1 Abstract Vein grafts are a form of treatment for vascular diseases but often only provide short term improvement due to the possibility of grafts failure. The primary mechanism of failure is due to intimal hyperplasia and inward remodeling of the graft. T he placement of a vein into an arterial circulation leads to changes in shear stress due to the graft being placed into a different circulation or shear environment. Thus, it is hypothesized that the shearing force across the endothelium leads to prolifera tion of cells within the vessel and causes biochemical and morphological changes to the graft. In this study, we sought to find a correlation between the rate and timing of cell proliferation with intimal hyperplasia development. In order to find a correla tion, vein graft were generated, harvested, and stained for bromodeoxyuridine (BrdU) and Ki 67, immunohistochemical stain s used to assess cell proliferation at 2 hours, 1, 3, 7, 14, and 28 day(s) after implantation in rabbits. Positive cells are visibly s tained and counted to determine the proliferation rate. We found that while the low and high flow vein grafts had a similar adaptati on in proliferation and intimal growth, the 7 day time point showed the most variation and difference between the two flows These data will be used as part of a model to understand the biological and physical effects of the vein graft and the process of vascular adaption from the injury.
2 1 ) Introduction 1.1 Peripheral Arterial Disease Cardiovascular disease is the le ading cause of death in the United States and a significant contributor to deaths globally. In the United States, cardiovascular diseases account for approximately 30% of all deaths 1 Cardiovascular disease s like peripheral arterial disease (PAD) have a high prevalence of coexisting with othe r diseases such as coronary artery and cerebrovascular disease thus causing additional problems for patients 2 PAD is a lower extremity arterial occlusive disease 3 and most commonly forms as a result of atherosclerosis, but can also have other causes such as embolisms, thrombosis, and vasculitis. PAD affects 8 10 mill ion people in the United States and affects 20% of people over the age of 70 years Presentation of PAD can be asymptomatic or can have a range of symptoms including leg pain and ischemia. Prognosis of PAD ranges from amputation, claudication, other cardio vascular events (stroke or myocardial infarction), and mortality 4 Exercise and drugs (cilostazol and pentoxifylline) are recommended treatments for PAD, but patients may need a surgical intervention if the vessel occlusion is too advanced. Surgical revascularization interventions like balloon angioplasty, bypass vein grafts, endarterectomy, or stent can help prolong the life of the limb or patient 3 Though vein grafts help, they do not cure or add ress the initial problem of occlusion. For example, i n a representative study on lower extremity vein grafts, the primary patency rate was only 59.5% after the first year surgery ( Figure 1) 5 Primary patency is defined as uninterrupted patency with no procedure or intervention directly performed on the graft 6 Figure 1 Patency of vein bypass graft procedure 5
3 1.2 Vein Graft Failure An exaggerated injury response often leads to restenosis, or luminal narrowing, tha t can result in failure of the vein graft. 7 In lower extremity grafts vein graft stenosis or occlusion occurs in 30% to 50% of cases within 5 years and 70 to 80% of high grade stenosis will progress to vein graft failure leading to significant morbidity and mortality 6 For many patients with advanced PAD, failure of a bypass graft may lead directly to major limb amputation and diminished quality of life. 8 The p athophysiology of vein graft failure involves complex interactions ( Figure 2 ) between a vascular injury response and a hemodynamic adaptation of the vein exposed to arterial forces 9 Figure 2 Complex integration of responses for adaptation 10 0hr 24 hr 1 wk 1 mo 3 mo 1 yr EC Injury & Repair Hemodynamic Adaptation Fibrosis / Lipi d Accumulation SMC Migration/Proliferation Ischemia/ Fibrosis Hemodynamic Adaptation Matrix Accumulation
4 1.3 Intimal Hyperplasia and Vessel Remodeling I ntimal hyperplasia ( IH ) c an be defined as excess intima formation. IH appears in the intima which is the innermost layer of the arterial wall ; the intima c onsist s of a single cell layer of endothelial cells and is in direct contact with blood flow 4;11 The media, the layer between the intima and adventitia, consists primarily of smooth muscle cells (SMC) and extracellular matrix (EMC) As in Figure 3, internal elastic lamina (IEL), a connective tissue layer lies between the intima and med ia lies and the external elastic lamina (EEL) lies between the media and adventitia the outermost layer of the vessel 11 Figure 3 Cross section of venous wall 11 There is a lack of understanding about the mechanisms of graft failure and about the balance between vessel remodeling and IH leading to atherosclerotic development in the graft. 8;9 What is known about atherosclerotic developmen t is that cellular proliferation ( Figure 4 ) inflammation, and matrix metabolism play a role in post implantation remodeling of the vessel 9;12 These remo d el ing responses of intimal thickening are the adaptatio n process for v ein grafts to the arterial circulation. 2 When the vein graft is implanted in the arterial system, the vein graft is exposed to higher blood pressure and int raluminal pressure leading to increased wall shear stress and intramural wall tension than the venous system. 13 The morphology of the implanted vein graft changes causing the vessel wall to become thicker and the graft diameter to increase 11 C hanging of the hemodynamic factors of the vessel is closely linked to IH L ow velocity of blood flow, low flow rate and/or low shear stress can also enhance IH. 8;14
5 Figu re 4 Map of interacting element involved in vein graft adaptation 15 During graft implantation, there is often damage to the endothelial cell layer from handling of the graft and changes in shear. Thus, t here is also a response to repair the injury to the endothelial cells caused by the vein graft pr ocedure, initiating local synthesis of cytokines and growth factors, inflammatory mechanisms that drive the IH response. 16 These mechanisms induce the recruitment of neutrophils and mononuclear cells, further amplifying recruitment of inflammatory cells and thickening of the intima l area. Activated by these chemoattractants SMC begin to dedifferentiate from a contractile to a synthetic phenotype and re plicate within the media. 14 The SMC also migrate out of the media into the intima where they prolif erate and deposit ECM proteins. 17;18 These synthetic cells produce 4 to 5 times more ECM than the contractile phenotype and have an increased proliferation rate of at least 10%. 19 This matrix thickens the intima which narrows the lumen and limits the durability of vein bypass grafts 14;16
6 The IH process can proceed to significant stenosis and t he majority of vei n graft stenosis from human peripheral bypass grafts can be classified as IH, as shown in Figure 5 18 As of now, the only treatment approach for IH is angioplasty and possibly brachytherapy ; therefore, additional studies are needed to understand the problem and create better therapies 19 Figure 5 IH development leading to luminal narrowing in a vein graft 1 2 1.4 Immunohistochemistry and Proliferation Immunohistochemistry can be used to assess the amount of proliferation within a sample. 5 bromodeoxyuridine (BrdU) is a halogenated derivative of thymidine and injected into the organism, BrdU is incorporated into the newly synthesizing strands of DNA in the nuclei during the S phase of the cell cycle and can be detected by an anti BrdU antibody 20 Since BrdU is responds to it and how much is left by the time to harvest the graft 21 BrdU is also incorporated in n on proliferating cells experiencing DNA repair, leading to problems with non specific staining. When using diaminobenzidine (DAB), p roliferating cells that have incorporated BrdU have a visible brown pigment over the nuclei. 8 Ki S and G2 phas es of the cell cycle but not in the G0 phase. 20 The length and localization of expression for Ki 67 is unknown, which is a factor that can aff ect the amount of proliferation shown 20 P roliferating cells that have Ki 67 expression have a visible brown pigment over the nuclei as well 8 Given that BrdU can be incorporated into DNA only during the S phase of the mitotic process, whereas Ki 67 is expressed for its whole duration, both methodologies allow for assay and comparison of proliferation rates. The comparison of the data from the BrdU and Ki 67 stain s is
7 useful because of the different ways of incorporation could potentially lead to different results for proliferation comparison. 1.5 Hypothesis My hypothesis is that the rate and timing of cellular proliferation correlates with intimal hyperplasia development in vein grafts Rabbit vein grafts were performed as part of a larger study, and my goal is to evaluate cell proliferation in these grafts and determine the relationship with intimal hyperplasia development. Previous studies suggest that that the same factors that stim ulate SMC proliferation simultaneously increase ECM production. 19 Therefore I would expect greater proliferation would lead to increased intimal hyperplasia since the SMC deposit cell matrix when the i ntima is inflicted with injury. A low flow graft experiences lower force shear compared to a high flow graft within the IH model I hypothesize that the decrease in shear leads to more turbulent blood flow and thus a higher proliferative response. I am als o comparing the proliferation that occurs at different time points after the vein graft surgery and I anticipate that the density will be higher in the low flow graft for the rabbits at the 3 day, 7 day, and 14 day time points. I think that the earlier tim e intervals would be too early for a response to occur in the intima and the latest time interval is when the graft is undergoing other adaptations to injury.
8 2 ) Methods 2 .1 Intimal Hyperplasia Model There are several models of intimal hyperplas ia development. Our laboratory created a rabbit vein graft model which creates a difference in hemodynamic flow to enhance the amount of intimal thickening Since manipulating the hemodynamics can promote IH, our laboratory develop ed an in vivo model for I H development in male New Zealand white rabbits. 22 The rabbits undergo a bilateral vein graft procedure to implant the bilateral external jugular vein segment i nto the common car otid arteries using a n anastomotic cuff technique ( Figure 6 ) As shown in Figure 6 3 of the 4 distal branches are ligated to create a low flow side and as a result, the opposite side beco mes a high flow side. The low flow side graft develops more IH than the high flow side graft from the hemodynamic changes. 11;22 Figure 6 Bilateral vein graft model with differ ential flow environments 23 Rabbits were injected with 30mg/mL of bromodeoxyuridine (BrdU) 2 hours before vein graft harvest at time points of 2 hours, 1 day, 3 days, 7 days, 14 days, or 28 days following the ve in graft implantation. Before harvest, the flow rate in the two grafts was measured ( 2.0 mm flow probe; T106, Transonic Systems) From the harvest, sections of the grafts were c ut into cross sections 5 micrometers thick and formalin fixed in paraffin. Secti ons from both sides of graft were cut and treated as independent samples due to the assumption of different exposure and
9 morphology for the two sides. Morphological analyses were completed with both in vivo external graft diameter and cross sectional measu rements ( Axiovision version 3.1, Zeiss). Masson 2 .2 Bromodeoxyuridine One set of the rabbit vein graft samples from each time point were stained with BrdU (Invitrogen) Before staining the slides, the samples must be de waxed and rehydrated using a xylene ethanol gradient in order to expose the antigens to the tissue To inhibit endogenous peroxidase activity, the samples were quenched in 3% hydrogen peroxide which is needed to prevent background staining of hemoproteins like hemoglobin in erythrocytes. 24;25 Before adding each reagent, the previous reagent was cleared by washing the slides with phosphate buffered saline (PBS) for 6 minutes. Before adding the primary antibody, the slides were incubated with tryp sin at 37C for 1 hour a denaturing solution at room temperature, and a blocking solution at room temperature. Biotinylated mouse anti BrdU was the primary antibody which then attached to streptavidin pe roxidase. This is a very strong non covalent interac tion used for immunohistochemical staining. T he marker, DAB attached to the streptavidin peroxidase. Slides were dehydrated before observing with microscope. 2 .3 Ki 67 The samples wer e de waxed and rehydrated us ing a xylene ethanol gradient, t he n quenche d in 1% hydrogen peroxide. Before adding each reagent, the previous reagent was cleared by washing the slides with PBS for 6 minutes. The samples were incubated in high heat (steamer) in a 1x target retr ieval solution (Dako) because antigen retrieval helps to eliminate cytoplasmic and nuclear background in immunohistochemical stains 24 To continue blocking for non specific staining, the samples were incubated in goat serum. After, the samples were inc ubated in the primary antibody, monoclonal mouse anti rat ki 67 antigen (clone Mib 5) (Dako). The primary antibody was incubated at 4C and all other reagents incubated at room temperature. To attach the visible marker, peroxidase substrate DAB (Vector), t he secondary antibody, biotinylated goat anti mouse (Vector), and the n Avadin Biotin Complex (ABC) (Vector) is added. ABC is needed to link the antibody with the marker as it is a very strong covalent bond allowing for a higher affinity between the two. 24 Slides were dehydrated be fore observing with microscope.
10 2 .4 Controls Controls were made for both the BrdU and the Ki 67 immunohistochemistry from mouse intestine. One mouse was injected with BrdU 2 hours before harvest (BrdU +) and one was not (BrdU ). From the harvest, sections were cut into cross sections 5 micrometers thick and formalin fixed in paraffin. Mouse intestine was used because intestinal epithelia cells proliferate, turn over quickly, and are under constant renewal 26;27 The intestine allows us to make the assumption that proliferating cells will be present in all samples. The BrdU positive and negative controls were stained with the above ( 2 .2) BrdU procedure with each batch of stains. For the positive Ki 67 control, a mouse intestine sample went through the above Ki 67 procedure ( 2 .3). For the negative Ki 67 control, a mouse intestine sample went through the above Ki 67 procedure, but was i ncubated with goat serum instead of the primary antibody ( 2 .3). 2 .5 Imaging Images of the stained sections were taken from a compound microscope (1.25 40x objectives, Zeiss Axiolab) and a digital camera interface (Zeiss). Images were taken at a magnificat ion of 5X, 1 0X and 2 0X. The entire vein graft cannot be visualized at high enough resolution to quantitate positive cells; therefore, the vein graft is recreated from several high resolution images of each slide. The images will be stitched in order to eva luate the amount of proliferation in each sample using an image software (Photoshop). 2 .6 Cellular Count We defined IH as the growth in area between the lumen and the IEL. Overall proliferation in the sample was viewed in the intimal and medial area s to l ook at the migration and the overall presence of SMC. To distinguish these layers, a masson staining was taken of a nearby section of the vein graft. Using photoshop, a digital copy of the IEL and EEL layer was transposed on the samples. The cells within a stitched image sample were counted with coordinates assigned to each positive cell (Axiovision). The count of positive cells per section is how much proliferation (# positive nuclei) is occurring. Density of actively proliferating cells was measured by Br dU positive cells per unit area (nuclei/mm 2 ) with the area being between the EEL and the lumen.
11 2 .7 Statistics All data is expressed as mean standard error of the mean except where using individual rabbit sections Differences were determined using a post hoc one way and t wo way ANOVA (Sigmaplot 12 .5 ) then a p airwis e multiple comparison procedure ( Holm Sidak method) among significant differences. To determine correlation, the Pearson product moment c orrelation was used. Significant d ifferences were measured at less than p < 0.05
12 3 ) Results 3 .1 Surgical and Morphometric Data Vein graft procedures were performed on 54 rabbits and vein grafts were harvested and cut on both ends of the graft. Samples were categorized by time point at 2 hours (n =30), 1 Day (n=29), 3 Days (n=30), 7 Days (n= 29), 14 Days (n=29), and 28 Days (n= 19) and by flow exposure at A comparison (one way ANOVA) of the normal flow ra tes (all time points) versus the low flow rates (all time points) versus the high flow rates (all time points) was significant (p <0.001) (Table 1) The low was significantly different than the high and normal flow, but the high flow was not significantly different than the normal flow Table 1. Pairwise comparison for flow rates (all time points) Comparison Difference of Means Significant? H vs. L 35.4 Yes N vs. L 29.814 Yes H vs. N 5.586 No When flow rate is graphed over time, both the normal and high flow rates have a similar shaped curve and similar mean flow rates. The low flow rate has a similar shape, but has a much lower average flow rate. When analyzing the flow rates over time (two way ANOVA), there is no significant difference.
13 Figure 7 Average flow rate for vein graft Given that the low flow graft was manipulated to have a lower flow rate than the normal and high flow graft, we then examined the averages of the intimal area s to determine the growth of the intima or the IH development. T he amount of intima area increases as time passes (Figure 8). The earlier time points did not have significant differences in intimal growth over time while the later time times did have significant differences (Table 2) The low flow graft had the highest IH development overall and the normal and high flow graft had significantly smaller development At the earlier time points, the intimal area development was similar but started differentiating in growth after 3 days. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days Flow rate (mL/min) Time after Implantation Low flow Normal flow High flow
14 Figure 8 Comparison of i ntimal area Table 2. Intimal area pairwise comparison for time Comparison P Significant? 28.000 vs. 0.080 <0.001 Yes 28.000 vs. 3.000 <0.001 Yes 28.000 vs. 1.000 <0.001 Yes 28.000 vs. 7.000 <0.001 Yes 28.000 vs. 14.000 <0.001 Yes 14.000 vs. 0.080 <0.001 Yes 14.000 vs. 1.000 <0.001 Yes 14.000 vs. 3.000 <0.001 Yes 14.000 vs. 7.000 <0.001 Yes 7.000 vs. 0.080 0.012 Yes 7.000 vs. 1.000 0.02 Yes 7.000 vs. 3.000 0.025 Yes 3.000 vs. 0.080 0.979 No 1. 000 vs. 0.080 0.978 No 3.000 vs. 1.000 0.87 No In addition comparing the low flow with normal and high flow there are significant changes in intimal growth (Table 3). The high flow and the low flow graft did not have significant difference in growth. 0.000 0.200 0.400 0.600 0.800 1.000 1.200 1.400 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days Intimal Area (mm 2 ) Low flow Normal flow High flow Time after Implantation P <0.001
15 Table 3. Intimal area pairwise comparison for flow Comparison P Significant? H vs. L <0.001 Yes N vs. L <0.001 Yes H vs. N 0.607 No 3 .2 Proliferation BrdU The grafts at the 2 hours and 1 day time points did not show much proliferation or IH devel opment ( Figure 9 a) d ) ). At 3 days, proliferation can be detected within all of the samples. Though it is difficult to tell between low and high flow by just visual observation, it can be determined that the highe r proliferation rates are between 3 days and 14 days by the increased qunatity of brown positive cells
16 a) c) e) g) i) b) d) f) h) j)
17 When comparing the the proliferation growth over time, there is a signficant difference in the amount of prolif eration (Table 4) To measure proliferation density, the proliferation was normaliz ed using the intima and media area. The normalization is needed to account for the varying sizes in graft sections and how different sizes can contribute different amounts o f proliferation. This normalization did not add any additional significance to the comparisons over time. Table 4. Pairwise Comparison with Proliferation and Proliferation Density Proliferation Proliferation Density Comparison P Significant? Comparis on P Significant? 7.000 vs. 0.080 <0.001 Yes 7.000 vs. 0.080 <0.001 Yes 3.000 vs. 0.080 <0.001 Yes 3.000 vs. 0.080 <0.001 Yes 7.000 vs. 1.000 <0.001 Yes 7.000 vs. 1.000 <0.001 Yes 7.000 vs. 28.000 <0.001 Yes 7.000 vs. 28.000 <0.001 Yes 3.000 vs. 1.000 <0.001 Yes 3.000 vs. 1.000 <0.001 Yes 3.000 vs. 28.000 <0.001 Yes 3.000 vs. 28.000 <0.001 Yes 14.000 vs. 0.080 <0.001 Yes 7.000 vs. 14.000 <0.001 Yes 7.000 vs. 14.000 0.006 Yes 14.000 vs. 0.080 0.001 Yes 14.000 vs. 28.000 0.029 Yes 3.000 vs. 14.000 0. 002 Yes 14.000 vs. 1.000 0.031 Yes 14.000 vs. 28.000 0.033 Yes 3.000 vs. 14.000 0.127 No 14.000 vs. 1.000 0.063 No 1.000 vs. 0.080 0.354 No 1.000 vs. 0.080 0.53 No 7.000 vs. 3.000 0.538 No 1.000 vs. 28.000 0.896 No 28.000 vs. 0.080 0.555 No 28.000 vs. 0.080 0.826 No 1.000 vs. 28.000 0.66 No 7.000 vs. 3.000 0.774 No k) l) Figure 9 Representative sections of BrdU stains in rabbit vein g rafts (20X) a) High flow 2 Hours, b) Low flow 2 Hours, c) High flow 1 Day, d) Low flow 1 Day, e) High flow 3 Days, f) Low flow 3 Days, g) High flow 7 Days, h) Low flow 7 Days, i) High flow 14 Days, j) Low flow 14 Days, k) High flow 28 Days, l) Low flow 2 8 Days
18 F or the proliferation count, the high and normal flow grafts closly mirrored each other except at 14 days ( Figure 10 ) The normal and high flow proliferation and proliferation density was highest at showed a curve of proliferation growth then a declining amount of growth as time progressed. Though the different flows showed a similar prolife ration pattern, the low flow graft has a distinguished peak at 7 days. Figure 10 Comparison of BrdU positive cells The proliferation density showed a similar trend when compared to the proliferation count. The high and normal f low grafts closly mirrored each other except at 14 days ( Figure 11 ). The normal and high flow proliferation and proliferation density was highest at 3 days while the low All flows showed a similar curve and similar values of proliferation density though the low flow graft was different with peak of proliferation at 7 days. 0 50 100 150 200 250 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days Proliferation (# of Positive Nuclei) Time after Implantation Low Normal High P <0.001
19 Figure 11 Comparison of Proliferation Density When looking at each indivi dual rabbit, we compare of the proliferation density between the low and high flow graft ( Figure 12 ) For 2 hours, 1 day, and 28 days, there is little change between the low and high flow graft within the rabbit. There is variatio n between the low and high flow within 3 and 14 day time points, but there is a greater difference within 7 days. Also at the 7 day time point, more of the rabbits increase in proliferation with low flow exposure as opposed to the other time points where t he proliferation within a rabbit can increase, decrease, or remain the same. There is a general trend where proliferation increase, then decreases over time. 0 5 10 15 20 25 30 35 40 45 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days Proliferation Density (Nuclei/mm2) Time after Implantation Low Normal High
20 Figure 12 Pairwise comparison of prolifera tion density 3 .3 Correlation within Flow Rate, Intimal Area, and Proliferation Density Given that proliferation seemed to peak at 7 days in the low flow graft samples, we wanted to evaluate how flow rate influenced the proliferation density for the individ ual samples. Looking at individual samples instead of an average allows us to see if there is a correlation with the samples. There was not a significant correlation between flow rate and proliferation density It appears as though there is a trend that l ower flow rates contribute to higher proliferation density ( Figure 13 ). It also appears that there is a burst of proliferation in the 7 day samples. There is a cluster of increased proliferation a t day 3 and day 7 regardless of the flow rate There is a lot of heterogeneity within the 7 day time point where the 2 hours, 1 day, and 28 day time points remain fairly stagnant regardless of flow rate. H L 2 Hours
21 Figure 13 Flow Rate vs. Proliferation D ensity We next compared flow rate and intimal area for which there was no significant correlation For 7, 14, and 28 days, it looks like lower flow rates contribute to higher intimal areas ( F i g u r e 1 4 ) At 28 days there is the highest growth of the intima and at earlier time points, there is little to no growth. Figure 14 Flow Rate vs. Intimal Area 0 10 20 30 40 50 60 70 80 0.000 10.000 20.000 30.000 40.000 50.000 60.000 70.000 80.000 90.000 100.000 Proliferation Density (Nuclei/mm2) Flow Rate (mL/min) 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 10 20 30 40 50 60 70 80 90 100 Intimal Area (mm 2 ) Flow Rate (mL/min) 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days
22 In addition, we looked at the at proliferation density with intimal area which has no si gnificant correlation. At 7, 14 and 28 days there is a trend that suggests intimal area increases as proliferation density increase s ( Figure 15 ). There is more variation within the 7 day time point compared to the other time poi nts. There is also a cluster of samples with low area and low proliferation for the earlier time points. Figure 15 Proliferation Density vs. Intimal Area 3 .4 Proliferation Ki 67 Positive Ki 67 is not visibly present in the 2 h ours, 1 day, and 28 days sections ( Figure 16 ). Though it is difficult to compare the amount of proliferation, the 3, 7, and 14 days have a much greater amount of proliferation with both low and high flow when compared to the 2 ini tial time points. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 0 10 20 30 40 50 60 70 80 Intimal Area (mm2) Proliferation Density (Nuclei/mm 2 ) 2 Hours 1 Day 3 Days 7 Days 14 Days 28 Days
23 a) c) e) g) b) d) f) h) i) j)
24 3 .5 Comparison between BrdU and Ki 67 These stains are n ot specific for SMC proliferation but for the overall proliferation occurring within the section The assumption is also made that there is not m uch difference between sections 5 15 micrometers apart. When comparing 2 adjacent sections at 14 days, they both have as visible amount of proliferation ( Figure 17 ). There seems to be more positive cells within the Ki 67 section than the BrdU section. k) l) Figure 16 Representative sections of Ki 67 stains in rabbit vein grafts (10X) a) High fl ow 2 Hours, b) Low flow 2 Hours c) High flow 1 Day d) Low flow 1 Day, e) High flow 3 Days, f) Low flow 3 Days, g) High flow 7 Days, h) Low flow 7 Days, i) High flow 14 Days, j) Low flow 14 Days, k) High flow 28 Days, l) Low flow 28 Days
25 a) Figure 17 Comparison of proliferat ion stains at a low flow 14 day rabbit vein graft (5X) a) BrdU b) Ki 67 b)
26 4) Discussion My hypothesis was that the rate and timing of cellular proliferation correl ates with intimal hyperplasia development in vein grafts Though previous studies show th at low shear forces exhibit more IH at 28 days we wanted to determine how flow rate correlated with IH development and proliferation First, w e needed to ensure that there was a significant difference between the low and the high flow for our study. Thoug h we included normal flow samples to set a standard, or normal, response, we see that the high flow and normal flow rates were very similar. The similar flow rates may be the reason that we see a similar adaptation response with proliferation and intimal a rea in high and normal flow samples. The slight difference in flow rate may be the cause of the slightly higher proliferation and intima area at 14 days though all other time points are very similar. This suggests that at earlier time points, the flow rate is not a factor for adaptation and intimal growth for all grafts and that they undergo a similar adaptive response to the surgical intervention In addition, it could be that physiologically it is difficult to achieve flows t hat are much higher than norm al and that the body adapts to this increase in flow rates Our analysis of intimal area demonstrates that intima l growth increases over time, with significant growth increases after 7 days By lowering the flow of the graft, there was increased growth to the intimal area in the lower flow rabbits than with normal or high flow. The low flow may provide prolonged exposure of inflammatory cells to the graft and also allow for the ECM to accumulate without additional disruption by blood flow or increased shea r There is also a possibility that circulating SMC cells have the time to also contribute to the local response with a lower blood flow. Although, the low flow graft shows more intima growth, the flow rate does not appear to manifest an adaptation or remo deling response until 7 days. Though an increase in B rd U proliferation in the grafts begins at 3 days, only at the 7 day time point does the proliferation vary markedly between the low and the high flow graft. This also suggests that the rate and timing of the proliferation as opposed to the overall amount of proliferation is what correlates to the IH development. In addition to increased IH and proliferation at 7 days, Figure 12 shows that there is more heterogeneity and variation with the rabbit samples a t 7 days compared to other time points. Because the trend in both proliferation and intima area deviate at 7 days, this is considered a significant time point for IH development and vein graft adaptation.
27 It is possible that the normal and high flow grafts are undergoing a normal response of thickening to accommodate a vein in an arterial circulation. Proliferation of SMC causes matrix deposition and will make the intima bigger regardless of flow. The low flow grafts could be undergoing that same adaptation but is mal adapted or over responsive, and thus there is additional IH growth. This burst of growth that starts at 7 days is most likely when proliferation contributes to the IH growth. Before and after the 7 day time point are most likely when other ada ptations to the injury and shear stress occur. The BrdU and Ki 67 were different in the amount of visible proliferation though these studies were limited in the amount of Ki67 samples This observable differ ence could be due to the fact Ki 67 is an endoge nous protein and does not rely on what is taken in by metabolism as B rd U uptake My initial o bservations suggest that there is an increase in cellular proliferation 3 days post implant It is difficult to make a judgment on whether the low or high flow had more proliferation for these samples based on the images for Ki 67 since staining is not completed for these samples Though there were no significant differences between flow rate and intima area, flow rate and proliferation density, and IH and prolifera tion density, there were visible trend s that suggest differences Potential limitations for the correlations might be due to the small sample size since it is possible to perform a type II statistical error with small samples meaning that we did not see a difference because our sample sizes were small In addition, there is a large amount of variability and heterogeneity within the rabbit samples, thus it sometimes difficult to detect a trend in small sample sizes with large variability A final limitation with these analyses is that we do not examine the full length of the graft, and often vein graft stenosis occurs, at least in humans, in focal points within the graft rather than homogeneously along the graft
28 5 ) Relevance and Future Goals 5 .1 Relev ance Intimal hyperplasia threatens almost every vascular reconstructive procedure and yet there is no treatment. Operations like angioplasty, stents, and heart transplants are all measures that delay morbidity and mortality, but they do not address the pro blem fundamentally. 4;27 There is a need for understanding balance between remodeling and IH in order to further studies on treatment 9 5 .2 Ki 67 Analysis In addition to doing a visual comparison of the Ki 67 to the BrdU, we plan on doing a full quantitative analysis of the Ki 67 stains to the BrdU. We will count the proliferation on each of the sections and do the same comparisons as the BrdU. I n addition to a comparison, Ki 67 serves as a way to verify our findings with the BrdU. 5 .3 Additional analysis In addition to finding the proliferation density using the area, we would like to find the density using the overall nuclear cell count of the sample. We also feel that if we acquire addition vein graft samples, we will be able to see additional trending behavior. The current data is normalized using the intima and media area. We can obtain more information about the adaptation by analyzing the intima and media as separate parts for comparison. 5 .4 Modeling The data from each of the samples will be integrated into a computational model by evaluating the time factor of cell proliferation along with changes in intimal thickness and changes in blood flow rate. The model will se rve as an in silico representation of the adaptation that occurs in a vein graft. There is a complex nature of processes that occur during adaptation ( Figure 18 ) and we would like to understand the mechanisms that. It is understoo d that there is no "magic bullet" approach by affecting one factor. Studying and manipulating SMC proliferation is unlikely to reduce the incidence of restenosis by itself. 28
29 Figure 18 Multi scale nature of adaptation model A model will help to identify previously unidentified targets for treatment and identify patients and regions most at risk for accelerated graft f ailure. Because of the looping nature of the adaptation, if we look at proliferation of cells, we hope to further gain understanding of the genes influencing that response and the response to the vessel. We plan on looking more at apoptosis, ECM, hemodynam ics, and the genes that regulate the adaptation response in order to develop this model.
30 Referenc e s (1) Santulli G. Epidemiology of cardiovascular disease in the 21st century: updated numbers and updated facts. Journal o f Cardiovascular Disease 2013;1:1 2. (2) Allaqaband S, Kirvaitis R, Jan F, Bajwa T. Endovascular treatment of peripheral vascular disease. Curr Probl Cardiol 2009;34:359 476. (3) Ouriel K. Peripheral arterial disease. Lancet 2001;358:1257 1264. (4) S ubbotin VM. Analysis of arterial intimal hyperplasia: review and hypothesis. Theor Biol Med Model 2007;4:41. (5) Conte MS, Bandyk DF, Clowes AW et al. Results of PREVENT III: a multicenter, randomized trial of edifoligide for the prevention of vein graft failure in lower extremity bypass surgery. J Vasc Surg 2006;43:742 751. (6) Owens CD, Ho KJ, Conte MS. Lower extremity vein graft failure: a translational approach. Vasc Med 2008;13:63 74. (7) Heckenkamp J, Lamuraglia GM. Intimal Hyperplasia, Arterial Remodeling, and Restenosis: An Overview. Perspectives in Vascular Surgery and Endovascular Therapy 1999;11:71 94. (8) Itoh H, Komori K, Funahashi S, Okadome K, Sugimachi K. Intimal hyperplasia of experimental autologous vein graft in hyperlipidemic rabb its with poor distal runoff. Atherosclerosis 1994;110:259 270. (9) Conte MS. Molecular engineering of vein bypass grafts. J Vasc Surg 2007;45 Suppl A:A74 A81. (10) Conte MS, Mann MJ, Simosa HF, Rhynhart KK, Mulligan RC. Genetic interventions for vein b ypass graft disease: a review. J Vasc Surg 2002;36:1040 1052. (11) Tran Son Tay R, Hwang M, Berceli SA, Ozaki CK, Garbey M. A model of vein graft intimal hyperplasia. Conf Proc IEEE Eng Med Biol Soc 2007;2007:5807 5810. (12) Garbey M, Berceli SA. A dyn amical system that describes vein graft adaptation and failure. J Theor Biol 2013;336:209 220. (13) Tran Son Tay R, Hwang M, Garbey M, Jiang Z, Ozaki CK, Berceli SA. An experiment based model of vein graft remodeling induced by shear stress. Ann Biomed E ng 2008;36:1083 1091. (14) Berceli SA, Tran Son Tay R, Garbey M, Jiang Z. Hemodynamically driven vein graft remodeling: a systems biology approach. Vascular 2009;17 Suppl 1:S2 S9.
31 (15) Hwang M, Berceli SA, Garbey M, Kim NH, Tran Son Tay R. The dynamics of vein graft remodeling induced by hemodynamic forces: a mathematical model. Biomech Model Mechanobiol 2012;11:411 423. (16) Jiang Z, Berceli SA, Pfahnl CL et al. Wall shear modulation of cytokines in early vein grafts. J Vasc Surg 2004;40:345 350. (1 7) Jiang Z, Yu P, Tao M, Ifantides C, Ozaki CK, Berceli SA. Interplay of CCR2 signaling and local shear force determines vein graft neointimal hyperplasia in vivo. FEBS Lett 2009;583:3536 3540. (18) Davies MG, Hagen PO. Reprinted article "Pathophysiolog y of vein graft failure: a review". Eur J Vasc Endovasc Surg 2011;42 Suppl 1:S19 S29. (19) Lemson MS, Tordoir JH, Daemen MJ, Kitslaar PJ. Intimal hyperplasia in vascular grafts. Eur J Vasc Endovasc Surg 2000;19:336 350. (20) Muskhelishvili L, Latendres se JR, Kodell RL, Henderson EB. Evaluation of cell proliferation in rat tissues with BrdU, PCNA, Ki 67(MIB 5) immunohistochemistry and in situ hybridization for histone mRNA. J Histochem Cytochem 2003;51:1681 1688. (21) Kee N, Sivalingam S, Boonstra R, W ojtowicz JM. The utility of Ki 67 and BrdU as proliferative markers of adult neurogenesis. J Neurosci Methods 2002;115:97 105. (22) Jiang Z, Wu L, Miller BL et al. A novel vein graft model: adaptation to differential flow environments. Am J Physiol Heart Circ Physiol 2004;286:H240 H245. (23) Jiang Z, Shukla A, Miller BL et al. Tumor necrosis factor alpha and the early vein graft. J Vasc Surg 2007;45:169 176. (24) Marc Key, Helle Grann Wendelbo. Immunohistochemical Staining Methods Fifth Edition (Dako) , 57 120. 2009. Ref Type: Online Source (25) Jared Snider PhD. Blocking Endogenous Targets for IHC. 2013. 11 25 2013. Ref Type: Online Source (26) Creamer B, SHORTER RG, BAMFORTH J. The turnover and shedding of epithelial cells. I. The turnover in the gastro intestinal tract. Gut 1961;2:110 118. (27) Cliffe LJ, Humphreys NE, Lane TE, Potten CS, Booth C, Grencis RK. Accelerated intestinal epithelial cell turnover: a new mechanism of parasite expulsion. Science 2005;308:1463 1465. (28) Glagov S. Intimal hyperplasia, vascular modeling, and the restenosis problem. Circulation 1994;89:2888 2891.