Endoglin differentially regulates TGF-ß-induced Smad2/3 and Smad1/5 signalling and its expression correlates with extracellular matrix production and cellular differentiation state in human chondrocytes.

OBJECTIVE: Transforming growth factor-ß (TGF-ß) plays a critical role in cartilage homeostasis and deregulation of its signalling is implicated in osteoarthritis (OA). TGF-ß isoforms signal through a pair of transmembrane serine/threonine kinases known as the type I and type II TGF-ß receptors. Endoglin is a TGF-ß co-receptor that binds TGF-ß with high affinity in the presence of the type II TGF-ß receptor. We have previously shown that endoglin is expressed in human chondrocytes and that it forms a complex with the TGF-ß signalling receptors. However, the functional significance of endoglin expression in chondrocytes is unknown. Our objective was to determine whether endoglin regulates TGF-ß/Smad signalling and extracellular matrix (ECM) production in human chondrocytes and whether its expression varies with chondrocyte differentiation state. METHOD: Endoglin function was determined by overexpression or antisense morpholino/siRNA knockdown of endoglin in human chondrocytes and measuring TGF-ß-induced Smad phosphorylation, transcriptional activity and ECM production. Alterations in endoglin expression levels were determined during subculture-induced dedifferentiation of human chondrocytes and in normal vs OA cartilage samples. RESULTS: Endoglin enhances TGF-ß1-induced Smad1/5 phosphorylation and inhibits TGF-ß1-induced Smad2 phosphorylation, Smad3-driven transcriptional activity and ECM production in human chondrocytes. In addition, the enhancing effect of endoglin siRNA knockdown on TGF-ß1-induced Smad3-driven transcription is reversed by ALK1 overexpression. Furthermore, endoglin levels are increased in chondrocytes following subculture-induced dedifferentiation and in OA cartilage as compared to normal cartilage. CONCLUSION: Together, our results suggest that endoglin regulates the balance between TGF-ß/ALK1/Smad1/5 and ALK5/Smad2/3 signalling and ECM production in human chondrocytes and that endoglin may represent a marker for chondrocyte phenotype.

Expression and function of TbetaRII-B, a variant of the type II TGF-beta receptor, in human chondrocytes.

OBJECTIVE: Transforming growth factor-beta (TGF-beta) has profound effects on chondrocyte proliferation and matrix production, and dysregulation of TGF-beta action has been implicated in osteoarthritis. The mechanisms by which the diverse actions of TGF-beta are regulated in chondrocytes are unclear. Although it is well documented that TGF-beta signaling is transduced by types I and II receptors, other TGF-beta receptors may play critical roles by regulating signaling receptor activity. Our objective was to examine the expression of TbetaRII-B, a splice variant of the type II TGF-beta receptor, and to analyze its role in regulating TGF-beta signaling in human chondrocytes. METHODS: TbetaRII-B expression was examined in human cartilage tissue specimens, human chondrocyte cell lines C28/I2 and tsT/AC62, and human primary chondrocytes by Western blot and reverse-transcriptase-polymerase chain reaction. Ligand binding and heteromerization of TbetaRII-B with other TGF-beta receptors on the cell surface were analyzed by affinity labeling, immunoprecipitation, and two-dimensional SDS-PAGE. Regulation of TGF-beta responses by TbetaRII-B was determined by examining Smad2 phosphorylation, Smad3-specific signaling, transcriptional activity, and type II collagen levels. RESULTS: TbetaRII-B is expressed in normal and osteoarthritic human cartilage. Furthermore, it is a dynamic component of the TGF-beta receptor system in human chondrocytes, forming heteromeric complexes with the types I and II TGF-beta receptors, betaglycan and endoglin. Importantly, overexpression of TbetaRII-B leads to enhanced TGF-beta signaling and responses in chondrocytes. CONCLUSIONS: These results suggest that TbetaRII-B may play a key role in the regulation of TGF-beta action in human chondrocytes.

Effect of T(3) treatment and food ration on hepatic deiodination and conjugation of thyroid hormones in rainbow trout, Oncorhynchus mykiss.

We studied the 7-day effects of 3,5,3'-triiodothyronine (T(3)) hyperthyroidism (induced by 12 ppm T(3) in food) and food ration (0, 0.5, or 2% body weight/day) on in vitro hepatic glucuronidation, sulfation, and deiodination of thyroxine (T(4)), T(3), and 3,3', 5'-triiodothyronine (rT(3)). T(3) treatment doubled plasma T(3) with no change in plasma T(4), depressed hepatic low-K(m) (1 nM) outer-ring deiodination (ORD) of T(4), induced low-K(m) (1 nM) inner-ring deiodination (IRD) of both T(4) and T(3) but did not alter high-K(m) (1 microM) rT(3)ORD, glucuronidation, or sulfation of T(4), T(3), or rT(3). Plasma T(4) levels were greater for 0 and 2% rations than for a 0.5% ration. Fasting decreased low-K(m) T(4)ORD activity and increased high-K(m) rT(3)ORD activity but did not alter T(4)IRD or T(3)IRD activities. T(4), T(3), and rT(3) glucuronidation were greater for 0 and 0.5% rations than for a 2% ration. T(3) glucuronidation was greater for a 0.5% ration than for a 0% ration. T(3) and rT(3) sulfation were greater for a 2% ration than for a 0 or a 0.5% ration; ration did not change T(4) sulfation. We conclude that (i) modest experimental T(3) hyperthyroidism induces T(3) autoregulation by adjusting hepatic low-K(m) ORD and IRD activities but not high-K(m) rT(3)ORD or conjugation activities; (ii) in contrast, ration level changes both deiodination and conjugation pathways, suggesting that the response to ration does not solely reflect altered T(3) production; (iii) deiodination and conjugation appear complementary in regulating thyroidal status in response to ration; and (iv) high-K(m) rT(3)ORD in trout differs from rat type I deiodination in that it does not respond to T(3) hyperthyroidism and it increases, rather than decreases, its activity during fasting.

Deiodination and deconjugation of thyroid hormone conjugates and type I deiodination in liver of rainbow trout, Oncorhynchus mykiss.

We studied the hepatic in vitro deconjugation and deiodination of glucuronide (G) and sulfate (S) conjugates of the thyroid hormones (TH) thyroxine (T(4)), 3,5,3'-triiodothyronine (T(3)), and 3,3', 5'-triiodothyronine (rT(3)) in trout. These conversions have not been studied in nonmammals. Deconjugation of T(4)G, T(3)G, rT(3)G, or rT(3)S was negligible in all subcellular fractions. Some T(4)S desulfation occurred but T(3)S was desulfated to the greatest extent by freshly isolated hepatocytes and by the mitochondrial/lysosomal and microsomal fractions. Deiodination of T(4)G, T(3)G, rT(3)G, T(4)S, T(3)S, and rT(3)S (1 or 1000 nM) was negligible in control trout and in trout treated with T(3) to induce inner-ring deiodination (IRD) but simultaneously tested rat microsomes rapidly deiodinated T(4)S, T(3)S, and rT(3)S. Furthermore, T(4)S, T(3)S, and rT(3)S (1-100 nM) were less effective than their unsulfated forms in competitively inhibiting trout hepatic outer-ring deiodination (ORD) of T(4) (0.8 nM), and rT(3)ORD (100 nM) was not competitively inhibited by T(4)S, T(3)S, or rT(3)S (100 nM) or by T(4) or T(3) (1 microM). Thus, there is no evidence in trout liver for THS deiodination, which is a key property of rat type I deiodination. We therefore studied other properties of trout hepatic high-K(m) deiodination, which has been considered homologous to rat type I deiodination. We found that it resembled rat type I deiodination in its rT(3)ORD ability, its optimum pH (7.0), and its requirement for dithiothreitol (DTT). However, it differed from rat type I deiodination not only in its negligible deiodination of T(4) and THS but also in its low DTT optimum (2.5 mM), its low apparent K(m) for rT(3) (200 nM), its lack of IRD ability, its extremely weak propylthiouracil inhibition (IC(50), 1 mM), its weaker inhibition by iodoacetate (IC(50), 10 microM) and aurothioglucose (IC(50), <3 microM), its activation by fasting, and its unresponsiveness to T(3) hyperthyroidism. We conclude that most conjugated TH are neither deconjugated nor deiodinated by trout liver and are therefore eliminated in bile. However, T(3)S can be desulfated. Substrate preference and other properties suggest that trout hepatic high-K(m) ORD shares some properties with rat type I deiodination but differs functionally in several other respects and may contribute negligibly to hepatic T(3) production in trout.

Sulfation of thyroid hormones by liver of rainbow trout, Oncorhynchus mykiss.

We studied hepatic sulfation of thyroid hormones (TH) in rainbow trout, Oncorhynchus mykiss. Sulfation of thyroxine (T4) and 3,5,3'-triiodothyronine (T3) was detected in the cytosolic (63-67%), microsomal (12-16%), nuclear (12-14%) and mitochondrial/lysosomal (7-8%) fractions. Using 3'-phosphoadenosine 5'-phosphosulfate (PAPS) as a sulfate donor, sulfation of T4 and T3 by the cytosolic fraction depended on protein concentration and time. The pH profiles for T4- and T3-sulfation were broad and overlapping with optimal pH values of about 6.5 and 7.0 U respectively. At pH 7.0, apparent K(m) (microM), Vmax (pmol/mg cytosolic protein per hour) and catalytic efficiency (Vmax/K(m)) values were 3,5',3'-triiodothyronine (reverse T3, rT3) = 0.7, 583 and 832; T4 = 1.7, 46 and 27; T3 = 11.5, 840 and 73. Inhibitor profiles for both T4- and T3-sulfation were not significantly different with a common inhibitor preference of rT3 > pentachlorophenol > triiodothyroacetic acid > tetraiodothyroacetic acid T4 = T3 = 3,5-diiodothyronine. T4-, T3- and rT3-sulfation activity decreased with increasing pre-incubation temperature (12, 24, 36 degrees C); however, there were no significant differences in T4-, T3- and rT3-sulfation activity at each pre-incubation temperature. We conclude that: (i) in trout, hepatic sulfation of TH is enzymatic and obeys Michaelis-Menten kinetics; (ii) like mammalian hepatic sulfotransferases (STs), trout hepatic STs are heat-sensitive cytosolic proteins using PAPS as a sulfate donor; (iii) unlike mammalian sulfation of TH, trout hepatic sulfation of T4, T3 and rT3 may be catalyzed by a single form of ST preferring rT3 as substrate and with a catalytic efficiency of rT3 >>> T3 > T4.

Glucuronidation of thyroxine and 3,5,3'-triiodothyronine by hepatic microsomes in rainbow trout, Oncorhynchus mykiss.

We studied the glucuronidation of thyroxine (T4) and 3,5,3'-triiodothyronine (T3) in the hepatic microsomal fraction of rainbow trout, Oncorhynchus mykiss. Uridine diphosphoglucuronosyl transferase (UDPGT) activity toward T4 depended on both protein concentration (linear up to about 0.75 mg/ml) and time (linear up to 60 min). The optimal pH for glucuronidation was 6.8-7.8 units for T4 and > or = 8.5 units for T3. At a common pH of 7.8, the apparent Km and Vmax values were, respectively, 6.0 muM and 42 nmol/mg protein/hr for T4, and 1.1 muM and 4.3 nmol/mg protein/hr for T3. At concentrations of 100 muM, T4 inhibited T3-glucuronidation, but T3 did not inhibit T4-glucuronidation. T4-glucuronidation was inhibited by 3,3', 5'-triiodothyronine (rT3) and tetraiodothyroacetic acid (Tetrac); T3-glucuronidation was inhibited by rT3, T4, Tetrac, and triiodothyroacetic acid. Therefore, analogues with two outer-ring iodines were the most effective inhibitors, showing that outer-ring iodine substitution influences UDPGT substrate specificity. Following a 15-min pre-incubation at 24 degrees C, UDPGT thermal stability was higher for T4 than T3. Polyoxyethylene 10 cetyl ether (Brij 56) maximally stimulated UDPGT activity for both T4 and T3 about two-fold at 0.0125% (w/v) detergent, suggesting that the UDPGTs are transmembrane proteins with the active site facing the lumen of the microsomal vesicles. Clofibrate did not affect either T4- or T3-UDPGT activity. On a per mg microsomal protein basis, T4-glucuronidation in the rat liver exceeded that in the trout 3-fold. We conclude that (a) trout liver microsomes have more than one form of UDPGT for the glucuronidation of T4 and T3 and (b) these trout UDPGTs share several properties of with those of the rat.

Identification of thyroid hormone conjugates produced by isolated hepatocytes and excreted in bile of rainbow trout, Oncorhynchus mykiss.

We studied thyroid hormone (TH) conjugation in fasted trout by incubating isolated hepatocytes with either [125I]T4 or [125I]T3, and by analyzing bile from trout injected with either [125I]T4 or [125I]T3. Glucuronide conjugates were identified by hydrolysis with beta-glucuronidase and sulfate conjugates by acid solvolysis with ethyl acetate/trifluoroacetic acid (1%). We used Sephadex LH-20 chromatography to concentrate the conjugate fractions from hepatocyte incubates prior to HPLC analysis. Glucuronide conjugates of T4 and T3 were produced in vitro and glucuronides of T4, T3, and 3,3'-T2 were found in vivo. Sulfation of T4 occurred in vitro and in vivo. T3 sulfation was not established in vitro, but sulfate conjugates of T3 and T2 were found in bile. Significant proportions of unconjugated T4 and T3 also occurred in the bile. We conclude that (i) as in other vertebrates, iodothyronines undergo hepatic glucuronidation and are excreted as such in the bile, (ii) T4 and T3 undergo sulfation, and in contrast to mammals, are excreted in significant amounts in the bile, (iii) 3,3'-T2, a prominent deiodination product of T3, is excreted as both glucuronide and sulfate conjugates, and (iv) the isolated hepatocyte system is appropriate for studying aspects of TH metabolism in trout.