Dulbecco’s modified Eagle’s medium (DMEM), antibiotics, trypsin, insulin, Bodipy 493/503 and Lipofectamine 2000 were obtained from Invitrogen (Carlsbad, CA, USA). Methylisobutylxanthine, dexamethasone were obtained from Sigma-Aldrich (St. Louis, MO, USA). Puromycin was purchased from Invivogen (San Diego, CA, USA). 3T3-L1 fibroblasts were obtained from Asociación Banco Argentino de Células (origin: ATCC), NIH 3T3 from ATCC and HEK293FT from Invitrogen. 3T3-L1 cells could not be authenticated due to the lack of an STR DNA profile database for mouse cells. However, the adipocyte phenotype that we observed only after differentiation is consistent with the fact that the cells used were indeed 3T3-L1 fibroblasts. Fetal Bovine Serum (FBS) was purchased from Natocor (Argentina).

3T3-L1 and NIH 3T3 fibroblasts were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 0.1 mg/mL streptomycin, 100 U/mL penicillin and 250 ng/mL fungizone (DMEM + 10% FBS). In differentiation experiments, plated cells were allowed to reach confluence and two days later differentiation was induced with a mixture containing: 10 μg/mL insulin (I), 0.1 μM dexamethasone (D), and 0.5 mM methylisobutylxanthine (M) in DMEM with 10% FBS (IDM). After 3 days medium was replaced with DMEM + 10% FBS supplemented with 10 μg/mL insulin and cells were fed every 2 days thereafter with DMEM + 10% FBS.

In some experiments, 2-day postconfluent 3T3-L1 fibroblasts were stimulated with individual inducers (I, D, M) or combinations of two inducers (ID, DM, IM) at the concentrations indicated above. After 3 days culture medium was replaced with DMEM + 10% FBS alone or supplemented with 10 μg/mL insulin in those assays induced in the presence of insulin.

A fragment coding a small hairpin RNA (shRNA) that targeted murine Mcam was subcloned into the pLKO.1 plasmid [16] according to the protocol available from Addgene (http://www.addgene.org/tools/protocols/plko/). The shRNA sequence 5’-CCGACTCGTAAGAGTGAATTT-3’, which targets the coding DNA sequence (CDS) of the murine Mcam mRNA, was selected from the TRC RNAi Consortium library (http://www.broad.mit.edu/genome_bio/trc/rnai.html), clone ID TRCN0000113082. The empty pLKO.1 vector, without the shRNA construct and the stuffer sequence, was used as a negative control.

For lentivirus production, exponentially growing HEK293FT cells cultured in 6-well plates were transfected in 1.5 ml of medium using 6 μl of Lipofectamine 2000 with 1.2 μg of either the shRNA containing vector or the empty vector, in combination with 1 μg of the pCMV-ΔR8.9 packaging plasmid and 0.3 μg of the pCMV-VSVG envelope plasmid. Media was replaced with fresh one after 6 hours. Lentivirus containing media were harvested at 48 h and 72 h after transfection and filtered through a 0.45 μm filter.

To establish cell lines stably expressing shRNA or control vector, exponentially growing 3T3-L1 fibroblasts in 6-well plates were transduced with virus-containing media supplemented with 8 μg/mL polybrene. After 24 h, 3T3-L1 cells were transduced again with lentivirus containing media for another 24 h. Then, transduced cells were subjected to selection with culture media supplemented with 5 μg/mL puromycin for 4–6 days.

3T3-L1 adipocytes were washed three times with phosphate-buffered saline (PBS) and then fixed for 30 min with 4% formaldehyde in PBS. Three volumes of Oil-Red-O (0.4% in isopropanol) were diluted with two volumes of water, filtered, and added to the fixed cell monolayers for 30 min at room temperature. Cells were then washed with water and the stained lipid droplets were visualized by light microscopy and photographed.

3T3-L1 cells grown in 6-well plates were harvested with a cell scraper in PBS 24 h after induction of differentiation for C/EBPβ or 72 h for β-catenin, vortexed and before the addition of Laemmli sample buffer [17] an aliquot was obtained for protein quantification using the method of Bradford with serum albumin as standard [18]. For ERK determination, two-day postconfluent 3T3-L1 cells were treated for 1 h with DMEM + 1 % FBS then cells were induced to differentiate for 30 min. After the treatments, cells were harvested in Laemmli sample buffer and vortexed. Samples were loaded in SDS-polyacrylamide gels, electrophoresed and transferred to nitrocellulose membranes. After blocking with 5% non-fat dry milk for 1 h, membranes were incubated overnight at 4 °C with primary antibody at a 1:1000 dilution, and then with HRP-conjugated secondary antibody at a 1:1000 dilution for 1 h at room temperature. Detection was performed with a chemiluminescent reagent. GAPDH was used as the loading control. C/EBPβ (C-19, cat # sc-150), GAPDH (6C5, cat # sc-32233), p-ERK (E-4, cat # sc-7383), ERK 2 (C-14, cat # sc-154) antibodies were from Santa Cruz (Santa Cruz Biotechnology, CA, USA) and β-catenin (D10A8, cat# 8480) was from Cell Signaling Technology (Danvers, MA, USA). Secondary antibodies used, anti-mouse IgG-HRP (C2011, cat# sc-2005) and anti-rabbit IgG-HRP (cat# sc-2004), were both from Santa Cruz.

3T3-L1 cells were cultured on glass coverslips. 7 days after induction of differentiation cells were fixed for 15 min with 4% formaldehyde in PBS at room temperature, permeabilized with 0.5% saponin in PBS for 10 min and blocked with 1% BSA − 0.3 M glycine in PBS for 1 h at room temperature. Cells were incubated with 1:75 dilution of anti-MCAM primary antibody (EPR 3208, cat # ab75769, Abcam, Cambridge, MA, USA) or 1:100 anti-PPARγ (C26H12, cat# 2435, Cell Signaling Technology, Danvers, MA) for 1 h at 37 °C in a humidified chamber, and then incubated for 1 h with 1:500 dilution of goat anti-rabbit Alexa 555-conjugated secondary antibody (cat# A21428, Invitrogen). Finally, nuclei were stained with Hoechst 33258 and, when indicated, lipid droplets with Bodipy 493/503. Coverslips were mounted on slides and cells were visualized and photographed in an Olympus IX-71 inverted microscope equipped with a QImaging EXi Aqua monochrome camera. In any given experiment all images were acquired using the same settings and 8–10 random fields were photographed from each sample. Images were processed and analyzed using Fiji software [19]. Background correction was performed by using the Rolling Ball algorithm for Hoechst images or by substracting the mean intensity value from a region without cells for MCAM and PPARγ images. Blue, green or red lookup tables (LUTs) were used to colour grayscale images. Then, for better visualization, minor adjustements to maximum intensity were made to the whole figure. For cell counting studies, the number of Hoechst-stained nuclei and lipid droplets-containing cells in multiple fields were determined. At least 1,000 nuclei were scored in each replicate.

Total RNA purification from 3T3-L1 fibroblasts was performed with TRI Reagent (Molecular Research Center) according to the manufacturer’s instructions, pellets were dissolved in nuclease-free water and RNA concentration was determined with NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). 2 μg of RNA and 0.5 μg of Oligo (dT)₁₅ were used for reverse transcription with RevertAid Reverse Transcriptase (Thermo Fisher Scientific). mRNA relative expression was determined by quantitative real-time PCR using a Bio-Rad MyIQ2 thermal cycler. Each PCR reaction was performed using Platinum Taq Polymerase (Invitrogen) in a final volume of 25 μL, containing 5 μL of a 1:10 dilution of first-strand cDNA and 5 pmol of each primer. The following cycling conditions were used: 94°C for 5 min, followed by 40 cycles of 94°C for 20 s, 58°C for 30 s and 72°C for 40 s. Target gene mRNA expression was normalized to acidic ribosomal protein (Rplp0) as a reference gene. MCAM and Perilipin 1 (Plin1) primers were designed using Primer3 v0.4.0 software [20], Rplp0, Peroxisome proliferator activated receptor gamma isoform 2 (PPARγ2), and Adiponectin (Adipoq) primer sequences were obtained from published reports [21–23]. The sequences of primers used are shown in Table 1.

The experiments were carried out three times unless otherwise stated. All data are expressed as mean ± SD. Statistical analysis was performed using GraphPad Prism 6 software. More than two samples were compared by One-Way ANOVA followed by Bonferroni post hoc test and two samples by Student’s t-test, p values below 0.05 were considered significant.

In the present investigation, we first analyzed the dependence of MCAM upregulation with the inducers of the mixture used to differentiate 3T3-L1 fibroblasts to adipocytes. In accordance with our previous microarray data, we found that MCAM mRNA level was increased in 3T3-L1 fibroblasts treated with differentiation mixture (IDM). However, treatment with any of the inducers alone failed to stimulate MCAM expression (Fig. 1A). Furthermore, none of the inducers alone was able to induce adipogenesis (Fig. 1B). When 3T3-L1 fibroblasts were treated using different combinations of inducers, only in the case of DM, differentiation was attained although in a lower extent than when the three inducers were used. This is in agreement with previous reports [24, 25]. Interestingly, MCAM was only upregulated by DM and neither ID nor IM were able to induce MCAM or differentiation (Fig. 1).

The increase of MCAM after induction of differentiation of 3T3-L1 fibroblasts was confirmed by immunofluorescence. In keeping with the results shown in Figure 1, we found an increase in the amount of this protein in the differentiated 3T3-L1 cells which were also stained with Bodipy 493/503 to visualize accumulation of cytosolic lipids (Fig. 2). As previously reported, we observed that this protein is localized at the plasma membrane (9).

We also evaluated the time-course of MCAM induction and found that MCAM mRNA continuously increased from day 3 after the addition of IDM and the increase in expression was maintained in mature adipocytes. In contrast, the mRNA of PPARγ, the master gene of adipogenesis, was increased after 24 h and Perilipin 1, a gene which is known to be upregulated by PPARγ [26, 27] after 48 h (Fig. 3).

To further investigate the importance of differentiation to adipocytes in the induction of MCAM, we used NIH 3T3 fibroblasts as a cell model that do not differentiate to adipocytes after the addition of the differentiation mixture. After treatment of NIH 3T3 cells with IDM, we did not find lipid containing cells that resembled adipocytes neither did we find a significant increase in Perilipin 1 or MCAM mRNA expression as we observed in 3T3-L1 cells (Fig. 1A). These results support the importance of adipogenesis in the induction of MCAM.

To characterize the role of MCAM during 3T3-L1 fibroblasts differentiation we generated 3T3-L1 cells that stably expressed a shRNA against MCAM mRNA. We analyzed MCAM mRNA levels in knockdown cells (MCAM-KD) and found that MCAM mRNA upregulation was reduced on day 4 and 7 after induction of differentiation compared to control cells (Fig. 4A). When we evaluated the ability of these cells to differentiate, we found that after induction of differentiation the cytosolic accumulation of lipids which was evaluated by Oil-Red-O staining was inhibited in the MCAM-KD cells (Fig. 4B). In keeping with this, as shown in Fig. 5A, we also found impairment of MCAM at the protein level in MCAM-KD cells which was evaluated by immunofluorescence. When we counted under the microscope the number of adipocytes relative to the number of Hoechst-stained cells, we consistently observed a significant decrease in the amount of adipocytes in knockdown cells (0.33±0.09 vs. 0.09±0.02, p<0.01, n=7).

Then, the effects of MCAM knockdown on the amount of PPARγ, Perilipin 1 and Adiponectin mRNA was evaluated in the MCAM-KD cells with respect to those treated with the empty vector. Consistent with the inhibition of the adipogenesis there was a decrease in PPARγ mRNA, as well as in the mRNA expression of Perilipin 1 and Adiponectin, which are both upregulated by PPARγ (Fig. 4A). The decrease of nuclear PPARγ was also confirmed by immunofluorescence (Fig. 5B).

To further investigate the effect of MCAM knockdown on adipogenesis, we analyzed two proteins which are regulated in early adipogenesis and are involved in PPARγ upregulation [28, 29]: C/EBPβ, a transcription factor upregulated in the first 24 h [4] and β-catenin, a protein involved in canonic Wnt signaling, which is known to be decreased in adipogenesis.

First, we analyzed the amount of C/EBPβ in MCAM-KD cells treated with or without differentiation mixture and compared the increase in the amount of this protein after induction of differentiation with that obtained in cells transduced with an empty vector. We found that 24 h after induction of differentiation the expected increase in C/EBPβ was still present in MCAM-KD cells (Fig. 6A).

On the other hand, in keeping with previous reports, it was found that in control cells induction of differentiation provoked a decrease in the content of β-catenin and this decrease was not affected in MCAM-KD cells (Fig. 6B).

Finally, we also found that another early event involved in this differentiation process, the activation of a MAP kinase, ERK, which takes place in the first hours after the addition of the differentiation mixture [30], was maintained in the MCAM-KD cells in spite of the inhibition in differentiation (Fig. 6C).

Since our results suggest that the early events in adipogenesis that precedes and are necessary for PPARγ activation were not impaired in MCAM-KD cells, we analyzed if the early increase of PPARγ mRNA at 24 h post differentiation induction (as shown in figure 3) was affected in MCAM-KD cells. On the contrary, we found that PPARγ expression was only impaired 3 days after the induction of differentiation in MCAM-KD cells when MCAM downregulation was also evident (Fig. 6D).