Vitamin K-dependent (VKD) proteins are modified by the VKD carboxylase as they transit through the endoplasmic reticulum. In a reaction required for their activity, clusters of Glu's are converted to Gla's, and fully carboxylated VKD proteins are normally secreted. In mammalian cell lines expressing high levels of r-VKD proteins, however, under- and uncarboxylated VKD forms are observed. Overexpression of r-carboxylase does not improve carboxylation, but the lack of effect is not understood, and the intracellular events that occur during VKD protein carboxylation have not been investigated. We analyzed carboxylation in 293- and BHK cell lines expressing r-factor IX (fIX) and endogenous carboxylase or overexpressed r-carboxylase. The fIX secreted from the four cell lines was highly carboxylated, indicating fIX-carboxylase engagement during intracellular trafficking. The r-carboxylase was functional for carboxylation: overexpression resulted in a proportional increase in fIX-carboxylase complexes that yielded full fIX carboxylation. Interestingly, the carboxylated fIX product was not efficiently released from the carboxylase in r-fIX/r-carboxylase cells, resulting in decreased fIX secretion. r-Carboxylase overexpression changed the ratios of intracellular fIX to carboxylase, and we therefore developed an in vitro assay to test whether fIX levels affect release. FIX-carboxylase complexes were in vitro carboxylated with or without excess VKD substrate or propeptide. These analyses are the first to dissect the rates of release versus carboxylation and showed that release was much slower than carboxylation. In the absence of excess VKD substrate/propeptide, fIX in the fIX-carboxylase complex was fully carboxylated by 10 min, but 95% was still complexed with carboxylase after 30 min. The presence of excess VKD substrate/propeptide, however, led to a significant increase in VKD product release, possibly through a second propeptide binding site in the carboxylase. The intracellular analyses also showed that the fIX carboxylation rate was slow in vivo and was similar in r-fIX versus r-fIX/r-carboxylase cells, despite the large differences in carboxylase levels. The results suggest that the vitamin K cofactor may be limiting for carboxylation in the cell lines.

The vitamin K-dependent (VKD) carboxylase binds VKD proteins via their propeptide and converts Glu's to gamma-carboxylated Glu's, or Gla's, in the Gla domain. Multiple carboxylation is required for activity, which could be achieved if the carboxylase is processive. In the only previous study to test for this capability, an indirect assay was used which suggested processivity; however, the efficiency was poor and raised questions regarding how full carboxylation is accomplished. To unequivocally determine if the carboxylase is processive and if it can account for comprehensive carboxylation in vivo, as well as to elucidate the enzyme mechanism, we developed a direct test for processivity. The in vitro carboxylation of a complex containing carboxylase and full-length factor IX (fIX) was challenged with an excess amount of a distinguishable fIX variant. Remarkably, carboxylation of fIX in the complex was completely unaffected by the challenge protein, and comprehensive carboxylation was achieved, showing conclusively that the carboxylase is processive and highly efficient. These studies also showed that carboxylation of individual fIX/carboxylase complexes was nonsynchronous and implicated a driving force for the reaction which requires the carboxylase to distinguish Glu's from Gla's. We found that the Gla domain is tightly associated with the carboxylase during carboxylation, blocking the access of a small peptide substrate (EEL). The studies describe the first analysis of preformed complexes, and the rate for full-length, native fIX in the complex was equivalent to that of the substrate EEL. Thus, intramolecular movement within the Gla domain to reposition new Glu's for catalysis is as rapid as diffusion-limited positioning of a small substrate, and the Gla domain is not sterically constrained by the rest of the fIX molecule during carboxylation. The rate of carboxylation of fIX in the preformed complex was 24-fold higher than for fIX modified by free carboxylase, which supports carboxylase processivity and which indicates that binding and/or release is the rate-limiting step in protein carboxylation. These data indicate a model of tethered processivity, in which the VKD proteins remain bound to the carboxylase throughout the reaction via their propeptide, while the Gla domain undergoes intramolecular movement to reposition new Glu's for catalysis to ultimately achieve comprehensive carboxylation.

No abstract

To explore the variability in biosensor studies, 150 participants from 20 countries were given the same protein samples and asked to determine kinetic rate constants for the interaction. We chose a protein system that was amenable to analysis using different biosensor platforms as well as by users of different expertise levels. The two proteins (a 50-kDa Fab and a 60-kDa glutathione S-transferase [GST] antigen) form a relatively high-affinity complex, so participants needed to optimize several experimental parameters, including ligand immobilization and regeneration conditions as well as analyte concentrations and injection/dissociation times. Although most participants collected binding responses that could be fit to yield kinetic parameters, the quality of a few data sets could have been improved by optimizing the assay design. Once these outliers were removed, the average reported affinity across the remaining panel of participants was 620 pM with a standard deviation of 980 pM. These results demonstrate that when this biosensor assay was designed and executed appropriately, the reported rate constants were consistent, and independent of which protein was immobilized and which biosensor was used.

A novel form of tetrameric streptavidin has been engineered to have reversible biotin binding capability. In wild-type streptavidin, loop3–4 functions as a lid for the entry and exit of biotin. When biotin is bound, interactions between biotin and key residues in loop3–4 keep this lid in the closed state. In the engineered mutein, a second biotin exit door is created by changing the amino acid sequence of loop7–8. This door is mobile even in the presence of the bound biotin and can facilitate the release of biotin from the mutein. Since loop7–8 is involved in subunit interactions, alteration of this loop in the engineered mutein results in an 11° rotation between the two dimers in reference to wild-type streptavidin. The tetrameric state of the engineered mutein is stabilized by a H127C mutation, which leads to the formation of inter-subunit disulfide bonds. The biotin binding kinetic parameters (koff of 4.28×10−4 s−1 and Kd of 1.9×10−8 M) make this engineered mutein a superb affinity agent for the purification of biotinylated biomolecules. Affinity matrices can be regenerated using gentle procedures, and regenerated matrices can be reused at least ten times without any observable reduction in binding capacity. With the combination of both the engineered mutein and wild-type streptavidin, biotinylated biomolecules can easily be affinity purified to high purity and immobilized to desirable platforms without any leakage concerns. Other potential biotechnological applications, such as development of an automated high-throughput protein purification system, are feasible.