Our present findings provide the evidence that the levels of some essential trace elements that contribute to hepatic inflammation, hepatic fibrosis, and hepatic reserve were common between the present NASH and CLD-C patients. However, other essential trace elements involved in hepatic steatosis and insulin resistance differed between the NASH and CLD-C patients, although these levels were approximately equivalent in the two groups. To the best of our knowledge, this is the first study to compare the involvement of these essential trace elements in the pathogenesis of NASH with that of CLD-C.
It is well established that iron deposition in the liver can initiate reactive oxygen species (ROS) and subsequently lead to hepatic inflammation in both NASH and CLD-C patients . The Attenuation of iron overload by phlebotomy thus resulted in the improvement of serum transaminase levels in such patients . Our present results have confirmed the strong correlation between serum ALT and ferritin levels in both NASH and CLD-C patients.
We did not observe a correlation between serum ALT and Zn levels in either NASH or CLD-C patients. However, we have confirmed a favorable effect of polaprezinc, a complex of Zn and L-carnosine, on the serum ALT levels in patients with CLD-C through an anti-oxidant action . Likewise, supplementation with zinc gluconate resulted in the improvement of ALT levels in patients with NAFLD . In contrast, Wang et al. documented a close correlation between serum Se and ALT levels in adult NAFLD patients with Se concentration over 130µg/L , which indicated that a higher circulating Se concentration was harmful in the liver. Our present findings did not support their results.
Hypoalbuminemia derived from an unfavorable hepatic reserve results in a relative increase in α2 macroglobulin, which more strongly binds to Zn, eventually causing a substantial increase in the urinary excretion of Zn . Serum Zn levels thus seem to be associated with serum Alb levels in both NASH and CLD-C patients.
A decline in serum Se levels is frequently observed in patients with decompensated liver cirrhosis [33, 34]. However, lower serum Se levels do not indicate Se deficiency, rather, they indicate an unfavorable hepatic reserve in these patients . Therefore, Se supplementation does not lead to the improvement of hepatic reserve in patients with decompensated liver cirrhosis. Our present study also revealed the positive correlation between serum Se and Alb in NASH and CLD-C patients, suggesting that lower serum Se levels reflect an unfavorable hepatic reserve in these patients.
Our earlier investigation revealed a close correlation between serum ferritin levels and the HOMA-IR values in patients with CLD-C . The present study showed a positive correlation between serum ferritin levels and the HOMA-IR values in patients with NASH. We thus speculated that a very similar mechanism that observed in patients with CLD-C may be exerted in patients with NASH. It is well established that the hepatic extraction and metabolism of insulin are decreased as the deposition of iron in the liver becomes more severe, leading to hyperinsulinemia .
We have confirmed that serum Zn levels were inversely correlated with the HOMA-IR in patients with CLD-C [8–10]. Zn has been demonstrated to play a crucial role in the stabilization of the insulin-like growth factor-1 (IGF-1) transcript . Zn deficiency led to an impairment of IGF-1 synthesis and subsequently an increase in insulin release from beta cells in patients with CLD-C . Our present analyses also elucidated an inverse correlation between serum Zn levels and the HOMA-IR values in patients with NASH, which supports the previous findings .
We suggest that the putative mechanism by which Zn deficiency causes insulin resistance in patients with NASH is probably the same as that in patients with CHD-C. Indeed, several studies revealed a decrease in IGF-1 release in NASH patients . We also confirmed that NASH patients had a significant correlation between serum Zn and IGF-1 levels (r = 0.762, p = 0.0171, n = 9, data not shown in the text).
We previously revealed an inverse correlation between serum Se concentrations and the HOMA-IR values of patients with CLD-C . Lower serum Se levels may impair the activation of mitogen-activated protein kinase, leading to insulin resistance in such patients . Unexpectedly, our present study did not show a significant correlation between serum Se levels and HOMA-IR values in patients with NASH. It is of interest that the high prevalence of type 2 diabetes mellitus (T2DM) was observed in individuals with relatively high levels of serum Se or relatively low Se levels , suggesting that serum Se levels were not correlated with the HOMA-IR values in those individuals. This result may explain the reason why the inverse correlation was not found between the serum Se concentrations and the HOMA-IR values in the present study’s NASH patients.
Other studies have focused on selenoprotein P (SeP) in patients with NASH. SeP is an extracellular glycoprotein that is synthesized in primarily hepatocytes, and it acts as a Se-transporter from the liver to other organs. SeP also seems to play an important role in insulin resistance . However, circulating SeP levels in patients with NASH are conflicting: some studies revealed that serum SeP levels were higher in NASH patients than controls and patients with nonalcoholic fatty liver (NAFL) , whereas other studies documented lower SeP levels in NASH patients than in controls and patients with NAFL .
The grade of hepatic steatosis turned out to be enhanced more intensively as the serum Zn levels were decreased in patients with CLD-C . Lower serum Zn levels may cause the inactivation of peroxisome proliferator-activated receptor-α (PPAR-α), and subsequent facilitation of lipid peroxidation in such patients . Nonetheless, the reason why the negative correlation was not found between the grade of hepatic steatosis and serum Zn levels in the NASH patients remains uncertain. It is of interest that administration of zinc gluconate did not result in the improvement of hepatic steatosis in NAFLD patients .
Our present study also showed that the severity of hepatic steatosis was increased in proportion to the serum Se levels in the patients with NASH. In db/db rats (an animal model of T2DM), long-term supplementation with selenite resulted in the exacerbation of hepatic steatosis by reducing the antioxidant defense capacity . However, selenite supplementation improved hyperglycemia via an increase in insulin synthesis in the rats. Bonnefont-Rousselot et al. documented that serum Se levels were independent of the grade of hepatic steatosis in patients with NAFLD . In contrast, another study indicated that the administration of selenoneine, which contains Se on the imidazole ring, alleviated the degree of hepatic steatosis in an experimental animal model of NASH . As described above, the correlation of serum Se levels with the degrees of hepatic steatosis remains controversial. Further investigations are required to clarify the role of Se in the pathogenesis of hepatic steatosis.
It is well established that a decrease in the activity of collagenase derived from Zn deficiency results in the progression to more advanced hepatic fibrosis . The stage of hepatic fibrosis thus became more severe as the serum Zn levels gradually decreased in the present study’s NASH and CLD-C patients.
A recent report revealed that supplementation with Se inhibited the procollagen synthesis in the animal model of hepatic fibrosis via a decrease in oxidative stress and subsequent reduced collagen formation and an enhancement of collagen degradation . These results may indicate that a decline in serum Se levels lead to the development of hepatic fibrosis in NASH patients  and CLD-C patients . Concerning the serum Se concentration in the present NASH patients, the serum Se levels were increased in proportion to the degree of hepatic steatosis, and they decreased in parallel with the degree of hepatic fibrosis. These results may indicate that lower serum Se levels contribute to the progression from NAFL to NASH.
We observed that serum Cu levels were also increased as the stage of hepatic fibrosis was more advanced in the CLD-C patients, whereas these levels did not affect the stage of hepatic fibrosis in the NASH patients at all. Cu is likely to play a pivotal role in hepatic fibrosis as a cofactor . Nobil et al. revealed that serum Cu levels in NAFLD patients with more severe activity were significantly lower than those without activity . They speculated that lower serum ceruloplasmin levels might cause a high susceptibility to oxidative stress in the hepatocytes, eventually leading to lower circulating Cu concentrations in NAFLD patients with severe activity. Aigner et al. also observed lower hepatic Cu content in NASH patients compared to that in NAFL patients . They suggested that lower Cu availability might contribute to the development of NAFLD.
Serum ferritin levels might predict the severity of hepatic fibrosis in NASH patients . Excessive iron may activate hepatic stellate cells by increasing α-smooth muscle actin (α-SMA), collagen, and transforming growth factor-β (TGF-β) . However, we could not confirm this finding in the present NASH and CLD-C patients.
The correlation of each trace element with other trace elements is another aspect of this study. Serum Zn levels tended to be correlated with serum Se levels in both the NASH and CLD-C patients, although we previously elucidated a close correlation between serum Zn and Se in patients with CLD-C . An inverse correlation was not apparent between the serum Zn and ferritin levels in the NASH patients. Whereas, a significant correlation was present between the serum Se and Cu levels in only the NASH patients, despite the fact that the reason remains unclear. These results may indicate that the associations of each trace element with other trace elements were extremely distinct between the NASH and CLD-C groups except for the correlation between Zn and Se.
There were several limitations to consider in this study. First, the sample size was small, especially in the NASH group. Some of the results obtained by this study did not reach the statistical significance for this reason. NASH patients had a tendency to be female predominant for the small sample size, although the distribution was not statistically significant. A large scaled study is required to confirm whether the same results as this study are provided. Second, other factors such as dietary intake of these essential trace elements were not evaluated at all in this study. Therefore, the results might be biased. Dietary assessments by the use of three day diet diary are necessary in the enrolled patients.
In summary, some essential trace elements commonly contributed to hepatic inflammation, the hepatic reserve, and hepatic fibrosis in NASH and CLD-C patients. Other essential trace elements involved in the pathogenesis of hepatic steatosis and insulin resistance were distinct between these two groups. Further studies are necessary to demonstrate the putative mechanisms by which the disorders of these essential trace elements metabolism evoke hepatic steatosis and insulin resistance in patients with NASH.