Close this search box.

Metalloproteomic analysis of liver proteins isolated from broilers fed with different sources and levels of copper and manganese – Scientific Reports

Cu and Mn concentrations in liver tissue and protein pellet

The concentrations of Cu and Mn in the liver tissue and protein pellet of the high Cu/Mn SO4 and high Cu/Mn (OH)Cl groups were higher than those of the low Cu/Mn SO4 and low Cu/Mn (OH)Cl groups, probably due to the greater supplementation of these minerals. Kim and Kil10 also observed an increase in Cu concentrations in the liver of broiler chickens with increasing inclusion of CuSO4 and tribasic copper chloride (TBCC) in the diets. Similarly, Sun et al11. reported higher Mn concentrations in various broiler tissues as the supplemental Mn hydroxychloride level was increased.

Comparing the supplementation sources, it was observed that Cu hydroxychloride proved to be more bioavailable than Cu sulfate, since its use resulted in higher concentrations of Cu in the hepatic tissue at the level of 15 ppm and in the protein pellet at both levels (15 and 150 ppm). Although the Mn concentrations did not differ between the sources in the liver tissue samples, in the protein pellet, there was a higher concentration of Mn in the group supplemented with Mn hydroxychloride at a level of 80 ppm compared to the sulfate source with the same level. In this sense, the analysis of mineral concentration in the protein pellet proved to be more sensitive to estimate bioavailability in relation to the concentration in tissue samples. These results suggest that in the present study, there was a greater incorporation of Cu and Mn in the liver proteins of broiler chickens supplemented with hydroxychlorides. In fact, the analysis by GFAAS indicated a greater number of protein spots with the presence of Cu and Mn in the gels of the high Cu/Mn (OH)Cl group compared to the high Cu/Mn SO4 group, which reinforces the hypothesis that hydroxychlorides may be more efficient in making absorbed minerals available for use by proteins in relation to sulfates.

Cu and Mn mapping in protein spots

Analysis by GFAAS indicated the unique presence of Cu in 19 spots. When analyzing the proteins expressed in these spots, it was observed that 5 spots showed the expression of the copper-binding protein albumin (spots 02, 06, 20, 21 and 32), and 1 spot showed the expression of the metalloenzyme of Cu and Zn, superoxide dismutase [Cu–Zn] (spot 17). Other proteins known to bind to copper, such as ATOX1 copper chaperones and CCs, were not identified, possibly due to the limitations presented by the 2D-PAGE technique in resolving low-abundance proteins and the dye used to reveal the protein spots. According to Smith et al12., low abundance proteins are hardly detected by Coomassie Blue staining, whose limit is between 50 and 100 ng of protein.

On the other hand, in 13 spots with the sole presence of Cu and 5 spots with the concomitant presence of Cu and Mn, the expression of proteins was observed that, as reported in the literature thus far, do not depend exclusively on Cu to perform their functions. However, some expressed proteins are metalloproteins activated by divalent metals, such as enolases (ENO1, ENO2 and ENO3) and alcohol dehydrogenase 1 (ADH1). Enolases are metalloenzymes that participate in the glycolytic pathway and are preferentially activated by Mg2+. ADH1 is part of a class of Zn2+-dependent enzymes responsible for the oxidation and reduction of a wide variety of alcohols and aldehydes13. These enzymes are extremely abundant in the liver, and in the present study, they were identified in several protein spots that showed divergent abundance. For this reason, it was not possible to evaluate the difference in the expression of these proteins between the groups studied using the 2D-PAGE technique. However, the presence of Cu in the spots where enolases (spots 02, 03, 10, 11, 14 and 15) and ADH1 (spots 01, 05, 28 and 19) were identified may indicate a possible replacement of Mg2+ and Zn2+ by Cu2+ when supplemented at levels higher than recommended.

Curiously, in 7 spots (spots 07, 12, 13, 18, 23, 27 and 29) with the sole presence of Cu and 2 spots (spots 25 and 34) with the simultaneous presence of Cu and Mn, proteins that normally do not are known to exhibit metal binding. These include proteins involved in carbohydrate metabolism (GAPDH, ALDOB, MDH1), retinol metabolism (ALDH1A1, ALDH1A2), protein folding and refolding (HSP90B1, HAP90AB1, HSPD1, HSPA9, HSPA5, HSPA8, Heat shock 70 kDa protein), detoxification (P4HB, GSTM2, Glutathione-s-transferase), and other processes (GLUD1, GOT2, RPLP0). This may have occurred due to the affinity of Cu+ (reduced state) for thiol and thioether groups found in cysteine and methionine residues and of Cu2+ (oxidized state) for oxygen groups found in aspartic and glutamic acid or imidazole nitrogen in residues of histidine14. This property enables the interaction of copper with a wide range of proteins, establishing their functions and structural states, in addition to causing harmful effects in cases of excess metal15.

Smith et al12. performed a study using metal affinity chromatography (IMAC), 2D-PAGE and mass spectrometry to identify human hepatocellular proteins with copper binding capacity. The authors identified 19 microsomal proteins and 48 cytosolic proteins with copper binding capacity. Among them are some proteins homologous to those reported in this study, such as protein disulfide isomerase, glyceraldehyde-3-phosphate dehydrogenase, heat shock protein 60 kD, heat shock cognate 71 kD protein, endoplasmic reticulum chaperone BiP and aspartate aminotransferase. In addition to these proteins, proteins commonly known to bind to metals were detected, such as albumin (Cu), alcohol dehydrogenase (Zn), α-enolase (Mg) and annexin V (Ca), which were also identified in our study. According to the authors, it is unlikely that all identified proteins bind to copper under normal physiological conditions; however, some proteins may be targets for copper under conditions of elevated levels.

Manganese was identified in only 10 spots and in 5 spots simultaneously with Cu. In 3 spots (spots 04, 09 and 31) with the sole presence of Mn and 2 spots (14 and 15) with Cu and Mn together, the previously mentioned enolases were characterized, indicating that, similar to Cu2+, Mn2+ can replace Mg2+ in these proteins. Other studies have shown that Mn2+ can replace Mg2+ in active sites of many proteins16,17,18,19,20. This substitution can occur because Mn2+ is a hard Lewis acid, similar to Mg2+, which allows the inverse to also occur21. In spots 19 and 24, ADH1 was characterized, indicating that it is a protein with the potential to bind Mn, although Cu has shown more affinity due to the greater number of spots associated with this metal and the respective protein.

Regarding the known Mn-binding proteins, 5 proteins were identified, namely, regucalcin (spot 04), trifunctional purine biosynthetic protein adenosine-3 (spot 04), glutamine synthetase (spot 08 e 26), phosphoenolpyruvate carboxykinase [GTP], and mitochondrial (spot 26). Albumin was identified in 3 spots with the unique presence of Mn (spots 16, 22 and 26). Although albumin is best known for binding to physiological Cu2+ and Zn2+ and to toxic Ni2+ and Cd2+22, there are reports in the literature that Mn2+ has two binding sites in albumin, where the secondary binding of Mn2+ corresponds to the primary binding site of Zn2+23, which justifies the identification of albumin in spots with Mn.

Similar to the spots with Cu, in the spots associated solely with Mn, proteins that have no known direct relationship with the metal were also characterized, such as HSPA5, HSPA8, Heat shock 70 kDa protein, ALDH1A1, ALDH1A2, GAPDH, ALDOB, GLUD1, LDHB, TPI1 and ATIC (spots 09, 16, 22, 24, 30 and 33). Normally, Mn2+ forms relatively weak complexes with many ligands compared to Cu2+21, which was reflected in the smaller number of spots associated with Mn. However, the concentration of 120 ppm Mn in the diet may have favored its binding to different proteins and consequently its detection in protein spots.

Changes in protein regulation in response to different sources and elevated levels of Cu and Mn

In the vast majority of protein spots, more than one protein was characterized, and many proteins were identified in more than one spot. According to Zhang et al24., a protein can be identified in several spots distributed in different positions on the gel due to posttranslational modifications that cause changes in pI and molecular mass. This fact explains the identification of the same protein in multiple spots in the present study and suggests that these posttranslational modifications may be involved in the response to Cu and Mn supplementation above nutritional levels. However, this factor made it difficult to assess the regulation of most of the proteins (upregulated or downregulated) between the studied groups, since divergent abundance of the respective spots were found. For example, analysis of the KEGG pathways revealed that one of the altered pathways was the glycolysis and gluconeogenesis pathways. However, the wide distribution of the proteins involved in these pathways in the protein spots does not allow any inference about their relationship with the evaluated treatments. Thus, the discussion was based on proteins identified in the spots that showed consistent abundance.

From the results of the present study, it is not possible to assume that the levels of 150 ppm Cu and 120 ppm Mn caused toxic effects in the broilers. However, Cu and Mn supplementation above the nutritional recommendation altered the abundance of spots containing proteins involved in many metabolic pathways, indicating that a homeostatic imbalance may have occurred that triggered the activation of several mechanisms for restoring homeostasis.

The heat shock proteins (HSP90B1, HSP90AB1, HSPD1, HSPA9, HSPA5, HSPA8, Heat shock 70 kDa protein) known as HSPs were upregulated (spots 12, 13 and 16) in the group supplemented with Cu and Mn sulfate at levels higher (high Cu/Mn SO4) than the group supplemented according to dietary requirements (low Cu/Mn SO4). HSPs are involved in the stress response and are sensitive to various stressors, such as oxidants, toxins, toxic metals, and free radicals. These molecular chaperones play a key role in the correct folding of proteins, refolding of misfolded proteins, prevention of cytotoxic aggregates or elimination of damaged proteins during cellular stress25. In this sense, HSPs may have been induced to suppress the effects of possible oxidative stress generated by Cu and Mn supplementation above the nutritional recommendation, since high levels can trigger the formation of reactive oxygen species (e.g., Fenton’s reaction and Haber–Weiss) and alter the normal functioning of many proteins for which they have binding sites.

Glutathione-s-transferases (GSTs) are enzymes that have multiple functions and play a key role in protecting against oxidative stress, acting mainly in the detoxification of various compounds in the liver, such as xenobiotics, lipid peroxidation products and metal ions26,27,28,29. In the group supplemented with high levels of Cu and Mn from the hydroxychloride source, an upregulation of GST and its isoforms was observed in relation to the group supplemented with normal levels (spots 18 and 23). These results suggest that increased exposure to Cu and Mn induced the expression of GSTs, which in turn activated cytochrome P450 drug and xenobiotic metabolism pathways to control the harmful effects of mineral supplementation above requirements. Thus, considering the results of the present study, GSTs and HSPs seem to be involved in tolerance to Cu and Mn stress, proving to be potential candidates for biomarkers of high exposure of these metals in broiler chickens. Further studies are needed to investigate the sensitivity and response of these proteins to metals individually, as well as to different sources of mineral supplementation.

The results of the mapping of Cu and Mn in the protein spots of the high Cu/Mn SO4 and high Cu/Mn (OH)Cl groups demonstrated that the hydroxychlorides allowed the greater incorporation of these minerals in the proteins in relation to the sulfates. On the other hand, a lower abundance of most protein spots in the high Cu/Mn (OH)Cl group was observed when compared to the high Cu/Mn SO4 group. Considering that the levels of Cu and Mn were above the requirements in both groups and that the hydroxychlorides proved to be more bioavailable than the sulfates, it is likely that the greater availability of minerals by the hydroxychlorides caused changes in the proteins in which Cu and Mn bound and consequently altered the functioning of these proteins, resulting in lower expression when compared to sulfates. Thus, the effects of high levels of metals from more bioavailable sources, such as hydroxychlorides, become prominent in relation to less bioavailable sources.


In the present study, Cu and Mn hydroxychlorides showed greater bioavailability compared to sulphates, as indicated by the concentrations of these minerals in liver tissue samples, pellets and protein spots. In addition, supplementation above nutritional requirements induced the expression of heat shock proteins (HSPs) and detoxification proteins (GSTs), suggesting the involvement of these proteins in metal tolerance and stress. However, it is worth noting that these findings were obtained under controlled laboratory conditions. Therefore, further studies are needed taking into account the practical conditions of poultry production, since the bioavailability of minerals and protein expression can be influenced by a variety of factors, including environmental, dietary and genetic interactions. It is also recommended to study additional supplementary levels and combine different molecular techniques to obtain complementary data and a more comprehensive understanding of the results.