Copper is an essential micronutrient. As with iron (Fe), it can undergo valency changes, from Cu (II) to Cu (I), and this ability to either accept or donate electrons makes it an important part of many catalytic processes. Some enzymes and biological processes where it plays a central role are given in Table 1. Perturbations in cuproenzyme activities are largely responsible for the clinical features of Cu deficiency, whereas overt signs of Cu overload stem from intracellular oxidative damage, particularly in the liver. Not surprisingly, Cu deficiency has a wide spectrum of consequences. In different species, these are manifest in a different order, with cardiac effects being seen first in ruminants(Reference Frank, Wibom and Danielsson2, Reference Klevay3), for example, while changes in glucose and cholesterol metabolism are observed first in humans(Reference Klevay, Inman, Johnson, Lawler, Mahalko, Milne, Lukaski, Bolonchuk and Sandstead4–Reference Klevay, Canfield, Gallagher, Henriksen, Lukaski, Bolonchuk, Johnson, Milne and Sandstead6).
The importance of Cu means that it has been the subject of intensive investigation over several decades, but despite these efforts, the ideal biomarker remains elusive. A comprehensive review covering the search for Cu biomarkers has recently been published(Reference Danzeisen, Araya, Harrison, Keen, Solioz, Thiele and McArdle1), and this paper summarises some of that material, whilst also considering whether there are other methodological approaches, not yet tested, which may generate data suggesting potential new status indicators. Despite the unmistakeable importance of Cu in maintaining health, there remains on-going difficulty with setting dietary recommendations due to the lack of sensitive and specific Cu biomarkers. Whilst severe deficiency and toxicity are relatively easy to recognize due to the obvious clinical signs, it is virtually impossible to identify marginal deficiency.
Cu deficiency can result from both primary and secondary causes. Primary causes usually relate to diet, though there are inherited disorders of Cu metabolism, such as Menkes' and Wilson's diseases, that result in systemic deficiency and overload respectively(Reference Cox and Moore7). Despite the adverse health consequences of these rare diseases, both have provided fundamental information for understanding the molecular basis of human Cu metabolism and status. Dietary Cu bioavailability undoubtedly influences Cu status, and whilst factors that affect the former are not fully characterized, nutrient-Cu interactions play a significant role. In infants the interaction of Cu with Fe is potentially the most important, with a reduction in Cu absorption demonstrated in formula-fed infants given high dietary levels of Fe (10·8 mg/L) compared with lower levels (1·8 mg/L)(Reference Haschke, Ziegler, Edwards and Fomon8). Use of zinc (Zn) supplements also increases the risk of Cu deficiency, since Zn blocks Cu absorption by up-regulating metallothionein transcription in enterocytes(Reference Coyle, Philcox, Carey and Rofe9). Several case studies have been reported with Cu deficiency occurring as a result of taking high levels of over-the-counter Zn supplements(Reference Rowin and Lewis10). This interaction is exploited in the treatment of Wilson's disease patients who are given pharmacological doses of zinc to avoid the accumulation of copper in the tissues(Reference Brewer11, Reference Brewer, Dick, Johnson, Brunberg, Kluin and Fink12).
Growing children and pregnant women are particularly vulnerable to mild/moderate Cu deficiency. The developing fetus accumulates significant Cu stores during the third trimester to provide for the first 3–4 months of life when dietary Cu intake is minimal(Reference Gambling, Danzeisen, Fosset, Andersen, Dunford, Srai and McArdle13). In order to meet this demand, maternal Cu absorption is up-regulated as demonstrated by a stable isotope study conducted during pregnancy(Reference Turnlund, Swanson and King14). Studies in children have shown that malnutrition commonly induces Cu deficiency, though of course the symptoms are confounded by other nutritional problems(Reference Castillo-Duran and Uauy15, Reference Cordano16). Many foods high in Cu, especially offal such as liver, are less commonly consumed now, and others, such as chocolate are high in fat, and hence are not considered beneficial for a healthy lifestyle. These factors also contribute to the risk of deficiency, especially in young women.
Cu deficiency can also arise as a consequence of other disorders and treatments. For example, coeliac disease(Reference Goyens, Brasseur and Cadranel17), Crohn's disease(Reference Spiegel and Willenbucher18) and other gut absorption problems all increase the risk of Cu deficiency, as do diseases of the immune system, such as AIDS and autoimmune diseases(Reference Stambullian, Feliu and Slobodianik19). The long-term consumption of high doses of antacids and other cation chelating agents reduce absorption, whilst excessive losses of caeruloplasmin–bound Cu may be experienced by patients undergoing ambulatory peritoneal dialysis(Reference Becton, Schultz and Kinney20).
Cu overload is less frequent but it also carries risks. Brewer and colleagues have campaigned for some time about the dangers of high Cu intake, and suggested that it may be associated with an increased risk of diseases such as Alzheimer's disease(Reference Brewer21). It has also been implicated in the development of prion diseases such as Creutzfeld-Jacob disease and kuru(Reference Pauly and Harris22). The data are equivocal, but provide further support for the need for sensitive and specific biomarkers of Cu status.
Current biomarkers
Most current approaches use cuproenzymes of one form or another. Many studies have used caeruloplasmin (Cp), for example, which is an acute phase protein, affected by the age and hormonal status of the individual. Cu homeostasis is tightly maintained by changes in both the absorptive efficiency and biliary excretion in the gut. At low and high intakes the efficiency of absorption is up- and down-regulated, respectively(Reference Turnlund, Keyes, Anderson and Acord23), but is predominantly controlled via endogenous excretion(Reference Harvey, Majsak-Newman, Dainty, Lewis, Langford, Crews and Fairweather-Tait24); however this control mechanism is imperfect at extremes of intake. Consequently, intervention studies have shown little or no effect of marginal or short-term Cu deficiency on either plasma Cp concentrations or activity(Reference Milne, Johnson, Klevay and Sandstead25, Reference Milne and Nielsen26). Cp also does not respond to high levels of dietary copper at the level of either mRNA transcription or protein translation. However, its activity is reported to decrease in response to severe Cu deficiency, so that it has value for indicating moderate/severe Cu deficiency(Reference Feillet-Coudray, Coudray, Bayle, Rock, Rayssiguier and Mazur27).
Other cupro-enzymes that have been tested, with greater or lesser success, include (SOD1), platelet CCO, lysyl oxidase and peptidylglycine α-amidating monooxygenase (refer to recent review for further information(Reference Danzeisen, Araya, Harrison, Keen, Solioz, Thiele and McArdle1)).
Recent developments in copper biomarkers
More recently, several groups have examined the expression of CCS, a Cu ‘chaperone’. When Cu is taken up by cells, it binds to one of a series of proteins (termed chaperones) which transport the metal to its target protein (Table 1). One of these, CCS, has been shown to change expression in response to Cu levels in a variety of models. Initial experiments carried out in rat models demonstrated that CCS protein levels were inversely proportional to Cu status and, that regulation appeared to act through degradation by the 28S proteosome(Reference Bertinato, Iskandar and L'Abbé28). Subsequently it was shown that Cu deficiency induced by feeding rats increased Zn in the diet could also de detected by erythrocyte CCS(Reference Iskandar, Swist, Trick, Wang, L'Abbé and Bertinato29). Interestingly, at a high level of Zn intake, Cu deficiency was actually improved, and this was correlated with a decrease in CCS expression. These data have been confirmed in mice on Cu deficient diets, supporting the idea that CCS or possibly the CCS:SOD1 ratio is a good indicator of Cu deficiency. Whether it will act as a good indicator of Cu excess remains yet to be tested, although data obtained in Drosophila melanogaster S2 cells suggest it may not be(Reference Southon, Burke, Norgate, Batterham and Camakaris30).
There is a body of evidence relating Cu deficiency to bone metabolism at all life stages. Skeletal defects such as osteopenia and spontaneous rib fractures are common features of Menkes disease in young children(Reference Ashkenazi, Levin, Djaldetti, Fishel and Benvenisti31–Reference Al-Rashid and Spangler33), bone defects in pre-term infants respond to Cu supplementation(Reference Allen, Manoli and LaMont34), and Cu deficiency is reportedly a factor in age-related osteoporosis(Reference Conlan, Korula and Tallentire35). Urinary pyridinoline and deoxypyrodinoline (biomarkers of bone resorption) may be useful functional indicators of Cu status. Studies have demonstrated increased bone resorption associated with Cu depletion in adult males(Reference Baker, Harvey, Majask-Newman, Fairweather-Tait, Flynn and Cashman36), and a reduced rate of bone loss at the lumbar spine in Cu-supplemented middle-aged women(Reference Eaton-Evans, McIlwrath, Jackson, McCartney and Strain37). However, the complex nature of bone metabolism suggests that these biomarkers are non-specific for Cu status and will be influenced by a variety of nutritional and environmental factors. Other suggested Cu biomarkers include immune and blood lipoprotein biomarkers (further information can be found in the recent review(Reference Danzeisen, Araya, Harrison, Keen, Solioz, Thiele and McArdle1)).
The potential for multiple markers, using high throughput methods such as proteomics, transcriptomics and other methods is being investigated. The current state-of-the-art for these techniques in relation to human nutrition is reviewed elsewhere in this supplement. Whilst several papers have reported the identification of suites of potential biomarkers in experimental models, application to the human situation is only just beginning to be investigated. Proteomics technology offers significant potential for the identification of novel Cu biomarkers particularly in relation to the analysis of Cu-transporting or Cu-binding proteins in both healthy individuals and those with Cu-related conditions such as Menkes' or Wilson's disease. There are specific technological problems associated with the investigation of metalloproteins, including analysis at low concentrations and the inherent instability in response to environmental changes. Consequently, isolation of Cu-containing proteins in physiological conformations is particularly challenging. A comprehensive summary of the proteomics of metal transport can be found in the review by Kulkarni and colleagues(Reference Kulkarni, She, Smith, Roberts and Sarkar38). The ability of these techniques to screen the entire proteome of a cell may ultimately facilitate the identification of biomarker(s) with no obvious role in Cu metabolism. Potentially, a protein-product substantially down-stream from processes clearly related to Cu metabolism may provide an unexpected component of the ‘suite’ of Cu biomarkers. Ultimately, a combination of ‘standard’ proteomics and transcriptomics technologies in conjunction with a range of innovative metal detection techniques will be required to drive the search for robust copper biomarkers.
Conclusions
In the absence of robust sensitive and specific biomarkers, it is difficult to know whether Cu status, either in relation to deficiency or excess, is a significant public health problem. Nonetheless, given the intake data that suggest levels may be lower than optimal and given the serious consequences of deficiency, there is a strong argument for developing such markers of status. After many years of searching, we believe that success is not too far away. CCS and the other chaperones, high throughput methods and identification of mechanisms of regulation all add to our knowledge and will hopefully contribute, so that one day we will be able to accurately assess an individual's Cu status and determine whether he or she is at risk of deficiency or overload.
Acknowledgements
HJM is funded by the Scottish Government Rural and Environmental Analysis Directorate, the European Union (NuGO and EARNEST) and the International Copper Association.
LJH is funded by an EU FP6 Network of Excellence (EURRECA, grant no. FP6-036196-2) (UEA), and the Biotechnology and Biological Sciences Research Council (IFR).