Summary:
L-asparaginases (EC 3.5.1.1) catalyze the hydrolysis of L-asparagine to aspartic acid and ammonia. Asparaginases of bacterial origin have been used for over 40 years in the treatment of acute lymphatic leukemia (ALL). Asparaginase isoenzyme II from Escherichia coli (EcA2) is especially effective in cancer therapy. The enzyme, a homotetramer (138 kDa) made up from equivalent dimers, is also an ideal model compound to study folding/unfolding and associa¬tion/dissociation phenomenon of a large oligomeric protein: It is readily amenable to site-directed mutagenesis, a single tryptophan residue per subunit simplifies spectroscopic analysis of conformational changes, and it spontaneously refolds after chemical denaturation. Our aim was to probe the molecular basis of the interactions between the EcA2 dimers by mutations, and to construct EcA2 variants with improved stability for possible application in leukemia treatment.
Using guanidine hydrochloride (Gu.HCl) equilibrium denaturation methods, we analyzed unfolding of wild-type EcA2 and a number of variants with amino acid replacements in the interior of the subunits or at the interfaces between the so-called intimate dimers. Enzymatic activity, fluorescence, and CD spectroscopy were used to monitor chemical denaturation by Gu.HCl. For wild-type EcA2 we found a highly cooperative transition from the folded to the denatured state. By gel filtration and sedimentation equilibrium ultracentrifugation, we showed that unfolding of the wild-type enzyme was preceded by dissociation into dimers and monomers. EcA2(WT) showed little change in stability between pH 5-8 but a rapid loss in stability below pH 5.
As compared to wild-type enzyme, variants with amino acid exchanges that weaken dimer-dimer interactions exhibited more complex denaturation profiles with one or two stable intermediate states. So, for instance, variants EcA2(Y176S) and EcA2(W66Y/Y181W) dissociate into fully active dimers at low Gu.HCl concentrations. On the other hand, variant EcA2(Y176F) shows higher structural stability and specific activity than the wild-type protein.
Thermal unfolding of EcA2 and its mutants was examined by differential scanning calorimetry (DSC) and CD. Correlation of DSC, CD, and light scattering data showed a single cooperative heat unfolding transition accompanied by irreversible protein aggregation, which precluded quantitative thermodynamic analysis of the excess heat capacity data. DSC data of EcA2 (WT) suggested that the increase in protein stability upon an increase in pH from 5-8 is mainly enthalpy-driven. Consistent with the chemical unfolding data, mutant EcA2(Y176F) showed higher thermal stability (a higher melting temperature, Tm) than wild type enzyme while EcA2(Y176S) and EcA2(W66Y/Y181W) were much more heat-sensitive, indicating the preference for large hydrophobic side chains at this site. In addition, two salt bridges, D156•••R191 and D188 •••R195, markedly contribute to the tetramer stability.
Our results demonstrate that even small changes at a subunit interface may markedly enhance or impair EcA2 stability without compromising its catalytic properties. Engineered enzyme variants with enhanced stability, such as EcA2(Y176F), are promising candidates for an improved asparaginase therapy of leukemias.