LMP1 regulates periodontal ligament progenitor cell proliferation and differentiation

Abstract

LMP1 is an intracellular scaffold protein that contains a PDZ domain and three LIM domains. LMP1 has multiple functions including regulating mesenchymal stem cell (MSC) osteogenesis. Gene delivery of LMP1 induces bone formation in vivo in heterotopic and orthotopic sites. However, little is known about the physiological function and gene regulatory mechanisms of LMP1 in MSCs at the molecular level. Periodontal ligament (PDL) cells are a unique progenitor cell population that can differentiate into multiple cell types, including osteoblasts, adipocytes, or chondrocytes. This study sought to determine the physiological function and gene regulatory mechanisms of LMP1 in PDL cells at the molecular level. We show that LMP1 is upregulated in early stage of PDL cell osteogenic differentiation. Stable gene knockdown of LMP1 by shRNA inhibits DNA synthesis and corresponding cell proliferation in PDL cells, and further leads to decreased mineralization in vitro. Overexpression of LMP1 increases cell proliferation, and PDZ and ww-interacting domains are not sufficient to mediate this effect. Further, we found that in PDL cells, LMP1 is a downstream target gene of TGF-beta1 that is an early signal critical in preosteoblast proliferation and differentiation. TGF-beta1 stimulates PDL cell proliferation, however, this effect is compromised when LMP1 is knocked down. We further identified that the activation of TAK1-JNK/p38 kinase cascade is involved in the LMP1 gene regulation by TGF-beta1. We conclude that LMP1 is a downstream gene of TGF-beta1, involved in PDL cell proliferation. Our findings advance the understanding of the physiological function of LMP1 and define a regulatory mechanism of LMP1 in PDL progenitor cells and other MSCs.

Fig. 1. LMP1 is upregulated at early stage of osteogenesis in PDL cells

Primary PDL cells were induced for osteogenic differentiation. (A) RT-qPCR was used to evaluate LMP1 gene expression. A representative result of three independent experiments is shown. **, p<0.01; *, p<0.05 vs. non-induced control. (n=3 per group) (B) The LMP1 protein expression is shown by Western Blot. Relative expression ratios after normalization to GAPDH are shown at the bottom.

Fig. 2. LMP1 is required for PDL cell proliferation

Two double-stranded shRNAs targeting LMP1 and a scramble shRNA were designed and cloned into a retroviral system. After retrovirus infection and puromycin selection, all the survival cells were pooled. (A) LMP1 mRNA expression was evaluated by RT-qPCR.(B) LMP1 stably knocked down and control PDL cells were seeded in 6-well-plates at low density (3× 10³/cm²). Osteogenic media were added to the cells, and media were changed every 3-4 days. At day 3, 6, and 9, cells were harvested and counted by hemocytometry, n=6 per group. (C) At day 10, cells were fixed and stained with crystal violet. (D) Subsequently, the crystal violet staining was washed and quantified. The optical density readout which correlates to cell numbers are shown. (E) LMP1 stably knocked down and control PDL cells were seeded in 12-well-plate at 3 × 10³/cm² and cultured in osteogenic induction media. ³H methyl thymidine was added after overnight attachment. At 5 d, the DNA was harvested and the ³H methyl thymidine incorporation was measured by scintillation counter. (n=4 per group). (F) PDL cells were cultured in 6-well-plates in serum free medium. 10% FBS was added, and RT-qPCR was used to examine the expression of Cyclin D1 and Cyclin B1 (n=3 per group).

Fig. 3. LMP1 silencing decreases osteogenic differentiation in PDL cells. LMP1

stably knocked down and control PDL cells were seeded in 12-well-plates at low density (3 × 10³/cm²). Osteogenic medium was added to the cells, and media were changed every 3-4 days. (A) At indicated time points, ALP activity was measured by ALP staining (left panel), and quantified assay (right panel). (B) Mineralization was assessed by Alizarin Red staining (left), and extracellular calcium concentration was quantified (right). (C) RT-qPCR was performed at d 7, 14, and 21 to evaluate the gene expression of several gene markers (C). a: p<0.01 compared to scramble shRNA in the same time point; b: p<0.05 compared to scramble shRNA in the same time point. n=3 per group.

Fig. 4. PDZ and ww-interacting domains are not enough to induce PDL cell proliferation

Full length LMP1 cDNA and a truncated form without any LIM domain were constructed into retroviral expression vector. PDL cells were transfected by retrovirus and selected by G418 for 10 days. Survived cells were pooled for the following experiments. Stable cell lines overexpressing LMP1 and LMP1-t were established in PDL cells from two different individuals. Representative data from 1 patient were shown here. (A) Truncated LMP1 only contains the first 144 aa including PDZ and ww interacting domains. (B) Endogenous and exogenous LMP1 proteins were detected by western blot. This antibody can detect the truncated form LMP1-t as well. (C) PDL cells were seeded in 6-well-plates at low density (3 × 10³/cm²). Cells were harvested by trypsin and counted using hemocytometry, n=6 per group. (D) PDL cells were seeded in 6-well-plate at 3 × 10³/cm² and ³H methyl thymidine was added. At 5 d, the DNA was harvested and the ³H methyl thymidine incorporation was measured by scintillation counter. (n=4 per group)

Fig. 5. LMP1 knockdown attenuates the TGF-β1 effect on PDL cells proliferation

(A) PDL cells were stimulated with TGF-β1 (2ng/ml), BMP2 (100ng/ml), or BMP6 (100ng/ml). phosphorylated-Smad1/5/8 and phosphorylated-Smad2 was examined by Western blot. (B) RNA was extracted after 24 h and RT-qPCR was used to evaluate LMP1 gene expression. LMP1 mRNA expression values was normalized to 18s RNA relative to that of serum-free controls. (C) After incubation in serum-free medium for 24 hours, human PDL cells were stimulated with TGF-β1 (2 ng/ml, 10 ng/ml, and 20ng/ml). 24 hours later, RNA was extracted and RT-qPCR was use to measure the expression level of LMP1. **: P<0.01 compared to serum free control; *: P<0.05 compared to serum free control. n=3 per group. (D) PDL cells were treated by TGF-β1 at 2 ng/ml. At various time points, LMP1 mRNA expression was measured by RT-qPCR. Open bar: PDL cells in serum free condition without TGF-β1. Closed bar: with TGF-β1. **: p<0.01 compared to serum-free control at the same time point. (E) PDL cells were cultured in DMEM with low serum concentration (2% FBS), and some cells were treated by TGF-β1. At day 3 and day 6, cells were fixed and stained with crystal violet. The optical density readout which correlates to cell numbers are shown, n=3 per group. *: p<0.05, **: p<0.01. (F) LMP1 knockdown PDL cells and control cells were cultured in 6-well-plates in 2% FBS, with/without TGF-β1. At day 3, 6, 9, cells were fixed and stained with crystal violet. The data showed the percentage change of crystal violet measurement between no TGF-β1 and TGF-β1 treatment (n=3 per group).

Fig. 6. TAK1-JNK/p38 cascade is involved in TGF-β1 induction of LMP1

(A) Confluent PDL cells were free for serum for 24 h. Prior to adding TGF-β1 (2 ng/ml), PDL cells were pretreated by SB431542 for 1 h. 24 h after TGF-β1 treatment, LMP1 gene expression was measured by RT-qPCR, normalized to 18s mRNA, and given relative to that of serum-free control. Control: serum-free. SB(1): SB431532 in DMSO, 1μM. SB(10): SB431532 in DMSO, 10μM. n.s.: no significant difference. (B) ALK5 siRNA was transfected into PDL cells for 72 h in serum-free media, followed by TGF-β1 stimulation. At 24 hours, RNA was extracted and qRT-PCR was used to examine the expression of ALK5 and LMP1. ALK5 knockdown compromised the effect of TGF-β1 on LMP1 gene expression. (C) Before adding TGF-β1 (2 ng/ml), PDL cells were pretreated by MAPK kinase inhibitors for 2 h. 24 h after TGF-β1 treatment, LMP1 gene expression was measured by RT-qPCR. DMSO was used as the solvent for all the inhibitors. SB203580: p38 inhibitor, 25μM. SP600125: JNK inhibitor, 25μM. (D) PDL cells were transiently transfected with 100nM siRNA (targeting TAK1 or scramble control) for 72 h, in serum-free and antibiotics-free DMEM. After treatment, media were changed (serum and antibiotics-free), and TGF-β1 was added. After 24 h, RT-qPCR was performed to measure gene expression. Left panel: TAK1 gene expression. Right panel: LMP1 gene expression. siCtrl: scramble siRNA; siCtrl + T: scramble siRNA and TGF-β1 treatment; siTAK1: TAK1 siRNA; siTAK1 + T: TAK1 siRNA and TGF-β1 treatment. (E) Confluent PDL cells were cultured in serum-free media for 24 h. Before adding TGF-β1 (2 ng/ml), PDL cells were pretreated by cycloheximide (CHX) for 2 h. 24 h after TGF-β1 treatment, LMP1 gene expression was measured by RT-qPCR, normalized to 18S mRNA, and given relative to that of serum free control. CHX(10): cycloheximide 10μM; CHX(25): cycloheximide 25μM. (F) Schematic overview of the regulation of LMP1 gene expression by TGF-β1. (n=3 per group in each experiment).