Cu(II) and advanced glycation end products

A travel back in time [1,2]

We have known that Cu(II) is potentially toxic for a long time. The year is 2002. We have only recently appreciated the role of nitric oxide in relaxing our blood vessels. The Nobel Prize in medicine for this discovery was awarded in 1998.

1. Glycated albumin binds three times as much Cu(II) and
2. Copper bound to glycated albumin remains redox active. [1]
3. Cu(II) glycated albumin causes the degradation of nitric oxide. [1] So what? We’ve covered Fenton/Haber-Weiss chemistry on this site. NADPH oxidase, ACE inhibitors for high blood pressure, the opposite of NO relaxing our blood vessels…
4. Plasma copper of diabetic rats is approximately twice that in normal rat plasma. [1]
5. The same is true for tail tendons of diabetic rats. [1]
6. Implants coated with glycated albumin can chelate 5x as much Cu(II) as implants coated with non glycated albumin. [1]

We’ve covered the so called Maillard reaction in the polyol post. The same reaction that leads to advanced glycation end products (AGE) also leads to the browning of bread and other food chemistry sorts of things. In 2004 food scientists in Germany compared binding of Cu(II) and Zn(II) to N(epsilon)-fructose-lysine and N(epsilon)-carboxymethyl-lysine. The regions in squares were chemically blocked so that XCu(II) and Zn(II) could only bind to the side chains. [2] Cu(II), but not Zn(II), bound to both of these compounds. [2]

The regions in the boxes are part of the peptide bond in proteins. The modified lysine side chains bind Cu(II) In vitro adducts of Cu(II) to glycated lysines was identified in 2014. [3] The following paper may be difficult for non biochemists to follow. Short lay summaries will be given after each set of figures because this is important.

1 Cu binding to collagen and albumin

0.4 mM bovine serum albumin (BSA) was reacted with 100 mM D-glucose in a pH 7.2 buffer under sterile conditions and nitrogen gas in capped vials at 37 °C for seven weeks. Solutions of 1 mM, 3 mM of copper (CuCl₂) or zinc(ZnCl₂) ions were added during the glycation process. The proteins were dialyzed against MES buffer, pH 7.2 and stored at -20 °C. AGE-related modifications were determined by measuring glycophore fluorescence at 425 nm under excitation at 345 nm.

• Panel A We are looking at a fluorescent probe that binds to glycosyl groups. Glycophore absorbs in the 345nm long UV region of the spectrum. When light of this wavelength is used to illuminate the sample
• Panel B We see an emission in the 425 nm blue region of the spectrum. With the 3 mM Cu(II) sample we are seeing a decrease in groups that react with glycophore.
• Panel C Mock 37^oC incubated samples have a fair decrease in the amount of free amino groups. What is binding to them, we do not know.
• Panel D For the case of BSA incubated with glucose and 3mM Cu(II) we are starting to suspect that Cu(II) might be blocking glycation..

The interactions between copper and glucose with this abundant blood protein are complicated.

2 Cu(II) changes protein structure

• Panel A When tryptophan is deep inside the core of a protein, it fluoresces in the long UV region of the spectrum when illuminated by shorter wavelength UV light. When the protein unfolds these tryptophans become solvent exposed and less fluorescent. All modified albumins underwent quenching of tryptophan fluorescence as a consequence of protein incubation with glucose. These data suggest that the protein is funfolding.
• Panel B When the tryptophan fluorescence was normalized to “normal” glycated BSA, a “blue shift” was seen in the maximal emission in the Cu(II) bound variety. This is considered indicative an the structure being different, perhaps due to a more exposed tryptophan.
• Panel C Analysis of momentum 1 (M1) relative to the tryptophan emission for different albumin samples was interpreted as indicative of the aggregation process. Aggregation was attributed to important conformational changes int the protein protein in the presence of copper(II) or zinc(II). A cartoon has been added of a protein unfolding, aggregating, and undergoing larger conformational changes.
• Panel D FTIR is used to measure secondary structure, i.e. alpha helices, beta sheets. Cartoons of alpha helices and beta sheets have been added. The idea is that Cu(II) is affecting global structure without affecting local structures.

The Cu(II) and Zn(II) glucose combination can cause this abundant blood protein to aggregate.

3 Cu(II) causes glycated albumin aggregation

This figure is a follow up to Figure 2c.

• Panel 3A  Rayleigh scattering is a means of monitoring the formation of larger particles.
• Panel 3B this is an SDS PAGE gel that sorts proteins by size and charge.  The charge is established by the charge of the amino acids and the size, which determines how much charged SDS detergent can coat it.  In this publication “native” probably means without the reducing agent.  BSA can form disulfide bonds with essentially double the apparent molecular weight.     
• Panel 3C Thioflavin T is a fluorescent probe that fluoresces when it interrelates into amyloid structures. A similar concept has been covered with Cu(II) induced IgG light chain amyloid AL post.

These aggregates also exhibit amyloid structures.

4 Cu(II) albumin aggregation part 2

Atomic force microscopy (AFM) is a type of scanning probe microscopy that uses a mechanical probe that touches or feels the surface with Piezoelectric elements that allow the touching to be recorded.  . The information is gathered by “feeling” or “touching” the surface with a mechanical probe. ATM is  more than 1000 times better than the optical diffraction limit.

These Cu(II) and Zn(II) aggregates are clearly different.

5 Loss of free thiols and gain of carbonyls

While BSA has only one free thiol, this thiol can form a disulfide bond with the other single free thiol of a second BSA. This is suggested in Panel 3B. Protein carbonylation is another modification that can cause a protein to unfold and aggregate.

We don’t know if the loss of free thiols in the presence of Cu(II) and Zn(II) is due to direct binding and competition for the assay reagents or if some of these thiols have been oxidized to sulfinic acid and related compounds. It is interesting to note that redox inactive Zn(II) also increases protein carbonyls

The 3 mM Cu(II) and glucose combination is the most potent inducer of BSA carbonyls.

6 In vitro toxicology

The trolox equivalent anti-oxidant capacity (TEAC) assay is very much a Cu(II) dependent assay that measures anti-oxidant capacity relative to Trollox, a vitamin E analog.

7 Glycated BSAs kill cells

A moused microglia cell line, BV2 cells, were used for all assays. The MTT assay , in simple terms, indicates cells with healthy amounts of NADH. Cell viability was also measured by surface characteristics (forward light scattering) using a technique called flow cytometry.

Recall that the unbound glucose and Cu(II) or Zn(II), as the case may be, have been removed from the BSA mixture. This modified BSA is greatly reducing cell viability.

8 Necrosis vs Apoptosis

Propidium iodine (PI) is a dye that becomes fluorescent when it binds to DNA. It can only get to the cell nucleus when the cell membrane has been damaged. PI is the red dye outside the scell waiting to enter holes in the membrane in the second image. High fluorescence is indicative of necrotic cells. Healthy cells should have little or no PI fluorescence, lower half. Annexin V is a protein that binds to phosphatidyl serine, a phospholipid that appears of cells undergoing apoptosis, programmed cell death. Healthy cells should appear in the lower left quandrant.

These data indicate that BSA treated with 3mM Cu(II) and Zn(II) are more toxic than just the 1mM treatment groups. It should be remembered that this is not the raw metal but BSA treated with the raw metal and glucose! 3 mM is really not that much.

9 Reactive oxygen species and IL-6

The rationale for looking at IL-6 is not that clear. IL-6 was not mentioned more than twice in the Kamalov study: once in the methods and once in the results saying that 3ZnBSA induced it. [3]

The confusing thing about use of IL-6 is that it is not a direct product of the inflammasome that we have discussed on this site. Some of these results are unexpected because Zn does not redox cycle like Cu, yet glycated BSA treated with Zn also causes the generation of ROS in a microglia cell line. These results are suggestive of an unfolded protein response that is speculated on this post, not necessarily by Kamalov and coauthors [3], to recruit macrophage.

Glycated albumin in the presence of Cu(II) and Zn(II) can induce cells to generate reactive oxygen species. A followup question(s) would be do Co(II), Mg(II), and Mn(II) have the same interaction with glycated albumin?

10 Cu sites on glycated BSA

Bovine serum albumin, BSA as we have been calling it, has several potential divalent cation binding sites that were discussed in the Kamalov publication. [3] These include a cysteine at position #34 in the N-terminus and several histidines in the metal binding site. The source of the images came from a publication on ischemia conditioned albumin.

This is bad news for diabetic cows. What about people?

A follow up study with human albumin

The followup study used human albumin and methylglyoxal as the lysine, arginine, and cysteine modifying small molecule. [4] In this publication the position was taken that about 10–15% of the total copper in blood is bound to albumin. The position was also taken that Cu(II) bound to albumin does not engage in Fenton chemistry. A technique called mass spectrometry was used to detect AGE in peptides from treated HSA. Cu(II ) at physiological and sub-physiological concentrations inhibited HSA glycation compared to Cu(II) free HSA. [4] At concentrations above 5 mg Cu(II) glycation was facilitated. [4]

Conclusion

Thank you for following us through this rather complicated study. While there is some evidence that copper may help those with T2D modulate their blood sugar, perhaps it would be better to avoid Cu(II) just in case the serum concentration exceeds that which is protective against albumin glycation. Too much of a Zn(II) supplement may also be bad. It is best to discuss these things with one’s healthcare provider.

References

1. Eaton J.W., Qian M. Interactions of copper with glycated proteins: possible involvement in the etiology of diabetic neuropathy. Mol Cell Biochem. 2002;234-235(1-2):135–142. [PubMed]
2. Seifert S.T., Krause R., Gloe K., Henle T. Metal complexation by the peptide-bound maillard reaction products N(epsilon)-fructoselysine and N(epsilon)-carboxymethyllysine. J Agric Food Chem. 2004;52(8):2347–2350. [PubMed]
3. Baraka-Vidot J, Navarra G, Leone M, Bourdon E, Militello V, Rondeau P. Deciphering metal-induced oxidative damages on glycated albumin structure and function. Biochim Biophys Acta. 2014 Jun;1840(6):1712-24. free article
4. Ramirez Segovia AS, Wrobel K, Acevedo Aguilar FJ, Corrales Escobosa AR, Wrobel K. Effect of Cu(ii) on in vitro glycation of human serum albumin by methylglyoxal: a LC-MS-based proteomic approach. Metallomics. 2017 Feb 22;9(2):132-140. [PubMed]