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

Immaturity of microvessels in haemorrhagic plaques is associated with proteolytic degradation of angiogenic factors

Interview with Dr. Jean-Baptiste Michel, MD, PhD.
Inserm Unit 698, CardioVascular Remodeling
CHU Xavier Bichat, 46 rue Henri Huchard
75018 Paris, France.
Basic Sciences, Pharmacology, Genomics and Cardiovascular Pathology

Q: Dr. Michel, your group published a very interesting paper on plaque neovascularization, a topic that has gained momentum in atherosclerosis research over the past years. What are the key important new findings you obtained?

A: Our study suggests that intraplaque hemorrhage is a self-perpetuating vicious circle that accelerates plaque destabilization and evolution towards rupture. In fact, we showed that in addition to the probable direct induction of vascular damages, hemorrhage-associated blood-borne proteolytic activities promote the degradation of vascular stabilizing factors including various angiogenic factors, an effect that is likely to destabilize plaque neovessels and to prevent their maturation.
- Leclercq A, Houard X, Loyau X, Philippe M, Sebbag U, Meilhac O, Michel JB. Topology of protease activities reflects atherothrombotic plaque complexity. Atherosclerosis 2007 Mar;191(1):1-10.
- Leclercq A, Houard X, Philippe M, Ollivier V, Sebbag U, Meilhac O, Michel JB. Involvement of intraplaque hemorrhage in atherothrombosis evolution via neutrophil protease enrichment.J Leukoc Biol. 2007 Dec;82(6):1420-9.

Q: Your focus was the carotid artery plaque as an endarterectomy sample with two areas of interest: the stenosing and the adjacent segment. Studying a disease spectrum, how diseased were the adjacent segments, i.e. what was the relative contribution of Stary type II and III lesions? What was the microvessel density and prevalence of hemorrhage in these adjacent plaques and was this significantly different from the stenosis areas? 

A: In accordance with the classification established by Stary, histologic analysis revealed that non-stenosing segments included fatty streaks (type II lesions) and fibrolipid lesions (type III and IV). The majority of fatty streaks were devoid of microvessels. Nethertheless, atherosclerosis-associated angiogenesis is an early event that could be detected in some early fibrolipid lesions. Microvessel density within fibrolipid lesions was decreased when compared to stenosing non-hemorrhagic and hemorrhagic lesions. Hemorrhage was not found in fatty streaks or fibrolipid lesions. Erythrodiapedesis could be seen sometimes in fibrolipidic lesions, in relation to neo-capillaries.

Q:  All stenosis areas had an intact fibrous cap but were of two subtypes: hemorrhagic and non-hemorrhagic plaques. How did you define micro- and macro-hemorrhage and how often did you encounter it? What methodology did you use to distinguish old from recent hemorrhage and how often was either type seen?

A: Macro-hemorrhage was easily detected during macroscopic examination of the sample and corresponded to coagulated large blood lake areas. Micro-hemorrhage was detected upon microscopic observation and corresponded to small areas of extravasated red blood cells in close proximity of microvessels. Recent hemorrhage was identified by the presence of intact extravasated red blood cells and was quantified by colorimetric assay for heme while older hemorrhage was detected by positive Prussian blue staining, which reveals hemosiderin, a degradation product of phagocytosed hemoglobin. In carotid endarterectomy samples, old or recent intraplaque hemorrhages (or mix of old and recent) were very frequent (70-80% of culprit lesions), associated or not with calcifications.

Q: Was there a difference between hemorrhagic and non-hemorrhagic plaques with regards to microvessel density or structure?  

A: Yes. Microvessels density was increased in hemorrhagic plaques as compared to non-hemorrhagic plaques. Also, the majority of these microvessels were lacking -smooth muscle actin-positive cells in their wall. The poor mural cell coverage of hemorrhagic plaque microvessels indicates that those vessels are structurally immature and fragile. Another important difference between non-hemorrhagic and hemorrhagic plaques was their leukocyte content that was increased in hemorrhagic plaques. Plaque-infiltrating leukocytes are also likely to favor intraplaque hemorrhage through the induction of vascular injuries and proteolysis of vascular-stabilizing factors such as angiopoietin-1.

Q: Certainly a fascinating aspect is the fragility of plaque neovessels. It is fascinating as we more or less though of it as a general feature of all neovessels in the very plaque area even though conventional views also taught us that these neovessels primarily sprout as a an organized process from the adventitial vasa vasorum network through the media into the plaque where they lose maturity rather than being formed in situ. Your data now suggest a change in direction, would you mind reflecting on it more closely?

A: In our study, we reported that the majority of microvessels in hemorrhagic plaques were lacking -smooth muscle actin-positive cells in their wall as compared to vessels from non-hemorrhagic plaques. However, the coverage of vessels with -actin-positive cells in non-hemorrhagic plaques is also far from being normal and would be reduced in terms of thickness/density if compared to adventitial blood vessels. Thus, we do think that fragility is an inherent and general feature of neovessels in both non-hemorrhagic and hemorrhagic lesions. Our data suggest that in hemorrhagic plaques, the proteolytic degradation of angiogenic growth factors is likely to destabilize plaque vessels and to further decrease the potential recruitment of mural cells.
Regarding the origin of plaque vessels, based on our analysis of early aortic atherosclerotic lesions in an ongoing study, we agree that plaque neovessels originate from the adventitial vasa vasorum network through the media. In fact, in aortas bearing early atherosclerotic lesions (fatty streaks and fibrolipid lesions), we observed that microvessels were rare in fatty streaks and more prominent in more advanced fibrolipid lesions. However, even when lesions were devoid of neovessels, the microvessel density was systematically increased in medias subjacent to both fatty streak and fibrolipid lesions compared to media from healthy aortas. Furthermore, medial microvessels extended much deeper towards the intima in atherosclerotic aortas than in healthy aortas. This ongoing study supports the idea that the growth of a fragile vascular network is a prominent feature of the earliest stages of atherosclerosis that is initiated in the media subjacent to lesions. In this context the proteolytic environment could be the major determinant of neo-vessel fragility.

Q: Just to focus in on it, are plaque neovessels formed very maturely initially and then become fragile leading to the hemorrhagic plaque phenotype or do plaque neovessels in hemorrhagic plaques never reach a mature and secure state?

A: As mentioned above, we consider that fragility may be an inherent feature of atherosclerosis-associated neovessels. To our opinion, this fragile structure may directly relate to the function of these vessels. If one considers that the primary function of plaque nevessels is to convey leukocytes within the lesion in order to eliminate subintimal lipid deposits, then loose endothelial junctions, sluggish blood flow and poor mural cell coverage appear as real assets for leukocyte extravasation.
However, immaturity of plaque microvessels may not be sufficient to induce intraplaque hemorrhage. Extravasation of red blood cells does not occur only through the endothelium but also through the basement membrane, which implies degradation, rupture or abnormalities of the basal lamina surrounding vessels. It is likely that in addition to neovessel fragility, factors capable of degrading the basal lamina are required to lead to the hemorrhagic plaque phenotype. Our data suggest that blood-borne proteases and leukocytes within plaques also play a crucial role in the induction of intraplaque hemorrhage by further destabilizing plaque neovessels.  

Q: In your paper you relate three processes to neovessel fragility: imbalance of angiogenic factors, proteolytic activity, and reduced smooth muscle cell migration. How are they at work, especially in the bigger picture of the previous question?

A: It is difficult to establish a hierarchy between these intertwined phenomena. The main message is that a serine protease environment constitutes a major impediment to the healing process in atherothrombotic disease. In an earlier study we have demonstrated for example, that blood-borne neutrophil elastase is able to inhibit tissue healing by mesenchymal cells.
- Fontaine V, Touat Z, Mtairag M, Vranckx R, Louedec L, Houard X, Andréassian B, Sebbag U, Palombi T, Jacob MP, Meilhac O, Michel JB. Role of leukocyte elastase in preventing cellular recolonization of the mural trombus. Am. J. Pathol. 2004; 164: 2077-2087.

Q: Regarding angiogenic factors, as mentioned before, prevailing theory is that most plaque neovessels stem from the advential vasa vasorum network and sprout into the plaque area, which requires a certain level of pro-angiogenic factors. If they is true, would we have to assume then that hemorrhagic plaques do not start out with a lower level of pro-angiogenic factors including VEGF? Furthermore, does the neovascularization process eventually slow down in the hemorrhagic plaques due to the change in the milieu of angiogenic factors and do these plaques “burn out” then?

A: As you suggested, hemorrhagic plaques probably do not start out with lower levels of pro-angiogenic factors but rather with high levels of pro-angiogenic factors resulting in an increased microvessel density. The intensity of the initial angiogenic response in plaques remains an open question and may depend on plaque composition (oxidation level and type of the subendothelial lipids deposits) and subsequent responses of stromal organizer cells. We also agree with your idea that the angiogenic process eventually slows down in hemorrhagic plaques due to the degradation of angiogenic factors at later stages. Angiogenic factors do not only regulate vessel growth but also their stability, leakage and survival.

Q: Regarding proteolytic activity, which proteases are involved, what is their source, what are their targets and what determines their activity? Also, an important concept seems to be that these proteases do not digest the vessel causing fragility but rather the pro-angiogenic factors – is this correct?

A:  We showed that plasmin and elastase activities were increased in hemorrhagic plaques. Plasmin formation results from the activation of plasminogen by its activators, tPA and uPA in presence of fibrin. The plasminogen leaking within plaques through immature vessels and during hemorrhage can be activated by the tPA and uPA present on the plasma membrane of smooth muscle cells and plaque-infiltrating leukocytes. Also, potential intraplaque fibrin deposits also provide a proper surface for plasminogen, Thus, intraplaque plasmin activity depends on the availability of blood-borne plasminogen, levels of plasminogen activators and inhibitors, and fibrin and cell content within plaques.
Elastase derives from neutrophils and its intraplaque activity is directly linked to the occurrence of hemorrhage that conveys neutrophils. As cited above, in an earlier study on intraluminal thrombus of abdominal aortic aneurysm, we demonstrated that elastase released by thrombus-trapped neutrophils impeded mesenchymal cell healing.
To our opinion, proteolytic activities within hemorrhagic plaques contribute to destabilization of plaque vessels through both direct induction of vascular injuries and degradation of vascular-stabilizing angiogenic/growth factors. Other proteases such as matrix metalloproteases and cathepsins probably also contribute to these processes.

Q: Regarding and returning to the initial observation of a lack of neovessel coverage by alpha-smooth muscle cell actin-positive cells, you found very intriguing insight from experiments on conditioned medium from the various plaque types which influences smooth muscle cell migration, adhesion, and survival in a very distinct ways. At the end, is this all a reflection of the change in the milieu of angiogenic factors or cytokines and is this the consequence of protease activity or change in the expression profile of these factors?     

A: The impediments to cell healing process due to proteases are numerous, including degradation of growth factor but also of cell adhesion molecules and of extracellular matrix components. We have demonstrated earlier that leukocyte elastase (ref) and pericellular plasminogen activation (ref) were able to provoke cell detachment of apoptosis. Thus, an excess of proteolytic activities can impair the healing process through direct and indirect mechanisms and most likely contributes to the decreased ability of conditioned medium from hemorrhagic plaques to stimulate smooth muscle cell migration and spreading. However, changes in the expression profile of angiogenic factors and cytokines in hemorrhagic plaques may also contribute to this phenomenon and cannot be excluded.

- Mtairag M, Houard X, Rais S, Pasquier C, Oudghiri M, Jacob MP, Meilhac O, Michel JB. Pharmacological potentiation of natriuretic peptide limits polymorphonuclear neutrophil-vascular cell interactions. Arterioscl. Tromb. Vasc. Biol. 2002;  22: 1824-1831.

- Meilhac O, Ho-Tin-Noé B, Houard X, Philippe M, Michel JB. Pericellular plasmin induces smooth muscle anoikis. FASEB J. 2003; 17: 1301-1303.

- Rossignol P, Ho-Ti-Noé B, Vranckx R, Bouton MC, Meilhac O, Lijnen HR, Guillin MC, Michel JB, Angles-Cano E. protease-nexin-1 inhibits plasminogen activation-induced apoptosis of adherent cells. J. Biol. Chem. 2004; 279: 10346-10356.

- Michel JB. Anoikis in the cardiovascular system, known and unknown extracellular mediators. Arterioscler. Thromb. Vasc. Biol. 2003; 23: 2155-2163.

Q: A final and in a way broader look at the very underlying mechanisms of these changes. As not all advanced plaques develop a hemorrhagic phenotype, we cannot conclude that this is a generic phenomenon of natural disease progression. Why then do some plaques develop a hemorrhagic phenotype? Is it all in the neovasculature? Even if it is, what are the driving factors and are these transient or persistent? Is the biological activity of the plaque, namely inflammation and oxidative stress, a central influencing factor?

A: We know from observational studies that plaque angiogenesis plays a central role in the development of a hemorrhagic phenotype. High microvessel density is indeed one of the main features of hemorrhagic lesions. However, we think that microvessel density and fragility may not be sufficient to induce plaque hemorrhage. Additional factors such as infiltration and activation of leukocytes within plaques may also play a crucial role in the induction of intraplaque hemorrhage. Thus, any factor regulating plaque angiogenesis and/or leukocyte recruitment and activity may potentially be a central influencing factor of plaque evolution. There are data showing that oxidized lipids are able to induce angiogenic signals in vivo. To us, this underlies the fact that as you suggest, the biological activity of the plaque is a main influencing factor of plaque evolution towards hemorrhage or not.

Q: What are the clinical implications of your findings? How should they influence our care for patients with atherosclerosis?

A: Our data reinforce the concept that intraplaque hemorrhage is one of the main determinants of vulnerability leading to clinical expression as recently demonstrated by MRI in patients. We propose an explanation of the phenomenon due to a detrimental blood-borne proteolytic environment within the core of the lesion.

Q: How should the current findings and in fact the entire topic be taken further? What do you see as a future direction?

A: New paradigms are now emerging in the understanding of human atherothrombotic diseases. Identification of the mechanisms that trigger atherothrombotic angiogenesis and a better understanding of the mechanisms regulating neovessel stability appear as crucial future directions for the prevention of intraplaque hemorrhage. What is important is to develop observational studies on human samples without a priori assumptions. Discoveries always come from deeper and realistic observations.



Q: Thank you so much for this interview, Dr. Michel, and the privilege to discuss your outstanding work in greater detail. 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:  Thank you and of course, do not hesitate to contact me if you have any other questions.

Benoit Ho-Tin-Noé, PhD
Julien Le Dall, MSci.
Jean-Baptiste Michel, MD, PhD.


Cardiovasc Res. 2010 Jan 1;85(1):184-93.
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