A Central Bioactive Region of LTBP-2 Stimulates the Expression of TGF-β1 in Fibroblasts via Akt and p38 Signalling Pathways

Abstract

Latent transforming growth factor-β-1 binding protein-2 (LTBP-2) belongs to the LTBP-fibrillin superfamily of extracellular proteins. Unlike other LTBPs, LTBP-2 does not covalently bind transforming growth factor-β1 (TGF-β1) but appears to be implicated in the regulation of TGF-β1 bioactivity, although the mechanisms are largely unknown. In experiments originally designed to study the displacement of latent TGF-β1 complexes from matrix storage, we found that the addition of exogenous LTBP-2 to cultured human MSU-1.1 fibroblasts caused an increase in TGF-β1 levels in the medium. However, the TGF-β1 increase was due to an upregulation of TGF-β1 expression and secretion rather than a displacement of matrix-stored TGF-β1. The secreted TGF-β1 was mainly in an inactive form, and its concentration peaked around 15 h after addition of LTBP-2. Using a series of recombinant LTBP-2 fragments, the bioactivity was identified to a small region of LTBP-2 consisting of an 8-Cys motif flanked by four epidermal growth factor (EGF)-like repeats. The LTBP-2 stimulation of TGF-β expression involved the phosphorylation of both Akt and p38 mitogen-activated protein kinase (MAPK) signalling proteins, and specific inactivation of each protein individually blocked TGF-β1 increase. The search for the cell surface receptor mediating this LTBP-2 activity proved inconclusive. Inhibitory antibodies to integrins β1 and αVβ5 showed no reduction of LTBP-2 stimulation of TGF-β1. However, TGF-β1 upregulation was partially inhibited by anti-αVβ3 integrin antibodies, suggestive of a direct or indirect role for this integrin. Overall, the study indicates that LTBP-2 can directly upregulate cellular TGF-β1 expression and secretion by interaction with cells via a short central bioactive region. This may be significant in connective tissue disorders involving aberrant TGF-β1 signalling.

Keywords: Akt; LTBP-2; TGF-β; fibroblast; fibrosis; p38 MAPK.

Figure 1

Exogenous latent transforming growth factor-β-1 binding protein-2 (LTBP-2) increases transforming growth factor-β1 (TGF-β1) in conditioned medium, which is independent of extracellular matrix. (A) MSU-1.1 cells were cultured for 24 h or 3 weeks post-confluence, then incubated in serum-free medium overnight containing 10 µg/mL LTBP-2 (black columns) or bovine serum albumin (BSA) control (grey columns) prior to TGF-β1 assay. Total TGF-β in the conditioned medium and cell layer was measured by ELISA (see Materials and Methods). LTBP-2 caused a significant increase in TGF-β1 without the presence of extensive extracellular matrix; (B) The secreted TGF-β1 is mainly in an inactive form. The medium from 3-week post-confluence LTBP-2-treated cultures was analysed with and without acid treatment for total and active TGF-β1, respectively. Approximately 70% of the TGF-β1 in the conditioned medium was inactive. Quantified data are expressed as mean ± SD (standard deviation). Statistical significance was determined by paired Student’s t test, * p < 0.05.

Figure 2

LTBP-2 upregulates TGF-β1 expression in MSU-1.1 cells. (A) MSU-1.1 cells were cultured for 24 h post-confluence, then incubated for 16 h in serum-free medium with 10 µg/mL LTBP-2 or both LTBP-2 and 10 µg/mL cycloheximide (CHX). The negative controls included 10 µg/mL BSA or cycloheximide only. The conditioned medium was analysed for TGF-β1 content (see Materials and Methods); (B) MSU-1.1 cells were grown to 24 h post-confluence, then incubated for 16 h with 10 µg/mL full length LTBP-2 or 10 µg/mL BSA control. Total RNA was harvested from the cells, then reverse transcribed into cDNA for use in qPCR. The cDNA was analysed for TGF-β1, and normalized to RNAPolII. LTBP-2 stimulated a fourfold increase in the levels of TGF-β1 mRNA compared to the BSA control.

Figure 3

Time course for LTBP-2 stimulation of TGF-β1 production. MSU-1.1 cells were grown on 12-well plates to 24 h post-confluence and incubated with LTBP-2 (10 µg/mL) for various times (3–24 h). BSA at the same concentration was added to control wells. (A) TGF-β1 accumulation in the medium was measured by ELISA as described in materials and methods. The TGF-β1 concentration peaked at around 15 h. The mean values of two independent experiments are shown (±SD); (B) Total cell counts at selected time points. Note that LTBP-2 had no effect on cell proliferation above the BSA controls.

Figure 4

A short incubation of MSU-1.1 cells with LTBP-2 is sufficient to upregulate TGF-β1 expression and secretion. (A) Quantitation of TGF-β1 in the medium. MSU-1.1 cells were grown to 24 h post-confluence, then incubated in serum-free medium with LTBP-2 (10 µg/mL) for various time periods (10–60 min). After each incubation period, the medium was discarded and replaced with fresh serum-free medium lacking LTBP-2. The conditioned medium was collected at 15 h and analysed for TGF-β1 content (see Materials and Methods). The results were compared to cells incubated for 15 h with LTBP-2 or a BSA control. A significant increase in TGF-β1 was detected with as little as 10 min of exposure to LTBP-2. The mean values of three independent experiments are shown (±SD); (B) Quantitation of TGF-β1 mRNA. At the end of each 15 h incubation above, total RNA was harvested from the cells, then reverse transcribed into cDNA. Quantitative PCR was performed to determine cellular TGF-β1 mRNA levels as described in materials and methods. Values were expressed relative to the RNAP2 housekeeping gene. The control consisted of cells exposed to BSA instead of LTBP-2 for 15 h. A significant increase in TGF-β1 mRNA was detected after 15 h with as little as 10 min of LTBP-2 exposure.

Figure 5

A central region of LTBP-2 consisting of an 8-cys motif flanked by pairs of epidermal growth factor (EGF)-like repeats (fragment LTBP-2C F3) contains the TGF-β1 stimulatory activity. (A) Schematic diagram of recombinant LTBP-2 fragments. Protein fragments produced specifically for this study (LTBP-2C(H) F1, F2, and F3) are highlighted within the blue box; (B) MSU-1.1 cells were grown to 24 h post-confluence, then incubated for 16 h with full-length LTBP-2 (50 nM), molar equivalents of each of three fragments spanning LTBP-2 (LTBP-2NT, LTBP-2C, LTBP-2CT) or BSA control. The mean values of two independent experiments are shown (±SD); (C) Fresh cells were subsequently incubated with each of three sub-fragments F1, F2, and F3 spanning central fragment LTBP-2C(H). In both (A) and (B), the conditioned medium was analysed for TGF-β1 content (see Materials and Methods). The mean values of two independent experiments are shown (±SD).

Figure 6

Exogenous LTBP-2 stimulates phosphorylation of AKT and p38 mitogen-activated protein kinase (MAPK) in human fibroblasts. (A) MSU-1.1 cells (1 × 10⁵ cells/well) were treated with or without LTBP-2 (10 µg/mL) for 30 min. Total cell lysates were immunoblotted for phosphorylation of candidate signalling molecules, including phospho-serine, phospho-tyrosine, phospho-threonine, phospho-p38 MAPK, phospho-Akt1/2/3, phospho-ERK, c-FOS, c-JUN, and for β-actin internal control as described in materials and methods. Note that there was major phosphorylation of p38 MAPK and AKT1/2/3, but no stimulation of ERK or cFOS; (B) Cells were treated for 30 min with full-length LTBP-2 or with molar equivalents of fragments containing TGF-β1 stimulating activity, LTBP-2C(H), or LTBP-2C(H) F3. Cell lysates were immunoblotted for total and phosphorylated p38 MAPK, Akt1/2/3, and ERK; (C) The ratio of phospho-protein to total protein for each signal molecule from each treatment is expressed relative to the average value from no LTBP-2 control cells (given an arbitrary value of 1.0). Similar results were observed in replicate experiments.

Figure 7

LTBP-2 stimulation of TGF-β upregulation involves Akt and p38 MAPK signalling pathways. MSU-1.1 cells were grown to 24 h post-confluence, then incubated in serum-free medium with or without inhibitor (10 µM) of Akt (GSK690693 or AZD536310) or p38 MAPK (VX-702 or SB202190) for 2 h, followed by an addition of 50 nM of LTBP-2 or bioactive fragment LTBP-2C F3. Incubation was continued for a further 15 h, and the conditioned medium was analysed for TGF-β1 content (see Materials and Methods). Controls involved incubations with individual inhibitors or BSA in the absence of LTBP-2. The mean values ± SD from duplicate experiments are shown.

Figure 8

Blocking of integrin αVβ3 receptors partially attenuates TGF-β1 production induced by LTBP-2. (A) Effects of blocking antibodies to integrins on LTBP-2 stimulated TGF-β1 expression. MSU-1.1 cells were pre-treated with blocking antibody for integrins β1 (10 µg/mL), αVβ3 (10 µg/mL), and αVβ5 (10 µg/mL) for 2 h. Incubation was continued for 15 h with the addition of a bioactive LTBP-2 fragment (LTBP-2C(H) or LTBP-2C(H) F3) 50 nM final concentration), and the conditioned medium was analysed for TGF-β1 content (see Materials and Methods). Control incubations included LTBP-2 or a bioactive fragment without antibody and BSA in the absence of LTBP-2. The mean values of two independent experiments are shown (±SD); (B) Increasing the concentration of anti-integrin αVβ3 antibody caused little further attenuation of LTBP-2-induced TGF-β1 upregulation. The above experiment was repeated including two concentrations of anti-integrin αVβ3 (10 µg/mL and 20 µg/mL) during a 2 h pre-incubation and for a further 15 h following an addition of a LTBP-2 C F3 fragment (5 nM final concentration). The conditioned medium was then analysed for TGF-β1 content (see Materials and Methods). Controls involved incubation of cells with anti-integrin antibody only or BSA. Mean values ±SD from triplicate wells are shown. Doubling of the anti αvβ3 antibody concentration caused only a very slight additional reduction in TGF-β stimulation, suggesting that the blocking effect was close to saturation. The mean values of two independent experiments are shown (±SD).

Figure 9

Schematic model of possible signalling events involved in the upregulation of TGF-β1 expression by LTBP-2. LTBP-2 binds to an unidentified receptor (perhaps complexed with αVβ3 integrin and enhanced by binding to cell surface heparan sulphate proteoglycans (HSPGs)) and induces the phosphorylation of Akt and p38β MAPK pathways. The inhibition of either pathway inhibits TGF-β1 expression and secretion. The precise mechanisms remain to be elucidated, but in other systems blocking p38 MAPK has been shown to inhibit TGFB1 gene transcription and blocking Akt pathways can inhibit TGF-β1 transcription and/or translation. Arrows indicate positive (stimulatory) effects. The inhibitors used to specifically block respective signalling molecules are also depicted.