Selenium in fish nutrition: a role in gill membrane integrity

Selenium (Se) is an interesting atom, falling into Class VI of the Periodic Table, along with sulfur (S). Molecular compounds exist in three valent states: Se+2, Se+4, and Se+6, similar to sulfur. Selenium has ionizing energies of 940, 2000, and 3000 kJ/mole, again similar to sulfur at 1000, 2000, and 3300 kJ/mole. Selenium often occurs in physiological tissues, replacing sulfur in the structure of biologically active compounds. This phenomenon is critically apparent in the enzymes involved in controlling or reducing oxidative and free-radical damage to tissue membranes.

Fish have a special dependency upon minimizing membrane integrity in gill and liver cell membranes, which separate the circulating blood from adjacent water or liquid environments by only a 2-cell membrane wall structure. Osmoregulatory function, between hypotonic exterior water surface areas and more hypertonic circulating blood fluids, is critical to minimize influx of water molecules in Donnan Equilibrium conditions, and resultant excess interior tissue flooding.

Marine fish have increased reliance on gill membrane integrity to minimize water loss and subsequent dehydration of tissues in the hypertonic sea water state. The general medical and biological effects of selenium were reviewed in an NRC bulletin (1976). Selenium in terrestrial animal nutrition was summarized in a later bulletin (NRC, 1983).


Selenium and sea survival

The role of selenium in ameliorating the stress of smolting and subsequent conversion into the hypertonic environment of the sea was first recognized in a large scale testing of salmonid smolts in 1992 (USFWS, 1992), when the US Fish and Wildlife Service moved large samples of smolts from nine different hatcheries from their fresh water environments to the sea.

Four to nine groups of 200-300 smolts in each were transported from the different hatcheries of the Columbia River Chinook salmon rehabilitation program, to the Marrowstone Point anadromous station, and were there, over a 5-day period, converted into a sea water environment. The same diets previously used continued to be fed to each group for six weeks.

Survival was measured, with dramatically conflicting results, varying from 20-80% with only one group achieving 90% of the population surviving by the end of the 6-week test period. Two additional groups of fish were also tested, one Rainbow trout and the other Coho salmon, that had been fed diet H440 which had both an ascorbic acid and a selenium supplement. These two species of anadromous salmonids both had over 90% survival in the sea.

Analysis of the hatchery water supply of the high surviving Chinook smolt group showed higher selenium and copper values than the other hatchery water supplies tested (USFWS, 1992).

From these tests the more specific role of selenium on sea survival and fish health was emphasized.


The glutathione peroxidase enzymes

Glutathione peroxidase (GSH-Px) was investigated as one probable mediating factor, since this enzyme has selenium substituted for sulfur as selenocysteine in its structure.

GSH-Px was discovered in 1957 in terrestrial animals (Mills, 1957). The biochemical function of GSH-Px is to reduce lipid organic peroxide-hydroperoxides to their corresponding alcohols, and to reduce free hydrogen peroxide to water. A typical reaction catalyzed by GSH-Px follows. Ascorbic acid can provide the indicated proton.

2GSH+H2O2 GSSG+2HOH then in the cell:

GSSG+NADPH+H+ 2GSH+NADP

Other GSH-Px enzymes were soon found. GSH-Px-1 is the major form present in the cyctosol of cells. GSH-Px-2 and GSH-Px-3 are other forms in cell plasma. All are tetramers with four selenocysteine atoms in their structures of about 84,000 Daltons.

GSH-Px-4 was later discovered to be a monomer of about 24 kD, with a more specific function to reduce peroxidation of the lipid moiety in the phospholipids of cell membranes.

Again, vitamin C can provide the proton for completion of the cycle. An interesting phenomenon of the ionization state of the enzyme reveals that it must be in the reduced state for activity. The major inter- and intracellular reducing capability of ascorbic acid needs to be considered to expedite the reduction of the GSH-Px from its oxidized form back to the reduced form necessary to attack another phospholipid peroxide and minimize further polyunsaturated fatty acid cell membrane structural damage.


Ascorbic acid

The dietary deficiency signs of vitamin C are manifold, but generally are related to hydroxylation reactions. One of the most frequently encountered is collagen abnormalities due to proline to hydroxyproline failure, resulting in abnormal protein synthesis and in the bizarre scurvy symptoms often seen in fish. Fish soon show hyperplasia of collagen and cartilage, then scoliosis, lordosis, internal hemorrhage, resorbed opercles, and abnormal support cartilage in gills, spine, and fins, with hyperplasia of the jaw and snout. These same symptoms have been observed in trout, salmon, yellowtail, carp, guppies, catfish, snakehead, tilapia, minnows, mullet, char, and other fish (Halver, 2002).

Several different forms of vitamin C are used in dietary supplements, each with different degrees of storage stability. Ascorbic acid (C₁) is prone to rapid oxidation to inactive dehydroascorbic acid (DHA). Ascorbic-2-monophosphate (C2MP), ascorbic-2- diphosphate (C2DP), ascorbic-2-polyphosphate (C2PP), are readily hydrolyzed by ingredient phosphorylases. Ascorbate-2-glycoside (C2G) and ascorbate-6-phosphate (C6P) are slowly hydrolyzed and oxidized to C₁ and dehydroascorbic acid. Quantitative analysis for these vitamin C sources and derivatives face an enigma.

Total vitamin C can be measured by extraction with strong acids that hydrolyze the derivatives, then deliberate oxidation and measurement of DHA. Mild solvent extraction and subsequent HPLC analysis requires special columns to detect ingredient and tissue stored forms of biologically active ascorbate (Halver and Felton, 2001). C2MP is relatively stable and is the most common form used as a supplement in fish diets. When hydrolyzed by phosphorylase in the digestive tract it provides the C₁ needed to keep the GSH-Px enzymes in the reduced state to assure activity.


Selenium and stress

One large scale test for the effects of stress on GSH-Px activity was conducted in bargetransported Chinook salmon smolts. Two salmon smolt barge-transport experiments were conducted to measure tissue selenium loss during the 30-hr transport 500 km downriver past seven hydroelectric dams on the Columbia River. Carcass selenium was measured before and after the barge trip. Liver GSH-Px activity and total ascorbate concentrations were assayed to correlate selenium loss with GSH-Px activity.

Hatchery reared salmon smolts fed diets supplemented with selenium and C2MP were PIT tagged, released into the river at the hatchery and then collected downriver at Lower Granite Dam, where they were confined in five compartments in a transport barge equipped with a water recirculation system. They were towed for 30 hrs to below Bonneville Dam and then released.

Samples were collected before introduction to the barge and were collected by grab net from one of the holds just prior to release. Fish were immediately anesthetized, blood samples were removed from the caudal peduncle, gills and viscera were removed, then liver and carcass were frozen on dry ice and transported to the laboratory.

Frozen eviscerated carcasses were weighed, measured, partially thawed and homogenized in two volumes of distilled water, then dried in a vacuum oven to constant weight. The dried tissue was pulverized in a mortar, placed in capped plastic tubes and archived until used for the selenium assays. The dried fish powder (250 mg) was weighed into a 100 mL Kjeldahl flask and 10 mL of sulfuric acid mix (4H₂SO₄ to 10HNO₃) was added.

The flask was heated until digestion was completed and faint white fumes appeared. The flask was cooled and 5 mL of hydrogen peroxide-nitric acid mixture (30% H₂O₂ and HNO₃ 1:1) was very slowly added. When foaming subsided the flask was reheated until white fumes reappeared. The flask was cooled and concentrated HCL (15 mL) was slowly added to the hydrolyzate and heating was continued until the volume was reduced to 6-8 mL. The flask was then cooled and volume adjusted to 10 mL with distilled water.

The selenium assay followed the Adeloju and Bond procedure (1983). A square wave voltimeter was used with the following conditions: at the hanging drop mercury electrode: initial potential -0.50v, final potential -0.84v; scan increment 2mV, pulse height 20mV, purge 30s, deposit time 180x, drop size large. The assays showed 20-30% of carcass selenium was mobilized and lost during the 30-hr trip. GSH-Px was concomitantly increased.

The GSH-Px enzyme activity was measured with an Oxis kit following the general procedure of Folke and Gunsler (1984). Frozen livers were pooled (three fish per sample) from each group. The sample was homogenized in a Polytron homogenizer for 1 min.

Homogenate was divided into two equal weight aliquots. One aliquot was used for the vitamin C assay, and the other for the GSH-Px assay. The GSH-Px aliquot was centrifuged for 20 min at 10,000 g at 5° C. The supernatant was divided into two equal aliquots, one for the enzyme activity and the other for the protein determination. The GSH-Px kit protocol was followed for the first aliquot. The second aliquot was used in the Lowry protein content determination. The liver GSH-Px activity was recorded as μM/mg protein.

The vitamin C aliquot was thawed and diluted 1:1 with 10% trichloracetic acid, bringing the solution to 1:10 dilution. The solution was homogenized for 30 sec and centrifuged in a refrigerator maintained at 2° C for 4 min. The supernatant solution was decanted and filtered through a 0.45 μm syringe filter. The filtrate was analyzed by HPLC using the procedure of Felton et al. (1994). HPLC conditions were as follows. Instrument: Perkin-Elmer model 250 Binary pump with LV 290 UV-Vis detector and PE Nelson software. Column: Alltech Associate Altima C-18 5μm, 250 x 4/6 mm column. Solvent: 0.1 M ammonium acetate pH 5.0 solution. Flow rate 0.75 mL at 2200 psi.

GSH-Px assays show a concomitant increase in activity as the selenium stores were reduced. The ascorbic acid levels were reduced but remained sufficient for the GSH-Px activity. A one-way analysis of variance (ANOVA) was used to indicate significant levels of differences. Differences among mean values were detected using the Tukey’s test. Significant differences were observed for selenium and GSH-Px mean values between start and end of barging (Halver et al., 2004).


Selenium requirements

A general selenium requirement for fish was first demonstrated by Poston (1976) when Atlantic salmon showed reduced survival of growing fish. The Se-deficient fish showed severe muscle dystrophy, which was cured when a combination of selenium and vitamin E was used. Se-deficient fish had depressed activities of GSH-Px in plasma. Later work showed that dietary supplementation with ascorbic acid increased the activities of GSHPx in plasma and improved the growth of Se-deficient Atlantic salmon.

The general roles of selenium in other animals have been reviewed in Selenium in Nutrition, Revised Edition (NRC, 1983). Lall has reviewed selenium functions, metabolism, deficiency, and requirements in the treatise Fish Nutrition (2002).

The polyunsaturated fatty acids in fish membrane lipids determine membrane flexibility and are prone to attack by reactive oxygen species. The resulting lipid peroxidation can cause injury and death to cells and certainly alter membrane flexibility and physiological function (Dey et al., 1993; Farkas et al., 1994; Farkas and Halver, 1996).


Gene expression

The recent flood of activity in the study of GSH-Px enzymes in neurobiology has focused attention upon the role of GSH-Px-4 and its role in protecting neurons from oxidative injury.

Prasade et al. (2002) reviewed the role of oxidative injury to brain cells and emphasized the role of antioxidants and protective agents in minimizing oxidative damage. The interrelationship of vitamin C and selenium was emphasized. Using neuroblastoma cell culture to mimic possible normal brain tissues, Yan et al. (2005) showed that the cell culture media, when insulted with prostaglandin, stimulated gene expression of GSHPx- 1 and superoxide dismutase, possibly as a response to generate more enzyme for better protection of existing DHA and EPA in the neurocell polyunsaturated fatty acid structure.

One of the most recent reports by Ran et al. (2006) has shown that overexpression of GSH-Px-4 is a stress response of neurons to oxidative injury. Their results show that primary culture cortical neurons derived from GSH-Px transgenic mice had increased cell survival after exposure to oxidants. They suggest that over-expression of GSH-Px-4 from the insult protects the neurons against oxidative injury and ß-amyloid cytotoxicity of early Alzheimer’s disease.

References

Adeloju, S.B. and A.M. Bond. 1983. Stripping voltammetric determination of selenium in biological materials by direct calibration. Anal. Chem. 55:2070-2082.

Dey, I., C. Buda, T. Wiik, J.E. Halver and T. Farkas. 1993. Molecular and structural composition of phospholipid membranes in livers of marine and freshwater fish in relation to temperature. Proc. Natl Acad. Sci. USA 90:7498-7502.

Farkas, T., I. Dey, C. Buda and J.E. Halver. 1994. Role of phospholipid molecular species in maintaining lipid membrane structure in response to temperature. Biophy. Chem. 50:147-155.

Farkas, T. and J.E. Halver. 1996. Involvement of phospholipid molecular species in controlling structural order of vertebrate brain synaptic membranes during thermal evolution. Lipids 31:1045-1050.

Felton, S.P., S.R. Grace and J.E. Halver. 1994. A non-ion pairing HPLC method for measuring new forms of ascorbate and ascorbic acid. J. Liquid Chrom. 17:123-131.

Felton, S.P, M.L. Landolt, S.R. Grace and A. Palmissano. 1996. Effects of selenium dietary enhancement on hatchery reared Coho when compared with wild Coho: hepatic enzymes and seawater adaptation evaluated. Aqua. Res. 27:135-142.

Folke, L. and W.A. Gunzler. 1984. Assays of glutathione peroxidase. In: Methods of Enzymology (L. Packer, ed). Vol. 104, Elsevier, New York, NY, USA, pp. 1114-1120.

Halver, J.E. 2002. The Vitamins in Fish Nutrition (J.E. Halver and R.W. Hardy, eds). 3rd Ed. Academic Press, San Diego, CA, USA, pp. 61-141.

Halver, J.E. and S.P. Felton. 2001. Analytical enigmas in assays for vitamin C. In: Ascorbic Acid in Aquatic Organisms (K. Dabrowski, ed). CRC Press, New York, NY, USA, pp. 7-12.

Halver, J.E., S.M. Felton and R. Zbanyszek. 2004. Carcass selenium loss as an indicator of stress in barge transported Chinook salmon (Oncorhynchus tshawytscha Walbaum) smolts. Aqua. Res. 35:1099-1103.

Lall, S. 2002. The minerals. In: Fish Nutrition (J.E. Halver and R.W. Hardy, eds). 3rd Ed, Academic Press, San Diego, CA, USA, pp. 259-308.

Lowry, O.H., N.J. Rosebrough, A.L. Farr and J. Randall. 1951. Protein measurement with the Folin-phenol reagent. J. Biol. Chem. 193:265-27.

Mills, G.C. 1957. Glutathione peroxidase. J. Biol. Chem. 229:189-197.

Poston, H.A, G.F. Combes and L. Leibovitz. 1976. Vitamin E and selenium interrelations in the diet of Atlantic salmon (Salmo salar): gross histological and biochemical deficiency signs. J. Nutr. 106:892-904.

Prasade, K.N., W.C. Cole and K.C. Prasade. 2002. Risk factors for Alzheimer’s Disease: Role of multiple antioxidants, non-steroidal anti-inflammatory and cholinergic agents alone or in combination in prevention and treatment. J. Amer. Coll. Nutr. 21(6):506- 522.

Ran, Q., M. Gu, H. Van Remmen, R. Strong, J.L. Roberts and A. Richardson. 2006. Glutathione peroxidase 4 protects cortical neurons from oxidative injury and amyloid toxicity. J. Neurosci. Res. 84:202-208.

National Research Council. 1976. Medical and Biologic Effects of Environmental Pollutants: Selenium. National Acedemy press, p. 206.

National Research Council. 1983. Selenium in Nutrition. Rev. ed., National Academy Press, p. 174.

USFWS Reports Marrowstone Station. 1992. Salmon smolt survival test for 1991 and 1992. US Fish & Wildlife Service Annual Report, Portland, OR, USA.

Yan, X.D., B. Kumar, P. Nahreini, A.J. Hanson, J.E. Prasade and K.N. Prasade. 2005. Prostaglandin-induced neurodegeneration is associated with increased levels of oxidative markers and reduced by a mixture of antioxidants. J. Neurosci. Res. 81:85-90.