The velocity of the oxidative renaturation of reduced ribonuclease A catalyzed by protein disulfide isomerase (PDI) is strongly dependent on the composition of a glutathione/glutathione disulfide redox buffer. As with the uncatalyzed, glutathione-mediated oxidative folding of ribonuclease, the steady-state velocity of the PDI-catalyzed reaction displays a distinct optimum with respect to both the glutathione (GSH) and glutathione disulfide (GSSG) concentrations. Optimum activity is observed at [GSH] = 1.0 mM and [GSSG] = 0.2 mM. The apparent kcat at saturating RNase concentration is 0.46 +/- 0.05 mumol of RNase renatured min-1 (mumol of PDI)-1 compared to the apparent first-order rate constant for the uncatalyzed reaction of 0.02 +/- 0.01 min-1. Changes in GSH and GSSG concentration have a similar effect on the rate of both the PDI-catalyzed and uncatalyzed reactions except under the more oxidizing conditions employed, where the catalytic effectiveness of PDI is diminished. The ratio of the velocity of the catalyzed reaction to that of the uncatalyzed reaction increases as the quantity [GSH]2/[GSSG] increases and approaches a constant, limiting value at [GSH]2/[GSSG] greater than 1 mM, suggesting that a reduced, dithiol form of PDI is required for optimum activity. As long as the glutathione redox buffer is sufficiently reducing to maintain PDI in an active form [( GSH]2/[GSSG] greater than 1 mM), the rate acceleration provided by PDI is reasonably constant, although the actual rate may vary by more than an order of magnitude. PDI exhibits half of the maximum rate acceleration at a [GSH]2/[GSSG] of 0.06 +/- 0.01 mM.

Protein disulfide isomerase (PDI), a very abundant protein in the endoplasmic reticulum, facilitates the formation and rearrangement of disulfide bonds using two nonequivalent redox active-sites, located in two different thioredoxin homology domains [Lyles, M. M., & Gilbert, H. F. (1994) J. Biol. Chem. 269, 30946-30952]. Each dithiol/disulfide active-site contains the thioredoxin consensus sequence CXXC. Four mutants of protein disulfide isomerase were constructed that have only a single active-site cysteine. Kinetic analysis of these mutants show that the first (more N-terminal) cysteine in either active site is essential for catalysis of oxidation and rearrangement during the refolding of reduced bovine pancreatic ribonuclease A (RNase). Mutant active sites with the sequence SGHC show no detectable activity for disulfide formation or rearrangement, even at concentrations of 25 microM. The second (more C-terminal) cysteine is not essential for catalysis of RNase disulfide rearrangements, but it is essential for catalysis of RNase oxidation, even in the presence of a glutathione redox buffer. Mutant active sites with the sequence CGHS show 12%-50% of the kcat activity of wild-type active sites during the rearrangement phase of RNase refolding but < 5% activity during the oxidation phase. In addition, mutants with the sequence CGHS accumulate significant levels of a covalent PDI-RNase complex during steady-state turnover while the wild-type enzyme and mutants with the sequence SGHC do not. Since both active-site cysteines are essential for catalysis of disulfide formation, the dominant mechanism for RNase oxidation may involve direct oxidation by the active-site PDI disulfide. Although it is not essential for catalysis of RNase rearrangements, the more C-terminal cysteine does contribute 2-8-fold to the rearrangement activity. A mechanism for substrate rearrangement is suggested in which the second active-site cysteine provides PDI with a way to "escape" from covalent intermediates that do not rearrange in a timely fashion. The second active-site cysteine may normally serve the wild-type enzyme as an internal clock that limits the time allowed for intramolecular substrate rearrangements.

At low concentrations of a glutathione redox buffer, the protein disulfide isomerase (PDI) catalyzed oxidative renaturation of reduced ribonuclease A exhibits a rapid but incomplete activation of ribonuclease, which precedes the steady-state reaction. This behavior can be attributed to a GSSG-dependent partitioning of the substrate, reduced ribonuclease, between two classes of thiol/disulfide redox forms, those that can be converted to active ribonuclease at low concentrations of GSH and those that cannot. With catalytic concentrations of PDI and near stoichiometric concentrations of glutathione disulfide, approximately 4 equiv (2 equiv of ribonuclease disulfide) of GSH are formed very rapidly followed by a slower formation of GSH, which corresponds to an additional 2 disulfide bond equiv. The rapid formation of RNase disulfide bonds and the subsequent rearrangement of incorrect disulfide isomers to active RNase are both catalyzed by PDI. In the absence of GSSG or other oxidants, disulfide bond equivalents of PDI can be used to form disulfide bonds in RNase in a stoichiometric reaction. In the absence of a glutathione redox buffer, the rate of reduced ribonuclease regeneration increases markedly with increasing PDI concentrations below the equivalence point; however, PDI in excess over stoichiometric concentrations inhibits RNase regeneration.

Although myeloma light chains are known to undergo receptor-mediated endocytosis in the kidney, the molecular identity of the receptor has not been characterized. We examined the interaction between cubilin (gp280) and four species of light chains isolated from the urine of patients with multiple myeloma. Four lines of evidence identify cubilin, a giant glycoprotein receptor, which is restricted in distribution to endocytic scavenger pathways and which has potent effects on endosomal trafficking, as a potentially physiologically relevant binding site for light chains: 1) light chains coeluted during immunoaffinity purification of cubilin; 2) polyclonal antisera to cubilin but not control sera, displaced human light chain binding from rat renal brush-border membranes; 3) cubilin bound to multiple species of light chains during surface plasmon resonance; 4) anti-cubilin antiserum interfered with light chain endocytosis by visceral yolk sac epithelial cells. However, both binding of light chains to brush-border membranes and endocytosis of light chains by yolk sac epithelial cells were only partially inhibited by anticubilin antibodies, suggesting presence of additional or alternate binding sites for light chains. Excess light chain had a potent inhibitory effect on endosomal fusion in vitro. Binding showed dose and time-dependent saturability with low-affinity, high-capacity equilibrium binding parameters. These data demonstrate that cubilin plays a role in the endocytosis and trafficking of light chains in renal proximal tubule cells.

Recent work with interleukins has shown a convergence of tyrosine phosphorylation signal transduction cascades at the level of the Janus and Src families of tyrosine kinases. Here we demonstrate that activation of the seven-transmembrane AT 1 receptor by angiotensin II induces a physical association between Jak2 and Fyn, in vivo. This association requires the catalytic activity of Jak2 but not Fyn. Deletion studies indicate that the region of Jak2 that binds Fyn is located between amino acids 1 and 240. Studies of the Fyn SH2 and SH3 domains demonstrate that the SH2 domain plays the primary role in Jak2/Fyn association. Not surprisingly, this domain shows a marked preference for tyrosinephosphorylated Jak2. Surface plasmon resonance estimated the dissociation equilibrium constant (K d ) of this association to be 2.36 nM. Last, in vivo studies in vascular smooth muscle cells show that, in response to angiotensin II, Jak2 activation is required for Fyn activation and induction of the c-fos gene. The significance of these data is that Jak2, in addition to serving as a critical angiotensin II activated signal transduction kinase, also functions as a docking protein and participates in the activation of Fyn by providing phosphotyrosine residues that bind the SH2 domain of Fyn.The Jak family of nonreceptor tyrosine kinases includes Jak1, Jak2, Jak3, and Tyk2. Each protein is approximately 130 kDa and contains seven conserved Jak homology domains (JH1 to JH7) (1, 2). The Jak kinases induce gene regulation through the signal transducers and activators of transcription. Unlike almost all other protein-tyrosine kinases, members of the Jak family bear no SH2 1 or SH3 domains. In contrast to the Jaks, members of the Src family of protein-tyrosine kinases are approximately 55-62 kDa in mass and do possess SH2 and SH3 domains. There are nine known members of the Src kinase family. While the expression of most members is restricted to hematopoietic cells, Fyn and Src are widely expressed by many cell types. There appears to be some functional redundancy of these two proteins, since knockout mice lacking either gene are born alive, while disruption of both the fyn and src alleles results in embryonic lethality (3-5). This redundancy is also seen in the activation of similar signaling pathways by Fyn and Src (6).Recent studies have demonstrated various levels of crosstalk between Jak2 and other signaling pathways. For example, activation of gp130 by the hematopoietic cytokine, interleukin-11, induces protein complex formation between the Jak and Src family tyrosine kinases. Specifically, treatment of 3T3-L1 cells with interleukin-11 leads to a transient complex of Grb2, Jak2, and Fyn (7). Subsequent studies by Yang et al. demonstrated an interleukin-11-dependent association of Jak2 with other signaling molecules including protein phosphatase 2A, phosphatidylinositol-3-kinase, and the Src family kinase Yes (8). To date, the regions that mediate these interactions, as well as the hierarchy of the signal transduction cascades, hav...