Microvascular Research 97 (2015) 159–166

Contents lists available at ScienceDirect

Microvascular Research journal homepage: www.elsevier.com/locate/ymvre

Comparison between endothelial progenitor cells and human umbilical vein endothelial cells on neovascularization in an adipogenesis mouse model Valentin Haug 1, Nestor Torio-Padron 1, G. Bjoern Stark, Guenter Finkenzeller, Sandra Strassburg ⁎ Department of Plastic and Hand Surgery, University of Freiburg Medical Center, Hugstetter Str. 55, 79106 Freiburg, Germany

a r t i c l e

i n f o

Article history: Accepted 17 October 2014 Available online 27 October 2014 Keywords: Endothelial progenitor cells Adipose-derived stem cells Human umbilical vein endothelial cells Adipose tissue engineering Vascularization Angiogenesis

a b s t r a c t Volume stability and growth of tissue engineered adipose tissue equivalents using adipose-derived stem cells (ASCs) rely strongly on angiogenesis and neovascularization to support the maintenance of cells. An attractive cellular approach is based on coimplantation of endothelial cells to create a vascular network. Endothelial progenitor cells (EPCs) are a promising cell population, since they can be easily isolated from autologous human peripheral blood and thus represent a clinically feasible option. We have previously shown in in vitro and semi-in vivo studies that ASCs exert beneficial effects on EPCs in terms of enhanced tube formation and formation of blood vessels, respectively. In this study, we investigated the in vivo effects of coimplantation on endothelial cell-mediated neovascularization and ASC-mediated adipose tissue formation. For this purpose, human ASCs and human EPCs (or HUVECs as direct comparison to EPCs) were suspended alone or in coculture in fibrin and subcutaneously injected into the back of athymic nude mice and explanted after 1, 3 or 6 months. Our results show that monocultures of EPCs or HUVECs were not able to perform vasculogenesis and constructs exhibited complete resorption already after 1 month. However, a remarkable difference between EPCs and HUVECs was detected when coimplanted with ASCs. While coimplanted HUVECs gave rise to a stable neovasculature which was characterized by perfusion with erythrocytes, coimplanted EPCs showed no ability to form vascular structures. In the case of HUVEC-derived neovasculature, coimplanted ASCs displayed perivascular properties by stabilizing these neovessels. However, formation of human adipose tissue was independent of coimplanted endothelial cells. Our results indicate that HUVECs are superior to EPCs in terms of promoting in vivo neovascularization and recruiting perivascular cells for vessel stabilization when coimplanted with ASCs. © 2014 Elsevier Inc. All rights reserved.

Introduction In reconstructive and plastic surgery, there is a tremendous demand for biologically functional adipose tissue aiming at restoring contour defects after soft tissue removal. Standard approaches for the reconstruction of soft tissues include alloplastic implants and autologous tissue flaps which have the disadvantages of foreign body reaction and donor site morbidity, respectively. Autologous free fat grafting represents a minimally invasive alternative to this but results in unpredictable graft resorption, especially in the long term (Kaufman et al., 2007), most likely due to insufficient vascularization. Tissue engineering of adipose tissue equivalents by means of a suitable cell population in combination with an adequate scaffold would provide a strategy to perform de novo adipogenesis and angiogenesis (Tanzi and Fare, 2009). Adipose-derived stem cells (ASCs) are considered to be such a suitable cell population. ASCs can be easily harvested ⁎ Corresponding author. E-mail address: [email protected] (S. Strassburg). 1 Both authors contributed equally.

http://dx.doi.org/10.1016/j.mvr.2014.10.005 0026-2862/© 2014 Elsevier Inc. All rights reserved.

from human fat tissue and have the potential to differentiate into lineages of adipogenic, osteogenic, chondrogenic and myogenic cells (Zuk et al., 2001). Several in vivo studies reported the formation of human adipose tissue after the implantation of human ASCs in different scaffolds with or without the addition of growth factors (Kimura et al., 2003; Patrick et al., 1999; Torio-Padron et al., 2007; von Heimburg et al., 2001) which were also characterized by ingrowth of host vasculature. ASCs are known to secrete several proangiogenic growth factors (Kilroy et al., 2007; Rehman et al., 2004) and therefore the use of ASCs may improve angiogenesis, because adipogenesis is regulated by factors that also drive angiogenesis (Christiaens and Lijnen, 2010). However, it is still one of the main challenges in adipose tissue engineering to provide a sufficient and functional vasculature. As oxygen diffusion is limited to a distance of about 200 μm (Malda et al., 2007), long-term survival of tissue engineered adipose tissue relies on rapid development of blood vessels throughout the tissue graft. Although ASCs stimulate angiogenesis by migration of endogenous endothelial cells, for adipose constructs of larger size this vascularization is not rapid enough and an earlier vascular supply from the center throughout the whole graft with

160

V. Haug et al. / Microvascular Research 97 (2015) 159–166

connections to host vasculature is a crucial requirement for the longterm survival of the graft. Therefore, formation of a capillary network by co-implantation with endothelial cells within a construct might be beneficial for tissue engineering applications. Recently, we and others conducted studies in which human ASCs were cocultured with human endothelial cells to mimic the environment found in vivo. In vitro, we found a stimulatory effect of ASCs on endothelial tube formation (Borges et al., 2007; Strassburg et al., 2013b) as well as the formation of functional blood vessels in short-term semi-in vivo studies on the chorioallantoic membrane (Strassburg et al., 2013a). Some in vivo studies coimplanted ASCs with human umbilical vein endothelial cells (HUVECs) in different scaffolds (Frerich et al., 2012; Koike et al., 2004; Verseijden et al., 2010) or with human endothelial progenitor cells (EPCs) in Matrigel (Melero-Martin et al., 2007, 2008). Even if these studies showed promising results in regard to adipose tissue formation and formation of stable long lasting blood vessels, neither HUVECs nor Matrigel are suitable for clinical applications. Therefore, a coimplantation of ASCs with an easy accessible autologous cell population such as endothelial progenitor cells from human peripheral blood in a clinically applicable biomaterial would be of desire. In the present study, we investigated the potential of human ASCs and EPCs (and HUVECs as positive control) for the formation of vascularized human adipose tissue. To test this, ASCs and EPCs/HUVECs were suspended alone or in combination in fibrin and subcutaneously injected into the back of athymic nude mice. After 1, 3 and 6 months, neovascularization and adipogenesis were evaluated by histology over time. Our results show that only coimplanted HUVECs gave rise to a functional human neovasculature, which was associated with humanderived perivascular cells and that (co)implanted ASCs effectively support adipogenesis. Material and methods Cell culture All cell culture was maintained at 37 °C with 5% CO2 and 20% oxygen in a humidified environment with medium change of 2–3 times a week. Human adipose-derived stem cells (ASCs; 2 females, 1 male, mean age: 34 years) were isolated from human subcutaneous fat tissue by collagenase-II digestion and erythrocytes lysis (17 mM tris(hydroxymethyl)aminomethane, 16 mM ammonium chloride) as described in detail before (Torio-Padron et al., 2007). ASCs were cultured in EGM-2 Bullet Kit (Lonza) supplemented with additional 10% fetal calf serum (Biochrom AG) and 1% penicillin/streptomycin (PAA) (complete EGM-2) for a maximum of 3 passages. Adipogenic differentiation potential of ASCs was routinely determined using medium supplementation (DMEM/Hams F-12 (1:1) with 3% FCS and 100 nM insulin, 1 μM dexamethasone and 0.25 mM IBMX) and Oil Red O staining following standard protocols. Human endothelial progenitor cells (EPCs) were isolated from buffy coats of healthy donors (obtained from the Department of Transfusion Medicine of the University Hospital Freiburg) after Biocoll (Biochrom AG) gradient centrifugation and seeding onto collagen-I (BD Biosciences) coated tissue culture plastic with complete EGM-2. Confluent EPC colonies were seeded onto fibronectin (Sigma) coated tissue culture plastic and used at passage 6 or less. The endothelial origin of isolated human EPCs was determined by morphology, incorporation of Dil-labeled acetylated low density lipoprotein and immunostaining for CD14, CD45, CD31, CD144 and Von Willebrand factor. Human umbilical vein endothelial cells (HUVECs) were purchased from Promocell, cultivated in complete EGM-2 and used for experiments at passage 6 or less. Prior to subcutaneous injection into mice, EPCs and HUVECs were aggregated to spheroids composed of 100 cells/spheroid in hanging drops of 25 μl in complete EGM-2 containing 20% methocel solution (6 g methyl cellulose (Sigma) in 500 ml medium supplemented with

10% FCS) as previously described (Korff and Augustin, 1998). After 24 h, the spheroids were harvested for in vivo experiments. Fibrin matrix The final concentration of the fibrin matrix was composed of 10 mg/ml fibrinogen and 0.5 U thrombin/mg fibrinogen (Baxter). The fibrinogen component was prepared with Aprotinin to 70– 110 mg/ml and further diluted to 20 mg with 0.9% NaCl. The thrombin component was prepared with 40 mM CaCl2 to 500 U/ml and further diluted to 25 U/ml with thrombin dilution buffer (40 mM CaCl2, 171 mM NaCl, 40 mM glycine, 50 g HAS/l, pH 7.4). For 1 ml fibrin matrix, ASCs as single cells and/or endothelial cell spheroids were suspended in 300 μl 40 mM CaCl2 and mixed with 200 μl 25 U/ml thrombin. After addition of 500 μl 20 mg/ml fibrinogen, polymerization of the construct started. Animal experiment Male athymic nude mice Balb c nu/nu, 6–8 weeks old were obtained from Charles River Laboratories. All animal experiments and animal care were performed after ethical approval by the regional council of Freiburg. All experimental procedures were performed under sterile conditions in a laminar flow. Subcutaneous injections of cell/fibrin constructs were carried out under general anesthesia (100 mg/kg Ketamin and 5 mg/kg Rompun). For simultaneous administration of both fibrin components (thrombin component with cells and fibrinogen component), a double syringe system was used. Each animal got two constructs (one left and one right on the back). Fibrin constructs were either loaded without cells (control) or with 30 mio ASCs as a single cell suspension or with 2000 spheroids composed of 100 endothelial cells (EPCs or HUVECs, respectively) or a co-injection of both cell types. ASCs as well as EPCs were pooled from three different donors to overcome possible limitations of donor variability. The animals were categorized into following groups with explantation times after 1, 3 or 6 months: Group 1: 1 ml fibrin (control) Group 2: ASCs (30 mio as a single cell suspension) Group 3: EPC spheroids (2000 spheroids composed of 100 EPCs/ spheroid) Group 4: ASC–EPC coimplantation (30 mio ASCs + 2000 spheroids composed of 100 EPCs/spheroid) Group 5: HUVEC spheroids (2000 spheroids composed of 100 HUVECs/spheroid) Group 6: ASC–HUVEC coimplantation (30 mio ASCs + 2000 spheroids composed of 100 HUVECs/spheroid). 4 animals were used for each group and for each time point. As one animal got two constructs, each group per time point consisted of 8 constructs. The area of the fibrin construct was tattooed to facilitate explantation in case of graft resorption. Explantation At 1, 3 or 6 months after injection, the mice were sacrificed with CO2. The constructs were explanted and subsequently cut into halves. One half of the construct was fixed in 4% formalin and embedded in paraffin. The other half of the construct was embedded in tissue freezing medium and snap frozen in liquid nitrogen for cryosections. Serial sections were cut at 5 μm for histological and immunohistological staining. Histology For Oil Red O staining, cryosections were stained in filtered 0.5% Oil Red O in isopropanol for 10 min and weakly counterstained with hematoxylin before mounting in glycerine. For hematoxylin and eosin (H&E)

V. Haug et al. / Microvascular Research 97 (2015) 159–166

staining, deparaffinized and rehydrated paraffin sections were routinely stained with H&E for gross morphology and host blood vessel ingrowth.

161

A total of 72 mice were operated without complications. The animals were healthy during the experimental period and no sign of inflammation or any infection was noted at the implant sites.

Immunostaining Identification of cell–fibrin constructs The human origin of cells within the constructs was verified by a mouse anti-human vimentin antibody for ASCs and a mouse antihuman CD31 antibody for endothelial cells. Sections were deparaffinized and rehydrated and antigen retrieval with citrate buffer pH 6 (Dako) was performed for 15 min, cooling down to room temperature for 15 min, followed by incubation in 3% H2O2 for 20 min to inhibit endogenous peroxidase and blocking of endogenous biotin for 1 h in mouse IgG blocking reagent (Vector Laboratories). The sections were incubated for 1 h at room temperature with the primary anti-human vimentin antibody (1:75; Dako) or the primary anti-human CD31 antibody (1:50; Dako) in a humid chamber. A ready to use HRP-labeled goat anti-mouse secondary antibody (Dako), was applied for 30 min and developed with Histogreen (Linaris) or 3,3′-diaminobenzidine (DAB; Vector Laboratories). Cells were counterstained with hematoxylin and the sections were permanently mounted. Immunofluorescence staining was performed with anti-human CD34. Deparaffinized and rehydrated sections were blocked in 10% goat serum (Dako) for 1 h followed by incubation with the primary anti-human CD34 antibody (1:50; Novocastra) for 2 h at room temperature. Subsequently, the sections were incubated for 45 min with the secondary goat anti-mouse antibody/Alexa Fluor 488 (1:200; Invitrogen). For immunofluorescence double staining, the sections were additionally incubated for 1 h with anti-smooth muscle alpha actin (α-SMA)/Cy3 (1:100; Sigma). Negative controls were treated according to the same protocol, omitting the primary human-specific antibody. Sections of human fat tissue were used as positive controls. Images of histological staining were taken with an Axioplan 2 (Zeiss) microscope.

The tattoo clearly marked the implantation and thus explantation site. The specimens were harvested after 1, 3 and 6 months postimplantation. Each biopsy was cut into 5 μm sections from regions of the entire construct (Figs. 1A–C). Microscopically, we could only identify the cell–fibrin constructs in groups containing ASCs at all time points. After 1 month, almost all constructs of groups containing ASCs (ASC monoculture [n = 7], ASC–EPC coculture [n = 8] and ASC–HUVEC coculture [n = 7]) could be clearly identified underneath the panniculus carnosus both in H&E staining and in human-specific anti-vimentin staining (Fig. 2A). Statistical analysis of construct size revealed no difference between groups containing ASCs. Constructs containing fibrin only or EPC or HUVEC monoculture showed complete resorption and could not be detected (Fig. 2B). Analysis of angiogenesis and neovascularization In order to assess whether (co)implanted endothelial cells were able to form functional blood vessels, tissue sections were analyzed by H&E staining and by human-specific anti-CD31 staining. Numerous blood vessels could be identified within the implant by H&E staining after 1 month in all groups containing ASCs (Fig. 3A). However, humanspecific anti-CD31 staining demonstrated that blood vessels in groups

Histomorphometric analysis The size of cell–fibrin constructs or the size of newly formed adipose tissue was calculated from sections stained with human-specific antivimentin. To obtain a representative overview, the explanted construct was cut entirely and 5 μm sections from the beginning, middle and end of the construct were analyzed. Histomorphometric analysis was carried out using Image J software. For calculation of construct size, the area of anti-vimentin positive cells was analyzed and for calculation of size of newly formed adipose tissue, the cells with adipocytic appearance were analyzed. For quantification of murine blood vessels within de novo human adipose tissue after 6 months, sections stained with anti-vimentin (to localize the human adipose tissue) and H&E (to visualize ingrown murine blood vessels) from the beginning, middle and end of the construct were analyzed. To determine the microvascular density, the number of ingrown blood vessels within the entire anti-vimentin positive cell– fibrin construct was counted and normalized to mm2. Statistical analysis Statistically significant differences between groups were determined by applying unpaired Student's t-test and significance was set at p b 0.05. Values are represented as mean +/− standard deviation or sem. Results Characteristics and thus suitability of utilized ASCs and EPCs are extensively described in our former publications (Strassburg et al., 2013a, 2013b).

Fig. 1. Identification of injected human cells in fibrin into athymic nude mice. A) Macroscopic appearance of subcutaneous cell–fibrin construct shortly before explantation. B) Explanted construct. C) Schematic of construct for histological analysis.

162

V. Haug et al. / Microvascular Research 97 (2015) 159–166

Fig. 2. Volume of explanted constructs containing ASCs after 1 month. A) Exemplary overview pictures of H&E staining (top row) and human-specific anti-vimentin staining (green, bottom row) showing the construct underneath the panniculus carnosus. B) Graph showing volume of explants. ASC: adipose-derived stem cells, EPC: endothelial progenitor cells, HUVEC: human umbilical vein endothelial cells, and Vim: vimentin. Results are represented as mean values +/− - stdev. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. Angiogenesis and neovascularization of constructs containing human cells and injected into athymic nude mice. A) Exemplary pictures of H&E staining after 1 month showing blood vessels (black arrows) within the fibrin construct containing human cells, as proven by human specific anti-vimentin staining (green, inlay). B) Exemplary pictures of H&E and human specific anti-CD31 staining (brown) showing the formation of human blood vessels in the ASC–HUVEC group only after 1, 3 and 6 months. ASC: adipose-derived stem cells, EPC: endothelial progenitor cells, and HUVEC: human umbilical vein endothelial cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

V. Haug et al. / Microvascular Research 97 (2015) 159–166

consisting of ASC monoculture and ASC–EPC coculture were comprised of ingrown mouse blood vessels only, since human-specific anti-CD31 staining was negative for these groups (data not shown). Histological analysis of groups consisting of EPC or HUVEC monocultures showed also negative CD31 staining (data not shown), demonstrating no formation of human blood vessels and complete resorption of the cell–fibrin construct. The formation of human blood vessels which were perfused with erythrocytes was only observed in explants consisting of ASC– HUVEC coculture after 1, 3 and 6 months (Fig. 3B). Although no vessel formation by EPCs was observed, at least we observed the existence of single EPCs after 6 months in constructs coimplanted with ASCs by anti-human CD34 staining (Fig. 4A). ASCs adopt a vessel stabilizing function In order to investigate whether HUVEC-derived blood vessels were stabilized by perivascular cells, we performed smooth muscle actin (α-SMA) staining. First, anti-vimentin staining was performed to localize the construct. Secondly, α-SMA/CD34 double staining was performed (Fig. 4A). In the ASC–HUVEC coculture group, numerous anti-CD34 positive blood vessels were observed within the construct, which were associated with α-SMA positive cells. However, since α-SMA staining was not human-specific, we performed humanspecific CD31/vimentin double staining for this coculture group showing that the newly formed human blood vessels were colocalized with human ASCs (Fig. 4B). Indeed we calculated from representative pictures that after 1 month 81% +/− 19%, after 3 months 100% and after 6 months 94% +/− 6% of all human vessels were associated with human ASCs. Analysis of de novo adipogenesis Despite the detection of ASCs in explants after 1 month (Fig. 2) histological analysis by human-specific anti-vimentin staining revealed the formation of human adipose tissue at the earliest at 3 months. The differentiation of ASCs to adipocytes was shown by cells with an adipocytic appearance (Fig. 5A). Positive Oil Red O staining of vimentin-positive cells confirmed the accumulation of lipids within the vacuoles (Fig. 5B). Histomorphometric analysis of representative pictures of anti-vimentin staining summarizes the mean size of de novo human adipose tissue after 3 and 6 months. After 3 months, the size of adipose

163

tissue was significantly increased in the ASC–HUVEC coculture group compared to the ASC monoculture or ASC–EPC coculture group, an effect which was not observed anymore after 6 months. Surprisingly, the size of human adipose tissue in the ASC–HUVEC coculture group was decreasing over time as opposed to the ASC monoculture and ASC–EPC coculture group (Fig. 5C). Additionally, numerous blood vessels could be detected within de novo human adipose tissue after 6 months (Fig. 5D). At least in the ASC group and the ASC–EPC group, these blood vessels were probably mouse-derived since we were not able to find human-derived blood vessels in explants from these groups. Quantification of blood vessels revealed a microvascular density of 52.6 +/− 12.9 vessels per mm2 for the ASC group, 12.1 +/− 3.8 vessels per mm2 for the ASC–EPC group and 152.8 +/− 80.4 vessels per mm2 for the ASC-HUVEC group (Fig. 5E).

Discussion Several in vivo approaches are described in the literature to induce neovascularization of tissue engineered adipose tissue equivalents, for example by the application of angiogenic growth factors (Kawaguchi et al., 1998; Masuda et al., 2004; Tabata et al., 2000), scaffold design (Stosich et al., 2007), integration of a vascular pedicle (Dolderer et al., 2007; Stillaert et al., 2007; Walton et al., 2004) or cell based strategies by the addition of endothelial cells (Borges et al., 2003). In the present study, we have conducted a coimplantation of ASCs and EPCs/HUVECs into athymic nude mice in order to investigate whether ASCs have positive effects on endothelial cells in terms of enhanced formation of a vasculature and to investigate the effects of enhanced vascularization on adipogenesis. Former in vitro coculture studies elucidated a paracrine communication between ASCs and endothelial cells with regard to an improved vascular network formation by endothelial cells. This cellular communication depended on angiogenic growth factors, such as VEGF (Strassburg et al., 2013b), HGF and PDGF (Merfeld-Clauss et al., 2010). Encouraged by our own previous studies demonstrating a beneficial effect of ASCs on endothelial cell tube formation in vitro as well as formation of functional blood vessels in semi-in vivo experiments (Merfeld-Clauss et al., 2010; Strassburg et al., 2013a, 2013b), long term in vivo animal studies were subsequently conducted in athymic nude mice.

Fig. 4. ASCs support stability of newly formed HUVEC-derived blood vessels. A) Exemplary pictures of human specific anti-vimentin staining (green) and human specific anti-CD34 (green)/anti-α-SMA (red)/DAPI (blue) triple staining. Human specific anti-vimentin staining shows the construct containing human cells and corresponding immunofluorescence shows the existence of EPCs or the formation of HUVECs into blood vessels. B) Exemplary pictures of human specific anti-vimentin (green) and anti-CD31 (brown) staining of a fibrin construct containing ASCs and HUVECs. Vimentin positive ASCs are colocalized with newly formed HUVEC-derived blood vessels. ASC: adipose-derived stem cells, EPC: endothelial progenitor cells, and HUVEC: human umbilical vein endothelial cells. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

164

V. Haug et al. / Microvascular Research 97 (2015) 159–166

Fig. 5. Adipogenic differentiation of injected human cells and neovascularization of de novo adipose tissue. A) After 3 and 6 months, human-specific anti-vimentin positive cells (green) with adipogenic appearance were shown in groups containing ASCs. B) Corresponsive Oil Red O (ORO) staining of human specific anti-vimentin positive cells after 6 months showing lipid storage in vacuoles. C) Graph showing histomorphometric analysis of anti-vimentin pictures. Area of anti-vimentin positive cells after 3 and 6 months was determined for groups containing ASCs using ImageJ software. Results are represented as mean values +/− stdev. D) Vascularization of de novo adipose tissue after 6 months. Exemplary pictures of human specific anti-vimentin staining (top row) and corresponding H&E staining (bottom row) with arrows highlighting blood vessels. E) Graph showing the number of blood vessels per mm2 of de novo adipose tissue after 6 months. Results are represented as mean values +/− sem. ASC: adipose-derived stem cells, EPC: endothelial progenitor cells, and HUVEC: human umbilical vein endothelial cells, *p ≤ 0.05. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

In this actual study, however, it became clear by human-specific anti-CD31 staining that only HUVECs form blood vessels and this phenomenon was only observed when HUVECs were coimplanted with ASCs. Although some other studies reported the formation of a vascular network in vivo after implantation of endothelial cells alone (Koike et al., 2004, Finkenzeller et al., 2009, Traktuev et al., 2009, Alajati et al., 2008), the reader should bear in mind that these studies were performed either by additional administration of angiogenic growth factors (such as VEGF/bFGF), the usage of a different matrix or these neovessels showed minimal perfusion with low stability. Furthermore, when a coimplantation with supporting mural cells or mesenchymal cells was performed, this experimental setting was always superior in terms of quantity of neovessels compared to monocultures. We also identified that human EPCs from peripheral blood do not possess the same angiogenic potential as HUVECs. This finding is in accordance with Finkenzeller et al., who also found that HUVECs were superior to EPCs in terms of promoting in vivo vascularization of engineered tissues (Finkenzeller et al., 2009). The study by Finkenzeller et al. was performed in monoculture and ours in coculture with ASCs, since studies postulated a potential positive effect of mesenchymal stem cells on the vasculogenic potential of EPCs by paracrine means (Melero-Martin et al., 2007, 2008; Traktuev et al., 2009). However, we could not confirm any positive effects of ASCs on EPCs, although ASCs are recognized to actively secrete cytokines and growth factors thereby creating a microenvironment for endothelial cells beneficial for angiogenesis and vasculogenesis (Verseijden et al., 2010). Our result on the impaired ability of implanted EPCs to form functional vessels is in agreement with Au et al., where peripheral blood-derived EPCs were coimplanted with 10T1/2 cells and tested for their in vivo vasculogenic potential. Coimplantation of both cell populations did not significantly

increase vessel densities compared to EPC implantation alone (Au et al., 2008). In general, EPCs formed only unstable blood vessels that disappeared almost completely after 1 month. This raises the question of whether the lower ability of EPCs to form vascular structures is the result of impaired vessel assembly or of impaired neovessel stability. When the formation of human blood vessels was observed, neovasculature was associated with α-SMA positive cells. Double staining for human CD34/α-SMA and vimentin/CD31, respectively, revealed that ASCs represent these α-SMA positive cells. Therefore, we can conclude that ASCs contribute to vessel stabilization. Other studies have also shown that ASCs were able to stabilize vascular networks (Rasmussen et al., 2011; Verseijden et al., 2010). As mentioned above, we can only speculate that in the case of neovasculature, ASCs not only perform adipogenesis but also differentiate to α-SMA positive cells to stabilize these neovessels, which is a key requirement for the formation of a long-lasting vasculature. The origin of ASCs has not been clearly identified. Some research groups have suspected that ASCs are a subpopulation of progenitor cells located in the stromal fraction (Guilak et al., 2006), and others have hypothesized that ASCs are of perivascular origin (Amos et al., 2008; Cai et al., 2011; Zimmerlin et al., 2010), which may explain their ability to differentiate into perivascular cells. In addition, Goerke at al., showed that in vitro cocultivation of endothelial cells with bone marrow-derived mesenchymal stem cells (MSCs) leads to differentiation of MSCs to perivascular cells as indicated by an increase in α-SMA expression (Goerke et al., 2012), an option which is also reasonable in our experimental setting. In the context of adipogenesis we observed that implanted ASCs were able to develop mature adipose tissue. This finding was not unexpected, since ASCs have been reported to promote adipogenesis, when subcutaneously injected into mice (Cho et al., 2007; Mizuno et al.,

V. Haug et al. / Microvascular Research 97 (2015) 159–166

2008; Torio-Padron et al., 2007). However, after 6 months, implanted ASCs perform adipogenesis with no improved performance when coimplanted with endothelial cells. One possible reason could be that coimplanted EPCs did not participate in human vessel formation and thus it is not surprising, that we could not observe the formation of larger human adipose tissues. For the ASC–HUVEC coculture group, after 3 months, coculture with HUVECs seems to have a positive effect on human adipose tissue size, possibly due to formation of a human neovasculature (Fukumura et al., 2003; Nishimura et al., 2007). However, after 6 months, we observed a decrease in human adipose tissue size. Our results imply that in the ASC–HUVEC coculture group not all implanted ASCs differentiate into adipocytes but also into pericytes. This could explain the reduced size of adipose tissue after 6 months in this group. In fact when we observed less formation of human adipose tissue, we simultaneously observed increased formation of a human vasculature. This is possibly due to the phenomenon that ASCs rather associate with human neovessels than undergo adipogenic differentiation. Since former studies have examined that endothelial cells can control proliferation and differentiation of adipose cells (Hutley et al., 2001; Varzaneh et al., 1994), another possible phenomenon could be that HUVEC-derived blood vessels repress adipogenic differentiation of ASCs. However, the molecular mechanisms underlining mature adipose tissue formation during in vivo coculture with different types of endothelial cells have yet to be determined. In our experiments we have used a fibrin matrix to suspend the cells for subcutaneous injection. It is well accepted in the literature that a suitable biomaterial which encourages adipose tissue formation and angiogenesis is crucial for adipose tissue engineering. In this context, fibrin – among many other naturally-derived and injectable biomaterials – seems to be a good candidate matrix (Cho et al., 2007; Gentleman et al., 2006; Hemmrich et al., 2008; Kawaguchi et al., 1998; Torio-Padron et al., 2007). However, it is important to note that degradation of fibrin is not simultaneous to adipose tissue formation, since resorption of fibrin was observed already after 1 month but adipose tissue formation at the earliest after 3 months. In summary, our results show that only coimplanted HUVECs formed a complex network of perfused neovessels and thus show a superior vasculogenic potential compared to human peripheral blood-derived EPCs with no beneficial effects on adipogenesis. In addition, ASCs serve as both, building blocks by differentiating into mature adipocytes and into perivascular cells and cellular factories that secrete mediators to modulate the local environment and thus stimulate angiogenesis of the de novo adipose tissue and vasculogenesis of coimplanted HUVECs. Disclosure statement No competing financial or personal interests exist. Acknowledgment The authors thank Birgit Scherer for excellent technical assistance. This work was supported by funding through the Deutsche Forschungsgemeinschaft (TO 614/2-1). References Alajati, A., Laib, A.M., Weber, H., Boos, A.M., Bartol, A., Ikenberg, K., Korff, T., Zentgraf, H., Obodozie, C., Graeser, R., et al., 2008. Spheroid-based engineering of a human vasculature in mice. Nat. Methods 5, 439–445. Amos, P.J., Shang, H., Bailey, A.M., Taylor, A., Katz, A.J., Peirce, S.M., 2008. IFATS collection: the role of human adipose-derived stromal cells in inflammatory microvascular remodeling and evidence of a perivascular phenotype. Stem Cells 26, 2682–2690. Au, P., Daheron, L.M., Duda, D.G., Cohen, K.S., Tyrrell, J.A., Lanning, R.M., Fukumura, D., Scadden, D.T., Jain, R.K., 2008. Differential in vivo potential of endothelial progenitor cells from human umbilical cord blood and adult peripheral blood to form functional long-lasting vessels. Blood 111, 1302–1305. Borges, J., Mueller, M.C., Padron, N.T., Tegtmeier, F., Lang, E.M., Stark, G.B., 2003. Engineered adipose tissue supplied by functional microvessels. Tissue Eng. 9, 1263–1270.

165

Borges, J., Muller, M.C., Momeni, A., Stark, G.B., Torio-Padron, N., 2007. In vitro analysis of the interactions between preadipocytes and endothelial cells in a 3D fibrin matrix. Minim. Invasive Ther. Allied Technol. 16, 141–148. Cai, X., Lin, Y., Hauschka, P.V., Grottkau, B.E., 2011. Adipose stem cells originate from perivascular cells. Biol. Cell. 103, 435–447. Cho, S.W., Song, K.W., Rhie, J.W., Park, M.H., Choi, C.Y., Kim, B.S., 2007. Engineered adipose tissue formation enhanced by basic fibroblast growth factor and a mechanically stable environment. Cell Transplant. 16, 421–434. Christiaens, V., Lijnen, H.R., 2010. Angiogenesis and development of adipose tissue. Mol. Cell. Endocrinol. 318, 2–9. Dolderer, J.H., Abberton, K.M., Thompson, E.W., Slavin, J.L., Stevens, G.W., Penington, A.J., Morrison, W.A., 2007. Spontaneous large volume adipose tissue generation from a vascularized pedicled fat flap inside a chamber space. Tissue Eng. 13, 673–681. Finkenzeller, G., Graner, S., Kirkpatrick, C.J., Fuchs, S., Stark, G.B., 2009. Impaired in vivo vasculogenic potential of endothelial progenitor cells in comparison to human umbilical vein endothelial cells in a spheroid-based implantation model. Cell Prolif. 42, 498–505. Frerich, B., Winter, K., Scheller, K., Braumann, U.D., 2012. Comparison of different fabrication techniques for human adipose tissue engineering in severe combined immunodeficient mice. Artif. Organs 36, 227–237. Fukumura, D., Ushiyama, A., Duda, D.G., Xu, L., Tam, J., Krishna, V., Chatterjee, K., Garkavtsev, I., Jain, R.K., 2003. Paracrine regulation of angiogenesis and adipocyte differentiation during in vivo adipogenesis. Circ. Res. 93, e88–e97. Gentleman, E., Nauman, E.A., Livesay, G.A., Dee, K.C., 2006. Collagen composite biomaterials resist contraction while allowing development of adipocytic soft tissue in vitro. Tissue Eng. 12, 1639–1649. Goerke, S.M., Plaha, J., Hager, S., Strassburg, S., Torio-Padron, N., Stark, G.B., Finkenzeller, G., 2012. Human endothelial progenitor cells induce extracellular signal-regulated kinase-dependent differentiation of mesenchymal stem cells into smooth muscle cells upon cocultivation. Tissue Eng. Part A 18, 2395–2405. Guilak, F., Lott, K.E., Awad, H.A., Cao, Q., Hicok, K.C., Fermor, B., Gimble, J.M., 2006. Clonal analysis of the differentiation potential of human adipose-derived adult stem cells. J. Cell. Physiol. 206, 229–237. Hemmrich, K., Van de Sijpe, K., Rhodes, N.P., Hunt, J.A., Di Bartolo, C., Pallua, N., Blondeel, P., von Heimburg, D., 2008. Autologous in vivo adipose tissue engineering in hyaluronan-based gels—a pilot study. J. Surg. Res. 144, 82–88. Hutley, L.J., Herington, A.C., Shurety, W., Cheung, C., Vesey, D.A., Cameron, D.P., Prins, J.B., 2001. Human adipose tissue endothelial cells promote preadipocyte proliferation. Am. J. Physiol. Endocrinol. Metab. 281, E1037–E1044. Kaufman, M.R., Bradley, J.P., Dickinson, B., Heller, J.B., Wasson, K., O'Hara, C., Huang, C., Gabbay, J., Ghadjar, K., Miller, T.A., 2007. Autologous fat transfer national consensus survey: trends in techniques for harvest, preparation, and application, and perception of short- and long-term results. Plast. Reconstr. Surg. 119, 323–331. Kawaguchi, N., Toriyama, K., Nicodemou-Lena, E., Inou, K., Torii, S., Kitagawa, Y., 1998. De novo adipogenesis in mice at the site of injection of basement membrane and basic fibroblast growth factor. Proc. Natl. Acad. Sci. U. S. A. 95, 1062–1066. Kilroy, G.E., Foster, S.J., Wu, X., Ruiz, J., Sherwood, S., Heifetz, A., Ludlow, J.W., Stricker, D.M., Potiny, S., Green, P., et al., 2007. Cytokine profile of human adipose-derived stem cells: expression of angiogenic, hematopoietic, and pro-inflammatory factors. J. Cell. Physiol. 212, 702–709. Kimura, Y., Ozeki, M., Inamoto, T., Tabata, Y., 2003. Adipose tissue engineering based on human preadipocytes combined with gelatin microspheres containing basic fibroblast growth factor. Biomaterials 24, 2513–2521. Koike, N., Fukumura, D., Gralla, O., Au, P., Schechner, J.S., Jain, R.K., 2004. Tissue engineering: creation of long-lasting blood vessels. Nature 428, 138–139. Korff, T., Augustin, H.G., 1998. Integration of endothelial cells in multicellular spheroids prevents apoptosis and induces differentiation. J. Cell Biol. 143, 1341–1352. Malda, J., Klein, T.J., Upton, Z., 2007. The roles of hypoxia in the in vitro engineering of tissues. Tissue Eng. 13, 2153–2162. Masuda, T., Furue, M., Matsuda, T., 2004. Photocured, styrenated gelatin-based microspheres for de novo adipogenesis through corelease of basic fibroblast growth factor, insulin, and insulin-like growth factor I. Tissue Eng. 10, 523–535. Melero-Martin, J.M., Khan, Z.A., Picard, A., Wu, X., Paruchuri, S., Bischoff, J., 2007. In vivo vasculogenic potential of human blood-derived endothelial progenitor cells. Blood 109, 4761–4768. Melero-Martin, J.M., De Obaldia, M.E., Kang, S.Y., Khan, Z.A., Yuan, L., Oettgen, P., Bischoff, J., 2008. Engineering robust and functional vascular networks in vivo with human adult and cord blood-derived progenitor cells. Circ. Res. 103, 194–202. Merfeld-Clauss, S., Gollahalli, N., March, K.L., Traktuev, D.O., 2010. Adipose tissue progenitor cells directly interact with endothelial cells to induce vascular network formation. Tissue Eng. Part A 16, 2953–2966. Mizuno, H., Itoi, Y., Kawahara, S., Ogawa, R., Akaishi, S., Hyakusoku, H., 2008. In vivo adipose tissue regeneration by adipose-derived stromal cells isolated from GFP transgenic mice. Cells Tissues Organs 187, 177–185. Nishimura, S., Manabe, I., Nagasaki, M., Hosoya, Y., Yamashita, H., Fujita, H., Ohsugi, M., Tobe, K., Kadowaki, T., Nagai, R., et al., 2007. Adipogenesis in obesity requires close interplay between differentiating adipocytes, stromal cells, and blood vessels. Diabetes 56, 1517–1526. Patrick Jr., C.W., Chauvin, P.B., Hobley, J., Reece, G.P., 1999. Preadipocyte seeded PLGA scaffolds for adipose tissue engineering. Tissue Eng. 5, 139–151. Rasmussen, J.G., Frobert, O., Pilgaard, L., Kastrup, J., Simonsen, U., Zachar, V., Fink, T., 2011. Prolonged hypoxic culture and trypsinization increase the pro-angiogenic potential of human adipose tissue-derived stem cells. Cytotherapy 13, 318–328. Rehman, J., Traktuev, D., Li, J., Merfeld-Clauss, S., Temm-Grove, C.J., Bovenkerk, J.E., Pell, C.L., Johnstone, B.H., Considine, R.V., March, K.L., 2004. Secretion of angiogenic and antiapoptotic factors by human adipose stromal cells. Circulation 109, 1292–1298.

166

V. Haug et al. / Microvascular Research 97 (2015) 159–166

Stillaert, F., Findlay, M., Palmer, J., Idrizi, R., Cheang, S., Messina, A., Abberton, K., Morrison, W., Thompson, E.W., 2007. Host rather than graft origin of Matrigel-induced adipose tissue in the murine tissue-engineering chamber. Tissue Eng. 13, 2291–2300. Stosich, M.S., Bastian, B., Marion, N.W., Clark, P.A., Reilly, G., Mao, J.J., 2007. Vascularized adipose tissue grafts from human mesenchymal stem cells with bioactive cues and microchannel conduits. Tissue Eng. 13, 2881–2890. Strassburg, S., Nienhueser, H., Stark, G.B., Finkenzeller, G., Torio-Padron, N., 2013a. Human adipose-derived stem cells enhance the angiogenic potential of endothelial progenitor cells, but not of human umbilical vein endothelial cells. Tissue Eng. Part A 19, 166–174. Strassburg, S., Nienhueser, H., Bjorn Stark, G., Finkenzeller, G., Torio-Padron, N., 2013b. Coculture of adipose-derived stem cells and endothelial cells in fibrin induces angiogenesis and vasculogenesis in a chorioallantoic membrane model. J. Tissue Eng. Regen. Med. http://dx.doi.org/10.1002/term.1769 (Epub ahead of print). Tabata, Y., Miyao, M., Inamoto, T., Ishii, T., Hirano, Y., Yamaoki, Y., Ikada, Y., 2000. De novo formation of adipose tissue by controlled release of basic fibroblast growth factor. Tissue Eng. 6, 279–289. Tanzi, M.C., Fare, S., 2009. Adipose tissue engineering: state of the art, recent advances and innovative approaches. Expert Rev. Med. Devices 6, 533–551. Torio-Padron, N., Baerlecken, N., Momeni, A., Stark, G.B., Borges, J., 2007. Engineering of adipose tissue by injection of human preadipocytes in fibrin. Aesthet. Plast. Surg. 31, 285–293.

Traktuev, D.O., Prater, D.N., Merfeld-Clauss, S., Sanjeevaiah, A.R., Saadatzadeh, M.R., Murphy, M., Johnstone, B.H., Ingram, D.A., March, K.L., 2009. Robust functional vascular network formation in vivo by cooperation of adipose progenitor and endothelial cells. Circ. Res. 104, 1410–1420. Varzaneh, F.E., Shillabeer, G., Wong, K.L., Lau, D.C., 1994. Extracellular matrix components secreted by microvascular endothelial cells stimulate preadipocyte differentiation in vitro. Metabolism 43, 906–912. Verseijden, F., Posthumus-van Sluijs, S.J., Pavljasevic, P., Hofer, S.O., van Osch, G.J., Farrell, E., 2010. Adult human bone marrow- and adipose tissue-derived stromal cells support the formation of prevascular-like structures from endothelial cells in vitro. Tissue Eng. Part A 16, 101–114. von Heimburg, D., Zachariah, S., Heschel, I., Kuhling, H., Schoof, H., Hafemann, B., Pallua, N., 2001. Human preadipocytes seeded on freeze-dried collagen scaffolds investigated in vitro and in vivo. Biomaterials 22, 429–438. Walton, R.L., Beahm, E.K., Wu, L., 2004. De novo adipose formation in a vascularized engineered construct. Microsurgery 24, 378–384. Zimmerlin, L., Donnenberg, V.S., Pfeifer, M.E., Meyer, E.M., Peault, B., Rubin, J.P., Donnenberg, A.D., 2010. Stromal vascular progenitors in adult human adipose tissue. Cytometry A 77, 22–30. Zuk, P.A., Zhu, M., Mizuno, H., Huang, J., Futrell, J.W., Katz, A.J., Benhaim, P., Lorenz, H.P., Hedrick, M.H., 2001. Multilineage cells from human adipose tissue: implications for cell-based therapies. Tissue Eng. 7, 211–228.