Contreras Rubén G, Shoshani Liora, Flores-Maldonado Catalina, Lázaro Amparo, Monroy Alma O, Roldán María Luisa, Fiorentino Rosana, Cereijido Marcelino
Center for Research and Advanced Studies, Dept. of Physiology, Biophysics and Neurosciences, Av. Instituto Politécnico Nacional 2508, Mexico, DF 07300, Mexico.
Pflugers Arch. 2002 Jul;444(4):467-75. doi: 10.1007/s00424-002-0827-8. Epub 2002 Apr 5.
The assembly and permanent sealing of tight junctions (TJs) depend crucially on cell-cell contacts containing E-cadherin. This poses a puzzling problem because, while TJs can be established between epithelial cells from different tissues and even different animal species ("heterotypic TJs"; Gonzalez-Mariscal et al. 1989, J Membr Biol 107:43), the cell-cell binding mediated by E-cadherin is a highly specific one (Takeichi 1995, Curr Opin Cell Biol 7:619). Yet the demonstration that TJs can be established at heterotypic borders is open to two distinct challenges. First, it is based on transepithelial electrical resistance (TER) and restriction to ruthenium red permeation only, which today are known to be just two of the many characteristics of TJs; and second some attributes of the TJs (e.g. the presence of specific molecules) have been found even in cells that do not establish these structures. This raised the question of whether heterotypic TJs were not true or full TJs. In the present work we demonstrate that heterotypic TJs in mixed monolayers of MDCK cells with a different cell type (LLC-PK1) are true TJs through several criteria, such as TER, the ability to stop the membrane diffusion of fluorescent sphingomyelin from the apical to the lateral domain, the presence of ZO-1, ZO-2, occludin, claudin-1 and claudin-2. We then turn to the presence of E-cadherin at heterotypic borders, and observe that it cannot be detected by the highly specific DECMA-1 antibody, in spite of the fact that this antibody does reveal the presence of E-cadherin at homotypic contacts of the same cell. Yet, ECCD-2, an antibody against another domain of E-cadherin, reveals that this molecule may be present at both types of borders. Thus, E-cadherin is present at heterotypic borders, yet it seems to be in a conformation unable to bind DECMA-1. Our results suggest: (1) that heterotypic borders can establish fully developed TJs; (2) that the sealing of these heterotypic TJs depends on E-cadherin; (3) but that this dependence is mediated through a cascade of chemical reactions involving two different G-proteins, PLC, PKC and calmodulin, which we have characterized elsewhere (Balda et al. 1991, J Membr Biol 122:193); and (4) hence molecules of E-cadherin that trigger junction formation can act from a distant homotypic contact.
紧密连接(TJ)的组装和永久密封关键取决于包含E-钙黏蛋白的细胞间接触。这带来了一个令人困惑的问题,因为尽管不同组织甚至不同动物物种的上皮细胞之间可以建立紧密连接(“异型紧密连接”;冈萨雷斯-马里斯卡尔等人,1989年,《膜生物学杂志》107:43),但由E-钙黏蛋白介导的细胞间结合是高度特异性的(竹内,1995年,《细胞生物学当前观点》7:619)。然而,异型边界处可建立紧密连接这一观点面临两个不同的挑战。首先,它仅基于跨上皮电阻(TER)和对钌红渗透的限制,而如今已知这只是紧密连接众多特征中的两个;其次,即使在未形成这些结构的细胞中也发现了紧密连接的一些属性(例如特定分子的存在)。这就引发了异型紧密连接是否并非真正或完整紧密连接的问题。在本研究中,我们通过多个标准证明,MDCK细胞与另一种不同细胞类型(LLC-PK1)的混合单层中的异型紧密连接是真正的紧密连接,这些标准包括TER、阻止荧光鞘磷脂从顶端向侧面区域膜扩散的能力、ZO-1、ZO-2、闭合蛋白、紧密连接蛋白-1和紧密连接蛋白-2的存在。然后我们转向异型边界处E-钙黏蛋白的存在情况,观察到尽管该抗体确实能揭示同一细胞同型接触处E-钙黏蛋白的存在,但高度特异性的DECMA-1抗体却无法检测到异型边界处的E-钙黏蛋白。然而,针对E-钙黏蛋白另一个结构域的抗体ECCD-2表明,该分子可能存在于两种类型的边界处。因此,E-钙黏蛋白存在于异型边界处,但似乎处于无法结合DECMA-1的构象。我们的结果表明:(1)异型边界可以建立发育完全的紧密连接;(2)这些异型紧密连接的密封依赖于E-钙黏蛋白;(3)但这种依赖性是通过涉及两种不同G蛋白、磷脂酶C、蛋白激酶C和钙调蛋白的一系列化学反应介导的,我们在其他地方已对其进行了表征(巴尔达等人,1991年,《膜生物学杂志》122:193);(4)因此,触发连接形成的E-钙黏蛋白分子可以从远处的同型接触起作用。