In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

We use cookies to optimise the design of this website and make continuous improvement. By continuing your visit, you consent to the use of cookies. Learn more

Red cells, hemoglobin, heme, iron, and atherogenesis.

Interview with Dr. Balla
Basic Sciences, Pharmacology, Genomics and Cardiovascular Pathology

Q: Dr. Balla, you and a number of collaborators obtained quite remarkable new data on the impact of red cell sequestration on atherosclerotic plaques. The study was published in ATVB in July. Would you mind summarizing the highlights of your study for us?

A: We investigated whether red cell infiltration of atheromatous lesions promotes the later stages of atherosclerosis. It is well established that hematomas are frequently formed either by fissures at the atherosclerotic lesion surface, or may begin within the lesions as hemorrhages from neovasculature that sprout from the vasa vasorum. Since we previously demonstrated that heme and hemoglobin can act as mediators of low-density lipoprotein oxidation and endothelial cell damage, and of all sites in the body, the atheromatous lesions are at greatest risk of exposure to heme, we raised the hypothesis that heme-iron may accumulate in atherosclerotic lesions by intrusion and lysis of erythrocytes. Liberated hemoglobin is oxidized to ferri (FeIII) hemoglobin (methemoglobin) known to release its heme moiety, and as a result the released heme provokes a feed-forward process of iron-driven plaque lipid oxidation and endothelial cell damage.

Examining the hemoglobin derived from disrupted plaques with hematomas we found that oxidation of ferro (FeII) hemoglobin intensively occurs generating ferrihemoglobin and via more extensive oxidation (FeIII/FeIV=O) ferrylhemoglobin. Hemoglobin derived from ruptured complicated lesions was mainly oxidized to ferrihemoglobin (51%) and hemichrome (29%). Protein oxidation marker, dityrosine, was accumulated in complicated lesions accompanied by the formation of crosslinked hemoglobin – a hallmark of ferrylhemoglobin – resulting in hemoglobin multimers.

Plaque material is known to contain lipid oxidation products including lipid hydroperoxide. We therefore tested whether atheroma lipids have hemolytic activity. Indeed, lipids of atheroma and ruptured complicated lesions caused significant lysis of red cells. Moreover, liberated hemoglobin underwent extensive oxidation generating ferrihemoglobin. In addition to ferrihemoglobin, these reactions generated ferrylhemoglobin, an unstable oxidized form of hemoglobin. This highly unstable oxidized form of hemoglobin (ferryl state, Fe IV) rapidly returns to the ferric (Fe III) state through protein electron-transfer in which the chain tyrosine 42 acts as a redox center, cycling between the tyrosine and the tyrosyl radical while delivering electrons to ferryl heme. Tyrosyl radicals can react with each other to generate dityrosine leading to inter- and intramolecular cross-linking and formation of hemoglobin multimers. As we chemically reduced the lipid hydroperoxide content of plaque lipids the lytic effect and oxidation of liberated hemoglobin was significantly lowered.

The importance of hemoglobin oxidation in disrupted plaques with hematomas is emphasized by our recent finding demonstrating that contrary to other forms of oxidized hemoglobin, ferrylhemoglobin acts as a potent pro-inflammatory agonist in endothelial cells, leading to the up-regulation of adhesion molecules that support the recruitment of macrophages into the vessel wall.

In the past two decades our laboratories revealed that heme and hemoglobin oxidized to ferrihemoglobin can threaten vascular endothelium via sensitizing cells to reactive oxygen and mediating oxidative modification of low-density lipoprotein. Drawing upon our previous observations we studied the effect of heme and hemoglobin on atheroma lipids. We found that oxidation of ferrohemoglobin to ferryl- and subsequently ferrihemoglobin by the plaque lipid materials destabilizes the heme moiety promoting its release from the globin and the free hydrophobic heme group readily enters atheroma lipids. Oxidative scission of the porphyrin ring of heme occurred leading to iron release and a feed-forward process of plaque lipid oxidation. Accordingly, lipids of ruptured plaques were highly oxidized and iron was accumulating. This, in turn, provoked a remarkable endothelial cell cytotoxicity.
Because haptoglobin stabilizes the binding of heme to globin and inhibits the heme release from hemoglobin, and hemopexin binds heme and prevents its oxidative scission, we studied their effect on the interactions between hemoglobin and atheroma lipids. These serum proteins, presenting at remarkably high concentrations in plasma, binds hemoglobin and heme with extraordinary avidity and promote their clearance.

The inhibition of heme release from globin by haptoglobin and sequestration of heme by hemopexin suppressed hemoglobin-, or heme-mediated oxidation of lipids of atheromatous lesions and attenuated endothelial cytotoxicity. These two plasma constituents might exhibit beneficial function against advanced plaque development.
In our earlier published studies, we revealed that as a defense against heme and oxidative stress, endothelial cells upregulate heme oxygenase-1 and ferritin. Heme oxygenase opens the porphyrin ring, producing biliverdin, carbon monoxide, and a most dangerous product—redox active iron. The latter can be effectively controlled by ferritin via sequestration and ferroxidase activity. These homeostatic adjustments have been shown to be effective in the protection of endothelium against the damaging effects of heme and oxidants; lack of adaptation in an iron-rich environment led to extensive endothelial damage in humans. Now we demonstrate in this study that atheroma lipids also induce heme oxygenase-1 in endothelial cells when exposed in sublethal doses, and the extent of induction is much more pronounced when lipids derives from ruptured complicated lesion. Such an induction can be mimicked by lipids of atheromatous lesions exposed to heme and hemoglobin. The degree of heme oxygenase-1 induction is partially a function of lipid hydroperoxide levels of the lipid.

Overall, our results support the concept that interior of advanced atheromatous lesions is a pro-oxidant environment in which the invading erythrocytes are lysed, hemoglobin is readily oxidized to ferri- and ferrylhemoglobin. In these interactions between hemoglobin and atheroma lipids, hemoglobin and released heme amplify lipid oxidation and subsequently provoke endothelial reactions such as upregulation of cytoprotective heme oxygenase-1 and ferritin or cytotoxicity to endothelium. The inhibition of heme release from hemoglobin by haptoglobin and capturing of heme by hemopexin suppress hemoglobin-mediated oxidation of lipids of atheromatous lesions and attenuate subsequent endothelial cell damage (Figure).

Figure legend: A model for red cell-mediated progression of atherogenesis. 1, Infiltration of the atheromatous lesion by red blood cells. 2, Erythrocyte lysis and liberation of ferrohemoglobin (Ferro-Hb) by lipids of atheroma. 3, Oxidation of ferrohemoglobin to ferrylhemoglobin (Ferryl-Hb) (a) and to ferrihemoglobin (Ferri-Hb) (b). 4, Release of heme from ferrihemoglobin. 5, Heme uptake by atheroma lipid. 6, Amplification of lipid oxidation in atheroma. 7, Damage and activation of endothelium induced by reactive lipid metabolites of atheroma. 8, Induction of HO-1 and ferritin by atheroma lipid and heme (with permission of Arterioscler Thromb Vasc Biol).

Q: As you started out, you obtained extracts from normal vessels and atherosclerotic plaques and incubated them with healthy donor red blood cells. How exactly did you extract the lipid content and how di you ascertain that it was the lipid and nothing else that made the difference?

A: Lipids of normal vessels, atheromatous and ruptured complicated lesions were extracted from tissue by chloroform-methanol (2:1 v/v). The organic phase was evaporated under N2 and the weight of the lipid extract was measured. The extract was redissolved in a small volume of chloroform and suspended in Hanks’ balanced salt solution to produce a suspension of 2 mg lipid/mL solvent. Lipid aggregates were dispersed by the evaporation of chloroform coupled with vigorous vortexing. Our samples were confirmed to be free of chloroform and methanol.

We also collected lipids from atheroma and complicated lesions by simply scraping the affected areas and did not use any solvent. Exposure of normal red cells to such lipid materials derived from lesions caused hemolysis and oxidation of liberated hemoglobin to ferri- and ferrylhemoglobin. As we followed the interactions between hemoglobin and such atheroma lipids, hemoglobin and heme also resulted in lipid oxidation and subsequently endothelial cell cytotoxicity. During the course of lipid oxidation we detected degradation of porphyrin ring.

Q: Quite remarkable is the fact that the red cell lysis potential of complicated atherosclerotic plaques is higher than that of non-complicated atheroma. What was the composition of these two types of lesions and what could have accounted for the difference?

A:Indeed, the red cell lysis potential of complicated atherosclerotic plaques is much higher than that of non-complicated atheroma. Moreover, oxidation of hemoglobin more intensively occurs in lipids of complicated lesions compared to that in lipids of non-complicated atheroma. Concentrations of lipid peroxidation products including conjugated dienes, lipid hydroperoxides and thiobarbituric acid-reactive substances were significantly higher in ruptured complicated lesions, as compared to atheromatous lesions, and iron was markedly elevated in the complicated lesions. Importantly, preincubation of lipid extract derived from atheroma and complicated lesion with reduced glutathione and glutathione peroxidase (which specifically reduced lipid hydroperoxide to alcohol) significantly lowered the lytic effect. We note that this inhibition was not absolute. Oxidation leads to formation of a wide range of biologically active products, and some of these, such as 7-oxysterols, have been reported to be cytotoxic. Previous chemical investigations of the gruel from advanced lesions revealed that it contains large amounts of organic soluble carbonyls, ceroid-like material and aldehydes that are also have cytotoxic effect.

Q: You were able to show that the hemoglobin released from the red blood cells undergoes oxidation. This could be reduced by the antioxidant system of glutathione/glutathione peroxidase, which also reduces red blood cell lysis in the first place – is that correct?   

A: That is correct. Ferrohemoglobin released from the red blood cells is oxidized to ferryl- and subsequently ferrihemoglobin by the same oxidized materials of plaques. Such an oxidation of hemoglobin must occur in vivo, since hemoglobin isolated from ruptured complicated lesions was mainly oxidized to ferri- and ferrylhemoglobin as well as to hemichrome. Protein oxidation marker, dityrosine, was present in complicated lesions accompanied by the formation of crosslinked hemoglobin leading to hemoglobin multimers. It has been described that hemoglobin is readily oxidized by oxidants such as hydrogen peroxide, and we previously observed that it can also be oxidized by reacting with lipid hydroperoxides associated with low-density lipoprotein. Thereby we have reasoned that a source of oxidants for hemoglobin oxidation is plaque lipids. Depletion of lipid hydroperoxides in plaque material reduced hemoglobin oxidation.

Q:  In the supplemental figures you showed very nicely that over time Heme content decreases and lipid peroxidation production increases in atherosclerotic plaques only. Please tell us more about it - is there a positive interaction, and if so, does the cycle start with lipid peroxidation of the red cell membranes? 

A:  When in the early 90s we described that heme mediates oxidative modification of low-density lipoprotein we observed the same phenomenon - after a rapid heme uptake during the oxidative reactions between heme and low-density lipoprotein, the heme ring (protoporphyrin IX) is degraded, with resultant release of free iron. Once heme is lodged within the plaque lipid, spontaneous oxidative reactions involving small amounts of preformed lipid hydroperoxides (or other oxidizing equivalent) will lead to oxidative scission of the porphyrin ring and release of heme iron. The cycle starts with lipid peroxidation of plaque lipid. Depletion of plaque lipid associated hydroperoxide or adding butylated hydroxytoluene prior to heme exposure prolonged the degradation of porphyrin ring.
We note that heme is associated with red cell membrane as well. Therefore it is likely that there are interactions between lipids of lesions and red cell membranes containing the inserted heme. The scission of the ring can be detected spectrophotometrically by the decrease in heme absorption at 412 nm. Both the destruction of the porphyrin ring and the release of ferrozine-trappable free iron are evidently involved in the oxidation of atherosclerotic plaque lipids. The importance of this degradation is emphasized by the fact that conjugated diene formation and accumulation of lipid hydroperoxides occur in parallel with the release of iron. Hemopexin must be critical in preventing heme-mediated alteration of atherosclerotic lesions because it inhibits both the oxidative modification of plaque lipid and the degradation of heme. This serum protein binds heme with extraordinary avidity (Kd less than 1 pmol/L) and is present at remarkably high concentrations in plasma (≈1g/L).

Q: You also assessed the effects on endothelial cells, looking at stress response markers including HO-1 activity and cytotoxicity. Which assays did you use and where these difficult to implement?

A:It was not an accident to choose heme oxygenase-1 and cytotoxicity as stress response markers. In 1991 we observed that heme and ferrihemoglobin after heme is taken up by the cells acutely sensitize endothelium to reactive oxygen such as superoxide anion, hydroperoxides, oxidized low-density lipoprotein and oxidants derived from leukocytes. On the contrary, if the cells are exposed to heme for prolonged period of time they convert from hypersusceptible to hyperresistance to oxidative damage.
The molecular basis of this protection or rather adaptation was shown to be the induction of heme oxygenase-1 and ferritin, and the ultimate cytoprotectant was identified to be H-ferritin via its ferroxidase activity. It is well established that both heme oxygenase-1 and ferritin are remarkably induced by heme and ferrihemoglobin in endothelium.
While oxidized low-density lipoprotein is as good inducer for hemoxigenase-1 as heme, it much weeker inducer for ferritin compared to heme. Lipids from atheromatous lesions also substantially enhanced heme oxygenase-1 mRNA (5 fold), and that was more pronounced when lipids were pretreated with heme or hemoglobin (30 and 15 fold increase, respectively). The induction of heme oxygenase-1 by plaque lipids was the measure of lipid peroxide content. We employed real time RT-PCR to follow heme oxygenase-1 mRNA expression. Western blot was used for heme oxygenase-1 protein expression, and the enzyme activity was measured using the bilirubin generation assay as in our previous studies.
The induction of heme oxygenase-1 at protein level was similar to that at mRNA level. Cytotoxicity was determined by 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyl-tetrazolium-bromide reduction.

Q: Would you mind specifying further if the toxic effect was mediated by lipids, modified lipids, heme-modified lipids or heme alone?

A:Atheroma lipid was slightly cytotoxic, and intriguingly heme or hemoglobin added to lipids dramatically enhanced the toxic effect on endothelial cells. Both hemopexin and haptoglobin reduced this cytotoxicity. Furthermore, inhibition of oxidative reactions by butylated hydroxytoluene almost prevented endothelial damage in spite of the fact that the lipid associated heme was not degraded.
Thus heme and hemoglobin provoked lipid oxidation and its products mediated the cytotoxic effect. Heme taken up by endothelium amplifies such an insult.

Q:  Any guess if endothelial cells in atherosclerotic plaque are more resistant or more vulnerable to this stress?

A: Elevated expressions of heme oxygenase-1 and ferritin were found by us and others in cells of human atherosclerotic lesions that reflect exposures of cells to heme and reactive lipid metabolites generated by heme within the lesions. These cells armed with the activated heme oxygenase-1/ferritin system live and survive in a hostile environment. My guess is that endothelial cells in atherosclerotic plaque are resistant to this stress.

Q: Despite the debate on antioxidants we cannot pass by without talking about. How did you address this question on your study?
A:Preventing the release of heme from ferrihemoglobin by haptoglobin or capturing the released heme by hemopexin inhibited lipid oxidation and reduced cytotoxicity. The chemical changes exerted by hemoglobin and heme on lipids of atheromatous lesions were attenuated by antioxidants and iron chelators such as butylated hydroxytoluene, α-tocopherol and the iron chelator deferoxamine. We recently published that fungal siderophors found in mold-ripened food products function as protective agents against oxidation of low-densitiy lipoprotein and atheroma lipids.

Q: With the nice conceptual framework that you developed how, would you say, can we modulate this sequence to stabilize atherosclerotic plaques? Would the best step be to actually take a step back and suppress neovascularization in the first place? 

A: Research on haptoglobin and haptoglobin receptor (CD163), hemopexin and its receptor, heme oxygenase/ferritin system and its products are opening novel approach in vascular biology. They act as antioxidants, but their functions are far beyond this feature. They are present in atherosclerotic plaques and their actions are very specific. Erythrocytes can indeed enter developing atherosclerotic lesions through neovasculature in the vasa vasorum underlying atherosclerotic plaques. Until now this event was not considered as a significant step in promoting atherogenesis.
Preventing the changes exerted by red blood cells, hemoglobin and heme on lipids and cells of atherosclerotic plaques by “suppression of neovascularization in the first place to stabilize atherosclerotic plaques” is a very novel thought.

Q: Is your group performing any work on reversing the sequence?

A: We investigate whether hydrophobic fungal siderophores – hexadentate trihydroxamates desferricoprogen, desferrichrome, desferrirubin, and desferrichrysin – are taken up by the gastrointestinal tract and function as anti-atherosclerotic metabolites in functional food.

Q: Finally, considering the vulnerable plaque and vulnerable patient topic, stabilization and destabilization, does your study hold potential to trace these developments by biomarkers or (novel) imaging tools?

A: Detection of hemoglobin oxidation markers, products of heme catabolism by heme oxygenase/ferritin system within vessels, alteration in the level of haptoglobin, haptoglobin receptor and hemopexin in blood are candidates for being biomarkers, although there must be major obstacles. These chemical and cell reactions are limited to tiny portion of the affected arteries lowering the sensitivity of markers. Other disease states also alter their plasma levels decreasing the specificity. Gradient echo sequence MRI is capable of detecting oxidation of hemoglobin in situ and hold potential to trace destabilization of plaques.



Q: Thank you so, so much again for this interview. It is always great, a real privilege, to discuss studies with you. As it has been our practice, in case someone has further discussion points, may we forward them via my E-mail ( or may they contact you directly at your E-mail address?

A: It is an honor for us to contribute to such a valuable discussion and we offer help and our knowledge in this field (

Notes to editor

Expert Comment from Dr. Renu Virmani, President and Medical Director CVPath, international Registry of Pathology in Gaithersburg, MD:

The above discussion provides a clearer understanding of how red bloods cells in atheromatous plaques undergo hemolysis and oxidation of the liberated hemoglobin, to heme and ferrihemoglobin which promote lipid oxidation and endothelial reactions which upregulation of heme oxygenase-1 and therefore prevent cytotoxicity to endothelial cells. Also, the toxicity of heme can be controlled by haptoglobin and sequesteration of heme by hemopexin can lead to suppression of oxidation of lipids thus protection of endothelial damage.
This type of understanding will lead to development of imaging techniques that will identify such lesions in patients and perhaps even provide treatment to prevent the harmful effects of hemorrhage.

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.

Contact us


European Society of Cardiology

Les Templiers
2035 Route des Colles
CS 80179 BIOT

06903Sophia Antipolis, FR