Cell-cycle-dependent expression of transforming growth factor beta type I receptor correlates with differential proliferative effects of TGFbeta1 in articular chondrocytes

TGFBR2 mediated phosphorylation of BUB1 at Ser-318 is required for transforming growth factor-β signaling

BUB1 (budding uninhibited by benzimidazoles-1) is required for efficient TGF-β signaling, through its role in stabilizing the TGFBR1 and TGFBR2 complex. Here we demonstrate that TGFBR2 phosphorylates BUB1 at Serine-318, which is conserved in primates. S318 phosphorylation abrogates the interaction of BUB1 with TGFBR1 and SMAD2. Using BUB1 truncation domains (1–241, 241–482 and 482–723), we demonstrate that multiple contact points exist between BUB1 and TGF-β signaling components and that these interactions are independent of the BUB1 tetratricopeptide repeat (TPR) domain. Moreover, substitutions in the middle domain (241–482) encompassing S318 reveals that efficient interaction with TGFBR2 occurs only in its dephosphorylated state (241–482 S318A). In contrast, the phospho-mimicking mutant (241–482 S318D) exhibits efficient binding with SMAD2 and its over-expression results in a decrease in TGFBR1-TGFBR2 and TGFBR1-SMAD2 interactions. These findings suggest that TGFBR2 mediated BUB1 phosphorylation at S318 may serve as a switch for the dissociation of the SMAD2-TGFBR complex, and therefore represents a regulatory event for TGF-β signaling. Finally, we provide evidence that the BUB1-TGF-β signaling axis may mediate aggressive phenotypes in a variety of cancers.

A role for age-related changes in TGFbeta signaling in aberrant chondrocyte differentiation and osteoarthritis

Transforming growth factor beta (TGFβ) is a growth factor with many faces. In our osteoarthritis (OA) research we have found that TGFβ can be protective as well as deleterious for articular cartilage. We postulate that the dual effects of TGFβ on chondrocytes can be explained by the fact that TGFβ can signal via different receptors and related Smad signaling routes. On chondrocytes, TGFβ not only signals via the canonical type I receptor ALK5 but also via the ALK1 receptor. Notably, signaling via ALK5 (Smad2/3 route) results in markedly different chondrocyte responses than ALK1 signaling (Smad1/5/8), and we postulate that the balance between ALK5 and ALK1 expression on chondrocytes will determine the overall effect of TGFβ on these cells. Importantly, signaling via ALK1, but not ALK5, stimulates MMP-13 expression by chondrocytes. In cartilage of ageing mice and in experimental OA models we have found that the ALK1/ALK5 ratio is significantly increased, favoring TGFβ signaling via the Smad1/5/8 route, changes in chondrocyte differentiation and MMP-13 expression. Moreover, human OA cartilage showed a significant correlation between ALK1 and MMP-13 expression. In this paper we summarize concepts in OA, its link with ageing and disturbed growth factor responses, and a potential role of TGFβ signaling in OA development.

Chondrocyte proliferation in a new culture system

OBJECTIVE

This study has aimed to study different culture systems that might stimulate an increase in cell proliferation of normal and osteoarthritis chondrocytes from articular cartilage in rat model.

MATERIAL AND METHODS

Three culture systems using chondrocytes embedded in alginate beads were tested: chondrocytes cultured in Dulbecco's modified Eagle's medium (DMEM) as control, a co-culture system consisting of a monolayer of de-differentiated chondrocytes as a source of mitotic factors, and an enriched medium containing culture medium obtained from a monolayer of chondrocytes and DMEM. Normal and osteoarthritis chondrocytes were stained with 5-carboxyfluorescein diacetate succinimidyl ester and were cultured in each of the three systems. After 5 days of culture cell, proliferation was detected by flow cytometry. Chondrocyte phenotype was confirmed by collagen type II and MMP-3 expression. To determine possible molecules released into the medium by the cultured chondrocyte monolayer and which would probably be involved in cell proliferation, a study of mRNA and expression of transforming growth factor-beta1 (TGF-beta1), fibroblastic growth factor-2 (FGF-2), epidermal growth factor (EGF), platelet derived growth factor-A (PDGF-A) and insulin-like growth factor-1 (IGF-1) proteins was conducted.

RESULTS AND CONCLUSIONS

Chondrocytes in the co-culture system or in enriched medium showed an increase in proliferation; only when osteoarthritis chondrocytes were cultured in enriched medium would they display a statistically significant increase in their proliferation rate and in their viability. When chondrocytes from the monolayer were analysed, differential mRNA expression of TGF-beta1 and IGF-1 was found during all passages, which suggests that these two growth factors might be involved in chondrocyte proliferation.

Effect of Stratified Culture Compared to Confluent Culture in Monolayer on Proliferation and Differentiation of Human Articular Chondrocytes

With conventional tissue culture of cells, it is generally assumed that when the available 2D substrate is fully occupied, growth ceases or is greatly reduced.However, in nature wound repair mostly involves proliferation of cells that are attracted to the defect site in a 3D environment.Hence, proliferation continues in 3D until the defect site is filled with cells contributing to repair tissue. With this in mind,we examined the growth behavior of human articular chondrocytes during stratified culture as opposed to routine culture to confluency. Additionally, we studied the influence of growth factors on proliferation during stratified culture and differentiation thereafter. Chondrocytes were cultured in monolayer on tissue culture plastic to confluency or stratified for an additional 7 days. Culture medium was based on DMEM with 10% serum and either supplemented with high concentrations of nonessential amino acids (NEAA) and ascorbic acid (AsAP), or instead with basic fibroblastic growth factor (bFGF), platelet-derived growth factor (PDBF-BB), and/or transforming growth factor beta1 (TGF-beta). After expansion, cells were harvested, counted, and their differentiation capacity was examined in pellet culture assay. It was shown that chondrocytes, cultured stratified proliferate exponentially for up to an additional 4 days and that cell yield increased 5-fold. Furthermore, during stratified culture the number of cells increased further in the presence of bFGF, PDBF-BB, and TGFbeta1 or high concentrations of NEAA and AsAP. Depending on donor variation and factors supplemented the cell yield ranged from 0.06 up to 1.1 million cells/cm2 at the second passage. During stratified culture in the presence of either bFGF and PDGF or high concentrations of NEAA and AsAP, exponential growth continued for up to 7 days. Finally, cells maintained their differentiation capacity when cultured stratified with or without growth factors (bFGF, TGF-beta, and PDGF), but not when cultured with high levels of AsAP and NEAA. In contrast to other 3D culture techniques like microcarrier or suspension culture, nutrient consumption remained the same as with conventional expansion. Because this allows culturing of clinically relevant amounts of chondrocytes without increasing the amount of serum, chondrocytes can be fully expanded in the presence autologous serum, avoiding the risk of viral and/or prion disease transmission associated with the use of animal-derived serum or serum replacers with animal-derived constituents.

Ontogeny of rat chondrocyte proliferation: studies in embryo, adult and osteoarthritic (OA) cartilage

The aim of this work was to study the ontogeny of chondrocyte cell division using embryo, adult and osteoarthritic (OA) cartilage. We searched for mitosis phases and performed a comparative evaluation of mitotic index, basic fibroblast growth factor b (FGFb), transforming growth factor beta1 (TGF-beta1) receptors, cyclin dependent kinase (CDK1) and Cyclin-B expression in fetal, neonate, 3, 5, 8 weeks old rats and experimental OA. Our results showed that mitosis phases were observed in all normal cartilage studied, although, we found a decrease in mitotic index in relation to tissue development. No mitosis was detected in OA cartilage. We also found a statistical significant reduction in cell number in OA cartilage, compared with the normal tissue. Furthermore, FGFb and TGF-beta1 receptors diminished in relation to tissue development, and were very scarce in experimental OA. Western blot assays showed CDK-1 expression in all cases, including human-OA cartilage. Similar results were observed for Cyclin-B, except for 8 weeks, when it was not expressed. Our results suggest that cell division seems to be scarce, if not absent within the OA cartilage studied. Nevertheless, the existence of factors essential for cell division leaves open the question concerning chondrocyte proliferation in OA cartilage, which is likely to be present in the early stages of the disease.

Immunohistochemical analysis of transforming growth factor beta isoforms and their receptors in human cartilage from normal and osteoarthritic femoral heads

OBJECTIVE

Osteoarthritis (OA) is characterized by erosion of cartilage and formation of osteophytes. Since transforming growth factor beta (TGF-beta) is known to be involved in chondrogenesis and osteogenesis, we studied by immunochemistry the expression of TGF-beta isoform types 1, 2, and 3 and their receptor types I and II in slightly and strongly altered areas of human OA cartilage and in osteophytes.

METHODS

Specimens were collected from femoral heads at the time of hip arthroplasty, selecting osteophytic regions and areas of slight or severe degradation according to the Mankin score. Cryostat sections were prepared and submitted to immunohistochemistry using appropriate antibodies to TGF-beta(1-3) and TGF-beta receptors I and II.

RESULTS

TGF-beta1 expression was shown to be depressed in strongly degraded cartilage, compared to normal and slightly altered areas. TGF-beta2 was barely detectable in all samples studied. In osteophytes, a marked overexpression of TGF-beta1 and -beta3 was observed. An important decrease in TGF-beta receptor II was found in fibrillated cartilage areas.

CONCLUSIONS

The three major isoforms of TGF-beta are expressed in human OA cartilage, albeit the TGF-beta2 level is very low. Their expression patterns and the ratio of receptors I and II varies according to the degree of OA severity. The decrease in TGF-beta1 production and marked downregulation of receptor II in fibrillated cartilage may lead to reduced chondrocyte responsiveness to TGF-beta and contribute to the irreversibility of the disease. Overexpression of TGF-beta1 and -beta3 in osteophytes suggests that the two isoforms are involved in the formation of these structures.

Fluid-induced shear stress stimulates chondrocyte proliferation partially mediated via TGF-beta1

There is growing evidence that a hydrodynamic environment is beneficial for growing cartilage tissue-engineered constructs; however, the mechanisms by which fluid shear provides for a better construct are not well understood. In this study, we investigated one possible mechanism by which constructs grow faster under fluid shear: fluid shear upregulates chondrocyte proliferation. Further, we investigated if this effect is mediated by TGF-beta1, a known mediator of fluid shear effects in other cell types and a mitogen for chondrocytes. To test the hypotheses, primary bovine articular chondrocytes were cultured in monolayers (approximately 40,000 cells/cm(2)) to 80-85% confluency. After 24 h of growth arrest, cells were exposed to 3.5 Pa fluid shear stress for 96 h. Total DNA was compared between flow and static culture slides. Total TGF-beta1 was quantified in flow-conditioned media (CM) and static culture-CM. Mitogenic capacity of the CM, with or without anti-TGF-beta1 or anti-TbetaRII (TGF beta receptor type II) antibodies, was also assessed. Results show that fluid shear significantly up-regulates chondrocyte proliferation (p < 0.02). Further, total TGF-beta1 in the flow-CM was more than 3.5-fold higher (p < 0.03) and its mitogenicity significantly higher (p < 0.007) as compared to static culture-CM. Adding excess anti-TGF-beta1 or anti-TbetaRII antibodies partially, but significantly depressed mitogenicity (approximately 20% decrease) of the flow-CM. These results show that fluid shear stress upregulates chondrocyte proliferation and that this effect is partially mediated by TGF-beta1.

The transcription factor SOX9 regulates cell cycle and differentiation genes in chondrocytic CFK2 cells

SOX9 is a transcription factor that is essential for chondrocyte differentiation and cartilage formation. We stably overexpressed SOX9 cDNA in the rat chondrocytic cell line CFK2. Compared with the vector control, a greater proportion of SOX9-transfected cells accumulated in the G0/G1 phase. This was associated with an increase in mRNA and protein expression of p21(cip1), an inhibitor of cyclin-dependent kinase activity. SOX9 enhanced p21(cip1) promoter activity in a luciferase reporter assay. CFK2 cells overexpressing SOX9 became more elongated and adhesive and demonstrated a shift in cytoplasmic F-actin distribution. N-cadherin mRNA levels were elevated in the SOX9-transfected cells, and SOX9 enhanced N-cadherin promoter activity. By electrophoretic mobility shift assay, nuclear extracts of SOX9-transfected CFK2 cells specifically bound an oligonucleotide comprising an N-cadherin promoter region containing a consensus SOX9-binding motif. The transcriptional activity of SOX9 depended upon nuclear localization signals required for SOX9 nuclear entry. Differentiation of transfected CFK2 cells was accelerated as evidenced by more rapid accumulation of alkaline phosphatase activity, increased production of proteoglycans, and increased calcium accumulation, and this was associated with decreased ERK1 expression. These studies demonstrate that SOX9 alters the rate of cell cycle progression of chondrocytes and their differentiation by enhancing or inhibiting the expression of selected genes, including p21(cip1) and ERK1, and that N-cadherin is an additional direct target of this transcriptional regulator.

Cell density-dependent proliferative effects of transforming growth factor (TGF)-beta 1, beta 2, and beta 3 in human chondrosarcoma cells HCS-2/8 are associated with changes in the expression of TGF-beta receptor type I

In this study, the growth properties of the human chondrosarcoma cell line HCS-2/8, its response to transforming growth factor (TGF)-beta isoforms 1, 2, and 3, and its expression of TGF-beta receptors I and II were examined. We demonstrated that these tumor cells are not contact-inhibited and that they can proliferate in the absence of additional serum growth factors. In sparse cultures, all TGF-beta forms inhibited the growth of HCS-2/8 cells, whereas they induced a 2-fold increase of DNA synthesis in serum-fed confluent cultures. In serum-free confluent conditions only TGF-beta 1 stimulated the proliferation rate, whereas TGF-beta 2 was without effect and TGF-beta 3 was rather inhibitory. This bimodal effect of TGF-beta forms was associated with a greater level of TGF-beta receptor 1 mRNA in confluent HCS-2/8 than in sparse cultures, suggesting that the growth response to TGF-beta forms is dependent on the receptor profile expressed.

Activin A and activin receptors in thyroid cancer

Proliferation is controlled by a network of mitogenic and growth inhibitory factors. Transforming growth factor-beta1 (TGF-beta1) and activin A are the most important growth inhibitors of benign follicular epithelial cells of the human thyroid. The effects of these substances on malignant primary thyrocytes are not known. We have examined the growth regulatory effects of activin A and TGF-beta1 in primary cultures derived from four papillary cancers, two follicular thyroid cancers, and three benign thyroid tissues. Malignant cells demonstrated resistance to activin and TGF-beta1 or reversal to a weak but significant mitogenic effect (p < 0.001). We also evaluated the activin receptor transcription pattern. Isoforms alk4-1, 4-2, and 4-3 were found in benign (n = 12) and malignant (n = 22) tissues. Two subtypes of type I and type II activin receptors were demonstrated. Semiquantitative reverse transcription-polymerase chain reaction (RT-PCR) demonstrated a significant threefold downregulation of alk4-1 receptors in papillary (n = 25) and follicular (n = 18) thyroid cancers as compared to normal thyroids (n = 12) (p < 0.001). To our knowledge these are the first data to demonstrate reversal of activin and TGF-beta1 effects in thyroid malignancy and to demonstrate changes of the type Ib activin receptor expression in thyroid malignancy.