0 Universidade Federal Do Paraná Eberhard Karls

UNIVERSIDADE FEDERAL DO PARANÁ EBERHARD KARLS UNIVERSITÄT TÜBINGEN

ADRIAN RICHARD SCHENBERGER SANTOS

REGULATION OF THE NAD+ METABOLISM IN BACTERIA BY L-GLUTAMINE, NAD+ AND PII PROTEINS&NBSP

CURITIBA 2020 0

UNIVERSIDADE FEDERAL DO PARANÁ EBERHARD KARLS UNIVERSITÄT TÜBINGEN

ADRIAN RICHARD SCHENBERGER SANTOS

REGULATION OF THE NAD+ METABOLISM IN BACTERIA BY L-GLUTAMINE, NAD+ AND PII PROTEINS&NBSP Tese apresentada ao Curso de Pós- Graduação em Ciências (Bioquímica), Setor de Ciências Biológicas, Universidade Federal do Paraná, e à Universidade de Tübingen como requisito parcial à obtenção do título de doutor em Bioquímica e em Microbiologia. Orientador: Prof. Dr. Luciano Fernandes Huergo e Prof. Dr. Karl Forchhammer

CURITIBA 2020

AGRADECIMENTOS

Aos meus orientadores, Prof. Dr. Luciano Fernandes Huergo e Prof. Dr. Karl Forchhammer pelo exemplo de competência, dedicação e entusiasmo com a ciência.

Às professoras Dra. Maria Berenice Reynaud Steffens e Rose Adele Monteiro pelo acompanhamento e correção de relatórios durante todo o período do doutorado, além de valiosas sugestões na qualificação.

Aos técnicos de laboratório Rosely Prado, Valter Antônio de Baura e Heinz-Peter Grenzendorf pelo trabalho indispensável no funcionamento de todo o laboratório.

A todos os meus colegas do Brasil, em especial aqueles que dividimos laboratório e dormitório. Edileusa, Fernanda, Erick, Andrey, Heloísa, Thiago, Ana Paula, Ana Goedert, Alex Tramontin Almeida, Manuel e Matheus.

A todos os meus colegas da Alemanha, em especial aqueles do lunch group. Rokhsareh, Jörg, Michael, Sofía, Ritu e Alexandra. Um agradecimento mais que especial para Louisa Perelo por toda incrível ajuda durante meu 1 ano e meio em Tübingen. E minhas colegas de dormitório Luíza, Peter, Anasuya, Klaudia e Xixi.

À minha mãe Luciane por todos esses anos de dedicação e por me ensinar como persistir.

Ao meu marido, Leandro Guimarães Ferreira.

Ao financiamento da CAPES e do DAAD.

ACKNOWLEDGEMENTS

To my supervisors Prof. Dr. Luciano Fernandes Huergo and Prof. Dr. Karl Forchhammer, by the example of competence, dedication, and excitement with science.

To the Professors Dr. Maria Berenice Reynaud Steffens and Rose Adele Monteiro by the close attendance and the correction of the annual reports during my Ph.D., besides the valuable suggestions during preparation process of this thesis.

To the laboratory technicians Rosely Prado, Valter Antônio de Baura, and Heinz-Peter Grenzendorf by the indispensable work and dedication.

To all my colleagues from Brazil, in special those from bench sharing and dormitory. Edileusa, Fernanda, Erick, Andrey, Heloísa, Thiago, Ana Paula, Ana Goedert, Alex, Manuel, and Matheus.

To all my colleagues from Germany, in special the ‘lunch group’. Rokhsareh, Jörg, Michael, Sofía, Ritu and Alexandra. A special thanks to Louisa Perelo for all the incredible help during my 1 and a half year in Tübingen. Also, to my colleagues of dormitory Luíza, Peter, Anasuya, klaudia and Xixi.

To my mother Luciane for all these years of dedication and for teaching me how to persist.

To my husband Leandro.

To the financing agencies CAPES and DAAD.

Dans les champs de l'observation, le hasard ne favorise que les esprits préparés.

LOUIS PASTEUR

RESUMO

Nicotinamida adenina dinucleotídeo (NAD)+ é um metabólito central que participa das principais reações redox junto com diversas enzimas oxidorredutases. Ainda, NAD+ é usado como substrato por ADP-ribosil transferases, sirtuínas e DNA ligases NAD+- dependentes. A biossíntese de NAD+ é uma das rotas fundamentais de biossíntese e a ubíqua NAD+ sintetase (NadE) catalisa seu último passo. Duas diferentes classes da NadE foram descritas até agora: dimérica com domínio único e dependente de amônio chamada de NadENH3 e a octamérica com dois domínios, um de hidrólise da L-glutamina e outro de síntese do NAD+, chamada de NadEGln. A presença de múltiplas isoformas é relativamente comum em procariotos. Neste trabalho, identificamos um novo grupo de NadE diméricas dependentes de glutamina em bactéria. A determinação da preferência de substrato e a análise estrutural sugerem que as enzimas NadEGln podem constituir intermediários evolutivos entre as NadENH3 diméricas e as NadEGln octaméricas. A caracterização de isoformas adicionais no organismo diazotrófico Azospirillum brasilense junto com a determinação dos níveis de glutamina em resposta ao choque de amônio nos levou a propor um modelo em que essas diferentes isoformas da NadE se tornam ativas de acordo com a disponibilidade de nitrogênio. Isso pode explicar a pressão seletiva que corrobora a coexistência de múltiplas isoformas da NadE em diversos procariotos. Adicionalmente, análise de vizinhança gênica realizada anteriormente indica que o gene nadEGln é frequentemente agrupado com o gene que codifica para as proteínas transdutoras de sinal PII, sugerindo uma relação funcional entre PII e NadE em resposta ao estado nutricional e a razão carbono/nitrogênio na célula bacteriana. Proteínas PII são reguladoras que sinalizam os níveis de energia e de nitrogênio através da ligação dos metabólitos ATP, ADP e 2-oxoglutarato. Para investigar essa hipótese, cromatografia de afinidade e ensaios de interferometria biolayer foram realizados para mostrar que a PII e a NadEGln interagem in vitro. Ensaios de atividade da enzima NadEGln demonstram que a formação do complexo alivia a inibição por retroalimentação negativa do produto da reação, o NAD+. Além do mais, 2- oxoglutarato, um efetor alostérico de PII e indicador do nível de nitrogênio celular, inibiu a formação do complexo PII-NadEGln dentro de níveis fisiológicos. Análise bioinformática sugere que esse mecanismo é conservado em bactérias distantes filogeneticamente. Os resultados dessa tese indicam uma relação funcional entre os efetores de PII ATP, ADP e 2-oxoglutarato e o inibidor da NadEGln, o NAD+, representando um novo eixo de regulação metabólica, que balança os níveis de nitrogênio e carbono. Concluindo, nossas descobertas apontam que PII age como uma subunidade regulatória dissociável da NadEGln, desse modo permitindo o controle da biossíntese de NAD+ de acordo com o estado nutricional da célula bacteriana.

Palavras chave: NAD+, metabolismo de NAD+, balanço de nitrogênio, balanço de carbono, ATP, ADP, 2-oxoglutarato, NadE, PII, glutamina.

ABSTRACT

NAD+ is a central metabolite participating in core metabolic redox reactions together with several oxidoreductase enzymes. Likewise, NAD+ is used as a substrate by ADP- ribosyl transferases, sirtuins, and the NAD+-dependent DNA ligase. NAD+ biosynthesis is one of the most fundamental biochemical pathways and the ubiquitous NAD+ synthetase (NadE) catalyzes its final step. Two different classes of NadE have been described to date: dimeric single-domain ammonium-dependent NadENH3 and octameric two-domain glutamine-dependent NadEGln. The presence of multiple NadE isoforms is relatively common in prokaryotes. Here, we identified a novel dimeric group of NadEGln in bacteria. Substrate preferences and structural analyses suggested that dimeric NadEGln enzymes may constitute evolutionary intermediates between dimeric NadENH3 and octameric NadEGln. The characterization of additional NadE isoforms in the diazotrophic bacterium Azospirillum brasilense along with the determination of intracellular glutamine levels in response to an ammonium shock led us to propose a model in which these different NadE isoforms became active accordingly to the availability of nitrogen. These data may explain the selective pressures that support the coexistence of multiple isoforms of NadE in several prokaryotes. Additionally, previous gene neighborhood analysis indicated that bacterial nadEGln gene is frequently clustered with the gene encoding the PII signal transduction proteins, suggesting a functional relationship between these proteins in response to the nutritional status and the carbon/nitrogen ratio of the bacterial cell. PII proteins are regulators that sense energy and nitrogen levels through the binding of the metabolites ATP, ADP, and 2-oxoglutarate. To investigate this hypothesis, ligand-fishing affinity chromatography and a biolayer interferometry assay were performed to show PII and NadEGln interaction in vitro. NadEGln enzyme activity demonstrates that this complex relieves NadEGln from negative feedback inhibition by its product NAD+. Moreover, 2- oxoglutarate, a PII allosteric effector and cellular nitrogen level indicator, inhibited the formation of the PII–NadEGln complex within a physiological range. Bioinformatic analyses suggest that this mechanism is conserved in distantly related bacteria. Our results indicate an interplay between the PII effectors ATP, ADP, and 2-OG, and the NadEGln inhibitor NAD+ that represents a novel metabolic hub, balancing the levels of nitrogen and carbon. Our findings support that PII proteins act as a dissociable regulatory subunit of NadEGln, thereby enabling the control of NAD+ biosynthesis according to the nutritional status of the bacterial cell.

Key words: NAD, NAD metabolism, nitrogen balance, carbon balance, ATP, ADP, 2- oxoglutarate, NadE, PII, glutamine.

ZUSAMMENFASSUNG

NAD+ ist ein zentrales Metabolit, welches zusammen mit mehreren Oxidoreduktase- Enzymen an Kernstoffwechsel-Redox-Reaktionen teilnimmt. Ebenso wird NAD+ von ADP-Ribosyltransferasen, Sirtuinen und der NAD+-abhängigen DNA-Ligase als Substrat verwendet. Die NAD+-Biosynthese ist einer der grundlegendsten biochemischen Wege, in welchem die ubiquitäre NAD+-Synthetase (NadE) den letzten Schritt katalysiert. Bisher wurden zwei verschiedene Klassen von NadE Enzymen beschrieben: NadENH3 - von Ammonium abhängige Dimere mit einer enzymatischen Domäne und NadEGln - von Glutamin abhängige Oktamere mit zwei Domänen. Das Vorhandensein mehrerer NadE-Isoformen ist bei Prokaryonten relativ häufig. In dieser Arbeit wurde eine neue dimere Gruppe von NadEGln Enzymen in Bakterien identifiziert. Substratpräferenzen und Strukturanalysen legen nahe, dass dimere NadEGln-Enzyme evolutionäre Zwischenstufen zwischen dimeren NadENH3 und oktameren NadEGln darstellen. Die Charakterisierung zusätzlicher NadE-Isoformen in dem diazotrophen Bakterium Azospirillum brasilense zusammen mit der Bestimmung des intrazellulären Glutaminspiegels in Reaktion auf einen Ammoniumschock führte uns dazu, ein Modell vorzuschlagen, in dem diese verschiedenen NadE-Isoformen entsprechend der Verfügbarkeit von Stickstoff aktiv werden. Diese Daten können die selektiven Faktoren erklären, die die Koexistenz mehrerer NadE-Isoformen in Prokaryonten hervorgerufen haben. Darüber hinaus deuten Analyse zur genetischen Nachbarschaft darauf hin, dass das bakterielle nadEGln-Gen häufig mit einem Gen geclustert ist, das ein PII- Signaltransduktionsprotein kodiert, was auf eine funktionelle Beziehung zwischen diesen Proteinen als Reaktion auf den Ernährungszustand und des Kohlenstoff/Stickstoff-Verhältnisses der Bakterienzelle hindeutet. PII-Proteine sind Regulatoren, die den Energie- und Stickstoffgehalt durch die Bindung der Metaboliten ATP, ADP und 2-Oxoglutarat wahrnehmen. Zur Untersuchung dieser Hypothese wurden „ligand-fishing affinity” Chromatografie und ein „Biolayer interferometry assay” durchgeführt, um PII- und NadEGln-Interaktion in vitro zu zeigen. Die Enzymtests zeigten, dass dieser Komplex NadEGln von der negativen Rückkopplungshemmung durch sein Produkt NAD+ schützt. Darüber hinaus zeigte sich, dass 2-Oxoglutarat, ein allosterischer PII-Effektor und zellulärer Indikator der Stickstoffkonzentration, die Bildung des PII-NadEGln-Komplexes schon in physiologischen Konzentrationen hemmt. Bioinformatische Analysen deuten darauf hin, dass dieser Mechanismus auch in entfernt verwandten Bakterien konserviert ist. Diese Ergebnisse deuten auf ein Zusammenspiel zwischen den PII-Effektormolekülen ATP, ADP und 2-OG und dem NadEGln-Inhibitor NAD+ hin, welches einen neuartigen metabolischen Knotenpunkt im Stickstoff- und Kohlenstoff-Stoffwechsel darstellt. Außerdem weisen die Ergebnisse darauf hin, dass PII-Proteine als eine dissoziierbare regulatorische Untereinheit von NadEGln wirken und dadurch die Steuerung der NAD+-Biosynthese entsprechend dem Ernährungszustand der Bakterienzelle bewerkstelligen.

Schlüsselwörter: NAD, NAD-Stoffwechsel, Stickstoffbilanz, Kohlenstoffbilanz, ATP, ADP, 2-Oxoglutarat, NadE, PII, Glutamin.

FIGURE SUMMARY

INTRODUCTION

Figure 1: Molecular structure representation of Nicotinamide adenine dinucleotide in the oxidized form...... 16 Figure 2: Interconversion of NAD+, NADP+, NADH, and NADPH...... 18 Figure 3: Structure of the E. coli CobB (PDB 1S5P)...... 23 Figure 4: L-tryptophan Kynurenine pathway and L-aspartate pathway routes to produce quinolinate...... 29 Figure 5: Scheme of the NAD+ metabolic pathway...... 32 CHAPTER I: Kinetics and structural features of dimeric Gln-dependent bacterial NAD+ synthetases suggest evolutionary adaptation to available metabolites

Figure 1: Sequence alignment of different NadEGln...... 48 Figure 2: Gel filtration analysis of the different NadE...... 51 Figure 3: Intracellular levels of glutamine and NifH and GS modification after an ammonium shock in A. brasilense...... 53 Figure 4: Phylogenetic analysis of selected NadEGln...... 55 Figure 5 Comparison between the structures of dimeric and octameric NadEGln...... 58 Figure 6: Structure of the dimeric NadE2Gln from B. thailandensis...... 60 Figure 7: Model for the role of the three NadE in A. brasilense...... 63

Figure S1: Gel filtration profile of HsNadE2...... 68 Figure S2: Determination of kinetic parameters for NAD synthetases from A. brasilense and H. seropedicae...... 69 Figure S3: SDS-PAGE analysis of A. brasilense NadE1Gln, NadE2Gln and NadENH3. 70 Figure S4: Gel filtration profile of AbNadE2...... 71 Figure S5: Superposition of the NAD+ synthetase domain of NadE2Gln from Burkholderia thailandensis (red) and NadENH3 from Bacillus subtilis (green)...... 72

CHAPTER II: NAD+ biosynthesis in Bacteria is subject to global Carbon/Nitrogen control via PII signaling

Figure 1: Reaction scheme of NAD synthetases...... 75 Figure 2: In vitro complex formation between AbNadE2 and GlnZ...... 85 Figure 3: The AbNadE2 activity is down regulated by NAD+ and GlnZ relives AbNadE2 from NAD+ inhibition...... 87 Figure 4: Effect of 2-oxoglutarate in the dissociation of GlnZ-NadE2Ab complex and enzymatic activity of AbNadE2...... 89 Figure 5: Effect of PII protein on NadE2 activity depends on the PII variant and is conserved in different Bacteria...... 91 Figure 6: Similarity based grouping of NadE orthologues identified in the PII-NadE genomic islands...... 93 Figure 7: The PII proteins act as a dissociable regulatory subunit of NadE2Gln to pace NAD+ production accordingly to the availability of L-glutamine and 2-OG...... 95

Figure S1 Putative binding partners of the A. brasilense PII protein GlnZ...... 120 Figure S2: Determination of the dissociation constant of the His-GlnZ and AbNadE2 complex in the presence of ADP by Bio-layer Interferometry...... 121 Figure S3: The NadEGln type 2 are activated in reductive environment when using L- glutamine as N-donor...... 122 Figure S4: Effect of NAD+ in the kinetic parameters of AbNadE2 reaction...... 123 Figure S5: In vitro formation of the NadE2-GlnZ complex does not require the GlnZ T loop and AbNadE2 preferentially interacts with the PII protein GlnZ...... 124

TABLE SUMMARY

CHAPTER I: Kinetics and structural features of dimeric Gln-dependent bacterial NAD+ synthetases suggest evolutionary adaptation to available metabolites

Table 1 Kinetic properties in different Gln-dependent NadE ...... 50 Table 2 Ammonium versus glutamine competition in H. seropedicae NadE1 and NadE2 ...... 50

Table S1 Primers used in this work ...... 67

CHAPTER II: NAD+ biosynthesis in Bacteria is subject to global Carbon/Nitrogen control via PII signaling

Table S1 Microsoft Excel Worksheet file (attached) ...... 118 Table S2 Bacterial strains and plasmids ...... 118

SUMMARY

1. INTRODUCTION ...... 16

NICOTINAMIDE ADENINE DINUCLEOTIDE IN THE LIVING CELL ...... 16 NAD+ AS A SUBSTRATE OF REGULATORY ENZYMES ...... 18 ADP-ribosylating enzymes consume NAD+ ...... 19 Bacterial NAD-dependent DNA ligases consume NAD+ for growth ...... 25 RNA polymerases use NAD+ as non-canonical initiating nucleotide ...... 27 NAD+ BIOSYNTHESIS PATHWAYS ...... 28 De novo NAD+ biosynthesis pathways ...... 28 The salvage NAD+ biosynthesis pathway...... 31 NAD SYNTHETASE ENZYMES ...... 33 REGULATION OF NAD+ BIOSYNTHESIS ...... 35 PII PROTEINS ARE SIGNAL TRANSDUCTION PLAYERS IN BACTERIA ...... 37 2. CHAPTER I: KINETICS AND STRUCTURAL FEATURES OF DIMERIC GLN- DEPENDENT BACTERIAL NAD+ SYNTHETASES SUGGEST EVOLUTIONARY ADAPTATION TO AVAILABLE METABOLITES ...... 41

ABSTRACT ...... 41 INTRODUCTION ...... 41 MATERIAL AND METHODS ...... 44 Plasmid construction ...... 44 Protein purification ...... 44 Gel filtration and MW determination ...... 45 NadE kinetic analysis...... 45 LC/MS analysis ammonium versus glutamine substrate competition ...... 46 Intracellular levels of glutamine and protein post-translational modification analysis ...... 46 RESULTS ...... 48 Characterization of ??H. seropedicae NadE2Gln (HsNadE2) ...... 48 Characterization of the three NadE from A. brasilense ...... 50 Intracellular glutamine levels in A. brasilense ...... 52 Phylogeny analysis of dimeric NadEGln ...... 54 Structural analysis of dimeric NadE2Gln...... 56 DISCUSSION ...... 61 ACKNOWLEDGEMENTS ...... 64 REFERENCES ...... 64 SUPPORTING INFORMATION ...... 67 FRONT PAGE OF THE PUBLISHED ARTICLE IN JBC (2018) ...... 73 3. CHAPTER II: NAD+ BIOSYNTHESIS IN BACTERIA IS SUBJECT TO GLOBAL CARBON/NITROGEN CONTROL VIA PII SIGNALING ...... 74 ABSTRACT ...... 74 INTRODUCTION ...... 74 METHODS ...... 77 Ligand fishing His-GlnZ affinity chromatography ...... 77 Construction of the plasmid expressing NAD synthetase from Synechocystis sp, GlnD and GlnZ∆Tloop from A. brasilense...... 79 Protein expression and purification ...... 80 NAD synthetase activity assays ...... 81 In vitro protein complex analysis ...... 81 Bio-layer interferometry assays ...... 81 Redox titrations of NadE ...... 82 Bioinformatic analysis of PII-NadE genomic islands ...... 83 RESULTS ...... 84 Identification of the dimeric NadEGln as a novel target of the signal transduction protein GlnZ ...... 84 In vitro complex formation between purified GlnZ and A. brasilense NadE2 ...... 84 The AbNadE2 is feedback inhibited by NAD+ ...... 85 The feedback inhibition of AbNadE2 by NAD+ is relieved by GlnZ ...... 87 Effects of GlnZ uridylylation and the GlnZ T-loop on AbNadE2 activity ...... 90 The regulatory interplay between NAD+ and PII proteins is conserved among bacterial dimeric glutamine dependent NadEGln ...... 92 Bioinformatic analysis of the PII-nadE gene islands ...... 92 DISCUSSION ...... 94 ACKNOWLEDGMENTS ...... 97 REFERENCES ...... 98 SUPPORTING INFORMATION ...... 118 Supporting Figures ...... 120 FRONT PAGE OF THE PUBLISHED ARTICLE IN JBC (2020) ...... 125 4. CONCLUSION ...... 126

5. BIBLIOGRAPHY ...... ERRO! INDICADOR NÃO DEFINIDO.

1. INTRODUCTION

Nicotinamide Adenine Dinucleotide in the living cell

Nicotinamide adenine dinucleotide (NAD) is found in all living cells and is composed of two nucleotides bound through their phosphate groups. One nucleotide contains an adenine nucleobase and the other a nicotinamide base (Figure 01). NAD is a pyridine dinucleotide that may exist in an oxidized (NAD+) or a reduced form (NADH) (Figure 02). The interconversion between NAD+ and NADH is catalyzed by dehydrogenases. NAD+ can continuously cycle between the NAD+ and NADH, serving as the electron carrier in most of the catabolic reactions, including the ubiquitous glycolysis and the tricarboxylic acid (TCA) cycle reactions.

Figure 1: Molecular structure representation of Nicotinamide adenine dinucleotide in the oxidized form. In black the phosphate and ribose groups together with the nucleotide rings of NAD+. In blue the nicotinamide nucleobase and green the adenine nucleobase. Source: The author (2020)

To maintain the redox homeostasis the NAD+ which is reduced to NADH during catabolism must be re-oxidized. The two major pathways that provide NADH oxidation

16 are: 1) The respiratory chain in which electron transfer from NADH to a final acceptor such as oxygen (aerobic respiration) or other compounds such as nitrate (anaerobic respiration) is coupled to ATP production; 2) Fermentation reactions in which electrons from NADH are delivered to organic compounds to release a reduced final product along with NAD+ (PLUMBRIDGE; DEUTSCHER, 2019). While the NADH/NAD+ pair is primarily used as an electron carrier in catabolic reactions. Electron transfer in anabolic reactions occurs using the phosphorylated form of the co-factor, namely, the NADPH/NADP+ pair. NAD+ is phosphorylated to NADP+ by NAD kinases (NADK). NADK activity responds to the cellular REDOX state1 and intracellular NAD+ levels (HOLM et al., 2010; ISHIKAWA; KAWAI-YAMADA, 2019). NADK is essential in E. coli, Mycobacterium tuberculosis, and Bacillus subtilis (GERDES et al., 2002; KOBAYASHI et al., 2003; SASSETTI; BOYD; RUBIN, 2003). This enzyme is also important to phototrophic cyanobacteria given the key role of NADPH in photosynthesis (HURLEY et al., 2002). The extra phosphate group on NADPH does not affect the estimated standard redox potentials of NADPH compared with NADH (HUANG et al., 2012), but rather gives to NADP+ a slightly different shape, making it possible for multiple redox reactions to discriminate this co-factor within the cell (Figure 02). Generally, the NADP/NADPH is used in anabolic reactions and is produced, in most cases, by the pentose pathway, in the step mediated by glucose-6-phosphate dehydrogenase (G6PD) reviewed by (POLLAK; OLLE; ZIEGLER, 2007). Other NADPH-producing enzymes worth to be mentioned are NADP-specific forms of isocitrate dehydrogenase (IDP), malic enzyme (ME), and aldehyde dehydrogenase (ALDH). Due to adenine moiety, NAD+/NADH and NADP+/NADPH absorb ultraviolet light (UV). NAD+ peak absorption is at a wavelength of 259 nanometers (nm), with an extinction coefficient of 16.9 mM−1cm−1 (HALD; LEHMANN; ZIEGENHORN, 1975). NADH also absorbs at higher wavelengths, with a second peak in UV absorption at 340 nm with an extinction coefficient of 6.22 mM−1cm−1 (HALD; LEHMANN; ZIEGENHORN, 1975). This difference in the ultraviolet absorption spectra between the oxidized and reduced forms at higher wavelengths makes it simple to measure

1 The redox state is defined here as the NAD+/NADH together with NADP+/NADPH ratio (SCHAFER; BUETTNER, 2001). 17 their enzymatic interconversion by monitoring UV absorption at 340 nm (BERGMEYER; BERNT, 1974).

Figure 2: Interconversion of NAD+, NADP+, NADH, and NADPH. The hydrogen atom bound to the nicotinamide nucleobase ring (green). In the reduced forms, a second hydrogen atom is present. In red the position of the phosphorylated ribose from the adenine nucleobase. Note the reduction position on NADH is far from the phosphorylation site for NADP(H). The arrows indicate a reversible redox reaction catalyzed by NAD(H)-dependent dehydrogenases (several EC) from NAD+ to NADH, NADP(H)-dependent dehydrogenases (several EC), NAD kinases (EC 2.7.1.23) from NAD+ to NADP+, and NADH kinases (EC 2.7.1.86) from NADH to NADPH.

Source: The author (2020)

NAD+ as a substrate of regulatory enzymes

Nucleotides belonging to the NAD family (NAD+, NADP+, NADH, and NADPH) are obligatory metabolites in all organisms. They serve as electron carriers for a diverse

18 set of enzymes in a plethora of metabolic pathways. Perturbations in NAD+/NADH and NADP+/NADPH ratios affect growth and have a general effect in the coupling between catabolic and anabolic activities in prokaryotes (CANELAS; VAN GULIK; HEIJNEN, 2008; EBERT et al., 2011; HOLM et al., 2010; HOU et al., 2009; KIM et al., 2009). In addition to the well-established role of NAD+ as a co-factor in core metabolic reactions, NAD+ is used as a substrate for several enzymes in plants, mammals (as detailed reviewed for plants here FERNIE et al. (2020) including humans STRØMLAND et al. (2019)). In bacteria, NAD+ acts as a substrate for diverse enzymes including lysine deacetylases, DNA ligases, protein ADP-ribosylating enzymes, and enzymes involved in RNA modification (SORCI; RUGGIERI; RAFFAELLI, 2014). The activity of NAD+-consuming enzymes presumably reduces NAD+ levels while augmenting the levels of NAD+ consumption byproducts (usually NAD+ biosynthesis intermediates, such as nicotinamide and nicotinamide mononucleotide). Perturbations in NAD+ levels may have significant negative effects on cellular metabolism (HOLM et al., 2010). Hence, it is not surprising that nature developed a range of sensor proteins that monitor the levels of NAD+ as well as of the NAD+ enzymatic byproducts and elicit the appropriate metabolic response to reestablish NAD+ homeostasis. Here we review the various NAD-consuming enzymes described in Bacteria. We then describe the different de novo and salvage NAD+ biosynthesis pathways. Finally, we consider the regulatory aspects of NAD+ biosynthesis with an emphasis on potential connections with the activity of NAD consuming enzymes and metabolic regulation.

ADP-ribosylating enzymes consume NAD+

ADP-ribosylation is the transfer of a single or multiple ADP-ribose unit(s) from NAD+ to a variety of molecules including amino acid residues on a target protein. ADP- ribosylation not only modifies proteins but also DNA and tRNA, further supporting the versatility of this modification (JANKEVICIUS et al., 2016; MUNNUR et al., 2019). In several pathogenic bacteria, ADP-ribosylation promotes the modification and inactivation of rifamycin, thus, a mechanism of antibiotic resistance (LI et al., 1996). ADP-ribosylation is distributed in all domains of life and is also described in viruses (PERINA et al., 2014). ADP-ribosylation reactions are catalyzed by two main superfamilies of enzymes: the ADP-ribosyltransferases (ARTs) and the sirtuins (ARAVIND et al., 2015).

ARTs are an extremely diverse superfamily of proteins and their activity results mostly in inactivation of a target protein. Such a mechanism was primarily studied in pathogenic bacteria that release ARTs as toxins to modify host proteins and gain an advantage in the infection biochemistry (COLLIER, 2001; CORDA; GIROLAMO, 2003). The ARTs are phylogenetically classified in three main clades accordingly to their characteristic catalytic motif: (1) the H-H-h clade is divided into two families, the tRNA 2´ phosphotransferase and the CC0527 family; (2) the H-Y-[QED] clade that comprises several families, including the diphtheria toxin-like ARTs, poly-ADP-ribose polymerases (PARPs) and mono-ARTs; (3) R-[ST]-E clade that comprises several bacterial toxins such as heat-labile toxin of E. coli ETEC strains, the cholera toxin, bacteriophage ARTs (Alt, ModA and ModB) and extracellular ARTs (ARAVIND et al., 2015). Additionally, the R-[ST]-E clade has several subclades that are in deep reviewed by PALAZZO, MIKOČ, and AHEL (2017). This broad and diverse superfamily of ARTs is clear evidence of the importance of NAD+ as a substrate rather than only acting as a redox cofactor in bacteria. The single protein reversible ADP-ribosylation that act as a metabolic regulation mechanism is the DraT/DraG enzymatic pair, which regulates nitrogenase activity in diazotrophic bacteria (MOURE et al., 2014). The dinitrogenase reductase ADP-ribosyl- transferase (DraT) enzyme is part of the H-Y-[QED] ARTs and is present on organisms such as Azospirillum brasilense, Rhodobacter capsulatus, and Rhodospirillum rubrum (HUERGO et al., 2012). Under unfavorable environmental conditions, nitrogenase is rapidly inactivated by DraT catalyzed arginine mono-ADP-ribosylation, thereby preventing nitrogen fixation during energy-limiting or nitrogen-sufficient conditions (BATISTA; DIXON, 2019). The inactivation reaction is reversed by dinitrogenase reductase-activating glycohydrolase (DraG) which hydrolyzes arginine ADP-ribose N- glycosidic thereby reactivating nitrogenase (MOURE et al., 2014). In A. brasilense, the activation and deactivation of nitrogenase are regulated by PII signal transduction proteins by an elegant mechanism. When ammonium is available, the PII family member GlnZ protein sequestrates DraG from the cytosol by forming a ternary complex with the ammonium transporter AmtB. The GlnZ protein not only sequesters DraG to the membrane but also acts to block ammonium transport through AmtB (HUERGO et al., 2006c). At the same time, the second PII family member, GlnB, interacts and activates DraT, promoting the nitrogenase ADP-ribosylation (HUERGO

20 et al., 2006a, 2006c; MOURE et al., 2013). When nitrogen is limiting, PII proteins are post-translationally modified by uridylylation (PII-UMP), which prevents GlnZ to interact to AmtB, DraG became active to remove of ADP ribose from nitrogenase. Meanwhile, GlnB-UMP does not interact with DraT thereby resulting in inactive DraT. These processes allow full reversion of nitrogenase activity under nitrogen limitation (HUERGO et al., 2006a, 2006c). The rationale of the PII signaling control over nitrogenase switch-on/off is to avoid nitrogen fixation, a very energy-demanding process, while external ammonium uptake is available. Beyond the well-established response to nitrogen status, signaling regulation is been suggested in Rhodospirillum rubrum to control nitrogen fixation accordingly to energy levels. The drop in NAD(P)H levels, that occur in the transition from light to darkness in phototrophic organisms was shown to inhibit nitrogenase activity (BROSTEDT et al., 1997). The proposed explanation was the reduced pyrimidine nucleotides from the NAD(P)H pool are oxidation in the electron transport pathway to nitrogenase, therefore, lower levels of NAD(P)H would decrease nitrogenase turn-over (BROSTEDT et al., 1997). On the other hand, the accumulation of NAD(P)+ in the darkness was also proposed as a factor for activation of the NAD+-dependent DraT (AKENTIEVA, 2018). Additionally, it has been shown that energy shifts induce DraG membrane sequestration independently of nitrogen status (WANG et al., 2018). Measurements of NAD(P)+/NAD(P)H levels in the transition from switch-on to switch- off of the nitrogenase would help to unravel novel regulations for nitrogen fixation. Lysine deacetylation as an NAD+-consuming Post-Translational Modification (PTM) The reversible acetylation/deacetylation of the Lysine side chain acts as ubiquitous reversible regulatory PTM in bacteria (WEINERT et al., 2013). The recent development of techniques that allow the identification/quantification of protein acetyl-lysine modification at proteome-scale, coined as “the acetylome”, is helping to unravel the metabolic significance of this PTM (MACEK et al., 2019). Acetylation neutralizes the positive charge of the lysine residues, thereby altering protein electrostatic surface and structure to regulate processes such as protein- protein interactions, DNA-protein interactions, protein cellular localization, and/or enzyme activity (CARABETTA; CRISTEA, 2017). The acetylation can be mediated by lysine acetyltransferases (KATs). Alternatively, lysine acetylation may occur non-enzymatically in the presence of acetyl-

21 phosphate/acetyl-CoA (ABOUELFETOUH et al., 2015) which is likely to explain most of the in vivo protein lysine acetylation. In Bacillus subtilis, in vitro analysis showed that acetyl-CoA synthetase (AcsA) is directly acetylated by acetyl phosphate or by acetyl- CoA. However, under low acetyl-CoA levels, acetylation of AcsA only occurs when catalyzed by the AcuA acetyltransferase (WANG; YOU; YE, 2017). Regardless of the enzymatic or non-enzymatic nature of the processes, the number of acetyl/succinyl- modified proteins surpasses the number of phosphorylated- proteins in proteomic studies, further supporting that lysine acetylation is of interest for bacterial metabolism regulation (MACEK et al., 2019). Lysin deacetylation in bacteria is mediated by lysine deacetylases. Three families have been described in bacteria: Zn2+-dependent deacetylases, NAD+-dependent sirtuins, and YcgC deacetylases which do not require Zn2+ or NAD+ (CARABETTA; CRISTEA, 2017). Zn2+-dependent deacetylases are hydrolases that release acetate, whereas sirtuins use NAD+ as an acetyl acceptor, producing nicotinamide and 2´-O- acetyl-ADP-ribose. Sirtuins are broadly distributed in all domains of life (FRYE, 2000). The different families of deacetylases seem to target a distinct set of protein substrates, including metabolic enzymes and translation regulators. Lysin acetylation seems to be intimately linked to energy metabolism (ZHANG et al., 2009). CobB is a bacterial NAD+-dependent sirtuin, belonging to the same group of the Sir2 protein in yeast, a well-characterized sirtuin involved in histone lysine deacetylation (FRYE, 2000). CobB is a monomeric protein, composed of two domains. A large domain adopts a canonical Rossmann fold for the NAD-binding and a small domain that contains a structural zinc ion and is composed of three antiparallel β- strands (β5-β7), a short β-strand (β2) and three helices (α2-α4) and presents a highly flexible structure (Figure 03). The binding of the NAD+ molecule occurs at the α1-β2 loop, located between the large and the small domains. A second loop, β7-α7, binds to the acetyl-lysine reside at the target protein (ZHAO; CHAI; MARMORSTEIN, 2004).

Figure 3: Structure of the E. coli CobB (PDB 1S5P). The zinc ion domain is represented in green and the Rossmann domain is represented in blue. The structure of CobB is composed of nine α-helices and ten β-strands. Two Loops are indicated Loop α1- β2 and β7-α7, for NAD+ and acetyl-lysine binding, respectively. Figure was generated by PyMOL Molecular Graphics System.

Source: The author (2020), modified from ZHAO; CHAI; MARMORSTEIN (2004).

In E. coli, CobB was characterized as the main lysin deacetylase, a cobB mutant strain exhibited no deacetylase activity when the cells are grown in tryptone broth supplemented with glucose (ABOUELFETOUH et al., 2015). The cobB deletion altered the global acetylome profile, confirming the role of CobB in regulation lysin acetylation in vivo (BAEZA et al., 2014; CASTAÑOǦCEREZO et al., 2014). The acetylome seems to vary greatly accordingly to culture conditions and the methodology employed for acetylome analysis (ABOUELFETOUH et al., 2015; BAEZA et al., 2014; CASTAÑOǦCEREZO et al., 2014; WEINERT et al., 2013). In Bacillus subtilis, the acetylation rate depends on the carbon source (KOSONO et al., 2015). cobB gene deletion in Yersinia pestis affects several cell functions such as chemotaxis, flagellar assembly, gene expression, and contributed to lower virulence and lower stress responses (LIU et al., 2018). Y. pestis strains defective in the production of the NAD-dependent deacetylase, named YfiQ, showed similar phenotypes, probably as a

23 result of the imbalance of lysin acetylation/deacetylation. Broad range metabolic effects were also observed in cobB knockouts in B. subtilis and Mycobacterium smegmatis (GARDNER; ESCALANTE-SEMERENA, 2009; GU et al., 2015). In addition to this physiological evidence, in vitro studies further support the role of lysine acetylation in regulating bacterial metabolism. One of the best studied examples is the conserved regulation of acetyl-CoA synthetase (AcsA) by YfiQ/CobB regulatory system. AcsA is a central enzyme, synthesizing acetyl-CoA from acetate, ATP, and CoA (STARAI et al., 2003). In S. enterica, AcsA is regulated by Pat acetylase (the lysine acetyltransferase from S. enterica) and a NAD-dependent deacetylase. The K609 residue is the sole acetylation site in S. enterica AcsA (STARAI et al., 2002). The mechanism seems to be a simple on-off switch when AcsA is acetylated by Pat (in response to high acetyl-CoA levels), AcsA activity is inhibited thereby decreasing acetyl-CoA biosynthesis (STARAI et al., 2002, 2003). Conversely, when CobB deacetylates AcsA, the AcsA activity is restored (STARAI et al., 2002, 2003). Remarkably, bacterial CobB regulation is still poorly understood. In E. coli, CobB is inhibited by nicotinamide in vitro and in vivo (GALLEGO-JARA et al., 2017). Nicotinamide is one of the products of the deacetylase reaction, and in eukaryotes nicotinamide signals NAD+ depletion (BOCKWOLDT et al., 2019). Nicotinamide acts as a noncompetitive inhibitor of CobB (GALLEGO-JARA et al., 2017). In + Saccharomyces cerevisiae, the CobB orthologue Sir2 has KM for NAD of 30 μM, the NAD+ concentration in yeast ranges from 1.5 to 2 mM, which means that NAD+ substrate does not limit Sir2 activity (BORRA et al., 2004). However, Sir2 has a nicotinamide inhibition constant of a60 μM. If one considers the estimated nicotinamide levels in yeast in the 10-150 μM (YANG; SAUVE, 2006) then it is very likely that nicotinamide levels control lysin deacetylation. The control of acetyl-CoA synthetases by sirtuins deacetylases is also conserved in mammals (STRØMLAND et al., 2019). In mice, two acetyl-CoA synthetases are present: AceCS1, localized in the cytoplasm; and AceCS2 localized in the mitochondria. The sirtuins SIRT1 and SIRT3 deacetylate to activate AceCS1 and AceCS2, respectively (HALLOWS; LEE; DENU, 2006). The ubiquitous role of NAD+ dependent sirtuins in the control of acetyl-CoA synthetases activity suggest an important interplay between acetyl-CoA and NAD+ levels. In fact, these metabolites

24 are well characterized as allosteric regulators of major metabolic enzymes (KLIMOVA et al., 2019). It is not completely clear which factors from the cell regulate lysine acetylation/deacetylation cycle in bacteria. However, lysine acetylation is related to the carbon source used for growth. As the acetylation process is highly dependent on acetyl-CoA(P) levels, and the deacetylation is mediated by NAD+-dependent sirtuins, the levels of acetyl-CoA/P and NAD+ metabolites may control the rate of acetylation/deacetylation. Indeed, in E. coli acetyl-P correlates with the acetylation rate (WEINERT et al., 2013) Levels of acetyl-CoA(P) and NAD+/NADH are allosteric regulators of carbon metabolism, therefore, it is reasonable to think they would also play a role in lysine acetylation to regulate protein function. To our knowledge, there is no data demonstrating a link between NAD+ and protein acetylation in bacteria. On the other hand, Eukaryotic NAD+-dependent sirtuins are shown to respond with different deacetylation rates in response to the cellular NAD+ levels (VASSILOPOULOS; WANG; GIUS, 2018).

Bacterial NAD-dependent DNA ligases consume NAD+ for growth

DNA ligase catalyzes the formation of phosphodiester bond connecting the 3´end of one Okazaki fragment to the 5’ beginning of the next fragment during DNA replication. DNA ligase also acts in DNA recombination and repair. DNA ligases are divided into two families according to the cofactor required; ATP-dependent or NAD+- dependent (LEHNMAN, 1974). The ATP-dependent DNA ligases are widespread in Archaea, viruses, Eukarya, and found in some eubacteria. On the other hand, NAD+- dependent DNA ligases are widespread in eubacteria. Few exemptions are found in nature such as the NAD-dependent DNA ligases in a poxvirus (SRISKANDA; MOYER; SHUMAN, 2001); and the human DNA ligase IV, a key enzyme in the DNA double- strand break repair, which uses NAD+ as an alternative adenylation donor (CHEN; YU, 2018). NAD-dependent DNA ligases are present in all bacteria and are essential for growth (GOTTESMAN; HICKS; GELLERT, 1973; KACZMAREK et al., 2001; KONRAD; MODRICH; LEHMAN, 1973; PARK et al., 1989; PETIT; DUSKO EHRLICH, 2000). There is little sequence identity between the DNA-ligases families. However, high conservation of five structural motifs (Motif I, III, IIIa, IV, and V) and overall protein 25 structure suggests the use of similar reaction mechanisms. Both types of DNA ligases catalyze the sealing of 5′-phosphate and 3′-hydroxyl termini at nicks in duplex DNA in three sequential reactions: (1) the attack on the α-phosphorus of ATP or NAD+ by ligase results in the release of pyrophosphate or nicotinamide mononucleotide (NMN) and formation of an adenylate intermediate, linked to a lysine residue; (2) the AMP is transferred to the 5′ end of the 5′-phosphate-terminated DNA strand to form DNA- adenylate (AppDNA); (3) ligase catalyzes attack by the 3′-OH of the nick on DNA- adenylate to join the two polynucleotides and release AMP (SRISKANDA; SHUMAN, 2002). In the case of the NAD-dependent DNA ligases, an additional domain named Ia is responsible for the interaction between NAD+ and LigA (E. coli DNA-ligase) (SRISKANDA; SHUMAN, 2002). The binding of NAD+ to Ia domain substitutes de Motif VI from ATP-dependent DNA ligases with a significant structural change in the enzyme structure (GEORLETTE et al., 2003; SRISKANDA; SHUMAN, 2002). The biochemical characterization of NAD+-dependent DNA-ligases are of special interest for medicine, its restriction to eubacteria support that they may be promising targets for the development of antibiotics (GEORLETTE et al., 2003; SRISKANDA; SHUMAN, 2002). In E. coli, both ATP and NAD-dependent DNA ligases are present (SRISKANDA, 2001). Evolutionary, the co-presence of ATP/NAD-dependent enzymes in E. coli and other distant related groups suggests several horizontal gene transfers events (WILKINSON; DAY; BOWATER, 2001). It is not clear yet if there is a biochemical or regulatory reason for the cofactor specificity. The E. coli LigA is sufficient to support mitotic growth in Saccharomyces cerevisiae lig4 knockout (coding for ATP-dependent DNA ligase in yeast) (SRISKANDA et al., 1999). + Given the low KM for NAD (in the range of 50 μM) and the reported intracellular levels of NAD+ of 2.6 mM, it is unlikely that NAD-dependent DNA ligases respond to + NAD levels in vivo. The E. coli LigA is feed-back inhibited by its product, NMN (IC50 of 31.3 μM), although no information on the type of inhibition was provided (ALOMARI, 2018). The accumulation of byproducts generated by the activity of the various NAD+- consuming enzymes may affect general NAD+ metabolism. For instance, NMN, ADP- ribose, and nicotinamide levels are suggested to control DNA ligase, transcription factors, and the NAD+ biosynthesis, respectively (BOCKWOLDT et al., 2019). A

26 potential cross-regulation between NAD+ consuming enzymes and NAD+ biosynthesis is reinforced by gene neighborhood analysis. In some bacteria, the NAD-biosynthetic gene cluster is frequently located in the same loci as ligA, encoding NAD+-consuming DNA-ligase (SORCI; RUGGIERI; RAFFAELLI, 2014).

RNA polymerases use NAD+ as non-canonical initiating nucleotide

The adenine moiety of the NAD+ structure can be used as initial nucleoside during transcription, resulting in an ab initio mechanism of RNA capping (BIRD et al., 2016). In addition to NAD+, other nucleotide-containing metabolites serve as substrates for non-canonical initiating nucleotide (NCIN) transcription, such as 5´-desphospho coenzyme A, flavin adenine dinucleotide, uridine diphosphate glucose (UDP-glucose), and uridine diphosphate N-acetylglucosamine (UDP-GlcNAc) (BIRD et al., 2016; JULIUS; YUZENKOVA, 2017). Although their roles in vivo are not yet confirmed. These NCINs (NAD+, NADH, dpCoA, and FAD) compete with ATP during the transcription initiation process. The 5’ NAD+-modified RNAs have a promoter consensus sequence (VVEDENSKAYA et al., 2018). After the transcription, the resulting RNA carries NAD+ conjugate at the 5´terminal (CHEN et al., 2009). In E. coli, the technology of modified NAD+ capture and sequencing leads to the identification of a subset of specific regulatory small RNAs (sRNA) and sRNA-like 5´-terminal fragments of some mRNAs (CAHOVÁ et al., 2015; WINZ et al., 2017). The physiological role of NAD-capping is still not completely understood. Analogous to the eukaryotic capping (7-methyl guanosine modification to the 5´ terminal), it was tested in vitro if the prokaryotic NAD+-capping affects RNA decay. The presence of 5’ NAD+ increases RNA stability against endonucleolytic degradation by RNase E and RppH in vitro (CAHOVÁ et al., 2015). Furthermore, degradation of 5’ NAD-RNA starts with a decap reaction, catalyzed by Nudix phosphohydrolase NudC resulting in 5´-monophosphate RNA and NMN (CAHOVÁ et al., 2015). 5´- monophosphate RNA, is susceptible to rapid degradation by the RNA degradosome (CARPOUSIS, 2007). It has been speculated that the use of NADH instead of NAD+ may link the redox status directly to RNA processing (JÄSCHKE et al., 2016). Indeed, NudC recognizes only NAD-RNAs, not NADH-RNAs (CAHOVÁ et al., 2015). Moreover, it is tempting to speculate if NudC or other putative nudix phosphohydrolases are allosterically

27 regulated or respond to signaling factors in the cell. Regardless of NudC function, NAD- capping is being related to RNA stability in B. subtilis (FRINDERT et al., 2018), and toxin production in Staphylococcus aureus (MORALES-FILLOY et al., 2019). Another example of novel the NAD+ function is that the E. coli DNA primase DnaG can use NAD+ adenine moiety to initiate its primase activity (JULIUS; YUZENKOVA, 2019). Altogether, the data described in this section illustrates that NAD+ function is far beyond its well-established role as a simple electron carrier.

NAD+ biosynthesis pathways

In most organisms NAD+ biosynthesis occurs through de novo and salvage pathways, most of the NAD+ intermediates can be interconverted in a complex manner (HACHISUKA; SATO; ATOMI, 2018; SORCI et al., 2009). De novo biosynthesis pathway can start with either L-tryptophan or L-aspartate, these pathways converge upon the formation of Quinolinate (Figure 4), which is then funneled to a more conserved set of the enzymatic pathway to generate NAD+ (Figure 5).

De novo NAD+ biosynthesis pathways

L-Tryptophan is a precursor for non-phototrophic Eukaryotes Fungi, animals, and some bacteria produce quinolinate from the amino acid L- tryptophan. In this route, also called the kynurenine pathway, 6 enzymatic steps are necessary to convert tryptophan into quinolinate. The enzymes operating are: tryptophan 2,3-dioxygenase (TDO) (EC:1.13.11.11)/indoleamine 2,3-dyoxygenase (IDO) (EC:1.13.11.52), arylformamidase (IAFM) (EC:3.5.1.9), kynurenine 3- monooxygenase (KMO) (EC:1.14.13.9), kynureninase (KYU) (EC:3.7.1.3), and 3- hydroxy-anthranilate 3,4-dioxygenase (HAO) (EC:1.13.11.6) (Figure 04).

Figure 4: L-tryptophan Kynurenine pathway and L-aspartate pathway routes to produce quinolinate. In the Kynurenine pathway (yellow box), tryptophan 2,3-dioxygenase (TDO) (EC:1.13.11.11)/indoleamine 2,3-dyoxygenase (IDO) (EC:1.13.11.52) catalyzes the first step, arylformamidase (IAFM) (EC:3.5.1.9) the second, kynurenine 3-monooxygenase (KMO) (EC:1.14.13.9) the third, kynureninase (KYU) (EC:3.7.1.3) the forth, and 3-hydroxy-anthranilate 3,4-dioxygenase (HAO) (EC:1.13.11.6) the fifth. The last step is spontaneous. For the aspartate pathway, aspartate oxidase (AO) (EC:1.4.3.16) participates in the first step and the second quinolinate synthase (QS) (EC:2.5.1.72).

Source: The author (2020).

The rate-limiting step in this pathway is the formation of N-formylkynurenine by incorporation of dioxygen across the C2-C3 bond of tryptophan. This reaction in mammals is catalyzed by TDO and IDO, both are heme-containing enzymes. TDO is found also in some bacterial species such as Xanthomonas campestris and Ralstonia metallidurans (FOROUHAR et al., 2007; ZHANG et al., 2007), and yeast (IWAMOTO

29 et al., 1995). In humans, TDO is more expressed in the liver and skin, meanwhile, IDO is ubiquitously expressed (RAFICE et al., 2009). Another aspect worth to mention is the role of the second step mediated by IAFM. The intermediate L-kynurenine is the precursor for the synthesis of kynurenic acid, an important metabolite with neuroprotective and anticonvulsant properties (WALCZAK et al., 2019). The biological role of kynurenic acid is still not completely understood, studies showed a correlation between the levels of kynurenic acid and carcinogenesis in peripheral tissues (WALCZAK et al., 2019). More studies are necessary to determine the biochemical properties of this pathway. One could speculate that the advantage of the tryptophan pathway would be that it may provide intermediate metabolites for other metabolic routes yet to be discovered. The evolutionary origin of the tryptophan pathway is still uncertain. It is more likely that it appeared first in prokaryotes and was acquired by the common Eukaryotic ancestor. A second horizontal gene transfer resulted in the transfer of the aspartate pathway to phototrophic eukaryotes by endosymbiont cyanobacteria (TERNES; SCHÖNKNECHT, 2014). Limiting aspects of the evolutionary studies of the L- tryptophan NAD+ biosynthetic pathway are uncertainties related to gene annotation in bacterial genomes and lack of deeper structural/biochemical studies.

L-Aspartate is an NAD+ precursor in the majority of prokaryotes The de novo NAD+ biosynthesis pathway from L-aspartate is present in the majority of prokaryotes and green plants (KURNASOV et al., 2003). It comprises two steps: aspartate oxidase (AO) (EC:1.4.3.16 - in E. coli named NadB) which converts L-aspartate to iminoaspartate, using FAD+ as a cofactor and oxygen or fumarate as electron acceptors, under aero or anaerobiosis, respectively (BEGLEY et al., 2001) (Figure 04). The AO enzyme is a dimer that is inhibited by NAD+ due to competition with the FAD cofactor (TEDESCHI et al., 1999). This provides an NAD+ negative feedback loop in the rate-limiting step of the NAD+ biosynthesis. In Thermotoga maritima, there is no AO but an NAD+-dependent aspartate dehydrogenase (TM1643) (EC:1.4.1.21) which also produces iminoaspartate (YANG et al., 2003). As iminoaspartate is unstable (decompose to pyruvate and oxaloacetate in aqueous solution) it is very likely that TM1643 or AO form a protein complex with quinolinate synthase (QS) (OSTERMAN, 2009). Thought, the complex formation has not been confirmed experimentally. 30

In the second step QS (EC:2.5.1.72) (in E. coli named NadA) catalyzes the condensation of iminosuccinate and dihydroxyacetone phosphate (sub-product of the glycolysis and Calvin cycle in plants) with the removal of phosphate and two water molecules to produce quinolinic acid (Figure 04) (KURNASOV et al., 2003). QS was shown to contain one [4Fe-4S] cluster per polypeptide, making this enzyme oxygen- sensitive (CICCHILLO et al., 2005). The Fe of the iron-sulfur cluster is used to form an intermediate with the carboxylate group of iminoaspartate (ESAKOVA et al., 2019). Nonetheless, several groups of bacteria appear to have a functional L-aspartate NAD+ biosynthetic pathway but lack the QS coding gene (OSTERMAN, 2009). This suggests the existence of another, yet to be discovered, enzyme that substitute QS function.

Tryptophan and aspartate pathways converge in the final steps of de novo NAD+ biosynthesis Quinolinate produced from either by the L-tryptophan or L-aspartate routes is phophoribosylated to nicotinate adenine mononucleotide (NaMN) by the quinolinic acid phosphoribosyltransferase enzyme (EC:2.4.2.19 - named in E. coli NadC). NadC is highly conserved within a sample of 400 genomes analyzed Osterman (2009). The structure of NadC from yeast (DI LUCCIO; WILSON, 2008; PANOZZO et al., 2002) and Mycobacterium tuberculosis (SHARMA; GRUBMEYER; SACCHETTINI, 1998) showed that NadC is a homohexameric enzyme which uses Mg2+ as a cofactor to catalyze the transfer of the phosphoribosyl moiety from 5-phosphoribosyl-1- pyrophosphate to quinolinate producing NaMN, pyrophosphate, and CO2.

The salvage NAD+ biosynthesis pathway

NAD+ regeneration from nicotinamide In the previous section, we discussed the role of NAD+ as a consumable substrate from important regulatory processes. Both protein ADP-ribosylation and lysine deacetylation processes can produce significant amounts of nicotinamide (GAO et al., 2019). In order to maintain a balance between the redox function and the consumption of NAD+ as an enzyme-substrate, NAD+ should be replenished and this occurs using not only de novo but also NAD+ salvage pathways. The majority of the NAD+ salvage pathways described to date converge, along with de novo biosynthetic pathways, at the point of Nicotinic acid mononucleotide (NaMN) (OSTERMAN, 2009) (Figure 5). It worth mentioning that some pathogenic bacteria lack the de novo NAD+

31 biosynthetic routes and completely rely on host NAD+ or intermediates (GAZZANIGA et al., 2009a).

Figure 5: Scheme of the NAD+ metabolic pathway. Source: The author (2020). Salvage from nicotinamide (Nm) starts with hydrolytic deamination to nicotinic acid (Na) catalyzed by nicotinamidase (PncA) (Figure 05) (EC:3.5.1.19). Na is adenylylated to NaMN by Na phosphoribosyltransferase (PncB) (EC:6.3.4.21). PncB uses the energy of ATP hydrolysis to drive the synthesis of NaMN and pyrophosphate from Na and phosphoribosyl pyrophosphate (Figure 05) (GROSS; RAJAVEL; GRUBMEYER, 1998). There is a different Nm salvage pathway that is apparently restricted to a few groups of bacteria and its evolutionary origin is unknown. This pathway was better characterized in Synechocystis sp. where Nm is phosphorylated to NMN by Nm phosphoribosyltransferase coded by slr0788 gene (named nadV) (EC: 2.4.2.12), and 32 in the second step, NMN is adenylated directly to NAD+ by the bifunctional enzyme NMN adenylyltransferase/ADP-ribose pyrophosphatase (coded by slr0787 named nadM) (EC: 2.7.7.1). NadM uses NMN and ATP, yielding ADP and NAD+ (HUANG et al., 2008; RAFFAELLI et al., 1999b). NAD+ regeneration from nicotinamide mononucleotide Nicotinamide mononucleotide NMN levels are correlated to bacterial growth by the action of the LigA enzyme. During replication, LigA uses NAD+ as a substrate for the ligase reaction, yielding significant amounts of NMN. It is known that LigA activity is inhibited by NMN (ALOMARI, 2018). The most common pathway for NMN salvage starts with the enzyme NMN deaminase (named PncC) (EC:3.5.1.42) which catalyzes the deamidation of NMN to NaMN (SORCI et al., 2014). PncC is widely distributed among bacteria, absent only in obligate pathogens, symbionts, and few exceptional groups, being substituted in those organisms by NadV-NadM, that is able to produce NAD+ directly from the scavenged Nm to NMN and then to NAD+ (SORCI; RUGGIERI; RAFFAELLI, 2014).

The classical Preiss-Handler pathway In 1958, Preiss and Handler described the biosynthesis of NAD+ from Na in extracts of red blood cells, yeast, and liver (PREISS; HANDLER, 1958). In the Preiss- Handler pathway, Na is converted to NaMN by PncB. At this point, all de novo and most of the described salvage pathways converge (Figure 05). The NaMN is adenylated to nicotinic acid adenine dinucleotide (NaAD) by the nicotinic acid mononucleotide adenylyltransferase (named NadD) (EC:2.7.7.18). NadD is essential for the synthesis of NAD+, thus, is a promising target for the design of antibiotics (RODIONOVA et al., 2014). The last step in the NAD+ biosynthesis is catalyzed by an amidotransferase enzyme NadE which converts NaAD to NAD+ (Figure 05). Quite remarkably, the two last steps in NAD+ biosynthesis are well conserved among distant taxonomic groups (GAZZANIGA et al., 2009a; GERDES et al., 2006; RODIONOVA et al., 2014).

NAD synthetase enzymes

The NAD synthetases (NadE) are ATP-dependent amidotransferases that catalyze the last step of the NAD+ biosynthesis, using ammonia to amidate NaAD to NAD+ at the expense of ATP hydrolysis yielding pyrophosphate (PPi) (PACE;

BRENNER, 2001). In the NadE reaction ATP is hydrolyzed to AMP and PPi and the pyridine carboxylate group in NaAD nicotinamide is activated by adenylation, forming a NaAD-AMP intermediate. This allows the nucleophilic attack by ammonia to amidate NaAD, releasing NAD+ and AMP. Several NadE enzymes have the synthetase activity together with an additional glutaminase domain that hydrolyzes L-glutamine to L-glutamate and ammonia. In those two-domain NadE enzymes, L-glutamine is used as the nitrogen donor to provide ammonia for the synthetase domain. On the other hand, single domain NadE, lacking the glutaminase domain, is believed to recruit ammonia directly from the medium (DE INGENIIS et al., 2012). Alternatively, these single domain NadE may form a protein complex with glutaminase enzymes in order to provide ammonia for NadE synthesis. A likely scenario for the evolutionary origin of the two-domain NadE suggests an initial recruitment of a glutaminase enzyme to function as a subunit of NadE by protein- protein interaction. This was followed by a domain fusion between the two enzymes (DE INGENIIS et al., 2012). Single-domain NadE, which is thought to be the archetypical member and are named ammonia-dependent NadE (NadENH3). NadENH3 are present in several organisms, such as Salmonella enterica serovar Typhimurium (DE INGENIIS et al., 2012), Bacillus anthracis (MCDONALD et al., 2007), Helicobacter pylori (KANG et al., 2003), Escherichia coli (JAUCH et al., 2005), and Bacillus subtilis (SYMERSKY et al., 2002a). All bacterial single domain ammonia-dependent NadE described to date are homodimeric (JAUCH et al., 2005; PARK; YEO; LEE, 2014). The second class of NadE are those which have the additional N-terminal glutaminase domain (RESTO; YAFFE; GERRATANA, 2009). This group of enzymes is named glutamine dependent NadE (NadEGln). The glutaminase domain of glutamine-dependent NadE belongs to the superfamily of nitrilases, characterized to have a conserved Glu-Lys-Cys catalytic triad with the cysteine being the nucleophile of the glutamine reaction (BRENNER, 2002; PACE; BRENNER, 2001). NadEGln was described in Mycobacterium tuberculosis (BELLINZONI et al., 2005), Herbaspirillum seropedicae (LASKOSKI et al., 2016), Thermus thermophilus (DE INGENIIS et al., 2012), and Thermotoga maritima (RESTO; YAFFE; GERRATANA, 2009). Evolutionarily, it has been suggested that NadEGln emerged in a common ancestor

34 between Eubacteria and Eukaryota, being lately horizontally transferred to Archaea (DE INGENIIS et al., 2012). Analyses of the M. tuberculosis NadE (MtNadE) structure revealed that MtNadE evolved a synergistic mechanism to allow communication between glutaminase and synthetase domains in such way that glutamine hydrolysis is coupled to NAD+ production, avoiding wasteful glutamine hydrolyzes in the absence of synthetase domain substrates (LARONDE-LEBLANC; RESTO; GERRATANA, 2009a). This mechanism is triggered by the formation of the NaAD-AMP intermediate in the synthetase domain, which in turn, activates the glutaminase domain (CHUENCHOR et al., 2012). The ammonia released within the glutaminase domain is then canalized to the synthetase domain trough a particular ammonia tunnel (CHUENCHOR et al., 2012). The MtNadE has a similar overall structure to human NadE, it is a homoctamer composed of two parallel arranged tetramers (CHUENCHOR et al., 2020). The subunits of the octameric NadE are connected by an ammonia tunnel of 40 Å (1 Å=0.1 nm) with mixed hydrophilic and hydrophobic properties, a feature that seems to be exclusive for the NadEGln (CHUENCHOR et al., 2012). The characterization of the Thermotoga maritima NadEGln indicated different kinetic properties regarding the interdomain synergism. While MtNadE has a nearly 1:1 efficiency in the conversion of 1 L-glutamine to 1 NAD+ (LARONDE-LEBLANC; RESTO; GERRATANA, 2009a), in T. maritima NadE, glutaminase domain performs wasteful hydrolysis of L-glutamine without the equimolar NAD+ formation (RESTO; YAFFE; GERRATANA, 2009). T. maritima NadEGln was suggested to be an ancestral form of NadEGln (RESTO; YAFFE; GERRATANA, 2009). Structural data and characterization of other NadEGln members are needed to confirm such hypothesis

Regulation of NAD+ biosynthesis

NAD+ and NADH are key signals of the redox balance and energy metabolism. The levels of these cofactors are regulated by fine-tuning mechanisms that vary accordingly to bacterial groups. Some gamma-proteobacteria employ the tri-functional enzyme NadR to regulate NAD+ levels (GERASIMOVA; GELFAND, 2005). NadR has three domains including nicotinamide nucleotide adenylyltransferase (EC:2.7.7.1), a ribosyl nicotinamide kinase (EC:2.7.1.22), and a transcriptional regulator for NAD+ related genes (MAGNI et al., 2004; RODIONOV et al., 2008a). 35

The nucleotide adenylyltransferase activity converts NMN directly to NAD+ (EC:2.7.7.1) at the expense of ATP hydrolysis. This activity is observed for NadR in E. coli (RAFFAELLI et al., 1999a), Salmonella enterica (GROSE; BERGTHORSSON; ROTH, 2005) and Haemophilus influenzae (KURNASOV et al., 2002). H. influenzae lacks the de novo NAD+ biosynthesis, thus, the scavenging of NMN from the host may be crucial for H. influenzae metabolism. The NadR of H. influenzae KM for NMN is 0.14 mM (KURNASOV et al., 2002), while in E. coli the Km for NMN is 0.7mM (RAFFAELLI et al., 1999a). One the other hand, Salmonella enterica NadR KM for NMN is 12.8 mM, which is unlikely to be functional physiologically (GROSE et al., 2005). The second NadR activity is the phosphorylation of ribosyl nicotinamide (RNm) to NMN catalyzed by the kinase domain (Figure 05), RNm is the preferred substrate of NadR from Salmonella enterica (GROSE; BERGTHORSSON; ROTH, 2005). The third activity of NadR is the transcriptional repression which is operated by an N- terminal helix-turn-helix (HTH) domain. The HTH domain is only present in enterobacterial NadR (GERASIMOVA; GELFAND, 2005). All three functions of NadR are controlled by NAD+ levels. The enzymatic activities are inhibited by high NAD+ levels. This negative feedback mechanism reduces NAD+-precursor uptake and usage (MAGNI et al., 2004). Furthermore, NadR represses NAD+ biosynthetic genes (including nadB, nadA, nadC, and pncB) under high NAD+ levels allowing negative feedback inhibition of de novo biosynthesis at the transcriptional level (RODIONOV et al., 2008a). In Bacillus, Clostridium, and Thermotogales, a second transcriptional regulator, named NiaR, controls the expression of NAD+ biosynthetic genes. NiaR from Bacillus subtilis is a repressor of operon nadABC and the nicotinic acid transporter, niaP. Transcription repression occurs in the presence of nicotinic acid (Na) (RODIONOV et al., 2008b). Nicotinic acid is transported into the cell and then converted directly into NaMN by PncB. When Na is abundant, the de novo biosynthetic genes are repressed, whereas the salvage pathway using Na remains active. In other organisms, such as Streptococcus pneumoniae, NiaR controls the transcription of pncAB, and several other NAD+-related genes (AFZAL; KUIPERS; SHAFEEQ, 2017). The transcription regulator NrtR was found and characterized in Synechocystis sp., Shewanella oneidensis (RODIONOV et al., 2008a), and Mycobacterium tuberculosis (GAO et al., 2019). NrtR senses levels of ADP-ribose, one of the NAD+

36 byproducts produced by enzyme activity such as ARTs. It was proposed that the cell interprets the accumulation of ADP-ribose as a signal of NAD+ depletion (RODIONOV et al., 2008a). In Synechocystis sp. NrtR is a repressor that binds to the promoter region of nadE, nadM-nadV, and nadA genes. ADP-ribose acts as an antirepressor, releasing the transcription of NAD-biosynthetic genes (RODIONOV et al., 2008a). The transcriptional regulation of NAD+ biosynthetic genes in enterobacteria, Bacillus, Clostridium, Thermotogales, Synechocystis sp., Shewanella oneidensis, and Mycobacterium tuberculosis support a strategy to regulate futile of NAD+ biosynthesis. Proteins that sense NAD intermediates operate to reduce the flow from the NAD de novo pathway allowing the reestablishment of the NAD pool using the more economic salvage pathways. In several groups of bacteria, the mechanisms by which NAD pools are controlled remains elusive. It was reported a possible NadQ transcription regulator in alfa-proteobacteria and beta-proteobacteria, but no studies were successful (RODIONOV et al., 2008a). Given the vital importance of NAD for general metabolism and protein PTM it is highly likely that other regulatory loops controlling the NAD pool remain to be discovered.

PII proteins are signal transduction players in Bacteria

In order to maintain the carbon and nitrogen homeostasis bacteria use signaling metabolites and signal transduction proteins to interpret the nutritional status of the cell and launch the appropriate metabolic response. The PII proteins are widespread signal transduction proteins present in a broad range of prokaryotes and the chloroplast of eukaryotic phototrophs (FORCHHAMMER; LÜDDECKE, 2016a; HUERGO; CHANDRA; MERRICK, 2013a). Sequence and structure of PII proteins are highly conserved throughout all domains of life (SANT’ANNA et al., 2009). PII proteins are composed of a homotrimeric barrel-like structure with clefs located between each subunit. Additionally, three characteristic loop regions, named B-,C-, and T-loops, are important for effector molecule binding (TRUAN et al., 2010). The extraordinary ability of PII to sense and integrate metabolic signals is due to the allosteric binding of key metabolites ATP, ADP, 2-oxoglutarate (2-OG), and, in certain cases, L-glutamine (FORCHHAMMER; LÜDDECKE, 2016a). These metabolites indicate cellular nutritional status related to energy (the ATP/ADP ratio); nitrogen (L-glutamine and 2-OG levels are associated with the rate of nitrogen assimilation) and carbon/nitrogen balance as 2-OG is used both as 37 a TCA cycle intermediate, and the main carbon skeleton for nitrogen assimilation (FOKINA et al., 2010; HUERGO; DIXON, 2015). The interplay between allosteric effectors is conserved in most canonical PII proteins studied (HUERGO; CHANDRA; MERRICK, 2013a). Allosteric effector binding induces conformational changes in the PII structure thereby affecting the ability of PII to engage protein-protein interaction with other target proteins (ZETH; FOKINAS; FORCHHAMMER, 2014). The nucleotides ATP and ADP bind to PII in a competitive manner (GERHARDT et al., 2012). The 2-OG binding site is formed when PII is previously occupied by Mg2+-ATP (TRUAN et al., 2010). This implies in cooperative binding between Mg2+-ATP and 2-OG such that Mg-ATP binding affinity increases in the presence of 2-OG and vice-versa (JIANG; NINFA, 2007). PII has three effector- binding sites located in the lateral clefts between each of its three subunits. In addition to the allosteric effectors ATP, ADP and 2-OG, some bacterial PII proteins respond indirectly to L-glutamine by suffering reversible post-translational modification. In proteobacteria, PII proteins are subject to PTM mediated by the glutamine-sensitive bifunctional uridylyl-transferase/removing enzyme (GlnD) (EC:2.7.7.59) (MERRICK, 2015). GlnD promotes the reversible uridylylation of a conserved tyrosine (Y51) residue located in the flexible T-loop signalized by nitrogen restriction (ZHANG et al., 2010). As a bifunctional enzyme, GlnD also promotes PII- UMP deuridylylation when the levels of 2-OG decrease and the levels of L-glutamine increase (ARAUJO et al., 2004; BONATTO et al., 2012; PALANCA; RUBIO, 2017). In some cyanobacteria, PII is phosphorylated instead of uridylylated in response to nitrogen levels. The T-loop phosphorylation was detected in Synechococcus elongatus (FORCHHAMMER; DE MARSAC, 1994, 1995) and Synechocystis sp. PCC 6803 (SPÄT; MACEK; FORCHHAMMER, 2015). So far, a single PII kinase was not identified (FORCHHAMMER; SELIM, 2020). By contrast, the PII-specific phosphatase (named PphA) was characterized in Synechocystis PCC 6803. PphA dephosphorylate PII accordingly to the effector binding in vitro (RUPPERT et al., 2002). Compared to no effector, PII bound only to ATP or ADP is less dephosphorylated, and in the presence of 2-OG, the dephosphorylation is almost completely inhibited (RUPPERT et al., 2002). The expression of PphA is controlled by the nitrogen source, for instance, the presence of nitrate or nitrite stimulated the expression of PphA (KLOFT; RASCH; FORCHHAMMER, 2005).

If one considered only the allosteric effector binding, PII proteins may exist theoretically in up to 21 different structural conformations (DA ROCHA et al., 2013). The presence of reversible PTM significantly augments the possible range of different PII conformations. The structural changes in response to allosteric effector binding and PTM affect the PII ability to interact and regulate key metabolic enzymes, transcriptional regulators, and transporters (FORCHHAMMER, 2008; HUERGO; DIXON, 2015). Among the best described PII targets are enzymes involved in nitrogen metabolism including nitrogen assimilation and nitrogen fixation (HUERGO; CHANDRA; MERRICK, 2013a; HUERGO; DIXON, 2015; NINFA; JIANG, 2005). The best studied PII target is the ammonium transport protein AmtB (JAVELLE; MERRICK, 2005). In several organisms, the gene encoding AmtB is co-transcribed with the PII gene (SANT’ANNA et al., 2009). Previously, we described the role of PII in the PTM of the nitrogenase enzyme through the DraG-DraT system in A. brasilense (see ADP-ribosylation consumes NAD+ for enzyme deactivation). Under high nitrogen, the 2-OG level is low and PII is bound to ADP (HUERGO; DIXON, 2015). Such conformation facilitates DraT-PII complex formation while promoting the formation of the ternary between AmtB, PII, and the DraG-GlnZ-AmtB complex is thought to be required for DraG inactivation (MOURE et al., 2019). The complexes PII-DraT, and PII- DraG dissociate when 2-OG increases when nitrogen becomes available (GERHARDT et al., 2012). The PII proteins have also been implicated in the control of fatty acid biosynthesis and this process seems to be ubiquitous from Bacteria to plants (FERIA BOURRELLIER et al., 2010; GERHARDT et al., 2015; HAUF et al., 2016; RODRIGUES et al., 2014, 2019). The unique sensory properties of PII support that it may have a broader range of protein targets. The conservation of gene context is frequently used as a fingerprint of putative proteins interacting partners (DANDEKAR et al., 1998). In prokaryotes, the gene encoding NadEGln is frequently co-localized with the gene encoding the regulatory signal transduction protein PII (glnB) (SANT’ANNA et al., 2009). Conservation of the glnB-nadE gene pair leads to the hypothesis that NadEGln and PII could physically interact and/or that PII proteins could somehow affect the NadEGln function (HUERGO; CHANDRA; MERRICK, 2013a). The investigation of these functional links is the major goal of this Thesis.

Genome analysis showed that some prokaryotes can encode up to three different NadE (DE INGENIIS et al., 2012). Two different NadEGln have been identified in the β- proteobacterium Herbaspirillum seropedicae and the α-proteobacterium Azospirillum brasilense. In a previous study, we presented the biochemical characterization of HsNadE1Gln, which resembled the properties of the archetypical NadEGln from M. tuberculosis (LASKOSKI et al., 2016). Here we describe the biochemical characterization of a second NadEGln in H. seropedicae (named HsNadE2Gln) and A. brasilense (named AbNadE2Gln) and provide a novel hypothesis for the co-occurrence of different isoforms of NadE (SANTOS et al., 2018) (CHAPTER I). We show that NadE2Gln enzymes are negatively feedback-inhibited by physiological levels of NAD+ and that such a regulatory loop is conserved among distantly related bacteria. We show that PII protein act as a dissociable regulatory subunit of dimeric NadE2Gln in bacteria. Complex formation between PII and NadE2Gln relieves the NadE2Gln inhibition by NAD+, thereby acting as a switch to coordinate NAD+ production with nutrient availability in prokaryotes (SANTOS et al., 2020) (CHAPTER II).

2. CHAPTER I: Kinetics and structural features of dimeric Gln-dependent bacterial NAD+ synthetases suggest evolutionary adaptation to available metabolites

This research was originally published in the Journal of Biological Chemistry. Adrian Richard Schenberger Santos, Edileusa Cristina Marques Gerhardt, Vivian Rotuno Moure, Fábio Oliveira Pedrosa, Emanuel Maltempi Souza, Riccardo Diamanti, Martin Högbom and Luciano Fernandes Huergo. Kinetics and structural features of dimeric Gln-dependent bacterial NAD+ synthetases suggest evolutionary adaptation to available metabolites. J Biol Chem. 2018; Vol 293:7397-7407. © Abstract

NAD+ and its reduced form NADH serve as cofactors for a variety of oxidoreductases that participate in many metabolic pathways. NAD+ also is used as substrate by ADP- ribosyl transferases and by sirtuins. NAD+ biosynthesis is one of the most fundamental biochemical pathways in nature, and the ubiquitous NAD+ synthetase (NadE) catalyzes the final step in this biosynthetic route. Two different classes of NadE have been described to date: dimeric single-domain ammonium-dependent NadENH3 and octameric glutamine-dependent NadEGln, and the presence of multiple NadE isoforms is relatively common in prokaryotes. Here, we identified a novel dimeric group of NadEGln in bacteria. Substrate preferences and structural analyses suggested that dimeric NadEGln enzymes may constitute evolutionary intermediates between dimeric NadENH3 and octameric NadEGln. The characterization of additional NadE isoforms in the diazotrophic bacterium Azospirillum brasilense along with the determination of intracellular glutamine levels in response to an ammonium shock led us to propose a model in which these different NadE isoforms became active accordingly to the availability of nitrogen. These data may explain the selective pressures that support the coexistence of multiple isoforms of NadE in some prokaryotes.

Introduction

NAD+ and its reduced form NADH serve as cofactors for a variety of oxidoreductases that participate in a diverse range of metabolic pathways. More recently, NAD+ has also been implicated in signal transduction mechanisms. Studies in mammals showed that a conserved Nudix domain NAD+ sensor protein regulates DNA repair mechanisms, thereby relating NAD+ levels to cancer and aging (LI et al., 2017). NAD+ is also used as substrate for signaling pathways by mono- and poly-ADP- 41 ribosyl-transferases and by sirtuins (ARAVIND et al., 2015; BHEDA et al., 2016; MOURE et al., 2014). Furthermore, manipulation of the NAD+/NADH levels in microorganisms can help to develop or improve a plethora of biotechnological processes. NAD+ biosynthesis is a fundamental biochemical process in all cells, it can occur de novo or through salvage pathways. The last reaction step in de novo pathway and in some of the NAD+ salvage pathways is the amidation of nicotinic acid adenine dinucleotide (NaAD) to form NAD+, a reaction catalyzed by the ubiquitous NAD+ synthetase (NadE) (GAZZANIGA et al., 2009b). In the first reaction step, NadE uses NaAD and ATP as substrates, forms a NaAD-AMP intermediate (thereby activating the carboxyl group of nicotinamide) and releases PPi. In the second reaction + step, NH3 acts as a nucleophile to attack NaAD-AMP releasing NAD and AMP as final products (BRENNER, 2002). Two different classes of NadE are found in nature. Single domain ammonium dependent NadENH3 (EC 6.3.5.1) is present in Bacteria and Archaea and uses external ammonium as the N donor. These enzymes are homodimers and have been extensively characterized biochemically and structurally (JAUCH et al., 2005; RIZZI et al., 1996). The second class is the glutamine-dependent NadEGln (EC 6.3.1.5) which is present in Eukarya, Bacteria, and Archaea and is able to use L-glutamine as N donor because of the presence of an extra N-terminal glutaminase domain (CN hydrolase domain) (DE INGENIIS et al., 2012). Just a few NadEGln have been characterized in prokaryotes (BELLINZONI et al., 2002; LASKOSKI et al., 2016; RESTO; YAFFE; GERRATANA, 2009). The best studied NadEGln is from Mycobacterium tuberculosis (MtNadEGln), this enzyme is arranged as a homoctamer where the N-terminal glutaminase domain from one subunit connects with the C-terminal synthetase domain from another subunit. The ammonia released from glutamine hydrolysis is directed to the synthetase domain through a 40 Å intersubunit ammonia tunnel (LARONDE-LEBLANC; RESTO; GERRATANA, 2009b). The MtNadEGln exhibits strong synergism between the two catalytic domains such that efficient glutamine hydrolysis only occurs in the presence of NaAD and ATP and depends on the formation of the NaAD-AMP intermediate.

Conversely, the presence of glutamine significantly decreases the Km for NaAD and ATP.

Extensive genomic context analysis revealed that a few prokaryotic single- domain NadE (presumably NadENH3 as judged by domain organization) are actually clustered with a separate gene encoding a glutaminase and, at least in the case of Thermus thermophilus, these domains can interact to form a holoenzyme that is able to utilize glutamine to direct NAD+ formation in vitro (DE INGENIIS et al., 2012). Based on these studies, an evolutionary scenario has been postulated where a single- domain NadENH3 was present in last universal common ancestor (LUCA), and the two- domain NadEGln evolved via recruitment and fusion of the N-terminal glutaminase domain (DE INGENIIS et al., 2012). Genome analysis revealed that some prokaryotes can encode up to three different NadE (DE INGENIIS et al., 2012). Two different genes encoding two-domain NadEGln have been identified in the diazotrophic endophytic β- Proteobacterium Herbaspirillum seropedicae. In a previous study we presented the biochemical characterization of HsNadE1Gln, which resembled the properties of the archetypical NadEGln from M. tuberculosis; it is also arranged as a homoctamer and preferentially uses glutamine and N donor (LASKOSKI et al., 2016). Here we describe the biochemical characterization of the second NadEGln from H. seropedicae (named HsNadE2Gln). Surprisingly, we observed that HsNadE2Gln forms a lower oligomeric structure (probably a homodimer) and can use both ammonium and glutamine as N donor. From now on we will use the term NadENH3 to describe single domain homodimeric enzymes. NadEGln will be used to describe enzymes carrying the additional N-terminal glutaminase domain with numbers 1 and 2, indicating quaternary structures arranged as octamers or dimers, respectively. Oligomerization studies of the two NadEGln detected in the diazotrophic α- Proteobacterium Azospirillum brasilense also showed that AbNadE1Gln forms an octamer, whereas AbNadE2Gln is a dimer. Furthermore, PDB searches identified two structures of dimeric NadE2Gln: one in the β-Proteobacterium Burkholderia thailandensis (PDB code 4F4H) and the other in the δ-Proteobacterium Acinetobacter baumannii (PDB code 5KHA). Phylogenetic analysis showed that all these dimeric NadE2Gln form a separate group distinct from the octameric NadE1Gln. Structural comparisons between dimeric A. baumannii and octameric MtNadE1Gln show that the dimeric and octameric proteins possess similar domain interaction geometry between

43 the glutaminase and synthetase domains achieved through an interprotein domain swapping in the octameric enzymes. In addition to the substrate-binding sites, a conserved intersubunit ammonia tunnel could be detected. We speculate that dimeric NadE2Gln are evolutionary intermediates between single-domain dimeric NadENH3 and two-domain octameric NadE1Gln.

Material and Methods

Plasmid construction

Isolation of plasmid DNA, gel electrophoresis, bacterial transformation, and cloning were performed using standard protocols. T4 DNA ligase and the restriction enzymes NdeI and BamHI were purchase from Fermentas and were used following the manufacturers instructions. The genes nadE2 from H. seropedicae SMR1, nadE1, nadE2, and nadE3 from A. brasilense FP2 were amplified by PCR using DNA from boiled culture as template and high-fidelity polymerase Pfu (Fermentas). PCR fragments and expression vectors pET28a, pET29a, and pTEV5 were digested with NdeI and BamHI and ligated. The primers used in PCR are listed on Table S1. DNA sequencing was performed using dye- labeled terminators (BigDye® ABI PRISM) in an automated DNA sequencer ABI 3500 from Applied Biosystems. Recombinant plasmids obtained in this work were: Gln pASnade2 (Express H. seropedicae NadE2 with a N-terminal His6 tag in pET28a), NH3 pASnade3 (Express A. brasilense NadE with a N-terminal His6 tag in pTEV5), pLHnade2 (Express native A. brasilense NadE2Gln in pET29a), and pASnade1

(Express A. brasilense NadE1 with a N-terminal His6 tag in pTEV5).

Protein purification

NadE proteins were overexpressed in freshly transformed E. coli BL21 (λDE3) and purified adapting previous protocols (LASKOSKI et al., 2016). Typically, the cells were grown in LB medium to an A600 nm of 0.7 at 37 °C. The incubation temperature was changed to 16 °C, and protein expression was induced by adding 0.5 mM isopropyl β-D-thiogalactopyranoside. The cells were harvested after 12 h of shaking at 200 rpm. For His-tagged protein purification (AbNadE1Gln, AbNadENH3, HsNadE1Gln, and HsNadE2Gln), the cell pellets from 0.8 liter of culture were resuspended in 40 ml of buffer A: 50 mM Tris-HCl, pH 8.0, 100 mM NaCl, 20% (v/v)

44 glycerol. The cells were lysed by sonication and centrifuged. The soluble fraction used for His-NadE purification using Ni2+ affinity chromatography on a Hi-trap chelating column (GE Healthcare); the proteins were eluted using an imidazole gradient. Protein concentration was measured at 595 nm by Bradford reaction assay (Bio-Rad).

When indicated, TEV protease was used to remove the N-terminal His6 tag fusion as follows: purified AbNadE1Gln and AbNadENH3 proteins were mixed with His- TEV protease at a 9:1 ratio in 50 mM Tris-HCl, pH 8, 100 mM NaCl, 1 mM DTT. The mixture was incubated at 8 °C for 20 h. His-TEV and uncleaved NadE protein were removed using a Ni2+ Hi-trap chelating column (GE Healthcare). Cleavage was verified by SDS-PAGE electrophoresis. The final preparation has four additional residues (Gly– Ala–Ser–His) at the N terminus of the TEV protease cleavage site. The A. brasilense NadE2Gln was purified without tags. The plasmid pLHnadE2 was used to express untagged NadE2Gln in E. coli BL21 (λDE3) as described above. The cell pellet was resuspended in buffer A (50 mM Tris-HCl, pH 8.0, 100 mM NaCl, 20% (v/v) glycerol) and sonicated. The soluble fraction was loaded onto a Q- Sepharose column, and proteins were eluted with 0.1–1 M linear gradient of NaCl. Fractions containing NadE were pooled and loaded onto a gel filtration HiLoad Superdex 200 16/60 column (GE Healthcare).

Gel filtration and MW determination

Analytical size-exclusion chromatography was performed in a 24 ml Superose 6 column (GE Healthcare) using 50 mM Tris-HCl, pH 8, and 100 mM NaCl as buffer. The column was calibrated with the molecular weight standards: thyroglobulin, bovine γ-globulin, chicken ovalbumin, and equine myoglobin (Bio-Rad). The log of molecular weight was plotted against the Kav (partition coefficient of each standard) to build a linear calibration curve. The molecular weight of each protein was determined by comparing the Kav against the calibration curve.

NadE kinetic analysis

The NadE activity was monitored by coupling the NAD+ production to its reduction to NADH using alcohol dehydrogenase from yeast (Sigma) and monitoring NADH formation at 340 nm as described previously (LASKOSKI et al., 2016). The assays were carried out at 30 °C in 62 mM Tris-HCl, pH 8.5, 20 mM KCl, 20 mM MgCl2, 1.6% (v/v) ethanol, 10 mM DTT, 6 units of yeast alcohol dehydrogenase (Sigma) and 45

50 nM of NadE (monomer concentration). To determine the kinetic constants, the substrates ATP and NaAD were kept at saturating concentrations (10 and 1 mM, respectively), whereas glutamine or ammonium (NH4Cl) varied from 2.5 to 10,000 μM and from 50 to 100,000 μM, respectively. The assays were performed in triplicate, and the initial velocity data were fitted to the Michaelis–Menten equation using GraphPad Prism software package.

LC/MS analysis ammonium versus glutamine substrate competition

Competition assays were performed combining different amounts of glutamine 15 and N-labeled NH4Cl (Sigma). Reactions containing the two substrates were performed in 20 mM Tris-HCl, pH 8, 2 mM ATP, 2 mM NaAD, 5 mM MgCl2, and 1 mM DTT. The reactions were started by adding 50 nM of NadE (monomer concentration) and incubated at 30 °C for 5 min. Acetic acid to 10% (v/v) was added to quench the reactions. The samples were centrifuged at 20,000 × g for 5 min at 4 °C, and 10 μL of the supernatant was used for LC/MS analysis as described (RODRIGUES et al., 2014). Metabolites were separated using a UFLC Prominence (Shimadzu) in a C18 column of 2.6 μm and 50 × 2.1 mm (Phenomenex) kept at 40 °C. The mobile phases were composed of 15 mM acetic acid and 10 mM tributylamine (solvent A) or methanol (solvent B). Samples were placed into vials and kept in an auto-sampler at 4 °C; 10 μL of each sample was injected at 0.2 ml min−1 in duplicate. Metabolites were eluted from the column using a linear gradient (0–100% of solvent B) in 30 min. The detection of the metabolites was performed by coupling the LC with a MicroTOF-QII (Bruker Daltonics), equipped with an electrospray ionization source. The retention time of NAD+ was confirmed using a standard solution. The 15N incorporation was determined by the formula: intensity peak m/z 663 − (0.227 × intensity peak m/z 662). This value represents the increase in the m/z 663.1 signal because of 15N ammonium incorporation subtracted from the theoretical 22.7% isotopic contribution of 1-[13C-14N] NAD+). The value obtained (which corresponds only to [12C-15N] NAD+) was divided by the signal of [14N] glutamine incorporation at 15 + 14 662 m/z to give the N NH4 /[ N] Gln incorporation ratio.

Intracellular levels of glutamine and protein post-translational modification analysis

A. brasilense FP2 was cultured in 110 ml of NFbHPN medium containing 5 mM glutamate as nitrogen source, and nitrogenase activity was determined by the 46 acetylene reduction method using GC exactly as described previously (HUERGO et al., 2006b). The A. brasilense cells were subjected to an ammonium shock (NH4Cl 200 μM) while being kept in a rotatory shaker incubator at 120 rpm and 30 °C. The samples were collected at −5, 0, 0.5, 1, 2, 5, 10, and 20 min after ammonium addition. To analyze the effectiveness of the ammonium shock, the post-translational modification of the GS and NifH proteins were monitored by Western blotting exactly as described previously (HUERGO et al., 2006b). For the extraction of metabolites, 2-ml samples were rapidly collected and quickly cooled in liquid nitrogen. The samples were centrifuged for 2 min at 20,000 × g at 4 °C, and the cell pellets were resuspended in 60 μL of ice-cold solvent (40:40:20 acetonitrile:methanol:water and 0.1 M formic acid) and kept for 20 min at −20 °C as described in (BENNETT et al., 2009). The extracts were centrifuged at 20,000 × g at 4 °C for 3 min, the supernatant was transferred to a new tube, and the pellet was subjected to another round of metabolite extraction. The supernatant from the two extractions were combined and neutralized with 4 μL of ammonium hydroxide. The samples were centrifuged (20,000 × g, 20 min, 4 °C), and supernatant was subject to LC/MS analysis. Separation and detection of metabolites was performed using a LC/MS equipment as described for NadE glutamine versus ammonium completion assays (see above). The high resolution MicroTOF-QII (Bruker Daltonics) was set to an acquisition interval of 50–1000 m/z and sampling rate of 1 s operating on negative MS mode. The capillary voltage was maintained at 3,500 V, and end-plate offset was −500 V. The nebulizer gas flow was 2.0 bar, and the dry gas was kept at 6.0 liters min−1 at 180 °C. All samples were subject to technical replicate runs. The equipment was calibrated with a standard curve of glutamine (R2 > 0.98) before each analysis. All samples were analyzed within the linear range of the calibration curve. The data were processed using QuantiAnalysis (version 4.0 SP1 Bruker). The metabolites were extracted using calculated m/z with an error window of ± 0.01 m/z, and retentions times were confirmed by comparison with standards. To calculate intracellular glutamine concentrations, the number of viable cells in each experiment was counted by serial dilution plating. Metabolite intracellular concentration was calculated based on the number of cells and the estimated intracellular volume of A. brasilense of 2.35 μm3 (1-μm diameter × 3-μm length).

Results

Characterization of ??H. seropedicae NadE2Gln (HsNadE2)

A previous report indicated the presence of two putative glutamine-depended NAD+ synthetase encoded by the genome of H. seropedicae. One of these enzymes, namely HsNadE1Gln, has been characterized previously, and its oligomerization and biochemical properties resemble the characteristics of MtNadE1Gln (LASKOSKI et al., 2016). The second glutamine-dependent NadE from H. seropedicae (Uniprot D81X05), namely HsNadE2Gln, has the two domains typically present in other types of NadEGln according to Pfam: a N-terminal glutaminase domain (CN hydrolase, amino acids 3–252) and a C-terminal NAD synthase domain (amino acids 281–542). Sequence alignments with other characterized NadEGln showed that HsNadE2Gln carries the conserved glutamine catalytic triad EKC in the N-terminal region and the conserved residues involved in NaAD and ATP binding in the C-terminal region (Fig. 1).

Figure 1: Sequence alignment of different NadEGln. The sequences of M. tuberculosis, H. seropedicae NadE1Gln and NadE2Gln, A. brasilense NadE1Gln and NadE2Gln, and B. thailandensis were aligned using Clustal W. The threshold for shading was set as 48

60%. The conserved catalytic triad EKC present in the glutaminase domain is indicated in red. Conserved residues involved in NaAD and ATP binding are indicated in yellow and orange, respectively. Residues forming the ammonia tunnel constriction are indicated in blue. The interdomain loop in MtNadE1Gln is underlined. Source: SANTOS et al.,(2018) JBC. The gene encoding HsNadE2Gln was cloned into pET28a, and the protein was expressed as a N-terminal His6 tag fusion protein using Escherichia coli BL21 (λDE3) as host. HsNadE2Gln was purified by Ni2+-affinity chromatography. Analytical gel filtration was performed, and HsNadE2Gln eluted as a single homogeneous peak with the elution volume corresponding to 85 kDa (Figs. S1 and S2). Because the calculated mass for each monomer is 61 kDa, the gel-filtration analysis supports that HsNadE2 is arranged in a different oligomeric state, contrasting with the octameric MtNadE1Glnand HsNadE1Gln. The activity of HsNadE2Gln was determined by coupling NAD+ formation with the activity of alcohol dehydrogenase and following the absorbance of NADH at 340 nm. The substrates NaAD and ATP were kept at saturating levels, and the experiments were performed with increasing concentrations of ammonium or glutamine to determine kinetic parameters for these two substrates. For both ammonium and glutamine, the initial velocity versus substrate concentration exhibited typical

Michaelis–Menten hyperbolic response (Fig. S2). The determined Km for ammonium and glutamine were in the same order of magnitude, being 240 and 130 μM, respectively (Table 1). This is in contrast MtNadE1Gln and HsNadE1Gln, which have a Km for glutamine at least 1 order of magnitude lower that for ammonium (Table 1). Gln The Km values of HsNadE2 are more related to the values obtained for a NadEGln characterized in the deep-branching Bacteria Thermotoga maritima (Table 1). Note that no information about the oligomerization state is available for TmNadEGln.

Comparison of the kcat/Km values, which are indicative of substrate preference, showed that HsNadE2Gln and TmNadEGln have no significant preference N donor substrates. On the other hand, HsNadE1Gln and MtNadE1Gln showed a clear preference to use glutamine as N donor (Table 1).

Table 1 Kinetic properties in different Gln-dependent NadE

Source: SANTOS et al.,(2018) JBC. (Ref. 13, 11, 9 and 12 are in order LARONDE-LEBLANC; RESTO; GERRATANA, 2009; LASKOSKI et al., 2016; RAFFAELLI et al., 1999b; RESTO; YAFFE; GERRATANA, 2009.).

The preference for the utilization of ammonium or glutamine was compared between HsNadE1Gln and HsNadE2Gln using LC/MS. For this analysis, unlabeled 15 15 + 14 + + glutamine is mixed with NH4Cl, and the ratio of N NAD and N NAD (NH4 /Gln) formed is determined using LC coupled to high resolution MS. The results confirmed that HsNadE1Gln has a stronger preference for glutamine than HsNadE2Gln (Table 2). When both ammonium and glutamine were available at equimolar concentration (2 mM), HsNadE2Gln was able to be incorporated 11.9-fold more ammonium in relation to glutamine than HsNadE1Gln (Table 2).

Table 2 Ammonium versus glutamine competition in H. seropedicae NadE1 and NadE2

Source: SANTOS et al.,(2018) JBC.

Characterization of the three NadE from A. brasilense

An inspection on the genome of the diazotrophic plant associative α- Proteobacterium A. brasilense Sp245 revealed the presence of three NadE-like genes: first a single-domain NadENH3 type (Uniprot G8ATC0), namely AbNadENH3, and the other two belonging to the NadEGln type, namely AbNadE1Gln (Uniprot G8AIW8) and AbNadE2Gln (Uniprot G8ATT3). The genes encoding these proteins were cloned and

50 expressed as recombinant proteins in E. coli BL21 (λDE3). The proteins were purified to homogeneity as judged by SDS-PAGE (Fig. S3). Analytical gel filtration revealed that all three NadE eluted as a homogeneous symmetric peak with elution volumes corresponding to 600, 103, and 72 kDa for AbNadE1Gln, AbNadE2Gln, and AbNadENH3, respectively (Fig. 2 and Fig. S4). The determined molecular weight fits with dimer oligomerization for AbNadE2Gln and AbNadENH3 (calculated masses are 120 kDa for dimer AbNadE2 and 74 kDa for AbNadE3). On the other hand, AbNadE1Gln is arranged as an octamer (calculated mass for octamer is 605 kDa); hence AbNadE1Gln resembles the oligomeric organization found in HsNadE1Gln and MtNadE1Gln (all octameric).

Figure 2: Gel filtration analysis of the different NadE. Gel filtration was performed on a Superose 6 column (GE Healthcare), which was calibrated with a range of molecular mass standards (Bio-Rad): point 1, thyroglobulin; point 2, bovine γ-globulin; point 3, chicken ovalbumin; and point 4, equine myoglobin. The arrows indicate the estimated molecular weight of each NadE protein. Source: SANTOS et al.,(2018) JBC. The AbNadENH3 belongs to the NadENH3 type and showed a dimeric organization. As expected, AbNadENH3 was not able to use glutamine as N donor for NAD+ production (data not shown). On the other hand, the identification of a dimeric NadEGln also in A. brasilense supports the presence of a novel subgroup of NadEGln, which is different from the archetypical octameric NadEGln described in M. tuberculosis and eukaryotes such as Saccharomyces cerevisiae. 51

The kinetic characterization of the two A. brasilense NadEGln revealed the same trend as observed for the H. seropedicae enzymes. The dimeric AbNadE2Gln showed a Km for ammonium and glutamine in the same order of magnitude being 1,850 and 820 μM, respectively (Table 1). On the other hand, the octameric AbNadE1Gln had a Km for glutamine more than 2 orders of magnitude lower that for ammonium (Table Gln 1). The kcat/Km values clearly show that AbNadE1 has a strong preference for glutamine as N donor, whereas AbNadE2Gln can use both substrates with nearly similar preference (Table 1).

Intracellular glutamine levels in A. brasilense

It is well-established that the intracellular glutamine levels in Proteobacteria correlates with the availability of ammonium in the culture media (HUERGO; DIXON, 2015; VAN HEESWIJK; WESTERHOFF; BOOGERD, 2013). Particularly, in some nitrogen fixing organisms, such as A. brasilense and H. seropedicae, nitrogen fixation only occurs when the intracellular glutamine levels are low (ARCONDÉGUY; JACK; MERRICK, 2001; DIXON; KAHN, 2004). We used untargeted LC/MS to determine the intracellular levels of glutamine in A. brasilense. Under nitrogen-fixing conditions, the glutamine levels ranged between

74 and 79 μM (Fig. 3A). After the addition of 200 μM of NH4Cl to the cells, the intracellular levels of glutamine augmented to 830 μM in just 30 s and remained at similar levels after 1 min (Fig. 3A). 2 min after the ammonium shock, the ammonium added is consumed by the cellular metabolism, and the intracellular glutamine levels declined to 200 μM, decreasing even further after 20 min (Fig. 3A). These fluctuations in intracellular glutamine were well-synchronized with the covalent modification of the nitrogenase component NifH and glutamine synthetase GS (Fig. 3B). Both NifH and GS are known to suffer reversible covalent modification that inactivates both enzymes in response to an ammonium shock (HUERGO et al., 2006b; HUERGO; CHANDRA; MERRICK, 2013b).

Figure 3: Intracellular levels of glutamine and NifH and GS modification after an ammonium shock in A. brasilense. A, average intracellular glutamine levels from triplicate experiments ±S.D. B, western blotting analysis showing of the reversible post-translational modification of NifH and GS. The upper bands represent the ADP-ribosylated NifH (NifH-ADPR) or adenylylated GS (GS-AMP).

Source: SANTOS et al.,(2018) JBC. Under nitrogen-fixing conditions, the intracellular levels of glutamine were nearly

Gln identical to the AbNadE1 Km for glutamine (׽80 μM) and ׽10-fold below the Gln AbNadE2 Km for glutamine (Table 1 and Fig. 3A). These data suggest that AbNadE1Gln is much more active than AbNadE2Gln under nitrogen-fixing conditions. In contrast, upon an ammonium shock, the intracellular levels of glutamine reached the

Gln ,AbNadE2 Km for glutamine (׽800 μM). Hence, upon an ammonium shock AbNadE2Gln should become more active, whereas AbNadE1Gln should be operating at Vmax (Fig. S2) The LC/MS metabolome data were submitted to XCMS software (TAUTENHAHN et al., 2012) to identify other potential metabolites that fluctuate in response to the ammonium shock. However, only glutamine showed statistically significant fluctuations (p < 0.05). Even though other metabolites such as NADH, NAD+, ATP, ADP, AMP, and 2-oxoglutarate could be detected in the analysis (as suggested by comparison of retention times and m/z to authenticated standards), their low signal and/or high variation among biological replicates did not allow their quantification. 53

Phylogeny analysis of dimeric NadEGln

The presence of other representatives of dimeric NadE2Gln in nature is supported by at least two structures deposited in the PDB. The NadE2Gln from the β- Proteobacterium B. thailandensis (PDB code 4F4H) and from the δ- Proteobacterium A. baumannii (PDB code 5KHA). A Neighbor-joining phylogenetic tree was constructed using the alignment of NadEGln sequences derived from organisms belonging to all domains of life (Fig. 4). The sequences were separated into three distinct groups; 1) the eukaryotic NadE1Gln; 2) a group containing the octameric NadE1Gln from M. tuberculosis, H. seropedicae and A. brasilense; and 3) a more diverse group comprising all the NadE2Gln described as dimeric. This last group includes the sequences of deep branching Bacteria (Thermotoga and Aquifex) and Archaea (Fig. 4).

Figure 4: Phylogenetic analysis of selected NadEGln. The sequence of NadEGln was retrieved from NCBI and aligned using Clustal W. The phylogentic three was constructed by Neighbor-joining using MEGA 7. All positions containing gaps and missing data were eliminated from the data set. Bootstrap values were adjusted to 1000 replicates. Three groups were observed: 1) eukaryotic representatives; 2) NadE1-like octameric NadEGln; and 3) NadE2-like dimeric NadEGln. The relevant proteins with experimentally determined quaternary structure are indicated by arrows. A. baumannii, WP_065718975.1; Akkermansia muciniphila ATCC BAA-835, ACD04457.1; Aquifex aeolicus, NP_213654.1; Arabidopsis thaliana, NP_175906.1; Ardenticatena maritima, KPL88222.1; A. brasilense, AIB10872.1; A. brasilense, AIB14429.1; Azotobacter vinelandii DJ, YP_002798395.1; B. thailandensis, AOJ56104.1; Chloroflexus aurantiacus, ABY34602.1; Clostridium cellulolyticum H10, YP_002505537.1; Danio rerio, NP_001092723.1; Dehalococcoides mccartyi 195, YP_181837.1; Drosophila melanogaster, NP_572913.1; Fervidobacterium nodosum Rt17-B1, YP_001410277.1; Gluconacetobacter diazotrophicus PA1 5, YP_001601200.1; Haloferax mediterranei, AHZ23047.1; H. seropedicae, AKN65438.1; H. seropedicae, AKN67808.1; Homo sapiens, NP_060631.2; Methanoregula formicica, 55

AGB01500.1; Methanosaeta thermophila PT, YP_843157.1; Methylobacterium extorquens CM4, YP_002423109.1; Mus musculus, NP_084497.1; Mycobacterium bovis AF2122/97, NP_856111.1; Ralstonia eutropha H16, YP_725265.1; S. cerevisiae S288C, NP_011941.1; Schizosaccharomyces pombe 972h-, NP_587771.1; Theionarchaea archaeon, KYK35595.1; Thermosipho africanus TCF52B, YP_002335497.1; T. maritima, AKE27162.1; Thermotoga neapolitana DSM 4359, YP_002534863.1; and M. tuberculosis, AMP30329.1. Source: SANTOS et al.,(2018) JBC.

Structural analysis of dimeric NadE2Gln

Given the availability of a 1.7 Å resolution structure of dimeric NadE2Gln from B. thailandensis (PDB code 4F4H) obtained by the Burkholderia structome project (BAUGH et al., 2013), we used this protein to analyze the structural features of the subgroup of dimeric NadE2Gln. Structural alignments with other available NadEGln structures revealed that B. thailandensis is highly related to the dimeric NadE2Gln from A. baumannii (data not shown) as anticipated by their degree of phylogenetic relationship (Fig. 4), hence supporting that B. thailandensis NadE2Gln may serve as a prototype for this group of enzymes. Alignments with NAD synthetase NadENH3 from Bacillus subtilis (SYMERSKY et al., 2002b) revealed structural conservation at the NAD synthetase domain structures (Fig. S5). Comparison between the dimeric B. thailandensis NadE2 and the octameric M. tuberculosis NadE1 (LARONDE-LEBLANC; RESTO; GERRATANA, 2009b) revealed that the individual domains are structurally very similar with conserved residues involved in substrate binding and domain folding (Fig. 1). However, they display a rearrangement of the relative domain positions within the protomers (Fig. 5). In B. thailandensis NadE2, the glutaminase and synthetase domains within the same polypeptide make extensive contacts and are located in the same plane (Fig. 5, A–C). On the other hand, in the case of M. tuberculosis NadE1, the glutaminase domain of one polypeptide extends to make contact with the NaAD synthetase domain of the next polypeptide in the octamer through a flexible loop comprising resides 313–331 (MtNadE1Gln numbering) (Fig. 5, A–C). This particular segment is shorter in the dimeric subgroup of proteins (Fig. 1). This flexible loop facilitates the extensive intersubunit contacts in the octameric structure of M. tuberculosis. The superimposed picture of the two structures (Fig. 5A) shows how the glutaminase domain of M. 56

.tuberculosis translates by ׽40 Å compared with the same domain in B thailandensis and makes most of the surface contact with the next chain, which suggests that the glutaminase and the NaAD synthetase function of the domain likely occur within domains of two different polypeptides. This arrangement indicates that the relative orientations between the N and G domains remain the same though through a domain-swap interaction in the octameric case (Fig. 5D). This is illustrated by the high structural similarity of the dimeric protein to the octameric when allowing for structural superpositions over different monomers. In this case the dimeric (PDB code 4F4H) structure can be superposed on the octameric (PDB code 3DLA) structure with an RMSD of 1.85 Å over 741 aligned residues (i.e. superposing the boxed domains in Fig. 5D).

Figure 5 Comparison between the structures of dimeric and octameric NadEGln. A, superimposition of the single chain of NadE2Gln from B. thailandensis (cyan) and NadE1Gln from M. tuberculosis (gray). The PDB structures 4FH4 and 3DLA were superimposed using COOT and visualized using PyMOL. Structures are colored by chain. B and C, comparison between the organization of dimeric NadEGln from B. thailandensis (B) and octameric NadEGln from M. tuberculosis (C). Only one dimer of the octameric M. tuberculosis NadE is shown without transparency for comparison with B. thailandensis. D, schematic organization of the NAD synthetase domain N and the glutaminase domain G in the dimeric and in the octameric conformation of NadEGln. Note that the relative orientations between the N and G domains remain the same though through a domain-swap interaction in the octameric case. The PDB structures 4FH4 and 3DLA were visualized using PyMOL. Source: SANTOS et al.,(2018) JBC.

Surface representation of B. thailandensis NadE2Gln showed that the conserved residues forming the glutaminase catalytic triad and the intersubunit NaAD–ATP bind sites are located in solvent-exposed pores (Fig. 6, A and B). A phosphate is present in 58 the NaAD binding site of B. thailandensis NadE (not shown) and makes similar contacts as observed for the NaAD phosphates in the M. tuberculosis NadE structure (residues Asn-471 and Lys-635, Mt numbering) (CHUENCHOR et al., 2012), hence suggesting similar substrate binding modes between octameric and dimeric NadEGln enzymes. Given that the NaAD- and glutamine-binding sites are separated 40–50 Å in space (within the same protomer or between the two different promoters), we hypothesized that B. thailandensis NadE2Gln should have an internal tunnel to translocate ammonia between the two catalytic domains as observed in MtNadE1Gln. We used CAVER to search for internal tunnels within the B. thailandensis NadE2Gln structure. This analysis identified two continuous tunnels that are potential routes for ammonia translocation. A 57.6 Å tunnel connects the glutaminase site from one promoter to the synthetase site in the other promoter (Fig. 6C). Another 58.3 Å tunnel connects the two catalytic domains within the same protomer (Fig. 6D). The interprotomer tunnel is U-shaped resembling the ammonia tunnel identified in the M. tuberculosis NadE, whereas the intrasubunit tunnel is Z- shaped (Fig. 6). Considering the protein symmetry, there are actually four possible pathways for ammonia translocation (G1 to S1, G1 to S2, G2 to S1, and G2 to S2); all these routes are interconnected by a major empty space located within the interprotomer/interdomain interface (Fig. 6).

Figure 6: Structure of the dimeric NadE2Gln from B. thailandensis. A and B, surface (A) and cartoon (B) representations; the monomers are colored green and cyan. The glutaminase catalytic triade residues are in red sticks. Residues involved in NaAD and ATP binding are shown as yellow and orange sticks, respectively. The dashed line indicates the separation of the NAD synthetase domain on the top and the glutaminase domain on the bottom. C and D, the ammonia tunnel connecting the glutaminase and NaAD synthetase domain in different protomers (C) and within the same protomer (D). The detailed structure of the intersubunit ammonia tunnel of the dimeric NadE2Gln from B. thailandensis is shown. E, residues forming the intersubunit ammonia tunnel from different promoters are indicated as green and cyan sticks, respectively. F, conserved residues forming the major constriction within the ammonia tunnel are show in cyan, and the glutaminase catalytic residues are in red. The figures were generated using PyMOL and the PDB entry 4F4H. The ammonia tunnel was identified by CAVER.

Source: SANTOS et al.,(2018) JBC. A major bottleneck constriction (0.94 Å) in the ammonia tunnel is formed just after the glutaminase catalytic triad (Fig. 6, E and F). Given that the ammonia van der Waals radius is ׽1.6 Å (FLOQUET et al., 2007), it is likely that BtNadE2Gln must go through conformational changes to allow ammonia transfer between the two catalytic 60 domains. The bottleneck constriction is mostly hydrophobic and formed by residues that are conserved in both octameric and dimeric forms of NadEGln (Figs. 1 and 6). The hydrophobic nature of the constriction suggests that ammonia rather than ionic ammonium may be the translocated species. A hydrophobic ammonia bottleneck or gate is also present in other ammonia-translocating proteins such as AmtB (ammonia channel) and GlmS (glucosamine-6-phosphate synthase) (FLOQUET et al., 2007; JAVELLE et al., 2008). In AmtB, conserved amino acid residues help to deprotonate ammonium before translocation (JAVELLE et al., 2008), an important step to drive the + equilibrium toward NH3 formation under physiological pH considering the NH4 pKa of 9.2. More studies will be necessary to determine the nature of the translocated species in NadE2Gln.

Discussion

In a previous study we reported the characterization of a NadE1Gln from H. seropedicae, HsNadE1Gln. We found that HsNadE1Gln oligomerization and biochemical properties resemble the characteristics reported for the intensively studied MtNadE1Gln and the S. cerevisiae NadE1Gln. Here we report the characterization of a second type of NadE2Gln encoded by the genome of H. seropedicae, HsNadE2Gln. Surprisingly, we found that HsNadE2Gln is arranged in a completely novel oligomeric state (Fig. 2). Even though the experimentally determined molecular mass for HsNadE2Gln was 85 kDa (which is between a monomer and dimer), it is more likely that HsNadE2Gln is arranged as a dimer based on the phylogenetic relationship with other dimeric NadE2Gln synthetases (Fig. 4). Additional studies in A. brasilense revealed that this organism also encodes a dimeric form of NadE2Gln, AbNadE2Gln (Fig. 2). Further inspection in the PDB revealed the presence of dimeric NadE2Gln in A. baumannii and B. thailandensis. These data suggest the presence of a novel group of dimeric NadE2Gln. Phylogenetic analysis of NadEGln sequences from various organisms with diverse lifestyles support the presence two distinct groups of NadEGln in prokaryotes (Fig. 4). It is tempting to speculate that the group containing MtNadE1Gln, AbNadE1Gln, and HsNadE1Gln (all octameric), is likely to be formed mostly by octameric types of NadE1Gln, whereas the group containing HsNadE2Gln, AbNadE2Gln, A. baumannii NadE2Gln, and B. thailandensis NadE2Gln (all dimeric) should be mostly composed of dimeric forms of NadE2Gln. 61

Structural analysis of B. thailandensis dimeric NadE2Gln showed that this enzyme shares some features with octameric MtNadE1Gln: the residues involved in substrate binding and catalysis and the residues forming a long intersubunit tunnel that are likely to carry ammonia from the glutaminase to the synthetase domain (Figs. 1 and 6). However, although BtNadE2Gln has an intensive intrachain glutaminase- synthetase domain contact, such contacts are virtually absent in the MtNadE1Gln enzyme (Fig. 5). The octameric NadE1Gln-like forms presumably evolved to make much more intersubunit contacts, a property that may explain the strong domain synergism detected in MtNadE1Gln (LARONDE-LEBLANC; RESTO; GERRATANA, 2009b) and the little efficiency of NadE1Gln-like in using ammonium as N source (Tables 1 and 2). The sequences of deep branching Bacteria (Thermotoga and Aquifex) and Archaea cluster in the NadE2Gln-like dimeric group (Fig. 4). A previous phylogenetic analysis suggested that a single-domain NadENH3 was present in last universal common ancestor and the two-domain NadEGln evolved via recruitment and fusion of the N-terminal glutaminase domain (DE INGENIIS et al., 2012). If this assumption is correct, we speculate that the dimeric NadE2Gln could be an intermediate evolutionary link between the single-domain dimeric NadENH3 and two domain octameric NadE1Gln. Indeed, the enzymatic characteristics of the NadE2Gln enzymes (Tables 1 and 2) show intermediate properties between dimeric NadENH3 and octameric NadE1Gln (i.e. NadE2Gln can use both ammonium and glutamine with similar efficiencies). The biogeochemical history of earth support that although ammonium was readily available to the ancient life forms (that could support NadENH3 activity), ammonium is very limited in most habitats nowadays, and this could explain the need to recruit a glutaminase domain to support NadE function during evolution. More than one copy of NadE has been detected in the genome of a range of prokaryotes (see Table S3 in DE INGENIIS et al.(2012)). One intriguing question remaining to be answered is why some organisms encode multiple isoforms of NadE. Different expression in response to nitrogen availability would be one possible explanation; however, at least in the case of A. brasilense, proteomic analysis revealed that AbNadE1Gln, AbNadE2Gln, and AbNadENH3 levels did not change in response to ammonium availability (Kukolj C. and Souza E.M., personal communication). Hence, the different kinetic properties and/or substrate preferences may constitute the only

62 selective pressure to maintain three NadE isoforms within the A. brasilense genome. Our data allowed us to propose a model for the role of these isoforms in A. brasilense. ,(Under nitrogen-fixing conditions, the intracellular levels of glutamine are low (׽80 μM Gln and NadE1 operates at Vmax/2 and thereby prompts responses to small fluctuations in glutamine availability within this range accordingly to the NadE1 Michaelis–Menten + curve (Fig. 7, A and D). Upon an ammonium shock of NH4 200 μM, the intracellular

Gln glutamine raises to ׽800 μM, NadE1 operates at Vmax, whereas NadE2 operates at Vmax/2 and thereby prompts responses to small fluctuations in glutamine availability within this range (Fig. 7, B and D). Hence, the presence of two glutamine-dependent NadEs would allow better adjustment of NAD+ production over a broad range of glutamine concentrations (Fig. 7).

Figure 7: Model for the role of the three NadE in A. brasilense. A, under nitrogen-fixing conditions nitrogenase feeds ammonium into GS, and intracellular levels of Gln glutamine are low (~80 μM). NadE1 operates at Vmax/2 and promptly responds to small fluctuations in 63 glutamine availability within this range (see the blue bar with Michaelis–Menten curves in D). NadE2Gln and NadENH3 activities are negligible leading to slow production of NAD+. B, upon an ammonium shock + of NH4 200 μM, the intracellular glutamine rises to ~800 μM, leading to total inhibition of nitrogenase Gln by ADP-ribosylation and nearly full inhibition of GS by adenylylation. NadE1 operates at Vmax, whereas Gln NadE2 operates at Vmax/2 and promptly responds to small fluctuations in glutamine availability within this range (see the green bar with Michaelis–Menten curves in D). NAD+ production is elevated. C, when external levels of ammonium are high, the high intracellular glutamine will lead to NadE1Gln and NadE2Gln to operate at Vmax using glutamine as substrate. Furthermore, because GS is fully inactive due adenylylation, free intracellular ammonium could be directly assimilated into NAD+ by NadE2Gln and presumably also by NadENH3 bypassing GS activity. The concert action of all three isoforms of NadE is likely to further enhance NAD+ production. D, velocities of the different A. brasilense NadE isoforms accordingly to the intracellular glutamine levels. Source: SANTOS et al.,(2018) JBC. Acknowledgements

This work was supported by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação Araucária, Coordenação de Aperfeiçoamento de Pessoal de Ensino Superior (CAPES), and CNPq-Instituto Nacional de Ciência e Tecnologia (INCT). The authors declare that they have no conflicts of interest with the contents of this article.

References

ARAVIND, L. et al. The natural history of ADP-ribosyltransferases and the ADP- ribosylation system. Current topics in microbiology and immunology, v. 384, p. 3– 32, 2015. ARCONDÉGUY, T.; JACK, R.; MERRICK, M. P(II) signal transduction proteins, pivotal players in microbial nitrogen control. Microbiology and molecular biology reviews : MMBR, v. 65, n. 1, p. 80–105, mar. 2001. BAUGH, L. et al. Combining Functional and Structural Genomics to Sample the Essential Burkholderia Structome. PLoS ONE, v. 8, n. 1, 2013. BELLINZONI, M. et al. Heterologous expression, purification, and enzymatic activity of Mycobacterium tuberculosis NAD(+) synthetase. Protein expression and purification, v. 25, n. 3, p. 547–57, ago. 2002. BENNETT, B. D. et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nature Chemical Biology, v. 5, n. 8, p. 593–599, 2009. BHEDA, P. et al. The Substrate Specificity of Sirtuins. Annual Review of Biochemistry, v. 85, n. 1, p. 405–429, 2016. BRENNER, C. Catalysis in the nitrilase superfamily. Current opinion in structural biology, v. 12, n. 6, p. 775–82, dez. 2002. CHUENCHOR, W. et al. Regulation of the intersubunit ammonia tunnel in Mycobacterium tuberculosis glutamine-dependent NAD+ synthetase. The Biochemical journal, v. 443, n. 2, p. 417–26, 15 abr. 2012.

DE INGENIIS, J. et al. Glutamine versus ammonia utilization in the NAD synthetase family. PloS one, v. 7, n. 6, p. e39115, jan. 2012. DIXON, R.; KAHN, D. Genetic regulation of biological nitrogen fixation. Nature reviews. Microbiology, v. 2, n. 8, p. 621–31, ago. 2004. FLOQUET, N. et al. Ammonia channeling in bacterial glucosamine-6-phosphate synthase (Glms): Molecular dynamics simulations and kinetic studies of protein mutants. FEBS Letters, v. 581, n. 16, p. 2981–2987, 26 jun. 2007. GAZZANIGA, F. et al. Microbial NAD metabolism: lessons from comparative genomics. Microbiology and molecular biology reviews : MMBR, v. 73, n. 3, p. 529–41, Table of Contents, set. 2009. HUERGO, L. F. et al. ADP-ribosylation of dinitrogenase reductase in Azospirillum brasilense is regulated by AmtB-dependent membrane sequestration of DraG. Molecular microbiology, v. 59, n. 1, p. 326–37, jan. 2006. HUERGO, L. F.; CHANDRA, G.; MERRICK, M. P(II) signal transduction proteins: nitrogen regulation and beyond. FEMS microbiology reviews, v. 37, n. 2, p. 251–83, mar. 2013. HUERGO, L. F.; DIXON, R. The Emergence of 2-Oxoglutarate as a Master Regulator Metabolite. Microbiology and Molecular Biology Reviews, v. 79, n. 4, p. 419–435, dez. 2015. JAUCH, R. et al. Structures of Escherichia coli NAD synthetase with substrates and products reveal mechanistic rearrangements. The Journal of biological chemistry, v. 280, n. 15, p. 15131–40, 15 abr. 2005. JAVELLE, A. et al. Substrate binding, deprotonation, and selectivity at the periplasmic entrance of the Escherichia coli ammonia channel AmtB. Proceedings of the National Academy of Sciences of the United States of America, v. 105, n. 13, p. 5040–5045, 1 abr. 2008. LARONDE-LEBLANC, N.; RESTO, M.; GERRATANA, B. Regulation of active site coupling in glutamine-dependent NAD + synthetase. Nature Structural and Molecular Biology, 2009. LASKOSKI, K. et al. In vitro characterization of the NAD+ synthetase NadE1 from Herbaspirillum seropedicae. Archives of Microbiology, 2016. LI, J. et al. A conserved NAD+ binding pocket that regulates protein-protein interactions during aging. Science (New York, N.Y.), v. 355, n. 6331, p. 1312–1317, 24 mar. 2017. MOURE, V. R. et al. Regulation of nitrogenase by reversible mono-ADP-ribosylation. In: Current Topics in Microbiology and Immunology. [s.l: s.n.]. RESTO, M.; YAFFE, J.; GERRATANA, B. An ancestral glutamine-dependent NAD(+) synthetase revealed by poor kinetic synergism. Biochimica et biophysica acta, v. 1794, n. 11, p. 1648–53, nov. 2009. RIZZI, M. et al. Crystal structure of NH3-dependent NAD+ synthetase from Bacillus subtilis. The EMBO journal, v. 15, n. 19, p. 5125–5134, 1996. RODRIGUES, T. E. et al. Search for novel targets of the PII signal transduction protein 65 in Bacteria identifies the BCCP component of acetyl-CoA carboxylase as a PII binding partner. Molecular microbiology, v. 91, n. 4, p. 751–61, mar. 2014. SYMERSKY, J. et al. NH3-dependent NAD+ synthetase from Bacillus subtilis at 1 Å resolution. Acta Crystallographica Section D: Biological Crystallography, v. 58, n. 7, p. 1138–1146, 1 jul. 2002. TAUTENHAHN, R. et al. XCMS online: A web-based platform to process untargeted metabolomic data. Analytical Chemistry, v. 84, n. 11, p. 5035–5039, 5 jun. 2012. VAN HEESWIJK, W. C.; WESTERHOFF, H. V; BOOGERD, F. C. Nitrogen assimilation in Escherichia coli: putting molecular data into a systems perspective. Microbiology and molecular biology reviews : MMBR, v. 77, n. 4, p. 628–95, 1 dez. 2013.

Supporting information

Table S1 Primers used in this work Uniprot Organism NCBI ID Gene Primers 5'-3' ID F GGATGCCATATGCTTCGCATCGCCATCG WP_041812749 nadE1 G8ASI0 H. R AACAGGGATCCATGGCAATCGGGTCC seropedicae F GGATGCCATATGCTTCGCATCGCCATCG YP_003777950.1 nadE2 D8IX05 R AACAGGGATCCATGGCAATCGGGTCC F GGATCCATATGACGGTTGACGGCG WP_041812749 nadE1 G8ASI0 R GGGCGGATCCTTACTCCTTGGGCAC A. F ATGCCCATATGACCGACCGCCTTTCC WP_014239482 nadE2 G8AIW8 brasilense R GGGCGGATCCTCACGCAATGCTCCCG F CCGACACCATATGCTGAACGCCCTG YP_004985433.1 nadE3 G8ATC0 R CATTGGATCCGTCATGCGTCGCTCC

Source: SANTOS et al.,(2018) JBC.

Figure S1: Gel filtration profile of HsNadE2. A) Elution profile of HsNadE2 on a Superose 6 column, the elution volume corresponds to 85 kDa. B) SDS-PAGE analysis of gel filtration chromatography fractions.

Source: SANTOS et al.,(2018) JBC.

Figure S2: Determination of kinetic parameters for NAD synthetases from A. brasilense and H. seropedicae. Initial velocity (Vo) for the substrates ammonium (A, C and E) and glutamine (B, D and F). Average data from triplicate experiments ± SD.

Source: SANTOS et al.,(2018) JBC.

Figure S3: SDS-PAGE analysis of A. brasilense NadE1Gln, NadE2Gln and NadENH3. MW molecular markers. His-tag proteins: AbNadE1Gln-His (Lane 1), AbNadE2Gln-His (Lane 3), AbNadE3NH3-His (Lane 5). Tag less proteins used in this work: AbNadE1Gln (Lane 2), AbNadE2Gln (Lane 4), and AbNadE3NH3 (Lane 6).). The gel was Coomassie blue stained. The line indicates the splicing in the image of the same gel.

Source: SANTOS et al.,(2018) JBC.

Figure S4: Gel filtration profile of AbNadE2. A) Elution profile of AbNadE2 on a Superose 6 column, the elution volume corresponds to 103 kDa. B) SDS-PAGE analysis of gel filtration chromatography fractions.

Source: SANTOS et al.,(2018) JBC.

Figure S5: Superposition of the NAD+ synthetase domain of NadE2Gln from Burkholderia thailandensis (red) and NadENH3 from Bacillus subtilis (green). View from the top of the NAD+ synthetase domain. The PDB structures 4FH4 (amino acids 281 to 542) and 1KQP (full length) were aligned and visualized using PyMOL.

Source: SANTOS et al.,(2018) JBC.

Front page of the published article in JBC (2018)

3. CHAPTER II: NAD+ biosynthesis in Bacteria is subject to global Carbon/Nitrogen control via PII signaling

This research was originally published in the Journal of Biological Chemistry. Adrian Richard Schenberger Santos, Edileusa Cristina Marques Gerhardt, Erick Parize, Fabio Oliveira Pedrosa, Maria Berenice Reynaud Steffens, Leda Satie Chubatsu, Emanuel Maltempi Souza, Luciane Maria Pereira Passaglia, Fernando Hayashi Sant'Anna, Gustavo Antônio de Souza, Luciano Fernandes Huergo and Karl Forchhammer. NAD+ biosynthesis in Bacteria is subject to global Carbon/Nitrogen control via PII signaling. J Biol Chem. 2020; Vol 295:6165-6176. © Abstract NAD+ is a central metabolite participating in core metabolic redox reactions. The prokaryotic NAD synthetase enzyme, NadE, catalyzes the last step of the NAD+ biosynthesis, converting nicotinic acid adenine dinucleotide (NaAD) into NAD+. Some members of the NadE family use L-glutamine as a nitrogen donor and are named NadEGln. Previous gene neighborhood analysis has indicated that bacterial nadE gene is frequently clustered with the gene encoding the regulatory signal transduction protein PII, suggesting a functional relationship between these proteins in response to the nutritional status and the carbon/nitrogen ratio of the bacterial cell. Here, using affinity chromatography, bioinformatic analysis, NAD synthetase activity, and biolayer interferometry assays, we show that PII and NadEGln physically interact in vitro, and that this complex relieves NadEGln negative feedback inhibition by NAD+. This mechanism is conserved in distant related Bacteria. Of note, the PII protein allosteric effector and cellular nitrogen level indicator 2-oxoglurate (2-OG) inhibited the formation of the PII-NadEGln complex within a physiological range. These results indicate an interplay between the levels of ATP, ADP, 2-OG sensed by PII with glutamine, and NAD+, representing a metabolic hub that may balance the levels of core nitrogen and carbon metabolites. Our findings support the notion that PII proteins act as a dissociable regulatory subunit of NadEGln enabling the control of NAD+ biosynthesis accordingly to the cell nutritional status of the bacterial cell.

Introduction NAD+ is a crucial metabolite participating as a cofactor in dozens of core metabolic redox reactions. Furthermore, NAD+ is used as substrate for lysine deacetylases, ADP-ribosylating enzymes, DNA ligase, and for NADP+ biosynthesis.

Increasing evidence supports that NAD+ also acts as an important signaling metabolite and that NAD+ is used for 5´-modified mRNA forming a cap-like structure in bacteria (HOUTKOOPER et al., 2010; JÄSCHKE et al., 2016; SORCI; RUGGIERI; RAFFAELLI, 2014). In prokaryotes, NAD+ biosynthesis can occur by more than one route, including de novo, using L-aspartate as precursor, and salvage pathways. In most cases, these routes converge before the final reaction, catalyzed by NAD synthetase enzymes (NadE) in which nicotinic acid adenine dinucleotide (NaAD) is amidated to produce NAD+ (KURNASOV et al., 2003). The NadE enzymes use ATP as a substrate to produce an NaAD-AMP intermediate, thereby activating the carboxylate of nicotinic acid to react with ammonia, releasing NAD+, AMP, and PPi as products (Fig. 1).

Figure 1: Reaction scheme of NAD synthetases. Last step of the NAD+ biosynthesis catalyzed by NadE using L-glutamine or free ammonia as N donor for the amidation of the precursor nicotinic acid adenine dinucleotide (NaAD). The hydrolysis of L- glutamine occurs in the glutaminase domain (in cyan). The amidation of NaAD is performed by the synthetase domain (in purple).

Source: SANTOS et al.,(2020) JBC.

There are two different classes of NadE enzymes identified to date. A shorter version of the enzymes carries only the NAD synthetase domain and uses ammonia directly as N-donor. These enzymes are known as ammonia dependent or NadENH3. Longer versions of NadE enzymes carry an extra N-terminal glutaminase domain and use glutamine as N-donor to deliver ammonia to the synthetase domain through an ammonia tunnel (DE INGENIIS et al., 2012; LARONDE-LEBLANC; RESTO; GERRATANA, 2009b) (Fig. 1). These enzymes are known as glutamine dependent or NadEGln. Regarding the quaternary structure arrangement, NadENH3 enzymes are found as dimers, while NadEGln exists in two different subtypes to date: type 1 are 75 octamers, and type 2 are dimers. Kinetic and structural analysis suggests that the dimeric type 2 NadEGln is an evolutionary intermediate between dimeric NadENH3 and octameric, or type 1 NadEGln (SANTOS et al., 2018). There is very little knowledge regarding the regulation of NAD+ biosynthesis pathways in bacteria. In Salmonella enterica, the expression of the enzymes involved in de novo NAD+ biosynthesis is regulated by NadR, which is an NAD+ sensor repressing the biosynthesis of NAD+ when the levels of this metabolite are high (GROSE; BERGTHORSSON; ROTH, 2005). There are other transcriptional regulators of NAD+ biosynthesis described in bacteria. However, their sensory properties and mode of action have not been studied in detail (RODIONOV et al., 2008c). The conservation of gene order can be used as a fingerprint of proteins that physically interact (DANDEKAR et al., 1998). In prokaryotes, the gene encoding NadEGln is frequently co-localized with the gene encoding the regulatory signal transduction protein PII (usually termed glnB) (SANT’ANNA et al., 2009). Conservation of the glnB-nadE gene pair leads to the hypothesis that NadEGln and PII could physically interact, and PII proteins could somehow affect the NadEGln function (HUERGO; CHANDRA; MERRICK, 2013a). The PII proteins are widespread signal transduction proteins present in a broad range of prokaryotes and in the chloroplast of eukaryotic phototrophs (HUERGO; CHANDRA; MERRICK, 2013a). In addition to the glnB gene, some organisms can encode additional PII gene paralogues. In proteobacteria, the second PII gene is named glnK and is encoded in an operon along with the amtB gene (ARCONDÉGUY; JACK; MERRICK, 2001). In the special case of Azospirillum brasilense, the glnK-like gene is not co-transcribed with amtB and is named glnZ (SANT’ANNA et al., 2009). The PII protein structure is highly conserved and forms a compact homotrimeric barrel with an extraordinary ability to sense and integrate the levels of key metabolites such as ATP, ADP, 2-oxoglutarate (2-OG) and in certain cases L-glutamine (FORCHHAMMER; LÜDDECKE, 2016b). These metabolites represent critical signals of the nutritional status of the cell as they reflect the availability of energy (ATP/ADP ratio), nitrogen (glutamine acts as a signal of nitrogen availability), and the carbon/nitrogen ratio (2-OG acts as a signal of the carbon/nitrogen balance) (HUERGO; DIXON, 2015).

The nucleotides ATP and ADP bind competitively to the three nucleotide-binding sites located in the clefts formed between each PII subunit (JIANG; NINFA, 2007). The three 2-OG binding sites in the PII trimer are formed only when PII is pre-occupied with MgATP. This cooperative binding of MgATP and 2-OG results in enhanced ATP affinity in the presence of 2-OG (TRUAN et al., 2010). Thus, in a competition between ATP and ADP, the presence of high levels of 2-OG favors the ATP binding to PII (FOKINA et al., 2010; OLIVEIRA et al., 2015). The interplay between the allosteric effectors 2- OG, ATP and ADP is a conserved feature of the PII protein family (HUERGO; CHANDRA; MERRICK, 2013a). In response to varying levels of these allosteric effectors, PII may exist theoretically in up to 21 different structural conformations (DA ROCHA et al., 2013). Although not all of them may play a physiologic role, structural changes indeed affect the ability of PII to interact and regulate essential metabolic proteins, thereby pacing the overall cellular metabolism accordingly to nutrient availability (HUERGO; DIXON, 2015). The PII proteins also respond to the levels of glutamine. However, the mechanism of regulation by glutamine is not universal. In Proteobacteria, PII proteins are subject to a cycle of reversible uridylylation of a Tyr residue located at the apex of a solvent-exposed loop, namely T-loop. This response is mediated by the glutamine- sensitive bifunctional uridylyl-transferase/removing enzyme, GlnD (MERRICK, 2015). On the other hand, plants and eukaryotic algae evolved PII proteins that are regulated by glutamine due to direct allosteric binding (CHELLAMUTHU et al., 2014). Despite the mechanism used for glutamine sensing, glutamine levels affect the PII protein function to coordinate cellular metabolism accordingly to the availability of nitrogen. Here we show that type 2 NadEGln enzymes are negatively feedback-inhibited by physiological levels of NAD+, and this regulatory mechanism is conserved in distantly related bacteria. We show that PII proteins act as a dissociable regulatory subunit of dimeric NadE2Gln in bacteria. Complex formation between PII and NadE2Gln relieves the NadE2Gln inhibition by NAD+, thereby acting as a switch to coordinate NAD+ production with nutrient availability in prokaryotes.

Methods

Ligand fishing His-GlnZ affinity chromatography

To prepare the A. brasilense protein extract, 400 mL of the A. brasilense 2812 strain was cultivated on NFbHP to an A600nm of 2. Cells were collected by centrifugation, resuspended in 20 mL of buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl, 20 mM

Imidazole, 5 mM of MgCl2, 1 mM of ATP and disrupted by sonication. The soluble fraction (after centrifugation at 30,000 xg, 4°C for 30 minutes) was divided into two aliquots, one was the control fraction and the other the GlnZ affinity ligand fishing. For the preparation of the His-GlnZ extract, 300 mL of E. coli BL21(DE3) carrying the pMAS3 plasmid that expresses the A. brasilense His-GlnZ protein was cultured in LB medium at 37°C until A600nm of 0.5. Then, 0.3 mM of IPTG was added and after 4 hours, cells were harvested by centrifugation and resuspended in buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl and 20 mM Imidazole. Cells were disrupted by sonication and the soluble fraction obtained after centrifugation (30,000 xg at 4°C for 30 minutes). For interaction assay, two Hi-trap chelating columns (1 mL) (GE healthcare) charged with Ni2+ were used. The first one was equilibrated with buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl and 20 mM Imidazole and loaded with His-GlnZ extract, washed with 20 mL of buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl and 50 mM of imidazole to remove unspecific bound proteins and equilibrated with 5 mL of buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl, 50 mM Imidazole, 5 mM of MgCl2, 1 mM of ATP. This column is prepared to bind the putative targets of GlnZ. The second column was equilibrated with 5 mL of buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl, 5 mM of MgCl2, 1 mM of ATP and 50 mM of imidazole. This column is a control for the unspecifically binding of proteins to the column resin, thus, providing a control fraction of proteins eluted in the ligand fishing assay but not bound to GlnZ. Both columns were loaded with A. brasilense protein extract and washed with 20 mL of buffer containing 50 mM Tris-HCl pH 8, 100 mM KCl, 5 mM of MgCl2, 1 mM of ATP and 50 mM of imidazole. Proteins bound to both columns were eluted with 3 mL of buffer containing 1.5 mM of 2-OG and 50 mM Tris-HCl pH 8, 100 mM KCl, 5 mM of MgCl2, 1 mM of ATP, 50 mM of imidazole. Two fractions of 1.5 mL (namely Fraction 1 and 2) were collected from the control and His-GlnZ columns and analyzed by label free quantitative LC- MS/MS as described previously (GRAVINA et al., 2018). Proteins were identified using an Uniprot A. brasilense database from June 2012 (8122 sequence entries). The log

78 enrichment ratio of each protein in the His-GlnZ column/control column in each fraction was plotted using GraphPad Prism 7 (Fig. S1).

Construction of the plasmid expressing NAD synthetase from Synechocystis sp,

GlnD and GlnZ∆Tloop from A. brasilense.

The sequence of the slr1691 gene from Synechocystis sp. strain PCC 6803, encoding dimeric NadE2Gln, was retrieved from CyanoBase (GERDES et al., 2006). Amplification of the gene by PCR was performed using Synechocystis genomic DNA as a template with the forward primer 5’- ATCACCATCACCATCACGATTACGATATCCCAACTAGTGAAAACCTGTATTTTCA GGGCGCTAGCCATATGTTTACCATTGCCCTTGCCCAGCTTAATC and the reverse primer 5’ - CAGCAAAAAACCCCTCAAGACCCGTTTAGAGGCCCCAAGGGGTTATGCTAGTTA T TGCTCAGCGGCCGCGGATCTTAACTGCCTTGGGGATGGAAAG. The glnD sequence of from A. brasilense was retrieved from NCBI database. The amplification by PCR was performed using A. brasilense genomic DNA as a template with the forward primer 5´- CTAGTGAAAACCTGTATTTTCAGGGCGCTAGCCATATGCTCTCCACCCGCGCCG CCTCCGCCGACGCGTCCGACGCCAAGGACGCCGGCACAGCCAACATCCCCAA CAAG and the reverse primer 5’ – CCAAGGGGTTATGCTAGTTATTGCTCAGCGGCCGCGGATCCCTCATGCGGACG GATCGGCGAGCGCGTGCAGCAGCCGCTCGCGGATCTGGGCCAGCTTGTTC. Both amplified fragments were Gibson assembled into the pTEV5 vector previously cut with NdeI and BamHI (NEB). The gene glnZ∆Tloop was obtained from pMSA4∆42-54 (RAJENDRAN et al., 2011) and subcloned into pET28a. The pMSA4∆42-54 plasmid was cut with NdeI and BamHI (NEB). The resulting fragments were separated by 1% agarose gel electrophoresis and the fragment extracted using Monarch DNA Gel Extraction Kit. Gene fragment was then cloned into pET28a using NdeI/BamHI sites and transformed into DH10B cells. The kanamycin-resistant selected clone was checked by restriction pattern and it was named pGlnZ∆42-54. The obtained plasmids (pASnadESc, pASGlnDAb and pGlnZ∆42-54) were sequenced to confirm the integrity of the inserted genes (Table S2) (ARAUJO et al.,

2004; HUERGO et al., 2006a, 2006c; MOURE et al., 2012a; PEDROSA; YATES, 1984; RIPPKA; DERUELLES; WATERBURY, 1979; ROCCO et al., 2008).

Protein expression and purification

Recombinant protein expression was performed in E. coli Lemo21 (DE3) cells (New England Biolabs) in LB medium. The TEV protease was overexpressed in E. coli Rosetta (DE) pLysS (Novagen). Typically, E. coli strains carrying the expression plasmids (Table S2) (ARAUJO et al., 2004; HUERGO et al., 2006a, 2006c; MOURE et al., 2012a; PEDROSA; YATES, 1984; RIPPKA; DERUELLES; WATERBURY, 1979;

ROCCO et al., 2008) were grown to an A600nm of 0.7 at 37ºC. Protein expression was induced by adding 0.5 mM isopropyl β-D-thiogalactopyranoside and the cell culture was incubated for 12 h at 120 rpm at 20ºC. For purification of His-tagged PII proteins (His-GlnZ, His-AbGlnB, His-GlnZ ΔTloop (ΔQ42-S54), His-HsGlnK or His-ScPII), cell pellets were resuspended in sonication buffer (50 mM Tris-HCl pH 8.0, 100 mM KCl, 20 mM Imidazol) and cells were disrupted by sonication (two times for 4 minutes, output 5 at 50% duty cycle in a Bronson sonifier), followed by centrifugation at 30,000 x g for 30 minutes to remove cell debris and insoluble material. The soluble fraction (approximately 40 mL) was loaded onto a 1 mL HisTrap HP Ni-NTA column (GE Healthcare). The column was washed with 10 mL (50 mM Tris-HCl pH 8.0, 100 mM KCl, 50 mM Imidazol) and proteins were eluted using a linear gradient of imidazole (100 to 500 mM) in the same buffer. Eluted fractions were analyzed by 15% SDS-PAGE and the fractions containing the protein of interest were pooled and dialyzed in storage buffer (50 mM Tris-HCl pH 8.0, 100 mM KCl, 10% glycerol). For purification of His-tagged NAD synthetase and GlnD proteins (HsNadE2, ScNadE2 and AbGlnD) the same procedure was applied with the exception that the buffers contained 100 mM of NaCl instead of KCl. To obtain untagged HsNadE2, the purified His-tagged protein was submitted to a thrombin cleavage using Thrombin CleanCleave Kit accordingly to manufacturer’s instructions (Sigma-Aldrich). TEV protease was purified as described previously (VAN DEN BERG et al., 2006). The His-tagged ScNadE2 had its N-Terminal tag removed with the TEV protease, as described previously for other NAD synthetases (SANTOS et al., 2018). Native AbNadE2, GlnB and GlnZ from A brasilense were purified as described previously (LASKOSKI et al., 2016; SANTOS et al., 2018). The GlnZ-UMP3

80 protein was obtained by in vitro uridylylation of GlnZ using purified A. brasilense GlnD as described (ARAÚJO et al., 2008). All the proteins were stored in small aliquots at - 80°C.

NAD synthetase activity assays

Reactions were performed at 30oC in 50 mM Tris-HCl pH 8, 50 mM KCl, 10 mM Gln MgCl2 and indicated substrate concentrations. NadE activity was determined by coupling the production of NAD+ to the NADH-forming oxidation of ethanol using alcohol dehydrogenase (ADH), and photometric detection NADH (A340nm). For enzyme assays, in which NAD+, NADH, NADP+ and NADPH were tested as potential effectors of NadEGln, the enzyme activity was measured using a discontinuous assay over the linear phase of the reaction. Reactions were started by the addition of L-glutamine and stopped at different times on ice by adding 100 mM EDTA. The pyrophosphate product was detected using the EnzChek Pyrophosphate Assay Kit (Thermo Fischer). All the assays were performed in triplicate using a Spark microplate reader (Tecan). The initial velocity data was fitted the equation indicated in each figure using GraphPad Prism software package.

In vitro protein complex analysis

In vitro protein complex formation was assayed using His-Magnetic beads (Promega). All reactions were conducted in interaction buffer containing 50 mM Tris-

HCl pH 8, 100 mM NaCl, 5 mM MgCl2, 0.05% Tween 20 (v/v), 10% glycerol (v/v), 20 mM imidazole in the presence or absence of effectors (ATP, ADP and/or 2-OG) as indicated in each experiment. Five microliters of beads were equilibrated in 200 μL of interaction buffer. The beads were recovered and resuspended in 400 μL of buffer followed by the addition of 20 μg of His-PII and 40 μg of untagged NadE protein, in this order. After 5 min at room temperature with gentle mixing, the beads were washed three times with 200 μL of interaction buffer. Bound proteins were eluted by boiling the beads in SDS-PAGE sample buffer. Proteins were analyzed by Coomassie blue stained 15% SDS-PAGE.

Bio-layer interferometry assays

To quantify the kinetic parameters of the PII-NadE complex an Octet K2 Biolayer Interferometry System (FortéBIO) was used. The purified proteins His-GlnZ and 81 untagged AbNadE2 were diluted in the interaction buffer (50 mM Tris-HCl pH 7, 50 mM + KCl, 10 mM MgCl2, 2 mM ATP or ADP, 2.5 mM NAD and 1 mg/mL BSA). The Ni-NTA Biosensor was first dipped into a solution of 2 μg/mL of 6xHis-tagged GlnZ for 150 s until a binding signal of approximately 2 nm was obtained (Loading). The sensor was briefly washed in binding buffer and then transferred to the analyte solution containing AbNadE2 at different concentrations for 150 s to record the association curve. Finally, the sensor was dipped into the interaction buffer for 300 s to monitor the dissociation. In parallel, a reference sensor was subjected to the same procedure, except that no His-tagged protein was bound, to determine the background of unspecific analyte binding. In a series of binding assays, AbNadE2 concentrations of 30 to 2000 nM

(dimer concentration) were used to determine the KD of the GlnZ-AbNadE2 protein complex.

To determine the inhibition constant (Ki) for 2-OG on GlnZ-NadE complex formation, interaction assays using 2μg/mL of 6xHis-tagged GlnZ and 400 nM AbNadE2 solutions were performed in the presence of different concentrations of 2- OG (0 to 480 μM). Therefore, the Ni-NTA Biosensor was loaded with His-GlnZ solution and then transferred into the AbNadE2 solution to measure the association. In a series of experiments, increasing concentrations of 2-OG were added to the interaction buffer to determine the inhibitory effect on association. The experiments were carried out in duplicates and analyzed with the Octet Data Analysis software using Savitzky-Golay filtering. The fitting of the curve was done with a 2:1 heterogeneous ligand model. Curves were then plotted in GraphPad Prism7 software.

Redox titrations of NadE

To determine the dependency of NadE activity on the redox environment, continuous enzymatic assays were performed in the presence of saturating concentrations of 2 mM L-glutamine or 10 mM ammonium chloride. When indicated, 10 mM of freshly diluted DTT (Dithiothreitol) was added 10 min prior to the start of the reactions. A redox titration using different ratios of DTT and DTTOXI (trans-4,5- dihydroxy-1,2-dithiane) at 10 mM total concentration was performed. The reaction buffer consisted of 50 mM Tris-HCl pH 8, 50 mM KCl, 10 mM MgCl2, 2 mM ATP and 0.2 μM of the indicated enzyme (AbNadE2, HsNadE2 and ScNadE2, monomer concentration).

Bioinformatic analysis of PII-NadE genomic islands

Metadata of genome sequences from Archaea and Bacteria available in the NCBI Assembly database were obtained in the following addresses: ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/bacteria/assembly_summary.txt and ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/archaea/assembly_summary.txt (Accessed 13 July, 2019). Only one genome sequence per species (or unknown species isolate) was downloaded to prevent data redundancy, giving priority to sequences obtained from type-strains and to those that represented complete genomes. Positions of PII protein homologs were identified in the genome sequences through tblastn searches, using the GlnB protein of A. brasilense as query (ACF77123.1). Only hits presenting e-values equal or lower than 1e-5 were considered. A sequence window containing the PII homolog sequence with its respective 10 kb upstream and 10 kb downstream sequences was defined as a “PII island”. Genome assemblies not presenting this sequence interval (contigs too short to contain 20 kb of adjacent sequences) were discarded. All protein sequences encoded in the “PII islands” were recovered and they composed a database named as “PII neighbors DB”. Homologs of the glutamine dependent-NadE proteins of A. brasilense (Uniprot G8AIW8) and (Uniprot G8ASI0) were searched in the “PII neighbors DB” with blastp, using 1e-5 as e-value cutoff. All NadE-like protein sequences were recovered using fasgrep from FAST tools (LAWRENCE et al., 2015), and dereplicated using rmdup from Seqkit package (SHEN et al., 2016). The distance between the midpoint of each nadE homolog gene and its closest neighboring PII homolog (midpoint of the PII island) was computed. Only nadE homologous sequences with the midpoint to PII homolog lower than 2000 bp were kept. Duplicated NCBI protein entries were removed. The remaining sequences were retrieved from NCBI and subjected to clustering analysis based on all against all BLAST+ similarities to detect orthologous groups using CLANS (FRICKEY; LUPAS, 2004). The taxonomic affiliation of each NadE homolog was recorded using the data available in the NCBI2lin repository (https://github.com/zyxue/ncbitax2lin), which contains pre-converted lineages from the NCBI taxonomy database.

Results

Identification of the dimeric NadEGln as a novel target of the signal transduction protein GlnZ

We used Ni-NTA column loaded with N-terminal His-tagged GlnZ in an attempt to identify novel PII binding proteins in the diazotrophic α-Proteobacterium A. brasilense. This ligand fishing GlnZ-affinity column was then challenged with A. brasilense protein extracts in the presence of MgATP. A blank nickel column, without His-GlnZ, was used as a negative control. After extensive washes with buffer containing MgATP, both columns were washed with buffer containing MgATP plus 1.5 mM of 2-OG. Two consecutive fractions of 1.5 ml (fractions 1 and 2) eluted with 2-OG were collected and analyzed by label-free LC/MS/MS. The rationale of this approach is that, with the addition of 2-OG, GlnZ would assume a different conformation and, therefore, release proteins that were specifically retained in the column due to direct physical interaction with GlnZ. The top 5 proteins that were specifically enriched in the fractions eluted with MgATP and 2-OG from the His-GlnZ affinity column in comparison to the control column were two putative uncharacterized proteins; orotate phosphoribosyltransferase, which is part of the pyrimidine biosynthetic pathway; pyruvate-phosphate dikinase regulatory protein, a bifunctional serine/threonine kinase/phosphatase that regulates pyruvate-phosphate dikinase; type 2 glutamine- dependent NAD synthetase, namely AbNadE2Gln, that has been studied previously by our group (Fig. S1, Uniprot G8AIW8) (SANTOS et al., 2018).

In vitro complex formation between purified GlnZ and A. brasilense NadE2

The formation of a PII-NadEGln complex has been previously suggested by bioinformatic analysis (SANT’ANNA et al., 2009). Hence, we focused our efforts to confirm the interaction between A. brasilense GlnZ and AbNadE2 using the purified components in vitro. The AbNadE2 enzyme was purified to homogeneity and tested for interaction with the N-terminal His-tagged GlnZ protein using pull-down assays. When ATP or ADP was present in the buffers, AbNadE2 was co-purified with His-GlnZ using nickel beads (Fig. 2A). When the pull-down assays were performed in the

84 presence of MgATP and saturating 2-OG concentrations, no interaction between GlnZ and AbNadE2 could be detected (Fig. 2A).

The interaction between GlnZ and AbNadE2 was then quantitatively analyzed using Biolayer interferometry. The His-tagged GlnZ was mobilized on Ni-NTA tips and exposed to various AbNadE2 solutions in presence of different effector molecules. The binding curves obtained with increasing concentrations of AbNadE2 enabled the estimation of a Kd of 300 nM in the presence of ATP and a Kd of 150 nM, in the presence of ADP (Fig. 2B and Fig. S2, respectively).

Figure 2: In vitro complex formation between AbNadE2 and GlnZ. A) Protein complex formation was assessed by pull-down using Ni2+ beads. Assays were performed in the presence of 5 mM MgCl2 and the effectors ATP, ADP and/or 2-OG at 1 mM, as indicated. Binding reactions were conducted in 400 μl of buffer, adding purified His-GlnZ mixed with native AbNadE2. After extensive washes, bound proteins were eluted with SDS-PAGE loading buffer and analyzed by SDS- PAGE stained with Coomassie Blue. Lanes 1, 4 and 7, HisGlnZ only. Lanes 2, 5 and 8, AnNadE2 only. Lanes 3, 6 and 9, a mixture of HisGlnZ and AbNadE2. B) Biolayer interferometry quantification of GlnZ- AbNadE2 interaction. His-GlnZ was immobilized on the Ni-NTA sensor tip in a concentration of 2 μg/mL. The sensor tip was then challenged in a solution containing the indicated AbNadE2 concentrations in the presence of 5 mM MgCl2 and 1mM ATP. Plots reporting the 'λ spectral shift in nm vs. AbNadE2 concentration are shown in the insert. The binding curves and Kd were determined and calculated using the manufacturer´s software (Fortébio) with the determined association and dissociation rates (kON= 04 -02 9.09x10 1/Mxs, kOFF= 2.72x10 1/s).

Source: SANTOS et al.,(2020) JBC.

The AbNadE2 is feedback inhibited by NAD+

To determine if the GlnZ-AbNadE2 interaction would affect the AbNadE2 activity, enzymatic assays were performed by combining the substrates L-glutamine, ATP, and NaAD. The formation of NAD+ by AbNadE2 was determined spectrophotometrically after its conversion to NADH by alcohol dehydrogenase (ADH), as described previously (SANTOS et al., 2018). Strikingly, despite complex formation, GlnZ had no effect on AbNadE2 activity using saturating or sub-saturating concentrations of the AbNadE2 substrates (data not shown). We reported previously and confirmed here that AbNadE2 is more active in a reducing environment in the presence of dithiothreitol using glutamine as N-donor (Fig. S3). This feature seems to be a general characteristic of dimeric NadE2Gln as we also noted this effect with orthologous enzymes from the E-Proteobacterium Herbaspirillum seropedicae (HsNadE2Gln) and from the Cyanobacteria Synechocystis sp. (ScNadEGln) (Fig S3A). On the other hand, the presence of DTT had no effect when L-glutamine was substituted by ammonium as N-donor for the reaction (Fig. S3B). These data suggest that NadE2Gln enzymes require a reducing environment for the full activity of the glutaminase domain. Given the role of NadE in NAD+ homeostasis and its possible regulation by the redox environment, we thought that some of the cell redox metabolites and the final products of NAD+ biosynthetic pathway (NAD+, NADH, NADP+, and NADPH) could play a role in the regulation of AbNadE2. Hence, AbNadE2 activity was measured by monitoring PPi released in the presence of these metabolites. We found that NAD+ acted as a potent inhibitor of AbNadE2, diminishing the enzyme activity to a17%. The other tested metabolites showed no effect (Fig. 3A). The presence of NAD+ inhibited

AbNadE2 in a dose-dependent hyperbolic curve with an estimated Ki of 1 mM (Fig. 3B).

Figure 3: The AbNadE2 activity is down regulated by NAD+ and GlnZ relives AbNadE2 from NAD+ inhibition. NadE2 discontinuous assays were performed to measure the formation of PPi in the presence of 4 mM L-glutamine, 4 mM ATP and 2 mM NaAD. Data were plotted as relative percent activity considering the reaction without NAD+ as 100 % activity. A) Effect of redox metabolites, when indicated 2.5 mM NAD+ or 1 mM of NADH, NADP+ or NADPH were added. An asterisk indicates statistical difference (T test, p<0.01), n = 3. B) Reactions were carried out in the presence of 100 nM NadE2Ab (monomer), GlnZ was added at 2 μM (trimer) when indicated. Reactions without GlnZ were contained 2 μM BSA. The + calculated AbNadE2 NAD Ki was 1 mM and 2.5 mM in the absence and presence of GlnZ, respectively. Data were analyzed and the IC50 were calculated by non-linear regression in Graph Pad Prism 7.

Source: SANTOS et al.,(2020) JBC.

The structural similarities between NaAD and NAD+ (Fig. 1) lead us to investigate if the presence of NAD+ could inhibit NadE activity by competing with the NaAD substrate (competitive inhibition). The type of inhibition can be distinguished by the kinetic constants, which were, therefore, determined. The presence of NAD+ did not affect the AbNadE2 NaAD KM, but instead, this metabolite altered AbNadE2 Vmax (from 0.09 μmol. s-1 to 0.04 μmol. s-1 in the presence of NAD+) (Fig. S4A). The L- + glutamine KM was also unaffected in the presence of 1 mM NAD (Fig. S4B). These data indicate the presence of an allosteric NAD+ inhibitory binding site on AbNadE2.

The feedback inhibition of AbNadE2 by NAD+ is relieved by GlnZ

The presence of GlnZ did not affect AbNadE2 activity in the presence or absence of 5 mM NAD+ (Fig. 3B). However, GlnZ was able to partially relief the inhibitory effect on AbNadE2 activity when NAD+ was present under physiological relevant levels (between 0.5 and 2.5 mM). The AbNadE2 NAD+ inhibition curve changed from hyperbolic to sigmoidal in the presence of GlnZ (Fig. 3B) increasing the + IC50 from 1 mM in the absence of GlnZ to 2.5 mM in its presence. The AbNadE2 NAD + IC50 in the presence of GlnZ is near the reported intracellular NAD concentration in E. coli of 2.6 mM (BENNETT et al., 2009). When NAD+ was kept constant at the physiologically relevant concentration of 2.5 mM, GlnZ was able to activate AbNadE2 in a hyperbolic GlnZ dose-dependent curve (Fig. 4A). The estimated Kact was 640 nM, which is in the same range as the Kd for the GlnZ-AbNadE2 complex estimated using Biolayer interferometry (Fig 2B). Given the negative effect of 2-OG on the formation of the GlnZ-AbNadE2 complex (Fig. 2A), we sought whether 2-OG could impair the ability of GlnZ to relive AbNadE2 NAD+ inhibition. When AbNadE2 activity was measured in the presence of 2.5 mM NAD+, and GlnZ was first equilibrated in buffer containing 2-OG before combining with AbNadE2, 2-OG impaired the ability of GlnZ to relive AbNadE2 NAD+ inhibition in a 2-OG dependent manner (Fig. 4B). The response to 2-OG was hyperbolic with a Ki for 2-OG of 15 μM (Fig. 4B). The Ki for 2-OG is in the same range than the Ki for 2-OG dependent inhibition of GlnZ-AbNadE2 complex formation, as determined by pull-down or Biolayer interferometry (Fig. 4C and 4D). When AbNadE2 activity was measured by mixing GlnZ and AbNadE2 before the addition of 2-OG, a different response to increasing 2-OG levels was observed.

The effect of 2-OG was much less pronounced, with 2-OG showing a Ki of 1 mM (compare Fig. 4B and E and Fig. 4C and F). Note that while the experiments depicted in Fig 4B reflect the ability of 2-OG to inhibit GlnZ-AbNadE2 complex formation, the assay in Fig. 4F reflects the ability of 2-OG to dissociate a pre-formed GlnZ-AbNadE2 complex. Pull-down assays indicated that concentrations of 2-OG up to 2 mM was not enough to fully dissociate a pre-formed GlnZ-AbNadE2 complex (Fig. 4E). On the other hand, 2 mM 2-OG completely abolished GlnZ-AbNadE2 complex formation (Fig. 4C). These data imply that once the GlnZ-AbNadE2 complex is formed, 2-OG is much less effective to impair the interaction between these proteins and, hence, to relieve AbNadE2 NAD+ inhibition.

Figure 4: Effect of 2-oxoglutarate in the dissociation of GlnZ-NadE2Ab complex and enzymatic activity of AbNadE2. NadE2 discontinuous assays were performed to measure the formation of PPi in the presence of 4 mM L-glutamine, 4 mM ATP and 2 mM NaAD. Data were plotted as relative percent activity considering the reaction without NAD+ as 100 % activity. A) Reactions contained 100 nM NadE2Ab and 2.5 mM of NAD+.

GlnZ was added at 0.25, 1, 2, 4 and 10 μM (trimer). The estimated Kd for the AbNadE2-GlnZ complex was 640 nM were calculated using non-linear regression in Graph Pad Prism 7. B) The activity of AbNadE2 in the presence of GlnZ is regulated by the levels of 2-OG. Data were plotted as percent inhibition relative to the AbNadE2 activity in the presence of NAD+, which was considered as 100% inhibition. Assays were performed in the presence of 2.5 mM NAD+, 2 μM GlnZ and increasing 2-OG concentrations. The estimated half-maximal inhibitory concentration (Ki) of 2-OG was 15.5 μM. C) The formation of the GlnZ-AbNadE2 complex was assessed by pull-down using Ni2+ beads. Reactions were performed in the presence of MgCl2 5mM, 1 mM ATP and increasing concentrations of 2-OG as indicated. Binding reactions were conducted in 400 μl of buffer, adding purified His-GlnZ and untagged AbNadE2. After extensive washes, bound proteins were eluted with SDS-PAGE loading buffer and 89 analyzed by SDS-PAGE stained with Coomassie Blue. D) Biolayer interferometry analysis of the GlnZ- AbNadE2 complex. The His-GlnZ was immobilized in the Ni-NTA sensor tip. The tip was then steeped in a solution containing 400 nM of AbNadE2, MgCl2 5mM, 1 mM ATP and different concentrations of 2- OG. Plots reporting the 'λ spectral shift in nm vs. 2-OG concentrations are shown in the insert. The binding curves and Ki were determined using the manufacturer´s software (Fortébio). E) Curve of AbNadE2 relative inhibition in response to increasing 2-OG concentrations. The AbNadE2 inhibition in the presence of 5mM NAD+ is considered 100% inhibition. Reactions were carried in the presence of 2.5 mM of NAD+ and 2 μM of GlnZ. The GlnZ and NadE2 proteins were preincubated in MgATP before the addition of 2-OG and the start of the reaction. Hence, this assay reflects the ability of 2-OG to dissociate a pre-formed GlnZ-AbNadE2 complex. The estimated Ki for 2-OG was 1 mM was calculated non-linear regression in Graph Pad Prism 7 . F) The HisGlnZ-NadE2 complex was immobilized on Ni2+ beads in the presence of 1mM ATP. Beads were washed (Wash) with buffer containing ATP (1mM) and the indicated 2-OG concentrations. Bound proteins were eluted in SDS-PAGE sample buffer (Elution). The samples were applied to SDS-PAGE 15%, and the gels were stained with Coomassie Blue.

Source: SANTOS et al.,(2020) JBC.

Effects of GlnZ uridylylation and the GlnZ T-loop on AbNadE2 activity

Under nitrogen limiting conditions, GlnZ is predominantly found in fully uridylylated form in vivo (HUERGO et al., 2006a). Using in vitro uridylylated GlnZ protein, we could show that, in contrast to non-uridylylated GlnZ, GlnZ-UMP3 was not + able to relieve AbNadE2 NAD inhibition (Fig. 5A). Furthermore, GlnZ-UMP3 could not interact with AbNadE2, as indicated by pull-down assays (data not shown). Hence, the uridylylation of GlnZ abrogates its ability to interact with AbNadE2 and to relieve the NAD+ inhibition.

Figure 5: Effect of PII protein on NadE2 activity depends on the PII variant and is conserved in different Bacteria. + A) The activity of AbNadE2 was measured in the presence of MgCl2 5mM, 1 mM ATP, 2.5 mM NAD and the indicated PII protein (GlnZ, GlnZΔloopT, GlnB and GlnZ-UMP3) at 2μM concentration (trimer), 1.5 mM of 2-OG was added when indicated. Treatment without PII was carried with 2 μM of BSA. Activity measured following PPI production. An asterisk indicates the statistical difference in comparison to GlnZ (T test, p<0.01), n = 3. B and C) NadE2 activity data were plotted as relative percent activity considering the reaction without NAD+ as 100 % activity. All reactions were carried out in the presence of 100 nM of HsNadE2 (B) or ScNadE2 (C), 4 mM ATP, 2 mM NaAD and 4 mM L-Glutamine. When indicated, the corresponding PII protein was added at 2 μM (trimer). Curves without PII were carried out using 2 μM

BSA. Activity measured following PPI with EnzChek Pyrophosphate Assay Kit. An asterisk indicates a statistical difference compared (T test, p<0.01); n = 3.

Source: SANTOS et al.,(2020) JBC.

Most of the known interactions between PII proteins and their target proteins occurs via the T-loop of the PII protein. Indeed, this is the region that suffers the most prominent conformational change upon 2-OG binding (TRUAN et al., 2010). We tested the ability of a GlnZ variant carrying a deletion of the T-loop (GlnZ ΔQ42-S54) to interact with AbNadE2. The delta T-loop variant was still able to interact with GlnZ, but the complex formation was unresponsive to 2-OG (Fig. S5A). In agreement with the pull-down data, the GlnZ variant ΔQ42-S54 was able to relieve the AbNadE2 NAD+ inhibition despite the presence of 2-OG (Fig. 5A). 91

A. brasilense encodes a second PII paralogue, namely GlnB, which is 67% identical to GlnZ at the amino acid sequence (DE ZAMAROCZY, 1998). To determine if GlnB could also interact with AbNadE2, pull-down assays were performed using His- GlnB or His-GlnZ as a bait in the presence of ATP. Even though AbNadE2 could be co-purified with His-GlnB, the amount of AbNadE2 protein recovered was higher using His-GlnZ as bait (Fig. S5B). In agreement, GlnB was much less effective than GlnZ in relieving the AbNadE2 NAD+ inhibition (Fig. 5A). These data suggest that both A. brasilense PII paralogues can interact with AbNadE2, but a more stable complex is apparently formed with the GlnZ paralogue.

The regulatory interplay between NAD+ and PII proteins is conserved among bacterial dimeric glutamine dependent NadEGln

In order to determine if regulatory interplay between NAD+ and PII proteins is conserved among NadE2Gln in bacteria, the NadE2Gln enzymes from the E- Proteobacterium H. seropedicae and the Cyanobacteria Synechocystis sp. were purified to homogeneity and tested for enzymatic activity in the presence of NAD+ and the GlnZ/GlnB orthologue from the respective organism. In both cases, NadE2Gln was inhibited by NAD+ in a dose-dependent manner within the physiological range (Fig. 5B and C). For both organisms, the presence of PII relieved NadE2Gln NAD+ inhibition, although the PII effect was less pronounced in the case of Synechocystis sp. NadE2 (Fig. 5B). These data suggest that both the inhibition by NAD+ and regulation by PII are conserved mechanisms to regulate NadE2Gln in bacteria.

Bioinformatic analysis of the PII-nadE gene islands

Inspection of the PIRSF database (WU, 2004) identified 19,848 sequences homologous and homeomorphic to AbNadE2 (i.e., containing similar domain architecture, a glutaminase domain fused to NAD synthetase domain, PIRSF accession 006630). These sequences are distributed in 1,772 eukaryotes, 17,578 bacteria, and 222 Archaea. Hence, glutamine dependent NadE enzymes are widespread throughout nature. The conservation of gene order can be used as a fingerprint of proteins that physically interact (DANDEKAR et al., 1998). We have set a bioinformatic approach to identify PII-nadE genetic islands in prokaryotes. A total of 1300 non-redundant PII-

92 nadE islands were identified (Table S1). Most of the PII-nadE islands identified belonged to β- and γ- Proteobacteria, 834 and 436 counts, respectively (Table S1). Interestingly, in A. brasilense and Synechocystis sp., the PII gene is not linked nadE; PII nevertheless regulated the NadEGln function (Figs. 3B and 5C). Hence, NadEGln regulation by PII can occur despite the absence of a genetic link of respective genes in these cases. The bioinformatic analysis reinforces that the formation of the PII-NadE complex may be a conserved feature in bacteria and may also be present in Archaea. The NadE sequences identified within the PII-nadE islands were separated into three major groups using CLANS (Fig. 6). Most of the retrieved NadE sequences belonged to dimeric NadEGln. Interestingly, a significant number of octameric glutamine dependent NadEGln synthetases was also retrieved in this analysis, along with few members of ammonium dependent NadENH3 enzymes (Fig. 6). These data suggest that regulation of NAD+ biosynthesis by PII may be even more widespread in nature and not only restricted to dimeric type 2 NadEGln.

Figure 6: Similarity based grouping of NadE orthologues identified in the PII-NadE genomic islands. All the NadE sequences retrieved from the identified PII-NadE islands were compared using all-against- all BLAST searches using CLANS. Each protein sequence is displayed as a dot and the pairwise similarities is presented in a graph accordingly to the similarity distances. Three major clusters were formed. The sequences of previously characterized dimeric and octameric glutamine dependent NadE were included as function group guide in the analysis and are indicated. The green cluster included the octameric NadE1Gln from Mycobacterium tuberculosis (M.tuberculosis), NadE1Gln from Herbaspirillum seropedicae (HsNadE1) and NadE1Gln from Azospirillum brasilense (AbNadE1). The blue cluster included the dimeric glutamine-dependent NadE enzymes from Herbaspirillum seropedicae (HsNadE2), NadE2Gln from Azospirillum brasilense (AbNadE2), NadEGln from Burkholderia thailandensis (B. thailandensis) and Synechocystis sp. (ScNadE2). The red cluster include NadE sequences lacking the N-terminal glutaminase domain and thus were considered putative ammonia dependent NadE.

Source: SANTOS et al.,(2020) JBC.

Discussion The biosynthesis of NAD+ is a vital metabolic pathway in bacteria such that the suppression of the enzyme catalyzing the last step of NAD biosynthesis, NadE, resulted in bactericidal effects (RODIONOVA et al., 2014) due to nicotinamide nucleotides being essential cofactors in redox catalysis, and the total concentration of these molecules must cover their requirement for cell metabolism. In addition to this well-established role, NAD+ also serves as a substrate for sirtuins (lysine deacetylases), ADP-ribosylating enzymes, DNA ligase, and NAD+ kinase and may act as a signaling metabolite (HOUTKOOPER et al., 2010; SORCI; RUGGIERI; RAFFAELLI, 2014). For these reasons, the production of NAD+ must be coordinated with its consumption and cell growth to keep the nicotinamide nucleotide levels homeostatic. Here we show that the glutamine-dependent dimeric NadE2 from distantly related bacteria such as from A. brasilense (α-Proteobacteria), H. seropedicae (E- Proteobacteria) and Synechocystis sp. (Cyanobacteria) are negative feedback inhibited by physiologically relevant concentrations of NAD+ (Fig. 3B, 5B and C). We speculate that bacterial NadE2Gln may be universally inhibited by NAD+. Kinetic analysis with AbNadE2 showed that NAD+ does not compete with the binding of substrate NaAD but reduces the enzymes Vmax (Fig. S4). This clearly indicates that NadE2Gln is feedback-inhibited through an allosteric inhibitory binding site for NAD+. Initially, we used a PII protein affinity column to successfully identify AbNadE2 as a novel target of the PII protein GlnZ in the diazotrophic α-Proteobacteria A. brasilense (Fig. 2A and Fig. S1). The presence of GlnZ changed the response of AbNadE2 to its feedback inhibitor NAD+. Instead of a hyperbolic inhibition curve with a

Ki of 1 mM, the GlnZ-AbNadE2 complex displayed a sigmoidal inhibition curve with an + IC50 of 2.5 mM. Thereby, GlnZ relived the AbNadE2 NAD inhibitory effect in particular under physiologically relevant NAD+ concentrations (Fig. 3B). The sigmoidal dose- response curve indicates the cooperativity of the NAD+ allosteric binding sites in the GlnZ-AbNadE2 complex (Hill coefficient 2.2). The presence of high 2-OG levels or GlnZ uridylylation impaired the formation of the AbNadE2-GlnZ complex, thereby preventing GlnZ to revert AbNadE2 NAD+ inhibition (Fig. 4D and 5A). From these data, we propose a model where the PII protein acts as a dissociable regulatory subunit of NadE2Gln to regulate NAD biosynthesis according to

94 the metabolic state sensed by GlnZ (Fig. 7). Under low nitrogen availability, glutamine levels are low, and 2-OG levels are high (BENNETT et al., 2009; VAN HEESWIJK; WESTERHOFF; BOOGERD, 2013), PII (GlnZ) is uridylylated resulting in diminished flux through NadE2 due to NAD+ feedback inhibition and low availability of the L- glutamine substrate (Fig. 7A). Upon an ammonium shock, glutamine levels rise (SANTOS et al., 2018), and PII is rapidly deuridylylated (HUERGO et al., 2006a), 2- OG levels reduce due to its use in ammonium assimilatory reactions (HUERGO; DIXON, 2015; YUAN et al., 2009). These conditions favor the formation of the NadE2- PII complex, thereby relieving the NAD+ inhibition over NadE2, allowing increased flux through NadE2 (Fig. 7B). This is of particular importance in the nitrogen-fixing bacterium A. brasilense, which possesses a mechanism to inactivate the abundant enzyme nitrogenase by ADP-ribosylation as an immediate response to ammonium shock (HUERGO et al., 2012). The nitrogenase ADP-ribosyl-transferase DraT uses NAD+ as a substrate and is activated upon interaction with the PII protein GlnB (MOURE et al., 2013, 2014) precisely under the conditions where GlnZ activates NadE2 (Fig. 7). Hence, the concerted activation of the NAD+ consuming enzyme, DraT, with the NAD+ producing enzyme, NadE2, by the paralogue PII proteins, would ensure NAD+ homeostasis during the enzymatic consumption of this cofactor.

Figure 7: The PII proteins act as a dissociable regulatory subunit of NadE2Gln to pace NAD+ production accordingly to the availability of L-glutamine and 2-OG. A) When nitrogen is limited, cell growth is arrested, the intracellular levels of 2-OG are high, and glutamine are low. PII is fully uridylylated and does not interact with NadE2, resulting in strong feedback negative inhibition of NadE2 by NAD+. B) When ammonium becomes available, glutamine rises and 2-

OG drops, PII is de-uridylylated (reaction catalyzed by the GlnD enzyme in response to high glutamine) and bound to MgATP and/or ADP (orange dots). Under this condition, PII interacts with NadE2 relieving NAD+ inhibition and thus allowing higher NAD+ production rates to feed the demands of active cell growth.

Source: SANTOS et al.,(2020) JBC.

Our in vitro data indicate that once the PII-NadE2 complex is formed, higher levels of 2-OG are needed to dissociate the complex in comparison to the 2-OG levels required to inhibit complex formation (Fig. 4E and F). Apparently, the PII affinity for 2- OG binding is reduced within the PII-NadE2 complex in comparison to free PII. This may result in a “memory effect” such that once the GlnZ-AbNadE2 complex is formed, it would become resistant to small and transient fluctuations in 2-OG concentrations. That the affinity of PII proteins towards the effector molecules is differentially modulated through complex formation with various targets has already been described at a structural level for cyanobacterial PII protein (FORCHHAMMER; LÜDDECKE, 2016c; ZETH; FOKINAS; FORCHHAMMER, 2014). The action of PII to tune down feedback-inhibition of NadE2 by NAD+ is conserved in distantly related bacterial such as H. seropedicae (E-Proteobacteria) or Synechocystis sp. (Cyanobacteria) (Fig. 5B and C). Hence, the model depicted in Fig. 7 may be applied for other prokaryotes, as well. The conservation of gene order is a fingerprint of proteins that physically interact (DANDEKAR et al., 1998). The fact that the PII-nadE pair is conserved in widely distant prokaryotes of bacterial and archaeal origin (Supporting Information, Table S1) suggests that the PII-NadE interaction may be widespread in prokaryotes. This is in analogy to the conserved PII-amtB genetic pair, whose protein products were shown to physically interact in a broad range of prokaryotes, including bacteria and Archaea (COUTTS et al., 2002; HUERGO; CHANDRA; MERRICK, 2013a). Our gene neighborhood analysis also suggests that other types of NAD synthetases, NadE1Gln and NadENH3, could be regulated by interaction with PII (Fig. 6). This hypothesis is currently under investigation. The unique ability of the PII protein to sense and integrate the levels of essential metabolites such as ATP, ADP. 2-OG and glutamine were capitalized by nature in such a way that PII regulates a range of key metabolic enzymes, transporters and transcriptional regulators by direct protein- interactions (HUERGO; CHANDRA; MERRICK, 2013a). In addition to the well-established control of the nitrogen

96 metabolism, recent data showed that PII regulates other key metabolic pathways, such as those involved in fatty acid production (GERHARDT et al., 2015; RODRIGUES et al., 2019). The identification of NadE2Gln as a novel PII target-protein supports the notion that PII may function as a central orchestration unit of metabolism to tune the flux trough key metabolic pathways according to the metabolic state determined by the availability of nutrients and energy supply. When nutrients and energy are plentily available, microbes achieve maximum growth rates, and under these conditions, the production of NAD+ and its derivatives (NADH, NADP and NADPH) must increase to keep pace with cell growth, providing enough cofactors for daughter cells. Indeed, single-cell fluorescence measurements indicate a peak of NADH accumulation during the progress to cell division in E. coli (ZHANG; MILIAS-ARGEITIS; HEINEMANN, 2018). As the new findings pointed to regulation through NAD+ levels while NADH did not affect NadE2Gln, we speculate that oscillations in the NADH/NAD+ ratio may play a role in the NAD+ biosynthesis and thus in regulation through PII transduction proteins. The mechanism responsible for coordinating NAD+ production with nutrient availability has remained elusive. The PII proteins, with their extraordinary sensory properties, are ideally suited to fulfill this role. The NAD+ levels play important signaling roles in eukaryotes by regulating processes such as DNA repair, cell cycle regulation, gene expression and calcium signaling (LARONDE-LEBLANC; RESTO; GERRATANA, 2009b; STRØMLAND et al., 2019). Impaired biosynthesis and increased NAD+ consumption are associated with aging and pathologies in mammals (STRØMLAND et al., 2019). The use of NAD+ as a substrate for a range of enzyme prokaryotic enzymes, including those involved in RNA modification (CAHOVÁ et al., 2015), sirtuins (BURCKHARDT; BUCKNER; ESCALANTEǦSEMERENA, 2019) and ADP-ribosylation (LIN, 2007), suggest that NAD+ also play important regulatory roles in prokaryotes. The key metabolic function makes NAD+ well-suited for signaling. The interplay between the levels of ATP, ADP, 2-OG, glutamine (sensed by PII) and NAD+ (sensed by NadE2), through the formation of the PII-NadE2 complex, constitute a novel metabolic hub that may act to balance the levels of core cellular metabolites.

Acknowledgments This work was supported by UFPR, CNPq, Fundação Araucária, CAPES, CNPq-INCT, as well as by DFG grant Fo195/9-2. Adrian Santos acknowledges the 97 support of a CAPES-DAAD cotutelle Ph.D. fellowship and as an associate member of the research training group GRK 1708. Luciano Huergo acknowledges the support of the Humboldt Foundation.

References ABOUELFETOUH, A. et al. The E. coli sirtuin CobB shows no preference for enzymatic and nonenzymatic lysine acetylation substrate sites. MicrobiologyOpen, v. 4, n. 1, p. 66–83, 1 fev. 2015.

AFZAL, M.; KUIPERS, O. P.; SHAFEEQ, S. Niacin-mediated gene expression and role of NiaR as a transcriptional repressor of niaX, nadC, and pnuC in Streptococcus pneumoniae. Frontiers in Cellular and Infection Microbiology, 2017.

AKENTIEVA, N. Posttranslational modification of dinitrogenase reductase in Rhodospirillum rubrum treated with fluoroacetate. World Journal of Microbiology and Biotechnology, v. 34, n. 12, p. 1–10, 1 dez. 2018.

ALOMARI, A. Biophysical and Kinetic Analysis of Escherichia coli DNA Ligase Activity and Inhibition. [s.l.] University of Portsmouth, 2018.

ARAÚJO, L. M. et al. Different responses of the GlnB and GlnZ proteins upon in vitro uridylylation by the Azospirillum brasilense GlnD protein. Brazilian Journal of Medical and Biological Research, v. 41, n. 4, p. 289–294, 2008.

ARAUJO, M. S. et al. In vitro uridylylation of the Azospirillum brasilense N-signal transducing GlnZ protein. Protein Expression and Purification, v. 33, n. 1, p. 19–24, 2004.

ARAVIND, L. et al. The natural history of ADP-ribosyltransferases and the ADP- ribosylation system. Current topics in microbiology and immunology, v. 384, p. 3– 32, 2015.

ARCONDÉGUY, T.; JACK, R.; MERRICK, M. P(II) signal transduction proteins, pivotal players in microbial nitrogen control. Microbiology and molecular biology reviews : MMBR, v. 65, n. 1, p. 80–105, mar. 2001.

BAEZA, J. et al. Stoichiometry of site-specific lysine acetylation in an entire proteome. Journal of Biological Chemistry, v. 289, n. 31, p. 21326–21338, 1 ago. 2014.

BATISTA, M. B.; DIXON, R. Manipulating nitrogen regulation in diazotrophic bacteria for agronomic benefitBiochemical Society TransactionsPortland Press Ltd, , 30 abr. 2019.

BAUGH, L. et al. Combining Functional and Structural Genomics to Sample the Essential Burkholderia Structome. PLoS ONE, v. 8, n. 1, 2013.

BEGLEY, T. P. et al. Cofactor Biosynthesis. [s.l.] Elsevier, 2001. v. 61

BELLINZONI, M. et al. Heterologous expression, purification, and enzymatic activity of Mycobacterium tuberculosis NAD(+) synthetase. Protein expression and purification, v. 25, n. 3, p. 547–57, ago. 2002.

BELLINZONI, M. et al. Glutamine amidotransferase activity of NAD+ synthetase from Mycobacterium tuberculosis depends on an amino-terminal nitrilase domain. Research in microbiology, v. 156, n. 2, p. 173–7, mar. 2005.

BENNETT, B. D. et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nature Chemical Biology, v. 5, n. 8, p. 593–599, 2009.

BERGMEYER, H. U.; BERNT, E. Lactate Dehydrogenase. In: Methods of Enzymatic Analysis. [s.l.] Elsevier, 1974. p. 574.

BHEDA, P. et al. The Substrate Specificity of Sirtuins. Annual Review of Biochemistry, v. 85, n. 1, p. 405–429, 2016.

BIRD, J. G. et al. The mechanism of RNA 5′ capping with NAD+, NADH and desphospho-CoA. Nature, v. 535, n. 7612, p. 444–447, 2016.

BOCKWOLDT, M. et al. Identification of evolutionary and kinetic drivers of NAD- dependent signaling. Proceedings of the National Academy of Sciences, v. 116, n. 32, p. 15957–15966, 2019.

BONATTO, A. C. et al. Uridylylation of Herbaspirillum seropedicae GlnB and GlnK proteins is differentially affected by ATP, ADP and 2-oxoglutarate in vitro. Archives of microbiology, v. 194, n. 8, p. 643–52, ago. 2012.

BORRA, M. T. et al. Substrate specificity and kinetic mechanism of the Sir2 family of NAD +-dependent histone/protein deacetylases. Biochemistry, v. 43, n. 30, p. 9877–

9887, 3 ago. 2004.

BRENNER, C. Catalysis in the nitrilase superfamily. Current opinion in structural biology, v. 12, n. 6, p. 775–82, dez. 2002.

BROSTEDT, E. et al. Electron transport to nitrogenase in Rhodospirillum rubrum: the role of NAD(P)H as electron donor and the effect of fluoroacetate on nitrogenase activity. FEMS Microbiology Letters, v. 150, n. 2, p. 263–267, 17 jan. 1997.

BURCKHARDT, R. M.; BUCKNER, B. A.; ESCALANTEǦSEMERENA, J. C. Staphylococcus aureus modulates the activity of acetylǦCoenzyme A synthetase (Acs) by sirtuinǦdependent reversible lysine acetylation . Molecular Microbiology, 2019.

CAHOVÁ, H. et al. NAD captureSeq indicates NAD as a bacterial cap for a subset of regulatory RNAs. Nature, v. 519, n. 7543, p. 374–377, 2015.

CANELAS, A. B.; VAN GULIK, W. M.; HEIJNEN, J. J. Determination of the cytosolic free NAD/NADH ratio in Saccharomyces cerevisiae under steady-state and highly dynamic conditions. Biotechnology and Bioengineering, v. 100, n. 4, p. 734–743, 1 jul. 2008.

CARABETTA, V. J.; CRISTEA, I. M. Regulation, function, and detection of protein acetylation in bacteriaJournal of BacteriologyAmerican Society for Microbiology, , 1 ago. 2017.

CARPOUSIS, A. J. The RNA Degradosome of Escherichia coli : An mRNA-Degrading Machine Assembled on RNase E. Annual Review of Microbiology, v. 61, n. 1, p. 71– 87, 26 out. 2007.

CASTAÑOǦCEREZO, S. et al. Protein acetylation affects acetate metabolism, motility and acid stress response in Escherichia coli . Molecular Systems Biology, v. 10, n. 11, p. 762, nov. 2014.

CHELLAMUTHU, V. R. et al. A widespread glutamine-sensing mechanism in the plant kingdom. Cell, v. 159, n. 5, p. 1188–1199, 2014.

CHEN, S.-H.; YU, X. Human DNA ligase IV is able to use NAD + as an alternative adenylation donor for DNA ends ligation. Nucleic Acids Research, v. 47, n. 3, p. 1321–1334, 2018.

100

CHEN, Y. G. et al. LC/MS analysis of cellular RNA reveals NAD-linked RNA. Nature Chemical Biology, v. 5, n. 12, p. 879–881, 2009.

CHUENCHOR, W. et al. Regulation of the intersubunit ammonia tunnel in Mycobacterium tuberculosis glutamine-dependent NAD+ synthetase. The Biochemical journal, v. 443, n. 2, p. 417–26, 15 abr. 2012.

CHUENCHOR, W. et al. Different ways to transport ammonia in human and Mycobacterium tuberculosis NAD+ synthetases. Nature Communications, v. 11, n. 1, p. 1–12, 1 dez. 2020.

CICCHILLO, R. M. et al. Escherichia coli quinolinate synthetase does indeed harbor a [4Fe-4S] cluster. Journal of the American Chemical Society, v. 127, n. 20, p. 7310– 7311, 25 maio 2005.

COLLIER, R. J. Understanding the mode of action of diphtheria toxin: A perspective on progress during the 20th centuryToxiconElsevier Ltd, , nov. 2001. Disponível em: . Acesso em: 16 mar. 2020

CORDA, D.; GIROLAMO, M. DI. Functional aspects of protein mono-ADP-ribosylation. The EMBO Journal, v. 22, n. 9, p. 1953–1958, 1 maio 2003.

COUTTS, G. et al. Membrane sequestration of the signal transduction protein GlnK by the ammonium transporter AmtB. The EMBO journal, v. 21, n. 4, p. 536–45, fev. 2002.

DA ROCHA, R. A. et al. Mathematical model of the binding of allosteric effectors to the escherichia coli PII signal transduction protein GlnB. Biochemistry, v. 52, n. 15, 2013.

DANDEKAR, T. et al. Conservation of gene order: A fingerprint of proteins that physically interact. Trends in Biochemical Sciences, 1998.

DE INGENIIS, J. et al. Glutamine versus ammonia utilization in the NAD synthetase family. PloS one, v. 7, n. 6, p. e39115, jan. 2012.

DE ZAMAROCZY, M. Structural homologues PII and PZ of Azospirillum brasilense provide intracellular signalling for selective regulation of various nitrogen-dependent functions. Molecular Microbiology, v. 29, n. 2, p. 449–463, 1998.

DI LUCCIO, E.; WILSON, D. K. Comprehensive X-ray structural studies of the

101 quinolinate phosphoribosyl transferase (BNA6) from Saccharomyces cerevisiae. Biochemistry, v. 47, n. 13, p. 4039–4050, 1 abr. 2008.

DIXON, R.; KAHN, D. Genetic regulation of biological nitrogen fixation. Nature reviews. Microbiology, v. 2, n. 8, p. 621–31, ago. 2004.

EBERT, B. E. et al. Response of Pseudomonas putida KT2440 to increased NADH and ATP demand. Applied and Environmental Microbiology, v. 77, n. 18, p. 6597– 6605, 15 set. 2011.

FERIA BOURRELLIER, A. B. et al. Chloroplast acetyl-CoA carboxylase activity is 2- oxoglutarate-regulated by interaction of PII with the biotin carboxyl carrier subunit. v. 107, n. 1, 2010.

FERNIE, A. et al. Editorial: NAD Metabolism and Signaling in Plants. Frontiers in Plant Science, v. 11, p. 146, 26 fev. 2020.

FLOQUET, N. et al. Ammonia channeling in bacterial glucosamine-6-phosphate synthase (Glms): Molecular dynamics simulations and kinetic studies of protein mutants. FEBS Letters, v. 581, n. 16, p. 2981–2987, 26 jun. 2007.

FOKINA, O. et al. Mechanism of 2-oxoglutarate signaling by the Synechococcus elongatus PII signal transduction protein. Proceedings of the National Academy of Sciences, 2010.

FORCHHAMMER, K. P(II) signal transducers: novel functional and structural insights. Trends in microbiology, v. 16, n. 2, p. 65–72, mar. 2008.

FORCHHAMMER, K.; DE MARSAC, N. T. The P(II) protein in the cyanobacterium Synechococcus sp. strain PCC 7942 is modified by serine phosphorylation and signals the cellular N-status. Journal of Bacteriology, v. 176, n. 1, p. 84–91, 1 jan. 1994.

FORCHHAMMER, K.; DE MARSAC, N. T. Phosphorylation of the P(II) protein (glnB gene product) in the cyanobacterium Synechococcus sp. strain PCC 7942: Analysis of

102 in vitro kinase activity. Journal of Bacteriology, v. 177, n. 20, p. 5812–5817, 1 out. 1995.

FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the PII signalling protein family. FEBS Journal, v. 283, n. 3, p. 425–437, 2016a.

FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the PII signalling protein family. The FEBS journal, v. 283, n. 3, p. 425–37, fev. 2016b.

FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the P II signalling protein family. FEBS Journal, v. 283, n. 3, p. 425–437, fev. 2016c.

FORCHHAMMER, K.; SELIM, K. A. Carbon/nitrogen homeostasis control in cyanobacteria. FEMS Microbiology Reviews, v. 025, p. 33–53, 2020.

FOROUHAR, F. et al. Molecular insights into substrate recognition and catalysis by tryptophan 2,3-dioxygenase. Proceedings of the National Academy of Sciences of the United States of America, v. 104, n. 2, p. 473–478, 9 jan. 2007.

FRICKEY, T.; LUPAS, A. CLANS: A Java application for visualizing protein families based on pairwise similarity. Bioinformatics, 2004.

FRINDERT, J. et al. Identification, Biosynthesis, and Decapping of NAD-Capped RNAs in B. subtilis. Cell Reports, v. 24, n. 7, p. 1890- 1901.e8, 14 ago. 2018.

FRYE, R. A. Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins. Biochemical and Biophysical Research Communications, v. 273, n. 2, p. 793–798, 5 jul. 2000.

GALLEGO-JARA, J. et al. Characterization of CobB kinetics and inhibition by nicotinamide. PLoS ONE, v. 12, n. 12, 1 dez. 2017.

GAO, R. et al. A single regulator nrtr controls bacterial NAD+ homeostasis via its acetylation. eLife, 2019.

GARDNER, J. G.; ESCALANTE-SEMERENA, J. C. In bacillus subtilis, the sirtuin protein deacetylase, encoded by the srtN gene (formerly yhdZ), and functions encoded by the acuABC genes control the activity of acetyl coenzyme a synthetase. Journal of Bacteriology, v. 191, n. 6, p. 1749–1755, mar. 2009.

GAZZANIGA, F. et al. Microbial NAD metabolism: lessons from comparative 103 genomics. Microbiology and molecular biology reviews : MMBR, v. 73, n. 3, p. 529–41, Table of Contents, 2009a.

GAZZANIGA, F. et al. Microbial NAD metabolism: lessons from comparative genomics. Microbiology and molecular biology reviews : MMBR, v. 73, n. 3, p. 529–41, Table of Contents, set. 2009b.

GEORLETTE, D. et al. Cofactor Binding Modulates the Conformational Stabilities and Unfolding Patterns of NAD+-dependent DNA Ligases from Escherichia coli and Thermus scotoductus. Journal of Biological Chemistry, v. 278, n. 50, p. 49945– 49953, 12 dez. 2003.

GERASIMOVA, A. V.; GELFAND, M. S. Evolution of the NadR regulon in enterobacteriaceae. Journal of Bioinformatics and Computational Biology, 2005.

GERDES, S. Y. et al. From Genetic Footprinting to Antimicrobial Drug Targets: Examples in Cofactor Biosynthetic Pathways. Journal of Bacteriology, v. 184, n. 16, p. 4555–4572, 15 ago. 2002.

GERDES, S. Y. et al. Comparative genomics of NAD biosynthesis in cyanobacteria. Journal of bacteriology, v. 188, n. 8, p. 3012–23, 2006.

GERHARDT, E. C. M. et al. Influence of the ADP/ATP ratio, 2-oxoglutarate and divalent ions on Azospirillum brasilense PII protein signalling. Microbiology (Reading, England), v. 158, n. Pt 6, p. 1656–63, jun. 2012.

GERHARDT, E. C. M. et al. The Bacterial signal transduction protein GlnB regulates the committed step in fatty acid biosynthesis by acting as a dissociable regulatory subunit of acetyl-CoA carboxylase. Molecular Microbiology, v. 95, n. 6, 2015.

GOTTESMAN, M. M.; HICKS, M. L.; GELLERT, M. Genetics and function of DNA ligase in Escherichia coli. Journal of Molecular Biology, v. 77, n. 4, p. 531–547, 15 jul. 1973.

GRAVINA, F. et al. Proteome analysis of an Escherichia coli ptsN-null strain under different nitrogen regimes. Journal of Proteomics, v. 174, n. July 2017, p. 28–35, 2018.

GROSE, J. H. et al. Assimilation of nicotinamide mononucleotide requires periplasmic

104

AphA phosphatase in Salmonella enterica. Journal of bacteriology, v. 187, n. 13, p. 4521–30, 2005.

GROSE, J. H.; BERGTHORSSON, U.; ROTH, J. R. Regulation of NAD synthesis by the trifunctional NadR protein of Salmonella enterica. Journal of Bacteriology, v. 187, n. 8, p. 2774–2782, 2005.

GROSS, J. W.; RAJAVEL, M.; GRUBMEYER, C. Kinetic Mechanism of Nicotinic Acid Phosphoribosyltransferase: Implications for Energy Coupling † . Biochemistry, v. 37, n. 12, p. 4189–4199, mar. 1998.

GU, L. et al. Functional Characterization of Sirtuin-like Protein in Mycobacterium smegmatis. Journal of Proteome Research, v. 14, n. 11, p. 4441–4449, 6 nov. 2015.

HACHISUKA, S. ICHI; SATO, T.; ATOMI, H. Hyperthermophilic archaeon Thermococcus kodakarensis utilizes a four-step pathway for NAD+ salvage through nicotinamide deamination. Journal of Bacteriology, v. 200, n. 11, 1 jun. 2018.

HALD, E.; LEHMANN, P.; ZIEGENHORN, J. Molar Absorptivities of β-NADH and β- NAD at 260 nm. Clinical Chemistry, v. 21, n. 7, p. 884–887, 1 jun. 1975.

HALLOWS, W. C.; LEE, S.; DENU, J. M. Sirtuins deacetylate and activate mammalian acetyl-CoA synthetases. [s.l: s.n.]. Disponível em: . Acesso em: 16 abr. 2020.

HAUF, W. et al. Interaction of the Nitrogen Regulatory Protein GlnB (PII) with Biotin Carboxyl Carrier Protein (BCCP) Controls Acetyl-CoA Levels in the Cyanobacterium Synechocystis sp. PCC 6803. Frontiers in Microbiology, v. 7, n. OCT, p. 1700, 26 out. 2016.

HOLM, A. K. et al. Metabolic and transcriptional response to cofactor perturbations in Escherichia coli. Journal of Biological Chemistry, 2010.

HOU, J. et al. Metabolic impact of redox cofactor perturbations in Saccharomyces cerevisiae. Metabolic Engineering, v. 11, n. 4–5, p. 253–261, 1 jul. 2009.

HOUTKOOPER, R. H. et al. The secret life of NAD+: An old metabolite controlling new metabolic signaling pathways. Endocrine Reviews, v. 31, n. 2, p. 194–223, 2010.

HUANG, H. et al. Electron bifurcation involved in the energy metabolism of the

105 acetogenic bacterium Moorella thermoacetica growing on glucose or H2 plus CO2. Journal of Bacteriology, v. 194, n. 14, p. 3689–3699, jul. 2012.

HUANG, N. et al. Bifunctional NMN Adenylyltransferase/ADP-Ribose Pyrophosphatase: Structure and Function in Bacterial NAD Metabolism. Structure, v. 16, n. 2, p. 196–209, 12 fev. 2008.

HUERGO, L. F. et al. ADP-ribosylation of dinitrogenase reductase in Azospirillum brasilense is regulated by AmtB-dependent membrane sequestration of DraG. Molecular Microbiology, v. 59, n. 1, 2006a.

HUERGO, L. F. et al. ADP-ribosylation of dinitrogenase reductase in Azospirillum brasilense is regulated by AmtB-dependent membrane sequestration of DraG. Molecular microbiology, v. 59, n. 1, p. 326–37, jan. 2006b.

HUERGO, L. F. et al. Interactions between PII proteins and the nitrogenase regulatory enzymes DraT and DraG in Azospirillum brasilense. FEBS letters, v. 580, n. 22, p. 5232–6, 2 out. 2006c.

HUERGO, L. F. et al. P II signal transduction proteins: Pivotal players in post- translational control of nitrogenase activity. Microbiology, v. 158, n. 1, 2012.

HUERGO, L. F.; CHANDRA, G.; MERRICK, M. PII signal transduction proteins: Nitrogen regulation and beyond. FEMS Microbiology Reviews, v. 37, n. 2, 2013a.

HUERGO, L. F.; CHANDRA, G.; MERRICK, M. P(II) signal transduction proteins: nitrogen regulation and beyond. FEMS microbiology reviews, v. 37, n. 2, p. 251–83, mar. 2013b.

HUERGO, L. F.; DIXON, R. The Emergence of 2-Oxoglutarate as a Master Regulator Metabolite. Microbiology and Molecular Biology Reviews, v. 79, n. 4, p. 419–435, dez. 2015.

HURLEY, J. K. et al. Structure-function relationships in Anabaena ferredoxin/ferredoxin:NADP+ reductase electron transfer: Insights from site- directed mutagenesis, transient absorption spectroscopy and X-ray crystallographyBiochimica et Biophysica Acta - BioenergeticsElsevier, , 22 abr. 2002.

106

ISHIKAWA, Y.; KAWAI-YAMADA, M. Physiological significance of NAD kinases in cyanobacteria. Frontiers in Plant Science, v. 10, n. June, p. 1–8, 2019.

IWAMOTO, Y. et al. Tryptophan 2,3-dioxygenase in Saccharomyces cerevisiae. Canadian Journal of Microbiology, v. 41, n. 1, p. 19–26, 1 jan. 1995.

JANKEVICIUS, G. et al. The Toxin-Antitoxin System DarTG Catalyzes Reversible ADP-Ribosylation of DNA. Molecular Cell, v. 64, n. 6, p. 1109–1116, 15 dez. 2016.

JÄSCHKE, A. et al. Cap-like structures in bacterial RNA and epitranscriptomic modification. Current Opinion in Microbiology, v. 30, p. 44–49, 2016.

JAUCH, R. et al. Structures of Escherichia coli NAD synthetase with substrates and products reveal mechanistic rearrangements. The Journal of biological chemistry, v. 280, n. 15, p. 15131–40, 15 abr. 2005.

JAVELLE, A; MERRICK, M. Complex formation between AmtB and GlnK: an ancestral role in prokaryotic nitrogen control. Biochemical Society transactions, v. 33, p. 170– 172, 2005.

JAVELLE, A. et al. Substrate binding, deprotonation, and selectivity at the periplasmic entrance of the Escherichia coli ammonia channel AmtB. Proceedings of the National Academy of Sciences of the United States of America, v. 105, n. 13, p. 5040–5045, 1 abr. 2008.

JIANG, P.; NINFA, A. J. Escherichia coli PII signal transduction protein controlling nitrogen assimilation acts as a sensor of adenylate energy charge in vitro. Biochemistry, 2007.

JULIUS, C.; YUZENKOVA, Y. Bacterial RNA polymerase caps RNA with various cofactors and cell wall precursors. Nucleic Acids Research, v. 45, n. 14, p. 8282– 8290, 2017.

JULIUS, C.; YUZENKOVA, Y. Metabolic cofactors act as initiating substrates for primase and affect replication primer processing. bioRxiv, p. 741967, 21 ago. 2019.

KACZMAREK, F. S. et al. Cloning and functional characterization of an NAD+- dependent DNA ligase from Staphylococcus aureus. Journal of Bacteriology, v. 183, n. 10, p. 3016–3024, 15 maio 2001.

107

KANG, G. B. et al. CRYSTALLIZATION AND X-RAY ANALYSIS OF NH3- DEPENDENT NAD + SYNTHETASE FROM HELICOBACTER. v. 10, n. 4, p. 418–421, 2003.

KIM, S. et al. Comparison of the effects of NADH- and NADPH-perturbation stresses on the growth of escherichia coli. Current Microbiology, v. 58, n. 2, p. 159–163, 25 fev. 2009.

KLIMOVA, N. et al. Interplay between NAD+ and acetylǦCoA metabolism in ischemia- induced mitochondrial pathophysiology. Biochimica et Biophysica Acta - Molecular Basis of Disease, v. 1865, n. 8, p. 2060–2067, 2019.

KLOFT, N.; RASCH, G.; FORCHHAMMER, K. Protein phosphatase PphA from Synechocystis sp. PCC 6803: The physiological framework of PII-P dephosphorylation. Microbiology, v. 151, n. 4, p. 1275–1283, 1 abr. 2005.

KOBAYASHI, K. et al. Essential Bacillus subtilis genes. Proceedings of the National Academy of Sciences of the United States of America, v. 100, n. 8, p. 4678–4683, 15 abr. 2003.

KONRAD, E. B.; MODRICH, P.; LEHMAN, I. R. Genetic and enzymatic characterization of a conditional lethal mutant of Escherichia coli K12 with a temperature-sensitive DNA ligase. Journal of Molecular Biology, v. 77, n. 4, p. 519– 529, 15 jul. 1973.

KOSONO, S. et al. Changes in the Acetylome and Succinylome of Bacillus subtilis in Response to Carbon Source. PLOS ONE, v. 10, n. 6, p. e0131169, 22 jun. 2015.

KURNASOV, O. et al. NAD Biosynthesis: Identification of the Tryptophan to Quinolinate Pathway in Bacteria. Chemistry and Biology, 2003.

KURNASOV, O. V. et al. Ribosylnicotinamide kinase domain of NadR protein: Identification and implications in NAD biosynthesis. Journal of Bacteriology, 2002.

LARONDE-LEBLANC, N.; RESTO, M.; GERRATANA, B. Regulation of active site coupling in glutamine-dependent NAD(+) synthetase. Nature structural & molecular biology, v. 16, n. 4, p. 421–429, 2009a.

LARONDE-LEBLANC, N.; RESTO, M.; GERRATANA, B. Regulation of active site

108 coupling in glutamine-dependent NAD + synthetase. Nature Structural and Molecular Biology, 2009b.

LASKOSKI, K. et al. In vitro characterization of the NAD+ synthetase NadE1 from Herbaspirillum seropedicae. Archives of Microbiology, 2016.

LAWRENCE, T. J. et al. FAST: FAST Analysis of Sequences Toolbox. Frontiers in Genetics, v. 6, n. MAY, p. 172, 19 maio 2015.

LEHNMAN, I. R. DNA Ligase: Structure, Mechanism, and Function. Science, v. 186, n. 4166, p. 790–797, 29 nov. 1974.

LI, J. et al. A conserved NAD+ binding pocket that regulates protein-protein interactions during aging. Science (New York, N.Y.), v. 355, n. 6331, p. 1312–1317, 24 mar. 2017.

LI, M. et al. Crystal structure of the catalytic domain of Pseudomonas exotoxin A complexed with a nicotinamide adenine dinucleotide analog: Implications for the activation process and for ADP ribosylation. Proceedings of the National Academy of Sciences of the United States of America, v. 93, n. 14, p. 6902–6906, 9 jul. 1996.

LIN, H. Nicotinamide adenine dinucleotide: Beyond a redox coenzyme. Organic and Biomolecular Chemistry, 2007.

LIU, W. et al. Protein acetylation mediated by YfiQ and CobB is involved in the virulence and stress response of Yersinia pestis. Infection and Immunity, v. 86, n. 6, 1 jun. 2018.

MACEK, B. et al. Protein post-translational modifications in bacteria. Nature Reviews Microbiology, 2019.

MAGNI, G. et al. Structure and function of nicotinamide mononucleotide adenylyltransferase. Current medicinal chemistry, v. 11, n. 7, p. 873–885, 2004.

MCDONALD, H. M. et al. Structural adaptation of an interacting non-native C-terminal helical extension revealed in the crystal structure of NAD+ synthetase from Bacillus anthracis. Acta crystallographica. Section D, Biological crystallography, v. 63, n. Pt 8, p. 891–905, 2007.

MERRICK, M. Post-translational modification of PII signal transduction proteins. Frontiers in Microbiology, v. 5, p. 763, jan. 2015.

109

MORALES-FILLOY, H. G. et al. The 5′-NAD cap of RNAIII modulates toxin production in Staphylococcus aureus isolates. . Journal of Bacteriology, v. 202, n. 6, 23 dez. 2019.

MOURE, V. R. et al. Heat stability of Proteobacterial P II protein facilitate purification using a single chromatography step. Protein Expression and Purification, v. 81, n. 1, 2012a.

MOURE, V. R. et al. Heat stability of Proteobacterial PII protein facilitate purification using a single chromatography step. Protein expression and purification, v. 81, n. 1, p. 83–8, jan. 2012b.

MOURE, V. R. et al. The nitrogenase regulatory enzyme dinitrogenase reductase adpribosyltransferase (DraT) is activated by direct interaction with the signal transduction protein glnb. Journal of Bacteriology, v. 195, n. 2, 2013.

MOURE, V. R. et al. Regulation of nitrogenase by reversible mono-ADP-ribosylation. In: Current Topics in Microbiology and Immunology. [s.l: s.n.].

MOURE, V. R. et al. The ammonium transporter AmtB and the PII signal transduction protein GlnZ are required to inhibit DraG in Azospirillum brasilense. The FEBS Journal, v. 286, n. 6, p. 1214–1229, 12 mar. 2019.

MUNNUR, D. et al. Reversible ADP-ribosylation of RNA. Nucleic Acids Research, v. 47, n. 11, p. 5658–5669, 2019.

NINFA, A. J.; JIANG, P. PII signal transduction proteins: sensors of alpha-ketoglutarate that regulate nitrogen metabolism. Current opinion in microbiology, v. 8, n. 2, p. 168–73, abr. 2005.

OLIVEIRA, M. A. S. et al. 2-Oxoglutarate levels control adenosine nucleotide binding by Herbaspirillum seropedicae PII proteins. FEBS Journal, v. 282, n. 24, 2015.

OSTERMAN, A. Biogenesis and Homeostasis of Nicotinamide Adenine Dinucleotide Cofactor. EcoSal Plus, v. 3, n. 2, 2009.

PACE, H. C.; BRENNER, C. The nitrilase superfamily: classification, structure and function. Genome biology, v. 2, n. 1, p. REVIEWS0001, jan. 2001.

PALANCA, C.; RUBIO, V. Effects of T-loop modification on the PII-signalling protein:

110 structure of uridylylated Escherichia coli GlnB bound to ATP. Environmental Microbiology Reports, v. 9, n. 3, p. 290–299, 1 jun. 2017.

PALAZZO, L.; MIKOČ, A.; AHEL, I. ADP-ribosylation: new facets of an ancient modificationFEBS JournalBlackwell Publishing Ltd, , 1 set. 2017. Disponível em: . Acesso em: 14 mar. 2020

PANOZZO, C. et al. Aerobic and anaerobic NAD + metabolism in Saccharomyces cerevisiae. FEBS Letters, v. 517, n. 1–3, p. 97–102, 24 abr. 2002.

PARK, U. E. et al. DNA ligase and the pyridine nucleotide cycle in Salmonella typhimurium. Journal of Bacteriology, v. 171, n. 4, p. 2173–2180, 1 abr. 1989.

PARK, Y. W.; YEO, H. K.; LEE, J. Y. Structural Analysis of the NH 3 -dependent NAD + Synthetase from Deinococcus radiodurans. v. 35, n. 9, p. 2863–2866, 2014.

PEDROSA, F. DE O.; YATES, M. G. Regulation of nitrogen fixation ( nif ) genes of Azospirillum brasilense by nifA and ntr ( gln ) type gene products. FEMS Microbiology Letters, v. 23, p. 95–101, 1984.

PERINA, D. et al. Distribution of protein poly(ADP-ribosyl)ation systems across all domains of life. DNA Repair, v. 23, p. 4–16, 1 nov. 2014.

PETIT, M.-A.; DUSKO EHRLICH, S. The NAD-dependent ligase encoded by yerG is an essential gene of Bacillus subtilisNucleic Acids Research. [s.l: s.n.]. Disponível em: . Acesso em: 17 mar. 2020.

PLUMBRIDGE, J.; DEUTSCHER, J. Regulation of Carbon Assimilation in Bacteria. [s.l.] Elsevier Inc., 2019.

POLLAK, N.; OLLE, C. D. ¨; ZIEGLER, M. The power to reduce: pyridine nucleotides- small molecules with a multitude of functions. Biochem. J, v. 402, p. 205–218, 2007.

PREISS, J.; HANDLER, P. Biosynthesis of diphosphopyridine nucleotide. Journal of Biological Chemistry, 1958.

RAFFAELLI, N. et al. The Escherichia coli NadR Regulator Is Endowed with Nicotinamide Mononucleotide Adenylyltransferase Activity The Escherichia coli NadR Regulator Is Endowed with Nicotinamide Mononucleotide Adenylyltransferase Activity.

111

Jornal of Bacteriology, v. 181, n. 17, p. 1–4, 1999a.

RAFFAELLI, N. et al. Synechocystis sp. slr0787 protein is a novel bifunctional enzyme endowed with both nicotinamide mononucleotide adenylyltransferase and ‘Nudix’’ hydrolase activities’. FEBS Letters, v. 444, n. 2–3, p. 222–226, 12 fev. 1999b.

RAJENDRAN, C. et al. Crystal structure of the GlnZ-DraG complex reveals a different form of P II -target interaction. Proceedings of the National Academy of Sciences, 2011.

RESTO, M.; YAFFE, J.; GERRATANA, B. An ancestral glutamine-dependent NAD(+) synthetase revealed by poor kinetic synergism. Biochimica et biophysica acta, v. 1794, n. 11, p. 1648–53, nov. 2009.

RIPPKA, R.; DERUELLES, J.; WATERBURY, J. B. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology, v. 111, n. 1, p. 1–61, 1979.

RIZZI, M. et al. Crystal structure of NH3-dependent NAD+ synthetase from Bacillus subtilis. The EMBO journal, v. 15, n. 19, p. 5125–5134, 1996.

ROCCO, C. J. et al. Construction and use of new cloning vectors for the rapid isolation of recombinant proteins from Escherichia coli. Plasmid, v. 59, n. 3, p. 231–237, maio 2008.

RODIONOV, D. A. et al. Transcriptional regulation of NAD metabolism in bacteria: NrtR family of Nudix-related regulators. Nucleic Acids Research, v. 36, n. 6, p. 2047–2059, 2008a.

RODIONOV, D. A. et al. Transcriptional regulation of NAD metabolism in bacteria: genomic reconstruction of NiaR (YrxA) regulon. Nucleic Acids Research, v. 36, n. 6, p. 2032–2046, abr. 2008b.

RODIONOV, D. A et al. Transcriptional regulation of NAD metabolism in bacteria: NrtR family of Nudix-related regulators. Nucleic acids research, v. 36, n. 6, p. 2047–59,

112 abr. 2008c.

RODIONOVA, I. A. et al. Metabolic and bactericidal effects of targeted suppression of NadD and NadE enzymes in mycobacteria. mBio, 2014.

RODRIGUES, T. E. et al. Search for novel targets of the PII signal transduction protein in Bacteria identifies the BCCP component of acetyl-CoA carboxylase as a PII binding partner. Molecular microbiology, v. 91, n. 4, p. 751–61, mar. 2014.

RODRIGUES, T. E. et al. Fatty acid biosynthesis is enhanced in Escherichia coli strains with deletion in genes encoding the PII signaling proteins. Archives of Microbiology, 2019.

RUPPERT, U. et al. The novel protein phosphatase PphA from Synechocystis PCC 6803 controls dephosphorylation of the signalling protein PII. Molecular Microbiology, v. 44, n. 3, p. 855–864, 7 maio 2002.

SANT’ANNA, F. H. et al. The PII superfamily revised: a novel group and evolutionary insights. Journal of molecular evolution, v. 68, n. 4, p. 322–36, abr. 2009.

SANTOS, A. et al. NAD+ biosynthesis in bacteria is controlled by global carbon/nitrogen levels via PII signaling. v. 295, n. 9, p. 1–12, 2020.

SASSETTI, C. M.; BOYD, D. H.; RUBIN, E. J. Genes required for mycobacterial growth defined by high density mutagenesis. Molecular Microbiology, v. 48, n. 1, p. 77–84, 25 mar. 2003.

SCHAFER, F. Q.; BUETTNER, G. R. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione coupleFree Radical Biology and Medicine, 1 jun. 2001. Disponível em: . Acesso em: 12 abr. 2020

SHARMA, V.; GRUBMEYER, C.; SACCHETTINI, J. C. Crystal structure of quinolinic acid phosphoribosyltransferase from Mycobacterium tuberculosis: A potential TB drug

113 target. Structure, v. 6, n. 12, p. 1587–1599, 15 dez. 1998.

SHEN, W. et al. SeqKit: A Cross-Platform and Ultrafast Toolkit for FASTA/Q File Manipulation. PLOS ONE, v. 11, n. 10, p. e0163962, 5 out. 2016.

SORCI, L. et al. Nicotinamide mononucleotide synthetase is the key enzyme for an alternative route of NAD biosynthesis in Francisella tularensis. Proceedings of the National Academy of Sciences of the United States of America, v. 106, n. 9, p. 3083–8, 3 mar. 2009.

SORCI, L. et al. Characterization of bacterial NMN deamidase as a Ser/Lys hydrolase expands diversity of serine amidohydrolases. FEBS Letters, v. 588, n. 6, p. 1016– 1023, 18 mar. 2014.

SORCI, L.; RUGGIERI, S.; RAFFAELLI, N. NAD homeostasis in the bacterial response to DNA/RNA damage. DNA repair, 12 ago. 2014.

SPÄT, P.; MACEK, B.; FORCHHAMMER, K. Phosphoproteome of the cyanobacterium Synechocystis sp. PCC 6803 and its dynamics during nitrogen starvation. Frontiers in Microbiology, v. 6, n. MAR, p. 248, 31 mar. 2015.

SRISKANDA, V. et al. Mutational analysis of Escherichia coli DNA ligase identifies amino acids required for nick-ligation in vitro and for in vivo complementation of the growth of yeast cells deleted for CDC9 and LIG4Nucleic Acids Research. [s.l: s.n.].

SRISKANDA, V. A second NAD+-dependent DNA ligase (LigB) in Escherichia coli. Nucleic Acids Research, v. 29, n. 24, p. 4930–4934, 2001.

SRISKANDA, V.; MOYER, R. W.; SHUMAN, S. NAD+-dependent DNA Ligase Encoded by a Eukaryotic Virus. Journal of Biological Chemistry, v. 276, n. 39, p. 36100–36109, 28 set. 2001.

SRISKANDA, V.; SHUMAN, S. Conserved residues in domain Ia are required for the reaction of Escherichia coli DNA ligase with NAD+. Journal of Biological Chemistry, v. 277, n. 12, p. 9695–9700, 22 mar. 2002.

STARAI, V. J. et al. Sir2-dependent activation of acetyl-CoA synthetase by deacetylation of active lysine. Science, v. 298, n. 5602, p. 2390–2392, 20 dez. 2002.

114

STARAI, V. J. et al. Short-Chain Fatty Acid Activation by Acyl-Coenzyme A Synthetases Requires SIR2 Protein Function in Salmonella enterica and Saccharomyces cerevisiae. [s.l: s.n.].

STRØMLAND, Ø. et al. Keeping the balance in NAD metabolism. Biochemical Society Transactions, n. December 2018, p. BST20180417, 2019.

SYMERSKY, J. et al. NH 3 -dependent NAD + synthetase from Bacillus subtilis at 1 Å resolution. Acta Crystallographica Section D Biological Crystallography, v. 58, n. 7, p. 1138–1146, 2002a.

SYMERSKY, J. et al. NH3-dependent NAD+ synthetase from Bacillus subtilis at 1 Å resolution. Acta Crystallographica Section D: Biological Crystallography, v. 58, n. 7, p. 1138–1146, 1 jul. 2002b.

TAUTENHAHN, R. et al. XCMS online: A web-based platform to process untargeted metabolomic data. Analytical Chemistry, v. 84, n. 11, p. 5035–5039, 5 jun. 2012.

TEDESCHI, G. et al. Structural characterization of l-aspartate oxidase and identification of an interdomain loop by limited proteolysis. European Journal of Biochemistry, v. 260, n. 3, p. 896–903, 5 mar. 1999.

TERNES, C. M.; SCHÖNKNECHT, G. Gene transfers shaped the evolution of de novo NAD+ biosynthesis in eukaryotes. Genome Biology and Evolution, v. 6, n. 9, p. 2335–2349, 2014.

TRUAN, D. et al. A New PII Protein Structure Identifies the 2-Oxoglutarate Binding Site. Journal of Molecular Biology, v. 400, n. 3, 2010.

VAN DEN BERG, S. et al. Improved solubility of TEV protease by directed evolution. Journal of Biotechnology, v. 121, n. 3, p. 291–298, 2006.

VAN HEESWIJK, W. C.; WESTERHOFF, H. V; BOOGERD, F. C. Nitrogen assimilation in Escherichia coli: putting molecular data into a systems perspective. Microbiology and molecular biology reviews : MMBR, v. 77, n. 4, p. 628–95, 1 dez. 2013.

VASSILOPOULOS, A.; WANG, R.-H.; GIUS, D. Mammalian Sirtuins, Cellular Energy Regulation, and Metabolism, and Carcinogenesis. [s.l.] Elsevier Inc., 2018.

VVEDENSKAYA, I. O. et al. CapZyme-Seq Comprehensively Defines Promoter-

115

Sequence Determinants for RNA 5′ Capping with NAD+. Molecular Cell, v. 70, p. 553–564, 2018.

WALCZAK, K. et al. Kynurenic acid and cancer: facts and controversiesCellular and Molecular Life SciencesBirkhauser Verlag AG, , 28 out. 2019.

WANG, H. et al. Energy shifts induce membrane sequestration of DraG in Rhodospirillum rubrum independent of the ammonium transporters and diazotrophic conditions. FEMS Microbiology Letters, v. 365, n. 16, 1 ago. 2018.

WANG, M. M.; YOU, D.; YE, B. C. Site-specific and kinetic characterization of enzymatic and nonenzymatic protein acetylation in bacteria. Scientific Reports, v. 7, n. 1, 1 dez. 2017.

WEINERT, B. T. et al. Acetyl-Phosphate is a critical determinant of Lysine Acetylation in E.coli. Molecular Cell, v. 51, n. 2, p. 265–272, 2013.

WILKINSON, A.; DAY, J.; BOWATER, R. Bacterial DNA ligases. Molecular Microbiology, v. 40, n. 6, p. 1241–1248, 1 jun. 2001.

WINZ, M. L. et al. Capture and sequencing of NAD-capped RNA sequences with NAD captureSeq. Nature Protocols, v. 12, n. 1, p. 122–149, 1 jan. 2017.

WU, C. H. PIRSF: family classification system at the Protein Information Resource. Nucleic Acids Research, 2004.

YANG, T.; SAUVE, A. A. NAD metabolism and sirtuins: Metabolic regulation of protein deacetylation in stress and toxicity. AAPS Journal, v. 8, n. 4, p. E632–E643, 6 out. 2006.

YANG, Z. et al. Aspartate dehydrogenase, a novel enzyme identified from structural and functional studies of TM1643. Journal of Biological Chemistry, v. 278, n. 10, p. 8804–8808, 7 mar. 2003.

YUAN, J. et al. Metabolomics-driven quantitative analysis of ammonia assimilation in E. coli. Molecular Systems Biology, 2009.

ZETH, K.; FOKINAS, O.; FORCHHAMMER, K. Structural basis and target-specific modulation of ADP sensing by the Synechococcus elongatus PII signaling protein. Journal of Biological Chemistry, v. 289, n. 13, p. 8960–8972, 2014.

116

ZHANG, J. et al. Lysine acetylation is a highly abundant and evolutionarily conserved modification in Escherichia coli. Molecular and Cellular Proteomics, v. 8, n. 2, p. 215–225, fev. 2009.

ZHANG, Y. et al. Crystal structure and mechanism of tryptophan 2,3-dioxygenase, a heme enzyme involved in tryptophan catabolism and in quinolinate biosynthesis. Biochemistry, v. 46, n. 1, p. 145–155, 9 jan. 2007.

ZHANG, Y. et al. Mutagenesis and functional characterization of the four domains of GlnD, a bifunctional nitrogen sensor protein. Journal of Bacteriology, v. 192, n. 11, p. 2711–2721, jun. 2010.

ZHANG, Z.; MILIAS-ARGEITIS, A.; HEINEMANN, M. Dynamic single-cell NAD(P)H measurement reveals oscillatory metabolism throughout the E. coli cell division cycle. Scientific Reports, 2018.

ZHAO, K.; CHAI, X.; MARMORSTEIN, R. Structure and Substrate Binding Properties of cobB, a Sir2 Homolog Protein Deacetylase from Escherichia coli. Journal of Molecular Biology, v. 337, n. 3, p. 731–741, 26 mar. 2004.

117

Supporting Information

Table S1 Microsoft Excel Worksheet file (attached) Table S2 Bacterial strains and plasmids Strain/plasmid Genotype/phenotype Source/reference

Synechocystis sp.

(RIPPKA; DERUELLES; PCC 6803 Wild-type WATERBURY, 1979) A. brasilense

FP2 Wild-type (PEDROSA; YATES, 1984)

LFH3 Nalr, ΔglnB Nif− (HUERGO et al., 2006c) 7611 Nif+ glnZ ::Ω, Spr Smr (DE ZAMAROCZY, 1998) glnB::kan/ glnZ:: Ω , Smr 2812 (DE ZAMAROCZY, 1998) Kmr E . coli DH10B Smr ; F’ [proAB+ lacZM15] Invitrogen F- ompT hsdSB(rB- mB-) Rosetta (DE) pLysS gal dcm (DE3) Novagen pLysSRARE Expresses T7 RNA Lemo21 (DE3) New England Labs Inc. polymerase Plasmid Characteristics Reference pET28a Kmr Expression vector Agilent pET29a Kmr Expression vector Agilent Ampr Expression vector, pTEV5 His tag followed by TEV (ROCCO et al., 2008) cleavage site Kmr. Express H. seropedicae NadE2 with a pASnade2 (SANTOS et al., 2018) N-terminal 6x His tag in pET28a Kmr. Express native A. pLHnade2 brasilense NadE2 in (SANTOS et al., 2018) pET29a

118

Ampr. Express Synechocystis sp. strain pASnadESc PCC 6803 NadE with a N- This work terminal 6x His tag in pTEV5 Kmr. Express native A. pMSA4∆42-54 brasilense GlnZ∆42-54 in (RAJENDRAN et al., 2011) pET29a Kmr. Express A. brasilense GlnZ∆42-54 with a N- pGlnZ∆Tloop This work terminal 6x His tag in pET28a Kmr. Express A.brasilense pMSA3 GlnZ with a N-terminal 6x (ARAUJO et al., 2004) His tag in pET28a Kmr. Express native A. pMSA4 (MOURE et al., 2012b) brasilense GlnZ in pET29a Kmr. Express A.brasilense pLH25 GlnB with a N-terminal 6x (HUERGO et al., 2006b) His tag in pET28a Ampr. Express A. brasilense GlnD with a N- pASGlnDAb This work terminal 6x His tag in pTEV5

Source: SANTOS et al.,(2020) JBC.

119

Supporting Figures

Figure S1 Putative binding partners of the A. brasilense PII protein GlnZ. A) A 1 ml Ni-NTA column charged with His-tagged GlnZ protein was loaded with A. brasilense protein extracts in the presence of MgATP. After extensive washes, proteins were eluted with buffer containing MgATP and 1.5 mM of 2-OG and separated in 1.5 ml fractions (fractions 1 and 2). The amount of the proteins in each fraction was compared to the amount of proteins eluted from a control column, under the same conditions, by quantitative label free LC/MS/MS in technical triplicates. The graphic shows the log protein enrichment of the signal of each protein eluted the His-GlnZ column / control column in their respective 1.5 mL fractions. B) Details of the top 5 enriched proteins that were eluted from the His-GlnZ affinity column.

Source: SANTOS et al.,(2020) JBC.

120

Figure S2: Determination of the dissociation constant of the His-GlnZ and AbNadE2 complex in the presence of ADP by Bio-layer Interferometry. A) The His-GlnZ was immobilized in the Ni-NTA sensor tip in a concentration of 2 μg/mL until saturation. The tip with His-GlnZ was then challenged in solution containing the indicated AbNadE2 concentrations in the presence of 1mM ADP. B) Plot reporting the 'λ spectral shift in nm vs AbNadE2 concentration. The binding was measured in an Octet K2 (FortéBIO) and the Kd determined using the manufacturer´s software.

Source: SANTOS et al.,(2020) JBC.

121

Figure S3: The NadEGln type 2 are activated in reductive environment when using L-glutamine as N-donor. NadEGln type 2 activity was continuously measured by determining NADH formation in assays coupled with alcohol dehydrogenase in the presence or absence of 10 mM Dithiothreitol (DTT) using 2 mM of L- glutamine as a N-donor (A) or 10 mM of ammonium as a N-donor (B). Proteins from A. brasilense (AbNadE2), Herbaspirillum seropedicae (HsNadE2) and Synechocystis sp. (ScNadE2) were assayed.

Source: SANTOS et al.,(2020) JBC.

122

Figure S4: Effect of NAD+ in the kinetic parameters of AbNadE2 reaction. A) Determination of initial velocities accordingly to varying NaAD concentrations in the presence and + -1 absence of 1mM NAD . A NaAD KM of 0.08 mM and Vmax of 0.09 μmol.s were obtained in the absence + + -1 of NAD . In the presence of NAD the values were KM 0.09 mM and Vmax 0.04 μmol.s . B) Determination of initial velocities accordingly to varying L-glutamine concentrations in the presence and + -1 absence of 1 mM NAD . A L-glutamine KM of 0.72 mM, and Vmax of 0.15 μmol.s were obtained in the + + -1 absence of NAD . In the presence of NAD the values were KM 0.92 mM and Vmax 0.028 μmol.s .

Source: SANTOS et al.,(2020) JBC.

123

Figure S5: In vitro formation of the NadE2-GlnZ complex does not require the GlnZ T loop and AbNadE2 preferentially interacts with the PII protein GlnZ. A) Complex formation was assessed by pull-down using Ni2+ beads. Reactions were performed in the presence of the ATP and 2-OG at 1mM, as indicated. Binding reactions were conducted in 400 μl of buffer (50 mM Tris HCl; 100 mM NaCl; glycerol 10%; 20 mM Imidazole; 0,05% Tween 20; 5 mM MgCl2). The fractions eluted from the Ni2+ beads were analyzed by SDS-PAGE and the gel was stained with Coomassie Blue. Lanes 1 and 4, HisGlnZ only. Lanes 2 and 5, AbNadE2 only. Lanes 3 and 6, mixture of HisGlnZ and AbNadE2. B) Complex formation was assessed by pull-down using Ni2+ beads. Reactions were performed under fixed concentration ATP 1 mM. Binding reactions were conducted in 400 μl of buffer (50 mM Tris HCl; 100 mM NaCl; glycerol 10%; 20 mM Imidazole; 0,05% Tween 20; 5 2+ mM MgCl2). The fractions eluted from the Ni beads were analyzed by SDS-PAGE 15% and the gel was stained with Coomassie Blue. Lanes 1 and 4, His-PII only (GlnB or GlnZ as indicated). Lanes 2 and 5, AnNadE2 only. Lanes 3 and 6, mixture of His-PII and AbNadE2.

Source: SANTOS et al.,(2020) JBC.

124

Front page of the published article in JBC (2020)

125

4. CONCLUSION x Azospirillum brasilense has two isoforms of NadE, one similar to the octameric NadE from Mycobacterium tuberculosis, named here NadE1. The second has a distinct sequence from the octameric NadE group, hypothesized as a novel type of NadE. x The novel NadE was characterized and named NadE2. This enzyme, differently from the previous described octameric NadEGln, present dimeric structure and use L-glutamine in vitro. x Differently from the octameric NadEGln, NadE2 has a low preference for L- glutamine over ammonia in competition substrate assay. x Analysis of the NadE2 from Burkholderia thailandensis showed that NadE2 presents a 60 Å (1 Å=0.1 nm) U-shaped internal tunnel, connecting the glutaminase domain to the synthetase site. Differently from the octameric enzyme from Mycobacterium tuberculosis, B. thailandensis presents an empty site located within the interdomain interface, enables four possible pathways for ammonia translocation. x Together with the L-glutamine concentration measurement, we propose a model where the two isoforms of NadE are active in different metabolic moments, responding to shifts in L-glutamine levels. x Functional analysis, together with structural comparison allowed us to propose a model where NadE type 2 is an evolutionary intermediate between the NadENH3 and the octameric NadEGln. PII proteins and NadEGln from type 2 interact in vitro in the absence of the 2-oxoglutarate metabolite. x Kinetic analyses showed that NadE2 activity is controlled by the redox environment and is inhibited by physiological concentrations of its product, NAD+. x Other forms of the NAD(H)(P) pool are not inhibitors of the NadE2. x The formation of the NadE2-PII complex prevents the feed-back inhibition of the NAD+. x In A. brasilense GlnZ rather than GlnB is more effective in the effect of relieving the feed-back inhibition.

126 x In a condition where carbon increase and nitrogen decrease, the accumulation of 2-oxoglutarate prevents the protein-protein complex formation, although its capacity to dissociate a pre-formed complex is rather limited. x The uridylylation of GlnZ prevents the complex formation. x Most of the NadE sequences that are in the PII genomic context are from the NadEGln and dimeric. x Overall, the data was used to formulate a mechanism of the PII control over NAD- metabolism. When nitrogen is limited, and cell growth is arrested, the intracellular levels of 2-OG are high, and glutamine are low. PII is fully uridylylated and does not interact with NadE2, resulting in strong feedback negative inhibition of NadE2 by NAD+. Once ammonium becomes available, glutamine rises and 2-OG drops, PII is deuridylylated (reaction catalyzed by the GlnD enzyme in response to high glutamine) and bound to MgATP and/or ADP. Under this condition, PII interacts with NadE2 relieving NAD+ inhibition and, thus, allowing higher NAD+ production rates to feed the demands of active cell growth.

127

5. REFERENCES

ABOUELFETOUH, A. et al. The E. coli sirtuin CobB shows no preference for enzymatic and nonenzymatic lysine acetylation substrate sites. MicrobiologyOpen, v. 4, n. 1, p. 66–83, 1 fev. 2015. AFZAL, M.; KUIPERS, O. P.; SHAFEEQ, S. Niacin-mediated gene expression and role of NiaR as a transcriptional repressor of niaX, nadC, and pnuC in Streptococcus pneumoniae. Frontiers in Cellular and Infection Microbiology, 2017. AKENTIEVA, N. Posttranslational modification of dinitrogenase reductase in Rhodospirillum rubrum treated with fluoroacetate. World Journal of Microbiology and Biotechnology, v. 34, n. 12, p. 1–10, 1 dez. 2018. ALOMARI, A. Biophysical and Kinetic Analysis of Escherichia coli DNA Ligase Activity and Inhibition. [s.l.] University of Portsmouth, 2018. ARAÚJO, L. M. et al. Different responses of the GlnB and GlnZ proteins upon in vitro uridylylation by the Azospirillum brasilense GlnD protein. Brazilian Journal of Medical and Biological Research, v. 41, n. 4, p. 289–294, 2008. ARAUJO, M. S. et al. In vitro uridylylation of the Azospirillum brasilense N-signal transducing GlnZ protein. Protein Expression and Purification, v. 33, n. 1, p. 19–24, 2004. ARAVIND, L. et al. The natural history of ADP-ribosyltransferases and the ADP- ribosylation system. Current topics in microbiology and immunology, v. 384, p. 3– 32, 2015. ARCONDÉGUY, T.; JACK, R.; MERRICK, M. P(II) signal transduction proteins, pivotal players in microbial nitrogen control. Microbiology and molecular biology reviews : MMBR, v. 65, n. 1, p. 80–105, mar. 2001. BAEZA, J. et al. Stoichiometry of site-specific lysine acetylation in an entire proteome. Journal of Biological Chemistry, v. 289, n. 31, p. 21326–21338, 1 ago. 2014. BATISTA, M. B.; DIXON, R. Manipulating nitrogen regulation in diazotrophic bacteria for agronomic benefitBiochemical Society TransactionsPortland Press Ltd, , 30 abr. 2019. BAUGH, L. et al. Combining Functional and Structural Genomics to Sample the Essential Burkholderia Structome. PLoS ONE, v. 8, n. 1, 2013. BEGLEY, T. P. et al. Cofactor Biosynthesis. [s.l.] Elsevier, 2001. v. 61 BELLINZONI, M. et al. Heterologous expression, purification, and enzymatic activity of Mycobacterium tuberculosis NAD(+) synthetase. Protein expression and purification, v. 25, n. 3, p. 547–57, ago. 2002. BELLINZONI, M. et al. Glutamine amidotransferase activity of NAD+ synthetase from Mycobacterium tuberculosis depends on an amino-terminal nitrilase domain. Research in microbiology, v. 156, n. 2, p. 173–7, mar. 2005.

128

BENNETT, B. D. et al. Absolute metabolite concentrations and implied enzyme active site occupancy in Escherichia coli. Nature Chemical Biology, v. 5, n. 8, p. 593–599, 2009. BERGMEYER, H. U.; BERNT, E. Lactate Dehydrogenase. In: Methods of Enzymatic Analysis. [s.l.] Elsevier, 1974. p. 574. BHEDA, P. et al. The Substrate Specificity of Sirtuins. Annual Review of Biochemistry, v. 85, n. 1, p. 405–429, 2016. BIRD, J. G. et al. The mechanism of RNA 5′ capping with NAD+, NADH and desphospho-CoA. Nature, v. 535, n. 7612, p. 444–447, 2016. BOCKWOLDT, M. et al. Identification of evolutionary and kinetic drivers of NAD- dependent signaling. Proceedings of the National Academy of Sciences, v. 116, n. 32, p. 15957–15966, 2019. BONATTO, A. C. et al. Uridylylation of Herbaspirillum seropedicae GlnB and GlnK proteins is differentially affected by ATP, ADP and 2-oxoglutarate in vitro. Archives of microbiology, v. 194, n. 8, p. 643–52, ago. 2012. BORRA, M. T. et al. Substrate specificity and kinetic mechanism of the Sir2 family of NAD +-dependent histone/protein deacetylases. Biochemistry, v. 43, n. 30, p. 9877– 9887, 3 ago. 2004. BRENNER, C. Catalysis in the nitrilase superfamily. Current opinion in structural biology, v. 12, n. 6, p. 775–82, dez. 2002. BROSTEDT, E. et al. Electron transport to nitrogenase in Rhodospirillum rubrum: the role of NAD(P)H as electron donor and the effect of fluoroacetate on nitrogenase activity. FEMS Microbiology Letters, v. 150, n. 2, p. 263–267, 17 jan. 1997. BURCKHARDT, R. M.; BUCKNER, B. A.; ESCALANTEǦSEMERENA, J. C. Staphylococcus aureus modulates the activity of acetylǦCoenzyme A synthetase (Acs) by sirtuinǦdependent reversible lysine acetylation . Molecular Microbiology, 2019. CAHOVÁ, H. et al. NAD captureSeq indicates NAD as a bacterial cap for a subset of regulatory RNAs. Nature, v. 519, n. 7543, p. 374–377, 2015. CANELAS, A. B.; VAN GULIK, W. M.; HEIJNEN, J. J. Determination of the cytosolic free NAD/NADH ratio in Saccharomyces cerevisiae under steady-state and highly dynamic conditions. Biotechnology and Bioengineering, v. 100, n. 4, p. 734–743, 1 jul. 2008. CARABETTA, V. J.; CRISTEA, I. M. Regulation, function, and detection of protein acetylation in bacteriaJournal of BacteriologyAmerican Society for Microbiology, , 1 ago. 2017. CARPOUSIS, A. J. The RNA Degradosome of Escherichia coli : An mRNA-Degrading Machine Assembled on RNase E. Annual Review of Microbiology, v. 61, n. 1, p. 71– 87, 26 out. 2007. CASTAÑOǦCEREZO, S. et al. Protein acetylation affects acetate metabolism, motility and acid stress response in Escherichia coli . Molecular Systems Biology, v. 10, n. 11, p. 762, nov. 2014.

129

CHELLAMUTHU, V. R. et al. A widespread glutamine-sensing mechanism in the plant kingdom. Cell, v. 159, n. 5, p. 1188–1199, 2014. CHEN, S.-H.; YU, X. Human DNA ligase IV is able to use NAD + as an alternative adenylation donor for DNA ends ligation. Nucleic Acids Research, v. 47, n. 3, p. 1321–1334, 2018. CHEN, Y. G. et al. LC/MS analysis of cellular RNA reveals NAD-linked RNA. Nature Chemical Biology, v. 5, n. 12, p. 879–881, 2009. CHUENCHOR, W. et al. Regulation of the intersubunit ammonia tunnel in Mycobacterium tuberculosis glutamine-dependent NAD+ synthetase. The Biochemical journal, v. 443, n. 2, p. 417–26, 15 abr. 2012. CHUENCHOR, W. et al. Different ways to transport ammonia in human and Mycobacterium tuberculosis NAD+ synthetases. Nature Communications, v. 11, n. 1, p. 1–12, 1 dez. 2020. CICCHILLO, R. M. et al. Escherichia coli quinolinate synthetase does indeed harbor a [4Fe-4S] cluster. Journal of the American Chemical Society, v. 127, n. 20, p. 7310– 7311, 25 maio 2005. COLLIER, R. J. Understanding the mode of action of diphtheria toxin: A perspective on progress during the 20th centuryToxiconElsevier Ltd, , nov. 2001. Disponível em: . Acesso em: 16 mar. 2020 CORDA, D.; GIROLAMO, M. DI. Functional aspects of protein mono-ADP-ribosylation. The EMBO Journal, v. 22, n. 9, p. 1953–1958, 1 maio 2003. COUTTS, G. et al. Membrane sequestration of the signal transduction protein GlnK by the ammonium transporter AmtB. The EMBO journal, v. 21, n. 4, p. 536–45, fev. 2002. DA ROCHA, R. A. et al. Mathematical model of the binding of allosteric effectors to the escherichia coli PII signal transduction protein GlnB. Biochemistry, v. 52, n. 15, 2013. DANDEKAR, T. et al. Conservation of gene order: A fingerprint of proteins that physically interact. Trends in Biochemical Sciences, 1998. DE INGENIIS, J. et al. Glutamine versus ammonia utilization in the NAD synthetase family. PloS one, v. 7, n. 6, p. e39115, jan. 2012. DE ZAMAROCZY, M. Structural homologues PII and PZ of Azospirillum brasilense provide intracellular signalling for selective regulation of various nitrogen-dependent functions. Molecular Microbiology, v. 29, n. 2, p. 449–463, 1998. DI LUCCIO, E.; WILSON, D. K. Comprehensive X-ray structural studies of the quinolinate phosphoribosyl transferase (BNA6) from Saccharomyces cerevisiae. Biochemistry, v. 47, n. 13, p. 4039–4050, 1 abr. 2008. DIXON, R.; KAHN, D. Genetic regulation of biological nitrogen fixation. Nature reviews. Microbiology, v. 2, n. 8, p. 621–31, ago. 2004. EBERT, B. E. et al. Response of Pseudomonas putida KT2440 to increased NADH and ATP demand. Applied and Environmental Microbiology, v. 77, n. 18, p. 6597– 6605, 15 set. 2011. 130

ESAKOVA, O. A. et al. An Unexpected Species Determined by X-ray Crystallography that May Represent an Intermediate in the Reaction Catalyzed by Quinolinate Synthase. Journal of the American Chemical Society, v. 141, n. 36, p. 14142– 14151, 11 set. 2019. FERIA BOURRELLIER, A. B. et al. Chloroplast acetyl-CoA carboxylase activity is 2- oxoglutarate-regulated by interaction of PII with the biotin carboxyl carrier subunit. v. 107, n. 1, 2010. FERNIE, A. et al. Editorial: NAD Metabolism and Signaling in Plants. Frontiers in Plant Science, v. 11, p. 146, 26 fev. 2020. FLOQUET, N. et al. Ammonia channeling in bacterial glucosamine-6-phosphate synthase (Glms): Molecular dynamics simulations and kinetic studies of protein mutants. FEBS Letters, v. 581, n. 16, p. 2981–2987, 26 jun. 2007. FOKINA, O. et al. Mechanism of 2-oxoglutarate signaling by the Synechococcus elongatus PII signal transduction protein. Proceedings of the National Academy of Sciences, 2010. FORCHHAMMER, K. P(II) signal transducers: novel functional and structural insights. Trends in microbiology, v. 16, n. 2, p. 65–72, mar. 2008. FORCHHAMMER, K.; DE MARSAC, N. T. The P(II) protein in the cyanobacterium Synechococcus sp. strain PCC 7942 is modified by serine phosphorylation and signals the cellular N-status. Journal of Bacteriology, v. 176, n. 1, p. 84–91, 1 jan. 1994. FORCHHAMMER, K.; DE MARSAC, N. T. Phosphorylation of the P(II) protein (glnB gene product) in the cyanobacterium Synechococcus sp. strain PCC 7942: Analysis of in vitro kinase activity. Journal of Bacteriology, v. 177, n. 20, p. 5812–5817, 1 out. 1995. FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the PII signalling protein family. FEBS Journal, v. 283, n. 3, p. 425–437, 2016a. FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the PII signalling protein family. The FEBS journal, v. 283, n. 3, p. 425–37, fev. 2016b.

FORCHHAMMER, K.; LÜDDECKE, J. Sensory properties of the P II signalling protein family. FEBS Journal, v. 283, n. 3, p. 425–437, fev. 2016c. FORCHHAMMER, K.; SELIM, K. A. Carbon/nitrogen homeostasis control in cyanobacteria. FEMS Microbiology Reviews, v. 025, p. 33–53, 2020. FOROUHAR, F. et al. Molecular insights into substrate recognition and catalysis by tryptophan 2,3-dioxygenase. Proceedings of the National Academy of Sciences of the United States of America, v. 104, n. 2, p. 473–478, 9 jan. 2007. FRICKEY, T.; LUPAS, A. CLANS: A Java application for visualizing protein families based on pairwise similarity. Bioinformatics, 2004. FRINDERT, J. et al. Identification, Biosynthesis, and Decapping of NAD-Capped RNAs in B. subtilis. Cell Reports, v. 24, n. 7, p. 1890- 1901.e8, 14 ago. 2018. FRYE, R. A. Phylogenetic classification of prokaryotic and eukaryotic Sir2-like proteins. Biochemical and Biophysical Research Communications, v. 273, n. 2, p. 793–798, 131

5 jul. 2000. GALLEGO-JARA, J. et al. Characterization of CobB kinetics and inhibition by nicotinamide. PLoS ONE, v. 12, n. 12, 1 dez. 2017. GAO, R. et al. A single regulator nrtr controls bacterial NAD+ homeostasis via its acetylation. eLife, 2019. GARDNER, J. G.; ESCALANTE-SEMERENA, J. C. In bacillus subtilis, the sirtuin protein deacetylase, encoded by the srtN gene (formerly yhdZ), and functions encoded by the acuABC genes control the activity of acetyl coenzyme a synthetase. Journal of Bacteriology, v. 191, n. 6, p. 1749–1755, mar. 2009. GAZZANIGA, F. et al. Microbial NAD metabolism: lessons from comparative genomics. Microbiology and molecular biology reviews : MMBR, v. 73, n. 3, p. 529–41, Table of Contents, 2009a. GAZZANIGA, F. et al. Microbial NAD metabolism: lessons from comparative genomics. Microbiology and molecular biology reviews : MMBR, v. 73, n. 3, p. 529–41, Table of Contents, set. 2009b. GEORLETTE, D. et al. Cofactor Binding Modulates the Conformational Stabilities and Unfolding Patterns of NAD+-dependent DNA Ligases from Escherichia coli and Thermus scotoductus. Journal of Biological Chemistry, v. 278, n. 50, p. 49945– 49953, 12 dez. 2003. GERASIMOVA, A. V.; GELFAND, M. S. Evolution of the NadR regulon in enterobacteriaceae. Journal of Bioinformatics and Computational Biology, 2005. GERDES, S. Y. et al. From Genetic Footprinting to Antimicrobial Drug Targets: Examples in Cofactor Biosynthetic Pathways. Journal of Bacteriology, v. 184, n. 16, p. 4555–4572, 15 ago. 2002. GERDES, S. Y. et al. Comparative genomics of NAD biosynthesis in cyanobacteria. Journal of bacteriology, v. 188, n. 8, p. 3012–23, 2006. GERHARDT, E. C. M. et al. Influence of the ADP/ATP ratio, 2-oxoglutarate and divalent ions on Azospirillum brasilense PII protein signalling. Microbiology (Reading, England), v. 158, n. Pt 6, p. 1656–63, jun. 2012. GERHARDT, E. C. M. et al. The Bacterial signal transduction protein GlnB regulates the committed step in fatty acid biosynthesis by acting as a dissociable regulatory subunit of acetyl-CoA carboxylase. Molecular Microbiology, v. 95, n. 6, 2015. GOTTESMAN, M. M.; HICKS, M. L.; GELLERT, M. Genetics and function of DNA ligase in Escherichia coli. Journal of Molecular Biology, v. 77, n. 4, p. 531–547, 15 jul. 1973. GRAVINA, F. et al. Proteome analysis of an Escherichia coli ptsN-null strain under different nitrogen regimes. Journal of Proteomics, v. 174, n. July 2017, p. 28–35, 2018. GROSE, J. H. et al. Assimilation of nicotinamide mononucleotide requires periplasmic AphA phosphatase in Salmonella enterica. Journal of bacteriology, v. 187, n. 13, p. 4521–30, 2005.

132

GROSE, J. H.; BERGTHORSSON, U.; ROTH, J. R. Regulation of NAD synthesis by the trifunctional NadR protein of Salmonella enterica. Journal of Bacteriology, v. 187, n. 8, p. 2774–2782, 2005. GROSS, J. W.; RAJAVEL, M.; GRUBMEYER, C. Kinetic Mechanism of Nicotinic Acid Phosphoribosyltransferase: Implications for Energy Coupling † . Biochemistry, v. 37, n. 12, p. 4189–4199, mar. 1998. GU, L. et al. Functional Characterization of Sirtuin-like Protein in Mycobacterium smegmatis. Journal of Proteome Research, v. 14, n. 11, p. 4441–4449, 6 nov. 2015. HACHISUKA, S. ICHI; SATO, T.; ATOMI, H. Hyperthermophilic archaeon Thermococcus kodakarensis utilizes a four-step pathway for NAD+ salvage through nicotinamide deamination. Journal of Bacteriology, v. 200, n. 11, 1 jun. 2018. HALD, E.; LEHMANN, P.; ZIEGENHORN, J. Molar Absorptivities of β-NADH and β- NAD at 260 nm. Clinical Chemistry, v. 21, n. 7, p. 884–887, 1 jun. 1975. HALLOWS, W. C.; LEE, S.; DENU, J. M. Sirtuins deacetylate and activate mammalian acetyl-CoA synthetases. [s.l: s.n.]. Disponível em: . Acesso em: 16 abr. 2020. HAUF, W. et al. Interaction of the Nitrogen Regulatory Protein GlnB (PII) with Biotin Carboxyl Carrier Protein (BCCP) Controls Acetyl-CoA Levels in the Cyanobacterium Synechocystis sp. PCC 6803. Frontiers in Microbiology, v. 7, n. OCT, p. 1700, 26 out. 2016. HOLM, A. K. et al. Metabolic and transcriptional response to cofactor perturbations in Escherichia coli. Journal of Biological Chemistry, 2010. HOU, J. et al. Metabolic impact of redox cofactor perturbations in Saccharomyces cerevisiae. Metabolic Engineering, v. 11, n. 4–5, p. 253–261, 1 jul. 2009. HOUTKOOPER, R. H. et al. The secret life of NAD+: An old metabolite controlling new metabolic signaling pathways. Endocrine Reviews, v. 31, n. 2, p. 194–223, 2010. HUANG, H. et al. Electron bifurcation involved in the energy metabolism of the acetogenic bacterium Moorella thermoacetica growing on glucose or H2 plus CO2. Journal of Bacteriology, v. 194, n. 14, p. 3689–3699, jul. 2012. HUANG, N. et al. Bifunctional NMN Adenylyltransferase/ADP-Ribose Pyrophosphatase: Structure and Function in Bacterial NAD Metabolism. Structure, v. 16, n. 2, p. 196–209, 12 fev. 2008. HUERGO, L. F. et al. ADP-ribosylation of dinitrogenase reductase in Azospirillum brasilense is regulated by AmtB-dependent membrane sequestration of DraG. Molecular Microbiology, v. 59, n. 1, 2006a. HUERGO, L. F. et al. ADP-ribosylation of dinitrogenase reductase in Azospirillum brasilense is regulated by AmtB-dependent membrane sequestration of DraG. Molecular microbiology, v. 59, n. 1, p. 326–37, jan. 2006b. HUERGO, L. F. et al. Interactions between PII proteins and the nitrogenase regulatory enzymes DraT and DraG in Azospirillum brasilense. FEBS letters, v. 580, n. 22, p. 5232–6, 2 out. 2006c.

133

HUERGO, L. F. et al. P II signal transduction proteins: Pivotal players in post- translational control of nitrogenase activity. Microbiology, v. 158, n. 1, 2012. HUERGO, L. F.; CHANDRA, G.; MERRICK, M. PII signal transduction proteins: Nitrogen regulation and beyond. FEMS Microbiology Reviews, v. 37, n. 2, 2013a. HUERGO, L. F.; CHANDRA, G.; MERRICK, M. P(II) signal transduction proteins: nitrogen regulation and beyond. FEMS microbiology reviews, v. 37, n. 2, p. 251–83, mar. 2013b. HUERGO, L. F.; DIXON, R. The Emergence of 2-Oxoglutarate as a Master Regulator Metabolite. Microbiology and Molecular Biology Reviews, v. 79, n. 4, p. 419–435, dez. 2015. HURLEY, J. K. et al. Structure-function relationships in Anabaena ferredoxin/ferredoxin:NADP+ reductase electron transfer: Insights from site- directed mutagenesis, transient absorption spectroscopy and X-ray crystallographyBiochimica et Biophysica Acta - BioenergeticsElsevier, , 22 abr. 2002. ISHIKAWA, Y.; KAWAI-YAMADA, M. Physiological significance of NAD kinases in cyanobacteria. Frontiers in Plant Science, v. 10, n. June, p. 1–8, 2019. IWAMOTO, Y. et al. Tryptophan 2,3-dioxygenase in Saccharomyces cerevisiae. Canadian Journal of Microbiology, v. 41, n. 1, p. 19–26, 1 jan. 1995. JANKEVICIUS, G. et al. The Toxin-Antitoxin System DarTG Catalyzes Reversible ADP-Ribosylation of DNA. Molecular Cell, v. 64, n. 6, p. 1109–1116, 15 dez. 2016. JÄSCHKE, A. et al. Cap-like structures in bacterial RNA and epitranscriptomic modification. Current Opinion in Microbiology, v. 30, p. 44–49, 2016. JAUCH, R. et al. Structures of Escherichia coli NAD synthetase with substrates and products reveal mechanistic rearrangements. The Journal of biological chemistry, v. 280, n. 15, p. 15131–40, 15 abr. 2005. JAVELLE, A; MERRICK, M. Complex formation between AmtB and GlnK: an ancestral role in prokaryotic nitrogen control. Biochemical Society transactions, v. 33, p. 170– 172, 2005. JAVELLE, A. et al. Substrate binding, deprotonation, and selectivity at the periplasmic entrance of the Escherichia coli ammonia channel AmtB. Proceedings of the National Academy of Sciences of the United States of America, v. 105, n. 13, p. 5040–5045, 1 abr. 2008. JIANG, P.; NINFA, A. J. Escherichia coli PII signal transduction protein controlling nitrogen assimilation acts as a sensor of adenylate energy charge in vitro. Biochemistry, 2007. JULIUS, C.; YUZENKOVA, Y. Bacterial RNA polymerase caps RNA with various cofactors and cell wall precursors. Nucleic Acids Research, v. 45, n. 14, p. 8282– 8290, 2017. JULIUS, C.; YUZENKOVA, Y. Metabolic cofactors act as initiating substrates for primase and affect replication primer processing. bioRxiv, p. 741967, 21 ago. 2019.

134

KACZMAREK, F. S. et al. Cloning and functional characterization of an NAD+- dependent DNA ligase from Staphylococcus aureus. Journal of Bacteriology, v. 183, n. 10, p. 3016–3024, 15 maio 2001. KANG, G. B. et al. CRYSTALLIZATION AND X-RAY ANALYSIS OF NH3- DEPENDENT NAD + SYNTHETASE FROM HELICOBACTER. v. 10, n. 4, p. 418–421, 2003. KIM, S. et al. Comparison of the effects of NADH- and NADPH-perturbation stresses on the growth of escherichia coli. Current Microbiology, v. 58, n. 2, p. 159–163, 25 fev. 2009. KLIMOVA, N. et al. Interplay between NAD+ and acetylǦCoA metabolism in ischemia- induced mitochondrial pathophysiology. Biochimica et Biophysica Acta - Molecular Basis of Disease, v. 1865, n. 8, p. 2060–2067, 2019. KLOFT, N.; RASCH, G.; FORCHHAMMER, K. Protein phosphatase PphA from Synechocystis sp. PCC 6803: The physiological framework of PII-P dephosphorylation. Microbiology, v. 151, n. 4, p. 1275–1283, 1 abr. 2005. KOBAYASHI, K. et al. Essential Bacillus subtilis genes. Proceedings of the National Academy of Sciences of the United States of America, v. 100, n. 8, p. 4678–4683, 15 abr. 2003. KONRAD, E. B.; MODRICH, P.; LEHMAN, I. R. Genetic and enzymatic characterization of a conditional lethal mutant of Escherichia coli K12 with a temperature-sensitive DNA ligase. Journal of Molecular Biology, v. 77, n. 4, p. 519– 529, 15 jul. 1973. KOSONO, S. et al. Changes in the Acetylome and Succinylome of Bacillus subtilis in Response to Carbon Source. PLOS ONE, v. 10, n. 6, p. e0131169, 22 jun. 2015. KURNASOV, O. et al. NAD Biosynthesis: Identification of the Tryptophan to Quinolinate Pathway in Bacteria. Chemistry and Biology, 2003. KURNASOV, O. V. et al. Ribosylnicotinamide kinase domain of NadR protein: Identification and implications in NAD biosynthesis. Journal of Bacteriology, 2002. LARONDE-LEBLANC, N.; RESTO, M.; GERRATANA, B. Regulation of active site coupling in glutamine-dependent NAD(+) synthetase. Nature structural & molecular biology, v. 16, n. 4, p. 421–429, 2009a. LARONDE-LEBLANC, N.; RESTO, M.; GERRATANA, B. Regulation of active site coupling in glutamine-dependent NAD + synthetase. Nature Structural and Molecular Biology, 2009b. LASKOSKI, K. et al. In vitro characterization of the NAD+ synthetase NadE1 from Herbaspirillum seropedicae. Archives of Microbiology, 2016. LAWRENCE, T. J. et al. FAST: FAST Analysis of Sequences Toolbox. Frontiers in Genetics, v. 6, n. MAY, p. 172, 19 maio 2015. LEHNMAN, I. R. DNA Ligase: Structure, Mechanism, and Function. Science, v. 186, n. 4166, p. 790–797, 29 nov. 1974. LI, J. et al. A conserved NAD+ binding pocket that regulates protein-protein interactions 135 during aging. Science (New York, N.Y.), v. 355, n. 6331, p. 1312–1317, 24 mar. 2017. LI, M. et al. Crystal structure of the catalytic domain of Pseudomonas exotoxin A complexed with a nicotinamide adenine dinucleotide analog: Implications for the activation process and for ADP ribosylation. Proceedings of the National Academy of Sciences of the United States of America, v. 93, n. 14, p. 6902–6906, 9 jul. 1996. LIN, H. Nicotinamide adenine dinucleotide: Beyond a redox coenzyme. Organic and Biomolecular Chemistry, 2007. LIU, W. et al. Protein acetylation mediated by YfiQ and CobB is involved in the virulence and stress response of Yersinia pestis. Infection and Immunity, v. 86, n. 6, 1 jun. 2018. MACEK, B. et al. Protein post-translational modifications in bacteria. Nature Reviews Microbiology, 2019. MAGNI, G. et al. Structure and function of nicotinamide mononucleotide adenylyltransferase. Current medicinal chemistry, v. 11, n. 7, p. 873–885, 2004. MCDONALD, H. M. et al. Structural adaptation of an interacting non-native C-terminal helical extension revealed in the crystal structure of NAD+ synthetase from Bacillus anthracis. Acta crystallographica. Section D, Biological crystallography, v. 63, n. Pt 8, p. 891–905, 2007. MERRICK, M. Post-translational modification of PII signal transduction proteins. Frontiers in Microbiology, v. 5, p. 763, jan. 2015. MORALES-FILLOY, H. G. et al. The 5′-NAD cap of RNAIII modulates toxin production in Staphylococcus aureus isolates. . Journal of Bacteriology, v. 202, n. 6, 23 dez. 2019. MOURE, V. R. et al. Heat stability of Proteobacterial P II protein facilitate purification using a single chromatography step. Protein Expression and Purification, v. 81, n. 1, 2012a. MOURE, V. R. et al. Heat stability of Proteobacterial PII protein facilitate purification using a single chromatography step. Protein expression and purification, v. 81, n. 1, p. 83–8, jan. 2012b. MOURE, V. R. et al. The nitrogenase regulatory enzyme dinitrogenase reductase adpribosyltransferase (DraT) is activated by direct interaction with the signal transduction protein glnb. Journal of Bacteriology, v. 195, n. 2, 2013. MOURE, V. R. et al. Regulation of nitrogenase by reversible mono-ADP-ribosylation. In: Current Topics in Microbiology and Immunology. [s.l: s.n.]. MOURE, V. R. et al. The ammonium transporter AmtB and the PII signal transduction protein GlnZ are required to inhibit DraG in Azospirillum brasilense. The FEBS Journal, v. 286, n. 6, p. 1214–1229, 12 mar. 2019. MUNNUR, D. et al. Reversible ADP-ribosylation of RNA. Nucleic Acids Research, v. 47, n. 11, p. 5658–5669, 2019. NINFA, A. J.; JIANG, P. PII signal transduction proteins: sensors of alpha-ketoglutarate that regulate nitrogen metabolism. Current opinion in microbiology, v. 8, n. 2, p. 136

168–73, abr. 2005. OLIVEIRA, M. A. S. et al. 2-Oxoglutarate levels control adenosine nucleotide binding by Herbaspirillum seropedicae PII proteins. FEBS Journal, v. 282, n. 24, 2015. OSTERMAN, A. Biogenesis and Homeostasis of Nicotinamide Adenine Dinucleotide Cofactor. EcoSal Plus, v. 3, n. 2, 2009. PACE, H. C.; BRENNER, C. The nitrilase superfamily: classification, structure and function. Genome biology, v. 2, n. 1, p. REVIEWS0001, jan. 2001. PALANCA, C.; RUBIO, V. Effects of T-loop modification on the PII-signalling protein: structure of uridylylated Escherichia coli GlnB bound to ATP. Environmental Microbiology Reports, v. 9, n. 3, p. 290–299, 1 jun. 2017. PALAZZO, L.; MIKOČ, A.; AHEL, I. ADP-ribosylation: new facets of an ancient modificationFEBS JournalBlackwell Publishing Ltd, , 1 set. 2017. Disponível em: . Acesso em: 14 mar. 2020 PANOZZO, C. et al. Aerobic and anaerobic NAD + metabolism in Saccharomyces cerevisiae. FEBS Letters, v. 517, n. 1–3, p. 97–102, 24 abr. 2002. PARK, U. E. et al. DNA ligase and the pyridine nucleotide cycle in Salmonella typhimurium. Journal of Bacteriology, v. 171, n. 4, p. 2173–2180, 1 abr. 1989. PARK, Y. W.; YEO, H. K.; LEE, J. Y. Structural Analysis of the NH 3 -dependent NAD + Synthetase from Deinococcus radiodurans. v. 35, n. 9, p. 2863–2866, 2014. PEDROSA, F. DE O.; YATES, M. G. Regulation of nitrogen fixation ( nif ) genes of Azospirillum brasilense by nifA and ntr ( gln ) type gene products. FEMS Microbiology Letters, v. 23, p. 95–101, 1984. PERINA, D. et al. Distribution of protein poly(ADP-ribosyl)ation systems across all domains of life. DNA Repair, v. 23, p. 4–16, 1 nov. 2014. PETIT, M.-A.; DUSKO EHRLICH, S. The NAD-dependent ligase encoded by yerG is an essential gene of Bacillus subtilisNucleic Acids Research. [s.l: s.n.]. Disponível em: . Acesso em: 17 mar. 2020. PLUMBRIDGE, J.; DEUTSCHER, J. Regulation of Carbon Assimilation in Bacteria. [s.l.] Elsevier Inc., 2019. POLLAK, N.; OLLE, C. D. ¨; ZIEGLER, M. The power to reduce: pyridine nucleotides- small molecules with a multitude of functions. Biochem. J, v. 402, p. 205–218, 2007. PREISS, J.; HANDLER, P. Biosynthesis of diphosphopyridine nucleotide. Journal of Biological Chemistry, 1958. RAFFAELLI, N. et al. The Escherichia coli NadR Regulator Is Endowed with Nicotinamide Mononucleotide Adenylyltransferase Activity The Escherichia coli NadR Regulator Is Endowed with Nicotinamide Mononucleotide Adenylyltransferase Activity. Jornal of Bacteriology, v. 181, n. 17, p. 1–4, 1999a. RAFFAELLI, N. et al. Synechocystis sp. slr0787 protein is a novel bifunctional enzyme endowed with both nicotinamide mononucleotide adenylyltransferase and ‘Nudix’’ hydrolase activities’. FEBS Letters, v. 444, n. 2–3, p. 222–226, 12 fev. 1999b. 137

RAFICE, S. A. et al. Oxidation of L-tryptophan in biology: A comparison between tryptophan 2,3-dioxygenase and indoleamine 2,3-dioxygenase. Biochemical Society Transactions, v. 37, n. 2, p. 408–412, 1 abr. 2009. RAJENDRAN, C. et al. Crystal structure of the GlnZ-DraG complex reveals a different form of P II -target interaction. Proceedings of the National Academy of Sciences, 2011. RESTO, M.; YAFFE, J.; GERRATANA, B. An ancestral glutamine-dependent NAD(+) synthetase revealed by poor kinetic synergism. Biochimica et biophysica acta, v. 1794, n. 11, p. 1648–53, nov. 2009. RIPPKA, R.; DERUELLES, J.; WATERBURY, J. B. Generic assignments, strain histories and properties of pure cultures of cyanobacteria. Journal of General Microbiology, v. 111, n. 1, p. 1–61, 1979. RIZZI, M. et al. Crystal structure of NH3-dependent NAD+ synthetase from Bacillus subtilis. The EMBO journal, v. 15, n. 19, p. 5125–5134, 1996. ROCCO, C. J. et al. Construction and use of new cloning vectors for the rapid isolation of recombinant proteins from Escherichia coli. Plasmid, v. 59, n. 3, p. 231–237, maio 2008. RODIONOV, D. A. et al. Transcriptional regulation of NAD metabolism in bacteria: NrtR family of Nudix-related regulators. Nucleic Acids Research, v. 36, n. 6, p. 2047–2059, 2008a. RODIONOV, D. A. et al. Transcriptional regulation of NAD metabolism in bacteria: genomic reconstruction of NiaR (YrxA) regulon. Nucleic Acids Research, v. 36, n. 6, p. 2032–2046, abr. 2008b. RODIONOV, D. A et al. Transcriptional regulation of NAD metabolism in bacteria: NrtR family of Nudix-related regulators. Nucleic acids research, v. 36, n. 6, p. 2047–59, abr. 2008c. RODIONOVA, I. A. et al. Metabolic and bactericidal effects of targeted suppression of NadD and NadE enzymes in mycobacteria. mBio, 2014. RODRIGUES, T. E. et al. Search for novel targets of the PII signal transduction protein in Bacteria identifies the BCCP component of acetyl-CoA carboxylase as a PII binding partner. Molecular microbiology, v. 91, n. 4, p. 751–61, mar. 2014. RODRIGUES, T. E. et al. Fatty acid biosynthesis is enhanced in Escherichia coli strains with deletion in genes encoding the PII signaling proteins. Archives of Microbiology, 2019. RUPPERT, U. et al. The novel protein phosphatase PphA from Synechocystis PCC 6803 controls dephosphorylation of the signalling protein PII. Molecular Microbiology, v. 44, n. 3, p. 855–864, 7 maio 2002. SANT’ANNA, F. H. et al. The PII superfamily revised: a novel group and evolutionary insights. Journal of molecular evolution, v. 68, n. 4, p. 322–36, abr. 2009. SANTOS, A. et al. NAD+ biosynthesis in bacteria is controlled by global carbon/nitrogen levels via PII signaling. v. 295, n. 9, p. 1–12, 2020.

138

SANTOS, A. R. S. et al. Kinetics and structural features of dimeric glutamine- dependent bacterial NAD synthetases suggest evolutionary adaptation to available metabolites. Journal of Biological Chemistry, v. 293, n. 19, 2018. SASSETTI, C. M.; BOYD, D. H.; RUBIN, E. J. Genes required for mycobacterial growth defined by high density mutagenesis. Molecular Microbiology, v. 48, n. 1, p. 77–84, 25 mar. 2003. SCHAFER, F. Q.; BUETTNER, G. R. Redox environment of the cell as viewed through the redox state of the glutathione disulfide/glutathione coupleFree Radical Biology and Medicine, 1 jun. 2001. Disponível em: . Acesso em: 12 abr. 2020 SHARMA, V.; GRUBMEYER, C.; SACCHETTINI, J. C. Crystal structure of quinolinic acid phosphoribosyltransferase from Mycobacterium tuberculosis: A potential TB drug target. Structure, v. 6, n. 12, p. 1587–1599, 15 dez. 1998. SHEN, W. et al. SeqKit: A Cross-Platform and Ultrafast Toolkit for FASTA/Q File Manipulation. PLOS ONE, v. 11, n. 10, p. e0163962, 5 out. 2016. SORCI, L. et al. Nicotinamide mononucleotide synthetase is the key enzyme for an alternative route of NAD biosynthesis in Francisella tularensis. Proceedings of the National Academy of Sciences of the United States of America, v. 106, n. 9, p. 3083–8, 3 mar. 2009. SORCI, L. et al. Characterization of bacterial NMN deamidase as a Ser/Lys hydrolase expands diversity of serine amidohydrolases. FEBS Letters, v. 588, n. 6, p. 1016– 1023, 18 mar. 2014. SORCI, L.; RUGGIERI, S.; RAFFAELLI, N. NAD homeostasis in the bacterial response to DNA/RNA damage. DNA repair, 12 ago. 2014. SPÄT, P.; MACEK, B.; FORCHHAMMER, K. Phosphoproteome of the cyanobacterium Synechocystis sp. PCC 6803 and its dynamics during nitrogen starvation. Frontiers in Microbiology, v. 6, n. MAR, p. 248, 31 mar. 2015. SRISKANDA, V. et al. Mutational analysis of Escherichia coli DNA ligase identifies amino acids required for nick-ligation in vitro and for in vivo complementation of the growth of yeast cells deleted for CDC9 and LIG4Nucleic Acids Research. [s.l: s.n.]. SRISKANDA, V. A second NAD+-dependent DNA ligase (LigB) in Escherichia coli. Nucleic Acids Research, v. 29, n. 24, p. 4930–4934, 2001. SRISKANDA, V.; MOYER, R. W.; SHUMAN, S. NAD+-dependent DNA Ligase Encoded by a Eukaryotic Virus. Journal of Biological Chemistry, v. 276, n. 39, p. 36100–36109, 28 set. 2001. SRISKANDA, V.; SHUMAN, S. Conserved residues in domain Ia are required for the reaction of Escherichia coli DNA ligase with NAD+. Journal of Biological Chemistry, v. 277, n. 12, p. 9695–9700, 22 mar. 2002. STARAI, V. J. et al. Sir2-dependent activation of acetyl-CoA synthetase by deacetylation of active lysine. Science, v. 298, n. 5602, p. 2390–2392, 20 dez. 2002.

139

STARAI, V. J. et al. Short-Chain Fatty Acid Activation by Acyl-Coenzyme A Synthetases Requires SIR2 Protein Function in Salmonella enterica and Saccharomyces cerevisiae. [s.l: s.n.]. STRØMLAND, Ø. et al. Keeping the balance in NAD metabolism. Biochemical Society Transactions, n. December 2018, p. BST20180417, 2019. SYMERSKY, J. et al. NH 3 -dependent NAD + synthetase from Bacillus subtilis at 1 Å resolution. Acta Crystallographica Section D Biological Crystallography, v. 58, n. 7, p. 1138–1146, 2002a. SYMERSKY, J. et al. NH3-dependent NAD+ synthetase from Bacillus subtilis at 1 Å resolution. Acta Crystallographica Section D: Biological Crystallography, v. 58, n. 7, p. 1138–1146, 1 jul. 2002b. TAUTENHAHN, R. et al. XCMS online: A web-based platform to process untargeted metabolomic data. Analytical Chemistry, v. 84, n. 11, p. 5035–5039, 5 jun. 2012. TEDESCHI, G. et al. Structural characterization of l-aspartate oxidase and identification of an interdomain loop by limited proteolysis. European Journal of Biochemistry, v. 260, n. 3, p. 896–903, 5 mar. 1999. TERNES, C. M.; SCHÖNKNECHT, G. Gene transfers shaped the evolution of de novo NAD+ biosynthesis in eukaryotes. Genome Biology and Evolution, v. 6, n. 9, p. 2335–2349, 2014. TRUAN, D. et al. A New PII Protein Structure Identifies the 2-Oxoglutarate Binding Site. Journal of Molecular Biology, v. 400, n. 3, 2010. VAN DEN BERG, S. et al. Improved solubility of TEV protease by directed evolution. Journal of Biotechnology, v. 121, n. 3, p. 291–298, 2006. VAN HEESWIJK, W. C.; WESTERHOFF, H. V; BOOGERD, F. C. Nitrogen assimilation in Escherichia coli: putting molecular data into a systems perspective. Microbiology and molecular biology reviews : MMBR, v. 77, n. 4, p. 628–95, 1 dez. 2013. VASSILOPOULOS, A.; WANG, R.-H.; GIUS, D. Mammalian Sirtuins, Cellular Energy Regulation, and Metabolism, and Carcinogenesis. [s.l.] Elsevier Inc., 2018. VVEDENSKAYA, I. O. et al. CapZyme-Seq Comprehensively Defines Promoter- Sequence Determinants for RNA 5′ Capping with NAD+. Molecular Cell, v. 70, p. 553–564, 2018. WALCZAK, K. et al. Kynurenic acid and cancer: facts and controversiesCellular and Molecular Life SciencesBirkhauser Verlag AG, , 28 out. 2019. WANG, H. et al. Energy shifts induce membrane sequestration of DraG in Rhodospirillum rubrum independent of the ammonium transporters and diazotrophic conditions. FEMS Microbiology Letters, v. 365, n. 16, 1 ago. 2018. WANG, M. M.; YOU, D.; YE, B. C. Site-specific and kinetic characterization of enzymatic and nonenzymatic protein acetylation in bacteria. Scientific Reports, v. 7, n. 1, 1 dez. 2017. WEINERT, B. T. et al. Acetyl-Phosphate is a critical determinant of Lysine Acetylation in E.coli. Molecular Cell, v. 51, n. 2, p. 265–272, 2013. 140

WILKINSON, A.; DAY, J.; BOWATER, R. Bacterial DNA ligases. Molecular Microbiology, v. 40, n. 6, p. 1241–1248, 1 jun. 2001. WINZ, M. L. et al. Capture and sequencing of NAD-capped RNA sequences with NAD captureSeq. Nature Protocols, v. 12, n. 1, p. 122–149, 1 jan. 2017. WU, C. H. PIRSF: family classification system at the Protein Information Resource. Nucleic Acids Research, 2004. YANG, T.; SAUVE, A. A. NAD metabolism and sirtuins: Metabolic regulation of protein deacetylation in stress and toxicity. AAPS Journal, v. 8, n. 4, p. E632–E643, 6 out. 2006. YANG, Z. et al. Aspartate dehydrogenase, a novel enzyme identified from structural and functional studies of TM1643. Journal of Biological Chemistry, v. 278, n. 10, p. 8804–8808, 7 mar. 2003. YUAN, J. et al. Metabolomics-driven quantitative analysis of ammonia assimilation in E. coli. Molecular Systems Biology, 2009. ZETH, K.; FOKINAS, O.; FORCHHAMMER, K. Structural basis and target-specific modulation of ADP sensing by the Synechococcus elongatus PII signaling protein. Journal of Biological Chemistry, v. 289, n. 13, p. 8960–8972, 2014. ZHANG, J. et al. Lysine acetylation is a highly abundant and evolutionarily conserved modification in Escherichia coli. Molecular and Cellular Proteomics, v. 8, n. 2, p. 215–225, fev. 2009. ZHANG, Y. et al. Crystal structure and mechanism of tryptophan 2,3-dioxygenase, a heme enzyme involved in tryptophan catabolism and in quinolinate biosynthesis. Biochemistry, v. 46, n. 1, p. 145–155, 9 jan. 2007. ZHANG, Y. et al. Mutagenesis and functional characterization of the four domains of GlnD, a bifunctional nitrogen sensor protein. Journal of Bacteriology, v. 192, n. 11, p. 2711–2721, jun. 2010. ZHANG, Z.; MILIAS-ARGEITIS, A.; HEINEMANN, M. Dynamic single-cell NAD(P)H measurement reveals oscillatory metabolism throughout the E. coli cell division cycle. Scientific Reports, 2018. ZHAO, K.; CHAI, X.; MARMORSTEIN, R. Structure and Substrate Binding Properties of cobB, a Sir2 Homolog Protein Deacetylase from Escherichia coli. Journal of Molecular Biology, v. 337, n. 3, p. 731–741, 26 mar. 2004.

141