Wilson disease (WD) is a rare, autosomal-recessively inherited disorder of copper metabolism due to mutations of the WD gene ATP7B [1, 2]. The WD gene ATP7B codes for a membrane-bound, copper-binding protein that is expressed primarily in the liver [1, 2]. Disease causing mutations in the ATP7B gene result in a reduced biliary copper excretion [3]. WD is clinically characterized by hepatic (e.g. liver cirrhosis) and neurological manifestations related to the accumulation of copper in the liver and the brain [3]. The onset of signs and symptoms of WD typically is in childhood and young adulthood but may even be much later [4].

In addition to copper accumulation the disturbed function of ATP7B results in an impaired incorporation of copper into apoceruloplasmin leading to the formation of an unstable ceruloplasmin which is rapidly degraded [5]. Although this might be of minor importance in copper homeostasis, impaired holo-ceruloplasmin synthesis is of diagnostic relevance as serum ceruloplasmin concentration is considered a useful laboratory test for diagnosis of WD [6]. Serum ceruloplasmin concentration is determined in most routine laboratories immunologically, but can also be measured enzymatically by its oxidase activity [7–9]. Recently, we could show that measurement of the enzymatic oxidase activity of serum ceruloplasmin with o-dianisidine dihydrochloride as substrate is of greater diagnostic value than its immunologically measured concentration [10].

WD typically begins with a pre-symptomatic period, during which copper accumulation in the liver leads to subclinical hepatitis, and progresses to liver cirrhosis and development of neuropsychiatric symptoms. The type of hepatic and neurological symptoms can be highly variable. The wide variation in the clinical phenotype of WD has still to be elucidated and may be related to the ATP7B genotype. Although some authors have tried to establish whether the ATP7B genotype determines the phenotype of the disease, the data are conflicting and no definitive association has been established [11–14]. Reasons for the difficulties in assessing genotype-phenotype correlations in WD are the high genetic heterogeneity with the large number of mutations and the rareness of the disease. About 300 mutations have been described (Wilson Disease Mutation Database compiled by the Department of Medical Genetics at the University of Alberta is accessible at the web page http://www.wilsondisease.med.ualberta.ca/database.asp). Protein-truncating nonsense mutations and frameshift mutations are presumed to have a different impact on the functional impairment of the ATP7B protein than missense mutations and are hypothetically more 'severe' mutations than these. In line with this assumption, a more severe impairment of copper metabolism parameters and an earlier age of disease onset in WD patients with protein-truncating mutations could recently be demonstrated for different European cohorts [15, 16].

The aim of our study conducted in a German WD cohort was to evaluate whether the type of mutation correlates (1) with serum ceruloplasmin levels measured enzymatically with o-dianisidine as substrate, and (2) with age of onset of the disease.

Although considered a useful diagnostic test in WD, immunologically measured serum ceruloplasmin has been shown to be in the normal range in 35% of patients with hepatic WD [20] and a low immunologically measured serum ceruloplasmin has a positive predictive value of only 5.9% [21]. As apoceruloplasmin may interfere with the immunologic assay but not with the enzymatic ceruloplasmin measurement, an overestimation of serum ceruloplasmin can be avoided by use of the oxidase method for ceruloplasmin determination. Recently we could provide evidence that serum ceruloplasmin oxidase activity measured enzymatically with o-dianisidine as substrate is superior to immunoreactive ceruloplasmin in diagnosing WD.

In our present study we analyzed if serum ceruloplasmin oxidase levels correlate with the ATP7B genotype. The patients were grouped according to the mutations detected in protein truncating and missense mutations. This grouping was based on the hypothesis that truncating and non-truncating mutation groups are likely to have a different impact in the functional properties of the mutated protein. As serum ceruloplasmin levels can increase in response to pregnancy, estrogen therapy, or acute inflammation, patients fulfilling these criteria were excluded from the study cohort [18, 19].

For the first time we could show that serum ceruloplasmin oxidase activities measured enzymatically with o-dianisidine as substrate differ significantly between WD patients with protein-truncating and non-truncating ATP7B mutations. In our German WD cohort of 59 patients protein-truncating mutations were associated with significantly lower oxidase levels than missense mutations. In addition to ceruloplasmin oxidase activities we determined immunoreactive ceruloplasmin serum levels and could show the same relation. Our observation is in line with previously published studies that were conducted in other European study cohorts: one Greek, one Polish, and one Italian [15, 16, 22]. In these studies patients with protein-truncating mutations had lower ceruloplasmin concentrations than patients with missense mutations. This is of note, because it augments the line of evidence for an association of protein-truncating mutations with a more severe impairment of holo-ceruloplasmin synthesis. It might explain normal or borderline ceruloplasmin levels in some WD patients as protein-truncating mutations disrupt any copper-transporting function, while certain missense mutations might allow residual function in holo-ceruloplasmin synthesis.

In parallel to the effect on serum ceruloplasmin oxidase activities we observed an earlier onset of WD in patients with truncating ATP7B mutations when compared to missense mutations. This finding in our German study cohort is in accordance with two previously published studies conducted in European cohorts, a Polish and a Greek [15, 22]. However, other reports did not find a younger age of onset in patients with truncating mutations [16, 23, 24]. The effect of truncating mutations observed in our present study might be influenced by the p.H1069Q mutation, the most frequent mutation in European WD patients. For this mutation an association with a late and predominantly neurological presentation in WD could be shown in a recent meta-analysis [11], although the underlying studies are quite controversial [12–14]. Thus the predominance of the homozygous pH1069Q mutation in the MMs group in our study cohort might have biased the 'group effect' seen for MMs when compared to truncating mutations. Nevertheless, our results support the line of evidence that at least for European WD patients an influence of the type of mutation on the phenotype can be assumed.

Because ATP7B mutation analysis, due to the great heterogeneity of mutations, is quite laborious, we asked if serum ceruloplasmin levels might allow prediction of truncating mutations. By ROC analysis ceruloplasmin level cut-points with the highest sum of sensitivity and specificity for prediction of truncating mutations were identified. At an optimal cut-point ceruloplasmin oxidase activity allowed prediction of SMs with a sensitivity of 80% and a specificity of 79.5%. Based on this result, mutation analysis in European WD patients might be facilitated by considering serum ceruloplasmin oxidase level before starting mutation analysis, as assuming the more probable type of mutation before starting mutation analysis might shorten the mutation identification process. However, as discussed above, the results of ROC analysis might only apply for European patients and not for populations with a different mutation spectrum.

In conclusion, our data show that in our German study cohort truncating ATP7B mutations were associated with lower serum ceruloplasmin oxidase levels and an earlier age of disease onset when compared to MMs. At least in European populations measurement of serum ceruloplasmin oxidase might help to predict presence of truncating ATP7B mutations and might facilitate the mutation analysis.