Pleuni Van Den Borne

The goal of this research was to better understand the process of arteriogenesis and discover new therapeutic targets involved. With this knowledge it is possible to develop new therapies for Peripheral Artery Disease (PAD). PAD is characterized by a reduction in blood flow to the lower extremities. About 10 to 20% of PAD patients develop critical limb ischemia with leg amputation as a consequence. According to the latest studies, 200 million patients suffer from PAD. Risk factors for PAD are similar to those of cardiovascular disease (CVD) in general and include smoking, diabetes, hypertension, unhealthy diet choices, lack of exercise et cetera. Although most arterial occlusions are currently removed by surgery, such as percutaneous intervention or bypass surgery, this is not always possible. For these patients, there are no further treatment options. New treatment options are being developed, but these mostly failed in clinical trials. There is a great need for new therapeutic options for these patients. A possible option is the stimulation of collateral artery growth (arteriogenesis) in order to restore blood flow to their lower extremities. In order to develop new treatments, there is a need to better understand the mechanism of arteriogenesis.

The main goal of this thesis is to provide new insights into arteriogenesis and to discover new therapeutic targets. In chapter 2 and 3 we studied chemokine (C-X-C motif) ligand 10 (CXCL10). CXCL10 gets upregulated in response to, for instance, interferon gamma production after Toll-like receptor (TLR) activation. CXCL10 is involved in chemotaxis of myeloid cells (MC) and proliferation of VSMCs, cells expressing the receptor CXCR3. In chapter 2, the functions of CXCL10 in CVD in both pre-clinical and clinical studies are highlighted and discussed. Atherosclerosis, aortic aneurysm formation, myocardial infarction (MI) and PAD share underlying mechanisms in which CXCL10 is involved. However, the exact function of CXCL10 in CVD and some studies showed contradictory results. Especially in PAD, the role of CXCL10 is not fully elucidated. Chapter 3 describes the role of CXCL10 during arteriogenesis in a murine hindlimb ischemia model. CXCL10 showed to be crucial during this process as observed by a reduced perfusion recovery in CXCL10-/- mice compared to wildtype mice. A discrepancy was observed between wildtype and CXCL10 -/- mice. One explanation for this discrepancy could be the difference between collateral vessel formation and growth (arteriogenesis) and capillary formation (angiogenesis).

CXCL10 was involved in collateral vessel formation and growth, as shown by the increase in collateral vessel number and size, which were hampered in CXCL10 -/- mice compared to wildtype mice. We sought for a mechanism for CXCL10 on collateral vessel formation and observed that CXCL10 was involved in VSMC migration. During arteriogenesis, both VSMC migration and proliferation are important aspects. We have demonstrated that CXCL10-/- mice showed reduced VSMC migration. In addition, CXCL10 was shown to be involved in the activation and migration of MCs. CXCL10 promotes MC recruitment and activation. The process of arteriogenesis is dependent on the expression of circulating cells and not on resident tissue cells, as shown by studies focusing on TLR signaling pathways. However, in this study CXCL10 expression on circulating and resident cells is equally important during arteriogenesis. Just as CXCL10, the CXCR3 receptor exists in different isoforms. It is important to keep in mind that the expression of CXCR3 between mice and humans is critically different. In humans, three isoforms exist, CXCR3-A, CXCR3-B and CXCR3-alt, while in mice only one isoform is known, expressed similar to the CXCR3-A receptor in humans. CXCR3-B is expressed by human endothelial cells and it functions as an anti-angiogenic receptor. These properties of CXCL10 only occur in humans and not in mice. This emphasizes that translating experimental data to a human setting is challenging and merits careful consideration.

The CD200-CD200R inhibitory pathway has earlier been described to inhibit TLR responses and is involved in downregulation of myeloid cell functions. The pathway gets activated after cell-cell interaction resulting in CD200 ligand-CD200 receptor (CD200R) binding. Activation of this pathway results in inhibition of cell function, for instance cytokine production. It is an important pathway to prevent excessive inflammation or tissue damage during autoimmune disease and is frequently studied in tumor growth. Within cardiovascular disease, the role of this axis has not been studied extensively. In chapter 4, we studied the CD200-CD200R inhibitory axis in arteriogenesis. We observed that CD200 -/- mice showed a reduced perfusion recovery compared to wildtype mice. Since the CD200-CD200R axis is involved in myeloid cell function, we did expect a change in monocyte or macrophage recruitment. However, no differences were observed in myeloid cell influx in vivo. Enhanced T lymphocyte responses were already observed in CD200 -/- mice in other studies, but the role of T lymphocytes in arteriogenesis is not as extensively described. Monocytes and macrophage functions are thought to have the largest contribution to this process. However, the interplay of monocytes/macrophages and T lymphocytes (T helper cells) is critically important during arteriogenesis. It is already known that this axis is not involved in normal myelopoiesis. In addition, no differences between wildtype and CD200 -/- were observed regarding activation state or subset presence. As shown in the CXCL10 -/- study, the role of VSMCs could also be important. Although CD200 is expressed by VSMCs, there is no evidence on the role of CD200-CD200R axis for VSMC proliferation or migration. Future research needs to be conducted to elucidate a clear mechanism of CD200-CD200R axis in arteriogenesis.

TLR4 signaling has been studied in arteriogenesis in the past. For other CVDs, such as atherosclerosis or MI, these pathways are well known. In addition, new therapeutic agents are being developed and tested for their ability to inhibit the disease progression and tissue damage. In chapter 5, a specific TLR4 inhibitor (TAK-242) was tested for its effect on arteriogenesis in a murine hindlimb ischemia model. TAK-242 was tested ex vivo on its ability to inhibit TLR4 activation in isolated murine and human leukocytes. We observed that TAK-242 was able to inhibit TLR4 responses ex vivo, both on the level of cell activation and cytokine production, as measured by NF-kB. NF-kB activation is known to be involved in arteriogenesis in mice and rabbits and is therefore also suitable as read-out for stimulation of arteriogenesis. To test the in vivo effect of TLR4 inhibition by TAK-242, mice were treated with the inhibitor. However, TLR4 inhibition did not show a significant effect on perfusion recovery. A limitation of this study is the lack of information regarding dose-dependency and timing of TAK-242 treatment. NF-kB activation was reduced after local injection, but this response occurred late after the onset of ischemia, which might have influenced the outcome. Compared to the results in TLR4 -/- mice to TLR4 inhibition in wildtype mice, the results obtained in this study could partly be explained by poor timing of TAK-242 treatment.

Currently, researchers are investigating the use of mesenchymal stem cells (MSCs) for treating hindlimb ischemia. In the past, we have demonstrated the protective effect of MSC-conditioned medium on reducing injury after MI and have shown that exosomes are the therapeutic element of the conditioned medium. Systemic exosome treatment resulted in reduced infarct size after ischemia/reperfusion injury. Therefore, we hypothesized that exosome treatment could promote perfusion recovery after hindlimb ischemia. In chapter 6, we studied the effect of MSC-exosomes. However, we observed that systemic treatment with exosomes inhibited perfusion recovery accompanied by smaller collateral vessels in a murine model. These results were unexpected and contradict the findings in myocardial ischemia/reperfusion injury studies. We did not observe this after systemic exosome treatment in the MI model. However, the exact mechanism of exosome treatment in this study remains unclear. Since we do not have any data on the local effect of exosomes, it is possible that systemic treatment led to a different response which hampered perfusion recovery. In the future, it is critical to validate the mechanism by which these exosomes influence collateral artery growth. Furthermore, the timing and route of administration might be critical.

In chapter 7, we focused on a different aspect of cardiovascular research, namely advanced human atherosclerosis and abdominal aortic aneurysm (AAA). In patients with advanced atherosclerosis or AAA, we investigated if leukotriene B4 (LTB4) expression is associated with a vulnerable plaque phenotype and if LTB4 expression levels could predict future cardiovascular events. Numerous studies have been performed in experimental and clinical settings to study the role of LTB4 in CVD. Even though the overall consensus was that LTB4 is involved in disease initiation and progression, this does not guarantee involvement in the advanced disease state. In this study, we could not provide evidence for this association or the predictive role of LTB4 expression for secondary clinical manifestations in patients that underwent carotid endarterectomy or AAA surgery. Clinical trials testing inhibitors for LTB4 formation to prevent disease progression need to be evaluated carefully.

Future perspectives
In the past years, collateral artery growth research has taken important steps towards clinical implementation. As our understanding of the biological components grows, new technical approaches are developed simultaneously. However, there are still hurdles to take, since translation of pre-clinical data to clinical studies is not yet successful. Animal models used in this thesis and often used in cardiovascular research in general do not include co-morbidities like aging, diabetes or hypercholesterolemia. A potential risk of promoting arteriogenesis is the possible stimulation of plaque neovascularization. Especially systemic treatment could stimulate plaque growth and destabilization by promoting the formation of new blood vessels within the plaque. Therefore, local delivery of therapeutic compounds might be necessary. In general, this concept is promising, but until now no delivery systems are found to be effective in clinical settings. Furthermore, the experimental models used to simulate clinical arterial occlusion do not fully resemble the clinical situation. Patients often suffer from progressive narrowing of the arteries over years or even decades. In experimental models, ischemic disease is simulated by surgical intervention to enable acute arterial occlusion, which does not represent the clinical setting.

Within this thesis, the most suitable candidate for enhancing arteriogenesis would be CXCL10. However, there are differences between murine and human CXCL10 signaling, which makes translation of research extremely challenging and should therefore be done with care. Although we observed a role for CD200/CD200R axis in arteriogenesis, its therapeutic potential seems small and would not be the ideal candidate for further investigations to stimulate collateral artery growth. In future research, possible therapeutic candidates need to be chosen carefully. Combining both the biological translation from animal to human together with development of technical approaches needs to be the focus for future research.