Structural studies on Porphyromonas gingivalis proteins

Abstract

[eng] We investigated the structure function relationship of PorX by combined biophysical experiments and enzymatic assays. As a regulator of the type 9 secretion system (T9SS) of the human pathogen Porphyromonas gingivalis (PG), PorX is responsible for PG virulence[1] and as drug target could be the key in disease prevention. We started with crystals of wild type PorX dimers, which were unsuitable for structure solution as well as further PG proteins for potential cocrystallization. We successfully optimized the crystallization conditions for selenomethionine derivative PorX by adding beryllium fluoride as phosphate mimic, zinc for higher occupancies in the active site, as well as additives and optimizing buffers, concentrations, crystallization- and crystal manipulation techniques. By X-ray diffraction and single anomalous dispersion we solved several structures of dimeric PorX with and without substrate (pGpG) up to a resolution of 1.9 Å, characterized its domains, active site and substrate binding pocket. The PorX N-terminal domain is a typical response regulator receiver domain (RD) of the OmpR[2] subfamily. PorX has the characteristic α5β5 Rossmann-like fold, conserved phosphorylation site and the YT pair responsible for conformational change during phosphorylation which leads to dimerization. A helical bundle domain (HBD) connects the RD to the C-terminal enzymatic active PglZ domain which was previously uncharacterized in structure and function. The PglZ domain has the characteristic α/β-fold of the alkaline phosphatase superfamily (APS), the metal coordination sphere of phosphodiesterase, and an extra domain capping the active site (CAP[3]).[4], [5] By small-angle X-ray scattering we investigated PorX dimers in solution, which did not significantly differed from the crystal structure, and PorX monomers showing significant conformational changes. We build and validated a SAXS model of monomeric PorX with a χ² of 1.15 and explained the mechanism of PorX dimerization: Phosphorylation leads to RD dimerization, their conformational changes lead to an exchange of surfaces between RDs and HBDs, which then bring both PglZs into contact, followed by PglZ dimerization. The newly formed PglZ dimeric interface (DI) folds into the catalytic active APS bi-metallo active site, and the capping subdomain opens the substrate binding pocket due to interaction with DI. These we confirmed by structure based mutational studies, SEC-MALLS as well as phosphorylation assays, SEC and enzymatic assays performed by our collaboration partners. Comparing PglZ subfamily homologs, we found out that the active center as well as its DI is conserved with most BREX-PglZs, involved in phage defense[6]. In PorX the DI folds upon dimerization and thus activates the protein, which thereby is likely also true for BREX-PglZs. Our collaborators determined the enzymatic activity by enzymatic assays confirming PorX as

phosphodiesterase and not a phosphatase even though APS active sites often have promiscuous activity[7], [8]. Phosphodiesterase substrate affinities are regulated by the substrate binding pockets, so in order to narrow down potential substrates to screen, we compared PorX to its closest known structures and enzyme classes. As results we found the most similar substrate cavities in ecto- nucleotide pyrophosphatases which cleave linear and cyclic dinucleotides as well as mono- and dinucleotide polyphosphates. This led to cocrystallization trials yielding our pGpG bound structure with the PorX substrate binding pocket only partial filled. This led us screen activity against a large polynucleotide library and product detection using ultra- high performance liquid-chromatography, coupled with mass spectrometry (UHPLC-MS). As result PorX cleaved linear and cyclic oligonucleotides as well as linear dinucleotides. We did not found any activity against cyclic mono- or dinucleotides, mono- or dinucleotide polyphosphates. Therefore it is likely that PorX is involved in oligonucleotide signaling in PG and by designing of PorX PglZ inhibitors it might be possible to stop PG T9SS transcriptional activation and render PG avirulent.

Summary Literature 1: Veith, Paul D.; Glew, Michelle D.; Gorasia, Dhana G.; Reynolds, Eric C., "Type IX secretion", Molecular microbiology 106, 1, p35–53, 2017. 2: Gao, Rong; Bouillet, Sophie; Stock, Ann M., Structural Basis of Response Regulator Function, 2019 3: Kim, A; Benning, MM; OkLee, S; Quinn, J Martin, BM; Holden, HM; Dunaway-Mariano, D, "Divergence of chemical function in the alkaline phosphatase superfamily", Biochemistry 50, 17, p3481–3494, 2011. 4: Galperin, Michael Y.; Koonin, Eugene V., "Divergence and convergence in enzyme evolution", The Journal of biological chemistry 287, 1, p21–28, 2012. 5: Sunden F, AlSadhan I, Lyubimov A, Doukov T, Swan J, Herschlag D., "Differential catalytic promiscuity of the alkaline phosphatase superfamily bimetallo core reveals mechanistic features underlying enzyme evolution.", J Biol Chem. 292, 51, p20960–20974, 2017. 6: Chaudhary, Kulbhushan, "BacteRiophage EXclusion (BREX)", Journal of cellular physiology 233, 2, p771– 773, 2018. 7: Mohamed, MF.; Hollfelder, F., "Efficient, crosswise catalytic promiscuity among enzymes that catalyze phosphoryl transfer", Biochimica et biophysica acta 1834, 1, p417–424, 2013. 8: Pabis, Anna; Kamerlin, Shina Caroline Lynn, "Promiscuity and electrostatic flexibility in the alkaline phosphatase superfamily", Current opinion in structural biology 37, p14–21, 2016.

[ger] In dieser Doktorarbeit beschreiben wir die Struktur und Funktion von PorX, einem essentiellen Protein des Typ-9-Sekretionssystems von Porphyromaonas gingivalis (PG).[1] PG ist Teil des dentalen Biofilms und gilt als Hauptauslöser für Gingivitis und Parodontitis.[2] PG befällt nicht nur orales Gewebe, sondern befällt auch u.a. Gehirn, Leber und Koronararterien. [3], [4], [5] Es ist an einer Vielzahl von systemischen Erkrankungen beteiligt, darunter Krebs[6] Arteriosklerose[7], [8] und Herzkreislauf-erkrankungen.[9] [10] Da die Gesamtheit dieser Krankheiten die Mehrheit der menschlichen Bevölkerung betrifft ist die Erforschung von PorX ein lohnendes Forschungsprojekt, nicht zuletzt, da das Ziel der Inhibierung ein lohnendes Folgeziel für Medikamentenentwicklung darstellt, welche Allgemeingesundheit und Lebensspanne erhöhen werden. Durch Röntgenkristallographie haben wir die dimere Kristallstruktur von PorX aufgeklärt, sowohl ohne, als auch im substratgebunden Zustand, in Komplex mit pGpG. Wir beschreiben die einzelnen PorX Domänen und ihr Zusammenwirken, das katalytische Zentrum sowie die Substratbindestellen. Durch Röntgenkleinwinkelstreuung haben wir die Struktur von PorX in Lösung analysiert, als Monomer sowie Dimer. Daraufhin und durch strukturbasierte Mutationsstudien haben wir den Mechanismus der Aktivierung und Dimerisierung von PorX erklärt und durch Phosphorylierungs- analysen, Größenausschlusschromatographie (SEC) und SEC-MALLS (multi-angle-laser-light- scattering) bestätigt. Die enzymatisch aktive C-terminale Domäne von PorX gehört zur PglZ-Familie, dessen Struktur und Funktion, trotz der relativen Häufigkeit in Prokaryonten,[11] bisher unaufgeklärt war. Daher wurde in dieser Arbeit die Struktur und das katalytische Zentrum von PorX mit seinen nächsten Homologen verglichen, sowie mit bekannten PglZ Klassen der BREX Systeme. Als Ergebnis konnten wir PorX-PglZ als Phosphodiesterase der Superfamilie der alkalinen Phosphatasen (APS) identifizieren. Sowohl das aktivitätsdeterminierende katalytische Zentrum, wie auch das bisher unbeschriebene Dimerisierungsmotiv, beteiligt an enzymatischer Aktivierung, sind zwischen BREX-PglZs hoch konserviert. In PorX faltet sich dieses Motiv während des Dimerisierungsvor- ganges und ist entscheidend für die Metallkoordinierung des katalytischen Zentrums. Weiterhin konnten wir über enzymatische Analysen die Substratspezifität für PorX aufklären und zeigen dass PorX spezifisch lineare und zyklische Oligonukleotide zu Nukleotidmonophosphaten hydrolysiert. Somit ist die Funktion von PorX an der Regulation von Oligonukleotidsignaltransduktion in PG beteiligt, und durch unsere Kristallstrukturen ist es möglich, dass, durch spezifisches Design von PorX-Inhibitoren, in Zukunft die Virulenz von PG verhindert werden kann.

Literaturverzeichnis 1: Veith, Paul D.; Glew, Michelle D.; Gorasia, Dhana G.; Reynolds, Eric C., Type IX secretion, 2017. 2: Hajishengallis, G.; Lamont, R. J., Beyond the red complex and into more complexity, 2012. 3: Olsen, Ingar; Yilmaz, Özlem, Modulation of inflammasome activity by Porphyromonas gingivalis in periodontitis and associated systemic diseases, 2016. 4: Huck, Olivier; You, Jian; Han, Xianxian; Cai, Bin; Panek, James; Amar, Salomon, Reduction of Articular and Systemic Inflammation by Kava-241 in a Porphyromonas gingivalis-Induced Arthritis Murine Model, 2018. 5: Jia, Lu; Han, Nannan; Du, Juan; Guo, Lijia; Luo, Zhenhua; Liu, Yi, Pathogenesis of Important Virulence Factors of Porphyromonas gingivalis via Toll-Like Receptors, 2019. 6: Gholizadeh, Pourya; Eslami, Hosein; Yousefi, Mehdi; Asgharzadeh, Mohammad; Aghazadeh, Mohammad; Kafil, Hossein Samadi, Role of oral microbiome on oral cancers, a review, 2016. 7: Reyes, Leticia; Herrera, David; Kozarov, Emil; Roldá, Silvia; Progulske-Fox, Ann, Periodontal bacterial invasion and infection, 2013. 8: Huck, O; Saadi-Thiers, K; Tenenbaum, H; Davideau, JL; Romagna, C; Laurent, Y; Cottin, Y; Roul, JG., Evaluating periodontal risk for patients at risk of or suffering from atherosclerosis, 2011. 9: Chistiakov, Dimitry A.; Orekhov, Alexander N.; Bobryshev, Yuri V., Links between atherosclerotic and periodontal disease, 2016. 10: Fiorillo, Luca; Cervino, Gabriele; Laino, Luigi; D'Amico, Cesare; Mauceri, Rodolfo; Tozum, Tolga Fikret; Gaeta, Michele; Cicciù, Marco, Porphyromonas gingivalis, Periodontal and Systemic Implications-A Systematic Review, 2019. 11: Goldfarb, T; Sberro, H; Weinstock, E; Cohen, O; Doron, Sy et al., BREX is a novel phage resistance system widespread in microbial genomes, 2015.