Targeted Delivery of Bone Marrow Mononuclear Cells by Ultrasound Destruction of Microbubbles Induces Both Angiogenesis and Arteriogenesis Re
http://www.100md.com
动脉硬化血栓血管生物学 2005年第10期
From the Department of Medicine II (T. Imada, Y.M., T.N., H. Masaki, T. Iwasaka) and Radiology (H.K.), Kansai Medical University, Osaka; the Department of Cardiovascular Medicine (T.T., M.O., H. Matsubara), Kyoto Prefectural University School of Medicine, Kyoto; the Department of Medical Biochemistry (M.Y., Y. Nishiwaki, N.N.), Graduate School of Medicine, Tokyo Medical and Dental University, Tokyo; and the Pharmacobioregulation Research Laboratory (Y. Nozawa), Taiho Pharmaceutical Co Ltd, Saitama, Japan.
Correspondence to Hiroaki Matsubara MD, Department of Cardiovascular Medicine, Kyoto Prefectural University School of Medicine, Kamigyo-ku, Kyoto, 602-8566, Japan. E-mail matsubah@koto.kpu-m.ac.jp
Abstract
Objective— Ultrasound (US)-mediated destruction of contrast microbubbles causes capillary rupturing that stimulates arteriogenesis, whereas intramuscular implantation (im) of bone marrow mononuclear cells (BM-MNCs) induces angiogenesis. We therefore studied whether US-targeted microbubble destruction combined with transplantation of BM-MNCs can enhance blood flow restoration by stimulating both angiogenesis and arteriogenesis.
Methods and Results— US-mediated destruction of phospholipid-coated microbubbles was applied onto ischemic hindlimb muscle and subsequently BM-MNCs were transfused. A significant enhancement in blood flow recovery after Bubble+US+BM-MNC infusion (34% increase, P<0.05) was observed compared with Bubble+US (25%). The ratio of capillary/muscle fiber increased by Bubble+US+BM-MNC-i.v (260%, P<0.01) than that in the Bubble+US group (172%), into which BM-MNCs were incorporated (angiogenesis). Smooth muscle -actin–positive arterioles were also increased, and angiography showed augmented collateral vessel formation (arteriogenesis). Platelet-derived proinflammatory factors activated by Bubble+US induces the expression of adhesion molecules (P-selectin and ICAM-1), leading to the attachment of transplanted BM-MNCs on the endothelium. Flow assay confirmed that the platelet-derived factors cause the adhesion of BM-MNCs onto endothelium under laminar flow.
Conclusions— This study demonstrates that the targeted delivery of BM-MNCs by US destruction of microbubbles enhances regional angiogenesis and arteriogenesis response, in which the release of platelet-derived proinflammatory factors activated by Bubble+US play a key role in the attachment of transplanted BM-MNCs onto the endothelial layer.
Ultrasound (US)-targeted microbubble destruction combined with transplantation of bone marrow mononuclear cells (BM-MNCs) enhances blood flow restoration by stimulating both angiogenesis and arteriogenesis. Release of proinflammatory factors from platelets activated by Bubble+US play a key role for adhesion of BM-MNCs on endothelium. This cell delivery system is efficient for therapeutic angiogenesis and arteriogenesis.
Key Words: angiogenesis ? vasculogenesis ? stem cell ? endothelial progenitor cell ? ultrasound
Introduction
Therapeutic angiogenesis, the ability to induce the formation of new blood vessels, is one of the most promising targets for regeneration therapy. To induce angiogenesis, investigators have delivered vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF/FGF2). or hypoxia-inducible factor-1/etoposide as recombinant proteins or genes.1 Intramuscular injection of bone marrow mononuclear cells (BM-MNCs) was shown to be feasible in patients or animals with ischemic limbs by supplying angiogenic factors and endothelial progenitors.2–4 A noninvasive cell delivery system that can target vascular endothelium would be a great advantage for manipulation of angiogenic cell therapy.
Ultrasound (US)-targeted microbubble destruction has been investigated as a new method for delivering drugs and genes to specific tissues.5–11 This method involves the attachment of drugs or genes to gas-filled microbubbles, which are then circulated through the intravascular space and mechanically destroyed within the target organ. Theoretically, one can target any anatomic site that is accessible by US, including selected damaged regions.5–11 Song et al have reported that US-targeted microbubble destruction causes capillary rupturing that stimulates arteriogenesis and an increase in blood flow in both normal12 and ischemic13 skeletal muscles, in which angiogenesis response is transient and unlikely contributes to chronic restoration of blood flow. We previously demonstrated that the recruitment of BM-MNCs and platelets stimulates angiogenesis response in ischemic muscles by releasing potent angiogenic factors, such as VEGF or bFGF, and supply of endothelial progenitors.3,14 Furthermore, we have recently reported that systemically transplanted BM-MNCs can be firmly attached onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore examined whether US-targeted microbubble destruction combined with intravenous transplantation of BM-MNCs causes angiogenesis response as well as arteriogenesis in a rat model with an ischemic hindlimb, leading to an enhancement of chronic blood flow restoration. Interestingly, we found that platelet-derived factors activated by US-targeted microbubble destruction induce the expression of adhesive molecules on the endothelium and subsequent attachment of transplanted BM-MNCs, resulting in an enhancement of formation of neocapillaries and new arterioles and an increase in regional blood flow recovery.
Materials and Methods
Microbubble Preparations
BR14 (Bracco Diagnostics) is a new ultrasound contrast agent, consisting of perfluorocarbon-containing microbubbles stabilized by a phospholipid monolayer.16,17 The suspension of gas microbubbles is reconstituted immediately before use by injecting 5 mL of 0.9% sodium chloride. The mean diameter of bubbles is 2.3 to 2.5 μm and the bubble concentration is 6x108 per ml. BR14 was administered as a 1 mL bolus injection via contralateral femoral vein.
Isolation of BM-MNC and Characterization of Endothelial-Lineage Cells
BM-MNCs were isolated from SEA/LEW rat femoral bone and centrifuged by density gradient (Lymphoprep; Nycomed). The MNC fraction was labeled with red-fluorescence cell linker (PKH26-GL; Sigma).2,3,14 Endothelial lineage cells were analyzed by fluorescence-activated cell sorter (FACS) using DiI-acetylated LDL (acLDL) incorporation (Biogenesis) and Ulex lectin binding (Sigma) as described previously.2,3,14
Hindlimb Ischemia, US Application, and Transfusion of BM-MNCs
Unilateral hindlimb ischemia was induced by resecting the left femoral artery as described.2,18 Arteriogenesis that restores the regional blood flow was reported to be observed 3 days after induction of limb ischemia of rats.12,13 We therefore performed the infusion of microbubble at day 3 after limb ischemia to efficiently deliver microbubble BR14 to the ischemic site. We divided the rats into the following five groups: (1) Control (saline injection, n=8); (2) Bubble (1 mL BR14)+US (n=8); (3) Bubble+US+BM-MNC-i.v. (n=8), BM-MNC (2x107 in 1 mL saline) injected from contralateral femoral vein 1 minute after Bubble+US; (4) BM-MNC-i.m. (n=8), BM-MNC (2x107) intramuscularly implanted into the ischemic limb; and (5) BM-MNC-i.v. (n=8); BM-MNC (2x107 in 1 mL saline). BM-MNC (2x107) included 0.3±0.04x107 platelets.
The skin overlying the ischemic thigh muscle to be treated was reflected back, ultrasound gel was placed over the ischemic muscle, and 1-MHz transducer (S-probe, Effective Radiating Area: 0.9 cm2; Ito Co Ltd) was held 3 mm over the muscle and to keep the skin temperature around 37°C. A continuous sinusoidal wave ultrasound (1MHz, 2W/cm2, Beam Nonuniformity Ratio: 3.6; ITO-US-700) was applied for 1 minute immediately after microbubble injection. To clarify whether the microbubbles really reach ischemic tissues after infusion from contralateral femoral vein, the delivery of injected microbubbles to the ischemic area was examined by a diagnostic ultrasound echography (Sonos-5500) equipped with an ultraband S12 sector transducer.
Evaluation of Neocapillary and Immunohistochemical Analysis
Four pieces of ischemic tissues from the adductor and semimembranous muscles were obtained 28 days after limb ischemia. Frozen sections were stained with antibodies for smooth muscle (SM) -actin (Sigma) and anti-factor VIII (DAKO) antibodies. The appropriate secondary antibodies conjugated with fluorescein isothiocyanate (FITC) or rhodamine were used. Ten fields from 2 muscle samples of each animal were randomly selected for the vessel count. To ensure that vessel densities were not overestimated as a consequence of myocyte atrophy or underestimated because of interstitial edema, the capillary/muscle fiber ratio was determined.3,18
Electron Microscope
Femoral arteries in skeletal muscles on which targeted US was applied were isolated after Bubble+US treatment (n=4), fixed by 2.5% glutaraldehyde and 1.5% osmium acid, dried, and viewed by electron microscope (HITATI S-700).15
Cell Culture and Platelet-Rich Plasma Isolation
Human umbilical vein endothelial cells (HUVECs; Kurabo) were cultured in HuMedia-EG2 medium. For use in the study apparatus, HUVECs (2nd or 3rd passages) were plated on 22-mm fibronectin-coated cover slips.15 Peripheral blood was drawn from healthy volunteers and mixed with 0.1 volume of citrate (108 mmol/L). Whole blood was centrifuged at 150g for 10 minutes to harvest platelet-rich plasma (PRP).
Adhesion Assay Under Laminar Flow and Immunofluorescence Study
HUVECs were incubated for 30 minutes with serum-free medium with or without freshly prepared 10% platelet-rich plasma, and then stimulated by (1) 10% Bubble+US (1MHz, 1.5W, 30 seconds) in 10% PRP (Bubble+US with platelet group), (2) 10% Bubble+US in PRP (Bubble+US without platelet), (3) incubation media (in which 10% platelet-rich medium was stimulated by 10% Bubble+US), or (4) US with platelet (in which 10% platelet-rich medium was stimulated by US alone), n=6 in each experiment. HUVECs were placed on the cold plate to prevent the heating by the US, and adhesion assay under laminar flow was performed as previously described.15 For further detail, please see the supplemental Methods, available online at http://atvb.ahajournals.org.
Stimulated HUVECs were fixed by 4% paraformaldehyde 1 hour after stimulation and incubated with anti–P-selectin (R&D Systems, Inc) and anti-platelet glycoprotein (GP)-Ib antibody (DAKO). The appropriate secondary antibodies conjugated with FITC or rhodamine were used. Nuclei were stained with 4',6'-diamidino-2-phenylindole dihydrochloride (DAPI) and viewed by a fluorescence microscope (Olympus OX71).
Laser Doppler Perfusion Image and Angiography
Laser doppler perfusion image (LDPI) and angiography were performed as previously described3,18 For further detail, please see the supplemental Methods.
Statistical Analysis
Statistical analyses were performed with 1-way ANOVA followed by pair-wise contrasts using the Dunnett test. Data (mean±SE) were considered statistically significant when P<0.05.
Results
Incidence of Endothelial-Lineage Cells in BM-MNCs
FACS analysis indicated that 28±1.8% and 31±1.5% of BM-MNCs incorporated DiI-acLDL and bound Ulex-lectin (n=5), respectively, and 20±1.2% of cells were positive for both markers. Endothelial-lineage cells were considered to be included in this fraction as reported.2,3,14
Laser Doppler Blood Perfusion
Subcutaneous blood perfusion was analyzed by LDPI imaging (Figure 1). Intravenous injection of BM-MNCs (BM-MNC-i.v.) did not cause a significant increase in the blood flow recover compared with the control (saline injection; Figure 1). Treatment with Bubble+US without BM-MNCs-i.v. showed a moderate increase (25±2% at Day 28 versus control, P<0.05), whereas the combination of Bubble+US and BM-MNCs-i.v. induced a further increase (34±2% versus control at Day 28, P<0.01), which was significantly higher than that of Bubble+US manipulation alone (P<0.05). The blood perfusion recovery by Bubble+US+BM-MNC-i.v. was comparable to blood perfusion by BM-MNC-i.m. (38±3%). Blood perfusion after US+BM-MNC-i.v. without Bubble did not significantly differ from those in BM-MNC-i.v. alone or control groups (3±1% versus control at Day 28, n=8; data not shown), suggesting that the combination of BM-MNCs infusion with Bubble+US significantly improves the blood flow recovery after limb ischemia compared with each manipulation alone, and that the efficient cell delivery system depends on US-mediated destruction of microbubbles.
Figure 1. Increase of blood flow by US-microbubble-mediated transfusion of BM-MNCs. A, Representative LDPI. Greater blood perfusion (red to yellow) in ischemic limbs was observed in Bubble+US+BM-MNC infusion, BM-MNC-i.m., and Bubble+US groups in contrast with the untreated control and BM-MNC-i.v. groups. B, Computer-assisted analyses of LDPI revealed a significantly greater blood perfusion in the Bubble+US+BM-MNC-i.v., BM-MNC-i.m., and Bubble+US groups than in the control group. The values shown are mean±SE (n=8) at each time point. *P<0.05, **P<0.01 vs control.
Neocapillary and Arteriole Formation
Formation of capillaries and arterioles was evaluated by factor VIII–positive and SM -actin–positive vessels. The ratio of capillary/muscle fiber (factor VIII–positive and SM -actin–negative vessel) was significantly increased in the Bubble+US+BM-MNC-i.v. group (260±15% versus control, P<0.01), which was greater (P<0.01) than that of Bubble+US group (172±11% versus control, P<0.05). The ratio of arteriole/muscle fiber (factor VIII–positive and SM -actin–positive vessel) was also increased by Bubble+US+BM-MNC-i.v. (188±10% versus control, P<0.01), which was greater (P<0.05) than that in the Bubble+US group (146±9% versus control, P<0.05; Figure 2A). Transfused BM-MNCs (labeled with red fluorescence) were found to be incorporated into microvessels by Bubble+US+BM-MNC-i.v. (arrows in Figure 2B). These findings suggest that Bubble+US+BM-MNC-i.v. enhances both angiogenesis as well as arteriogenesis response.
Figure 2. Formation of arterioles/capillaries and incorporation of transfused BM-MNCs. A, Vessels were immunostained with anti–SM -actin and anti–factor VIII antibodies. Numbers for arterioles (positive for both SM -actin and factor VIII staining) and capillaries (SM -actin–negative and factor VIII–positive) were evaluated as a vessel number/muscle fiber ratio (n=8, each group). *P<0.05, **P<0.01 vs the saline-injected control. Bar=50 μm. B, Incorporation of transfused BM-MNCs (labeled with red-fluorescence). Red-labeled cells were incorporated into factor VIII–positive capillaries (arrows in the merged image) in the Bubble+US+BM-MNC-i.v. group. Data shown are representative of 5 different animals. Bar=200 μm.
To examine the specificity of this cell delivery system, we examined the distribution of transfused BM-MNCs in other tissues (kidney, spleen, liver, heart, pancreas, small intestines, and brain) 3 weeks after targeted cell delivery by Bubble+US+BM-MNCs-iv. Small numbers of labeled BM-MNCs were detected in the kidney (mainly in the tubule) and greater numbers of BM-MNCs were observed in the spleen, whereas no labeled cells were detected in other tissues (Figure I, available online at http://atvb.ahajournals.org).
Arteriogenesis Response Evaluated by Angiography
Compared with angiogenesis formed by capillary sprouting, arteriogenesis is often studied with the use of angiography. On the postoperative day 28, all animals were subjected to iliac angiography. Representative angiograms (n=4 in each group) are shown in Figure II (available online at http://atvb.ahajournals.org), in which arrows indicate the ligated ends of femoral arteries. Collateral vessels in the thigh area were quantitatively counted using 5-mm2 grids.3,18 An apparent increase in collateral vessel formation was observed in the Bubble+US+BM-MNC-i.v. and BM-MNC-i.m. (4.2±0.2-fold and 4.3±0.2-fold, n=6, respectively; P<0.001) compared with that in the BM-MNC-i.v. group. The increase in the Bubble+US group was 1.8±0.1-fold (P<0.05) when compared with that in the BM-MNC-i.v. group, which was significantly smaller (P<0.01) than that of the Bubble+US+BM-MNC-i.v. group. There was no significant difference between control (saline infusion) and BM-MNC-i.v. groups.
Delivery of Microbubbles to the Ischemic Muscle
Arteriogenesis that restores the regional blood flow was observed 3 days after induction of limb ischemia of rats.12,13 We therefore performed the infusion of microbubble at day 3 after limb ischemia to efficiently deliver microbubble BR14 to the ischemic site. We further examined whether the transfused microbubbles really reach the hindlimb muscle at 3 days after induction of ischemia. Apparent increase in the contrast densities in the ischemic thigh muscle (indicated by arrows in Figure III, available online at http://atvb.ahajournals.org) was observed 15 seconds after injection of microbubbles in both control (normal) and ischemic limbs compared with the pre-images before injection of microbubbles. The increase in contrast shadow diminished after 1 minute of US stimulation, indicating that microbubbles really reach ischemic hindlimb muscles after venous infusion.
Electromicroscopy
A previous study reported the presence of small holes in the endothelial cells treated with Bubble+US,11 whereas we could not detect the presence of apparent small holes in the vascular endothelium in the skeletal muscle on which the Bubble+US was applied. Neither adhesion of MNCs nor formation of fibrin network including platelets was detected on the surface of normal endothelium or endothelium stimulated by US without microbubbles. Interestingly, we found the attachment of platelets associated with fibrin network and MNCs on the surface of endothelium in all arteries treated by Bubble+US+BM-MNCs infusion (n=6; Figure 3), whereas no adhesion of platelets or MNCs was detected in the US+BM-MNC group without microbubbles (n=6; data not shown).
Figure 3. Electron-microscopic appearance of endothelium stimulated by Bubble-US+BM-MNCs. Electron-microscopy showing normal endothelium covered with a monolayer coat (Normal) and appearance of Bubble+US+BM-MNCs i.v.-stimulated endothelium. Attachment of both platelets associated with fibrin network and MNCs was observed on all samples treated by Bubble+US+BM-MNCs i.v. Representative figures are shown (n=4 in each group). Bar=10 μm.
Induction of Adhesion Molecules
We have reported that transplanted BM-MNCs can firmly attach onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore examined the expression profile of adhesive molecules on HUVECs treated by Bubble+US. The expression of adhesive molecules (P-selectin or ICAM-1) was not induced when HUVECs were stimulated by Bubble+US in the medium without platelets. Because we found that adhesion of the fibrin network including platelets was consistently observed on the surface of endothelium in all Bubble+US-treated samples (Figure 3), we next examined the involvement of platelet-derived factors in the adhesion of BM-MNCs. Interestingly, the expression of P-selectin (red fluorescence, right panel in Figure 4) and ICAM-1 was markedly induced in both HUVECs and attaching glycoprotein Ib–positive platelets (merged, yellow), when HUVECs were stimulated by Bubble+US in the medium including platelets (only P-selectin data shown in Figure 4).
Figure 4. Adhesion of platelets and induction of adhesion molecules on Bubble+US-stimulated HUVECs. The presence of platelets and expression of P-selectin are shown by immunostaining with anti–GP-Ib (green fluorescence) and anti–P-selectin (red fluorescence) antibodies and DAPI staining (blue fluorescence). The expression of P-selectin and adhesion of platelets were observed on HUVECs, when HUVECs were stimulated by Bubble+US in the medium containing 10% platelets or exposed to the incubation medium (in which 10% platelet-including medium was stimulated by Bubble+US). The expression of P-selectin (red fluorescence) on HUVECs and attaching platelets was markedly induced. Bar=50 μm.
We also studied whether factors released from the platelets stimulated by Bubble+US are involved in the induction of adhesion molecules. Therefore, we examined the effect of the incubation medium, in which platelet-including medium was stimulated by Bubble+US, on the induction of adhesion molecules. The addition of incubation medium caused the apparent increase in P-selectin expression on HUVECs (red fluorescence, right panel in Figure 4), suggesting that the factors released from the platelets stimulated by Bubble+US are closely involved in the induction of adhesion molecules on HUVECs. Treatment of HUVECs by US alone without microbubbles in the medium including platelets did not induce any expression of adhesive molecules (data not shown). These findings suggest that activation of platelets by Bubble+US and release of platelet-derived proinflammatory factors play a key role in the induction of adhesion molecules in the endothelial cells.
Laminar Flow Assay
We next studied whether the adhesive activity of BM-MNCs on the endothelium was actually modulated by Bubble+US-activated platelets under laminar flow condition. HUVECs were stimulated by Bubble+US in the medium containing platelets and then the adhesion ratio of BM-MNCs on HUVECs was evaluated under laminar flow as previously reported.15 The presence of platelets in the medium markedly increased the adhesion ratio of BM-MNCs (3.5-fold, P<0.01) compared with Bubble+US without platelets (Figure 5). Moderate increase (1.9-fold, P<0.01) was observed in the incubation medium group (in which medium including platelets was stimulated by Bubble+US and then added to HUVECs) compared with the ratio in the untreated HUVECs (control). Treatment of HUVECs by US stimulation without microbubbles in the medium including platelets or addition of control medium including platelet alone (without US stimulation) did not cause a significant increase in the adhesion ratio of BM-MNCs compared with the ratio in the untreated HUVEC (data not shown).
Figure 5. Laminar flow assay of BM-MNCs adhesion. Adhesive activities of BM-MNCs were quantified on HUVEC monolayer under laminar flow. Adhesive cells are indicated by yellow arrows. Cells were considered to be adherent after 10 seconds of stable contact with the monolayer. HUVECs were stimulated by Bubble+US (1MHz, 1.5W, 30 seconds) in the medium with 10% platelets (n=6), without platelets (n=6), or in the incubation medium (in which 10% platelet-including medium was stimulated by Bubble+US, n=6). *P<0.05, **P<0.01 vs the untreated control. Bar=100 μm.
Discussion
Angiogenic cell therapy by intramuscular implantation of autologous BM-MNCs was shown to be feasible in patients with ischemic limbs.2 Because intramuscular implantation is invasive at the injected sites, the development of a noninvasive cell delivery system that can target vascular endothelium would be a great advantage for the manipulation of angiogenic cell therapy. A new delivery system of drugs or genes has been developed using US-targeted microbubble destruction; the drugs or genes that attach onto gas-filled microbubbles circulate through the intravascular space and are mechanically destroyed within the target organ by ultrasound,5–11 whereas no studies were reported to determine whether this method is feasible for delivering the "cells" to specific vascular sites.
US-targeted microbubble destruction was reported to cause an inflammatory action on the cell surface by making small holes that revert to a normal appearance within 24 hours.10,11 Song et al have reported that US-targeted microbubble destruction causes capillary rupturing that stimulates arteriogenesis and an increase in blood flow in both normal12 and ischemic13 skeletal muscles, in which angiogenesis response is transient and unlikely contributes to chronic restoration of blood flow. They concluded that arteriogenesis response rather than angiogenesis plays a major role in US+microbubble-stimulated blood flow recovery. We previously demonstrated that the recruitment of BM-MNCs and platelets stimulates angiogenesis response in the ischemic muscles by releasing potent angiogenic factors, such as VEGF or bFGF, and supply of endothelial progenitors.3,14 Furthermore, we have recently reported that systemically transplanted BM-MNCs can be firmly attached onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore expanded the previous studies by Song et al12,13 and examined whether US-targeted microbubble destruction combined with intravenous transplantation of BM-MNCs causes both angiogenesis and arteriogenesis response in an ischemic hindlimb model, leading to a greater enhancement of blood flow restoration.
We found that (1) intravenous infusion of BM-MNCs combined with US-mediated destruction of microbubbles markedly enhances the restoration of regional blood perfusion in ischemic hindlimbs by stimulating both chronic angiogenesis and arteriogenesis response, and (2) release of platelet-derived proinflammatory factors activated by US-mediated destruction of microbubbles causes the adhesion of transfused BM-MNCs on endothelium by inducing the expression of adhesion molecules (P-selectin and ICAM-1). We found that BM-MNCs transfused intravenously were trapped by the spleen and a few BM-MNCs were present in the renal tubules. Considering that BM-derived hematopoietic stem cells were reported to transdifferentiate to renal tubular cells and improve renal function in the ischemia-reperfusion injury model,19 the present study establishes for the first time that targeted delivery of BM-MNCs by US destruction of microbubbles is an efficient cell delivery system for therapeutic angiogenesis and arteriogenesis, and that the presence of platelets and/or platelet-derived proinflammatory factors activated by US+microbubbles play an important role in the targeted adhesion of BM-MNCs on vascular endothelium.
Previous studies have indicated that platelet attachment to an inter- or sub-endothelial matrix of endothelial cells promotes selectin-mediated leukocyte adhesion to the damaged endothelium under the flow assay condition.20,21 Inhibition of P-selectin caused a marked inhibition of leukocyte adhesion at a high shear stress.20 P-selectin is a receptor for leukocytes and monocytes when its expression is induced on activated platelets and endothelium. This property facilitates rapid adhesion of leukocytes to endothelium in injury tissue regions and enhances platelet-leukocyte interactions at sites of inflammation. Endothelial P-selectin is located on membranes of Weibel-Plade bodies, the secretory granules of endothelium in which large multimers of von Willebrand factor (vWF) are stored.22 After cellular stimulation with agonists such as thrombin or histamine, P-selectin is rapidly expressed on the endothelial cell surface, making it an excellent candidate for directing adherence of unstimulated leukocytes toward endothelium within minutes of tissue injury.23 Furthermore, we have reported that BM-MNCs have a higher rolling and adhesive activities because of the greater expressions of adhesive molecules such as P-selectin compared with peripheral blood-derived leukocytes.15 In this study, we found that factors released from platelets stimulated by microbubble destruction are responsible for the attachment of platelets and BM-MNCs onto the endothelium and the induction of endothelial P-selectin and ICAM-1. The present data from the flow assay also confirm that platelet-derived factors play a key role for adhesion of BM-MNCs onto the endothelial cells under laminar flow. Although we previously showed that platelet-derived VEGF is mainly associated with the angiogenesis response by platelet implantation,3 preincubation with antibodies for VEGF, bFGF, or PDGF-BB showed no influence on the induction of endothelial P-selectin and ICAM-1 (unpublished observation). Further studies will be required to identify the platelet-derived proinflammatory cytokines responsible for induction of endothelial adhesion molecules. Taken together, these findings suggest that release of platelet-derived proinflammatory factors and direct interaction of platelet onto the endothelial matrix, initiated by US-microbubble destruction, is an underlying mechanism for adhesion of the transfused BM-MNCs on the endothelium under shear stress.
The previous studies showed that US-mediated destruction of microbubbles induces arteriogenesis response in the skeletal muscle, whereas angiogenesis response is transient and unlikely contributes to the increase in the regional blood flow.12 The arteriogenesis response consists of the formation of new arterioles, which presumably occurs when preexisting capillaries acquire SM coating, and an increase in the diameter of these newly formed and/or preexisting arterioles into channels with larger diameters.13 Compared with angiogenesis formed by capillary sprouting, arteriogenesis is often studied with the use of conventional angiography. Our angiography finding is consistent with the arteriogenesis response, and the immunohistological data suggest the angiogenesis and arteriogenesis response as evaluated by the increases in capillary numbers and SM coated arterioles, respectively. The neocapillary formation was observed in the day 28 samples, suggesting that the angiogenesis is a chronic response in our study. The controversy with the studies by Song et al12,13 may be attributable to the difference in the used microbubbles (albumin-coated Optison versus phospholipids-coated BR14). Recruitment of monocytes triggered by monocyte chemoattractant protein-1 was shown to induce arteriogenesis in inflammatory ischemic sites.24 Because BM-MNCs contain monocyte-lineage progenitor cells,15 it is plausible that recruitment of BM-MNCs contributes to arteriogenesis together with inflammation response by US+microbubble-mediated capillary rupturing.
In conclusion, the present study demonstrates that intravenous transfusion of BM-MNCs combined with US-destruction of microbubbles is an efficient targeted cell delivery system for therapeutic angiogenesis as well as arteriogenesis, in which the release of platelet-derived proinflammatory factors activated by Bubble+US plays a key role in the attachment of transplanted BM-MNCs onto the endothelial layer.
Acknowledgments
This study was supported in part by research grants from the Ministry of Education, Science, Sports, and Culture, Japan, the Study Group of Molecular Cardiology, the Japan Medical Association, Japan Smoking Foundation, and the Japan Heart Foundation.
References
Isner JM, Vale PR, Symes JF, Losordo DW. Assessment of risks associated with cardiovascular gene therapy in human subjects. Circ Res. 2001; 89: 389–400.
Tateishi-Yuyama E, Matsubara H, Murohara T, Ikeda U, Shintani S, Masaki H, Amano K, Kishimoto Y, Yoshimoto K, Akashi H, Shimada K, Iwasaka T. Therapeutic angiogenesis for patients with limb ischemia by autologous transplantation of bone-marrow cells: a pilot study and randomized controlled trial. Lancet. 2002; 360: 427–435.
Iba O, Matsubara H, Nozawa Y, Fujiyama S, Amano K, Mori Y, Kojima H, Iwasaka T. Angiogenesis by implantation of peripheral blood mononuclear cells and platelets into ischemic limbs. Circulation. 2002; 106: 2019–2025.
Shintani S, Murohara T, Ikeda H, Ueno T, Sasaki K, Duan J, Imaizumi T. Augmentation of postnatal neovascularization with autologous bone marrow transplantation. Circulation. 2001; 103: 897–903.
Bekeredjian R, Chen S, Frenkel PA, Grayburn PA, Shohet RV. Ultrasound-targeted microbubble destruction can repeatedly direct highly specific plasmid expression to the heart. Circulation. 2003; 108: 1022–1026.
Price RJ, Skyba DM, Kaul S, Skalak TC. Delivery of colloidal particles and red blood cells to tissue through microvessel ruptures created by targeted microbubble destruction with ultrasound. Circulation. 1998; 98: 1264–1267.
Lawrie A, Brisken AF, Francis SE, Tayler DI, Chamberlain J, Crossman DC, Cumberland DC, Newman CM. Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro. Circulation. 1999; 99: 2617–2620.
Shohet RV, Chen S, Zhou YT, Wang Z, Meidell RS, Unger RH, Grayburn PA. Echocardiographic destruction of albumin microbubbles directs gene delivery to the myocardium. Circulation. 2000; 101: 2554–2556.
Lindner JR, Kaul S. Delivery of drugs with ultrasound. Echocardiography. 2001; 18: 329–337.
Teupe C, Richter S, Fisslthaler B, Randriamboavonjy V, Ihling C, Fleming I, Busse R, Zeiher AM, Dimmeler S. Vascular gene transfer of phosphomimetic endothelial nitric oxide synthase (S1177D) using ultrasound-enhanced destruction of plasmid-local microbubbles improves vasoreactivity. Circulation. 2002; 105: 1104–1109.
Taniyama Y, Tachibana K, Hiraoka K, Namba T, Yamasaki K, Hashiya N, Aoki M, Ogihara T, Kaneda Y, Morishita R. Local delivery of plasmid DNA into rat carotid artery using ultrasound. Circulation. 2002; 105: 1233–1239.
Song J, Qi M, Kaul S, Price RJ. Stimulation of arteriogenesis in skeletal muscle by microbubble destruction with ultrasound. Circulation. 2002; 106: 1550–1555.
Song J, Cottler PS, Klibanov AL, Kaul S, Price RJ. Microvascular remodeling and accelerated hyperemia blood flow restoration in arterially occluded skeletal muscle exposed to ultrasonic microbubble destruction. Am J Physiol. 2004; 287: H2754–H2761.
Kamihata H, Matsubara H, Nishiue T, Fujiyama S, Tsutsumi Y, Ozono R, Masaki H, Mori Y, Iba O, Tateishi E, Kosaki A, Shintani S, Murohara T, Imaizumi T, Iwasaka T. Implantation of autologous bone marrow mononuclear cells into ischemic myocardium enhances collateral perfusion and regional function via side-supply of angioblasts, angiogenic ligands and cytokines. Circulation. 2001; 104: 1046–1052.
Fujiyama S, Amano K, Uehira K, Yoshida M, Nishiwaki Y, Nozawa Y, Jin D, Takai S, Miyazaki M, Egashira K, Imada T, Iwasaka T. Bone marrow monocyte lineage cells adhere on injured endothelium in a monocyte chemoattractant protein-1-dependent manner and accelerate reendothelialization as endothelial progenitor cells Circ Res. 2003; 93: 980–989.
Fisher NG, Christiansen JP, Leong-Poi H, Jayaweera AR, Linder JR, Kaul S. Myocardial and microcirculatory kinetics of BR14, a novel third-generation intravenous ultrasound contrast agent. J Am Coll Cardiol. 2002; 39: 530–537.
Fisher NG, Leong-Poi H, Sakuma T, Rim SJ, Bin JP, Kaul S. Detection of coronary stenosis and myocardial viability using a single intravenous bolus injection of BR14. J Am Coll Cardiol. 2002; 39: 523–529.
Amano K, Matsubara H, Iba O, Okigaki M, Fujiyama S, Imada T, Kojima H, Nozawa Y, Kawashima S, Yokoyama M, Iwasaka T. Enhancement of ischemia-induced angiogenesis by eNOS overexpression. Hypertension. 2003; 41: 156–162.
Lin F, Cordes K, Li L, Hood L, Couser WG, Shankland SJ, Igarashi P. Hematopoietic stem cells contribute to the regeneration of renal tubules after renal ischemia-reperfusion injury in mice. J Am Soc Nephrol. 2003; 14: 1188–1199.
Kuijper PH, Torres HG, Linden JA, Lammers JW, Sixma JJ, Koenderman L, Zwaginga JJ. Platelet-dependent primary hemostasis promotes selectin-an integrin-mediated neutrophil adhesion to damaged endothelium under flow conditions. Blood. 1996; 87: 3271–3281.
Theilmeier G, Lenaerts T, Remacle C, Collen D, Vermylen J, Hoylaerts MF. Circulating activated platelets assist THP-1 monocytoid/endothelial cell interaction under shear stress. Blood. 1999; 94: 2725–2734.
McEver RP, Beckstead JH, Moore KL, Marshall-Carlson L, Bainton DF. GMP-140, a platelet alpha-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest. 1989; 84: 92–99.
McEver RP. GMP-140: a receptor for neutrophils and monocytes on activated platelets and endothelium. J Cell Biochem. 1991; 45: 156–161.
Heil M, Schaper W. Influence of mechanical, cellular, and molecular factors on collateral artery growth (arteriogenesis). Circ Res. 2004; 95: 449–458.(Takanobu Imada; Tetsuya T)
Correspondence to Hiroaki Matsubara MD, Department of Cardiovascular Medicine, Kyoto Prefectural University School of Medicine, Kamigyo-ku, Kyoto, 602-8566, Japan. E-mail matsubah@koto.kpu-m.ac.jp
Abstract
Objective— Ultrasound (US)-mediated destruction of contrast microbubbles causes capillary rupturing that stimulates arteriogenesis, whereas intramuscular implantation (im) of bone marrow mononuclear cells (BM-MNCs) induces angiogenesis. We therefore studied whether US-targeted microbubble destruction combined with transplantation of BM-MNCs can enhance blood flow restoration by stimulating both angiogenesis and arteriogenesis.
Methods and Results— US-mediated destruction of phospholipid-coated microbubbles was applied onto ischemic hindlimb muscle and subsequently BM-MNCs were transfused. A significant enhancement in blood flow recovery after Bubble+US+BM-MNC infusion (34% increase, P<0.05) was observed compared with Bubble+US (25%). The ratio of capillary/muscle fiber increased by Bubble+US+BM-MNC-i.v (260%, P<0.01) than that in the Bubble+US group (172%), into which BM-MNCs were incorporated (angiogenesis). Smooth muscle -actin–positive arterioles were also increased, and angiography showed augmented collateral vessel formation (arteriogenesis). Platelet-derived proinflammatory factors activated by Bubble+US induces the expression of adhesion molecules (P-selectin and ICAM-1), leading to the attachment of transplanted BM-MNCs on the endothelium. Flow assay confirmed that the platelet-derived factors cause the adhesion of BM-MNCs onto endothelium under laminar flow.
Conclusions— This study demonstrates that the targeted delivery of BM-MNCs by US destruction of microbubbles enhances regional angiogenesis and arteriogenesis response, in which the release of platelet-derived proinflammatory factors activated by Bubble+US play a key role in the attachment of transplanted BM-MNCs onto the endothelial layer.
Ultrasound (US)-targeted microbubble destruction combined with transplantation of bone marrow mononuclear cells (BM-MNCs) enhances blood flow restoration by stimulating both angiogenesis and arteriogenesis. Release of proinflammatory factors from platelets activated by Bubble+US play a key role for adhesion of BM-MNCs on endothelium. This cell delivery system is efficient for therapeutic angiogenesis and arteriogenesis.
Key Words: angiogenesis ? vasculogenesis ? stem cell ? endothelial progenitor cell ? ultrasound
Introduction
Therapeutic angiogenesis, the ability to induce the formation of new blood vessels, is one of the most promising targets for regeneration therapy. To induce angiogenesis, investigators have delivered vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF/FGF2). or hypoxia-inducible factor-1/etoposide as recombinant proteins or genes.1 Intramuscular injection of bone marrow mononuclear cells (BM-MNCs) was shown to be feasible in patients or animals with ischemic limbs by supplying angiogenic factors and endothelial progenitors.2–4 A noninvasive cell delivery system that can target vascular endothelium would be a great advantage for manipulation of angiogenic cell therapy.
Ultrasound (US)-targeted microbubble destruction has been investigated as a new method for delivering drugs and genes to specific tissues.5–11 This method involves the attachment of drugs or genes to gas-filled microbubbles, which are then circulated through the intravascular space and mechanically destroyed within the target organ. Theoretically, one can target any anatomic site that is accessible by US, including selected damaged regions.5–11 Song et al have reported that US-targeted microbubble destruction causes capillary rupturing that stimulates arteriogenesis and an increase in blood flow in both normal12 and ischemic13 skeletal muscles, in which angiogenesis response is transient and unlikely contributes to chronic restoration of blood flow. We previously demonstrated that the recruitment of BM-MNCs and platelets stimulates angiogenesis response in ischemic muscles by releasing potent angiogenic factors, such as VEGF or bFGF, and supply of endothelial progenitors.3,14 Furthermore, we have recently reported that systemically transplanted BM-MNCs can be firmly attached onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore examined whether US-targeted microbubble destruction combined with intravenous transplantation of BM-MNCs causes angiogenesis response as well as arteriogenesis in a rat model with an ischemic hindlimb, leading to an enhancement of chronic blood flow restoration. Interestingly, we found that platelet-derived factors activated by US-targeted microbubble destruction induce the expression of adhesive molecules on the endothelium and subsequent attachment of transplanted BM-MNCs, resulting in an enhancement of formation of neocapillaries and new arterioles and an increase in regional blood flow recovery.
Materials and Methods
Microbubble Preparations
BR14 (Bracco Diagnostics) is a new ultrasound contrast agent, consisting of perfluorocarbon-containing microbubbles stabilized by a phospholipid monolayer.16,17 The suspension of gas microbubbles is reconstituted immediately before use by injecting 5 mL of 0.9% sodium chloride. The mean diameter of bubbles is 2.3 to 2.5 μm and the bubble concentration is 6x108 per ml. BR14 was administered as a 1 mL bolus injection via contralateral femoral vein.
Isolation of BM-MNC and Characterization of Endothelial-Lineage Cells
BM-MNCs were isolated from SEA/LEW rat femoral bone and centrifuged by density gradient (Lymphoprep; Nycomed). The MNC fraction was labeled with red-fluorescence cell linker (PKH26-GL; Sigma).2,3,14 Endothelial lineage cells were analyzed by fluorescence-activated cell sorter (FACS) using DiI-acetylated LDL (acLDL) incorporation (Biogenesis) and Ulex lectin binding (Sigma) as described previously.2,3,14
Hindlimb Ischemia, US Application, and Transfusion of BM-MNCs
Unilateral hindlimb ischemia was induced by resecting the left femoral artery as described.2,18 Arteriogenesis that restores the regional blood flow was reported to be observed 3 days after induction of limb ischemia of rats.12,13 We therefore performed the infusion of microbubble at day 3 after limb ischemia to efficiently deliver microbubble BR14 to the ischemic site. We divided the rats into the following five groups: (1) Control (saline injection, n=8); (2) Bubble (1 mL BR14)+US (n=8); (3) Bubble+US+BM-MNC-i.v. (n=8), BM-MNC (2x107 in 1 mL saline) injected from contralateral femoral vein 1 minute after Bubble+US; (4) BM-MNC-i.m. (n=8), BM-MNC (2x107) intramuscularly implanted into the ischemic limb; and (5) BM-MNC-i.v. (n=8); BM-MNC (2x107 in 1 mL saline). BM-MNC (2x107) included 0.3±0.04x107 platelets.
The skin overlying the ischemic thigh muscle to be treated was reflected back, ultrasound gel was placed over the ischemic muscle, and 1-MHz transducer (S-probe, Effective Radiating Area: 0.9 cm2; Ito Co Ltd) was held 3 mm over the muscle and to keep the skin temperature around 37°C. A continuous sinusoidal wave ultrasound (1MHz, 2W/cm2, Beam Nonuniformity Ratio: 3.6; ITO-US-700) was applied for 1 minute immediately after microbubble injection. To clarify whether the microbubbles really reach ischemic tissues after infusion from contralateral femoral vein, the delivery of injected microbubbles to the ischemic area was examined by a diagnostic ultrasound echography (Sonos-5500) equipped with an ultraband S12 sector transducer.
Evaluation of Neocapillary and Immunohistochemical Analysis
Four pieces of ischemic tissues from the adductor and semimembranous muscles were obtained 28 days after limb ischemia. Frozen sections were stained with antibodies for smooth muscle (SM) -actin (Sigma) and anti-factor VIII (DAKO) antibodies. The appropriate secondary antibodies conjugated with fluorescein isothiocyanate (FITC) or rhodamine were used. Ten fields from 2 muscle samples of each animal were randomly selected for the vessel count. To ensure that vessel densities were not overestimated as a consequence of myocyte atrophy or underestimated because of interstitial edema, the capillary/muscle fiber ratio was determined.3,18
Electron Microscope
Femoral arteries in skeletal muscles on which targeted US was applied were isolated after Bubble+US treatment (n=4), fixed by 2.5% glutaraldehyde and 1.5% osmium acid, dried, and viewed by electron microscope (HITATI S-700).15
Cell Culture and Platelet-Rich Plasma Isolation
Human umbilical vein endothelial cells (HUVECs; Kurabo) were cultured in HuMedia-EG2 medium. For use in the study apparatus, HUVECs (2nd or 3rd passages) were plated on 22-mm fibronectin-coated cover slips.15 Peripheral blood was drawn from healthy volunteers and mixed with 0.1 volume of citrate (108 mmol/L). Whole blood was centrifuged at 150g for 10 minutes to harvest platelet-rich plasma (PRP).
Adhesion Assay Under Laminar Flow and Immunofluorescence Study
HUVECs were incubated for 30 minutes with serum-free medium with or without freshly prepared 10% platelet-rich plasma, and then stimulated by (1) 10% Bubble+US (1MHz, 1.5W, 30 seconds) in 10% PRP (Bubble+US with platelet group), (2) 10% Bubble+US in PRP (Bubble+US without platelet), (3) incubation media (in which 10% platelet-rich medium was stimulated by 10% Bubble+US), or (4) US with platelet (in which 10% platelet-rich medium was stimulated by US alone), n=6 in each experiment. HUVECs were placed on the cold plate to prevent the heating by the US, and adhesion assay under laminar flow was performed as previously described.15 For further detail, please see the supplemental Methods, available online at http://atvb.ahajournals.org.
Stimulated HUVECs were fixed by 4% paraformaldehyde 1 hour after stimulation and incubated with anti–P-selectin (R&D Systems, Inc) and anti-platelet glycoprotein (GP)-Ib antibody (DAKO). The appropriate secondary antibodies conjugated with FITC or rhodamine were used. Nuclei were stained with 4',6'-diamidino-2-phenylindole dihydrochloride (DAPI) and viewed by a fluorescence microscope (Olympus OX71).
Laser Doppler Perfusion Image and Angiography
Laser doppler perfusion image (LDPI) and angiography were performed as previously described3,18 For further detail, please see the supplemental Methods.
Statistical Analysis
Statistical analyses were performed with 1-way ANOVA followed by pair-wise contrasts using the Dunnett test. Data (mean±SE) were considered statistically significant when P<0.05.
Results
Incidence of Endothelial-Lineage Cells in BM-MNCs
FACS analysis indicated that 28±1.8% and 31±1.5% of BM-MNCs incorporated DiI-acLDL and bound Ulex-lectin (n=5), respectively, and 20±1.2% of cells were positive for both markers. Endothelial-lineage cells were considered to be included in this fraction as reported.2,3,14
Laser Doppler Blood Perfusion
Subcutaneous blood perfusion was analyzed by LDPI imaging (Figure 1). Intravenous injection of BM-MNCs (BM-MNC-i.v.) did not cause a significant increase in the blood flow recover compared with the control (saline injection; Figure 1). Treatment with Bubble+US without BM-MNCs-i.v. showed a moderate increase (25±2% at Day 28 versus control, P<0.05), whereas the combination of Bubble+US and BM-MNCs-i.v. induced a further increase (34±2% versus control at Day 28, P<0.01), which was significantly higher than that of Bubble+US manipulation alone (P<0.05). The blood perfusion recovery by Bubble+US+BM-MNC-i.v. was comparable to blood perfusion by BM-MNC-i.m. (38±3%). Blood perfusion after US+BM-MNC-i.v. without Bubble did not significantly differ from those in BM-MNC-i.v. alone or control groups (3±1% versus control at Day 28, n=8; data not shown), suggesting that the combination of BM-MNCs infusion with Bubble+US significantly improves the blood flow recovery after limb ischemia compared with each manipulation alone, and that the efficient cell delivery system depends on US-mediated destruction of microbubbles.
Figure 1. Increase of blood flow by US-microbubble-mediated transfusion of BM-MNCs. A, Representative LDPI. Greater blood perfusion (red to yellow) in ischemic limbs was observed in Bubble+US+BM-MNC infusion, BM-MNC-i.m., and Bubble+US groups in contrast with the untreated control and BM-MNC-i.v. groups. B, Computer-assisted analyses of LDPI revealed a significantly greater blood perfusion in the Bubble+US+BM-MNC-i.v., BM-MNC-i.m., and Bubble+US groups than in the control group. The values shown are mean±SE (n=8) at each time point. *P<0.05, **P<0.01 vs control.
Neocapillary and Arteriole Formation
Formation of capillaries and arterioles was evaluated by factor VIII–positive and SM -actin–positive vessels. The ratio of capillary/muscle fiber (factor VIII–positive and SM -actin–negative vessel) was significantly increased in the Bubble+US+BM-MNC-i.v. group (260±15% versus control, P<0.01), which was greater (P<0.01) than that of Bubble+US group (172±11% versus control, P<0.05). The ratio of arteriole/muscle fiber (factor VIII–positive and SM -actin–positive vessel) was also increased by Bubble+US+BM-MNC-i.v. (188±10% versus control, P<0.01), which was greater (P<0.05) than that in the Bubble+US group (146±9% versus control, P<0.05; Figure 2A). Transfused BM-MNCs (labeled with red fluorescence) were found to be incorporated into microvessels by Bubble+US+BM-MNC-i.v. (arrows in Figure 2B). These findings suggest that Bubble+US+BM-MNC-i.v. enhances both angiogenesis as well as arteriogenesis response.
Figure 2. Formation of arterioles/capillaries and incorporation of transfused BM-MNCs. A, Vessels were immunostained with anti–SM -actin and anti–factor VIII antibodies. Numbers for arterioles (positive for both SM -actin and factor VIII staining) and capillaries (SM -actin–negative and factor VIII–positive) were evaluated as a vessel number/muscle fiber ratio (n=8, each group). *P<0.05, **P<0.01 vs the saline-injected control. Bar=50 μm. B, Incorporation of transfused BM-MNCs (labeled with red-fluorescence). Red-labeled cells were incorporated into factor VIII–positive capillaries (arrows in the merged image) in the Bubble+US+BM-MNC-i.v. group. Data shown are representative of 5 different animals. Bar=200 μm.
To examine the specificity of this cell delivery system, we examined the distribution of transfused BM-MNCs in other tissues (kidney, spleen, liver, heart, pancreas, small intestines, and brain) 3 weeks after targeted cell delivery by Bubble+US+BM-MNCs-iv. Small numbers of labeled BM-MNCs were detected in the kidney (mainly in the tubule) and greater numbers of BM-MNCs were observed in the spleen, whereas no labeled cells were detected in other tissues (Figure I, available online at http://atvb.ahajournals.org).
Arteriogenesis Response Evaluated by Angiography
Compared with angiogenesis formed by capillary sprouting, arteriogenesis is often studied with the use of angiography. On the postoperative day 28, all animals were subjected to iliac angiography. Representative angiograms (n=4 in each group) are shown in Figure II (available online at http://atvb.ahajournals.org), in which arrows indicate the ligated ends of femoral arteries. Collateral vessels in the thigh area were quantitatively counted using 5-mm2 grids.3,18 An apparent increase in collateral vessel formation was observed in the Bubble+US+BM-MNC-i.v. and BM-MNC-i.m. (4.2±0.2-fold and 4.3±0.2-fold, n=6, respectively; P<0.001) compared with that in the BM-MNC-i.v. group. The increase in the Bubble+US group was 1.8±0.1-fold (P<0.05) when compared with that in the BM-MNC-i.v. group, which was significantly smaller (P<0.01) than that of the Bubble+US+BM-MNC-i.v. group. There was no significant difference between control (saline infusion) and BM-MNC-i.v. groups.
Delivery of Microbubbles to the Ischemic Muscle
Arteriogenesis that restores the regional blood flow was observed 3 days after induction of limb ischemia of rats.12,13 We therefore performed the infusion of microbubble at day 3 after limb ischemia to efficiently deliver microbubble BR14 to the ischemic site. We further examined whether the transfused microbubbles really reach the hindlimb muscle at 3 days after induction of ischemia. Apparent increase in the contrast densities in the ischemic thigh muscle (indicated by arrows in Figure III, available online at http://atvb.ahajournals.org) was observed 15 seconds after injection of microbubbles in both control (normal) and ischemic limbs compared with the pre-images before injection of microbubbles. The increase in contrast shadow diminished after 1 minute of US stimulation, indicating that microbubbles really reach ischemic hindlimb muscles after venous infusion.
Electromicroscopy
A previous study reported the presence of small holes in the endothelial cells treated with Bubble+US,11 whereas we could not detect the presence of apparent small holes in the vascular endothelium in the skeletal muscle on which the Bubble+US was applied. Neither adhesion of MNCs nor formation of fibrin network including platelets was detected on the surface of normal endothelium or endothelium stimulated by US without microbubbles. Interestingly, we found the attachment of platelets associated with fibrin network and MNCs on the surface of endothelium in all arteries treated by Bubble+US+BM-MNCs infusion (n=6; Figure 3), whereas no adhesion of platelets or MNCs was detected in the US+BM-MNC group without microbubbles (n=6; data not shown).
Figure 3. Electron-microscopic appearance of endothelium stimulated by Bubble-US+BM-MNCs. Electron-microscopy showing normal endothelium covered with a monolayer coat (Normal) and appearance of Bubble+US+BM-MNCs i.v.-stimulated endothelium. Attachment of both platelets associated with fibrin network and MNCs was observed on all samples treated by Bubble+US+BM-MNCs i.v. Representative figures are shown (n=4 in each group). Bar=10 μm.
Induction of Adhesion Molecules
We have reported that transplanted BM-MNCs can firmly attach onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore examined the expression profile of adhesive molecules on HUVECs treated by Bubble+US. The expression of adhesive molecules (P-selectin or ICAM-1) was not induced when HUVECs were stimulated by Bubble+US in the medium without platelets. Because we found that adhesion of the fibrin network including platelets was consistently observed on the surface of endothelium in all Bubble+US-treated samples (Figure 3), we next examined the involvement of platelet-derived factors in the adhesion of BM-MNCs. Interestingly, the expression of P-selectin (red fluorescence, right panel in Figure 4) and ICAM-1 was markedly induced in both HUVECs and attaching glycoprotein Ib–positive platelets (merged, yellow), when HUVECs were stimulated by Bubble+US in the medium including platelets (only P-selectin data shown in Figure 4).
Figure 4. Adhesion of platelets and induction of adhesion molecules on Bubble+US-stimulated HUVECs. The presence of platelets and expression of P-selectin are shown by immunostaining with anti–GP-Ib (green fluorescence) and anti–P-selectin (red fluorescence) antibodies and DAPI staining (blue fluorescence). The expression of P-selectin and adhesion of platelets were observed on HUVECs, when HUVECs were stimulated by Bubble+US in the medium containing 10% platelets or exposed to the incubation medium (in which 10% platelet-including medium was stimulated by Bubble+US). The expression of P-selectin (red fluorescence) on HUVECs and attaching platelets was markedly induced. Bar=50 μm.
We also studied whether factors released from the platelets stimulated by Bubble+US are involved in the induction of adhesion molecules. Therefore, we examined the effect of the incubation medium, in which platelet-including medium was stimulated by Bubble+US, on the induction of adhesion molecules. The addition of incubation medium caused the apparent increase in P-selectin expression on HUVECs (red fluorescence, right panel in Figure 4), suggesting that the factors released from the platelets stimulated by Bubble+US are closely involved in the induction of adhesion molecules on HUVECs. Treatment of HUVECs by US alone without microbubbles in the medium including platelets did not induce any expression of adhesive molecules (data not shown). These findings suggest that activation of platelets by Bubble+US and release of platelet-derived proinflammatory factors play a key role in the induction of adhesion molecules in the endothelial cells.
Laminar Flow Assay
We next studied whether the adhesive activity of BM-MNCs on the endothelium was actually modulated by Bubble+US-activated platelets under laminar flow condition. HUVECs were stimulated by Bubble+US in the medium containing platelets and then the adhesion ratio of BM-MNCs on HUVECs was evaluated under laminar flow as previously reported.15 The presence of platelets in the medium markedly increased the adhesion ratio of BM-MNCs (3.5-fold, P<0.01) compared with Bubble+US without platelets (Figure 5). Moderate increase (1.9-fold, P<0.01) was observed in the incubation medium group (in which medium including platelets was stimulated by Bubble+US and then added to HUVECs) compared with the ratio in the untreated HUVECs (control). Treatment of HUVECs by US stimulation without microbubbles in the medium including platelets or addition of control medium including platelet alone (without US stimulation) did not cause a significant increase in the adhesion ratio of BM-MNCs compared with the ratio in the untreated HUVEC (data not shown).
Figure 5. Laminar flow assay of BM-MNCs adhesion. Adhesive activities of BM-MNCs were quantified on HUVEC monolayer under laminar flow. Adhesive cells are indicated by yellow arrows. Cells were considered to be adherent after 10 seconds of stable contact with the monolayer. HUVECs were stimulated by Bubble+US (1MHz, 1.5W, 30 seconds) in the medium with 10% platelets (n=6), without platelets (n=6), or in the incubation medium (in which 10% platelet-including medium was stimulated by Bubble+US, n=6). *P<0.05, **P<0.01 vs the untreated control. Bar=100 μm.
Discussion
Angiogenic cell therapy by intramuscular implantation of autologous BM-MNCs was shown to be feasible in patients with ischemic limbs.2 Because intramuscular implantation is invasive at the injected sites, the development of a noninvasive cell delivery system that can target vascular endothelium would be a great advantage for the manipulation of angiogenic cell therapy. A new delivery system of drugs or genes has been developed using US-targeted microbubble destruction; the drugs or genes that attach onto gas-filled microbubbles circulate through the intravascular space and are mechanically destroyed within the target organ by ultrasound,5–11 whereas no studies were reported to determine whether this method is feasible for delivering the "cells" to specific vascular sites.
US-targeted microbubble destruction was reported to cause an inflammatory action on the cell surface by making small holes that revert to a normal appearance within 24 hours.10,11 Song et al have reported that US-targeted microbubble destruction causes capillary rupturing that stimulates arteriogenesis and an increase in blood flow in both normal12 and ischemic13 skeletal muscles, in which angiogenesis response is transient and unlikely contributes to chronic restoration of blood flow. They concluded that arteriogenesis response rather than angiogenesis plays a major role in US+microbubble-stimulated blood flow recovery. We previously demonstrated that the recruitment of BM-MNCs and platelets stimulates angiogenesis response in the ischemic muscles by releasing potent angiogenic factors, such as VEGF or bFGF, and supply of endothelial progenitors.3,14 Furthermore, we have recently reported that systemically transplanted BM-MNCs can be firmly attached onto the injured vascular endothelium in an adhesive molecule–dependent manner.15 We therefore expanded the previous studies by Song et al12,13 and examined whether US-targeted microbubble destruction combined with intravenous transplantation of BM-MNCs causes both angiogenesis and arteriogenesis response in an ischemic hindlimb model, leading to a greater enhancement of blood flow restoration.
We found that (1) intravenous infusion of BM-MNCs combined with US-mediated destruction of microbubbles markedly enhances the restoration of regional blood perfusion in ischemic hindlimbs by stimulating both chronic angiogenesis and arteriogenesis response, and (2) release of platelet-derived proinflammatory factors activated by US-mediated destruction of microbubbles causes the adhesion of transfused BM-MNCs on endothelium by inducing the expression of adhesion molecules (P-selectin and ICAM-1). We found that BM-MNCs transfused intravenously were trapped by the spleen and a few BM-MNCs were present in the renal tubules. Considering that BM-derived hematopoietic stem cells were reported to transdifferentiate to renal tubular cells and improve renal function in the ischemia-reperfusion injury model,19 the present study establishes for the first time that targeted delivery of BM-MNCs by US destruction of microbubbles is an efficient cell delivery system for therapeutic angiogenesis and arteriogenesis, and that the presence of platelets and/or platelet-derived proinflammatory factors activated by US+microbubbles play an important role in the targeted adhesion of BM-MNCs on vascular endothelium.
Previous studies have indicated that platelet attachment to an inter- or sub-endothelial matrix of endothelial cells promotes selectin-mediated leukocyte adhesion to the damaged endothelium under the flow assay condition.20,21 Inhibition of P-selectin caused a marked inhibition of leukocyte adhesion at a high shear stress.20 P-selectin is a receptor for leukocytes and monocytes when its expression is induced on activated platelets and endothelium. This property facilitates rapid adhesion of leukocytes to endothelium in injury tissue regions and enhances platelet-leukocyte interactions at sites of inflammation. Endothelial P-selectin is located on membranes of Weibel-Plade bodies, the secretory granules of endothelium in which large multimers of von Willebrand factor (vWF) are stored.22 After cellular stimulation with agonists such as thrombin or histamine, P-selectin is rapidly expressed on the endothelial cell surface, making it an excellent candidate for directing adherence of unstimulated leukocytes toward endothelium within minutes of tissue injury.23 Furthermore, we have reported that BM-MNCs have a higher rolling and adhesive activities because of the greater expressions of adhesive molecules such as P-selectin compared with peripheral blood-derived leukocytes.15 In this study, we found that factors released from platelets stimulated by microbubble destruction are responsible for the attachment of platelets and BM-MNCs onto the endothelium and the induction of endothelial P-selectin and ICAM-1. The present data from the flow assay also confirm that platelet-derived factors play a key role for adhesion of BM-MNCs onto the endothelial cells under laminar flow. Although we previously showed that platelet-derived VEGF is mainly associated with the angiogenesis response by platelet implantation,3 preincubation with antibodies for VEGF, bFGF, or PDGF-BB showed no influence on the induction of endothelial P-selectin and ICAM-1 (unpublished observation). Further studies will be required to identify the platelet-derived proinflammatory cytokines responsible for induction of endothelial adhesion molecules. Taken together, these findings suggest that release of platelet-derived proinflammatory factors and direct interaction of platelet onto the endothelial matrix, initiated by US-microbubble destruction, is an underlying mechanism for adhesion of the transfused BM-MNCs on the endothelium under shear stress.
The previous studies showed that US-mediated destruction of microbubbles induces arteriogenesis response in the skeletal muscle, whereas angiogenesis response is transient and unlikely contributes to the increase in the regional blood flow.12 The arteriogenesis response consists of the formation of new arterioles, which presumably occurs when preexisting capillaries acquire SM coating, and an increase in the diameter of these newly formed and/or preexisting arterioles into channels with larger diameters.13 Compared with angiogenesis formed by capillary sprouting, arteriogenesis is often studied with the use of conventional angiography. Our angiography finding is consistent with the arteriogenesis response, and the immunohistological data suggest the angiogenesis and arteriogenesis response as evaluated by the increases in capillary numbers and SM coated arterioles, respectively. The neocapillary formation was observed in the day 28 samples, suggesting that the angiogenesis is a chronic response in our study. The controversy with the studies by Song et al12,13 may be attributable to the difference in the used microbubbles (albumin-coated Optison versus phospholipids-coated BR14). Recruitment of monocytes triggered by monocyte chemoattractant protein-1 was shown to induce arteriogenesis in inflammatory ischemic sites.24 Because BM-MNCs contain monocyte-lineage progenitor cells,15 it is plausible that recruitment of BM-MNCs contributes to arteriogenesis together with inflammation response by US+microbubble-mediated capillary rupturing.
In conclusion, the present study demonstrates that intravenous transfusion of BM-MNCs combined with US-destruction of microbubbles is an efficient targeted cell delivery system for therapeutic angiogenesis as well as arteriogenesis, in which the release of platelet-derived proinflammatory factors activated by Bubble+US plays a key role in the attachment of transplanted BM-MNCs onto the endothelial layer.
Acknowledgments
This study was supported in part by research grants from the Ministry of Education, Science, Sports, and Culture, Japan, the Study Group of Molecular Cardiology, the Japan Medical Association, Japan Smoking Foundation, and the Japan Heart Foundation.
References
Isner JM, Vale PR, Symes JF, Losordo DW. Assessment of risks associated with cardiovascular gene therapy in human subjects. Circ Res. 2001; 89: 389–400.
Tateishi-Yuyama E, Matsubara H, Murohara T, Ikeda U, Shintani S, Masaki H, Amano K, Kishimoto Y, Yoshimoto K, Akashi H, Shimada K, Iwasaka T. Therapeutic angiogenesis for patients with limb ischemia by autologous transplantation of bone-marrow cells: a pilot study and randomized controlled trial. Lancet. 2002; 360: 427–435.
Iba O, Matsubara H, Nozawa Y, Fujiyama S, Amano K, Mori Y, Kojima H, Iwasaka T. Angiogenesis by implantation of peripheral blood mononuclear cells and platelets into ischemic limbs. Circulation. 2002; 106: 2019–2025.
Shintani S, Murohara T, Ikeda H, Ueno T, Sasaki K, Duan J, Imaizumi T. Augmentation of postnatal neovascularization with autologous bone marrow transplantation. Circulation. 2001; 103: 897–903.
Bekeredjian R, Chen S, Frenkel PA, Grayburn PA, Shohet RV. Ultrasound-targeted microbubble destruction can repeatedly direct highly specific plasmid expression to the heart. Circulation. 2003; 108: 1022–1026.
Price RJ, Skyba DM, Kaul S, Skalak TC. Delivery of colloidal particles and red blood cells to tissue through microvessel ruptures created by targeted microbubble destruction with ultrasound. Circulation. 1998; 98: 1264–1267.
Lawrie A, Brisken AF, Francis SE, Tayler DI, Chamberlain J, Crossman DC, Cumberland DC, Newman CM. Ultrasound enhances reporter gene expression after transfection of vascular cells in vitro. Circulation. 1999; 99: 2617–2620.
Shohet RV, Chen S, Zhou YT, Wang Z, Meidell RS, Unger RH, Grayburn PA. Echocardiographic destruction of albumin microbubbles directs gene delivery to the myocardium. Circulation. 2000; 101: 2554–2556.
Lindner JR, Kaul S. Delivery of drugs with ultrasound. Echocardiography. 2001; 18: 329–337.
Teupe C, Richter S, Fisslthaler B, Randriamboavonjy V, Ihling C, Fleming I, Busse R, Zeiher AM, Dimmeler S. Vascular gene transfer of phosphomimetic endothelial nitric oxide synthase (S1177D) using ultrasound-enhanced destruction of plasmid-local microbubbles improves vasoreactivity. Circulation. 2002; 105: 1104–1109.
Taniyama Y, Tachibana K, Hiraoka K, Namba T, Yamasaki K, Hashiya N, Aoki M, Ogihara T, Kaneda Y, Morishita R. Local delivery of plasmid DNA into rat carotid artery using ultrasound. Circulation. 2002; 105: 1233–1239.
Song J, Qi M, Kaul S, Price RJ. Stimulation of arteriogenesis in skeletal muscle by microbubble destruction with ultrasound. Circulation. 2002; 106: 1550–1555.
Song J, Cottler PS, Klibanov AL, Kaul S, Price RJ. Microvascular remodeling and accelerated hyperemia blood flow restoration in arterially occluded skeletal muscle exposed to ultrasonic microbubble destruction. Am J Physiol. 2004; 287: H2754–H2761.
Kamihata H, Matsubara H, Nishiue T, Fujiyama S, Tsutsumi Y, Ozono R, Masaki H, Mori Y, Iba O, Tateishi E, Kosaki A, Shintani S, Murohara T, Imaizumi T, Iwasaka T. Implantation of autologous bone marrow mononuclear cells into ischemic myocardium enhances collateral perfusion and regional function via side-supply of angioblasts, angiogenic ligands and cytokines. Circulation. 2001; 104: 1046–1052.
Fujiyama S, Amano K, Uehira K, Yoshida M, Nishiwaki Y, Nozawa Y, Jin D, Takai S, Miyazaki M, Egashira K, Imada T, Iwasaka T. Bone marrow monocyte lineage cells adhere on injured endothelium in a monocyte chemoattractant protein-1-dependent manner and accelerate reendothelialization as endothelial progenitor cells Circ Res. 2003; 93: 980–989.
Fisher NG, Christiansen JP, Leong-Poi H, Jayaweera AR, Linder JR, Kaul S. Myocardial and microcirculatory kinetics of BR14, a novel third-generation intravenous ultrasound contrast agent. J Am Coll Cardiol. 2002; 39: 530–537.
Fisher NG, Leong-Poi H, Sakuma T, Rim SJ, Bin JP, Kaul S. Detection of coronary stenosis and myocardial viability using a single intravenous bolus injection of BR14. J Am Coll Cardiol. 2002; 39: 523–529.
Amano K, Matsubara H, Iba O, Okigaki M, Fujiyama S, Imada T, Kojima H, Nozawa Y, Kawashima S, Yokoyama M, Iwasaka T. Enhancement of ischemia-induced angiogenesis by eNOS overexpression. Hypertension. 2003; 41: 156–162.
Lin F, Cordes K, Li L, Hood L, Couser WG, Shankland SJ, Igarashi P. Hematopoietic stem cells contribute to the regeneration of renal tubules after renal ischemia-reperfusion injury in mice. J Am Soc Nephrol. 2003; 14: 1188–1199.
Kuijper PH, Torres HG, Linden JA, Lammers JW, Sixma JJ, Koenderman L, Zwaginga JJ. Platelet-dependent primary hemostasis promotes selectin-an integrin-mediated neutrophil adhesion to damaged endothelium under flow conditions. Blood. 1996; 87: 3271–3281.
Theilmeier G, Lenaerts T, Remacle C, Collen D, Vermylen J, Hoylaerts MF. Circulating activated platelets assist THP-1 monocytoid/endothelial cell interaction under shear stress. Blood. 1999; 94: 2725–2734.
McEver RP, Beckstead JH, Moore KL, Marshall-Carlson L, Bainton DF. GMP-140, a platelet alpha-granule membrane protein, is also synthesized by vascular endothelial cells and is localized in Weibel-Palade bodies. J Clin Invest. 1989; 84: 92–99.
McEver RP. GMP-140: a receptor for neutrophils and monocytes on activated platelets and endothelium. J Cell Biochem. 1991; 45: 156–161.
Heil M, Schaper W. Influence of mechanical, cellular, and molecular factors on collateral artery growth (arteriogenesis). Circ Res. 2004; 95: 449–458.(Takanobu Imada; Tetsuya T)