Characterisation and expression of copper homeostasis genes in sea bream (Sparus aurata)

Abstract

The redox properties of Copper (Cu) make it both an ideal cofactor for many enzymes, and, in its free form, a highly toxic molecule capable of stimulating production of reactive oxygen species or binding to protein thiol groups. Therefore, living organisms have evolved homeostatic systems to “handle” Cu avoiding dangerous and wasteful aspecific interactions. These systems comprise uptake, carrier, storage and excretion proteins. The importance of Cu-homeostatic systems was initially discovered in humans where alterations of Cu-excretory proteins were shown to be responsible for two lethal genetic disorders; the Wilson and Menkes diseases. The levels of bioavailable Cu in the aquatic environment is important because concentrations in oceanic waters tend to be minute, whilst in some fresh and coastal waters, particularly around areas of mineral extraction, viniculture and farming operations, concentrations can be excessive. In contrast to terrestrial vertebrates, fish are not only exposed to dietary sources of copper but are also exposed to dissolved ionic copper that may enter via the skin and gills. Indeed, the latter route is important in fish and it has been demonstrated in physiological studies that under conditions of dietary deficiency, fish can satisfy their own body requirements by uptake from water. Therefore, fish must have systems relating to both gill and gut to enable maintenance of body homeostasis of this essential, yet toxic, metal. In an attempt to understand the mechanisms of Cu homeostasis in fish, whether under conditions of deficiency, adequacy or excess, it is essential to consider the expression of known Cu-homeostasis proteins. Thus, cDNAs for sea bream (Sparus aurata) homologues of copper transporter 1 (Ctr1), antioxidant protein 1 (Atox1), Menkes protein (ATP7A), Wilson protein (ATP7B), and metallothionein (MT), which are responsible for the uptake, delivery to the secretory pathway and scavenging of intracellular Cu, were cloned and their mRNA tissue expression levels measured. To investigate the molecular basis of the different homeostatic and toxic responses to waterborne or dietary Cu, sea bream were exposed to sub-toxic levels of Cu in the diet (130 mg/Kg of dry diet) or water (0.3 mg/L) and tissue mRNA and Cu levels were measured. Moreover, to discriminate between the effect of different metals on the transcriptional regulation of Cu homeostasis genes in fish, Sparus aurata fibroblast (SAF1) cells were exposed to sub-toxic levels of Cu (25 μM), Zn (100 μM) and Cd (10 μM). In addition, a microarray was used to gain a broader overview of the transcriptional response of SAF1 cells to Cu (25 μM). Waterborne or dietary Cu resulted in distinct expression profiles of Cu-homeostasis genes and markers of oxidative stress. After dietary exposure, Cu increased in intestine and liver, whilst after waterborne exposure Cu increased in gill and liver. Exposure to dietary Cu resulted in decreases in Ctr1 and ATP7A mRNA in both liver and intestine. Renal Ctr1 levels remained unchanged, whilst ATP7A mRNA decreased. In contrast, waterborne Cu exposure increased intestinal Ctr1 and ATP7A mRNA, and increased renal Ctr1 and decreased renal ATP7A mRNA. Both dietary and waterborne Cu increased ATP7B mRNA in liver. Metallothionein (MT) mRNA increased in liver and gill after waterborne Cu. Glutathione reductase (GR), a marker of oxidative stress, increased expression in liver and gill after waterborne Cu exposure, but decreased in intestine. Thus, exposure to Cu via water or diet has different, often opposite effects on Cu-homeostasis genes. The decrease in expression of both Cu-transport genes in intestine after dietary exposure may indicate a defensive mechanism to limit uptake of Cu. The opposite effects in intestine after waterborne exposure are more difficult to explain, but again may reflect a defence mechanism against excess bloodborne Cu coming from the gill. Since both dietary and waterborne Cu increased Cu levels in liver and increased hepatic ATP7B it is likely that well-characterised mammalian route of Cu excretion to bile is active in sea bream. However, only hepatic Cu derived from gill increased the expression of the stress markers MT and GR. This suggests that Cu is delivered to liver in a different form from gill as that from intestine, the intestinally derived pool being less toxic. Thus the increase in copper transport gene expression in intestine after gill exposure might be a mechanism to enable incorporation of excess bloodborne Cu into the intestinal pathway of Cu delivery to liver, thus minimizing toxicity. The in vitro exposure of SAF1 cells to Cu showed a similar response to liver of fish exposed to waterborne Cu indicating similar Cu availability and complexation. ATP7A mRNA levels were induced by Cu but not by Zn or Cd suggesting Cu-specific regulation. Conversely, MT and GR were induced by all metals tested. The transcriptomic analysis highlighted that the biological processes most significantly affected by Cu were secretion, protein trafficking and stress. Overall, these results show that in fish copper has distinct effects on tissue Cu transporter genes and oxidative stress depending on whether it is taken up via the gill or gut and that intestinal absorption may be required for normal uptake and metabolism of Cu, regardless of the route of uptake. Moreover, changes in mRNA levels indicate that Cu homeostasis genes, at least in fish, may be regulated at the transcriptional level. Although more work needs to be done to identify genes that are robust predictors of Cu toxicity, the microarray results presented here show a clear transcriptional fingerprint which may characterize Cu toxicity in fish.