Elevated Fluid Shear Stress Enhances Postocclusive Collateral Artery Growth and Gene Expression in the Pig Hind Limb
Objective— The role of fluid shear stress (FSS) in collateral vessel growth remains disputed and prospective in vivo experiments to test its morphogenic power are rare. Therefore, we studied the influence of FSS on arteriogenesis in a new model with extremely high levels of collateral flow and FSS in pig and rabbit hind limbs.
Methods and Results— A side-to-side anastomosis was created between the distal stump of one of the bilaterally occluded femoral arteries with the accompanying vein. This clamps the collateral reentry pressure at venous levels and increases collateral flow, which is directed to a large part into the venous system. This decreases circumferential wall stress and markedly increases FSS. One week after anastomosis, angiographic number and size of collaterals were significantly increased. Maximal collateral flow exceeded by 2.3-fold that obtained in the ligature-only hind limb. Capillary density increased in lower leg muscles. Immunohistochemistry revealed augmented proliferative activity of endothelial and smooth muscle cells. Intercellular adhesion molecule-1 and vascular cell adhesion molecule (VCAM)-1 were upregulated, and monocyte invasion was markedly increased. In 2-dimensional gels, actin-regulating cofilin1 and cofilin2, destrin, and transgelin2 showed the highest degree of differential regulation.
Conclusions— High levels of FSS cause a strong arteriogenic response, reinstate cellular proliferation, stimulate cytoskeletal rearrangement, and normalize maximal conductance. FSS is the initiating molding force in arteriogenesis.
Numerous studies have documented the influence of fluid shear stress (FSS) as an arterial molding force,1–3 but information on the actions of markedly increased in vivo FSS on the development of arterial collateral vessels after occlusion of a conduit artery is lacking. Such studies are needed because attempts at changing FSS often also alter the circumferential wall stress, another acknowledged molding force of growing collateral vessels. The formation of a collateral circulation after an arterial occlusion correlates well with the calculated increase in FSS because of the increased collateral flow caused by the pressure decrease along pre-existent collaterals. However, because of the fast increase in collateral diameter by cellular proliferation, FSS decreases quickly again, and the early termination of the growth process at an incomplete stage of adaptation is believed to be caused by the only transient action of FSS. One of the hypotheses to be tested was, therefore, whether prolonged action of FSS would also improve the final adaptation by continued growth. The present experiments were therefore undertaken to prospectively study the causal relations between arteriogenesis and FSS by a stepwise and lasting increase of collateral flow brought about by the creation of a side-to-side anastomosis between the distal stump of the occluded femoral artery and its accompanying vein. This maneuver abruptly increases FSS and decreases circumferential wall stress. Another question to be answered was which proteins are prominently involved in the switch to prolonged collateral growth. To the best of our knowledge, our study is the first to provide direct evidence for the morphogenic power of FSS in the shaping of the peripheral collateral circulation.
Materials and Methods
The present study was performed according to Section 8 of the German Law for the Protection of Animals, which conforms to the US National Institutes of Health (NIH) guidelines. Twenty hybrid pigs and 6 New Zealand White rabbits were used in this study. Bilateral femoral ligation was performed under anesthesia, and 1 week later an arteriovenous (AV) shunt was created unilaterally between distal femoral artery stump and femoral vein. The ligated contralateral side served as control (Figure 1). We did not observe any gangrene or gross impairment of hind limb function after femoral artery occlusion and AV fistula creation.
Fractional collateral flow (FCF) was calculated from hemodynamic measurements in vivo performed 7 days after shunt creation. Furthermore, the animals underwent magnetic resonance imaging (MRI) angiography for the direct estimation of collateral flow. Thereafter, contrast medium was injected for digital subtraction angiography in pigs and computed tomography scans in rabbits. Immunohistochemistry was performed in frozen tissue for the quantitative determination of proliferation rate and expression of adhesion molecules.
Two-dimensional polyacrylamide gel electrophoresis and mass spectrometry were used to evaluate differential protein expression.
Molecular probes for human actin depolymerizing factor/destrin-cDNA, human cofilin1-cDNA, rabbit cofilin2-cDNA,4 and a synthesized antisense oligonucleotide to 18S rRNA were used for Northern blot analysis. Results are presented as means±SEM. Comparisons between 2 mean values were performed using the unpaired Student t test. P<0.05 values were considered to be statistically significant.
For an expanded Methods section, please see http://atvb.ahajournals.org.
All animals were able to stand up and walk without limp after awaking from anesthesia. No signs of ischemia or necrosis were noted. In 4 animals, the shunt had closed spontaneously for unknown reasons, resulting in a success rate of 80%. Patency of the shunt was judged by palpable vibrations, typical ultrasound patterns, and, at the final experiment, by digital subtraction angiography and MRI in the other cases. Simultaneous ligature plus shunt was well-tolerated in rabbits.
The network of collateral vessels was markedly enlarged on the shunt side after 1 week of shunt opening and 2 weeks after femoral artery occlusion (Figure 2A and 2B). Point counting of angiograms showed many more collateral vessels after 2 weeks compared with ligation only (control; 13.4±1.4 versus 5.6±0.6; P<0.001). The number of visible collaterals was increased as well as the diameter of arteries feeding into the collateral network (Table 1). Point counting was performed in duplicate by 6 observers, of whom 4 were naive. Agreement between observers was excellent, ie, all gave higher ranks to the shunt side.
Magnetic Resonance Angiography
Electrocardiogram-triggered MR contrast medium injection was followed by data acquisition, which showed mainly the greatly increased flow to the shunted side, the enlargement of the feeder arteries, but no detailed images of the collaterals proper (Figure 2A). Quantification of iliac vein pixel intensity using the NIH Image 1.62 software for Macintosh revealed a 2.3-fold increase in collateral-related flow on the shunt side (P<0.001; Table 1).
Pressure data obtained in pigs at rest showed no significant deviation between the shunt- and control-ligated leg. After shunt closure, only at maximal vasodilatation a difference could be demonstrated. During the first 20 seconds of the reactive hyperemia, the peripheral pressure of the shunted side increased faster compared with control, arguing for a reduced collateral resistance of the shunted side (Table 1).5
Cell proliferation in shunt collateral vessels of the pig system, based on Ki 67 and Mef2b staining, was 2-fold higher than in control: mitotic index 14% versus 7% of the collaterals in the contralateral side (P<0.001). The mitotic index of smooth muscle cells in the rabbit system was 20% in shunt collaterals versus 8% in those of the control (ligated) side (P<0.001). Mitotic labeling of endothelial cells indicative for angiogenesis was significantly increased in lower leg muscles but was unchanged in upper leg muscles close to the growing collateral vessels (Table 2). The total lumen area of vessels on the shunt side in pigs was 1.5-fold larger than on the control side (P<0.05). Staining of rabbit collateral artery sections with the RAM-11 antibody revealed monocyte invasion into the intimal, medial, and adventitial layers, and staining with antibodies recognizing vascular cell adhesion molecule-1 (VCAM-1) showed strong overexpression in the endothelial and smooth muscles layers, as well as of adventitial cells (Figure 3).
Intercellular Adhesion Molecule-1 Expression Was Markedly Increased but Staining Was Restricted to Endothelial Cells
The calibrated fluorescence signal for intercellular adhesion molecule-1 in endothelial cells increased from zero (normal arterioles) to 75.2±7.2 AUs in control collaterals and to 120.5±3.6 in shunt collaterals. The VCAM-1 signal increased from 20.1±3.0 (normal arterioles) to 80.5±6.2 in normal collaterals and to 165.1±4.5 in shunt collaterals. The VCAM-1 signal in the adventitia of shunt collaterals increased 14-fold over control. All changes were statistically significant at the P<0.001 level.
The 2D-PAGE from pig tissue showed a protein pattern of collateral arteries that differed from that of normal arteries. So far, we identified 22 proteins with altered expression, of which destrin, cofilin1, cofilin2, and transgelin2 showed the most marked expression changes. On the mRNA level, one 1.9-kb transcript of destrin was detectable in untreated arteries. Destrin expression decreased in the growing collateral arteries from both sides to almost undetectable levels. Cofilin1 mRNA was upregulated (+36%), whereas cofilin2 mRNA is downregulated in the collateral arteries after AV shunting. Transgelin2 was upregulated. Northern blot analysis showed also an increased 1.3-kb mRNA amount (+61%) after AV shunting compared with the occluded-only side. All changes were significant at the P<0.05 level.
Previous experiments had shown that the collateral circulation in the canine and porcine heart,6,7 as well as in the rabbit, mouse, and pig hind leg, stops growing before an optimal adaptation is reached, ie, usually when the maximal conductance had reached 35% to 40% of normal.8 Experimental therapy with growth factors can improve on that by not more than several percentage points.9
Increased FSS is thought to be responsible for triggering arteriogenesis,10,11 because the sudden decrease in peripheral pressure after an arterial occlusion increases the velocity of flow in pre-existent collateral arterioles that interconnect the prestenotic high-pressure bed with the poststenotic low-pressure bed. The cause of the premature growth inhibition could be the decrease of FSS caused by the marked increase in collateral diameter, which is inversely related to shear stress by the cube of the arterial radius. FSS decreases also because normal autoregulation of blood flow sets in (and stump pressure increases) after the growth phase and restricts blood flow to levels far below maximal.
With the present experiments, we prevented the early increase in postocclusive pressure and the decrease in FSS by forcing the collateral flow to drain directly into the venous system. This almost triples collateral flow and markedly increases FSS. The highest estimate of collateral flow of 4-fold over control in the pig and 7-fold in the rabbit would almost reach normal maximal flow of an unoccluded vascular bed. With this experiment, we were able to test the hypothesis regarding whether it is possible to achieve a more complete restoration of vascular structure and function. The results show that this is indeed the case. The molecular mechanisms, apart from the increase in FSS, were a more pronounced invasion of monocytes via overexpression of adhesion molecules, a restart of cell proliferation, mainly of smooth muscle cells and endothelial cells, and a prolongation of the dedifferentiated state of the smooth muscle cells that is characterized by differential expression of the actin-(de)polymerizing proteins. Our new experiments indeed showed that prolonged action of markedly increased FSS produced a much more advantageous adaptation by collaterals and indirectly confirmed our hypothesis that the premature decrease of FSS was the cause for the inadequate collateral growth. That shear stress is of general great importance for the maintenance of arterial diameter has long been known.12–16 Numerous studies were subsequently published and the reader is referred to excellent reviews.11,17 Despite the general acceptance of FSS as a molding force in the arterial system, its role in the development of the collateral circulation had remained controversial. Our earlier experiments had favored tangential, or circumferential wall stress, as the more important force, because it is 2 orders of magnitude higher.13 Model experiments by Spaan’s group had recently shown that FSS cannot be the only molding force, because their mathematical model predicted that in a network of parallel arterioles with small variations of diameter, the larger ones would be favored and would grow, whereas the smaller ones would regress.18 However, regression of smaller collaterals is a typical feature in our models. Our present experiments clearly show that a primary change in FSS is the dominant mechanical force in collateral artery growth. However, we cannot rule out that other physical factors like heat dispersion, diffusion, or signaling from downstream vessels may also play a role.
Collateral artery formation and ischemia are often associated19 but not necessarily causally linked.10 It is highly probable that the venous drainage of collateral flow had induced transient ischemia in the lower leg, as visualized by the increased capillary density in the lower leg muscles (see Table 2). However, collateral vessels grow in the upper leg and are not under the direct influence of ischemic tissue, as visualized by the unchanged capillary density values in the adductor muscles. At the time of the terminal experiment, ie, 1 week after the creation of the shunt, ischemia was not present even in the lower leg, as shown by the absence of tissue atrophy (MR data), normal polarographic pO2 values in the gastrocnemious muscles, and no changes in the absolute values and right-to-left ratio of vascular endothelial growth factor mRNA tissue concentrations. Rapid capillary sprouting in the lower leg had prevented lasting and irreversible ischemia and lower leg skeletal muscle necrosis. That capillary sprouting is able to reduce minimal resistance was shown by us previously in the pig heart.20 We had also shown previously that ischemia is not a requirement for arteriogenesis from experiments in the rabbit10,21 and mouse hind leg. We therefore conclude that also under the conditions of venous drainage of collateral flow, ischemia had not influenced the development of collateral vessels in the upper leg.
Differential Gene and Protein Expression
Growing collateral vessels are characterized by a change from the contractile to the synthetic and proliferative phenotype of the smooth muscle cells. The latter 2 are characterized by the lack of actin filaments, which is the result of downregulation of actin transcription as well as the increased actin depolymerization. Proteins involved in actin binding are cofilin1 and cofilin2, as well as destrin and transgelin2. These are the proteins that showed the highest degree of expression change on the mRNA as well as on the protein level in FSS-transformed collateral vessels. They are necessary for cell mobility; however, apart from their role as structural proteins, they are also part of the ρ-signaling chain, which is involved in smooth muscle cell migration and proliferation.22
The Role of Monocytes
Antibodies recognizing monocyte-specific cell surface antigens in the pig system are scarce and those available did not function reproducibly in our hands. Therefore, we performed identical ligation shunt experiments in rabbits to study the involvement of monocytes. The induced increase in FSS produced an even more marked adhesion and invasion of monocytes that were found not only in the intima but also in the media and adventitia. They were attracted by a marked overexpression of intercellular adhesion molecule-1 and of VCAM-1, which later was not only expressed in endothelial but also expressed in smooth muscle and adventitial cells. Overexpression of VCAM is usually caused by abundance of oxygen-based radicals,23 and these may have been produced by an “uncoupled” eNOS.24 Our study confirms, again, the importance of monocytes in the process of arteriogenesis. They are attracted by FSS-activated endothelium and they digest the extracelluar matrix, and together with the activated endothelium they produce the mitogens for the endothelium and the smooth muscle cells. Only markedly increased shear stress can activate, again, the endothelium, leading to a new round of monocyte invasion and arterial growth. Monocyte invasion was strong enough for detection of the cytokine marker oncostatin M on 2-dimensional gels. Proteases of monocyte origin also digest elastin. That and the fibroblast growth factor–dependent downregulation of elastin expression in smooth muscle cells contributes also to smooth muscle cell proliferation.25,26
Age may influence the outcome of vascular studies.27 All pig studies were performed in young castrated male pigs that still gained weight but exhibited a mature cardiovascular system as judged by normal mitotic indices and normal heart weights. The rabbits used in parallel for this study were adult and sexually mature. Because no systemic differences between our pig and rabbit results were found, we conclude that age had not influenced the pig results.
We do not wish to leave the impression that the AV shunt method as used in our experiments is applicable in human patients, because it may increase already-existing ischemic states.
The authors thank Alf Theissen for animal care and Marianne Granz for technical assistance in molecular biology.
F.P. and S.B. contributed equally to this study.
- Received April 13, 2004.
- Accepted June 4, 2004.
Kamiya A, Togawa T. Adaptive regulation of wall shear stress to flow change in the canine carotid artery. Am J Physiol. 1980; 239: H14–H21.
Unthank JL, Fath SW, Burkhart HM, Miller SC, Dalsing MC. Wall remodeling during luminal expansion of mesenteric arterial collaterals in the rat. Circ Res. 1996; 79: 1015–1023.
Pijls NH, Van Son JA, Kirkeeide RL, De Bruyne B, Gould KL. Experimental basis of determining maximum coronary, myocardial, and collateral blood flow by pressure measurements for assessing functional stenosis severity before and after percutaneous transluminal coronary angioplasty. Circulation. 1993; 87: 1354–1367.
Schaper W, Flameng W, Winkler B, Wuesten B, Tûrschmann W, Neugebauer G, Carl M, Pasyk S. Quantification of collateral resistance in acute and chronic experimental coronary occlusion in the dog. Circ Res. 1976; 39: 371–377.
Hoefer IE, Van Royen N, Buschmann IR, Piek JJ, Schaper W. Time course of arteriogenesis following femoral artery occlusion in the rabbit. Cardiovasc Res. 2001; 49: 609–617.
Unger EF, Banai S, Shou M, Jaklitsch M, Hodge E, Correa R, Jaye M, Epstein SE. A model to assess interventions to improve collateral blood flow: continuous administration of agents into the left coronary artery in dogs. Cardiovasc Res. 1993; 27: 785–791.
Ito WD, Arras M, Winkler B, Scholz D, Htun P, Schaper W. Angiogenesis but not collateral growth is associated with ischemia after femoral artery occlusion. Am J Physiol. 1997; 273: H1255–H1265.
Gimbrone MA, Jr, Anderson KR, Topper JN, Langille BL, Clowes AW, Bercel S, Davies MG, Stenmark KR, Frid MG, Weiser-Evans MC, Aldashev AA, Nemenoff RA, Majesky MW, Landerholm TE, Lu J, Ito WD, Arras M, Scholz D, Imhof B, Aurrand-Lions M, Schaper W, Nagel TE, Resnick N, Dewey CF, Gimbrone MA, Davies PF. Special communication the critical role of mechanical forces in blood vessel development, physiology and pathology. J Vasc Surg. 1999; 29: 1104–1151.
Thoma R. Untersuchungen über die Histogenese und Histomechanik des Gefäβsystems. Stuttgart: F. Enke; 1893.
Murray CD. The physiological principle of minimum work applied to the angle of branching arteries. J Gen Physiol. 1926; 9: 835–841.
Cornelissne AJ, Dankelman J, Vanbavel E, Spaan JA. Balance between myogenic, flow-dependent, and metabolic flow control in coronary arterial tree: a model study. Am J Physiol Heart Circ Physiol. 2002; 282: H2224–H2237.
Matsunaga T, Weighrauch DW, Montz MC, Tessmer J, Warltier DC, Chilian WM. Angiostatin inhibits coronary angiogenesis during impaired production of nitric oxide. Circulation. 2002; 105: 2185–2191.
William C, Schneider R, Frei U, Eckardt KU. Increases in oxygen tension stimulate expression of ICAM-1 and VCAM-1 on human endothelial cells. Am J Physiol. 1999; 276: H2044–H2052.
Sedding DG, Seay U, Fink L, Heil M, Kummer W, Tillmanns H, Braun-Dullaeus RC. Mechanosensitive p27Kip1 regulation and cell cycle entry in vascular smooth muscle cells. Circulation. 2003; 108: 616–622.
Langille B. Remodeling of developing and mature arteries: endothelium, smooth muscle, and matrix. j Cardiovasc Pharmacol. 1993; 21: 11–17.