Abstract
Shear stress can enhance differentiation of human embryonic stem cells (hESC) to vascular cells. We tested the hypothesis that intra-arterial hESC injection will lead to arteriogenesis while intramuscular injection will have no effect on vascularization.
Methods and Results
The superficial femoral arteries were excised on both hind limbs in athymic rats. hESC (2×10 6 ) were injected intra-arterially (shear stress) or intramuscular (no shear stress) in one limb after arterial excision. Blood flow, muscle perfusion, and number of arteries/mm 2 muscle were studied at 10 and 21 days after injection.
Blood flow in the common iliac artery improved significantly at 10 days after intra-arterial injection of hESC (22±9%, P <.02), and tight muscle perfusion improved significantly both at 10 and 21 days (9±2%, 16±5% respectively, both P <.02). In comparison, intramuscular injection of hESC did not affect blood flow at 10 and 21 days (−3±10% and 4±6%, respectively), while perfusion showed no significant effect of hESC injection after 10 days (1±8%) and was increased 21 days after hESC injection (11±5%, P =.03). Arterial density did not improve after intra-arterial hESC injection at 10 days (15±13%, P =.15) and significantly improved at 21 days (13±4%, P <.05). No significant change was demonstrated after intramuscular injection.
Summary
Intra-arterial injection of hESC resulted in moderate improvement of flow and perfusion and increased number of arteries in the ischemic hind limb. No consistent change in perfusion, flow, and number of arteries was observed after intramuscular injection.
1
Introduction
Human embryonic stem cells (hESC) can potentially differentiate to multiple cell types such as endothelial and smooth muscle cells as well as cardiac myocytes . Given relevant stimuli, which can simulate the in vivo conditions, hESC differentiation can occur in tissue culture dish . To attain the complete milieu for induction of hESC differentiation, animal models may be more appropriate as multiple factors, part of which are still unknown, exists in such models.
Blood vessel formation in the embryo involves differentiation of ESC to multiple cell types. In the adult, on the other hand, two distinct processes involving a limited number of cells are identified: capillary sprouting involving endothelial cells and pericytes and arteriogenesis involving endothelial, smooth muscle cells, and macrophages . Both mechanisms are milieu dependent in which local expression of growth factors secondary to hypoxia (capillary sprouting) and mechanical shear stress (arteriogenesis) play an important role.
The present study was set to test the hypothesis that intra-arterial hESC can produce arteriogenesis in the adult animal while intramuscular injection of hESC will not have such an effect. We assumed that shear stress plays a fundamental role in ESC differentiation that eventually produces effective arteriogenesis.
To test the hypothesis, we injected undifferentiated hESC to athymic rats that underwent bilateral femoral artery occlusion. We compared the effects of hESC injected intramuscular to those of hESC injected intra-arterial on arteriogenesis as measured by changes in flow, perfusion, and number of arteries/mm 2 of muscle in the ischemic limb.
2
Methods
2.1
Overall study design
Bilateral hind limb ischemia was induced in athymic rats ( Foxn1 rnu , Harlan, Israel) by surgical disruption of the femoral artery continuity. Undifferentiated hESC or control solution where injected in one limb while the other limb served as a control. To test the effects of shear stress on hESC contribution to arteriogenesis, we injected hESC intra-arterially in athymic rat model of hind limb ischemia and compared the effects on arteriogenesis to rats injected with hESC into the muscles of the ischemic hind limb (no shear stress). At 10 and 21 days, flow, perfusion, and the number of arterial blood vessels/mm 2 muscle were measured in the two limbs. Seven groups of rats were studied. Rats injected intra-arterially with hESC (10 and 21 days), rat injected intramuscular with hESC (10 and 21 days), rats injected intra-arterially with saline (10 and 21 days), and rats injected intramuscular and intra-arterial saline (10 days). The last group of animals was studied based on preliminary observation that intra-arterial injection of saline alone may result in increased flow. Table 1 summarizes the groups comprising the study. Rats were kept under best conditions and all procedures were performed avoiding any unnecessary stress or pain, according to the Guide for the Care and Use of Laboratory Animals . Ethics approval for this study was given under the code 20-02-2004 by the Technion, IIT, Haifa, Israel, ethics commission.
| Group | 10 d i.a. | 21 d i.a. | 10 d i.m. | 21 d i.m. | 10 d i.m. control | 21 d i.m. control | 10 d i.a. control |
|---|---|---|---|---|---|---|---|
| i.a. injection (in 0.5ml) | 2×10 6 hESC | 2×10 6 hESC | – | – | – | – | Saline |
| i.m. injection (in 0.3ml) | – | – | 2×10 6 hESC | 2×10 6 hESC | Saline | Saline | Saline |
During the time of investigation, none of the rats had to be sacrificed due to complications after surgery, severe infections or weight loss. No limb loss was observed.
2.2
Cell culture and injection preparations
I4 cells were cultured with at least two passages on matrigel (Dickinson, Mountain View, CA, USA) before injection with mouse embryonic fibroblast conditioned media composed of 80% KnockOut DMEM medium (GIBCO-BRL, Grand Island, NY, USA), supplemented with 20% KnockOut SR (a serum-free formulation) (GIBCO-BRL), 1 mM glutamine (GIBCO-BRL), 0.1 mM β-mercaptoethanol (Sigma-Aldrich, St. Louis, MO, USA), 1% nonessential amino acids stock (GIBCO-BRL) and 4 ng/ml basic fibroblast growth factor (GIBCO-BRL).
Cells were routinely assayed for karyotype and mycoplasma infection (Biological industries, Beit-haeemek, Israel). Cells were live-stained for SSEA-4 (R&D Systems, Minneapolis, MN,USA) in order to assure more than 90% of the cells are undifferentiated and could be used for injection. Cells in culture were incubated 1 h with the primary antibody at 37°C and washed three times with phosphate-buffered saline. Cy3-conjugated antimouse secondary antibody was added for 30 min incubation in room temperature (Santa Cruz, CA, USA). Cells positive for Cy3 staining were counted and their percentage of all cells was calculated. hESC viability prior to and after injection through the 26G catheter was tested. Injection through the 26G catheter had no effect on cell number and viability (data not shown).
Prior to injection, cells were harvested using trypsin/EDTA (Biological industries, Beit-haeemek, Israel). Cell viability was assessed using trypan-blue (Sigma-Aldrich), and cells were injected only if viability was above 85%. Finally, 2×10 6 hESC were resuspended in ringer-lactate solution for no longer than 20 min prior to injection.
2.3
Athymic rat hind limb ischemia model
Hind limb ischemia was induced in athymic rats (males, average weight 450g) that were anesthetized using a subcutaneous injection of ketamine (60 mg/kg), xylasine (7.5 mg/kg) and acepromazine (0.75 mg/kg). Skin incision at the medial side of the two hind limbs was performed (incision <2.5 cm) and the epigastric artery and vein were isolated, ligated and cut. Next, we isolated 1 cm of the common femoral artery, proximal to the ligated epigastric branch but distal to the arteria circumflexa femoris. The femoral arteries on both sides were ligated proximal to the ligated epigastric branch and cannulated retrogradely on one side with a 26G catheter via arteriotomy. Two million hESC suspended in 0.5 ml ringer lactate solution ( n =14) were injected retrogradely, at a rate <0.2 ml/min (average flow in the common iliac artery at day 10 of control leg was 1.65 ml/min, n =29). The common femoral artery was afterwards ligated 1 cm proximal to the site of arteriotomy and, finally, was excised on both sides.
In the intramuscular groups, 2×10 6 hESC suspended in 0.3 ml ringer lactate solution ( n =14) was injected intramuscularly at six predetermined locations in limb muscles surrounding the site of arterial excision.
As controls we studied additional 16 rats that were injected with saline intra-arterially alone ( n =11) and both intra-arterially and intramuscularly ( n =5). The combination of intramuscular cell injection and intra-arterial saline injection was employed to exclude the potential arteriogenic effects of intra-arterial bolus injection in our model. To avoid anatomical bias, injected legs (right or left) were randomly chosen in all groups.
At 10 and 21 days, we used the same anesthesia protocol. After completion of flow and perfusion measurements, rats were killed by injection of 1 ml pentobarbital into the heart.
As part of validation of the hind limb ischemia model and injection protocol, we injected methylene blue intra-arterially using the same protocol we used for cell and saline injection. Thereby, the region of preexisting collateral vessels was detected and later used for intramuscular cell injection. In addition, as part of model validation, we performed flow and perfusion measurements prior to femoral artery occlusion and excision, and after occlusion.
2.4
Hemodynamic measurements
Hind limbs skin perfusion (PeriScan PIM II Perfusion Imager, Perimed, Sweden) was evaluated at baseline and at 10 or 21 days. Skin perfusion used for statistical analysis represents the mean of two consecutive perfusion measurements on the dorsal as well as the plantar side of the feet. As perfusion measurements can vary based on limb position, feet were precisely positioned at baseline and final measurements.
Muscle perfusion was evaluated after removal of the skin. Again, mean of two measurements of each side was used in the analysis in order to minimize positioning errors.
Flow in the common iliac artery as an indication for limb blood flow was measured after laparatomy using a 0.7 cm Doppler probe (Transonic Animal Research Flowmeter, Ithaca, NY, USA). Care was taken in order to evaluate flow at the same distance from the aortic bifurcation in all rats. Before performing measurements, the iliac arteries were exposed and in order to prevent arterial spam, lidocain 2% drops were applied directly on the exposed artery. Each measurement was evaluated after flow stabilized for 40 seconds in each side. For analysis, the mean of at least two measurements was used. The two measurements were recorded online, and in case of significant variation (higher than 20%), a third measurement was performed and the arithmetical average was used for analysis.
Generally, tests were performed either simultaneously on both sides to minimize influence of hemodynamic changes.
2.5
Histology and immunohistochemistry
Immediately after sacrifice, hind limb muscles were fixed in formalin 4% for at least 24 h and then embedded in paraffin blocks. To identify arteries, smooth muscle actin staining (Dako A/S, Denmark) was performed on 8-μm-thick sections.
Four samples from each hind limb thigh musculature were studied for number of arteries. Arteries were counted on 18 randomly chosen microscopic fields (100× magnification) by two independent observers who were blinded to group allocation. Each observer repeated counting 4 times per slide or until standard error of the counts was <20% (error calculation was done online while acquiring data).
2.6
Injected cell tracking
To identify the presence of the injected hESC, we injected hESC tagged with fluorescent dye (CellTrackerTM CM-DiI, Invitrogen, CA, USA) intramuscular. Frozen section from the injection site were tested 12 h and 10 days after injection using fluorescent microscopy. hESC aggregates were observed in the muscles at 12 h but could not be identified 10 days after injection (data not shown). As a second strategy for identification of injected cells, we stained representative muscle tissues from intramuscular and intra-arterial injected rats 10 days after injection using a human leukocyte antigen (HLA)-abc antibody (Dako A/S, Glostrup, Denmark). HLA-abc staining did not produce reproducible results, possibly due to low HLA expression by undifferentiated or partially differentiated hESC. Our third approach to identify the presence of injected cells was to extract genomic DNA from muscle tissue to detect the presence of human specific alu repeats using the method described by Walker et al. . One microgram of muscle DNA from rats of all experimental groups was used for DNA tracking. Level of sensitivity reported to detect as low as 0.5 ng of human DNA (equivalent to approximately 100 cells). In rats injected with hESC intramuscularly (10 and 21 days), polymerase chain reaction (PCR) analysis yielded positive signal for human alu repeats indicating the presence of injected hESC in the injected limbs in all rats and was negative in all control limbs. Analysis of the intra-arterial injected rats after 10 days or after 21 days has shown the presence of the cells in both limbs (injected and control).
2.7
Statistical analysis
Wilcoxon signed rank test was used to test the effects of cell injection comparing the control to the injected limb for each rat in each group. We also used the students paired t test to substantiate the statistical significance of our findings.
In order to rule out differences between the groups that may bias our findings, one-way analysis of variance (ANOVA) comparing mean of the absolute values of flow and perfusion of all the control legs was performed.
One rat in the intra-arterial 10 days group was excluded from muscle perfusion analysis because of wound infection on the non-injected limb, which would artificially increase perfusion measurements secondary to local hyperemia and formation of granulation tissue.
2
Methods
2.1
Overall study design
Bilateral hind limb ischemia was induced in athymic rats ( Foxn1 rnu , Harlan, Israel) by surgical disruption of the femoral artery continuity. Undifferentiated hESC or control solution where injected in one limb while the other limb served as a control. To test the effects of shear stress on hESC contribution to arteriogenesis, we injected hESC intra-arterially in athymic rat model of hind limb ischemia and compared the effects on arteriogenesis to rats injected with hESC into the muscles of the ischemic hind limb (no shear stress). At 10 and 21 days, flow, perfusion, and the number of arterial blood vessels/mm 2 muscle were measured in the two limbs. Seven groups of rats were studied. Rats injected intra-arterially with hESC (10 and 21 days), rat injected intramuscular with hESC (10 and 21 days), rats injected intra-arterially with saline (10 and 21 days), and rats injected intramuscular and intra-arterial saline (10 days). The last group of animals was studied based on preliminary observation that intra-arterial injection of saline alone may result in increased flow. Table 1 summarizes the groups comprising the study. Rats were kept under best conditions and all procedures were performed avoiding any unnecessary stress or pain, according to the Guide for the Care and Use of Laboratory Animals . Ethics approval for this study was given under the code 20-02-2004 by the Technion, IIT, Haifa, Israel, ethics commission.
