Interleukin-8 Receptors CXCR1 and CXCR2 Are Not Expressed by Endothelial Colony-forming Cells

Abstract

Endothelial colony-forming cells (ECFCs) are human vasculogenic cells described as potential cell therapy product and good candidates for being a vascular liquid biopsy. Since interleukin-8 (IL-8) is a main actor in senescence, its ability to interact with ECFCs has been explored. However, expression of CXCR1 and CXCR2, the two cellular receptors for IL-8, by ECFCs remain controversial as several teams published contradictory reports. Using complementary technical approaches, we have investigated the presence of these receptors on ECFCs isolated from cord blood. First, CXCR1 and CXCR2 were not detected on several clones of cord blood- endothelial colony-forming cell using different antibodies available, in contrast to well-known positive cells. We then compared the RT-PCR primers used in different papers to search for the presence of CXCR1 and CXCR2 mRNA and found that several primer pairs used could lead to non-specific DNA amplification. Last, we confirmed those results by RNA sequencing. CXCR1 and CXCR2 were not detected in ECFCs in contrary to human-induced pluripotent stem cell-derived endothelial cells (h-iECs). In conclusion, using three different approaches, we confirmed that CXCR1 and CXCR2 were not expressed at mRNA or protein level by ECFCs. Thus, IL-8 secretion by ECFCs, its effects in angiogenesis and their involvement in senescent process need to be reanalyzed according to this absence of CXCR-1 and – 2 in ECFCs.

Introduction

Endothelial colony-forming cells (ECFCs) are postnatal vasculogenic cells isolated in culture and committed to the endothelial lineage [1]. On the last two decades, the interest for these cells has been growing either as a putative cell therapy product for ischemic diseases and more recently, as a liquid biopsy to explore endothelial compartment [1, 2]. During the past, the cellular response of ECFCs after autocrine or paracrine stimulation by chemokines has been the subject of numerous studies in order to better understand the properties of these cells. Among those cytokines, ECFCs were shown to secrete interleukine-8 (IL-8, CXCL8). This secretion of IL-8 increase when ECFCs are senescent [3–5] and IL-8 could enhances the angiogenic properties of ECFCs [6, 7], depending of the expression of its receptors.

Effects of chemokines are mediated by seven-transmembrane-domain receptors. To date 23 receptors have been identified [8]. Among them, CXCR1 and CXCR2, which share 77% amino acid homology, were first characterized as two receptors for IL-8 presents on the surface of neutrophils [9, 10]. It was latter shown that CXCR1 could also bind CXCL6 whereas CXCR2 could interact with all ELR⁺ CXC chemokines [11]. However, presence of IL-8 receptors on mature endothelial cells or ECFCs remains controversial. Petzelbauer et al. were unable to detect CXCR1 or CXCR2 expression on HUVECs or DMEC by RT-PCR nor specific IL-8 binding to cultured HUVECs [12] whereas other teams detected CXCR1 or CXCR2 on HUVECs or HMEC [13, 14]. Similarly, several teams have described the presence of CXCR1 or CXCR2 at mRNA or protein level on ECFCs isolated from cord blood or adult peripheral blood [6, 7, 15, 16] whereas in our team we have never been able to detect those receptors on cord or adult blood derived ECFCs [3, 5, 17].

Because of discrepancies found in literature concerning ECFCs isolation methods (Sup table 1) but also expression or not of CXCR1 and CXCR2 by ECFCs, the possibility or not of direct interactions between ECFCs and IL-8 remains undetermined. The answer to this question may help to explain the pathophysiology of the pro-angiogenic effect of IL-8 in vivo or to determine whether IL-8 could be a candidate molecule to enhance the angiogenic properties of ECFCs as a product of cell therapy. Moreover, new CXCR1/2 inhibitors have been developed for the treatment of clear cell Renal Cell (RCC) and Head and Neck Squamous Cell Carcinomas (HNSCC). Since angiogenesis have a crucial role in RCC and in tumor in general, it’s important to appreciate potential involvement in ECFC expression of these two IL-8 receptors [18].

To answer theses questions, we explored the presence of CXCR1 and CXCR2 receptors on properly characterized ECFCs isolated from cord blood by combining different complementary techniques.

Materials and Methods

Cells Isolation and Culture

Cord-blood endothelial colony-forming cells (CB-ECFCs) were isolated from cord blood adherent mononuclear cell (MNC) fraction as previously described [19–23]. Human umbilical cord bloods were obtained from the Cell therapy Unit of Saint-Louis Hospital (responsible authorities from cell therapy unit: Pr Larghero, AP-HP, Paris, autorisation number AC-2016-2759). ECFCs were then expended on fibronectin (FN)-coated plates (Merck, Germany) using EGM-2MV (Lonza, USA) supplemented with 10% heat inactivated fetal bovine serum (FBS, Sigma-Aldrich, USA) and always used between passages 3 to 6. THP1 cells, purchased from the American Type Culture Collection (ATCC® TIB202™), were cultured in RPMI 1640 (Gibco, USA) supplemented with 10% FBS as previously described [24]. For peripheral blood mononuclear cells (PB-MNC) isolation, blood samples collected on EDTA were obtained from healthy volunteers from Etablissement Français du Sang (EFS, convention n°13/EFS/64). Cells were obtained by density gradient centrifugation with Pancoll (PAN Biotech, Germany). Human induced pluripotent stem cells (h-iPSCs) and human IPSC-derived endothelial cells (h-iECs) were generated from mesenchymal stromal cells isolated from normal discarded subcutaneous white adipose tissue obtained during clinically-indicated procedures in accordance with an Institutional Review Board-approved protocol as previously described [25]. In brief, h-iPSCs were generated via episomal transfection of selected reprogramming factors (Oct4, Sox2, Klf4, L-Myc and Lin28) into human mesenchymal stromal cells (MSCs). H-iPSCs were cultured in mTeSR1 medium (STEMCELL Technologies) on 6-well plates coated with Matrigel. To induce the differentiation of h-iPSCs to h-iECs, h-iPSCs were first dissociated and plated on Matrigel in mTeSR1 medium supplemented with 10 μM Y27632. After 24 hours, the medium was changed to basal medium (Advanced DMEM/F12, 1 × GlutaMax, and 60 μg/mL L-ascorbic acid) supplemented with 6 μM CHIR99021 for 48 hours. The medium was then changed to basal medium supplemented with 50 ng/mL VEGF-A, 50 ng/mL FGF-2, 10 ng/mL EGF and 10 μM SB431542 for another 48 hours. h-iECs were CD31 + cells selected at this step using magnetic beads coated with anti-human CD31 antibodies (DynaBeads, ThermoFisher). H-iECs were then expanded in culture using EGM-2 medium.

ECFCs were used between passage 4 to 8 and days 30 to 35. For the confluence, they were used between 80–100% confluence. Regarding the passage of iECs and iPSCs for RNA-Seq, iECs were sequenced right after differentiation and CD31+ selection (at passage 1). The confluence is around 80–90%. The passage of iPSCs is P18-P23 and the confluence is about 80%.

Flow Cytometry Immunophenotyping

Cultured cell were detached with trypsin, washed in PBS containing 10% FBS and resuspended in PBS/0.5% BSA (Bovine serum albumin, Sigma Aldrich, USA) at the concentration of 10⁶ cells/50 μL. After incubation with FcR Blocking (130-059-901, Myltenyi Biotech, USA) for 5 min at RT, cells were labeled with PE conjugated anti-CXCR1 (FAB330P, R&D, USA), PE conjugated anti-CXCR2 (FAB331P, R&D, USA), non-conjugated anti-CXCR1 (sc-7303, Santa Cruz Biotechnologies, USA) or non-conjugated anti-CXCR2 (sc-7307, Santa Cruz Biotechnologies, USA) for 30 min at a temperature of 4 °C away from light. Isotype-matched antibodies from the same manufacturer were used as negative control. For non-conjugated antibodies, a secondary antibody (F0479, Dako, USA) was added for 30 min at 4 °C away from light. Acquisition was performed on Attune acoustic flow Cytometer (Life Technologies, USA) and analyzed on Attune cytometer software (Life Technologies, USA). For specified experiments, CB-ECFCs were grown in EBM-2 0,5%FBS supplemented with 1 ng/mL, 10 ng/mL, 100 ng/mL IL-8 (618-IL, R&D, USA) for 2 h, 4 h, 24 h or 48 h before immunophenotyping.

For whole blood cytometry, samples collected on EDTA were obtained from healthy volunteers from Etablissement Français du Sang (EFS, convention n°13/EFS/64). Blood was washed three times with PBS 0.5% BSA then labeling was carried as described before. Just before acquisition, cells are resuspended in cell lysis buffer (High-Yield Lyse, Invitrogen, USA) 10 min at RT then acquisition was carried as previously described.

Migration Assay

Migration assay were performed as previously described [8].

Briefly, for ECFCs migration assay, 24-well Transwell PET membranes with 8 μm diameter pores (Corning, USA) were coated with 0.2% gelatin at 37 °C for 30 min. Then, 2 × 10⁴ ECFCs/well were loaded in EBM-2 0.5% FBS into the upper part of the chamber. For indicated experiment, ECFCs were incubated in EBM-2 0,5%SVF supplemented with 10 ng/mL IL-8 (618-IL, R&D, USA) 24 h before. The lower part of the chamber was filled with EBM-0,5%SVF supplemented with 1 ng/mL, 10 ng/mL, 100 ng/mL IL-8 (618-IL, R&D, USA or EGM-2MV 10% FBS used as a positive control. Migration was allowed for 5 h at 37 °C. After fixation and mounting with Vectashied mounting medium containing DAPI (H-1200, Vector Labs, USA), membranes were photographed with an epifluorescence microscope (magnification x10, Evos XL, Life technologies, USA). For each membrane, numbers of cells present on ten separate fields were counted with Image J software.

For neutrophils migration assay, blood samples collected on EDTA were obtained from healthy volunteers from Etablissement Français du Sang (EFS, convention n°13/EFS/64). Neutrophils (PMNs) were isolated with MACSxpress Neutrophil Isolation Kit (Miltenyi Biotech, Germany) according to manufacturer’s instruction. 24-well Transwell PET membranes with 3 μm pore diameter (Corning, USA) were blocked with EBM-2 0.5% FBS at 37 °C for 30 min. Then, 1 × 10⁵ PMNs/well were loaded in EBM-2 0,5%FBS into the upper part of the chamber. The lower part of the chamber was filled with EBM-0,5%SVF supplemented with 1 ng/mL, 10 ng/mL, 100 ng/mL IL-8 (618-IL, R&D, USA). A well without membrane was used as a positive control. Migration was allowed for 90 min at 37 °C. Then, medium in the lower part of the chamber was collected and neutrophils were counted with flow cytometer (Attune, Life Technologies, USA).

RNA-Seq Analysis

RNA-Seq analysis was performed as previously described [25]. The following groups with 3 biological replicates within each group were sequenced: human CB-ECFCs, h-iECs, and h-iPSCs. Total RNA was extracted from each sample using RNeasy Mini Kit (Qiagen) and RNA quantity and quality were checked with nanodrop and Agilent Bioanalyzer instruments. Libraries were prepared and sequenced using Illumina HiSeq2500 platform with 2 × 150 paired end configuration by GENEWIZ. Raw sequencing data (FASTQ files) was examined for library generation and sequencing quality using FastQC (Babraham Institute). Reads were aligned to UCSC hg38 genome using the STAR aligner [26]. Alignments were checked for evenness of coverage, rRNA content, genomic context of alignments, complexity, and other quality using a combination of FastQC and Qualimap [27]. The expression of the transcripts was quantified against the Ensembl release GRCh38 transcriptome annotation using Salmon. These transcripts abundances were then imported into R (version 3.5.1) and aggregated to the gene level with tximport. Differential expression at the gene level was called with DESeq2 [28]. All sample comparison was performed using Likelihood Ratio Test (LRT). Heatmap of enriched gene sets were generated with pheatmap package. The RNA-Seq datasets are deposited online with SRA accession number: PRJNA509218.

Statistics

All analysis were performed with Prism software 5 (GraphPad). Results were analyzed with non-parametric test of Kruskall-Wallis test. P value < 0.05 was considered as statistically significant.

Results

Expression of CXCR1 or CXCR2 by ECFCs Assessed by Flow Cytometry

We compared the abilities of two different sets of antibodies purchased form R&D (FAB330P and FAB331P) and Santa Cruz (sc-7303 and sc-7304) companies to detect expression of CXCR1 or CXCR2 respectively on several cell type including ECFCs. Antibodies used in previous publications (Sup table 2) were either not commercially available anymore or reference not precisely stipulated in publication.

As CXCR1 and CXCR2 were first identified on leukocytes in particular on neutrophils, we tested these antibodies at different concentrations on whole blood samples. Concentrations were chosen in order to flank the concentration advised by the manufacturer. For both set of antibodies, staining increased while increasing antibodies’ concentration and we observed a saturation of staining at the highest concentration of antibodies as expected (Fig. 1).

Fig. 1.

Expression of CXCR1 and CXCR2 on whole blood samples by flow cytometry with increasing concentration of antibodies purchased from R&D (a) or Santa-Cruz (b)

Then, we tested those antibodies at the same concentrations on mononuclear cells (PBMC) (Fig. 2). Antibodies purchased from Santa Cruz Company detected the presence of CXCR1 and CXCR2 on those cells whereas antibodies from R&D seemed a little less sensitive. We separated two populations of cells, one that did not express CXCR1 or CXCR2 and one that expressed at least one of those receptors. Further analysis revealed that the positive population matched with the monocyte population (data not shown). We also tested the different antibodies on THP1, a monocyte cultured cell line and we obtained comparable results as those with PBMC (Sup Fig. 1). Finally, we tested the antibodies, at the same concentrations as before, on cord-blood ECFCs (CB-ECFCs). No significant shift suggesting the presence of cell surface receptor could be observed neither for CXCR1 or CXCR2 with the different antibodies tested (n = 3, Fig. 3). To confirm this result, we compared the mean fluorescent intensity between isotype control antibody and anti-CXCR1 or CXCR2 antibody and we did not find any significant difference (Sup table 3, p = 0.63 and 0.084).

Fig. 2.

Expression of CXCR1 and CXCR2 on peripheral blood mononuclear cell by flow cytometry with increasing concentration of antibodies purchased from R&D (a) or Santa-Cruz (b)

Fig. 3.

Expression of CXCR1 and CXCR2 on CB-ECFCs by flow cytometry with increasing concentration of antibodies purchased from R&D (a) or Santa-Cruz (b)

As some data in the literature suggest that CXCR1 and CXCR2 expression on ECFCs is induced by IL-8 [4, 7], we performed immunophenotyping on CB-ECFCs grown in increasing concentrations of IL-8 (1 ng/mL, 10 ng/mL or 100 ng/mL) for 2 h, 4 h, 24 h or 48 h. We did not detect the presence of CXCR1 and CXCR2 in any of the conditions tested (n = 3, Sup Figs. 2 and 3). To complete these results we added a functional assay by studying the migration of neutrophils or ECFCs toward increasing concentrations of IL-8. For neutrophils, migration capacities improved with increasing concentration of IL-8 (Fig. 4a). At contrary, for ECFCs, we did not found any significant difference between the different concentrations of IL-8 (Fig. 4b). We obtained the same result when ECFCs were cultivated in presence of IL-8 during 24 h before migration (Fig. 4c).

Fig. 4.

Chemotactic migration assay induced by IL-8 at increasing concentrations. a Migration of neutrophils (PMNs) towards Il-8 (n = 9), % of positive control. b Migration of CB-ECFCs towards Il-8 (n = 9), % of positive control. c Migration of CB-ECFCs grown in 10 ng/mL IL-8 for 24 h towards Il-8 (n = 9), % of positive control

Comparison of RT-PCR Primers Used for CXCR1 and CXCR2 Expression Studies

Since no significant membrane expression of CXCR1 and CXCR2 was found on CB-ECFCs in contrast to leukocytes (with the same mAb from different manufacturers), we looked for the presence of CXCR1 or CXCR2 at a transcriptomic level. CXCR1 and CXCR2 have been studied by several groups and discrepancies exist about their specific expression in ECFCs from cord or adult blood [3, 5–7, 16, 17]. Since a lot of data is already available in the literature, we first compared the sequences of RT-PCR primers used in the different publications (Sup table 4 and Sup Fig. 4). CXCR1 has only one transcript possible. Primers used by Yoon or Kimura teams map within a single exon so they may amplify genomic DNA in case of a contamination of RNA. On the contrary, primers used by our team are designed across an intron-exon junction so there are specific of messenger RNA (Fig. 5). Concerning the CXCR2 gene, it has seven alternatives transcripts possible. Again, primers used by Yoon, Kimura or Kwon teams map within a single exon. In consequence, they may detect all alternatives transcripts of CXCR2. However, primers used by our team are again designed across an intron-exon junction. So they are again specific of RNA but they can’t detect two possible alternatives transcripts XM_01700399.2 and XM_01700399.1. Then, we looked for possible off-target amplifications of others transcripts by interrogation of the BLAST database. The primers used by Kimura team for CXCR2 quantification may also amplify the pseudogene CXCR2P1 because, for the reverse primer there is only a nucleotide different in 3’ for the transcript sequence and the 5’ nucleotides are different for the reverse and the forward primers. We did not find other possible off target amplification.

Fig. 5.

Schematic representation of the transcripts of CXCR1 and CXCR2 with the position of the primers used in the different papers (Kimura et al. [6], Yoon et al. [7], Kwon et al. [16], Audigier et al. [17]/Smadja et al. [3].)

Expression of CXCR1 or CXCR2 by ECFCs Assessed by RNA Sequencing

To confirm the absence of expression of CXCR1 and CXCR2 by ECFCs, we performed RNA sequencing on CB-ECFCs, h-iECs and h-iPSCs (Fig. 6a). Both CB-ECFCs and h-iECs express high levels of endothelial lineage markers eNOS (NOS3) and VE-Cadherin (CDH5) compared with h-iPSCs. Since CXCR1 and 2 belong to the chemokine receptor family and are involved in the signaling of SASP molecules, we studied the expression of other chemokine receptors and SASP components. We confirmed that cytokines belonging to the senescence associated secretory phenotype (SASP) such as HGF, IL-6 or CDKN2 were expressed by CB-ECFCs but with few cytokines receptors expression [17]. This profile differs from that observed for h-iECs that expressed another panel of inflammatory cytokines such as CCL2, CCL3, IL-1β and the chemokines receptors CCR3 and CCR5. The two types of cells share few markers in common: CDH5, Serpine-1 (PAI-1) and CDKN1A. On the other hand, h-iPSCs, that are in undifferentiated pluripotent state, secreted few cytokines, with the exception of VEGFA. CXCR1 and CXCR2 were not detected in CB-ECFCs. CXCR1 was not detected in any of the cell populations studied whereas h-iEC and h-iPSC were the only cells to express CXCR2 (Fig. 6b and c).

Fig. 6.

a Relative level of expression of HGF, IL-6, CDKN2A, NOS3, VWF, SERPINE-1, CDKN1A, CDH5, VEGFA, CCR8, CXCR2, CCR5, CCL3, CCR3, CCL2, IL-1B, CXCL-1, CCL5 by CB-ECFCs, h-iECs and h-iPSCs evaluated by RNA sequencing. b Expression of CXCR1 by CB-ECFCs, h-iECs and h-iPSCs evaluated by RNA sequencing. c Expression of CXCR2 by CB-ECFCs, h-iECs and h-iPSCs evaluated by RNA sequencing

Discussion

In this study, we have demonstrated that cord blood derived ECFCs do not express IL-8 receptors CXCR1 and CXCR2.

First, we explored ECFCs’ expression of CXCR1 and CXCR2 by flow cytometry. We validated the two antibodies chosen for each receptor on cells known to express CXCR1 and CXCR2, leucocytes (of total blood sample and after separation by Ficoll) and THP1, a monocyte cell line. All staining were specific on these cells. We secondly tested those antibodies on CB-ECFCs. We studied fresh and frozen cells, because freezing can alter expression of cell markers. We could not detect CXCR1 or CXCR2 on ECFCs. Then, we compared the primers of RT-qPCR used for detection of CXCR1 and CXCR2 on ECFCs in the literature. When designing primers for quantitative PCR, it is usually advised to span an intron-exon junction in order to improve the specificity of the test [29]. Otherwise, amplification of DNA that may contaminate the sample after RNA extraction remains possible. Our team never detected CXCR1 or CXCR2 transcripts in PB or CB-ECFCs and is the only one to have primers that spans exon-intron junction. This may reflect a better specificity of the primers tested. This result was confirmed by RNA sequencing which did not detect CXCR1 or CXCR2 in CB-ECFCs whereas we could detect them in h-iEC or h-iPSC. h-iECs were chosen as control cells because of their in vitro and in vivo pro-angiogenic properties that resemble those of ECFCs [25]. IL-8 is a chemokine that belongs to the CXC chemokines ELR⁺ group. IL-8 was first identified as a modulator of the immune response because of its abilities to attract neutrophils in a context of inflammation [30]. It was secondly shown to play a role in angiogenesis like all ELR⁺ chemokines [31]. So far, two receptors for IL-8 have been identified on neutrophils, CXCR1 and CXCR2. Various reports about expression of those receptors by mature endothelial cells have been published. Several teams have described their presence at the surface of human umbilical vein endothelial cell (HUVEC) [13] or lung and dermal microvascular endothelial cell (HMEC) [32]. On the contrary, CXCR1 and CXCR2 were not detected on HUVEC or DMEC according to Petzelbauer et al. [12]. According to ECFCs, the same discrepancy is found in the literature. Expression of CXCR1 and CXCR2 by ECFCs is described by several teams, that mediate the effect of IL-8 [6, 7, 15] or other agonist such as N-acetylated proline-glycine-proline (AC-PGP) [16]. On the contrary, our team never detected those receptors on CB-ECFCs or PB (peripheral blood)-ECFCs even after activation of PAR-1 thrombin receptor on CB-ECFCs [3, 5, 17]. Thus, we concluded that IL-8 secreted by ECFCs has no direct effect on those cells but it can have a paracrine effect on neighbor cells.

Several hypotheses could explain these conflicting results. Petzelbauer et al. [12] suggest that some cell cultures may contain a subpopulation of IL-8 responsive cells such as mast cells or T-cells. Mast cells and T lymphocytes secrete proangiogenic factors that can induce angiogenesis in vitro [33]. The presence of inflammatory cells in endothelial cell culture may depend on cell culture medium used. Expression of CXCR1 and/or CXCR2 by endothelial cells could also be modulated by cell culture conditions [34]. Another possible explanation is the hypothesis of variable expression according to tissue or endothelial cell population. CB-ECFCs is not a well-defined population. Endothelial progenitor cells (EPC) were first described in 1997 and since then this appellation has been used for completely different populations of cells identified either by flow cytometry or cell culture approaches. Among EPC isolated by cell culture were later identified two types of cells, “early” endothelial progenitor cells of myeloid origin and ECFCs, the true endothelial progenitors [35]. However, even among the ECFCs population remains a certain variability explained by difference in ECFCs isolation protocols or cell culture procedure [36]. We cannot refute the hypothesis that characteristic of the cells studied in the papers are slightly different. This may explain why we could not detect expression of CXCR1 and CXCR2 on CB-ECFCs whereas other teams did. A final explanation could be represented by the “state” of the ECFCs. Indeed, the various studies carried out do not describe the age of the cells or their state of senescence. Therefore, it cannot be excluded that the expression of CXCR1 and CXCR2 increases with age or senescence since our data concern only young and apparently no senescent cells.

Besides its chemoattractant properties, IL-8 is a chemokine secreted among others by senescent cells. It was shown that young CB-ECFCs do not secret IL-8 if no activation is provided, in contrast to PB-ECFCs [3]. In idiopathic pulmonary fibrosis, a pathologic condition associated with senescence of ECFCs an increase of IL-8 secretion has been observed [5]. Senescence is described as an irreversible growth arrest that can be triggered by stresses such as telomere shortening, DNA damage or reactive oxygen species [37]. Above the initial triggers, some data suggest that components of the SASP could modulate the process of senescence. Medina’s group suggested that IL-8 can have autocrine effect on CB-ECFCs as depletion of IL-8 in those cells by RNA interference approaches markedly delay their senescence and improved their functionality [4]. Their hypothesis is a potential activation of CXCR2 by IL-8 and a recruitment of NAPDH oxidases to produce reactive oxygen species inducing senescence [38]. However, opposite observations have also been published. Depletion of IL-8 in placenta-derived mesenchymal stem cell paradoxically leads to an increase of their senescence [38] and in HUVECs IL-8 prevents senescence induced by H₂O₂ via telomerase activation [39]. Finally, addition of exogenous IL-8 to ECFCs does not seem to induce senescence and has modest effect on angiogenic test depending of ECFCs age and IL-8 secretion [4].

However, the secretion of other factors among other factors that belongs to the «SASP» senescent-associated secreted phenotype could be confounding factor. For example, besides IL-8, IGFBP-3 and – 5, or VEGI are secreted by senescent HUVECs [40]. Some of the biological effect attributed to IL-8 in some papers quoted before could be due to others cytokines or proteins whose signaling path depends on another cellular receptor. ECFCs express most of the cellular receptor for CCL cytokines (CCR1, -2, -3, -4, -5, -8, -9, -10) [17] and some cytokines that have an angiogenic effect such as CCL2 (MCP-1) [41]. Modulating IL-8 can also lead to modification of other cytokines concentration. For example, depletion of IL-8 is accompanied by a decrease in MCP-1 and IL-6 concentration [4], two cytokines that could also influence senescence. Similarly, in in vivo models, effects attributed to the action of CXCR1 or CXCR2 ligands on endothelial cells/ECFCs could actually be due to other cell types. For example, Kwon et al. described that N-acetylated proline-glycine-proline (AC-PGP) promotes also revascularization in a rat model of cutaneous wound [16]. However, this revascularization is associated with an infiltration of monocytes/macrophages, cells that are known to enhance the process of vascularization, mostly in a paracrine manner [33].

In summary, our data show an absence of IL-8 receptors ECFCs isolated from cord blood by several techniques. IL-8 secretion in senescent or activated ECFCs needs to be reanalyzed according to paracrine effect and cell cooperation driven by ECFCs. Absence of CXCR-1 and 2 receptors combined to the use of ECFCs as liquid biopsy could provide a better understanding of ECFCs involvement in angiogenic related disorders.