Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
Contents lists available at ScienceDirect
Journal of Steroid Biochemistry & Molecular Biology journal homepage: www.elsevier.com/locate/jsbmb
The origin and metabolism of vitamin D in rainbow trout S.L. Pierens, D.R. Fraser * Faculty of Veterinary Science, The University of Sydney, NSW 2006, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 2 September 2014 Accepted 5 October 2014 Available online 11 October 2014
An explanation for the origin and the high concentration of vitamin D (cholecalciferol) in some species of fish is still not apparent. Because fish may live in deep water and may, thus, not be exposed to solar ultraviolet (UV) light, it is commonly assumed that vitamin D found in their livers and adipose tissue has been derived from a food chain, originating in zooplankton exposed to UV light at the water surface. To investigate the metabolism and possible origin of vitamin D in fish, rainbow trout were reared from eggs, in the absence of light, and were fed a vitamin D-free diet. When small quantities of radioactivelylabelled vitamin D were injected or fed to these trout, much of the radioactivity was found as excreted metabolites in bile. Hence, even when they are vitamin D deficient, trout vigorously catabolise and excrete exogenous vitamin D. The main vitamin D metabolite found in plasma of non-deficient trout was 1,25-dihydroxycholecalciferol [1,25(OH)2D3]. This was produced in the liver by an enzyme process that was strongly stimulated in vitamin D deficiency. When vitamin D was fed for several weeks to vitamin D-deficient trout, plasma 1,25(OH)2D3 levels rose to 180 pg/ml and the fish became hypercacemic. When vitamin D-deficient fish were inadvertently exposed to 60 W incandescent light for 24 h, they became moribund and died. It was subsequently found that vitamin D-deficient trout can produce vitamin D in skin when exposed to blue light at wavelengths between 380 and 480 nm. It is concluded that trout, like terrestrial vertebrates, produce 1,25(OH)2D3 as the functional form of vitamin D and that this has an effect on calcium homeostasis. Furthermore, vitamin D is formed in the skin of these fish by the photochemical action of visible light on 7-dehydrocholesterol. Elucidation of the physicochemical mechanism of this process requires further research. ã 2014 Elsevier Ltd. All rights reserved.
Keywords: Vitamin D deficiency Rainbow trout Metabolism 7-Dehydrocholesterol Skin Visible light
1. Introduction One of the early findings in the history of vitamin D research was that some species of fish contain concentrations of vitamin D in liver or in other soft tissues that are much higher than in the tissues of any terrestrial vertebrates [1]. The chemical form of vitamin D in fish tissues is cholecalciferol. For any fish species the concentration may vary over a wide range. For example, the cholecalciferol content of the oil extracted from cod liver is reported to vary from 0.7 to 12.5 mg/g [2]. Much of the cholecalciferol in cod liver is esterified with long chain fatty acids [3], and hence, may be considered as an inactive or possibly a storage form. In contrast, vitamin D fatty acid esters are a very minor component of the vitamin D molecules in mammalian tissues [4]. Therefore, the accumulation of esterified cholecalciferol in lipid may explain why some species of fish, unlike mammals, have high concentrations of this molecule in liver and other tissues.
* Corresponding author at: Faculty of Veterinary Science, RMC Gunn Building (B19), The University of Sydney, NSW 2006, Australia. Tel.: +61 2 9351 2139. E-mail address: [email protected] (D.R. Fraser). http://dx.doi.org/10.1016/j.jsbmb.2014.10.005 0960-0760/ ã 2014 Elsevier Ltd. All rights reserved.
The origin of cholecalciferol in fish is uncertain. For most terrestrial vertebrates, vitamin D is obtained by photochemical conversion of 7-dehydrocholesterol in skin, using the energy of solar ultraviolet light in the wavelength range of 290–320 nm. In comparison to terrestrial animals, many species of fish live in water at depths where little solar ultraviolet light would penetrate. Therefore, cutaneous synthesis of cholecalciferol in fish by the action of ultraviolet light on 7-dehydrocholesterol would seem to be unlikely. To explain the high cholecalciferol content of fish it has been assumed that it is either produced endogenously by some undiscovered enzymatic synthesis or else it is accumulated from the diet in a food chain that originated in organisms containing 7-dehydrocholesterol, which are exposed to ultraviolet light at the water surface. Attempts to demonstrate enzymatic synthesis of cholecalciferol in fish liver from radioactively labelled 7-dehydrocholesterol [5] or from labelled acetate [6] have given no indication of metabolic synthesis. On the other hand, vitamin D has been found in zooplankton at the ocean surface [7] which could be the source of the vitamin D that is eventually found in the liver lipid of deep sea fish such as cod. This report describes a wide range of studies on laboratoryreared rainbow trout (Oncorhynchus mykiss) which were
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
undertaken over a period of about 30 years. To investigate both the metabolism of exogenous vitamin D and the possibility of synthesis of endogenous vitamin D, rainbow trout, as a convenient experimental fish species, were reared from eggs in darkness on a vitamin D-deficient diet. 2. Materials and methods Rainbow trout “eyed ova” from a regional trout hatchery were placed in an aerated, re-circulating laboratory aquarium, maintained at 10–15 C with a biological filter to detoxify faecal ammonia. Lipid was extracted from an additional 2050 rainbow trout eggs to determine their vitamin D content by a chick toe–ash bioassay [8]. The hatched fish were reared on a semi-purified, vitamin D-free diet, prepared in the form of soft gelatin pellets (Table 1). The trout were kept in total darkness except for 10 min each day, during feeding, when the aquarium room was lit with a dim 60 W incandescent ceiling light with a minimum wavelength emission of about 400 nm. Blood samples were collected by cardiac puncture from trout, anaesthetised in a solution of MS222 (tricaine methanesulfonate, 100 mg/l). Analysis for cholecalciferol and 7-dehydrocholesterol in lipid extracts of trout tissues was performed by high performance liquid chromatography (HPLC) according to standard methods [10]. Determination of 25-hydroxy-cholecalciferol [25(OH)D] or 1,25dihydroxycholecalciferol in lipid extracts was done either by competitive protein binding assays or by HPLC [10]. HPLC analysis of cholecalciferol had a minimum detection limit of 1 ng/g wet tissue. Trout were irradiated in a solar light simulator (Full Spectrum F35 simulator model 1 with scanning system, CSIRO, Australia). The wavelength and intensity of light was controlled by the use of different light sources and filters. Four different irradiation wavelength ranges were produced: (1) full spectrum sunlight, (2) ultraviolet B (290–320 nm), (3) ultraviolet A (350–400 nm), (4) visible blue light (380–480 nm). Radioactively-labelled cholecalciferol: [1a-3H] cholecalciferol (2 Ci/mmol) and [1,2-3H] cholecalciferol (276.9 mCi/mmol) were kindly prepared by P.A. Bell and B. Pelc by the method of Callow et al. [11]. Determination of the distribution of 3H on carbon-1 of [1a-3H] cholecalciferol has shown that 85% is in the 1a-position and 15% is in the 1b-position [12]. Therefore, in the enzymemediated 1a-hydroxylation of this tritium-labelled molecule, about 85% of the 3H is lost as 3H2O. [4-14C] Cholecalciferol (32.3 mCi/mmol) and 25-hydroxy[26,27-3H] cholecalciferol (11 Ci/mmol) were obtained from The Radiochemical Centre, Amersham, U.K. [4-14C, 1a-3H] 25-hydroxycholecalciferol was prepared by standard Sephadex LH20 chromatography, from lipid extracts of the blood of pigs which had been injected intravenously with either [1a-3H] cholecalciferol or [4-14C] cholecalciferol.
Table 1 Vitamin D-free fish diet. Water
1l
Gelatin Ethanol-extracted casein Powdered dextrin Powdered cellulose Inorganic saltsa Corn oil Linolenic acid Vitaminsa
60 g 190 g 140 g 35 g 35 g 40 g 5g
a Inorganic salts, water-soluble and fat-soluble vitamins (without cholecalciferol) were added according to the defined requirements of micronutrients for rainbow trout [9].
59
2.1. Trout liver homogenates were incubated for 3–4 h at 17–18 with radioactive 25-Hydroxycholecalciferol in an NADPH generating medium [13]. The incubation mixture was then extracted with chloroform/ methanol and residual radioactive substrate and any metabolic products were separated and quantified by Sephadex LH20 lipid chromatography. 3. Results Each rainbow trout egg, weighing 80–85 mg, contained 5–6 ng of vitamin D activity as determined by chick toe–ash bioassay. No 25(OH)D was detected by competitive protein binding analysis in the trout egg lipid extract. It was, therefore, assumed that each fish at hatching, contained up to 6 ng cholecalciferol. HPLC analysis of lipid extracts of the fish diet was unable to detect any cholecalciferol or 25(OH)D. The absence of detectable 25(OH)D or 1,25(OH)2D in blood plasma as the fish grew, suggested also that these fish were vitamin D deficient and that they were not making vitamin D by a metabolic process. There was a significant decrease (p < 0.05) in total Ca in blood plasma (1.6 0.48 mmol/l, mean SD, n = 15) of the vitamin D-deficient trout compared to that from non-deficient commercial trout of 2.70 0.06 mmol/l. There was also a significant decrease (p < 0.05) in plasma protein concentration from 35.00 0.27 mg/ml in commercial trout to 22.80 0.46 mg/ml in the deficient trout. No differences were found in plasma phosphorus concentration or in bone calcium or bone phosphorus content between the vitamin D-deficient trout and non-deficient commercial trout. There were no behavioural or other visible signs of abnormality demonstrated by these trout. Vitamin D-deficient trout, weighing about 200 g, were fasted for 24 h and then injected intraperitoneally with 30 ng [1,2-3H] cholecalciferol in 100 ml propylene glycol. After a further 24 h without food, the trout were killed and various tissues were collected to measure the distribution of radioactivity. Surprisingly, about 90% of the tritium dose was found as more polar metabolites of vitamin D in the bile which had accumulated in the gall bladder during the time of starvation. Hence, rather than using this small dose of vitamin D to perform some functional role, it appeared that it had been metabolically changed and excreted in bile. In comparable experiments, vitamin D-deficient trout were injected intraperitoneally with 25 ng (0.65 mCi) [26,27-3H] 1,25-dihydroxycholecalciferol dissolved in 100 ml propylene glycol. The fish were killed 3 h later and tissues were collected for determination of radioactivity. Again, the concentration of excreted radioactivity in bile at 6.2 ng/ml was greater than in blood plasma (0.17 ng/ml), kidneys (0.3 ng/g) and liver (0.9 ng/g), and only about 20% of the radioactivity in bile appeared to be unchanged 1,25(OH)2D3. Vitamin D, as [1,2-3H] cholecalciferol, was incorporated into the fish diet at a concentration of 50 mCi (70 mg) per 100 g dry ingredients. This was fed to a group of vitamin D-deficient trout for 4 months. The animals were then fed the vitamin D-deficient diet for a further 4 weeks, after which they were killed and the nature of the radioactivity in liver was determined by Sephadex LH20 chromatography of lipid extracts. Only one peak of radioactivity was found in liver and this was less polar than that of cholecalciferol (Fig. 1). On saponification and re-chromatography of the non-saponifiable lipid, the radioactivity was found to cochromatograph with cholecalciferol. It was therefore, concluded that the radioactivity extracted from the trout liver after long term feeding of [1,2-3H] cholecalciferol, was the fatty acid ester of the labelled cholecalciferol. Vitamin D-deficient trout at a body weight of about 100 g were injected intraperitoneally with 1 mCi (19.4 ng) [26,27-3H]
60
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
Fig. 1. Sephadex LH20 chromatograph of radioactivity in lipid extract of the liver of vitamin D-deficient rainbow trout after feeding for 4 months [1,2-3H] cholecalciferol, incorporated into the fish diet at a concentration of 50 mCi (70 mg) per 100 g dry ingredients and then feeding the vitamin D-free diet for a further 4 weeks before analysing the radioactive material remaining in liver.
25-hydroxycholecalciferol in 100 ml propylene glycol and blood was collected 19 h later. Lipid extracts of blood plasma when chromatographed on Sephadex LH20 revealed two peaks of radioactivity (Fig. 2). About 55% of the tritium in plasma was
Fig. 2. Sephadex LH20 chromatograph of radioactivity in lipid extract of blood plasma from vitamin D-deficient rainbow trout, collected 19 h after being injected intraperitoneally with 1 mCi (19.4 ng) [26,27-3H] 25-hydroxycholecalciferol. 55% of the radioactivity in plasma co-chromatographed with 25-hydroxycholeclaciferol and 44% co-chromatographed with 1,25-dihydroxycholecalciferol.
present as the injected 3H-25(OH)D3 while about 44% was in a 3H-labelled metabolite, co-chromatographing with 1,25(OH)2D3. Analysis of the blood plasma of vitamin D-replete commercial rainbow trout by competitive protein binding assays, showed that the 25(OH)D concentration was below the level of detection of 1.2 ng/ml. However, the concentration of 1,25(OH)2D was 128 10 pg/ml. Incubation of [4-14C, 1a-3H] 25(OH)D3 (0.7 mg; 3H/14C = 5) with the homogenate of 1.9 g liver from vitamin D-deficient rainbow trout (body weight = 240 g) for 4 h at 17 converted 61% of the substrate to a more polar product with a 3H/14C ratio of 2.56, thus confirming by the loss of the 1a-tritium atom, its identity as 1,25 (OH)2D3 (Fig. 3). Vitamin D-deficient rainbow trout at 150–200 g body weight were fed the semi-purified diet which had been fortified with 612 mg cholecalciferol per kg dry ingredients. Four trout were then sacrificed at weekly intervals and the concentrations of calcium and 1,25(OH)2D3 in blood plasma were measured. After 1 week on the vitamin D-fortified diet the fish had become hypercalcemic but blood calcium returned to the normal range after 2 weeks (Fig. 4). The concentration of 1,25(OH)2D3 in plasma rose gradually to a maximum of about 180 pg/ml after about 4 weeks and thereafter declined to a range of 60–80 pg/ml (Fig. 5). When aquarium tanks containing rainbow trout, which had been reared in darkness on the vitamin D-free diet for 2 years, were inadvertently exposed for 24 h to 60 W incandescent light, the trout rapidly became moribund and died. Because it had been shown that the hepatic 1-hydroxylase of these vitamin D-deficient fish was very active and that feeding dietary cholecalciferol led to hypercalcemia, it was postulated that exposure to visible light had in some way induced the formation of vitamin D in skin which resulted in a lethal overproduction of 1,25(OH)2D3 and consequent fatal hypercalcemia. HPLC analysis of lipid extracts from the skin of a number of species obtained from a fish market, demonstrated the presence not only of 7-dehydrocholesterol but also measurable quantities of cholecalciferol (Table 2). Rainbow trout and Atlantic salmon had higher levels of 7-dehydrocholesterol in skin than
Fig. 3. Sephadex LH20 chromatography of lipid extract from vitamin D-deficient rainbow trout liver homogenate incubated at 17 for 4 h with [4-14C, 1a-3H] 25(OH) D3 (0.7 mg; 3H/14C = 5). 30% of the 14C radioactivity co-chromatographed with 25 (OH)D3 with a 3H/14C ratio of 4.96 and 61% of the 14C radioactivity cochromatographed with 1,25-dihydroxycholecalciferol with a 3H/14C ratio of 2.56, confirming its identity as the 1-hydroxylated metabolite of 25(OH)D3.
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
61
Fig. 4. Calcium concentration in the blood plasma of vitamin D-deficient rainbow trout, with time, while feeding a diet containing 612 mg cholecalciferol/kg DM. Four fish were killed at each time point and the plotted values are means SD.
other sea water species, but all the fish examined had 20 ng or more cholecalciferol per gram wet weight of skin. To test the hypothesis that visible light was able to induce vitamin D formation in skin, rainbow trout were irradiated in a variable wavelength solar simulator, with light passing through 5 cm water at 12 . Trout were exposed either to full spectrum
simulated solar light (290–1200 nm) or to blue light (380–480 nm) (Fig. 6). Exposure for two 30 min periods resulted in the appearance of vitamin D in skin with both light sources (Fig. 7), with full spectrum simulated sunlight producing more vitamin D than blue light. Furthermore, increasing amounts of 1,25(OH)2D3 were detected in blood of these fish after irradiation (Fig. 8). The identity of vitamin D obtained from the skin of trout irradiated with 380–480 nm blue light was confirmed by mass spectrometry (Fig. 9). In other experiments in which dead rats, with the hair coat closely clipped to the skin, and dead trout were both exposed to 380–480 nm blue light, vitamin D was again found in the trout skin but there was no change in the small quantities of vitamin D in rat skin. It was, therefore, concluded that some characteristic of trout skin structure had enabled the energy of blue light to convert 7-dehydrocholesterol into cholecalciferol. 4. Discussion
Fig. 5. Concentration of 1,25(OH)2D3 in the blood plasma of vitamin D-deficient rainbow trout, with time, while feeding a diet containing 612 mg cholecalciferol/kg DM. Each point represents the 1,25(OH)2D3 concentration in the pooled plasma of four trout.
It has been shown by several investigators that the liver of rainbow trout and of other fish species, is capable of synthesising 1,25(OH)2D3 from 25(OH)D3 [14–17]. It also appears that the kidneys of several fish species have the 1-hydroxylase enzyme [18,19]. In comparison to terrestrial vertebrates, the concentration of 25(OH)D in blood plasma is very low while that of 1,25(OH)2D is higher than found in mammalian blood [20]. In support of the concept that 1,25(OH)2D3 is the main circulating form of vitamin D, our unpublished measurements on the plasma vitamin D-binding protein (DBP) from goldfish (Carassius auratus) indicate that in contrast to mammalian, avian and amphibian DBP, the binding affinity for 1,25(OH)2D3 is higher than that for 25(OH)D3. From these many findings, it has been concluded that the liver of fish takes up cholecalciferol and converts some to 25(OH)D3 which in turn is further hydroxylated to 1,25(OH)2D3 and then secreted into blood.
62
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64 Table 2 Concentration of 7-dehydrocholesterol and cholecalciferol in the skin of various fish species. Fish species
7-Dehydrocholesterol mg/g skin wet weight
Cholecalciferol ng/g skin wet weight
Rainbow trout (Oncorhynchus mykiss) Atlantic salmon (Salmo salar) Yellowfin tuna (Thunnus albacares) Trevally (Pseudocaranx georgianus) Blackfish (Girella elevata) Bream (Acanthopagrus australis) Whiting (Sillago ciliata)
10.7 1.4 (n = 16) 26.0 3.2 (n= 4) 0.07 0.01 (n = 4) 0.95 0.35 (n = 5) 1.1 0.3 (n = 3) 0.3 0.04 (n = 5) 1.0 0.2 (n = 12)
300 20.7 (n = 15) 22.5 5.2 (n = 4) 72.1 15.7 (n = 3) 18.6 3.2 (n = 8) 23.9 9.1 (n = 4) 20.9 9.6 (n = 3) 24.9 5.4 (n = 9)
Fresh fish were purchased from the Sydney Fish Markets to obtain skin for analysis. Mean values SD obtained by HPLC analysis of lipid extracts of fish skin.
An alternative metabolic fate of hepatic cholecalciferol in fish, is to become esterified with long chain fatty acids. This vitamin D ester could possibly be a storage form of vitamin D. However, when radioactively-labelled cholecalciferol esters were formed in liver of vitamin D-deficient rainbow trout, which had been fed food containing [1,2-3H]-cholecalciferol, the esters remained even when the fish were again on a vitamin D-free diet (Fig 1). Vitamin D-deficient trout also catabolised most of injected doses of [1,2-3H] cholecalciferol and secreted the catabolites in bile. From these results it could be concluded that the liver of trout is especially capable of detoxifying exogenous vitamin D, both by metabolic inactivation and by sequestration as biologically inert fatty acid esters. Such mechanisms would be an advantage for carnivorous fish, the diet of which may be other fish, rich in vitamin D esters, which they themselves had accumulated throughout their life. The esterification of cholecalciferol to detoxify that received in food, may thus be an explanation why some species of fish, such as cod (Gadus morhua), contain large quantities of cholecalciferol esters in liver [3]. One potential toxic action of vitamin D could be the induction of hypercalcemia by some endocrine action of 1,25(OH)2D3, secreted
Fig. 6. Light intensity over the wavelength range of 380–480 nm used to irradiate vitamin D-deficient rainbow trout in a variable solar light simulator.
Fig. 7. Concentration of vitamin D in the skin (ng/g wet weight) of vitamin Ddeficient rainbow trout after they were irradiated in a solar light simulator. Irradiation was for two periods of 30 min, 2 h apart. During the 2-h interval between irradiations, the fish was kept in darkness. Each point represents the mean SD of four fish. A-1 and A-2 indicate the skin vitamin D content of trout irradiated with full solar spectrum light. B-1 and B-2 indicate the skin vitamin D content of trout irradiated with light over the wavelength range of 380–480 nm as illustrated in Fig. 6.
Fig. 8. Concentration of 1,25(OH)2D in blood plasma (pg/ml) of vitamin D-deficient rainbow trout after they were irradiated in a solar light simulator. Irradiation was for two periods of 30 min, 2 h apart. During the 2-h interval between irradiations, the fish was kept in darkness. Each point represents the value from the pooled plasma of four fish. A-1 and A-2 indicate the plasma 1,25(OH)2D concentration of trout irradiated with full solar spectrum light. B-1 and B-2 indicate the plasma 1,25 (OH)2D concentration of trout irradiated with light over the wavelength range of 380–480 nm as illustrated in Fig. 6.
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
63
Fig. 9. Mass spectra comparison of standard cholecalciferol with that of putative cholecalciferol extracted and chromatographically purified from the skin of vitamin Ddeficient rainbow trout after irradiation with light of the wavelength range of 380–480 nm.
by the liver [21–23]. In our studies with vitamin D-deficient trout, the restoration of vitamin D status by supplying vitamin D in food caused hypercalcemia and this was associated with high concentrations of 1,25(OH)2D3 in blood (Figs. 4 and 5). Vitamin D deficiency in trout apparently stimulates the activity of hepatic 1a-hydroxylase as demonstrated by incubation of liver homogenates with radioactively labelled 25(OH)D3 (Fig. 3). It is possible also that the output of 1,25(OH)2D3 by liver of normal fish may not be tightly regulated, in contrast to the well-regulated renal production of 1,25(OH)2D3 in terrestrial vertebrates [13]. The quantity secreted into blood may be determined by the production of 25(OH)D3. Therefore the liver of trout appears to be both a protective organ against toxicity from dietary vitamin D and also an endocrine organ, secreting the hormone 1,25(OH)2D3, which in some way affects blood calcium concentration. The big surprise from all of our studies, was the observation of the lethal effect of 24 h exposure of vitamin D-deficient trout to visible light from a 60 W incandescent bulb. These fish were in water with a high concentration of dissolved calcium carbonate, so if they had been producing toxic levels of 1,25(OH)2D3 in blood, their death may have been caused by increased uptake of calcium from water with consequent fatal hypercalcemia. However, in terrestrial vertebrates, visible light is not capable of converting 7-dehydrocholesterol in skin to cholecalciferol. We clearly demonstrated that shaved rat skin did not form vitamin D when exposed to visible light with the wavelength range of 380–480 nm. Nevertheless, cholecalciferol was indeed produced in the skin of both live and dead vitamin D-deficient rainbow trout exposed to visible light in the 380–480 nm wavelength range. This observation seems to defy the known energy requirement, (provided by UV light at 290–320 nm), to rupture the 9–10 bond of the B ring of 7-dehydrocholesterol to form pre-cholecalciferol. One possibility is that the structure of fish skin, with its metallic sheen caused by thin bony scales associated with intracellular anhydrous guanine
crystals [24] focuses the lower energy of multiple photons of blue light onto molecules of 7-dehydrocholesterol to induce breakage of the 9–10 bonds. Identification of the physicochemical mechanism for this process will require further research. In clear water, solar UV light at 290–320 nm can penetrate to about 20 m, by which depth about 90% of the UV energy has been absorbed [25]. Although rainbow trout, in their natural habitat of freshwater rivers, in water less than 20 m deep, could well be exposed to solar UV light, many other species of fish inhabit depths below the penetration range of UV light. If such deep water fish are able to produce cholecalciferol from 7-dehydrocholesterol in skin by exposure to visible blue light, then this may be a mechanism by which many species of fish, as well as rainbow trout, acquire vitamin D from exposure to sunlight. In contrast to the limited penetration range of solar UV light, 10% of the blue light range of 380–480 nm is still present at depths of 200 m [26]. 5. Conclusions Studies with vitamin D-deficient rainbow trout have demonstrated that radioactively labelled cholecalciferol is both metabolically inactivated and converted to biologically inert vitamin D esters in the liver. The liver also secretes 1,25(OH)2D3 into blood and the production of this hormone appears to be stimulated in vitamin D deficiency. Trout skin, unlike that of rats, is able to produce cholecalciferol when exposed to visible light in the wavelength range of 380–480 nm. These findings suggest that instead of obtaining vitamin D from a food chain, starting with zooplankton at the water surface, some fish may depend upon visible light from the sun, to induce cholecalciferol production in skin to meet their vitamin D requirements. Rather than using food as a source of vitamin D, these fish may need the vigorous hepatic inactivation mechanisms to protect against toxicity from high levels of vitamin D in the food chain.
64
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
Acknowledgement The authors are most grateful for the dedicated assistance of Angelika Trube in the care of the rainbow trout and the whole aquarium system, and for her expert help with laboratory analyses. References [1] A. Takeuchi, T. Okano, M. Ayama, H. Yoshikawa, S. Teraoka, Y. Murakami, T. Kobayashi, High-performance liquid chromatographic determination of vitamin D3 in fish liver oils and eel body oils, J. Nutr. Sci. Vitaminol. (Tokyo) 30 (1984) 421–430. [2] T. Yamakawa, T. Kinumaki, Determination of vitamin D in natural products, Vitamins 46 (1972) 167–177. [3] W. Von Müller-Mulot, G. Rohrer, K. Schwarzbauer, Native vitamin D3 esters in cod-liver oil: chemical determination of free, esterified and total vitamin D3,, Fette Seifen Abstrichmittel 81 (1979) 38–40. [4] D.R. Fraser, E. Kodicek, Investigations on vitamin D esters synthesized in rats. Turnover and sites of synthesis, Biochem. J. 106 (1968) 491–496. [5] G.A. Blondin, B.D. Kulkarni, W.R. Nes, A study of the origin of vitamin D from 7-dehydrocholesterol in fish, Comp. Biochem. Physiol. 20 (1967) 379–390. [6] D.S. Rao, N. Raghuramulu, Lack of vitamin D3 synthesis in Tilapia mossambica from cholesterol and acetate, Comp. Biochem. Physiol. A 114 (1996) 21–25. [7] A.M. Copping, Origin of vitamin D in cod-liver oil: vitamin D content of zooplankton, Biochem. J. 28 (1934) 1516–1520. [8] J.A. Campbell, B.B. Migikovsky, A.R.G. Emalie, Studies on the chick assay for vitamin D I. Precision of tibia and toe ash as criteria of response, Poult. Sci. 24 (1945) 3–7. [9] Fish Nutrition, in: J.E. Halver (Ed.), Academic Press, New York, 1989, pp. 1–772. [10] R.L. Horst, T.A. Reinhardt, B.W. Hollis, Improved methodology for the analysis of plasma vitamin D metabolites, Kidney Int. 38 (1990) 28–35. [11] E. Callow, Metabolism of tritiated vitamin D, Proc. R. Soc. London B 164 (1966) 1–20. [12] P.A. Bell, E. Kodicek, The stereospecificity of tritium distribution in [1-3H]- and [1,2-3H]-cholesterol and cholecalciferol, Biochem. J. 116 (1970) 755–757. [13] D.R. Fraser, E. Kodicek, Regulation of 25-hydroxycholecalciferol 1-hydroxylase activity in kidney by parathyroid hormone, Nat. New Biol. 241 (1973) 163–166.
[14] M.E. Hayes, D.F. Guilland-Cumming, R.G.G. Russell, I.W. Henderson, Metabolism of 25-hydroxycholecalciferol in a teleost fish: the rainbow trout (Salmo gairdneri), Gen. Comp. Endocrinol. 64 (1986) 143–150. [15] K. Sundell, J.E. Bishop, B.T. Björnsson, A.W. Norman, I,25-Dihydroxyvitamin D3 in the Atlantic cod: plasma levels, a plasma binding component, and organ distribution of a high affinity receptor, Endocrinology 131 (1992) 2279–2286. [16] A. Takeuchi, T. Okano, T. Kobayashi, The existence of 25-hydroxyvitamin D31a-hydroxylase in the liver of carp and bastard halibut, Life Sci. 48 (1991) 275–282. [17] A. Takeuchi, Comparative studies on vitamin D in vertebrates especially in fish, Vitamins (Jpn.) 68 (1994) 55–64. [18] H. Henry, A.W. Norman, Presence of renal 25-hydroxy-vitamin-D-1-hydroxylase in species of all vertebrate classes, Comp. Biochem. Physiol. B 50 (1975) 431–434. [19] I.E. Graff, O. Lie, L. Aksnes, In vitro hydroxylation of vitamin D3 and 25-hydroxy vitamin D3 in tissues of Atlantic salmon Salmo salar, Atlantic mackerel Scomber scombrus, Atlantic halibut Hippoglossus hippoglossus and Atlantic cod Gadus morhua, Aquac. Nutr. 5 (1999) 23–32. [20] L.V. Avioli, Y. Sonn, D. Jo, T.H. Nahn, M.R. Haussler, J.S. Chandler, 1,25Dihydroxyvitamin D in male, non spawning female, and spawning female trout, Proc. Soc. Exp. Biol. Med. 166 (1981) 291–293. [21] S.K. Srivastav, R. Jaiswal, A.K. Srivastav, Response of serum calcium to administration of 1,25-dihydroxyvitamin D3 in the freshwater carp Cyprinus carpio maintained either in artificial freshwater, calcium-rich freshwater or calcium-deficient freshwater, Acta Physiol. Hung. 81 (1993) 269–275. [22] K. Sundell, A.W. Norman, B.T. Björnsson, 1,25(OH)2 vitamin D3 increases ionized plasma calcium concentrations in the immature Atlantic cod Gadus morhua, Gen. Comp. Endocrinol. 91 (1993) 344–351. [23] A.J. Srivastav, G. Flik, S.E. Wendelaar Bonga, Plasma calcium and stanniocalcin levels of male tilapia Oreochromis mossambicus, fed calcium-deficient food and treated with 1,25 dihydroxyvitamin D3, Gen. Comp. Endocrinol. 110 (1998) 290–294. [24] D. Gur, Y. Politi, B. Sivan, P. Fratzl, S. Weiner, L. Addadi, Guanine-based photonic crystals in fish scales form from an amorphous precursor, Angew. Chem. Int. Ed. 52 (2013) 388–391. [25] C. Booth, J. Morrow, The penetration of UV into natural waters, Photochem. Photobiol. 65 (1997) 254–257. [26] E.M. Kampa, Underwater daylight and moonlight measurements in the eastern North Atlantic, J. Mar. Biol. Assoc. U. K. 50 (1970) 397–420.
Contents lists available at ScienceDirect
Journal of Steroid Biochemistry & Molecular Biology journal homepage: www.elsevier.com/locate/jsbmb
The origin and metabolism of vitamin D in rainbow trout S.L. Pierens, D.R. Fraser * Faculty of Veterinary Science, The University of Sydney, NSW 2006, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 2 September 2014 Accepted 5 October 2014 Available online 11 October 2014
An explanation for the origin and the high concentration of vitamin D (cholecalciferol) in some species of fish is still not apparent. Because fish may live in deep water and may, thus, not be exposed to solar ultraviolet (UV) light, it is commonly assumed that vitamin D found in their livers and adipose tissue has been derived from a food chain, originating in zooplankton exposed to UV light at the water surface. To investigate the metabolism and possible origin of vitamin D in fish, rainbow trout were reared from eggs, in the absence of light, and were fed a vitamin D-free diet. When small quantities of radioactivelylabelled vitamin D were injected or fed to these trout, much of the radioactivity was found as excreted metabolites in bile. Hence, even when they are vitamin D deficient, trout vigorously catabolise and excrete exogenous vitamin D. The main vitamin D metabolite found in plasma of non-deficient trout was 1,25-dihydroxycholecalciferol [1,25(OH)2D3]. This was produced in the liver by an enzyme process that was strongly stimulated in vitamin D deficiency. When vitamin D was fed for several weeks to vitamin D-deficient trout, plasma 1,25(OH)2D3 levels rose to 180 pg/ml and the fish became hypercacemic. When vitamin D-deficient fish were inadvertently exposed to 60 W incandescent light for 24 h, they became moribund and died. It was subsequently found that vitamin D-deficient trout can produce vitamin D in skin when exposed to blue light at wavelengths between 380 and 480 nm. It is concluded that trout, like terrestrial vertebrates, produce 1,25(OH)2D3 as the functional form of vitamin D and that this has an effect on calcium homeostasis. Furthermore, vitamin D is formed in the skin of these fish by the photochemical action of visible light on 7-dehydrocholesterol. Elucidation of the physicochemical mechanism of this process requires further research. ã 2014 Elsevier Ltd. All rights reserved.
Keywords: Vitamin D deficiency Rainbow trout Metabolism 7-Dehydrocholesterol Skin Visible light
1. Introduction One of the early findings in the history of vitamin D research was that some species of fish contain concentrations of vitamin D in liver or in other soft tissues that are much higher than in the tissues of any terrestrial vertebrates [1]. The chemical form of vitamin D in fish tissues is cholecalciferol. For any fish species the concentration may vary over a wide range. For example, the cholecalciferol content of the oil extracted from cod liver is reported to vary from 0.7 to 12.5 mg/g [2]. Much of the cholecalciferol in cod liver is esterified with long chain fatty acids [3], and hence, may be considered as an inactive or possibly a storage form. In contrast, vitamin D fatty acid esters are a very minor component of the vitamin D molecules in mammalian tissues [4]. Therefore, the accumulation of esterified cholecalciferol in lipid may explain why some species of fish, unlike mammals, have high concentrations of this molecule in liver and other tissues.
* Corresponding author at: Faculty of Veterinary Science, RMC Gunn Building (B19), The University of Sydney, NSW 2006, Australia. Tel.: +61 2 9351 2139. E-mail address: [email protected] (D.R. Fraser). http://dx.doi.org/10.1016/j.jsbmb.2014.10.005 0960-0760/ ã 2014 Elsevier Ltd. All rights reserved.
The origin of cholecalciferol in fish is uncertain. For most terrestrial vertebrates, vitamin D is obtained by photochemical conversion of 7-dehydrocholesterol in skin, using the energy of solar ultraviolet light in the wavelength range of 290–320 nm. In comparison to terrestrial animals, many species of fish live in water at depths where little solar ultraviolet light would penetrate. Therefore, cutaneous synthesis of cholecalciferol in fish by the action of ultraviolet light on 7-dehydrocholesterol would seem to be unlikely. To explain the high cholecalciferol content of fish it has been assumed that it is either produced endogenously by some undiscovered enzymatic synthesis or else it is accumulated from the diet in a food chain that originated in organisms containing 7-dehydrocholesterol, which are exposed to ultraviolet light at the water surface. Attempts to demonstrate enzymatic synthesis of cholecalciferol in fish liver from radioactively labelled 7-dehydrocholesterol [5] or from labelled acetate [6] have given no indication of metabolic synthesis. On the other hand, vitamin D has been found in zooplankton at the ocean surface [7] which could be the source of the vitamin D that is eventually found in the liver lipid of deep sea fish such as cod. This report describes a wide range of studies on laboratoryreared rainbow trout (Oncorhynchus mykiss) which were
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
undertaken over a period of about 30 years. To investigate both the metabolism of exogenous vitamin D and the possibility of synthesis of endogenous vitamin D, rainbow trout, as a convenient experimental fish species, were reared from eggs in darkness on a vitamin D-deficient diet. 2. Materials and methods Rainbow trout “eyed ova” from a regional trout hatchery were placed in an aerated, re-circulating laboratory aquarium, maintained at 10–15 C with a biological filter to detoxify faecal ammonia. Lipid was extracted from an additional 2050 rainbow trout eggs to determine their vitamin D content by a chick toe–ash bioassay [8]. The hatched fish were reared on a semi-purified, vitamin D-free diet, prepared in the form of soft gelatin pellets (Table 1). The trout were kept in total darkness except for 10 min each day, during feeding, when the aquarium room was lit with a dim 60 W incandescent ceiling light with a minimum wavelength emission of about 400 nm. Blood samples were collected by cardiac puncture from trout, anaesthetised in a solution of MS222 (tricaine methanesulfonate, 100 mg/l). Analysis for cholecalciferol and 7-dehydrocholesterol in lipid extracts of trout tissues was performed by high performance liquid chromatography (HPLC) according to standard methods [10]. Determination of 25-hydroxy-cholecalciferol [25(OH)D] or 1,25dihydroxycholecalciferol in lipid extracts was done either by competitive protein binding assays or by HPLC [10]. HPLC analysis of cholecalciferol had a minimum detection limit of 1 ng/g wet tissue. Trout were irradiated in a solar light simulator (Full Spectrum F35 simulator model 1 with scanning system, CSIRO, Australia). The wavelength and intensity of light was controlled by the use of different light sources and filters. Four different irradiation wavelength ranges were produced: (1) full spectrum sunlight, (2) ultraviolet B (290–320 nm), (3) ultraviolet A (350–400 nm), (4) visible blue light (380–480 nm). Radioactively-labelled cholecalciferol: [1a-3H] cholecalciferol (2 Ci/mmol) and [1,2-3H] cholecalciferol (276.9 mCi/mmol) were kindly prepared by P.A. Bell and B. Pelc by the method of Callow et al. [11]. Determination of the distribution of 3H on carbon-1 of [1a-3H] cholecalciferol has shown that 85% is in the 1a-position and 15% is in the 1b-position [12]. Therefore, in the enzymemediated 1a-hydroxylation of this tritium-labelled molecule, about 85% of the 3H is lost as 3H2O. [4-14C] Cholecalciferol (32.3 mCi/mmol) and 25-hydroxy[26,27-3H] cholecalciferol (11 Ci/mmol) were obtained from The Radiochemical Centre, Amersham, U.K. [4-14C, 1a-3H] 25-hydroxycholecalciferol was prepared by standard Sephadex LH20 chromatography, from lipid extracts of the blood of pigs which had been injected intravenously with either [1a-3H] cholecalciferol or [4-14C] cholecalciferol.
Table 1 Vitamin D-free fish diet. Water
1l
Gelatin Ethanol-extracted casein Powdered dextrin Powdered cellulose Inorganic saltsa Corn oil Linolenic acid Vitaminsa
60 g 190 g 140 g 35 g 35 g 40 g 5g
a Inorganic salts, water-soluble and fat-soluble vitamins (without cholecalciferol) were added according to the defined requirements of micronutrients for rainbow trout [9].
59
2.1. Trout liver homogenates were incubated for 3–4 h at 17–18 with radioactive 25-Hydroxycholecalciferol in an NADPH generating medium [13]. The incubation mixture was then extracted with chloroform/ methanol and residual radioactive substrate and any metabolic products were separated and quantified by Sephadex LH20 lipid chromatography. 3. Results Each rainbow trout egg, weighing 80–85 mg, contained 5–6 ng of vitamin D activity as determined by chick toe–ash bioassay. No 25(OH)D was detected by competitive protein binding analysis in the trout egg lipid extract. It was, therefore, assumed that each fish at hatching, contained up to 6 ng cholecalciferol. HPLC analysis of lipid extracts of the fish diet was unable to detect any cholecalciferol or 25(OH)D. The absence of detectable 25(OH)D or 1,25(OH)2D in blood plasma as the fish grew, suggested also that these fish were vitamin D deficient and that they were not making vitamin D by a metabolic process. There was a significant decrease (p < 0.05) in total Ca in blood plasma (1.6 0.48 mmol/l, mean SD, n = 15) of the vitamin D-deficient trout compared to that from non-deficient commercial trout of 2.70 0.06 mmol/l. There was also a significant decrease (p < 0.05) in plasma protein concentration from 35.00 0.27 mg/ml in commercial trout to 22.80 0.46 mg/ml in the deficient trout. No differences were found in plasma phosphorus concentration or in bone calcium or bone phosphorus content between the vitamin D-deficient trout and non-deficient commercial trout. There were no behavioural or other visible signs of abnormality demonstrated by these trout. Vitamin D-deficient trout, weighing about 200 g, were fasted for 24 h and then injected intraperitoneally with 30 ng [1,2-3H] cholecalciferol in 100 ml propylene glycol. After a further 24 h without food, the trout were killed and various tissues were collected to measure the distribution of radioactivity. Surprisingly, about 90% of the tritium dose was found as more polar metabolites of vitamin D in the bile which had accumulated in the gall bladder during the time of starvation. Hence, rather than using this small dose of vitamin D to perform some functional role, it appeared that it had been metabolically changed and excreted in bile. In comparable experiments, vitamin D-deficient trout were injected intraperitoneally with 25 ng (0.65 mCi) [26,27-3H] 1,25-dihydroxycholecalciferol dissolved in 100 ml propylene glycol. The fish were killed 3 h later and tissues were collected for determination of radioactivity. Again, the concentration of excreted radioactivity in bile at 6.2 ng/ml was greater than in blood plasma (0.17 ng/ml), kidneys (0.3 ng/g) and liver (0.9 ng/g), and only about 20% of the radioactivity in bile appeared to be unchanged 1,25(OH)2D3. Vitamin D, as [1,2-3H] cholecalciferol, was incorporated into the fish diet at a concentration of 50 mCi (70 mg) per 100 g dry ingredients. This was fed to a group of vitamin D-deficient trout for 4 months. The animals were then fed the vitamin D-deficient diet for a further 4 weeks, after which they were killed and the nature of the radioactivity in liver was determined by Sephadex LH20 chromatography of lipid extracts. Only one peak of radioactivity was found in liver and this was less polar than that of cholecalciferol (Fig. 1). On saponification and re-chromatography of the non-saponifiable lipid, the radioactivity was found to cochromatograph with cholecalciferol. It was therefore, concluded that the radioactivity extracted from the trout liver after long term feeding of [1,2-3H] cholecalciferol, was the fatty acid ester of the labelled cholecalciferol. Vitamin D-deficient trout at a body weight of about 100 g were injected intraperitoneally with 1 mCi (19.4 ng) [26,27-3H]
60
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
Fig. 1. Sephadex LH20 chromatograph of radioactivity in lipid extract of the liver of vitamin D-deficient rainbow trout after feeding for 4 months [1,2-3H] cholecalciferol, incorporated into the fish diet at a concentration of 50 mCi (70 mg) per 100 g dry ingredients and then feeding the vitamin D-free diet for a further 4 weeks before analysing the radioactive material remaining in liver.
25-hydroxycholecalciferol in 100 ml propylene glycol and blood was collected 19 h later. Lipid extracts of blood plasma when chromatographed on Sephadex LH20 revealed two peaks of radioactivity (Fig. 2). About 55% of the tritium in plasma was
Fig. 2. Sephadex LH20 chromatograph of radioactivity in lipid extract of blood plasma from vitamin D-deficient rainbow trout, collected 19 h after being injected intraperitoneally with 1 mCi (19.4 ng) [26,27-3H] 25-hydroxycholecalciferol. 55% of the radioactivity in plasma co-chromatographed with 25-hydroxycholeclaciferol and 44% co-chromatographed with 1,25-dihydroxycholecalciferol.
present as the injected 3H-25(OH)D3 while about 44% was in a 3H-labelled metabolite, co-chromatographing with 1,25(OH)2D3. Analysis of the blood plasma of vitamin D-replete commercial rainbow trout by competitive protein binding assays, showed that the 25(OH)D concentration was below the level of detection of 1.2 ng/ml. However, the concentration of 1,25(OH)2D was 128 10 pg/ml. Incubation of [4-14C, 1a-3H] 25(OH)D3 (0.7 mg; 3H/14C = 5) with the homogenate of 1.9 g liver from vitamin D-deficient rainbow trout (body weight = 240 g) for 4 h at 17 converted 61% of the substrate to a more polar product with a 3H/14C ratio of 2.56, thus confirming by the loss of the 1a-tritium atom, its identity as 1,25 (OH)2D3 (Fig. 3). Vitamin D-deficient rainbow trout at 150–200 g body weight were fed the semi-purified diet which had been fortified with 612 mg cholecalciferol per kg dry ingredients. Four trout were then sacrificed at weekly intervals and the concentrations of calcium and 1,25(OH)2D3 in blood plasma were measured. After 1 week on the vitamin D-fortified diet the fish had become hypercalcemic but blood calcium returned to the normal range after 2 weeks (Fig. 4). The concentration of 1,25(OH)2D3 in plasma rose gradually to a maximum of about 180 pg/ml after about 4 weeks and thereafter declined to a range of 60–80 pg/ml (Fig. 5). When aquarium tanks containing rainbow trout, which had been reared in darkness on the vitamin D-free diet for 2 years, were inadvertently exposed for 24 h to 60 W incandescent light, the trout rapidly became moribund and died. Because it had been shown that the hepatic 1-hydroxylase of these vitamin D-deficient fish was very active and that feeding dietary cholecalciferol led to hypercalcemia, it was postulated that exposure to visible light had in some way induced the formation of vitamin D in skin which resulted in a lethal overproduction of 1,25(OH)2D3 and consequent fatal hypercalcemia. HPLC analysis of lipid extracts from the skin of a number of species obtained from a fish market, demonstrated the presence not only of 7-dehydrocholesterol but also measurable quantities of cholecalciferol (Table 2). Rainbow trout and Atlantic salmon had higher levels of 7-dehydrocholesterol in skin than
Fig. 3. Sephadex LH20 chromatography of lipid extract from vitamin D-deficient rainbow trout liver homogenate incubated at 17 for 4 h with [4-14C, 1a-3H] 25(OH) D3 (0.7 mg; 3H/14C = 5). 30% of the 14C radioactivity co-chromatographed with 25 (OH)D3 with a 3H/14C ratio of 4.96 and 61% of the 14C radioactivity cochromatographed with 1,25-dihydroxycholecalciferol with a 3H/14C ratio of 2.56, confirming its identity as the 1-hydroxylated metabolite of 25(OH)D3.
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
61
Fig. 4. Calcium concentration in the blood plasma of vitamin D-deficient rainbow trout, with time, while feeding a diet containing 612 mg cholecalciferol/kg DM. Four fish were killed at each time point and the plotted values are means SD.
other sea water species, but all the fish examined had 20 ng or more cholecalciferol per gram wet weight of skin. To test the hypothesis that visible light was able to induce vitamin D formation in skin, rainbow trout were irradiated in a variable wavelength solar simulator, with light passing through 5 cm water at 12 . Trout were exposed either to full spectrum
simulated solar light (290–1200 nm) or to blue light (380–480 nm) (Fig. 6). Exposure for two 30 min periods resulted in the appearance of vitamin D in skin with both light sources (Fig. 7), with full spectrum simulated sunlight producing more vitamin D than blue light. Furthermore, increasing amounts of 1,25(OH)2D3 were detected in blood of these fish after irradiation (Fig. 8). The identity of vitamin D obtained from the skin of trout irradiated with 380–480 nm blue light was confirmed by mass spectrometry (Fig. 9). In other experiments in which dead rats, with the hair coat closely clipped to the skin, and dead trout were both exposed to 380–480 nm blue light, vitamin D was again found in the trout skin but there was no change in the small quantities of vitamin D in rat skin. It was, therefore, concluded that some characteristic of trout skin structure had enabled the energy of blue light to convert 7-dehydrocholesterol into cholecalciferol. 4. Discussion
Fig. 5. Concentration of 1,25(OH)2D3 in the blood plasma of vitamin D-deficient rainbow trout, with time, while feeding a diet containing 612 mg cholecalciferol/kg DM. Each point represents the 1,25(OH)2D3 concentration in the pooled plasma of four trout.
It has been shown by several investigators that the liver of rainbow trout and of other fish species, is capable of synthesising 1,25(OH)2D3 from 25(OH)D3 [14–17]. It also appears that the kidneys of several fish species have the 1-hydroxylase enzyme [18,19]. In comparison to terrestrial vertebrates, the concentration of 25(OH)D in blood plasma is very low while that of 1,25(OH)2D is higher than found in mammalian blood [20]. In support of the concept that 1,25(OH)2D3 is the main circulating form of vitamin D, our unpublished measurements on the plasma vitamin D-binding protein (DBP) from goldfish (Carassius auratus) indicate that in contrast to mammalian, avian and amphibian DBP, the binding affinity for 1,25(OH)2D3 is higher than that for 25(OH)D3. From these many findings, it has been concluded that the liver of fish takes up cholecalciferol and converts some to 25(OH)D3 which in turn is further hydroxylated to 1,25(OH)2D3 and then secreted into blood.
62
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64 Table 2 Concentration of 7-dehydrocholesterol and cholecalciferol in the skin of various fish species. Fish species
7-Dehydrocholesterol mg/g skin wet weight
Cholecalciferol ng/g skin wet weight
Rainbow trout (Oncorhynchus mykiss) Atlantic salmon (Salmo salar) Yellowfin tuna (Thunnus albacares) Trevally (Pseudocaranx georgianus) Blackfish (Girella elevata) Bream (Acanthopagrus australis) Whiting (Sillago ciliata)
10.7 1.4 (n = 16) 26.0 3.2 (n= 4) 0.07 0.01 (n = 4) 0.95 0.35 (n = 5) 1.1 0.3 (n = 3) 0.3 0.04 (n = 5) 1.0 0.2 (n = 12)
300 20.7 (n = 15) 22.5 5.2 (n = 4) 72.1 15.7 (n = 3) 18.6 3.2 (n = 8) 23.9 9.1 (n = 4) 20.9 9.6 (n = 3) 24.9 5.4 (n = 9)
Fresh fish were purchased from the Sydney Fish Markets to obtain skin for analysis. Mean values SD obtained by HPLC analysis of lipid extracts of fish skin.
An alternative metabolic fate of hepatic cholecalciferol in fish, is to become esterified with long chain fatty acids. This vitamin D ester could possibly be a storage form of vitamin D. However, when radioactively-labelled cholecalciferol esters were formed in liver of vitamin D-deficient rainbow trout, which had been fed food containing [1,2-3H]-cholecalciferol, the esters remained even when the fish were again on a vitamin D-free diet (Fig 1). Vitamin D-deficient trout also catabolised most of injected doses of [1,2-3H] cholecalciferol and secreted the catabolites in bile. From these results it could be concluded that the liver of trout is especially capable of detoxifying exogenous vitamin D, both by metabolic inactivation and by sequestration as biologically inert fatty acid esters. Such mechanisms would be an advantage for carnivorous fish, the diet of which may be other fish, rich in vitamin D esters, which they themselves had accumulated throughout their life. The esterification of cholecalciferol to detoxify that received in food, may thus be an explanation why some species of fish, such as cod (Gadus morhua), contain large quantities of cholecalciferol esters in liver [3]. One potential toxic action of vitamin D could be the induction of hypercalcemia by some endocrine action of 1,25(OH)2D3, secreted
Fig. 6. Light intensity over the wavelength range of 380–480 nm used to irradiate vitamin D-deficient rainbow trout in a variable solar light simulator.
Fig. 7. Concentration of vitamin D in the skin (ng/g wet weight) of vitamin Ddeficient rainbow trout after they were irradiated in a solar light simulator. Irradiation was for two periods of 30 min, 2 h apart. During the 2-h interval between irradiations, the fish was kept in darkness. Each point represents the mean SD of four fish. A-1 and A-2 indicate the skin vitamin D content of trout irradiated with full solar spectrum light. B-1 and B-2 indicate the skin vitamin D content of trout irradiated with light over the wavelength range of 380–480 nm as illustrated in Fig. 6.
Fig. 8. Concentration of 1,25(OH)2D in blood plasma (pg/ml) of vitamin D-deficient rainbow trout after they were irradiated in a solar light simulator. Irradiation was for two periods of 30 min, 2 h apart. During the 2-h interval between irradiations, the fish was kept in darkness. Each point represents the value from the pooled plasma of four fish. A-1 and A-2 indicate the plasma 1,25(OH)2D concentration of trout irradiated with full solar spectrum light. B-1 and B-2 indicate the plasma 1,25 (OH)2D concentration of trout irradiated with light over the wavelength range of 380–480 nm as illustrated in Fig. 6.
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
63
Fig. 9. Mass spectra comparison of standard cholecalciferol with that of putative cholecalciferol extracted and chromatographically purified from the skin of vitamin Ddeficient rainbow trout after irradiation with light of the wavelength range of 380–480 nm.
by the liver [21–23]. In our studies with vitamin D-deficient trout, the restoration of vitamin D status by supplying vitamin D in food caused hypercalcemia and this was associated with high concentrations of 1,25(OH)2D3 in blood (Figs. 4 and 5). Vitamin D deficiency in trout apparently stimulates the activity of hepatic 1a-hydroxylase as demonstrated by incubation of liver homogenates with radioactively labelled 25(OH)D3 (Fig. 3). It is possible also that the output of 1,25(OH)2D3 by liver of normal fish may not be tightly regulated, in contrast to the well-regulated renal production of 1,25(OH)2D3 in terrestrial vertebrates [13]. The quantity secreted into blood may be determined by the production of 25(OH)D3. Therefore the liver of trout appears to be both a protective organ against toxicity from dietary vitamin D and also an endocrine organ, secreting the hormone 1,25(OH)2D3, which in some way affects blood calcium concentration. The big surprise from all of our studies, was the observation of the lethal effect of 24 h exposure of vitamin D-deficient trout to visible light from a 60 W incandescent bulb. These fish were in water with a high concentration of dissolved calcium carbonate, so if they had been producing toxic levels of 1,25(OH)2D3 in blood, their death may have been caused by increased uptake of calcium from water with consequent fatal hypercalcemia. However, in terrestrial vertebrates, visible light is not capable of converting 7-dehydrocholesterol in skin to cholecalciferol. We clearly demonstrated that shaved rat skin did not form vitamin D when exposed to visible light with the wavelength range of 380–480 nm. Nevertheless, cholecalciferol was indeed produced in the skin of both live and dead vitamin D-deficient rainbow trout exposed to visible light in the 380–480 nm wavelength range. This observation seems to defy the known energy requirement, (provided by UV light at 290–320 nm), to rupture the 9–10 bond of the B ring of 7-dehydrocholesterol to form pre-cholecalciferol. One possibility is that the structure of fish skin, with its metallic sheen caused by thin bony scales associated with intracellular anhydrous guanine
crystals [24] focuses the lower energy of multiple photons of blue light onto molecules of 7-dehydrocholesterol to induce breakage of the 9–10 bonds. Identification of the physicochemical mechanism for this process will require further research. In clear water, solar UV light at 290–320 nm can penetrate to about 20 m, by which depth about 90% of the UV energy has been absorbed [25]. Although rainbow trout, in their natural habitat of freshwater rivers, in water less than 20 m deep, could well be exposed to solar UV light, many other species of fish inhabit depths below the penetration range of UV light. If such deep water fish are able to produce cholecalciferol from 7-dehydrocholesterol in skin by exposure to visible blue light, then this may be a mechanism by which many species of fish, as well as rainbow trout, acquire vitamin D from exposure to sunlight. In contrast to the limited penetration range of solar UV light, 10% of the blue light range of 380–480 nm is still present at depths of 200 m [26]. 5. Conclusions Studies with vitamin D-deficient rainbow trout have demonstrated that radioactively labelled cholecalciferol is both metabolically inactivated and converted to biologically inert vitamin D esters in the liver. The liver also secretes 1,25(OH)2D3 into blood and the production of this hormone appears to be stimulated in vitamin D deficiency. Trout skin, unlike that of rats, is able to produce cholecalciferol when exposed to visible light in the wavelength range of 380–480 nm. These findings suggest that instead of obtaining vitamin D from a food chain, starting with zooplankton at the water surface, some fish may depend upon visible light from the sun, to induce cholecalciferol production in skin to meet their vitamin D requirements. Rather than using food as a source of vitamin D, these fish may need the vigorous hepatic inactivation mechanisms to protect against toxicity from high levels of vitamin D in the food chain.
64
S.L. Pierens, D.R. Fraser / Journal of Steroid Biochemistry & Molecular Biology 145 (2014) 58–64
Acknowledgement The authors are most grateful for the dedicated assistance of Angelika Trube in the care of the rainbow trout and the whole aquarium system, and for her expert help with laboratory analyses. References [1] A. Takeuchi, T. Okano, M. Ayama, H. Yoshikawa, S. Teraoka, Y. Murakami, T. Kobayashi, High-performance liquid chromatographic determination of vitamin D3 in fish liver oils and eel body oils, J. Nutr. Sci. Vitaminol. (Tokyo) 30 (1984) 421–430. [2] T. Yamakawa, T. Kinumaki, Determination of vitamin D in natural products, Vitamins 46 (1972) 167–177. [3] W. Von Müller-Mulot, G. Rohrer, K. Schwarzbauer, Native vitamin D3 esters in cod-liver oil: chemical determination of free, esterified and total vitamin D3,, Fette Seifen Abstrichmittel 81 (1979) 38–40. [4] D.R. Fraser, E. Kodicek, Investigations on vitamin D esters synthesized in rats. Turnover and sites of synthesis, Biochem. J. 106 (1968) 491–496. [5] G.A. Blondin, B.D. Kulkarni, W.R. Nes, A study of the origin of vitamin D from 7-dehydrocholesterol in fish, Comp. Biochem. Physiol. 20 (1967) 379–390. [6] D.S. Rao, N. Raghuramulu, Lack of vitamin D3 synthesis in Tilapia mossambica from cholesterol and acetate, Comp. Biochem. Physiol. A 114 (1996) 21–25. [7] A.M. Copping, Origin of vitamin D in cod-liver oil: vitamin D content of zooplankton, Biochem. J. 28 (1934) 1516–1520. [8] J.A. Campbell, B.B. Migikovsky, A.R.G. Emalie, Studies on the chick assay for vitamin D I. Precision of tibia and toe ash as criteria of response, Poult. Sci. 24 (1945) 3–7. [9] Fish Nutrition, in: J.E. Halver (Ed.), Academic Press, New York, 1989, pp. 1–772. [10] R.L. Horst, T.A. Reinhardt, B.W. Hollis, Improved methodology for the analysis of plasma vitamin D metabolites, Kidney Int. 38 (1990) 28–35. [11] E. Callow, Metabolism of tritiated vitamin D, Proc. R. Soc. London B 164 (1966) 1–20. [12] P.A. Bell, E. Kodicek, The stereospecificity of tritium distribution in [1-3H]- and [1,2-3H]-cholesterol and cholecalciferol, Biochem. J. 116 (1970) 755–757. [13] D.R. Fraser, E. Kodicek, Regulation of 25-hydroxycholecalciferol 1-hydroxylase activity in kidney by parathyroid hormone, Nat. New Biol. 241 (1973) 163–166.
[14] M.E. Hayes, D.F. Guilland-Cumming, R.G.G. Russell, I.W. Henderson, Metabolism of 25-hydroxycholecalciferol in a teleost fish: the rainbow trout (Salmo gairdneri), Gen. Comp. Endocrinol. 64 (1986) 143–150. [15] K. Sundell, J.E. Bishop, B.T. Björnsson, A.W. Norman, I,25-Dihydroxyvitamin D3 in the Atlantic cod: plasma levels, a plasma binding component, and organ distribution of a high affinity receptor, Endocrinology 131 (1992) 2279–2286. [16] A. Takeuchi, T. Okano, T. Kobayashi, The existence of 25-hydroxyvitamin D31a-hydroxylase in the liver of carp and bastard halibut, Life Sci. 48 (1991) 275–282. [17] A. Takeuchi, Comparative studies on vitamin D in vertebrates especially in fish, Vitamins (Jpn.) 68 (1994) 55–64. [18] H. Henry, A.W. Norman, Presence of renal 25-hydroxy-vitamin-D-1-hydroxylase in species of all vertebrate classes, Comp. Biochem. Physiol. B 50 (1975) 431–434. [19] I.E. Graff, O. Lie, L. Aksnes, In vitro hydroxylation of vitamin D3 and 25-hydroxy vitamin D3 in tissues of Atlantic salmon Salmo salar, Atlantic mackerel Scomber scombrus, Atlantic halibut Hippoglossus hippoglossus and Atlantic cod Gadus morhua, Aquac. Nutr. 5 (1999) 23–32. [20] L.V. Avioli, Y. Sonn, D. Jo, T.H. Nahn, M.R. Haussler, J.S. Chandler, 1,25Dihydroxyvitamin D in male, non spawning female, and spawning female trout, Proc. Soc. Exp. Biol. Med. 166 (1981) 291–293. [21] S.K. Srivastav, R. Jaiswal, A.K. Srivastav, Response of serum calcium to administration of 1,25-dihydroxyvitamin D3 in the freshwater carp Cyprinus carpio maintained either in artificial freshwater, calcium-rich freshwater or calcium-deficient freshwater, Acta Physiol. Hung. 81 (1993) 269–275. [22] K. Sundell, A.W. Norman, B.T. Björnsson, 1,25(OH)2 vitamin D3 increases ionized plasma calcium concentrations in the immature Atlantic cod Gadus morhua, Gen. Comp. Endocrinol. 91 (1993) 344–351. [23] A.J. Srivastav, G. Flik, S.E. Wendelaar Bonga, Plasma calcium and stanniocalcin levels of male tilapia Oreochromis mossambicus, fed calcium-deficient food and treated with 1,25 dihydroxyvitamin D3, Gen. Comp. Endocrinol. 110 (1998) 290–294. [24] D. Gur, Y. Politi, B. Sivan, P. Fratzl, S. Weiner, L. Addadi, Guanine-based photonic crystals in fish scales form from an amorphous precursor, Angew. Chem. Int. Ed. 52 (2013) 388–391. [25] C. Booth, J. Morrow, The penetration of UV into natural waters, Photochem. Photobiol. 65 (1997) 254–257. [26] E.M. Kampa, Underwater daylight and moonlight measurements in the eastern North Atlantic, J. Mar. Biol. Assoc. U. K. 50 (1970) 397–420.
