Ciclopirox olamine promotes the angiogenic response of endothelial cells and mesenchymal stem cells

Abstract

BACKGROUND:

Prolyl hydroxylase inhibitors (PHIs) are promising compounds to promote angiogenesis by stabilizing hypoxia-inducible factor-1α (HIF-1α), a master regulator of angiogenesis. Increased HIF-1α presence induces expression of proangiogenic genes such as vascular endothelial growth factor (VEGF).

OBJECTIVE:

We investigated the pharmacological induction of hypoxia via the PHI ciclopirox olamine (CPX) as angiogenesis strategy on human dermal microvascular endothelial cell (hd-mvEC) spheroids directly and indirectly via activating human mesenchymal stem cells (hMSCs).

METHODS:

HMSCs were isolated from bone marrow and hd-mvECs from foreskin biopsies. MSC-conditioned medium after CPX stimulation (MSC-CM CPX) was analyzed by VEGF ELISA and Proteome Profiler™ Human Angiogenesis Array. Direct stimulation with CPX and indirect stimulation via MSC-CM CPX were compared in sprouting assays of hd-mvEC spheroids.

RESULTS:

Direct stimulation with CPX significantly increased sprouting of hd-mvEC spheroids. MSC-CM CPX also induced sprouting from hd-mvEC spheroids, which was mediated by angiogenic VEGF and other proangiogenic factors that had been produced by stimulated hMSCs.

CONCLUSIONS:

The stimulation with CPX increased the proangiogenic response of hd-mvECs and hMSCs. The direct stimulation of hd-mvECs with CPX has the potential to replace external VEGF supplementation. Thus, CPX can induce angiogenesis in ECs even in the absence of auxiliary cells demonstrating a promising proangiogenic approach.

Keywords

prolyl hydroxylase inhibitor ciclopirox olamine mesenchymal stem cells endothelial cells sprouting angiogenesis vascularization

1 Introduction

Angiogenesis is the formation of capillaries from pre-existing blood vessels; it occurs with expansion and the remodeling of the vascular network [1, 2]. The underlying principle is the proliferation of endothelial cells (ECs) of existing blood vessels and involves two mechanisms: the endothelial sprouting and the intussusceptive microvascular growth. Sprouting comprises EC migration, proliferation and tube formation [1, 3]. Typically, ECs are quiescent and divide less than once in 100 to 300 days [4]. However, they are activated in the presence of proangiogenic signals such as vascular endothelial growth factor (VEGF), angiopoietin, platelet-derived growth factor (PDGF) and fibroblast growth factor (FGF) [4, 5]. Sprouting is initiated when a single EC becomes a tip cell and starts a new sprout in the direction of the angiogenic growth factor gradient [6]. To achieve this, the tip cell has to break through the basement membrane, whereas neighboring cells simultaneously act as stalk cells [6]. These sprouts do not contain a lumen. After tip cell selection and migration [7], sprouts from neighboring vessels fuse and build up a functional vessel linkage, an anastomosis [6]. To provide vessel stabilization and to regulate vessel perfusion in parallel, mural cells such as pericytes in capillaries and vascular smooth muscle cells in larger blood vessels are recruited to the immature vasculature. Acting as mural cells, mesenchymal stem cells (MSCs) are also able to support the formation of microvascular networks by ECs [8]. MSCs secrete a variety of cytokines and growth factors which have paracrine and autocrine activities [9]. Through paracrine actions via extracellular microvesicles or secreted factors, MSCs are known to modulate angiogenesis [10], apoptosis [11], differentiation [12] and immune responses [13, 14] of surrounding cells. It has been shown that interactions between MSCs and endothelial progenitor cells induce differentiation towards a pericyte-like phenotype [15]. However, not all MSCs are able to act as pericyte [16, 17]. Nevertheless, such interactions can support the maturation of newly formed luminal structures [18]. Hence, co-cultures of ECs and MSCs are of interest for the development of vascularization strategies.

The hypoxia-inducible factor-1α (HIF-1α) is a transcription factor and master switch of angiogenesis. Under normoxic conditions, HIF-1α is almost immediately degraded after synthesis by hydroxylation of one or two proline residues of HIF-1α by the prolyl-hydroxylase domain containing enzymes (PHDs) (Fig. 1). The degradation of HIFs is inhibited by a hypoxic environment or pharmacologically by a prolyl hydroxylase inhibitor (PHI) (Fig. 1). Proline hydroxylases are iron-containing dioxygenases. Thus, they can be inhibited by substrate competition at the active site of the enzyme or iron chelation. Inhibition leaves two prolines of HIF-1α unhydroxylated. Thus, HIF-1α is not marked for degradation by the proteasome, and so intranuclear levels of HIF-1α increase, leading to an upregulation of various genes in tissues facing hypoxia. One set of genes is involved in angiogenesis such as VEGF, PDGF and FGF [4, 19].

Fig.1

Role of prolyl hydroxylase inhibitors (PHIs) such as ciclopirox olamine (CPX) in the oxygen-dependent regulation of HIF-1α pathway. Under normoxic condition, HIF-1α is almost immediately degraded after synthesis. Hydroxylation of one or two proline residues of HIF-1α by the prolyl-hydroxylase domain containing enzymes (PHDs) initiates the degradation. Then, hydroxyprolines enable the binding of von Hippel Lindau tumor suppressor protein (pVHL) to HIF-1α. Thus, HIF-1α is tagged by ubiquitination and marked for degradation via the proteasome. Under hypoxic condition, which may also be pharmacologically induced by PHI such as CPX, the prolyl hydroxylase is inhibited and fails to oxidize the two prolines of HIF-1α. As it is not marked for degradation and escapes from ubiquitination, levels of HIF-1α rapidly increase. Consequently, accumulated HIF-1α translocates to the nucleus, dimerizes with metabolically stable HIF-1β and binds to hypoxia-response elements (HREs) within the promoters of target genes. Thereby, gene expression of angiogenic factors such as VEGF, PDGF and FGF is induced.

The broad effects at gene expression level make PHIs such as ciclopirox olamine (CPX) attractive compounds in promoting local angiogenesis by HIF-1α stabilization [20]. Under normoxic conditions CPX, a lipophilic bidentate iron chelator, stabilizes HIF-1α [20, 21]. Thereby, CPX prevents the degradation of HIF-1α imitating an artificial hypoxia state, which results in an increased gene expression and an upregulation of angiogenic factors such as VEGF, PDGF and FGF which have been shown to have synergistic effects on sprouting [22]. Lim et al., 2016 showed that CPX in combination with the lysophospholipid sphingosine-1-phosphate promoted diabetic wound closure by an increase of functional vessels, accelerated granulation tissue formation and an increase in collagen fibers [21]. These findings correlate with those of Ko et al., 2011 who reported significantly accelerated wound closure, increased angiogenesis, and increased dermal cellularity by CPX [23].

So far, there are no studies directly investigating the influence of CPX on primary ECs and their sprouting capability. Moreover, while indirect stimulation with CPX via fibroblasts has shown promising results regarding the angiogenic response of human umbilical vein endothelial cells (HUVECs) [24], indirect stimulation via MSCs has not been studied up to date. Here, we analyzed CPX as stimulation agent by in vitro sprouting assays of human dermal microvascular endothelial cell (hd-mvEC) spheroids.

In vitro sprouting assays are based on spheroids of ECs providing a unique 3D approach to study angiogenesis. From literature we know that the surface ECs of spheroids become quiescent and respond to angiogenic factors such as VEGF [25]. These spheroids are embedded in collagen gels to further enable sprout formation by proangiogenic stimuli [26, 27]. In addition to sprouting assays, there are several other in vitro assays to analyze the efficacy of pro- and antiangiogenic stimulation agents such as in vitro cell migration assays towards pro- or antiangiogenic factors, tube formation assays or organ assays [28]. The tube formation assay is a straightforward method to analyze the tube formation ability in Matrigel. However, other non-endothelial cells such as fibroblasts have been shown to respond to Matrigel by forming tubes [29, 30]. In contrast, the rat aortic ring assay as organ assay emulates more closely the in vivo situation by monitoring the outgrowth of ECs by the addition of stimulation agents often in a matrix such as fibrin [30, 31]. A disadvantage of organ assays is the use of non-human tissue [30].

For this study, we chose the sprouting assay of mvECs as it is an in vitro physiologically relevant method to investigate the early process of microvascular angiogenesis. ECs of microvascular origin have been chosen to imitate the microvascular environment of sprouting by hd-mvEC spheroids. HUVECs are often used for sprouting assays. However, HUVECs are derived from the endothelium of the umbilical cord not demonstrating an autologous cell type compared to mvECs isolated out of skin. In this study, we assessed the angiogenic response to CPX in hd-mvECs and hMSCs by quantifiying sprouting out of hd-mvEC spheroids. The indirect stimulation via hMSCs was studied by the usage of MSC-conditioned medium (MSC-CM).

2 Methods

2.1 Cell culture and characterization

Human primary cells were isolated from tissue biopsies (bone marrow and foreskin). Bone marrow biopsies were kindly provided by the Orthopedic Clinic Koenig-Ludwig-Haus Wuerzburg. All donor(s) and legal representative(s) provided full informed consent in writing. The study was approved by the local ethical board of the University of Wuerzburg (local ethics approval: 182/10, 25.11.2015).

2.1.1 Human mesenchymal stem cell pool

HMSCs were isolated from human bone marrow as previously described [32]. The hMSC pool was obtained by combining hMSCs obtained from one female and two male donors, whose age was between 35 to 76 years. HMSCs were used between passage 1 and 5 and cultured in DMEM high glucose (DMEM HG, Gibco) supplemented with 5 ng/mL hbFGF (R&D Systems), 50μg/mL ascorbic acid 2-phosphate (Wako), 10% fetal calf serum (FCS, Bio & Sell GmbH) and 1% penicillin/streptomycin (PAA). Pooled hMSCs were characterized by flow cytometry using CD markers as stipulated by the International Society for Stem Cell Therapy (ISCT), namely CD11b, CD19, CD31, CD34, CD44, CD45, CD73, CD90, CD105. Moreover, the trilineage differentiation (adipogenic, chondrogenic and osteogenic) potential of hMSCs was confirmed by culture in respective differentiation medium (DM) as described in the supplemental information.

2.1.2 Human dermal microvascular endothelial cell pool

Hd-mvECs were isolated from human foreskin as previously described [33]. The hd-mvEC pool was obtained by combining cells derived from five male donors, whose age was between 3 to 6 years. Hd-mvECs were used between passage 1 and 3 and cultured in VascuLife^® (CellSystems) supplemented with 1% penicillin/streptomycin. Characterization of the hd-mvEC pool was performed via immunocytochemistry using vWF (DAKO), VE-Cadherin (Sigma-Aldrich) and CD31 (DAKO).

2.2 Flow cytometry

Pooled hMSCs were characterized by flow cytometry using antibodies (Thermo Fisher Scientific) for CD11b, CD19, CD31, CD34, CD44, CD45, CD73, CD90, CD105 as listed in the supplemental information. Flow cytometry staining was performed under light protection and all centrifugation steps occurred at 400 g for 6 minutes. 5×10⁵ cells per tube were washed once with PBS and then with PBS supplemented with 1% FCS. The cells were resuspended in 50μL antibody mixture and incubated for 30 minutes at 4°C. Then, 1 ml PBS supplemented with 1% FCS was added followed by centrifugation and resuspension in 200μL PBS supplemented with 1% FCS. Fluorescent measurement was performed with BD LSR-2 (BD Biosciences) and unstained cells were used for the flow cytometry setting. Analysis was done with the FlowJo software.

2.3 Immunocytochemistry/Immunohistochemistry

HMSC pellets (chondrogenic differentiation) were fixed with 4% paraformaldehyde (PFA, AppliChem), dehydrated, embedded in paraffin (Leica EG1150 Modular Tissue Embedding Center) and 4μm thick sections were cut with a sliding microtome (Leica SM 2010R). Stainings of hMSC pellets with Alcian blue and immunohistochemical staining for Col II (0.5 mg/mL with 2 mg/mL hyaluronidase (Sigma-Aldrich) pretreatment for 12 minutes at room temperature, AF5710, Acris) were performed. For the immunohistochemical staining, the sections were pretreated as indicated and incubated with the primary antibody overnight. Then, the Super-Vision 2 HRP-Polymer Kit (DCS Innovative Diagnostik-Systeme) was used for the immunohistochemical detection as described in the manufacturer’s instructions.

Hd-mvECs seeded on glass slips were fixed with ice-cold 1 : 1 EtOH/acetone (Carl Roth) and immunocytochemically stained for vWF (0.12 mg/mL, M0616, DAKO) and VE-Cadherin (0.01 mg/mL, V1517, Sigma-Aldrich). Moreover, they were immunofluorescently stained for CD31 (0.41 mg/mL, M0823, DAKO) overnight followed by 60 minutes incubation with the secondary antibody (donkey anti-mouse Alexa Fluor 647, A-31571, Invitrogen).

2.4 Production and characterization of MSC-conditioned medium (MSC-CM)

MSC-conditioned medium after CPX stimulation (MSC-CM CPX) was based on low FCS MSC medium, DMEM HG supplemented with 2% FCS. After reaching 80% confluence, the MSC medium was changed to the low FCS MSC medium supplemented with 10μM CPX (Sigma-Aldrich). After 24 hours incubation at 37°C and 5% CO₂, the medium was collected, filtrated (0.2μm) and stored at –80°C until analysis and further use for sprouting assays. As control, MSC-CM without CPX was collected after 24 hours. The CPX stimulated and unstimulated hMSCs were harvested for gene expression analysis of HIF-1α and VEGF. MSC-CM was analyzed for hVEGF by ELISA (R&D Systems) and for the expression profiles of angiogenesis related proteins by the Proteome Profiler™ Human Angiogenesis Array (R&D Systems). These two methods were performed according to the manufacturer’s instructions. Determined hVEGF amounts and expression profiles of angiogenesis related proteins were normalized by the total protein content. The total protein concentration of each sample was measured by the DC Protein Assay Kit (Bio-Rad) performed according to the manufacturer’s instructions.

2.5 Gene expression analysis

Cell pellets of 5×10⁵ unstimulated and CPX stimulated hMSCs were snap-frozen in liquid nitrogen and stored at –80°C until RNA isolation by the RNeasy Micro Kit (Qiagen) as described in the manufacturer’s instructions. Next, cDNA was synthesized out of RNA samples with the iScript™ cDNA Synthesis Kit (Bio-Rad) according to the manufacturer’s protocol. For quantitative real-time polymerase chain reaction (qRT-PCR), cDNA samples were analyzed each containing 15 ng cDNA in a volume of 2μl. The PCR mastermix was composed of 10μL SsoFast™EvaGreen^® Supermix (Bio-Rad), 2μL forward primer, 2μL reverse primer and 4μL RNase free water. In each well of the qRT-PCR 96 well plate, 18μL PCR mastermix and 2μL cDNA were added. The qRT-PCR was performed with the qPCR cycler (Bio-Rad) with the following program: 1×30 seconds at 95°C, 40× steps of 5 seconds at 95°C + 5 seconds at 60°C, 1×10 seconds at 95°C, 1×5 seconds at 65–95°C by increasing temperature in 0.5 increments. Subsequently, data acquisition and analysis by the Pfaffl method were performed with the CFX Manager software (Bio-Rad) to determine the gene expression levels. The gene expression of hHIF-1α and hVEGF were normalized to three different reference genes: hGAPDH (human glyceraldehyde-3-phosphate dehydrogenase), hEF1α (human elongation factor 1α) and hHRP3 (human hypoxanthine-guanine phosphoribosyltransferase). All used primers (Eurofins) are listed in the supplemental information.

2.6 Sprouting assays

The influence of CPX by direct stimulation and indirect stimulation via MSC-CM CPX was studied in sprouting assays of hd-mvEC spheroids. Hd-mvEC spheroids were prepared by mixing of 3×10⁵ hd-mvECs with 12 mL methylcellulose solution (Sigma-Aldrich) and 48 mL VascuLife^®. Subsequently, 100μL of this cell suspension (5,000 cells per mL) were pipetted per well of a non-adherent round bottom 96 well plates. By incubation for 18 to 24 hours at 37°C and 5% CO₂, one spheroid composed of approximately 500 hd-mvECs formed per well. These hd-mvEC spheroids were harvested and embedded in a Col I gel (50 spheroids per mL). The collagen solution was prepared on ice by mixing 4 mL 0.4% Collagen R solution (Serva) with 2 mL 0.1% acetic acid (Carl Roth), 600μL medium 199 (Sigma-Aldrich) and an appropriate amount of 0.2 N NaOH (Carl Roth) to neutralize the Col I mixture. After mixing with the harvested spheroids, 1 mL of the spheroid-containing Col I gel solution were pipetted per well of a pre-warmed 24 well plate. After polymerization for 30 minutes at 37°C and 5% CO₂, the stimulation of the spheroids was performed with a volume of 200μl per well with the following experimental groups: (1) VascuLife^® basal medium without (w/o) stimulation agent, (2) VascuLife^® basal medium with final concentration of 25 ng/ml hVEGF (R&D Systems) as positive control, (3) VascuLife^® basal medium with final concentration of 10μM CPX; (4) MSC-CM and (5) MSC-CM CPX. After 24 hours at 37°C and 5% CO₂, the sprouting of spheroids was studied by taking pictures of ten randomly selected spheroids per condition (EVOS XL microscopy, Life Technologies). Digital quantification of the number of sprouts (NOS) and the cumulative sprouting length (CSL) was performed with ImageJ software (National Institutes of Health, NIH).

2.7 Statistical analysis

Quantitative data is presented as mean +/– standard deviation. Statistical data analysis was performed by nonparametric Kruskal-Wallis test using IBM SPSS software (Version 23.0, IBM corporation) to detect significant differences between the experimental groups of the sprouting assays. Differences were considered as statistically significant for p-values <0.05 (n = 4). Follow-up analysis by homogenous subsets using IBM SPSS software was performed to confirm significant differences.

3 Results

3.1 Characterization of the hMSC and the hd-mvEC pool

HMSCs were characterized according to the International Society for Stem Cell Therapy (ISCT) criteria [34]. First, the ability to adhere on plastic was confirmed by brightfield microscopy showing the characteristic large and flattened spindle-shaped morphology of primary hMSCs (Fig. 2A). Second, a strong expression (>96%) of positive MSC surface markers CD44, CD73, CD90 and CD105, respectively, was confirmed (Supplement Figure 1A). CD11b, CD19, CD31, CD34 and CD45 as negative surface markers were not detectable (Supplement Figure 1B). As last criterion, trilineage differentiation potential of hMSC differentiation was ascertained (Fig. 2A), demonstrating adipogenesis after 14 days, osteogenesis after 28 days, and chondrogenic differentiation after 21 days. The endothelial cell phenotype of the established hd-mvEC pool was confirmed by immunocytochemistry. Hd-mvECs were positively stained for VWF and VE-cadherin (Fig. 2B) and positive CD31 staining of the hd-mvECs was confirmed by immunofluorescence (Fig. 2B).

Fig.2

Characterization of the cell pools. A The hMSC pool was characterized by brightfield microscopy (cell attachment) and trilineage differentiation potential (adipogenic on day 14: Oil Red O, osteogenic on day 28: Alizarin Red S, chondrogenic on day 21: Alcian blue and Col II immunohistochemical staining with isotype control in the upper right corner. B The hd-mvEC pool was characterized by endothelial surface marker (vWF, VE-Cadherin, CD31) stained by immunocytochemistry (isotype controls in the upper right corner) and immunofluorescence, respectively. All scale bars, 200μm.

3.2 Proangiogenic factor production by hMSCs under CPX stimulation

HMSCs were stimulated with CPX for 24 hours. Gene expression analysis of CPX stimulated hMSCs revealed a decrease of HIF-1α expression on gene level, whereas VEGF expression was upregulated 8.4 fold compared with unstimulated hMSCs as control (Fig. 3A).

Fig.3

Effects of CPX stimulation on hMSCs determined by gene expression analysis of hMSCs and VEGF protein quantification of MSC-conditioned medium (MSC-CM). A Gene expression of HIF-1α and VEGF of hMSCs under CPX stimulation in comparison with unstimulated hMSCs (control). B VEGF quantification of MSC-CM CPX (CPX) in comparison with MSC-CM (control) by ELISA.

The medium of these cells was harvested as conditioned medium (MSC-CM) for analysis and application in the sprouting assay. MSC-conditioned medium of 10μM CPX stimulation (MSC-CM CPX) was analyzed in comparison with the control (MSC-CM). MSC-CM CPX contained 1.5 fold more VEGF as the control (Fig. 3B). Angiogenic factors were further analyzed by the Proteome Profiler™ Human Angiogenesis Array (Fig. 4) and a number of proangiogenic factors were substantially increased in conditioned medium under CPX (MSC-CM CPX). Angiopoietin-2, Endothelin-1, IL-8, Vasohibin and VEGF-C were just above the detection threshold under CPX but not detectable in the control. MSC-CM CPX contained 2.5 fold more Angiogenin and 12.5 fold more FGF-7 compared to the control (Fig. 4 left). Decreased protein levels of Angiopoietin-1, IGFBP-1, IGFBP-2, IGFBP-3, MCP-1, Pentraxin, Serpin E1, Serpin F1, TIMP-1, Thrombospondin-1 and uPA were detected for the MSC-CM CPX compared with the control (Fig. 4 right). MMP-8, MMP-9, PD-ECGF, PDFG-AA, Persephin, TIMP-4 were only detected in the control but not in MSC-CM CPX (Fig. 4 right).

Fig.4

Proangiogenic factor production by hMSCs under CPX stimulation. Analysis of angiogenic factors by the Proteome Profiler™ Human Angiogenesis Array of MSC-CM CPX (CPX) in comparison with MSC-CM (control). Left: Increased expression (+) of factors for CPX, right: decreased expression (–) of factors for CPX in comparison with the control.

3.3 CPX stimulation induced the sprouting of hd-mvEC spheroids

CPX was used as a stimulation agent in sprouting assays of hd-mvEC spheroids by direct application and via MSC-CM. After 24 hours, spheroids stimulated with 25 ng/ml VEGF, 10μM CPX and MSC-CM CPX showed pronounced sprouting by brightfield microscopy, whereas the untreated control (w/o) showed no sprouting and the MSC-CM control comparable less sprouting (Fig. 5A). The number of sprouts (NOS) was significantly increased by stimulation with 25 ng/mL VEGF and 10μM CPX compared with the untreated control (p = 0.004) (Fig. 5B). The NOS of 10μM CPX stimulation was 2 fold higher than the untreated control, whereas stimulation with 25 ng/mL VEGF led to an 1.6 fold increase. MSC-CM CPX showed a slight NOS increase compared with its control MSC-CM and the untreated control. The NOS of 10μM CPX and 25 ng/ml VEGF was higher than the NOS of conditioned medium (MSC-CM and MSC-CM CPX). The cumulative sprouting length (CSL) describes the sum of sprouting lengths of all measured sprouts of ten randomly selected spheroids per condition. The CSL of direct stimulation of hd-mvEC spheroids with 25 ng/mL VEGF and 10μM CPX was significantly increased in contrast to the untreated control (p = 0.001) (Fig. 5C). Direct stimulation with 10μM CPX even showed a significant higher CSL increase which was 1.5 fold compared to the positive control of 25 ng/mL VEGF. Indirect stimulation with CPX via MSC-CM CPX led to a significant increase of CSL which was almost 1.5 fold more as its control MSC-CM. Additionally, indirect stimulation with CPX (MSC-CM CPX) led to a significant CSL increase compared to the untreated control. In contrast, the CSL of direct stimulation of the spheroids with 25 ng/ml VEGF or 10μM CPX was higher compared to the indirect stimulation via CPX stimulated hMSCs (MSC-CM CPX).

Fig.5

CPX induced sprouting of hd-mvEC spheroids. A Morphology of spheroids after 24 hours of stimulation visualized by brightfield microscopy (scale bar as depicted, same magnification of all images), B number of spheroids (NOS) and C cumulative sprouting length (CSL) after 24 hours of stimulation (direct stimulation: 25 ng/mL VEGF and 10μM CPX versus indirect stimulation MSC-CM (control) and MSC-CM CPX (CPX) are depicted: ^*p < 0.05, n = 4, error bars represent standard deviation (Kruskal-Wallis test).

4 Discussion

PHIs are promising compounds to promote angiogenesis by stabilizing HIF-1α [20]. Increased HIF-1α presence induces expression of proangiogenic genes such as VEGF [4, 20]. In this study, the FDA-approved CPX was chosen as PHI. Compared with other PHIs, CPX has a higher lipophilicity, allowing better membrane penetration, and a higher affinity for iron, allowing lower concentrations for HIF-1α stabilization [20]. Originally, CPX was developed as an antifungal agent for the treatment of mycoses of the skin and nails [35, 36]. The proangiogenic effect of CPX by HIF-1α stability and VEGF expression was discovered by serendipity when a mouse skin wound model was treated with commercially available CPX containing dermal cream [20].

Here, we investigated the pharmacological induction of hypoxia via CPX as angiogenesis strategy on hd-mvEC spheroids directly and indirectly via activating hMSCs. Frequently, HUVEC spheroids are used for angiogenesis sprouting assays [26, 27]. HUVECs grow well in defined culture media and can be sourced easily [37]. Though, it was shown in angiogenesis sprouting assays using spheroids that HUVECs are more sensitive to proangiogenic stimuli than ECs derived from adult blood vessels [38]. This might lead to an overestimation of the potential of angiogenesis strategies tested with HUVECs. We therefore chose to work with freshly isolated hd-mvECs, as ECs from different tissues have been shown to respond differently [30]. Earlier work had shown that HUVEC monocultures do not appear to respond substantially to PHI [39].

In this study, we show that in contrast to HUVECs [39], hd-mvECs display a direct and strong sprouting response in 3D multi-cellular spheroids by CPX stimulation in the absence of VEGF or other proangiogenic agents in the basal medium. It became even evident that 10μM CPX in the basal medium showed an even stronger effect than 25 ng/mL VEGF. This means that CPX stimulation can replace external VEGF supplementation because sufficient levels of VEGF and other proangiogenic factors are induced by pharmacological hypoxia.

It was hitherto assumed that PHIs such as CPX work best via a mesenchymal partner to indirectly stimulate the proangiogenic response of ECs (bankshot approach); Lim et al., 2013 dissected the sprouting response of HUVECs to CPX in a microfluidic system in the presence of fibroblasts [24]. Sprouts only occurred in the presence of fibroblasts, which were the main source of VEGF in this system [24]. Earlier it had been shown, that in mixed cultures of fibroblasts and HUVECs, only fibroblasts expressed HIF-1α much more strongly in the presence of a PHI [39]. Therefore, it was assumed that in a mixed cell population, for example in a wound context, the mesenchymal cells would be the true targets of PHIs to effect angiogenesis in ECs [20, 23]. The results obtained in this study open up the possibility that ECs can be direct targets for PHIs, as well, and that this depends on the origin of the EC and their ability to get stimulated via autocrine action or inducing neighbors via paracrine mechanism.

In accordance with Lim et al., 2013 [24], exposure of hMSCs with CPX led also to an upregulation of proangiogenic factors in this study. Gene expression analysis revealed a drop of HIF-1α mRNA, while VEGF mRNA increased after 24 hours. As HIF-1α transcription normally has a short half-life, stabilization via CPX might lead to a negative feedback and thus to the downregulation of HIF-1α mRNA expression. On the protein level, VEGF production was increased by 1.5 fold. In addition to VEGF, other proangiogenic factors were found increased in CPX-treated MSC culture medium (MSC-CM CPX). HIF-1α directly facilitates transcription of VEGF, Angiopoietin-2 and stem cell factor (SCF) genes [40]. Accordingly, increased levels of angiogenic factors Angiogenin [41], chemokine IL-8 [42], Endothelin-1, Angiopoietin-2 (important for neovascularization [43]) and FGF (important for endothelial capillary initiation [43]) were found in CPX-treated MSC culture medium. However, the CPX-treated MSC culture medium showed only a slight proangiogenic effect on sprouting. This might be attributed to dilution effects related to the application of conditioned media in sprouting assays (five-times diluted). Thus, a much lower concentration of proangiogenic factors in CPX-treated MSC culture medium might cause the slight sprouting increase.

In general, VEGF is the master regulator of physiological but also pathological angiogenesis and the most common target for therapeutic angiogenesis approaches [44]. However, as a single agent in supraphysiological doses, VEGF leads to the formation of tortuous and leaky microvessels and tumor formation [45]. A local delivery and combination with other proangiogenic factors would be desirable and therefore using a PHI to locally induce a portfolio of angiogenic factors would be safer and more efficacious.

Taken together, the direct stimulation of ECs with CPX replaced external VEGF supplementation by HIF-1α stabilization via a pharmacological hypoxia resulting in sufficient levels of VEGF and other proangiogenic factors. The strategy of ECs as direct target for a PHI such as CPX represents an innovative angiogenesis approach independent of auxiliary cells such as MSCs.

Footnotes

Acknowledgments

Antje Kremer acknowledged the Studienstiftung des deutschen Volkes for her PhD scholarship. Moreover, the authors kindly thank the “Fraunhofer Gesellschaft” and the “Bayern Fit Programm” of the Bavarian Ministry of Economic Affairs, Energy and Technology for financially supporting the study. Marietta Herrmann is supported by the Interdisciplinary Center for Clinical Research (IZKF) at the University of Wuerzburg (Project D-361).

References

Patan

. Vasculogenesis and angiogenesis. Cancer Treat Res. 2004;117:3–32.

Risau

. Mechanisms of angiogenesis. Nature. 1997 Apr 17;386(6626):671–4. doi: 10.1038/386671a0.

Drake

. Embryonic and adult vasculogenesis. Birth Defects Res C Embryo Today. 2003 Feb;69(1):73–82. doi: 10.1002/bdrc.10003.

Nauta

, van Hinsbergh

, Koolwijk

. Hypoxic signaling during tissue repair and regenerative medicine. Int J Mol Sci. 2014 Oct 31;15(11):19791–815. doi: 10.3390/ijms151119791.

Carmeliet

. Angiogenesis in health and disease. Nat Med. 2003 Jun;9(6):653–60. doi: 10.1038/nm0603-653.

Lee

, Bautch

. Ups and downs of guided vessel sprouting: The role of polarity. Physiology (Bethesda, Md). 2011 Oct;26(5):326–33. doi: 10.1152/physiol.00018.2011.

Nitzsche

, Muller

, Lukomska

, Jolkkonen

, Deten

, Boltze

. Concise Review: MSC Adhesion Cascade-Insights into Homing and Transendothelial Migration. Stem cells (Dayton, Ohio). 2017 Jun;35(6):1446–60. doi: 10.1002/stem.2614.

Pedersen

, Blois

, Xue

, Xing

, Sun

, Finne-Wistrand

, et al. Mesenchymal stem cells induce endothelial cell quiescence and promote capillary formation. Stem Cell Res Ther. 2014 Feb 17;5(1):23. doi: 10.1186/scrt412.

Caplan

, Dennis

. Mesenchymal stem cells as trophic mediators. J Cell Biochem. 2006 Aug 1;98(5):1076–84. doi: 10.1002/jcb.20886.

10.

Pacini

, Petrini

. Are MSCs angiogenic cells? New insights on human nestin-positive bone marrow-derived multipotent cells. Frontiers in cell and developmental biology. 2014;2:20. doi: 10.3389/fcell.2014.00020.

11.

Plumas

, Chaperot

, Richard

, Molens

, Bensa

, Favrot

. Mesenchymal stem cells induce apoptosis of activated T cells. Leukemia. 2005 Sep;19(9):1597–604. doi: 10.1038/sj.leu.2403871.

12.

Phinney

, Prockop

. Concise review: Mesenchymal stem/multipotent stromal cells: The state of transdifferentiation and modes of tissue repair—current views. Stem cells (Dayton, Ohio). 2007 Nov;25(11):2896–902. doi: 10.1634/stemcells.2007-0637.

13.

Chamberlain

, Fox

, Ashton

, Middleton

. Concise review: Mesenchymal stem cells: Their phenotype, differentiation capacity, immunological features, and potential for homing. Stem cells (Dayton, Ohio). 2007 Nov;25(11):2739–49. doi: 10.1634/stemcells.2007-0197.

14.

Uccelli

, Moretta

, Pistoia

. Mesenchymal stem cells in health and disease. Nature reviews Immunology. 2008 Sep;8(9):726–36. doi: 10.1038/nri2395.

15.

Loibl

, Binder

, Herrmann

, Duttenhoefer

, Richards

, Nerlich

, et al. Direct cell-cell contact between mesenchymal stem cells and endothelial progenitor cells induces a pericyte-like phenotype in vitro. Biomed Res Int. 2014;2014:395781. doi: 10.1155/2014/395781.

16.

Blocki

, Wang

, Koch

, Peh

, Beyer

, Law

, et al. Not all MSCs can act as pericytes: Functional in vitro assays to distinguish pericytes from other mesenchymal stem cells in angiogenesis. Stem cells and development. 2013 Sep 1;22(17):2347–55. doi: 10.1089/scd.2012.0415.

17.

Blocki

, Beyer

, Jung

, Raghunath

. The controversial origin of pericytes during angiogenesis - Implications for cell-based therapeutic angiogenesis and cell-based therapies. Clinical hemorheology and microcirculation. 2018;69(1-2):215–32. doi: 10.3233/ch-189132.

18.

Herrmann

, Binder

, Menzel

, Zeiter

, Alini

, Verrier

. CD34/CD133 enriched bone marrow progenitor cells promote neovascularization of tissue engineered constructs in vivo. Stem cell research. 2014 Nov;13(3 Pt A):465–77. doi: 10.1016/j.scr.2014.10.005.

19.

Krock

, Skuli

, Simon

. Hypoxia-induced angiogenesis: Good and evil. Genes Cancer. 2011 Dec;2(12):1117–33. doi: 10.1177/1947601911423654.

20.

Linden

, Katschinski

, Eckhardt

, Scheid

, Pagel

, Wenger

. The antimycotic ciclopirox olamine induces HIF-1alpha stability, VEGF expression, and angiogenesis. Faseb j. 2003 Apr;17(6):761–3. doi: 10.1096/fj.02-0586fje.

21.

Lim

, Sham

, Chee

, Chan

, Raghunath

. Combination of ciclopirox olamine and sphingosine-1-phosphate as granulation enhancer in diabetic wounds. Wound Repair Regen. 2016 Sep;24(5):795–809. doi: 10.1111/wrr.12463.

22.

Kim

, Lee

, Song

, Kim

, Hwang

, Jang

, et al. Combination of three angiogenic growth factors has synergistic effects on sprouting of endothelial cell/mesenchymal stem cell-based spheroids in a 3D matrix. Journal of biomedical materials research Part B, Applied biomaterials. 2016 Nov;104(8):1535–43. doi: 10.1002/jbm.b.33498.

23.

, Nauta

, Morrison

, Zhou

, Zimmermann

, Gurtner

, et al. Antimycotic ciclopirox olamine in the diabetic environment promotes angiogenesis and enhances wound healing. PLoS One. 2011;6(11):e27844. doi: 10.1371/journal.pone.0027844.

24.

Lim

, Kim

, Aref

, Kamm

, Raghunath

. Complementary effects of ciclopirox olamine, a prolyl hydroxylase inhibitor and sphingosine 1-phosphate on fibroblasts and endothelial cells in driving capillary sprouting. Integr Biol-Uk. 2013;5(12):1474–84. doi: 10.1039/c3ib40082d.

25.

Korff

, Augustin

. Integration of endothelial cells in multicellular spheroids prevents apoptosis and induces differentiation. The Journal of cell biology. 1998 Nov 30;143(5):1341–52.

26.

Laschke

, Menger

. Spheroids as vascularization units: From angiogenesis research to tissue engineering applications. Biotechnology advances. 2017 Nov 1;35(6):782–91. doi: 10.1016/j.biotechadv.2017.07.002.

27.

Blocki

, Wang

, Koch

, Goralczyk

, Beyer

, Agarwal

, et al. Sourcing of an alternative pericyte-like cell type from peripheral blood in clinically relevant numbers for therapeutic angiogenic applications. Molecular therapy: The journal of the American Society of Gene Therapy. 2015 Mar;23(3):510–22. doi: 10.1038/mt.2014.232.

28.

Auerbach

, Lewis

, Shinners

, Kubai

, Akhtar

. Angiogenesis assays: A critical overview. Clinical chemistry. 2003 Jan;49(1):32–40.

29.

Obeso

, Weber

, Auerbach

. A hemangioendothelioma-derived cell line: Its use as a model for the study of endothelial cell biology. Lab Invest. 1990 Aug;63(2):259–69.

30.

Staton

, Stribbling

, Tazzyman

, Hughes

, Brown

, Lewis

. Current methods for assaying angiogenesis in vitro and in vivo. International journal of experimental pathology. 2004 Oct;85(5):233–48. doi: 10.1111/j.0959-9673.2004.00396.x.

31.

Nicosia

, Ottinetti

. Growth of microvessels in serum-free matrix culture of rat aorta. A quantitative assay of angiogenesis in vitro. Lab Invest. 1990 Jul;63(1):115–22.

32.

Kremer

, Ribitsch

, Reboredo

, Durr

, Egerbacher

, Jenner

, et al. Three-Dimensional Coculture of Meniscal Cells and Mesenchymal Stem Cells in Collagen Type I Hydrogel on a Small Intestinal Matrix-A Pilot Study Toward Equine Meniscus Tissue Engineering. Tissue Eng Part A. 2017 May;23(9-10):390–402. doi: 10.1089/ten.TEA.2016.0317.

33.

Groeber

, Engelhardt

, Lange

, Kurdyn

, Schmid

, Rucker

, et al. A first vascularized skin equivalent as an alternative to animal experimentation. ALTEX. 2016;33(4):415–22. doi: 10.14573/altex.1604041.

34.

Dominici

, Le Blanc

, Mueller

, Slaper-Cortenbach

, Marini

, Krause

, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy. 2006;8(4):315–7. doi: 10.1080/14653240600855905.

35.

Bohn

, Kraemer

. Dermatopharmacology of ciclopirox nail lacquer topical solution 8% in the treatment of onychomycosis. Journal of the American Academy of Dermatology. 2000 Oct;43(4 Suppl):S57–69.

36.

Jue

, Dawson

, Brogden

. Ciclopirox olamine 1% cream. A preliminary review of its antimicrobial activity and therapeutic use. Drugs. 1985 Apr;29(4):330–41. doi: 10.2165/00003495-198529040-00002.

37.

Maciag

, Hoover

, Stemerman

, Weinstein

. Serial propagation of human endothelial cells in vitro. The Journal of cell biology. 1981 Nov;91(2 Pt 1):420–6.

38.

Heiss

, Hellstrom

, Kalen

, May

, Weber

, Hecker

, et al. Endothelial cell spheroids as a versatile tool to study angiogenesis in vitro. Faseb j. 2015 Jul;29(7):3076–84. doi: 10.1096/fj.14-267633.

39.

Raghunath

, Sy Wong

, Farooq

, Ge

. Pharmacologically induced angiogenesis in transgenic zebrafish. Biochemical and biophysical research communications. 2009 Jan 23;378(4):766–71. doi: 10.1016/j.bbrc.2008.11.127.

40.

Rey

, Semenza

. Hypoxia-inducible factor-1-dependent mechanisms of vascularization and vascular remodelling. Cardiovascular research. 2010 May 1;86(2):236–42. doi: 10.1093/cvr/cvq045.

41.

Sheng

, Xu

. Three decades of research on angiogenin: A review and perspective Acta Biochim Biophys Sin (Shanghai). 2016 May;48(5):399–410. doi: 10.1093/abbs/gmv131

42.

Ning

, Manegold

, Hong

, Zhang

, Pohl

, Lurje

, et al. Interleukin-8 is associated with proliferation, migration, angiogenesis and chemosensitivity in vitro and in vivo in colon cancer cell line models. Int J Cancer. 2011 May 1;128(9):2038–49. doi: 10.1002/ijc.25562.

43.

Fagiani

, Christofori

. Angiopoietins in angiogenesis. Cancer letters. 2013 Jan 1;328(1):18–26. doi: 10.1016/j.canlet.2012.08.018.

44.

Martino

, Brkic

, Bovo

, Burger

, Schaefer

, Wolff

, et al. Extracellular matrix and growth factor engineering for controlled angiogenesis in regenerative medicine. Frontiers in bioengineering and biotechnology. 2015;3:45. doi: 10.3389/fbioe.2015.00045

45.

Ozawa

, Banfi

, Glazer

, Thurston

, Springer

, Kraft

, et al. Microenvironmental VEGF concentration, not total dose, determines a threshold between normal and aberrant angiogenesis. The Journal of clinical investigation. 2004 Feb;113(4):516–27. doi: 10.1172/jci18420.