Metalloclusters of the HD-domain superfamily of enzymes

HD proteins are omnipresent and belong to a superfamily of metalloenzymes counting presently > 137,000 members with diverse and unknown functions affecting the human health and environment, including HIV-1 immunoresponse, anti-virulence, DNA/RNA unwinding and degradation, and signaling. Whereas their initial functionality was predicted to be solely hydrolytic, novel diiron oxygenases have emerged carrying out chemically difficult small molecule activations.

These proteins are typified by a helical fold harboring the H…HD…D residue quartet, known to bind a divalent metal ion (most commonly Zn²⁺). The presence of two additional histidines inbetween the aspartate residues extends their metal binding capacity and supports formation of di- or even tri-nuclear clusters. Though members of this superfamily were originally annotated as (phospho)hydrolases, less than a decade ago, a diiron HD enzyme involved in the catabolism of inositol associated with type I diabetes mellitus, namely myo-inositol oxygenase (MIOX) was demonstrated to carry out a radically different reaction using molecular oxygen to afford transformation of its substrate. The only recently recognized HD enzyme PhnZ, was also shown to follow the paradigm of MIOX, employing oxygen for the acquisition of phosphate from an organophosphonate by marine microorganisms. Biochemically uncharacterized dinuclear HD domain proteins have been selected as targets for discovering their substrates, identifying key residues that favor hydrolase or oxygenase activity and examining the catalytic potency of the types of metals incorporated. Considering the emergence of MIOX and PhnZ as evolutionary means to carry out reactions affecting the human health and environment, as well as the emergence of hydrolytic HD enzymes orchestrating immunoresponse in eukaryotes and prokaryotes, our work aims to serve as a paradigm for discovering new antiviral (and therapeutic) factors and functions within the largely uncharacterized HD superfamily.

Oxygenases

Hallmarks of HD-oxygenases and identification of new MVDO candidates. HD-domain oxygenases constitute a functional outlier of the HD hydrolytic enzymes and employ a rather unconventional strategy for the O₂-dependent 4 e- oxidation of their primary substrates involving a Fe(III)-Fe(III)-superoxo intermediate for C-H activation, reminiscent of that of mononuclear non-heme enzymes such as the hydroxyethylphosphonate dioxygenase (HEPD). They employ a mixed-valent Fe(II)-Fe(III) cofactor that is regenerated after each turnover. The kinetic mechanism of PhnZ was delineated and the redox potentials for the stabilization of the Fe(II)-Fe(III) cofactor were measured. The findings from these experiments aim to serve as a descriptor for determining a) the oxygenase activity of dinuclear HD proteins and b) the redox state role of the metallocofactor mediating hydrolytic/redox activities. In addition, the syntenic occurrence of phnZ-like and phnY-like genes, the latter encoding for α-KG dependent oxygenases, will be explored as an operon architecture in ‘sequence and species space’ to provide HD-MVDO candidates that catalyze novel reaction(s).

Hydrolases

Catalytic and activating (co)factors of mono/dinuclear HD (phospho)hydrolases and phosphodiesterases. The majority of the HD hydrolases have functions that are only crudely understood but are predicted to catalyze (phospho)hydrolytic reactions on a broad range of substrates including nucleotides, nucleosides and nucleic acids (DNA, RNA). The overall aim is to establish the operant mechanisms regarding the catalytic potency of the type of metals they employ, the nuclearity of their cofactors and tertiary structure, kinetics of reaction and inhibition, thermodynamic, electronic and structural parameters directing the respective functions.

Cofactor evolution and metal activity dependence of HD-GYP PDEs. C-di-GMP and the novel hybrid c-GAMP nucleotide are important bacterial signaling molecule involved in virulence response and biofilm formation, the latter linked with several chronic bacterial infections. C-di-GMP and c-GAMP are degraded by phosphodiesterases (PDEs); HD-GYP HD-proteins represent a relative novel addition to this PDE functional superfamily. A specific is to solve structural conundrums about the nuclearity of the active cofactor (di- or tri-nuclear), the unexpected Fe- dependent hydrolytic activity and the role of accessory metallocofactors. Support from bioinformatics approaches will potentially portray the structural evolution of these catalytic sites in enzymes with similar function.

Function and substrates of the YqeK, ThpO and YedJ HD domain proteins

Three phylogenetically close relatives of PhnZ (i.e. YqeK, ThpO and YedJ) were structurally characterized by the Joint Center of Structural Genomics. YqeK and ThpO had diiron sites, whereas YedJ was devoid of any metals. Presently, there is only scant biochemical information about the role of YqeK in biofilm formation, and almost no functional characterization about any of the other proteins exists. We employ a a combination of structural, kinetic and biochemical approaches to establish a) the metal type of their active cofactors, the unanticipated role of Fe in their hydrolytic activity (if not oxygenases) and their natural substrates.

Active sites and specific activities of the HD-domain nucleases in CRISPR systems

The Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-associated (Cas) defense system is the only adaptive and inheritable immunity found in prokaryotes. HD nucleases are suggested to be involved in the third stage of action of the cascade complex (known as interference), which consists cleavage of viral DNA (and RNA). The Type I and Type III CRISPR-Cas systems (distinguished on the basis of the constitutive protein elements) harbor HD nuclease domains. The focus of our research is on the functional and evolutionary diversification of the enzymes with respect to the chemical nature of their metallocofactor, for which there is contradictory structural/functional information available. Identification and establishment of the latter is critical for modulating the function of Type-I systems; the study is combined with bioinformatic studies to identify possible (evolutionarily) correlations of the specific active sites with stress-response factors and metal availability.