Angiogenesis is defined as the growth of new blood vessels from pre-existing ones, in a process that involves endothelial and supporting cells, all guided by pro-angiogenic growth factors that will stimulate the appropriate signaling pathways.1 Horses are known to suffer from ischemic conditions such as laminitis, where dysfunction of the microvasculature of the laminal dermis has been implicated as an underlying cause.2 Similarly, other ischemic conditions such as delayed wound healing, or some forms of colic, can have severe implications in their career or be life-threatening. Because of this, research has been done on identifying cell-based therapies that can accelerate and improve angiogenesis in horses; however, most of the studies have been done in vitro.3,4 Endothelial colony-forming cells (ECFCs), a type of endothelial progenitor cell, have the important function of stimulating blood vessel formation and repair, making them good candidates to study cell-based therapy in horses.5 However, the lack of consensus in their phenotypic definition makes in vitro studies difficult to interpret.6 Because angiogenesis involves many cellular and molecular processes, in vivo studies are warranted, but this comes with the limitations of being more expensive, invasive, oftentimes difficult to analyze, and may raise ethical concerns.5
Ex vivo models of angiogenesis using vascular explant cultures are useful in evaluating angiogenesis because they overcome the limitations of in vitro techniques while reducing the complexity of in vivo models, thus, bridging the gap between these techniques.7,8 Ex vivo models can mimic most of the steps of angiogenesis seen in vivo, such as initial sprouting, matrix remodeling, and lumen formation. Paracrine signaling between endothelial cells, fibroblasts, macrophages, and pericytes is critical in the formation of vascular tubes, which can be mimicked in ex vivo assays.8,9 This signaling can be stimulated by equine-specific angiogenic growth factors, such as vascular endothelial growth factor A (VEGF-A). Fetal bovine serum (FBS) is widely used to supplement endothelial growth media (EGM); however, its xenogeneic origin increases the risks for immune reactions, and there are ethical concerns that arise from its harvesting.10–13 When culturing equine ECFCs, horse serum (HS) has been successfully used as a supplement to EGM for cell expansion.3,14 However, studies in humans are contradictory regarding the efficacy of human serum or plasma for the expansion of endothelial cells.10
Human platelet lysate is a rich source of growth factors, including VEGF-A, and has been used for the propagation of different cell types, including endothelial cells. In human-cultured ECFCs, platelet lysate enhanced the formation of vascular structures both in vitro and in vivo.15 Equine platelet lysate (ePL) has been shown to improve culture conditions for equine mesenchymal stem cells (MSCs), without altering proliferation rates, phenotype, or function, resulting in superior cell viability.16 Moreover, the concentration of growth factors such as PDGF-BB from pooled ePL is higher when compared with HS.17 To our knowledge, there are no reports on the effects of ePL as a supplement to EGM to study ex vivo angiogenesis in horses.
The mouse aortic ring assay is a common ex vivo model to study the vascular formation, being suitable in both anti- and pro-angiogenic studies. Aortic explants can form branching microvessels when embedded in the extracellular matrix. Furthermore, it is relatively low cost with the benefit of no interventions performed in live subjects.18 Although there is inherent variability in the angiogenic response of rings from the same animal, this system can be adapted to different experimental conditions, and the endothelium of the explants behaves similarly to endothelial cells in vivo.9,19
This assay has been used in other species such as chick embryos, dogs, humans, cows, and a single report of the use in horses.5,8 Equine arterial rings were used to evaluate the effects of cortisol on angiogenesis, by adapting the mouse aortic ring to equine facial and laminar arteries.20 However, there are no reports of wider use of equine arterial rings to evaluate angiogenesis and how this ex vivo model responds to different endothelial growth factors.
This study aimed to describe the arterial ring assay for the evaluation of ex vivo angiogenesis in horses by using equine facial arteries and to evaluate the effects of various types of growth media on vascular network formation. We hypothesized that equine arterial rings serve as an ex vivo model to study angiogenesis in horses, and sprouting angiogenesis will be enhanced when using ePL vs the standard HS-supplemented EGM.
Materials and Methods
Animals
For harvesting of platelet-rich plasma (PRP), 1 L of whole blood was collected from each of 6 healthy university-owned horses (American Quarter Horse [n = 4], and Thoroughbred [n = 2]; ages 5–16 years). A complete blood count was performed to determine the baseline platelet concentrations.
For the ex vivo angiogenesis assay, tissue containing facial arteries was collected within 5 minutes post-euthanasia from healthy horses euthanized for reasons unrelated to this study. To evaluate the dynamics of sprouting angiogenesis and the effect of growth factors in vascular network formation, facial arteries from 5 horses (American Quarter Horse [n = 2], Warmblood [n = 1], and Thoroughbred [n = 2]; ages 2–20 years) were collected post-euthanasia.
To evaluate the angiogenic effects of ePL, facial arteries from 6 additional horses (American Quarter Horse [n = 4], and Thoroughbred [n = 2]; ages 4–15) were used. Euthanasia was performed by IV injection of pentobarbital sodium and phenytoin sodium solution (Euthasol) at a dose of (78 mg/kg). The use of animals was approved and monitored by the Auburn University Institutional Animal Care and Use Committee (protocols 2020-3745, 2020-3809).
Facial artery dissection and arterial ring preparation
Tissues were maintained at 4 °C in serum-free endothelial basal media (EBM). A branch of the transverse facial artery of 1 mm diameter, was dissected and cleaned of surrounding adipose tissue. The lumen of the arteries was rinsed to remove blood clots with cold (4 °C) EBM using a 25 gauge needle attached to a 1 mL syringe. Arteries were cut into 1 mm rings using a scalpel blade and placed in cold (4 °C) EBM. Each artery yielded between 15 to 20 rings.
Matrigel (BD Biosciences) was prepared as per the manufacturer’s instructions. Briefly, it was thawed overnight at 4 °C, and 40 μL was added to each well of a chilled (−20 °C) 96-well plate, and allowed to polymerize for 15 minutes in an incubator at 37 °C, 5% CO2. Each ring was dried using task wipers and placed over the Matrigel, with the lumen parallel to the gel surface. An additional 40 μL of Matrigel was placed on top of the ring so it was fully covered. Matrigel was allowed to polymerize before exposing the rings to the study conditions.
Preparation of equine platelet lysate
Platelet-rich plasma was obtained based on a simple tube centrifugation protocol.21 Briefly, 1 L of blood was collected aseptically from the jugular vein and into a blood collection bag containing citrate phosphate dextrose adenine anticoagulant (126 mL). Blood was centrifuged at 345 X g, 4 °C for 10 minutes, and the plasma was further centrifuged at 615 X g 4 °C for 5 minutes. The supernatant was centrifuged at 1,160 X g 4 °C for 5 minutes. The platelet-poor plasma (PPP) was decanted and saved for future use. The remaining PRP pellet from the 6 horses was pooled, yielding a mean platelet concentration of 2,758 X 103/μΛ. PRP was stored at −80 °C.
To produce the lysate, platelets were fractured using 3 freeze-thaw cycles. Samples frozen at −80 °C were thawed at 37 °C and 2 more cycles were completed. Confirmation of cell lysis and absence of WBC contamination was done by measuring the platelet concentration (1 X 103/μL) and WBC count (0 X 103/μL). Based on a mean baseline platelet concentration of 120 X 103/μL, and the mean PRP platelet concentration between the 6 horses, a mean of 24-fold increase was obtained. The ePL was diluted using pooled PPP to a concentration of 10-fold from the baseline platelet concentration and centrifuged once at 4,800 X g 4 °C for 60 minutes, followed by 2 centrifugations of 30 minutes each at 4,800 X g room temperature. The supernatant was collected and stored at −80 °C for future use. The ePL at 10-fold concentration was diluted by adding pooled PPP to produce 2-, and 5-fold increases from baseline platelet concentrations. Samples were processed using a sterile technique within a biosafety cabinet.
Preparation of endothelial growth media
Endothelial growth media (EGM-2 with Bullet Kit, Lonza), free of FBS was prepared as per the manufacturer’s instructions by mixing the growth factor-free EBM with the Bullet Kit (the Bullet Kit contains human growth factors). For the different conditions, 10% of either HS or ePL at 2-fold, 5-fold, and 10-fold increases from baseline platelet concentrations were added to the EGM, and filtered to remove solids, bacteria, or debris using a 0.45 μm cellulose acetate membrane filter.
Dynamics of sprouting angiogenesis
Five rings from each of the 3 horses (n = 15) were exposed to EGM with 10% HS (EGM + HS, standard equine endothelial cell growth media). Rings were analyzed and photographed daily using an inverted phase contrast microscope at a 20X magnification to evaluate the time of appearance of the first sprout (FS), matrix lysis (ML), and vascular regression (VR). FS was defined as the first vessel-like protrusion observed. ML corresponded to the evidence of Matrigel degradation, seen as a halo on the periphery of the ring. VR was observed as the breakage of the vascular network occurring at the periphery. The time (day) of appearance of the FS, VR, and ML were compared between horses.
Effect of growth factors in angiogenesis
Fifteen rings from each of the 2 horses were embedded in Matrigel as previously described. From each horse, 3 rings were randomly assigned to each condition (n = 6): (1) EGM+HS (endothelial cell growth media containing human growth factors and equine serum, positive control), (2) EBM (free of growth factors and serum), (3) EBM + human vascular endothelial growth factor (hVEGF), (4) EBM + HS, or (5) EGM + EDTA (negative control). Media was replaced every 48 hours until analysis on day 7. Vascular network area (VNA) was determined using Fiji software and represented the area in micrometers square (μm2) contained between the perimeter of the vascular network and the arterial ring. The maximum network growth (MNG) in micrometers (μm) was calculated from the average of the 8 maximum lengths measured from the center of the ring to the furthest angiogenic sprout. VNA and MNG were compared among groups on day 7. Day 7 was chosen based on results from the previous phase.
Effect of equine platelet lysate on angiogenesis
Arterial rings from the facial artery were dissected and embedded in Matrigel as previously described. Rings from 6 horses (20–30 rings per horse) were supplemented with EBM (free of growth factors and serum) for 6 days. For inclusion in the study, rings should have evidence of sprouting angiogenesis after culture with EBM (days −6 to 0). Out of 174 rings, 111 met the inclusion criteria. Rings were randomly selected and assigned by triplicate to each of the groups: EGM-10xePL (EGM + 10% 10-fold ePL) (n = 18), EGM-5xePL (EGM + 10% 5-fold ePL) (n = 18), EGM-2xePL (EGM + 10% 2-fold ePL) (n = 18), EGM-HS (EGM+10% HS) (n = 18), EGM-PPP (EGM + 10% PPP) (n = 9), EBM-PPP (EBM + 10%PPP) (n = 15), and EBM (n = 15). Once sprouting was established (day 0), EBM was removed and replaced with 200 μL of each of the study growth medias. Photomicrographs at a 20X magnification were obtained at baseline (day 0), days 1, 2, and 3 (Figure 1).
Analysis of the effects of equine platelet lysate on angiogenesis. (A) Diagram showing the study timeline. (B) For determination of the vascular network area, the background was removed and a binary image created to quantify the area in black. (C) The binary image was converted into a single-pixel skeleton; thus, allowing for pixel count of branches and determination of vascular density. (D) The Strahler analysis of the previous figure assigned different colors to each itineration of the vascular tree to count branches at each level. EBM = endothelial basal media. VEGF-A = vascular endothelial growth factor A. Bar = 200 µm.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0023
Equine VEGF-A concentrations
Supernatants from arterial ring explants (n = 12) supplemented with EGM-10xePL, EGM-5xePL, EGM-2xePL, EGM-HS, EBM-PPP, and EBM, were collected on days 0 (following 6 days of culture in EBM) and at day 3 for measurement of VEGF-A concentrations by equine-specific ELISA. VEGF-A concentrations were also determined in the different growth medias before being used and defined as a baseline. Samples were stored at −80 °C until analysis.
ELISA was performed as per the manufacturer’s instructions (Equine VEGF-A, Kingfisher Biotech, Inc). Capture antibody (2.5 μg/mL) was prepared by diluting the anti-equine VEGF-A polyclonal antibody (1 mg/mL) in phosphate buffer solution, and 100 μL was added to each well of a 96-well plate. The plate was incubated for 12–24 hours, followed by the addition of 100 μL of 4% BSA as the blocking buffer, and incubated at room temperature for 1–3 hours. Recombinant equine VEGF-A was added as the standard. Samples were added to the wells and incubated for 1 hour at room temperature. A detection antibody (biotinylated anti-equine VEGF-A polyclonal antibody) at a concentration of 0.2 μg/mL was added to the wells after washing (0.05% Tween-20). Following another wash, streptavidin-HRP was added and incubated for 30 minutes. Tetramethylbenzidine substrate solution was added to the wells and developed in the dark for 30 minutes. Immediately after, the stop solution was added and absorbance was measured at a wavelength of 450 nm using a plate reader (SpectraMax iD5).
Image analysis
Image processing and analysis were further refined, and photomicrographs were modified using Fiji image processing software. Images were converted into 8-bit, and the low-intensity background was removed by adjusting the gamma feature to values between 1–1.5. A bandpass filter between 2 and 50 pixels was applied to filter out large and small structures. Following this, the edges were highlighted by using a variance filter. To reduce background artifacts, a subtraction of 100 was applied.22
The image was converted to binary with pixel intensities of 0 for white and 255 for black. A cycle of close-dilate was done to smooth objects and fill in small holes. Finally, any artifact was manually deleted to isolate the vascular tree. To obtain the VNA in micrometers square (μm2), the threshold was adjusted to 255. The binary image was skeletonized, and using histogram analysis, the number of pixels in black (255) was recorded for the vascular density. Finally, a Strahler analysis (http://fiji.sc/Strahler_Analysis) was applied to obtain the number of branches (Figure 1).
Statistical analysis
Statistical analysis was done using commercial software (SAS 9.5M7 or GraphPad Prism 9.3.1). Normality was assessed by the Shapiro-Wilk test and QQ plots. Medians of FS, VR, and ML between horses and VNA and MNG between conditions were compared by Kruskal Wallis tests with Dunn’s post hoc tests. The time of appearance of FS, VR, and ML was recorded for descriptive statistics.
To compare the effects of EGM supplemented with ePL, HS, or PPP, EBM-PPP, and EBM, rings exposed to the 3 different concentrations of ePL (EGM-2xePL, EGM-5xePL, and EGM-10xePL) were combined into a single group (n = 54). Following normalization for baseline, means were compared among groups on days 1, 2, and 3 by 2-way repeated measures ANOVA with Tukey’s post hoc test. VEGF-A concentrations between rings exposed to the different growth medias were compared by 1-way ANOVA with Tukey’s post hoc test. The effect of the different concentrations of ePL on vascular network formation and the growth rate between rings exposed to the different growth medias were compared by linear regression. The effect of VEGF-A in the number of branches, density, and VNA on day 3 was determined by Pearson correlation; P < .05 was considered statistically significant.
Results
Assessment of sprouting angiogenesis
All equine arterial rings exposed to EGM + HS were able to produce new blood vessels after 5 days, with the earliest being 3 days (median = 4.47 days). Vascular regression was observed as early as day 9, with a maximum of 13 days (median = 10.73 days). Finally, ML was not observed until day 11, with a maximum of 13 days (median = 12.07 days; Figure 2). A window between days 5 and 9, based on the maximum for FS and the first evidence of VR was determined to be optimum for ring selection in further experiments. Within the different groups, no statistically significant differences were observed between the 3 horses for FS (P = .3934), VR (P = .0607), or ML (P = 0.2364).
Violin plots representing the distribution of dynamics of sprouting angiogenesis for appearance of (A) first sprout, (B) vascular regression, and (C) Matrigel lysis in days from 15 arterial rings of 3 horses (n = 5) supplemented with EGM-HS (endothelial growth media + horse serum). The median is represented by the solid line, and the 25th and 75th percentiles are represented by the dashed line.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0023
Effect of growth factors in angiogenesis
Arterial rings exposed to EBM and EGM + EDTA with growth media change every 48 hours did not show any vascular growth. VNA in EGM + HS was significantly larger than EBM (P = .0015) and EGM + EDTA (P = .0015). Significant differences in MNG were observed between EGM + HS when compared with EBM + hVEGF (P < .0001), EBM (P < .0001), and EGM + EDTA (P < .0001). Both EBM + HS and EBM + hVEGF had significantly larger MNG when compared with EBM and EGM + EDTA (P < .0001). A trend toward higher MNG was observed for the groups containing HS (EGM + HS and EBM + HS) when compared with EBM + hVEGF (Figure 3).
Effect of growth factors on vascular network formation. Side-by-side box plots comparing the (A) vascular network area and (B) maximum network growth, from the arterial rings after 7 days of exposure to the different growth medias (n = 6). (C) Vascular network area was the difference between the area inside the perimeter of the vascular growth (solid line) and the perimeter of the ring (dotted line). (B) Maximum network growth was determined by dividing the arterial ring into quadrants (solid lines). For each quadrant, the 2 maximum growths from the surface of the ring (dotted lines) were recorded. Finally, the 8 maximum growths obtained were averaged. The lower and upper limits of the box represent the 25th and 75th percentiles, the median is represented by the horizontal line within the box, and the whiskers delimit the range. Statistically significant differences are present when the letters above each of the boxes differ between groups. EGM = endothelial growth media. EBM = endothelial basal media. HS = horse serum. hVEGF = human vascular endothelial growth factor. Bar = 200 µm.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0023
Effect of equine platelet lysate on angiogenesis
After 3 days of exposure to the different conditions, no overall significant differences between treatments were observed at any time point for VNA (P = .6271), number of branches (P = .5014), or density (P = .4394). Some treatments significantly increased the number of branches, VNA, and density over time more than others, and a significant effect of time in the different angiogenic parameters was observed (P < .0001; Figure 4). However, when comparing growth rates between the different treatments, no significant differences were observed for VNA (P = .9825), branches (P = .4082), and density (P = .3048).
Effects of equine platelet lysate on vascular network formation. (A) vascular network area, (B) number of branches, and (C) vascular density, based on the number of pixels, for rings exposed to the different conditions after 3 days with no growth media change. No overall statistically significant differences were observed for any of the treatments. Mean ± SEM is represented in each bar. The effect of increasing concentrations of ePL on vascular growth at each time point is represented by the means for (D) vascular network area, (E) number of branches, and (F) density. The concentration of ePL (x axis) corresponds to the fold increase in platelets from the mean baseline platelet concentration, with 2-fold, 5-fold, and 10-fold being 2x, 5x, and 10x, respectively, and indicated by the vertical dashed lines. EGM-PPP was used as the negative control (x = 0). EGM = endothelial growth media. EBM = endothelial basal media. HS = horse serum. ePL = equine platelet lysate. PPP = platelet-poor plasma.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0023
The R2 values were positive indicating increases in VNA, branches, and density with increasing concentrations of ePL; however, no difference was observed between slopes on the different days for VNA (P = .1554), number of branches (P = .2923), or density (P = .1964), and the slopes were not significantly different from 0 by day 3 (Figure 4).
Equine VEGF-A concentrations
On day 3, VEGF-A concentration was significantly different between groups (P = .0351), with a higher trend for the rings exposed to EGM containing ePL at any concentration and HS (Figure 5).
Concentrations of vascular endothelial growth factor A (VEGF-A) in supernatants of media of arterial rings. (A) Side-by-side box plots comparing VEGF-A concentrations in supernatants after 3 days of exposure of the arterial rings to the different growth medias (n = 12). The lower and upper limits of the box represent the 25th and 75th percentiles, median is represented by the horizontal line within the box, and the whiskers delimit the range. The dot represents the mean. Statistically significant differences are present when the letters above each of the boxes differ between groups. (B) Mean ± SEM of the change in vascular network area about VEGF-A concentrations. Note the different scales for the two segments of the x axis. EBM = endothelial basal media. HS = horse serum. ePL = equine platelet lysate. PPP = platelet-poor plasma. 2x, 5x, and 10x represent the ePL concentration based on the fold increase from the baseline platelet concentration.
Citation: American Journal of Veterinary Research 84, 7; 10.2460/ajvr.23.02.0023
As VEGF-A concentration increased, particularly for the groups containing ePL and HS, there was a positive effect on VNA (P = .0243; Figure 5). Descriptive statistics for concentrations of VEGF-A are summarized (Table 1).
Mean ± SD for vascular endothelial growth factor A (VEGF-A) concentrations in the supernatants of arterial rings exposed to different conditions.
| Condition | Baseline | Day 3 | |
|---|---|---|---|
| Mean ± SD (ng/mL) | Mean ± SD (ng/mL) | Range (ng/mL) | |
| 10xePL | 0.074 ± 0.006* | 0.574 ± 0.111 | 0.414 – 0.653 |
| 5xePL | 0.074 ± 0.006* | 0.583 ± 0.092 | 0.450 – 0.666 |
| 2xePL | 0.251 ± 0.350 | 0.597 ± 0.165 | 0.391 – 0.795 |
| HS | 0.096 ± 0.040 | 0.576 ± 0.109 | 0.424 – 0.682 |
| PPP | 0.074 ± 0.006* | 0.328 ± 0.207 | 0.050 – 0.504 |
| EBM | 0.074 ± 0.006* | 0.373 ± 0.105 | 0.217 – 0.434 |
VEGF-A concentrations on day 0 (baseline) and day 3 are provided. The asterisk (*) indicates that concentrations were below the assay’s limit of detection of 0.074 ng/mL. EBM = endothelial basal media. HS = horse serum. ePL = equine platelet lysate. PPP = platelet-poor plasma. 2x, 5x, and 10x represent the ePL concentration based on the fold increase from the baseline platelet concentration.
Discussion
Here we described the ex vivo angiogenic response of equine arteries when exposed to EGM supplemented with different sources of growth factors by the use of an arterial ring assay. This ex vivo model using equine arteries has been described in a report that studied the effect of glucocorticoids on angiogenesis; however, FBS was used as a supplement to the EGM.20 Our results showed that equine arterial rings supplemented with EGM-HS every 48 hours can sprout new vessels. Furthermore, equine arterial rings supported sprouting angiogenesis in serum-free media, allowing for later testing of different biological products as candidates for regenerative angiogenesis. An important finding during the initial phase was the fact that equine arterial rings can respond differently depending on the presence of growth factors, and greater vascular growth was observed in the group exposed to EBM-HS compared with EBM alone, suggesting that equine-specific growth factors stimulate the angiogenic response of equine arteries. Although not statistically significant, this could be supported by the fact that rings exposed to EBM-HS had larger VNA and MNG when compared with rings exposed to human VEGF (EBM + hVEGF).
Due to the variability in angiogenic response between rings, preselecting samples with similar angiogenic responses after sprouting has started, similar to studies in rat models, is important.23 The determination of parameters such as the appearance of FS was critical as a starting point for determining the best time for ring selection. Our results showed that the appearance of the FS in horses, around day 5, was similar to reports using rats, where new blood vessels were present between days 3–5 of culture.23,24 When using Matrigel, we observed that neovessels arising from equine arteries will undergo regression as early as day 9; thus, analysis between days 5 and 9 of culture was optimal. Although a different type of matrix was used, the dynamics observed with equine arterial rings explants are similar to reports using rat aortas in a collagen matrix, where the neovessel growth phase starts after 2 to 3 days of culture and continues until regression between day 7 to 10. This regression consists of a process of fragmentation, disintegration, and retraction, and is often associated with degradation of the basement membrane.9
This study demonstrated that sprouting angiogenesis in the equine arterial ring assay is dependent on the growth media used. Using this model with equine arteries allowed for both stimulation of angiogenesis through the addition of growth factors, as well as inhibition by the addition of EDTA. Thus, this assay can be used to test both promotions of equine angiogenesis as well as regression. As expected by using equine tissues, supplementation of EBM with equine-specific growth factors contained in HS resulted in better vascular network area than EBM containing human VEGF-A, although this difference was not statistically significant. Moreover, the lack of difference in VNA and MNG between rings exposed to EGM-HS and EBM-HS denotes the major influence of HS in angiogenesis, independently of other factors contained in the company provided growth factor cocktail with EGM. The equine serum contains angiogenic growth factors such as epidermal growth factor, TGF-β, insulin-like growth factors, and most importantly VEGF-A. VEGF-A is highly implicated in the recruitment of tip cells and initiation of sprouting angiogenesis; thus, providing equine VEGF-A through equine-derived blood products will stimulate VEGF receptor 2 with high specificity.25,26 This is important because this assay can be implemented to study the allogenic use of different biological products for regenerative medicine.
We tested the use of alternative allogenic sources of vascular growth factors, and ePL was demonstrated to be a good alternative to provide equine-specific factors to induce sprouting angiogenesis. The treatment type and time that the rings were exposed to the different conditions had an effect on vascular network formation. Furthermore, using ePL seemed to be comparable with the use of HS to support vascular growth. Interestingly, the EBM-PPP group also trended toward a higher number of branches, VNA, and density, which was an unexpected result because PPP is considered to have significantly less VEGF concentration when compared with PRP or platelet lysate.27 Even though there are no reports of concentrations of growth factors in equine PPP, it has been observed that mouse PPP enhances angiogenesis when used in mouse wounds compared with the use of PRP, evidenced by a higher blood vessel density and expression of VEGFR-2.28 Moreover, human umbilical cord endothelial cells cultured for 48 hours in the presence of human PPP had greater expression of the endothelial cell marker CD34 and VEGFR-2 compared with cells exposed to PRP.28 The effects of PPP in our study could be explained by the fact that PRP at high concentrations can elicit an inhibitory and cytotoxic effect. Moreover, PPP is also metabolically active and may also stimulate tissue regeneration by its bioactive molecules such as PDGF and IGF-1.29 Future studies assessing the cytotoxic effects of blood products in equine endothelial cells are warranted to determine the ideal concentrations to use in cell culture. Our results demonstrated a trend toward an increase in vascular growth with higher ePL concentrations. However, a study evaluating equine MSCs observed a dose-dependent response to the addition of ePL up to a concentration of 30%, with further increases resulting in the inhibition of cell proliferation.30 Therefore, further studies with a higher sample size evaluating a wider range of ePL concentrations are required to determine the maximum concentration that will support vascular growth without the potential to inhibit angiogenesis.
There is limited information regarding the concentration and composition of growth factors in HS and ePL; however, it has been reported that concentrations in ePL are higher than previously reported in HS, specifically for TGF-β1 and PDGF-BB.17,31 In our study we obtained a lower concentration of equine VEGF-A in ePL than previously reported; however, a different assay was used and a smaller number of animals.17 Even though we did not observe large variations in VEGF-A concentrations, a previous report found the opposite, with large ranges observed between horses, hence the importance of using pooled ePL.17 Our study is consistent with others, demonstrating that ePL may be an alternative source of growth factors for cell culture, improving the angiogenic signaling pathway toward an endothelial phenotype.
Other growth factors, besides VEGF-A, contained in ePL, PPP, or HS were not measured in the present study; therefore, the VEGF-A concentrations observed on day 3 could be influenced by other factors upregulating the release of VEGF-A from the explants. Such is the case of PDGF-BB contained in biological products, which can upregulate the release of VEGF-A by endothelial cells.32 Even though our results showed a positive correlation between VEGF-A concentrations and VNA, no group was exposed to growth media depleted of VEGF-A; therefore, other growth factors contained in the biological products could have enhanced vascular network formation. This study also showed evidence of the endogenous production of VEGF-A by equine arterial rings, and this might explain the fact that sprouting occurs in rings exposed to EBM for 6 days when growth media is not changed. This has been explained in rat arteries, where the release of growth factors and cytokines is responsible for paracrine stimulation of sprouting angiogenesis.23 This finding is of importance since equine arterial rings can be used to study the effects of different growth factors, without the interference from other factors used to start sprouting. Interestingly, for the first phase of our study, rings exposed to EBM that had media change every 48 hours did not show sprouting angiogenesis, which could be due to the removal of endogenous growth factors from the old growth media.
Different approaches have been described to analyze angiogenesis using this ex vivo model. Historically manual sprout count has been done; however, due to the exponential growth leading to a high number of branches and the 3D nature of the set-up, the visual real-time count is complicated.33 Different software has capabilities for counting blood vessels; however, image modifications are required to reduce background artifacts.8,34 Here we modified a protocol used by Rohban et al, 2013 to remove the background and isolate the vascular tree as a binary image, which proved to be an effective approach to have objective measurements for VNA and density.33 Furthermore, we demonstrated the utility of the Strahler analysis, originally developed for neural cells, for counting terminal branches of a vascular tree.
Some limitations of this ex vivo assay using rats or mice include the variability in angiogenic responses among different animals or rings, as well as differences in ring dimensions.5 Variability among rings in this study is illustrated by the wide error bars (Figure 4). This is likely due to the use of animals within a wide range of ages (2–20 years old), and the high genetic diversity of horses versus similar ring models in mice or rats. It has been determined that older mice and rats produce fewer vessels compared with younger ones.19,35 With high variability, achieving statistical power was a limitation of this study. This study aimed for an overall medium effect size of 0.5 (Cohen’s d), which was achieved based on the data. With a medium effect size and no overall statistical difference observed, the existence of a true direct effect of ePL on vascular growth should be interpreted with caution. Our data only achieved a Cohen’s d effect size of 0.6 when comparing EBM alone with the ePL groups, where a larger difference was expected. Therefore, increasing the sample size may increase the magnitude, or at least the precision, of the effect size, and statistically significant effects of ePL on vascular growth may be detected. Another limitation was the use of phase contrast microscopy alone with no immunostaining for endothelial cells, which could have aided in discriminating endothelial cells from other cells at the initial stages of angiogenesis.1 However, the exponential growth observed in the microvessel-like structure, and the long sprouts were most likely consistent with sprouting angiogenesis.
Matrigel is of xenogenic origin when used with equine arteries, and will contain mouse growth factors, making difficult the isolation of equine-specific growth factors as players in angiogenesis. To account for this, all arterial rings were grown in EBM (no growth factors) before exposure to the study conditions; furthermore, the data were corrected for baseline to reduce the intrinsic effects of Matrigel in vessel growth. Despite these limitations, this study is a starting point for the evaluation of ePL on vascular growth and as a potential source of growth factors in cell expansion.
We concluded that the method established for the mouse aortic ring assay could be used with equine arteries to study the effect of allogenic sources of growth factors. This method provides a large sample size, high reproducibility, and low cost, and it is an option to consider before moving into in vivo experiments.8 The results show that there is variability in responses of rings, that are likely inherent to the horse, despite selecting rings that demonstrate sprouting activity for experimental conditions. Nonetheless, it was demonstrated that supplementation with PPP, ePL, and HS can support sprouting angiogenesis in this model. The results obtained are important for the future selection of the most appropriate culture conditions for equine ECFCs, and it serves as a starting point for studies using the coculture of progenitor cells with an ex vivo model.
Acknowledgments
Supported by Animal Health and Disease Research Funds, Auburn University, and the Birmingham Racing Commission.
The authors have nothing to declare.
The authors thank Jessica Brown, Qiao Zhong, and Katelyn Kahler for their technical support.
References
- 1.↑
Baker M, Robinson S, Lechertier T, et al. Use of the mouse aortic ring assay to study angiogenesis. Nat Protoc. 2012;7:89–104. doi:10.1038/nprot.2011.435
- 2.↑
van Eps AW, Belknap JK, Schneider X, et al. Lamellar perfusion and energy metabolism in a preferential weight bearing model. Equine Vet J. 2021;53:834–844. doi:10.1111/evj.13356
- 3.↑
Sharpe AN, Seeto WJ, Winter RL, et al. Isolation of endothelial colony-forming cells from blood samples collected from the jugular and cephalic veins of healthy adult horses. Am J Vet Res. 2016;77:1157–1165. doi:10.2460/ajvr.77.10.1157
- 4.↑
Pascucci L, Alessandri G, Dall’Aglio C, et al. Membrane vesicles mediate pro-angiogenic activity of equine adipose-derived mesenchymal stromal cells. Vet J. 2014;202:361–366. doi:10.1016/j.tvjl.2014.08.021
- 5.↑
Katakia YT, Duddu S, S N, et al. Ex vivo model for studying endothelial tip cells: revisiting the classical aortic-ring assay. Microvasc Res. 2020;128:103939. doi:10.1016/j.mvr.2019.103939
- 6.↑
Urbich C, Dimmeler S. Endothelial progenitor cells functional characterization. Trends Cardiovasc Med. 2004;14:318–322. doi:10.1016/j.tcm.2004.10.001
- 8.↑
Kapoor A, Chen CG, Iozzo RV. A simplified aortic ring assay: a useful ex vivo method to assess biochemical and functional parameters of angiogenesis. Matrix Biol Plus. 2020;6–7:100025. doi:10.1016/j.mbplus.2020.100025
- 9.↑
Aplin AC, Fogel E, Zorzi P, et al. The aortic ring model of angiogenesis. Methods Enzymol. 2008;443:119–136.
- 10.↑
Bauman E, Granja PL, Barrias CC. Fetal bovine serum-free culture of endothelial progenitor cells-progress and challenges. J Tissue Eng Regen Med. 2018;12:1567–1578. doi:10.1002/term.2678
- 11.
Sundin M, Ringdén O, Sundberg B, et al. No alloantibodies against mesenchymal stromal cells, but presence of anti-fetal calf serum antibodies, after transplantation in allogeneic hematopoietic stem cell recipients. Haematologica. 2007;92:1208–1215. doi:10.3324/haematol.11446
- 12.
Even MS, Sandusky CB, Barnard ND. Serum-free hybridoma culture: ethical, scientific and safety considerations. Trends Biotech. 2006;24:105–108. doi:10.1016/j.tibtech.2006.01.001
- 13.↑
Horwitz EM, Gordon PL, Koo WK, et al. Isolated allogeneic bone marrow-derived mesenchymal cells engraft and stimulate growth in children with osteogenesis imperfecta: implications for cell therapy of bone. Proc Natl Acad Sci USA. 2002;99:8932–8937. doi:10.1073/pnas.132252399
- 14.↑
Salter MM, Seeto WJ, DeWitt BB, et al. Characterization of endothelial colony-forming cells from peripheral blood samples of adult horses. Am J Vet Res. 2015;76:174–187. doi:10.2460/ajvr.76.2.174
- 15.↑
Kim H, Prasain N, Vemula S, et al. Human platelet lysate improves human cord blood derived ECFC survival and vasculogenesis in three dimensional (3D) collagen matrices. Microvasc Res. 2015;101:72–81. doi:10.1016/j.mvr.2015.06.006
- 16.↑
Naskou MC, Sumner SM, Chocallo A, et al. Platelet lysate as a novel serum-free media supplement for the culture of equine bone marrow-derived mesenchymal stem cells. Stem Cell Res Ther. 2018;9:75. doi:10.1186/s13287-018-0823-3
- 17.↑
Sumner SM, Naskou MC, Thoresen M, et al. Platelet lysate obtained via plateletpheresis performed in standing and awake equine donors. Transfusion. 2017;57:1755–1762. doi:10.1111/trf.14124
- 18.↑
Nowak-Sliwinska P, Alitalo K, Allen E, et al. Consensus guidelines for the use and interpretation of angiogenesis assays. Angiogenesis. 2018;21:425–532. doi:10.1007/s10456-018-9613-x
- 19.↑
Aplin AC, Fogel E, Zorzi P, et al. The aortic ring model of angiogenesis. Methods Enzymol. 2008;443:119–136.
- 20.↑
Morgan R, Keen J, Halligan D, et al. Species-specific regulation of angiogenesis by glucocorticoids reveals contrasting effects on inflammatory and angiogenic pathways. PLoS ONE. 2018;13(2):e0192746. doi:10.1371/journal.pone.0192746
- 21.↑
Fontenot RL, Sink CA, Werre SR, et al. Simple tube centrifugation for processing platelet-rich plasma in the horse. Can Vet J. 2012;53:1266–1272.
- 22.↑
Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Methods. 2012;9:671–675. doi:10.1038/nmeth.2089
- 23.↑
Blot G, Sartoris T-M, Sennlaub F, et al. Modifications to the classical rat aortic ring model to allow vascular degeneration studies. STAR Protoc. 2021;2:100281–100281. doi:10.1016/j.xpro.2020.100281
- 24.↑
Iqbal F, Gratch YS, Szaraz P, et al. The aortic ring co-culture assay: a convenient tool to assess the angiogenic potential of mesenchymal stromal cells in vitro. J Vis Exp. 2017;(127):56083. doi:10.3791/56083
- 26.↑
Naito H, Iba T, Takakura N. Mechanisms of new blood-vessel formation and proliferative heterogeneity of endothelial cells. Int Immunol. 2020;32:295–305. doi:10.1093/intimm/dxaa008
- 27.↑
Zhou B, Ren J, Ding C, et al. Rapidly in situ forming platelet-rich plasma gel enhances angiogenic responses and augments early wound healing after open abdomen. Gastroenterol Res Pract. 2013;2013:926764. doi:10.1155/2013/926764
- 28.↑
Shahidi M, Vatanmakanian M, Arami MK, et al. A comparative study between platelet-rich plasma and platelet-poor plasma effects on angiogenesis. Med Mol Morphol. 2018;51:21–31. doi:10.1007/s00795-017-0168-5
- 29.↑
Chellini F, Tani A, Zecchi-Orlandini S, et al. Influence of platelet-rich and platelet-poor plasma on endogenous mechanisms of skeletal muscle repair/regeneration. Int J Mol Sci. 2019;20:683. doi:10.3390/ijms20030683
- 30.↑
Russell KA, Koch TG. Equine platelet lysate as an alternative to fetal bovine serum in equine mesenchymal stromal cell culture – too much of a good thing? Equine Vet J. 2016;48:261–264. doi:10.1111/evj.12440
- 31.↑
Giraldo CE, López C, Álvarez ME, et al. Effects of the breed, sex and age on cellular content and growth factor release from equine pure-platelet rich plasma and pure-platelet rich gel. BMC Vet Res. 2013;9:29–29. doi:10.1186/1746-6148-9-29
- 32.↑
Apte RS, Chen DS, Ferrara N. VEGF in signaling and disease: beyond discovery and development. Cell. 2019;176:1248–1264. doi:10.1016/j.cell.2019.01.021
- 33.↑
Rohban R, Reinisch A, Etchart N, et al. Identification of an effective early signaling signature during neo-vasculogenesis in vivo by ex vivo proteomic profiling. PLoS ONE. 2013;8(6):e66909. doi:10.1371/journal.pone.0066909
- 34.↑
Iqbal F, Szaraz P, Librach M, et al. Angiogenic potency evaluation of cell therapy candidates by a novel application of the in vitro aortic ring assay. Stem Cell Res Ther. 2017;8:184. doi:10.1186/s13287-017-0631-1
- 35.↑
Zippel N, Ding Y, Fleming I. A modified aortic ring assay to assess angiogenic potential in vitro. Methods Mol Biol. 2016;1430:205–219.
