Stromal Vascular Fraction Reverses the Age-Related Impairment in Revascularization following Injury

The microvasculature primarily controls and regulates blood flow through an organ, but impairment of blood flow can result in tissue ischemia [1‒3]. As aging progresses, aberrant physiological remodeling responses can serve as the foundation that eventually limit tissue plasticity during pathophysiological remodeling, such as following injury [4]. The increase in systemic oxidative stress [5], cellular senescence [6], and chronic inflammation [7] that are characteristics of advancing age has been linked to endothelial cell dysfunction and a decline in the capacity of a microvascular network to sprout new vessels via angiogenesis and collateralization.

Current medications and surgical interventions utilized after an ischemic event act by decreasing organ workload, mitigating inflammation, or grafting of large vessels to redirect blood flow, but none target the multifactorial pathology of endothelial cell dysfunction to promote growth of the microvasculature [8]. These treatments stand in contrast to the gains observed in cell-based therapies [9]. Adipose tissue contains an abundant source of multipotent cell populations such as mesenchymal stem cells (MSCs), endothelial cells, pericytes, and hematopoietic cells including macrophages, T cells, and other cells of the perivascular niche [10, 11]. Of note, T cells have been shown to play a role in stimulating angiogenesis in ischemic areas [12] due to migration to the area of injury [13, 14] and collateralization specifically by homing to sites of active angiogenesis [15].

In recent years, studies have demonstrated equal efficacy for stromal vascular fraction (SVF) and bone marrow MSCs in treating myocardial ischemia [16, 17], promoting angiogenesis [18], and reducing inflammation [19]. Availability of autologous and HLA-matched allogeneic SVF sources confers inherently minimal risk of rejection [20], offering an ideal candidate in vascular therapeutics due to the ability to leverage an easy-to-harvest, real-time, point-of-care approach [21]. Despite these benefits of cell-based therapies, there are few studies conducted using advanced age as the model of pathology and examining the impact of donor age on the vasculogenic properties of cell therapy as utilized in the current study.

To investigate the effect of age on SVF cell populations as well the potential to increase angiogenesis in an aged microvasculature, we utilized an ex vivo model composed of connective tissues located along the intestinal mesentery, termed mesenteric windows [22]. The ex vivo mesentery culture model developed by the Murfee laboratory was utilized in this study for ease of access, time-lapse imaging capabilities, and survivability in culture [23]. This model has shown the functional effect of pericytes on tip cell formation and maintenance of in vivo-like endothelial cell phenotypes along capillary sprouts during active angiogenesis [24‒27]. Furthermore, exteriorization injury (EI) of the tissue has been shown to be a reproducible model to induce acute hypoxia and stimulate angiogenesis [28, 29] but has not been previously used in a setting of advanced age. Prior studies from our laboratory have shown that systemically delivered SVF from young donors can improve coronary microvascular perfusion in aged rats [30]. Furthermore, a single intravenous (i.v.) injection of SVF from young donors results in a biodistribution of cellular engraftment across various organs and blood vessels [30]. However, neither the therapeutic potential of SVF from aged donors on tissue perfusion nor the changes in vascular density as a result of cell administration have been studied. This study aims to leverage the benefits of the mesenteric ex vivo prep to investigate the impact of donor age on vasculogenic potential in a setting of advanced age and injury with intravenously administered SVF therapy. Compared to SVF from young rats, we hypothesized that SVF from aged rats would exhibit increased inflammatory cell populations (M1 macrophages and T cells), limit the ability of the microvasculature to collateralize following hypoxic injury, and reduce the T cell-mediated angiogenic potential.

The Institutional Animal Care and Use Committees of the University of Florida and the University of Louisville approved this study, and all procedures followed the NIH Guide for the Care and Use of Laboratory Animals [31]. Young (3–9 months) and aged (23–24 months) male and female Fischer 344 rats (Harlan Laboratories, Indianapolis, IN, USA and National Institute of Aging, Bethesda, MD, USA, respectively) were group housed with free access to food and water and maintained on regular 12-h light/dark cycles. Green fluorescent positive (GFP+) transgenic Fischer 344 male and female rats, maintained in-house (Louisville), aged 3–9 and 24 months were utilized as young and old SVF donors, respectively. All rats were acclimated to facility conditions for a minimum of 1 week and then randomly divided into the experimental groups described in detail below and depicted in Figure 1.

SVF cells were isolated as previously described [2, 27]. Briefly, epididymal/ovarian fat pads from young (4–9 months) or aged (24 months) Fischer 344 donor rats were harvested, mechanically homogenized, and digested with a collagenase and deoxyribonuclease-1 enzymatic solution. Following centrifugation, the buoyant adipocyte layer was removed. SVF cells were washed, filtered through a 20-μm nylon screen to remove large cell or tissue aggregates, and then prepared for flow cytometry analysis, treatment of cultured mesenteric windows, intravenous (i.v.) tail vein injection into experimental animals, or modification using magnetic antibody cell sorting (MACS) prior to i.v. injection into experimental animals.

Following isolation, the fresh SVF from young and aged donors was incubated in rat Fc Block Mix (BD Biosciences 550271) at 4°C for 20 min protected from light. Cells were then stained for cell surface markers with the appropriate monoclonal antibodies at 4°C for 30 min also protected from light. Online supplementary Table 1 (see www.karger.com/doi/10.1159/000526002 for all online suppl. material) contains the antibodies included in each staining panel used in this study along with dilutions. During the incubation, anti-mouse IgG kappa compensation beads and negative control beads (BD Biosciences 552843) were added to each compensation tube; antibodies (positive compensation) or fluorophore-conjugated IgG (negative compensation) were added to the appropriate tubes. Following antibody incubation, red blood cells were lysed with diluted Lyse buffer (BD Bioscience 5558999); tubes were vortexed, incubated for 3 min at 37°C, and spun at 350 g for 5 min; and then cells washed twice with wash buffer. For the tubes requiring CD68 staining, cells were fixed and stained with MACS Inside Stain Kit (Miltenyi Biotec, 130-090-477) according to the manufacturer’s instructions. Data were collected on an LSR II cytometer (BD Biosciences) and analyzed with FlowJo 7.6 software (Tree Star, Ashland, OR, USA).

SVF was isolated as described [2, 27]. Young or aged Fischer 344 rats were anesthetized with an intramuscular injection of ketamine (80 mg/kg body weight) and xylazine (8 mg/kg body weight). Rat mesenteric tissues were harvested and cultured according to a previously established protocol [32‒34]. Briefly, an incision was made along the linea alba, and the mesentery was placed onto a sterile plastic stage using cotton tip applicators to first remove the cecum and subsequently the ileum and jejunum. The rat was euthanized with an intracardiac injection of 0.2 mL Beuthanasia. Vascularized mesenteric tissues were excised and rinsed once in sterile saline (Baxter 2B1324X) and immersed in minimum essential media (MEM; Gibco 11095-072) containing 1% penicillin-streptomycin (PenStrep; Gibco 15140122) at 37°C.

To determine the effects of age on SVF transplantation therapy, the following groups were utilized: (1) one million young SVF (from donor rats aged 4–9 months) + young mesenteric windows (from rats aged 4–9 months), (2) one million young SVF (from donor rats aged 4–9 months) + aged mesenteric windows (from rats aged 24 months), (3) one million aged SVF (from donor rats aged 24 months) + young mesenteric windows (from rats aged 4–9 months), and (4) one million aged SVF (from donor rats aged 24 months) + aged mesenteric windows (from rats aged 24 months) (n = 8 tissues per group; tissues per group were harvested from 2 rats). Freshly isolated SVF cells were suspended in 10% FBS + MEM solution at a concentration of 10 million cells/mL. Tissues were the spread onto a polycarbonate filter membrane (pore size = 5 μm) on a cell crown insert (Millipore Z681792-3EA) where 100 μL of SVF cell solution (one million cells) was transferred to the surface of each mesenteric window. Windows and cells were incubated for 20 min to allow attachment of SVF cells to the tissue. Cell crowns were then inverted into 6-well culture dishes (Cellstar 657 160) with the fat of the tissue pressing against the bottom surface. Three milliliters of culture medium composed of 10% FBS + MEM was added to each well on top of the filter. Tissues were cultured in standard incubation conditions (5% CO₂, 37°C) for 3 days.

After 3 days in culture, tissues were spread on glass slides before being rinsed 3 times for 10 min in PBS + 0.1% saponin (Sigma-Aldrich S7900). Fat around window was removed and a hydrophobic marker was applied around the tissue. The tissue was then labeled with PECAM at 1:200 (BD Pharmigen 555026) and coverslipped. Images were acquired using ×4 (dry, NA = 0.1), ×10 (dry, NA = 0.3), and ×20 (oil, NA = 0.8) objectives on an inverted microscope (Nikon Eclipse Ti2) coupled with an Andor Zyla sCMOS camera. Image analyses and quantification were performed in ImageJ 2.0.0-rc-54 (US National Institutes of Health, Bethesda, MD, USA). Images were quantified as percent vascularized area for each tissue in order to evaluate angiogenic effect of SVF transplantation on networks. Vascularized areas were measured by drawing the perimeter along the edge of the vascular networks. Avascular regions within the network smaller than 130 µm² were subtracted from the vascularized area metric. The final vascularized area was divided by the total tissue area.

Data are represented as box and whisker plots. Statistical analyses were performed in SigmaPlot 14.0 (Systat, San Jose, CA, USA) with p ≤ 0.05 as the significance level. A two-way ANOVA with pairwise Tukey’s multiple comparisons post hoc tests with interactions was used to test for statistical significance between the groups (p ≤ 0.05).

Previously isolated frozen GFP+ SVF cells were reconstituted 10 million/mL in 1 mL Lactated Ringers solution (warmed to 37°C) as previously described [1, 2]. Aged female rats (24 months) received SVF from young donors (O + YSVF, n = 4) or SVF from aged donors (O + OSVF, n = 4) via tail vein injection. Three days after injection, rats were anesthetized with inhaled 5% isoflurane with 1.5–2.0 L/min oxygen flow through an induction chamber. After confirmation of a lack of a tactile reflex, the thoracic cavity was opened and rats euthanized via cardiac removal. Mesenteric windows were isolated (6 tissues per animal) and placed individually in a 6-well dish on a crown insert with a membrane as previously described [32‒34]. Windows were then cultured in MEM media +1% PenStrep (Sigma-Aldrich PO7881) and cultured for up to 3 days at 5% CO₂ and 37°C.

After 1 or 3 days in culture, the windows were affixed to glass slides, fixed in 4% paraformaldehyde overnight, then rinsed in 1× PBS before being dried, and stored at −20°C. Windows were washed ×3 with PBS before being permeabilized in 0.5% Triton X-100/0.1 M glycine at room temperature (RT) for 30 min. After rinsing ×3 in PBS slides were blocked with 5% normal donkey serum (Jackson ImmunoResearch 017-000-121) and 1.5% bovine serum albumin in 1× PBS at RT for 1 h. Rabbit anti-rat CD3 (1:200; Abcam ab16669) and the fluorescent probe Griffonia (Bandeiraea) Simplicifolia Lectin 1 (GSL-1) rhodamine (1:100; Vector Laboratories RL-1102) were diluted in 1.5% BSA-PBS to be incubated on the tissue at 4°C overnight. After ×3 washes in PBS, the slides were incubated in secondary antibody solutions made in 1.5% BSA-PBS containing donkey anti-rabbit IgG Alexa Fluor 647 (1:400; Abcam ab150063) for 1 h at RT. Slides were then washed, incubated with DAPI (Invitrogen P36962), and coverslipped using Fluoromount-G (SouthernBiotech 0100-01).

Image stacks were acquired using a Nikon C2+ confocal microscope (Melville, NY, USA) with ×20 objective + Nyquist resolution (final magnification ∼×70), and 0.85 μm Z-step size with 10 slices/stack. GFP signal was excited with 488 laser. One mesenteric window from an old control (no cell injection) was utilized as a negative control. Visualization of colocalization of GFP+, CD3+, and DAPI (nuclei) was done using ImageJ (n = 1/group).

The methods for this surgical procedure have been adapted from the protocol described by Yang et al. [28]. Aged female Fischer 344 rats were anesthetized with inhaled 5% isoflurane with 1.5–2.0 L/min oxygen flow through an induction chamber. The animal’s abdomen was prepped by shaving and then sterilized, and the animal was placed on heating pad. Under sterile conditions, a 3-cm opening was made through abdominal skin and muscle. Gently, a section containing 8–10 mesenteric windows was removed from the body cavity and placed onto a sterile plastic stage placed on top of the abdomen. This exteriorization of the windows from the abdominal cavity leaves all blood flow to the intestines and mesentery intact, but results in a hypoxic EI. The windows were bathed in sterile physiological salt solution warmed to 37°C throughout the 20-min exteriorization procedure. The borders around the exteriorized zone were marked with 5-0 silk loops to ensure proper identification upon explant, and then the intestines and windows were placed back into the abdominal cavity and the incision closed with 5-0 silk sutures. Previously frozen SVF cells were thawed and prepared as described above and injected via tail vein, 10 million cells in 1 mL warmed Lactated Ringer’s solution. The following groups were utilized: (1) aged rats (24 months) + young (4–9 months) SVF (O + YSVF, n = 4), (2) aged rats (24 months) + aged (24 months) SVF (O + OSVF, n = 4), and (3) aged rats (24 months) with no cells (NCC, n = 4). The animals were subcutaneously injected once with a 1.2 mg/kg body weight buprenorphine slow release and monitored daily to evaluate postoperative recovery.

Three days after EI, mesenteric windows were isolated as described above both from EI regions and nonexteriorized (NEI) regions of the mesentery (6 tissues per condition per animal). Isolated windows were cultured individually in a 6-well dish on a crown insert with membrane as previously described [32‒34]. During the mesenteric window culture, at days 0, 3, and 5, MEM + 1% PenStrep media was supplemented with 3 mL of GSL-I rhodamine (1:40 dilution) in MEM. Windows were incubated at 37°C for 20 min and rinsed with MEM media, and then 3 mL of fresh MEM + 1% PenStrep was replaced. The 6-well plates were imaged on a Zeiss Z.1 Axio Observer with a sterile-heated incubation chamber; ×10 images were acquired in the same location for each mesenteric window over time. Images were processed in ImageJ by thresholding and conversion to binaries. Extraneous cells not a part of the microvascular network were excluded based upon circularity. Percent vascular area, or the area the microvessels occupy in the image, was measured via analysis of mean area. Conversion of the binary to a skeleton allowed for pruning of false branches. Analysis of the pruned skeleton images yielded total vessel length (summation of length of all vessel segments), average vessel length per segment (sum of total vessel length divided by number of segments), number of segments (segment between endpoint and branchpoint or segment between two branch points), and number of branches (number of bifurcations). Image analysis protocol is modified from Strobel et al. [35]. Data are represented as box and whisker plots.

Statistical analyses were performed in SigmaPlot 14.0 (Systat) with p ≤ 0.05 as the significance level. Multiple two-way repeated measures ANOVAs with Bonferroni pairwise comparisons allowed for comparison of group, day, and injury model on the following angiogenic metrics: percent vascular area, total vessel length, average vessel length per segment, number of segments, and number of branches.

The methods for this cell sorting have been adapted from the protocol provided by Miltenyi Biotec and as previously described in [36]. Previously frozen GFP+ SVF was thawed, washed, counted, and resuspended in chilled MACS buffer with rabbit anti-rat CD3 (1:500; Abcam ab16669) to incubate at 4°C for 30 min. The cell solution was then washed and resuspended in MACS buffer with anti-rabbit IgG microbeads (1:10; Miltenyi Biotec 130-048-602) and incubated at 4°C for 15 min. The cell solution was then washed and added to MS columns (Miltenyi Biotec 130-042-201), which were placed in the magnetic stand (Miltenyi 130-042-108). Once the flow-through CD3-depleted fraction was collected, the magnetic column was washed with MACS buffer. To collect the CD3-enriched fraction, the column was removed from the magnetic field, 500 μL MACs buffer added to the column, and then the solution gently expelled using the plunger. Once sorted, final cell counts of depleted and enriched populations were performed.

Sorted cells were reconstituted at 1.1 million cells in 1 mL Lactated Ringers solution (warmed to 37°C) as previously described [1, 2]. Aged female rats (24 months) received EI as described above and then were randomly divided into the following groups: (1) aged + CD3-depleted SVF from aged donors (O + OSVF CD3−, n = 4), (2) aged + CD3-enriched SVF from aged donors (O + OSVF CD3+, n = 4). Three days after EI, mesenteric windows were isolated as described above from EI regions and NEI regions of the mesentery (6 tissues per condition per animal). Isolated windows were cultured individually as previously described [32‒34]. During the culture on days 0, 3, and 5, the media was supplemented with GSL-I rhodamine in MEM and imaged as described above.

Statistical analyses were performed in SigmaPlot 14.0 (Systat) with p ≤ 0.05 as the significance level. Multiple two-way repeated measures ANOVAs with Bonferroni pairwise comparisons allowed for comparison of group, day, and injury model on the following angiogenic metrics: percent vascular area, total vessel length, average vessel length per segment, number of segments, and number of branches as described above.

SVF isolated from young donors included a significantly higher percentage of hematopoietic stem cells (CD34+) and MSCs (CD45−/CD34−/CD90+) (Fig. 2a). Young SVF had a significantly higher stromal fraction, which is composed of fibroblasts, pericytes, and preadipocytes (Fig. 2a). SVF from young donors contained significantly larger populations of total macrophages as well as immature and transitional macrophages (Fig. 2b). SVF from old donors had a higher percentage of lymphocytes (Fig. 2a) and M1 inflammatory macrophages (CD45+/CD11b+/CD68+/CD86+/CD163−) (Fig. 2b). Young SVF showed a marked increase in thymocytes (CD45−/CD90+) compared to SVF from aged rats (approaches significance, p = 0.057) (Fig. 2a). SVF isolated from old donors had a significantly higher percentage of CD3+ T cells as well as helper and CD4+ memory T cells compared to young SVF. Activated and CD8+ memory T cells were higher in aged SVF compared to young SVF (approaches significance, p = 0.057) (Fig. 2c).

The effect of varying combinations of host and donor ages on percentage of vascular area in isolated mesenteric windows was assessed. There was a significant increase in percentage of vascular area in young SVF on young mesenteric tissue (young SVF + young tissue) compared to all other combinations when seeded and cocultured (Fig. 3a–e). There was an equal level of reduced vascularized area in the other three combinations: aged SVF + young tissue, aged SVF + aged tissue, and young SVF + aged tissue after 3 days in culture (Fig. 3b–e).

Following i.v. injection of GFP + SVF cells from both donor groups could be detected in and around the mesenteric microvasculature 6 days after administration (Fig. 4a). Neither the preparation (freshly isolated vs. previously frozen) nor the age of donor of SVF alters the ability of these cells to incorporate into host tissues. Furthermore, some of the present SVF cells were co-positive for the T-cell marker CD3.

To examine the effect of advanced age on vascular area following injury and the angiogenic therapeutic effect of SVF, mesenteric windows subjected to EI from NCC, O + YSVF, and O + OSVF-treated animals were cultured and imaged for up to 5 days. Representative images are shown in Figure 5a. In the NEI control windows of the NCC group, the aged and untreated tissues showed a significant increase in the vascular area by day 5 compared to day 0. By day 3 in the NEI windows of aged rats treated with young SVF (O + YSVF), there was a significant increase in the vascular area compared to day 0. However, in the EI windows of the NCC, there is no change in the vascular area throughout the time points. EI windows of aged rats treated with young SVF (O + YSVF) had a significantly higher vascularized area compared to the NCC. While the O + OSVF-treated group showed an increase in the vascular area in response to injury, it was not significantly different compared to the other EI groups; however, the O + OSVF day 0 EI was significantly higher than its NEI counterpart (*) (Fig. 5b). Online supplementary Figure 2 displays other angiogenic metrics such as total vessel length (A, B) and average vessel length per segment (C, D), number of vessel segments (E, F), and number of branch points (G, H) in noninjured (NEI) and injured (EI) mesenteric microvascular networks.

The implication of modifying the T-cell populations in SVF therapy was examined by administering T cell-depleted (CD3−) or -enriched (CD3+) SVF to aged rats following injury, and mesenteric windows were cultured and imaged for up to 5 days (Fig. 6a). In the NEI condition, CD3− SVF had a significantly higher vascular area compared to day 0 at days 3 and 5 (Fig. 6b). Furthermore, the CD3− group at days 3 and 5 also had a significantly higher vascular area than the CD3+ group at the same time points (Fig. 6b). Treatment with CD3− or CD3+ SVF in the setting of injury (EI) did not produce differences between the groups at any time points. The CD3+ group in the EI condition at day 5 had a significantly higher percent vascular area compared to day 0. This is in contrast to no difference in the vascular area observed in the CD3− group for day 5 versus day 0 comparison (p = 0.126) (Fig. 6b). Online supplementary Figure 3 displays other angiogenic metrics such as total vessel length (A, B) and average vessel length per segment (C, D), number of vessel segments (E, F), and number of branch points (G, H) in noninjured (NEI) and injured (EI) mesenteric microvascular networks.

The overall goal of this study was to examine how composition in cell populations of SVF is altered with age of donor, its impact on angiogenic potential in an aged vasculature, and to determine whether T cells contribute to this regenerative effect as part of a broader effort to elucidate which cell population is responsible for the therapeutic gains seen in SVF treatment. The first major finding from this study is that aging affects the cellular composition of SVF in adipose tissue, leading to increased T-cell and decreased MSC populations in older animals. Second, SVF can increase the vascular area of the mesenteric microvasculature via in vitro coculture as well as in vivo when i.v. injected; in both conditions, SVF cells can incorporate into these networks. Lastly, there is an age-related impairment in revascularization following injury; i.v. SVF therapy from young SVF donors improves angiogenic potential over non-cell-treated controls. While SVF from aged donors showed a higher percentage of vascularized area compared to non-cell-treated controls, this difference was not significant. Contrasting our initial hypothesis, SVF increases revascularization following injury, but this therapeutic potential is not lost when T cells are removed from the fraction. In fact, there appears to be a T cell-mediated impedance on in vitro angiogenesis in the noninjured windows, as the CD3− group exhibited an increased vascular area compared to the CD3+ group.

A novel strategy that can holistically mend the pathological microvessels themselves or promote new growth, rather than merely treating the symptomatic consequence, is of significant clinical interest [9]. The SVF isolation process has become automated so that it can be done within the operating suite for immediate use or frozen and stored for later use or as a potential allogeneic use [11]. Obviously, the translational potential of frozen SVF would dramatically increase if it is equally effective as freshly isolated cells. Consistent with a prior study that showed comparable bone healing via angiogenesis using fresh or frozen SVF [37], our results showed that either fresh (Fig. 2, 2, 3) or frozen SVF (Fig. 4, 4, 5) can both increase angiogenesis in coculture or following i.v. injection, making SVF an ideal candidate in vascular therapeutics [20, 21, 38]. The heterogeneous cell population containing both immune and vascular cell components is thought to be an optimal milieu of environmental cues to develop and enhance vascular function [39‒41]. Adipose-derived SVF is composed of a small number of MSCs, endothelial cells, pericytes, and hematopoietic cells such as macrophages and T cells [11, 42]. While a few studies have examined how age of the SVF donor [43] or the culturing of SVF to confluency [44] can alter cell surface markers, these studies did not examine cell populations. We show age-related decreases in the MSC and stromal cell populations with increases in lymphocytes, M1 macrophages, and various T cells in the cellular composition of SVF (Fig. 2). Our results from young SVF donors are congruent with the current literature in total macrophage [45] and total T-cell percentages [46] with our data, showing the breakdown of the constituents (Fig. 2). The age-related changes to cell populations within SVF are important to consider for use as an autologous cell therapy or its application in an aged setting.

The ex vivo mesenteric-window model was selected for its ease of access, comprisal of an intricate microvascular network, and sustainability in culture. While the mesenteric-window model is not an established model for studying vascular diseases, it has demonstrated age-related decreased angiogenesis [47] (online suppl. Fig. 1) and impairments in vascular function [48], indicating that this model can be beneficial to studying an intact, aged microvascular network. Previously, it has been shown that SVF from young donors, when seeded and cocultured with mesenteric windows from young hosts, forms aggregates and demonstrates the beginnings of vasculogenesis within 36 h of plating [49]. Furthermore, the SVF formed vessels, indicating the presence of endothelial cells and vascular pericytes; these neovessels integrated into the mesenteric tissue in vitro [50]. The presence of young SVF on a young window stimulated angiogenesis compared to a window without SVF [50]. Here, we show that coculture with SVF from young and aged donors has a similar angiogenic potential resulting in comparable vascularized areas of aged microvascular networks (Fig. 3). This is in contrast to our previous study, which examined angiogenic potential of young versus old SVF-embedded collagen gels, which reported decreased vessel formation and host perfusion after subdermal implantation with those containing aged SVF [43]. However, it is important to note that these age-variable SVF constructs were implanted into young mice, emphasizing that the age of the environment primarily determines angiogenic capability of SVF.

With clinical applications of cell therapies typically involving systemic administration, it is important to examine the biodistribution and engraftment potential of SVF. Our previous study showed that a one-time i.v. injection of 10 million young SVF cells into an aged rat resulted in the largest percentage of SVF being retained in the aorta and carotid arteries (50% and 40%, respectively) 4 weeks later [30]. However, the microcirculation was not examined for SVF retention, and the engrafting cell type was not determined. Scherberich et al. [51] showed that SVF cells injected subcutaneously were able to physically contribute to the vasculature as endothelial cells and were further supported by SVF-derived pericytes [52], but these experiments were not conducted in an aging model. Here we show that i.v. injection of SVF cells from young and aged donors resulted in SVF within and around vessels of the microvasculature and that T cells from the donor fraction can also be found in the microvasculature (Fig. 4). Understanding what cell types are retained in the host tissues could aid in improving efficacy of administered cell therapies. Similar studies with chronic cell therapies such as MSCs administered via i.v. injection demonstrated that incorporation is as low as 10%, but this cell type was not shown to aggregate and form new vessels [53, 54]. Future studies will be needed to quantify the percentage of integration per host vessel, the varying degrees of integration for each cell type found in SVF, and what cytokines may be altered as a result of administering SVF.

In vitro studies have shown that SVF has an intrinsic angiogenic potential [49], which when added to mesenteric windows increases the host vasculature [50] (Fig. 3). While SVF therapy has been utilized in models of ischemic injury and showed functional improvements, few studies integrate ischemic models with aged models [55]. Exteriorization of mesenteric windows has been shown to be a stimulus for angiogenesis in young rats [28]. We showed that EI in aged rats produced little change in the vascular area of the mesentery, but i.v. administration of young SVF therapy markedly increased vascular area, while aged SVF administration was nonsignificant (Fig. 5). Aging has long been considered an unmodifiable risk factor for the development of vascular pathologies [3, 56, 57]. Microvascular dysfunction can lead to organ perfusion/demand mismatch and ultimately organ failure or loss of limb [58, 59]. Therefore, an autologous cell therapy that could be administered to increase angiogenesis in an area following an ischemic injury is of vast clinical significance. A better understanding of how cell therapies achieve their gains, either through paracrine release or cell engraftment/incorporation, and which cell types are crucial to their function still needs to be explored further.

T cells have been implicated as crucial cells needed for tissue and vascular repair following injury, as they activate other cells such as macrophages and fibroblasts, stimulate apoptosis and clearance of dead cells, and are responsible for switching from a proinflammatory state to a reparative state [60]. T cells have also been shown to migrate to the injured area [13], secrete factors that enhance angiogenesis [12], and target the angiogenic front following injury [14, 15]. In a knockout mouse model depleted of regulatory T cells, there was a marked reduction in angiogenesis following lung ischemia [61]. When we depleted aged SVF of the CD3+ T cells and administered during EI, we did not detect a difference in the vascular area between T cell-depleted and T cell-enriched SVF following injury (Fig. 6). Interestingly, we did see differences between the T cell-depleted and -enriched groups in the noninjured windows, as SVF depleted of T cells resulted in higher vascular area at days 3 and 5 of culture compared to the enriched group (Fig. 6). These data suggest that while T cells may not be responsible for the increase in angiogenesis following injury, a lack of T cells in SVF appears to be beneficial in a noninjury setting via increased microvascular angiogenesis. This is especially applicable to aging populations whose density of microvascular networks is decreased with age [62, 63]. Whether T cell-depleted SVF might preserve organ function in preparation for an ischemic insult is an important future direction.

We hypothesized that T cells may mediate the regenerative effect of SVF. On the contrary, our results show that vascular regeneration occurs steadily over 5 days with SVF even when depleted of CD3 cells. In fact, CD3-depleted SVF outperformed CD3-enriched SVF for vascular recovery in noninjury settings. It is possible that a multitude of cells confer vascular regenerative properties to SVF (MSCs, fibroblasts, pericytes, other immune cells, etc.), which would explain why vascular regeneration still occurred in the CD3− SVF group during EI. Interestingly, with onset of injury, the ability to achieve significant vascular regeneration is lost in the CD3− group but is preserved in the CD3+-enriched group. This could signify that CD3+ T cells are a contributing component of SVF-mediated vascular regenerative response to injury. Correspondingly, potential angiogenic contributions of other SVF cells cannot preserve significant angiogenic response after injury. Contrasting this notion, we found that aged SVF has a significantly greater proportion of T cells than young SVF, yet this increased T-cell population did not confer significant vascular regenerative ability following aged SVF donation. Aged SVF yielded less stem cells, fibroblasts, and pericytes. Therefore, future studies should also examine changes in these cell populations contained within the SVF. It is possible, perhaps even likely, that synergy between regenerative potential of multiple heterogeneous cell types is required for optimal angiogenic response to SVF. Further elucidating this complex symphony of cells within SVF represents an important future direction. This could be achieved by isolating other cell populations in SVF to assess their contribution to angiogenic potential by measuring vascular changes over time in conjunction with cell tracking. Taken together, this study presents novel insight into age-related changes to cell populations isolated from the adipose-derived SVF, visualization of immune cells in SVF residing in and around the host microcirculation after i.v. injection, effects of SVF therapy from young and aged donors on microvasculature after ischemic injury, and evidence that the angiogenic effects of SVF cell therapy are not wholly T cell-mediated.

The aging population is at greater risk of ischemic injury with less capacity for vascular repair following injury. Therefore, the elderly stand to benefit greatly from cell therapies as a remedy for vascular dysfunction. The field of cellular therapeutics is in its relative infancy with current emphasis on optimization. In the context of this study, SVF represents a heterogeneous cell population shown to increase angiogenic regeneration in our novel aged mesenteric injury model. This study provides others with a new tool for tracking vascular remodeling and can be used in conjunction with study of cell therapies or pharmacologics in a setting of advanced age. Furthermore, we show how age of the donor should be considered not only for cellular differences but functionality as a vascular therapeutic. Age-related changes to cell dynamics and function in providing therapeutic gains – that is, secretion of anti-inflammatory cytokines, increasing sensitivity to VEGF, increasing migration and engraftment potential of injected cells, and endothelial cell division – are further avenues to be explored in cellular therapeutics that could be done using the aged mesenteric model.

The authors wish to thank the University of Missouri Rat Resource & Research Center (P40OD011062) for the initial GFP+ Fischer 344 breeding pairs. The authors also wish to thank Daniel Benson, Rajeev Nair, and Dr. Natia Kelm for their support and assistance in conducting the experiments detailed in this study.

All animal surgeries were performed in accordance with protocols approved by the University of Florida and University of Louisville Institutional Animal Care and Use Committee (IACUC-approved protocol #201710060 and #19635) and the NIH Guide for the Care and Use of Laboratory Animals.

The authors have no disclosures to declare. The authors have no conflicts of interest to declare.

This research was funded by the National Institutes of Health (P30ES030283 and R01AG053585 to AJL, and R01AG049821 to WLM), the Gheen’s Foundation (AJL), and funding from the University of Florida’s Division of Research Program Development.

Dr. Amanda J. LeBlanc, Gabrielle Rowe, Jason E. Beare, and Dr. Walter L. Murfee conceived of and designed the research. Gabrielle Rowe, David S. Heng, Jason E. Beare, and Nicholas A. Hodges conducted the experiments. Gabrielle Rowe, Nicholas A. Hodges, and Jason E. Beare analyzed the data. Gabrielle Rowe, David S. Heng, Evan P. Tracy, and Dr. Amanda J. LeBlanc wrote the manuscript. Gabrielle Rowe, David S. Heng, Jason E. Beare, Evan P. Tracy, Dr. Walter L. Murfee, and Dr. Amanda J. LeBlanc edited the manuscript. All authors read and approved the manuscript.

All data generated or analyzed during this study are included in this article and its online supplementary material. Further inquiries can be directed to the corresponding author.