We present a series of QM/MM calculations aimed at understanding the mechanism of the biological dehydration of glycerol. Strikingly and unusually, this process is catalyzed by two different radical enzymes, one of which is a coenzyme-B-dependent enzyme and the other which is a coenzyme-B-independent enzyme. We show that glycerol dehydration in the presence of the coenzyme-B-dependent enzyme proceeds via a 1,2-OH shift, which benefits from a significant catalytic reduction in the barrier. In contrast, the same reaction in the presence of the coenzyme-B-independent enzyme is unlikely to involve the 1,2-OH shift; instead, a strong preference for direct loss of water from a radical intermediate is indicated. We show that this preference, and ultimately the evolution of such enzymes, is strongly linked with the reactivities of the species responsible for abstracting a hydrogen atom from the substrate. It appears that the hydrogen-reabstraction step involving the product-related radical is fundamental to the mechanistic preference. The unconventional 1,2-OH shift seems to be required to generate a product-related radical of sufficient reactivity to cleave the relatively inactive C-H bond arising from the B cofactor. In the absence of B, it is the relatively weak S-H bond of a cysteine residue that must be homolyzed. Such a transformation is much less demanding, and its inclusion apparently enables a simpler overall dehydration mechanism.

Molecular dynamics (MD) simulations have been employed for the first time to gain insight into the geometry of glycerol (GOL) bound within the active site of B 12 -dependent diol dehydratase (B 12 -dDDH). A peculiar feature of the B 12 -dDDH enzyme is that it undergoes suicidal inactivation by the substrate glycerol. To fully understand the inactivation mechanism, it is crucial to identify all possible interactions between GOL and the surrounding amino acid residues in the enzyme−substrate complex. Particularly important is the orientation of the C3-OH group in GOL since the presence of this OH group is the only difference between GOL and propanediol (PDO), a substrate for B 12 -dDDH that does not induce suicidal inactivation. The MD simulations indicate that glycerol can adopt two conformations that differ with respect to the orientation of the C3-OH group; in one conformer, the C3-OH group is oriented toward Ser301 (C3-OH•••Ser301), and in the other toward Asp335 (C3-OH••• Asp335). Although the former configuration is consistent with the crystal structure of B 12 -dDDH crystallized with cyanocobalamin (CNCbl) as the cofactor, MD simulations of this system suggest a substantial predominance of the latter conformer. A similar result with an even higher preference for the latter conformer is obtained for B 12 -dDDH with 5′deoxyadenosylcobalamin (AdoCbl) as a cofactor. Employing QM/MM calculations it is found that the energy difference between the two conformers of GOL is very small in CNCbl B 12 -dDDH, where the slightly preferred conformer is C3-OH••• Ser301. However, in AdoCbl B 12 -dDDH, this energy difference is higher, implying that GOL exists predominantly as the C3-OH•••Asp335 conformer. These findings offer a new perspective for investigations of substrate-induced inactivation of the B 12 -dDDH enzyme.

Diol dehydratase, dependent on coenzyme B 12 (B 12 -dDDH), displays a peculiar feature of being inactivated by its native substrate glycerol (GOL). Surprisingly, the isofunctional enzyme, B 12 -independent glycerol dehydratase (B 12 -iGDH), does not undergo suicide inactivation by GOL. Herein we present a series of QM/MM and MD calculations aimed at understanding the mechanisms of substrate-induced suicide inactivation in B 12 -dDDH and that of resistance of B 12 -iGDH to inactivation. We show that the first step in the enzymatic transformation of GOL, hydrogen abstraction, can occur from both ends of the substrate (either C1 or C3 of GOL). Whereas C1 abstraction in both enzymes leads to product formation, C3 abstraction in B 12 -dDDH results in the formation of a low energy radical intermediate, which is effectively trapped within a deep well on the potential energy surface. The long lifetime of this radical intermediate likely enables its side reactions, leading to inactivation. In B 12 -iGDH, by comparison, C3 abstraction is an endothermic step; consequently, the resultant radical intermediate is not of low energy, and the reverse process of reforming the reactant is possible.

No abstract

The aim of this study was to compare osteotomy for malunited distal radius fracture with embedment of a corticospongious graft into a periosteal flap of the recipient bone (test) with the standard procedure (control) with respect to graft resorption. A retrospective assessor-blind analysis of consecutive patients (test: n=19, control: n=30) was performed. Ulnar tilt, palmar tilt and capitate-ulna distance were assessed from radiographs taken before, two to four days after and over three months after the surgery to determine loss of correction achieved by the surgery and estimate graft resorption during the postoperative period. In both unadjusted and adjusted comparisons, loss of correction of all parameters was lower in the test group (P<0.05). The odds of "none to mild" resorption were greater in the test group with an adjusted odds ratio of 5.43 (95% confidence interval: 1.32-26.5, P=0.025). Total graft collapse occurred in five of 30 controls and in none of 19 test patients. Graft embedment into the periosteum may improve its preservation.