Insulin-like growth factor 2 regulates the proliferation and differentiation of rat adipose-derived stromal cells via IGF-1R and IR

CHAO WANG1,2,3, XIAOMING LI4, HONGXING DANG1,2,3, PING LIU4, BO ZHANG4 & FENG XU1,2,3
1Department of Pediatric Intensive Care Unit, Children’s Hospital of Chongqing Medical University, Ministry of Education Key Laboratory of Child Development and Disorders, Chongqing, 400014 China,2China International Science and Technology Cooperation Base of Child Development and Critical Disorders, Chongqing, 400014 China,3Chongqing Engineering Research Center of Stem Cell Therapy, Children’s Hospital of Chongqing Medical University, Chongqing, China, and 4State Key Lab of Trauma, Burns and Combined Injury, Institute of Surgery Research, Daping Hospital, Army Medical University, Chongqing 400042, China

Abstract
Background: Insulin-like growth factor 2 (IGF2), an essential component of the stem cell niche, has been reported to modu- late the proliferation and differentiation of stem cells. Previously, a continuous expression of IGF2 in tissues was reported to maintain the self-renewal ability of several types of stem cells. Therefore, in this study, we investigated the expression of IGF2 in adipose tissues and explored the effects of IGF2 on adipose-derived stromal cells (ADSCs) in vitro. Methods: The expression pattern of IGF2 in rat adipose tissues was determined by gene expression and protein analyses. The effect of IGF2 on proliferation, stemness-related marker expression and adipogenic and osteogenic differentiation was systematically investigated. Furthermore, antagonists of IGF2-specific receptors—namely, BMS-754807 and picropodophyllin—were added to explore the underlying signal transduction mechanisms. Results: IGF2 levels displayed a tendency to decrease with age in rat adipose tissues. After the addition of IGF2, isolated ADSCs displayed higher proliferation and expression of the stemness-related markers NANOG, OCT4 and SOX2 and greater differentiation potential to adipocytes and osteoblasts. Additionally, both type 1 insulin-like growth factor receptor (IGF-1R) and insulin receptor (IR) participated in the IGF2- mediated promotion of stemness in ADSCs. Conclusions: Our findings indicate that IGF2 could enhance the stemness of rat ADSCs via IGF-1R and IR and may highlight an effective method for the expansion of ADSCs for clinical application.

Key Words: adipose-derived stromal cells (ADSCs), insulin-like growth factor 2 (IGF2), multipotency, stem cell therapy

Introduction
Mesenchymal stromal cells (MSCs), which are capa- ble of self-renewal and multilineage differentiation, display great potential for tissue engineering and cell-based therapy. In addition, their strong para- crine and immunomodulatory functions have been recently demonstrated to result in a more restorable local and systematic niche that can facilitate tissue repair and regeneration [1]. Adipose tissue that is easily accessible and contains abundant MSCs, also called adipose-derived stromal cells (ADSCs), has gained special interest in recent decades [2]. It is believed that ADSCs residing in adipose tissues
account for as much as 0.01—0.1% of total cells, compared with 0.001—0.01% of bone marrow

stromal cells (BMSCs) [3]. Unfortunately, the num- ber and proliferative capacity of MSCs have been reported to decrease with increasing age, which remains a hurdle for in vitro expansion, an indispens- able step before clinical use, given that the amount of cells required for effective therapy is approXimately 106/kg body weight [4 6]. Moreover, the age-related decrease in the ability of MSCs to differentiate into osteoblasts, chondrocytes and endothelial cells partly restricts the application of MSCs in tissue engineer- ing for bone, cartilage and blood vessels [4,7]. Therefore, finding a method to maintain or enhance the self-renewal and multipotency of MSCs remains a promising approach for their clinical application, especially for older adults.

Correspondence: Feng Xu, MD, PhD, Department of PICU, Children’s Hospital of Chongqing Medical University, Ministry of Education Key Laboratory of Child Development and Disorders, 136 Zhongshan No. 2 Road, Yuzhong, Chongqing 400014, People’s Republic of China. E-mail: [email protected]

(Received 10 August 2018; accepted 20 November 2018)
ISSN 1465-3249 Copyright © 2018 International Society for Cell and Gene Therapy. Published by Elsevier Inc. All rights reserved. https://doi.org/10.1016/j.jcyt.2018.11.010

176 C. Wang et al.

Insulin-like growth factor 2 (IGF2), a protein hormone, shares structural similarity with insulin and potentially regulates many developmental pro- cesses, including growth, differentiation and senes- cence [8]. IGF2 is expressed in a number of tissues throughout development and adulthood. As previ- ously reported, IGF2 is persistently present at a high level in the fetus and in a glycosylated form in adult rat serum, and with the exception of that in brain tis- sues in rodents, its expression in the muscle, liver and bone decreases sharply 3 weeks after birth [9 11]. Interestingly, an increasing number of stud- ies have demonstrated that IGF2 is highly beneficial for maintaining self-renewal and differentiation capacities in embryonic stem cells (ESCs) and neural stromal cells (NSCs). For example, IGF2 alone is sufficient for maintaining ESC cultures in the absence of a feeder layer in vitro [12]. In addition, IGF2 is more potent in promoting the expansion of NSCs than their standard growth medium [13,14]. These data suggest that the continuous expression of IGF2 may be an essential environmental factor for preserving the stemness of several types of stem cells. However, little is known regarding the change in the expression pattern of IGF2 in rat adipose tissues with increasing age and the role of this variation in the regulation of ADSC biology. Consequently, the primary aim of this study was to determine the con- tribution of age-related changes in IGF2 expression to the self-renewal and multilineage differentiation of ADSCs.
IGF2 is a short, single-chain peptide of 67 70 amino acids in length with a molecular weight of 7.5 kD. Its activity is regulated by the expression of sev- eral receptors and IGF-binding proteins (IGFBPs) [8,14]. IGF2 binds to IGF2 receptors to trigger intrinsic tyrosine kinase activities and subsequently activates the PI3K/AKT pathway or the MAPK/ ERK pathway; these pathways have been proven to be critical in stemness modulation [15,16]. Nonethe- less, by initially binding to distinctive cell membrane receptors, IGF2 can exert different effects on biologi- cal processes in MSCs. In general, IGF2 has a high affinity for three receptors: insulin receptor (IR), especially IR-A; type 1 IGF receptor (IGF-1R); and type 2 IGF receptor (IGF-2R), with IGF-2R display- ing no intrinsic kinase activity and regulating extra- cellular IGF2 levels through receptor-mediated endocytosis followed by IGF2 degradation in the lysosome [8,14,17]. As demonstrated by some stud- ies, IGF-1R is involved in maintaining ESC proper- ties, and blocking IGF2/IGF-1R signaling reduces the survival and clonogenicity of ESCs [12,18]. Fur- thermore, IR is the most highly expressed receptor in NSCs, and the self-renewal capacity of these cells depends exclusively on IR [13]. Overall, IGF2 may

engage different receptors in modulating the stem- ness of various types of stem cells. Therefore, the exact signaling pathways mediated by IGF2 via spe- cific receptors in ADSCs need to be elucidated.
In the present study, we hypothesized that IGF2 expression in adipose tissues changes with age, and this change may be involved in an age-associated var- iation in the stemness of ADSCs. Therefore, we first examined the pattern of IGF2 expression in adipose tissues from rats of different ages. Then, we aimed to explore the effect of IGF2 on the proliferation, stem- ness-related marker expression and multilineage dif- ferentiation of ADSCs. We also investigated the potential signaling pathways of IGF2 in ADSCs.

Methods
Animals
All rats used in the study were purchased from the EXperimental Animal Center of Chongqing Medical University (Chongqing, China). All animal proce- dures related to tissue isolation in the present study were in accordance with the Guide for the Care and Use of Laboratory Animals and were also approved by the Animal Ethics Committee of Chongqing Medical University (Chongqing, China).

Isolation and culture of ADSCs
Adipose tissues were harvested from bilateral subcu- taneous inguinal areas of female Sprague Dawley rats (8 10 weeks, 200 230 g) anesthetized with 1.5% pentobarbital sodium solution. The tissues were thoroughly washed with phosphate-buffered saline (PBS). After carefully removing vessels and fascias, we minced the tissue into small sections ( 0.1 mm3) and digested them with 0.075% type II collagenase (Sigma, St. Louis, MO, USA) at 37˚C with gentle agitation for 40 min. Four volumes of PBS were then added to dilute collagenase, followed by filtration through 100-mm cell strainers and cen- trifugation at 500g for 10 min. The remaining pellets were washed and resuspended in DMEM/F12 (HyClone, Logan, UT, USA) containing 10% fetal bovine serum (FBS) (Gibco Invitrogen, Carlsbad, CA, USA) and 1% penicillin-streptomycin solution. The obtained cells were immediately plated onto T75 tissue culture flasks and incubated at 37˚C in a humidified atmosphere containing 5% CO2. The flasks were gently washed with PBS every 24 h for
2 3 days to eliminate non-adherent cells. Adherent cells were passaged until 80 90% confluence with TrypLE EXpress solution (Gibco) at a ratio of 1:2 (passage 0). The medium was changed every 2 days. ADSCs at passage 3 were used in experiments.

IGF2 regulates the proliferation and differentiation of ADSCs 177

ADSCs at passage 3 were plated at a density of
4 103 cells/cm2 in triplicates. Upon reaching
70 80% confluence, the cells were harvested and counted. Generation time (GT) was calculated using the formula: GT t= log N=N_0 = log2 , where t is time from seeding to counting the cells, N is the final number of cells N0 is the number of plated cells.

Flow cytometry
Surface biomarkers of ADSCs were characterized using flow cytometry. Briefly, ADSCs were trypsi- nized and washed to obtain a single-cell suspension. Subsequently, the cells were incubated with the fol- lowing primary antibodies at room temperature for 30 min: PE-conjugated anti-rat CD29 (BioLegend, San Diego, CA, USA), PE-conjugated anti-rat/ mouse CD90 (BioLegend, USA), fluorescein iso- thiocyanate (FITC)-conjugated anti-rat CD34 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and FITC-conjugated anti-rat CD45 (BD Bioscien- ces, San Jose, CA, USA). Unstained cells were used as negative controls. The samples were then washed twice with PBS and characterized using a Novocyte flow cytometer (ACEA Biosciences, San Diego, CA, USA). The raw data were further analyzed using NovoEXpress software (ACEA Biosciences).

In vitro multilineage differentiation assay
To detect the differentiation ability of ADSCs, cells at a density of 5 104 per well were seeded in six- well plates and cultured in the respective differentia- tion medium.
For adipogenic differentiation, ADSCs at 90% con- fluence were switched to adipogenic induction medium

FBS supplemented with 10 mmol/L b-glycerophos- phate, 100 nmol/L dexamethasone and 50 mg/mL ascorbic acid-2-phosphate (all from Sigma) for 21 days. The induced cells were cultured with or without IGF2, BMS-754807 and PPP. The medium was replaced every 3 days. Calcium deposition was analyzed with alizarin red staining (Sigma), and the expression of osteogenesis-associated genes was eval- uated with qRT-PCR.
For chondrogenic differentiation, a total of
1 106 ADSCs were centrifuged at 500g for 10 min in 15-mL tubes, and the pellets were then incubated for 21 days using a Chondrogenesis Differentiation Kit (Gibco). The medium was changed every 3 days. The cells were then fiXed with 4% paraformaldehyde and embedded in paraffin. Sulfated glycosaminogly- can-rich extracellular matrix was detected by staining with alcian blue solution (Sigma).

MTT assay
Proliferation assays were performed using 3-(4,5- dimethyl-2-thiazolyl)-2,5-diphenyl-2-H-tetrazolium bromide (MTT) according to the manufacturer’s protocol (Beyotime, Shanghai, China). ADSCs were seeded in 96-well cell culture plates at a density of
3 103 cells per well. After overnight attachment, the cells were starved for 6 h in medium without FBS, and then IGF2 at different concentrations (0, 50, 100 and 200 ng/mL) was added to DMEM/F12 containing 10% or 1% FBS. BMS-754807 or PPP was added 2 h before IGF2 stimulation. At 24 and 48 h, 20 mL of MTT solution (5 mg/mL) was added to each well, and the plates were incubated for 4 h in a 5% CO2 incubator at 37˚C. Next, the supernatants were carefully discarded, and 200 mL dimethyl sulf-

containing 1 mmol/L dexamethasone, 10 mg/mL insu-

oXide

was added, and the plates were shaken for

lin, 200 mmol/L indomethacin and 0.5 mmol/L 3-isobutyl-1-methylXanthine (all from Sigma) in Dul- becco’s Modified Eagle’s Medium (DMEM) HG sup- plemented with 10% FBS for 3 days and then changed to adipogenic maintenance medium contain- ing 10 mg/mL insulin for 2 days. The cells were treated with or without IGF2 (Peprotech, Rocky Hill, NJ, USA), the IGF-1R/IR antagonist BMS-754807 (Selleck, Houston, TX, USA) and the IGF-1R antag- onist picropodophyllin (PPP, Selleck). After three cycles of induction, the formation of intracellular lipid droplets was evaluated with oil red O staining (Sigma) after fiXation in 4% paraformaldehyde (Sigma), and the expression of adipogenesis-related genes was mea- sured by quantitative reverse transcription polymerase chain reaction (qRT-PCR).
For osteogenic differentiation, cultured ADSCs at 70% confluence were changed to osteogenic induction medium in DMEM/F12 containing 10%

5 min. The absorbance of formazan was measured at a test wavelength of 490 nm using an enzyme-linked immunoassay reader (BioTek, Winooski, VT, USA).

Western blot analysis
ADSCs cultured in six-well plates were starved for 6 h and pretreated with BMS-754807 or PPP for 2 h and subsequently incubated with IGF2 at different con- centrations in DMEM/F12 containing 10% FBS or 1% FBS for 48 h. The treated cells were then washed with cold PBS and lysed with RIPA lysis buffer con- taining a protease inhibitor cocktail. For extraction of total protein from adipose tissues of rats of different ages (1 day, 8 days, 10 days, 25 days and 8 weeks), we added RIPA lysis buffer to the homogenized tissues. After removal of debris by centrifugation and exami- nation of protein concentration with the BCA assay, 20 mL protein lysate was loaded for SDS-PAGE and

178 C. Wang et al.

then transferred onto PVDF membranes. The mem- branes were incubated with blocking solution (3% w/ v bovine serum albumin) and the following primary antibodies at room temperature for 1 h and at 4˚C overnight, respectively: anti-GAPDH (Abcam, Cam- bridge, UK), anti-NANOG (Abcam, UK), anti- OCT4 (Novus, Littleton, CO, USA), anti-SOX2 (Bioworld, St. Louis, MO, USA) and anti-IGF2 (Bio- world, USA). The membranes were then rinsed with TBST (TBS with 0.01% Tween), incubated with horseradish peroXidase conjugated secondary anti- bodies (Abcam, UK) at room temperature for 1 h and visualized by SuperSignal West Pico Chemilumines- cent Substrate (Thermo, Waltham, MA, USA) using an enhanced chemiluminescence detection system. GAPDH served as an internal control.

RNA isolation and qRT-PCR
Total RNA was extracted from differentiated cells and homogenized adipose tissue samples using TRIzol reagent (Invitrogen, Carlsbad, CA, USA). RNA con- centration was determined by a Nanodrop 2000 spec- trometer. After the removal of residual DNA with a TURBO DNA-free Kit (Thermo), 1 mg total RNA was used to synthesize cDNA with a Transcriptor First Strand cDNA Synthesis Kit (Roche, Basel, Switzer- land). Subsequently, qRT-PCR was performed using FastStart Essential DNA Green Master Mix (Roche, Switzerland) and the LightCycler 96 System (Roche, Switzerland). The expression level of genes was ana- lyzed and normalized to Gapdh. The fold change in
gene expression was evaluated using the 2¡44CT method. Primer sequences are provided in Table I.

Immunofluorescence
The cultured cells were fiXed with 4% paraformalde- hyde for 30 min and washed three times with PBS. After incubation with 0.3% Triton X-100 (Sigma) for 10 min, the samples were washed and blocked with

5% bovine serum albumin for 1 h at room tempera- ture. The primary antibodies, anti-IGF-1R (1:200, CST, Danvers, MA, USA) and anti-IR (1:200, CST), were then added, and the cells were incubated over- night at 4˚C. Next, the samples were thoroughly rinsed with PBS and incubated with FITC-conjugated goat anti-rabbit IgG (Abcam) for 2 h at room temper- ature in the dark. 40,6-diamidino-2-phenylindole was then used for staining cell nuclei. The immunofluores- cence images were captured with a fluorescence microscope (Olympus, Shinjuku, TKY, Japan).

Statistical analysis
Statistical analyses were performed using GraphPad Prism software v5.04 (GraphPad Software Inc.). The results are expressed as mean SD from at least three individual experiments. All data were analyzed using one-way analysis of variance, followed by a Bonferroni multiple comparison post hoc test, where P < 0.05 was considered statistically significant.

Results
IGF2 expression in inguinal adipose tissues from rats of different ages
To investigate the change in the expression pattern of IGF2 in rat adipose tissues with age, we isolated inguinal adipose tissues from rats aged 1 day, 8 days, 10 days, 25 days and 8 weeks. qRT-PCR analysis showed that IGF2 mRNA expression in adipose tis- sues isolated from 25-day-old and 8-week-old rats was dramatically lower than that in tissues isolated from 1-, 8- and 10-day-old rats, and IGF2 expres- sion in adipose tissues from 8-week-old rats was barely detectable (Figure 1B; P < 0.05). Similarly,
IGF2 protein expression in adipose tissues from
8-week-old rats was lower than that in adipose tissues from 1- and 8-day-old rats (Figure 1A,C; P < 0.05), demonstrating that IGF2 expression in rat adipose tissues decreases with increased age.

Table I. Primers used in the present study.
Gene Forward primer (50 to 30) Reverse primer (50 to 30) Length (bp) Accession number

Fabp4 AAATCACCCCAGATGACAG GACACATTCCACCACCAG 140 NM_053365
Lpl TTGACTCCCTGCTGAATG CTCTTGGCTCTGACCTTGT 145 NM_012598
Ppar-g CAGGAGCAGAGCAAAGAG TGGTCATTCAAGTCAAGGT 131 NM_013124.3
Alp CATCGGACCCTGCCTTAC GGAGACGCCCATACCATC 169 NM_013059
Bsp CAGCACGGTTGAGTATGG AAGCTCGGTAAGTGTCCC 113 NM_012587
Runx2 AGGCATTTCATCCCTCACTG TATGGAGTGCTGCTGGTCTG 201 NM_001278483
Igf2 AGCTCGAAGCGTTCAGAGAG CACTGATGGTTGCTGGACAT 109 NM_031511.2
Igf-1r CCAACGGATTGATTCTAATG CACAGGATCTGTCCACGAC 188 NM_052807.2
Ir CAGTTTGTGGAACGGTGCTG TGGTAGGGTCATCGGGTTCT 142 NM_017071.2
Gapdh GGCAAGTTCAACGGCACAGTC CGGAAGGGGCGGAGATGAT 217 NM_017008.4
Gapdh = glyceraldehyde 3-phosphate dehydrogenase.

IGF2 regulates the proliferation and differentiation of ADSCs 179

Figure 1. IGF2 expression in adipose tissues of rats of different ages. (A) Western blot analysis of IGF2 expression in adipose tissues of rats of different ages (1 day, 8 days, 10 days, 25 days and 8 weeks), and (C) quantitative analysis of protein levels. (B) qRT-PCR measurement of IGF2 mRNA expression in rat adipose tissues. Levels in 1-day-old rats were taken as 1.0. *P < 0.05 compared with 1-day-old rat group.

Characterization of rat ADSCs
To evaluate whether the age-related decrease in IGF2 expression is associated with the modulation of ADSC properties, we separated ADSCs from inguinal adipose tissues of 8-week-old rats. Cultured ADSCs adhered to plastic dishes after 24 h and exhibited spindle- and round-shaped morphologies. After three passages, the cells showed a homoge- neous fibroblastic morphology (Figure 2A). The pri- mary cells reached 80 90% confluence in approXimately 4 5 days and the mean GT of third passage cells was 46.08 2.4 h. In addition, ADSCs showed the ability to differentiate into adipogenic, osteogenic and chondrogenic lineages after being cultured in corresponding induction medium in vitro, as confirmed by oil red O, alizarin red and alcian blue staining, respectively (Figure 2B). Flow cytometry demonstrated that ADSCs were positive for MSC- specific surface markers (CD90 and CD29) and neg- ative for hematopoietic cell markers (CD34 and CD45) (Figure 2C), indicating that the isolated rat ADSCs met the standard recommended by the International Society for Cell Therapy [19] and were suitable for subsequent experiments.

Effect of IGF2 on the proliferation and stemness-related marker expression of ADSCs
To explore the proliferative potential and stemness- related marker expression, we seeded ADSCs in cul- ture medium containing different concentrations of IGF2 in the presence of 1% or 10% FBS in vitro. MTT analysis showed that none of the concentra- tions of IGF2 had any effect on cell proliferation in the presence of 10% FBS (Figure 3A). In contrast,

in the presence of 1% FBS, IGF2 at concentrations ranging from 50 to 200 ng/mL promoted the prolif- eration of ADSCs relative to that of control cells (Figure 3B; P < 0.05). The relative protein levels of the stemness-related markers NANOG, OCT4 and SOX2 were higher in the 100 ng/mL IGF2 stimula- tion group in the presence of 1% FBS than in the other three groups (control, 50 ng/mL and 200 ng/ mL) (Figure 3E,F; P < 0.05). Consistent with the MTT assay results, IGF2 had no significant effect on the expression of stemness-associated proteins in the presence of 10% FBS (Figure 3C,D). These findings indicate that IGF2 promotes the proliferation and stemness properties of ADSCs, and 100 ng/mL
IGF2 in the presence of 1% FBS represents the opti- mal stimulation condition.

Effect of IGF2 receptor inhibitors on the proliferation and stemness-related marker expression of ADSCs
To further confirm the function of IGF2 in regulat- ing the self-renewal potential and its related recep- tors, we employed two specific inhibitors of IGF2 receptors, namely, BMS-754807 and PPP. As detected by qRT-PCR and immunofluorescence assays, purified ADSCs expressed both IGF-1R and IR (Figure 4A). The results of cytotoXicity assays showed that the proliferation of ADSCs was reduced in the presence of BMS-754807 concentrations higher than 100 nmol/L and PPP concentrations higher than 200 nmol/L (supplementary Figure 1;
P < 0.05). Therefore, 100 nmol/L BMS-754807 and 200 nmol/L PPP were used for subsequent experi-
ments. Additionally, the proliferation of ADSCs in the BMS-754807-treated group was lower than the proliferation of ADSCs in the IGF2- and PPP-

180 C. Wang et al.

Figure 2. Characterization of ADSCs isolated from 8-week-old rats. (A) Phase-contrast images of primary passage ADSCs after adhesion for 24 h (a) and third passage ADSCs (b, c). (B) Multilineage differentiation ability of ADSCs in vitro. Adipogenic differentiation of ADSCs was identified by oil red O staining after 15 days of induction (left). Alizarin red staining of ADSCs after 21 days of osteogenic induction (middle) and alcian blue staining of ADSCs after 21 days of chondrogenic induction (right). (C) Flow cytometric analysis of surface markers of ADSCs. The cells were positive for CD90 and CD29 and negative for CD34 and CD45. Bar = 200 mm.

treated groups; however, the proliferation of ADSCs in the PPP-treated group was not significantly differ- ent from the proliferation of ADSCs in the IGF2 group but was higher than that of ADSCs in the con- trol group (Figure 4B; P < 0.05). Western blot assays demonstrated that addition of both BMS-754807 and PPP blocked the promotion of NANOG, OCT4, SOX2 protein expression induced by
100 ng/mL IGF2, with no significant difference between the BMS-754807- and PPP-treated groups (Figure 4C F; P < 0.05). Collectively, these results show that the effect of IGF2 on the proliferation of ADSCs is mainly mediated by IR, and IGF-1R is more potent at increasing the IGF2-induced expres- sion of stemness markers in ADSCs.

Effect of IGF2 and its receptor inhibitors on the adipogenic differentiation ability of ADSCs
One of the essential characteristics of MSCs is their ability to differentiate into cells of adipogenic lineage

in the presence of adipogenic induction medium. To determine the adipogenic differentiation capacity of ADSCs, we cultured ADSCs with adipogenic induc- tion medium for 15 days with or without IGF2 and its receptor inhibitors BMS-754807 and PPP. Oil red O staining indicated higher numbers of positive cells in the IGF2-treated group than in the control group; however, the effect was inhibited by the addition of BMS-754807 and PPP (Figure 5A). Interestingly, the percentage of positive cells in the PPP-treated group was higher than that in the BMS- 754807-treated group but showed no significant difference relative to the control group. qRT-PCR analysis of adipogenic markers demonstrated that the expression of Fabp4, Lpl and Ppar-g was higher
in the IGF2-treated group than in the control group (Figure 5B; P < 0.05). However, the increase was abrogated by the addition of BMS-754807 and PPP, and the level of adipogenic markers in the BMS- 754807-treated group was lower than that in the PPP-treated group (P < 0.05). These findings

IGF2 regulates the proliferation and differentiation of ADSCs 181

Figure 3. Effect of IGF2 on the proliferation and stemness-related marker expression of ADSCs in vitro. (A and B) Proliferation of ADSCs was measured by MTT assays. IGF2 at concentrations of 50, 100 and 200 ng/mL was added to the culture medium containing 10% FBS
(A) or 1% FBS (B) for 48 h. (C—F) Western blot analysis of the expression of stemness-related markers in ADSCs after IGF2 stimulation for 48 h in the presence of 10% FBS (C and D) or 1% FBS (E and F). Quantitative measurement of protein levels (NANOG, OCT4 and SOX2) in the presence of 10% FBS (D) or 1% FBS (F). *P < 0.05.

suggest that IGF2 increases the adipogenic differen- tiation ability of ADSCs mainly through IGF-1R/IR.

Effect of IGF2 and its receptor inhibitors on the osteogenic differentiation ability of ADSCs
To examine the role of IGF2 and its receptor inhibi- tors in the osteogenic differentiation ability of ADSCs, we cultured ADSCs with or without IGF2, BMS-754807 and PPP in osteogenic induction medium for 21 days. Alizarin red staining showed that calcium deposition in the IGF2-treated group was higher than that in the control group. However, only the addition of BMS-754807 inhibited the cal- cium deposition induced by IGF2, whereas the addi- tion of PPP had no impact on calcium deposition in comparison with the addition of IGF2 (Figure 6A). As detected by qRT-PCR, the expression of Alp, Bsp and Runx2 in the IGF2-treated group was
higher than that in the control group (P < 0.05). However, only the addition of BMS-754807 weak-
ened the effects of IGF2, as the addition of IGF2 and PPP enhanced the expression of osteogenic

markers similar to that in the IGF2-treated cells (Figure 6B; P < 0.05). These results imply that IGF2 enhances the osteogenic differentiation capac- ity of ADSCs mainly via the activation of IR.

Discussion
The characteristic properties of ADSCs—namely, self-renewal and multilineage differentiation—make them attractive candidates for clinical use [2,3]. However, an adequate cell number is necessary for effective application, especially in cell-based therapy. Nevertheless, with increasing age, the proliferation and multilineage differentiation capacities of ADSCs are reduced, which may largely restrict their potential therapeutic effect [6,20]. Therefore, the mainte- nance of ADSC properties during aging is of great importance and remains highly critical to address.
The stem cell niche plays an essential role in stem cell biology, and variations in the niche with age rep- resent one of the major causes of changes in stemness [21 23]. Previously, IGF2, as one of the niche fac- tors, was reported to decrease in bone, muscle and

182 C. Wang et al.

Figure 4. Effect of IGF2 receptor inhibitors on the proliferation and stemness-related marker expression of ADSCs. (A) Immunofluorescence assay (a and b) and qRT-PCR analysis (c and d) of IGF-1R and IR expression in ADSCs. Bar = 20 mm. (B) Proliferation of ADSCs was detected by MTT assays. ADSCs were treated with 100 ng/mL IGF2 in the presence of BMS-754807 (100 nmol/L) or PPP (200 nmol/L) for 48 h. (C—F) Western blot analysis of NANOG, OCT4 and SOX2 levels after IGF2 stimulation. ADSCs were treated with BMS-754807
(C and E) or PPP (D and F) for 48 h. (E and F) Quantitative measurement of protein levels. *P < 0.05.

serum with increasing age [9,10]. Moreover, IGF2, which is highly expressed in embryonic and brain tis- sues, has been demonstrated to be critical for regu- lating the self-renewal capacities of ESCs and NSCs [12 14]. IGF signaling has also been shown to regu- late the self-renewal ability of BMSCs with consecu- tive passages in vitro [24]. Accordingly, in this study, we found that IGF2 expression decreased with age in rat adipose tissues. Furthermore, we reported that

exogenous IGF2 promoted the proliferation, stem- ness-related marker expression and multilineage dif- ferentiation of ADSCs isolated from 8-week-old rats, indicating that IGF2 might be related to the age- related loss of stemness in ADSCs. In addition, we subsequently showed that the effects of IGF2 were mainly mediated via IGF-1R and IR.
ADSCs have been recently proven to be advanta- geous in tissue engineering of bone and skin because

IGF2 regulates the proliferation and differentiation of ADSCs 183

Figure 5. Adipogenic differentiation of ADSCs. (A) Lipid formation was measured by oil red O staining after 15 days of induction in vitro. ADSCs in adipogenic induction medium were treated with or without 100 ng/mL IGF2, BMS-754807 and PPP. Bar = 100 mm. (B) qRT-PCR analysis of Fabp4, Lpl and Ppar-g expression in ADSCs. Results were normalized against glyceraldehyde 3-phosphate dehydrogenase. *P < 0.05.

of their adequate quantity and low immunogenicity [25]. In fact, primary ADSCs isolated from the stro- mal vascular fraction (SVF) of adipose tissues con- tain several other types of cells, including pre- adipocytes, fibroblasts and endothelial cells [2,26]. It is now widely accepted that the expression of specific surface antigens is a distinctive characteristic of ADSCs, and these antigens can serve as markers dur- ing the purification of ADSCs. Consistent with pre- vious reports [19,26], the ADSCs used in this study were positive for CD90 and CD29 and negative for CD34 and CD45. More importantly, ADSCs showed the ability to differentiate into adipocytes, osteoblasts and chondrocytes, which are unique characteristics of MSCs. Taken together, these results indicate that the current technique for isolat- ing ADSCs is sufficient for functional assays.
Previous studies have proven that IGF2 can enhance the proliferation of ESCs, NSCs and pre- adipocytes [12,13,27]. In addition, it has been shown that the proliferation-promoting effect of IGF2 on osteoprogenitors is affected by serum con- centration in cuture medium [28]. In our study, we consistently found that the proliferation and

stemness-associated marker expression of ADSCs remained unaltered with a range of IGF2 concentra- tions in the presence of culture conditions with 10% FBS. However, with 1% FBS, the IGF2-stimulated groups showed increased optical density values as measured by the MTT assay and elevated expression of stemness-related markers. A possible explanation is that bovine serum contains a high concentration of IGFBPs and soluble IGF2/M6P receptor, which can inhibit the biological activity of IGF2 by forming inactive compounds [29,30]. Toshiyuki et al. found that the survival rate of neurons increased with the addition of 100 ng/mL or higher concentrations of IGF2 when cultured with low FBS concentrations [31]. Our data similarly showed that the proliferation of ADSCs gradually increased with increasing IGF2 concentrations in the presence of 1% FBS.
Several core transcription factors including NANOG, OCT4 and SOX2 are essential for main- taining the stemness-related properties of ADSCs. Previous studies have demonstrated that an increase in age is accompanied by a decrease in the expression of these factors, whereas the overexpression of these three factors results in enhanced proliferation and

184 C. Wang et al.

Figure 6. Osteogenic differentiation ability of ADSCs. (A) Alizarin red staining of ADSCs after culturing in osteogenic induction medium supplemented with or without 100 ng/mL IGF2, BMS-754807 and PPP for 21 days. Bar = 100 mm. (B) qRT-PCR analysis of Alp, Bsp and Runx2 expression in ADSCs. Results were normalized against glyceraldehyde 3-phosphate dehydrogenase. * P < 0.05.

multilineage differentiation [32,33]. In the present study, we found that the expression of these stem- ness-related markers was increased by IGF2 stimula- tion under culture conditions with 1% FBS. Furthermore, enhanced adipogenic and osteogenic differentiation of ADSCs was achieved by the addi- tion of IGF2, which was verified by increased oil red O and alizarin red staining. Additionally, evidence presented by other studies has demonstrated that NANOG and OCT4 are downstream effectors of IGF2 [13,34]. Hence, these data indicate that the IGF2-induced increase in the self-renewal and multi- potency of ADSCs is mediated, at least in part, via NANOG, OCT4 and SOX2 signaling.
The biological activity of IGF2 is mainly medi- ated by IGF-1R and IR. Both IR and IGF-1R have been shown to regulate the self-renewal capacity of cells, although the effects differ among various types of stromal cells [8,12,13]. In our study, we reported that ADSCs expressed both IGF-1R and IR. As pre- viously reported, IR is highly expressed in adipose tissues and contributes more to glucose metabolism, whereas IGF-1R primarily mediates mitogenic effects [8,35]. Other studies have shown that IGF- 1R is activated by IGF2 during preadipocyte differ- entiation, while IR is involved in maintaining the self-renewal of NSCs [13,28]. Our data are in line with previous results, showing that the improved proliferation and adipogenic differentiation of

ADSCs induced by IGF2 is mainly mediated by IR and IGF1R/IR, respectively. Furthermore, our find- ings showed that the IGF-1R/IR antagonist BMS- 754807 could inhibit the adipogenic potential induced not only by basal induction medium but also by the addition of IGF2. The induction medium used in this study included a high concentration of insulin, which has been reported to bind with IGF- 1R/IR to promote glucose uptake and lipogenesis during the adipogenic differentiation process [36]. Our results are consistent with other data, showing that elimination of insulin from the adipogenic induction medium reduces adipogenesis to back- ground levels [37]. We also found that IGF-1R was more potent in maintaining the expression stemness- related markers, whereas IR preferentially reversed the osteogenic differentiation of ADSCs enhanced by IGF2. In general, there are two IR isoforms, including IR-A (with exclusion of exon 11) and IR-B (with inclusion of exon 11). IR-A binds IGF2 with high affinity and has been demonstrated to regulate multipotency, whereas IR-B is responsible for cell differentiation [35]. Recent studies have found that undifferentiated placental MSCs (PMSCs) and mature osteoblasts highly express IR-B, and the IR- B:IR-A ratio increases during osteogenic differentia- tion [38,39]. Therefore, the effect of IGF2 on the osteogenic differentiation of ADSCs may mainly depend on IR, especially IR-B, which makes little

IGF2 regulates the proliferation and differentiation of ADSCs 185

contribution to the multipotency in ADSCs. How- ever, the specific mechanisms regulated by these two IR isoforms need to be further investigated in the future. In addition, the downstream signaling path- ways activated by different IGF2 receptors remain to be elucidated.
In conclusion, in this study, we report that IGF2 expression in rat adipose tissue decreases as age increases and that exogenous IGF2 supplementation can enhance the proliferation and multilineage dif- ferentiation of ADSCs, indicating that IGF2 may contribute to the age-associated decrease in the stemness of ADSCs. These findings provide new insights into methods for culturing ADSCs and may facilitate the subsequent application of ADSCs in tis- sue engineering and cell-based therapy.

Acknowledgments
This work was supported by the National Program on Key Basic Research Project of China (2011CB964701).
Disclosure of interest: The authors have no commercial, proprietary, or financial interest in the products or companies described in this article.

Author disclosure statement
The authors declare that no conflict of interest exists.

References
[1] Samsonraj RM, Raghunath M, Nurcombe V, Hui JH, van Wijnen AJ, Cool SM. Concise Review: Multifaceted Charac- terization of Human Mesenchymal Stem Cells for Use in Regenerative Medicine. Stem Cells Transl Med 2017;6: 2173–85.
[2] Wankhade UD, Shen M, Kolhe R, Fulzele S. Advances in adipose-derived stem cells isolation, characterization, and application in regenerative tissue engineering. Stem Cells Int 2016;2016:3206807.
[3] Mohamed-Ahmed S, Fristad I, Lie SA, Suliman S, Mustafa K, Vindenes H, et al. Adipose-derived and bone marrow mesenchymal stem cells: a donor-matched comparison. Stem Cell Res Ther 2018;9:168.
[4] Lee CC, Ye F, Tarantal AF. Comparison of growth and dif- ferentiation of fetal and adult rhesus monkey mesenchymal stem cells. Stem Cells Dev 2006;15:209–20.
[5] Guillot PV, Gotherstrom C, Chan J, Kurata H, Fisk NM. Human first-trimester fetal MSC express pluripotency markers and grow faster and have longer telomeres than adult MSC. Stem Cells 2007;25:646–54.
[6] Bunpetch V, Wu H, Zhang S, Ouyang H. From “bench to bedside”: current advancement on large-scale production of mesenchymal stem cells. Stem Cells Dev 2017;26:1662–73.
[7] Barreto S, Gonzalez-Vazquez A, Cameron AR, Cavanagh B, Murray DJ, O’Brien FJ. Identification of the mechanisms by which age alters the mechanosensitivity of mesenchymal stro- mal cells on substrates of differing stiffness: Implications for osteogenesis and angiogenesis. Acta Biomater 2017;53:59–69.
[8] Youssef A, Aboalola D, Han VK. The roles of insulin-like growth factors in mesenchymal stem cell niche. Stem Cells Int 2017;2017:9453108.
[9] Qiu Q, Jiang JY, Bell M, Tsang BK, Gruslin A. Activation of endoproteolytic processing of insulin-like growth factor-II in fetal, early postnatal, and pregnant rats and persistence of circu- lating levels in postnatal life. Endocrinology 2007;148:4803–11.
[10] Brown AL, Graham DE, Nissley SP, Hill DJ, Strain AJ, Rechler MM. Developmental regulation of insulin-like growth factor II mRNA in different rat tissues. J Biol Chem 1986;261:13144–50.
[11] Sullivan KA, Feldman EL. Immunohistochemical localiza- tion of insulin-like growth factor-II (IGF-II) and IGF-bind- ing protein-2 during development in the rat brain. Endocrinology 1994;135:540–7.
[12] Bendall SC, Stewart MH, Menendez P, George D, Vijayara- gavan K, Werbowetski-Ogilvie T, et al. IGF and FGF coop- eratively establish the regulatory stem cell niche of pluripotent human cells in vitro. Nature 2007;448:1015–21.
[13] Ziegler AN, Schneider JS, Qin M, Tyler WA, Pintar JE, Frai- denraich D, et al. IGF-II promotes stemness of neural restricted precursors. Stem Cells 2012;30:1265–76.
[14] Ziegler AN, Levison SW, Wood TL. Insulin and IGF recep- tor signalling in neural-stem-cell homeostasis. Nat Rev Endo- crinol 2015;11:161–70.
[15] Li Z, Fei T, Zhang J, Zhu G, Wang L, Lu D, et al. BMP4 Signaling Acts via dual-specificity phosphatase 9 to control ERK activity in mouse embryonic stem cells. Cell Stem Cell 2012;10:171–82.
[16] Watanabe S, Umehara H, Murayama K, Okabe M, Kimura T, Nakano T. Activation of Akt signaling is sufficient to maintain pluripotency in mouse and primate embryonic stem cells. Oncogene 2006;25:2697–707.
[17] Baxter RC, Twigg SM. Actions of IGF binding proteins and related proteins in adipose tissue. Trends Endocrinol Metab 2009;20:499–505.
[18] Wang L, Schulz TC, Sherrer ES, Dauphin DS, Shin S, Nel- son AM, et al. Self-renewal of human embryonic stem cells requires insulin-like growth factor-1 receptor and ERBB2 receptor signaling. Blood 2007;110:4111–9.
[19] Dominici M, Le BK, Mueller I, Slaper-Cortenbach I, Marini F, Krause D, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 2006;8: 315–7.
[20] Mare˛dziak M, Marycz K, Tomaszewski KA, Kornicka K, Henry BM. The influence of aging on the regenerative poten- tial of human adipose derived mesenchymal stem cells. Stem Cells Int 2016;2016:2152435.
[21] Oh J, Lee YD, Wagers AJ. Stem cell aging: mechanisms, reg- ulators and therapeutic opportunities. Nat Med 2014;20: 870–80.
[22] Lynch K, Pei M. Age associated communication between cells and matrix: a potential impact on stem cell-based tissue regeneration strategies. Organogenesis 2014;10:289–98.
[23] Morrison SJ, Spradling AC. Stem cells and niches: mecha- nisms that promote stem cell maintenance throughout life. Cell 2008;132:598–611.
[24] Schilling T, Schonefeldt C, Zeck S, Meissner-Weigl J, Schneider D, Jakob F, et al. Local expression of insulin-like growth factor (IGF) signaling components during in vitro aging of mesenchymal stem cells. J Stem Cells Regen Med 2007;2:41–2.
[25] Bacakova L, Zarubova J, Travnickova M, Musilkova J, Pajor- ova J, Slepicka P, et al. Stem cells: their source, potency and use in regenerative therapies with focus on adipose-derived stem cells—a review. Biotechnol Adv 2018;36:1111–26.
186 C. Wang et al.

[26] Bourin P, Bunnell BA, Casteilla L, Dominici M, Katz AJ, March KL, et al. Stromal cells from the adipose tissue- derived stromal vascular fraction and culture expanded adi- pose tissue-derived stromal/stem cells: a joint statement of the International Federation for Adipose Therapeutics and Science (IFATS) and the International Society for Cellular Therapy (ISCT). Cytotherapy 2013;15:641–8.
[27] B€ack K, Br€annmark C, Stra◦ lfors P, Arnqvist HJ. Differential effects of IGF-I, IGF-II and insulin in human preadipocytes
and adipocytes—role of insulin and IGF-I receptors. Mol Cell Endocrinol 2011;339:130–5.
[28] Jia D, Heersche JN. Insulin-like growth factor-1 and -2 stim- ulate osteoprogenitor proliferation and differentiation and adipocyte formation in cell populations derived from adult rat bone. Bone 2000;27:785–94.
[29] Tu C, Fiandalo MV, Pop E, Stocking JJ, Azabdaftari G, Li J, et al. Proteomic analysis of charcoal-stripped fetal bovine serum reveals changes in the insulin-like growth factor signal- ing pathway. J Proteome Res 2018. https://doi.org/10.1021/ acs.jproteome.8b00135.
[30] Honegger A, Humbel RE. Insulin-like growth factors I and II in fetal and adult bovine serum. Purification, primary struc- tures, and immunological cross-reactivities. J Biol Chem 1986;261:569–75.
[31] Motoike T, Unsicker K. Identification of a potent neurotrophic substance for ciliary ganglionic neurons in fetal calf serum as insulin-like growth factor II. J Neurosci Res 1999;56:386–96.
[32] Han SM, Han SH, Coh YR, Jang G, Chan RJ, Kang SK, et al. Enhanced proliferation and differentiation of Oct4- and SoX2-overexpressing human adipose tissue mesenchymal stem cells. EXp Mol Med 2014;46:e101.
[33]
Fan J, Sun Z. The antiaging gene Klotho regulates prolifera- tion and differentiation of adipose-derived stem cells. Stem Cells 2016;34:1615–25.
[34] Yao C, Su L, Shan J, Zhu C, Liu L, Liu C, et al. IGF/ STAT3/NANOG/Slug Signaling axis simultaneously con- trols epithelial-mesenchymal transition and stemness mainte- nance in colorectal cancer. Stem Cells 2016;34:820–31.
[35] Belfiore A, Malaguarnera R, Vella V, Lawrence MC, Sciacca L, Frasca F, et al. Insulin receptor isoforms in physiology and disease: an updated view. Endocr Rev 2017;38:379–431.
[36] Nahum N, Forti E, Aksanov O, Birk R. Insulin regulates Bbs4 during adipogenesis. IUBMB Life 2017;69:489–99.
[37] Han H, Wei W, Chu W, Liu K, Tian Y, Jiang Z, et al. Muscle Conditional Medium reduces intramuscular adipocyte differ- entiation and lipid accumulation through regulating insulin signaling. Int J Mol Sci 2017;18.
[38] Youssef A, VKM H. Regulation of osteogenic differentiation of Picropodophyllin placental-derived mesenchymal stem cells by insulin-like growth factors and low oXygen tension. Stem Cells Int 2017;2017:4576327.
[39] Avnet S, Perut F, Salerno M, Sciacca L, Baldini N. Insulin receptor isoforms are differently expressed during human osteoblastogenesis. Differentiation 2012;83:242–8.

Supplementary materials
Supplementary material associated with this article can be found in the online version at doi:10.1016/j. jcyt.2018.11.010.