Effect of the liquorice root derivatives on salt and water balance in a teleost fish, rainbow trout (Oncorhynchus mykiss)

Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part A journal homepage: www.elsevier.com/locate/cbpa

Effect of the liquorice root derivatives on salt and water balance in a teleost fish, rainbow trout (Oncorhynchus mykiss) Chun Chih Chen ⁎, Dennis Kolosov, Scott P. Kelly Department of Biology, York University, Toronto, Ontario, Canada M3J 1P3

a r t i c l e

i n f o

Article history: Received 19 June 2014 Received in revised form 27 August 2014 Accepted 3 September 2014 Available online 12 November 2014 Keywords: Glycyrrhizic acid Glycyrrhetinic acid Cortisol 11β-Hydroxysteroid dehydrogenase type 2 Gill Ion transport Tight junction

a b s t r a c t The effect of liquorice root derivatives (LRDs) glycyrrhizic acid (GL) and glycyrrhetinic acid (18βGA) on salt and water balance and end points of gill ion transport in a freshwater teleost, (rainbow trout) was examined after feeding fish diets containing GL or 18βGA (0, 5, 50 or 500 µg/g diet) for a two week period. Serum cortisol levels and gill 11β-hydroxysteroid dehydrogenase type 2 mRNA abundance decreased in fish fed GL but increased (at select doses) in fish fed 18βGA. At higher doses of GL, gill Na+-K+-ATPase and H+-ATPase activity increased, while cystic fibrosis transmembrane conductance regulator type II mRNA abundance significantly decreased at the lowest dose of GL. End points of gill transcellular ion transport were not significantly altered in fish fed 18βGA, except for a reduction in Na+-K+-ATPase activity at a 50 µg/g dose. In contrast, high doses of GL and 18βGA increased gill transcript abundance of the tight junction protein claudin-31 (cldn-31). Other end points of gill paracellular transport differed in fishes fed LRDs. Tricellulin mRNA abundance was increased by high dose GL and decreased by high dose 18βGA, and cldn-23a and cldn-27b mRNA abundance significantly decreased in response to GL irrespective of dose. Despite the above observations, systemic end points of salt and water balance (i.e. serum [Na+] and [Cl−] as well as muscle moisture) were unaffected by LRDs. Therefore data suggest that LRDs can alter end points of ion transport in fishes but that overall salt and water balance need not be perturbed. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Liquorice plants (Glycyrrhiza spp.) are perennial legumes that possess a root with numerous medicinal properties (for reviews, see Davis and Morris, 1991; Nassiri Asl and Hosseinzadeh, 2008). Liquorice root (LR) and liquorice root derivatives (LRDs) have well-documented effects on the endocrine system of mammals (Davis and Morris, 1991). The saponin-glycoside glycyrrhizic acid (GL) is a major bioactive constituent of LR, and upon ingestion, GL can be hydrolyzed into a single molecule of glycyrrhetinic acid (18βGA) and two molecules of glucuronic acid by gastrointestinal β-glucuronidase (Kim et al., 1996; Nassiri Asl and Hosseinzadeh, 2008). Both GL and 18βGA are steroid compounds that can interact with corticosteroid receptors of vertebrate tissues. While some evidence suggests that GL and 18βGA bind directly to both the glucocorticoid receptor (GR) and mineralocorticoid receptor (MR) (Ulmann et al., 1975; Armanini et al., 1983), the corticosteroidmediated effect of LR in mammals is primarily thought to occur indirectly through the modulated actions of aldosterone on the MR. Specifically, 18βGA has been demonstrated to inhibit 11β-hydroxysteroid

⁎ Corresponding author at: Department of Biology, York University, 4700 Keele Street, Toronto, ON, Canada M3J 1P3. Tel.: +1 416 736 2100x77835; fax: +1 416 736 5698. E-mail address: [email protected] (C.C. Chen).

http://dx.doi.org/10.1016/j.cbpa.2014.09.041 1095-6433/© 2014 Elsevier Inc. All rights reserved.

dehydrogenase type 2 isoform (11β-HSD2), an enzyme co-localized with MR and responsible for converting cortisol into inactive cortisone, which in turn prevents overstimulation of MR by cortisol (Tanahashi et al., 2002). The actions of LRDs on 11β-HSD2 thereby lead to an increase in cortisol availability in mineralocorticoid target tissues (Armanini et al., 2004). As a consequence, although LR is a broadly acknowledged botanical with many medicinal benefits, the harmful consequences of excess consumption of LR or LRDs manifest as pseudohyperaldosteronism that can cause hypertension as well as salt and water imbalance (Ruszymah et al., 1995; Armanini et al., 2004; Ferrari, 2010; Al-Dujaili et al., 2011; Lin et al., 2012). To date, work that has examined the effect of LR or LRDs on the vertebrate endocrine system or endocrine-mediated physiology of vertebrates has focused almost exclusively on mammalian models or mammalian derived model systems. Nevertheless, 18βGA has been reported to potentiate aldosterone-mediated active Na+ transport (uptake) across frog skin (Ishikawa and Saito, 1980), and more recently Alderman and Vijayan (2012) have reported that in zebrafish 18βGA decreased brain 11β-HSD2 activity with an associated increase in whole body cortisol levels. This suggested that 11β-HSD2 can potentially affect the negative feedback loop of corticosteroid synthesis in zebrafish. Therefore, there is evidence to suggest that LRDs can interact with corticosteroid sensitive tissues in aquatic (or semi-aquatic) vertebrates as well as terrestrial vertebrates, and in teleost fishes, this

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

would seem to occur despite the functional absence of aldosterone (Prunet et al., 2006). The gills occupy a central role in the maintenance of homeostasis in teleost fishes. The gill epithelium directly interfaces with a surrounding medium of water and is simultaneously involved in respiration, nitrogenous waste excretion, acid/base balance and iono/osmoregulation (Perry, 1997; Evans et al., 2005). Therefore, the gill epithelium is one of the first tissues to experience environmental perturbation (or an external source of stress) as well as a key tissue that copes with, or compensates for systemic change as a consequence of an internal stress. As a result, the influence of corticosteroid hormones on teleost fish gill physiology is well documented, and in particular, cortisol-mediated changes in the contribution of the gill epithelium to iono/osmoregulatory homeostasis (Laurent and Perry, 1990; McCormick, 2001; McCormick and Bradshaw, 2006). In addition, it has been proposed that the effects of cortisol on the paracelllular transport properties of the gill also play an important role in the regulation of teleost fish salt and water balance (Kelly and Wood, 2001a, 2002a, 2002b; Tipsmark et al., 2009; Bui et al., 2010; Chasiotis et al., 2010). In both cases, cortisol signaling via teleost fish GRs and the MR are suggested to contribute to alterations in epithelial solute transport or epithelial transport end points (Kelly and Chasiotis, 2011). At the tissue-level (local) regulator of circulating cortisol, 11β-HSD2 has been described in teleost fishes (Kusakabe et al., 2003; Baker, 2004). Furthermore, a role for gill 11β-HSD2 in protecting the MR from cortisol activation has also been proposed in teleost fishes (Kiilerich et al., 2007). Therefore, it seems reasonable to contend that if LR or LRDs were to influence salt and water balance in a teleost fish, this will occur at least in part, through their direct or indirect actions on the gill tissue. To consider this idea further, it was hypothesized that if teleost fishes are sensitive to LRDs in a manner that is consistent with other vertebrates (despite the absence of aldosterone), then teleost fish salt and water balance as well as gill ionoregulatory elements may be altered by the presence of LRDs. If aquatic vertebrates such as fishes were to be considered a model system for studying the effects of LR or LRDs on the vertebrate endocrine system (with the goal of expanding our understanding of the medicinal benefits of these botanicals), a broader understanding of the effects of LR or LRDs on the endocrinemediated physiology of these organisms would be very useful. Indeed, it also seems reasonable to consider the idea that in a commercial aquaculture setting, LR or LRDs could be used for their medicinal benefits to fish health itself. Therefore, the specific objective of the current study was to examine the effect of the LRDs GL and 18βGA on salt and water balance in freshwater rainbow trout as well as end points of epithelial ion transport in the gill epithelium. To the best of our knowledge, this is the first study to consider the effect of LR or LRDs on salt and water balance in fishes. 2. Materials and methods 2.1. Experimental animals Rainbow trout (Oncorhynchus mykiss) were obtained from Humber Springs Trout Club and Hatchery, Mono, ON, Canada. Rainbow trout used for feeding experimental diets were ~15 g. Following transfer to the laboratory, these fish were held in 200 L opaque polyethylene aquaria (n = 24 fish per tank) supplied with flow-through dechlorinated City of Toronto tap water (approximate composition in μM: Na+ 590; Cl− 920; Ca2+ 900; K+ 50; pH 7.4). Water was aerated and during acclimation to laboratory conditions fish were fed ad libitum once daily with commercial trout pellets (Martin Profishent, Elmira, ON, Canada). Water temperature was 11 ± 1 °C, and fish were held under a constant photoperiod of 12 h light:12 h dark. Rainbow trout used for primary cell culture were ~ 450 g. These animals were held in 600 L opaque polyethylene aquaria under conditions identical to those described previously. Fish husbandry, animal experiments, and tissue

87

collection methods were conducted in accordance with an approved York University Animal Care Protocol that conformed to the guidelines of the Canadian Council on Animal Care. 2.2. Preparation and administration of experimental diets Glycyrrhizic acid (GL; Glycyrrhizic acid ammonium salt from Glycyrrhiza root) and 18β-glycyrrhetinic acid (18βGA) were obtained from Sigma-Aldrich, Oakville, ON, Canada. Diets containing various doses of GL (5, 50, and 500 μg GL/g diet) or 18βGA (5, 50, and 500 μg 18β GA/g diet) were prepared by dissolving GL or 18βGA in absolute ethanol and evenly spraying ethanol solutions onto commercial trout pellets (Martin Profishent, Elmira, ON, Canada). The ethanol solution was allowed to evaporate at room temperature and diets were stored at − 20 °C until use. Control diets (0 μg GL/g diet or 0 μg 18β GA/g diet) were prepared using absolute ethanol only but otherwise treated identically to experimental diets. Fish were fed experimental diets over a 14-day period at a 3% body weight ration once daily on a scheduled regime (i.e., at the same time each day). 2.3. Tissue sampling Rainbow trout were rapidly net-captured and anesthetized in 1.0 g∙L− 1 tricaine methanesulfonate (MS-222; Syndel Laboratories Ltd., Canada). Blood was sampled by multiple individuals to reduce sampling time (typically complete within 5 min). Fish were then killed by spinal transection. Blood was allowed to clot at 4 °C for 30 min, then it was centrifuged (4 °C, 10,600 g, 10 min), and the resulting serum was aliquoted and stored at −80 °C until further analyses. Gill arches from the left branchial basket were collected for RNA extraction and enzyme activity assays. All samples were flash frozen in liquid nitrogen and stored in −80 °C until further analyses. A gill arch from the right basket was collected for scanning electron microscopy (SEM). A flank of epaxial white muscle (dorsal to the lateral line) was collected to determine muscle moisture content. 2.4. Serum analyses, muscle moisture content, and gill enzyme activities Serum cortisol levels were quantified using a cortisol EIA kit (Oxford Biomedical Research, Oakland County, Michigan). Serum [Na+] was determined by atomic spectroscopy using an AAnalyst 200 spectrometer (PerkinElmer Life and Analytical Sciences, Woodbridge, ON, Canada). Serum [Cl−] was determined according to methods outlined by Zall et al. (1956) using a Multiskan™ Spectrum microplate reader (Thermo Electron Corp, Nepean, ON, Canada). Muscle moisture content was determined gravimetrically by drying a known weight of fresh muscle tissue to a constant weight at 60 °C and expressing the weight loss (i.e., moisture content) as a relative percentage of the initial wet weight. Gill Na+-K+-ATPase (NKA) activity was determined according to McCormick (1993). In brief, enzyme activity was examined by recording the NADH oxidation rate of samples either in the absence or presence of specific enzyme inhibitors. To inhibit NKA activity, 0.5 mM oubain (Sigma-Aldrich) was used. To inhibit vacuolar-types H+-ATPase (VA), 25 nM bafilomycin A1 (BioShop Canada Inc., Burlington, ON, Canada) was used. Enzyme assays were run at room temperature using a Multiskan™ Spectrum microplate reader, and following this, protein concentration in supernatant was determined using Bradford Reagent (Sigma Life Science, MO, USA). Enzyme activities are expressed as μmol ADP∙mg protein−1∙h−1. 2.5. RNA extraction, cDNA synthesis and qRT-PCR analyses Total RNA was extracted from rainbow trout gill tissue using TRIzol reagent (Life Technologies, Burlington, ON, Canada) according to the manufacturer instructions. Total RNA yield from each sample was determined using a Multiskan™ Spectrum spectrophotometer and

88

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

treated with DNase I (Amplification Grade; Life Technologies) prior to cDNA synthesis. First-strand cDNA synthesis was carried out using SuperScript™ III Reverse Transcriptase and Oligo(dT)12–18 primers (Invitrogen Canada Inc.). Quantitative real-time PCR (qRT-PCR) was used to assess transcript abundance of genes encoding corticosteroid

receptors and 11β-HSD2 as well as transcript abundance of transcellular and paracellular transport genes in rainbow trout gill tissue. Transcellular ion transport genes examined were NKA-α1a subunit (nka α1a), NKA-α1b subunit (nka α1b), vacuolar type H +-ATPase β subunit (vaβ), cystic fibrosis transmembrane conductance regulator II

Table 1 Primer sets and corresponding amplicon sizes, annealing temperatures and GenBank accession numbers for genes encoding 11β-hydroxysteroid dehydrogenase type 2 (11β-hsd2), corticosteroid receptors (gr1, gr2 and mr), transcellular transport proteins (nka α1a, nka α1b, vaβ, cftr II, nbc, nhe2, and nkcc1a) and paracellular transport proteins (ocln, tric, and cldns), and reference genes (actb, ef1α) in rainbow trout. Gene

Primer sequence (5′➞3′)

Amplicon size (bp)

Annealing temperature (°C)

Accession number

11β-hsd2

FOR: GGCTGTGGTTCTTCTGTGTG REV: GATGAGGGGAGCAGGTCTTC FOR: GGACTGAAACACAGCAAGGAC REV: GCAATACTCGCCTCCAACAG FOR: AGAACACGTCTGCCATGC REV: CTGGAGAAAGCGGAGGTAG FOR: TGTGTCTGGGTAATGGTAGC REV: CGTTGTTGTTGTTCTCTTGG FOR: AGAAAGCCAAGGAGAAGATG REV: AGCCCGAACCGAGGATAGAC FOR: AGCAAGGGAGAAGAAGGACA REV: GAGGAGGGGTCAGGGTG FOR: CAACCCTCAGTGCCGTATC REV: GAAGAAGCGAGCAGTTTCC FOR: CTCAAGGCATCACCATCG REV: CTCGGGCACCACAGAGAAC FOR: GATGAGGAGGAGGTGGAAGG REV: GCAGAGATGAGGGGTTTGAG FOR: GTAGCCTCCTTCACCACACG REV: ATGTAGCCCCAGTTCCACTC FOR: CTGTATTCCACGCCTCTCTG REV: GTTCACATCGGGCTTCTTG FOR: CAGCCCAGTTCCTCCAGTAG REV: GCTCATCCAGCTCTCTGTCC FOR: GTCACATCCCCAAACCAGTC REV: GTCCAGCTCGTCAAACTTCC FOR: AAGGAAGGTCTGGAGGAAGG REV: CAGCTTGCCGTTGTAGAGG FOR: GAGGACCAGGAGAAGAAGG REV: AGCCCCAACCTACGAAC FOR: TGGATCATTGCCATCGTGTC REV: GCCTCGTCCTCAATACAGTTGG FOR: CTCACCGTCATCTCGTGTG REV: CATCCAGCAGAGGGGAAC FOR: CGTCCTGCTGATTGGATCTC REV: CAAACGTACTCCTTGCTGCTG FOR: GCAGTGTAAAGTGTACGACTCTCTG REV: CACGAGGAACAGGCATCC FOR: CCTGGTCTGCTCTACAATGC REV: CCCGAAGAATCCCAAAATAA FOR: CTGCTTTGTGTCGTGTCTGG REV: AGCGAAGAACCCAAGGATG FOR: CTGTCACCAACTGCCAAGAC REV: CCAGAAAGCCACTGATGATG FOR: CTTCATCATCGCCTTCATCTC REV: GAGCCAAACAGTAGCCCAGTAG FOR: ATCCTAAACCTCACAGCGACA REV: CGGTCTTTCCAGCACCTTAC FOR: GCCAACATCGTAACAGGACA REV: CCAGAAGAGCACCAATGAGC FOR: CTTTCATCGGAGCCAACATC REV: CAGACAGGGACCAGAACCAG FOR: CTTTCATCGGCAATAACATC REV: CAGCAATGGAGAGCAGG FOR: CGGCGAGAACATAATCACAG REV: GGGATGAGACACAGGATGC FOR: TCGGCAACAACATCGTGAC REV: CGTCCAGCAGATAGGAACCAG FOR: ATTGTGTGCTGTGCCATCC REV: AGACACCAACAGAGCGATCC FOR: GTCCACAGACCTCTTGCTCAC REV: TCCTGCCACCTTCATAATCC FOR: GGACTTTGAGCAGGAGATGG REV: GACGGAGTATTTACGCTCTGG FOR: GGCAAGTCAACCACCACAG REV: GATACCACGCTCCCTCTCAG

257

59

NM_001124218.1

335

59

NM_001124730

346

57

NM_001124482

369

56

AY495584

133

56

NM_001124461.1

176

59

NM_001124460.1

259

57

AF140022.1

209

59

BK008785

203

57

AF434166.1

299

56

EF446605.2

197

58

BK008786

341

58

GQ476574

170

60

KC603902

291

60

HQ656020

182

60

BK008768

285

60

BK007964

171

59

BK008769

261

61

BK007965

200

60

BK007966

223

58

BK008772

253

60

BK008773

216

54

KF445437

255

60

BK007967

270

60

BK008775

283

60

BK008776

310

60

EU921670

201

60

BK008777

297

59

BK007968

311

61

BK007969

321

60

BK007970

268

60

BK008778

354

58

AF157514

159

60

AF498320.1

gr1 gr2 mr nka α1a nka α1b vaβ cftr II nbc nhe2 nkcc1a ocln tric zo-1 cldn-1 cldn-3a cldn-5a cldn-7 cldn-8d cldn-10c cldn-10d cldn-10e cldn-12 cldn-23a cldn-27b cldn-28b cldn-29a cldn-30 cldn-31 cldn-32a cldn-33b actb ef1α

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97 + + (cftrII), Na+/HCO− exhanger 2 (nhe2), 3 cotransporter (nbc), Na /H and Na+ /K +/Cl− cotransporter 1a (nkcc). Paracellular transport, or tight junction (TJ), protein genes examined were occludin (ocln), tricellulin (tric), zonula occludens-1 (zo-1), and claudin (cldn)-1, -3a, -5a, -6, -7, -8d, -10c, -10d, -10e, -12, -23a, -27b, -28b, -29a, -30, -31, -32a, and -33b. The primer sets used, specific annealing temperatures, amplicon sizes, and GenBank accession numbers for all genes can be found in Table 1. Analysis was conducted using SYBR Green I Supermix (Bio-Rad Laboratories Canada Ltd., ON, Canada) in a Chromo4™ Detection System (CFB-3240, Bio-Rad Laboratories Canada Ltd.) under the following conditions: 1 cycle denaturation (90 °C, 4 min) followed by 40 cycles of denaturation (95 °C, 30s), annealing (54–61 °C, 30s), and extension (72 °C, 30s), respectively. A final melting curve was generated from 1 °C below primer annealing temperature up to 95 °C. Gene of interest qRT-PCR results were normalized using transcript abundance of reference genes [βactin (actb) and elongation factor 1α (ef1α)] that did not significantly alter in response to the treatment regimes (P ≥ 0.05).

2.6. Scanning electron microscopy (SEM) and gill morphometrics SEM samples were prepared using methods outlined previously (Chasiotis et al., 2012). Dehydrated gill filaments were mounted onto SEM stubs using a double-sided adhesive tape. Samples were then sputter coated (Hummer VI Au/Pd 40/60, Anatech USA; 2 min) and subsequently examined using a Hitachi S-520 SEM (Hitachi HighTechnologies Canada, Toronto, Canada) attached to a Quartz PCI Version 6 image capture system (Quartz Imaging, Vancouver, Canada). Image analysis was performed using ImageJ software (National Institutes of Health, Java 1.6.0_45 (64-bit)). Gill morphometric measurements were determined according to methodology previously reported (Chasiotis et al., 2012).

89

(Ω cm2) after correcting for background resistance (TER measured across 'blank' culture inserts bathed with L15 on both sides). Flux of the paracellular permeability marker polyethylene glycol (molecular mass 400 Da; [3H]PEG-400; PerkinElmer, Woodbridge, ON, Canada) was determined once the epithelia had reached a plateau in TER. [3H] PEG-400 was added to the basolateral side of epithelia, and its rate of movement to the apical side ascertained using methods and calculations previously described (Wood et al., 1998). 2.8. Statistical analyses All data are expressed as mean values ± SEM, where n represents the number of fish per treatment group or number of cultured gill epithelium inserts per treatment group. Significant differences (P ≤ 0.05) between treatment groups were detected using a one-way analysis of variance (ANOVA) followed by a Holm–Sidak multiple comparison procedure unless otherwise indicated (SigmaPlot Build 11.0.0.77, Systat Software Inc., www.sigmaplot.com). 3. Results 3.1. Effect of GL or 18βGA on serum [Na+], [Cl−], cortisol, and muscle moisture content in rainbow trout Following a 2-week experimental period, varying doses of dietary GL or 18βGA did not significantly alter serum [Na+], [Cl−], or muscle moisture content in fish (Table 2). GL treatment significantly decreased circulating cortisol levels at the highest dietary dose (500 μg/g diet, Fig. 1A). Significantly increased serum cortisol levels were observed in fish fed the lowest dietary level of 18βGA (5 μg/g diet, Fig. 1B), while higher doses did not significantly alter serum cortisol levels.

2.7. The in vitro effects of GL and 18βGA on a primary cultured rainbow trout gill epithelium

3.2. Effect of GL or 18βGA on mRNA abundance of 11β-HSD2 and corticosteroid receptors in rainbow trout gill tissue

Procedures for the preparation and culture of rainbow trout gill epithelia were based on those first reported by Wood and Part (1997) for the preparation and culture of reconstructed gill epithelia composed of gill pavement cells only. Detailed methodology has been outlined by Kelly et al. (2000). In brief, rainbow trout gill cells were isolated by trypsination and cultured in 25 cm2 cell culture flasks until they reached confluence. Flasks were held in an air atmosphere at 18 °C, and cells were bathed in Leibovitz's culture medium supplemented with 6% fetal bovine serum and antibiotics (hereafter referred to as L15). At confluence, flask grown cells were harvested by trypsination and seeded into cell culture inserts (polyethylene terephthalate filters, 0.9 cm2 growth area, 0.4 μm pore size, and 1.6 × 106/cm2 pore density; BD Falcon TM; BD Biosciences, Mississauga, ON, Canada). Inserts were housed in companion cell culture plates (Falcon BD), and the apical and basolateral compartments of the culture system contained L15 medium. Cultured gill epithelia were composed of pavement cells only. Cultured gill epithelia were exposed to either 0 (control), 50, 500 or 5000 ng/ml GL or 18βGA, cortisol (500 ng/ml; hydrocortisone 21-hemisuccinate sodium salt; Sigma-Aldrich), dexamethasone (100 ng/ml; Sigma-Aldrich), or combinations of these factors immediately after seeding gill cells on to cell culture inserts. Cortisol, dexamethasone, and/or LRDs were present on basolateral side of cultured gill epithelium preparations only and renewed at each media change (i.e., first media change after 24 h in culture and thereafter every 48 h). Epithelia were exposed to cortisol, dexamethasone, and/or GL or 18GA until a plateau in transepithelial electrical resistance (TER) was measured (~ 6 days after seeding on to culture inserts). Measurements of TER were obtained using chopstick electrodes (STX-2) connected to a custom-modified EVOM epithelial volt ohmmeter (World Precision Instruments, Sarasota, FL) and expressed as ohms per centimeter squared

GL treatment significantly decreased 11β-HSD2 mRNA abundance at all doses examined (5, 50 and 500 μg/g diet, P ≤ 0.01) (Fig. 2A), but there was no significant difference between GL doses. In contrast, 18βGA did not significantly alter 11β-HSD2 mRNA abundance at lower dietary doses (5 and 50 μg/g diet) and significantly elevated 11β-HSD2 mRNA abundance (P = 0.013) at the highest dietary dose of 500 μg/g (Fig. 2B). Relative to the control fish, GL or 18βGA treatments did not significantly change mRNA abundance of corticosteroid receptors (Table 3). 3.3. Effect of GL or 18βGA on rainbow trout gill ionomotive enzyme activity, transcellular transport protein mRNA abundance, and surface ionocyte morphometrics GL treatment significantly increased gill NKA activity at the highest dietary dose used (500 μg/g diet, P = 0.05; Fig. 3A) and VA activity at 50 and 500 μg/g diet (P ≤ 0.01; Fig. 3B). 18βGA significantly reduced gill NKA activity (P b 0.01), but only at a dietary dose of 50 μg/g (Fig. 3C). 18βGA did not significantly change gill VA activity (Fig. 3D). In contrast to the alterations observed in ionomotive enzyme activity, the abundance of mRNA encoding NKA-α1a and NKA-α1b subunits as well as VAβ did not significantly alter following dietary GL or 18βGA treatment (Table 4). The effect of dietary GL or 18βGA treatment on mRNA abundance of cftrII, nbc, nhe2, and nkcc in rainbow trout gill tissue only revealed a significant decline in cftrII in the 5 μg/g GL-treated group (P b 0.01) (Table 4). Dietary GL or 18βGA treatment also did not significantly alter ionocyte morphometrics (i.e., ionocyte fractional surface area, individual surface area, or number of ionocytes exposed) in rainbow trout gills (Fig. 4).

90

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

Table 2 Effect of glycyrrhizic acid or glycyrrhetinic acid on muscle moisture content, serum [Na+], and serum [Cl−] in rainbow trout. μg/g diet

Glycyrrhizic acid

0 5 50 500

Glycyrrhetinic acid

Muscle moisture (%)

Serum [Na+] (mM)

Serum [Cl−] (mM)

Muscle moisture (%)

Serum [Na+] (mM)

Serum [Cl−] (mM)

80.48 80.11 80.45 79.56

158.44 158.10 155.51 160.99

123.78 123.21 122.22 125.67

79.98 79.91 80.48 79.16

153.37 150.85 155.75 147.93

120.56 118.33 123.20 120.39

± ± ± ±

0.26 0.32 0.19 0.30

± ± ± ±

3.21 3.50 2.98 3.58

± ± ± ±

2.78 3.26 2.99 2.75

3.4. Effect of GL or 18βGA on mRNA abundance of tight junction (TJ) proteins in the gill of rainbow trout Dietary GL or 18βGA treatment did not significantly alter ocln or zo-1 mRNA abundance in the gill of rainbow trout (Fig. 5). In contrast, GL caused an apparent increase in tric mRNA abundance that was significant (P = 0.014) in the 500 μg/g treatment group (Fig. 5C), while 18βGA treatment significantly reduced (P = 0.027) tric mRNA abundance in the 500 μg/g treatment group (Fig. 5D). Of the 18 members of the cldn superfamily of TJ proteins examined following dietary GL or 18βGA treatment, transcript abundance of the majority was not significantly affected (Fig. 6). However, significant changes were observed in cldn-12 (P b 0.01), cldn-23a (P b 0.05), cldn-27b (P b 0.01), and cldn-31 (P b 0.05) in GL dietary treatment groups (Fig. 6A). Changes in cldn

0.23 0.51 0.23 0.50

3.64 2.22 3.28 1.75

± ± ± ±

2.15 0.95 2.19 2.56

3.5. The in vitro effects of GL or 18βGA on primary cultured rainbow gill epithelia Exposure of primary cultured trout gill epithelia to GL did not significantly alter transepithelial electrical resistance (TER) (Fig. 7A) or flux rates of the paracellular permeability marker [3H]PEG-400 (Fig. 7B). In epithelia exposed to 18βGA, a small but significant increase in TER was observed at the lowest treatment dose of 50 ng/ml, but no differences in TER were observed at other doses (Fig. 7C). In the same preparations, [3H]PEG-400 flux did not significantly alter in any of the

A 150

6 5 4 3 2 1 0

0

5

50

125 100 75 50 25 0

500

0

5

50

500

GL (μg/g diet)

GL (μg/g diet)

B

B

150

11 β -hsd2 Abundance (%)

6

Serum Cortisol (ng/mL)

± ± ± ±

mRNA abundance in 18βGA-treated trout were detected for cldn-6 (P b 0.05), cldn-8d (P b 0.01), cldn-10e (P b 0.05), and cldn-31 (P b 0.05) (Fig. 6B).

11 β -hsd2 Abundance (%)

Serum Cortisol (ng/mL)

A

± ± ± ±

5 4 3 2 1 0

125 100 75 50 25 0

0

5

50

500

18β GA (μg/g diet) Fig. 1. Effect of dietary (A) glycyrrhizic acid (GL) and (B) glycyrrhetinic acid (18βGA) on serum cortisol levels in rainbow trout. Data are expressed as mean values ± SEM (n = 10). Significant differences between treatment groups is indicated by different letters (P b 0.05).

0

5

50

500

18β β GA (μg/g diet) Fig. 2. Effect of dietary (A) glycyrrhizic acid (GL) and (B) glycyrrhetinic acid (18βGA) on gill 11β-hydroxysteroid dehydrogenase type 2 (11β-hsd2) mRNA abundance. Data are expressed as mean values ± SEM (n = 10). Significant differences between treatment groups is indicated by different letters (P b 0.05).

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

91

Table 3 Effect of glycyrrhizic acid or glycyrrhetinic acid on mRNA abundance of genes encoding glucocorticoid receptor 1 and 2 (gr1 and gr2), and the mineralocorticoid receptor (mr) in gill tissue of rainbow trout. Gene

gr1 gr2 mr

Glycyrrhizic acid (μg/g diet)

Glycyrrhetinic acid (μg/g diet)

0

5

50

500

0

5

50

500

100 ± 10.7 100 ± 9.6 100 ± 8.9

101.2 ± 8.8 91.5 ± 5.4 91.9 ± 7.6

110.6 ± 17.4 86.8 ± 5.7 87.0 ± 8.9

126.4 ± 8.3 101.3 ± 7.1 120 ± 4.9

100 ± 10.1 100 ± 8.6 100 ± 11.0

95.5 ± 5.2 82.1 ± 6.1 91.4 ± 7.2

109.3 ± 7.7 83.2 ± 7.0 88.6 ± 9.3

107.4 ± 9.6 94.5 ± 8.8 105.9 ± 8.3

18βGA doses used (Fig. 7D). Combining GL or 18βGA with cortisol (Fig. 8A, B) or dexamethasone (a GR agonist) (Fig. 8C, D) treatment of cultured gill epithelia did not significantly alter the effect of either steroid on TER or [3H]PEG-400 permeability. 4. Discussion 4.1. Overview The current study provides a first look at the effect of LRDs on salt and water balance in a teleost fish as well as end points of ion transport

in teleost fish gills. The data from this set of experiments do not allow the original hypothesis that LRDs will alter salt and water balance in fishes to be accepted in so far as systemic ion and water levels appear to be unperturbed. That is, serum Na+ and Cl− as well as muscle moisture content typically change when fishes experience ionoregulatory disequilibrium (e.g., Kelly et al., 1999), but no significant alterations were observed in serum [Na+], [Cl−], and muscle moisture content in fish fed diets containing GL or 18βGA. Nevertheless, LRDs did significantly alter circulating cortisol levels, gill mRNA abundance of 11β-hsd2, as well as select molecular and/or biochemical end points of salt and water balance in gill tissue. This would suggest that LRDs can

A

C

B

D

Fig. 3. Effect of dietary (A, B) glycyrrhizic acid (GL) and (C, D) glycyrrhetinic acid (18βGA) on trout gill Na+-K+-ATPase and H+-ATPase activity. Data are expressed as mean values ± SEM (n = 10). Significant differences between treatment groups is indicated by different letters (P b 0.05).

92

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

Table 4 Effect of glycyrrhizic acid or glycyrrhetinic acid on mRNA abundance of genes encoding transcellular transport proteins in gill tissue of rainbow trout. Gene

Glycyrrhizic acid (μg/g diet) 0

nka α1a nka α1b Vaβ cftr II nbc nhe2 nkcc1a

100 100 100 100 100 100 100

Glycyrrhetinic acid (μg/g diet)

5 ± ± ± ± ± ± ±

8.6 7.8 7.8 7.1a 12.9 9.9 14.1

83.7 79.2 87.5 69.3 90.7 81.1 111.9

50 ± ± ± ± ± ± ±

7.1 6.6 5.7 5.4b 10.1 6.1 16.1

98.6 85.3 97.8 78.7 96.6 90.5 114.3

500 ± ± ± ± ± ± ±

10.3 8.3 7.4 8.6a 11.7 5.8 14.9

109.5 74.8 91.3 90.2 121.7 98.1 142.6

0 ± ± ± ± ± ± ±

10.0 3.6 6.6 4.5a 19.8 4.1 15.5

100 100 100 100 100 100 100

5 ± ± ± ± ± ± ±

14.1 5.9 4.8 6.5 13.8 10.6 16.7

86.9 95.8 109.9 89.9 97.8 108.1 102.3

50 ± ± ± ± ± ± ±

8.7 7.0 4.8 7.2 12.8 8.8 15.9

76.8 89.3 103.5 85.4 116.7 99.6 76.2

500 ± ± ± ± ± ± ±

8.7 9.8 8.1 6.4 12.8 10.4 8.6

86.7 90.8 95.6 98.2 148.2 102.3 71.4

± ± ± ± ± ± ±

8.2 6.9 4.0 7.8 34.3 7.3 7.2

Different letter superscripts denote significant differences between dietary doses of glycyrrhizic acid. No significant differences were observed in fish fed different doses of glycyrrhetinic acid.

modulate cortisol metabolism and alter the abundance and/or activity of ionoregulatory transport end points in teleost fishes. In this regard, the hypothesis that LRDs will alter gill ionoregulatory elements can be accepted, but in this study, these changes do not impact overall salt and water balance. The view that LRDs can alter cortisol metabolism is consistent with Alderman and Vijayan (2012) who showed that 18βGA decreased 11β-HSD2 activity and elevated whole body cortisol levels in zebrafish. The latter may have been influenced by LRD-induced corticosteroid-mediated events in ionoregulatory tissues of fishes in part due to changes in circulating cortisol levels. However, the consequences of this do not lead to perturbed salt and water balance, possibly because changes in gill biochemistry may be compensated by other ionoregulatory tissues. Nevertheless, even though LRDs did not change overall hydromineral status in these experiments, changes brought about by LRDs in the ionoregulatory properties of the gill may become significant to salt and water balance in more challenging environmental circumstances. This idea will require further study.

A

B

C

D

E

4.2. The systemic effects of GL and 18βGA in trout The current suite of experiments revealed changes in circulating cortisol levels in rainbow trout following a 2-week period of being fed either GL or 18βGA. In the case of GL-treated fish, reductions in serum cortisol levels were generally observed, being significant at the highest dose of 500 μg/g diet. At first, this seems curious as GL might be expected to increase circulating cortisol levels. This is based on the assumption that GL is converted to 18βGA in fishes by β-glucuronidase as it is in mammals, and that the resulting systemic exposure to 18βGA would inhibit 11β-HSD2 activity, thus reducing the conversion of cortisol to cortisone. This is not an unreasonable assumption as studies have demonstrated hepatic and/or intestinal β-glucuronidase in teleost fishes, which should allow them to convert ingested GL into 18βGA (Daniel and Kahle, 1989; Chilke, 2012). Furthermore, we see an increase in circulating cortisol levels in fish fed 18βGA, albeit at the lowest dose, and a recent study has reported that 18βGA increases whole body cortisol levels and reduces 11β-HSD2 activity in zebrafish (Alderman

G

H

I

J

F

Fig. 4. Morphometrics and scanning electron micrographs of rainbow trout gill following dietary administration of glycyrrhizic acid (GL) or glycyrrhetinic acid (18βGA). Mitochondrionrich cell (MRC) (A, B) fractional surface area, (C, D) individual cell surface area, and (E, F) number of exposed cells can be seen. Representative scanning electron micrographs for (G) GL control and (H) 500 μg/g GL dose, as well as (I) 18βGA control and (J) 50 μg/g doses are shown. All data are expressed as mean values ± SEM (n = 5). Scale bars indicate 10 μm.

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

A

B

C

D

E

F

93

Fig. 5. Effect of dietary (A, C, E) glycyrrhizic acid (GL) and (B, D, F) glycyrrhetinic acid (18βGA) on gill mRNA abundance of genes encoding tight junction proteins occludin (ocln), tricellulin (tric), and ZO-1 (zo-1) in rainbow trout. All data are expressed as mean values ± SEM (n = 10). Significant differences between treatment groups is indicated by different letters (P b 0.05).

and Vijayan, 2012). However, the data support the view that GL and 18βGA may be having different, and in this case opposite, effects in trout with respect to cortisol metabolism. Therefore, this brings into question whether GL is being converted to 18βGA or not. Another layer of complexity is the potential CR binding properties of GL and 18βGA. In mammals, both GL and 18βGA are corticosteroid receptor (CR) ligands, although 18βGA has a higher affinity for CRs than GL (Ulmann et al., 1975; Armanini et al., 1983). Therefore, it is possible that both GL and 18βGA may be acting as CR ligands in trout, but recent evidence suggests that only 18βGA will inhibit 11β-HSD2 activity (Makino et al., 2012). If the effects of GL are primarily confined to

direct CR binding, and little or no GL ends up being converted to 18βGA, a reduction in serum cortisol levels may be a consequence of negative feedback. The low affinity of GL for CRs may also explain why only the highest dietary dose of GL resulted in a significant decrease in circulating cortisol levels. In contrast to GL, 18βGA treatment elevated serum cortisol levels, but only at lowest dietary dose used. Why cortisol levels were not elevated when higher levels of dietary 18βGA were fed to the fish is not entirely clear, but this could be a result of the direct actions of 18βGA on CRs. Nevertheless, and irrespective of circulating cortisol levels, the LRDs examined in this study do not alter overall salt and water balance. This contrasts with previous reports

94

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

A)

B)

Fig. 6. Effect of dietary (A) glycyrrhizic acid (GL) and (B) glycyrrhetinic acid (18βGA) on gill mRNA abundance of genes encoding claudin (cldn) tight junction proteins in rainbow trout. All data are expressed as mean values ± SEM (n = 10). Significant differences between treatment groups is indicated by different letters (P b 0.05).

using mammalian models where excess LRD consumption leads to hyper-activation of the MR by cortisol resulting in LRD-induced pseudohyperaldosteronism manifesting as hypernatremia and altered MR transcript abundance (Ruszymah et al., 1995; Kusano, 2004; Ohtake et al., 2007). 4.3. Effect of GL and 18βGA on 11βhsd2, gr1, gr2, and mr mRNA abundance in trout gills A reduction in 11βhsd2 mRNA abundance was observed in the gills of fishes fed GL. This is consistent with observations in mammals where GL has also been reported to reduce the mRNA and protein abundance of 11βHSD2 (Tanahashi et al., 2002). In trout, this may partly be in response to decreased circulating cortisol levels as 11β-HSD2 had been proposed to play a role in protecting the gill MR from cortisol activation (Kiilerich et al., 2007). Therefore, as the systemic cortisol level decreased, less 11β-HSD2 enzymes would be required for homeostasis at the target tissue. In contrast, 18βGA increased 11βhsd2 mRNA abundance, but only at the highest dietary dose. Therefore, once again we see opposite effects of GL and 18βGA on trout that will require further exploration, possibly an understanding of how GL and 18βGA affect 11β-HSD2 enzyme activity in an isolated ligand binding assay. Yet despite the changes observed in 11βhsd2 mRNA, no significant changes were seen in CRs mRNA abundance, suggesting that at the transcriptional level at least, gill CRs are unaffected by the presence of, or changes brought about by LRDs. 4.4. Effect of GL and 18βGA on end points of transcellular ion transport in trout gills Both GL and 18βGA altered biochemical and molecular end points of transcellular ion transport in the gill tissue of trout. Transcellular ion

transport in the gill of fishes is typically associated with the gill mitochondrion-rich cell (MRC) (for a review, see Evans et al., 2005). However, none of the observed LRD-induced changes in ion transporter activity or transcript abundance were coupled with alterations in MRC (ionocyte) surface morphometrics, because gill MRC morphometrics did not significantly change in response to GL or 18βGA. Typically, differences in the surface area of MRC exposure, be it an increase in the number of cells or an increase in the apical exposure area of an unchanged cell populace, correlates well with differences in ion transport rates across the gill epithelium (Perry et al., 1992a, 1992b). However, if the cells are not exposed to the external environment, no matter what changes take place within MRCs, there will be a limited capacity for these to manifest as changes in ion transport rates. Therefore, one view of the current data is that LRDs are altering the abundance of select transcellular ion transporters, but that unchanged MRC surface morphometrics are not allowing alterations in ion transport machinery to radically alter ion transport rates. This may partly explain why serum ion levels and muscle moisture content remain relatively stable. An alternate view could be that LRDs are changing ion transport rates across the gill epithelium but that this may be to compensate for ion losses elsewhere (in another ionoregulatory tissue that we have not yet examined). For example, GL caused a significant increase in gill NKA and VA activity. These are the two major ionomotive enzymes in the gill epithelium of teleost fishes. Therefore, it is difficult to imagine that an increase in their activity did not impact ion transport rates. However, if an increase in gill ion transport rates is compensating for an LRD-induced compromise in the ion transport activity of another tissue (e.g., the kidney), the overall effect may be unchanged circulating ion levels. Therefore, the effects of LRDs on other ionoregulatory tissues would be an interesting avenue for future consideration. Moreover, the significant decrease observed in cftrII transcript abundance at the lowest GL dose will also

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

A

C

B

D

95

Fig. 7. Effect of (A, B) glycyrrhizic acid (GL) as well as (C, D) glycyrrhetinic acid (18βGA) on transepithelial electrical resistance (TER) and [3H]PEG-400 flux rates across primary cultured rainbow trout gill epithelia. In each set of experiments, cortisol (Co; 500 ng/mL) was added as a positive control. All data are expressed as mean values ± SEM (n = 6–8). Significant differences between treatment groups is indicated by different letters (P b 0.05).

deserve further exploration, as CFTRII is suggested to play a role in salinity transfer in salmonids (Mackie et al., 2007). 4.5. Effect of GL and 18βGA on end points of paracellular ion transport in trout gills At various doses, GL-treated fish gill exhibited a decrease in transcript abundance of cldn-12, cldn-23a, and cldn-27b as well as an increase in the transcript abundance of cldn-31 and tric. Previous studies have reported gill cldn-12, cldn-31, and tric to be dexamethasone and/or cortisol sensitive, while gill cldn-27b has been found to be influenced by environmental pH (Chasiotis and Kelly, 2011; Kelly and Chasiotis, 2011; Kumai et al., 2011; Chasiotis et al., 2012; Kolosov and Kelly, 2013; Kolosov et al., 2013). In addition, increased abundance of cldn-23a, cldn-27b, and cldn-31 are associated with the development of resistive properties in primary cultured trout gill epithelia, which supports the notion that these Cldns contribute to the barrier properties of the gill epithelium (Kolosov et al., 2014). In cultured gill epithelia composed of pavement cells, cldn-23a and cldn-27b are more abundant than cldn-31, and even though cldn-31 is present in pavement cells, a 60-fold increase in its abundance in cultured epithelia composed of

both pavement cells and mitochondrion-rich cells suggest that it is may be highly expressed in the latter (Kolosov et al., 2014). Therefore, a reduction in the presence of barrier forming cldn-23a and cldn-27b, which are otherwise abundant in gill tissue, may indicate a reduction in resistive properties (i.e., increased permeability) of the gill epithelium, while increased cldn-31 may be partly compensating for this in association with elevated NKA and VA activity. In the gills of fish fed a high dose of 18βGA (500 μg/g diet), a decrease in mRNA abundance of tric coupled with an increase in cldn-31 and cldn-8d was observed. In previous studies, cldn-8d and cldn-31 have been demonstrated to increase in response to corticosteroid treatment in cultured trout gill epithelia (Kelly and Chasiotis, 2011), and Tric mRNA abundance has been reported to decrease (Kolosov and Kelly, 2013). Therefore, these results seem consistent if we take 18βGA as an exogenous corticosteroid mimetic that also has the potential to elevate systemic cortisol levels. The results would also suggest that 18βGA may be reducing gill epithelium paracellular permeability. However, these observations are made at the highest dietary dose. In contrast, 18βGA treatment at 50 μg/g diet decreased NKA activity as well as cldn-6 and cldn-10e transcript abundance in the gill. In the spotted green puffer fish, cldn-6 and cldn-10e are

96

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

A

C

B

D

Fig. 8. Effect of (A, B) cortisol (500 ng/mL) and (C, D) dexamethasone (100 ng/mL) with glycyrrhizic acid (GL) and glycyrrhetinic acid (18βGA) on transepithelial electrical resistance (TER) and [3H]PEG-400 flux rates across primary cultured rainbow trout gill epithelia. All data are expressed as mean values ± SEM (n = 6–8). Significant differences between treatment groups is indicated by different letters (P b 0.05).

exclusively associated with gill MRCs (Bui et al., 2010, Bui and Kelly, 2011, Bui and Kelly, 2014). Therefore, the combined reduction of NKA (which is abundant in MRCs), cldn-6 and cldn-10e may indicate a reduction in gill MRC numbers in this group. Gill MRC morphometrics do not support this view, but it is possible that MRCs are reduced in areas of the gill not covered by the morphometric analysis. 4.6. Direct effects of GL and 18βGA on gill epithelium permeability To examine whether GL or 18βGA have direct effects on gill tissue, and more specifically on the permeability properties of the gill epithelium, GL and 18βGA were used to treat primary cultured trout gill epithelia. Primary cultured gill epithelium preparations are very responsive to hormone treatment and have been shown to alter their permeability characteristics in response to cortisol, dexamethasone, 3,5′,3′-triiodo-L-thyronine and prolactin (Kelly and Wood, 2001a, 2001b, 2002a, 2002b; Chasiotis et al., 2010; Chasiotis and Kelly, 2011). In the current study, cortisol and dexamethasone altered the permeability characteristics of cultured gill epithelia as expected by increasing TER and reducing [3H]PEG-400 flux (i.e., paracellular permeability) (Kelly

and Wood, 2001a, 2002b; Kelly and Chasiotis, 2011). However, GL had no effect on TER or [3H]PEG-400 flux, and while 18βGA had a slight, but significant effect, on TER at the lowest dose of 50 ng/ml, no significant effect of 18βGA was seen on [3H]PEG-400 flux at any dose. These data suggest that GL and 18βGA are not capable of directly affecting the permeability of the gill epithelium on their own. Furthermore, treating cultured epithelia with cortisol or dexamethasone in combination with GL and 18βGA indicated that the effect of these steroids on gill epithelium permeability was unperturbed by LRDs in vitro. These results suggest that the effects observed in vivo may be indirect actions of LRDs, potentially arising from corticosteroid negative-feedback. 5. Conclusion and perspectives In mammals, LR and its derivatives have well-documented medicinal properties. However, LR excess has the capacity to deleteriously impact salt and water balance. The harmful effects of LR excess are well understood in mammals but have not been considered to any real extent in other vertebrate groups. Perhaps this is not surprising

C.C. Chen et al. / Comparative Biochemistry and Physiology, Part A 180 (2015) 86–97

since LR is not, to the best of our knowledge, typically ingested by other vertebrates such as fishes. However, there is an increasing interest in the use of botanicals in commercial settings such as the aquaculture industry, where they may be viewed as safe alternatives for promoting fish health, food intake and growth (Citarasu, 2010; Chakraborty and Hancz, 2011). In addition, because of the ever increasing popularity of zebrafish as a model organism, the teleost is rapidly becoming an archetype for vertebrate endocrine research. Therefore, there is no reason why teleosts cannot be used to examine the effects of phytochemicals on the vertebrate endocrine system as well as their potential medical benefits. Finally, while select biochemical and molecular changes were observed in the ion transport properties of rainbow trout gills in response to LRDs, overall salt and water balance was unperturbed. As a result, further study should consider whether this effect of LRDs alters the ability of teleosts to cope with to sudden environmental change (e.g., changes in environmental ion levels) and thus whether LRDs may be beneficial in preparing fishes and/or allowing fishes to cope with environmental insult. References Alderman, S.L., Vijayan, M.M., 2012. 11β-Hydroxysteroid dehydrogenase type 2 in zebrafish brain: a functional role in hypothalamus-pituitary-interrenal axis regulation. J. Endocrinol. 215, 393–402. Al-Dujaili, E.A.S., Kenyon, C.J., Nicol, M.R., Mason, J.I., 2011. Liquorice and glycyrrhetinic acid increase DHEA and deoxycorticosterone levels in vivo and in vitro by inhibiting adrenal SULT2A1 activity. Mol. Cell. Endocrinol. 336, 102–109. Armanini, D., Karbowiak, I., Funder, J.W., 1983. Affinity of liquorice derivatives for mineralocorticoid and glucocorticoid receptors. Clin. Endocrinol. (Oxf) 19, 609–612. Armanini, D., Mattarello, M.J., Fiore, C., Bonanni, G., Scaroni, C., Sartorato, P., Palermo, M., 2004. Licorice reduces serum testosterone in healthy women. Steroids 69, 763–766. Baker, M.E., 2004. Evolutionary analysis of 11beta-hydroxysteroid dehydrogenase-type 1, -type 2, -type 3 and 17beta-hydroxysteroid dehydrogenase-type 2 in fish. FEBS Lett. 574, 167–170. Bui, P., Kelly, S.P., 2011. Claudins in a Primary Cultured Puffer Fish (Tetraodon nigroviridis) Gill Epithelium. In: Turksen, K. (Ed.), Claudins: Methods and Protocols. Methods in Molecular Biology. Humana Press, Totowa, pp. 179–194. Bui, P., Kelly, S.P., 2014. Claudin-6, -10d and -10e contribute to seawater acclimation in the euryhaline puffer fish Tetraodon nigroviridis. J. Exp. Biol. 217, 1758–1767. Bui, P., Bagherie-Lachidan, M., Kelly, S.P., 2010. Cortisol differentially alters claudin isoforms in cultured puffer fish gill epithelia. Mol. Cell. Endocrinol. 317, 120–126. Chakraborty, S.B., Hancz, C., 2011. Application of phytochemicals as immunostimulant, antipathogenic and antistress agents in finfish culture. Rev. Aquac. 3, 103–119. Chasiotis, H., Kelly, S.P., 2011. Effect of cortisol on permeability and tight junction protein transcript abundance in primary cultured gill epithelia from stenohaline goldfish and euryhaline trout. Gen. Comp. Endocrinol. 172, 494–504. Chasiotis, H., Wood, C.M., Kelly, S.P., 2010. Cortisol reduces paracellular permeability and increases occludin abundance in cultured trout gill epithelia. Mol. Cell. Endocrinol. 323, 232–238. Chasiotis, H., Kolosov, D., Kelly, S.P., 2012. Permeability properties of the teleost gill epithelium under ion-poor conditions. Am. J. Physiol. Regul. Integr. Comp. Physiol. 302, R727–R739. Chilke, A., 2012. Ontogenic variation of ß-glucuronidase in the liver and intestine of Labeo rohita. Recent Res. Sci. Technol. 4, 6–9. Citarasu, T., 2010. Herbal biomedicines: a new opportunity for aquaculture industry. Aquacult. Int. 18, 403–414. Daniel, W.L., Kahle, E.J., 1989. Arylsulfatase and beta-glucuronidase expression in green sunfish, bluegill, and their reciprocal interspecific hybrids. Biochem. Genet. 27, 167–181. Davis, E.A., Morris, D.J., 1991. Medicinal uses of licorice through the millennia: the good and plenty of it. Mol. Cell. Endocrinol. 78, 1–6. Evans, D.H., Piermarini, P.M., Choe, K.P., 2005. The multifunctional fish gill: dominant site of gas exchange, osmoregulation, acid–base regulation, and excretion of nitrogenous waste. Physiol. Rev. 85, 97–177. Ferrari, P., 2010. The role of 11β-hydroxysteroid dehydrogenase type 2 in human hypertension. Biochim. Biophys. Acta 1802, 1178–1187. Ishikawa, S., Saito, T., 1980. The effect of glycyrrhetinic acid on the action of aldosterone in stimulating sodium-transport in frog-skin. Endocrinol. Jpn. 27, 697–701. Kelly, S.P., Chasiotis, H., 2011. Glucocorticoid and mineralocorticoid receptors regulate paracellular permeability in a primary cultured gill epithelium. J. Exp. Biol. 214, 2308–2318. Kelly, S.P., Wood, C.M., 2001a. Effect of cortisol on the physiology of cultured pavement cell epithelia from freshwater trout gills. Am. J. Physiol. Regul. Integr. Comp. Physiol. 281, R811–R820. Kelly, S.P., Wood, C.M., 2001b. The physiological effects of 3,5′,3′-triiodo-L-thyronine alone or combined with cortisol on cultured pavement cell epithelia from freshwater rainbow trout gills. Gen. Comp. Endocrinol. 123, 280–294.

97

Kelly, S.P., Wood, C.M., 2002a. Prolactin effects on cultured pavement cell epithelia and pavement cell plus mitochondria-rich cell epithelia from freshwater rainbow trout gills. Gen. Comp. Endocrinol. 128, 44–56. Kelly, S.P., Wood, C.M., 2002b. Cultured gill epithelia from freshwater tilapia (Oreochromis niloticus): Effect of cortisol and homologous serum supplements from stressed and unstressed fish. J. Membr. Biol. 190, 29–42. Kelly, S.P., Chow, I., Woo, N.Y.S., 1999. Haloplasticity of black seabream (Mylio macrocephalus): hypersaline to freshwater acclimation. J. Exp. Zool. 283, 226–241. Kelly, S.P., Fletcher, M., Part, P., Wood, C.M., 2000. Procedures for the preparation and culture of “reconstructed” rainbow trout branchial epithelia. Methods Cell Sci. 22, 153–163. Kiilerich, P., Kristiansen, K., Madsen, S.S., 2007. Hormone receptors in gills of smolting Atlantic salmon, Salmo salar: expression of growth hormone, prolactin, mineralocorticoid and glucocorticoid receptors and 11β-hydroxysteroid dehydrogenase type 2. Gen. Comp. Endocrinol. 152, 295–303. Kim, D.H., Jang, I.S., Lee, H.K., Jung, E.A., Lee, K.Y., 1996. Metabolism of glycyrrhizin and baicalin by human intestinal bacteria. Arch. Pharm. Res. 19, 292–296. Kolosov, D., Kelly, S.P., 2013. A role for tricellulin in the regulation of gill epithelium permeability. Am. J. Physiol. Regul. Integr. Comp. Physiol. 304, R1139–R1148. Kolosov, D., Bui, P., Chasiotis, H., Kelly, S.P., 2013. Claudins in teleost fishes. Tissue Barriers 1 (3), e25391. Kolosov, D., Chasiotis, H., Kelly, S.P., 2014. Tight junction protein gene expression patterns and temporal changes in transcript abundance in model fish gill epithelia. J. Exp. Biol. 217, 1667–1681. Kumai, Y., Bahubeshi, A., Steele, S., Perry, S.F., 2011. Strategies for maintaining Na+ balance in zebrafish (Danio rerio) during prolonged exposure to acidic water. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 160, 52–62. Kusakabe, M., Nakamura, I., Young, G., 2003. 11beta-Hydroxysteroid dehydrogenase complementary deoxyribonucleic acid in rainbow trout: cloning, sites of expression, and seasonal changes in gonads. Endocrinology 144, 2534–2545. Kusano, E., 2004. How to diagnose and treat a licorice-induced syndrome with findings similar to that of primary hyperaldosteronism. Intern. Med. 43, 5–6. Laurent, P., Perry, S.F., 1990. Effects of cortisol on gill chloride cell morphology and ionic uptake in the freshwater trout, Salmo gairdneri. Cell Tissue Res. 259, 429–442. Lin, D., Sun, W., Wang, Z., Chen, L.G., Chen, X.L., Wang, S.H., Li, W.S., Ge, R.S., Hu, G.X., 2012. The effect of glycyrrhetinic acid on pharmacokinetics of cortisone and its metabolite cortisol in rats. J. Biomed. Biotechnol. 2012, 856324. Mackie, P.M., Gharbi, K., Ballantyne, J.S., McCormick, S.D., Wright, P.A., 2007. Na+/K+/2Cl− cotransporter and CFTR gill expression after seawater transfer in smolts (0+) of different Atlantic salmon (Salmo salar) families. Aquaculture 272, 625–635. Makino, T., Okajima, K., Uebayashi, R., Ohtake, N., Inoue, K., Mizukami, H., 2012. 3Monoglucuronyl-glycyrrhretinic acid is a substrate of organic anion transporters expressed in tubular epithelial cells and plays important roles in licorice-induced pseudoaldosteronism by inhibiting 11β-hydroxysteroid dehydrogenase 2. J. Pharmacol. Exp. Ther. 342, 297–304. McCormick, S.D., 1993. Methods for nonlethal gill biopsy and measurement of Na K-ATPase activity. Can. J. Fish. Aquat. Sci. 50, 656–658. McCormick, S.D., 2001. Endocrine control of osmoregulation in teleost fish. Integr. Comp. Biol. 41, 781–794. McCormick, S.D., Bradshaw, D., 2006. Hormonal control of salt and water balance in vertebrates. Gen. Comp. Endocrinol. 147, 3–8. Nassiri Asl, M., Hosseinzadeh, H., 2008. Review of pharmacological effects of Glycyrrhiza sp. and its bioactive compounds. Phytother. Res. 22, 709–724. Ohtake, N., Kido, A., Kubota, K., Tsuchiya, N., Morita, T., Kase, Y., Takeda, S., 2007. A possible involvement of 3-monoglucuronyl-glycyrrhetinic acid, a metabolite of glycyrrhizin (GL), in GL-induced pseudoaldosteronism. Life Sci. 80, 1545–1552. Perry, S.F., 1997. The chloride cell: structure and function in the gills of freshwater fishes. Annu. Rev. Physiol. 59, 325–347. Perry, S.F., Goss, G.G., Fenwick, J.C., 1992a. Interrelationships between gill chloride cell morphology and calcium uptake in freshwater teleosts. Fish Physiol. Biochem. 10, 327–337. Perry, S.F., Goss, G.G., Laurent, P., 1992b. The interrelationships between gill chloride cell morphology and ionic uptake in four freshwater teleosts. Can. J. Zool. 70, 1775–1786. Prunet, P., Sturm, A., Milla, S., 2006. Multiple corticosteroid receptors in fish: from old ideas to new concepts. Gen. Comp. Endocrinol. 147, 17–23. Ruszymah, B.H.I., Nabishah, B.M., Aminuddin, S., Khalid, B.A.K., 1995. Effects of glycyrrhizic acid on right atrial pressure and pulmonary vasculature in rats. Clin. Exp. Hypertens. 17, 575–591. Tanahashi, T., Mune, T., Morita, H., Tanahashi, H., Isomura, Y., Suwa, T., Daido, H., Gomez-Sanchez, C.E., Yasuda, K., 2002. Glycyrrhizic acid suppresses type 2 11 beta-hydroxysteroid dehydrogenase expression in vivo. J. Steroid Biochem. Mol. Biol. 80, 441–447. Tipsmark, C.K., Jorgensen, C., Brande-Lavridsen, N., Engelund, M., Olesen, J.H., Madsen, S.S., 2009. Effects of cortisol, growth hormone and prolactin on gill claudin expression in Atlantic salmon. Gen. Comp. Endocrinol. 163, 270–277. Ulmann, A., Menard, J., Corvol, P., 1975. Binding of glycyrrhetinic acid to kidney mineralocorticoid and glucocorticoid receptors. Endocrinology 97, 46–51. Wood, C.M., Part, P., 1997. Cultured branchial epithelia from freshwater fish gills. J. Exp. Biol. 200, 1047–1059. Wood, C.M., Gilmour, K.M., Part, P., 1998. Passive and active transport properties of a gill model, the cultured branchial epithelium of the freshwater rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. A: Mol. Integr. Physiol. 119, 87–96. Zall, D.M., Fisher, D., Garner, M.Q., 1956. Photometric determination of chlorides in water. Anal. Chem. 28, 1665–1668.