Maeda S, Dean D D, Gomez R, Schwartz Z, Boyan B D
The University of Texas Health Science Center at San Antonio, San Antonio, TX 78229-3900, USA.
Calcif Tissue Int. 2002 Jan;70(1):54-65. doi: 10.1007/s002230010032. Epub 2001 Dec 21.
Transforming growth factor beta-1 (TGF-beta1) is secreted in a biologically inactive form and stored in the extracellular matrix as a 290 kDa complex consisting of the mature TGF-beta1 homodimer (Mr 25 kDa), the latency-associated peptide (LAP; Mr 75 kDa), and the latent TGF-beta1 binding protein-1 (LTBP1; Mr 190 kDa). Latent TGF-beta1, composed of these three components, is known as the "large latent TGF-beta1 complex." In contrast, latent TGF-beta1 without LTBP1 is known as "small latent TGF-beta1." For all latent forms, dissociation of the TGF-beta1 homodimer from LAP is necessary for growth factor activation and acquisition of biological activity. Matrix vesicles produced by growth plate chondrocytes contain matrix metalloproteinases that can activate small latent TGF-beta1. The enzyme responsible for this is matrix metalloproteinase-3 (MMP-3), although matrix vesicles also contain MMP-2 and plasminogen activator. The present study tested the hypothesis that matrix vesicle enzymes are also involved in the release of the large latent TGF-beta1 complex stored in the extracellular matrix. Matrix vesicles were isolated from cultures of resting zone and growth zone chondrocytes and metalloproteinases present in the matrix vesicles extracted with guanidine-HCl. Chondrocyte extracellular matrices were prepared by lysing confluent cultures and removing the lysed cells. The matrices were incubated with matrix vesicle extracts and the release of total and active TGF-beta1 was determined. To determine if MMP-2 or MMP-3 was involved in the release, matrix vesicle extracts were preincubated with anti-MMP-2 antibody or anti-MMP-3 antibody to selectively deplete the enzyme activity. Matrices were also treated with rhMMP-2 or rhMMP-3. To determine the identity of the released protein(s), digests were separated on SDS-polyacrylamide gels and Western blotting analysis was performed using a specific antibody to LTBP1. Matrix vesicle extracts released both active and total (=latent + active) TGF-beta1 in a time-dependent manner, with peak release after 1 hour of incubation. The amount of total TGF-beta1 released was 10 times higher than the release of active TGF-beta1. The effect of the matrix vesicle extracts was dose-dependent; in addition, the amount and ratio of active to total TGF-b1 released was very similar, irrespective of the source of matrix or matrix vesicle extracts. Pre-incubation of matrix vesicle extracts with anti-MMP-3 antibody blocked the release of active and total TGF-beta1, whereas pre-incubation with pre-immune IgG or anti-MMP-2 antibody had no effect. The addition of rhMMP-3, but not rhMMP-2, caused a dose-dependent increase in the release of total, but not active, TGF-beta1. Western analysis confirmed that both matrix vesicle extracts and rhMMP-3 released the large latent TGF-beta1 complex from the matrix. In addition to the expected 290, 230, and 190 kDa bands, samples run without reduction also contained proteins of molecular weights 110 and 50 kDa that reacted with the anti-LTBP1 antibody. When these same samples were electrophoresed after reduction, the high molecular weight immunoreactive bands disappeared and three bands of molecular weight 75, 32, and 25 kDa were observed. These results indicate that matrix vesicles contain enzymes, especially MMP-3, which are responsible for the release of TGF-beta1 from the matrix, most of which is in latent form. Further, the data suggest that release of the large complex occurs via cleavage at several novel sites in the 130 kDa LTBP1 molecule. Since matrix vesicle MMP-3 is also able to activate small latent TGF-beta1, these results suggest that the large latent TGF-beta1 complex protects against activation of the small latent TGF-beta1. Thus, the data suggest that release of the large latent TGF-bl complex from the matrix and activation of the latent growth factor are only two steps of what must be at least a three-step process.
转化生长因子β-1(TGF-β1)以生物无活性形式分泌,并作为一种290 kDa的复合物储存于细胞外基质中,该复合物由成熟的TGF-β1同二聚体(Mr 25 kDa)、潜伏相关肽(LAP;Mr 75 kDa)和潜伏TGF-β1结合蛋白-1(LTBP1;Mr 190 kDa)组成。由这三种成分组成的潜伏TGF-β1被称为“大潜伏TGF-β1复合物”。相比之下,不含LTBP1的潜伏TGF-β1被称为“小潜伏TGF-β1”。对于所有潜伏形式而言,TGF-β1同二聚体与LAP解离是生长因子激活和获得生物活性所必需的。生长板软骨细胞产生的基质小泡含有能激活小潜伏TGF-β1的基质金属蛋白酶。负责此作用的酶是基质金属蛋白酶-3(MMP-3),尽管基质小泡中也含有MMP-2和纤溶酶原激活剂。本研究检验了基质小泡酶也参与细胞外基质中储存的大潜伏TGF-β1复合物释放的假说。从静止区和生长区软骨细胞培养物中分离出基质小泡,并用盐酸胍提取基质小泡中存在的金属蛋白酶。通过裂解汇合培养物并去除裂解细胞来制备软骨细胞外基质。将这些基质与基质小泡提取物一起孵育,并测定总TGF-β1和活性TGF-β1的释放量。为了确定MMP-2或MMP-3是否参与释放过程,将基质小泡提取物与抗MMP-2抗体或抗MMP-3抗体预孵育以选择性地耗尽酶活性。基质也用重组人MMP-2(rhMMP-2)或重组人MMP-3(rhMMP-3)处理。为了确定释放的蛋白质的身份,将消化产物在SDS-聚丙烯酰胺凝胶上分离,并使用抗LTBP1特异性抗体进行蛋白质印迹分析。基质小泡提取物以时间依赖性方式释放活性和总(=潜伏+活性)TGF-β1,孵育1小时后释放达到峰值。释放的总TGF-β1量比活性TGF-β1的释放量高10倍。基质小泡提取物的作用呈剂量依赖性;此外,无论基质或基质小泡提取物的来源如何,释放的活性TGF-β1与总TGF-β1的量和比例非常相似。将基质小泡提取物与抗MMP-3抗体预孵育可阻断活性和总TGF-β1的释放,而与免疫前IgG或抗MMP-2抗体预孵育则无影响。添加rhMMP-3而非rhMMP-2会导致总TGF-β1(而非活性TGF-β1)释放呈剂量依赖性增加。蛋白质印迹分析证实,基质小泡提取物和rhMMP-3均能从基质中释放大潜伏TGF-β1复合物。除了预期的290、230和190 kDa条带外,未还原情况下运行的样品还含有分子量为110和50 kDa且与抗LTBP1抗体反应的蛋白质。当这些相同样品在还原后进行电泳时,高分子量免疫反应条带消失,观察到分子量为75、32和25 kDa的三条条带。这些结果表明,基质小泡含有酶,尤其是MMP-3,其负责从基质中释放TGF-β1,其中大部分以潜伏形式存在。此外,数据表明大复合物的释放是通过在130 kDa LTBP1分子中的几个新位点处切割而发生的。由于基质小泡MMP-3也能够激活小潜伏TGF-β1,这些结果表明大潜伏TGF-β1复合物可防止小潜伏TGF-β1的激活。因此,数据表明从基质中释放大潜伏TGF-β1复合物和潜伏生长因子的激活只是至少一个三步过程中的两个步骤。