch.(Adapted from Viborg and Hillyard, 2005).

Fig. 9. A. Rehydration from 50 mM NaCl by B. marinus was not affected byblockage of epithelial Na+ uptake by 100 μM amiloride. B. Rehydration from120 mM NaCl was reduced in the presence of 10 μM amiloride. (From Hillyardand Larsen, 2001).

Fig. 10. Rehydration by B. punctatus was similarly stimulated in the presence of50 mM NaCl, 10 mM CaCl2 and the combination of 35 mM NaCl plus 10 mMCaCl2. Asterisk indicates significant stimulation relative to deionized water butno differences between any of the salt solutions.


with the peptide completely blocked labeling with the antibody(Fig. 5B).

There was considerable variation between the expression ofmRNA for AQP 1–3 in epithelial cells isolated with KI with noobvious differences between hydration states. Representativegels from both treatments are shown in Fig. 6 and the results ofsix experiments summarized in Table 1. Most important, AQP 1was not observed in cells isolated with collagenase while,with one exception, AQP 3 was the only aquaporin consistentlydetected with this treatment and was seen in both pelvic andpectoral skin.


The rate of rehydration by toads immersed in the dilute saltsolutions was greater than from DI water (Fig. 7). Rehydrationfrom 120 mM NaCl was less than that from DI water but greaterthan predicted from the van′t Hoff equation. When gluconatereplaced chloride as the anion, rehydration from the 50 mM saltwas not elevated above the predicted value and was similar torehydration from 100 mM sucrose. BCF of toads (B. punctatus)rehydrating from 50 mM NaCl was not greater than from DIwater (Fig. 8). This figure also shows that BCF of the pectoralskin of dehydrated toads was not increased after water contact.Finally, amiloride had no effect on rehydration from 50 mMNaCl (Fig. 9A) but significantly reduced rehydration rate from120 mM NaCl (Fig. 9B, experiments with B. marinus). Recentexperiments showed similar increases in water absorption byB. punctatus rehydrating from 50 mM NaCl, 10 mM CaCl2 and35 mM NaCl plus 10 mM CaCl2 (Fig. 10).


4. Discussion


As noted in Methods, BCF is measured by phase shifts inlaser light reflected from blood cells in the cutaneousvasculature and is a function of both blood cell density andvelocity. The pattern for increased BCF following cutaneouswater exposure is consistent among all species tested butthe time course is variable among species. B. punctatus (ca 20–25 g) achieves maximal BCF in approximately 20 seconds(Viborg and Hillyard, 2005) while B. alvarius (200–400 g)require 2–3 min (Fig. 1) as do B. marinus of similar size(Viborg et al., 2006). B. alvarius are found in xeric habitatswhile B. marinus are found in more mesic environments so thedifference appears to be size rather than habitat related. Onthe other hand, the magnitude of BCF is greater in desertspecies, B. punctatus and B. alvarius than in mesic species,B. marinus and B. bufo which is consistent with morphologicalobservations by Roth (1973) that xeric species have a moredeveloped vasculature in the ventral abdominal skin. Fig. 1Bshows that neither water contact nor dehydration alonestimulate BCF. Rather, cutaneous water potential receptors ofsome kind stimulate BCF in dehydrated toads (Viborg andRosenkilde, 2004). For all species, the water absorptionresponse posture (WR) is initiated during or after the rise inBCF which indicates additional neural or hormonal processingof sensory information before the skin is applied to a hydrationsource (reviewed by Hillyard et al., 1998).

Fig. 2 shows the parallel time course for stimulation ofBCF and water uptake in rehydrating B. bufo (Water up-take=1.32·BCF+4.69; Pb0.001). Similar results were observedwhen hydrated toads were injected with the beta adrenergicagonist isoproterenol and both BCF and water gain of dehydratedtoads were inhibited by the beta adrenergic antagonist, propran-olol (Viborg and Rosenkilde, 2004). These results indicate thestimulation of BCF by dehydrated toads is mediated by asympathetic reflex in response to chemosensory input from theskin (discussed by Hillyard et al, this symposium). Betaadrenergic stimulation also increases water permeability ofisolated frog skin (Rajerison et al., 1972) so the concomitantincrease in BCF and water absorption may be mediated by thesame reflex. The initial increase in water absorption wasaccompanied by but not correlated with an increase in BCF.Similar results were obtained with several Bufonid species:B. punctatus (Fig. 3), B. woodhouseii, B. marinus and B. alvarius(Viborg and Hillyard, 2005; Viborg et al., 2006) indicating bothdiffusion and perfusion contribute to the capacity of toads torehydrate quickly. A lack of blood flow results in dilution of theextracellular fluid and a reduction in the osmotic gradient forwater absorption (Christensen, 1975; Parsons and Mobin, 1991).Thus, blood flow needs to be adjusted according to waterabsorption which is regulated by neural, hormonal and behavioralfactors. The relationship between blood flow and waterabsorption was estimated by Viborg et al. (2006). This wasaccomplished by measuring aortic blood flow (central arterialflow=CAF) of dehydrated B. marinus that were given access to

water. CAF values, measured with an implanted flow probe,were approximately 7 ml/min/kg at rest and rose by an averageof 80% to 12.5 ml/min/kg when water was provided. Althoughshunting probably occurred (Slivkoff and Warburton, 2001),the difference, 5.5 ml/min x kg provides an estimate ofcutaneous blood flow. Typical water uptake rates in dehydratedBufo marinus amounts to ∼0.5 ml/min/kg (Hillyard and Larsen,2001; Viborg et al., 2006), and the major fraction is takenup across the seat patch. This is about 9% of the cutaneousblood flow and since the toads are dehydrated by approximately10%, will not cause an excessive hemodilution. As noted above,BCF and water absorption decline in parallel as the toadsrehydrate which could allow plasma osmolality to reach thehydrated value.

In contrast, hydrated toads injected with AVT showed asignificant increase in water absorption without an increase inBCF (Fig. 4). It may therefore be concluded that although trans-cutaneous water uptake is dependent on the seat patch bloodflow, there is no obligate correlation between the twoparameters. This could be the result of rate limiting steps inthe epithelial cells or endothelial cells of the cutaneous vas-culature (see below).


The presence of AQP 1 in the subepidermal capillariesprovides a high conductance pathway for water absorbed acrossthe epidermis to enter the circulation. These results provide amechanism for observations by Word and Hillman (2005) whoconcluded that the vasculature is the primary route for wateruptake. This conclusion is based on experiments that showedrehydrating B. marinus to gain weight after the lymphatic fluidwas removed and the lymphatic fluid was not osmoticallydiluted as would be expected if absorbed water accumulated inthe lymph sacs as proposed by (Wentzell, et al., 1993). Insupport of this latter hypothesis, tritiated water accumulates inthe lymphatic fluid and the lymphatic sacs become distendedduring rehydration.

The absence of mRNA for AQP 1 in collagenase-preparedcells indicates contamination of dermal tissue when cells arescraped from the epidermis following treatment with KI.Although preliminary, our detection of mRNA for AQPs 2and 3 in epidermal cells from B. bufo is consistent with thedemonstration of similar aquaporins, AQP-h3 and AQP-h3 inskin of the Japanese tree frog Hyla japonica (Hasegawa et al.,2003; Tanaka et al., 2005; Suzuki, 2006). Alignment of ourPCR fragments for AQPs 1, 2 and 3 with homologoussequences from H. japonica shows 83 %, 65% and 87.5 %similarity for AQPs h1, h2 and h3, respectively (Willumsenet al., 2003). It should be noted that AQPs h2 and h3 werelocalized only in the seat patch of H. japonica and becameinserted into the apical membrane of granular cells uponstimulation with AVT. We detected AQP 3 in both pelvic andpectoral skin. B. bufo is a relatively mesic toad and its seat patchis not well developed or easily distinguished from the pectoralskin. Either the two areas were inadvertently sampled or B. bufomight have regulated aquaporins in the more anterior skin


region. It should be noted that our AQP 3 and the similar AQP-h3 are not closely homologous to the mammalian AQP 3 whichis located in basolateral membranes of renal collecting duct cellsand is believed to be a constitutive water channel that is alsopermeable to glycerol (Ecelbarger et al., 1995).


In addition to B. marinus (Fig. 7), our laboratories haveshown that B. punctatus (Sullivan et al., 2000), and B. alvarius(Viborg et al., 2006) also absorb water more rapidly from dilutesalt solutions than from DI water. This phenomenon appears tobe common among the Bufonidae and perhaps other anuranfamilies as well but themechanism remains speculative. Hillyardand Larsen (2001) suggested this might result from an increasein capillary perfusion of the skin. However, Viborg and Hillyard(2005) showed cutaneous blood flow in B. punctatus wasequally stimulated by contact with 50 mM NaCl or DI water(Fig. 8) as was the case for B. alvarius and B. marinus (Viborget al., 2006). A second hypothesis is that water absorption iscoupled to Na+ and Cl− transport. It is well known that NaClcoupled water transport occurs in isolated skin when the outersurface is bathed with full strength Ringer's (Steinbach, 1967,summarized by Nielsen, 2006). As predicted, amilorideinhibition of Na+ transport inhibited water absorption byB. marinus immersed in 120 mM NaCl (Fig. 9B). For diluteNaCl solutions, Guo et al. (2003) calculated that the additionalrate of water absorption can be accounted for by coupling toNaCl transport. However, Hillyard and Larsen (2001) found thatamiloride, in concentrations as high as 100 μM, failed tosignificantly reduce water absorption by toads immersed in50 mM NaCl (Fig. 9A). It appears that solute coupling of waterabsorption plays a minor role when the osmotic gradient isreduced by the dilute salt solution. More recently (Fig. 10) wehave found a similar stimulation of water absorption byB. punctatus rehydrating from 10 mM CaCl2, 50 mM NaCland a combination of 10 mM CaCl2 and 25 mM NaCl. Analternative to circulation or solute coupling was investigated byKatz (1987) who found cutaneous water transport, in vivo and inisolated skin, to be greater in salt acclimated B. viridis andsuggested exposure to the higher salinity induced a greaterdensity of intramembranous particles that are now known to beaquaporins. It was proposed that the decreased osmotic gradientwas compensated for by a greater osmotic permeability. Thesimilarity of the response to NaCl and CaCl2, but not Nagluconate suggests Cl− might be a signaling factor for theactivation/insertion of aquaporins as has also been suggested byGrosso et al. (1994).

5. Summary

1. The stimulation of BCF is a sympathetic reflex in response toosmotic/ionic receptors in the skin and is much morepronounced in dehydrated toads.

2. The increase in water absorption by dehydrated toads isaccompanied by an increase in BCF but the two are notcorrelated suggesting that both water transport mechanisms

(diffusion and perfusion) contribute to rapid rehydration byterrestrial toads.

3. AVT stimulates water absorption without stimulating BCF. Thepresence of AQP 1 in the endothelial cells of cutaneous capil-laries may provide for some degree of stimulated water uptake.

4. Water uptake from dilute NaCl solutions is greater than fromdeionized water despite the reduced osmotic gradient. This isnot associated with changes in BCF or solute coupling butmay be related to Cl− as a signaling factor for increasedactivation/insertion of aquaporins in the skin. The mecha-nism remains unresolved (confusion).

Acknowledgements

We would like to thank Lene Niemann Nejsum at the Waterand Salt Research Center at University of Aarhus for designingthe primers and providing the AQP 1 antibody. SDH wassupported in part by grant IBN 9215023 from the US NationalScience Foundation.

References

Baker, C.A., Hillyard, S.D., 1992. Capacitance, short-circuit current andosmotic water flow across different regions of the isolated toad skin.J. Comp. Physiol. B 162, 707–713.

Baldwin, R.A., 1974. The water balance response of the pelvic “patch” of Bufopunctatus and Bufo boreas. Comp. Biochem. Physiol. 47A, 1285–1295.

Baldwin, G.F., Bentley, P.J., 1981. A role for skin in Ca metabolism of frogs?Comp. Biochem. Physiol. 68A, 181–195.

Bentley, P.J., 1966. Adaptations of amphibians to arid environments. Science152, 619–623.

Christensen, C.U., 1975. Correlation between net water flux, osmoticconcentration of the interstitial fluid and osmotic water permeability of theisolated skin of Bufo bufo bufo. J. Comp. Physiol. 96, 95–100.

Ecelbarger, C.A., Terris, J., Frindt, G., Echevarria, M., Marples, D., Nielsen, S.,Knepper, M.A., 1995. Aquaporin-3 water channel localization andregulation in rat kidney. Am. J. Physiol. 269, F663–F672.

Ferreira, H.G., Jesus, C.H., 1972. Salt adaptation in Bufo bufo. J. Physiol. 228,583–600.

Grosso, A., Jaquet, P., Brawand, P., DeSousa, R.C., 1994. Anion-induceddynamic behavior in vasopressin-sensitive epithelia exposed to mercury.Am. J. Physiol. 266, C1577–C1585.

Guo, P., Hillyard, S.D., Fu, B.M., 2003. A two-barrier compartment model forvolume flow across amphibian skin. Am. J. Physiol. Regul. Integr. Comp.Physiol. 285, R1384–R13894.

Hasegawa, T., Tanii, H., Suzuki, M., Tanaka, S., 2003. Regulation of waterabsorption in the frog skins by two vasotocin-dependent water-channelaquaporins, AQP-h2 and AQP-h3. Endocrinology 144, 4087–4096.

Helman, S.I., Cox, T.C., Van Driessche, W., 1983. Hormonal control of apicalmembrane Na transport in epithelia. Studies with fluctuation analysis.J. Gen. Physiol. 82, 201–220.

Hillyard, S.D., Larsen, E.H., 2001. Lymph osmolality and rehydration fromNaCl solutions by toads, Bufo marinus. J. Comp. Physiol. B 171, 283–292.

Hillyard, S.D., Hoff, KvS., Propper, C.R., 1998. The water absorption response:a behavioral assay for physiological processes in terrestrial amphibians.Physiol. Zool. 71, 127–138.

Jørgensen, C.B., 1994. Water economy in a terrestrial toad (Bufo bufo), withspecial reference to cutaneous drinking and urinary bladder function. Comp.Biochem. Physiol. 109A, 311–324.

Katz, U., 1987. The effect of salt acclimation of the water uptake and osmoticpermeability of the skin of the toad (Bufo viridis, L.). J. Physiol. 82, 183–187.

Kawada, J., Taylor, R.E., Barker, S.B., 1969. Measurement of Na–K ATPase inthe separated epidermis of Rana catesbeiana frogs and tadpoles. Comp.Biochem. Physiol. 30, 965–975.


Konno, N., Hyodo, S., Takei, Y., Matsuda, K., Uchiyama, M., 2005. Plasmaaldosterone, angiotensin II and arginine vasotocin concentrations in the toad,Bufo marinus following osmotic treatments. G. Comp. Endocrinol. 140,86–93.

McClanahan Jr., L., Baldwin, R., 1969. Rate of water uptake through theintegument of the desert toad, Bufo punctatus. Comp. Biochem. Physiol. 28,381–389.

Nielsen, R., Larsen, E.H., 2007. Beta-adrenergic activation of solute coupledwater uptake by toad skin epithelium results in near-isosmotic transport.Comp. Biochem. Physiol. A 148, 64–71. doi:10.1016/j.cbpa.2006.12.041.

Parsons, R.H., Mobin, F., 1991. Water flow across the pectoral and pelvic patchin Rana catesbeiana. Physiol. Zool. 64, 812–822.

Rajerison, R.M., Montegut, M., Jard, S., Morel, F., 1972. The isolated frog skinepithelium: presence of a and b adrenergic receptors regulating activesodium transport and water permeability. Pfluegers Arch. 332, 313–331.

Roth, J.J., 1973. Vascular supply to the ventral pelvic skin of anurans as relatedto water balance. J. Morphol. 140, 443–4760.

Ruibal, R., 1962. The adaptive value of bladder water in the toad, Bufocognatus. Physiol. Zool. 35, 218–223.

Slivkoff, M.D., Warburton, S.J., 2001. Angiotensin II alters blood flowdistribution in amphibians. Physiol. Biochem. Zool. 50, 2377–2389.

Steinbach, H.B., 1967. On the ability of isolated frog skin to manufactureRinger's fluid. J. Gen. Physiol. 50, 2377–2389.

Sullivan, P.A., Hoff, KvS., Hillyard, S.D., 2000. Effects of anion substitution onhydration behavior and water uptake of the red-spotted toad, Bufo punctatus:is there an anion paradox in amphibian skin? Chem. Senses 25, 167–172.

Suzuki, 2006. From the SEB Symposium Abstract.Tanaka, S., Hasegawa, T., Tanii, H., Suzuki, M., 2005. Immunocytochemical

and phylogenetic distribution of aquaporins in the frog ventral skin andurinary bladder. Ann. N. Y. Acad. Sci. 1040, 483–485.

Viborg, A.L., Hillyard, S.D., 2005. Cutaneous blood flow and water absorptionby dehydrated toads. Physiol. Biochem. Zool. 78, 394–404.

Viborg, A.L., Rosenkilde, P., 2004. Water potential receptors in the skin regulateblood perfusion in the ventral pelvic patch of toads. Physiol. Biochem. Zool.77, 39–49.

Viborg, A.L., Wang, T., Hillyard, S.D., 2006. Cardiovascular and behaviouralchanges during water absorption in toads, Bufo alvarius and Bufo marinus.J. Exp. Biol. 209, 834–844.

Wentzell, L.A., McNeil, S.A., Toews, D.P., 1993. The role of the lymphaticsystem in water balance processes in the toad Bufo marinus (L.). Physiol.Zool. 66, 307–321.

Willumsen, N.J., Amstrup, J., Nejsum, L.N., Larsen, E.H., Nielsen, S., Hillyard,S.D., 2003. Differential localization of aquaporins 1–3 in amphibian skin.FASEB J. 17, A919.

Word, J.M., Hillman, S.S., 2005. Osmotically absorbed water preferentiallyenters the cutaneous capillaries of the pelvic patch in the toad Bufo marinus.Physiol. Biochem. Zool. 78, 40–47.


vascular aspects of water uptake mechanisms in the toad skin: perfusion, diffusion, confusion

Documents