An octanoyl moiety was added to LMSG (mw, 7.87 kDa) to obtain Oct-LMSG following a previously described protocol (17). The structure of Oct-LMSG, analyzed by Fourier transform infrared and NMR spectroscopy (Fig. S1), consists of the alternative 3-linked β-D-galactopyranose and 4-linked 3,6-anhydro-α-L-galactopyranose or α-L-galactopyranose-6-sulfate containing an octanoyl ester moiety (Fig. 1).

L929 fibroblasts purchased from the American Type Culture Collection were cultured in Gibco Minimum Essential Medium (Gibco; Thermo Fisher Scientific, Inc.) containing 2.2 g/l sodium bicarbonate, 10% fetal bovine serum and 1% antibiotic-antimycotic (Gibco; Thermo Fisher Scientific, Inc.) in a humidified incubator containing 5% CO₂ at 37˚C. To determine cell viability, cells were seeded in a 96-well plate (3x10⁴ cells/well) and incubated with various concentrations of Oct-LMSG (10-1,000 µg/ml) for 24 and 48 h at 37˚C. Subsequently, cell viability was measured with an MTT assay. In brief, the wells containing cells were incubated with 10 µl of medium containing 5 mg/ml MTT (final concentration in the well, 0.5 mg/ml) for 4 h at 37˚C in the dark. The medium was removed and 100 µl dimethyl sulfoxide was added to each well to solubilize formazan crystals in the cells. The absorbance at 595 nm was measured using a Varioskan LUK multimode microplate reader (Thermo Fisher Scientific, Inc.). The viability of cells was expressed as a percentage of the control.

Fibroblasts (3x10⁵ cells/well) were cultured overnight on poly-L-lysine-coated round-glass coverslips in a 24-well plate and treated with LMSG and Oct-LMSG at a concentration of 100 µg/ml for 12 h at 37˚C. Cells with no treatment served as the control. The coverslips were washed three times (3x10 min) with 100 µl PBS, fixed with 100 µl 4% paraformaldehyde in PBS for 20 min and blocked with 100 µl 1% BSA in PBS for 30 min at room temperature (RT). After washing, the coverslips were incubated with anti-LM5 primary antibody (monoclonal-rat IgG, specific to β-D-galactan; dilution, 1:250; cat. no. LM5-050; Plant Probes) overnight at 4˚C, followed by incubation with the secondary antibody, FITC-conjugated goat anti-rat IgG (dilution, 1:500; cat. no. 31629; Thermo Fisher Scientific, Inc.), for 1 h at RT. The cell nuclei were specifically stained with DAPI. The cytoplasmic and plasma membranes of the cells were stained with CellMask™ Deep Red Plasma Membrane Stain (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocols. Cellular uptake of LMSG and Oct-LMSG was observed under a confocal laser-scanning microscope (Zeiss LSM800; Carl Zeiss AG). In addition, the fluorescent label concentration was assessed by measuring the FITC intensity using a Varioskan LUK multimode microplate reader (Thermo Fisher Scientific, Inc.) to further investigate the cellular uptake of Oct-LMSG in fibroblasts.

After 12 h of treatment, fibroblasts were immediately fixed in Karnovsky's fixative for 30 min at RT. Cells were washed with PBS, centrifuged at 1,500 x g for 1 min at 25˚C and then post-fixed with 1% OsO₄ in 0.1 M phosphate buffer for 1 h at RT. Fixed cells were washed again and centrifuged at 1,500 x g for 1 min at 25˚C. The collected cell pellets were placed in a small cavity in solidified 2% agarose and filled with 2% liquefied agarose. After solidification, the agarose containing the embedded cells was cut into small cubes and dehydrated through a series of ethyl alcohol gradients (at 50, 70, 80, 90 and 95% for 5 min each and 100% for 10 min). The agarose cubes were gently infiltrated with propylene oxide, followed by a mixture of propylene oxide and Epon 812 (1:1), and a mixture of propylene oxide and Epon 812 (1:3) for 10 min each. They were finally infiltrated with pure Epon 812 at RT overnight. The agarose cubes were then embedded in fresh Epon 812 and polymerized at 60˚C for 48 h. Ultra-thin sections were obtained using an Ultracut N ultramicrotome (Reichert-Nissei; Leica Microsystems, Inc.). Sections were stained with 5% uranyl acetate and 2% lead citrate prior to observation under a JEM1010 transmission electron microscope (TEM; JEM-1010; Jeol, Co., Ltd.).

TEM results indicated an increase in vesicles or granules within the cytoplasm of fibroblasts. Therefore, immunofluorescent staining for type I collagen was performed. Cells cultured on coated round-glass coverslips were divided into three groups: i) Normal control-cells without treatment; ii) Oct-LMSG-cells treated with 100 µg/ml of Oct-LMSG; and iii) TGF-β-cells treated with 10 ng/ml of TGF-β (Sigma-Aldrich; Merck KGaA), a known pro-fibrogenic growth factor that binds with the cell membrane receptors of fibroblasts and mediates type I collagen biosynthesis (6). After 12 h of treatment, cells were incubated with anti-collagen type I α1 chain (Col1A1) antibody (dilution, 1:250; cat. no. A22089; ABclonal Science, Inc.) overnight at 4˚C, followed by incubation with the secondary antibody, FITC-conjugated goat anti-rabbit IgG (dilution, 1:500; cat. no. 701078; Thermo Fisher Scientific, Inc.), for 1 h at RT. As a negative control for immunofluorescent staining, the sample was prepared by omitting the secondary antibody. The cells were counter-stained with DAPI (nuclei) and CellMask™ Deep Red Plasma Membrane (cytoskeleton) according to the manufacturer's protocols. Immunofluorescent images were observed under a fluorescence microscope (Leica DFC 7000T; Leica Microsystems, Inc.).

Fibroblasts (3x10⁶ cells/well) were cultured in a 6-well plate for 24 h and then divided into three groups: i) Normal control-cells without treatment; ii) Oct-LMSG-cells treated with 100 µg/ml Oct-LMSG; and iii) TGF-β-cells treated with 10 ng/ml TGF-β as a positive control. Cells were treated with an Oct-SG or TGF-β solution (2 ml) and further incubated for 6, 12 and 24 h before being harvested for RNA and protein extractions.

RNA was extracted from cultured cells using 200 µl of TRI Reagent^® RNA Isolation Reagent (Sigma-Aldrich; Merck KGaA) according to the manufacturer's protocol. The concentration and quality of RNA was measured by determining the absorbance ratio at 260/280 nm using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.). RNA (1 µg) was converted to cDNA using the Revert Aid First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, Inc.). Subsequently, mRNA transcription of Col1A1, Col1A2 and GAPDH was amplified by quantitative (q)PCR using 2 µl of cDNA with specific primers and conditions. The qPCR conditions included initial denaturation at 95˚C for 10 min, followed by 40 cycles at 95˚C for 10 sec, 60˚C for 30 sec and 72˚C for 30 sec. Col1A1-specific primers were 5'-GAGAGGTGAACAAGGTCGCG-3' (forward) and 3'-AAACCTCTCTCGCCTCTTGC-5' (reverse), Col1A2-specific primers were 5' CCCAGAGTGGAACAGCGATT-3' (forward) and 3'-ATGAGTTCTTCGCTGGGGTG-5' (reverse), and GAPDH-specific primers were 5'-GGTGAAGGTCGGTGTGAA-3' (forward) and 3'-CTCGCTCCTGGAAGATGGTG-5' (reverse) (14). The amplification and analysis were performed on a QuantStudio™ 6 Flex Real-Time PCR System device (Applied Biosystems; Thermo Fisher Scientific, Inc.). The relative gene expression in the different samples was calculated using the 2^-ΔΔCq method (18). The relative mRNA transcription levels were normalized to the GAPDH internal control and presented as a fold-change relative to the control.

Proteins were extracted from the cells using a lysis buffer containing 100X protease inhibitor solution (MedChemExpress, LLC). The protein concentration was determined using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.). Proteins (40 µg/lane) were then separated by 12%-SDS-PAGE and transferred to a nitrocellulose membrane (Sigma-Aldrich; Merck KGaA). The membrane was incubated with specific primary antibodies for Col1A1 (cat. no. A22089; ABclonal Science, Inc.), Col1A2 (cat. no. A21059; ABclonal Science, Inc.), Smad2/3 (cat. no. 8685; Cell Signaling Technology, Inc.), Smad4 (cat. no. 46535; Cell Signaling Technology, Inc.), p-Smad2/3 (cat. no. 8828; Cell Signaling Technology, Inc.) and p-Smad4 (cat. no. AF8316; Affinity Biosciences, Inc.) at a dilution of 1:1,000 at 4˚C overnight, followed by incubation with HRP-conjugated secondary antibodies, anti-mouse IgG (cat. no. 62-6520; AB_2533947) and anti-rabbit IgG (cat. no. 31460; AB_228341; Thermo Fisher Scientific, Inc.) at a dilution of 1:2,000 for 1 h at RT. Immunoreactive protein bands were visualized using the Clarity™ Western ECL substrate (Bio-Rad Laboratories, Inc.). The relative protein expression was normalized to β-actin as the internal control (cat. no. AF7018; Affinity Biosciences, Inc.) and presented as a fold-change relative to the control. Protein expression was quantified by ImageJ 1.46r analysis software (National Institutes of Health).

Data are presented as the mean ± standard error of the mean from three or more independent experiments. All treated groups were compared with the control groups. Statistical significance was analyzed using one-way ANOVA followed by Tukey's multiple-comparisons test. P<0.05 was considered to indicate a statistically significant difference.

The cytotoxicity of LMSG and Oct-LMSG to fibroblasts, as determined by the MTT assay, indicated that there was no significant cytotoxicity, and no difference between groups was observed after 24 and 48 h of incubation (Fig. 2A). To validate the hypothesis that addition of the octanoyl moiety could mediate increases in LMSG uptake into fibroblasts, an indirect immunofluorescent labelling technique was performed and the FITC intensity was quantified. As presented in Fig. 2C, the cells treated with LMSG and Oct-LMSG followed by FITC labelling displayed green fluorescence bound to the cells and localized within the cytoplasm (orange color dispersed in the cytoplasm shown in merged micrographs). No green fluorescence was detected in the control cells. The FITC-LMSG fluorescence intensity of LMSG- and Oct-LMSG-treated cells was 4.69±0.89x10⁴ and 9.71±0.64x10⁴ arbitrary units, respectively (Fig. 2B). Comparison between LMSG and Oct-LMSG revealed that the introduction of the octanoyl moiety increased the uptake of LMSG into fibroblasts, suggesting that the cellular uptake of LMSG was enhanced by octanoyl supplementation.

TEM analysis was performed to assess the ability of Oct-LMSG to affect the ultrastructure of L929 fibroblasts. It was noted that cultured fibroblasts in the control group exhibited a small oval body occupied by the nucleus of a heterochromatin structure and encircled by a small amount of cytoplasm (Fig. 3A). In addition, high-magnification images of these controls showed that the cytoplasm was filled with sparse vesicles containing pale particles, rough endoplasmic reticulum (rER) and mitochondria (Fig. 3D). By contrast, fibroblasts cultured with Oct-LMSG and TGF-β displayed primarily euchromatin nuclear features and were surrounded by an extensive cytoplasmic medium containing abundant vesicles, vacuoles and free ribosomes (Fig. 3B and C). Under high magnification, these vesicles, which appeared to be dilated cisternae of the rER, exhibited morphological differences from those of the control group, containing dense and dark materials (Fig. 3E and F). The presence of euchromatic nuclear features, increased rER-dilated cisternae and free ribosomes in the cytoplasm of fibroblasts treated with Oct-LMSG indicated that Oct-LMSG could activate fibroblasts in a similar manner to TGF-β.

To investigate the effect of Oct-LMSG on the synthesis of type I collagen in L929 fibroblasts, cells were initially exposed to Oct-LMSG (100 µg/ml) for 6, 12 and 24 h. Cells were also treated with TGF-β (10 ng/ml), a growth factor that induces the synthesis of collagen in fibroblasts (19). Col1A1 and Col1A2 mRNA transcription following treatment were measured via reverse transcription-qPCR analysis. As presented in Fig. 4, Oct-LMSG and TGF-β significantly elevated mRNA transcription levels of Col1A1 and Col1A2. The stimulatory effect of Oct-LMSG on collagen mRNA transcription was more pronounced at 6-12 h of incubation and tapered off to control levels at the 24-h mark. Following this, the effect of the Oct-LMSG on protein expression was determined by western blot analysis. As with mRNA transcription levels, exposure to Oct-LMSG increased expression of type I collagen proteins (Col1A1 and Col1A2; Fig. 5), particularly Col1A1 protein expression, which demonstrated a higher expression level that was sustained throughout the first 24 h of incubation compared to the control. In addition, Col1A1-immunofluorescent staining was performed to confirm collagen synthesis in the fibroblasts. The result showed heightened fluorescein activity in Oct-LMSG and TGF-β-treated fibroblasts compared to the control (Fig. 6), indicating that increased collagen synthesis was induced by Oct-LMSG. Taken together, these results suggest that Oct-LMSG enhances the expression of type I collagen in vitro.

Given that the TGF-β/Smad signaling pathway is crucial in regulating Col1A1 synthesis (6), the phosphorylation status of Smad proteins after Oct-LMSG treatment was analyzed. The results indicated that p-Smad2/3 and p-Smad4 were upregulated after Oct-LMSG treatment relative to controls and that the upregulated phosphorylation had no effect on total protein levels, except for Smad2/3 at 12 h of incubation (Fig. 7). These findings were consistent with those of fibroblasts treated with TGF-β. Oct-LMSG significantly upregulated p-Smad2/3 levels to 3.5±0.7-fold of the control at 6 h, 9.3±1.2-fold of control at 12 h and 2.9±0.3-fold of the control at 24 h of incubation. Concurrently, p-Smad4 levels also increased significantly with Oct-LMSG treatment to 1.9±0.1-fold of the control at 6 h, 4.9±0.9-fold of the control at 12 h and 1.8±0.2-fold of the control at 24 h of incubation. These results indicated that Oct-LMSG enhances type I collagen synthesis, at least in part, mediated by the phosphorylation of the Smad signaling pathway in fibroblasts.