The presequence translocase of the inner mitochondrial membrane (TIM23 complex) operates at a central junction of protein import. It accepts preproteins from the outer membrane TOM complex and directs them to inner membrane insertion or, in cooperation with the presequence translocase-associated motor (PAM), to the matrix. Little is known of how the TIM23 complex coordinates these tasks. We have identified Tim21 (YGR033c) that interacts with the TOM complex. Tim21 is specific for a TIM23 form that cooperates with TOM and promotes inner membrane insertion. Protein translocation into the matrix requires a switch to a Tim21-free, PAM bound presequence translocase. Tim17 is crucial for the switch by performing two separable functions: promotion of inner membrane insertion and binding of Pam18 to form the functional TIM-PAM complex. Thus, the presequence translocase is not a static complex but switches between TOM tethering and PAM binding in a reaction cycle involving Tim21 and Tim17.

The vast majority of mitochondrial proteins are synthesized in the cytosol and are imported into mitochondria by protein machineries located in the mitochondrial membranes. It has become clear that hydrophilic as well as hydrophobic preproteins use a common translocase in the outer mitochondrial membrane, but diverge to two distinct translocases in the inner membrane. The translocases are dynamic, high-molecular-weight complexes that have to provide specific means for the recognition of preproteins, channel formation and generation of import-driving forces.

Mitochondrial preproteins destined for the matrix are translocated by two channel-forming transport machineries, the translocase of the outer membrane and the presequence translocase of the inner membrane. The presequence translocase-associated protein import motor (PAM) contains four essential subunits: the matrix heat shock protein 70 (mtHsp70) and its three cochaperones Mge1, Tim44 and Pam18. Here we report that the PAM contains a fifth essential subunit, Pam16 (encoded by Saccharomyces cerevisiae YJL104W), which is selectively required for preprotein translocation into the matrix, but not for protein insertion into the inner membrane. Pam16 interacts with Pam18 and is needed for the association of Pam18 with the presequence translocase and for formation of a mtHsp70-Tim44 complex. Thus, Pam16 is a newly identified type of motor subunit and is required to promote a functional PAM reaction cycle, thereby driving preprotein import into the matrix.

The serine protease urokinase-type plasminogen activator (uPA) interacts with a specific receptor (uPAR) on the surface of various cell types, including tumor cells, and plays a crucial role in pericellular proteolysis. High levels of uPA and uPAR often correlate with poor prognosis of cancer patients. Therefore, the specific inhibition of uPA with small molecule active-site inhibitors is one strategy to decrease the invasive and metastatic activity of tumor cells. We have developed a series of highly potent and selective uPA inhibitors with a C-terminal 4-amidinobenzylamide residue. Optimization was directed toward reducing the fast elimination from circulation that was observed with initial analogues. The x-ray structures of three inhibitor/uPA complexes have been solved and were used to improve the inhibition efficacy. One of the most potent and selective derivatives, benzylsulfonyl-D-SerSer-4-amidinobenzylamide (inhibitor 26), inhibits uPA with a K i of 20 nM. This inhibitor was used in a fibrosarcoma model in nude mice using lacZ-tagged human HT1080 cells, to prevent experimental lung metastasis formation. Compared with control (100%), an inhibitor dose of 2 ؋ 1.5 mg/kg/day reduced the number of experimental metastases to 4.6 ؎ 1%. Under these conditions inhibitor 26 also significantly prolonged survival. All mice from the control group died within 43 days after tumor cell inoculation, whereas 50% of mice from the inhibitor-treated group survived more than 117 days. This study demonstrates that the specific inhibition of uPA by these inhibitors may be a useful strategy for the treatment of cancer to prevent metastasis.The detachment of malignant cells from the primary tumor and their subsequent migration within the surrounding tissue, including intravasation and extravasation into blood and lymph vessels, leads to tumor dissemination and the formation of metastases at distant loci. Whereas a solid tumor can be removed by surgery or treated by radio-, chemo-, or hormone therapy, invasive tumor cells that have spread over the whole body can form secondary tumors leading to poor prognosis or death of cancer patients. Several proteases, such as matrix metalloproteases, the cysteine proteases cathepsin B and L, the aspartyl protease cathepsin D, and serine proteases, e.g. plasmin and uPA, 1 are involved at multiple stages during growth, invasion and progression of tumors, including metastases formation (1). High levels of expression of these proteases often correlate with poor prognosis for cancer patients (2). However, in several clinical cancer trials with different types of nonspecific matrix metalloprotease inhibitors, disappointing results with poor benefit and severe side effects were observed (3). This stimulated the search for alternative proteases with matrix degrading activity as new targets for anti-cancer drugs. An important role in metastasis has been recently ascribed to the plasmin-plasminogen activation system and especially to uPA.Both, uPA and the second endogenous plasminogen activator tPA are...