Beneit et al. Cardiovasc Diabetol (2016) 15:161 DOI 10.1186/s12933-016-0477-3 Cardiovascular Diabetology ORIGINAL INVESTIGATION Open Access Expression of insulin receptor (IR) A and B isoforms, IGF‑IR, and IR/IGF‑IR hybrid receptors in vascular smooth muscle cells and their role in cell migration in atherosclerosis N. Beneit1,2,3, C. E. Fernández‑García4, J. L. Martín‑Ventura4, L. Perdomo1,2,3, Ó. Escribano1,2,3, J. B. Michel5, G. García‑Gómez1,2,3, S. Fernández1,2,3, S. Díaz‑Castroverde1,2,3, J. Egido3,4, A. Gómez‑Hernández1,2,3*† and M. Benito1,2,3† Abstract Background: Abnormal proliferation and migration of vascular smooth muscle cells (VSMCs) is a major contributor to the development of atherosclerotic process. In a previous work, we demonstrated that the insulin receptor isoform A (IRA) and its association with the insulin-like growth factor-I receptor (IGF-IR) confer a proliferative advantage to VSMCs. However, the role of IR and IGF-IR in VSMC migration remains poorly understood. Methods: Wound healing assays were performed in VSMCs bearing IR (IRLoxP+/+ VSMCs), or not (IR−/− VSMCs), expressing IRA (IRA VSMCs) or expressing IRB (IRB VSMCs). To study the role of IR isoforms and IGF-IR in experimental atherosclerosis, we used ApoE−/− mice at 8, 12, 18 and 24 weeks of age. Finally, we analyzed the mRNA expression of total IR, IRB isoform, IGF-IR and IGFs by qRT-PCR in the medial layer of human aortas. Results: IGF-I strongly induced migration of the four cell lines through IGF-IR. In contrast, insulin and IGF-II only caused a significant increase of IRA VSMC migration which might be favored by the formation of IRA/IGF-IR receptors. Additionally, a specific IGF-IR inhibitor, picropodophyllin, completely abolished insulin- and IGF-II-induced migration in IRB, but not in IRA VSMCs. A significant increase of IRA and IGF-IR, and VSMC migration were observed in fibrous plaques from 24-week-old ApoE−/− mice. Finally, we observed a marked increase of IGF-IR, IGF-I and IGF-II in media from fatty streaks as compared with both healthy aortas and fibrolipidic lesions, favoring the ability of medial VSMCs to migrate into the intima. Conclusions: Our data suggest that overexpression of IGF-IR or IRA isoform, as homodimers or as part of IRA/IGF-IR hybrid receptors, confers a stronger migratory capability to VSMCs as might occur in early stages of atherosclerotic process. Keywords: Atherosclerosis, Insulin receptor, Migration, Vascular smooth muscle cells Background streaks to more advanced plaques is a complex process Atherosclerosis is the leading cause of mortality world- [1] where vascular smooth muscle cells (VSMCs) plays wide. The progression of vascular lesions from early fatty a main role. The presence of a large number of intimal VSMCs has been taken as evidence that VSMC migration *Correspondence: algomezh@ucm.es from the media plays an important role in early stages of † A Gómez-Hernández and M Benito are co-senior authors atherogenesis [2]. These VSMCs that migrated into the 1 Biochemistry and Molecular Biology II Department, School intima exhibit an abnormally increased proliferation and of Pharmacy, Complutense University of Madrid, Plaza Ramón y Cajal s/n, 28040 Madrid, Spain extracellular matrix production, leading to the formation Full list of author information is available at the end of the article of the fibrous cap in atherosclerotic lesions [3]. Growth © The Author(s) 2016. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 2 of 14 factors, including insulin-like growth factors (IGFs), have associate with IGF-IR favoring atherogenic actions of been implicated in the regulation of VSMC migration [4, IGF-II [21]. However, the specific role of each IR isoform 5]. in VSMC migration remains unknown. We therefore ana- The insulin and IGFs (IGF-I and IGF-II) signaling lyzed, in the current study, the effect of insulin or IGFs on is mediated by hormone interaction with the insulin the migration of murine aortic VSMC lines, as well as the receptor (IR) and the IGF-I receptor (IGF-IR) which are contribution to this process of IR isoforms, IGF-IR and members of subclass II of the tyrosine kinase receptor hybrid receptors (IRA/IGF-IR and IRB/IGF-IR). Subse- super-family [6, 7]. Both receptors are expressed at the quently, we wondered whether some of the mechanisms cellular surface as preformed disulfide-linked dimers in studied in vitro could be of any in vivo relevance. For this α2β2 configuration. The extracellular α subunit of each purpose, we firstly analyzed IRA, IRB and IGF-IR mRNA hemireceptor contains the ligand binding sites, while β expression by qRT-PCR in aorta from ApoE−/− mice at subunits include a large cytoplasmic region with tyrosine 8, 12, 18 and 24 weeks of age. Then, we studied compara- kinase activity [8]. Because of the high degree of homol- tively IR, IGF-IR and α-smooth muscle actin (α-SMA) ogy of the two receptors, hybrid receptors formed by an expression in aortic roots from 8- and 24-week-old IR αβ-hemireceptor and an IGF-IR αβ-hemireceptor are ApoE−/− mice. Finally, we analyzed the mRNA expres- also found in cells co-expressing IR and IGF-IR [9–11]. sion of total IR, specifically IRB isoform, IGF-IR and IGFs Alternative splicing of the IR gene gives rise two iso- in the medial layer of human aortic segments. Our results forms, IRA and IRB [12], whereas there is only a single strongly suggest that the overexpression of IGF-IR or IRA isoform of IGF-IR. Indeed, IRB differs from IRA by the isoform, as homodimers or as part of IRA/IGF-IR hybrid inclusion of exon 11 which encodes a 12-amino acid receptors, confers a migratory advantage to VSMCs that sequence at the C-terminus of the IR α-subunit. The IR might be of relevance in early stages of atherosclerotic isoforms show different functional features. Although process. both isoforms have similar affinity for insulin, IRA exhib- its a higher affinity for IGFs, especially for IGF-II [13], Methods as well as a greater internalization and recycling rate Tissue samples than IRB [14]. Because of these differences, IRB is pref- Twenty-eight human infradiaphragmatic aortic seg- erentially associated with metabolic and differentiating ments (medial layers of 8 healthy, 9 fatty streaks and signals. Conversely, IRA mainly favors cell growth, prolif- 11 fibrolipidic lesions) each harvested from a different eration, and survival. In 32D cells, IRA induces mitogenic donor after organ transplantation with the authorization and antiapoptotic signals in response to IGF-II, whereas of the French Biomedicine Agency (authorization num- IRB tends to send differentiation signals [15]. In mouse ber PFS09-007), were included in this study. There were beta cell lines, IRA confers a stronger proliferative capac- no significant differences in terms of age and gender. The ity favoring the mitogenic effects of IGF-I and increasing investigation conforms to the principles outlined in the glucose uptake [16], and it also might provide an explana- Declaration of Helsinki. tion for the beta cell hyperplasia induced by liver insulin resistance in iLIRKO mice [17]. Additionally, long-term Experimental model AAV-mediated hepatic expression of IRA, but not IRB, Male mice were maintained in the Animal Care Facility improves glucose homeostasis in iLIRKO mice, preclud- under the standard conditions of temperature and 12 h ing beta cell mass expansion and, therefore, avoiding the light/dark cycle. All animals used are under C57BL/6 final beta cell failure [18]. IRA is also more efficient than genetic background. Six week-old male ApoE−/− knock- IRB at increasing glycogen synthesis, glycogen synthase out mice and their control (C57BL/6 mice) were fed a activity and glycogen storage in hepatocytes and in vivo Western type diet (A04+ 21% kcal from fat) for 2, 6, 12 in the liver [19]. Furthermore, IRA has been reported to or 18 weeks respectively. All animal experimentation be the predominant IR isoform expressed in cancer cells, described in this manuscript was conducted accord- such as in a variety of carcinomas or breast cancer cell ing with accepted standards of human animal care, as lines [11, 13, 20]. approved by the corresponding institutional committee. A further understanding of the molecular mechanisms The investigation also conforms to the Guide for the Care involved in early atherosclerosis is critical for identifying and Use of Laboratory Animals published by the National strategies to limit disease progression before it leads to Institutes of Health (NIH Publication No. 85–23, revised clinical consequences. In this regard, we previously dem- 1996) and in accordance with The ARRIVE Guideline for onstrated that IRA, but not IRB, confers a proliferative Reporting Animal research. advantage to VSMCs in response to several proathero- Plasma levels of insulin were analyzed using ELISA kits genic stimuli. Additionally, we found that IRA might (Millipore). Cholesterol and triglycerides were tested in Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 3 of 14 plasma samples from fasted mice (Spinreact, Barcelona, TScratch program) the percentage of wound healing clo- Spain). Blood glucose level was determined in fasted ani- sure at the different time points in relation to time 0. mals using an automatic monitor (Roche Molecular Bio- chemicals GmbH, Mannheim, Germany). RNA extraction and real‑time quantitative polymerase chain reaction Cell cultures Total RNA was isolated from tissue samples using TRI- Generation of immortalized IRLoxP+/+, IR−/−, IRA and zol reagent (Invitrogen, Carlsbad, CA, USA) and quanti- IRB VSMC lines was previously described [17]. Briefly, fied by absorbance at 260 nm. One microgram of RNA primary VSMCs were obtained from thoracic aorta arter- was used to perform the reverse transcription with a High ies of 3 male 8-week-old IRLoxP+/+ mice. Anesthetized Capacity cDNA Archive kit (Applied Biosystems, Foster mice (Avertin, 250 mg/kg, ip) were saline perfused and City, CA, USA). Real-time quantitative PCR (qRT-PCR) thoracic aorta arteries were submitted to collagenase was performed on an ABI Prism 7500 sequence detection dispersion and primary culture. Then, primary culture PCR system (Applied Biosystems) according to the manu- of IRLoxP+/+ VSMCs were immortalized by transfection facturer’s protocol, using the ΔΔCt method as previously with pBabe retroviral vector encoding SV40 Large T anti- described [17]. Quantification of IGF-IR, and IRA and IRB gen and selected with 1 μg/mL puromycin for 3 weeks. isoform mRNA levels in aorta from experimental model Immortalized IRLoxP+/+ VSMCs were infected with were performed by amplification of cDNA with TaqMan adenoviruses encoding Cre recombinase to obtain IR−/− probes. The quantification of human IR, IRB isoform, IGF- fication of cDNA using SYBR® Premix Ex Taq™ (Takara VSMCs. Finally, IR−/− VSMCs were transfected with IR, IGF-I and IGF-II mRNA levels was performed by ampli- pBABE retroviral vector encoding the individual spliced isoforms of the human IR, IRA or IRB, and selected with Bio Inc., Otsu, Japan). The mRNA levels of target genes 200 μg/mL hygromycin for 2 weeks to obtain IRA or IRB were normalized to 18S mRNA content. Thus, the amount VSMCs respectively. of target, normalized to 18S and relative to the control, is Four cell lines were cultured to subconfluence (70– given by real-time quantitative (RQ) = 2−ΔΔCt; ΔCt (cycle 80%) with 10% (fetal bovine serum) FBS-DMEM for threshold) = Ct (target gene) − Ct (18S); ΔΔCt = ΔCt for signaling studies and were serum and glucose starved any sample − ΔCt for the control. Sequences of primers are for 4–5 h and then stimulated with insulin (10 or the following: IR: forward primer: 5´-CGAGAAGACCA 100 nmol/L, Sigma-Aldrich Corp.), IGF-I (10 nmol/L, TCGACTCGG-3´ and reverse primer 5´-GACACCA- Millipore) or IGF-II (10 nmol/L, Millipore). GAGCGTAGGATCG-3´; IRB: forward primer: 5´-GAGG ATTACCTGCACAACGTG-3´ and reverse primer 5´-TAG Western blot analysis GGTCCTCGGCACCAG-3´; IGF-IR: forward primer: Western blot analyses were performed on protein 5´-ATGCGGTGTCCAATAACTAC-3´ and reverse primer extracts from VSMCs as previously described [17]. The 5´-TTGTTGATGGTGGTCTTCTC-3´; IGF-I: forward antibodies used were anti-IRβ and anti-IGF-IRβ from primer: 5´-TTTCAACAAGCCCACAGGGT-3´ and reverse Santa Cruz Biotechnology (Dallas, TX, USA); anti-β- primer 5´-TTGAGGGGTGCGCAATACAT-3´;IGF-II: for- actin from Sigma-Aldrich Corp. (St. Louis, MO); anti- ward primer: 5´-GTCATGGCAGACGCCACATT-3´ and phospho-Akt (Thr308), anti-phospho-p70 S6 Kinase reverse primer 5´-CGAAGGCTCTGCCCTTCTTA-3´; and (Ser389) and anti-phospho-p44/42 MAPK (Thr202/ Housekeeping gene as 18S: forward primer: 5´-CCGTCG- Tyr204) from Cell Signaling (Danvers, MA, USA). TAGTTCCGACCATAA-3´ and reverse primer 5´-CAGCTTTGCAACCATACTCCC-3´. Wound healing assays Cells were cultured to confluence in 10% FBS-DMEM Histological analysis and serum deprived for 18 h. Then, the confluent mon- Aortic roots were OCT-embedded and sections of 7 μm olayer was scratched with a sterile pipette tip and cells interval were Oil-Red-O/hematoxylin stained to meas- were washed twice with PBS and fresh medium was ure lipid depot. The lesion size on aortic root was meas- added. Cells were stimulated with insulin (10 nmol/L), ured as described [21]. IR and IGF-IR were detected by IGF-I (10 nmol/L) or IGF-II (10 nmol/L) and maintained immunoperoxidase with rabbit anti-IRβ (sc-711) and for 24 h at 37 °C and 5% CO2. Migration was followed by anti-IGF-IRβ (sc-713) polyclonal antibodies. We also Cy3™ antibody, mouse monoclonal (C6198, Sigma- phase-contrast microscopy (Eclipse TE300 Nikon micro- measured α-SMA with Anti-Actin, α-Smooth Muscle- scope coupled to a digital sight DS-U2 camera) at differ- ent time points (0, 6, 12 and 24 h) up to wound healing Aldrich) to localize smooth muscle cells in aortic roots closure. Photographs were taken to quantify (using and check whether such cells had migrated from media Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 4 of 14 to intima. To localize smooth muscle cells with positive IR−/− VSMCs responded to IGFs, but did not to insu- staining for IR or IGF-IR, we performed immunofluores- lin, in regards to Akt activation. However, a significant cence against IR or IGF-IR using anti-rabbit FITC (green increase of p-Akt, p-p70S6 K, and p-p44/42 MAPK Actin, α-Smooth Muscle-Cy3™ antibody (red staining) staining) followed by immunofluorescence against Anti- was induced by the three stimuli in IRLoxP+/+ VSMCs (Additional file 1: Figure S1B). More importantly, IRA and finally, incubation with DAPI to stain nuclei (blue VSMCs showed a significantly higher activation of Akt, staining). p70S6 K and p44/42 MAPK in response to insulin, IGF-I or IGF-II, than in IRB VSMCs (Additional file 1: Gelatin zymography Figure S1B). To determine MMP-2 and MMP-9 activities, 80% con- fluent cells were serum-deprived for 12–24 h and the Differential contribution of IRA, IRB and IGF‑IR to VSMC culture medium of different experimental conditions migration in response to insulin or IGFs was used for an electrophoresis in 10% SDS–polyacryla- Proliferation and migration of VSMCs contribute to mide gels polymerized in the presence of 0.1% gelatin atherosclerotic process. In this regard, we previously (Sigma-Aldrich) under non-reducing conditions. Gels demonstrated that IRA isoform confers a proliferative were washed with 2.5% Triton X-100 for 30 min to advantage to those cells [17]. Then, we hypothesized remove SDS, rinsed with substrate buffer (0.2 M NaCl, that a differential role of IR isoforms might also occur 5 mM CaCl2, 1% Triton X-100, 0.02% NaN3, 50 mM Tris in the migration of VSMCs. To address this issue, we pH 7.5) and incubated in this buffer at 37 °C overnight to measured the migration rate of four VSMC cell lines at allow protein renaturation and MMP activation. To visu- 6, 12 and 24 h by wound healing assays. Maximal migra- alize gelatin degradation, gels were stained with Coomas- tion was observed in IRLoxP+/+ VSMCs as compared sie Brilliant Blue R-250 (Bio-Rad) for 60 min. MMP-2 and with IR−/− and IRB VSMCs. IRA VSMCs showed a sig- MMP-9 activities were quantified by Image J Program. nificantly higher migration than IRB VSMCs (Fig. 1a). Moreover, we assessed the effect of insulin or IGFs stimu- Statistical analysis lation on VSMC migration. Firstly, we found a significant All values are expressed as mean ± SEM. Differences increase of insulin-induced migration in IRA VSMCs at between two groups were assessed using unpaired two- 12 and 24 h after wounding. IRLoxP+/+ and IRB VSMCs tailed t-tests. Data involving more than two groups were exhibited a slightly response to insulin and no effect analyzed using a one-way ANOVA followed by a Bonfer- was observed in IR−/− VSMCs (Fig. 1b). IGF-I strongly roni test if differences were noted (GraphPad Prism). The induced migration of all cell lines studied with significant null hypothesis was rejected when p < 0.05. increases in the percentage of wound closure at 12 and 24 h (Fig. 2a). Given the fact that IGF-I enhanced migra- Results tion of cells lacking IR, the effect exerted by IGF-I might Differential insulin and IGFs signaling by IRA or IRB be mediated through IGF-IR. In contrast, IGF-II showed isoforms in VSMCs a discrete effect on VSMC migration. However, IGF-II To study the role of IR isoforms in the migration of vas- induced IRA VSMC migration significantly faster than cular smooth muscle cells four murine aortic VSMC lines IRB VSMCs at 12 h (Fig. 2b). These results suggest a dif- were used: bearing IR (IRLoxP+/+ VSMCs), lacking IR ferential contribution of IR and IGF-IR to VSMC migra- (IR−/− VSMCs), specifically expressing IRA isoform (IRA tion, the latter being the main mediator of IGF-I-induced VSMCs) or alternatively expressing IRB isoform (IRB migration. Conversely, insulin and IGF-II seem to require VSMCs). Western blot analysis of IRβ confirmed deletion IR, and more specifically IRA isoform, to induce migra- in IR−/− VSMCs and expression of IR in reconstituted tion of VSMCs. IRA and IRB VSMC lines (Additional file 1: Figure S1A). One of the mechanisms that may also partially explain Regarding IGF-IR, a significantly higher expression the differential migration between IRA and IRB VSMCs was found in IR−/− and IRB VSMCs compared to is a differential activation of certain matrix metallopro- IRLoxP+/+ VSMCs, while cells exclusively expressing IRA teinases (MMPs), specifically the gelatinases such as showed lower level of IGF-IR protein (Additional file 1: MMP-2 and MMP-9, which have been involved in VSMC Figure S1A). migration [22–24]. Thus, we observed by gelatin zymog- We studied the signaling pathways induced by insulin raphy that MMP-2 activity was much higher in IRA than or IGFs. Cells were stimulated with 10 nmol/L insulin, IRB VSMCs under basal conditions or upon stimulation IGF-I or IGF-II for 10 min and phosphorylation of Akt, with insulin, IGF-I or IGF-II for 12 and 24 h (Additional p70S6 K and p44/42 MAPK was analyzed. As expected, file 2: Figure S2). Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 5 of 14 IRLoxP+/+ IR -/- IRA IRB a 0h † ### 100 IRLoxP+/+ # IR-/- † 80 IRA ### (% wound closure) 6h VSMC Migration IRB ## 60 † ### ### 12 h 40 20 24 h 0 100 µm 100 µm 100 µm 100 µm 6h 12 h 24 h Control b IRLoxP+/+ IR -/- IRA IRB 120 Control †# *# 0h Insulin, 10 nmol/L # 100 †# * ## (% wound closure) 80 VSMC Migration 6h 60 † ## ## 40 12 h 20 0 +/+ -/- A B +/+ - /- A B +/+ - /- A B 24 h xP IR IR IR xP IR IR IR xP IR IR IR Lo Lo Lo IR IR IR 6h 12 h 24 h Insulin, 10 nmol/L Fig. 1 Insulin induces migration in IRA VSMCs. a Analysis of basal migration of four VSMC lines at 6, 12 and 24 h. The experiments were performed by wound healing assays and the percentage of wound healing closure quantified (using TScratch program) at different time points. Experiments were performed 4–7 times. b Effect of insulin on wound healing closure in four VSMC lines at 6, 12 and 24 h. Photomicrographs (10x magnification) were selected the more representative. Experiments were performed at least 3 times. *p < 0.05 vs. each control; †p < 0.05 vs. IR−/− VSMCs; #p < 0.05, ## p < 0.005, ###p < 0.001 vs. IRB VSMCs Inhibition of IGF‑IR strongly reduces VSMC migration migration of IRLoxP+/+ or IRA VSMCs was partially To further investigate the role of IGF-IR in VSMC migra- inhibited (Fig. 4a), suggesting that IRB, but not IRA, tion, we utilized a highly specific inhibitor of IGF-IR requires the presence of IGF-IR to induce migration in tyrosine phosphorylation, picropodophyllin (PPP), with- response to insulin. Regarding IGF-I-mediated migra- out affecting IR. A PPP dose–response curve revealed tion, a reduction of approximately one half was the result a complete inhibition of IGF-IR phosphorylation at 5 of IGF-IR inhibition in all cell lines analyzed and no sig- and 10 µmol/L in IRLoxP+/+ VSMCs upon stimulation nificant differences were noted between IRA and IRB with IGF-I. Similarly, 5 µmol/L PPP completely avoided VSMCs (Fig. 4b). Finally, PPP caused a much higher inhi- IGF-IR tyrosine phosphorylation in all others cell lines bition of IGF-II-induced migration in IRB VSMCs than (Fig. 3a). Therefore, cells were treated with 5 µmol/L PPP that observed in IRLoxP+/+ or IRA VSMCs (Fig. 4c). for 12 h to address migration experiments. We found a notable and significant decrease of VSMC migration Insulin and IGF‑II induces the formation of IRA/IGF‑IR induced by PPP treatment, reaching roughly 50% of inhi- hybrid receptors in VSMCs bition in all the four cell lines migration (Fig. 3b). These To address the issue of hybrid receptor formation, lysates data indicate that IGF-IR mostly contributes to VSMC from IRLoxP+/+, or IRA, or IRB VSMCs, were immuno- migration under basal conditions. precipitated with anti-IRβ antibody followed by Western A step further, we studied the effect of IGF-IR inhibi- blot against IGF-IR. All cell lines showed a basal associa- tion on the VSMC migration stimulated by insulin or tion between IR and IGF-IR. Nevertheless, expression of IGFs. Interestingly, PPP treatment completely abolished IRA/IGF-IR hybrid receptors was significantly increased insulin-induced migration in IRB VSMCs, whereas only in IRA VSMCs upon stimulation with insulin or Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 6 of 14 a IRLoxP+/+ IR -/- IRA IRB 120 Control IGF-I, 10 nmol/L * * * 0h *# *# 100 *# 80 (% wound closure) # VSMC Migration # 6h 60 # 40 12 h 20 0 +/+ - /- A B +/+ - /- A B +/+ -/- A B 24 h xP IR IR IR xP IR IR IR xP IR IR IR Lo Lo Lo 100 µm 100 µm 100 µm 100 µm IR IR IR IGF-I, 10 nmol/L 6h 12 h 24 h b IRLoxP+/+ IR -/- IRA IRB 120 Control †† # 0h IGF-II, 10 nmol/L * 100 *† 80 †# ## (% wound closure) VSMC Migration 6h 60 †# 40 12 h 20 0 +/+ - /- A B +/+ -/- A B +/+ - /- A B 24 h xP IR IR IR xP IR IR IR xP IR IR IR Lo Lo Lo 100 µm 100 µm 100 µm 100 µm IR IR IR 6h 12 h 24 h IGF-II, 10 nmol/L Fig. 2 IGF-I strongly induces VSMC migration and IGF-II has a lesser effect on IRA VSMC migration. Effect of IGF-I (a) or IGF-II (b) on wound healing closure in four lines of VSMCs at 6, 12 and 24 h. Photomicrographs (10x magnification) were selected the more representative. Experiments were performed at least 3 times. *p < 0.05 vs. each control; †p < 0.05, ††p < 0.005 vs. IR−/− VSMCs; #p < 0.05, ##p < 0.005 vs. IRB VSMCs IGF-II, but not with IGF-I, for 12 h (Fig. 4d). In addition, ApoE−/− mice. However, IRB was diminished in aorta expression of both IR and IGF-IR was measured in total from ApoE−/− mice at all weeks of age (Fig. 5a). After protein extracts and detected an increase of IRA isoform, that, we studied the lesion area and lipid deposits with an equal expression of IGF-IR, in response to insulin together with IR and IGF-IR protein levels only in aortic or IGF-II (Additional file 3: Figure S3). These findings are roots from 8- and 24-week-old ApoE−/− mice and their consistent with the greater migration observed in IRA respective controls, because it was previously described VSMCs upon stimulation with insulin or IGF-II for 12 h that at 15–20 weeks of age, ApoE−/− mice might have (Figs. 1b, 2b). fibrous plaques in which the presence of a fibrous cap containing VSMCs would indicate their migration from Role of IR isoforms and IGF‑IR in VSMC migration in an the media [25]. A progression of atherosclerosis was experimental model of atherosclerosis observed by OilRedO staining as a progressive increase To assess the role of IR isoforms and IGF-IR in experi- of lesion area, stenosis and lipid deposits in aortic roots mental atherosclerosis, we used ApoE−/− mice at 8, 12, from 24-week-old ApoE−/− vs. 8-week-old ApoE−/− 18 and 24 weeks of age. This model showed hypercho- mice (Fig. 5b). By immunohistochemistry against total lesterolemia and hypertriglyceridemia at all weeks of age IR, we found a significant decrease of IR in aortic roots without alterations in body weight, glucose or insulin from 8-week-old ApoE−/− mice (Fig. 5b), consistent with plasma levels (see Table 1). the decreased mRNA expression of both IRA and IRB Firstly, we studied the mRNA expression of IR isoforms isoforms in aorta from ApoE−/− mice at 8 weeks of age and IGF-IR by qRT-PCR in aorta. A significant increase (Fig. 5a). In aortic roots from 24-week-old ApoE−/− mice, of IRA and IGF-IR mRNA was noted in 24-week-old protein levels of total IR were similar to their controls Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 7 of 14 a IRLoxP+/+ IR -/- IRA IRB IGF-I (5 min, 10 nmol/L): - + + + + + - + + + - + + + - + + + PPP (1 h, µmol/L): - - 1 2 5 10 - - 1 5 - - 1 5 - - 1 5 IP: IGF-IRβ WB: p-Tyr IP: IGF-IRβ WB: IGF-IRβ IgG heavy chain b 80 Control IRLoxP+/+ IR-/- IRA IRB 70 PPP (5 µmol/L, 12h) (% wound closure) 60 VSMC Migration 0h 50 40 *** ** Control 30 ** 20 * 12 h 10 100 µm 100 µm 100 µm 100 µm 0 IRLoxP+/+ IR -/- IRA IRB 0h 70 Inhibition of VSMC Migration (%) (PPP 5µmol/L, 12h) PPP 60 5 µmol/L 50 12 h 40 100 µm 100 µm 100 µm 100 µm 30 20 10 0 IRLoxP+/+ IR-/- IRA IRB Fig. 3 IGF-IR contributes to basal VSMC migration. a Effect of PPP (IGF-IR inhibitor) at different doses on IGF-IR tyrosine phosphorylation of four VSMC lines. We performed the immunoprecipitation against IGF-IRβ followed by Western blot against p-tyrosine. b Effect of PPP on basal VSMC migration at 12 h by wound healing closure and the percentage of inhibition of VSMC migration induced by PPP. Experiments were performed 4–7 times. *p < 0.05, **p < 0.005, ***p < 0.001 vs. each control (Fig. 5b), and this can be correlated with the increased IGF-IR and α-SMA (Fig. 6b), we found that those VSMCs expression of IRA at mRNA level, together with the present in fibrous plaques from 24-week-old ApoE−/− decreased expression of IRB mRNA (Fig. 5a). In addi- mice expressed IR or IGF-IR (see arrows and merged of tion to the increased IGF-IR mRNA levels, a significant Fig. 6b). increase of IGF-IR protein was observed in aortic roots from ApoE−/− mice at 24 weeks of age (Fig. 5a, b). How- mRNA expression of IGF‑IR and IGFs in human aortic ever, besides VSMCs, many inflammatory cells can be medial layers present in those plaques and they might also express IR Based in our in vitro and in vivo results and those pre- or IGF-IR. viously published in ApoE−/− or BATIRKO mice, two In order to localize VSMCs in aortic roots, we checked models showing vascular damage [21], we hypothesized by immunofluorescence the expression of α-SMA, a that IR and its isoforms and IGF-IR could play a differ- well-known marker of smooth muscle cells. Very simi- ential role in human early atherosclerosis. For this pur- lar levels of α-SMA were found in aortic roots from pose, we analyzed the mRNA expression of total IR, ApoE−/− mice at 8 and 24 weeks of age in relation to their specifically IRB isoform, IGF-IR and IGFs by qRT-PCR respective controls (Fig. 6a). We also observed medial in the medial layer of human aortas (Additional file 4: VSMCs migrated into the intima of fibrous plaques from Figure S4). Among 28 aortic samples, 8 were devoid 24-week-old ApoE−/− mice (see white arrows of Fig. 6a). of grossly visible signs of lesions and were classified as Moreover, by double immunofluorescence against IR or healthy, whereas all other samples showed macroscopic Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 8 of 14 a 40 b 50 IGF-I (10 nmol/L, 12h) Ins (10 nmol/L, 12h) (% increase vs. control) 40 IGF-I (10 nmol/L, 12h) + PPP Ins (10 nmol/L, 12h) + PPP (% increase vs. control) 30 VSMC Migration VSMC Migration 30 20 20 *# * ## 10 # 10 0 0 IRLoxP +/+ IRA IRB IRLoxP+/+ IRA IRB -10 IGF-II, 10 nmol/L IGF-I, 10 nmol/L Ins, 10 nmol/L d c Control IGF-II (10 nmol/L, 12h) 40 IGF-II (10 nmol/L, 12h) + PPP IP: IRβ (% increase vs. control) WB: IGF-IRβ 30 IRLoxP+/+ VSMC Migration IgG heavy Control chain Ins (10 nmol/L, 12h) 20 IGF-I (10 nmol/L, 12h) (IP:IRβ)(N-Fold vs. Control) IGF-II (10 nmol/L, 12h) IGF-IR;I Rβ/IgG heavy ch. 10 IP: IRβ * WB: IGF-IRβ *** ** 0 IRA IgG heavy chain -10 IRLoxP+/+ IRA IRB * * IP: IRβ WB: IGF-IRβ IRB IgG heavy IRLoxP+/+ IRA IRB chain 12 h Fig. 4 Effect of IGF-IR inhibition on VSMC migration induced by insulin or IGFs. Quantification of PPP effect on VSMC migration stimulated with insulin (a), IGF-I (b) or IGF-II (c) at 12 h by wound healing closure. *p < 0.05 vs. each control; #p < 0.05, ##p < 0.005 vs. IRB VSMCs. d Formation of hybrid receptors (IRA/IGF-IR or IRB/IGF-IR) in IRLoxP+/+, IRA and IRB VSMCs stimulated with insulin, IGF-I or IGF-II. We performed the immunopre‑ cipitation against IRβ followed by Western blot against IGF-IRβ. Experiments were performed at least 3 times. *p < 0.05, **p < 0.005, ***p < 0.001 vs. each control Table 1 Characterization of an experimental model of atherosclerosis 8 weeks 12 weeks 18 weeks 24 weeks −/− −/− −/− Control ApoE Control ApoE Control ApoE Control ApoE−/− (n = 5) (n = 5) (n = 8) (n = 8) (n = 8) (n = 8) (n = 8) (n = 8) Body weight 20.4 ± 0.1 21.3 ± 0.6 24.7 ± 0.5 24.3 ± 0.5 27.7 ± 1 26.5 ± 0.3 28.3 ± 1 30.5 ± 1 (g) BAT (mg)/BW 3.9 ± 0.02 2.8 ± 0.6 4.3 ± 0.4 3.7 ± 0.4 4.1 ± 0.7 3.0 ± 0.1 4.7 ± 0.5 3.4 ± 0.5 (g) WAT (mg)/BW 10.5 ± 0.8 11 ± 0.6 12.2 ± 2.7 17 ± 0.2 14.7 ± 2 17.5 ± 1 22 ± 4 27 ± 1 (g) Glucose (mg/ 80.5 ± 6.5 79 ± 13 83 ± 4 104 ± 15 110.9 ± 12 106 ± 4 117 ± 1 101 ± 11 dL) Insulin (ng/mL) 0.24 ± 0.003 0.3 ± 0.1 0.25 ± 0.02 0.32 ± 0.02 0.26 ± 0.02 0.35 ± 0.04 0.24 ± 0.02 0.3 ± 0.01 Cholesterol 79.1 ± 13 461 ± 39* 109.4 ± 8.4* 565.3 ± 48.8* 122.05 ± 12* 569.8 ± 18* 120 ± 5* 603.9 ± 31** (mg/dL) TG (mg/dL) 43.4 ± 1.6 53.9 ± 25* 36.6 ± 5 56.9 ± 8* 62.3 ± 12 73.2 ± 5* 57.2 ± 8 110 ± 12** −/− −/− −/− Male ApoE mice were fed with Western type diet since 6 week of age for 2 weeks (control and ApoE at 8 weeks of age), 6 weeks (control and ApoE at 12 weeks of age), 12 weeks (control and ApoE−/− at 18 weeks of age) or 18 weeks (control and ApoE−/− at 24 weeks of age). We measured the body weight, BAT (mg)/ BW (g), WAT (mg)/BW (g), glucose, insulin, cholesterol and triglycerides plasma levels the same day of sacrifice in fasted mice. ApoE−/− mice at all weeks of age showed hypercholesterolemia and hypertriglyceridemia and the other parameters with no significant differences in relation to each their controls BW body weight, BAT brown adipose tissue, WAT white adipose tissue, TG triglycerides * p < 0.05, ** p < 0.005 vs. each control Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 9 of 14 8 weeks 24 weeks a b 60 % lesion area /total area Control Control IRB Control ApoE-/- *** 1,4 ApoE-/- 1,2 40 100 µm 100 µm 1 20 RQ 0,8 ApoE -/- * 0,6 0 0,4 100 µm 100 µm 8 weeks 24 weeks 0,2 OilredO & Hematoxylin 0 2,5 8 weeks 24 weeks Control IRA 2 ApoE-/- * * Control 40 Control ApoE-/- % IR / aorta area 1,5 30 100 µm 100 µm * RQ 1 20 0,5 ApoE-/- 10 0 0 1<00 µm 100 µm 8 weeks 24 weeks IR ICH 3 Control IGF-IR * 8 weeks 24 weeks 2,5 ApoE-/- 2 Control 40 *** % IGF-IR / aorta area Control ApoE-/- 1,5 30 1 RQ 100 µm 100 µm 20 * 0,5 10 0 ApoE-/- 8 weeks 12 weeks 18 weeks 24 weeks 0 1<00 µm 100 µm 8 weeks 24 weeks IGF-IR ICH Fig. 5 IR and IGF-IR expression in aorta from an experimental model of atherosclerosis. a By qRT-PCR, we studied IRA, IRB and IGF-IR mRNA expres‑ sion in aorta from ApoE−/− mice at 8, 12, 18 and 24 weeks of age and their respective controls. *p < 0.05 vs. each control. b Representative photomi‑ crographs (10× magnification) and quantifications of OilRedO staining and immunohistochemistry of IR and IGF-IR in aortic roots from ApoE−/− at 8 and 24 weeks of age and their respective controls. *p < 0.05 vs. each control; ***p < 0.001 vs. each control. AE ApoE−/− mice, C control mice. C 8ws (n = 5); AE 8ws (n = 5); C12 ws (n = 5); AE12 ws (n = 8); C18 ws (n = 8); AE 18ws (n = 8); C 24ws (n = 8); AE 24ws (n = 8) features of early atheromatous lesions, including 9 fatty Discussion streaks and 11 fibrolipidic lesions. We found that total IR Vascular injury results in changes in the vascular smooth mRNA expression was significantly lower in media from muscle cell environment, including increases in growth fibrolipidic lesion-bearing aortas as compared to healthy factor availability and cell alterations, leading to pro- aortas. However, no significant changes were observed liferation and migration of VSMCs and thereby to an when we analyzed specifically IRB mRNA level. More organized atherosclerotic plaque [26]. Consequently, importantly, a marked and significant increase in gene elucidating the underlying regulators of these processes expression of IGF-I and IGF-IR was observed in medial could provide a novel therapeutic strategy targeting layer from aortas bearing fatty streaks when compared VSMCs to prevent atherosclerotic progression and its with healthy aortas and fibrolipidic lesions. Furthermore, clinical consequences. In the present paper, we addressed IGF-II mRNA expression was notably higher in fatty the issue of a better understanding of molecular mecha- streaks than in healthy aortas and fibrolipidic lesions. nisms involved in VSMC migration, specifically the part These results strongly suggest that increased expression played by IR isoforms and IGF-IR in this process. We of IGF-IR and IGFs in tunica media, mainly composed of used four aortic VSMC lines: IRLoxP+/+ VSMCs (bear- VSMCs, might contribute to migration of those cells into ing IR), IR−/− VSMCs (lacking IR), IRA VSMCs (express- the intima. ing IRA isoform) and IRB VSMCs (expressing IRB Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 10 of 14 C 8ws AE 8ws C 24ws AE 24ws a DAPI 100 µm 100 µm 100 µm 100 µm Control ApoE-/- α-SMA 50 % α-SMA / aorta area 40 100 µm 100 µm 100 µm 100 µm 30 20 10 Merged 0 8 weeks 24 weeks 100 µm 100 µm 100 µm 100 µm b DAPI α-SMA IR Merged Merged AE 24ws 50 µm 50 µm 50 µm (α-SMA + IR) (α-SMA + IR + DAPI) DAPI α-SMA IGF-IR Merged Merged AE 24ws 50 µm 50 µm 50 µm (α-SMA + IGF-IR) (α-SMA + IGF-IR + DAPI) Fig. 6 VSMCs migrated into the intima of atherosclerotic plaques express IR or IGF-IR. a Representative photomicrographs (10× magnification) and quantifications of immunofluorescence of α-SMA in aortic roots from ApoE−/− mice at 8 and 24 weeks of age and their respective controls. b Representative photomicrographs (20× magnification) of double immunofluorescence against IR or IGF-IR (green staining) and α-SMA (red staining) in aortic roots from ApoE−/− mice at 24 weeks of age. DAPI staining was performed to localize nuclei of cells presented in aortic roots (blue staining). White arrows indicate VSMCs in the intima of fibrous plaque from 24-week-old ApoE−/− mice. AE ApoE−/− mice, C control mice. C 8ws (n = 5); AE 8ws (n = 5); AE 18ws (n = 8); C 24ws (n = 8); AE 24ws (n = 8) isoform) and we firstly studied the activation of PI3 K/ a lower IRA autophosphorylation than insulin, IGF-II Akt or MAPK, the major signaling pathways by which induced a higher p70S6K/Akt activation ratio compared insulin and IGFs exert their effects on cellular metabo- with insulin. IGF-I elicited a similar signaling pattern lism, growth and survival [27]. We previously described than IGF-II, although it binds IRA with much lower affin- that insulin induced a more sustained signaling in IRA ity than IGF-II [29]. compared with IRB neonatal pancreatic beta cell lines Type 2 diabetes and the metabolic syndrome are well [28] and a higher stimulation of Akt or MAPK pathways known risk factors for atherosclerosis in part due to insu- in IRA than IRB VSMCs [21]. Now, we show that IRA lin resistance and/or hyperinsulinemia [30, 31]. Addition- isoform in VSMCs mediates a greater activation of Akt, ally, accelerated atherosclerosis in diabetic patients has p70S6K and p44/42 MAPK than IRB in response not only been associated with the fact that diabetic VSMCs exhibit to insulin, but also to IGFs. Furthermore, phosphoryla- significantly increased rates of proliferation, adhesion and tion of those kinases was highly pronounced when IRA migration [32]. In this sense, several authors have exam- VSMCs were stimulated with IGFs, especially with IGF- ined the effect of insulin on VSMC migration, but the II. In this regard, IRA has been reported to exhibit high results are controversial. Yang and Kahn reported that affinity for insulin, intermediate affinity for IGF-II and physiological concentrations of insulin did not affect the low affinity for IGF-I, whereas IRB is a highly specific migration of cultured rat VSMC, but increased it when receptor for insulin [13]. Moreover, in mouse fibroblast cells were also stimulated with Ang II or inhibited it in expressing only IRA and not IGF-IR, despite producing the presence of NO [33]. Furthermore, in bovine aortic Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 11 of 14 VSMCs 10 nmol/L insulin moderately stimulated migra- that IRA expression and thereby IRA/IGF-IR hybrid tion mainly through the MAPK pathway, whereas insulin receptor formation, are increased by insulin or IGF-II, maintained VSMC quiescence and differentiation via PI3K- and this might favor the migration of VSMCs. dependent signaling [34]. In others studies, a high dose of Human atherosclerotic disease has been histologically insulin (100 nmol/L) was found to induce migration and classified by the AHA into eight types of lesions, from proliferation of both human and rat VSMCs [35, 36]. Nev- early lesions, including initial and fatty streak lesions, to ertheless, this supraphysiological concentration of insulin advanced atherosclerotic lesions [45]. In early athero- can stimulate IGF-IR, which is known to mediate VSMC sclerotic lesions, IGF-IR might contribute to atheroscle- migration [4]. In our work, migration studies showed that rotic progression by mediating proatherogenic actions of insulin stimulation led to a higher increased migration in IGFs. Other studies have described an increased level of IRA than IRB VSMCs and had no effect on IR−/− VSMCs, IGF-IR expression in VSMCs of atherosclerotic plaques indicating that insulin exerts its effect on VSMC migra- of rabbit aortas [46], as we have observed in both aorta tion by binding IR, mainly IRA isoform. IGF-I, by contrast, and aortic roots from ApoE−/− mice at 24 weeks of age. strongly promoted the migration of VSMCs and this effect Moreover, increased IGF-IR and IGF-I mRNA tran- appears to be mediated mainly through IGF-IR, since a sig- scripts were observed in asymptomatic than in sympto- nificant increase in migration was observed in cells lacking matic plaque VSMCs [47]. In contrast, IGF-IR and IGF-I IR. Consistent with this, it is well established that IGF-I is a are decreased in advanced atherosclerotic plaques and potent stimulator of VSMC proliferation and migration as a this is consistent with the increased apoptotic rates of result of binding its own receptor [4, 37, 38]. IGF-II has also VSMCs, potentially leading to plaque weakening, plaque been found to induce migration in VSMCs through IGF-IR, rupture and acute coronary events [48, 49]. In the same but it was sixfold less potent than IGF-I [4]. Because IGF- way, a recent study showed that a monocyte/macrophage II stimulated migration only in IRA VSMCs, our findings specific IGF-IR knockout mice on ApoE−/− background indicate that IGF-II requires the presence of IRA isoform, significantly increased atherosclerotic lesion formation as homodimers or as part of IRA/IGF-IR hybrid receptors, and changed plaque composition to a less stable pheno- to induce the migration of VSMCs. type [50]. Regarding IGF-II, it has been described as a Migration assays in the presence of PPP, a highly spe- pivotal promoter of growth of the atherosclerotic lesion cific inhibitor of IGF-IR, demonstrated the implication of in ApoE−/− mice and local overexpression of IGF-II can this receptor in VSMC migration. PPP specifically blocks induce the appearance of aortic focal intimal masses phosphorylation of tyrosine 1136 in the IGF-IR activa- [51]. In this sense, we describe that IRA isoform expres- tion loop, which is necessary for autophosphorylation sion is significantly increased in aorta from 24-week- of others tyrosine residues of the β subunit, but do not old ApoE−/− mice, which it might favor proatherogenic interfere with IR activity [39, 40]. We found that IGF-IR actions of IGF-II. Consistently, we show for the first inhibition abolished insulin- and IGF-II-induced migra- time that gene expression of IGF-I, IGF-II and IGF-IR is tion in IRB VSMCs, whereas migration of IRA VSMCs notably increased in the medial layer, mainly composed was partially inhibited. However, IGF-I-stimulated migra- of VSMCs, of human aortas bearing early atheroscle- tion was inhibited by PPP at a similar level in both IRA rotic lesions. Additionally, in ApoE−/− mice we observed and IRB cell lines. Taking into account that IR/IGF-IR that VSMCs migrated into the intima of fibrous plaques hybrid receptor autophosphorylation occurs by an intra- showed positive staining for IR and IGF-IR. Over- molecular transreaction in which both β subunits transfer all, increased expression of IRA and IGF-IR would be phosphate to each other [41], it might be considered that required to support the migration of medial VSMCs into signal transduction through hybrid receptors, like IGF-IR the intima. homodimers, is prevented with PPP treatment. Therefore, remaining migration when IGF-IR phosphorylation was Conclusions inhibited appears to be mediated by IRA, but not by IRB In conclusion, our in vitro results strongly suggest that homodimers, upon stimulation with insulin or IGF-II. IGF-I is a potent inducer of VSMC migration mainly Other authors have described the presence of IR/IGF-IR through IGF-IR, and to a lesser extent through IR. Insu- hybrid receptors in human or rat VSMCs [42, 43], and we lin and IGF-II are less potent and require the presence have previously demonstrated that both IRA/IGF-IR and of IRA isoform to induce migration, by binding IRA IRB/IGF-IR hybrid receptors are expressed in our VSMC homodimers or IRA/IGF-IR hybrid receptors. Therefore, lines [21]. Insulin and IGFs are able to stimulate IRA/ overexpression of IGF-IR or IRA isoform might occur in IGF-IR hybrid receptors, although insulin is less effective. early stages of atherosclerosis favoring the migration and Regarding IRB/IGF-IR activation, insulin and IGF-II are proliferation of VSMCs and thereby atherosclerotic pro- much less effective than IGF-I [44]. We currently describe gression of relevance in experimental mouse models and Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 12 of 14 IGF-II IGF-I Ins IRA/ IGF-IR IGF-IR IRA IRB IRB/ IGF-IR AKT VSMC ERK Gene transcription MIGRATION ECs Inma VSMCs Normal Artery Early Atherosclerosis Fig. 7 Proposed model of role of IR isoforms, IGF-IR and IR/IGF-IR hybrid receptors in VSMC migration. We propose that IGF-I mainly through IGF-IR, while insulin and IGF-II by binding IRA homodimers or IRA/IGF-IR hybrid receptors might favor VSMC migration and contribute to plaque growth in early states of atherosclerosis. ECs endothelial cells, VSMCs vascular smooth muscle cells in humans (Fig. 7). Whether IRA/IGF-IR hybrid recep- Authors’ contributions NB: has participated in the acquisition of data, analysis and interpretation of tors play a direct role at the early stage of atherogenesis data, and writing the manuscript. CE F-G: has participated in the acquisition in humans remains to be established. of data in human samples. JL M-V: has participated in the acquisition of data in human samples and helped to draft the manuscript. LP: has participated in the in the experiments of migration VSMCs. OE: has participated in the Additional files development of experimental model and helped to draft the manuscript. JB M: has participated in the collection of human samples. G G–G and SF: have participated in the development of experimental model. SD-C: has partici‑ Additional file 1. Characterization and insulin signaling in VSMC lines. pated in the acquisition of data. JE: helped to draft the manuscript. AG-H: has (A) Western blot analysis of IR and IGF-IR in four VSMC lines. *p<0.05 vs participated in the acquisition of data, analysis and interpretation of data and IRLoxP+/+ VSMCs; †p<0.05, ††p<0.005 vs. IRA VSMCs. Immunoprecipita‑ in the design of study and its coordination and in the statistical analysis as well tion of IRβ in IRLoxP+/+ or IR-/- (B) Representative gels of Western blot as writing the paper. MB: helped to draft the manuscript and leads as PI the analysis and its quantifications of Akt (T308), p70S6K (T389) and p44/42 projects that provided financial support for experiments. All authors read and MAPK (T202/Y204) phosphorylation in four cell lines stimulated with insu‑ approved the final manuscript. lin, IGF-I or IGF-II. Experiments were performed at least 3 times. *p<0.05, **p<0.005, ***p<0.001 vs each control. Author details 1 Additional file 2. Measure of MMP-2 and MMP-9 activities by gelatin Biochemistry and Molecular Biology II Department, School of Pharmacy, Com‑ zymography in IRA and IRB VSMCs stimulated with insulin, IGF-I or IGF-II plutense University of Madrid, Plaza Ramón y Cajal s/n, 28040 Madrid, Spain. 2 for 12 or 24h. FBS was used as a positive control. Health Research Institute of San Carlos Clinic Hospital (IdISSC), Madrid, Spain. 3 CIBER of Diabetes and Associated Metabolic Diseases (CIBERDEM), Barcelona, Additional file 3. Western blot analysis of IR and IGF-IR expression in Spain. 4 Vascular Research Lab, IIS-Fundación Jiménez Diaz-Autonoma Univer‑ IRLoxP+/+, IRA and IRB VSMCs stimulated with insulin, IGF-I or IGF-II for sity, Madrid, Spain. 5 Inserm, U698, Universite Paris 7, CHU X-Bichat, Paris, France. 12h. Additional file 4. Increased expression of IGF-IR and IGFs in media from Acknowledgements human aortas bearing fatty streaks. Analysis of mRNA expression of total Not applicable. IR, IRB isoform, IGF-IR and IGFs by qRT-PCR in the medial layer of 28 human aortas (8 healthy, 9 fatty streaks and 11 fibrolipidic lesions). *p<0.05 vs. Competing interests healthy aorta; †p<0.05 vs. fatty streak. The authors declare that they have no competing interests. Availability of data and materials Abbreviations Data sharing not applicable to this article as no datasets were generated or α-SMA: α-smooth muscle actin; Akt: protein kinase B; ApoE−/−: apolipoprotein analysed during the current study. E knockout; BATIRKO: brown adipose tissue-specific insulin receptor knockout; IGF-IR: insulin-like growth factor-I receptor; IGFs: insulin-like growth factors; IR: Ethics approval and consent to participate insulin receptor; MMPs: matrix metalloproteinases; qRT-PCR: real-time quanti‑ Human tissue samples were harvested after organ transplantation with the tative PCR; VSMCs: vascular smooth muscle cells. authorization of the French Biomedicine Agency (authorization number Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 13 of 14 PFS09-007). Animal studies conform to the Guide for the Care and Use of 17. Escribano O, Guillén C, Nevado C, Gómez-Hernández A, Kahn CR, Benito Laboratory Animals published by the National Institutes of Health (NIH M. Beta-cell hyperplasia induced by hepatic insulin resistance: role of a Publication No. 85–23, revised 1996) and accord with The ARRIVE Guideline for liver-pancreas endocrine axis through insulin receptor A isoform. Diabe‑ Reporting Animal research. tes. 2009;58:820–8. 18. Diaz-Castroverde S, Gómez-Hernández A, Fernández S, García-Gómez Funding G, Di Scala M, González-Aseguinolaza G, Fernández-Millán E, González- This work was supported by grants SAF2007/60058, SAF2008/00031, Rodríguez Á, García-Bravo M, Chambon P, Álvarez C, Perdomo L, Beneit SAF2011/22555 and SAF2014-51795-R from Ministerio de Ciencia e N, Escribano O, Benito M. Insulin receptor isoform A ameliorates Innovación to MB, Comunidad de Madrid (S2010/BMD-2423) and CIBER de long term glucose intolerance in diabetic mice. Dis Model Mech. Diabetes y Enfermedades Metabólicas Asociadas (CIBERDEM, ISCIII, Spain) to 2016;9:1271–81. MB, Spain. 19. Diaz-Castroverde S, Baos S, Luque M, Di Scala M, González-Aseguinolaza G, Gómez-Hernández A, Beneit N, Escribano O, Benito M. Prevalent role Received: 22 September 2016 Accepted: 22 November 2016 of the insulin receptor isoform A in the regulation of hepatic glycogen metabolism in hepatocytes and in mice. Diabetologia. 2016;59:2702–10. 20. Sciacca L, Costantino A, Pandini G, Mineo R, Frasca F, Scalia P, Sbraccia P, Goldfine ID, Vigneri R, Belfiore A. Insulin receptor activation by IGF-II in breast cancers: evidence for a new autocrine/paracrine mechanism. Oncogene. 1999;18:2471–9. References 21. Gómez-Hernández A, Escribano Ó, Perdomo L, Otero YF, García-Gómez G, 1. Libby P, Ridker PM, Hansson GK. Progress and challenges in translating Fernández S, Beneit N, Benito M. Implication of insulin receptor A isoform the biology of atherosclerosis. Nature. 2011;473:317–25. and IRA/IGF-IR hybrid receptors in the aortic vascular smooth muscle cell 2. Bennett MR, Sinha S, Owens GK. Vascular smooth muscle cells in athero‑ proliferation: role of TNF-α and IGF-II. Endocrinology. 2013;154:2352–64. sclerosis. Circ Res. 2016;118:692–702. 22. Kuzuya M, Kanda S, Sasaki T, Tamaya-Mori N, Cheng XW, Itoh T, Itohara S, 3. Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Iguchi A. Deficiency of gelatinase A suppresses smooth muscle cell inva‑ Nature. 1993;362:801–9. sion and development of experimental intimal hyperplasia. Circulation. 4. Bornfeldt KE, Raines EW, Nakano T, Graves LM, Krebs EG, Ross R. Insulin- 2003;108:1375–81. like growth factor-I and platelet-derived growth factor-BB induce 23. Cho A, Reidy MA. Matrix metalloproteinase-9 is necessary for the regula‑ directed migration of human arterial smooth muscle cells via signal‑ tion of smooth muscle cell replication and migration after arterial injury. ing pathways that are distinct from those of proliferation. J Clin Invest. Circ Res. 2002;91:845–51. 1994;93:1266–74. 24. Johnson C, Galis ZS. Matrix metalloproteinase-2 and -9 differentially 5. Bayes-Genis A, Conover CA, Schwartz RS. The insulin-like growth factor regulate smooth muscle cell migration and cell-mediated collagen axis: a review of atherosclerosis and restenosis. Circ Res. 2000;86:125–30. organization. Arterioscler Thromb Vasc Biol. 2004;24:54–60. 6. Ebina Y, Edery M, Ellis L, Standring D, Beaudoin J, Roth RA, Rutter WJ. 25. Nakashima Y, Plump AS, Raines EW, Breslow JL, Ross R. ApoE-deficient Expression of a functional human insulin receptor from a cloned cDNA in mice develop lesions of all phases of atherosclerosis throughout the arte‑ Chinese hamster ovary cells. Proc Natl Acad Sci USA. 1985;82:8014–8. rial tree. Arterioscler Thromb. 1994;14:133–40. 7. Ullrich A, Gray A, Tam AW, Yang-Feng T, Tsubokawa M, Collins C, Henzel 26. Johnson JL. Emerging regulators of vascular smooth muscle cell function W, Le Bon T, Kathuria S, Chen E, Jacobs S, Francke U, Ramachandran J, in the development and progression of atherosclerosis. Cardiovasc Res. Fujita-Yamaguchi Y. Insulin-like growth factor I receptor primary structure: 2014;103:452–60. comparison with insulin receptor suggests structural determinants that 27. Siddle K. Signaling by insulin and IGF receptor: supporting acts and new define functional specificity. EMBO J. 1986;5:2503–12. player. J Mol Endocrinol. 2011;47:R1–10. 8. Malaguarnera R, Belfiore A. The insulin receptor: a new target for cancer 28. Bartolomé A, Guillén C, Benito M. Role of the TSC1-TSC2 complex in the therapy. Front Endocrinol (Lausanne). 2011;2:93. integration of insulin and glucose signaling involved in pancreatic beta- 9. Moxham CP, Duronio V, Jacobs S. Insulin-like growth factor I receptor cell proliferation. Endocrinology. 2010;151:3084–94. β-subunit heterogeneity. Evidence for hybrid tetramers composed of 29. Sacco A, Morcavallo A, Pandini G, Vigneri R, Belfiore A. Differential signal‑ insulin-like growth factor I and insulin receptor heterodimers. J Biol ing activation by insulin and insulin-like growth factors I and II upon Chem. 1989;264:13238–44. binding to insulin receptor isoform A. Endocrinology. 2009;150:3594–602. 10. Soos MA, Whittaker J, Lammers R, Ullrich A, Siddle K. Receptors for insulin 30. Stout RW. Insulin as a mitogenic factor: role in the pathogenesis of and insulin-like growth factor-I can form hybrid dimers. Characterisation cardiovascular disease. Am J Med. 1991;90:62S–5S. of hybrid receptors in transfected cells. Biochem J. 1990;270:383–90. 31. Nigro J, Osman N, Dart AM, Little PJ. Insulin resistance and atherosclerosis. 11. Belfiore A, Frasca F, Pandini G, Sciacca L, Vigneri R. Insulin receptor iso‑ Endocr Rev. 2006;27:242–59. forms and insulin receptor/insulin-like growth factor receptor hybrids in 32. Faries PL, Rohan DI, Takahara H, Wyers MC, Contreras MA, Quist WC, King physiology and disease. Endocr Rev. 2009;30:586–623. GL, Logerfo FW. Human vascular smooth muscle cells of diabetic origin 12. Seino S, Bell GI. Alternative splicing of human insulin receptor messenger exhibit increased proliferation, adhesion, and migration. J Vasc Surg. RNA. Biochem Biophys Res Commun. 1989;159:312–6. 2001;33:601–7. 13. Frasca F, Pandini G, Scalia P, Sciacca L, Mineo R, Costantino A, Goldfine ID, 33. Yang M, Kahn AM. Insulin-inhibited and stimulated cultured vascular Belfiore A, Vigneri R. Insulin receptor isoform A, a newly recognized, high- smooth muscle cell migration are related to divergent effects on protein affinity insulin-like growth factor II receptor in fetal and cancer cells. Mol phosphatase-2A and autonomous calcium/calmodulin-dependent Cell Biol. 1999;19:3278–88. protein kinase II. Atherosclerosis. 2008;196:227–33. 14. Vogt B, Carrascosa JM, Ermel B, Ullrich A, Häring HU. The two isotypes of 34. Wang CC, Gurevich I, Draznin B. Insulin affects vascular smooth muscle the human insulin receptor (HIR-A and HIR-B) follow different internaliza‑ cell phenotype and migration via distinct signaling pathways. Diabetes. tion kinetics. Biochem Biophys Res Commun. 1991;177:1013–8. 2003;52:2562–9. 15. Sciacca L, Prisco M, Wu A, Belfiore A, Vigneri R, Baserga R. Signaling 35. Abhijit S, Bhaskaran R, Narayanasamy A, Chakroborty A, Manickam N, differences from the A and B isoforms of the insulin receptor (IR) in Dixit M, Mohan V, Balasubramanyam M. Hyperinsulinemia-induced vas‑ 32D cells in the presence or absence of IR substrate-1. Endocrinology. cular smooth muscle cell (VSMC) migration and proliferation is mediated 2003;144:2650–8. by converging mechanisms of mitochondrial dysfunction and oxidative 16. Escribano O, Gómez-Hernández A, Díaz-Castroverde S, Nevado C, García stress. Mol Cell Biochem. 2013;373:95–105. G, Otero YF, Perdomo L, Beneit N, Benito M. Insulin receptor isoform 36. Yu C, Wang Z, Han Y, Liu Y, Wang WE, Chen C, Wang H, Jose PA, Zeng C. A confers a higher proliferative capability to pancreatic beta cells Dopamine D4 receptors inhibit proliferation and migration of vascular enabling glucose availability and IGF-I signaling. Mol Cell Endocrinol. smooth muscle cells induced by insulin via down-regulation of insulin 2015;409:82–91. receptor expression. Cardiovasc Diabetol. 2014;13:97. Beneit et al. Cardiovasc Diabetol (2016) 15:161 Page 14 of 14 37. Gockerman A, Prevette T, Jones JI, Clemmons DR. Insulin-like growth 46. Polanco JI, Albajar M, Pocoví M, Rodríguez Rey JC. Induction of insulin-like factor (IGF)-binding proteins inhibit the smooth muscle cell migration growth factor receptor (IGF IR) mRNA levels by low density lipoproteins. responses to IGF-I and IGF-II. Endocrinology. 1995;136:4168–73. Biochem Biophys Res Commun. 1996;226:917–22. 38. Duan C, Bauchat JR, Hsieh T. Phosphatidylinositol 3-kinase is required for 47. Jia G, Cheng G, Soundararajan K, Agrawal DK. Insulin-like growth factor-I insulin-like growth factor-I-induced vascular smooth muscle cell prolifera‑ receptors in atherosclerotic plaques of symptomatic and asymptomatic tion and migration. Circ Res. 2000;86:15–23. patients with carotid stenosis: effect of IL-12 and IFN-gamma. Am J 39. Girnita A, Girnita L, del Prete F, Bartolazzi A, Larsson O, Axelson M. Physiol Heart Circ Physiol. 2007;292:H1051–7. Cyclolignans as inhibitors of the insulin-like growth factor-1 receptor and 48. von der Thüsen JH, Borensztajn KS, Moimas S, van Heiningen S, Teeling P, malignant cell growth. Cancer Res. 2004;64:236–42. van Berkel TJ, Biessen EA. IGF-1 has plaque-stabilizing effects in athero‑ 40. Vasilcanu D, Girnita A, Girnita L, Vasilcanu R, Axelson M, Larsson O. The sclerosis by altering vascular smooth muscle cell phenotype. Am J Pathol. cyclolignan PPP induces activation loop-specific inhibition of tyrosine 2011;178:924–34. phosphorylation of the insulin-like growth factor-1 receptor. Link to 49. Okura Y, Brink M, Zahid AA, Anwar A, Delafontaine P. Decreased expres‑ the phosphatidyl inositol-3 kinase/Akt apoptotic pathway. Oncogene. sion of insulin-like growth factor-1 and apoptosis of vascular smooth 2004;23:7854–62. muscle cells in human atherosclerotic plaque. J Mol Cell Cardiol. 41. Frattali AL, Pessin JE. Relationship between alpha subunit ligand 2001;33:1777–89. occupancy and beta subunit autophosphorylation in insulin/insulin-like 50. Higashi Y, Sukhanov S, Shai SY, Danchuk S, Tang R, Snarski P, Li Z, Lobelle- growth factor-1 hybrid receptors. J Biol Chem. 1993;268:7393–400. Rich P, Wang M, Wang D, Yu H, Korthuis R, Delafontaine P. Insulin-like 42. Chisalita SI, Arnqvist HJ. Expression and function of receptors for insulin- growth factor 1 receptor deficiency in macrophages accelerates athero‑ like growth factor-I and insulin in human coronary artery smooth muscle sclerosis and induces an unstable plaque phenotype in Apolipoprotein E cells. Diabetologia. 2005;48:2155–61. deficient mice. Circulation. 2016;133:2263–78. 43. Johansson GS, Arnqvist HJ. Insulin and IGF-I action on insulin receptors, 51. Zaina S, Pettersson L, Ahrén B, Brånén L, Hassan AB, Lindholm M, Matts‑ IGF-I receptors, and hybrid insulin/IGF-I receptors in vascular smooth son R, Thyberg J, Nilsson J. Insulin-like growth factor II plays a central role muscle cells. Am J Physiol Endocrinol Metab. 2006;291:E1124–30. in atherosclerosis in a mouse model. J Biol Chem. 2002;277:4505–11. 44. Pandini G, Frasca F, Mineo R, Sciacca L, Vigneri R, Belfiore A. Insulin/insulin- like growth factor I hybrid receptors have different biological character‑ istics depending on the insulin receptor isoform involved. J Biol Chem. 2002;277:39684–95. 45. Stary HC, Chandler AB, Dinsmore RE, Fuster V, Glagov S, Insull W, Rosen‑ feld ME, Schwartz CJ, Wagner WD, Wissler RW. A definition of advanced types of atherosclerotic lesions and a histological classification of athero‑ sclerosis. Circulation. 1995;92:1355–74. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit